AI War 2:The Paradigm Shift

Next Release Notes

AI War 2: Finalizing Multiplayer

What was this phase all about?

With multiplayer progressing, and DLC2 something that we were pouring increasing effort into, Chris found himself fairly dissatisfied with the state of the base game. After a few rounds of potential design shifts to make the game flow better, this is what we settled on. Ultimately this was supposed to last for about a month, but it wound up taking over four months and touching almost every aspect of the game. A more detailed writeup can be viewed here.

The amount of refinement and polish in this phase was also really intense, and the final product is something we're super proud of.

New Video Tutorials

• Be sure to check these out if you're coming back after a long break!

• Strategic Sage's Taste It Again, For The First Time: https://youtu.be/fp0PSvZlKso

• eXplorminate Starter's Guide: https://www.youtube.com/watch?v=noDhOL51rPA&list=PL8cNlze3fGQj1SaZWDqcVck_RtM7_t-ui&index=1

2.800 The Paradigm Shift

(Released April 14th, 2021)

• The Fallen Spire campaign now has real lore in the game, displayed via the Journal
  • This was done by adapting much of the AIWC lore (thanks Keith!) to ai war 2.

• If you are just tuning in now, be sure to check the videos above or our writeup here to get you up to speed.

Beta 2.775 Prepare For Beta Exit

(Released April 11th, 2021)

• Add Settings to allow people to modify the starting hotkeys for Battlestations and Support fleets for people who prefer certain keybindings for fleet types
  • Thanks to Badger for adding.

• In the Esc Menu, details about each AI's waves are no longer visible unless you hold down CTL.
  • Thanks to Badger for adding.

• Some improvements to the How To Play section to explain deepstrikes and AI Reserves and Hunter better
  • Thanks to some discussion with several people including Apthorpe and Ithuriel, and to Badger for adding.

Balance

• Balance adjustment to two citadels:
  • Impactor reload time 9 -> 6 seconds
  • Drainer reload time 3 -> 5 seconds and health drain percent raised from 50% to 70%
  • End result of changes is that the impactor actually has better damage versus high value targets than the drainer which has held the advantage of being able to heal itself on top of previously also being the best overall damage in that category.
  • Thanks to CRCGamer for updating.

• Raptor reduced metal cost from 1800 to 1200
  • Velociraptor health increased to from 3700 to 4200, 50% more than raptor.
  • Raptor got nerfed a long time ago by the anti-cheese changes to cloaking, which led raptor to break stealth quickly when attacking planet. With their expensive cost compared to their health, it was unwise to use them offensively. This change should make ambush much more usable.
• Railpod cap increased from 19 to 70. Health reduced from 1200 to 400. Damage reduced from 160 to 40. Added x4 damage if on planet for less than 15 seconds. Cost reduced from 250 to 200. Railpod tech changed from "Ambush, Light" to "Ambush, Artillery"
  • These various changes gives Ambush an offensive-minded ship. It's meant to be used as an artillery battery on enemies planets, with a somewhat strong alpha-strike when exiting a wormhole. On defense building the full cap is less useful as it'll lose 3/4 of its firepower after 15 seconds of existing.
  • Thanks to ArnaudB for updating.

• Botnet damage reduced by 33%. It gains a x2 modifier against strikecraft, lose it's x1/100 multiplier against non-strikecrafts and start at 40 shots per salvo increasing by 20 with every mark. Description update to tell player of this.
  • After various changes to make this more usable, this will be used till zombification mechanic get an update. Current botnet is too pricy for its limitation. This should make it a valuable asset worth paying 40 AIP for.
  • The numerous units immune to zombification are enough of a counter against it. It having less damage against non-strikecraft is another balance lever, since with its base damage reduction it won't be good against big threats.
  • Thanks to ArnaudB for updating.

• Nanocaust Center (and Home version) now give 3x speed to its allies on planet. This should help it with dying from attrition once it gets big enough. It was a very silly way for our commonplace galatic plague to be defeated.
  • there is plenty of counters against this: distraction, sniping the Center, transport and so on.
  • This should make the Nanocaust much more unique.
  • Thanks to ArnaudB for updating.

• Makeshift drone has its health reduced from 10k hull 10k shield to 3k hull and 0 shield. At mark 1 those values were ridiculous for a kamikaze drone.
  • Thanks to ArnaudB for updating.

• Science cost for direct upgrade flattened across the board. Culled a number of 50 science for non-transports upgrade, and make sure it you take multiple upgrade it'll be neat again.
  • Max science cost for direct upgrade is now 3000. With diminishing return in play, upgrade for more than 3k science were pointless options.
  • Battlestation cost 500, 1500, 2500 science to directly upgrade instead of 450, 1400, 2800. This should be neater.
  • Command Stations third direct upgrade is now 2000 instead of 3000. This should make scaling easier to manage.
  • Thanks to ArnaudB for updating.

Bugfixes

• Make it harder for player-allied marauders to kill ai command stations by mistake
  • Thanks to Ecthelon for reporting, and to Badger for adding.

• The 'watched fleets above ships' setting for the Planet menu now persists through application close/reopen. This was very annoying
  • Thanks to Daw11 for reporting and Badger for fixing.

• The AI no longer sends bonus hunter ships against human-allied marauders at AI difficulties <= 5. At higher difficulties, this mechanic is still enabled but it is nerfed significantly
  • Thanks to a number of people for a balance discussion on discord. I believe this was prompted by some gameplay by Ithuriel. And thanks to Badger for fixing.

• Some cleanup for Objectives
  • The collect science tooltip no longer mentions hacking AI-owned planets.
  • The "collect science and hacking for free" objectives no longer appear if capturing the planet would cost AIP.
  • Remove some extraneous uses of the word "lets"
  • Thanks to Badger for doing this.

• Fixed some harmless but annoying error messages that could pop up when wormhole invasion data was saved, if it was in just the wrong state at the time.
  • Thanks to Ithuriel0 for reporting.

• Fixed an exception that could happen in CalculateEffectiveFullFleetStrength_OutOfAbsoluteTotal_PlayerFleetsOnly() in rare cases due to thread racing.
  • Thanks to Zweihander2021 for reporting.

Mod Updates

• AMU:
  • Put in a fix for the Auto-Juggle Energy Assistant not working. Apparently this was due to some changes in the Vanilla code for placing self-building structures, but it's hard to say exactly.
    • Modifying AJEA to re-adjust to the game command made it work - as well as the More System Defenders auto-placing code as well, which was still using the same method, despite it not working before. Extremely strange, but it's tested and works again.
    • Thanks to Zane for reporting on Discord.
• Kaizers Marauders:
  • Finally (hopefully) killed the bug where saving a game throws an exception due to the Grace Period End being out of its bounds. Hopefully also won't kill savegames.
    • Thanks to a lot of people for reporting, most recently Zachariath on Discord.
    • I apologize for taking so long to fix this. It was a very minor bug and I was always shy to put in serialization gates due to being unable to check whether the fix worked, didn't work or totally killed savegames until the next game version dropped, but hopefully it will work. If not please notify me.

Beta 2.773 Civvies Hotfix

(Released April 9th, 2021)

• Fixed a civilian industries error in the last two builds.
  • Thanks to Zane for reporting.

Beta 2.772 Wormhole Hotfix

(Released April 9th, 2021)

• Fix an issue that crept into the last version where player units could not go through wormholes and would error if you tried.
  • Amusingly, enemies and NPCs in general could still use wormholes just fine.
  • Thanks to lots of folks for reporting.

Beta 2.771 Critical Performance Mass

(Released April 8th, 2021)

• Some more planet names in the standard set, in a bit of different linguistic styles to balance out the overabundance of Japanese stylings that had been piling up. Should be more balanced now in terms of feel from a cross section of cultures.

• The Macrophage Histiocytes mod (both the base one and the spire integration one) have been updated to work with the latest beta code, and also to avoid code exceptions on MP clients.
  • Civilian Industries has also been updated to avoid exceptions on MP clients.
  • Thanks to Zegma for reporting the Histiocytes MP exception.

• Fixed a bug in the last couple of versions (since we added networking channels), where LiteLetLib was broken because it was not configured to have channels. It is now set up to have up to 8 channels, which should be sufficient (we only use 4), and so that should work again. Requires testing, however.
  • Thanks to Dontgiveitup for the report.

• Fixed a pretty minor issue where the targeting background thread was often allocating far more SetTargetList and SetFRDTarget commands (and then returning them to the pool) than it needed to.
  • There was some extra CPU load and cross-thread contention from this, mainly in really large games.

• The "GetPraetorianGuardExternal was null on faction AI Praetorian Guard" error should now be fixed.
  • Thanks to ParadoxSong for reporting.

• AMU:
  • Adds new rates of fire to be used by ships, scaling it steps of 50 seconds from 50 to 1000.
  • Fixed a bug in the stringification of the planet name and optionally the mark level as well, and made the function much more efficient.
    • This fixes the Kaizers Marauders Superstructure Notifier short text still displaying the mark level of the planet even if AI-owned. It was usually way too long for the tiny tooltip text space.
• Kaizers Marauders:
  • Reduced the energy cost of defensive structures:
    • Consumer Turret: 2500 -> 2000
    • Breach Torpedo Turret: 3000 -> 2750
    • Beam Sniper Turret: 4000 -> 1750
    • Fortified Forceshield Generator: 20000 -> 9000
    • Phase Cannon Turret: 18000 -> 13000
    • Defense Post: 55000 -> 30000

Major Performance Regression Fix

• Shots are no longer saved into savegames under any circumstances (though they are sent across the network still), and in existing saves they are not loaded from disk.
  • There's no situation where these are anything but a waste of space, really. And in the case of a memory leak with shots not being cleared up like we have in the prior build, they are REALLY a waste of space.
  • Thanks to Sigma7 for the save that really demonstrated the need for this.

• Okay, wow! The ConcurrentPool<> from last build has been returned to instead using the TimeBasedPool<>. The switch away from that was causing immense amounts of lag, and potentially some gameplay errors as well.
  • Essentially, there is a lot more cross-thread contention when you have everything adding to and pulling from the same list. Additionally, the extra pressure of using a ConcurrentQueue rather than a ConcurrentBag internally was apparently a huge hit.
  • This is one of those things where we just had to try it, and it rears its head a lot more in certain games than others. This was also certainly an issue that tied into the "way too many shots hanging around" problem, but even in the absence of all shots the game was extremely laggy without any shots at all!
  • Thanks to Metrekec and others for reporting.

• Shots now die after 20 seconds of existing for any reason, not 45.
  • There's no reason for a shot to last more than 20 seconds.
  • Additionally, if a shot has a null target, it now only lasts 5 more seconds, not 10.
  • In general, however, the huge numbers of extra shots (70,000 in one save!) have disappeared in the new build of the game. It looks like a huge portion of that was really just the ConcurrentPool<> issue.

Further Performance Hunting

• Added a new galaxy option under the performance tab (this is advanced): Ship Systems To Calculate Per Targeting Pass. min 50, default 500, max 1000
  • The lower this is, the slower ships are to react to firing at new targets that are in their range, but overall this is something that matters more in very huge games. The higher this is, the faster the ships will react, but the more likely you are to have hitching in your gameplay if there are truly tons of ships all getting new orders all once.
  • The CPU load from calculating this is only on the host in multiplayer, but the CPU load of actually integrating the data into the main game exists on the host and all clients.
  • The default has been 500 for a super long time, but if you're in a very large game you might consider lowering this.

• Added a new galaxy option under the performance tab (this is advanced): Max Targets Per Ship System Per Targeting Pass. min 3, default 15, max 50
  • The lower this is, the shorter a working list of targets each ship has; this keeps things moving more quickly, but if all their targets are exhausted, then they have to wait for a new targeting pass. The higher this is, the more redundancy there is for if a lot of their targets die before the next targeting pass, but also the more CPU load there is.
  • The CPU load from calculating this negligible on any machine, but the CPU load of integrating the data into the main game exists on the host and all clients.
  • The default has been 25 for a super long time, but we have now lowered this to 15 as the default.

• Added a new galaxy option under the performance tab (this is advanced): Extreme: Shots All Insta Hit
  • This vastly alters gameplay, so you should only turn it on if you're having dire performance issues or need to test something. With this setting on, shots are never fired; instead, they instantly just hit their targets.
  • This makes battles look very unspectacular, visually, and it also changes balance in a variety of subtle ways. This is mostly for playing on true toasters or testing something when shots are working improperly.

• Updated AMU to have all local code file paths and references, so that it can be compiled against the latest version of the game on anyone's machine.
  • Recompiled AMU so that if there were any inconsistencies with some of the changes that have been made to the base game, it won't break.

• The "MoveManyToOnePoint" command has been split into a ton of different sub-groups, like we previously did with "SetWormholePath"
  • The idea is that if something is flooding the game with these sort of commands, we want to have a bit more of an idea where that's coming from.
  • This may break some code-oriented mods, but we want them to update themselves to specify what kind of movement orders they are giving so that we can tell if one is misbehaving.
  • All of the StarKelp mods have been updated to handle this (Macrophage stuff, Civilian Industries), as has AMU. The SirLimbo mods are also confirmed updated.

• There was a substantial amount of "queue a journal entry log" command spam that was happening every second. The more factions you had, the more it would spam. This was slowing things down in both multiplayer and singleplayer, actually to a surprising degree even in single player.

• Minor performance improvement to MoveManyToOnePoint_NPCPatrol so that it doesn't log extra entries and then put them back.

• The frequency with which stacks are evaluated (split or combined) is now every 4 seconds, or with 1 second gaps, whichever was slower. It used to be every 1 second, or every 0.5 seconds, whichever was slower.
  • There was absolutely no good reason for it to run that frequently, and in large games all it was doing was creating a lot of churn and notable slowdown for no real tangible benefit. The stack splitting and combination code is great, but it just doesn't need to be that frequent.

Beta 2.770 Multiplayer Metal

(Released April 7th, 2021)

• Changed things up so that FCEs in existing saves won't be removed. Mainly important for non-beta-branch ongoing games.
  • Thanks to a number of folks for reporting.

• New Planet and Fleet naming style options that are Roman names.
  • Thanks to Badger for adding.

• Changed description of Engineer and Interplanetary Engineer to state they are given only to Command Stations and Citadels/Battlestations
  • Thanks to ArnaudB for adding.

• Added a new ConcurrentPool<> class, which is a lot like our TimeBasedPool<> class, except it only has one ConcurrentBag inside of it, and it doesn't hold its contents for 1+ seconds before using them.
  • This is now being added to metal flows, entity orders, shots, and beam weapon visual displays.
  • This should notably lower the CPU use of processing large numbers of those, and also keep the pools themselves smaller, particularly in the case of shots.
  • It is possible that we will start having multiple threads grabbing a single object of one of those types and thinking that it is two different things.
    • To minimize the chance of THAT, we now are using an internal ConcurrentQueue for these rather than a ConcurrentBag.
  • If we see wild and wooly new bugs, it's easy for us to swap back to using some or all of these the way they used to be. All of these only had a 1 second delay except for shots, which had a 5 second delay. We'll see what happens.

Balance

• Area minefield has its strength multiplier reduced from 2 to 0.2.
• Paralysis minefield has its strength multiplier reduced from 20 to 0.2.
  • Might have been a way to prevent the AI from underestimating minefields, but the numbers are nonsense. Test with this.
  • Thanks to ArnaudB for adjusting.

• Autobomb and all its variants (including ram and Nanoswarm) have their strength multiplier reduced from 0.7 to 0.3.
  • This should hopefully lead AI forces to be less confident about lauching hundreds strength worth of kamikaze ships into an hopeless battles. Especially during waves.
  • This should hopefully cut down on eyes overreacting to a melee-focusing fleet, as currently it makes melee mostly worthless.
  • Thanks to ArnaudB for adjusting.

• *Increased reward for killing an Overlord phase 2 to 10000 Science and 60 Hacking Points, like it used to before.
  • Thanks to ArnaudB for adjusting.

• For player/minor faction allied scourge, make builders more slowly
  • The need to make lots of builders immediately meant that the first bunch of fireteams would always be on escort duty. This made the scourge not really do much for the first 10-30 minutes or so, if there aren't already a bunch of clear planets near the Scourge.
  • Thanks to Badger for adjusting.

• Updated the Marauder Outpost weapon system to have a 3 second reload instead of 30. Additionally adjusted base damage per shot to 360 from 120. Furthermore increased its attack range from 3K to 4K. It now has just as much DPS as a single one of the Marauder Mk1 Concussion Turrets (600) instead of a practically non-existent 20 DPS.
  • Thanks to CRCGamer for adjusting.

• Adjusted Marauder Outpost spawning time forward a few seconds ahead given a report from Badger about the game complaining about discrepancies in a MP game.
  • Thanks to CRCGamer for adjusting.

• Remembered that a Spire variant of the Raider frigate exists. Now updated said variant to have the same increased health benefits the other two versions got. Additionally increased the damage on its weapon by 20%.
  • Thanks to CRCGamer for adjusting.

• Gravity Generator has 0.5 slow effect at all marks. The effect of multiple generators still stacks.
  • This was an excessively powerful anti-melee for both players and AI. Now you need 3 or 4 generators together to achieve the minimum 80 speed on all ships. Hacking generators on the ODSS is more useful, and melee won't be utterly useless on any Mark VI or VII AI world.
• Effect range still increase with marks.
  • Thanks to ArnaudB for adjusting.

• Gravity guardian has 0.5 slow at all marks.
  • To mirror gravity generator which no longer scale per mark.
  • This was also overpowered since the effect stack.
  • Thanks to ArnaudB for adjusting.

• Concussion Corvette
  • Metal cost per ship 2700 -> 1800, MK7 HP to metal ratio now 7.2 instead of 4.8
  • Speed raised from 400 -> 700
  • Now cheaper to replace and not as glacially slow. They only situationally can deal big damage.
  • By leaving the individual HP fairly low they are still glass cannons.
  • Thanks to CRCGamer for adjusting.

• Inhibiting Tesla Corvette
  • Max targets hit per shot 20 -> 10
  • Reload cycle 17 seconds -> 9 seconds
  • Weapon jam efficiency per "hit" cut in half since its being applied twice as often.
  • Resulting change in DPS is the range went from 5-101K to 9.5-95K and is less likely to be wasted on less than half of max targets hit.
  • Still needs to be transport dropped on target or helped with FRS Turbo Raider to get efficient.
  • Rate of fire also carries over to the Zapper variant but that still remains nigh unusable for human players because of costs.
  • Thanks to CRCGamer for adjusting.

• MLRS Corvette
  • Weapon damage bonus has been reworked. Previous bonus was x4 versus targets <= 30% hull health
  • Now is x2 versus targets <= 50% hull health and 1.5x versus targets hull with <1% shield remaining
  • From testing that makes the MLRS Corvette oppressive against other strikecraft but it remains a glass cannon with a HP to metal ratio of 6.17
  • Thanks to CRCGamer for adjusting.

• Changes to seeding in multi-empire games:
  • Previously, distribution nodes only would spawn in single-empire games. They now spawn with any number empires.
  • Previously, co-processors would only seed if you had 3 or fewer empires. Now they seed with any number of empires.
    • NOTE! These only seed in games with multiple AIs, so they still may not show up in general if you're only facing one AI.
  • Previously, 2 data centers were removed for every human empire beyond the first. This is no longer the case. That turned out to be needlessly punitive rather than a proper balance lever.
  • Same thing for major data centers; previously it removed 1 for every human empire beyond the first. Now it's the same regardless of the number of players.
  • If there's anything we need to do to make multiplayer harder to compensate for this, then we can, but hopefully most people will play against a larger number of AIs, which is inherently harder. But that is an option for the player, versus being railroaded into higher AIP games.
  • Note: all of these changes will only affect campaigns that are started new in the new version. Existing maps have whatever they already had.
  • Thanks to Arides for raising this issue.

Forcefields

• Forcefields are taking a nerf, as they gained 3x shield health at Mark II and had 5.6x shield health at Mark III. That allowed logistical planets to tank anti-shield wave without tractor or gravity being used, or for logistical to survive a 250 strength wave, while having half that strength, for far too long.
• Forcefields shield increases per mark changed from "1,3.2,5.8,7.2,8.2,9,10" to "1,2,3,4.5,6.0,8,10". Science cost changed from "2500,4750,10500,11500" to "1500,1500,2250,2250,3000,3000".
  • You get the shield value of Mark III at Mark V. It's a good indication of how busted early marks were, and how bad later marks were.
  • Unlike most tech, forcefields get better at higher. With the higher cost it just means the tech remain cost-effective at all marks.
  • The reason behind the huge values at high mark is so players can have very powerful forcefields during high-AIP games, like in Fallen Spire campaigns.
  • Thanks to ArnaudB for adjusting.

Command Stations And Energy

• Logistical Command Station got a HUGE engineering buff, similar to the "Powerful Command Station" mod. This makes the station very helpful as an adjacent planet to reinforce fleets.
  • Factories count increased from 2 to 4.
  • Engineers count increased from 14 to 21. (Engineer hack gives it half of military station, this helps it out.)
  • Thanks to ArnaudB for adjusting.

• Military Station now has regular Forcefields again, since harmonic bypassing effect should make immobile forcefields redundant.
  • This is to make the nerf to forcefields less painful to players too.
  • Thanks to ArnaudB for adjusting.

• Military Command Station has factory count decreased from 3 to 2.
  • Thanks to ArnaudB for adjusting.

• Changed Economical Command tech science costs to be cheaper early on and scale linearly.
  • Changed Military Command tech science costs to scale linearly.
  • In both cases, doesn't actually reduce the total science costs of all marks, but make the techs competitive with direct upgrade, even for low-middle AIP games.
  • Thanks to ArnaudB for adjusting.

• Economic Command Station reduced energy gain per mark from 240k to 180k. It's however cheaper to mark up with techs.
  • Thanks to ArnaudB for adjusting.

• High Energy Transport reduced Hacking Points cost to transform into from 60 to 30.
  • Thanks to ArnaudB for adjusting, and a variety of folks for noting the HaP cost being too high.

Bugfixes

• Fix a bug where there could be too few Praetorian Guards in multi-AI games
  • Reported by Metrekec, fixed by Badger.

• Fixed a bug in the prior version that was still seeding Fleet Capacity Extenders.
  • Thanks to Badger for noting and fixing.

• Fixed a bug where ships despite having Forcefield Harmonics could not move through wormholes if a forcefield covered the wormhole.
  • Thanks to NR SirLimbo for fixing.

• ArnaudB fixed a math bug of ours where Coilbeams were dealing 100x as much damage as intended to any targets beyond the first. Essentially, the math forgot to divide by 100 at the end of doing some percentage calculations.
  • Thanks to ArnaudB for fixing, and Ithuriel0, Lord Of Nothing, Zweihander2021, zharmad, and others for reporting.

• In the event that you have a software firewall running, or something to that effect, then the networking tab will now show "Error Checking Public IP" as your public IP address rather than popping up an endless string of errors.
  • Thanks to wildtiger97 for reporting.

• We have had occasional reports of the list of tutorials being empty. This looks to have likely been a problem with some of the multithreaded way we parse the xml files that contain these. These are loaded very differently from quick starts or savegames, turns out. The solution we have in place for now is to have it trigger a rebuild of the tutorial data if it is not found when you open the tutorial menu. If there are ever any times where that is still blank, then we have a larger problem.
  • Thanks to wildtiger97 for reporting.

• There was a memory leak of sorts relating to shots, and it was a severe performance impact on games after a while. Exactly when this memory leak appeared is hard to pin down, but it's likely been knocking around for a few months at minimum, and potentially far longer.
  • I had to redo quite a bit of logic about shots in order to make them work well in multiplayer, and my suspicion is that I introduced this bug during that time.
  • Essentially, there are various states that a shot can be in that are semi-invalid, and to some extent that's okay and the shot needs to wait for things to become valid and then actually do its job.
    • A great example is when a shot is fired earlier on the host and then sent to the client before the frame that it is supposed to exist. This happens very frequently, because of the nature of our sim cycles and the "fast blast" way of getting shots over the client. In those cases, the shot is supposed to wait around and then appear at the appropriate time and look just like it did on the host, a few dozen ms prior. Looking at it with the naked eye, even with the computers right next to one another (but routing way out through Valve's servers and back), it's hard to tell a difference between them. But this logic was backwards, and so most likely it was drawing shots too early, and in cases where they arrived on time it was not drawing them at all. That was a bit of a second bug/observation by players, and that should also now be fixed, knock on wood.
  • As part of this, we made the target on shots into a lazy load squad wrapper, and added a new method for lazy load squad wrappers that does not cache null references when it finds them. It's entirely valid for a shot to appear on a client with a target that does not yet exist on the client, but will in a few milliseconds. This should make THOSE cases properly work, and the shots not just be-absent.
  • That said, the bigger problem was certainly with the shots that were sticking around too long. There were shots that seemed to think they were waiting for a target, and so just sitting around, but in reality their target was long dead. This was happening in single player, and on hosts and clients. This was leading to a buildup of hundreds or thousands of shots just sitting there being processed every frame, but doing nothing.
    • To combat this, alongside the other fixes we put in for shots to work better with "Delayed do something," we also now have some "if it's been too long, then die" functions. Specifically two:
      • If the shot has no target, and has existed for more than 10 seconds, then kill it. It's just going to be sitting there with nowhere to go, otherwise, and this is a big part of the leak.
    • If the shot does have a target somehow, but it has existed for more than 45 seconds, then good grief kill it anyway. Probably this is overkill, but overkill is how you prevent edge cases.
  • At the moment, my observation of a test case is that the shot counts stay reactive and very low like I expect, so even the 10-second case is not being hit that I can tell. This might mean that some of my other fixes to help them better wire themselves up to targets (the lazy squad wrapper extensions) or the correct flipping of the one piece of logic, has solved the root problem here. Either way, even if that's the case, we have two levels of overflow valve here for catching strays.
  • Thanks to NR SirLimbo for noticing the leak, debugging to find out that it was shots without any target, and suggesting action.

Multiplayer Improvements

• Put in a variety of debug tracing logic in DoWormholeTraversalLogic(), because that could apparently have rare exceptions on MP clients.

• Clarified one of the error messages for when network messages are received that are garbled.

• Various fleet-edit commands could throw errors on multiplayer clients, sometimes just due to order of operations of the sync cycle. This is one of those problems that is less of an outright bug and more of an order of operations issue. We now have it set up so that if the client finds that the situation is impossible to update properly, it just silently ignores those problems. It will have the correct data within 5 seconds, but odds are fair that the reason it has a problem is that it already has the correct data.
  • Thanks to Badger for reporting.

• "P2P connection to host failed" should now do a better job of actually letting the player know that this happened, as well as stopping them from endlessly waiting to connect after it already timed out, etc.
  • This won't solve any underlying issues (nor can it), because those are typically network-oriented or Steam-oriented, but it will eliminate the confusion and allow for a quicker retry.
  • Thanks to Badger for reporting.

• Fixed an exception that could happen on MP clients in DoSuccessfulCompletionLogic_CalledFromMainSimOnly().
  • Thanks to Badger for reporting.

• Fixed a couple of threading issues that could come up in our linked list classes in multiplayer in particular.
  • Thanks to Badger for reporting.

• Fixed some indexing errors that could happen in deserializing ships in multiplayer at times. Honestly if these were having an exception in this way, then there was already a more serious error that already happened. But we may as well not have this also erroring on us.
  • Thanks to Badger for reporting.

• When an exception happens during deserialization of a condensed-style string, the game now gives a much more coherent response. It looks likely that a few canaries were giving exceptions during the attempt at even reading the canary.
  • Thanks to Badger for reporting.

• Fixed a couple of potential exceptions that could happen in Window_InGameSidebarShips if your game was in an invalid state.
  • Thanks to Badger for reporting.

• Fixed a couple of annoying error messages that could pop up if you WERE just connected in multiplayer, and some last commands were being sent, but you just left MP either willingly or because of a disconnect.
  • Thanks to Badger for reporting.

• Fixed an exception that hosts could run into in MP if they tried to send certain broadcasts when no clients were connected.

• Our network "canary code" has been built in more deeply to our serialization and deserialization process, which lets us get exceptions more reliably when they happen. These are also contextual, so we can start embedding these into actual entities and get more information without the savegames ballooning in size. What we know right now is that some squads do not serialize properly across the network, but they all do serialize properly to disk, but it could be in any of several dozen files because it may involve external data specific to a given faction, or something else along those lines. The goal of an extended canary system will be to narrow it down for further code review. Thus far, the prior canaries seem to have narrowed it down specifically to squads, which is already useful.

• Split apart our UniversalMessageHandler handler method into multiple sub-methods (but did not change any code in them) so that we will now have full stack traces that tell us what kind of message was involved even when the error is caught further inside. As it was, because of nested error handling, that information was not available to us, which made certain things harder to debug than they needed to be.

• Fixed some multiplayer bugs that were introduced with the rewrite of speed groups on March 31st.
  • Essentially, it made it kind of a roll of the dice as to whether a client could decode a world from a host.
  • It also made it a roll of the dice as to whether ship data sent from the host to the client would arrive in an intelligible fashion.
  • As you might expect, this led to all sorts of errors during gameplay, but it also made it so that various savegames could not be transmitted at all once they got a certain amount large and started having fireteams with speed groups.
  • The fix now actually transmits less data, and it leaves the speed group IDs entirely on the host. This is way more efficient, works perfectly in our current design, and solves all of the related problems.
  • It's possible that there is a second cause of ships failing to be sent over the network, in which case we'll hit canaries again, but at this point it seems very likely that this is the only one; until pretty recently, ship syncs were happening perfectly, so looking for two bugs is probably overkill.
  • Thanks to Badger, Bummeri, Caseycc, Zegma, Zachariath, and others for reporting.

• As part of the fix noted above with the shots memory leak, we fixed the fact that way fewer shots were appearing on MP clients than were happening on MP hosts.
  • Battles now look properly epic for all parties!

OSX/Linux Metal/Vulkan Support Improvements

• Fixed a number of shaders not displaying properly on vulkan or metal, specifically because they were set to only work on directx and opengl.
  • In recent OSX versions this has meant that the ships on the main menu were not glossy or metallic at all (falling back on the diffuse shader), and the game loss screen had a number of pink textures. Same with engineers.
  • If you are playing vulkan on linux, or metal on OSX, and you see any more pink materials that look like they are missing some important stuff, please let us know!

• Dramatically cleared out some unused shaders and objects in the main project, thus making the end projects folder smaller as well as making it so that we can compile the main game build faster.
  • Short term savings: another 64MB of RAM and disk space, which is not huge, but is from a part of the program that most benefits a fast startup.

• Fixed a bug where beam weapons and engineering repair beams and things like that were showing as pink on OSX metal or Vulkan in general. The shaders for those were not set up for those architectures.

• Several further shader glitches with the planets on the loss screen have been fixed up for metal and vulkan.
  • These actually also now look nicer and more modern on opengl and directX, too. The sun now has this really cool but suble animation, versus before it was just a steady glow.

• Good grief! We had to re-code our entire shader for drawing background nebulas.
  • We were able to do a basic recompile of our one that brings together disparate parts to create bespoke nebulas; that one was originally created in Amplify Shader Editor a few years ago, and so it just had to tick the box to allow metal and vulkan support. Very easy.
  • However, we have a double-cubemap skybox shader that we had adapted from unity's base skybox shader (which is open source), and it had some very complicated rotation math as well as some very unusual HSV transformation math (basically inline HSV translations on RGB data, versus our usual translation of RGB to hsv, modify, and return to RGB). This one was not easy at all to upgrade.
    • Ultimately what we did with this one was convert it into a format that works in ASE, using the legacy unlit shader path. We almost had it working as a surface shader, but there's no way to update the vertex offset (SV_POSITION) and the texture coordinates (TEXCOORD0) separately in a vertex shader that we can tell. So that led to a ton of very odd-looking distortion. If we were able to assign directly into TEXCOORD0 in a surface shader, then we could move to the newer style.
    • As it is, the shader has been updated to include the latest ShaderLab syntax, and since the only parts that are off the usual reservation is the super-funky math, we should be able to maintain compatibility with future versions of unity with this shader so long as Amplify keeps updating their Legacy Unlit shader path (or if there comes a way to set TEXCOORD0 in a surface shader). Either way, that seems likely to work just fine. We've worked with the Amplify folks before as customers requesting new features, and they were extremely accommodating, so we really like working with them.
  • TLDR: This was an epic journey we didn't want to go on, and don't look forward to going on again in the future when Unity inevitably changes something in some future version. But it's done, and this was by far the most strange shader we have in our overall kit.

• Can now confirm that all of our shaders and such work properly on Metal, so you won't have any missing ships, or any wrong-colored artifacts, etc.
  • My guess is that Vulkan now works on linux as well, but if it was literally crashing on your system then obviously not. But if it was heavily garbled, it might be working great now.

Last RAM Use Improvements For Now

• Dramatically cut down on the GPU load (and RAM size) of the Flenser.
  • The model itself was one of the largest things in the base game bundle, larger than any normal texture. Wow. At any rate, it's much more efficient now all around.

• Took this quick chance to go ahead and cut down on RAM use once again, this time with the ship models.
  • This is something that also will reduce VRAM load, AND more importantly will also reduce the GPU pipeline load.
  • The most notably change in this was that I took all of the extra cubemaps that we have and are using for models, and moved them from being 2028 in size to being 256 in size, and removing all of the mipmaps. This drops them from being 8MB (or 32MB in one case) down to being like 300kb.
    • In terms of end quality that you see, the result is basically identical. Even swapping back and forth right in front of my face, I could hardly tell a difference. The reason is that these are not crisp reflections already, by design. They give general colors to certain sides of reflective models, but the added detail was frankly wasted before.
  • Total savings: aiwar2goodiebox (base game models, etc) down from 1025MB to 786MB. The spire rises bundle down from 714 MB to 411 MB. Total savings of 542 MB.
  • If you combine this with the changes from last week, then the entire RAM load of the game has dropped by about 1.2GB compared to the month before. DLC2 is likely to erase all those gains, but that's okay because the amount of content is epic and deserves to be really pretty. But the main thing is we're not going to be seeing an overall spike that is too crazy when it comes to DLC2 partly because of reducing the base game and DLC1 to more efficient levels.

Beta 2.768 RAM Reduction

(Released April 2nd, 2021)

• Fix a bug where toggling fleets between Watched and Unwatched did not work through any interface.
  • Thanks to Badger for fixing!

Balance

• Buffs to vanilla marauders:
  • Raider Frigates both variants old values 100K hull and 32K shield.
  • Standard Raider Frigate now 350K hull and 125K shield.
  • Drone Raider Frigate now 250K hull and 100K shield.
  • Adjusted Marauder constants to give them a better shot at living long enough to establish themselves.
    • Minimum allowed wave size value has been quadrupled.
    • Time between outposts being built was 6 minutes, now 5.
    • New outposts come in with 10 mark one turrets instead of 5.
    • New outposts upgrade to mark two within 5 minutes instead of 10.
    • Mark three outpost lifetime requirement lowered to 15 minutes from nearly 17.
    • With these changes Marauders field larger initial forces and should fortify a position much better in a shorter time frame but will still take quite a while to ramp up to producing raider frigates to attack with.
  • Thanks to CRCGamer for making these changes.

• Both raider frigates now do 10x as much damage with Plasma Torpedo system, functionally this means a spread damage weapon that was doing under 50 DPS might be actually doing something at 500ish DPS.
  • Thanks to CRCGamer for making these changes.

• Both raider frigates had their Heavy Beam Cannon damage doubled. Might be better at killing small strikecraft now instead of tickling them.
  • Thanks to CRCGamer for making these changes.

• Fleet Capacity Extenders (FCEs) have been removed from the game.
  • The short reason is that they were stupidly unbalanced, and they also created a lot of edge cases and other problematic situations.
  • Another way of looking at this is that, since the Paradigm Shift, you now have FAR more ways to get more of what you want... even if it's not exactly 100% what you particularly want.
  • There's always a tug of war in this game between the concept of "play it as it lies" (as you find it, whatever it is), and "construct an empire of your own design."
  • Prior to the paradigm shift, the big threat to this was that you couldn't refund science, and rerolls existed, and overall you had a hard time getting what you wanted... unless you save-scummed a lot or did other un-fun things.
  • Now after the paradigm shift, overall I feel like we're pretty balanced in terms of things you find versus things that you can choose, but FCEs were a definite exception to this. They were basically a magic "wish for more of your favorite thing" button, which is antithetical to the larger design and the way that ARSes work now.
  • Even just the uncertainty of "has this ship line been doubled or is it just unusually large" is not great. But there were other balance edge cases like this allowing you to effectively stuff extra ship lines in a smaller transport, thus making the FRS units stupidly powerful when combo'd with this. This was also an enormously problematic unit when stacked with the new High Capacity Transport that was added during the beta period.
  • Any FCEs in existing games will be scrubbed.
  • Thanks to Strategic Sage for pointing out the problems here, although honestly this has already been on my hit-list for a while.

• The High Capacity Transport is now the High Energy Transport.
  • As noted above, the High Capacity Transport was big trouble when combined with the FCE or with FRS members, but all of the same commentary that applies to the FCE also applies to the HCT flagship, frankly.
  • The NEW flagship actually represents the filling of a gap in our abilities thus far in the new paradigm: the ability to spend hacking points for more energy, since energy is one of the big limiting factors.
  • This hack still costs 60 to do, which is a huge amount, but this transport gives you 400k energy in return at mark 1, and an extra 110k energy per mark level after that. For comparison, and economic command station starts at 500k and moves up by 240k each mark... but even logistical only starts out with 175k.
  • Any HCTs you had are now translated automatically over to HETs.
  • Thanks to Strategic Sage for pointing out the problems extend to here, and that we needed a way to hack for more energy. That's a super useful new function (since before you could not get more energy without also getting more hacking points, it was a bit of a vicious cycle if you were short on energy but had extra HaP).

Visual and Performance Improvements

• Fixed the Loss screen having some pink "missing text" materials on its particle effects for burning embers.
  • Wound up just creating entirely new particle materials, and they look better than ever!
  • Thanks to Daw11 for reporting.

• Performance improvements to the main menu:
  • Drastically reduced the polygon count (to about 1/3 the prior value) on the ship hangar scene of the main menu (this is looking out from inside a factory).
    • This has always been the most visually intensive scene in the entire game, and scares some folks if their computer can't handle it. More computers are now able to handle it, and with very little visual difference in quality. Thanks, PolyFew! This was not possible with Mesh Simplify.
  • Removed the spotlight on the main menu scene, which was responsible for the harsh white light on the ceiling. This often would kind of blink out and then come back as the light of ships passed by, and that always looked strange. This was the engine trying to conserve resources, and in general it was rough on older GPUs.
  • Removed about half of the room on the main menu scene, which means you can see out through some girders to the either bright, dim, or somewhere-in-between sky. This is WAY more dynamic and interesting than the boring spotlight, and requires far less computation.
  • This is a monster improvement on lower-end cards, in particular my OSX machine from 2011 that is under the min system requirements now runs the main menu smoothly rather than at an awful stutter.

• Due to the way that high quality texture compression is just dumped and completely decoded on this one platform, OpenGL for OSX has been removed from being an option. It will always use way more RAM (like 8x as much sometimes) and perform worse than Metal.
  • This should keep even quite old computers running the game much better in OSX.

• We have updated the way that we are compressing the background planets in the main planet view, and the result is that the asset bundles for those have dropped from 632MB to 160MB, and the background nebulas from 379MB to 239MB.
  • This makes for less for you to download, faster loading times on game start, and better performance during the game. There's a very miniscule difference in quality if you zoom super in and stare at the planet up close, but even then it is hard to tell.
  • We also removed most of the mipmaps that these have, since those were not needed in any practical scenario for these particular objects. It was just a waste of RAM and disk space, and not for any performance benefit, unlike many other situations where mipmaps are awesome.

• Visuals of the Red Desert World improved in general.

• A few commonly-used background images for the UI have been made power-of-two and now are able to be stored in a BC7 high quality compressed format. This makes them smaller and more efficient, while actually having the same amount of detail.

• The core game files have been optimized to save 150MB in RAM for everyone, which also means that much less to load on startup.
  • This has dropped from 410.2MB to 261.8 MB, and represents data that is loaded before it starts the "loading timer."

• The "Material doesn't have _StencilComp property" harmless warning logs in the player log are no longer a thing. The shaders in question now have thsoe properties, although they don't use them.
  • This stuff is required for UI Mask capabilities, but this particular UI shader never needs to be masked.

• The arcenui asset bundle has been optimized a bit more, with a variety of its textures getting updates.
  • This brings it down from 124 MB to 113MB. There is a vague chance that a few of the mipmaps would have been useful, but honestly most of them were used in ways that were almost certainly more slowing than helpful on any semi-modern computer.

Beta 2.766 Huge Multiplayer Speed Boost

(Released April 1st, 2021)

• Updated the civilian industries mod to be more correct in its description of when it was last updated as well as what level of tech support Arcen provides.

• Fixed a bug where if you had some expansions or mods on, and used those to select certain lobby settings (AI types, battlestation types, etc), then you could get exceptions if you reloaded into the lobby after disabling the expansions or mods in question.

• A few more balance changes for frigates.
  • Now gain an extra 10% range at marks 2, 4, and 6. Capping at 130% base range at mark 6.
  • Durability upgraded by an extra 0.2 base stats per mark level for a gain of 1.2 by mark 7.
  • This means that frigates at Mark 7 now have 7.7x base hull and shield values.
  • Burst-Fire weapon jammer added to the Raid Frigate as well as the Apparition, Bounty Hunter, and Vicious Raid Frigate variants.
  • Thanks to CRCGamer for making these!

New Mod! Frigates Focus by CRCGamer

• This mod effectively rebalances a few things within the game for a different experience.
• You will start with a fleet of frigates with revised starting fleets to choose from.
• Some of the additional starting fleet options have requirements.

• Please enable The Spire Rises (DLC 1), Zenith Onslaught (DLC 2), and More System Defenders (MOD)
• This will grant all available options at the moment.

• ARS Hacks will only contain available frigates.
• Pre-seeded fleets will lean towards frigates heavily. Some strikecraft will appear now and then.

• FRS strikecraft still remain available for fleet buffs.
• Is Democracy's worst nightmare.

HUGE Multiplayer Speed Improvements

• Our implementation of the main steam networking library now uses the no-nagle approach, so data sends more directly.
  • That's a judgement call for any game, but we need our data to get there faster and we already compact the messages that are okay waiting for up to 200ms (uhh... that's none of them).

• Up until now, for multiplayer on Steam, AI War 2 has been using the new "connection-oriented" network library from Steam. This is one of four distinct libraries (one of which is deprecated) that Valve provides for getting data from here to there, and it's their newest one.
  • For folks not all that into networking details, essentially imagine it's like Valve has a racetrack (their hardware and network), and they have four different vehicles that developers can choose to use on said racetrack. One of the vehicles is old and basically only there for legacy purposes: so for a new game, there are three vehicles, and a single racetrack.
  • Each vehicle has a variety of customization options unique to it, for depending on how you will drive it. For instance, think of this like four wheel drive mode (worse gas mileage, better traction), or environment mode (way better emissions output, but worse performance). Each vehicle has different of these options, and different performance characteristics in general, is the bottom line.
  • What we've been using up until now has been the Connection-Oriented protocol, which is a vehicle that is very... reliable, is I suppose the main thing. It tries to emulate the TCP protocol on top of UDP, which is a pretty common goal in networking libraries.
    • TCP is what runs all our websites, and it's an absolute terrible protocol that everyone hates and that is super inefficient. Games can't use TCP at all, but generally implement some form of "reliable UDP" (an oxymoron, but a useful concept) on top of UDP itself. Some of these work more like UDP, others work more like TCP, and all of them have their own quirks.
    • Anyhow: we have been running it with the Nagle algorithm enabled, which was actually a terrible thing for our particular game. I programmed our network code in a network-agnostic fashion, and I already do the batching that makes sense. Nagle is all about putting a bunch of small and inefficient packages into larger packages... and it creates up to 200ms of extra latency if you don't have enough to fill a full load.
    • All of these options are great, and none of them are "bad." But in the same way you don't try to race a dump truck in 4 wheel drive in the Indy 500, and you don't try to take a high performance racecar offroading, you have to pick the right options for your particular game. The Nagle algorithm is terrible for AI War 2, but probably a godsend for a lot of action games in particular. This connection-oriented protocol was not the best for AI War 2, but for a turn-based strategy game it might be perfect.
  • The particular problem with these tools, and trying to figure out which one to use, is that it's not always obvious exactly what the capabilities of each one are. And they are VERY different in terms of code. It's like thinking you can fly a small plane because you know how to drive a small car with a manual transmission; they both take some practice, and they don't quite equate.
  • We have now switched to providing both the now-Nagle-less Connection-oriented protocol from Steam, but also their more UDP-like "Peer to Peer" implementation.
    • The P2P framework is pretty interesting, because it doesn't have much of a concept of the host and client even being connected. One just starts talking, and the other can opt to say "no, be quiet," but beyond that everything is at the game level. This is very nice, because my hope is that some of the random disconnections that a few people were having will thus go away. There's no connection to drop; we were talking 3 seconds ago, there was a blip, now we're talking again. Easy. Hopefully.
    • The other benefit that the P2P framework has is that it supports multiple channels of data. In each channel, essentially there's a queue of messages. If you have packet loss or fragmentation, everything in the queue has to wait for the items before it to be re-assembled. This is the number one cause of stutter and lag.
    • The benefit of using multiple channels is that for all of the heavy data, like the sync stuff, we can shove that on a channel that does not interrupt game flow at all. Meanwhile, on the main channel, comparably minimal information is sent back and forth between the host and clients. This doesn't lessen the bandwidth requirements any, but this DOES greatly lessen the load on the core critical data that needs to get passed back and forth to keep the game moving. In router design and network programming, this sort of thing is often referred to as traffic control, or quality of service control. Aka, if someone is watching Netflix, don't let that destroy the internet for everyone else in the house. That doesn't mean throttling, if you have enough bandwidth; it means just making sure everyone has a chance to get their data in the sort of timeframe they expect it.
  • So! The P2P framework is now the default, and it already seems vastly faster. We are leaving the connection-oriented mode around, just in case someone has a problem with peer to peer and how it functions. But we anticipate that, especially as we start adding channels (that the connection-oriented version will just stuff all in a single channel, sadly), the gap will widen even more. If that's the case, and things are working well for everyone, then the connection-oriented version will just be phased out.

• The game now supports multiple network channels for networking data on every network framework EXCEPT for the older Steam Connection-Oriented framework (since that one does not support channels).
  • Steam P2P, GOG, and LiteNetLib all happily support network channels, and use them to great effect, now.
  • All of the network channels are in the format "reliable ordered immediate," so no Nagle, but also everything is still reliable and so acks and resends happen on all the channels.
  • The benefit of splitting out by channel like this is that if there are fragmented packets in one of them, they don't cause slowdown and waiting around on the other channels.
  • There are four overall channels (except on Steam C-O, which squashes them all into one, which is how it has been up until now but is horrible. Please us Steam P2P):
    • Main channel: this is the data that must be on time, and unblocked, for the game to keep proceeding. If this is streaming fast enough, your game will not be lagging at all.
      • This includes the original handshakes between clients and the host, initial world data from the host to each client, and then any commands issued by clients or the host. Beyond that, it's pretty much just frame timing updates for general game synchronization on the host's orders. This also includes all "fast blast" data from the host to the client on unit and shot creation.
    • Frequent channel: this is a fair chunk of data from the host to the client, and mostly involves data that should be updated every couple of sim frames. This is the broadest stuff that is kept in sync, but it's not absolutely required in order to run the game. Lobby chat is also sent via frequent (lobby chat runs at a lower level than in-game chat).
    • Bulky-1 channel: this is almost entirely devoted to analyzing ships, and the clients and host talk back and forth about what they find.
      • More or less: "Hey is this what you have for these 5 ships?" "Those three are right, but these other two don't match. Send me the diffs on those two please." "Here are the diffs on those two you asked about."
      • There are also some cases of the server knowing "hey a thing just changed on this ship (such as a speed group speed) that the client would not know about, so preemptively sync to them before it drifts.
    • Bulky-2 channel: this is entirely devoted to large-form rolling sync data about things other than ships. That mainly means world data, world externals, faction data, faction externals, planetfaction data, and the like.
      • This channel is super hefty, and is the vast bulk of the data that is sent in general. But if it's delayed a few seconds in getting to you, it doesn't matter. If the Main channel is delayed a few hundred milliseconds, it will feel awful, by contrast.
  • An example of some of the timings are here in discord: https://discordapp.com/channels/240637654717300736/753611766046523572/827334456695980052
    • In that one example, the host sent 23MB of data in about 15 minutes. 14MB of that was Bulky-2, 3.8MB Bulky-1, 3.3MB Frequent, and a mere 2.1MB on the Main channel. Bear in mind that about 300KB of the main channel was just the initial world send.
    • If you total all of that sort of thing up, this is way less data than a low quality internet video, spread out over more time. Our problem has never been the general volume of data, but lag has been present because of a couple of things lately.
      • First, we had a lot of thrashing on what would now be Bulky-1 because of sync data going vastly awry with speed groups. That was re-coded entirely in the last release, so there is far less data in that category.
      • Second, we had the Nagle algorithm on by accident, which caused frequent delays of 200ms to accumulate small messages into batches rather than passing them on. This only affected Steam, not GOG or LiteNetLib. This has been fixed in this release.
      • Third, the network framework used for Steam was suboptimal in a variety of ways, architecturally. This did not affect the other platforms, but it was dragging down Steam to a substantial degree. This has been fixed in this release.
      • Fourth, all of this other data -- literally 90% of which doesn't need to be on time for the simulation to run well -- has been moved off the Main channel. This is a reduction in 90% of where we might see congestion from acks and resends and problems of that nature. If your connection has any substantial packet loss, then you now get to dodge 90% of the impact of that.
  • Final verdict? Basically an order of magnitude better performance for everyone, but in particular for Steam, which was running worst but used most.

Beta 2.765 The Hunt For Frigates And Civilians

(Released March 31st, 2021)

• When you are searching for units on the galaxy map, it will now allow match on both the full name and the short name. So you can type ARS to find the Advanced Research Station, etc.
  • Thanks to Daw11 for suggesting.

• A few minor code efficiency improvements in some frequently-called code that will grant a small speed boost. Probably imperceptible on most machines, but why not.

• Our per-ship network checking code in multiplayer has been slimmed down to no longer check the type of ships, or the faction of ships. It is no longer possible for those to ever be wrong, so we can save something like 1/4 of the bandwidth from each ship. It's a tiny amount, but given how many times this happens and thus how much it adds up to over even a short period, a 25% savings is nothing to sneeze at.

Balance

• DLC2 shot attraction backport to Bulwark Turret implemented (making that DLC1 turret way more awesome)
  • Thanks to CRCGamer for implementing.

Balance adjustments to a number of frigates

• Assault Frigate + Station Keeping variant adjustments
  • Speed 600 -> 700
  • Point defense shots per salvo +2 mk/lvl
  • Point defense range from 5200 -> 6500

• Mugger, Cloaked Mugger
  • Speed 400 -> 600
  • Point defense shots per salvo +2 mk/lvl
  • Point defense range from 5200 -> 6500

• Brawler
  • Speed 400 -> 600
  • Repulsor Missile range from 4200 -> 5600
  • Repulsor Missile shots per salvo +3 mk/lvl

• Watchman Frigate
  • Additional sub-weapon of 300 base damage, x6 shots every 4 seconds and +1 mk/lvl. Range of 5600

• Siege Frigate, Devastator, Ion Disruptor
  • Speed 400 -> 500
  • Weapon system ranges 10100 -> 12000

• Additional adjustment to Siege Frigate where it now can only hit up to 10 targets per shot. Prevents excessive damage dilution due to its spread damage mechanic in some situations.

• Thanks to CRCGamer for this update to the frigates to make them more viable to main!

Civilian Industries Returns

• Civilian Industries should once again be functional! We have not tested it, but it's recompiled into the new beta code style, and so should continue working like it used to.
  • Thanks to StarKelp for sharing the source so we could get it to work, since this is quite a popular mod.

• All of the units in Civilian Industries have been updated to have proper short names for viewing in the new sidebar. Almost all of the previous ones were incomprehensible because of inheritance.

More Mod Updates

• Update v1.25 for More System Defenders - Speedy Frigates
  • Frigates within the mod had base speeds adjusted to a minimum of 800.
  • Augmented Artillery Destroyer also given the speed increase to 800.
  • Arming time of Cruise Missile Battery sub-units adjusted down to one second.
  • Subjective strength rating of said sub-units given a 20% multiplier.
  • Max targets hit within AoE adjusted for Shockwave Missiles.
  • AI Fortified Cruise Missile Base visual scales increased.
  • Shockwave missile max targets 10 still fires three at a time though, damage 2.5x previous.

• AMU:
  • Fixed the "AMU Debug For Faction Index" setting being fixed at -1 and, after that was fixed, found out it was not displaying anything. Fixed that as well, it now only debugs for that one specific Faction ID.
  • Fixed a potential exception in Auto-Juggle Energy Function from not clearing the list of entities to kill. Over time this would become a memory leak and leak.
  • Added new settings in cooperation with CRCGamer and his More System Defenders mod: If both mods are installed AMU, can now auto-build the following defensive units if enabled in the Automation tab:
    • Cruise Missile Bases (0-5), these are placed near your command station and if possible on facing away from the nearest wormhole.
    • Artillery Destroyers (0-2)
    • Flypaper Frigates (0-50)
    • Gorgon Aggregate Body Frigates (0-50)
    • Escort Carriers (0-50)
    • Ambush Carriers (0-50)
• Extended Ship Variants:
  • As it adds a lot of ships for medium hull and fills the void there it now also modifies the medium hull cost to be on par with light and heavy hulls.

• AMU:
  • Rewrote almost the entire description to be more clear and organized. Then added that AMU now cooperates with More System Defenders to allow the auto-building of its defensive units.

• AMU:
  • Tweaked AMU to also allow for Auto-Building More System Defenders units not only when the mod itself but also its indev-counterpart are enabled.
  • Confirmed actually killing the exception in the Auto-Juggle Energy Assistant (AJEA). It would occur when swapping any other command station in for Economic Command Stations due to a stale fleet membership for the E->M Converters. AJEA now only considers converter fleet memberships valid if they have a positive maximum amount of entities.
  • Reorganized the custom Exception Logging code in AMU to allow for more functions:
    • There now is an enum ExceptionThrowingMode. Only if it's set to BlockAllFurtherExecutionOfEntireThread it will kill the thread (and usually the whole faction along with that). Other options just show a popup but don't do anything else, or quietly debug log it.
    • This FINALLY fixes the issue of a single Executor Fake Faction script crashing taking down the entire thread with it. Now the script will be removed, but the rest of things can stay. 3-ish times before this was attempted and claimed, but never tested, this time it IS tested and works.
  • At the same time the error logging code now has a string passed into it to define what modder screwed up and how to contact them. In my case this is -NR-SirLimbo#0021 on Discord or [email protected]
  • Additionally the call of Exception Logging was simplified and QuickExLog() shortcut functions were created.
• Kaizers Marauders:
  • Integrated the new AMU logic and adapted it to changes in the Vanilla codebase.
  • Fixed the Rampage Heavy Frigate (drones) having the wrong icon.

• AMU:
  • Created the DynamicAveragerRing class, which essentially is a list of values that an average is created out of, and each time adding a new value it replaces the old one, assuming the required change threshold, if set, is exceeded.
  • Fixed another 2 bugs in the Auto-Juggle Energy code that made the mod exponentially overestimate the player's metal income at high game speed, and logarithmically underestimate it at low game speeds. This lead to all kinds of horribly wrong behavior.
• Kaizers Marauders:
  • Removed the defense spawn distance factoring from sniper turrets. In the hope that nobody will ever clutter their gravity well so much that the game begins to lag when trying thousands of locations to place something at the edge of the planet.
  • Used the new DynamicAveragerRing class to do calculations on the defense spawn distances.
  • Essentially when the last result distances were further away than within a margin it expands the next spawn attempt's base range, and if they were closer than that margin it inflates the next attempt's spawn distance.
  • This way it should be able to increase and decrease dynamically, but this is untested as of yet. I'd love to get someone's save with a built-up Kaiezrs Marauder Capital planet to see if it works, and by how much it grows better or worse.

Bugfixes

• Fixed a couple of places in the squad class where cross-threading issues could crop up with index out of range exceptions looking at unit orders. The side benefit of this is that it also makes this code a bit faster to run, because fewer array indices are happening.
  • Thanks to Zweihander2021 and Sigma7 for reporting.

• Fixed a number of additional cases where cross-threading exceptions could happen with orders to units, even though none of them had been hit yet. Better to be preventative.

• Game code is no longer allowed to directly touch the QueuedOrders list on entities. This may affect some code mods.
  • The code is now required to call GetQueuedOrderAtIndex_OrNull( index ), GetLastQueuedOrder_OrNull(), GetQueuedOrderCount(), RemoveOrder( Order ), QueueOrder(), and RemoveOrderAt( index ), as needed.
  • This was just absolutely ripe for endless cross-threading issues, and now we're able to condense all the possible errors into one place that is error-proof.
  • In general we have also switched to using more Remove rather than RemoveAt. This is slightly less efficient, but also less likely to junk or corrupt wrong data. If it causes a slowdown, we'll have to look into it.
  • It's worth noting that, in general, this is BY FAR the largest source of cross-threading issues, and we had literally like 30 different band-aids throughout the codebase. This removes and centralizes all of them, avoiding any chance of such issues. We've literally been having these for over a year, though, with no sign of stopping... and so an architecture change was in order.
    • Updated one line of code in civilian industries to make that mod work with this change.

• Properly fixed that index out of range exception that could still happen on WriteTechThatBenefits(). I was missing a -1 in there.
  • Thanks to Badger for reporting.

• Fixed some issues where death effect damage may have been overflowing, and in general where it would not serialize if it was negative. Now it just corrects to 0 damage rather than throwing an exception. This was during DeathEffectCausingDamageReceivedToEntity_Data mainly for multiplayer.
  • Thanks to Badger for reporting.

• For some strange reason, it was still possible for us to hit Steam's "LimitExceeded" error when sending certain messages that include the entire world file. This was intermittent with clients in general.
  • Overall it is true, we were trying to fragment savegames as little as possible, basically breaking them at around 500kb rather than something lower, knowing that Valve's limit on messages is 512k. But with our own headers being around 100 bytes or so, we still should have had plenty of room, so why exactly this was happening is not clear.
  • What we've now done is shifted it so that we now target 400kb, which is VASTLY more room than needed. However, in terms of the grand scheme of sending world data, this is really paranoid in a way that should avoid the issue while not letting edge cases slip through. No headers are remotely near 112kb in size, so if we still get the error then this is something to do with how much is queued up in the steam networking buffer or somesuch.
  • It's also worth noting that the difference in fragmenting 100kb one way or the other is essentially invisible in terms of performance, and this only affects initial connection.
  • Thanks to Badger for reporting.

• "CheckForShotAOEDetonation error at debugStage 4100" errors in multiplayer should now be fixed, or if not then it should give an even more specific error now. But knock on wood, should be fixed!
  • Thanks to Badger for reporting.

• Fixed a "BUG: entityTospawnat is null" on multiplayer clients, but in the process also stopped them from trying to spawn the contents of a wave at all. That was just going to lead to strangeness and lag whenever a wave arrives. That data will be arriving by fast-blast from the host within a matter of milliseconds, and there's nothing for the client to calculate on their own.
  • Thanks to Badger for reporting.

• A HandleAutobuild() exception was happening on some multiplayer clients. It probably still will, but it will give way more informative errors now. There's some slight extra protections in there already, but probably not enough for whatever was going on.
  • Thanks to Badger for reporting.

• Fixed a bug that would outright crash the game in DLC2 if you had a revenge-firing ship shooting another revenge shooting ship on a planet that no player is actively viewing.
  • Essentially, what happened was darker mirrors shooting each other when nobody was looking would not make a sound so much as they would silence all sound forever...
  • The change to make this work actually wound up being a lot more extensive in terms of how we process AOE damage, shots, and even knockback. The basic fix would have been faster, but more likely for complicated bugs like this to be possible in the future. If you see any strangeness with knockback in particular, please let us know!
  • Thanks to NR SirLimbo for the initial reports, and Democracy for the save that let us replicate this within 30 seconds of loading it. That was a lifesaver, as this still took something like 3.4 hours to find and fix.

• The "Mismatch on faction index" error that was popping up frequently in the opening of the lobby will now only happen if you are actually generating a final map for the game. It's a useful debugging tool, but the warning it's giving us on the "partial map" is about something that gets entirely erased when you actually start the game, so it causes no problems.

• Fixed an exception that could happen in some cross-threading situations in TargetEvaluator_ZombifyAsPrimary.
  • Thanks to Zweihander2021 for reporting.

• Increased the delay between fragmented Steam messages from 10ms to 50ms. Less likely for the valve to throw a congestion failure message that way, and it's not going to be noticeably different to users.

Multiplayer Canary Code

• "Canary code" has been introduced into several of the multiplayer messages. These may be taken back out later, but they really require very little processing and potentially a great amount of clarity, so we'll see.
  • Essentially, using our extremely efficient sub-ASCII text encoding, we send a string at certain points in the multiplayer messages, and on the other end it tries to read that exact string. The key point here is that the client KNOWS what it should be receiving. If it receives something other than what it expects, then it will yell right away and throw an exception and stop processing that message.
  • When it comes to our serialization format, it is blind and it is without markup or metadata. This is hyper efficient, but when we are reading naked bits... any string of 4 bites can maybe a byte, and any string of 16 bits can make some sort of small integer. We are reliant on clients and hosts knowing exactly what they are reading at any given time, and it's often not until we get a value that is clearly out of range later on that it actually dies. This can be VERY downstream from the actual divergence in data (we could have had 1, 10, 100, or even 1000s of reads that were invalid without knowing it).
  • This sort of thing gives us an existential problem in that when we know there's an error in some sort of message, we have no way to know how to narrow it down. The only way to find the divergence is to get the client and the host to both write a log of what they sent, and then diff the two. The problem is always immediately obvious once we do that. However, we have no way of knowing when to log, when the problem is a very intermittent one. Full time logging would make the game unplayable, and we also can't retroactively log after discovering we have a problem (the host never even knows there is a problem).
  • The one other way that we can find a problem like this is via manual code review, but that is extremely tedious and we could be looking through literally thousands of lines of code in dozens of files for a single send, since it could involve many external factions or even mods.
  • To solve this existential problem, at certain points we can send a known unique series of ASCII characters, which get converted into a hyper-specific series of bits. We're talking about... let's say maybe a series of 72 bits. In the canary check, it would be unimaginably unlikely for all 72 bits to happen to be the right combination of on and off unless the data that has been read prior to now is all exactly correct. Thus we basically cut down dramatically on the amount of code that we have to review in order to find the problem spot, since we don't have logs.
  • If the amount of code is still too much, then after the first round of canary checks, we may wind up needing to add more layers of more detailed canaries, potentially that are toggled on as a more lightweight form of message checking, since logs are really impossible to synchronize in a realtime multiplayer game. The data is just too voluminous for players to handle, and they have no way to pre-parse it for our use. The canary solution keeps things brief, to a single line of log code on just the client, and has minimal-to-no slowdown added (if we wind up adding lots of layers of canaries, that would be slowdown, and would be a toggle for when you suspect problems).
  • Two things to be aware of with canaries:
    • If I make a mistake when entering a canary, then I may make multiplayer unplayable by accident. That could happen, and would just require a code change to fix it. This is difficult to test in advance of someone else firing up multiplayer, unfortunately.
    • If my canaries are all correct, multiplayer still may become unplayable if there's a bug that a canary is now finding. Essentially, an area of multiplayer might ONLY have been functional because of the nature of "any row of bits is a number," and that in turn would lead to all sorts of strange effects like... well, potentially ships going to the wrong location, or having very mysterious orders, that sort of thing. Those are all just lots of series of numbers, so its entirely plausible to have a functional -- but wrong and fill with lots of sync data request -- multiplayer game going in those circumstances. The canaries would make such a game unplayable, however, because they actually would find at least the general areas of the code and then halt reading of the message (since they know it's garbage after that, even if it's theoretically-playable garbage). This may actually mean a short-term reduction in multiplayer playability, is the bottom line. But it leads to long-term correctness, and the removal of strange MP-specific divergences that wind up sending more and more and more sync data trying and failing to fix itself. Aka over time it will also lead to less bandwidth usage.
  • Thanks to Badger for the report that posed this conundrum.

• Lots more canary code, on clients sending gamecommands to the host, and the host doing the same to the clients.
  • This way, if any gamecommands are being corrupted, we will hear about it just the same as we would with all the various types of sync data, and the fast blast data.
  • We have long suspected that there is something amiss in the fast blast data for speed groups in particular, but never had a way to test it until now. With these changes, it's entirely likely the multiplayer is going to be extremely broken with various canaries squawking. If folks post the logs of what they run into, and savegames are also ideal, then we should be able to exorcise those bugs rapidly and the stability of multiplayer as a whole is likely to skyrocket. Apologies for any temporary blockages that this causes, though.

• So far in some limited multiplayer testing, we aren't seeing systematic issues yet!

Speed Groups Rewrite

• The way that speed groups are calculated, stored, updated, and most importantly shared via multiplayer has been completely rewritten.
  • This is something that fireteams use, and it's also the "group move" function on player ships.
  • Essentially, the old way was pretty inefficient in single player, and lost its data for fireteams when you saved and loaded your game, and also was completely broken in multiplayer.
  • The initial revised approaches we started with still involved sending speed groups over to the client, but this is just a huge amount of data and often a real pain in the rear to try to coordinate. A much better approach that we settled on is to actually keep the speed group data entirely on the host (rather like fireteams themselves), but then just share the resulting speed limits for any grouped ships.
  • We are sharing those speed limits in the form of sending the entire ship data, which is not a huge amount of data, but is nontrivial. Part of the reason for this is that we're aware that other slight divergences will likely happen in the few frames of movement after a group is created or disbanded, so a full sync is less likely to require a resync soon after. If we later find out something different, we can make it sync just the new speed limit in those cases.
  • Overall this new approach that we have now implemented should lead to a large bandwidth drop in multiplayer (all the other sync repair data), as well as the removal of certain ships seeming to "rubber band" to an area and then back as they try to move faster locally than the host says they can. This should also take less processing on the CPU on the host and in single-player.

Hunter Bugfixes

• Fix a bug where hunter ships against a Hostile To All faction might decide to attack one of your planets if you were on the way
  • Thanks to Zweihander2021 for reporting

• Fix a bug where Hunter Fleet ships that spawned in waves against minor factions (Nanocaust, Dyson, etc) were able to go after the player, instead of remaining focused on the faction the wave spawns against.
  • Thanks to a number of people for reporting, including Democracy and cml

• Huge thanks to Badger for fixing this!

• Remove the hunter_targeting_this_faction_dies_immediately="true" tag from the dyson sphere, since that is no longer needed now that the hunter doesn't turn rogue and attack you anymore.
  • Thanks to Arides for reporting the exploits that you could get up to with this tag on, and to Badger for fixing the hunter code up.

Beta 2.764 Command Station Hotfix

(Released March 26th, 2021)

• Improve the text of an Objective to remove incorrect information about forcefields

• The X map has been redone to make it more balanced and aesthetically pleasing. The old X map struggled badly with the Praetorian Guard mechanic (especially in the spire campaign) amonth other things.
  • The new map X map is more similar to AIWC's X map

• You now get 4 FFs rather than 3 on mil cmd stations.
  • You still get 3 on logi stations.
  • You now get 2 on econ stations rather than 1.
  • Thanks to ArnaudB for sorting this out.

• Fixed an error where until you clicked to the fleets tab, each fleet would say "Fleet type: ERR None". This was new since we started having the ability to click into fleets without going to the fleets tab first.
  • Thanks to GalacticZombie for reporting.

• Fixed a regression in the most recent build where, thanks to some other better error handling, the "you don't have X mods enabled right now, would you like to?" function for savegames was no longer working.
  • This now works again, and has been generalized out more so that it is easier for us to not have this break in the future.

• Remember in the last release notes, I said I was removing some fleet leader memberships and that it didn't seem to be a problem but also didn't solve the bug I was hunting at the time? Turns out it was a problem, in that it made it impossible to build command stations since their memberships would be auto-deleted as you attempted to do so.
  • That bit of code has been reverted, since it was wholly unneeded in the first place and just caused problems.
  • Thanks to Zweihander2021, zeusalmighty, Lord Of Nothing, and many others for reporting.

• Fixed a couple of regressions from the last month that were related (I think) to a fix put in to support certain flavors of linux mouse handling raycasts. All of the functionality should be correct for all OSes and flavors of those OSes, now.
  • One of the problems was that if you were hovering over the UI, it could still show the battlefield or placement cursor, either stuck in one spot, or locked up to the top right-hand corner of the screen. It now no longer shows those cursors at all when your mouse is over the UI, which is what you'd expect.
  • The second problem was that some combinations of clicks, and this seemed to vary by OS, could actually wind up activating orders to ships or placement of structures when you were still hovering over the UI. That could lead to insta-placement of command stations in the upper right of the grav well, or similar for turrets and such.
  • IF this was a longer-term issue than we thought, then it could also explain why some units were seeming to randomly path to the upper left right (you were accidentally giving them orders, if that's what this was -- we always suspected decollision, but maybe it was this all along). If this was indeed what was happening, then this issue has actually existed for years and was not related to our linux fix a month ago.
  • Thanks to Arquebus X and mashtong54 for reporting.

Beta 2.763 QoL Explosion

(Released March 25th, 2021)

• Whenever you click a button or icon or other clickable element in the UI, it now causes it to immediately update no matter what.
  • This makes interfaces way more responsive to player clicks, but without having to turn on the unrestricted updates method, which causes framerate drops.
  • Additionally, when it comes specifically to the planet sidebar, it had TWO update timers limiting it (the general purpose one, and then one that predated that which did the same function but was just for that ultra-heavy ui window).
    • The interplay between the two of them led to that specific sidebar to feel extremely laggy, and it was only somewhat better when you put the unrestricted ui update speeds on there. Now it flows vastly better and more consistently, and in particular collapsing and expanding sections happens instantly like you would expect.
  • Thanks to Daw11 for reporting the sluggishness.

• Added another batch of names to the standard naming list.

• The scourge are now smarter at using their builders to expand when there's available territory
  • Thanks to Ecthelon for pointing out some inefficiencies

Planet Sidebar Improvements

• In the event that a player's fleet leader on the sidebar does not have an overlay icon, but is assigned to a control group, then it will show the control group number in the sidebar directly.

• On the sidebar, fleet leaders now always have their own line item no matter what, so that you can hover over them and get their tooltip information for each specific leader.
  • Often you'll have a couple of support factories, or several identical transports, and those should not collapse down in that way.

• On the sidebar, added a new option under the options section, which lets you switch between the existing icon mode, and a new text mode.
  • Ships can either be listed as icons in a grid, or wider text entries in rows. The icons are the classic view, but the text is much easier to visually understand, and is the new default.
  • This turns out to be vastly less tall than we thought it would be.
  • Thanks to Arquebus X and GalacticZombie for being the most recent to request this.

• On the left-hand sidebar, when you have it broken out by faction, it now shows a color on the strength count for that faction that indicates their relationship to you.
  • When you hover over the header to see more about their strength breakdown, it then also explicitly says if they are your ally, enemy, or neutral, etc.
  • This was basically impossible to find in the recent builds unless you still used the by-relationship version of the sidebar.

Selected Ships Window Improvements

• The selected ships window in the bottom right corner now shows ship icons in your ship color, if you have actual ships selected (versus selecting and entire fleet).

• The selected ships window in the bottom right corner now uses the shortened name of ships (if there is one), for the same reason the build sidebar does. Essentially being able to see things that otherwise might be too long and wrap.
  • Thanks to NR SirLimbo for suggesting.

• In the event that you have selected an entire fleet, and the fleet is assigned to a control group number, that number now shows to the left of the fleet name in the selected window.

• The way that the selection window sorts ships and fleets is now completely different.
  • Overall it helps to ensure that the order makes sense and feel more natural, so that you can see things more quickly.

• Added a new "Selected Ships/Fleets Window" subsection to the Display section of settings.
  • Configuration relating to the window that shows the selected ships display in the bottom-right corner of the screen.

• Added a new setting: Max Lines To Show For Selected Ships Before Scrolling
  • When you have a lot of fleets or ships selected, eventually it needs to scroll so that it doesn't go off the top of your screen. But depending on your HUD scale, and your monitor size in general, you may feel like you have more or less room that you want to use for this. The old default was 8 prior to this setting existing, but the new default is 16. Min 4, max 60.

• Added a new setting: Include Selected Ships That Are Selected By Fleet
  • Normally the selected ships and fleets list only shows directly-selected ships, and fleets which are selected as a whole. With this setting on, if you select a fleet as a whole, it will still include all of the full list of ships from it down in the big list under the rest.

• For purposes of the selected ships window, if you have drones selected that are in whatever modes (pursuit, etc), it does not show those buttons as lit-up now. You can't control drones, and their state is irrelevant for larger UI purposes.
  • Apologies for not being able to find who suggested this, but it was a good suggestion.

• For tooltips over ship lines that are related to the selected window, it now includes the total count and strength for all ships of that type, even if some of them are part of a fleet that is not showing the details of what it is in itself.

Fleet Management Improvements

• Fixed a bug in the swap fleets screen where you could actually swap ships IN to unclaimed fleets, thus losing access to a ship line until you claimed the thing. Oops.

• The fleet management window now shows the full cap strength of each ship line (including centerpiece) next to it.
  • Thanks to Badger for suggesting.

• It is now possible to multi-swap several ship lines (from any number of fleets) into the fleet you are viewing.
  • The swap "in and out" is still single-item, but the "swap in from elsewhere" is now multi-swap. It's a straightforward difference when you see it in game, hopefully.
  • This is a huge timesaver, and it DOES properly handle multiple ship lines that are the same type.
  • We have not tested if this properly handles elites (cruisers) being swapped in and out of fleets that have them (supposed to be max one per fleet), but the code is in there to enforce it. We just don't know for sure if that works.
  • The multi-swap variant can never be used with fleet leaders, so that part is not a worry.
  • Thanks to TheIronBird, Zweihander2021, Strategic Sage, and Daw11 for requesting.

• It is now possible to swap flagships between fleets, including golems for transports, or arks for golems, etc.
  • This only works for officer and strike fleets, not any other kinds of fleets (not support, not lone wolf, etc).
  • If the flagships are on different planets, they will be set to 1 health rather than the usual of being destroyed.
  • This can swap across factions!
  • If there are drones in the fleet, it should move them, but this part has not been tested. Only a few golem types really have drones, most of the rest of that is with combat factories, which cannot be swapped in this fashion.
  • Thanks to TheIronBird, Zweihander2021, Strategic Sage, and Daw11 for requesting.

Settings And Options Improvements

• The top line item of the game tab of the settings menu is a new "show advanced settings" toggle. It notes:
  • There are a whole lot of these, friend. If you're not already super into the game or having a specific problem, it's a good idea to steer clear of these. The defaults are awesome 99% of the time.
  • Any setting with is_advanced_setting="true" is no longer shown unless this is on.
  • That wipes out most of the settings, except for the ones that people commonly change.
  • That gets rid of all the debug ones, most of the game ones, all of the performance ones except those that have a large impact, a few of the excessive display ones, a vast majority of the audio settings including the auto mixing tools, almost all of the camera tab, and nothing on the network tab (since that was already pared down).
  • Note that there's still a lot on the display tab since those are things people tend to like to tweak. Given that these are often the chief complaints someone has in a non-advanced fashion, these staying non-advanced seems like a good idea even though this tab is overwhelming.
  • Most of the automation tab is non-advanced, which seems... maybe a good idea? Really unsure there.
  • The audio tab probably still has more than some folks expect, but these bits allow people to get at what they are looking at. Maybe it could be pared down slightly, but I'd do that with caution.
  • The elements left on the camera screen are very key, given those are the things that most often get changed to suit tastes.
  • Thanks to Strategic Sage for inspiring this feature.

• Added a new General section to the Galaxy Options tab.
  • Inside that, added a new Show Advanced Galaxy And Faction Options setting.
  • There's a wild world of tweaks that you can make, many of which are going to feel mod-like in nature. If you like tuning things to your exact specifications, then these are for you. If you don't want to feel overwhelmed, this is probably safely left off.
  • Please note that this option not only affects the options tab, but also the factions tab.
  • This removes the vast majority of settings, most of which are truly advanced things for ultra-tuning mechanics. It leaves the things that are big decision-points that someone might consider while still playing the normal game experience.
  • Thanks to Strategic Sage for expressing how overwhelming this section had become.

• The scouting section on the galaxy options tab has been removed, and its contents folded into the general tab.
  • That said, basically all of them are counted as advanced, and so are invisible.

• All of the AMU settings have been marked as advanced, because by their very nature they are.

• is_advanced_setting="true" can now also be set on custom fields for factions, and the factions tab then shows or hides them based on the option in the galaxy options tab.
  • Here again, the amount of customization options was getting intense, and this really dials that back.
  • The starting economic options on the humans have been hidden unless advanced are turned on(they were already marked as advanced).
  • On AIs, the type of the warden and in particular PG are all hidden unless advanced is on. And the hunter, as it turns out.
  • The DLC2 AI options have provisionally also been hidden under advanced. Because... they are. Even though they are super cool. Same with wormhole borer.
  • The added options on players that AMU adds are also hidden under advanced, same reason.
  • The amount of relief even I feel when looking at these de-cluttered screens now is palpable, I have to say. It's like a giant weight off, especially with AMU in the mix. That said, having the power of all this by turning on advanced feels like a superpower, so we have a double win.

• In the single player lobby, it no longer shows the "humans controlling this faction" stuff. That's only in the MP lobby now.

• On screens where there are hidden advanced settings, it now tells you that those are there, and how many. That way if you are curious, you never need to worry if you're missing something.
  • There are some particularly cool AI settings in DLC2, for example. Those let you really customize how the flow of the map is. But... it's a huge amount of clutter if you're new, or not planning on looking at that right this second.

• Added a new advanced option for AIs: Universal Difficulty
  • Most of the time, having the AI, Hunter, Warden, and Praetorian Guard Difficulty difficulty levels all match within an AI faction is the desired setup.
  • With this setting enable, the subfaction difficulties will be set to match that of the main faction. Key note: that means if you do make changes to them individually, you need to disable this setting or else the central setting will override!
  • This now defaults to ON!

• The difficulty levels for the sub-factions of the AI are now hidden by default, behind the advanced settings, since Universal Difficulty will take care of them for you unless you want to break them out separately.

• The following settings are now stored in the "graphics settings" file that is NOT synced between computers during Steam or GOG cloud sync.
  • Essentially these are per-computer settings, versus settings that really should be shared between computers.
  • This already included all the things related to the screen resolution and windowed mode, and even SOME of the color grading settings. So none of those are changing. All of the rest of what is added will have its values reset to default in the next version if you've changed it from the defaults:
  • Galaxy map bloom, and hexagon intensity.
  • Sound mixing subsection.
  • Performance CPU Load settings including Use Unrestricted UI Update Speeds, Parallel Faction Processing (should leave this ON!).
  • Performance Visual Load settings.
  • Performance Visual And RAM Load settings.
  • Mouse scroll speeds and inversions.
  • HUD Scale (hey, different monitors and different needs).
  • All of the camera speed modifiers.
  • Thanks to marty651 for reporting, and pointing out how so many of these will still vary between machines.

• The little "there are x hidden advanced fields" text things are now clickable directly to give you an override of that feature on whatever tab you are on until you switch tabs. This makes it SO much more convenient to quickly examine just a few fields you are curious about, and/or make changes with them.

Ship Short-Name Improvements

• The combat factories, transports, and forcefield generators now all have shorter versions of their names for use in the sidebar and the selected ships window.
  • It's likely that there are other types of ships and structures that have long names that will wrap inappropriately, and just let us know as you find them, please.
  • For instance, human home command and some of the other structures there now also have shorter name versions.
  • Most capturables now use the acronym that people commonly use for them.
  • Various strikecraft need and now use abbreviations on themselves.
  • Turrets all have a shortening to "Tur" at minimum, if not more.
  • Guard posts now all just say GP, since that's easier to see. And guardians are now GU, since that's easier to tell apart than GD would be.
  • Overall where we can spare a word and still be clear, we're doing so. Orbital Mass Driver is just Mass Driver on the sidebar. Easier to see that way.
  • There's still plenty more we have missed, without a doubt. Volunteer helpers with this would be greatly appreciated.

• All of the various scourge units in DLC1 and DLC2 now have shortened name variants for the sidebar so that they don't wrap and can be understood.

• All drones now have a short sidebar name that uses the designation DRN.

• Frigates are now either in the Heavy or Light class, and for their abbreviation now say either FFL or FFH.
  • So far, the only frigates noted as heavy are the lost spire frigate, spire raider frigate, imperial spire frigate, darker mirror, phantasmal host frigate, heavy minelayer frigate, and the forcefield frigate and its variants. However, that will change.
  • Thanks to Zwei for suggesting, and for helping out with corvettes, destroyers, cruisers, and strikecraft.

• For cruisers, we're now using the designation CC, and for destroyers, the designation DD. Battleships are BB, dreadnoughts are DN.
  • Lesser and Greater are being denoted as Ls and Gr.
  • For corvettes, those are K, and if strikecraft are ever mentioned with a designation, it's SC.
  • For station-keepers, they get a P in front of the rest of their designation. So a station-keeping frigate is PFFL if it's a light frigate.
  • For turrets, the designation is TUR, and for golems it is GOL.
  • One of several good sources: https://en.wikipedia.org/wiki/Glossary_of_military_abbreviations

• Astro Trains now use the designation Astro T.
  • Astro Train Guards use the designation TGU.
  • Battlestations now use the designation BST.
  • Citadels now use the designation CIT.

• In the long name for ships, anything that started out as "Drone Whatever" is now "Whatever Drone."
  • This was the norm for most ships already, and helps keep things sorted together by function on lists. Now it's fully consistent.
  • In DLC2, we have some stuff that starts with Drone, but it's not actually a drone -- it's a thing like a Drone Factory or similar, which helps us tell that it's not a thing that IS a drone, but it ACTS on drones.

• Any minor faction beacons now use the short name designation of BEAC.

• For things like the AI Eyes, which change forms, their short names now start with the same word but put the form change at the end.
  • For sorting purposes, this keeps their order the same, and makes it so that they don't just around in the unit list as their alert status goes on and off.

• All of the various engineers are now denoted in short form as ENG, with a shortened modifier after that if need be. Again, so they all show right next to each other easily on the sidebar and similar.

• Fixed a variety of cases where the short name from a base unit was being applied inappropriately to a variant (among all the cases where the application was appropriate).
  • This was the big thing we were worried about, in the main game and in mods, but we've used the new export spreadsheet (see below) to weed this out for the main game, at least.

• Units that are Dire now have a D in front of their short-form designation (DGU, DGP).
  • Units that are Royal now have an R in front of their short-form designation (RDGU, RGU, etc).
  • The readability and organization of this turned out to be a lot more substantial.

• Anything with the word Gunboat in it now uses the acronym GBT in the short name form.

Critical Note For Modders

• The game now silently complains into your log on startup if any unit's effective "display name for the sidebar" is longer than 18 characters.
  • For anyone with a mod, xml or otherwise, this is a great time to boot up your mod and see if errors are going silently into the log when you do.
  • If so, this is a case where ideally you stick to the central conventions for shorter names, or make up new ones where you need to. This makes a lot of interfaces more consistent and tidy and readable (and prevents wrapping on certain screens that are otherwise unreadable).
  • The full names of ships can continue to be absurdly long, as much as you want.
  • Note that display_name_for_sidebar inherits separately from display_name, so if you are changing that as you inherit from things, you need to be sure to change that anyhow.
  • For convenience, it states the name of the expansions or mods in question for each item.

• Our "GameEntityReferenceData.csv" export has been updated to include far more information, including the long and short names of ships, their internal name, what mod or expansion or both they are from, and to be sorted by mod and expansion before name.
  • It also includes the length of the short name, and makes it easy for us to figure out if there are any mismatches between short and long names (those are bad). The internalname is never seen by players, so is often mismatched, and that's no problem.
  • Long name is display_name (DisplayName), and short name is display_name_for_sidebar (DisplayNameForSidebar). It's used on more than just the sidebar now.

• There is a new xlsx excel file that we are distributing with the game, which modders or players can paste data into on the second tab, from the csv file, and then have nicely-formatted data with conditional rulesets to help find problems, and things of that nature. This should make a lot of life much easier for folks looking to find inconsistencies.

New Mods! Tame DS and Powerful Cmd Stations

• Added a new Tame Dark Spire mod by ArnaudB, made thanks to the inspiration of FwiffoForce.
  • A mod that prevent Dark Spire from expanding. Either in normal games or during Fallen Spire campaign. They won't build locus.

• Added a new Powerful Command Stations mod by ArnaudB.
  • This changes the way Command Stations work, shifting forcefields to the stations themselves: you get a cheap economic station for 60k, a shield protected logistical one for 200k, and a VERY powerful military station that cost 500,000 metal.
  • A hull tech for Command Stations is added.
  • Stations have no bonus turrets.
  • In addition forcefields are rebalanced here. You start with one on every station, but you can hack for more at the ODSS.

Mod Updates

• The Shield Generators mod by cml has been updated to meet the new standards.

• The More System Defenders mod by CRCGamer has been updated to include the new short-name styles.

• Spire Railgun Shop by Lord of Nothing has been updated to include the new short-name styles.

• AMU:
  • Reorganized the base types: Instead of just Strike Craft and just Frigates there are now Light/Medium/Heavy variants of those similar to how Small/Medium/Big turrets work, with escalating "suggested base stats" and values.
    • Mods now using these types will likely have different icons, sizes, targeting priorites etc for some ships based on this. Too many minor changes to count...
  • Brought other types in line with that.
  • Added Destroyer and Cruiser base types.
  • Removed the LargeShipNonStackable base and replaced it with a Battleship and Dreadnought base type.
  • Added sidebar-display-names for all the AMU base entities based on what Vanilla uses.
  • All variants of the AddToShipGrantList() function now require a multiplier for frigates, since these are usually much much stronger than strike craft ship lines.
• Extended Ship Variants:
  • Added sidebar-display-names for all the AMU base entities based on what Vanilla uses.
  • Copies from the Cloaked and Parasitic variants of the Viral Shredder no longer state that they are copies. They are supposed to just "mix in" with the real ones.
  • The Shredder Frigate now uses the AMU Medium Frigate base type.
  • The Sharpener Frigate has been renamed into the Heavy Sharpener Frigate. It always was a heavy one, but didn't say so.
  • EMP Missiles no longer count as Strike Craft but as Frigates.
  • All of my mods put the short ship definition (FFL = Light Frigate, C-FAC = Combat Factory, etc) to the REAR of the game entity type. ESV changes the Vanilla Combat Factories to do this as well.
• Kaizers Marauders:
  • Began using the new AMU base types.
  • Added or changed sidebar-display-names where needed.
  • The Disrupter Corvette is now the Disrupter Heavy Corvette, and as such is a bit larger and has more mass and engine power.
  • The Light Sting Frigate was renamed into the Sting Light Frigate to put the size in front of the ship type.
  • The Rampage Frigate was renamed into the Rampage Heavy Frigate, and as such is a bit larger and has more mass and engine power.
  • The Raid Cruiser was renamed into the Raid Seeker, due to it not actually fitting in with DLC2 Cruisers at all.
  • The Raid Battleship was renamed into the Raid Super Battleship, due to it not actually being a Battleship, whenever those actually become a thing, of any sort.
  • Kaizer's [Spire] Dreadnought was renamed into Kaizer's [Spire] Super-Dreadnought, due to it not actually being a Dreadnought, whenever those actually become a thing, of any sort.
  • Frigates hacked now have 4x the strength cap they had before, leading to way higher and more significant counts, fitting in better with the Vanilla Frigates.
• Micro Mod Collection:
  • Added sidebar-display-names where needed.

• Kaizers Marauders:
  • Reduced the Hull of the Sting Light Frigate from 25k to 15k, Shield from 125k to 85k and Damage from 40k to 25k. In return the metal cost was reduced from 104k to 67.5k and the energy cost from 6k to 4k.
  • In return Marauders and players if they hack it now get about 35.1% more, and they are about that margin cheaper.
  • The purpose was to bring it in line with other Light Frigates, which have around 1-ish strength. The Sting Light Frigate now has 0.97s, down from before 1.51s. At that strength it would've been what I consider a Medium Frigate.

Balance

• Adjusted HRF special ARS ship grant caps to be actually competitive after recent adjustments to Dark Spire and Dyson.
  • Concussive Defender 4 -> 8
  • Tesla Defender 4 -> 6
  • Grenade Launcher Protector 4 -> 6
  • Nucleophilic Protector 3 -> 4
  • Pike Defender 13 -> 30
  • Raider Defender 13 -> 30
  • Vanguard Defender 13 -> 15
  • Pulsar Tank Defender 13 -> 15
  • Metal cost of Concussive Defender adjusted from 80K to 40K. (It has half the health of any other frigate offering here.)
  • Concussive Defender speed adjusted from 500 to 600. Same for Tesla Defender.
  • HRF Special ARS adjusted to allow up to three hacks. Expected costs of 12, 18, and 27 HaP for each hack.
  • Thanks to CRCGamer for noticing, and making these changes!

Bugfixes

• Put in one more check that could still happen when returning entity orders to the pool from multiple threads. It would cause a visible popup of "Mismatch on ClaimedByEntity_Internal" now, but was not a constant thing, so didn't slow down the game like before.

• We were still having some (legitimate) errors with entity orders going back into the cross-threading pool after being split between multiple entities improperly. It's rare, but now rather than showing an error message when that happens, we just discard the invalid order instead of putting it back into the pool where it might further compound errors. The garbage collector will pick it up, and all will be well. This has been an interesting exercise in cross-threading memory cleanup.

• The TSS and ODSS and similar structures now should automatically correct any too-high ship/turret/etc line grants that they have. This doesn't affect things like the ARS or FRS.

• Remove some debugging code that was spamming ArcenDebugLog messages about enemy king attackers
  • Thanks to Daw11 for reporting and Badger for fixing.

• Fixed a minor visual bug where on a bunch of windows, the start of the first line was 6px too high, and thus sometimes actually cut off.

• ArcenDynamicTableRows now properly are linked to expansions and mods that originate them (not ones that just alter them with a partial record).
  • This DOES include rows that originate from a copy_from from some other source like the main game or a different mod or expansion.

• The "skip_export" xml tag is no longer used, although it can be left in place on units.
  • We actually specifically want ALL of our units to export now.
  • Just for cleanliness, we have removed that tag from all of the units in the main game and expansions.
  • Also removed some of the last graphicsdone="true" references from our own xml.

• Fixed a bug in GetDisplayNameInternal() for factions that could still throw an exception during stringbuilder creation of names, because of the field we marked as ThreadStatic... not being static. Heh. This is more efficient now in general, as a side bonus.
  • Thanks to Badger for reporting.

• Fixed an exception that could happen when writing tooltips for units that were already maxed out at mark 7.
  • Thanks to Badger for reporting.

• A mysterious error in ActuallyFireSalvoAtTargetPriorityList() appeared in a multiplayer game, but we have no idea what it was. That code has now been fully instrumented so that next time it happens, we can actually fix it. It is either MP-specific, or related to a new feature on the beta branch, probably.
  • Thanks to Badger for reporting.

• Fixed an oversight that would lead to exceptions on EjectEntireStackFromMyselfIfPresent() on multiplayer clients. That code simply should not be run on clients.
  • Thanks to Badger for reporting.

• Fixed an exception that could happen in strength counting, mostly in a cross-threading fashion but maybe slightly more likely in multiplayer. A side effect of this change is that the code is also very slightly more efficient now, too.
  • Thanks to Badger for reporting.

• Put in an (untested) fix for the ability to watch fleets in multiplayer not working reliably. This was related to the player account ID and how it was sent, most likely. We now explicitly send it.
  • Thanks to Badger for reporting.

• The game now clears out old stale fleet leader slots on player fleets. This is essentially when you've transformed the type of the fleet leader, or switched what kind of command station you had, but it still had some old data.
  • This may not wind up being needed or a good idea, but it does remove some ambiguity and does also make savegames a tiny bit smaller. We thought this was responsible for a certain bug, but it was not.

• Fixed a minor bug in the fleet list of units where elites could wind up showing ahead of fleet leaders.

• Updated a number of journal entries and tips to be a bit more accurate to the new paradigm.
  • Thanks to Lord Of Nothing for pointing out some holes, and CRCGamer for filling them.

Beta 2.762 Performance Hotfix

(Released March 22nd, 2021)

• Changed the golem so it does damage again properly.
  • Since zombification isn't affected by bonus multiplier, Botnet can indeed zombify guardians
  • After considering, this was left in since Parasite Citadel can do the same for 30 AIP with only less efficiency. Botnet will (probably) be allowed to do the same for 40 AIP till zombification-mechanic get some update.
  • Thanks to ArnaudB for doing this.

• Dyson Sentinel had a 0.19 strength multiplier, making it looks much weaker than it was.
  • Totally not a bug made into a feature: Dyson units now have a 0.50 strength multiplier and their description reflect this.
  • This gives hackable Dysons an 'unique' ability useful to the player, to avoid triggering eyes for instance.
  • Thanks to ArnaudB for doing this.

• Updated the wording on the mark 2 and 3 increases for the AIs when attacking their homeworlds and nearby planets to say "if you have at least 10 strength outside of transports" rather than the more confusing "if you unload at least 10 strength from a transport."
  • Thanks to fighterman481 for suggesting.

• Fixed an exception that could happen in ClearOrders() at debug stage 400.
  • Thanks to alocritani for reporting.

• Fixed a bug from the last (non-hotfix) version that was causing the Player.log to be spammed with "Mismatch on ClaimedByEntity_Internal when returning to the pool!" errors.
  • This was causing massive amounts of lag in all sorts of unexpected situations, and the main predictor of just how bad the lag was, was how busy your hard drive was and how long you had been playing (this would get exponentially worse over time).
  • The root cause here was me moving when a variable was nulled-out in order to make cross-threading issues less likely. However, after that point it does a validity check and complains if there is a problem. This check is important, but it was always now failing against a now-null variable.
  • The validity check now uses a local copy of what is nulled out, so spurious errors aren't thrown. Additionally, if it DOES have an error, it will now go to the main log and the display, so we don't get surprise hidden problems like this.
  • Thanks to Sigma7, Metrekec, ArnaudB, alocritani, Badger, and others for reporting.

• Fixed some more hidden log messages that would only go to the unity Player.log instead of the usual logs we look in. The only ones that were left in the game beyond the one we fixed just a moment ago were some relating to wave debugging being on. This would exhibit a similar style of mysterious performance drain if someone had that logging turned on for some reason and didn't know what log to look in, but it was very unlikely to ever actually be a problem. Still, better safe than sorry!

Beta 2.761 Hacking Hotfix

(Released March 19th, 2021)

• Fixed a bug in the prior release that made all hacks not show up for players.
  • Thanks to Sol for reporting.

Beta 2.760 Xenophiles And Overlords

(Released March 19th, 2021)

• Minor faction allied Scourge is now allowed to do outright invasions like the Marauders. This complements but does not replace the existing "Join minor faction allies" behaviour.
  • So if you have just a scourge on a team, it will invade the galaxy sometimes
  • If the scourge are with the marauders then the scourge can see that they have allies and can join them (this is the current behavior)

• Added a new state_of_matter_to_become_on_wormhole_exit="YourStateOfMatterHere" and returns_to_default_state_of_matter_after_seconds_from_wormhole_exit="numberofseconds".
  • What it does is every time the ship exits a wormhole (this does NOT work when coming out of transports or exiting guard posts), it'll change to the other state of matter for X amount of seconds.
  • Completely untested so far.
  • Thanks to NR SirLimbo for suggesting.

Planet Sidebar Enhancements

• The data for which fleets you were watching in old savegames is now lost, sorry about that.
  • However, in new savegames where there is shared-faction multiplayer, the watchlists are now per-playeraccount, which means each person can just watch THEIR fleets that they are responsible for.
  • Thanks to Badger for requesting.

• When hovering a Fleet's entry in the Planet menu, there's now a hotkey to toggle between being Watched and Unwatched.

• Fleets with longer names now have their text size decreased

• Instead of showing the raw number of ships (which is not as important as the Strength of ships), it now shows "Enemy Strength on Flagship Planet"
  • This way you can monitor multiple battles on multiple planets at once just by looking at the sidebar

DLC2 Delay to May 18th, 2021

• Updated the release date for the second expansion to be May 18th, 2021. We all just needed more time to work on this and get it really in excellent shape. Apologies for the extended wait that keeps extending, but the added time has been a really positive thing for the base game (as you can see lately) as well as for the DLC itself (wow those units and factions are coming together). We really want to have something that is polished and awesome for you from day one for the DLC, versus scrambling to fix things on launch of it.

The Xenophile Update

• An update to shipline hacks from minor factions, making them use the new ARS/TSS multiple sectors system and aiming to make those hacks tempting instead of disappointing.
  • All shipline-Hacks on aliens cost more on hostile planets but less on friendly ones. Shiplines hacks are now much faster, but get much harder every-time you hack the same target (like ARS and TSS).
  • All shipline-Hacks on aliens now give +3 AIP per shipline hacked.

• Dark Spire Wraith increased speed from 400 to 500. Damage reduced from 2800 to 1200, added x2 bonus damage against enemies below 80 armor, which includes most turrets.
  • Ship cap increased from 8 to 20.
  • Now a close-combat high-damage ship that's pricy and moderately tough. Much more compelling to use. The changes also affects base version and is effectively a nerf.
• Dark Spire Phantom damage reduced from 12,000 to 4,000. Hackable version had 5000 before, it's now the same for both base and hackable version, like all others Dark Spire ships.
  • Ship cap increased from 9 to 14, like Bombard cap.
  • It's theoretially three times as powerful as a bombard, but has more weaknesses from bonus damage and lower range.
• Dark Spire Specter ship cap increased from 4 to 8.
  • It's a powerful multi-shot anti-shield but no tougher than most frigates.

• Dark Spire Constants changes:
  • Dark Spire Conversion ratio based on intensity is now 40/50/60 instead of 30/50/70 at low/mid/high intensities.
  • Dark Spire "Income per minute" multiplier is now 0.08/0.1/0.14 instead of 0.7/1.0/2.0 to reduce the impact of Dark Spire on its shiplines are hacked, as the shiplines have been balanced and the hacks have greater cost. Narrowed variance between intensity is safer for DLC3 rogue-like options. This also helps performance against vengeance strike caused by high energy income causing vengeance strike loops.

• Dyson Sentinel health increased by 100% from 1800 to 3600, damage reduced by 180% from 4150 to 750.
  • The exemple of something going horribly wrong. It had enough DPS to rivalize three human lines of strikecrafts and really wasn't supposed to. With FRS it was gamebreaking.
  • Ship cap increased from 20 to 30.
• Dyson Defender health increased by 200% from 2000 to 6000, damage increased by 100% from 1500 to 3000.
  • Another example of the messy balance of Dyson. It's now tougher than the Sentinel but slower with less range.
  • Ship cap reduced from 32 to 30.
• Dyson Bulwark shield increased from 10k to 100k, half its hull. Damage of AOE weapon reduced from 4400 to 4000, damage of second weapon increased from 4150 to 6600.
  • Ship cap increased from 4 to 6.
• Dyson Bastion shield increased from 10k to 100k, half its hull (not a hackable ship).

• Alien hull tech cost lowered to: "500,1000,2000,3000,4000"
  • The first two levels are meant to be cheap, so players can easily get them. There is less science to go around so it's still a commitment, and you need to seriously increase AIP or/and spend a lot of hacking points in order to get enough alien shiplines to matter.
• Sentries tech shifted to: "500,1000,2000,2500"
  • 250 science of Level 2 shifted to level 1, level 3 lowered by 250 for aesthetic. Pricier level 1 means less autopicking level 1 and level 2 look more tempting even though the cost for level 1+2 is the same.

• Destabilize Wormhole hack cost reduce from 120 to 20. A long delayed change due to being very low priority. It might be useful now.

• Fallen Spire Frigate primary weapon damage reduced by 50%, added a secondary 20 shots weapon with dps equal to 50% of that reduced damage. DPS effectively unchanged.
  • This makes FS frigate a proper escort ship for capital ships, cleaning chaff while bigger ships focus on the bigger targets.
  • The primary reason for this change is performance. Big ships are the first to go down on both side, however when fighting the AI it often means the AI retreat once its big units go down. Over multiple battles, more and more 'chaff' ends up surviving as big units die. However big units are much more performance-friendly. This should greatly help early-mid and particularly late Fallen Spire games.
• Fallen Spire Railfrigate from SpireShop mod has also been modified to fit. Its weapons values had been rounded upward, its secondary weapon hits 20 targets instead of 10.
  • Railfrigate now has 25k dps instead of 20k dps. Standard Spire Frigate has 30k dps and remains better stats-wise.
  • Its description mentions AI forces will focus it first, due to being it being a sniper. (This is why standard spire frigate survive MUCH better than railfrigate variant)

• Huge thanks to ArnaudB for coming up with this and surprising us out of the blue with it! This is very exciting.

Balance

• Nerfed Polarizer base damage from 150 to 125.
  • Thanks to CRCGamer for making the change.

• Added a new only_show_when_max_ai_difficulty_is_greater_than_or_equal_to and only_show_when_max_ai_difficulty_is_less_than to hacks (they default to 0 and 99, respectively).
  • The idea here is to let us have variants of the hacks for different difficulty levels, or different difficulty ranges.
  • Normally hacks already scale by difficulty, but some just don't scale well across difficulties, and trying to otherwise massage the numbers to make that work just is not always feasible.
  • This actually gives us substantially more freedom, because it allows us to completely swap things out at different levels if we want to.

• There are now three versions of the superterminal hack. One for difficulties 7 and down, one for diff 8, and one for diff 9+.
  • The 7-and-down version has been returned to historical values, which are much more severe. Strategic Sage pointed out just how weak the response was after our recent round of nerfs, when it comes to those difficulties.
  • Currently both difficulty 8, and difficulty 9+, use copies of the same new-nerfed values, to keep those responses from being so absolutely insane. It may be the diff 8 actually needs to be a bit stronger, not sure. But essentially the scaling on this particular hack was just so intense that at 8+, and 9+ even moreso, the hack was almost impossibly difficult based on testing by ArnaudB and others.
  • Please let us know what your experiences are at difficulties 8 and 9, so that we can tune those up or down a bit more, as needed.

• Further change: the difficulty 10 superterminal hack is now back to the pre-nerf values as well, and diff 8 and 9 can now be balanced independently. Those are both still at the nerfed values, but if they are too easy, please let us know.
  • Thanks to Strategic Sage and Metrekec for reporting.

• Botnet golem now works like Sabot Guard Post: x100 damage against targets below 100 tx. It can attack everything and its damage, for both base and AI-version, is 1% of its previous damage. So same damage with multiplier.
  • A realization this was a much better way to handle Botnet while working on DLC2 golems prompted this change. This should solve the weird targeting oddities Botnet had, particularly with its excessive tendency to rush into enemy forces.
  • AIP cost reduced from 50 to 40. It cannot be buffed by FRS and it's immensely pricier to repair, so like Hive Golem an AIP cost reduction is justified.
  • Thanks to ArnaudB for orchestrating.

• Two additions to the Watchman Frigate station-keeper.
  • First it has 33% reduced damage taken from anything that has arrived on the planet it is protecting within the last 12 seconds.
  • Secondly it always only takes half damage from opponents with an albedo of .7 or higher which is mostly stealthy cloaked types. It is named the Watchman after all.
  • Thanks to CRCGamer for coming up with and implementing this.

• Spire Dreadnought only gain +50% damage to its beam weapon compared to the battleship, however it gains a secondary 20 multi-shots weapon like the frigate, with as much dps.
  • It has effectively 150%+150% damage compared to battleship, it's 3x as powerful instead of 2x.
  • The Imperial Spire Dreadnought also benefit from this, and should be able to clean up the map much more easily, leading to better performance.
  • Thanks to ArnaudB for setting this up.

The New AI Overlord Phase 2

• Added forced_to_be_this_state_of_matter_when_on_former_or_current_ai_homeworld_or_bastion_worlds, which can make units be a certain state of matter when at current or former AI Homeworlds and Bastion worlds.
  • The AI Overlord Phase 2 now has this set to solo-phase, which means it has time to escape from its planets and for you then to be able to fight it.

• Normal AI threat (which doesn't last long, since usually it joins the hunter fleet) now considers the AI Overlord to be perpetually in danger if it is Phase 2, and so should move to protect it more.

• When the AI Overlord has moved into Phase 2 AND is no longer phased out, the AI sub-factions now abandon whatever they were planning and move in. This may mean that the PG starts striking at a very inopportune time for you, which should be pretty exciting.

• Added a new findNearestHumanKing() to the faction utility methods.

• There is a new should_not_be_given_normal_orders_by_ai_logic="true" that we can assign to certain units to exclude them from fireteams and other rational behaviors. For these we can then give them some sort of other logic.
  • There is should_always_be_given_orders_to_attack_player_king="true" that can be combined with the above, and when that has been done, it will send that unit on a path of destruction to the nearest human king.
  • Phase 2 of the AI Overlord now has both of these behaviors, while the rest of the AI forces that are subordinate to it are mostly focused on defending it and supporting it (so will come along kind of despite themselves).

• For the first time, the AI Overlord Phase 2 now properly shows up on the normal galaxy map as an icon. Very important now that it's running for you so fast.

• The AI Overlord Phase 1 no longer gives you any science or hacking points. That was making the end of the game too easy.

• The AI Overlord Phase 2 can now pass right through forcefields (in addition to having the Norris Effect), so your military commands can't just block it.

• The movement speed of the AI Overlord Phase 2 has been dropped from 400 to 300, and its health has been adjusted from 25 million to 64 million.
  • Its parasite bolts have also increased in salvo size from 60 to 80.
  • Hopefully this is an appropriate amount of terrifying; we shall see. In initial tests, it is definitely very Terminator-esque, which was the idea. It can still get caught and die on a single very heavily defended military homeworld choke point if there is one. Which is fine, good on you if you have that sort of defense!

• Thanks to Metrekec, ArnaudB, Badger, and Zeus for the discussion that led to this.

Player Command Station Balance Adjustments

• Logistical command station balance adjustments:
  • Tractors down from 14 to 10.
  • Forcefields down from 6 to 3.
  • Gravity generators down from 6 to 3.
  • Base allied speed multiplier down from 3.5x to 2.5x, plus 0.5x per mark rather than 1x per mark.
  • Metal production down from 120 base + 80 per mark to 90 base plus 50 per mark.
  • Energy production down from 250k base + 100k per mark to 175k base + 50k per mark.
  • These are no longer quite such a huge standalone middle ground between economic and military. The goal is for people to use all three kinds, just in various ratios. These are also just not quite as good at slowing down enemies to a ridiculous degree like they used to be.

• Military commadn station balance adjustments:
  • Tractors down from 22 to 16.
  • Immobile forcefields down from 4 to 3.
  • Allied attack multiplier down from 1.25x to 1.1x, and amount added per mark down to 0.15 rather than 0.25.
  • Defensive cap multiplier also down from 1.33 to 1.15.
  • Metal production down from 60 base + 40 per mark to 50 base plus 20 per mark.
  • Energy production alerted from 40k base + 20k per mark to 70k base + 10k per mark.
  • These planets were such insane roadblocks for the AI that the only thing offsetting them was player metal and energy. We want these to combo well with economic, but it was just a bit too much. The economic changes also make it more stand-alone on the energy side, but less so on the metal side.

• Economic command station balance adjustments:
  • Forcefields down from 2 to 1.
  • Gravity generators up from 0 to 1.
  • Metal generation up from 250 base + 200 per mark to 400 base + 220 per mark.
  • Energy generation up from 400k base + 200k per mark to 500k base + 240k per mark.
  • These changes make them slightly more vulnerable even than they were before, but good grief can these power a ton of other planets in terms of metal and energy.

• Thanks to Metrekec for the suggestions in this direction after his most recent difficulty 10 win.

Bugfixes

• When the nanocaust hive is destroyed the nanocaust stop doing anything; this is generally how hive-based creatures work. However, they now attrition much faster at this point, so we don't wind up having 3K nanocaust strength sitting around doing nothing
  • Thanks to TheIronBird for reporting.

• Hopefully fix a problem hovering fleets in the Planet sidebar
  • Thanks to alocritani for reporting

• Fixed a bug with the dark spire energy harvesting where, since the last build or so, it was not properly using the harvest ratios for the factions dark spire, macrophage, or devourer golem.
  • While we were at it, also made it so that it now properly harvests at a better ratio for the many subfactions of the AI.
  • Thanks to ArnaudB for discovering.

• Removed a couple of exceptions that we were throwing as a sort of self-check in Helper_SendExoGalacticAttack_SingleExoTarget.
  • Essentially, the only time that these could fire is if an exo was just launching against a target right as that target died. In those cases it was leading to an exception, but that wasn't actually an error state, so there was no real point to this check at this point. In the past, it helped identify some edge case errors, but we're all good now.
  • Thanks to Sigma7 and Daw11 for reporting.

• Updated beam weapons so that they no longer strike units that are invalid for them based on their normal targeting restrictions (aka, if they can only hit mobile units, then happening to pass over structures with the beam won't hit the structures anymore).
  • Also updated this for AOE in general, which had a code branch that was kind of mysteriously never being called. So if there is a minefield that a cruiser wanders over, for example, and the minefield only targets cruisers and things that are large like that, the minefield won't also damage nearby strikecraft anymore (essentially, that was likely to waste some of the power of the minefield in almost all cases, anyhow). If we need to make some further adjustments to this, we can, but right now the logic as it was before seemed to be a bit off to say the least.
  • Thanks to Sol for reporting.

• Fixed a remaining exceptions that could still happen in ClearOrders() due to some cross-threading issues.
  • Thanks to Sigma7 for reporting.

• The way that ship targeting works, if a target expects to be overkilled, no other ships will fire on it. This can sometimes backfire if the ship expecting to be overkilled is wrong about that. Or it can lead to ships that are closer who can fire faster shots to kill it faster waiting for a slow-moving shot from far away to arrive, which is also non-ideal.
  • We have now set this up where there is a two-pass system for ship targeting. In the first pass, things work like they have up until now. In the second pass, which only happens on systems that have 10s or less reload times (by default; not counting slowing or speeding modifiers), it now ignores overkill damage and if there are still any shots available within the current salvo, it fires on them anyhow.
  • This should solve a number of edge cases of "ships not shooting at targets that are in range," both in a short-term sense (big shot is incoming and it waits for it in a way that seems stupid) and in a long-term sense (something is wrong and nobody shoots anymore). However, this will need testing, since that was intermittent.
  • Thanks to CRCGamer and Democracy for reporting the latter case with Lanternfish Turrets.

• Ships that have incoming shots attached to them, which is based around the concept of preventing overkill, now evaluate their lists of shots each sim step and cull any that shouldn't be there.
  • This happens for a variety of reasons, like ships that change states of matter, change planets, shots that are intercepted in some fashion, and so on.
  • It's really unclear if this data was sitting around and getting stale for the long-term over the last few years (until you save and reload), but it sure looks like it could have been. The more recent changes to the game have made it more likely to actually trigger, but the base thing seems to have been kind of old.
    • Fun fact: we did have code for removing this sort of thing already, but it would ONLY trigger in the event that the shot that would be fired against the ship was not enough to kill it. So if a ship had a stale incomingshot, and it was at something low like 100 health left, and all enemies did at least 100 damage, then it would never reach the part of the code where it would clear the stale incomingshot data. This seems most likely to be what was going on with the Lanternfish Turret bug, but it's hard to be sure just yet. Please let us know what you find in the new version!
  • Thanks to CRCGamer and Democracy for reporting.

• Put in a secondary check for unit death on them taking damage. This might hopefully help with ships that were taking self-damage but not actually dying until something else shot them.
  • Thanks to Zeus for reporting.

• Added the following to the Zenith Trader description: When it is on one of your planets, you will see its wares in your Build menu in a section near the bottom.
  • Thanks to Lyrad8791 for alerting us to the fact that this was never specified anywhere in the game.

• Fixed a bug in the galaxy map filters where if you chose the "Normally invisible" option, it would just throw endless errors rather than showing you a value.
  • Thanks to alocritani for reporting.

• Some factions can request that threat against them go back to guarding (like zombies, dyson sphere, etc). This request was not being always honored, especially the Dyson, which could to vast quantities of threat against the Dyson that would eventually go after you.
  • Thanks to Badger for finding and fixing.

• Rather than the previous logic (from inwork on this release) trying to get rid of threat against other factions, which was either too aggressive or not aggressive enough, we now have a hunter_targeting_this_faction_dies_immediately="true" that has been added to the dyson sphere factions.
  • Essentially, if hunter are after them, it's time for those hunter units to die. Those were the ones coming after your homeworld erroneously. If it's other forms of threat against that faction, or threat in general against other factions, those seem okay.
  • Thanks to Arides, ArnaudB, and others for reporting. And big thanks to Badger for advice on how to best make this work.

• Similarly, for the warden and praetorian guard, if they are targeting a faction that is nowhere near them, then they now also get killed off.
  • This solves some issues where they were building up against a secondary faction, then could not reach them anymore (probably because they murdered the threat), but now they just sit around like idiots.
  • Killing these units allows them to respawn better ones that actually will do things like defend their homeworld from the players.
  • We don't know for sure this is why some PG were sitting off to the side, but we suspect it could be related.

• Fixed some issues that could cause the inability to deserialize a savegame to get you into a strange invalid state rather than just telling you it didn't work.

Beta 2.758 Threat Explanation

(Released March 16th, 2021)

• Tweaks to the new planet sidebar:
  • If Watched and Local fleets are both below the Ships, Watched fleets are above the Local fleets.
  • If a fleet is both watched and Local it now appears under Watched rather than local.
  • If a fleet is local, its hotkey is now shown in green. This way you can now tell which of your watched fleets are local.
  • Hovering the fleet in the sidebar now gives text similar to the Fleet Health menu
  • Thanks to Badger for proposing and for implementing these!

• Plasma Guardian lowered Albedo from 0.4 to 0.3
  • Absence of commonplace counters made it too problematic in waves, this makes it vulnerable to tractor which it does have bonus damage against.
  • Thanks to ArnaudB for adjusting.

Bugfixes

• Fix a bug where crippled flagships in the Planet bar were reporting 2 ticks of health instead of 1 tick

• Add some more debugging to errors in exogalactic strikeforce generation

• Fix a bug where you toggling fleets between Watched and Unwatched would fail
  • Thanks to Daw11 for reporting

• Added a fix to situations where the AIP was zero potentially causing GetSpendingRatios() to have a divide by zero error. Maybe this was just something that happened on initial game load for part of a second while data was being loaded, it's not clear, and maybe it's only on certain machines with (particularly fast?) types of multi-core setups. At any rate, it will now handle this gracefully instead of shutting down the AI faction, and it shouldn't even be an error anymore.
  • Thanks to Master Frost for reporting.

• QueueLogJournalEntryToSidebar() has been updated so that the old version is obsolete, and modders and similar need to update their code to use a new version that has an OnClient enum parameter.
  • This new parameter generally tells the game to either ignore this on clients -- almost always a good idea, unless this is something detected only on a single client only -- or run this on the host only.
  • The default for anything that is still using the deprecated versions is that it will run on the host only.
  • Essentially, these are quite midsize messages that should not be going out in large batches, but sometimes were. This would potentially swamp a multiplayer game, potentially if it actually created all those journal entries at the far end (you'd have tons of duplicates and savegames would be strangely large.

• Fixed the following journal entries that were at bare minimum happening too frequently in multiplayer because client were trying to do it, but may have happened even more than that.
  • First time met PG, hunter, AI reserves, warden, dark spire, astro trains, dark zenith, outguard.
  • All the various astro train secondary notices.
  • All of the "base lore" journals for the various races and similar.
  • Imperial spire arriving, and nanocasut invasion starting.
  • All of the other journal entries were fine. Things like the beginner journals, info on allies, zenith architrave stuff, extragalactic war starting, and many various other more specific situational ones were already MP-safe.
  • Thanks to Badger for noticing the insane number of these that were being queued in his multiplayer games.

• The AI Defenses galaxy map filter no longer shows data for unexplored planets.
  • Additionally, it now only shows the amount of defense that you actually know about, and if there are more that you don't know about it shows a + sign after it.
    • This in itself does reveal some information to you, but that feels appropriate. Is this one unit we're talking about, or 5000? Something more is there, but how much?
  • Thanks to Daw11 for reporting.

• Fixed an issue with the new planet sidebar where it was not properly caring about the amount of time units had been on a planet when deciding what of them to show in the new by faction and by faction and mobility views.
  • We haven't been able to replicate the issue, but we've seen the screenshots and also we saw the wrong code. So hopefully fixing the wrong code will fix the issue that others were seeing.
  • Thanks to Daw11 and Zweihander2021for reporting.

• Fixed a bug where, since just after the new states of matter were added, we've had an issue with scouting intel being wiped out on old planets any time a save was loaded (basically every ship was being told it had just arrived on the planet), AND entirely wiped out for multiplayer clients in general.
  • Thanks to Zweihander2021, alocritani, and RedPine for reporting.

• Fixed a bug where the tooltips would show you the AI homeworld without you having ever scouted there. This was new sometime recently in the beta.
  • Thanks to alocritani for reporting.

• Fixed a rare exception that could happen in HandleSpecialEffectRequests(), and made it a bit faster while we were at it.
  • It's possible that the exception will still happen, but if so we will now have more information on where it happens.
  • Thanks to Zweihander2021 for reporting.

• Sometimes Unity logs some spurious errors that start out "<RI.Hid>" and those would pop up and get in your error log as well. Those no longer will do that.
  • Those still go to the unity log if there is truly a problem, but almost universally it's something like "oh we can't find your other monitor that you disconnected since last time, well no worries we'll just use your main one." These are not things that should be throwing freaky errors for you to see.
  • Thanks to Zweihander2021 for reporting.

• When you are in a Drain situation in metal, it now should give you a more accurate percentage of what you have remaining, as well as a more accurate timer of how long it will take to complete all of your current work.
  • Previously, it was really just wrong in a variety of ways.
    • Now it only counts each thing being built in a fleet once, no matter how many factories were helping it. Previously, 3-4 factories would cause it to to be 3-4x as high.
    • However, now it also counts how long it will take to build the rest of the entire queue for a fleet item, if it's not disabled, so that might be 50 things rather than 1, 3, or 4.
  • Anything that was self-building, or being assisted by engineers was already correctly calculated.
  • Any drones being built were already correctly calculated, and same with rebuilding remains, in that those don't actually cost you any metal at all. Note that after something is rebuilt from remains, it has to be repaired, and THAT part has a cost, which is correctly tracked now.
  • Anything being claimed is now correctly tracked, whereas before it was estimating vastly too much. If a golem that cost 1m metal had 5 things working on it, it would think you needed 5m metal. Way off.
  • Engine repairs are free, and those are now properly ignored. The metal allocated for them is just for how fast they are repaired, but it was showing up as an actual amount you'd need to spend. Same thing for personal shields being repaired. That's also free.
  • For hull health and bubble forcefield repair, it now tracks the remaining cost correctly, versus before it was only including a tiny fraction of what was needed, plus also multiplying that by the number of repairing engineers. That just had no connection to reality, really.
  • Thanks to Sigma7 for reporting.

Accurate Threat Counts, Hunters, And Similar

• First off: what are we on about with the below changes?
  • Well, some funky things are happening with the hunter fleet, and that seemed like a symtom of a larger issue.
  • As part of that, I found some code that was quite inefficient (string comparisons), and also potentially sometimes wrong (turns out the flavor of hunter you selected could make certain code work or not).
  • Untangling all of this took more time than originally intended, but should solve some issues as well as fixing the potential for future mistakes.
  • Overall this was a giant rabbit hole that did not solve the original problem, but did uncover a couple of others and proactively solve them, hopefully without causing new ones. Some of the strangeness of the hunter and similar sometimes not behaving may now be fixed, but it's hard to say.

• GetSpecialFactionName() has been deprecated, as all the places that were using it were basically either using it inefficiently or using it when a nicer name should instead be showed.
  • There are a variety of places in the interface that are now more clear and less debug-text-looking, and other parts of the engine that simply run faster.

• It's possible that hacking the Vengeance Generator from the Dark Spire, and other similar actions, were just broken by the above change.
  • Same with Dyson Sphere hacks.
  • Same with Dark Zenith Library.
  • Same with Zenith Architrave Portal
  • If anyone can confirm if those various things still work -- in a fresh campaign, not an existing one -- that would be super helpful.
  • Below in the notes, we note a code and xml analaysis has looked at all of these, but just to be sure it would be nice to have eyes on.

• It's also possible that depot events targeting specific factions are no longer working with astro trains.
  • Same with waves launched at specific non-player factions.

• Looking at the above, the following divergences were found:
  • AI Risk Analyzers had a different name in the code and the xml. That would have busted them some, but the one bit that was using the old code-style name seems resolved now.
  • One of the two hunter fleet variants was inconsistent with the other and the xml... who the heck knows what this would have caused in the past. Probably bugs of the sort I am worried about happening now, but at any rate it's now fixed.
  • Wow, same thing happened with one of the two Praetorian Guard code instances. I wonder what problems this was causing.
  • Aaand the Warden Fleet had two different versions in code, and a completely other thing in xml. Yikes.
  • Dark Spire had a difference between code and xml. Same with Devourer Golem. And Dyson Sphere. Also Macrophage.
  • We had an extra SpecialFaction_NanobotZombie faction in code that was an old remnant. That's been cleared out.
  • Relentless AI Waves had a different name in code and xml, too.
  • Everything else was consistent. Not counting mods, this was 42 factions or subfactions, with 9 being inconsitent, and 1 being just an extra outdated thing. So that wa still 32 that were correct and consistent.
  • There were three places where this was definitely going to cause problems in the future with the Risk Analyzers, but those are fixed proactively.

• Looking at code based on the above, the following results were found:
  • Dark Zenith Library xml looks to still work.
  • Same with Zenith Architrave Portal, looks good.
  • Dark Spire were broken by these changes, but the xml should now be fixed.
  • Dyson Sphere was definitely also busted, but the xml should now be fixed.
  • Everything else seems to have already been consistent and no problems.

• Added is_always_a_direct_attack_against_players, which lets us track certain factions which are always basically threat regardless of the rest of their disposition (unless they are actively attacking you).
  • This applies to the relentless waves, the tsunami CPA units, the AI reserves units (those do go away, but while they are present they count), instigators, anti player zombies, anti everyone zombies, and that's it.

• is_considered_hunter, is_considered_warden, and is_considered_praetorian have been added to make it much easier to check which factions are of that type.
  • This is purely a processing efficiency thing, which is relevant the more ships there are in the galaxy. Mostly for backgroudn threads, but still.

• The way that the threat is calculated has been updated a bit to hopefully avoid some edge cases where it seems incorrectly under-counted.
  • This did not change much in the savegames that people gave us.

• The display now shows threat numbers that include even numbers on player planets -- aka, before, threat ships would become attacker ships, and they were either/or.
  • This was very messy and confusing, because attacker ships might just be something random passing through, and might not truly be a subset or even intersection with threat.
  • For the sake of consistency, threat is now threat even if it's on your planets. That way the threat doesnt' seem to diminish just because it's on your planet now. That looked like you were killing something when really you were not.
  • Similarly, the strength and number of ships attacking you is just all about enemy ships actually at your planets, and their status doesn't matter.

• Added a new RelativeToOtherFaction_ThreatStrength and RelativeToOtherFaction_ThreatStrengthVisible, which lets players have an idea of how much of the enemy threat forces are aimed at some other faction than them.
  • These actually give an enormously better picture, in the threat screen in particular, of what is threatening factions other than the player.

• Similarly, the monolithic number of the hunter fleet strength is now broken out by that part which is against humans, and the portion that is against other factions, and breaks that out on the threat window display.
  • This was the final piece of the clarity puzzle. Most of the actual numbers that were wrong were only a bit wrong (missing counting certain things, etc), but not being able to tell that huge parts of the hunter fleet have no interest in you was kind of a big deal.
  • Note taht this could also make the hunter fleet look insurmountably dificult.
    • In one example, there were appearing to be 275 threat against humans, but really there was 308. But at the same time, it said there was 646 hunter strength, which would be unbeatable... except only 69 of it was against humans.
    • In the other example, the total threat reported was 312, but the real threat was actually 347. It said 652 hunter strength, but only 74 was against humans.
  • Both of the examples of "impossible games" are actually more to do with what looked like impossible odds because of hunters doing things that won't bother you unless you specifically go and harass them.

• Thanks to Lord Mekton and Arides for reporting the things that led to this rabbit hole.

More Mode Icons, And Tracking Hunter Targets

• On the individual ships, we now have an "IsAfterANonHumanTeam_NonSim" that allows for us to keep track of which ships are not here for you right now.
  • Based on the name of the variable, you can probably guess, this is not meant to be used with the simulation itself, but only for display.

• Added about 15 new underlays that can be used for ships, complete with selection circle colors to match, etc.
  • Split these out into a new Official_SelectedStatus sprite dictionary, instead of them being part of the "central bits" with health bars, etc.
  • This actually makes these moddable for the first time, without them being incompatible with basic health bars and such.

• Updated our two main compound-icon shaders to allow for passing in this other image, and pulling the data from it.
  • The overall efficiency of this is actually a bit higher, too, now that we have more icon options. We're able to use two smaller 1024 images rather than one larger 2048 one (which is 4x the size, not 2x like the current approach).

• Fixed a bug in our existing calls to the compound-icon shader from half a year or so ago, where it was one of those "how did this even work?" things.

• We are still using the ActionStatus internally, it turns out, but we are also now using that in combination with the recently-added (and now filled-in) IconShipStatus.

• States of matter now define a icon_ship_status_to_draw_when_not_in_alt_mode, which lets us (or modders!) define custom status icons and rings for the various ones.
  • Multi-phase and single-phase and dark matter states all now have their icon states set up!
  • The only one you can really see yet is the multi-phase one, for the spire relics in DLC1, but those work well! Expect to see more of both solo-phase and multi-phase in DLC2.

• When AI hunter ships or similar are after NPC factions, rather than you, they now get a different icon status behind themselves.
  • One thing to note in this sort of situation is that the Threat galaxy map view will NOT show these ships, since they are not threat to you (unless you walk into them), but if you do the strength filters for the hunter you will see them.
    • It's worth knowing where the hunter is, even when it's not aimed at you, because if you happen to fly up to them while they are waiting to hurt some other faction, they will pass the time by hurting you instead.

• For hunter ships that are after another faction other than you, in their tooltips it now says that "NOT AFTER YOU" and then some added text explaining who they ARE after, and some general advice on dealing with them, depending on how detailed your current tooltip is based on what buttons you are holding down.

Mod Updates

• AMU:
  • The SmartFactionImplementationBase and AMU_Utils classes now have the Log() utility function, which is but a shortcut for ArcenDebugging.SingleLineQuickDebug(), but cleans up code a lot.
  • The InitializeExternalData() function now requires a Faction object passed into it. Thus the game no longer needs to even start for it to work, which is important if it's needed during mapgen already.
  • The InitializeSettings_AlwaysDoBase() function is now always executed upon creation of the Faction Implementation, thus the settings are available from the very start of the implementation, which is once again important during mapgen.
• Kaizers Marauders:
  • Used the new Log() function in AMU to clean up code.
  • Adjusted the Journals creation to the new method in Vanilla so it's up to date.

Beta 2.757 Super Mega Exciting Sidebar

(Released March 12th, 2021)

• Holding both W and S will now do nothing rather than moving forward, and same with A and D and not moving sideways.
  • Thanks to Feuermagier for reporting.

Planet Sidebar Improvements

• Completely re-coded the planet view sidebar under the hood so that it is more flexible and more efficient, and so that we can start making some new options on how it will actually show information.

• Step one after that was basically just to get it back to more or less what it was showing before, but with some basic improvements:
  • Ships of the same type from different factions that are in the same category (aka multiple groups of enemies with v-wings) now have their own icons with proper colors.
  • You can now see unit remains, under construction ships, and crippled ships broken out from their counterparts that don't have such a status, in the event that there are such ships. Before they were always lumped together, or just ignored entirely from the sidebar standpoint.
  • Some of the wording for when you're looking at a planet that you can see, but where there are no ships visible at all, has been improved.
  • In the event that there are some ships present in a category, but the strength is less than 0.1 (and thus would cause it to say 0 strength), it now says 0.1 instead. This was needlessly confusing when the strength was literally too small to register.
  • These bits are just the tip of the iceberg on all sorts of new stuff we can now do.

• The way that ships in each category (yours, allied, enemy, etc) show up is now vastly better sorted.
  • Previously it was stuff with a higher strength value first, and then a semi-arbitrary order for things with equal strength.
  • Mow it's command stations first (any faction, then major AI structures next (any faction), then noncombatants next (any faction -- probably these are capturables), then sort by faction within the category.
    • Within each faction, it's sorting by fleet leaders very first, then stationary after that, since these are thing that tend not to die as much and cause others to cascade around. Within those, turrets are last. In the mobile group, it goes elites, then large ships, and lastly strikecraft.
    • Within each tiny group, it sorts by name, and when there are duplicates of the same thing with different icon statuses (crippled, uncaptured, still building, etc), it sorts by those.
  • All in all, this should make things vastly easier to find when you go scanning through a list. It's the least invasive version of this sort of change, and potentially can be refined more, but it keeps things together in such a way that you can more easily find what you are looking for.
  • Thanks to lots of people suggesting this for years, but perhaps Badger was the original pioneer of requesting this.

• The planet sidebar tab now uses a global list of icon buttons, regardless of which category they go in, to make them a lot more efficient, particularly if we have way more than four categories for any reason.

• There are now a group of options that are shown at the bottom of the planet sidebar in a normally-collapsed section. These are personal settings that are saved between games and sessions, so you should not need to alter this often. However, if you are the sort who likes to toggle between a few different views on different planets or just to check something for a moment, these are conveniently placed so that you can do that without having to go into menus and back out.

Subsectioning By Faction

• An alternative mode for sectioning the planet sidebar has been added: by faction.
  • The prior one was by relationship, and just showed you, then all allies in one lump, all enemies in a lump, and then the nobody faction.
  • This new one shows each individual faction that has ships present, sorted first by those categories, on the sidebar.
  • The difference here is incredibly dramatic, and this is really powerful of a feature. For now it is still not the default, because the default relationship one has already been made so much better.

• There is now a short_name="Name Here" xml attribute available to all factions. It has been applied to all factions in the main game and DLCs, but modders should also be sure to apply it to their own factions.
  • The idea is to have a much shorter name that comfortably fits on the sidebar when viewing the sidebar by-faction, so abbreviations and lots of left-out stuff is good.
  • When hovering over the faction abbreviation on the sidebar, it shows the full name, so this really is just all about space (and sorting -- the short name is what is used for sorting the list).

• There is a third and final alternative mode for sectioning the planet sidebar, and that is "by faction and mobility."
  • This breaks out all of the things that are mobile or stationary per faction.
  • Please note that for the concept of mobile, if it's a ship that can't go through wormholes and thus can't leave the planet, it will still be counted as immobile for sorting here (so FFs that can move a bit but can't go through wormholes, engineers that can'go through wormholes, etc). Essentially these are forces that can't go attack enemies.
    • Your support factories, battlestations, and citadels are also included in the immobile camp, as they are able to go through wormholes BUT are primarily angled at setting up camp locally and deploying stuff for a while.
  • When thinking about mobility for yourself, this is mainly a separation of your stuff that would be a major undertaking to get to another planet.
  • When thinking about mobility for enemies, this is mainly a separation of the items that they can't use to attack you on other planets if you aggro them.

Local Fleets On Planet Sidebar Tab

• The fleets sidebar no longer has a top section, with blue colors, noting fleets that are on the current planet.
  • This has always been something that a lot of people find confusing, even though it was organizationally handy. But the confusion of this, mainly with it being on the fleet sidebar, was just too much. We can do this a better way.

• On the planets sidebar, you can now see the fleets that are at the local planet, listed by default under all the other ships.
  • The tooltips and controls on these buttons, and most of the button visuals, are identical to what you see on the fleets tab.
  • The main difference is that instead of showing the planet name of the fleet, it shows either blank space (when all is well) or the shield bar if the flagship of the fleet has taken shield damage, or the health bar if the flagship has taken damage.

• When the flagship of a fleet is crippled, it now shows up with the words CRIPPLED in orange on both the fleets tab sidebar and the planets sidebar.
  • This makes it super obvious when that's actually going on.

• All of the fleet management subscreens can now be open on either the fleets or planets tabs. Since you can now open them on the planet tab, this saves a lot of effort in terms of clicking over to the fleets tab, particularly when upgrading fleets that happen to be on the current planet.

• On the planet sidebar, you can now see the strength of fleets like you can on the fleets tab. However, it's shown by the overall fleet health meter (both of these get replaced by the crippled notice if the flagship is crippled.

• Also on the planet sidebar, if you hold shift you will notice that the ship cap numbers get replace with the colored name of the planet the flagship is on.
  • This is less crucial information, but having it in easy toggle range is nice.

• On both the fleet tab and planet tab, if a fleet is in transport mode, it now has a little cyan (T) behind its name.

Watched Fleets

• Added a new Fleet Watchlist feature in the fleet management details window:
  • Normally on the planet sidebar, you can see a list of all the fleets you have on that planet, including the health of the flagship. There are times where you want to keep an eye on a fleet at all times, even when it's on other planets, though. The fleet watchlist allows you to monitor specific fleets on the planet tab of the sidebar regardless of where you go.
  • If a fleet would show up on the watchlist AND the local planet fleet list, it just shows up on the local planet list, since that's more informative.

• if a watched fleet is on the current planet, and the current planet local fleets are shown and open, then it doesn't double-show the fleet in both. It just includes the fleet on the local planet list (but keeps the count of fleets correct in both categories).

• Thanks to Badger, Strategic Sage, Arides, and others for the requests that led to all these fleet additions on the main planet view, including watched and local fleets.

Cripple Mechanic Updates

• Ships no longer become crippled at 10% health. Instead, they become crippled when they are literally at 1 health. This gives more wiggle room, and is more intuitive in general.

• Major simplification of crippling penalties:
  • Rather than having a bunch of things configurable per type of ship, it's now global for any ship type that can be crippled.
  • Ships simply get a 2x multiplier to repair costs when they are crippled. This just undoes the normal repair discount versus building fresh.
  • The bail-out function is now turned off by default.
  • You can still have a hacking point cost to ships getting crippled if you wish to, but the default is now 0, and whatever you set is now global for all unit types.
  • Thanks to ArnaudB for succinctly summing up why this was a good way to go.

• Additionally, all of the cripplng-related features, and the human balance features in general, can now be edited mid-game.
  • So if you had an ongoing game that you loaded into a recent build that turned on bail-outs and such, you can turn that off if you prefer.
  • Thanks to Democracy for reporting.

Beta 2.756 Choice Of Injury

(Released March 10th, 2021)

• Added a note in the AI hacking response bit where it characterizes that stuff as Slothful or Terrifying or whatever:
  • "Note that the hack itself may have a response that is far more aggressive than the generalized AI response component."
  • Thanks to Leif for noting how the "slothful" hacking response to a superterminal hack can still be apocalyptic.

• The superterminal hack has been toned down in general, so that hopefully it's not quite so devastating. It turns out that simply tuning some of the numbers on the hack itself was probably the way to go.
  • This may still be too hard, or may now be too easy. We'd be very interested in feedback. Likely the response will be somewhere in the ballpark of 1/3 to 1/2 as strong, but it's hard to be sure and we've not tested the results yet.
  • Thanks to Arides for reporting.

• The Render Vengeance Generator Vulnerable hack was insanely too difficult and also pricey.
  • It now takes only 250 seconds rather than 300 (this prevents the tail end that is craziest).
  • The primary response strength has also been dropped from 2.5 to 1.8, and hopefully this does not make things too easy when paired with the above.
  • The secondary response has dropped from 1.5 to 1.1.
  • The cost in hacking points has been dropped from 60 to 35.
  • Thanks to Bionic for suggesting.

• Added a new "Human Balance" section to the Balance tab of galaxy options.
  • These settings change up some of the balance to human players in general, but not any one particular empire or type of human player in multiplayer.

• Fixed one oversight with the new ship states of matter stuff, where their "time on the planet" was not resetting when they change states of matter. Now that works as desired, and their time on the planet is indeed rest.
  • Thanks to Zeus for suggesting.

• Fixed an INCREDIBLY longstanding bug where fatal damage to a ship was not being noticed correctly some of the time when ships took damage from a source. This was denying them chances to do things with specialized "when I kill a unit" effects, and probably throwing off kill count credit, as well. Mostly it was not a huge thing, but it was clearly wrong.
  • Historically, we have had a function that goes in there and finds 0-health units and kills them later if we don't know why it's at zero health. But in the last couple of builds we took that out partly, mainly because with changes in mark level you could get units at 0 health based on them being partially damaged and then losing a mark level, and those should not die from wounds of that sort in our opinion.
  • That said, if this is continually a problem, we can reinstate that; the true problem was with command stations of players, and player flagships, and we can certainly make THEM not have this logic while everything else does. But having this work properly is probably best for all sorts of reasons with ships in general, and things seem to work properly now that we've fixed the core "notice we just killed an enemy" code being corrected. So we'll see how that goes.
  • Thanks to Strategic Sage, Zwei, and arl85 for reporting.

Citadel Overhaul

• Citadel rebalance.
  • Citadels no longer come with a bunch of turrets already attached to them. You only get whatever you have from the TSS.
  • Citadels themselves are of course very powerful still.
  • AIP to claim has been reduced from 60 to 25.
  • Defensive structure cap multiplier from 133% to 140%.
  • Citadels now always come with 8 interplanetary engineers.
  • Note that about half of these features will only apply on new savegames (AIP reduction, defensive structure cap increase).
  • Thanks to Arides for starting the discussion, lots of folks for weighing in, and ArnaudB for most of the design of the solution.

Forcefield Harmonics

• Added a new ability: can_pass_through_enemy_forcefields="true"
  • FORCEFIELD HARMONICS: Can match shield harmonics to pass through enemy forcefields, although any weapons fire it does will still impact on the forcefield itself.
  • Given to the following units:
    • Astro trains of all sorts, and relic trains.
    • Devourer Golem, Ravenous Shadow (ouch).
    • Zenith Trader and Cyborg Ambassador (the latter is mostly thematic).
    • All members of the Imperial Spire (good thing they're on your side).
    • Spire relics (handy for you).
  • Thanks to Zeus, CRCGamer, Bionic, and others for helping come up with this idea and name, mainly in response to the new immobile forcefields (always an arms race!).

Scrap Refunds

• New human balance subsection feature: Scrap Refunds On Friendly Planets
  • What percentage (0-50%, default 15%) of the cost of a scrapped unit do you get back? This only applies to units on your own or allied planets, and damaged or incomplete units return less.
  • Untested, but should work.
  • Thanks to Zeus, Strategic Sage, CRCGamer, and Metrekec for weighing in on this.

Crippling Mechanics

• Once a ship has been crippled, it will now stay crippled until it is back to 100% hull health.
  • This was a change we were going to make in general for some of the refleeting flow work, but it also puts an end to another exploit where you could upgrade golems (artillery for example) by direct science upgrades in order to un-cripple them.
  • Thanks to Strategic Sage, NR SirLimbo, and ArnaudB for the initial discussion on crippling, and Sigma7 for the report on artillery golem exploits.

• Metal expenditure changes for crippled units (some of this may already have been happening, but now it's double-sure):
  • Drones are no longer built inside crippled flagships.
  • Crippled factory battlestations will no longer produce any ships.
  • Crippled engineering battlestations and similar will no longer repair any targets.
  • Flagships in general won't be able to use claim beams, or repair remains, if they are crippled.

• New human balance setting: When Crippled: Bail Out To Nearest Friendly Planet
  • When a mobile ship belonging to a player is crippled, it immediately 'bails out' to the nearest allied planet. If there are no such planets, then it is sent to the nearest neutral planet. If there's still nothing, it stays where it is. Defaults to on.

• New human balance setting: When Crippled: Bail Out To Nearest Home Planet
  • When a mobile ship belonging to a player is crippled, it immediately 'bails out' to the nearest allied home planet. If there are none available, or it is already on the home planet, it will move to the nearest allied planet in general. If there are no such planets, then it is sent to the nearest neutral planet. If there's still nothing, it stays where it is. Defaults to off.

• No matter what, when a ship is crippled and belongs to you, it now gives you a chat message about that.
  • It gives information including the player name and planet name, and goes into the central chat log for later reference if need be.
  • This should be pretty helpful in multiplayer, particularly if someone is not noticing that their stuff gets crippled.
  • Thanks to Badger and his play group for inspiring this change.

• Ships will now fully-reliably do their "on crippled" logic in general, if they have any.
  • Sometimes before if they were crippled by passive effects or attrition or similar, they would not do whatever was unique to that faction for the crippling effect.
  • Note that in existing savegames where your ships were already crippled, this will now catch them and cause them to bail out or charge you HaP or whatever.

• balance_repair_extra_cost_for_seconds_after_crippling has been removed.
  • This was previously set to 120 seconds, and it made sense when the crippled status was something that you could just wait on.
  • Previously, if you could repair a ship back to 10% health and have it un-cripple, and then the other 90% of health would either be during or after this 120 second period, costing you more or not; if you micro'd it, you could get around it, which is never good.
  • Now the crippled status doesn't go away until a ship is fully repaired, so we have no need for a timer like this (at least not related to crippling).
  • Ships will simply cost more to repair while crippled, and not cost more to repair when not crippled, and that's much simpler.

• balance_multiplier_for_ships_that_have_extra_repair_cost_temporarily has been removed.
  • This was previously set to 3.5x, globally, which was really hard for golems and no bother at all for transports.

• Added a new section to balance in the galaxy options screen: Crippling Costs.
  • These settings change up how much it costs when one of your large units is crippled.

• Added Cost Multiplier Of Repairing Crippled Transports.
  • After they are crippled, and until they are repaired to full health, their repairs will cost this much more than normal. Default 4.5x for transports. Range of 1x (normal) to 6x, with lots of intermediate options.
  • Also identical options for a bunch of other categories:
    • Default 1.25x for officers (golems and arks).
    • Default 2.5x for battlestations (basic only; not citadels).
    • Default 1.75x for citadels.
    • Default 2x for support (usually combat factories).
    • Default 1.35x for lone wolves.
  • Normal 1x cost for anythinig that isn't one of those categories somehow.
  • Thanks to Strategic Sage, NR SirLimbo, and ArnaudB for suggesting this in general.

• When ships bail out due to crippling, they are also paralyzed for the max time allowed when they reach the destination, but they don't have any other new bad status effects on themselves.

• New group of options in the Crippling Costs section of the galaxy options: Lost Hacking Points When Transport Crippled
  • Each time they are initially crippled, how many (if any) hacking points do you lose? The default for transports is 1. The range for any of these is 0 to 20.
  • And the other similar settings:
    • The default for officers (golems and arks) is 4.
    • The default for battlestations (basic only; not citadels) is 2.
    • The default for citadels is 3.
    • The default for support (usually combat factories) is 2.
    • The default for lone wolves is 5.
  • These also pop up as part of the message visible to all players, but only the player who actually had a unit crippled takes any hacking point loss in multiplayer,
  • Thanks to Strategic Sage, NR SirLimbo, and ArnaudB for the discussion that led to this end result as well.

Technical Bits

• The WarpToPlanet code function from AMU has also been moved to the main game.
  • We made a slight change in that ALL guarding offsets are now cleared, not just those for melee, since the planet is new and thus a guarding offset is meaningless in the new place.
  • Thanks to NR SirLimbo for coding and contributing it!

• Added a new GetIsFriendlyTowardsSafe() on units, which allows us to check friendliness easily with any other faction, not just the local one.

• Added WarpToPlanetAtSafePointNearCommandStationIfPossible, which tries to put you near the command station, or another fleet leader if that's not possible, or just a friendly unit if none of those are possible.

• Added FindNearestAlliedPlanet_FairlyExpensive() on planets, which lets us find the nearest allied planet (by hops) from a given faction with a specific starting planet.
  • Also added FindNearestAlliedOrNeutralPlanet_FairlyExpensive, which does the same thing bug for allied, neutral, or owned-by-nobody planets.
  • And finally also added FindNearestAlliedKingPlanet_FairlyExpensive, which is the same as the first but also requires that there must be a king unit (generally a home command station) on said planet. This could be your home command, or an allied on in MP.

• GetFauxFIntValueFromSettingByName_DefaultOne_DuringGame now uses an internal dictionary to cache string results that it has already parsed in some fashion, which saves on processing when this is called frequently.

• Updated the AMU mod to not freak out with the new changes.

• Units that have not yet been fully claimed are unable to be crippled (they also don't die, but they don't get the cripple status, cost extra, or bail out).

• When ships are disabled for some reason, but can't normally be transported, it no longer shows a tooltip message saying that they can't be transported but normaly can. That was confusing!

• When a ship is disabled for multiple reasons, and one of them is being crippled, it now shows both reasons. It was possible to not be able to see the crippled info if it was also paralyzed or paused or similar.

• The width of the name column on the galaxy options screens are now wider, so that it's less likely to word-wrap. This is much easier to read.
  • The text has also been sized down slightly in order to make sure it fits more consistently, too.

Beta 2.755 Hotfixes

(Released March 10th, 2021)

• In the lobby interface, updated the Scourge impact rating from Hard to Brutal due to the rebalance.
  • Bear in mind that you can absolutely still play with these in a "hard" fashion, but this does reflect the fact that if you try them above their starting impact of 5, you can expect an outsized response.
  • Another way to think about this: the impact scale of 1-10 is a bit different from something that is inherently Hard versus inherently Brutal. You can still get a wide variety of difficulties, but at each step it's going to be more intense.
  • Thanks to ArnaudB for changing.

• Fixed some compiler errors in the last part of the code that was supposed to make the passthrough key (Z) work properly for the revised stationary flagship mode.
  • THEN actually finished actually implementing that code, and made sure it worked.
  • Apologies for this. I was extremely tired, apparently, and completely spaced out on even finishing the feature I was in the middle of right before release. Thank goodness for beta branch.

• Fixed a bug in the last release where ships that were in transports from a save from a prior build could not be unloaded in the new version. Essentially, the state of matter for the ships was being blanked out incorrectly.
  • Thanks to Kahuna, ArnaudB, Gdrk, Ithuriel, and Leif for reporting.

• Added some extra sanity-checking around states of matter, so that we're less likely to have any errors related to them being un-set or mis-set in the future.

• Fixed a couple of bugs in tutorial 4 on the beta branch that were keeping it from being completed.
  • Essentially, since we had changed up the type of bombers in the starting fleet, it no longer matched the tutorial and thus it could not be completed.
  • Fixed a typo saying Piercer tech rather than Piercing.
  • Added a new tutorial-specific version of the ARS that has more info, its own hack, and can only be hacked once rather than three times.
  • Thanks to GregC for reporting, down to the line numbers we needed to adjust in the xml!

• Fixed an issue that has been around since we introduced tech refunds (apparently). If your marking-down of a ship that is damaged would cause it to die, that now will just cause it to be at 1 HP remaining, instead.
  • This was actually bypassing ships that normally are only cripple-only and causing them to disappear, leading to entire fleets disappearing. It could also cause your home command station to disappear without a trace without you losing the game.
  • All of these are directly fixed.
  • Thanks to Sigma7 for the savegame that let us finally figure this one out! Thanks also to Ryulong for letting us know this could happen to kings.

• Also fixed things up in general so that any ship that is crippled won't ever be deleted even if it's health is 0 or lower for some reason. It will just go to 1 health.
  • This doesn't affect permadeath options, which still let them die.

• For player king units and command stations in general, if those are for some reason at or below 0 health at a random time, they just move to 1 health now.
  • These still are very much killable, but they die when they are actually shot or damaged in some way, or when a scrap command is issued. Not ever a "well, I see I have zero health for some reason, so I'll just be dying now."
  • This sort of thing could have actually happened from things like us or modders adjusting balance numbers in the xml to have lower health, and then damaged command stations or kings just kill themselves on save load. That had not happened yet, but we were aware it was the sort of thing that could.

Beta 2.754 Buffs And Bugfixes

(Released March 9th, 2021)

Immobile Forcefield Options

• Outguard forcefields have been updated to now be completely immobile, and thus immune to the norris effect.
  • Same for the human home command station forcefields, and the forcefields that you can buy from the zenith trader.
  • Regular forcefields now note their vulnerability to being moved by ships with the norris effect.
  • Added a new "Immobile Forcefield Generator" that military stations now have instead of the normal kind that logistical and economic have. This one can't be norrised, but is otherwise identical.
  • The Turtle Defenses battlestation fleet (from DLC1) now uses Immobile Forcefield Generators, which both protects against norris, and also prevents accidental movement of them when selecting them for use.
  • Thanks to stanazolol69 for the idea of the immobile forcefield generator in general for norris purposes, and to Bionic for the idea on them from a practical management standpoint.

Useful Stationary Flagships

• So! Chris completely misremembered how stationary flagship mode works, and this led to it being described as a way to set up a rally point.
  • Previously, this did not work because any orders that were given to a flagship in stationary flagship mode were refused, but any orders that it already had it would execute. This is backwards! It should accept all new orders, but sit there with them and hand them to any ships emerging from itself.
    • Given how much more desirable that is, that is how this mechanic now works.
  • Additionally, any move orders on the galaxy map work the same as any other orders. If you don't hold Z, your flagship will still just sit right where it is in stationary flagship mode. This makes it actually possible to set up cross-planet rally points using the galaxy map screen, whereas otherwise it was not possible at all.
  • And finally, the various tooltips have been updated to reflect this new functionality.
  • Thanks to alocritani for reporting the discrepancy.

Buffs To Underdogs

• A balance pass to help a few units unfairly treated by the many, many changes over the course of the beta.
  • Thanks to ArnaudB for doing all of these!

• Hive golem AIP gain reduced from 30 to 20.
  • With heightened metal cost, little benefit from its durability cost and eyes proving more troublesome, this golem no longer justify its horribly pricy cost.

• Spider range increased from 8k to 10k.
  • It should now be able to usefully applies it slow effect.
• Gunbot and porcupine health increased by 50% and metal cost reduced by 33%.
  • Too expensive in the beta where metal is precious, with so low health they were a metal sinkhole in the player's hand and poor jokes within AI's waves.
• Sniper metal cost reduced by 40%. It was too pricy for how unreliable it is for the player. Hopefully this should make it worthwhile to use.

• Tutelar tank FRS hull bonus reduced from 2.5 to 2.0. With all strikecraft health going into hull, it was way too strong. Sorry.
• Ireful MLRS Corvette shield bonus increased from 1.25 to 1.50 and gain an additional 1.50 bonus to hull.
• Encircling Spider FRS shield bonus increased from 1.50 to 2.0.
• All AIP costs for FRS ships is now 12, except Turbo Raider.
• Turbo Raider AIP cost reduced from 20 to 14. Speed is king, but for a bonus that doesn't applies to Flagships it was too pricy.

• Beam Turret now has 1/3 of its damage and 3 times more beams.
• Solar Flare Turret now has 1/3 of its damage and 3 times more beams. Its paralyze effect lasts 2 seconds instead of 3.
  • They were underperforming horribly in many situations, particularly with low stacks setting, including default ones, or against low health enemies.

Bugfixes

• Updated the autosave code so that even if you disable it, in ironman mode it will still work. And if you disable it in ironman mode, it should definitely actually be disabled.
  • Thanks to Sigma7 for reporting.

• Fixed kind of a cascade of issues that was preventing Interplanetary Engineers from being constructed, mainly if they were on battlestations or command stations.
  • These issues were mostly introduced on March 4th, when the other changes relating to repair delay timers, etc, were introduced.
  • That said, the way these worked was kind of accidental before, and it didn't show up in the interface with progress on constructing them like it would for other ship lines, etc. All of that now works much more properly, and you get proper interface feedback for the first time ever with these.
  • Thanks to ArnaudB for reporting.

• Fixed a variety of potential cross-threading bugs in EntityOrderCollection, including one that had popped up in particular in ClearOrders. The one in ClearOrders should not only be fixed, but if something else happens where it does recur, it should no longer corrupt savegames, and it should also give us more information about where exactly the error was.
  • Thanks to Sigma7 for reporting.

• Fixed a bug in the fallen spire where it would give you an exception after you beat the game with the imperial spire around, because they had no one left to go kill.
  • Thanks to Zweihander2021 for reporting.

• Fixed the remnants of an exception that could happen when there was a missing praetorian guard faction in loading the lobby settings. This is something that should be remedied on actually generating the new map in general, but there was some overzealous error checking that was getting in the way of that process.
  • This was something that happened in particular when editing older quick starts after loading them into the lobby.
  • Overall this involved some extra plumbing to make sure that older quick starts and saves self-repair new information in. At some point it may be required to explicitly link up some of the AI sub-factions to their parent AI in multi-AI games, which is where these problems seem to crop up. But for now this works, and so minimal changes seems like the wiser way to go.
  • Thanks to alocritani for reporting.

• The tooltips of strength-based AI eyes no longer talk about unit counts (super confusing!), and instead properly talk about the relative strengths of opposing forces.
  • Thanks to Strategic Sage for reporting.

• After poring through the code for a while, it was really not clear why the Raid Engine was so very wimpy in terms of the response it was sending. It seems like it is using the general pre-wave-buff amount of forces, multiplied by 1.5x. But the code would not really seem to support that being the case, it's very strange.
  • At any rate, the waves of the raid engine are now multiplied by the wave_budget_multiplier of the AI difficulty that is in question. So on difficulty 7, as in the example save that we had, this is an increase of 6x strength for the wave, leading them to be these 99-strength monstrosities popping out.
  • The hope is that they were universally wimpy, and so nobody gets crushed now by some giant 1000-strength wave instead of a few hundred in some very intense game. We haven't seen any evidence of that yet, but that sort of thing is always the worry when we don't understand what is going on and put in a new multiplier to compensate anyhow. At any rate, we're sure we'll hear about it; but the raid engine is a good thing to kill post-haste in any game now, for sure.
  • Thanks to Strategic Sage for reporting.

• Fixed an exception that could happen when doing the neutral planet science extraction hack (or any other non-instant planet-based hacks) against dyson sphere planets.
  • All of the other factions are checked and are okay.
  • Thanks to Metrekec for the report and the save for duplicating it.

• Fixed an oversight in the previous version where ships that were being transported would not properly keep their state of matter when going into and out of transports, or being saved.
  • This didn't matter in the prior build, but will matter in the near future for DLC2 and 3.

• In the event that a transported ship is removed after loading a savegame because it is deprecated, it now logs an entry in the game log so that this is not some mystery.

• Fixed up the vicious raid frigate so that it inherits its properties from the regular raid frigate and NOT from the deprecated AI Raid Frigate (AIs don't get frigates anymore). This wrong inheritance was causing it to have all of its ships mysteriously disappear whenever a savegame is loaded with them in it (existing saves, when loaded, will now load them properly and keep them).
  • We also checked our other frigates to make sure that there were no similar issues.
  • Thanks to Timerlane for reporting.

• Additionally, the ship cap for lines of vicious raid frigates has been increased from 6 to 9; it was an oversight that it was so low.
  • Thanks to Timerlane for reporting, and ArnaudB for confirming intent.

Beta 2.753 Interplanetary States of Matter

(Released March 8th, 2021)

• Fixed the Devourer Golem being allowed to attack literally everything. Now it only targets mobile ships again.
  • Thanks to NR SirLimbo for fixing.

• Battlestations now cost 250k to build (mainly, to repair), rather than 6k. This was a balance oversight. Their claim cost has always been 200k.
  • Thanks to ArnaudB for suggesting.

• Citadels now cost 400k metal to claim, rather than 200k. They already had construction (repair) costs of 1.5m, so this claim cost was just a copy of the cost on battlestations. Again a minor balance oversight.

• Added a new IconShipStatusTable, which replaces the old hardcoded enum ActionStatus.
  • This allows for us to, in general, let ships have new background indicator icons that denote their status.
  • Put slightly better, this actually allows for modders to do that.
  • This may break the AMU mod, so for now we're just putting in the table, and we'll finish the feature tomorrow after the other bits are out.
  • For now, any ships that are in an alternate state of matter just use the guard ActionStatus, which is fine for today.

• Put in extra debugging for when xml row nodes fail to parse for various reasons.

• Added a new is_disallowed_from_spawning_in_waves="true" flag, which can be applied to units that otherwise are vastly too strong when included in a wave from the AI.
  • This works on guardians, dire guardians, regular strikecraft, whatever.
  • This is now applied to temperamental/tantruming and fragmenting guardians, because their mechanics are really only designed around being on an AI planet, and they quickly become nightmares on player planets.
  • Thanks to Metrekec and Strategic Sage for reporting.

Interplanetary Engineers

• Reduced the time needed for players to build on AI planets across the board.
  • Expenditure at all marks below 7 is now at 70%, making things takes 142% longer to build than on allied planets.
  • Expenditure at marks 7 is now 60%, making things takes 170% longer to build.
  • The old numbers didn't take into account the removal of seeded battlestations and AIP increase of citadels to sixty, both of which leading to a dearth of Interplanetary Engineers to build beachheads at insane speed.
  • Marks no longer changes with every mark of the AI's planet, as that was both confusing and doubly punishing. On higher marks where you needed more beachhead it was slower, and it vice-versa. It's now much more intuitive and straightforward, with Mark VII having an extra difficult to make it "special".
  • Thanks to ArnaudB for implementing.

• Added hacks for Engineers and Interplanetary Engineers in the ODSS. You can hack Engineers for Command Stations or Interplanetary for Battlestations (and Citadels).
  • Thanks to ArnaudB for implementing.

• Added 4 Interplanetary Engineers to starting battlestations.
  • This is mostly so players can intuitively deduce the new engineers and interplanetary engineers hacks will give engineers to Command Stations only and Interplanetary to Battlestations only.
  • Thanks to ArnaudB for implementing.

• Made Interplanetary engineers self-build like drones. They keep getting killed and need to be manually rebuilt when used offensively, which is plain annoying and encourage tedious micromanagement.
  • This also means they don't show up in the build menu (good!)
  • They do show up in the lobby when you pick a station, and in the fleet menu, which is what matter.
  • Thanks to ArnaudB for implementing.

Mod Updates

• AMU:
  • Fixed a very stupid bug in the AddListSeparated() function that was breaking out Kaizers Marauders.
    • Thanks to Histidine on Discord for finding.
  • The getPlanetNameFormated() function now now also requires a bool whether or not the planetary mark level should be shown if the planet is an AI-owned planet. It was extending the planet names needlessly and especially for notifications got too long.
    • Thanks to a bunch of people for suggesting over time, most recently Histidine.
• Kaizers Marauders:
  • Adjusted all logic where planet names are printed out or displayed to the changes in AMU.

• AMU:
  • Fixed a bug where the PlacementCondition.GetLocation() in edge cases (that got really common with highly built-up Kaizers Marauders) allowed to spawn things literally on top of each other.
    • Thanks to Histidine for providing a save that lead to discovering this.
• Kaizers Marauders:
  • Fixed an error when Marauders had only very, very little space to spawn more Fortified Forceshield Generators or Defense Posts, very often when seeding outposts near the player, and they already had a ton of stuff close.
  • This then turned into a bad performance leak because it had to retry so many times all the time to spawn these entities.
    • Thus the distance that things are spawned at is now calculated and added to the next spawning attempt. The result is dynamically extending things out for as far as needed.
    • Note that this spawn-bonus-distance is not serialized. This means that, per planet, if conditions are this bad, the first spawn of something difficult to place may take quite a while - but every further spawn should take a lot less time.
    • This may also change the placement between saving and reloading, but only in edge cases where an Outpost survives but its huge amount of defenses has been crushed. It will still try to seed further away, at least for a while. An autocorrection to this has been implemented but not tested.
      • Thanks to Histidine again for the save that lead down this rabbit hole. This was something I wanted to find a "good" solution for, but had put it off because it was a rather complex issue.

DLC2 & 3 Features

• Added a new StateOfMatterTypeData, which is something that will come into play in a MAJOR way in DLC3, but also will have a substantial presence in DLC2, and even a bit of retroactive presence in DLC1.
  • Because all of these things can cross DLC, this is something that we're building into the base game.

• The first state of matter is just called "Normal", and it is set up so that ships that are in this state of matter (that's most of them) won't show a tooltip with the explanation of it. But here's an explanation anyway:
  • Objects in realspace, existing in three spatial dimensions and traveling along the axis of the fourth dimension (time). Aka, most of the observable universe.

• The second state of matter is not something you'll see until DLC3, but we may as well define it here: 5th Dimension
  • Units in this state of existence cannot see or interact with those in realspace, and vice-versa. Very few units can pass between the spatial planes of the 3rd and 5th dimensions. The relationship of this higher plane to our normal one is extremely hard to comprehend, and we are just starting to explore it.

• The third state of matter is also mostly a DLC3 thing, although it will retroactively play a bit into DLC2 and the base game when we get there: Dark Energy.
  • Units made up of dark energy still exist within realspace, but their nature is very poorly understood. We can detect one another, but crossing over is hard. Dark Energy makes up 68% of the observable universe, while dark matter makes up 27%, while normal matter is a mere 5%. Most space travel is made possible by dark matter reactors.

• The fourth state of matter is mainly a DLC2 thing: Solo-Phase
  • Objects temporarily beyond all reach, both hostile and friendly. They can be detected by our instruments, and even given orders, but for the most part all they can do is wait and heal. They could be deployed for a surprise attack on phasing back in to normal reality, as well, however.
  • Thanks to Zeus for suggesting.

• The fifth state of matter is another DLC2 thing: Multi-Phase
  • Objects in an alternate state of matter that can only interact with other multi-phase objects. Units in this state may be in such a state temporarily -- even unwillingly -- or may intentionally and permanently be set to guard this side-band of realspace.
  • Thanks to CRCGamer for suggesting.

• On all ships and shots, there is now a CurrentStateOfMatter, which defaults to Normal (realspace), but can be changed to other things.

• On ships, there is now a GameSecondWillExitStateOfMatter, which is normally 0. This can be set to future gameseconds to cause an entity to revert to normalspace, or combined with various other features yet to be described.

• If a shot is fired, it will automatically exist in whatever the state of matter its parent firing entity is.
  • Right now this won't apply for drone guns.

• If a shot's state of matter does not match its target's state of matter, the shot now ceases to have a target, which is the same as if the target has moved to a different planet.

• Added a new forced_to_always_be_this_state_of_matter field on ships. This can be defined to make a ship always a 5th-dimensional entity, as one example, or always dark matter or even multi-phase or whatever.
  • This does not normally need to be set, and anything that you want to be able to shift into something like solo-phase or multi-phase should definitely not have this set, since this would prevent them from phasing or being phased.

• On the world object, there is now a LocalPlayerViewingStateOfMatter, which defaults to normal. It's not until DLC3 that we would start swapping this out to let you view the 5th dimension in certain circumstances.
  • However, this does have an affect on visibility of certain things even in DLC2 and even DLC1 now, so we should go ahead and keep it consistent and nicely planned-ahead.

• Updated shots and units so that if they are supposed to be invisible to you (mainly 5th dimension right now), they should actually be invisible unless you are also looking at the 5th dimension or whatever, in which case it only shows you stuff from that other state of matter.

• Bubble forcefields now only protect or displace ships that are in the same state of matter, whatever that might be.
  • Gravity effects continue to cross all states of matter.
  • Tractor beams also can now only grab units that are in the same state of matter.
  • Also, tachyon beams can only hit units that are in the same state of matter.
  • AOE damage can now only hit units that are in the same state of matter, and any AOE damage that tries to fire off in the solo-phase state now just completely fizzles. It might be visible, but it's harmless in there.
  • Any shots that are impacting will not hit enemies that are in a different state of matter. This is partially relevant because the target may shift states of matter right before being hit by a shot that WAS in the same state of matter when it was fired.
  • Units in different states of matter still decollide with one another, except those that have different view modes (like the 5th dimension).
  • The "can a unit be placed here" logic also has been updated to account for this, and again mostly deals with those with separate views like the 5th dimension.
  • Targeting logic has also been updated so that ships can only target other ships in the same state of matter... and for those in the solo-phase, they can't target at all.

• If the current state of matter of a unit is not normal, then it shows what the state of matter is in the tooltip (assuming you can see the unit to hover over it), and the medium and full tooltips give increasingly lengthy explanations for it.

• The immune_to_all_damage_and_invisible_if_first_half_of_the_current_second feature that was on spire relics (DLC1) has been removed).
  • Now there are the following three: alternative_state_of_matter, phases_to_other_alternative_state_of_matter_after_seconds, and phases_back_to_default_state_of_matter_after_seconds.
  • These allow for the same sort of behavior, but also a LOT more flexibility both with the same phasing in and out (different intervals in one and in normal), and a choice of states.

• The spire relic (DLC1) now phases to the Multi-Phase state every 30 seconds. This is similar to what it did before, except before it would have really been in the Solo-Phase state (not that that was defined before).
  • So in some very rare edge cases, having DLC1 on and a relic chase in progress, you might run into some other multi-phase enemies that are a danger to it now. Not super likely, but it certainly is an interesting edge case.

• Ships in the dark energy, solo-phase, and multi-phase states are no longer visible to the naked eye, but you can see the icons for them.
  • Rather like how the spire relic was when it was phased out before.

Beta 2.752 Golden Devourer

(Released March 5th, 2021)

• A new ArcenArrays class has been added, and the Randomize methods have been moved from Engine_Universal over to there (with an obsolete tag so that modders have to do it when they recompile their mod, but things work prior to then).
  • A new ExpandAndReplaceArrayIfTooSmall has been added to ArcenArrays.

• Fixed a potential bug in GetCanUnlockTech() where MAYBE under some certain circumstances it could get an exception if you had already fully unlocked a tech that was being checked.
  • Probably our existing code was fine, but after Histidine got an error in here, we're being extra careful.

• The TechUnlocks and FreeTechUnlocks arrays are now expanded during calls to GetCanUnlockTech() if they need to be.
  • This is a bit of a confusing issue, as it seems like the only way that this should be able to happen is if mods or DLC or something are reloaded after a game has started, but there may be something there with techs that are added after a game was started and we just missed that.
  • At any rate, this is more likely where the bug was, so this will self-fix itself now.
  • Thanks to Histidine for reporting.

• One of the tips was still talking about GCAs, and has been updated to talk about the TSS and ODSS.
  • Thanks to Democracy not only for noticing it, but also for rewriting the whole thing for us!

• Fixed an issue in general with self-attrition damage being considered as "player is scrapping me" damage, which meant that a few on-death effects (most notably spawning units on death) would not happen if a unit self-attritioned to death.
  • Thanks to Zeus for reporting.

• When there is a range of DPS because of multiple beam weapons or similar, the second number in the range now has proper formatting.

AMU mod:

• Readjusted the mod to be up to date and work with the currentmost AIW2 Version.
  • The cost multiplier for certain entities (such as crippled flagship) if repaired within the set timeframe can now be adjusted.
  • Dire Guard Post AIP can now be modified.
• Source Code updated.

Fireteam Aggression

• The Praetorian Guard now have the same style of overconfidence ratios that the Warden was given a bit ago, making them more aggressive and more able to come defend their homeworlds if they really want to but you have cut them off. This in general just makes them more active, without going completely crazy. Their budgets and income have not been changed at all (unlike when the Wardens got an income buff a while ago because of their increase in suicidal tendencies). The PG are more of a goalie, so the risks that they now take are likely to be much more productive than some of those the Warden previously was doing.
  • Thanks to Democracy for reporting the ability to trap the PG way too easily away from the homeworlds of the AI.

• ALL of the various factions that use fireteams, in the base game, DLC1, and DLC2, now have greatly-increased thresholds for planets that they will pass through in order to join up with their groups.
  • This was a change that we made recently to the warden (allowing them to go through 80 strength enemy planets to join up), but now it's also applied to all the other factions in amounts that are usually more like 20-40 strength, but for some stronger factions (or reckless ones) is up to 80.
  • Previously, all of them were set to 1 strength, which was something that prevented a lot of movement of a lot of them. Given how strong player fleets often are now, players can still block these factions quite often, but it's no longer so black and white.
  • Thanks to Democracy and others for some observations related to the Praetorian Guard in particular.

• Previously, the hunter would not path through a path that was cumulatively 1/10th its own strength if it was running to go lurk.
  • Bear in mind that is the difficulty of the ENTIRE path, so at any given point the forces the hunter faced might be quite weak. This meant that it was hard for them to go anywhere there was even token resistance added up.
  • Now the hunter can head along paths that are up to 5x its current strength to get to a desired lurk spot.
    • We did something similar with the warden a few weeks ago, but made it a 30:1 ratio they would accept, rather than a 5:1 ratio like the hunter.
  • Most other factions that want to lurk will also now use the 5:1 ratio rather than the 1/10th value they all had.
  • This should all make for much more active factions.
  • This doesn't completely solve the PG being trapped from their own homeworld by forces in the way, but then again we don't want them to just throw their lives away. They are now much more aggressive about taking on the opportunity when you shift more to their homeworld, and they come crushing in to kill you versus just letting you pillage their boss.
  • Thanks to Democracy for reporting.

DLC2 Features

• New ability: Hardened
  • HARDENED: Any damage from a single source (explosion, shot, attrition, etc) will have its value reduced to X% of this unit's max hull health if it would be greater than that. Protects against very large guns, ion cannons, mass drivers, etc.
  • Added to a unit by giving it something like damage_taken_cannot_be_more_than_this_percentage_of_max_health="0.1" for a 10% hardening (this is very strong -- larger numbers are less of a reduction).
    • One way to look at this is that it means that it will never take fewer than 10 shots to kill something that is hardened at 10%. At 50%, it could in theory take two shots to kill a thing.
  • It is worth noting that there are various types of effects that sometimes hit more than once in a short span, and those would be capped to the hardened level per time they hit, meaning that their overall effectiveness against hardened units is going to be higher.
  • This is untested so far.
  • Thanks to Zeus for coming up with this idea and requesting it.

• Added a new aoe_and_beam_damage_multiplier_to_non_primary_target, which should have values set to something above 0. Something like 30 would do 30% of damage to all of the secondary targets of an AOE blast or beam weapon, rather than the full 100% that it does now. Fun fact! You can actually set the value to something like 400 to do 4x as much damage to secondary targets as you would to the primary target, if you wanted to for some reason.
  • This DOES work in conjunction with aoe_spreads_damage_among_available_targets, although the results are a bit funky. In those cases, the damage is first spread evenly, and then reduced for the non-primary target.
    • Please note that aoe_spreads_damage_among_available_targets only works for area of effect explosions, and not with beam weapons, hence the naming difference.
  • For beam weapons, this is only relevant for those that have hits_all_intersecting_targets set to true.
  • This also combines with coilbeam damage reductions for non-primary targets, as a direct multiplier against it.
  • This also works with chain lightning.
  • Note that when multiple beams are fired, as with beam-arrays, only the target that was directly aimed-at will actually be hit for the full damage if this setting is used for a ship.
  • This is a ridiculously versatile new feature, as it allows for a variety of new weapon types.
  • This feature is untested.
  • Thanks to CRCGamer for suggesting.

• Added a handy new FillRingOfPointsAtDistance to the Mat class, which lets us get a variable number of points in a radial arrangement given a distance from a center point.

• Added a new distance_from_center_for_beam_emission for beam weapons, as an optional feature.
  • This basically makes multiple beams turn into something that are emitted in a radial offset around the center of the ship, rather than coming out of the center of the ship.
  • This is best for things that are large, like golems, and a good amount might be something that is 80% of the radius of the ship itself.
  • By itself, this will lead to a bunch of beams that are coming out from a ring and focusing on a center point (like a death star), but if you combine this with degrees_offset_per_beam... it's still going to do that, because the alternative was almost impossible to code. You get to choose if it strikes all targets or just one target, though.
    • Note that this also does not work with the new rotating beams effect.
  • Note: when combined with chain lightning, if this is set to some number, AND the number_beams_to_fire is set to more than 1, then it will visually do the death-star effect to the first ship that then has the chain lightning after that.
    • No extra damage comes from this with chain lightning, and it's just a visual effect for how it strikes the first ship. There's no practical purpose to this, it's just for visual flair. Normally the number of beams is ignored for chain lightning.
  • Note: when used for a point-laser (aka not something that strikes multiple targets), this completely ignores any fanning-out offsets, and just focuses the lasers on the singular target. This is another one where it's purely a visual effect with no gameplay purpose.
  • Final note: this effect isn't reflected in the tooltip in any fashion. From a practical point of view, it's very obvious when it is present, and the behavior of whatever the tooltip already says is not all that much different (aside from not being a fan if you set it up to be a fan AND this sort of radial emission). If we wind up needing to make some small addendum for any of the variants, please let us know.
  • This is fully tested in all the forms except chain lightning.

• New gold and bright red "radial beam" effects added for the DLC2 golems, which we show off in this video: https://youtu.be/IckDa6s_7j0
  • Thanks to Zeus and CRCGamer and Badger for the suggestions that led to these new looks for them.

Devourer Upgrade

• The Devourer Golem's weapons range has been doubled, making it far more dangerous.
  • It also has been given a weapons adjustment so that it now uses one of the new DLC2-style Zenith ring beam weapons (the gold kind that is most dangerous).
  • This weapon is also a lot more dangerous than the old one it had, so the devourer is a lot more threatening in general.

• In general, updated the beam weapons so that they can visually be offset from their center firing point, which was previously not possible, turns out.

• Also made it so that as soon as a guard post is damaged at all, it immediately dumps its units.
  • Also also made it so that when guard posts are emptied, their ships try to come out unpacked around them in a decollided fashion, rather than all bunched up in one place. When a guard post is being bombarded by high energy beam weapons, this keeps all the contents from being evaporated in one hit.

Beta 2.751 Repairs And Clarity

(Released March 4th, 2021)

Please note: because of the changes to how repair works, the AMU mod and Kaizer's Raiders mods are both temporarily broken. If you have them enabled, expect exceptions during mapgen, and during savegame load, etc.

• Added a 1.1x damage modifier to the Maugrim for targets of 5 tX mass or more. Makes it more likely to use that 12 second reload Heavy Beam Cannon on something worth shooting rather than just a stray strike craft stack.
  • Thanks to CRCGamer for making this adjustment.

• In the event that you try to load one or more transports that are crippled, the message: "X of your transports are currently crippled and cannot be loaded. Repair them first!" will pop up for you.
  • Thanks to Badger for reporting the confusion that was otherwise happening for new players.

• The hotkey for "Hold To Give Orders To Stationary Flagships" is now Z instead of Ctrl. This means that when flagships are in stationary mode, you can set them up on a planet and just ctrl+click a wormhole like usual to give them movement orders through the wormhole (that they will not follow, but which work as a rally point.
  • Thanks to Chthonic_One, Badger, and Zeus for pointing out how hard it was to make these work as rally mode items from a neighboring planet because of this.

• The stationary flagship button tooltip now includes the text "This is a great way to set up a flagship on one planet with a rally point for its units to a neighboring planet, among other uses." in bold at the end of the first paragraph.
  • Thanks to Badger and Zeus for suggesting this.

Guardian Balance

• Guardians now all once again start at mark 1, which means they won't be absolutely insane when coming in waves against you.
  • Thanks to a variety of players for making us aware of the danger of this in waves, and to CRCGamer for counsel on it.

• For the same reason, royal guardians (just the Royal AI type) are now also mark 1 instead of mark 3 for the start.
  • Also he got the More System Defenders mod guardians taken care of.
  • Thanks to CRCGamer for the reminder.

• When guardians are seeded at a planet initially at game start, or when they are spawned as part of a reinforcement event, the game now tries to seed them at 1 mark higher than whatever the planet would normally have.
  • So if this is a mark 2 planet, its guardians are mark 3. Mark 6 planets should have mark7 guardians. Obviously mark7 still has mark7 guardians.
  • This gives guardians that bit of a boost around the galaxy, and is actually more effective than just a blanket "min mark level of 3" like we had in the past.
  • However, this won't affect waves, so it keeps waves and hacking responses and that sort of thing from being hilariously overpowered.
  • Thanks to Sol and CRCGamer for leading us more in this direction.

Repair Delay Sanity

• "RepairDelaySeconds" has been removed, and now there are two new fields: RepairImpossibleForSeconds and RepairExtraCostForSeconds.
  • When reading from older savegames, if the delay was less than 5 seconds it will go to the impossible field, and if it was more than 5 seconds it will go to the extra cost field.
  • ForceRepairDelay has been renamed to ForceRepairImpossibleFor_Max15, to be consistent with the new nomenclature. Also, the most it will make repair impossible for is 15 seconds, as noted in the name.
  • CustomRepairDelaySecondsAfterCrippling_Max511 has been renamed to CustomRepairExtraCostForSecondsAfterCrippling_Max511.
    • This was custom_repair_delay_seconds_after_crippling_max_511, and is now custom_repair_extra_cost_for_seconds_after_crippling_max_511.
  • CustomRepairDelaySecondsAfterDamagedByEnemy_Max511 has been renamed to CustomRepairImpossibleForSecondsAfterDamagedByEnemy_Max15.
    • This was custom_repair_delay_seconds_after_damaged_by_enemy_max_511, and is now custom_repair_impossible_for_seconds_after_damaged_by_enemy_max_15.
  • Balance_RepairDelaySecondsAfterCrippling is now Balance_RepairExtraCostForSecondsAfterCrippling.
    • This is balance_repair_extra_cost_for_seconds_after_crippling, and it has been dropped from 300 to 120.
    • The purpose behind this is to give some weight to ships being crippled, versus them just popping in and out of cripple status with no ill effects. However, a 5 minute malus is a really long time, versus 2 minutes is more appropriate.
  • Balance_RepairDelaySecondsAfterDamagedByEnemy is now Balance_RepairImpossibleForSecondsAfterDamagedByEnemy.
    • This is balance_repair_impossible_for_seconds_after_damaged_by_enemy, and the default remains 5.
    • This idea with this was always to prevent engineers from being able to infinitely sustain a golem or similar when it is actively in the middle of a firefight.
  • Balance_RepairDelaySecondsAfterTransformingCommandStation is now Balance_RepairImpossibleForSecondsAfterTransformingCommandStation.
    • This is balance_repair_impossible_for_seconds_after_transforming_command_station now, and the default value remains 5.
    • The idea with this was always to make it so that when players rapidly switch command station types, they have a short burst of vulnerability, since this can be a powerful tactic.
  • There is a new Balance_MultiplierForShipsThatHaveExtraRepairCostTemporarily.
    • This is balance_multiplier_for_ships_that_have_extra_repair_cost_temporarily, and for now we're going to try this at 3.5.
    • This gives a very nontrivial extra cost to ships being repair for 2 minutes after they are crippled, but of course they are not actually repaired for the entire 2 minutes, neccessarily.
    • The important thing is that players aren't prevented from playing, and it doesn't seem like engineers are just broken or something, but there is an economic hit that you take.
  • RepairExtraCostForSeconds is not able to be reduced by any means; not by being on your planets, or any other factors.
    • Please note that this does not block natural shield regen from happening, unlike RepairImpossibleForSeconds, which does. Same deal with personal regen of health in general.
  • RepairImpossibleForSeconds also is now not able to be reduced by any means, unlike even the past beta; however, it cannnot EVER be more than 15 seconds, rather than often being 300 before. And nothing in the game actually would set it to be more than 5 seconds at the moment for any player unit.
  • Thanks to Chthonic_One, ArnaudB, Badger, and others for weighing in on potential ideas and problems with the way the system was set up last beta release.

TLDR:

• 1. Ships that are shot can't be repaired for 5 seconds, same as always. No way to shorten that time period now, since that is kind of defeating the purpose and it's so darn short.
• 2. Ships that are crippled now cost 4.5x as much metal to repair for 120 seconds, and there's no way to shorten this either. This doesn't make ships that are constructed and sent to them more expensive, and it doesn't prevent them from being repaired, but it will put a dent in your economy if it keeps happening over and over.

• I had several more complicated variants in the pipeline, as well as some other options, but wound up simplifying it down to this.
• Actions need consequences, but that doesn't mean you sitting there bored, or even having to withdraw your transport and fiddle with things.

Bugfixes

• Ships still under construction (like a Zenith Power Generator bought from a Trader) should no longer cause AIP on death (scrapping or being killed or otherwise). Additionally, they should not cause a response from the Vengeful AI type. Please let us know if there are still problems with this, as we have not tested the fix.
  • Thanks to Ithuriel for reporting.

• Added some slight spacing after the range declaration on ships, to avoid text running into one another.

• Fixed a bug with the mirror damage on medium and up tooltips not showing the full text properly.

• Put in a fix that should prevent any situations where too few bastion worlds seed.
  • Tested on X maps and they were seeding the correct number, even when they were further away than adjacent.
  • Also verified that with randomized AI layouts (DLC2 feature), the bastion worlds work properly, but again are just seeded further away.
  • Thanks to Democracy for reporting.

• Fixed a bug in the prior beta where most tutorials were broken (unable to generate maps), and generating maps with custom AI layouts was also broken.
  • Thanks to Lowestech for reporting the tutorials.

• Fixed a rare random couple of exceptions that could happen in map generation in Helper_MakeSureFactionIsSetUpForPlayer, possibly mainly after coming from tutorials to the lobby?
  • Either way, whatever it was is fixed now, and that code will give more specific error messages if it has errors in the future.

• If you tried to load a tutorial and it errored, then sometimes it could make for errors then appearing in the lobby the next time you went there. Fixed that.
  • Essentially, if Praetorian Guard are missing in the lobby, it now adds those.

Beta 2.750 Bastions And Tuning

(Released March 3rd, 2021)

• Put in a new xml feature, will_not_regenerate_dying_ships_if_health_below_this_percentage, which should take a floating point number between 0 and 1. If it's set to 0.2, then if a regenerator ship has less than 20% health remaining, it will not actually regenerate any other ships. Normally the logic is only "don't kill yourself, regenerator!"
  • Applied this to the DLC2 Valkyrie Cruiser, at 0.5, but have not tested that it works.
  • Thanks to Zeus for requesting.

• For a long time, there has been a mechanic where if you outnumber enemies 10:1 on the local planet, or there are fewer than 0.2 strength on the current planet, then the repair lockout would immediately reset to zero. For hostile factions, their rule was if they had no enemies present, their repair lockout would reset to zero immediately.
  • The repair lockout is based around how long it has been since you were last shot, and the purpose of this mechanic was to make it so that with turrets and other structures on your own planets, you don't have an incentive to scrap and rebuild your turrets because of repair lockouts after you've beaten the enemy off your planet.
  • This mechanic now only applies to "planetary" forces, which means basically turrets, minefields, station-keepers, and other things like that which can't go through wormholes (most anything from the ODSS, and most of the small forcefields).
  • People had noticed that they could game this system with their mobile offensive fleets by transiting wormholes to adjacent safe zones and bypassing the repair delays on things like their transports, which was a source of minor cheese in a variety of cases.

Fixes And Tweaks

• The variable LastTimeTakenDamageFromPlayer on ships was a misnomer, it turns out, and it has been renamed to LastTimeTakenDamageFromBeingShotByAnyone to prevent all sorts of possible confusion.
  • This should affect nothing, unless some mod happens to use it. But in that case, there was a good chance the mod might have been using it wrong based on the name, so it would be worth a review anyhow.

• Fix an incorrect ARS hack tooltip
  • Thanks to Metrekec for reporting

• Add some defensive code to the hacking tooltips
  • Thanks to Chthon for reporting

• Fix a typo in the TSS hovertext
  • Thanks to Strategic Sage for reporting

• Updated a bunch of the code for how we find "external data" that is used by factions, mods, and various other semi-custom or fully-custom data.
  • The old methods have been marked as obsolete, which will force any code-based faction-related mod authors to update to the new style before they can compile again, but it will NOT break mods in the meantime.
  • For those mod authors please make sure and take a look at how we handle this in the central codebase, for instance in the dyson sphere faction or really any of them except AI Reserves (which is an odd case). There is a super-simple way for you to resolve the obsolete code, but it's actually leaving in more verbose code that does not have the extra error checking. If you swap over to instead using the new code that the main game uses, then when there are failures we will better know why.
  • This overhaul was rather time consuming (most of a morning to do), but it's something that keeps coming up again and again in both multiplayer and single-player. This new code makes it so that we should be less likely to have errors there, and that if we do it should be a lot more apparent WHY we are having an error.
  • Related to this, made some minor improvements to the dyson sphere check code when hacking them, but frankly it's unlikely to fix whatever the original issue was, since the original issue was actually buried down deep in this other code. The dyson bug itself may or may not be fixed, but at least if we have it happen again, we should know more about why. A savegame prior to doing a neutral planet science hack on a dyson world, if that does produce an error again, would be very welcome.
  • Thanks to Isiel and others for various reports that led to this.

• Set wormhole transit speed bonus modifier to 1 for Trader provided Tamed Macrophage. This should suppress text about the bonus since as a station-keeper it cannot leave the planet anyways.
  • Thanks to CRCGamer for fixing!

• Fixed a potential thread race condition when planets were created on not the main work thread.
  • This may never actually have happened, but let's not tempt fate. In multiplayer it was more likely to be a random error on clients, though we had not seen it yet.

• Fixed an error in ReactToPowerLevel() where it would throw an exception and disable a faction if the faction didn't own or influence any planets at the moment when it ran this.
  • This will wind up with these factions who have apparently aggro'ed the exo forces of an AI just not getting any exos against them, presumably because they have no territory for the exo to attack.
  • Thanks to Histidine for reporting.

• Fixed an issue that could happen in various circumstances when the name of factions was being created on the fly on multiple threads at once. This had to do with a static Stringbuilder in all cases, and now we have that marked as ThreadStatic and updated so that it uses one per thread. How did we ever manage with things like this prior to ThreadStatic??
  • Thanks to Badger for reporting.

• Put in a tweak that should probably fix the Raptor Drones showing ??? on the upgrade science list. Can't test for sure, since we don't have a savegame.
  • If this doesn't do it, we may just need to remove these specific ships from the upgrade list of info, since as CRCGamer notes they only come from drone guns. So they don't really have a cap, and however many there are at the moment is incidental to how much combat has been happening lately.
  • Thanks to samnainocard for reporting.

• Drones that were shot out of a drone gun belonging to the outguard would not properly be in the outguard faction, and thus would cause errors later. Fixed.
  • Thanks to Zeus for reporting.

• When the game is thinking about despawning outguard because they have been around for long enough, it no longer throws any errors with any members who are missing outguard faction info. Those simply won't despawn. Most likely these are just drones or similar who are going to die soon anyway.
  • Thanks to Zeus for reporting.

• Fireteams now use ConcurrentDictionary instead of ArcenSparseLookup, as apparently these could get a bit wires-crossed with multiple threads, and this may be slightly more efficient in general, anyway.
  • This prevents certain nullref exceptions that could happen in BuildShipsLookup, and probably also a few other things.
  • It's worth noting that it turns out that the foreach statement with a ConcurrentDictionary is fully thread-safe, and in general supports continuing iteration over the directionary while things are added or removed, which is a strange and wonderful superpower of this.
  • Thanks to Sigma7 for reporting an exception in BuildShipsLookup.

• Fixed a bug with some fireteam tracing always being on, and slow things down because of that, because of a tracing flag having a typo causing it not to be used. This may have only affected the scourge, not sure.

• In the tooltips, ships that are immune to repair no longer show the repair timer for when they are next able to be repaired -- that was contradictory and confusing.
  • However, if such a unit is able to regenerate itself, it shows that timer and says that's how long it will be before it can start regenerating health again.
  • Additionally, for ships that can regen AND be repaired, the wording for that has been updated.

• The "Deals half normal damage due to firing from under forcefield." warning is no longer shown for any units that get a bonus for firing out from under forcefields.
  • The actual mechanic was working fine, but the tooltip was still showing this one piece of info, which was wrong.
  • Thanks to Democracy for reporting.

• Fixed a bug where if units like the bastille turret were under a great-forcefield (which does not cause damage penalties to those firing out from under it), then the unit with a bonus would also not get the bonus for firing out from underneath it.

• Turtle AI types now have a special version of the Troop Accelerator that they can seed that causes no AIP when it dies.
  • This also prevents the adaptive AI from just causing tons of AIP that you would have to take whenever it cycles through the Turtle type.
  • The description on this is updated to note this, in much the same way that the Zenith Trader version of the troop accelerator is.
  • For now, this is still the only type of special structure seeded by the turtle type, but it will no longer be so dire when it happens.
  • Thanks to Daw11 for reporting.

• Peacemaker AIs are seeding tons of Orbital Mass Drivers, which was by design, but seemed confusing especially with adaptive AI types.
  • Now they are seeding Peacemaker Mass Drivers instead, which have half the durability in exchange for their prevalence.
  • Thanks to Daw11 for reporting.

• The Fortress Baron now seeds a special Baron Fortress that is weaker and has a unique name.
  • The fortresses purchased from the Zenith Trader are also now weaker and have a unique name.
  • Same with the black hole machines for the one-way doormaster.

• Added a new Planetary range setting with 70k base range (56k range after the default 0.8 range modifier). This serves the purpose of having weapons with a range that in theory can go from one edge of the planet to another (2x 26k = 52k) without being at the 80k range limit of ships stopping to move.
  • Modified a number entities for ZO with this range setting so that they would still move into range.
  • Thanks to NR SirLimbo for adding!

• Added a new can_be_stacked_even_if_would_not_normally for ships, which basically allows us to override anything else that would automatically prevent stacking (including cannot_be_stacked).
  • The general idea is that there are exceptions to generalities, and this allows us to override the edge cases.
  • This is now applied to the Pulsar Punk, Etherjet, Persuader, and variants so that they don't slow down performance to a crawl when a bunch of them are used. Normally strikecraft either don't have tractor beams, or don't have such a magnitude of them at once.
    • In the future, if we run into a case where we really need to differentiate ships like this, we can make stacking and non-stacking versions.
  • Thanks to Democracy for reporting the issues.

• While we are at it, just to be on the safe side, we are going ahead and reverting the change that made all tractor sources not-stackable. That is probably safer.
  • This might mean that things like mosquitos need to cannot_be_stacked set to true on themselves.

• Fixed a longtime bug that was preventing regular guardians from ever seeding in waves or wave-like things, even when they were supposed to (such as on hacking responses).
  • Dire guardians were able to seed on truly massive waves or wave-like things because they are handled differently, but guardians themselves have long been absent. For a third hack of an ARS, this could mean a CPU-grinding amount of tiny ships, rather than offsetting that at least some with guardians to keep things moving and interesting.
  • Thanks to Democracy for reporting.

• Fixed an issue since Feb 25th where all of the beam weapons and other lines other than the new chain lightning had their hdr glow removed, making them pretty hard to see.
  • Thanks to Democracy for reporting.

• Fixed a bug with waves where if they included any dire guardians or regular guardians, they would ONLY include those. This was a simple code error, and now it properly mixes them with strikecraft when relevant.
  • Thanks to Democracy for discovering.

• Previously, if you sabotage hacked a warp gate (and it died without costing you AIP), then if you captured the planet, you would still get the full AIP later. This should be fixed now, though the fix is untested. If anyone can confirm in the next build, that would be appreciated!
  • Thanks to Bionic for reporting.

• Fixed a bug in multi-AI games where it would complain about duplicate factions when you loaded in, with relentless AI wave factions and CPA logic factions.

Endgame Difficulty

• The AI Difficulty xml has a new feature called number_of_ten_minute_increments_before_homeworld_is_fully_defensible.
  • This was previously always set to 18, which is 3 hours of time.
  • Essentially, the AI homeworlds want to grow to be monsters of defenses by nature, but we introduced this mechanic a few months ago in order to hold them back from being too crazy at the start of the game.
  • The problem is, they would still often become crazy, more than was warranted, as they were often reaching a cap because of players alerting few planets. This doesn't make sense to be global for all difficulties, as CRC Gamer pointed out.
  • So, by difficulty, the timings are now:
    • 1-5: 60 (10 hours)
    • 6: 50 (8.3 hours)
    • 7: 40 (6.6 hours)
    • 8: 30 (5 hours)
    • 9: 24 (4 hours)
    • 10: 18 (3 hours, previously what everything was set to)
  • Thanks to ArnaudB and CRCGamer for helping us discover this!

• Added a new debug setting to the AI Logging subsection in settings: Track Guard Post Content Details
  • Any time ships are added to a guard post for any reason, it will log a 'reason code' on the guard post with why they were added. Useful for hunting down problems with planets with too many guard posts. Not a great thing to leave on long-term or for general play, as this will substantially bloat savegame files. No impact on game speed, however. You can see the results of this tracking data by hovering over the guard post you are curious about.
  • Thanks to ArnaudB and CRCGamer for inspiring this addition.

• The way that we check for external invincibility on units is now a bit more efficient.

• The way that the game checks for "must have at least this number of something for external invulnerability" is now based on "greater than or equal to" instead of just greater than.
  • This was probably leading to things being off by one in a way that folks noticed with AI Eyes.

• Dire guard posts now have their own SpecialEntityType, of DireGuardPost, rather than GuardPost.
  • All of the things that affect GuardPost still affect DireGuardPost, including things like the hacks to weaken them.
  • That said, they won't contribute to the invulnerability of AI Eyes anymore, so there's that!

• The AI Overlord now gets invulnerability only from dire guard posts, rather than all guard posts. However, it gets it from all of the dire guard posts that it owns in the galaxy, rather than just those on its home planet, now.
  • For existing savegames, this is effectively no difference.

• The text for invulnerability provided by outside sources has been cleaned up, allows for plurals properly, and will show on any entity now without need for a special description appender.
  • This should mostly apply to AI Eyes and the AI Overlord Phase 1, but it also applies to things like the Custodian with Astro Trains, and it would apply to modded or additional units that use this sort of mechanic.

• When a former AI homeworld is reconquered by an AI, it should now no longer seed dire guard posts.
  • Additionally, whenever any AI world is reconquered by an AI, it should have a different mix of guard posts than it used to.

• Three new fields on AI types:
  • number_of_dire_posts_on_homeworld: How many dire guard posts will there be on the AI homeworld? All other guard posts that are seeded there will just be regular armed guard posts.
    • Given that there is variability in the number of guard posts on AI homeworlds in general, this makes the dire part consistent at least for a given AI type. Dire guard posts costs AIP to destroy, and are way tougher. And also are the ones protecting the overlord.
    • Note that we are not toning down the total number of guard posts, since that would actually make things harder since the same number of units would be concentrated in a smaller number of receptacles.
  • number_of_bastion_worlds: how many mark 7 "bastion" worlds are there in the vicinity of the homeworld? No longer is this limited to purely just however many planets are adjacent to the AI homeworld.
    • If there would somehow have been more mark 7 worlds than this based on an extreme map type (lots of links), those will now be mark 6 worlds, instead. Aka, easier.
    • If there would have been very few based on very narrow map topology (x or snake), then the mark 7 bastion worlds will be seeded further out in a line. Aka, harder.
  • number_of_dire_posts_on_bastion_worlds: each bastion world will have this number of dire guard posts, allowing the general number of dire guard posts to be similar (or higher or lower, depending on the AI type), while still having fewer on the actual AI homeworld.
    • This makes it so that the AI homeworld is not always just this giant pit of doomstacks of AI units, with them instead being spread between a few nearby worlds. You have to kill the dires on these worlds before you can kill that overlord, after all. Same fight, but less concentrated in one small world, and thus a bit easier but also hopefully more interesting.
  • Thanks to ArnaudB and CRCGamer for inspiring this change.

• AI types adjusted:
  • Most of them: number_of_dire_posts_on_homeworld 6, number_of_dire_posts_on_bastion_worlds 2, number_of_bastion_worlds 3.
    • This spreads things out considerably.
  • Turtle: number_of_dire_posts_on_homeworld 10, number_of_dire_posts_on_bastion_worlds 5, number_of_bastion_worlds 1.
    • This gives that doomstack turtle effect.
    • The Vengeful AI in DLC2 also works like the Turtle in terms of its balance of these, since that's an interesting paradox.
  • Fortress Baron: number_of_dire_posts_on_homeworld 4, number_of_dire_posts_on_bastion_worlds 3, number_of_bastion_worlds 4.
    • This gives the effect of this really big fortress area from several worlds.
    • The Spire Hammer in DLC2 also works like the Fortress Baron in terms of its balance of these.
  • Peacemaker: number_of_dire_posts_on_homeworld 7, number_of_dire_posts_on_bastion_worlds 1, number_of_bastion_worlds 5.
    • A bit in the middle, with a slightly more intense homeworld, a LOT more bastion worlds than normal (that's scary), but way less to do on each one.
    • The Geneticist in DLC2 also works like the Peacemaker in terms of its balance of these.
  • Beast (DLC2): number_of_dire_posts_on_homeworld 10, number_of_dire_posts_on_bastion_worlds 3, number_of_bastion_worlds 2.
    • A bit of a different flavor, to go with a very brutal AI type.
  • Thanks to ArnaudB and CRCGamer and Democracy for various suggestions here.

• The way that guard posts are chosen for AI-allied Scourge (DLC1) has been updated so that it will never use dires, and is more proper in general.
  • If we accidentally broke it, then let us know that also, too, please.

• The number of mark 7 worlds used to be partly controlled by the AI difficulty, and partly based on map size, but this is now entirely based on the AI type.
  • This style of brokering difficulty (mainly lowering it for lower difficulties or smaller maps) is not really required. This will make smaller maps a bit harder, though.

• Bastion worlds are now unusually hyper metal-rich, which is a nice bonus if you choose to capture them (you certainly do not have to).

• The initial seeding of AI homeworld strength caps is no longer 2x what it would otherwise be.
  • That strength is now effectively spread out among the various Bastion worlds on which you now need to do at least a bit of fighting.

• Wardens now have a bit more affinity for bastion worlds, like they do for AI homeworlds.

• Spire cities can now not be built on AI Bastion worlds. That would have generally been a terrible idea, anyhow.

• AI Instigators can no longer spawn on AI Bastion worlds OR next to them (this can avoid some really nasty situations that previously you could get into).

• The ZM from DLC2 also can't spawn on AI Bastion worlds.

DLC2 Features

• Forcefields or regular ship hulls can now be "Electrotoxic"
  • This is set up with returns_this_percentage_of_damage_if_damaged_by_enemy_attack, which should be a float that is larger than 0 if you want to use it. 0.5 is 50%, 2 is 200%, etc.
  • Depending on how strongly electrotoxic a target is, it will deal back some portion of any damage that it takes from incoming weapons fire (beam weapons and AOE and so on are included).
  • If a bubble forcefield takes 8000 damage from some ship that has a bonus against bubble forcefields, as one example, and the forcefield was set up to be 150% electrotoxic, then it will deal back 12000 damage to the shooter.
  • This has some similarities to the mirror mechanic in the first AI War, but various different applications, and probably will only be used on some specialist units. This is something for in DLC2.
  • Not yet tested.

• On weapon systems, instead of setting firing_timing="OnlyInRange", you can set firing_timing="WhenParentEntityHit"
  • This makes it so that the weapon won't fire normally, but will instead only return fire when the ship holding the weapon system has been hit.
  • At the most basic level, if ONLY this is turned on, this makes for a "retaliatory" weapon system that can be anything.
  • The range and such of a retaliatory weapon is technically irrelevant, if the enemy can hit it, it can fire on the enemy. Reload timings are also irrelevant. It fires only as often as it is shot, but is able to fire every time it is shot.
  • That said, if this is set up as a beam weapon, or chain lightning, or some AOE burst, or a "drone gun," or whatever else... those things will all work just fine.
  • It's worth noting that the shots that are fired in this way are still able to be blocked by forcefields, and have damage bonuses, and all of the other regular things that a shot does.
    • The one main thing that a shot like this can NOT do is cause a retaliatory shot to be triggered on the other end (you'd get into infinite loops). But if the other end has an electrotoxic effect, that WILL be triggered.
  • Please note that at this point, we are not at what you would call a "mirror shot." Rather, this is just a passive weapon system that fires back on enemies whenever it gets shot by a non-retaliatory shot. This is interesting even before we take it to the "mirror shot" level.
    • It is also worth noting that this will not be triggered by semi-passive effects like electrotoxic effects, attrition damage, or similar. But AOE damage will trigger it.
  • It is also worth noting that this CAN be given to a ship that is a forcefield generator, and any time anything hits the forcefield -- aimed at the generator or something under it -- it will generate a retaliatory shot. Probably not something to do, but it can be done.
  • This is untested so far.
  • Thanks to Zeus and CRCGamer for requesting.

• Mirror time! Your weapon system has to have the above firing_timing="WhenParentEntityHit" set on it already. And then add a new returns_this_percentage_of_damage_when_firing_retalitatory_shot="0.5" or whatever ratio you want on the weapon system.
  • This works ONLY with return-fire shots that are not drone-gun style. If you use it with AOE style return shots, or a beam weapon return shot, it's uncertain what the result there would be, too. Ideally this is just used with a general projectile. This also only gets one shot per salvo, and anything specifying more is ignored. It's an entirely reactionary system.
  • If you set this to something like 0.5, then 50% of the damage that was just done to the ship with this system on it will be "reflected" back in a retaliatory shot at the ship that originally fired on it.
  • You can, of course, set that ratio to be 1 (for full damage) or more than 1 (for amplified return fire).
  • Things like forcefields can block these sorts of mirror shots, but other forms of damage reduction (other than temporary or permanent invincibility) will not work.
  • Additionally, any weapon damage bonuses are irrelevant with this sort of effect, so combining those is going to be confusing to players (don't do that). If you want that sort of thing, use retaliatory shots without the mirror component.
  • This is untested so far.
  • Thanks to Zeus and CRCGamer for requesting.

• Added in a new custom_repair_delay_seconds_after_damaged_by_enemy_max_511 and custom_repair_delay_seconds_after_crippling_max_511, which are on entities and allow for you to set custom amounts of time that a unit is locked out of regen and/or repair after it is either damaged or crippled, respectively.
  • Normally it is 300 seconds after crippling (5 minutes), or 5 seconds after being damaged. But for various units it is useful to be able to set other values for one or both things. The max this field can take is 511 seconds, because of the data type it is. Anything set higher will just be reduced to 511.
  • It's worth noting that as long as you set it to something more than 0, that will be the new default. So if a certain flagship should recover more quickly from crippling, say only 100 seconds instead of 300, you can set that sort of lower value and it will apply. Anything 1 second and up.
  • Thanks to CRCGamer for suggesting.

• Additionally, we now have tooltips that explain what these values are for each unit in the full tooltip, in grayer text with red for the actual time values, so that people are aware this mechanic exists at all, if they previously were not.

• There is, in general, a new large background thread of potential features all relating to area boosts of ships against allied ships.
  • These get subdivided into two categories: not-evaluated-at-time-of-being-fired-upon, and those which ARE evaluated when a ship is fired upon. Only the latter is used for now, but we can do various things in the future with the other.
  • Essentially this provides a bit of a framework for us to implement a number of features, as well as letting us actually implement the one feature we care about today.

• Added a new attract_range_for_shots_against_allies feature, which should be a range specified like for tachyon or tractor or gravity, and which should be on its own system on a ship.
  • This was implemented as a ship system so that we can turn that on and off by mark level of the ship, or via modules later on. All those nice system-based features that are optional based on certain contexts.
  • This feature is very simple. UI description:
    • ATTRACTANT FIELD: All shots fired against friendly units within range X are automatically redirected at this unit, instead.
  • Thanks to CRCGamer for requesting.
  • Untested so far!

Mod Updates

• AMU:
  • Fixed a bug where the new GetFactionListOrEmpty function would only find "dead" AIs or Players at the start of the game due to them having no king units. This was causing errors with AI-allied, potentially also player-allied Kaizers Marauders.
  • Fixed Battlestations and Citadels not being listed under Hackers, as they now TRULY ARE allowed to hack things in Vanilla.
  • Simplified the logic for setting up a rollup of abnormal hackers in the ExtendedHackingImplementation.
  • Moved and centralized the logic for Placement Condition location searches to the PlacementCondition class and made it static, which makes it possible to access it without any need for variable tags.
  • Created a new function within AMU in the galaxy settings under AMU - Topography: Capturable Repositioning:
    • This changes the locations of capturables (Zenith Power Generators, Zenith Matter Converters, Major Data Centers, etc) on a planet to not be near wormholes and somewhat on the outter edge of the system.
    • The result is that there will be less situations where important targets such as MDCs are located right on top of wormholes or in the travel line of Astro Trains, making it extremely difficult to defend them.
  • Vastly upgraded the Auto-Juggle Energy assistant:
    • It can now handle different types of converters, with different efficiencies at different marks, and will always find the most efficient way to re-allocate excess power, or make up for needed energy.
    • It also should be more efficient (in CPU and bandwidth, not so much in RAM I'm guessing).
  • Fixed a really bad bonehead programming bug where the PlacementCondition.GetLocation() function would not try to seek points at X distance to a specified location all the time - it would wrongfully update the specified location with the current location attempt, leading to greatly misplaced Kaizers Marauders turrets.
• Micro Mod Collection:
  • In conjunction with the upgraded Auto-Juggle Energy the Energy and Matter converters were changed:
    • Both can now increase in mark level with Orbital Mining and Economic Command Station techs, and by directly spending science on a command station.
    • Matter Converters were nerfed (initially) to only grant 40k energy for -500 metal/s, but gain 5k yield per mark. At mark 7 they now grant 70k energy.
    • Energy Converters were buffed to now grant 200 instead of 150 metal/s, further increased by 25 per mark. At mark 7 they now grant 350 metal/s. The cost stays at flat 50k energy.
    • At its peak the Human Matter-Energy conversion loss is 2%. So a chunk of 100g of metal converted to energy and back would return as 98g, or a power cell with 100W converted to metal and back would return as 98W. That's pretty amazing!
• Kaizers Marauders:
  • Instead of getting the outpost and capital hacks done by battlestations and citadesl through the above logic they now use the new Vanilla logic. This changes nothing for the player but cleans up the code.
  • Adjusted them to work with the new PlacementCondition logic.
  • Adjusted them to also have color-coded and internally recoded debug toggles.
    • Thus it was discovered that Defector Traces were internally still using Stage 3 General Tracing, and thus the toggle for the player to enable did nothing withle the general tracing one did more than it should.
    • It was also discovered that in no place of the code the Invasion-, Offensive- and Defensive Spawners were traced at all, and internally the debug toggles were entirely unused. This has been fixed.
• Extended Ship Variants:
  • Reduced the count of all player-found Etherjets, Ensnarer Etherjet and Thieves by 75% as they are they were made 4x stronger in Vanilla.

• Extended Ship Variants:
  • Fixed the Acidic Support Factory having built drones AND a drone gun for the same drone type, leading to infinite potential drone counts...
  • Normalized the amount of Sentry Frigates all the support fleets have to 2 like Vanilla fleets.

• AMU:
  • Updated a few variable overwrites:
    • Received Damage While Under Construction default is now 3, and it correctly stats that its minimum is 1x.
    • Fixed a bug where AMU would disable achievements if a player turned science per planet down, and then back up even if it was below the 2k Vanilla science count.
    • Added the settings to manipulate the amount of bonus hacking points for owning adjacent planets.
    • Added the settings to manipulate the AIP that AI Command Stations, Warp Gates and all stages of the AI Overlord inflict on the player. The first two combined also are the AIP cost for capturing a neutral planet that the player hasn't paid the price for yet.
    • Also added the settings to manipulate the AIP on death or claim of Data Centers, Fortified Data Centers or Major Data Centers.
      • Thanks to zeusalmighty428 for suggesting!
  • Updated the Source Code.

• AMU:
  • Fixed the amount of hacking points set in the player faction stating that the default is 100, it actually was 60.

AMU:

• Fixes and improvements to the Auto-Juggle Energy assistant:
  • Fixed the Auto-Juggle Energy not factoring in power gained through planetary resource boosts that even affect Energy and Matter Converters.
  • For the assistant to consider something a valid "converter" it now requires it to have 0 raw energy cost, and either positive energy and negative metal generation or the inverse. Structures that cost energy and produce metal no longer apply.
    • This was done to prevent any false positive on converters or modded converters that could be boosted with flagships to exceed the laws of energy converservation (see below in the Micro Mod Collection update notes).
  • The assistant now has 2 more "fail" conditions:
  • If, even after using all converters, a brownout cannot be prohibited all converters will be blown up. First the player needs to sort out their economy before the assistant can reasonably work, and they will most likely need metal to do so.
  • Ditto if using as many converters as needed turns the metal generation negative (even if not accounting for other metal uses) and the remaining time of spending all metal until it's gone is below 10 seconds it will stall until it's safe to convert again.
    • This is to prevent counteracting whatever the player may try to do and displays a warning. Without self-sabotaging on purpose it's quite hard to reach these states anyway.
  • Before it was unclear whether or not converters on planets with destroyed command stations were even considered at all. Now they will be considered, but nothing can be built, only destroyed if needed.
  • Updated the source code.

Micro Mod Collection:

• Changed the raw energy cost of the Energy Converter to instead be a negative power production.
  • This is so that it also is increased via planetary metal and energy income boosts. Otherwise at Mark 7 the player could begin producing metal and energy out of nothing, literally breaking the laws of energy conservation.
  • Known issue: When a game is saved while the player's energy production is negative (which can happen now, although it's not easy to do if not deliberate) it'll throw an exception, it should however have a fairly low impact, possibly UI-only.

• AMU:
  • Fixed the Max Waves From Wormhole Invasions having a default of only 2 when it states (and should have) 5.

AMU:

• Fixed the Metal Bleed notification from the Auto-Juggle Energy Assistant not working.
• As the planetary range setting is now in Vanilla it was removed in AMU.

Beta 2.748 Defaults And Fixes

(Released February 25th, 2021)

• New Show Faction Alliance Details In Escape Menu setting under Debug: Escape Menu Additions:
  • This will take up a LOT of processing while it draws, so it's not something to use frequently. But if the allegiances seem off for factions, this will let you see what they even are at a very detailed level.
  • Thanks to Bionic for inspiring looking into this, but thankfully all seems well.

• Fix to a typo in a recent release: Ambush Turret hull was reduced instead of metal cost. This has now been fixed.

• Fixed a pair of cross-threading issues that could happen when updating the amount of strength available per tech line, often on starting quickstarts but also other times.

• Fixed another trio of cross-threading errors that could happen in general when fleets were being analyzed while a fleet was still in the process of being added. This was happening at the start of quickstarts in particular as well.

• Fixed what seems to be the last of the cross-threading issues, potentially mainly on quickstarts, based on fleets calculating what planets they were on.
  • Thanks to Metrekec for reporting.

• Since basically the inception of this game, we've had the ability to say "if a keybind or a setting had its value set in a version older than X, reset it to the default value," which is extremely useful for when we are changing around defaults in ways that should be pushed to everyone without us renaming or removing a field. That's been a very useful tool, although it's not like it's used every day or something.
  • However, we have never had that actually in and functioning for galaxy options (the stub was there but it did nothing), or custom fields on faction types (like humans, for instance). Now we have that for both of those categories of data.
  • The first two fields to benefit from this are the number of cryogenic pods, and the number of home human settlements, both of which are meant to be a lot higher than they used to be, giving players way more metal income early in the game. But for anyone who was just firing up the game and not doing Reset To Defaults in the lobby (why would you), they were still having the old values.
    • Also, as a side note to this, the defaults as noted in the tooltips for these have both been corrected to the new values, whereas previously the stated defaults were the old defaults.
  • Thanks to Suzera for reporting the tooltip issue.

• Restructured the base sniper unit and children entities a bit.
  • Snipers, Trippers, and Railpods no longer get a x8 versus fast enemies moving faster than 1K speed. Instead get a 2x bonus against enemies that have been on planet for under 15 seconds.
  • Additionally basic Snipers get a 3x bonus against any enemy held within a tractor beam. From some testing this is *very* satisfying to watch them focus fire anything caught by things like Deathgrip Tackle Drones.
  • Additionally un-nerfed Tripper damage from being half that of a regular Sniper since the old potential x24 damage multiplier doesn't exist anymore.
    • This means they will be somewhat more exciting against higher powered engines than previously and a lot more exciting when they get to stack that bonus with the ambush bonus against things new to the planet.
  • Thanks to CRCGamer for implementing these changes.

• Added a new cheat to the game: grantalloutguard
  • All outguard are immediately considered contacted and show up on your sidebar, whether they were included in any beacons in the galaxy or not.
  • Thanks to Zeus for inspiring this change. It will make testing outguard much easier for everyone!

DLC2 Work

• Added new mechanics which can be used to create "chain lightning" out of beam weapons.
  • These are based on general beam weapons, with NOT having "hits all targets on, and having instead these filled in: beam_chains_out_to_targets_x_times, beam_chains_out_to_targets_x_range, and beam_chains_out_to_max_targets_from_each_source.
  • An example in DLC2 has been created in the form of Invincible Ohmu Chain Lightning, which will remain around longer-term as a cheat code for fun.
  • Video of this here: https://youtu.be/OqSfu8UoBTA
  • Thanks to CRC Gamer for suggesting the mechanic!

• Added a new target evaluator, LowestHealthAsMajorPrimary, which is also useful for large chain lightning weapons, and potentially other AOE items.
  • This was added post-video, and makes it work much better.

• Added a new only_targets_strikecraft_and_frigates, which is useful as a companion to large chain lightning weapons.
  • This was added post-video, and makes it work much better.

• Various chain lightning updates:
  • The Chain Lightning attacks now have their own visual effect that looks lightning-like and uses a new variant of the custom shader.
  • The chain lightning effect can now also have a minimum distance of travel, which is set to 200 by default in our test unit, so that units on top of one another don't get strange self-lightning.
  • The Chain lightning visual effects are now faster, and also last longer on the chain parts rather than the original blast.
  • Video of the results: https://youtu.be/wVGZhGtxtBM

Beta 2.747 Outguard Usability

(Released February 24th, 2021)

• Made the code for regenerator effects a bit more explicit in not being able to work if they are crippled, or remains, or various other things. I think this part was already working fine, but it's good to be sure.
  • However, it was definitely possible for the number of ships to be regenerated in excess of the ship cap, probably most likely when there was also a factory producing things at the same time as ships were dying, but really potentially in a variety of cases. These have now been fixed and that should not happen anymore.
  • Thanks to Sol for reporting.

• New bonus for players!
  • If you have captured a planet that is next to another human planet, it will yield an extra 5 hacking points. This is noted on the tooltip when looking at the planets.
  • Normally the big incentive is to planet-hop around, in order to reduce total AIP increases. And that's still definitely the overriding factor in play. But we wanted to provide a contradictory incentive for when you DO choose to take a pocket of planets. It's a minor thing, but can give a little bit of a boost to players who are thinking about mid-AIP or high-AIP playstyles.
  • Thanks to CRCGamer for suggesting.

• Long (tall style) popup text now has some extra whitespace under it to prevent it from ever being cut off. This is applicable for things like journal entries.

• Nerfed the Royal AI guardian variants to start at Mark 2 instead of 3. While certainly not an insurmountable challenge it definitely was far stiffer competition than the other brutal AIs with the general guardians being raised to mark 3 baseline.
  • Thanks to CRCGamer and Puffin for catching.

• Fixed the "Neutral Planet Science Extraction" description from being "old, do not use" to "Hack the AI to distract it as you deploy your science-related nanomachines to the planet."
  • Thanks to Metrekec and ArnaudB for reporting.

• The Warden fleet's ability to increase its strength is now back to what it was for the first 5 minutes of the game, and then its ability to boost itself due to low population cap is only 1/4 as strong until the 10 minute mark, and then only 1/2 as strong until the 20 minute mark. After 20 minutes, it is fully able to self-heal as much as before.
  • The Warden is still just as aggressive early on in, but this gives you a change to kick their butt and have them out of the way for the first few planets. It also keeps there from being deadly amounts of warden on early planets.
  • While we do want the AI to be more vibrant in the early game, we don't want it putting up this much of a fight until people have a bit of time to catch their breath. We definitely don't want to introduce any difficulty walls for new players.
  • Thanks to Daw11 for reporting, and Strategic Sage and TechSY730 for weighing in.

Outguard Updates

• Description of the outguard faction (not even sure if it's shown much of anywhere) now updated some to reflect them a bit better.

• Integrated Zeus's new journal entry with lore for the outguard into the game, and that now pops up the first time you meet any outguard. This is apparently going to see a few revisions, but it's nice to go ahead and have that in there.
  • As part of this, it also notes "Outguard sighted -- potential underlings!" in the chat, which is pretty blatant in terms of what these are.

• After you have hacked an outguard beacon, it now disappears, although it still exists. We're being tricky about this by setting its "time it was created" into the future continuously, which means that any of the scouting logic thinks that it was not seen in the last scan (since the last scan was the current second).
  • This lets us not disrupt how the outguard function from a technical standpoint, but makes the beacon not hang around in a confusing and annoying way.
  • Thanks to samnainocard for reminding us of this.

• In general, the text color for hack targets is now the mark level of the ship line, making it more obvious what is better to hack for.
  • The error text is now colorized next to them, saying "Not Possible" when they are not an option, rather than sometimes saying "Invalid."

• The hack "Access Outguard Beacon" has been renamed to "Make Contact With Outguard."
  • The description of the hack now says:
    • Hacking this beacon will allow you to contact some of the Outguard -- remnant anti-ai forces that can come to your aid. Since this is going to be routed through the AI communication channels, expect heavy AI response to this. However, once contacted, the Outguard will be at your disposal as-needed for the rest of your campaign. They can be called directly to planets far from your territory, they don't incur deepstrike or counterstrike penalties, and they can often reach places you may have trouble directly going yourself.
    • Many outguard can also turn the tide of a conventional battle, if you need them to. Some others have other functions that are more utilitarian or economic in nature.

• The intensity of the Outguard hack has been cut in half, making it far easier in the early game and not so apocalyptic in the later game.
  • Thanks to Strategic Sage for suggesting.

• Rather than getting two outguard groups when you hack an outguard beacon, it now shows you the two groups and allows you to choose one of them.

• All outguard groups now have a custom amount of AIP that it costs to hire them.
  • For anyone creating outguard, you MUST now specify aip_cost_on_contact="10" or whatever to set the amount.
  • For now, all of the outguard groups continue to cost 10 AIP -- but now the flexibility is there for many different amounts.
  • Note that, technically, all outguard groups used to cost 5 AIP since you got two at a time, but at any rate the transaction cost has not changed.
  • Thanks to Strategic Sage and others for inspiring this change.

• The tooltips for the outguard groups are now generated more efficiently, and include a bit more data.
  • Additionally, you can now actually see these tooltips BEFORE you hack for them, so you can see the details of a mercenary group's call conditions and so on when you choose which one to hack for.
  • Thanks to Strategic Sage and others for inspiring this change.

• The game now prevents you from accidentally hacking an outguard beacon to call an outguard faction that you have already contacted, as contacting them again has no benefit.

• When you C-click an outguard beacon to see the details of the ships in each potential outguard faction, it now also gives you the full details about the group like you could see on the sidebar. Previously there was no way to find this out prior to actually hacking and getting a group.

• You can now C-click individual outguard options when hacking, and see just what comes from them.

• You can also C-click outguard who are on-call on the outguard tab, and thus actually see what they contain. Previously once you had hacked them, you could only go back to the original beacon to find that information (and now the beacon is invisible, so that was also not even an option).

• Outguard beacon description updated:
  • Hack this beacon to make contact with extremely helpful groups of allies who you can later call to your aid. They can often strike where you cannot, and are great for emergencies. See hack details for more info.

• The outguard sidebar has had some color introduced, and does things on three lines instead of four, now. Overall much more attractive.

Beta 2.746 Release The Warden

(Released February 23rd, 2021)

• Number of typo fixes in starting fleets.
  • Thanks to ArnaudB for fixing.

• When the general mark level of the AI goes up because of AIP, it now gives you a chat message about that.

• Fixed a display bug where in various circumstances, it would say "pass" instead of "fail" for the "number of mark X cities other than me" on spire cities. This was purely a visual thing, but quite confusing.
  • Thanks to samnainocard and ArnaudB for reporting.

• Fixed a bug in the prior version that was causing elite ships to still count against the fleet-wide bonus limiter, when they were not supposed to.
  • Please note that elite ships already DO get any bonuses from fleet-wide effects, even those that say non-flagship (elites are not flagships). They don't benefit from something like the high capacity transport, though, since that explicitly only benefits strikecraft. But if there were an elite strikecraft line, that would benefit from the HCT.
  • Thanks to ArnaudB for reporting.

• Fixed a one-line typo in the prior release which broke all tachyon beams from actually revealing enemies.
  • Thanks to Democracy, Metrekec, and others for reporting.

• The AI Reserves no longer use the Relentless AI Wave faction, as that may have caused some strange logic and wasn't needed at any rate.

• The Exo waves generated by the Fallen Spire varioos things no longer use the Relentless AI Wave faction.
  • Not sure if this will help with their overall difficulty or not, but we'll see.

• Reconquest exo waves no longer use the Relentless AI Wave faction.
  • Same with extragalactic wormhole waves.
  • This caused them to sometimes come after players in a way that was not desirable, rather than actually reconquering the planet they were after.
  • Thanks to Democracy for reporting.

• Fixed a couple of annoying issues with the bandbox selection mode popping up even when you click on the UI as the start of what would be a bandbox. Not sure if this was a new issue or not, but it feels like one and was super irritating.

• Fixed an exploit where you could get a strange speed boost by putting ships into loading mode and then giving them a bunch of orders to attack something in the other direction by spamming the right-click button to get them to go the wrong way at the faster speed.
  • Thanks to Democracy for reporting.

Scourge Tuning

• Fixed the Scourge still going over the limit of builder. There was a division by zero dodged by defaulting to 1. I commented out the troublesome part.
  • Thanks to ArnaudB for fixing.

• Scourge warriors now requires the same experience requirement for each Mark (300).
  • A failure to realize they went back to Mark 1 after evolving meant they were short-changed, since lower marks are more expensive.
  • Thanks to ArnaudB for fixing.

Balance In The Player's Favor

• Reduced cost multiplier for hacking TSS/ARS from 2.0 to 1.5 With the cost increase, the 3rd hack was so expensive it was pointless to offer it, even on owned planet with half cost.
  • ODSS multiplier cost similarly reduced from 1.6 to 1.4 for coherence.
  • The later costs on repeat hacks were just getting so extreme as to never be worth it or affordable.
  • Thanks to ArnaudB for adjusting.

• Halved the cost of FRS, Cruisers and Destroyers hacks that were too costly, since they also had AIP cost.
  • Hacking the above structures only increases hacking response by 66% of the hack cost, instead of 100%.
  • Thanks to ArnaudB for adjusting.

• The number of human home settlements has changed from a default of 2 to a default of 4, with the range still being 0-5.
  • Also, the number of cryo pods has increased from 10 to 25, still with a range of 0 to 30.
  • Both of these will help folks have enough income in the early game, plus some extra spots of vulnerability.
  • Thanks to ArnaudB for suggesting.

• In a way that is almost certainly redundant given the existing code in place, we are now making sure that AI Instigators never spawn on AI homeworlds or planets adjacent to them.
  • The method we were using before (and still use) checks for an enemy king unit, which is generally sufficient, but sometimes the king moves (phase 2, for instance), or is already dead but the game is still going on (multiple AIs).
  • The new way prevents even former homeworlds from being a place that the instigator can spawn on or next to. The old method is still relevant and is also in place; as noted, this is largely redundant, but there are edge cases where they are not (AI king on the move, etc).

• Ships that are under construction now only take 3x as much damage as normal, rather than 10x as much.
  • Thanks to some folks on discord for requesting this. This will feel better on planets of yours that are under attack, and when you are trying to beachhead. That said, some counter changes to beachheading are also coming.

Player Nerfs

• For thematic and balance reasons, human homeworlds no longer have any science or hacking points on them. It's a strange idea that those would not have already been gathered, and there's no point making players wait for the extra starting science and hacking points. It also was giving even more of a boost in multi-faction MP, throwing those off from single-player.
  • Thanks to ArnaudB for noticing!

• The default number of hacking points for players starting out the game has been dropped from 100 to 60.
  • We do want folks to be able to do something right from the start, but it was a bit on the excessive side. Overall we're trying to make sure that players have plenty to hack, and plenty of hacking points, but at the same time that they are not drowning in hacking points.

• When you are building structures on an AI planet, or you have engineers helping to build a structure or boost ship production on an AI planet, there is now a slowdown in output based on the mark level of the AI planet. The raw output of factories is not slowed down (just engineers assisting the factory).
  • At mark 1-2, all is normal.
  • Mark 3: 90% expenditure (making things take 111% longer to create)
  • Mark 4: 70% expenditure (making things take 142% longer to create)
  • Mark 5: 58% expenditure (making things take 172% longer to create)
  • Mark 6: 50% expenditure (making things take 200% longer to create)
  • Mark 7: 40% expenditure (making things take 250% longer to create)
  • (Quick math reminder: just do something like 100 / 0.4 to get a number like 250, which is easy to read as to how much longer it will take).
  • Thanks to ArnaudB for suggesting, and Magedrifter for inspiring this change.

AI Buffs

• New pair of mechanics! If you make a minimally-aggressive move on certain AI planets, the mark level of that AI faction (and only that faction, in the case of multi-AI games) will go up permanently.
  • In some respects this sounds like a penalty to players, and you could certainly argue that it makes the endgame a bit harder. And that is true, to an extent.
  • However, what it also does is allow for players to not fear the cutoff mark for these mark levels from AIP rising, and thus starve themselves of planets during a tense game. The rise in general AI mark level to 3 is now inevitable when you are facing off against an enemy homeworld -- so, that gives you far more latitude to work, even on very high difficulties, with AIP rises without worrying about that trigger point to the next AIP level. It will happen, so part of your mission is now making yourself even more powerful rather than keeping yourself artificially weak to slide under that threshold.
  • The actual thresholds in general:
    • If you unload at least 10 strength on any mark 7 planet of an AI, it will gain mark 2 if it doesn't already have it. A message will pop in about this, and there is a note about this mechanic in the AIP tooltip now, too.
      • It's set at this level so that you can still fly transports through if you need to bypass AI planets to get to an instigator on the far side, but you probably can't fly a golem or ark through. Also, ships like etherjet tractors can't drag off substantial enough strength of yours to trigger this, either.
    • Similarly, if you unload at least 10 strength on the homeworld of an AI, then it will gain mark 3 if it doesn't already have it, and all the other same mechanics and tooltips and so on apply.
  • Thanks to ArnaudB for the discussion and helping hash this out.

• Fixed a couple of cases where certain AI logic was not triggering for ships belonging to the PG, the Relentless Wave faction, the CPA faction, or the AI Reserves.
  • This more or less meant that units from those subfactions would not trigger a variety of things, including things like the Shark B plot (but realistically probably plenty of other things as well).

• The Shark B plot of the AI now has a base of 4000 rather than 300, and increases by 20 per AIP rather than 10 per AIP.
  • This is substantially more challenging of an optional feature to enable now.

• The equivalent of shark B has been baked into the main game mechanics now. The level of intensity of that depends on the difficulty level of the highest-mark AI:
  • Nothing at diffs 1-4.
  • Diff 5: base 500, +1 per AIP.
  • Diff 6: base 750, +2 per AIP.
  • Diff 7: base 1300, +4 per AIP.
  • Diff 8: base 2000, +7 per AIP.
  • Diff 9: base 2800, +10 per AIP.
  • Diff 10: base 3600, +15 per AIP.
  • If you play on a difficulty that has these values, AND you have Shark B on, then those numbers will combine and it will be truly intense.
    • Note that the achievement for Shark B only gets triggered by actually having Shark B on, not by being on a difficulty level that has some of this mechanic in there.
  • So what is this feature, anyway?
    • Essentially, any time the humans lose a command station, the AI gets a new "go for the throat" sub-fleet of the hunter. Essentially they're trying to finish the job, and it gives a bit of an extra punch to when you lose some territory even temporarily. Note that this only happens when the AI itself, or an AI-subfaction kills the command station.
  • Thanks to ArnaudB for suggesting.

AI Warden Aggression

• Previously, the Warden was "overconfident" and would attack your ships when it was outnumbered only on LOW difficulties, but this would lead to the AI actually building up large amounts of warden ships that never did anything on higher difficulties.
  • The warden now is almost directly inverted in terms of its overconfidence, where it is now more likely to attack on higher difficulties even if it is outnumbered. Paired with some upcoming progressive income buffs, this should make for more fights with the warden while those fights not all being apocalyptic.
  • Thanks to ArnaudB, Magedrifter, and others for noting just how passive the warden is, even with huge advantages against the player.

• Made a number of changes to the warden danger evaluation for pathfinding and such for its fireteams, so it is far more likely to path through dangerous planets. This will cause minor scuffles and losses, and that's okay. We want the warden to be... brushing up against you a lot.
  • Rather than the warden thinking things like "if it's more dangerous than 1 strength, don't call a remote unit," it now only says "if it's more dangerous than 80 strength," which is a big difference. Sure that will get a lot of small units killed, but it also gives a chance for them to rebuild, and some will sneak by anyhow.
  • Rather than the warden going "if the total danger to where I want to lurk is stronger than 1/10th of my own strength, then just ignore that possibility," it now doesn't care about that sort of danger at all. The warden isn't here to cower.
  • Rather than the warden ignoring possible places to lurk if they "only" outnumber the enemy 10:1 on the way, they now will only go if the enemy outnumbers THEM by 30:1 now. This is basically assuming that the target planet is within about 3 hops, so on average they are outnumbered 10:1 (which is still a complete inversion from them previously needing to be 10:1 outnumbering YOU).
  • This means the wardens will give SOME consideration to not blundering into dangerous forces, but overall they will go where they want to. Again, more scuffles and skirmishes.

• As for potential targets for the warden fleet, they are already not interested if there are less than 2 hostile strength on them, but now there are a few other changes:
  • If it's not a planet that was originally a homeworld for the AI, and there is less than 6 friendly strength on the planet, then it's not interesting for the warden to try to defend.
  • Aka, if the planet has been neutered, and it's not currently being attacked by the hunter or something, then this is not a hill worth dying on.
  • Thanks to Democracy for suggesting.

• The Warden fleet gets its budget from "donations" from the main AI sentinels, but also sometimes from some other sub-factions that don't need their budget. That's the setup for this explanation.
  • However, now that we've got an intentionally-more-reckless warden, we want them to have essentially some progressive extra income based on just how depopulated they are. Rather than giving them a literal secondary revenue stream (which would be complicated), we're instead amplifying every donation they get if their population is heavily depleted at the moment.
    • Bear in mind that the warden is now going to be depleting itself in all sorts of ways that are not all that dangerous to you now, and so the idea is that you're getting more conflict (as with the AI waves), and thus having a more interesting time, but the warden will just evaporate and be gone forever in a game if it did that without having compensation to back it up. Also bear in mind the HUGE buffs that we've been giving to player economy and capturables, so this is also kind of balancing that out a bit, too. This isn't meant to just be cheap rubberband AI, but rather something that adds more flavor and challenge when the players themselves have been reaping buff after buff lately.
  • Okay, so when the warden's current population is (pick the first that applies only):
    • Under 1%, get a 10x bonus to donations.
    • Under 2%, get a 8x bonus to donations.
    • Under 4%, get a 7x bonus to donations.
    • Under 6.6%, get an 6x bonus to donations.
    • Under 10%, get a 5x bonus to donations.
    • Under 12.5%, get a 3x bonus to donations.
    • Under 16.6%, get a 2.25x bonus to donations.
    • Under 25%, get a 1.6x bonus to donations.
    • Under 33%, get a 1.25x bonus to donations.
    • Under 50%, get a 1.1x bonus to donations.
  • In general, this will hopefully help them maintain a minimal presence in the galaxy at the very least, while still getting into more overconfident scuffles with you on purpose.
    • Previously, the strategy of the warden was to wait and build up doomstacks that it would then use to crush you on planets if it could, but otherwise just lurk and avoid you. This was usually not super effective as a strategy, and when it was effective it did not feel good to be on the other end of that.
    • The general new strategy of the warden is aggression and chaos, and while it will spend some time lurking and growing, a lot of that is going to be situational. It is very unlikely to build a doomstack now (unless there's another place in code where we missed it being overly cautious), and it's highly unlikely that it will come and make a planet absolutely impossible for you to take unless you were already teetering on not being able to take that planet.
    • What the warden WILL do is generally hit you with small to slightly-huge forces, situationally, and it acts as a wildcard as you plot the destruction of various enemy forces. You can thin out the warden a bit, but overall they will just rebound from that, and frankly they will thin themselves out enough that, again, there's not the doomstacks there that you have to worry about in the same way.
  • Overall let's see how this does! If they feel too everpresent, or too weak, we can change numbers around. But more conflict, and more wildcards, are something we view as good in this sort of area.
  • Thanks to ArnaudB for suggesting a form of progressive income, although we twisted the concept a bit.
  • These numbers have been revised down after Democracy had a bunch of fights with the interesting new warden. Thanks for helping us tune those before unleashing them on everyone else!

Mod Updates

• AMU:
  • Adjusted the default amount of hacking points players can set to 60 to fit with the new Vanilla balance.

Beta 2.745 The Unsmothering

(Released February 22nd, 2021)

• Removed a duplicate file that caused an error on startup for Civilian Industries.

• Fixed an issue introduced in the prior beta build where clicks were often registering twice in quite an annoying manner. It turns out that our new adjustments to do raycasts on FixedUpdate in order to solve a rare problem on some OSX machines needed a bit more nuance.
  • Thanks to Timerlane for reporting.

• Fixed an issue with speed groups that could cause various inefficiencies, as well as also errors in multiplayer and now in single player, too. Not sure if this fixed the multiplayer sync problem with speed groups, but it does at least fix the single player one and the inefficiencies.
  • Thanks to Zweihander2021 for reporting.

• Fixed a couple of typos in the deepstrike description, and one strange wording.
  • Thanks to Ovalcircle for reporting in detail.

• Fixed a whole bunch of errors that could happen if you were viewing a tooltip of a unit right as it died. Command stations in particular.
  • Thanks to samnainocard for reporting.

• Fixed a typo from the prior version that was causing only reverse tractor beams to actually draw.
  • Thanks to Democracy for finding!

Fallen Spire City Upgrade Mechanics

• On the fleets sidebar popout, for any ship lines that have only a single entry, it now lists the details of the specific ship if there are any.
  • This is immediately useful for seeing the spire city details from hovering over the city hub entry in a given fleet, or for seeing flagship info on any fleet.

• The city-specific information for the fallen spire no longer requires you to unpause the game before it can be viewed.

• The tooltips for why a specific spire city can or cannot be upgraded are now way more explicit and clear on what is needed for this exact one.

• Spire cities no longer upgrade in an automated fashion at all.
  • For one, that was often happening in a strange way that was not correct.
  • For two, it was denying players the ability to choose which city would upgrade when multiple were possible, which was greatly annoying to them.
  • Instead, now at the top of the screen you see a notification with the Informational level of priority when you can upgrade spire cities. Clicking on it will actually do the upgrade.
  • Thanks to ArnaudB for reporting.

Volume Of Ships/Turrets Reduction

• ARS seeding adjustments:
  • There were 2 extra ARSes that tried to seed pretty close to human empires, in addition to 1 for each human empire, in addition to the 1 for each human empire that is VERY close.
  • Those extra 2 are now gone, while the other base of "2 per human empire" are still there. So in a single player game, this would move down from 5 close-ish ARSes to 2. In a 2-player game, that's down from 7 close-ish ones to 4.
  • There were just way too many choices for people, and an excess of options in general.

• There were then a bunch more ARSes way out in the galaxy at various points, and the number of those could be tremendously huge, too. We have tentatively cut those numbers in half, given how many extra choices and how much information overload people were having.

• Last change for ARS, that second one that normally would be close based on the number of humans empires is now just seeded kind of wherever.
  • This lets there be more ARSes out there in general in multiplayer, without the near-field being just absolutely riddled with them.

• For the ODSS, it was previoulsy seeding one near each human empire.
  • It now seeds that first one absolutely anywhere, instead of it being close by.

• For ODSS, it was previously trying to seed the other ones half in the near distance, and half anywhere.
  • The half that were going anywhere are now just gone, and then the half that were going in the middle distance now are within 3-11 hops, roughly.

• Lastly, the TSS. There are way too many of these, too.
  • There is still one put very near to each human empire, as these are very important.
  • There is another added for each human empire as well, but rather than being within 3-6 hops, it is now within 3-99 hops. Far less congestion in the early game of good things to capture.
  • The rest were put out in the main galaxy, some in the vaguely middle distannce, and some kind of whever. It now only seeds half as many, and mostly does them wherever.

• Overall, hopefully these changes lead to less information overload, and less overload of turrets held by players in general.

• The hacking point cost of hacking an FRS has doubled.

• The hacking point cost of hacking an ARS has doubled.
  • The hacking point rise per hack on ARSes is now 2x rather than 1.5x, and the response strength increase is now 2.8x rather than 2.2x.

• The hacking point cost of hacking a TSS has doubled.
  • The hacking point rise per hack on TSSes is now 2x rather than 1.8x, and the response strength increase is now 2.8x rather than 1.8x.
  • The maximum number of hacks on a TSS is now 3 rather than 4.
    • This eliminates some of the "large number of turrets left behind" that could otherwise happen when you take a planet and don't want to afford that super expensive last hack.

• Additionally, the TSS now gives you 5 options rather than 6.
  • And the ODSS gives you a max of 4 options rather than 6.
  • And the ARS gives you a selection of 4 strikecraft and 1 frigate rather than 6 and 2.
  • Again, all of these things come down to the sheer analysis paralysis that people could face, and the ability of people to get exactly what they want at all times, and therefore the increasing homogeneity of all galaxies.

• Bear in mind that we will adjust further if needed, but the goal is to give players meaningful choices that are different each game, and not SO many choices that it is either overwhelming or just a repetitive "choose your favorite thing" every game.
  • It's really important to remember that with ship upgrades now, getting to mark 4 is most of the battle, so having more diversity in unlocks (because you can't find exactly what you want) is a good thing and actually can lead to higher overall power versus just getting to mark 7 with a single tech class that you lean into completely. Put another way: if you have 6 ship or turret lines, you get more strength if you have 3 each in 2 groups that have been upgraded to mark 4, versus all 6 being in one group that has been upgraded to mark 7. You also get more strategic and tactical flexibility, but just the raw strength numbers are higher.

• • Thanks to Strategic Sage, CRCGamer, Zeus, Democracy, and Zweihander2021 for weighing in on this, among others.

Balance Updates

• Some updates, mostly nerfs to turrets in response to long-lasting issues for raid-melee ones, and nerf for exotic turrets after Magedrifter's recent Difficulty 10 victory.
  • Turrets changes due to TSS shift and balancing:
    • A number of turrets have adjusted costs:
    • Makeshift cost raised from 15 to 20k. Like Ruffian.
    • Scrap cost raised from 10k to 15k. It overshadowed Ambush far too much.
    • Ambush cost lowered from 22,5k to 20k. So its competes properly with raid turrets.
    • Crusher cost lowered from 225k to 150k. Hopefully it'll make it used more.
  • Also:
    • Makeshift lost metabolism damage since it has the drones, the description didn't even mention it anymore.
    • Shredder gained Makeshift's metabolism instead, where it'll be useful and make the turret attractive. Description changed to mention this.
    • Reduced Subverter zombification multiplier from 20 to 5. Its damage was buffed due to being too low, but its parasitism was far too strong even at low mark.
    • Harmonic maximum damage and damage per mark reduced by 25%.
    • Thanks to Magedrifter for the feedback on these last two, and ArnaudB for implementing all this.

• Prevented the Scourge from getting builders over the limits. Also fixed the experience needed for the initial armory and spawners.
  • Thanks to ArnaudB for discovering and fixing!

• Sentry Frigate reduction: with constant max caps they ate too much metal.
  • Reduced combat sentry frigate count on starting support fleets from 4 to 2
  • reduced combat sentry frigate count on seeded support fleets from 4 to 2
  • Reduced Sentry Frigate on command stations from 12 and 4 to 2.

• Adjusted all basic guardians to start at Mark 3.
  • Additionally bumped a small number of basic guardians that rely on weapons one speed step.
  • Note that units like Carriers or Tethuida with special properties were not included in the speed step increase.
  • Thanks to CRCGamer for identifying how weak guardians were in the new pantheon of player ships, and addressing this.

DLC2 Feature Items

• Cruisers and Destroyers have been added as new SpecialEntityType entries, and rollups.
  • They also have their own mark level scaling styles, although right now those are identical to how Frigates scale up.
  • These are primarily to be used in DLC2 and DLC3, respectively.
  • Usage: special_type="Cruiser" where you would use special_type="Frigate" or similar in the past.

• There is a new is_elite="true" that can be applied to entities directly (like we do with is_strikecraft="true").
  • Anything with a special_type="Cruiser" or Frigate will automatically also be marked as IsElite, so you don't need to also specify that one.
  • Any ship lines that are elite no longer count against fleet-wide bonus caps, and this is noted in the tooltips for the fleet-wide bonuses (along with the fact that flagships don't count).
  • The logic for making sure that only one elite line can be placed in a given fleet is not well-tested, but we'll get to that as we have units of that sort.

• Cruiser and Destroy Construction facilities now seed exactly two per any size of galaxy, plus an additional 1 per each 2 additional multiplayer human empires beyond the first (so a 3-player and 4-player game both have 3 cruiser facilities, and a 5-person game has 4).
  • These only seed if there are actually any cruisers or destroyers in the mods or expansions you have installed.

• Reorganized a fair bit of the order in which code is read in for ships, so that when they are being added to fleets or AI ship groups more of the info is present. This lets cruisers and destroyers actually be read in properly to their other categories.

• Added a new tractor_hits_engine_gx_greater_than, tractor_hits_albedo_greater_than, tractor_hits_mass_less_than, and tractor_hits_mass_greater_than.
  • The text on the tooltips has also been updated to dynamically write these various things for units affected.
  • We probably need some checks to make sure that none of the existing logic broke, let alone any new units that use these. This was complicated, but looking at one unit seems okay so far.
  • Thanks to Zeus and CRC Gamer for requesting.

Mod Updates

• Update to More System Defenders v1.23
  • Artillery Cruiser renamed to Artillery Destroyer in response to some other stuff elsewhere. Power cost and tX adjusted as well.
  • Basic AI Guardian units added by the mod have had their base starting mark level adjusted up to 3 to match base game changes.
  • Fixed the station keeping frigate entries that can be acquired via ODSS to not show up in the valid build entries for battle stations and citadels.

Beta 2.744 Scourge Industries, Inc

(Released February 20th, 2021)

• Fixed a one-line typo in the prior version that was causing tractor beams to not work in general, and exception popups on startup. Essentially the tractor beams would only work on ships with no engine strength at all, because tractor_hits_engine_gx_less_than was no longer being read from xml at all.

• Added wording in the tooltips for hacking ARS and TSS that says " The last sector can also be prohibitively expensive to hack, so unless you have no other options you may wish to leave it at just three." (or two in the case of ARS.
  • Thanks to Zweihander2021 for alerting us to folks running into confusion there on the pros and cons and expectations.

• The old hacking tooltip was really not accurate anymore. It read:
  • Hacking: Used to exploit flaws in the AI internal network. The more hacking points you use against an AI, the stronger that AI's response to your hacks will be. Only points spent against an AI owned structure count.
  • New version reads:
    • Hacking: general-purpose nanomachines used to improve your own units, sabotage enemies, or steal from enemies. Note that you are in a target-rich environment, and likely cannot afford even a third of the hacks available to you. Choose your hacks wisely, and capture more planets or destroy distribution nodes to get more hacking points. Also note: the more hacking points you use against an AI, the stronger that AI's response to your hacks will be. Only points spent against an AI-owned structure count.
  • Thanks to Zweihander2021 and ArnaudB for suggestions in this area.

• Fixed a bug with the new burst fire weapons that was causing them to not reset their timings properly, and thus generally switch back and forth every shot after their initial period.

• All of our camera handling logic and mouseover logic (mainly stuff that uses raycasts) is now being run on Update and on FixedUpdate. There was a rare issue where sometimes no raycasts were working for someone, and it is most likely the fact that we were only doing that in Update, when the main physics work is done in FixedUpdate instead.
  • We'll see if this solves the problem for the user in question, but it shouldn't cause any problems for anyone else and seems to be working fine so far.
  • Thanks to Badger and his play group for discovering.

• You can once again swap ship lines into/out of mobile support fleets.
  • Thanks to tadrinth for raising the issue.

• Fixed "Max Planets To Scout At Once" referring to the removed "Scout Adjacent Only" feature.
  • Thanks to Asirito for reporting.

• The max of "Max Planets To Scout At Once" is now 10 rather than 40, to keep things from getting too crazy. The default is still 8.

• The Other Defensive Schematic Server now has a unique icon separate from the Turret Schematic Server.
  • Thanks to Badger for suggesting.

• Put in some defensive code against tech vaults throwing errors when you mouseover them.
  • Thanks to Zweihander2021 for reporting.

• Fixed an issue where in recent beta builds, you could get seemingly-higher-level unclaimed golems on certain planets. They would then revert to their proper mark level of 1 when you actually capture them.
  • Thanks to NullZwei for reporting.

Tractor Balance Tweaks

• Any ship that produces tractor beams can no longer be stacked.

• Any ship that has a reverse tractor beam locked onto an enemy cannot be knocked back (aka knocked off the ship it latched onto). If the thing it is targeting is knocked around, then it will be knocked with it of course.

• Updated the way that tractor beams work in general (reverse or otherwise) so that if there is not a unique new ship to grab that no other tractor beams are grabbing, they will instead just grab onto one that someone else is grabbing. This is particularly relevant for reverse tractors.

• Fixed a number of bugs with reverse tractor beams. The now work, and they look AWESOME.

• Etherjet, Thieves, Persuaders, and similar all now have 1/4 the ship cap, cost 4x as much for players and the AI, have 4x the attack power and hull health, and have updated descriptions.
  • Existing savegames may still have too many of these on the side of players or the AI, which will make that local force far stronger than it should be.
  • Thanks to ussdefiant60 for pointing out how the fact that these will no longer stack is going to be problematic.

Fallen Spire Balance Tweaks

• Spire relic train reduced from 700 to 500, to give you more time to hack it.
  • Thanks to users on discord for suggesting.

• All of the various fallen spire notifications are now part of the Fallen Spire code file, which may not sound important (and it's not, directly), but shows how DLC or mod content can be added without having to update a central non-moddable file. At any rate, moving this over shows more easily how this is done, since we don't know of any mod factions that currently use notifications on the top bar (or maybe they do and we just don't know it).
  • We also have started on a new way of handling spire city upgrades, using the notifications on the top bar, but it's not ready yet.

Scourge Major Rebalance

• Scourge major rebalance:
  • Scourge mess of an economy got greatly simplified. Its buildings and units have the same experience cost at all intensities. Metal production is also the same for spawners and builders at all intensities. Now only two variables balances the Scourge's economy: Experience per second and Neophyte cost.
  • The Scourge does get nastier in others way at higher intensity, even without considering the two variables above.
  • Previously Scourge got the cheapest and most strongest upgrades first and the most expensive and weakest upgrades last. This is now changed. Scourge get the most expensive and strongest upgrades first, and the cheapest and weakest upgrades last.
  • This means Scourge stay at low marks much longer, but once units or buildings reach higher mark, they upgrade to the next faster and faster.
  • However this also means that killing high-marks buildings or builders will truly hurt the Scourge, whereas it was essentially meaningless before. At start, a spawner Mark VII has 300% the production of the Mark I. This gap narrows as the Scourge gain power.
  • Said power increase still happens as you spend science. The Scourge get a flat bonus to metal production every 3 planets worth of science. Do this enough time and, like AI mark, the Scourge will "mark up" and all its buildings will start Mark II. This means it gets the most expensive and strongest upgrade for free. The Scourge can "mark up" again to mark III if you play a high-enough AIP game. This gives it the 2nd most expensive and 2nd strongest upgrade. Watch out.
  • Overall the Scourge start up more slowly but scale up more compared to before, where it quickly reached a peak and stalled.
  • Killing builders is much more meaningful, destroying high-marks buildings is more meaningful, it's more clear on the galaxy maps that you're having an effect on the Scourge by killing them, as newly-built buildings stay at low marks much longer.
  • With the new intensity: Scourge first spawner can reach Mark VII at intensity 1/3/5/7/10 in 7h40/5h10/3h50/3h10/2h30. Those values are less extreme than the previous ones and more forgive, so that Rogue-like DLC2 options limits the number of scenarios unwinnable for most players.
  • All those changes affect all three Scourge: Ai-Allied, Friendly one, Minor-Faction one.
  • Thanks to ArnaudB for doing all this!

• Since the scourge now have fewer builders (which are more important), the scourge now dedicates units to escorting their builders.
  • Thanks to Badger for the awesome new fireteam mechanic!

Civilian Industries Updates

• Civilian Industries Update
  • Updated to support the new galaxy settings system.
  • Updated to support new pathfinding code.
  • Updated to support new hacking code.
  • Updated to use new tech names where applicable.
  • Added in some restrictions on Cargo Ship capacity.
    • Scales up based on trade stations, militia leaders, and intensity.
    • Previously there was no limit other than unfulfilled trade requests, meaning there could be some insane late game cargo ship counts.
  • Experimental unit stat update.
    • All regular entities spawns for a Civilian Industry faction will have increased hull and damage stats compared to their regular variants, but no shields.
    • This should hopefully help cut down on their unit counts late game, and make any damage they take mid fight more permanent, allowing them to be more easily sniped off whenever they attack something.
    • Turrets and Protectors are unchanged.
  • These changes combined may be a notable nerf to their power level, so please let StarKelp know if you feel like they need any complimentary buffs, or if they still deserve more nerfs.

Beta 2.743 Scourge Growth Chart

(Released February 19th, 2021)

• A number of updates to the text in the lobby and in the intel sidebar to make the advice on how to handle the Scourge more nuanced, particularly as it relates to higher intensity and/or AI difficulty combinations.
  • Thanks to ArnaudB for providing the insights and updates!

• Fixed the lobby still saying "Defensive Schematic Servers To Seed" rather than the TSS and ODSS variants.
  • Thanks to ArnaudB for noticing!

• Added lobby intensity description for Fallen Spire.
  • Thanks to ArnaudB for adding!

• Improved the efficiency of some of the tractor beam and tachyon beam code to a minor degree.
  • On planets where no player is looking, this effect is particularly pronounced now. So it might help some minor factions more than anything else.

• Fixed a bug where if you had tractor beams disabled visually, some of the functionality of them would actually be handled incorrectly.

• Updated the cloaked variant of the Mugger Frigate to follow the same dual-tech rules as its base version.
  • Thanks to CRCGamer for fixing.

• Fixed a bug in the drone-line removal checks that would remove a drone line and destroy all its members even if the centerpiece had a drone gun that would produce this kind.
  • This most notably lead to the Extended Ship Variants' Rescue Factory not working.
  • Thanks to NR SirLimbo for fixing.

• Starting Fleets received several buffs and corrections:
  • Classic trade the Fusion Bomber for the Heavy version, with a lightly oversized cap.
  • Doorkicker had its Siege Frigate cap raised from 4 to 6.
  • Consumer had its Fusion bomber cap raised from 30 to 40.
  • Sniper had its Tackle Drone Frigate cap raised from 4 to 6.
  • Undying had its Warbird Hydra Frigate cap lowered from 6 to 5.
  • Omniscient had its Apparition Frigate cap raised from 6 to 9.
  • Broadside had its Siege Frigate cap raised from 4 to 6.
  • Thanks to ArnaudB for finding and fixing.

• Other balance tweaks:
  • Heavy bomber default cap raised from 20 to 35
  • Parasitic bomber cap raised from 20 to 25
  • Thanks to ArnaudB for tuning.

• Scourge has been rebalanced to grow in strength a bit less exponentially than it used to, so that it still gets strong but is not an overwhelming death squad in the late game.
  • Made the Scourge builder and its cloaked version much tougher, as well.
  • Thanks to ArnaudB for the analysis and changes.

DLC2 Features

• Added is_reverse_tractor_beam="true" as two options on tractor beam systems.
  • This is was quite an interesting coding challenge, but essentially what it does is attach one of more ships to a target, and have them get pulled along with it, rather than them holding it in place. They do prevent if from leaving the current planet, even though they drag it around, though.
  • This needs a LOT more testing.
  • Thanks to zeus for requesting for some DLC2 content.

• If a ship is being tractored, it can no longer go through a wormhole unless the ship that is tractoring it drags it through.
  • Essentially if you have tractored an enemy ship and it is sitting on a wormhole, it can't escape your tractor by going through.
  • Or if you have reverse-tractored an enemy, it can't drag your attached ship through a wormhole.
  • But if an enemy has tractored you, and it wants to drag you off to another planet, that still works A-OK.

• Added use_alternate_rate_of_fire_after_x_shots, alternate_rate_of_fire, and use_alternate_rate_of_fire_for_x_shots_before_reverting.
  • If the alternate_rate_of_fire is specified, then the other two must be also specified to be at least 1 or more.
  • This allows for a new BURST FIRE mode.
  • More testing needed.
  • Thanks to zeus for requesting related to DLC2 ships.

Mod Updates

• Extended Ship Variants:
  • Increased the strength multiplier (not the actual combat strength) of the Sharpener Frigate from 1.5 to 2.5 and the Wedge Frigate from 1.5 to 2, as they had a lower displayed strength compared to the Shredder Frigate.
    • Thanks to tadrinth for reporting on Discord.

Beta 2.742 Hacking Balance

(Released February 17th, 2021)

• Add an xml option to have a weapons systems beam rotate over time, 'seconds_for_beam_to_rotate_fully'; if set to 360 then in 360 seconds the beam will sweep a full circle.

• Fix a bug where the Dyson Bastion was using the same model as the Dyson Bulwark.

• Put in yet more defensive code against a persistent cross-threading issue in the fleet factories-supporting code.
  • Thanks to Lictuel and Magedrifter for the most recent reports.

• Fixed a rare error that could happen in CheckForOwnershipChange() if the game was in just the wrong state when you exited to the main menu.
  • Thanks to NRSirLimbo for reporting.

• Fixed a bug where mobile support fleets were able to swap ship lines between fleets. Not sure how recent that is.
  • Also fixed a bug where the spire city lines could be swapped between fleets. Again not sure how recent that is.
  • Thanks to ArnaudB for reporting.

• When planets have mark levels swapped around because of that new randomization in mapgen, it now properly actually swaps the mark levels of all the ships and structures on the affected planets! This won't change existing savegames, but will keep it from happening in new ones.
  • Thanks to Democracy for reporting.

Revisions To Recent Beta Features

• Distribution nodes now cost 3 AIP to shoot rather than 1, and provide 35 HaP instead of 80. They now give 1200 science rather than 1000.
  • Thanks to ArnaudB for reporting how much they were skewing things.

• Superterminals are the one kind of hack that actually still destroys its target even when you cancel a hack. A superterminal hack is by definition a "hack until you can't take it anymore" sort of endeavor, so this not being there was quite unbalanced.
  • Thanks to Asirito and ArnaudB for reporting.

• hacking_does_not_increase_difficulty_of_future_hacks_against_this_faction has been removed from hacks, and replaced with hacking_points_against_this_faction_percentage_recorded_modifer
  • The ARS, TSS, and ODSS all now have a multiplier of 0.33 in this field.
  • Thanks to ArnaudB for suggesting.

• Fixed a typo in the prior build that was listing the ODSS as having four internal sectors and being hackable four times when it was really two.
  • Thanks to CRCGamer and ArnaudB for reporting.

• hacking_points_against_this_faction_percentage_recorded_modifer can actually be used to make hacks MORE noticed by the faction you are targeting, incidentally.
  • At this point, because of its extreme power and comparably low cost in HaP, we're having the superterminal hack actually now record twice the normal response rate.

• Direct battelstation science upgrades:
  • Were: 250,800,1800,2800,3600,5500
  • Now: 450,1400,2800,3800,4600,6500
  • Thanks to ArnaudB reluctantly suggesting this nerf to a thing he loves.

• Direct citadel science upgrades:
  • Were: 400,1100,2200,3300,4400,6500
  • Now: 500,1500,3000,4000,5000,7000
  • This value is now the same as a command station, for context.

• The "Neutral Planet Science Extraction" hack makes a return, as many people missed it. However, the balance is a bit different:
  • Previously, it gave you a discount in hacking point cost if you had already partly extracted the science from that world. It no longer does.
  • Previously, in multi-empire multiplayer games, it made you spend more hacking points for each empire beyond the first. It no longer does.
  • It used to take only 10 seconds (although more on higher difficulties), but now takes 60 seconds instead, making it one of the longer hacks around.
  • The response strength on completion is now 3x higher than before.
  • It now costs 30 HaP rather than 10 HaP.
  • Thanks to ArnaudB and others for suggesting.

• For hacks that require a local hacker, it now says what kinds they can be.
  • Hacks in general that require a local hacker can only be done by an officer or a transport, and the tooltip now says that.
  • We now have hacker_can_be_battlestation, hacker_can_be_support_fleet, and hacker_can_be_lone_wolf for hacks to make it so that we can use those if they are around.

• With hacks in general, it now tries to find them in the following order (which decreases the chance of you getting the annoying "you have multiple hackers and we can't tell who to use" message):
  • If it must be a battlestation, then first check selected citadels, then all citadels, then selected battlestations, then all battlestations.
  • If it can be something else:
    • If lone wolves are allowed, check them first (first selected, then any in general).
    • Then officers; then strike fleets.
    • Then if support fleets are allowed, check them.
    • Then if battlestations are allowed, check citadels, then battlestations/
  • For this whole process, it checks in this priority order JUST for selected ships, with no errors, and if you have one selected that's allowed, it will use that. Then it checks by group in selected, then all fashion.

• The following hacks can now be done by support fleets:
  • Weaken Turrets or Guard Posts
  • Hack GCA (if in old save)
  • Extract science from neutral world
  • TSS or ODSS hack for all fleets
  • Search for Spire Relic, Analyze Spire Debris, and Steal Macrophage Debris (DLC1)

• The following hacks can now be done by battlestations:
  • Weaken Turrets or Guard Posts
  • Hack GCA (if in old save)
  • Extract science from neutral world
  • TSS or ODSS hack for all fleets
  • Search for Spire Relic, Analyze Spire Debris, and Steal Macrophage Debris (DLC1)

• The following hacks can now be done by lone wolf fleets:
  • Spire Archive Extraction
  • Weaken Turrets or Guard Posts
  • Hack GCA (if in old save)
  • Extract science from neutral world
  • Subvert Super Terminal
  • TSS or ODSS hack for all fleets
  • Sabotage An Enemy
  • Reprogram An Enemy
  • Access Outguard Beacon and all the faction beacons
  • Provoke Vengeance Strike and Render Vengeance Generator Vulnerable
  • Dyson hacking (other than granting of ships)
  • Search for Spire Relic, Analyze Spire Debris, and Steal Macrophage Debris (DLC1)
  • Pretty much all the DLC2 hacks other than those that grant ship lines.

• Updated some code to allow support for the spirecraft fleet leaders from the fallen spire to act as hackers in general. This has never been possible before now, with any hack.
  • Thanks to ArnaudB for reporting.

Mod Updates

• AMU:
  • Updated the ExtendedHackingImplementation:
    • There no longer is a specific type for "this is done by battlestations or citadels". Instead this has been extended to literally EVERYTHING.
    • There now are specific filters for what unit type can hack: Transports, Arks, Fleet Golems (Golems that can have fleet ships), Lone Wolves, Mobile Support Factories, Battlestations, Citadels and others (in case there ever is).
      • These others can be defined in a rollup. It's now possible to make a hack that only a Battlestation, VWing, Metal Harvester, Spire Dreadnought or anything else can execute.
    • In case the hack requires a local hacker (even an abnormal one) this list of potential hacker types will be displayed in the tooltip if none are present on the planet.
    • This required a new approach to the duplicate hacking notifier, which now uses a ConcurrentBag to store hackers and uses them for making the notifiers.
      • This should remove any chance of cross-threading exceptions.
    • Hacks of this kind still need to be set up in XML as requiring no planetary hacker, thus the hacker initially will be Home Command. The special hack will simply transfer the hack onto the actual abnormal hacker at the first second.
      • It requires them to be non-crippled and non-destroyed, have no active hacks and prefers selected over all, though will use the closest hacker if more are present.
    • With this capability a lot of future hack ideas become possible...
  • Introduced new functions for faction filtering and caching:
    • Every SmartFactionImplementationBase now has 8 faction lists that are cached, and refreshed at most every game second, otherwise recycled, and if never used stay dormant.
      • These are AllHostileToMe, AllIAmHostileTo, AllFriendlyToMe, AllIAmFriendlyTo, AllLivingAIsHostileToMe, AllLivingPlayersHostileToMe, AllLivingAIsFriendlyToMe, AllLivingPlayersFriendlyToMe. Two more, AllLivingAIs and AllLivingPlayers are stored inside the GlobalImportantRollupHolder as these are universal for all factions.
        • These help to reduce CPU cost through constant recalculations, and even more in GC Churn. The overall effect will most likely be minor, but it's easy to do and costs very little memory, and most importantly: Makes the lists future-proof.
        • They support random draws, getting the first or null, average base marks, average power level and combined power level calculations, and potentially more in the future as needed.
        • They also replace the GetFirstLivingAIOrNull and similar functions, which are now available in the form of a very complex GetFactionListOrEmpty and similar functions within the AMU Core.
        • The filters include searching for living, dead or both, in combination with searching only for a specific type or all but a specific type, and all with a specific relation to a faction (or vice-versa), or all but this.
  • Updated the included source code.
• Kaizers Marauders:
  • Implemented all the above changes. Hacks should now work again, and better than ever.
  • Fixed a bug in displaying capturable ship or turret lines that could potentially occur when being a spectator in multiplayer.
  • Fixed a bug that potentially allowed for more Marauder Capital hacks than should be allowed.

Beta 2.741 Refining The Big One

(Released February 16th, 2021)

• This version breaks any savegames from the prior version, which is not something we usually do. However, the prior savegame balance was off enough that we don't want to test anymore with games started in that version (that will just be confusing data), and we are on the beta branch after all.

• Added Denial Fleet to the starting options, bringing the suppression frigate to the roster. As a visually impressive frigate, it was sorely lacking.
  • Thanks to ArnaudB for adding!

• There is a new Minefields category on the build sidebar, and all of the main-game mines are now in that. Any mod minefields will want to set build_sidebar_categories_i_am_part_of="Minefields".

• All of the remaining "Grant Ship Line" hacks have been renamed to be more thematic and clear, and the tutorial has also been updated to use the correct hack name.

• The annoying "Civilian" voice group is no longer used for any of the civilian-style structures. Instead those are now just silent.
  • This keeps spies, engineers, combat factories, battlestations, etc, from all being incredibly annoying.
  • This was meant to be a bit of an easter egg feature on a few civilian craft, but right now there are none using it anymore.
  • Thanks to NullZwei for reminding us.

• Decoy drones now use the annoyed civilian voice line. Other ships may in the future, but for now this is the only one. As has been pointed out, he's so annoying that the AI shoots him first.
  • Thanks to Puffin and others for suggesting.

• When there are members of a "relentless AI Wave" subfaction, if they are on a planet that is allied to them, and there are at least 3 strength of enemies to them at that planet, then they now fight at that planet before running off to attack somewhere else.
  • This applies for things like counterattacks, and hacks against the AI, etc. They now directly try to engage your forces at the planet where you are hacking them, unless your forces are pathetically weak. Previously, the challenge was fighting them before they could all run away to go destroy your home planet.
  • Thanks to AxiomExotic for the idea.

Revisions To Prior Beta Version

• Apparently the Warbird changes went missing amidst everyone's commits last patch.
  • So went with a somewhat different approach this time that is fairly similar overall but is somewhat less harsh overall and hits the truly OP parts more accurately.
  • Still gets a 15% hull and shield cut for base hull and all following variants.
  • Decloaks against the AI in 1 shots instead of 5.
  • Hydra heads of the Warbird have their health and shots cut in half since each is "half" a Warbird.
  • Undid the cap nerf to the Cloaked starting fleet since this time around the cap nerfs are only to the Warbird Hydra which is absolutely deserving of the nerf.
  • Additionally went over and applied the same balance logic to the Ambush Carrier.
  • Thanks to CRCGamer for implementing!

• Added a new setting for hacks: hacking_does_not_increase_difficulty_of_future_hacks_against_this_faction
  • This is now set to true for the ARS and DSS (but not the FRS or anything else).
  • When true, the overall response of a faction to being hacked does not increase from the hack in question.
  • Thanks to Democracy for reporting how crazy this made the late-game hacks.

• The DSS has been split into two structures: the Turret Schematic Server and the Other Defensive Schematic Server.
  • The hacking costs of both are higher than what the DSS was in the prior build, as things were coming too cheaply to players, making the game pretty trivial.
  • However, the TSS is substantially more expensive than the ODSS, which makes it so that the other items are actually viable to get cost-wise.
  • The ODSS can also only be hacked twice.
  • Thanks to a variety of players for feedback on this topic.

• Fixed a general bug in the game where the "did campaign start on at least this version" code was not working at all.
  • This made it so that things like showing the GCA basic tip was happening instead of showing the DSS/TSS tip.

• The defensive_structure_cap_multiplier was set rather high, it turns out. So we're adjusting these across the board:
  • For those that were 0.33, they are now 0.25. (That's basically just the economic command station).
  • For those that were 0.66, they are now 0.48. (That's just the logi command station).
  • 0.7 to 0.5 (shieldwall battlestation)
  • 1 to 0.7 (most battlestations)
  • 2 to 1.33 (most citadels and mil stations)
  • Thanks to a variety of players for griping about how plentiful turrets were, but special thanks to Ozone Grif. This was partly due to turrets being too cheap in HaP in general, and partly due to them being too prolific in the first place.

• The ship cap of all minefields has been doubled compared to what it was before (though, bear in mind, the other multipliers that are lower are still applied, so it comes out somewhere in the middle). We may need some further tuning to this.
  • Thanks to zeus and ArnaudB in particular on this.

• If players intentionally pick the same two starting battlestations, the game now makes one of them something else at random.
  • Thanks to CRCGamer for suggesting.

• Put in a fix that should cause planets that are "influenced" by a hostile faction (like those taken over by marauders) to still count as being hostile to you for purpose of hack costs.
  • Thanks to Eonfighter for reporting.

Bugfixes

• Fixed a cross-threading exception in removing invalid orders for ships.
  • Thanks to NullZwei for reporting.

• When warden fleet members can't find a way to regroup and thus join the hunter fleet, it no longer logs messages to your debug log. These were silent, but cluttered things up and were there just as a preventative/testing measure.

• Turned off some extra debugging that was left on when you were changing galaxy options during gameplay.

• Put in some extra protections against null entries in GetFromPoolOrCreate() for entity orders. This is again a cross-threading thing.
  • Thanks to NullZwei for reporting.

• A variety of our low-level pools now have prevention code in place for putting null data back into them. This could happen in certain cross-threading situations, and would yield random errors and memory usage problems.
  • Thanks to NullZwei for reporting.

• Our internal TimeBasedPool class now uses ConcurrentBag<> internally, instead of Queue<>.
  • This allows for more flexibility with cross-threading, including hopefully fewer/no errors as well as more speed ideally.
  • This is heavily used by GameCommands and EntityOrders, so having this perform really well on a variety of threads at once is a pretty big deal. It has been doing a pretty solid job for a few years now, but some of the newer hex-core-and-better AMD CPUs in particular are reliably causing many more cross-threading issues these days.
    • This is in no way a sleight against those AMD CPUs, incidentally; most likely it is happening because of their extra efficiency and speed allowing for subtle timing issues that were always there to come to light now that a new performance threshold has been hit. So we're just adapting to that.

• Our internal LoosePool class now uses ConcurrentBag<> internally, instead of List<>.
  • This has all the same benefits of what we just di to TimeBasedPool, although this is probably going to affect less since the LoosePool is not used across threads nearly so much. But better safe than sorry with it.

• Our internal ExternalizedPool class now uses ConcurrentBag<> internally, instead of Queue<>.
  • This has all the same benefits of what we just di to TimeBasedPool, although this is probably going to affect less since the LoosePool is not used across threads nearly so much. But better safe than sorry with it.

• Fixed the expected release date on the DLC2 expansion from saying January to saying March.
  • Thanks to Mac for reporting.

• Fixed some rare but possible bugs that could happen in ClearForReturningToPool().
  • Thanks to Kendrickorium for reporting.

• Fixed a bug where any hacks that were supposed to complete more quickly on non-hostile planets would say they would, but then would count into the negative seconds remaining and take the normal amount of time.
  • Thanks to Ryulong and ArnaudB for reporting.

Mod Updates

• Updated Histiocytes to support the new galaxy settings options.

• Adjusted the FRS version of the Artillery Cruiser in my More System Defenders mod to start at a higher mark level like the rest of the FRS options.

• Updated the Supercharged Raiders mod by ArnaudB to work with the new cap limits.

• Slotted the Laser Mine into the new proper sidebar build category. (More System Defenders)

Beta 2.740 The Big One

(Released February 15th, 2021)

New Hack! Transforming Fleet Leaders

• Added a new FleetLeaderType xml class, which lets us define certain types of fleet leaders which can transform between one another.
  • Not all fleet leader types need to be defined or use this at all. For instance, at the moment golems and arks are completely absent from this because they stay what they are.
  • But for transpors, battlestations, and citadels we want to be able to have them convert from one to another and back and forth with a new hack.
  • With this in mind, anything that is going to be able to use that sort of hack needs to have a fleet_leader_type="Transport" (for example, with transports).
    • Anything that has that type on it can be transformed AWAY to something else.
      • So if you are a modder and you designed a citadel that must stay what it is, don't assign this fleet_leader_type="Citadel" to it or else it can be changed into another type.
  • Anything that has this type and that is a target that we can transform TO needs to have a hacking_cost_for_other_fleet_leaders_of_same_type_to_become_me set to greater than zero.
    • For the things that we want to be able to transform FROM, but not TO (like there are actually seven or eight variants of basic transport), set this to 0. Out of the basic transports with identical stats, we only selected one to have a value greater than zero, so that you only see one entry for that type.
    • For all the other types that are unique targets, being able to swap over to those types can have a cost that is built in based on however unique and valuable that specific type is.
  • Note also that the costs are relative within a category. The citadels are relatively cheap to change types on (for the moment) because they are already more rare and you already paid a lot for them. But transports are in your face from the start of the game, and upgrading one can be a big deal.
  • Ah! This is also really useful for the support factories. So those now have this ability, as well.
  • And, surprisingly, the Spire Lost Frigate is also a great candidate for this. Did you know there are actually three different forms of it?
    • Now you can swap between the three forms for a really inexpensive amount of hacking points, which makes them all a lot more versatile!

• Added a new button in the fleet management window, under where you would give science upgrades to a fleet.
  • This button only applies for fleets with fleet leaders that can be transformed, and it shows information about the range of hacking point costs it might cost to do that.
  • Once it's clicked, you can see details about the types that this ship would become, and actually become that type if the ship isn't crippled.
  • Thanks to a variety of folks for suggesting this in many forms over the years! Fluffiest, Ecthelon, CRCGamer, and definitely some others.

• If a fleet leader transforms types, and that fleet leader's fleet has been upgraded, AND the amount that it costs to upgrade the old type of the fleet leader is different from the new type of the fleet leader (whew), then it will revert the upgrades on that fleet to 0 and refund you the science.
  • This prevents you from having a basic transport upgraded more cheaply, and then swapping it to the more-expensive-to-upgrade Cloaked Transport type and getting the higher mark level for free.
  • Thanks to NR SirLimbo for warning of this cheese.

ARS Expansion And Overhaul

• The reroll hack for ARSes has been removed. That sort of blind gambling for something good is something we want to avoid.

• The base hacking cost of ARSes has moved from 7 to 4 (double on hostile worlds).
  • A single ARS can be hacked 3 times, now, rather than just once. The price increases by 1.5x each time that specific ARS is hacked.
  • At each stage, an ARS now gives 4 strikecraft and 2 frigate options rather than 4 strikecraft and 1 frigate option, and does a complete reroll of its contents after you make a choice.
  • This gives you more choices, but not EVERY choice, and once you have chosen the thing that best fits with your empire, you get a new slate of options.

• Added a new response_strength_multiplier_per_time_hacked_same_target. We are now using this on the ARS to make it so that, as the cost increases for each of the 3 attacks you make on a single target, so too does the response in terms of enemies coming out of the structure.

• Actually, the hacking intensity increase of the ARS is now 2.2x rather than 1.5x, which means that on the third time hacking a single ARS the enemy forces are truly terrifying. You definitely have to want it, when it comes to that third hack. It's possible that you will fail the hack because your hacker gets crippled, and you'll lose that last chance to get a new ship line. But it doesn't charge you hacking points in that case, it just destroys the ARS, so it's a disappointment but not save-scum worthy.

Intra-Galactic Coordinators Removed

• Intra-Galactic Coordinators have been removed from the game (though they remain in existing savegames).
  • These were funky and a bit fiddly, and there are now better ways of accomplishing more or less the same thing.
  • These were either going to be wildly unbalanced (turrets and frigates), or extremely redundant and not worth it (strikecraft).
  • It's nice to have a bit less chaff out there to sort through, particularly one with a confusing name, as your deciding how to approach your empire.

Global Command Augmenter Out, Defense Schematic Server In

• The Global Command Augmenter (GCA) has been sunsetted for new games. It's still in existing campaigns, but new campaigns no longer seed them and likely never will again. We have a better replacement!

• The new Defense Schematic Server (DSS) has been added for the game, and is what seeds in place of GCAs in all new campaigns. It has the same icon and graphics, and uses basically the same seeding rules. However, these are owned by the AI (like an ARS) rather than being un-claimed neutral entities. These cannot be claimed, unlike a GCA.
  • The DSS would like to give you 6 turret lines to choose from when you hack it, and 6 "other defenses" lines. However, it will not give you more than 40% of the total options available to you based on the mods and DLC you have installed in the game for the turret category, and not more than 25% of those available in the other defenses category.
  • Unlike tha GCA, it also tries very hard to give you individual basic lines of turrets and defenses, rather than so commonly doubling up a line and giving you a 2x or 3x cap for a line. However, if there are very few turrets or defense types in your game, there is a chance that you might get lucky and get only a few options including some doubled ones.
  • Also unlike the GCA, the hacks from the GCA are NOT shared among all of the empires in multiplayer. The new lines are locked to the empire that hacked for them, and can never be traded or refunded or altered in much of any way.
    • There are, however, far more options for how to granularly hack for each line of defensive structures, and they provide an even broader amount of power per empire, so there's still plenty to go around.

• The "Global Command Augmentation Mulitplier" has been renamed to "Defensive Structure Cap Mulitplier."
  • The old commandstation_addedtocommandstation_multiplier xml tag has been removed, and now is called defensive_structure_cap_multiplier.
    • Note! If you are a modder and you are creating new battlestations or citadels, normally this defaults to 1. If you want your structure to grant more or fewer turrets/etc, then you can set this to something like 0.5 to make it give half as much or 1.75 to give 175% as many, etc. You can ALSO set it to zero or less, and your battelstation or citadel will get no bonuses whatsoever.
  • Refresher on amounts for various structures:
    • 2x on home command station, that seems fine.
    • 0.5x on economic command station, that still seems high. Let's take that down to 0.33x.
    • 1x on logistical command station, which again seems high. Given the insane usefulness of logi stations in general, let's set this at 0.66 instead.
    • 2x on military command station, that seems fine.
    • All of the battlestations are defaulting to 1x, and that's a good start.
      • For the shield wall battlestation, let's have 0.75 to balance that one out a bit.
    • For all the citadels, let's set this to 2x for now. Maybe some more nuance with some of them will be good at some point.
  • Thanks to tadrinth for the idea of applying this to things other than just command stations.

• There is a second version of the DSS hack that is slightly cheaper in hacking points and which gives a triple-powerful amount of the chosen defensive line to just the hacker (which must be battlestation or citadel).
  • This is probably usually the inferior option to pick... except wow, this can be pretty crazy with citadels.
  • Thanks to tadrinth for the idea.

• Added is_blocked_from_dss_grant_to_battelstations_and_citadels and is_blocked_from_dss_grant_to_commandstations, which, when set to true on a ship/structure type, will make it so that a DSS grant will not include them in the respective group. That way the DSS can provide some things for command stations only, or for battlestations/citadels only.
  • This will hopefully be used very sparingly, but it's handy in general as an option for certain high-value things.
  • When this is in use, it should show a message on the ship line that says where it won't be applied, right up at the top. This is not well tested yet.
  • Thanks to zeus for suggesting.

Turret And Defenses

• The turret counts on the starting battlestations are now a middle-ground between what they were previously and what they were last build.
  • Thanks to various players for giving input and suggestions on this, and to ArnaudB for making the changes.

• The starting defensive fleets from the More Starting Options mod have been updated to also follow the new guidelines.
  • Thanks to CRCGamer for implementing that.

Battlestations and Citadels

• The speed of citadels has been increased from 300 to 600.
  • The "Speedy Citadels" save-safe mod has now been removed, since this is what it did.
  • Thanks to ArnaudB in particular, but also others, for suggesting.

• The More Starting Options mod has been integrated into the main game itself (mostly as part of DLC1, but it varies).
  • Huge thanks to ArnaudB for creating this mod, and then also for allowing us to integrate this for everyone.

• Added a new XmlModNamesToIgnoreNow table, which lets us basically ignore certain mods that used to exist but now are part of the game (or for other reasons).
  • This applies now to the "More Starting Options" mod, so that any games started with that mod on will still work, since it was not a save-safe mod (but still has content that exists in the game itself, so is safe-safe in that NEW way).

• Players now get two battlestations at the start of the game, rather than just one!
  • The battlestations now default to Random, but the combat and support fleets continue to default to some basic easy types.
  • When at least one of your battlestations is set to random, then it ensures that the other one will not be set to a duplicate type of it.

• Citadels now all cost 60 AIP to claim. These things are MAJOR power boosts for you, and being able to potentially even stack them on a planet is incredibly potent.
  • The cost of these should all always be the same to claim, since once you have one you can switch it to any other type.
  • 60 AIP may be judged to be too expensive in the end, but this is something we want to have as a rather extreme defensive option that gives you such high defensive bonuses that the AI just has to take notice. These things are in mamy ways more powerful than any Golem now.
  • Note that when viewing a citadel for capture, it will include all of the turrets and other defenses that are relevant based on your past DSS hacks.
    • Worth noting that in multiplayer, this will look different for each player, which is fine. Gifting a citadel between player empires in multiplayer will also change the loadouts based on that.

• The game no longer seeds any battlestations at all for you to capture. You start with two now right off the bat, and because of the new DSS mechanics they are each far more powerful.
  • If there's a need for more of the sort of thing that battlestations provide, then you should look for a citadel to fill that role.

Other Hacking Updates

• The way that rerolls work is now wholly deterministic, meaning that you can't keep reloading and trying again in order to get a new value.

• When viewing the ARS tooltip for the ship, it now shows you how many hacks you have done on it out of how many times you can do it.
  • A bunch of other sidebar bits in the hacking menu also now show this information clearly and in the proper places.

• In the event that you cancel a hack, it no longer destroys the target (or counts as a full hack attempt).
  • So for instance, if you decide to hack an ARS for one ship, and then change your mind, you can cancel the hack and just try again from the same list.

• In the event that you fail a hack, it does the "counts as a hack attempt" full logic. So if you're about to fail, canceling will save you from that.

• A new feature under the hood lets us require that a hacker be a battelstation (that also includes citadels).

• Another new feature under the hood allows us to count the "number of completed hacks" against a target as being the number from any type of hack, not just a single hack type.
  • This is useful when there are multiple hacks that do a similar thing, and you want either option to count against a total that both have.

• A bunch of other little hacking sub-features that you can see on the new hacks, frankly.

Hacking Time/Intensity

• The multiplier_to_global_hacking_times has been removed from AI difficulty.
  • This made hacks take longer on higher difficulties, giving the AI response more time to hit you.
  • This actually is interesting as well because it did not affect something like the superterminal at all, since that's an at-will hack that lasts as long as you want.
  • Overall this could lead to tedium with hacks on high difficulties, and the sole purpose was for extra difficulty.
  • Previously, for difficulties 5 and down it took 1x the normal amount of time, at difficulty 7 it was 1.5x, 8 1.6x, and 10 was 2x.

• In place of that mechanic, we now have multiplier_to_global_hacking_event_intervals on the AI difficulty.
  • The overall length of time it takes to do a hack (say, 30 seconds) now stays the same. But the interval on which "things happen to you" is now multiplied by this number.
  • So, instead of having a longer, more drawn-out period on a high difficulty to get attacked by more enemies in response to your hack, you just get a (potentially much) more vigorous reponse in the original timespan.
  • We are actually making difficulties 1-4 easier by making difficulty 1 have an interval multiplier of 2, and then on down to a mult of 1 for diff 5.
    • Diff 6 is 0.8 now, diff 7 is 0.7, diff 8 is 0.6, diff 9 is 0.5, and diff 10 is 0.4. That's going to be nightmarishly hard, but hopefully survivable. We can tune as needed, but at any rate folks shouldn't be bored waiting for a hack to finish!
  • Thanks to ArnaudB for suggesting that the timeframes not be increased on hacks on higher difficulties, since that does get boring.

• Note to modders: if you were previously coding some hacks and directly used PrimaryHackResponseInterval, SecondaryHackResponseInterval, TertiaryHackResponseInterval, or BaseHackDuration, you now have to use the getter methods instead, which handle all this logic for you.

• If you have gone to the trouble of capturing an AI world (or at least it no longer belongs to an enemy), then some hacks will also go faster now on that world, which also means that the overall AI response from them are substantially weaker.
  • The ARS and FRS hacks are the first to have this, and just like they cost half as much HaP to do on non-hostile worlds, they now also go twice as fast. Getting a third hack out of an ARS is a lot easier on a world not controlled by the enemy, since they intensity of the ships coming out of the ARS won't have nearly as much time to ramp up.

Basic Transports

• The Agile Transport now has 85% of the health of a normal transport, rather than 50% of the health of a normal transport.
  • Thanks to CRCGamer for suggesting.

• The ability to build custom fleets from the build sidebar, which was something very few people did (and it was generally for purposes of cheese), has been removed.

• For regular transports (the most boring kind), if they sit around on a planet for 5 minutes without being crippled or leaving, they will start giving a 10% bonus (that stacks!) to any player-based metal and energy production happening at that planet.
  • For the awesome cloaked transports, they do not get this bonus at all.
  • For the agile transports, they provide only a 4% bonus.
  • Why have this? Well, sometimes you have extra transports sitting around, based on how you've organized your fleet, and it seems like they ought to be good for something. Rather than being an annoyance you want to delete, what if they were instead a commodity well worth protecting? So here we are.
  • Note that none of the officer fleets, or the combat factories or whatever else, provide this sort of bonus. They already have things to do. But this is now a mechanic that can be used in new ways in the future if we want to.
  • Thanks to Badger for inspiring this idea.

• New ship: High Capacity Transport Flagship
  • Upgraded version of the basic transport which can double the ship cap of up to three strikecraft lines contained within it.
  • This cannot be found anywhere by chance, but for 60 HaP you can switch one of your transports over to this type at any point, even right at the start of the game.
  • Why take time out to add this in right at this particular juncture? Well, the FRS mechanics already encourage having some smaller fleets, but the bonuses from idle basic transports encourage giant blobs. This is another argument in favor of small fleets.
    • Honestly we're fine with however people want to play, so long as there isn't one overwhelmingly obvious "best" way to play. At the moment, the best choice is probably an interesting mix of large and small fleets for different purposes.
  • Thanks to zeus for suggesting this new ability and new transport type!

UI And QoL Improvements

• Added a new Sidebar subsection to the Display section of settings.

• Added a new setting to that subsection (defaults to off): Build Menu By Unit Type Before Fleet Type
  • Normally the game sorts the build sidebar by fleet type, then fleet, then by unit type. Enable this if you would prefer it sort by unit type first (aka grouping all turrets of a specific sort together).
  • Note: this is flipping the old default behavior, which was unit-type first.
  • Thanks to tadrinth for suggesting.

• It is now possible to select "Random" for all of the starting fleet types (combat, battlestation, and support).
  • Thanks to cml and others for suggesting.

• Fixed a minor issue where on player home planets it was likely for a few types of the structures to be seeded on top of one another. They now try hard not to do that.

• Ships that produce metal or energy now properly include whatever bonuses they have from their local planet in what they show in their tooltip. Previously, that was not shown for any of the various kinds of bonuses that can cause this.

• The way that the fleet-wide bonuses are calculated is now based on JUST the ship line counts that are non-flagship style.
  • This is more what people were expecting anyhow, and with the ability to switch flagship types we were running into some trouble.
  • With that in mind, all of the FRS ships have had their lines lowered by 1

• Made a couple of minor adjustments that makes it so that ships move a bit more directly between positions, potentially when they are coming to a stop (easing into place less, and more just doing a hard stop).
  • There's a chance that on certain machines this may cause a bit of visual jitter as ships move. If that happens, we'll walk this back or make it optional. But for now this seems to make ship movement on 1x speed more smooth. If you play on a simulation speed of 1.5x or 2x or higher, the movement is incredibly smooth, and we'd like that for folks that play on the slower (aka normal) speeds.

• Hacking a superterminal previously just filled the chat log with spam of the AIP reductions. We now are only reducing AIP every 3 seconds instead, and thus having 1/3 as many messages.
  • We also are now putting a timestamp next to every AIP change message in general, so that when there are multiple in a long cycling run like this you can tell something is happening.

• The hacking section of the 'how to play' section has gotten a major overhaul, with a lot of more up to date information in there (from the new beta builds) as well as with a big lore dump explanation on what hacking points even are.
  • Thanks to Leif for asking about this.

• The vicious raider AI type has been updated to note its wave budget bonus in its description.
  • Thanks to ArnaudB for inspiring this change.

• Various player tips and journal entries that talk about GCAs still work for older savegames, while newer savegames it advises about DSSes instead.

• A few places have recently referred to the player as "Admiral." These have been corrected to now refer to the player as "Commander," as is the case in all prior direct addresses.

• Updated the "about science points" section of the retrieve science screen to no longer refer to the removed ability to hack to gain science from neutral planets.

Balance Adjustments

• Players now start with 100 hacking points, rather than zero. This is for much the same reason that they start with 15,000 science: let's get the game started, and let them make some interesting decisions right away.

• Revised Warbird Frigate balance:
  • Roughly 15% hull and shields cut. Also carries over to the Ripper and Hydra variant of the Warbird.
  • Paralysis duration on shots lowered from 4 to 3 seconds.
  • Shots per volley reduced from 16 to 10 but gains 1 extra shot per mark level to the full 16 again at mark 7.
  • Additionally ship cap reduced from 6 to 5 in most fleets and then additionally adjusted the cap from 6 to 5 in the cloaked starting fleet as well.
  • Thanks to Mac for raising the issue of how OP these are, and to CRCGamer for crunching the numbers and making the changes.

• Outguard updates
  • Updated Outguard to be "Expatriate" -- changed preexisting hull tech accordingly.
  • Also, buff to "outguard (merc) flagship stats, as this was always a mediocre flagship at best.
  • Thanks to zeus for updating for us!

Bugfixes

• Fixed a general bug where if flagships ever transformed, they would form a whole new fleet instead.

• Maybe made the Miffed and Angry achievements less likely to trip erroneously.
  • Thanks to Mac for reporting.

• Fixed weapons with restrictions based on whether or not the target must be mobile or must be static (i.e. the Botnet Golem) being able to circumvent the attack restrictions if the player orders so.
  • Thanks to NR SirLimbo for finding and fixing!

• Put in 2 more restrictions for weapons to ignore targets:
  • If the target is invulnerable by any means, skip it.
  • If any damage modifier has a multiplier of 0 (artificially used to NOT make something attackable) and that multiplier applies, skip it.
  • Thanks to NR SirLimbo for finding and fixing!

• Fixed a bug in the weapon attack-ignore code that was leading to not attacking any unit as long as any other unit was providing invulnerability, even though the first unit should've been vulnerable already (i.e. with AI Eyes).
  • Thanks to NR SirLimbo for finding and fixing!

• Fixed an issue where if you had multiple local IP addresses and view the networking tab, it could cause an endless string of errors.
  • Thanks to athros for reporting.

• Several static working lists on the Fleet object have been made ThreadStatic, so that if they are used by multiple threads they will not clash or cause any problems.
  • We had several reports of strange errors, thanks to Leif and Rifi for reporting.

• Fixed a bug where player melee ships were often going back to the same point over and over again when there were no enemies present, leading to what seemed like an instant dance.
  • This will mean that melee ships get caught out of position more after a fight, but that's fine.

• Fixed a bug in the prior version that was causing the Subvert Superterminal hack to show up as blank, and then throw errors if you hovered over it.
  • Thanks to Edicitsep and poljik2 for reporting.

• Fixed the reprogram hack cost of OMDs to be 24 instead of 5.
  • Thanks to tadrinth and CRCGamer for reporting.

• Fixed an issue in the hacking log where it was not showing the ship type you had hacked for properly.

• Fixed an issue where certain hacking errors just silently went to the log rather than popping up and showing you the error visibly.

• Removed the reinforcement point tag from the AI Plasma Eye to match the AI Ion Eye. The reasoning behind eyes not being reinforcement points still applies because of garrisoned units disappearing if still attached when the eye swaps to an alerted version.
  • Thanks to CRCGamer for catching and fixing!

Mod Updates

• AMU:
  • Fixed Drones from the AMU_DroneBase... Not actually being drones.
    • As they now count as Strike Craft adjusted the health and damage values from the Drones within Extended Ship Variants to compensate.
  • Fixed and ajdusted all manner of things related to hacking that broke through the recent changes in Vanilla AIW2.
  • Hacks that are done by Battlestations or Citadels will now, if but one is selected, honor that selection or if none are selected will use the closest one as the actual hacker again.
  • Hacks that are done by Battlestations or Citadels will now display the current actual hacker if any are found.
  • Hack Text Snippets, as they rarely change, have been made much more efficient for the Garbage Collector, hopefully improving performance a bit.
  • Shifted the additional shot speeds from Kaizers Marauders to AMU, and added one more EXTREMELY slow shot speed type.

• Kaizers Marauders:
  • Hopefully put in a fix for an error in GetDangerOfPathAndTargetIgnoreNegligible(). If not, next time it appears there will be more debug data.
    • Thanks to Leif for reporting.
  • Fixed the broken hacks.

• Micro Mod Collection:
  • Tripled the amount of Hacking Points in Distribution Nodes in light of the recent HaP "inflation" as Vanilla DNs now grant much more too, added to that the starter Hacking Points.

• Extended Ship Variants:
  • As the Vanguard is now a vastly better ship due to damage amplification the Vex Guard was a bit overpowered. Reduced its damage amplification to 5 (down from inherited 30).
  • All the transport and mobile factory variants can now be transformed into each other.
  • As the Agile variant has been buffed in Vanilla in return the radar dampening effect that ESV granted to it has been removed. It does retain the speed buff on entering a new planet, and the 21gx engine power.
  • Reduced the strength (and thus threat level) of the Target Painter variant by 20% as enemies were targeting it a bit too often.
  • Shortened the name of the Vanilla Rejuvinator support starter fleet so it doesn't have to break line any more once (due to the multitude of ESV support starter fleets) the scroll bar appears.

• Extended Ship Variants:
  • Implemented the "resource-bonus-if-idle" mechanic for the modded transport centerpieces:
    • The Engineering Transport has an improved bonus (15%) and gets it sooner (after 4 minutes).
    • The Tugboat Transport only gets the same 4% bonus that the Agile Transport has.
    • All other transports have no bonus.

• Extended Ship Variants:
  • Converted the Tugboat Drones into Drone Projectiles and made the Tugboat Transport Flagship launch them.
    • This was necessary due to the swap-centerpiece mechanic, as it was causing the original fleet to get "lost" without a flagship, literally costing the player the entire fleet.
    • This is only a temporary solution, as this "cheats" the player out of the buff when no Tugboat Drones have been present since loading the game. 90% of the time that won't be a problem, but it still is inconsistent.

• AMU + Extended Ship Variants:
  • Created a simple Executor Fake Faction-script in AMU that only activates itself if Extended Ship Variants is installed.
  • It will scan through all player-owned fleets with a Tugboat Transport as flagship and, if it doesn't have a fleet membership group for the Tugboat Drones, will add one.
  • Thus, at worst 1 second after unpausing the Tugboat Transports will gain their previous speed boost of 700 for all ships in it.
  • Inversely the script will remove the fleet membership if no more Tugboat Drones are alive AND the flagship is any other than a Tugboat Transport, thus also removing the speed buff.

• Updated AIShieldGenerators mod to work with the hacking time/intensity changes in Beta 2.735

• More System Defenders 1.22 - FRS Revision & Ark Hacks (by CRCGamer)
  • Most recent changes:
  • Revisions to the Augmented Artillery Cruiser available in FRS
    • Cap reduced from 3 to 2
    • Shield bonus boosted from 1.25x to 1.5x
    • Boost only works with 5 lines or less
    • AIP cost reduced from 30 to 10
  • Stormfront Ark given the same self-hacks for shield boosts as various vanilla ark choices

Beta 2.734 Bugfixes

(Released February 12th, 2021)

• The AI Shield Generators mod by cml has been updated to work with the new hacking mechanics.
  • Thanks, cml!

• The way the fleet bonus limits are written is now more brief, and if they are hit the entire fleet bonus text is shown in a different color. It's way easier to see what's going on at a glance.

• Fixed a bug in the prior version that was causing fleet-wide limits to sometimes not work because it thought it had more ship lines than it really did.
  • This may not have affected any players, but it affects the player potluck cheat, and could in theory affect some real fleet someday.

• Fixed a bug with the fleet-wide limiters in the prior beta build, where it would only turn on if you had exactly the number you were supposed to or MORE.
  • This was backwards from what the UI actually told you, for extra confusion.

• Fixed a bug with some of the caching of hull and shield values (for performance reasons) that often caused various things like FRS ships and hull/shield hacks to not apply properly to ships in various circumstances.
  • Thanks to CRCGamer for reporting the hack-related side of these.

• The FRS fleet-wide bonus for move speed does NOT apply to the flagship, and that was both working correctly and also stated correctly on the interface.
  • However, the other fleet-wide bonuses (shields, hull, attack power) DO apply to the flagship, and the interface was reporting that incorrectly. On those it now explicitly says that they work with flagships.
  • Thanks to Jordan K for reporting the inconsistency.

• Turbo Raiders now work on any fleet with 6 or fewer ship lines, since flagships are counted in the total line count, but this is the one kind of bonus that doesn't actually help flagships.
  • Thanks to ArnaudB for inspiring this change.

• Fixed an issue with the EarlyOfficers where we were seeding one per human empire rather than 1 overall.
  • Also fixed a bug where extra officer fleets in general (not just the early sort) could seed because of some random noise from things above them. On a standard single-player game, you should have about 4 on the map.
  • Thanks to CRCGamer and ArnaudB for reporting.

• Fixed a bug that could happen when you did a hack from a pop-out window, and then exited to the main menu and re-entered to do any other hack from a popout menu.
  • Thanks to CRCGamer for reporting.

• Fixed an issue in the last couple of versions that could lead to an infinite loop of "multiple hackers present" messages when you opened the hacking sub--menu for certain hacks.
  • Thanks to Strategic Sage, Firewoven, and Leif for reporting.

• The starting defensive fleets were indeed a bit over-tuned and have now been nerfed.
  • In the starting fleet, for instance, you mostly got about 60 of things that you got 30 of in the actual command station, and 30 of things you would get 15 or 16 of elsewhere. We've toned this down to half values.
    • Similarly, you had 30 tractor beams rather than something more normal like 15.
  • In the fleets with minefields, we also cut those numbers in half, because 70 was way too many -- 45 seems better in most cases.
  • Thanks to Daw11 for reporting.

• Fixed another issue that was causing turret counts to be drawn 5x too high when viewed in places like the lobby and looking at a starting fleet. Essentially some code that was in several places needed to be moved to a central consistent place.
  • Thanks to ArnaudB and Daw11 for reporting.

• One last fix to misleading tooltips, with GCAs reporting about 5x more turrets than they were actually going to grant you. Now they actually give you the proper numbers on things.
  • Thanks to CRCGamer and Lictuel for reporting.

Beta 2.733 Arks, Reprogramming, And FRS

(Released February 11th, 2021)

• In general, passive attrition from ships is now boosted by boost hacks against their ship, as well as from fleet-wide bonuses.
  • Previously, neither of those boosted attrition.

• Attrition is now better color-coded in tooltips, and it now only shows the full explanation if you are on the full tooltip. Instead it just shows a more concise "ATTRITIONER: Amount/s" for lower-detail tooltips.

• Fixed a minor issue in tooltips where the DPS per weapon was not shown with commas, making large numbers harder to read than they should have been.

• Updated the scouting tutorial to mention spy units and how those work, since that's relevant.
  • Thanks to Puffin for reminding us.

• If the mark level of a drone would naturally drop to be lower than the centerpiece of its fleet, then it now marks up to match the centerpiece of its fleet.
  • This is similar to how it worked in the past.

• Under the hood, there is a new CustomBaseMark on units, which can be set to a value at some time. Their mark level will then not drop below that.
  • For units that are reprogrammed, we now set this, so that you get the mark level of whatever the unit was when you captured it (it can still go higher based on techs, but it won't go lower).
  • Additionally, for ships fired from a drone-gun, they will now use this for the mark level of the firing ship (for example, makeshift turrets). This keeps behavior how it used to be prior to the new tech respec code.
  • Thanks to DEMOCRACY_DEMOCRACY for reporting.

• The achievement "FOR SCIENCE!" used to be related to hacking ARS or TVs to convert them into science, but that's no longer a thing.
  • Now it is: "Retrieve spent science so that you can re-spec part of your fleet in response to a changing situation."

• On the galaxy map filters, a few quality of life adjustments:
  • Normal is now shown with a golden text color, and is at the very top of the list, so you can easily scroll to it.
  • Deepstrike Danger and Threat are both shown in purple, as they are both unusually useful as well as being about dangers.
  • Strength and Spy Network are both shown in blue, as they are also unusually useful and about reading the topography of the map.
  • Capturables are shown in light green, as they are very useful and one of the things that is most goal-oriented.
  • Thanks to gaz and themouthofsauron for suggestions in this area.

• Fixed some exceptions that could happen because of cross-threading in EmitOnAOEParticles.
  • Thanks to CRCGamer for reporting.

• Silenced a couple of knockback errors that were harmless but that could happen in particular in multiplayer (but also in single-player) if things died in just the wrong order.
  • Thanks to Tydorius for reporting.

Hacking Updates

• The "Double Ship Line Cap" hack now only works on "basic strikecraft" ship lines, which means nothing that starts at a mark level higher than 1. The tooltip explains this, now.

• The sabotage hack now gets 1.25x more expensive to use for each mark level above 4 of a target.
  • So if something costs 10 to hack at mark 4 and down, at mark 5 it will cost 12.5. At mark 6 it will cost 15.6, and at mark 7 it will cost 19.5.
  • This keeps the bulk of the galaxy in one similar cost spectrum, but then for the really intense planets does make those more costly to do.

• In multiplayer, hacks that increase in cost for every attempt now use the hacking history of all players, not just whoever is doing the hack.
  • Also new? If you failed the hack, it now doesn't count that against you, even in single-player.
  • Thanks to Arides for reporting.

• The maximum amount of time a subvert-superterminal hack can last is now capped at 20 in-game minutes.
  • It seems like it was maybe capped at 3 in-game minutes before, but it's hard to tell from the code. It was supposed to be indefinite.
  • Our goal was for people to be able to push it for as long as they can hold out, though, so unless this winds up being unbalanced this might be better. Feedback welcome.

• Fully fixed it up so that now if you are unable to do a hack in general, but it has a pop-out for sub-options, you can still get the sub-options and see their details while at the same time seeing why the hack is invalid on each one.
  • This is most common when you don't actually have a hacker present, but this makes it so that you can easily peruse distant planets and look at the various options inside for instance an ARS or whatever.

• Fixed a number of places where the number of ships you will be getting was misrepresented -- either too high or too low.
  • There was an ApplyFleetCapMultiplier() that has been removed, as it was redundant with the newer GetResultingCapFromSquadCapMultiplierList().
  • In places where both were used, it was reporting too high. In places where just the former, it was showing too low and now uses the latter.

• Both the FRS and ARS now show the total strength of the lines that you would be getting in their options area.
  • They also show the mark level as a roman numeral and a color and not just a color alone. This is a lot more clear at a glance for this screen.

• A large number of hacks just said "Grant Ship Line," which was vague and annoying, particularly when there were a lot of structures of the same sort on a planet.
  • The FRS now says "Steal Experimental Ship Line" and has a much more specific and informative description.
  • The ARS now says "Steal AI Ship Line" and has more lore flavor in its description.
  • Others will be coming soon in terms of having better names and descriptions.

New Hacks

• Two new hacks!
  • Increase Shields
    • Increase the shield durability for one of your units at this planet (works for both personal shields an bubble forcefields).
    • In terms of the interface and so on, works just like Increase Hull Health.
  • Increase Weapon Power
    • Increase the attack power for all weapons on one of your units at this planet.
    • In terms of the interface and so on, works just like Increase Hull Health.

• Most (but not all) Arks now have the Increase Shields hack, similar to how all the golems that players can capture have the Increase Hull Durability hack.
  • Rorqual Hegira: cost 40 HaP, can do 1 time, gives 1.5x boost to the bubble forcefield.
  • Gyrn, the Voidhome: cost 20 HaP, can do 2 times at increasing cost, gives 2x boost to personal shields.
  • Orchid: cost 20 HaP, can do 2 times at increasing cost, gives 2x boost to personal shields.
  • Ark One: cost 20 HaP, can do 2 times at increasing cost, gives 2x boost to personal shields.
  • Great A'Thomek (DLC1): cost 20 HaP, can do 2 times at increasing cost, gives 2x boost to personal shields.
  • Sol Ater (DLC1): cost 20 HaP, can do 2 times at increasing cost, gives 2x boost to personal shields.

• The remaining Arks now have the Increase Weapon Power hack, since that's better suited for them.
  • Thanatos: cost 24 HaP, can do 2 times at increasing cost, gives 2x boost to attack from all weapons.
  • Belle Prime (DLC1): cost 24 HaP, can do 2 times at increasing cost, gives 2x boost to attack from all weapons.
  • Nodorian Tortoid (DLC1): cost 24 HaP, can do 2 times at increasing cost, gives 2x boost to attack from all weapons AND the passive attrition damage.
  • Grand Salvage (DLC1): cost 24 HaP, can do 2 times at increasing cost, gives 2x boost to attack from all weapons.

• New hack! Reprogram An Enemy
  • Hack into and turn one of a few enemy structures on this planet to become yours! Not many enemy units are vulnerable to capture in this way, but there are a few. If the structure would normally increase AI Progress on death, capturing it this way does not cause that increase. Hack runs very quickly, due to the likelihood that the target is trying to kill your hacker.
  • This works much like the sabotage hack in terms of how you select targets and all that sort of thing, but instead of destroying them you get to keep them.
  • Both ion cannons and mass drivers are the first ones to now support this.
    • They now have proper costs that rise as their mark levels go up, and you can hack even mark 7 ones, unlike before (it's just bloody expensive).

• AI Fortresses and Superfortresses can now both be reprogrammed (and the latter can also be sabotaged).
  • However, they are ludicrously expensive to do that with, so it may not be something you really do much.
  • These may need some metal/energy cost balance. Feedback welcome.

• Magnifiers can also now be reprogrammed, for a more middling cost.
  • So can Black Hole Machines, for an even more reasonable cost.
  • These may need some metal/energy cost balance. Feedback welcome.

Removed Hacks

• The generically-named "Hack Ion Cannon" and "Hack Mass Driver" hacks have been retired.
  • They both took over all of the ships of a certain sort of their respective planets, but only up to mark 6 planets, and the whole thing was a bit janky. Let's make the whole thing more robust and able to be used with other kinds of units, too.
  • See above for the better replacement version.

• The hack to convert an ARS to science has been removed.
  • This is another of those cases of the battle between what you design and what you "play it as you find it."

• The "Neutral Planet Science Extraction" hack has been removed.
  • The only reason to use this was to avoid aggro'ing neighboring enemies. But... well, please do meet the neighbors if you want to mine the planet for science and hacking points!

• The "Covert Science Extraction" hack has also been removed.
  • This is something that was just a straight up trade of hacking points for science points, while avoiding an AIP increase.
  • This is a bad combo with things like the new more powerful distribution nodes, and leads to some strange incentives when there.

Technical Details

• In general, the way that the code for how the costs for hacks work has been completely redone. It was super confusing before, and may have had some errors.
  • One thing that was happening for sure is that the "Weaken Turrets" hack was not actually charging you more for higher-level planets, for example.
  • Hacks in general were also charging you the full amount if you canceled them, for some reason.
  • If you canceled a hack, it also said you spent hacking points that you did not (in the log). Except for those times where it 100% did charge you for nothing.
  • It appears you may have been double-charged sometimes, or also got a result for free at other times. What a mess. All clean now.
    • Why was this such a mess? Well, it was the work of about 6 programmers over 5 years, with about 10 different ways of doing things.

• Added a new "Invincible Ohmu" ship.
  • Literally invincible crazy-firepower ship that is only brought into the game via cheats. Great for testing hacks, since it can't be killed.
  • Spawn it with the cheat "cmd:ohmu" (this is now in the wiki cheats page).

Balance Updates

• Updated the Human Resistance Fighters so that you get 3-4 frigates per hack from them, rather than just one. They were way too weak with just one, and not a good deal for hacking points.
  • Thanks to CRCGamer for finding and also fixing!

• Updated the frigates on the starting fleet from 4 to 6 for mugger and sniper frigates (it didn't make sense anymore that they had that low count.)
  • Also fixed the description of Consumer starting fleet
  • Thanks to ArnaudB for finding and fixing!

FRS Rebalance

• Major rebalance to the ships you can hack for from the FRS!
  • For a long time, many of them have been either too expensive to be useful (in terms of AIP costs), or have been too useful to pass up (because they benefit an entire giant fleet of 40 ship lines).
  • We want these to always be useful and exciting, and to encourage different styles of play. So, with that in mind, the general changes:
    • First, these will cost a lot less AIP, so that they are actually viable to take even on lower-AIP games.
    • Second, these will start at a high mark level (like the experimental fabricators in AIWC), so you can choose to just use them as an exciting high-mark combat fleet line.
    • Third, if they have a fleet-wide bonus (a fleet supercharge ability), then that will only apply to fleets with a smaller number of lines, now. So if you want a small buffed strike force, you can create that, but you can't just throw these in a giant blob fleet and still get the effect.
  • The descriptions for all of the ships that are possible also now give recommendations for how to use these to maximum effect.
  • The specific changes are:
    • Turbo Raider:
      • Starts at mk5, only supercharges fleets with 5 or fewer lines, only costs 20 AIP instead of 40.
    • Ultima Fusion Bomber:
      • Starts at mk5, only supercharges fleets with 5 or fewer lines, only costs 10 AIP instead of 80.
    • Ireful MLRS Corvette:
      • Starts at mk6, only supercharges fleets with 7 or fewer lines, only costs 10 AIP instead of 30. Notably awesome with officers and frigates.
    • Tutelar Pulsar Tank:
      • Starts at mk5, only supercharges fleets with 4 or fewer lines, only costs 15 AIP instead of 40. Now gives a hull bonus of 2.5x instead of 1.5x. Notably incredible with officers.
    • Encircling Spider:
      • Starts at mk6, only supercharges fleets with 5 or fewer lines, only costs 12 AIP instead of 50. Notably awesome with officers and frigates.
    • Inciting Parasite:
      • Starts at mk6, only supercharges fleets with 6 or fewer lines, only costs 16 AIP instead of 60.

• The hack to re-roll FRS contents has been removed. You still just get two choices with them.
  • These are specialty high-powered ships, and you get what you get. If you don't want what is there, then skip the FRS.
  • Overall we are working on removing most of the "re-roll" mechanisms, since those are gamble-y and don't feel great to use. Instead giving you more options (as appropriate -- unlike here in the case of FRS), and/or the ability to just skip things.

• FRS hacking balance updates:
  • The FRS now costs 10 HaP instead of 15, and has a response to the hack that is 3x stronger than an ARS.
  • Additionally, if the FRS is in territory that is hostile to you, then the hacking cost is doubled (so in an old-style typical fashion, the cost is now 20, but if you take the planet it's now 10).
  • The FRS ships had absolutely INSANE quantities (like almost 400 turbo raiders). This has been toned down substantially, to more like a third or so of what it was. That was way overkill.

• ARS hacking balance updates:
  • The ARS now costs 7 HaP on non-hostile planets, rather than 15, and 14 on hostile planets.

Beta 2.732 Recon, Golem Buffs, And Sabotage

(Released February 10th, 2021)

• The ranges of turrets no longer increase with mark level for human players. This messes with placement in several ways, doesn't work well with re-speccing science down, and in general adds very little for most players while adding some substantial headache for a subset of empire designer players.
  • However, all of the human turrets are now (from mark 1 on) at 1.3x of whatever their old mark 1 levels were.
  • Thanks to Strategic Sage, tadrinth, zeus, and ArnaudB for the discussion that led to this.

• Fixed a tooltip bug where the RECON hops watched by command stations was not actually showing for player units (that includes fallen spire stuff, too). Now that's properly shown for player units.
  • This certainly made it difficult to know that there was a benefit of this sort from logistical command stations (and now military as well!).

• Fixed an issue in the prior build where if you opened the in-game galaxy options without first opening the lobby galaxy options, you'd get endless errors until you closed that screen.
  • Thanks to Eonfighter for reporting.

• Distribution nodes now grant 80 HaP instead of 20, making them WAY more valuable.
  • Depending on how much you start doing with hacking, these may be some MVP targets now.

• The cheat for metal can now be done without an argument for how much, and it then gives you 10 million metal instead.

Hacking Updates

• The "Watch Planet" hack now costs 5 HaP instead of 10 HaP.
  • This is something that is devalued a lot when you consider the addition of spies in the last build, so this is mostly for more distant targets.
  • This hack's description now talks about the alternatives of recon abilities from logistical and military command stations, as well as spies, so that when someone sees this hack they have a good sense of what they can do with it.

• Added a new hack_completes_instantly="true" xml feature for various hacks to make them complete instantly, even when the game is paused, rather than them taking even one second.
  • This skips any sort of AI responses that there would be to a hack, too, so it's mainly for things that already had not response.
  • This is now used for the various scouting/exploration hacks, and will be used for some other upcoming hacks.
  • This makes the exploration hacks way more responsive-feeling, particularly when the game is paused and you want to examine things without unpausing and re-pausing over and over again.

• When you are in the middle of a hack and your target dies or hacker gets crippled or whatever, it now notes that the hack failed in a chat message.

• If a hack fails or is canceled, it now shows up in a much better way in the hacking history, making it a lot more clear in failure states in particular.

• When hacks finish successfully, it now gives a message in a consistent way, after it has probably already given a message about more detailed results.

• Even if you can't do a hack, it now lets you actually see into the list of options when you click the hack button.

• For instant hacks, if they require a hacker is present, they no longer care which hacker it is. That never matters, unlike hacks against structures that are done over time.
  • This reduces some interface fiddliness with hacking your own units to upgrade them in various ways.

New Hacks

• New hack!
  • You can now hack your own logistical command stations to increase their recon range by +2 for a cost of 60 HaP.
  • You can do this a second time for a cost of 150 to make them incredibly all-seeing.
  • This actually can scout planets for you, if the things in the new recon range are not scouted yet. This is super duper powerful... but also expensive in HaP.

• New hack!
  • You can now hack your own military command stations to increase their recon range by +1 for a cost of 40 HaP.
  • You can do that up to 3 times, for an added cost of 1.5x each time.
  • This is using the exact same hack as the logistical command station, but using all the new hacking tools to make it so that it is custom per unit!

• New hack! It is now possible to increase the hull durability of certain ships by hacking them.
  • At the moment, this is a feature mainly for golems. This gives them far more durability than before (on average about 2x), for on average around 20 hacking points.
  • This makes the golems feel a lot more like the golems in AI War Classic, but it's a costly investment to do so. You can hack them multiple times and really crank it up, but this gets even more super expensive.

Updated Hacks

• The sabotage hack now allows us to have a varied cost depending on what you are hacking. Normally the cost is 11 HaP for most targets (down from 15 previously).
  • Troop Accelerators are only 7 HaP.
  • Black Hole Machines, Alarm Posts, Magnifiers, and Raid Engines are 16 HaP (hey, you're dodging 5 AIP from them).
    • But the ones from the Zenith Trader are only 7 HaP now.
  • Scourge Fortresses are all now 18 HaP.
  • The AI Spire Fortress is now 19 HaP.
  • Wormhole Borer is now 25.
  • In general, there may be some new targets that folks want us to add for sabotage. Things that were previously not worth it (need low HaP to be worth it) or too powerful (need higher HaP to not make sabotage way too effective).
  • You can ALSO now sabotage warp gates, and thus get out of their 5 AIP cost (and/or not have to gate-raid) for 14 HaP.
    • This gives you some new options for controlling where waves can come into your planets, as well as allowing you to do it faster.
    • To some extent, this can let you take more planets for the same amount of AIP if you do this enough, which presents interesting options; but you'll forgo a lot of the extra HaP from that if you do so, which has its own downsides.
  • Bear in mind that these numbers may require tuning up or down in general, but in general the relative values seem nicely-spread.

• The sabotage hack's tooltips and text in general have been updated, and a lot of behind-the-scenes code bits for it.
  • The sabotage hack also now uses the thing where it pops up a list of options for you to hack, with info for each one. It does this instead of making you park really close to the thing you want to target.
  • It's worth noting that even when there is one option, it makes you choose it from the sub-list. This is because it gives you extra tooltip information that you can't get any other way, and it keeps the experience consistent so that you don't "click to see options" and accidentally trigger it if there's only one target.
  • In the hacking sidebar, if there is a variable amount of hacking points that something can cost depending on the options you chose, as in this case, it now shows the range.

Removed Hacks

• The ability to re-roll tech vault contents via a hack has been removed.
  • This might be a bit controversial, but we're giving you waaay more options now in each tech vault, and there is overall a tension in the game between "customize" and "play it as you find it."
    • The re-rolling, even if it includes a deterministic seed (which this now does), still encourages a form of gambling with hacking points, and/or save scumming. Neither of those things are something we want to encourage, and they're both at odds with the idea of "play it as you find it" without actually furthering the goal of "design your empire."
  • So essentially, this is on the other side of the spectrum from the ability to refund your techs to give you more control over building your empire. This sort of re-roll gives you too much power to adjust what you find, and in a gamble-y sort of way, and that's not great. Hopefully the increase in options at each tech vault are now sufficient that you can find something of interest, and if not... that's part of the decision-making of choosing which targets to capture or hack, and which to ignore.

• The "Watch Planet (local)" hack, which only cost two HaP, has been removed.
  • This was something that let you watch certain planets for a fraction of the cost, but only until the AI re-took the planet, etc.
  • This niche is now filled by the recon abilities of command stations, and by Spy units.
  • Having this as an option is ultimately something that is confusing and a bit of extra clutter when you consider the current options around.

Removed Galaxy Options

• The "Scout Adjacent Only" option from the galaxy options has been removed.
  • This was ludicrously masochistic and not something that we felt like anyone was likely using in the first place. This was something that we added a few years ago when we revamped scouting, but at the moment it's just detritus.

• The "Hops for planets adjacent to yours are watched" option from the galaxy options has been removed.
  • This was a semi-cheaty workaround for players who wanted more situational awareness than the game used to give them.
  • The same sort of functionality is now provided in-game via spies and command station buffs, and is done so in a non-cheaty fashion. This should satisfy the players who want this style, and be more discoverable as well.

Technical Details

• Added display_name_and_description_insert_target_ship_name as a feature for hacks.
  • When true, this allows us to put {Target} in the description and name and have it show the name of the unit being hacked.
    • There are a number of hacks that are now going to be used on a wide variety of ships/structures, so this lets us do that with them actually being identifiable.
  • There is a semi-similar feature for hacking event logs already, but not something for hacks yet-to-be-done that also works in event logs.

• There's a bunch of new code on fleet memberships for logging how many times one of those has been hacked in a fleet, and in general tracking some upgrades on that level.
  • This is the basis of a framework for hacking your own ships -- but always by fleet membership line as a whole -- and it's useful because it can survive command station destruction and reconstruction, and type changes, etc.
  • This is also, structurally, very very similar to what we will be doing for ship loadouts in the future for spire ships and certain other things. So this plays into that in terms of being useful as a guide for a lot of future features.

• Added a new Hacking_HackSelfUnit_Base abstract class, which forms a new basis for hacking our own ships.
  • This does not HAVE to be used as the underlying class for handling hacks against our own ships, but it has a lot of convenient helper methods that will make each of those other hacks a lot more brief to code and easier to know are correct.
  • Some of the more generalized stuff has also been integrated into the BaseHackingImplementation that basically every hack inherits from, so that is useful also. This is a pretty major extension of hacking, particularly in the ability to tune hacks via xml and also differentiate a given hack for multiple target types.

• Also added a new hack_data sub-node for units in general, which can let us define in xml a lot of dials about how a given unit responds to a specific hack.
  • This can be things like extra cost increases, or even custom data used by a given hack.
    • There is override_cost_in_hacking_points, and cost_in_hacking_points_multiplier_per_times_same_unit_hacked, as some basic items that allow for overrides.
    • max_times_single_unit_can_be_hacked is in there for controlling how many times a unit can be hacked in general by this one hack.
    • You can have arbirary key/value data as custom_hack_data child nodes.
  • This information DOES propagate to child units via the copy_from attribute. It is possible to do partial overrides of that info in the children.

• only_for_hostile has been removed from hacks, and there is a new only_for_target_type instead.
  • This can be Any (the default), AlliedUnits, AlliesAndMyUnits, MyUnits, NeutralUnits, EnemyUnits, or EnemyOrNeutralUnits.

• unit_tag_target has been removed from hacks, because we are directly assigning hacks already and this is redundant and confusing.

• Added a new Hacking_SelfUnitWithSubSelection_Base, which has a ton of things set up for us to be able to self-target various numbers of ships for certain hacks on planets. Useful when various things can self-upgrade.

Beta 2.730 Spies, Early Officers, And Randomness Options

(Released February 9th, 2021)

• Fixed a couple of ways in which savegames could fail to load because of removed hacks in the prior build.
  • Thanks to Sigma7, Arides, and Isiel for reporting.

• Added a new "EarlyOfficers" tag to the following Arks: Ark One, Orchid Ark, Rorqual Hegira Ark, Black Widow Golem, and in DLC1 Sol Ater Ark.
  • These have also had their max seeded per galaxy reduced from 2 to 1 (Sol Ater and Rorqual was already 1).
  • Their spawn likelihood in general has also been reduced to 1/20th of what it was before in most cases, compared to other Arks and Golems (for the usual ways of spawning that exist so far.
  • There is now a single EarlyOfficers ship spawned within 1-4 hops of one of the player homeworlds in each game. This is in addition to the other usual golems and arks that already seed.
    • This essentially gives one empire in each game a way to get an officer fleet early, and one which is likely to be particularly helpful in the early game and maybe less helpful later.
  • Thanks to various players on discord for suggesting we have an early officer like this, and to CRCGamer and zeusalmighty428 for helping us come up with the list.

• Like strikecraft, engineers no longer have personal shields and instead have that added directly to their hull health.

• Rather than costing 2 hacking points per line you retrieve science form, it now costs 5. This is still extremely cheap, but should hopefully make this mechanic in general feel a bit more balanced.
  • Yes we want you to re-spec when you need to, and to invest science early and then rework your empire as you find new things (because you do not have the ability to see through the fog of war in the preferred way to play), but we don't want wanton and constant re-speccing, or for people to feel like this is some sort of cheat. This is a part of game flow, and doesn't take away from "play it as you find it." The ships are what you find, the science is what you design.
  • Thanks to folks on discord for their general feelings.

Spies

• Added a new civilian unit: Spy
  • Provides visibility on an enemy planet with a very low chance of being detected. Able to cross through wormholes, unlike the rest of a command fleet.
  • Has a high amount of cloaking, a very high albedo so it is hard to decloak, is fast, and has decent health. Costs a lot of metal (12k) and uses a lot of energy (2k) for the fact that it has no weapons or other tools.
  • Logistical command stations get four of these, and military command stations get two. Home and economic command stations get none.
  • It's worth noting that you can get the same effect by hacking planets to watch them with spy nanites, but that uses up finite hacking points.
  • Thanks to zdrgn for suggesting.

• Logistical Command Stations now watch planets at 2 hops out, rather than 1. Military command stations now watch planets at 1 hop out.

• Lesser Spy Cradle now watches at 5 hops out rather than just 3.
  • Greater Spy Cradle now watches at 8 hops rather than 4.
  • The latter of these (both of these are DLC1) can now basically see almost the entire map, hopefully making it more worthwhile.

• On planets, there is a new GameSecondLastWatchedByCommandStation that we can use to tell the last time a planet was watched by a command station's power, or a spy cradle, or similar (many fallen spire ships have range-1 spy devices).

• New galaxy map display mode: Spy Network
  • Shows which planets have spies on them (and how many), or which have spy nanites, which ones are viewed by recon structures, and which ones are visible for other reasons. Shows the same icons as the Normal view.
  • You can use this to see easily where you have visibility and why, and make sure you don't have extra spies in place in pointless places where you already have recon view, or where multiple spies are accidentally on the same planet (maybe from different players, even).

More Randomization Options In MapGen

• Two new additions to the Roguelike Additions section of mapgen:
  • Planets To Randomize: Marks 2-6
    • Range from 0 to 300, new default of 6 (was previously the equivalent of 0).
  • Planets To Randomize: Marks 1-4
    • Range from 0 to 300, new default of 2 (was previously the equivalent of 0).
  • General description:
    • Normally there's a pretty sensible distribution of planets and mark levels, but this can cause a feeling of sameness in a lot of galaxies.
    • If you're ready to spice things up, and willing to have some potentially very challenging map designs, then this feature will go in and randomly swap the mark levels of planets in the ranges specified.
    • It will make however many swaps you ask for, up to 300 per category, and will overlap itself as it does the swaps without any hesitation. It will not swap any homeworlds, or what ships are on these planets.
  • This actually helps us get back some of the feeling of AIWC, where you could occasionally have a mark 4 (equivalent of mark 6) planet right next to your starting world.
  • Thanks to Sombre and others for requesting features for this sort of randomization.

• Added a three new Roguelike options to the Map Gen section of game options in the lobby: Extra Allowed Distance For Adajent-Seeded Items (range 0-99, default 0)
  • Certain things that you can capture are seeded on adjacent planets to players if at all possible, to give them something early to capture. If you would like to expand that limitation so that it needs to be within 2 planets, or 3 planets, or... 99 planets... you can do so.
  • The more you expand this, the harder the game MAY be, as things will be absolutely random in distribution. These items will tend to still be required to be at least 2 away from any AI homeworld, though.
  • This will also be applied to things that are intended to be seeded two hops out, so if you give +1 here, it will seed those within 3 hops.
  • Thanks to Sombre and others for requesting features for this sort of randomization.

• Added a three new Roguelike options to the Map Gen section of game options in the lobby: Extra Allowed Distance For Middle-Distance Items (range 0-99, default 0)
  • Certain things that you can capture are seeded on adjacent planets that are usually on the nearer half of the galaxy; generally within 6-8 hops of player homeworlds, but at least 3 hops out.
  • By increasing this count, you allow them to be seeded farther away (but they may still seed close, it's the nature of random).
  • Thanks to Sombre and others for requesting features for this sort of randomization.

• Added a three new Roguelike options to the Map Gen section of game options in the lobby: Reduced Distance Restriction For Any Items Items (range 0-6, default 0)

Various things that you can capture or that give some other benefit are normally seeded a minimum of something like 3 or even 5 hops from human planets (when the map allows for that).

• • This allows for you to relax those restrictions, which can make for surprise early-game boons. A Zenith Power Generator on your doorstep, and a SuperTerminal right next to it? A Golem really close by? Who knows!
  • It's worth noting that this only increases the restriction, but doesn't shift them all closer for sure. Most of the time if this is increased all the way to 6, that will just mean 'anywhere in the galaxy' for things that are meant to be late-game otherwise.
  • This can make parts of the game easier, but it's ultimately going to be very random.
  • Thanks to Sombre and others for requesting features for this sort of randomization.

Tech Vault Balance

• Tech Vaults no longer grant hull techs as one of their four options. They now grant only three options: two weapons, and then a third special one. The third special one is very likely to be quite valuable, still, but the hull one was always going to be the obvious pick.
  • Thanks to Strategic Sage, Lictuel, and others for suggesting.

• The game no longer seeds "however many human empires there are" (plus 1 if difficulty of the highest AI < 5) of tech vaults within a few hops of the player starting zones.
  • These tech vaults (it's just one in a normal single-player game against a difficulty 5+ AI) are simply lost to you.

• There was previously another 3 + "however human human empires there are" (plus another 1 if difficulty of the highest AI < 5) of tech vaults that were seeded in the "middle distance," aka randomly within 3-8 hops of player homeworlds if at all possible.
  • This has now been changed so that the number is based on a new "Base Tech Vaults To Seed" galaxy option, which you can adjust from 0 to 3, but which now defaults to 1.
  • For multiplayer, if the "Seed Extra Tech Vaults In Multiplayer" is on, then it will seed 1 extra tech vault per human empire beyond the first.
  • It no longer gives extra tech vaults for really low AI difficulties.
  • Thanks to Sombre for requesting more customization options, and others in general for registering the OP-ness of tech vaults in large volumes.

Galaxy Options Categories and Subcategories

• The galaxy options now also support full categories and subcategories, like the settings have for a week or two.
  • The destroyables, capturables, and galaxy design sections have been condensed into a new Map Generation section, with various subsections.
  • There is a new Allies section, which organizes a number of settings more clearly for you to be able to modify those.
  • In general a number of galaxy options moved around, but are easier to find now. It's also easier for mods to add well-organized options.

• The category buttons on the galaxy options screens are now a bit less tall, and closer together, so that more can be shown. Relevant if you have a lot of mods.

• Civilian Industries has been updated to include a galaxy option setting category, since that is now required to be explicitly created.

• Basic galaxy option categories are now in place for AMU, although NR SirLimbo can rearrange those to his liking in the future.

Beta 2.729 Tech Vaults, Bugfixes, And Major AMU Updates

(Released February 8th, 2021)

Tech Vaults

• TLDR of the below changes in the detailed notes:
  • Tech Vaults now give you an option of 4 different techs to choose from. Two weapons, one hull, and one special.
    • These options may differ between players in multiplayer, in terms of what each player sees, and that's okay. This is meant to be custom to the player now.
  • Because you get so many more options now, and because we don't want to generally translate hacking points into science points, you can no longer convert a tech vault into science points.
  • The rest is mostly a bunch of improvements to make hacks-with-options possible for modders to add, and easier for us devs to add.

• It is no longer possible to save a game before a tech vault reroll-hack and then reroll-hack it and then reload the save and reroll-hack to get different results. It now gives you consistent results every time you reroll-hack for a given tech vault, but each time you reroll-hack in a sequential fashion (aka without reloading the save and getting your HaP back) it will give you a fresh set of stuff.
  • Thanks to tadrinth for reporting.

Hacking Improvements Under The Hood

• The List<TechUpgrade> TechsGranted has been removed from ships, as it is now redundant and we are handling things more flexibly.

• A new in HackingSeed has been added to ships/structures, which will let us give consistent random results (including doing rerolls as needed) without ever having to fear that save-scumming for infinite non-deterministic random reroll will be used.

• TechUpgrades now support arbitrary tags, like ships/structures do. This is useful for various kinds of arbitrary filtering we might want to do.
  • Especially useful for mods that have custom techs and want to make something like a tech-vault that works only for specific subgroups. But really, this has a variety of functions.
  • All of the weapon techs now have the tag Weapon.
  • The hull techs all have the tag Hull, and then also have either HullStandard, HullTurret, or HullOtherFaction.
  • The civilian techs have the tag Civilian, and command station techs have the tag Command, adn defense has the tag Defense.
  • The turret hull tech, and the command stations and mobile fort all now have the tag SpecialBonus.

• Added GetFactionOrNull_Safe() to GameEntity_Base, since we use this sort of thing a lot.
  • Also added GetFleetOrNull_Safe() to GameEntity_Squad, for the same reason.

• On TechUpgradeTable, there is a new SortTechList() method, which lets us sort any list of techs in the same order that it would be in the game sidebar.
  • Useful after techs have been chosen from a pool in some fashion.

• Added a new HackingUtils class.
  • There is a static GetDeterministicRandomGeneratorForHackTarget_Threadsafe() on there, where you can pass in a squad that is to be the target of a hack, and it will give you a random number generator that gives you consistent results.
    • The nice thing is that this is also CONTEXTUALLY deterministic, so for instance if there were 5 options before and 1 of them is no longer valid because you already got it all or something, it will give you a somewhat revised set of 5. This is a lot more flexible than in the past, while at the same time not falling prey to save-scumming.
    • This works with the HackingSeed if that has been set to be above 0, or if that has never been set then it uses the PrimaryKeyID of the hacking target.
  • Added a new GetListOfTechsValidForFactionFromTag() method, which gets the list of techs from a tag that the faction can currently acquire (has not already maxed them out, etc).
  • Added a new StartListOfTechsForTechVaultStyleGranter_ByTag(), which lets us fill a list partially from one tag.
  • And added a new AddToListOfTechsForTechVaultStyleGranter_ByTag(), which lets us add to a list from another tag.

• Hacking_GrantTech has been made abstract, and there is now a Hacking_GrantTech_TechVault implementation for the tech vault for that. Any other hacks that mods or similar want to make out of this can be based on the same new underlying class.

• The tech vault is now set up so that it will give you options from 2 weapon techs, 1 standard hull type, and 1 special bonus type.
  • If there are no weapon types or standard hull types that you can still upgrade (WOW), then it gives you more of whatever the lower categories are.

• It is no longer possible to hack Tech Vaults to turn them into science. This is basically a way of converting hacking points into science, and that's something that we don't really want to let people have: in multiplayer it is unbalanced, and in single-player... well, it's less unbalanced, but still not great.
  • Now that tech vaults give you far more potentially-valuable items, and those items work alongside refunds to techs if you want to, there should always be something worth getting from them -- or if not, they can just be ignored like any other capturable that doesn't fit with a specific campaign.

• IHackingImplementation now has a WriteAnySpecialDisplayCodeForHackedShipTooltip method that can optionally be filled.
  • This is a major new addition that allows any arbitrary hack to write information to the tooltip of ships that can succumb to this hack.
  • It literally can write whatever you want, and it even includes what tooltip detail level players are looking at it from.

• IHackingImplementation now has a GetDoesHackRequireASinglularChoiceFromASubmenu that allows for any hack to specify that it wants give the player an arbitrary list of things to choose one from.
  • Then also added AddAllButtonsForSingularChoiceInSubMenu, which allows us to create buttons on the fly as needed for filling the sub-panel.
  • HackingUtils now has a CalculateNextBoundsForSingleHackingOption() method, which is helpful for anyone to be able to populate buttons of the correct size from AddAllButtonsForSingularChoiceInSubMenu.

• Renamed GetClosestHacker to GetPreferredHacker (since that is more accurate now) and moved it to HackingUtils. Did this with obsolete wrappers to not break existing mods.
  • Also moved CalculateHackerForHack, CalculateCanDoThisHack, and TryDoHack to HackingUtils. This makes so much more sense in terms of how we need to access this sort of thing, especially now.

• Fixed it so that when you make a choice from a hacking sub-options menu (as with the ARS in the past, but now also with tech vaults, etc), then it closes the popout choices versus leaving it hanging open.

Bugfixes

• Fixed a bug where the game was thinking there were 1 too many Human Empires for calculating the amount of hacking points that could be gathered per planet.
  • Thanks to NR SirLimbo for finding and fixing!

• Put in various fixes for Integer Text Boxes in Galaxy Settings:
  • The values now correctly appear in both the actual text boxes and the labels when viewing the settings.
  • The settings correctly alter when using Game Commands to change them.
    • Most of this was due to Integer Text Boxes being a "hybrid" between an integer and a string input.
  • Thanks to NR SirLimbo for finding and fixing!

• Fixed a number of exceptions that could happen if you were viewing the details of a command-station-style fleet and the command station dies while you are looking at it.
  • Thanks to Sigma7 for reporting.

• Fixed a seemingly-rare cross-threading exception that could happen in LongRangePlanningData_GameEntity_Squad. This is something that has probably been possible to hit for years, but we only saw our first report of it now, so it does seem quite rare.
  • Thanks to Tydorius for reporting.

• Fixed a few other possible nullref exceptions, at least one of which was hit, in LongRangePlanningData_GameEntity_Squad. It looks like these are more likely to be hit on the latest ultra-powerful CPUs with 8 cores and amazing multithreading, etc.
  • Thanks to Tydorius for reporting.

• Fixed starter fleets showing the wrong number of strike craft. They now correctly apply the 40% boost at mark 1 to their cap that was introduced in recent balance changes.
  • Thanks to NR SirLimbo for finding and fixing!

• Fixed a long-standing issue where a negative knockback (= pull) would always pull a ship onto the pull's source, unless the pull was so powerful it would pull the target over the source, in which case it would instead do so of limiting it.
  • This only bug if the target had but a single pull in this movement instance.
  • It might become a slight nerf the Crusher Turret, but very minor if at all.
  • Thanks AGAIN to NR SirLimbo for finding and fixing!

Mod Updates

• AMU:
  • Fixed AMU throwing an error when starting a new game due to the Hacking Points gatherable per planet being changed.
  • Improved FakeFactionExecutor Executors:
    • Actually created the mechanic for Executors of the ExecutorFakeFaction to not cause the entire thread to fail. This was falsely claimed during a previous release, but now actually happens.
    • Executors now also get an ArcenCharacterBuffer passed into their Execute() function to write data into that is dumped on exceptions.
    • Executors also reset when the ExecutorFakeFaction is created multiple times (for example when loading multiple saves in the same game session) instead of potentially accumulating endlessly.
      • Thanks to ArnaudB for wondering why he'd get 11x 3 AIP reduction per 20 minutes...
  • Cleaned up AMU's exception logging to be much more clear and better formatted.
  • Implemented a great amount of further game settings. In total there are now 43 settings in the Galaxy Settings, most of which can be altered ingame. A few examples:
    • Wormhole Invasion Duration
    • Shark AI plot AIP or strength
    • Multiplayer Hacking Point Scaling
    • Repair Delay After Being Crippled
    • Repair Cost For Hull/Shield/Forcefield/Engine
    • Recloaking Time
    • Maximum Paralysis/Weapon Delay/Engine Stun Time
    • Flagship Speed Multiplier While Hacking
    • Max Ship Lines Per Mobile Centerpiece
    • Brownout Duration
    • ... and many more.
  • Created the Severe Brownout mechanic:
    • Essentially, while browning out and still in the red with power any ship or structure the player affected owns may become paralyzed at any time for a random duration of 1-20 seconds.
    • The worse the brownout (based on energy produced and consumed) the worse this gets. However, once the energy balance is positive again this stops even while the brownout is still turning off forcefields.
    • This will also stall until the actual brownout happens, so the "grace period" and "allowed power debt" rules from normal brownouts still apply.
• Extended Ship Variants
  • Fixed the Rescue Beacon being auto-demoted in mark by the game due to having no techs. It now counts as a drone frigate, which means it always shares its centerpiece's rank. In addition its health now scales like that of a frigate.
    • Thanks to crawlers for reporting this.

Beta 2.728 Full Multiplayer Beta

(Released February 5th, 2021)

• You can give nicknames to (pretty much) any faction, now. Not sub-factions like the warden or whatever, but if you have multiple AIs you can name them, and if you have multiple traders or devourers you can give each one a name, too.
  • Thanks to Vic for suggesting.

• The way that the "closest hacker" is chosen is now completely redone.
  • For purposes of previews of hacks, it just chooses a random valid hacker if there are multiple, but it prefers ones that are selected to ones that are not.
  • For purposes of actually doing a hack, if you have multiple hackers selected (or no hackers selected and multiple viable ones here), then it will thow up an error window asking you to select one.
  • For the distant hacks that are done by your king unit, those don't involve this.
  • This whole thing simplifies a lot about hacking, and honestly in most cases you probably won't run into the error. In the cases where you do, just hitting the fleet hotkey or clicking the flagship in question takes care of things.
    • We were thinking about making menus to make you select things, but that was going to get a lot more fiddly what with other hacking menu options being a thing. Instead this handles the edge cases without even getting in your way in cases where there is just one hacker, or one selected hacker, around.
  • In the event that you have a normal flagship and a support flagship both selected, it will choose the normal flagship without bugging you, incidentally.

Multiplayer

• The central Balance_HackingPerPlanet, with its balance_hacking_per_planet, has been removed.
  • There is now a balance_hacking_per_planet_per_number_of_human_empires, which fills a list of Balance_HackingPerPlanet_PerNumberOfHumanEmpires.
  • Instead of this always being 30, it is now 20 with 2 human empires, and 15 with 3+ human empires.
  • This was another of those lingering balance issues with multi-faction multiplayer. You wind up taking many more planets in that scenario, so you should still have a good amount of HaP... but not SO much as before.

• New hotkey: Open Faction Window Without Pausing
  • Pressing this button will pop open the 'View /Edit Factions' window without pausing the game. This is extremely useful for in multiplayer, where you may wish to use the controls in there to set up rules for how you interact with other players, gift them metal, or to even switch which faction you are controlling.
  • Defaults to F2.
  • There is now a tooltip in the escape menu that explains what you can do in the factions window, and which also mentions this tooltip.

• As with AI War 1, if you hit your metal cap in multiplayer, the metal will automatically spill over to all other human empires.
  • This is done in a very simplistic fashion: if there are two other empires, they each get half of whatever your overage would have been (normally your overage is just wasted, recall). If there are three, they each get a third, etc.
  • If one of the other empires is ALSO at their metal cap, it does not reallocate their portion of your overage to those not at-cap. Their portion is just tossed away as wasteage.

• On the player faction screen, there is now a subsection for "Players Controlling." It then has detailed tooltips about what that section is for.

• In the in-game factions window only, the non-editable settings for a player faction are now shown further down below and have their own subsection shown with an explanation of that. These are basically just for reference, so moving them out of our way is a good idea since this screen is going to be doing more now.

• When you first open the in-game factions window, it now automatically opens to your faction. This makes it vastly easier to manage your settings related to your faction in multiplayer.

• Fixed up a couple of edge cases in how other player faction names could appear a bit wrong in some cases in multiplayer. And in general made uncontrolled factions show better names.

Gifting Resources Between Empires

• There is a new "Multiple Human Empires" section of the player faction window that appears when there are actually multiple human empires (in other words, in solo play it doesn't appear, and purely shared-faction multiplayer it doesn't appear, and a human empire and a champion doesn't see it, etc).
  • The first line item under this is a dropdown of a gift type, and a gift recipient, and an amount. Any player can actually set up gifting from any human faction to any other in here, but by default you're seeing it for your own faction.

• The first style of gifting now fully works: one-time metal gifts.
  • This lets one faction give a lump sum amount to another faction.
  • If the giver tries to give too much, it will just give whatever they have on hand and a message to that effect is shown.
  • If the recipient cannot store as much as the giver is trying to give, it will be reduced to the amount they can store, and a message to that effect is shown.

• It is now possible for empires in multiplayer to gift hacking points between one another.
  • This works just like the metal one, essentially, except that there's no max cap on how many HaP a player can store.
  • The tooltip notes that science is specialized to each empire and so never can be gifted. That's actually pretty critical for balance, since all the players are getting a copy of the science, so that's not something that will change in the future.
  • However, we can definitely see some situations in large multiplayer games where each empire is too poor in HaP to do some hack, but pooled together they can totally do it. Artificially limiting things so that people can't share HaP doesn't seem warranted in light of this. Most everything that you could hack for (fleets, doubled ship lines, new ship lines, etc) can be gifted between players after being hacked-for already, anyhow.

• You can now set up an ongoing gift of up to 25,000 metal per second from one player empire to another.
  • Once this is set up, on the faction screen you can see the record of it, and clear it if you want. You can also just overwrite it with a new gift if you prefer.
  • As the tooltip here explains, the gifting is done AFTER all of your other expenses for metal within your own empire, so gifting will never slow down or impede your own direct expenses; gifting will only give whatever it can out of the total you set, after your expenses.
  • The math on this all seems to be correct, and seems to show up properly in the metal flows history for both the giver and the recipient, but this is pretty complicated and so if we're actually off in terms of the numbers we show on the UI for the flows, that's probably a display thing. It's hard to be certain if those data points in the flows are at a scale of 1 per second, or 1 per tenth of a second. If the latter, then we're showing the flows from gifting as being 10x too high compared to all the other numbers, but the actual underlying math for how the flows work is correct either way.
  • This feature turned out to be... remarkably more robust than we had really been planning on creating. We never had anything remotely so powerful in the first AI War. Gifting metal harvesters or similar back and forth was a huge pain, and there was always a risk to your own production on top of it. This solution is far more user friendly in every respect!

• The last form of gifting between player empires is another ongoing one: this time it's energy.
  • Energy is something that constantly flows and is consumed, so this is very easy to make sure is consistent on the UI. It doesn't have a "per time" element.
  • You can send up to 2 million energy from one empire to another, and the setup is just like you would do for ongoing metal flows.
  • Also the same as the metal flows, it will ONLY give the energy that you can actually afford. So if you've set yourself up to give 100k to an ally, and your extra energy drops to 50k because you either built a bunch of new stuff or lost an energy producer, then your gifting will drop to 50k during that period. You don't ever have to worry about your energy going negative at all or risking a brownout because of gifting out.
  • Energy really wasn't on the list of things that we were going to make giftable in multiplayer for this game, but it turns out it is incredibly useful, and remembering how things were with the first game and trying to share economies in various ways, this is again far more straightforward and safe. The fact that you don't have to take down your gifting rules if your economy tanks is really useful.

Human Resistance Fighters

• Added a new hull tech type, right under Alien: Expatriate.
  • Description: Expatriate hulls are unlikely to be something you encounter at all unless you specifically hack certain factions for them. The ships that benefit may be very similar to your own, but there is a difference. Most humans who are not trapped on a planet are already a part of your war effort, but some groups have refused to fully join for various reasons. Collectively, these are the Expatriates.

• All of the human resistance fighter ships now solely get levels from the Expatriate tech, and not from any other hull or weapon tech types.
  • When it comes to any other human-style factions that you can hack to get ships from, we need to make that same adjustment; for all the alien ones, so far we seem to have hit all of them with the Alien tech.

• The Human Resistance Fighter version of the Assault Frigate is now called the Assault Frigate Defender.

• The Nucleophilic Defender, Concussive Defender, and Tesla Defender now all start at Mark 2 rather than Mark 1, to be like the rest of the Human Resistance Fighter units.
  • These are overall more powerful in the main, true, but shorting them a mark level was confusing at best.
  • Thanks to GreatYng for reporting.

Improvements To The Prior Build

• The new popup for science refunds now says 'item' or 'items' depending on whether you have 1 or > 1 things to refund. Also some additional colour has been added to the text

• The Yes/No order of buttons on most of the confirmation popups has been swapped to be No/Yes, which is now consistent with... all the other interfaces in the game.
  • Thanks to Badger for noting that we'd inverted this!

• The arcen thread manager now still checks for auto-killing threads even if the game is paused, and it spaces out its timing based on realtime and not game time. This probably will have zero effect on anything, but it's more correct.

• There is a new DoOnPerSecondWhilePausedNonSimUpdates_OnMainThread, which developers and modders can use to do things once-per-second while the game is paused.
  • We are now using this to make sure that the totals for the science strength amounts on the tech sidebar update properly even when the game is paused.
  • Thanks to Daw11 for reporting how this was not changing unless you unpaused.

• Added a few new names to the "Standard" planet name list.

• EntitySimLogic has a new DoPerFrameWhilePaused method that gets called every frame only while the game is paused.
  • This lets us do some various per-fleet calculations, such as what the intended mark levels of things are, even when the game is paused, so that things can react to mark level increases and decreases properly without having to unpause.
  • Thanks to Daw11 for reporting.

• Fixed an issue where refunding techs was not reducing the mark level of ships until the next time you unlocked a tech, or saved and reloaded, because of some caching happening at the faction level.
  • Thanks to Daw11 for reporting.

• In the fairly-rare cases where a player-controllable ship has a starting mark level higher than 1, we now state what that starting mark level is right next to where the tech upgrades are. Otherwise it seems like a very perplexing bug, because there seems to be no good reason for the ship to be at that higher mark level.
  • This mainly applies to ships that you get from Human Resistance Fighters at the moment, but it's something that could come into play with any kind of ship if we wanted it to, since the mechanic does exist and be applied however we want.

• Fixed several windows, including the science refund, fleet member swap, and fleet gifting screens, so that when you hit the escape key it closes them rather than opening the escape menu.
  • This is nice for the sake of consistency, but it also prevents players from existing back out to the main menu while these screens are still open.
  • Thanks to Badger for reporting.

Beta 2.727 Tech Refunds And Balance Tuning

(Released February 4th, 2021)

• Added to the engineering tech description: Once engineers reach mark 3, they also become cloaked, which is incredibly super-duper useful if you are using engineers in dangerous areas.
  • Thanks to Paradox Song for... ahem... reminding Chris this was even a thing.

• Fixed an issue in the prior build where the AI and other variant MLRS/Fusion turret renames were not carried over. Now all the AI, Ghost, Marauder, etc, turrets all have the proper names.
  • Thanks to CRCGamer for reporting.

• Fixed a couple of stupid typos that had broken counterattacks for the last while. They were counting down to zero but not actually launching any units. This was unrelated to any wave changes, but rather was just a matter of some typos in some code that we were making cross-thread-exception-safe.
  • Thanks to ussdefiant and CRCGamer for reporting.

• prevents_all_weapons_fire_on_planet has been renamed to creates_ceasefire_on_planet
  • Also added a new blocks_ceasefire_on_planet.
  • Ceasefire blockages override any and all ceasefires. They basically cut through no matter how many ceasefire-creators there are -- a single ceaasefire-blocker wins over however many ships are trying to create a ceasefire.
  • The following ships now have the ability to block ceasefires: Ravenous Shadow, Devourer Golem, Flenser, Planetcracker, and Mothership.
  • Thanks to Chthonic_One and Ovalcircle for requesting that the devourer still be able to eat the Trader. This will also lead to some interesting fun with DLC2 later.

• The game now detects a variety of statuses that should prevent you from swapping ship lines to/from fleets, and now prevents members of those fleets from showing up in swap lists. On the fleet in question, it shows a message explaining what is happening any why you can't do swaps.
  • If the flagship is crippled, you can no longer swap fleet members out from it.
  • If the flagship has not yet been fully claimed, you can also no longer swap fleet members out from it! That was a problem with items still in the process of being claimed (and even let you bypass AIP costs of golems while looting their ship lines for other fleets).
  • Thanks to ArnaudB for reporting.
• XML tags added for additional flexibility to KDL_VanillaRatesOfFire.xml and KDL_VanillaEntries.xml (on-death effects)

• Added three new window prefabs for various future purposes:
  • ModalScrollingContentsOKCancel, ModalDropdownWindow, and ModalTinyTextWindow.
  • On their own, these do nothing, but they pave the groundwork for a variety of future functions. Most directly we're going to use them for some hacking game flow options, but in the more distant future they can be used by us or modders for anything, since they're pretty freeform.

Multiplayer: Gifting Fleets/Planets

• In multi-empire multiplayer games, human players can now gift fleets from one to another.
  • This works for all kinds of fleets, even those that have not been fully claimed, or which are crippled, etc. Mobile fleets, starting fleets, lone wolves, support fleets, entire planet fleets, whatever.
  • The one exception on type of fleet you can't gift is your home planet's command fleet (with your home command station, etc).
  • And the one other exception in terms of fleet status that you can't gift is a fleet with a dead centerpiece. This is almost always a planet you used to own but your command station is dead.
  • The button for this is directly in the fleet management window, and only appears if there are multiple human empires. If there are only humans sharing one empire, or humans who have an empire and others who have a champion, then this button does not appear.
  • This is one of the very last features needed for multi-faction multiplayer to leave alpha status and go into beta status. It's not THE last feature, but it's definitely the most complicated to code. We weren't planning on doing this today, but when working the the ship line swap feature restrictions we suddenly had a clear picture of how this would work, so just knocked it on out.

More Tech Tweaks

• Stingray has been moved out of the Technologist/Exotic tech, and into the Fusion/Breakers tech.
  • These hit bubble forcefields, which is a better fit for that tech than the exotic one.

• Vanguard has been moved out of the Ambush/Opportunist tech, and into Technologist/Exotic.
  • This isn't actually a ship that does ambush attacks. It's actually RESISTANT to ambushes.

• The Pulsar Tank has been moved out of Fusion/Breakers tech and into Ambush/Opportunist, because it gets a bonus when coming out of wormholes, which is an ambush mechanic.

• The Opportunist tech has been renamed back to Ambush, because that's clear enough.
  • The description of the tech has been updated a bit to include the bonus of the pulsar tech in its mentions, too.

• The Fusion/Breakers tech has been renamed AGAIN (this is why we do a beta), this time to Piercing.
  • The new description is: Piercing ships are those like bombers or stingrays, which focus on shields in some fashion. Fusion-style piercer ships (bombers) are able to bypass personal shields of targets and hit hulls directly. Their overall theme is smashing large or shielded targets. The other piercer ships are largely about taking down bubble-forcefields, which bombers get no bonus against.

• Thanks to Puffin and zeus for discussions on this, and in particular to Puffin for the better slotting of some of these units.

• The Anti-Shield Pulsar Tank is now a ship that uses Ambush and Piercer techs, rather than Piercer and Heavy.
  • We already had some precedents of a variant ship that switched its weapon tech, but now this switches the variants to use the original weapon tech and the new-thematic weapon tech, rather than a hull+weapon tech combo. This helps identify them as outlier variants, which should be a good thing.
  • For the sake of consistency, especially since strange variant ship combos are likely to be on the rise in future DLCs and mods, the Stalker now uses Exploitative and Subterfuge weapon techs, rather than having the medium hull tech. (The base Eyebot the stalker is based on uses Subterfuge and Medium).
  • Paralyser, similarly, now uses Subterfuge and Piercing rather than Subterfuge and Light hull. This is a variant of the Stingray that also causes paralysis.
  • If there are other ship cross-combos that seem like they should have two weapon techs, please let us know and remind us. The hope is that this will make it really clear that this is a crossover unit that is part of two categories rather than just one.
  • Thanks to Puffin, zeus, and CRCGamer for reporting on these.

• Parasitic Fusion Bomber now uses Exotic and Piercing techs, rather than Piercing and Medium.
  • Parasitic Pike Corvette now uses Exotic and Disruptive techs, rather than Disruptive and Medium.
  • Mugger now uses Exotic and Piercing techs, rather than Piercing and Heavy.
  • Thanks to CRCGamer and cml for bringing up these ships.

• Exploitative has been returned to the name Disruptive, since that is clear enough especially with the tech description now.
  • Same with Close Quarters being renamed back to Melee.
  • Thanks to cml and others for weighing in on this.

• The fourth tutorial has been updated to use the (current) proper names of the techs you need to upgrade.
  • Thanks to Puffin for reminding us.

• Generalist/Mainline tech has been renamed to Core.
  • New description: Core ships tend to be all-rounders that are for the 'line of battle' or for screening purposes. Conventional armaments of ballistics and simple directed energy weapons aren't particularly flashy but they get the job done. Ships using these systems are great at supporting or becoming the backbone of effective fleets.
  • Thanks to Chthonic_One, CRCGamer, Metrekec, and cml for the discussion and wording this time.

• Moved the Tackle Drone Launcher Frigate from Exotic to Disruptive.

• Moved Sentinel Gunboats and Veteran Sentinel Gunboats from Exotic to Core.

• Moved 'Ranger' Sentinel Gunboats from Exotic to Subterfuge.
  • May add new thing for these. Need to see if possible.

• 'Nanoswarm' Auto-Bombs benefit from both Exotic and Melee, rather than Heavy and Melee.

• 'Bounty Hunter' Raid Frigates benefit from both Ambush and Raid, rather than Light and Raid.

• 'Fusion Saw' Metabolizing Gangsaws benefit from both Piercing and Melee, rather than Heavy and Melee.

• 'Barnacle' Metabolizing Gangsaw now benefits from both Subterfuge and Melee, rather than Heavy and Melee.

• 'Aggressor' Agravic Pod now benefits from both Ambush and Disruptive, rather than Ambush and Medium.

New Ship: Ambush Carrier Frigate

• Add the 'Ambush Carrier Frigate', to give Ambush tech a bit of a boost, as well as its first Frigate type.
  • Thanks to CRCGamer for contributing from his mod in order to help round this out!

Balance Tweaks

• Vanguards and Vanguard Hydras now have an acid effect.
  • Their base damage is reduced, so their total output is the same with the acid accounted for.

• Acid effects can now scale with Mark level.
  • damage_amplification_flat_added_per_mark="10" as an example.
  • This scaling is applied to Acid Turrets and both Vanguard types.

• Logistical command stations normally give a 3.5x speed boost to all allied ships on their planet. They previously gave an addition 0.5x speed boost to ships per added mark level (so mark 3 was 4.5x speed boost). Now they give a full 1.0x per mark level (so mark 3 is a 5.5x boost).
  • Additionally, the description of the command station and of the tech for logistical command stations now make mention of this speed boosting effect directly.
  • Thanks to ArnaudB for suggesting.

• Hacked Dark Specters now give you 4 ships rather than 1.
  • Thanks to CRCGamer for pointing out how overly harsh it previously was.

• The three types of Dyson ships that you can hack for now all use the alien hull tech rather than light or medium. They already had no weapon tech.
  • They also now have a max mark 7 level of 7 rather than 4.
  • And lastly, the Bulwark now grants you 4 ships rather than 1.
  • Thanks to CRCGamer for bringing this up.

• Grenade Launcher Corvette is now Medium hull instead of Heavy, armor reduced from 110 to 70, speed increased from 600 to 800, damage increased 20%, range increased from 4,200 before lobby range slider, to 6,500 before lobby range slider.

• 'Molotov' Grenade Launcher Corvette is now Light hull instead of Medium, armor reduced from 70 to 50, speed increased from 1,000 to 1,600, same damage and range increases as the Grenade Launcher Corvette, greatly increased its engine-stun and weapon-jam effects, including increasing the available "range" to work up to engine below 14, and armor below 70.

• MLRS Corvette is now Medium hull instead of Heavy, armor increased from 50 to 70, speed increased from 600 to 800.

• 'Harasser' MLRS Corvette speed increased from 1,000 to 1,600.

• Viral Shredder speed increased from 700 to 1,000, damage increased 50%, slightly increased replication rate on top of this.

• Vampire Claw and 'Absorber' Vampire Claw health increased 33%.

• 'Absorber' Vampire Claw is now Medium hull instead of Heavy, is now the same speed as a normal Vampire Claw.

• Swapped the health of the Pulsar Tank and 'Punk' Pulsar Tank around.

• Pulsar Punk is now Medium hull instead of Heavy, armor reduced from 120 to 70, speed increased from 500 to 800, now has temporary speed bonus upon entering a planet like Raiders.

• Replaced the Ion Disruptor Frigates weapon. It is now closer to the base Siege Frigate, with 66% of the total damage, hitting up to 20 targets. Has an inverted knock-back effect, pulling enemies together for other area attacks. Still has some paralysis effect. No longer has any targeting restrictions.

• Siege Frigate and its two variants (Devastators, Ion Disruptors) now benefit from Medium hull rather than Light. Armor increased from 40 to 70.

Astro Trains

• Shifted Astro Train project requirements:
  • Dire Guardian from intensity 7 to 6.
  • Hunter Killer from intensity 8 to 7.
  • Fenrir from intensity 5 to 8.
  • Shellshocker from intensity 6 to 8.
  • Custodian from intensity 7 to 9.

• Altered descriptions of the prototypes to be prototype variants of the Extragalactic units, rather than Guardians.
  • Fun fact: Some of the Extragalactics are based on the prototypes, originally as kind of a...working concept to be used, but worked well enough.
  • They are notably weaker than their children now, and thus are getting upgrades and a kind of revamping into being the...prototypes of the thing *ABOVE* their equivalent, which makes more sense than the AI building prototypes of things it has already started making a bunch of.
  • This has the exception of the Custodian, which is being rethemed as a smaller, more easily distributed Thunderchild, due to being otherwise too high a tier to use.
  • The prototypes were all generally themed on some kind of boss, excluding Umbra which is just based on an interesting word, and Shellshocker which has no base. Their upgrades generally attempt to go further into the theme.

• Swapped models of prototypes to their equivalent Extragalactic.

• All of the Prototypes (excluding Custodian, which was already granted directly to the Warden) are now granted directly to the Hunter Fleet.
  • Thanks to ArnaudB for mentioning something that led to this.

• Fenrir prototype now has 15 Million health, up from 6. Beam weapon retooled from single target high damage beam, into a focused 'fan' beam, with no target limit and a faster fire rate, but lower damage. Now has a 'Flak Battery' weapon.
  • Drones are currently the same.

• Reanimator prototype now has 30 Million health, up from 6. Now has a 'Reanimation Blast' weapon. Drones damage increased 400%.

• Removed the drones from the Umbra prototype - no models for it. Bit strange with a majorly cloaked unit regardless, it being possibly revealed but still spawning super cloaked drones.

• Fixed Umbra prototype decloaking in a single salvo.

• Umbra prototype now has the 'Fires Through Forcefields' mechanic. Now has the same weapon of the Poltergeist. Now has 15 Million health, up from 6.

• Shellshocker prototype now has 30 Million health, up from 6. Has the same damage output as the Wyrm, but spread out among more shots. Now does more engine-stun and weapon jam (immediately hits its cap).

• Custodian now has the attack and speed bonuses from the Thunderchild, as well as its tesla weapon (but weaker). Health increased from 7 Million to 45.

• Warspite and Warspite Artillery now have 15 Million health, up from 9 and 2.5 respectively. Artillery component now has AoE on its weapon, with no target limit. 'Hidden' component overall damage condensed into fewer shots, then overall DPS increased over that.

• Ravenous Shadow now has 60 Million health, up from 39.5. Now has the same damage as the Devourer Golem, a slight bit of Vampirism, a minor speed increase, and a small range increase.

Tech Refunds

• The tech history no longer has a note at the top of it that says it is a WIP.

• It is now possible, under the hood, for us to refund techs and fleet upgrades and give back the science. These now properly show up in the tech history, including the total science spent.

• When you gift a fleet to another player in multiplayer, any science upgrades that were applied to it are refunded to you.

• Player ships are now the appropriate mark level of whatever their techs and fleet upgrades say they should be.
  • Previously, they were whatever they were set to be only if that was higher. But now that we can refund science, and transfer ships between fleets, that sort of logic no longer works.
  • This should be perfectly fine unless there is some faction (in a mod maybe?) that is granting a player a ship that is supposed to be randomly at a higher mark despite not having the techs to back it up (and not based on a min_mark_level on the ship). In that sort of scenario, the saves will now drop the ship down to the level it was supposed to be based on its min mark level and whatever tech and fleet upgrades there are.
    • We're not actually aware of any such cases, so probably it's all fine. But it's an edge case that could theoretically affect a mod or an old faction.
  • What this also means is that when you swap fleet lines between players, or gift fleets, you'll potentially see mark levels either rise or fall based on the techs and upgrades of the recipient. Science and hacking points are both resources that aren't giftable between players.

• Added a new AlternativeHeightToUseInAutoSizing that were are now able to use to make non-uniform button heights where needed in lists of buttons on places like the left hand sidebar.
  • Also added a new ExtraSpaceBeforeInAutoSizing and a new ExtraSpaceAfterInAutoSizing, which allows us to put uneven gaps between items in a list.

• The tech sidebar tab now has extra spacing between groups of techs, to help visually differentiate them. The overall spacing between other items is also slightly smaller.

• The tech sidebar tab now also has an extra button all the way at the bottom, which is shorter than the rest, and which says "Retrieve Spent Science."

• At the top of the new "tech refunds" screen, there is an "Explanation, Please" button that you can click for a detailed lore report from engineering. This is a pretty lengthy narrative lore dump!
  • Thanks to cml for requesting that we explain the mechanic. Having lore details in place throughout the game, particularly as they relate to gameplay, is always a goal of ours. Often it's hard to find enough space to write out a full explanation if it's not in the wiki or the lengthy tips section, but this button provides a really easy access point for this one.

• On the retrieve spent science window, you now see a list of all of the techs that you have actually invested science into, with how much science goes along with each.
  • You can click on these to toggle them to be refunded, or leave them as "keep as-is."
  • The tooltips for these have been adapted to show just how much ship cap you are about to lose, and how much strength you will lose.

• Fixed a longstanding issue with upgrade tooltips not showing certain numbers because the ship caps were not explicitly set, and so it was using negative numbers rather than the number of ships you actually had.

• Added a new custom_hacking_display_string_instead_of_name, custom_hacking_display_string_includes_related_string, and custom_hacking_display_string_includes_related_int on hacking types.
  • This allows us to do custom entries that read differently from normal.
  • An example usage is Retrieved {RelatedInt} Science From {RelatedString}.

• The retrieve science window now:
  • Includes all the fleets you've invested into.
  • Shows those fleets by type, and with colors for each type for easy reading.
    • This also includes tooltips for them, and the ability to C-click them, and oh man there's so much UI depth in every aspect of this, good grief.
  • Shows the total accumulated science you will get back in the top bar.
  • Shows the total cost of hacking points you will spend on the Ok button.
  • Gives you a confirmation popup, or various appropriate error popups if you've made no choices yet or can't afford the choice you made.
  • The confirmation popup explains that there is now way to get back spent hacking points, as they are gone forever.
  • It sends out a series of gamecommands to actually execute the science retrieval in for each chosen entry.
  • The cost right now is a linear 2 HaP per line of stuff refunded, regardless of how many times a line was upgraded. If we later need to do more, we can, and if we need something like a cooldown period on refunds we can also do that. But the hope is that there are compelling-enough places to spend HaP that this is a self-limiting problem inherently.
  • And finally, the hacking log now also includes the full details of the hacking points spent in this way, even though they are not traditional hacks.
  • We also made a new hack that we load from xml to get the costs and display info for this, and a new Hacking_NullHackHandler that anyone can link against for odd cases like this where a hack is handled manually but needs to be logged like any other hack.

TLDR

• Thanks to Badger for inspiring the addition of re-speccing in general, because with all these tech changes in this build we really don't want to irrevocably break people's games. The ability to re-spec is needed from that angle alone.
  • But beyond that, this is actually a really major thing that we realized is missing in this game in general, leading to players hoarding science throughout the game, or feeling "hosed by the RNG" later if they are playing on a high level and did not get lucky, or feeling like they need to reveal the map early and plan their entire game out from moment one.
  • The flow that we'd rather have is people spending in the moment, and having to manage AIP and HaP as their long-term irreversible costs. The ability to fiddle with your fleet composition, and what is stronger or weaker, is kind of at the core of a lot of the fleets system in general. And since you can't possibly know everything that might be available to you out there in the fog of war, it's only natural to want to be able to rework your empire's design as it grows from the early game into the midgame and late game.

Mod Updates

• AMU:
  • Each player faction can now set a starting amount for metal, science and hacking. By default these are exactly the same as in Vanilla right now, though they can be set higher and lower.
    • Note that setting higher-than-normal amounts will result in achievements becoming locked due to cheating.
    • Setting negative starting AIP (= starting AIP reduction) or triggering negative AIP (NOT merely setting it so) per interval (= AIP reduction over time) will now also disable achievements.
  • Updated AMU's description and scource code.
  • Fixed a nasty bug with Fireteams based on the SmartFireteamedFactionImplementationBase inside FireteamMaintenance()
    • When "attacking" planets the Fireteams now check whether or not they outnumber the enemies at the target by at least 1.5:1 before declaring the order unnecessary and resetting the targeting.
      • This effectively stopped Marauders from helping out planets they owned but were under attack, no idea how that ever worked...
      • For this purpose created IsOutnumberedOn(Planet planet, FInt outnumberedBy), which does what it says.
    • Fireteam targeting will no longer wrongfully use the lurk planet as current target for individual ship movement. This is another case of no idea how that ever worked...
    • If fireteams already have arrived on the lurk planet or are ready to attack, the target is adjacent and Marauders are not outnumbered there the individual ships now should immediately attack it. This is to save time, especially when defending, with Fireteams not waiting to gather up before the attack.
    • Thanks to ussdefiant60 for a save that demonstrated the bugs, and leading to the adjustments.
• AMU, Kaizers Marauders
  • Rejigged a lot of logic on how the GlobalImportantRollupHolder and External Data are loaded, so that even at the very first sim step the data already is there in its entirety.
    • For this purpose created an abstract InitializeExternalData() method in SmartFactionImplementationBase that is automatically called by OnDeterministicThreadOnly_DoInitialLogic(), which itself is automatically called by DoOnLocalStartNonSimUpdates_OnMainThread(), which automatically calls GlobalImportantRollupHolder.Initialize().
    • SmartFireteamedFactionImplementationBase will additionally call the new abstract LoadFireteamValues() can be used to set min and max base fireteam sizes, but does not demand doing so.
      • The effect of this is that the initiailzation of data is now mostly handled at the moment the galaxy is created and no longer when the game is unpaused. The initial lag spike when a save is loaded for the first time, or when a new game is started thus no longer happens then, but during loading of the factions - whenever that may be.
      • End result: Better handling of data, less lag after unpausing, potential new bugs.

• AMU now fully supports DeterministicExecutors for launching simple scripts of any kind on game session start and mapgen.
  • They have already been secured against individual errors, even if one of the executors in the middle errors out the others will at least attempt to run through properly.
  • Planetary science and hacking points are now re-set on game session start. This fixes the amounts getting stuck on wrong numbers when restarting the game, loading saves or creating new games with other settings.

• More System Defenders 1.21 by CRCGamer - Escorts & Vanguards
  • Most recent changes:
  • Ambush Carrier removed after donation to base game. Mod still enables station-keeping and AI guardian variants however.
    • Note from devs: special thank you to CRCGamer for this! This helps round out a base-game tech that otherwise had too few things in it.
  • Escort Carrier and AI guardian variant reworked. Now has a parasitic main weapon and deploys Vanguard drones instead.
  • Because of the recent tech swaps of certain strikecraft in the base game the previously used Stingray drones are now deprecated.
  • Because of the more significant combat effectiveness of the Escort Carrier with said changes ARS and random Fleet counts normalized from 9 to 6.
  • To offset the overall reduced ship count the durability of the Escort Carrier is raised by about ~20%. Metal cost raised from 65K to 75K.

• AMU:
  • All Factions basing their Stage 0-3 and LRP threads off the SmartFactionImplementation will now gracefully handle ThreadAbortedExceptions instead of throwing errors.

Pathfinding Performance

• FindPath() is being retired and replaced by FindPathFreshOrFromCache().
  • This latter returns a new, cached, PathBetweenPlanetsForFaction objeect rather than just a list of planets.
  • If a programmer alters that list of planets after they get the object, they need to be stoned lightly because it will break all sorts of things.
  • The idea here is that we are calculating paths way too frequently, often thousands of times in a single second, and most of the paths we are calculating are the same during that span.
    • This is very true for things like checking the danger of a path (for some of the pathfinding variants), as well as for fireteams in general, as well as for ships that are undergoing fleet movement orders of certain sorts.
  • This overall change should lead to a dramatic improvement in performance in a number of areas of the game, but mostly in the background AI processing for factions.
  • The old FindPath() method is still there, so existing mods will still work, but it's been marked as obsolete so that mods will need to update to the newer, more efficient methods before they can be properly compiled/updated by the mod author for a new version.
  • Also added a new PathfindingCacheForFactionOnContextObject that gets cached on ArcenSimContext, so that we can have thread-safe separate caches per context per faction (main sim, long term planning, whatever).
  • In our innermost calculations for finding paths, we are also now reusing lists better, and we are using the [ThreadStatic] attribute for the first time in this program. This turns out to be extremely useful, and we just recently learned about it: https://docs.microsoft.com/en-us/dotnet/api/system.threadstaticattribute
  • Additionally, even if the number of paths generated was the same (it's very much not), we're also now saving a lot of transient GC allocations, which is nice.
  • Note to modders: if you were using FindPath, generally you are now using FindPathFreshOrFromCache(). This gives back a PathBetweenPlanetsForFaction object with a PathToReadOnly on there, and you should ONLY read from it, not ever alter it.
    • It is also possible for it to give back a null PathBetweenPlanetsForFaction object, so you do have to check for that now (that generally means an invalid pathing attempt, or already at the target, etc, and is there for efficiency's sake.)
  • Another note to modders: if you were getting a pathfinder directly and then trying to call FindPath on the pathfinder... that's really something to be avoided, now, because it won't cache at all.
    • To work around that, you can call InnerFindPath_RawSinglePathfinder() and pass in the info as well as the pathfinder of your choice, and you'll get back either a populated PathBetweenPlanetsForFaction or null, all of it cached properly and thus more efficient.
    • There's nothing that will FORCE you to use pathfinders this way -- you just can switch to passing in a list of planets to be filled rather than just getting one in return -- but unless you are absolutely positive that this pathfinding call gets called less frequently than once every few seconds, you really should use the cached versions.
  • In the escape menu, at the bottom of the performance section, we can all now see the number of path recalcs, and the number of paths from cache.
    • For reference, in a fairly large savegame with a fair number of late-game factions (5 hours in, near the end), running the game for about 20 seconds leads to 1200 path recalculations, and 4500 path pulls from cache. Each pathing call is pretty intense, and this is not remotely the most expensive scenario we have around, so this is a huge potential boost in performance, especially for older CPUs.
      • In the case of certain other savegames that we saw logs of, where the Hunter fleet thread was getting terminated for taking more than 30 seconds (yikes!) to do its work, it will be interesting to see what these numbers are like and if that is no longer a long-running thread. Either way, these performance improvements are worth it. But if not, then we can use that specific savegame to do some testing and tune things different ways for the hunter specifically.
  • Thanks to Lightjolly for the report.

• Fixed some bugs with outguard using the wrong variable to do things like check if they are able to get to a planet from where they are now. No idea how impactful the bug was, but it's fixed now.
  • Fixed another exception that could happen in the fallen spire if it was unable to path to a king unit, or had a path shorter than 1 hop to a king planet. Not sure anyone ever hit that, either, but it was also busted.

Beta 2.726 Tech Tuning

(Released January 30th, 2021)

• MLRS Turret has been renamed to Fusion Turret.
  • This was in the Fusion tech already, while MLRS Corvettes are in Splash.
  • To make this fit better, the range of the fusion turret has been increased a small bit, the damage doubled, shot count halved from 10 to 5, and it now does 80% damage through personal shields.
  • New description: Built for hitting small groups of ships or structures, ideally those that are large and shielded. Strikes past personal shields to do direct hull damage.
  • Thanks to Chthonic One for suggesting.

• Problem is, in DLC1 we already had a Fusion Turret, but that was not actually firing through shields, which is what fusion damage is. So that was already kind of mis-named.
  • This turret is now called the Dispersion Turret (after its Disperser Bolt weapon).
  • This turret already does a lot more damage to ships with personal shields, but does not cut through them at all.

• Revised some of the code around getting the tooltip text for ships/structures to use bitwise flags rather than a simple enum. This lets us make more complicated requests to it for various purposes.
  • When you are going to hack something to get a ship line, and that hack would cost you AIP (as with an FRS), the tooltip for each option now shows you the AIP cost that you would incur.
  • Previously, the only way to see this AIP cost was on the tooltip for the structure you would be hacking!
  • Thanks to Sigma7 for reporting.

• The warning notification at the top of the screen for the devourer golem(s) now actually shows the text "devourer golem" directly on it.

More Tech Name Tweaks

• After discussion on our new naming_and_balance section on our discord (https://discord.gg/snQBW4g9Gr), the following further changes are being made to tech names:
  • Atomic/Fusion tech is being renamed again, this time to Breakers.
    • Reasoning: the name fusion was not liked because it implied all were fusion weapons, which they were not. The name atomic was a bit on the unclear end for various reasons, and just felt thematically off. Breakers as a concept actually describes what these ships do.
    • Revised description: Breaker ships tend to be very powerful for whatever their size is, but can be expensive. Many of the tanks and bombers are in this category. Many, but definitely not all, breaker ships are able to bypass personal shields of targets and hit hulls directly. Their overall theme is smashing large or shielded targets.
  • Disruptive/Contravention is being renamed to Exploitative (only downside is that this seems pretty similar visually to Exotic).
    • Reasoning: both of the old names were a bit on the odd or generic side. These ships overall are about countering other ships, but another way to look at it is that they are built around exploiting weakness of other ships in unusual ways.
    • Revised description: Exploitative ships vary a lot, but all counter something. Some counter very low-armor ships, others counter very high-armor targets. Others use weapon jamming or tractor beams to prevent enemies from acting. They are all about exploiting unusual weaknesses.
  • The Raid/Blitz tech has been renamed back to Raid.
    • Reasoning: this is clear, visually distinct as a word in the list of other techs, and does not have any unfortunate historical connotations.
  • The Generalist tech has been renamed to Mainline.
    • Revised description: Mainline ships tend to be all-rounders that are for the 'line of battle' or for screening purposes. These tend to be an excellent backbone for a powerful fleet, anchoring it with solid firepower and reasonable defenses.
    • Reasoning: these are either screening ships or "ship of the line" or "wall of battle" or "line of battle" ships. We could have just gone with "Line" instead of "Mainline," but then talking about "ship lines" would become an Abbott and Costello sketch. Generalist was so vague and boring that it was hard for anyone to follow.
  • Splash/Area of Effect ships have been returned to the name Splash.
    • Reasoning: splash is clear enough, visually distinct, and single-word. Area of Effect is too long for many places it would be shown, and the acroynm AOE is potentially confusing.
    • Revised description: Splash ships are usually meant for crowd control, or hitting lots of targets in either a focused beam or area of effect. This might be shrapnel damage in an area that shreds low-armor targets, or it might be massive electrical discharges with various effects.
  • Mobile Orbital Fortifications have been renamed to Mobile Fort, because the other name fit in no windows.
  • Thanks to Captain, Tim_Fragmagnet, Rifi, cml, tadrinth, CRCGamer, CRZgatecrusher, Zdrgn, Ozone, Isiel, DEMOCRACY? DEMOCRACY!, and -NR-SirLimbo for weighing in on this round. We expect there might be a few more rounds in the short term as people get a feel for this.

• Techs are now sorted by their display name instead of their "internal name" so that we're sure that they are always alphabetical when name changes happen in a subcategory of them.

Mod Updates

• CRZgatecrusher's "The Reprocessers" mod updated to make targeting homeworlds super unlikely)

• More System Defenders v1.20 by CRCGamer now included:
  • Most recent changes:
    • Station-Keeping ships have had their techs remapped to match the changes with base station keeping ships of being Station-Keeping mono-tech.
    • Artillery Cruisers can be found in the wild rarely. You are allowed a pair of them at economic and military commands (so swaps don't suddenly eat 900K metal)
    • Flypaper Frigates are only found in the wild and are no longer provided immediately due to the nature of station-keeping ships being mono-tech.
    • The Gorgon frigates have been removed from the economic command as a default and moved to military. They can still be found in the wild to unlock for all.
    • Cruise Missile Battery paralysis weapon timer reduced to 17 seconds from 30 to match AI version. Frankly the uptime on the CC was too poor.
    • Augmented Artillery Cruiser AIP cost from FRS reduced to 30 from 50 since most fleet ships now don't use shields (and they buff shields).
    • Several ships AOE radius improved by 100-200 to more reliably hit full target allotment.

Beta 2.725 Techs, Metal, Energy And Ceasefires

(Released January 29th, 2021)

• Fixed a divide by zero exception that could happen in Helper_SendExoGalacticAttack_SingleExoTarget() when no units were actually being deployed.
  • Thanks to ussdefiant60 for reporting.

• If an exception happens when checking for whether an order is valid or not on a multiplayer client, it will just assume the order is invalid and get on with things without throwing an exception. If it happens on a host or in single-player, it will now give a bit more information before doing the same.
  • Thanks to Badger and his play group for reporting.

Limiting Coprocessors

• Coprocessors are a form of AIP reduction that reward you for destroying all of them, but penalize you in the short term if you only do some of them.
  • These give a bit too much power in terms of AIP reduction in average games with only a single AI. The original balance of these was really more based around AIWC, where there were always two AIs.
  • Starting with new games in this build, Coprocessors will only seed if there are at least two AI factions that are hostile to the first player (and, continuing with rules from the last while, also only seed if there are 3 or fewer human empires in a multiplayer game).

Metal Storage Improvements

• Added a new added_metal_storage_per_mark_level, which lets us give more metal storage to higher-mark versions of ships that give you metal storage.

• Home command stations continue to give 2.5m base metal storage, but now give 500k extra metal storage per mark they are upgraded.
  • All other command station types used to give a flat 300k metal storage, but that's no longer the case.
  • Economic now gives 300k, plus 100k extra per mark level.
  • Logistical now gives 600k of storage, plus 300k extra per mark level.
  • Military now gives only 50k of storage, and another 50k per mark level.
  • The spire mining facility in DLC1 continues to grant its base 1m extra metal storage, but now also gives an extra 100k per mark level upgraded.
  • Also!? The added metal storage amounts are now actually shown in tooltips for the first time.

Brownout Refinements

• The mechanics of brownouts has been made a lot more flexible, to avoid some of the "temporary accidental" situations that people were running into.
  • Previously, brownouts would start immediately as soon as you had a negative energy balance, and they would last for 60 seconds after power was restored.
  • Now you can run up to a -50k energy deficit with no ill effects.
  • Additionally, you have to run below -50k energy for at least 10 seconds straight before a brownout begins.
  • THAT said, it now will last for 2 minutes rather than 1 minute after you restore power to at least -50k.
  • We may need to explore more warnings to players during the 10-second period if they have gotten into the situation of being about-to-brownout, but it may not be an issue since most things that "put people over by a little" will still put them over by less than 50k. So that would prevent further building, and let people realize they need to build more energy.
    • In the event of a sudden dramatic energy loss, such as a command station, it's very likely that a brownout is unavoidable without deleting a ton of turrets and such, and we don't really want to give people a warning too much with this case. We're trying to avoid fluctuations causing forcefield outages, not an early warning system for players to pause and delete stuff after they lose a command station. Some will use it that way, which is probably fine, but let's see how this does.

Ceasefire Mechanics

• Added a new ability for ships: prevents_all_weapons_fire_on_planet="true"
  • CEASEFIRE: Prevents all units from firing weapons while it is on a planet with them.
  • This ability has been granted to the Zenith Trader.
  • Expect to see some more of this with a new unit in DLC2, as well!
  • Thanks to Leif, Tim_Fragmagnet, StarKelp, Chthonic_One, Kendrickorium, and Isiel for the discussion that led to this.

• A notification at the top of the screen is shown if a planet is under a ceasefire, and either 1) you have ships there, or; 2) you or another human player own the planet.

Tech And Metal/Energy Income Updates

• There is now a description field on techs, which lets us give explanations about them.

• The Metal Generation tech has had its per-unlock cost reduced from 4k for each of three levels to instead 3k for each of three levels.
  • It also now has the following description:
    • Early on, consider directly upgrading your home command station fleet (from the fleet sidebar tab) for stronger gains. The metal generation tech is more effective when you have a lot of planets.
  • Thanks to CRCGamer for suggesting.

• Previously, Metal Harvesters gave 60 metal per second at mark 1, plus 30 per mark as they went up in levels.
  • They now give 75 and 50, respectively.
  • Thanks to CRCGamer for suggesting.

• Command: Economic tech new description:
  • Useful when you have a lot of this kind of command station. Alternatively, consider investing directly in the fleet of each planet (fleet sidebar), since that also upgrades metal harvesters and turrets located there. This tech improves ALL of your economic command stations, primarily with more metal and energy and a bit of extra sturdines.
  • Costs adjusted down from 4k each for the first 3 mark levels, to 3k each instead. Fourth mark level remains 4k.
  • Thanks to zeus for helping figure out what to say here.

• Command: Military tech new description:
  • Useful when you have a lot of this kind of command station. Alternatively, consider investing directly in the fleet of each planet (fleet sidebar), since that also upgrades metal harvesters and turrets located there. This tech improves ALL of your military command stations, primarily with more attack strength and hull health.
  • Costs left alone for now.
  • Thanks to zeus for helping figure out what to say here.

• Command: Logistical tech new description:
  • Useful when you have a lot of this kind of command station. Alternatively, consider investing directly in the fleet of each planet (fleet sidebar), since that also upgrades metal harvesters and turrets located there. This tech improves ALL of your logistical command stations, with a mix of economic and military benefits.
  • Costs cut in half to 1.5k science for the first two marks, and left the same at 3k for the next two after that.
  • Thanks to zeus for helping figure out what to say here.

• Metal generated by economic command stations moved from 250 base and +100 per mark to being 250 base and +200 per mark.
  • Energy generated by economic command moved from 400k base and +100k per mark to being 400k base and +200k per mark.

• Metal generated by military command stations moved from 20 base and +20 per mark to being 60 base and +40 per mark.
  • Energy generated by military command stations adjusted up from 20k base and +5k per mark level to 40k base and +20k per mark level.

• Metal generated by logistical command stations moved from 60 base and +20 per mark to being 120 base and +80 per mark.
  • Energy generated by logistical command stations adjusted up from 200k base and +50k per mark level to 250k base and +100k per mark level.

• Science and hacking points of logistical command stations is now 10 per second rather than 2 per second like the other command stations.
  • You still get the same amount, but you get it faster now. This makes logistical a better choice for a place you don't plan to keep long.

• The "Citadels" tech has been renamed Mobile Orbital Fortifications, and now applies to Battlestations again (not just Citadels).
  • Thanks to "Z" for suggesting the name, and ArnaudB and loads of other people for suggesting that battlestations get this again.
  • New description: Upgrades any battlestations or citadels you may have or acquire. These structures allow you to build extra defensive turrets, mines, etc on either your own planets or as a 'beachhead' on enemy planets.

• The "Forcefields" tech has been given the following description:
  • Upgrades all bubble forcefields that you are able to build at your various command stations. Also improves mobile bubble forcefield generators (Forcefield Frigates, etc).
  • Thanks to Z for suggesting.

• New Minefields tech description:
  • Minefields are constructed by either command stations, battlestations, or citadels. They are not super common, so you may only want to invest once you are sure you are going to have them.

• The "Sentries" tech has been renamed to "Station-Keepers"
  • The Station-Keeping Assault Frigate and the Station-Keeping Watchman Frigate both now benefit from the Station-Keepers tech instead of other random techs. The Sentry Frigate already used this tech.
  • Alien station-keepers that you can capture for now do not use this tech. For station-keepers added in mods, they may or may not use this tech (that's up to the modder). But our intent would usually be that they do at least partially.
  • New description for this tech: Station-Keepers are typically mobile frigates that are bound to the gravity well of a particular planet. They are there to help defend a command station as part of its fleet.

• Renamed the "Metal Generation" tech to "Orbital Mining," since it is not remotely the only way to get metal.
  • Added this to its description: This upgrades the efficiency of a few structures on your home planet, and of all the metal harvesters you build on every planet, but that's all.
  • Thanks to Z for ideas on this.

• New description for Engineering: Upgrades the effectiveness and durability of the engineers that you can build from command station or from certain mobile factories.
  • Thanks to Z for ideas on this.

• The Ambush weapon tech has been renamed to Opportunist.
  • New description: Opportunist ships are largely characterized by temporal exploits. Things like 'Bonus damage to units that have been on planet X seconds.' Overall these ships take advantage of the environment in some way. They also include things like sniper units that use the speed of enemies against them.
  • Thanks to ArnaudB, Z, Zeus, and Kahuna for the discussion that hashed this out.

• The Concussion weapon tech has been renamed to Artillery.
  • New description: Artillery ships generally fire concussive shaped charges, generally over medium-to-long distances. Most of the units are siege-oriented, and they often specialize in killing low-armor targets. A number of them also employ knockback, which can keep smaller enemies from returning fire.
  • Thanks to ArnaudB, Zeus, and Kahuna for the discussion that hashed this out.

• The Disruptive weapon tech has been renamed to Contravention.
  • New descripion: Contravention ships vary a lot, but all counter something. Some counter very low-armor ships, others counter very high-armor targets. Others use weapon jamming or tractor beams to prevent enemies from acting.
  • Thanks to ArnaudB, Zeus, and Kahuna for the discussion that hashed this out.

• The Fusion weapon tech has been renamed to Atomic.
  • New descripion: Atomic ships tend to be very powerful for whatever their size is. Many of the tanks and bombers are in this category. Many, but definitely not all, atomic ships are able to bypass personal shields of targets and hit hulls directly.
  • Thanks to ArnaudB, Z, Zeus, and Kahuna for the discussion that hashed this out.

• The Generalist weapon tech is keeping its name, and has this new unhappy description:
  • Generalist ships tend to be all-rounders that are not overly good at any particular task. These tend to be an excellent backbone for a powerful fleet, anchoring it with solid firepower and reasonable defenses.

• The Melee weapon tech has been renamed to Close Quarters.
  • New descripion: Close Quarters ships include all of the melee-ranged ships that are focused on very quickly reaching targets and either exploding them or slicing into them. But it also includes a variety of very short-range dangerous turrets and other units that focus on point defense.
  • Thanks to Ath, Z, and Kahuna for the discussion that hashed this out.

• The Raid weapon tech has been renamed to Blitz.
  • New descripion: Blitz ships usually trade hull integrity for speed and attack power. These are not the units you want to use in a stand-up fight, but they're great for raiding and for hit-and-run attacks. When it comes to turrets, most of them are cheap and lightly defended for their size, but quick to build and pack a punch.
  • Thanks to ArnaudB, Z, Zeus, and Kahuna for the discussion that hashed this out.

• The Splash weapon tech has been renamed to Area Of Effect.
  • New descripion: Area Of Effect ships are usually meant for crowd control, or hitting lots of targets in some fashion. This might be shrapnel damage in an area that shreds low-armor targets, or it might be massive electrical discharges with various effects.
  • Thanks to Z for the discussion that hashed this out.

• The Subterfuge weapon tech now has this description:
  • New descripion: Subterfuge ships are often cloaked, but not always. Many of them apply debuffs, such as temporarily disabling engines. A few are giant-killers, focusing devastating blows on big targets.
  • Thanks to ArnaudB, Ath, and Kahuna for the discussion that hashed this out.

• The Technologist weapon tech has been renamed to Exotic.
  • New descripion: But they include gems like parasites (for taking over enemy ships) and stingrays (for specially targeting bubble forcefields). Overall these involve a lot of electronic warfare.
  • Thanks to ArnaudB, Z, Ath, and Kahuna for the discussion that hashed this out.

• The Light hull tech now has this description:
  • Light ships come in all sizes and capabilities, but tend to have low amounts of armor. This isn't a problem except when up against enemies that use weaponry that specifically targets that low armor.

• The Medium hull tech now has this description:
  • Medium ships come in all sizes and capabilities, but tend to have midlevel amounts of armor. There are fewer of these, but they are less likely to be the target of bonuses against either high or low armor ratings.
  • Additionally, its cost has gone down from 7k and 10k to 5k and 8k.

• The Heavy hull tech now has this description:
  • Heavy ships come in all sizes and capabilities, but tend to have large to VERY large amounts of armor. High amounts of armor does not confer any natural bonuses, but tends to coincide with high amounts of hull health. There are some specialist high-armor-killer units out there, but beyond that fewer ships get bonuses against them.

• The Turret hull tech now has this description:
  • Turrets are guns that nobody bothered to strap an engine to. They're typically built by your command stations, battlestations, or citadels. Usually in friendly territory, but you can also use them in offensive beachheads.

Mod Updates

• Made Kaizers Marauders compatible with Vanilla, no longer requiring DLC 1, and removed its requirement:
  • The CostBasedShipSpawner inside AMU now no longer throws an error but quietly remarks in the debug log if there's no ships for a tag (such as for example spire-raiders).
  • Fixed an exception inside the Kaizers Marauders Stage 2 logic where it was always trying to read the starting spire debris even though without DLC 1 the setting does not exist.

• AMU Prototypes:
  • Created multiple generalized GameEntity prototypes within AMU for: Strike Craft, Frigates, Other large non-stackable ships, Drones, Drone Projectiles, Small/Med/Big turrets.
  • Both Kaizers Marauders and Extended Ship Variants now use these. This will do very little (hopefully) but now makes a good attempt at "sorting out" targeting and protection priorities, collision priorities, drones not aggroing Guard Posts, Drone Projectiles not being swappable and a few other little things.
  • It will also speed up any further mod development and cut down on redundant work (including modding upkeep if things change or balance is adjusted), and can do so for any other modder potentially working with AMU.

• More System Defenders has been updated by CRCGamer in order to move the Station-Keepers over to that Station-Keeping tech.

• "Cosmetic" mods have been a misnomer for a while, and are now called "Save Safe" mods.
  • Essentially these are invisible to savegames whether they are on or not, but they may change more than cosmetic items.
  • The file ModIsConsideredCosmetic.txt can still be used, but ModIsSaveSafe.txt is the new standard. We'll continue triggering the appropriate flag from either.
  • Thanks to Puppet Master for suggesting.

Beta 2.724 Mod Bonanza

(Released January 28th, 2021)

• When waves are launched, they can come from a variety of sources and target a variety of other factions. In the prior build, it was trying to throw all of these into the relentless wave ai faction, but that's not actually correct.
  • In the case of something like the AI Reserves launching a wave, they should own their own ships that they launch at (presumably) you. Fixed.
    • Thanks to zdrgn for reporting the exception that this was causing.
  • Additionally, when an AI actually IS launching a wave, but it's targeting some NPC faction (another AI in a civil war, a marauder faction, whatever), it no longer uses the AI faction or the relentless wave faction -- instead, it just feeds these new units directly into the AI's hunter fleet, targeted at that NPC faction. This makes them the most effective, and skips a few steps that could be error-prone and maybe see these ships come after you instead, etc.
    • Thanks to Badger for suggesting this part.

• A surprising number of areas of faction code in particular would report fatal errors into the log, but not actually to the screen. This would often lead to you not knowing there was an error, and probably having laggy performance from all the silent logging. Now we'll all know what is happening and we can fix the root problems when they come up.

• In the event that the game has had some critical failures previously, it's possible that UpdateAllVisuals() will also give you some errors. Now it gives better errors, and is in general less likely to give errors in the first place.
  • Thanks to Kendrickorium for the report.

Settings Menu Improvements

• The performance section of the settings menu has been updated to have a variety of subcategories that make it easier to visually parse and find what you need.
  • Same for the Game section of the settings menu.

• Rather than the Display category automatically entirely going to the graphics settings file, there is now a new is_stored_in_graphics_settings_file="true" that is set that causes those settings to go there.
  • That allows us to add other things to the display category that might not have the same relevance.

• In the past, we had categories on settings that were simply defined as part of the setting. Now we explicitly define categories and even optional subcategories for settings.
  • The explicit categories allows us to make new categories that are empty of traditional settings, such as having expansions and mods show up as their own sections on the sidebar.
  • The subcategories allows us to visually organize really long sections of settings if we need to. Things like the debug window can otherwise just get plain overwhelming.

• The settings menu has in general gone through a bit of a redesign:
  • The HUD and Tooltips categories have been made subcategories of display.
  • There is one overall Camera category, with the "all cameras," "main camera" and "galaxy camera" bits as subsections.
  • The networking tab now has a subsection for the "Your IP Address Information" so that is more clear what it is about.
  • Kaizer's Marauders now implements a new mod-specific subcategory of the otherwise-invisible Custom sidebar category, which lets it appear without being in a strange position. Other mods can also easily do this.

• Mod descriptions can now only show 1000 characters in a tooltip. Beyond that, it cuts off with an ellipsis and gives a "read more" button over to the right of the mod's normal buttons. You can then click that for a full popup screen that lets the mod author write however much they want.
  • Thanks to Puppet Master for suggesting, and NR SirLimbo in particular for making it needed. ;)

• The automation and audio sections of the settings menu now have subsections.
  • Wow this makes things so much easier to find!
  • Additionally, the warning triggers have been moved into the Game section under warnings, so all of that is in one place.

• The display section of the settings menu has been substantially subsectioned, and its hud and tooltips sections now have more sensible subsections as well.
  • The show random AI types has also moved to the game section, rather than being in the HUD section.

• The camera section of the settings is further split up so that you can find options in there.
  • A few options for the galaxy map camera that are really mostly about color grading have been moved to that new subsection of the display section.
  • Here again, the increase in clarity is just astounding.

• Most of the networking section has been split out to be part of the debug menu, since that's really what most of that is.
  • The debug menu has then been broken into a ton of subsections by function, so that we can actually find things in there!

• Whenever you open the settings menu, if there are more categories than can be displayed without scrollbars, it scrolls back to the top of the category list rather than leaving it wherever you last left it.

• Whenever you change categories in the settings menu, it now scrolls the main content part of the window back to the top rather than leaving it wherever you last left it.

Mod Integration Improvements And Related

• The Expansions and Mods now each have their own category in settings, which has a count of how many of them you have enabled out of how many in total there are.
  • • Thanks to Isiel for suggesting.

• Clarified Civilian Industry tooltip to say "No expansions required" rather than "Requires no expansions."

• Mods can now declare that they have prerequisites in the form of other mods or expansions.
  • These are done by listing the other mods in a RequiredMods.txt file, one line at a time, or expansions in a RequiredExpansions.txt file, one at a time.
  • With these present, the case where a player enables a mod but not its prerequisites is now impossible (that caused all sorts of exceptions and confusion).
    • In the case of expansions, the game will not enable those because of a mod's request, but it will inform the player of all the mods that could not be enabled because of what expansions were missing or disabled.
    • In the case of mods, if the prerequisite mods are present and can themselves be enabled, then it will just auto-enable them for you.
      • In the case of a prerequisite mode that can't be found or can't be enabled, it just tells the player about that.
  • This whole process allows for prerequisites up to 30 deep, which would be an insane amount, and beyond that it won't auto-enable things for you because it assumes (probably correctly) that somebody set up circular references (two mods that both depend on each other in order to turn on). In the case of circular references, those can still be enabled manually.
  • An example use case of this is that if you enable Kaizer's Marauders but not AMU, it used to error out mysteriously with very funky errors. Now it instead will just turn on AMU.
  • Another example use case is that if you don't have DLC1 installed or enabled, but you turn on Spire Railgun Shop, then it would do another set of funky and unclear errors. Now it just turns Spire Railgun Shop off and tells you why.

• Mods that we distribute has been updated as follows:
  • Kaizer's Marauders now explicitly requires DLC1 and AMU.
  • MacrophageHistiocytesSpireIntegration now explicitly requires DLC1 and MacrophageHistiocytes.
    • It's possible that this could be re-coded to just be a part of the MacrophageHistiocytes mod that turns on and off, like what happened with MoreSystemDefenders and similar, but for now this is fine.
  • SpireRailgunShop now explicitly requires DLC1.

• Fixed a bug where we were doing the clear method of SpecialFactionData AFTER we had disabled a mod or expansion and right before a reload, which meant that they couldn't find their code and thus generated an error if they had custom code. The error was harmless but confusing, and sets a bad precedent of people ignoring errors that are not actually harmless.
  • Anyhow, now this all works properly, and we also have some added debugging info in place to find things like this in the future. But so far this works just fine. This was affecting both Kaizer's Marauders when they are disabled, and Civilian Industries when they were disabled.

• It is now possible to designate mods as being "framework mods" by giving them a ModIsConsideredFramework.txt file.
  • This basically lets us get them out of the main category of mods so that players aren't scrolling past them in confusion and wondering what functionality they give (aka, no function -- they are just used by other mods, and other mods will turn them on as-needed now).

• In the mods tab, there are now sub-categories for both cosmetic mods and framework mods.
  • We thought about having a separate tab for the cosmetic mods, but at this point that seems like it would make things more difficult rather than less.

• Fixed a bug where when you hot-reloaded the mods or expansions, it would make for doubling or tripling of the music background track, and also make all the volume progressively more quiet each time. This is because of us reloading the sound root and volume mixer, and that's just incredibly fraught as a thing to do.
  • Normally the sound mixer and root themselves are not something that would be modded, but in the event that someone chose to do that, we no longer support hot-reloading those particular things. Whatever the sound root and mixer are at the time the game is launched are the ones that will stay until you restart the program (whether it's a mod version or our main version).
  • Again, we really can't think of any possible reason to create a new sound mixer or sound root, since actually doing adjusting the mix is possible without that step, so this is probably entirely moot other than fixing this bug itself.
  • Thanks to Tim_Fragmagnet for reporting.

• If you try to load a savegame that you are missing some expansions and mods for, and you have those expansions and/or mods on your machine but just not enabled, then the game now asks if you want to enable them. It then does a hot-reload, including any prerequisite chains, and you're able to then quickly load the save after that.
  • This is particularly helpful for us developers, but really it's handy for anyone who plays with various configurations. Do note that this does not disable any extra expansions or mods you might have installed that the save does not.
  • In the event that the save is referencing a mod or expansion that you don't have on your machine, or it's referencing a mod with prerequisites that you are missing, then it will just give you the warning message and fail to load the save like it always has.
  • Note that this only works for single-player saves right now, not joining multiplayer games and trying to match the host.

Three New Mods By ArnaudB

• New cosmetic mod by ArnaudB: Raising The Floor: Multi AIs Adjustment
  • Fighting multiple AIs at the same time is a daunting challenge, often spoiled by the immense increase of threat once a single AI Homeworld dies, even when there are multiple others AIs around to lessen the importance of such a deed.
  • This mod causes the AI Overlord to reduce AIP by 200 when killed, with the total AIP gained from killing an AI Homeworld raised from 170 to 220. The reason behind this effective 20 AIP increase is to make the AIP Floor rise.
  • The remaining AIs 'forget' some of what you have done with the loss of one of their own, but they are less inclined to underestimate you.
  • The driving force behind this mod is the uninteresting dynamic forced by the current meta. With vanilla settings it's always better to clear the paths to all AI Homeworlds, then smash them in a row in a grinding match that seems to last forever.
  • With this change, the player is heavily incentivized to remove AIs progressively over the course of a game. The AIP floor will increase a lot, forcing a progressive heightening of the forces in the galaxy, but without the excessive and sudden all-out war that is the vanilla experience.
  • This has effectively no impact on games with a single AI. You do get more AIP when killing the first phase of the Overlord, but at this point you are so close to finishing the game that 25 more AIP from killing it won't do anything. You still get the AIP reduction when you win... but you have already won.
  • Enjoy your multi-AIs games and don't hesitate to give feedback!
  • (This doesn't disable achievements)

• New cosmetic mod by ArnaudB: Speedy Citadels
  • Increase the speed of citadels to 600, the same as battlestations. You can now mix them and clean up your controls group.

• New cosmetic mod by ArnaudB: Supercharge Raid
  • Add supercharge back to raider and their variants.

The Reprocessers mod by CRZgatecrusher

• New mod by CRZgatecrusher: The Reprocessers
  • The Reprocessers are a faction that inhabit deep space and occasionally attack places for research purposes. Comes with 4 unique phases:
  • Phase 1: spawn special 'bombs' and attempt to obtain metal
  • Phase 2: Build up attack. nothing happens in this phase as they are 'organizing' an attack
  • Phase 3: Launch the attack. launches an attack whose size is based on the metal obtained. the faction gains research points when it kills things
  • Phase 4: Attempt to research things. this is the time the faction attempts to unlock a new unit. returns to phase 1
  • T1 units are similar to player units to almost identical
  • T2 units are unique and have special features like a lifecycle or actual replication.

Beta 2.723 AI Accountant Fired, And Relentless Waves

(Released January 26th, 2021)

• Hunter fleet fireteams can now get a bit larger before making new ones.
  • Thanks to Badger for adjusting.

Wave Balance

• Wave adjustments:
  • The AIs are more likely to send a couple of different ship types rather than just one monolithic type when they send a wave.
  • The AIs are more likely to send guardians with waves.

• The "wave_power_ratio" field from the AI difficulty level has been removed.
  • This was something that was designed to make waves actually hit hard enough to be damaging. Aka, "all waves at difficulty 8 must be at least half as strong as the defending world's power."
  • In practice, with really well-defended homeworlds in particular, this was making it so that those would NEVER be attacked, and the AI would be incredibly predictable in that regard, with the new stronger wave balance.
  • Given that waves have been so weak as to be basically nonexistent for the last year or more, anything is better than that, and basically it's preferable to have more weaker waves that might interact interestingly with the Hunter rather than occasional apocalyptic waves that come without warning and also feel unfair.
  • This is a mechanic that only kicked in on difficulties 6 and up, and you could really see how much more poorly the game played on difficulty 8 in the new stronger-waves style partly because of this. It was a great idea when it was first introduced, but the game meta has moved on.
  • HOWEVER, this underlying mechanic is still useful for reconquest waves in particular, where the AI really intends to win a given battle in order to retake a given planet. So we've kept the underlying mechanic for those specific purposes.

• There is a new "Relentless AI Wave" faction for each AI, which works (for now) exactly like the Tsunami CPA faction does.
  • Aka, they attack mercilessly and try to push in more and more on you, and they never get donated over to the Hunter faction.
  • In previous versions of the game, for the last year or so at least, the waves were so small and were so ineffectual when it came to against-player waves that the Hunter was not getting fed, anyway.
    • If the new version of the waves winds up feeding the Hunter, then that just makes the Hunter suddenly unbalanced in terms of how hard it is.
  • In the new version of the game, we want these stronger waves to be a time-limited thing. They arrive, they try to hurt you (or their other target) very badly, and then they die. They pursue you until death.
    • The hunter, of course, can coordinate with these just fine. And we can refine their behavior over time as needed.
  • One thing that is not clear is how well this will work with multiple AIs launching waves at each other (when they are not allied).
  • Another thing that is not clear is how well this will work when there is an AI launching a wave against a third party faction like marauders. Reports in the prior build were that the third-party factions just got murdered pretty efficiently and then the AI forces would turn on you.
    • If you have savegames that demonstrate any of these things, if you run into behaviors that seem sub-par, then that would be really helpful.
  • It's also possible that some of the CPA Faction Logic itself is going to turn out to be very inappropriate for the relentless waves faction, but we figured it was best to start with a known quantity that is close to what we want and then revise from there.
    • One thing we did take out from the start was any concept of the relentless wave waiting for a time to strike. Since a wave arrives all together, it assumes that it is going to hit immediately.
  • So far so good in terms of the relentless AI wave forces spawning out of a wave and then doing what it says on the tin.

Praetorian Guard and Warden Balance

• The budgets for the Warden and Praetorian Guard were fine, but their CAPS have been off since 2.710. They were simply set too low for the AI to ever really field any of them.
  • Thanks to Puffin for discovering this!
    • Basically a "cap per AIP" of 22 for the PG at difficulty 5 would mean that at 100 AIP, that means a strength cap of 2,200. In visible-to-players game units, that means... 2.2 strength. Woooow that's a poor PG.
  • But just increasing the numbers was not going to go well, because a high "cap per AIP" means that the game gets linearly more difficult as the AIP rises. This is part of the problem we had before. So it's time for SEVERAL new mechanisms.

• On the PG, we still have praetorian_guard_strength_cap_per_aip, but now we also have praetorian_guard_base_strength_cap.
  • We also have a new praetorian_guard_strength_cap_per_aip_above_200, praetorian_guard_strength_cap_per_aip_above_400, and praetorian_guard_strength_cap_per_aip_above_600.
  • These numbers let us have the game curve the difficulty up over time in a sub-linear fashion for a while, and then gradually shift to something more than linear (not exponential, though).

• New PG changes:
  • Diff 1:
    • Was per-AIP: 3
    • Now base: 2000, per-AIP: 20, per-AIP-above-200: 10, per-AIP-above-400: 15, per-AIP-above-600: 30.
    • Quick note: these numbers are additive. So at 700 AIP, the formula for calculating the strength cap is as follows:
      • 2000 + ( 20 * 700 ) + ( 10 * (700-200) ) + ( 15 * (700-400) ) + ( 30 * (700-600) ). Aka a total of 28500, or 28.5k of max strength. This is still pathetic, but hey it's difficulty 1. The old number was 2100, or 2.1k of strength, which is nonexistent by comparison.
  • Rather than show you all the numbers in this hard to read format, here's a spreadsheet that has the new numbers for the PG and for the Warden, and a comparison to the old numbers: https://docs.google.com/spreadsheets/d/1Oqhe_evrtzDOKyVBfwZq27zVZu6WJw8oaa2CZWxgZyc/edit?usp=sharing

• In the difficulty tooltips for the warden and the PG, it now shows what the strength cap would be at AIP 100, 300, and 700 to give players an idea of the progression without showing them all the math directly.

• Thanks to Tim_Fragmagnet, ArnaudB, Puffin, and others on Discord for reporting this.

Beta 2.722 AI Wave Balance

(Released January 25th, 2021)

• Updated the Makeshift Turret description to hopefully be more clear.

• Fixed the range and damage boni from having more units on the planet (i.e. Scourge Peltians) not factoring in ships from stacks.
  • Thanks to NR SirLimbo for catching and fixing this one!

• Prevent players from going negative on hacking points. Should a player hit 0 or fewer hacking points, all active hacks are stopped immediately and display a chat message advising of the failure. This counts as a failed hack.
  • Thanks to Dominus Arbitrationis for fixing.

• Re-coded how the raycasts to hit planets and icons on the galaxy map works. It now does a non-allocating "hit all" raycast, rather than only hitting the first item.
  • In this new code path, it first tries to find a ship icon that is enabled and linked to an actual ship. If it does, then it hovers over that.
    • Our suspicion is that in the past, it was hitting some of these, but they were actually disabled or not linked to a ship for some reason, and so you'd get just blankness. This should no longer be possible.
  • If if doesn't find any ship icon that is valid where your cursor is, then it looks for planets. Assuming that your cursor is over a planet at all, then even if there's a partially-enabled icon, that won't matter.
    • Additionally, if there's a collider somehow in the way for the link between two planets, it doesn't matter since it isn't even checking for that yet, and won't do so unless it finds you are not over a planet itself. That was something that was getting in our way in the past, and while we thought we fixed all those cases, hopefully none of them will be possible as a problem now (even in edge cases).
  • If we still have found nothing (we're not over an icon or a planet), then it looks for the link between planets to see if you are hovering over them.
    • Knock on wood, this should solve all the strange edge cases of planets not being clickable but still showing the planet name when you hover over them, etc. Those cases were increasingly rare in general, but hopefully this gets the last of them.
  • Thanks to Chthon and NR SirLimbo for reporting the most recent cases.

AI Waves Work And Extra Forcefields For Player Homes

• Wave Warnings of Short, Medium (default), and Long have been changed from 60, 180, and 300 to 60, 120, and 180.

• Previously, waves could be declared on a really long interval, AND the wave itself was set to only arrive after a super long time. So it might be a 10 minute interval, and then another 10 minutes before the wave hits after it was calculated, and then give you a 5-minute warning.
  • This has been adjusted so that waves never have an underlying time difference of more than 3 minutes from when they are targeted and when they arrive. This makes it far more likely that the wave target is in remotely the state the AI was thinking it was when it sent the wave in the first place.

• (Revised in a minute) A new wave_budget_multiplier has been added to the AI Difficulty table, and this gives a free boost to the income of waves for any AI type just by difficulty.
  • Please remember that we've just given the players a TON of free buffs, especially in the early game, so this is in a lot of ways us trying to compensate for that.
  • The other thing to bear in mind is that players have been talking about how slow it is for waves to come and bother them, and therefore how much less a factor defending against waves is in general in this game compared to classic.
  • For difficulties 1-2, the multiplier is 1, so no change.
  • Difficulty 3: 1.1x
  • Difficulty 3: 1.1x
  • Difficulty 4: 1.3x
  • Difficulty 5: 2x
  • Difficulty 6: 2.5x
  • Difficulty 7: 3.5x
  • Difficulty 8: 4.5x
  • Difficulty 9 and 10: 5.5x
  • In the grand scheme of things, these are not extreme changes; potentially we need to give even more of a boost to wave budgets to make them truly interesting.
    • As it stands right now, testing on difficulty 5, an Ensemble AI gives us a 3 strength wave at 11 minutes into the game, against a planet with 5 strength in mobile forces, 22 strength in turrets, and 7 more mobile strength rampaging nearby -- plus 16k unspent science, so those strength numbers really could be 10x higher than this first test and still be woefully underpowered.
    • On the plus side, the wave was at least coming after only 11 minutes, and only with a 2 minute warning, so there is some mild urgency there.
    • That said, taking a moment to just spend the basic science that we had on hand from that scenario -- and bearing in mind a ton of turrets have not been bothered to be built, and there's 1.4 million metal just sitting there -- we transform our defensive forces into 82 strength of turrets, with 5 strength mobile and another 16 strength one planet away. Difficulty 5 is not supposed to be overwhelming, but this is just absolutely underwhelming to a crazy degree. 3 strength versus a "casual" force over 30x as strong. It's safe to say that waves are still not a factor at all in this.

• (Revised in a minute) So, let's adjust those numbers some more:
  • Difficulty 1: 1x
  • Difficulty 2: 2x
  • Difficulty 3: 3x
  • Difficulty 4: 5x
  • Difficulty 5: 10x
  • Difficulty 6: 20x
  • Difficulty 7: 40x
  • Difficulty 8: 80x
  • Difficulty 9 and 10: 140x
  • One reason it's worth noting that wave budgets need to be inflated so much is that, unlike reinforcement budgets, they are not duplicated by planet.
    • The reinforcement budget of X will be applied to something like 10-20 planets, and much more frequently over time. So in a lot of very real respects, we have always been shorting waves by a factor of 10 or more in terms of what you would think the AI gets.
  • Okay, testing this again on difficulty 5, at this point I'm getting a 16 strength wave hitting me at 11 minutes. I built more turrets this time, so even without upgrading a single tech, I have 33 strength waiting for them. Three quick upgrades later, still 10k science in the bank and 1.5 million spare metal again, and now I've got 73 strength in turrets waiting for them. I happened to bring my main offensive force home, which was a measly 12 strength since they had no upgrades yet, and the invaders were repulsed without me losing a single turret. But it was at least a fight that took a few moments.
    • This is at difficulty 5, which is not supposed to be particularly hard, so that seems fair enough. Let's see what difficulty 8 looks like with this. That gives us... 596 strength incoming. Hilarious. I'm going to die with no question. There are multiple dials here, and clearly they just exponentially increased. 1/8th of that (the equivalent of a multiplier of 10 line difficulty 5 had) would still have been a strength of 74.5, which is... more intense than the first wave feels like it should be.

• (Revised in a minute) So, let's redo those numbers from difficulty 6 on up:
  • Difficulty 6: 10x
  • Difficulty 7: 8x
  • Difficulty 8: 5x
  • Difficulty 9 and 10: 3x
  • Yes, the higher difficulties get less of a boost, but they also clearly needed less of one. This keeps the AI from being so ridiculously pathetic on the lower difficulties, and hopefully keeps them from being monstrous at higher.
  • Let's test out difficulty 8 again. This time I ran into more trouble, because the first planet I tried to attack put up a pretty good fight. I upgraded two techs twice, and so had 79 turret strength easily, and 23 mobile strength. The AI had time to reinforce its first planet, AND the hunter showed up with 20 strength, so it bled me dry for a bit before I neutralized the planet and the hunter ran off. It took much longer for the AI to decide to send a wave, but at 21 minutes and 15 seconds in, it decided to send a 69-strength wave at my homeworld. With 2 minutes of warning and my ships in shambles from my offensive efforts, this was a tough battle. Granted, I had 8.5k extra science just sitting there, but I also had the bad luck that the incoming wave was mostly bombers. I quickly knocked them down to 38 strength, but they took out my home command station while I still had 75 strength sitting around on that planet. I need more defense in depth, but beyond that I would have probably been fine.

• So, let's down this just a hair more from 6 on up:
  • Difficulty 6: 8x
  • Difficulty 7: 6x
  • Difficulty 8: 4x
  • Difficulty 9 and 10: 2.5x

• Also? Let's adjust some forcefield counts for players, because losing like that when you have a small planet is no fun:
  • Player home forcefield count: from 1x to 2x at mark 1 for players, and then +1 for each of the next two mark levels just like before.
  • A lot of players would not know to invest directly into their homeworld like this, but at difficulty 8 and 9 that being a needed thing seems like an okay situation.

• (Discarded in a minute) What on earth is this going to do to the meta on higher marks? That's hard to say. One thing that is certain is that the general mark level increases are something that players really fear.
  • Given that players have almost no mark 1 ships now (most of the time), it seems like giving the AI some more buffs here might be worthwhile.
  • Difficulty 5: all mark level thresholds move down a mark (so 120 AIP means mark 3, not mark 2), and mark 2 now comes at 60 AIP.
  • Difficulty 6: all mark level thresholds move down a mark, and mark 2 now comes at 55 AIP.
  • Difficulty 7: all mark level thresholds move down a mark, and mark 2 now comes at 5 AIP (yes, 5 -- so there's no early-game incentive to avoid taking a planet until the first wave arrives).
  • Difficulty 8: all mark level thresholds move down a mark, and mark 2 now comes at 5 AIP.
  • Difficulty 9: all mark level thresholds move down a mark, and mark 2 now comes at 5 AIP.
  • Difficulty 10: all mark level thresholds move down a mark, and mark 2 now comes at 5 AIP.
  • Let's see how we fare on difficulty 8 NOW. Six minutes in and I am playing for real, with 5k science in reserve but 3 overlapping home shields up, 92 turret strength sitting ready to eat my enemies on that homeworld, and a hard-won first planet from the AI but still a victory. At 17 minutes in, I have a second planet and it has 27 in turret strength while my home planet now has 165 in turret strength. I have 23 mobile strength between them, and two-layer forcefields on both. At this difficulty level, strong defenses cause the AI to wait before sending a wave, so let's see how far it goes. 22 minutes in, I've just smeared a third AI world to bits and they launch a 91 strength wave at my weaker non-homeworld planet. I opt for one more tech unlock, and now I'm sitting at 36 turret defenses and 18 mobile defenses there. It won't be enough, but that's okay. I need a GCA, but I'm not going to die from this. My economic second planet manages to get them down to 42 strength before biting it, but I'm still rich as heck. The hunter fleet shows up, because it correctly sees an opening, and darn if the warden doesn't also show up. I wind up retreating with 41 strength of the enemy sitting on that planet. I'd like for them to attack my homeworld, but they aren't taking the bait. 45 minutes in, and I'm at a stalemate with them, unable to retake my second world. It's not super interesting at this point.
  • As I had suspected from the amount of fear that players have of the leveling-up of the AI in mark level strengths, it's just too big a jump. That means that, early stall or not, the late game is just going to be impossible on the highest difficulties with the sort of shifts that I've made in today's balance adjustments.

• (Discarded) What we need for the AI mark levels that go up is a smoother balance curve, and something that has fewer scary cliffs all at one location.
  • To start with, we're getting rid of the concept of a "general mark level" for the AIs. Instead there is now a defensive, offensive, and wave mark level.
  • The "defensive" one is largely used like the older "general" one was, in cases where there is nothing else to indicate a mark level.
    • The defensive mark level should grow the slowest, by far.

• Actually that got insanely super complicated. Let's just put the AIP level-ups back where they were and try difficulty 8 again.
  • With this shift, and me playing properly, they send a 50-strength wave against my second planet, a military station this time, and I easily repel them, killing half of them. I have energy problems, but the game is moving and off to a solid start. My homeworld is so ludicrously defended that most of the AI forces won't try to attack it, so getting beyond that world and having proper defenses elsewhere that the strong AI waves don't come in and wreck is going to be the new challenge.
  • At high levels of play, I think that's going to be the new meta, if anything close to these balanace numbers stays: trying not to over-protect your homeworld so that you can properly defend other worlds (because of energy shortages), and in general dealing with waves that can be scary big, but ultimately just give you a fight before moving on. Late-game, it will be interesting to see the balance.
  • There have been comments that we need to give the AI more guardians now that the balance has shifted so much in the player's favor with the AI getting no frigates, etc -- but with the shifts to waves instead being so much more harsh, I'm inclined to keep the AI guardians where they are at (for now).
  • It's hard to know for certain, but it feels like the balance of power between the player and the AI is a lot more appropriate again, now that waves alone are so much stronger (and actually a factor for the first time in a long time).

The Last Of IMDraw And Shapes Work

• The way that range rings are drawn for ships that either already exist, or that are about to be placed by you using the build menu, now uses Shapes instead of IMDraw.
  • This new approach took one scene that was mid-heavy, and the old stats were that it dropped Chris from 120 fps to 70fps just to start showing the range circles (in the original AI War this was also a big performance hit for us, for different reasons). In the same scene, using the new code, the drop is from 120fps without the range circles shown, down to between 100 and 115 fps with ranges shown. The exact final framerate depends on how zoomed-in you are at the time, and thus how much of your screen the range circles individually take up (more coverage takes more processing power, but it's all still far better than the old approach).
  • Also the new rings look smooth and nice, rather than terrible and dated, especially when you zoom in on them.

• After experimenting with changing the battle-indicator circles on the galaxy map to use shapes instead of IMDraw, we found that the performance was basically the same, but the visuals with Shapes were substantially worse. We're just going to keep on using IMDraw for purposes of these parts of the galaxy map, and the ping effects, since that works so well.

• When you are in placement mode for building some structures (turrets, etc), it now uses the nicer-looking Shapes lines and circles. The performance difference is negligible, but it looks vastly cleaner and more professional, especially when you zoom in further to look at the ship you are placing.

• After looking at the performance of the line drawing for IMDraw with unit orders, it really takes an extraordinary number of lines before performance starts to substantially drop. It seems unlikely that we would be able to beat that with Shapes, and it might not even be possible to match it. So for the time being we are going to leave it where it is, and consider our work on the performance-related-toIMDraw to be complete.
  • There are some other places where we use IMDraw to draw some circles when some debugging is turned on for shots and similar, but we also don't really care about framerate drops with a situation like that -- that's the perfect sort of situation for IMDraw.

Mod Updates

• Extended Ship Variants (ESV)
  • Shortened the description. A lot.
  • Removed all min/max caps from starter fleets (both mobile and support) and used the default cap XML field instead.
  • Gave Turbo Stingrays, Etherjets, Ensnarer Etherjets and Thiefs the ability to spawn in fleets and ARSs.
  • Gave the AI the ability to spawn Turbo Stingrays.

Beta 2.721 Histiocytes And Circles

(Released January 22nd, 2021)

• Updates to turret counts in the new balance -- thanks to ArnaudB for making these shifts.
  • Updated cap counts for starting battlestations, seeded battle/citadel and GCAs.
  • Adjusted counts for military station, home station for pike, ambush, beams.
  • 30 small - 6 large on battlestations. 20 small - 4 large on seeded battlestation/citadel. 15 small - 3 large on GCA.
  • Adjusted the gravity generator, tachyton and tractor as well.
  • Thanks to ArnaudB for putting this together!

Mod Updates

• ESV
  • Fixed the Viral Shredder variants not using the cannot_actually_be_built_builds_self_indirectly XML field, meaning that the game would fail trying to load them.

• New version 1.18 version of More System Defenders mod by CRC Gamer.

New Included Mod: Macrophage Histiocytes by StarKelp

• Added the Macrophage Histiocytes mod.
  • Adds in a multitude of new features to the Macrophage faction, most of which can be enabled or disabled at will.
  • Histiocytes:
    • Drones created by Macrophage, Telia, and Infested Mines. These mimic the stats of regular Harvesters.
  • Metal Generator Infestation:
    • Harvesters can corrupt Metal Mines, causing them to create additional Histiocyte drones.
  • Dark Spire Rivalry
    • Macrophage are now capable of consuming Dark Spire Vengeance Generators.
    • Dark Spire take offense to this, and will wither the Telia on planets adjacent to their generators in response.
  • Tweaks Galore
    • A number of tweaks can be found in a new Macrophage section of Galaxy Settings, which include toggling off many of the above features.
    • Others include passive Telium regen, allowing Spores to also infest targets, and letting Extragalactic units wither Telium away like Dark Spire units.
  • Additionally included is a submod that can be activated to apply these effects to the Spire Infested Telium and Harvesters.

Drawing Lines And Circles

• Shapes, by Freya Holmér, has been integrated so that we can draw extremely efficient lines and circles in an HTML5-looking style.
  • https://acegikmo.com/shapes/docs/
  • Bear in mind that because of the component-based nature of this, we have to pool and alter shapes versus being able to request them drawing "on demand" as we have been doing with IMDraw in the past.

• We've implemented our own ArcenShapeRoot base class, which allows us to wrapper and carefully control edits to various subcategories of shapes.
  • At the moment we have implemented 5 specific sub-shapes for our purposes: disc transparent, disc sized, line, rotational ring, and thickness ring.

• We then have a new ArcenShapePool class, which manages pooling and initialization for each of the shape classes that we have.
  • This is all working for the five initial shape styles, but it's then up to other code to actually request the use of a shape, set where it should be and what it should look like, and ultimately return it to the pool when it finishes with it, etc.
  • Right now none of the other code does that yet (it's all still using IMDraw), but the initialization and pooling logic is fully up and going. Next step will be to convert over specific usages of IMDraw, one after the other, to use the new approach.
  • The performance of IMDraw is really really bad when it comes to drawing hundreds or thousands of circles, although it can draw thousands of lines without a problem. But in both cases, they wind up also looking pretty dated and unattractive.
    • Folks have been noticing that on the more recent drivers from nVidia on windows, we now wind up with a 50% drop in framerate just from selecting ships and drawing the selection circles around them. Those were not causing nearly so much overhead in the past, so it seems to be something to do with the newer hardware or newer drivers or both (for us internally, drivers from 2018 were fine, but on the newer 2021 drivers the performance tanks, without hardware changes).
  • Anyhow, for those folks who are NOT seeing performance degradation (either they are on older drivers or their hardware is just not affected), these sets of changes will result in prettier lines and circles but probably not any performance boost since IMDraw was previously lightning fast. For those on newer hardware, it will be both prettier and faster.

• The cursor that appears on the plane of gameplay has been updated to use Shapes rather than IMDraw.
  • Rather than two coarse thin line circles, it is now a dot in the center and a nicer lined circle around the outside.
  • This is both slightly more efficient (but so low in volume of draws as to not be a big deal) as well as a lot prettier. This is also the first example of us actually using the new Shapes code in actual practice.
  • In the cases of this one, we pull out some items from two different pools and then just hang onto them forever, since we're pretty frequently needing a cursor, after all.

• The way that circles are drawn around ships that are selected, or that you hover over, is now based on Shapes instead of IMDraw. This looks incredibly more polished now, and comes with a far less substantial performance hit than before when you have a bunch of ships that are all selected at once.
  • There are still various other parts of the game to update with this new style, but these were the really big ones that were causing massive slowdowns for folks on newer drivers.
  • For Chris in a standalone build, in the "terrible framerate" test case from Daw11, the framerate was previously getting cut in half from 130fps to around 60fps or even lower on his GTX 1070 and i7-6700HQ. Now it drops from about 130fps to about 120fps. This is way more appropriate for the type of operation being undertaken.
  • It's also far more attractive!
  • Thanks to Daw11 and others for reporting.

Beta 2.720 Linux Runs

(Released January 21st, 2021)

• Fixed the game up to not have splash screens for unity on any OS.

• Fixed up some internal things so that we can do new full-builds much faster.

• Improved some of the visuals from the loss screen so that they will look better in this new version of unity, as there were some shaders that did not translate well used on the asteroids and a few other elements there.

• Fixed an issue where the linux build would not launch at all in the prior beta, because there's a new UnityPlayer.so file that is needed. That is now included in this version and on our internal tests on Ubuntu we seem to be back in business.
  • Thanks to Daw11 for reporting.

Beta 2.719 Unity 2019.4, And Performance

(Released January 20th, 2021)

• Note: The current build has a "Made With Unity" splash screen that is not intended to stay. We have more to fix up with this in general, but it's good enough for testing on various platforms for now. There is now a current_beta branch on Steam, and a most_recent_stable_beta to go back to the prior one if this one gives you problems.

• The xml field cannot_actually_be_built has been replaced by two fields:
  • cannot_actually_be_built_is_not_a_ship_line for empty slots.
  • cannot_actually_be_built_builds_self_indirectly for viral shredder copies and similar.
  • Previously there were a variety of strange things that could happen because viral shredder copies were set up identically to "empty ship lines" in this.
    • It's possible that in some of the fleet sidebars and swap line popouts, viral shredder copies may show up inappropriately now. If you find any cases of that, please let us know.
  • One bug that was happening before was that viral shredder copies were having their fleet membership disconnected from their actual fleets, which made them not able to get in transports after that point or be selected properly with the fleet.
    • For viral shredder copies in existing savegames that had this problem, they now work just fine when you load the savegame (they don't have to suicide themselves or anything like that).
  • Thanks to GreatYng and Crabby for reporting.

• When viral shredders are transferred between fleet lines, if there are any viral shredder copies left behind, they will delete themselves.
  • Previously, they were instead throwing endless streams of errors.
  • Thanks to Chthon, Puffin, Crabby, and PetGhost for reporting.

Unity Upgrade To 2019.4.18f1

It's worth noting that a lot of file format stuff changed here (more efficient) due to Unity's internal database formats changing, so the updates are ludicrously big. The total when all three platforms are combined is a difference of 3GB to download on Steam. On any given single platform it's probably closer to 2GB.

• AI War 2 is being upgraded from unity build 2019.1.7f1 to 2019.4.18f1.
  • Overall this is an upgrade from the long-term support version of Unity from April of 2019 to the January of 2021 build.
  • This MAY fix the inverted scroll wheel bug on linux, finally, but we're not sure.
  • This does also come with some internal performance improvements to how jobs are handled on the CPU, most notably around rendering. So this hopefully will be at least a minor performance boost for folks.
  • We also have upgraded Amplify Shader and Shader Graph to the latest versions, so whatever automated improvements they have to the sorts of shaders that we created those can now be applied. Other shaders that we hand-coded are already probably as efficient as they can get.
  • We did look into "Addressables" as a potential replacement for Asset Bundles, hoping that we could get faster load times. But really all that would do is give us better asset management (which we already have in great shape) and still would be using asset bundles under the hood. So definitely not worth any development time there.
  • There may be some other various improvements that we are not aware of (seriously, almost two years of lengthy release notes on a weekly basis, 99% of which doesn't apply to AI War 2). Given that this is the "long term support" branch, it is unlikely that there are new bugs that were not there before, but stranger things have happened in the past.
  • This does also allow us to use more up to date assets and packages in general, to maximize performance on things like drawing lines and circles if that's a possibility.
  • DirectX11 in general might perform better now, but we'll see how that goes. If it performs better now than OpenGL on your machine, please do let us know!
  • Vulkan support on linux has also probably improved substantially, so let us know on that, too. That's really highly dependent on your specific drivers and hardware and OS flavor.
  • There are a bunch of other random things that unity fixed, like better support for multi-screen monitors on OSX, and things like that.
  • Looks like our build times for large asset bundles and such are also dramatically better, which lets us more easily make changes to art in the future. Seems to be at least a 400% improvement over what we've seen in the past, and honestly that is probably under-selling it. So that's a huge savings for us in time!
    • Actually, yeah -- one example case with small changes, it's improved from 400 seconds to 28 seconds. This is gigantic for our ability to iterate and test on art and ui elements.

• We are now using Graphics Jobs again in unity. There was a performance regression with them in 2019.1, which they fixed in 2019.3, so that is now a benefit again.
  • This should help to make various aspects of rendering less CPU-bound, but it will depend a lot on the machine in question.

• Our gimbal sprites are now using the geometry pipeline rather than the transparent pipeline, which makes drawing them vastly more efficient in two ways.
  • Firstly, it avoids depth sorting on the CPU, which is time-consuming and can be a bottleneck.
  • Secondly, it allows for full use of GPU Instancing, so that there should be no more than a single draw call per combination of sprite dictionaries, rather than potentially having more draw calls because of non-matching dictionary groups being non-adjacent in the sorted draw list.
  • This can cut the draw calls in a giant many-types-of-ships battle from being hundreds to being like 6 to 10.
  • It looks like we made the change to use the transparent pipeline on July 15, 2020, after more than two years of it being in the geometry pipeline.
  • This has a massive positive impact on both the galaxy map view and the planet view.
  • Thanks to Lightjolly for the report that helped us find this.

• We are making some adjustments to our camera stacks, and part of this is also using some new unity features:
  • We are now letting unity monitor CPU and GPU frame timings, and that can feed into adaptive screen resolutions for certain cameras.
  • The background nebula now uses adaptive screen resolutions if the framerate is poor, and it also now is set to NEVER use MSAA since that is useless on that camera.
  • Our UI camera is now set to use Forward rendering like all the others, rather than using Deferred.
  • MSAA is now off on the sprite cameras, overlay cameras, and on the effect cameras for stale intel and unexplored planet effects. MSAA is only enabled on the main camera that would be drawing actual ship geometry that would benefit from it (and the main menu camera).
  • Our pre-canvas LDR mapper camera is now set to not do occlusion culling, although it already was set to a culling mask of Nothing, so it's just doing image manipulation anyway. But just in case.
  • The main menu screen (which can be super heavy for some folks) now allows for adaptive resolution.

Beta 2.718 Tutorial And Tech Fixes

(Released January 19th, 2021)

• Tutorial refinements:
  • Notices of new journal entries no longer appear during tutorials (in the chat log), but the journal entries do still appear in the sidebar.
  • For folks with DLC3 installed, they were getting a necromancer faction in the tutorial, which is now blocked from happening. Actually, they are no longer added with a default configuration just because you have DLC3 installed in general.
  • Discoverable factions in general are no longer added during tutorials.
  • The Praetorian Guard are now entirely blocked from appearing in tutorials.
  • Thanks to blackdog904 for reporting some of these issues that were still there.

• The second tutorial now uses Raid Frigates instead of Assault Frigates so that it goes much more quickly.

• Fixed an intermittent issue where the tutorials category list could be empty, leading to no tutorials appearing.
  • This would mainly happen if you changed the installed expansions or mods after loading the game but before looking at the tutorials, but that was not the only way it could happen.
  • This bug was introduced when we implemented hot-reloading of expansions and mod content a couple of months ago.
  • Thanks to Kolbex for reporting.

• Fixed an issue where the tutorial text could wind up coloring some bits after its "step complete: proceed to next step" bits.

• Fixed a bug in our new (as of a few months ago) more efficient ArcenDoubleCharacterBuffer where it sometimes would return stale text.
  • Essentially, we were checking to see if any text that was written in this new pass was different from the text from the old pass, and that was working perfectly.
  • What was NOT working was that we did not check to see if last time we wrote more than this time. So if we wrote identical text, just less, then it would keep displaying whatever was there from before. This was extremely hard to pin down, but it affected things like the chat popups and tutorial step complete messages, and various parts of the galaxy map.
  • The additional check to fix this is extremely inexpensive and corrects the behavior without making the ArcenDoubleCharacterBuffer any less efficient than it already was.
  • Thanks to a variety of folks for reporting variants on this, but in particular Kahuna, Crabby, SortYa and others.

• When you're looking at a flagship that you've not captured yet, it no longer shows the text about how long it will take ships that are unloaded to wait before they can fire. That's not relevant there.

• When you C-click a fleet that you do not yet own, it now shows the proper mark levels and ship caps for each ship it talks about.
  • Thanks to Minotaar for reporting.

• Fixed another issue with the counting of ships where if there was a fleet that you could capture it would show the wrong number of ships for the current value that you're upgrading from on the tech tooltip.
  • Thanks to Minotaar for reporting.

• When you C-click a tech, it now properly shows the ship caps and how they would increase for any ships that you could capture that are part of that tech.
  • This is both a fix to a regression (it was showing nothing in the last two betas) as well as a fix to a longer-term issue (prior to the beta branch, it just showed a fixed number that did not increase).
  • Thanks to Minotaar for reporting.

• Previously the colors for the mark line counts and strength totals that you would get from upgrading a tech were kind of scrambled up. In a couple of places we had green and blue inverted (blue = ships, green = defense, but it was backwards on the actual science buttons), and it wasn't consistently applied in the tooltip.
  • Now this is shown consistently throughout the tooltip and the actual sidebar, and on the sidebar if it would show one but not the other, it now only shows the one that has a value (rather than showing zero for the one that has no improvement).
    • In the even that it does wind up showing a zero, it will be because there is less than 1 strength granted from some line.
  • Thanks to Tim_Fragmagnet for reporting.

Refinements To New Balance

• The starting support fleets now have 16, 20, and 14 combat engineers rather than 8, 10, and 6. This brings them into line with the new starting values in the new balance of the beta branch.
  • Thanks to ArnaudB for reporting.

Beta 2.717 Tuning And Optional Capture

(Released January 18th, 2021)

Refinements To The New Balance

• We now have the first example of a custom MarkLevelScaleStyle that is just for a single ship or a few ships that are assigned manually (in this case it's assigned manually).
  • For the player home forcefields, those now actually DO get some added cap added to them (up from the start of 1) as you upgrade your fleet. They start out with a cap of 1, and for mark 2 and 3 they each get +1. Beyond mark 3, they get no more cap.
  • Thanks to CRCGamer, Chthon, Crabby, and others for suggesting/discussing.

• The attack multiplier curve for human frigates has been nerfed HARD.
  • In the lower marks, you in particular really just get far less attack power, while still getting the more aggressive hull health. Frigates in general actually grow up a little slower than the average AI ship now in terms of attack power, and a lot slower than strikecraft. But they have such intense multishot that their power is pretty insane as it is.
  • That said, when you finally get up to mark 6 and 7, if you choose to, they actually accelerate slightly beyond what the AI would have on attack power. So their curve has been adjusted in general, but they still are overall buffed compared to prior builds before this beta branch.
  • Players were easily fielding 100+ strength in the first 20 minutes of the game with the prior beta build, so this is a first way to try to combat that. We want frigates to be good, but not god-tier.
  • Thanks to a variety of players on discord for discussing. More balance will be needed, but today was more about fixing bugs first.

Fixes To Prior Build Regressions

• Fixed an issue with the GCA additions to fleets not always being added into the caps of ship lines in a variety of places. It seems to properly include the results from a GCA hack in all of the tech previews now, as well as actually granting you the GCA ships at all (unlike in the prior build).
  • Thanks to a variety of players for reporting, including Chthon (thanks for the save!), DEMOCRACY_DEMOCRACY, Spaz, Crabby, and others on discord.

• Fixed a variety of places where absolute_max_cap_when_in_fleets was being set to values that were too low for the new balance. This led to players getting fewer forcefields from logistics stations than they were supposed to, for instance, but also affected tractor beams and tachyon beams and a variety of other units. They've all been properly updated now.
  • Thanks to Chthon for reporting.

• Fixed an issue that would let negative numbers show up in the tech upgrade tooltips for ship counts for various ship lines.
  • Also fixed another issue where turrets in particular would look like they were going down a few in count because of a rounding error that would happen when you had enough planets with turrets all on them.
  • Thanks to NRSirLimbo and others for reporting.

• Added can_be_empty_or_filled_and_no_complaints to AI ship groups.
  • Usually used for stuff that is in one or more expansions and mods, and so defined in the base game but empty unless a certain expansion is present.
  • CAN be used in conjunction with if_empty_copy_items_from without problems.
  • Is now used on some various features that were complaining if you did not have DLC1 or DLC2 installed in the prior build.
  • These errors that were there when you started up the game in the prior build were alarming, but harmless.
  • Thanks to Chthon, CRCGamer, zeusalmighty, and others for reporting.

Bugfixes

• Fixed a couple of rare race conditions that could cause exceptions while drawing or moving shots around as you switched planets.
  • Thanks to samnainocard for reporting.

• Fixed an exception that can happen when hitting Reset To Defaults while you were on the factions tab in the lobby. It was getting confused and asking for some nonsensical data on one frame, but then would be correct after that. At this point it just lets the one frame be wrong and correct itself after, which is how it has been prior to mid-last-year when we made some of the items more strictly verified.
  • Thanks Crabby for reporting.

• Fixed a rare exception that could happen in GetHasSelection() if a squad died just the wrong way.
  • Additionally, if we ever do have an exception in there, it will now give more useful errors and disrupt the main GUI view less.
  • Thanks to samnainocard for reporting.

Misc Improvements

• In the Tech Menu hovertext for a given tech, when displaying 'New Ships for ship line' and 'Old Ships for ship line', the older value is a smaller font to make it clearer what's happening.

• The intel objective for locating the AI Overlords that you have not found has been changed in a few ways:
  • It now just has one entry for however many you have not found, and does not specify how many there are. In some games, you don't know how many there are, and we don't want to tell you until you scout.
  • It is no longer linked to an actual overlord, so you can't change the priority of the objective and cause the position of the AI Overlord planet to be revealed (not that we could get this to work, for whatever reason).
  • The tooltip itself now states that there are one or more undiscovered AI king units, and that you'll need to find out how many.
  • Thanks to Tim_Fragmagnet for reporting.

• The keybind for "Hold For Yellow Pings" is now X, rather than Shift+Ctrl.
  • This prevents issues where anything that was "just ctrl" or "just shift" was not working anymore if you held Ctrl and Shift at the same time.
  • Basically, when we have modifier keys that are a combo of lesser things, we don't want to trigger this, right? If you think about a word processor, as a simple example, hitting Ctrl+A should select all and not ALSO type the letter A.
  • There weren't really any direct issues from this, except that anyone who reflexively holds Shift down wile using place-multiple (Ctrl or Alt) would find that it failed for them. Now that will work again.
  • Thanks to Astilious for reporting.

• Added a new is_intentionally_empty_beyond_centerpiece="true", which can be applied to fleet design templates.
  • When this is true, you won't get the "Fleet Design Template X never has any items added to it (centerpiece aside)! This will cause some problems." error for it being empty.
  • However, when this is true, if there is anything in any of the groups other than the Centerpeice slot, then it will complain about them since that's not what you indicated you wanted. Just a helpful self-check for modders and developers.
  • Thanks to zeus for reporting.

• Added a new can_be_empty_or_filled_and_no_complaints="true" flag for fleet design templates.
  • This one is more for testing style fleets, which may or may not have entries in them.
  • This is now applied to the non-distributed Precepts of the Precursors mod, to prevent it from erroring.
    • For most of the cases they are clearly test fleets, but in one case it looks like a dyson seedlings fleet was defined but just never finished being set up.

• If the "for science" cheat is used without any arguments, it no longer errors out but instead just gives you 10k science.

• The Poltergeist has been moved to tier 2 instead of being tier 1 in the AI Exogalactic War Front faction.
  • This makes it so that they are very unlikely to show up in any game that doesn't have Fallen Spire or something else truly crazy going on. The fact that these guys can shoot straight under forcefields is way overpowered and seems cheesy when it shows up out of context, but when it is properly pitted against the really overpowered forces you gather up in the Fallen Spire campaign, it fits right in.
  • Thanks to Arides for reporting how this doesn't fit with tier 1.

• Previously, in multi-faction multiplayer if an ally was having a brownout, that was impossible to tell apart from you having one.
  • Now the tooltips are different for you versus your allies, so you can actually tell what is going on.

• Eight various golems were actually still named "Golem Arks" from waaaay long back. Fixed.

Capturables - Clarity and Control

• The game now always shows if there is an AIP or metal cost to claim something, even if you are on the brief tooltips.
  • The colorization is also now better and more eye-catching, and on the brief tooltip it is more brief.
  • Thanks to Unit 744 for suggesting.

• Fixed an issue that was extremely confusing, where ships that belonged to your faction now but still had not been claimed yet (because of a metal cost over time, generally) would no longer show what the AIP cost (or metal cost for that matter) were going to be.
  • This led to confusion about whether you had already paid the AIP cost or not, since you could look in the AIP log and see that the entry was missing.

• Capturables that take a while to rebuild (like golems or arks) now can be selected like other units, but they will immediately reject any orders you give them.
  • You can, however, change their stances, and pause them, now.
  • When set into pause mode, your unclaimed ships will not actually BE claimed -- the tooltip will show that information, and it will avoid spending all that metal that your economy may or may not be able to afford, as well as potentially incurring AIP if that's related to this capturable.
  • Thanks to poljik2 and Crabby for suggesting.

Mod Updates

• AMU:
  • Multiple internal adjustments to work with changes to the Vanilla code again.
  • Fixed some rogue debug printouts that shouldn't occur unless they are turned on deliberately.
  • Adjusted the WarpToPlanet() method as it could have thrown missing method exceptions due to some Vanilla code changes. However, no unit should have had the requirements to trigger this function (yet).
  • Fixed yet another bug where Fireteams could get stuck on planets if the command station was armed, even though the planet was already conquered. The FireteamMaintenance() function now detects such planets and resets the team's objectives.
  • Fixed an issue with the canSpend() method from the CostBasedShipSpawner leading to an attempt at spawning infinite ships when being allowed to go into debt. This would only occur if Kaizers Marauders were set to intensity 10, and fixes them locking up the game permanently.
  • Updated the included source code.
• KM:
  • Prevented Marauders from building more than one Capital if the game was loaded from a save at just the right time.
  • Fixed the Raid Battleship not moving towards targets and instead staying at sniper range.
  • Marauders allied to players or AIs will generate only half the exowar response. Playtesting for an extended period of time revealed that this got out of hand too quickly.
  • Marauders will "split" their Exowar budget: If there's 2 hostile AIs they will divide it by 2, if there's 3 they will divide it by 3, and so on. This will lead to lower-rank but still more Exowar units thrown at Marauders.
  • Fixed Marauders at very high intensities building so many forcefield generators that they would start stacking and erroring out the threads because they could no longer find valid locations. The same treatment was given to the Defense Posts because they might have the same issue in the future.
  • Removed the NoDrones and EmptyPlaceholder fleet design templates. They were never used anyway and relics of previous changes when fleet ships and drones were "emptied out" and removed, but were now complaining that they had no ships in them (well duh).
  • The Fortified Forceshield Generator now no longer can stack with bad placements or when being pushed into one another, and now also uses the new forcefield scaling.
• ESV:
  • General balance pass, mostly about nerfing the power of FRS ships and the Melee-Frigates. The latter were just so powerful and fast that they could act as if they were better Raid Frigates. Plus more drone fleets and less drone projectiles.
    • Adjusted to no longer have min/max caps and instead use flat caps similar to Vanilla, shifted shields to hull and adjusted descriptions mentioning shields for Strike Craft, Drones and Drone Projectiles.
      • Excluded are Void Bombers which use special shields as a special mechanic for themselves.
      • This revealed some typos which lead to inconsistent ship line counts like no or too high variance between min/max in some cases, which are now fixed as well.
    • Halved the ship cap of all the FRS ship lines
    • Alpha Strike Wings now correctly empower their Damage Modifier based on being on the planet for a while, instead of adding an ambush damage modifier, it always was supposed to be a Raid ship.
      • Thanks to ArnaudB and probably a few others over an extended period of time for mentioning this.
    • Adjusted the Suppressor and Devestator ship counts from the starter fleets from 4 to 6 to match what the player can find in the galaxy.
    • Reduced the speed of the Melee Frigate ships from 2900 to 1800. They were, again, way better than Raid Frigates...
    • Reduced the damage multiplier for enemies with >= 5 tx on the Wedge Frigates from 7 to 4, but it now begins applying to enemies starting at 3 tx.
    • The Metabolizing Drones from the Metabolizing support fleets now have half the health, they were too durable for mere Drones.
    • The Metabolizing and Acidic factories no longer use drone guns but now use actual drones. Their self-attrition is now 35% instead of 3% per second but they no longer decay as long as their parent Mobile Factory is on the same planet.

• Updated More System Defenders to 1.17.
  • Includes a fix to loading very old savegames.
  • Also should properly delete some renamed files, now.

Beta 2.716 The Great Balance Curve

(Released January 11th, 2021)

• Subversion quickstart tooltip now mentions that the nanocaust invade later in the game.

• Make Astro Trains level up a bit more slowly

Mod Updates

• Updated the mod Extended Ship Variants to match the new unit amounts so that they won't be underpowered in the new version.

• Updated the mod More Starting Options to match the new unit amounts so that they won't be underpowered in the new version.

• More System Defenders Dev Build 1.16 - Ships rebalancing take one.
  • Updated mod by CRCGamer to have wider changes for fitting into the new balance.

• Updated the UtilityMethods class so that mods can more properly use it to get entities in an AI ship group and similar.

• Updated the Civilian Industries raid logic.
  • Making use of a new function created by Chris that gets all raid ship types without potentially erroring out.
  • Added a max attempts per loop per raid wave, meaning it will more quickly error out and will no longer break its entire thread for the rest of the game.
  • Added in some more detailed debugging for if this ever occurs again in the future.
    • Thanks to mashtong54 on steam for the report, and Chris for helping optimize the fix.

Interface Updates

• Added a new DoOnLocalStartNonSimUpdates_OnMainThread(), which lets us do some of the logic from DoPerSecondLogic_Stage3Main_OnMainThreadAndPartOfSim() prior to the game being unpaused.
  • We now use this in particular to make sure that the details on the tech menu are updated properly before you have unpaused the game. Aka it shows how many ship lines would be improved, and all that sort of thing.

• The way that the ship caps are calculated for purposes of previews when upgrading a tech is now vastly more accurate.
  • When you are looking at the list of ships lines that are improved by a tech, it now shows what their ship cap changes will be, as well as what their individual strength improvements will be per ship line. Previously you could only see this if you clicked into the C-click menu (though you could see aggregate numbers for strength, by tech, and still can).
  • There were a LOT of edge cases where various counting things for ships were wrong, and that code has all been completely rewritten.

• The strength estimates for improvements by techs have been wrong for a long time, since they were using a formulate that wound up using the old mark level's hull and shield amounts rather than the current one.
  • This has been corrected, and you can now accurately see what your benefits would be for each tech. Holy decision-making ability, Batman.

• When a tech will benefit ship lines that you can capture AND that you already have, it now includes both. Previously, it was ignoring any capturable ship lines that were duplicative of something you already had, which was not very helpful for decision-making.

• The "weak against" tooltip (hold R and hover over enemy) is now much more accurate in terms of what it lists in general, as well as also showing things that get multiple bonuses against a single unit in the proper multiplicative format. If it's two different weapons, then it may overestimate; that's a pretty rare case, if it even exists, though.
  • Additionally, these are now sorted from best against the target, to least strong against the target, instead of being sorted alphabetically. Anything that is the same amount good against a target is sorted alphabetically.

• The weak-against view now shows how many ships you have (or could have) of each given type. This is a handy little thing that gives you a view into how many more ships you can bring to bear of a certain sort against these enemies.

• In the strong against display, it now uses the max unlocked mark of the player ship lines, rather than the lowest mark. This rarely matters, but sometimes gives more accurate bonuses.

• All of the other improvements that were made to the weak-against display are also now on the strong-against display (hovering over your own units with R held down).
  • Being able to see how many ships of each enemy type you have discovered is actually pretty cool on this screen. It shows them all with the max discovered mark level, but breaking out them by mark level would be extremely painful in terms of how much screen space it would take.

Balance Adjustments

• Strikecraft (in the base game and DLC1) no longer have personal shields, but all of what was shields are now in hull.
  • Basically, strikecraft don’t need personal shields: they don’t live long enough for regen to kick in where that’s beneficial.
  • Instead, all that this means is that they have a new weakness (fusion weapons) that makes fusion overpowered.

• Updated Fusion Bombers, lowering their bonus and putting most of it into the weapon damage directly.
  • The bonus was meant to optimize targeting, should've been lower long ago, makes them less likely to get hindered by allies at all.
  • Thanks to Puffin for finding this!

• The Dark Spire ships that you can get from the Hacks no longer upgrade from human techs. That was overpowered for those ships, as well as emphasizing those techs way too much.
  • The names on these are now unique, showing "Hacked Dark Wraith" and similar rather than just showing the name from the main setup.
  • You now get 2 rather than just one Specter, when you hack for those. They also have half the metal cost of before, and a plasma torpedo that is substantially lower power, and slightly lower range. This keeps them from just absolutely outclassing Bombards in every way.
  • Thanks to CRCGamer directly for the bulk of these changes, and for the inspiring the rest.

No More Unlock Variance

• The galaxy option "Remove Unlock Variance in Capturables" has been removed from the game.
  • So has actual underlying ship line variance in capturables!
  • For fleet lines that are pre-existing, if they have min and max, it will now just use the max. So existing mods won't be hurt.
    • If you are a modder and wanting to set something more intentional, then please use the "cap" xml tag instead.
  • This particular style of variance added nothing to the strategic decision-making of the game, but could lead to you getting unintentionally hosed.

• On GameEntityTypeData, we now keep track of a MinFleetMembershipCap and MaxFleetMembershipCap, which is the largest number ever in a fleet versus the smallest ever in a fleet.
  • We had a handy "quality scoring" that previously we were using for situations where there was variance in a given fleet line (that is gone, as noted), but there are multiple fleet lines that can use the same ship in different quantities.
  • Essentially, there are a variety of types of fleet designs, and some of them give you more or fewer of certain types of ships. We are shifting our quality scoring to look across those types of lines, since the quality of a given fleet line purpose is always now fixed instead of variant.
  • Confusing? Yeah... sorry about that. Overall this is not relevant except to modders and developers.

• It's worth noting that there IS still variance in how many strikecraft and frigate lines there are in a given fleet, when you find them out in the wild.
  • The unlock variance setting already did nothing about that, and for now we're keeping that in place.

• Strikecraft random ranges are no more, so here's what we've settled on based on the prior ranges.
  • Bear in mind that even at mark 1 these are inflated now for player ships, so this is lower than you'd actually see in the game.
  • Range 10-18 is now 14.
  • Range 5-7 is now 6.
  • Range 5-9 is now 7.
  • Range 6-10 is now 8.
  • Range 6-12 is now 9.
  • Range 7-14 is now 10.
  • Range 8-15 is now 11.
  • Range 10-15 is now 13.
  • Range 10-18 is now 14.
  • Range 12-20 is now 15.
  • Range 12-25 is now 19.
  • Range 15-25 is now 20.
  • Range 25-40 is now 32.
  • Range 30-40 is now 35.
  • Range 20-35 is now 28.
  • Range 30-60 is now 45.
  • Range 40-55 is now 48.
  • Range 40-60 is now 50.
  • Range 60-80 is now 70.
  • Anything that already had the same min and max still has that same amount.
  • We are going to soon do some analysis of ships based on their strength at their ship caps and make sure that things look reasonable as divided out by tech, but this is the first step.
    • Previously it was pretty hard to do that sort of analysis since any given ship line could randomly be often 50% weaker than normal or 200% stronger than normal.

• All of the other types of ships that had a range of default values now have a cap in the middle, also.
  • Again, in all cases we can now balance from there, which is nice.

Mark Level Scaling Overhaul

• You'll want to read this for the detailed math behind the strikecraft: https://docs.google.com/spreadsheets/d/1PxvEq3jOEPDt7RWAZXaJEatLx_unwvsH2NgMFKoguac/edit?usp=sharing

• And this has the detailed math for everything else: https://docs.google.com/spreadsheets/d/1acg0Q7C0wyXz3tHEO21UMifb3EcsihGRSJB8m6yTbX8/edit?usp=sharing

• The following has been removed from the Balance_MarkLevel table:
  • attack_multiplier
  • most_everything_squad_cap_multiplier_list
  • frigate_squad_cap_multiplier_list
  • hull_points_multiplier
  • assist_range_multiplier
  • shield_radius_multiplier

• A new Balance_MarkLevelScaleStyle table has been added, which has most of those things.
  • human_squad_cap_multiplier_list replaces the two other squad cap multiplier lists.
    • This is named as such to make it clear that this really only applies to human caps, since pretty much all other factions use their own style of caps unrelated to fleet lines.
  • Everything else is the same in terms of the content, although they are now all expressed as lists in order to make these more concise to define.
  • While we were at it, we also split out the multipliers for shields and hull health, because why not. Flexibility is a good thing.

• GameEntity definitions now have a new, optional, mark_level_scale_style.
  • This allows a nonstandard MarkLevelScaleStyle to be assigned to any entity that you wish.

• There is also the ability to define defaults for entire categories, like strikecraft or frigates or turrets or whatever.
  • In the event that any of that is missing or not defined, then it just falls back on the defaults.

• The defaults are exactly what they were in the game up until now, although soon this will be applied to increasing minorities of ships.

• Strikecraft caps for human fleets have been dramatically rebalanced.
  • Please see v3 of this sheet for all the details: https://docs.google.com/spreadsheets/d/1PxvEq3jOEPDt7RWAZXaJEatLx_unwvsH2NgMFKoguac/edit?usp=sharing
  • With these changes, essentially:
    • You get more strikecraft in general, but the AI does not.
    • You get vastly more strikecraft at mark 2 and mark 3 compared to before.
    • Overall, upgrades through mark 4 are far more powerful and represent a much greater progress toward the total power of mark 7 (mark 4 for humans is now 62% as powerful as mark 7, rather than 35%, because of the increase in ship caps without changing any actual ship stats).
    • Once you hit mark 5+, there are no more additional ships added to your ship line. However, the total ships you have at mark 4 is already higher than what you had at mark 7 prior to now, so this is not exactly a punishment.
    • The total strength boost of a full cap of mark 7 ship now versus before, for humans, is 21%. Aka, if you fully upgrade a ship line to mark 7, it's now 21% more powerful than in prior versions of the game. This does not affect the AI in any way.
    • THAT said, you are still seeing diminishing returns on the power curve as you upgrade above mark 4. Two fleet lines both upgraded to mark 4 are now very likely to be far more powerful than a single mark 7 line.
  • These changes make it a lot less obvious that you should specialize in any single tech, and make a wide variety of tech strategies hopefully much more viable.
    • This is not the end of the story in terms of making that so, but for strikecraft this is the big item.

• Frigates now get progressively more durable in higher mark levels, and also ramp up their attack power a bit faster than before.
  • Overall very-high-mark frigates are considerably more fearsome than before, and they are also beefier in the middle.
  • Since we are not adjusting the caps of frigates up a bunch like we have for player ships, this was what made sense for this case.
  • Detailed math is here: https://docs.google.com/spreadsheets/d/1acg0Q7C0wyXz3tHEO21UMifb3EcsihGRSJB8m6yTbX8/edit?usp=sharing

No More Frigates For AIs

• Added a new is_deprecated="true" flag that allows the game to automatically remove any instances of a given ship in old games, as well as hiding it from any future encylopedias in the game.
  • This has been applied to the AI Raid Frigate.

• All of the frigates have been removed from any AI groups that were using them.
  • This essentially means that AIs cannot build their own frigates anymore. They might make zombies out of some, but that's about the extent of it.
  • This goes back to a much older design for the game, where we wanted AIs to only have Guardians, and humans to have Frigates as their own unique tech.
  • AIs have sometimes been using frigates excessively, particularly stacking too many of them in guard posts as if they were just strikecraft, so that's really not something we want to have happen.

• AI ship groups and fleet design templates now check to see if they are empty, and throw errors if they are.
  • The idea is that if these are used anywhere in the game, then you'll wind up with funky empty waves or planets or fleets or whatnot. The self-check is a help there.

• AI Ship Groups are sometimes intentionally left empty (usually to not have something large and scary on every planet, and use the weighting system to accomplish this).
  • These ship groups now have a is_intentionally_empty="true" tag on them, so that we can know that and not complain about them during automated checks.
  • Any ship groups that have that tag but also have ships will now be complained about, as well.
  • For ship groups that are essentially deprecated, but which are in existing savegames, we now have a if_empty_copy_items_from tag that lets us assign something general into it for cases where it was previously used.

• Added a new is_considered_visible_part_of_the_ai_faction_for_what_can_own_purposes flag for various AI subfactions.

• For any AI factions, or any factions with the above flag, any frigates that they have either directly out, or in guard posts, will be removed and a little notice is posted silently in the log when you load a savegame.
  • There were some truly insane numbers of frigates in guard posts in some places before, so this will be quite a nerf to some previously-unlucky savegames.
  • If there were frigates in an incoming wave, or something to that effect, those will still arrive and are not cleared out.
  • Thanks in particular to Democracy for alerting us to how crazy this could get.

Frigate Balance

• Frigates (in the base game and DLC1) now have a bit more hull, and about a 16% reduction in metal cost.
  • Previously they just evaporated too easily.
  • Additionally, these took too long to rebuild, and also were a poor deal in metal-to-health ratios compared to most strikecraft.

• Frigates no longer go up in ship cap at all, unlike in prior versions of the game.
  • For station-keepers, it was annoying to have to go back and fill in more.
  • For absolutely every kind of frigate, a ship cap increase of 1 on a cap of 2 (as an example) is a 50% increase in metal and energy costs, which is pretty darn turbulent on your economy.
  • Overall with the perception being that you need to have a certain number of frigates for them to be useful at all, it's better to just start at that number and then let them get beefier as they mark up. This gives higher-mark ones more staying power, without the added expense or gaps-in-turnover during large battles.
  • When you pair this with the buffs to frigates from mark level stats, it's hopefully enough to make them really interesting, but one reason we're headed into a beta branch is to find out if all the numbers check out.
  • So, caps for frigates have been adjusted as follows:
    • Previously 1, now 4.
    • Previously 2, now 6.
    • Previously 3, now 9.
    • Previously 4, now 12.
    • Forcefield frigates are the sole exception, and they get a total of 3 rather than 4.
  • We may need to make adjustments to some other specific frigate lines, but we'll just see how things play out.

• The max cap of the Chained Dark Spire Eidolon station-keeping frigate is now 2, rather than being the 4 it would be if it was with other frigates.
  • Same with Tamed Macrophage, another thing you can purchase from the Zenith Trader.
  • Same with Scrubbed Aberration and Scrubbed Abomination.
  • It's worth noting that all of these are markless frigates, so they always previously had a max cap of 1 no matter what. Having a max cap of 2 might make them a bit more fun.

Turret Balance

• There is a new attack_range_multiplier_list that we can use on mark level scaling styles to increase the range of certain units as they mark up.
  • This is not something we would ever have wanted to generalize previously, but now we are able to sub-categorize and so it's nice to have.
  • In most all cases, we will not actually use anything other than 1.

• The assist and attack range multiplier from mark levels now only applies when it's not already a sniper-level range, or if it's not melee-range.

• The game now automatically checks for human and AI styles of turrets, and differentiates them.
  • If something is trying to use a turret as both a human and AI type, then it now complains about that.

• All of the various ensnarer turrets were being used by the AI as well as being used by one specific type of human battlestation.
  • This was problematic, as the costs were too high for the ensnarer turrets, they did not die to remains, they had the word AI in their name, and a bunch of other things.
  • These have now all been split out, although if you happened to have the AI ones in a previous game, you still will now, and they will scale wrong and still behave wrong. These were already rare, though, so it should not affect too many games.

• The game now has different scaling multipliers set up for human, AI, and other turret types.
  • Right now, the AI and other turret types just keep using the "Original" style that give them the same stats they have had prior to this new build.
  • The human turrets, however, are different.
    • Human turrets no longer gain any ship cap as they go up in mark level.
    • The more aggressive scaling from the frigates scaling is now used, for much the same reason.
    • The assist range of these turrets is like the Original style, though, not like Frigates.
    • And finally, there is actually an increasing attack range on the human turrets per mark level, until mark 4. This is new and not something we will do much with other units, but is great for turrets since they cannot move.
  • Overall with human turrets, the goal here is again to change the curve of power like we did with frigates, but to stop increasing their ship caps as you level them up.
    • It's super annoying having to go back to all your planets and place new turrets every time you increase a tech that affects some of them, and now that's no longer a thing.

• Because turrets no longer increase in cap as they mark up, we needed to increase the base caps for all of them that you get.
  • Overall we were seeing around a 260% increase from mark 1 to mark 4 in the prior build of the game, so for now we're going with that in general.
  • Previous cap of 5 is now 13.
  • Previous cap of 7 is now 18.
  • Previous cap of 15 is now 40.
  • Previous cap of 19 is now 50.
  • Previous cap of 20 is now 52.
  • Previous cap of 24 is now 62.
  • Previous cap of 30 is now 78.
  • Previous cap of 38 is now 98.
  • Previous cap of 40 is now 104.
  • Previous cap of 60 is now 156.
  • Previous cap of 75 is now 195.
  • Previous cap of 90 is now 234.
  • Previous cap of 180 is now 468.
  • Wow some of these numbers look large. Well, you certainly won't be hurting for places to spend your extra energy, anymore.
    • With these numbers you can definitely set up whatever sort of coverage you want on a planet, and then how upgraded your turrets are determines how effective the coverage is.
    • In the past, it was possible that you wouldn't have enough turrets to actually get full coverage on a planet until you did some upgrades. Now that isn't a problem, energy-willing.
  • Puffin reminds me that the multiplier_to_all_fleet_caps="0.2" means that these will actually show up as 1/5 of the values you see here, so it's not going to be such a culture shock. That makes more sense!
    • This applies ONLY to turrets.

More Balance

• Minefield balance:
  • All minefields now have a bit aggressively more than 2.6x the ship cap that their average would have been before.
  • Any minefields with shield points now have those as hull points instead.
  • There is a new mark level scaling for minefields, which does the old "original style" hull scaling, but which keeps the more-aggressive style of attack scaling from frigates and turrets.
  • There are way more minefields in DLC2 than in the base game, and there are none in DLC1, so some of these are not visible to most folks yet.

• We no longer try to automatically calculate IsNonTurretDefense.
  • That had been returning quite a few false positives, such as all the ensnarer turrets and a lot of other factional structures.
  • That probably didn't matter until now, and it's been the same calculation for the last 3-4 years, but now we're actually doing some differentiation on this and so it needed to be corrected.
  • is_stationary_forcefield has been added to let us identify those, and is_non_turret_defense has been added to let us identify THOSE.
  • Essentially there are a few forcefields which are the latter, and the former consists of the tachyon array, tractor array, and gravity generator. We're excluding minefields from that latter category.

• Non-turret defenses now have their own scaling which is basically just the frigates for hull and shields, but the original style for weapons (of which they probably have none).

• Stationary forcefields do their own thing, with no weapon scaling at all per mark level (again they probably have none), but then with an increasing divergence in hull and shield multipliers.
  • Their amount of extra shielding gets really intense in the higher marks, and it's a pretty interesting effect
  • Details on their calculations: https://docs.google.com/spreadsheets/d/1acg0Q7C0wyXz3tHEO21UMifb3EcsihGRSJB8m6yTbX8/edit?usp=sharing

• The ship caps on the stationary forcefields and non-turret defenses both no longer increase with mark level, so they also get some extras.
  • Essentially for forcefields, logistical now gets 6, military 4, economic 2. Previously it was 2, 1, 1, with increases by mark level.
  • For the non-turret defenses, they get an approximate 2.6x-ish boost, but it's a bit varied since we already tuned it up a little. If it feels like too few or too many, please let us know!

• Then added a new is_engineer="true" for those style of ships, and a new is_default_for_engineers="true" to go with them.
  • The overall design is that engineers no longer increase in count as the mark level goes up, but their hull points and shield points go through the roof like a forcefield does (multiplier-wise, anyhow).
  • Engineers already have improvements to their ability to assist built into their designs, but now their increases per mark level are doubled from what they were. This makes individual engineers FAR more powerful with even just a single mark level upgrade.
  • Since the caps of engineers don't increase, we also have increased the caps of them.
    • Overall their numbers are generally doubled or tripled, depending on the kind of engineer and the context.

Guardians

• Guardians and Dire Guardians now have their own scaling, which you can see the math on here:
  • https://docs.google.com/spreadsheets/d/1acg0Q7C0wyXz3tHEO21UMifb3EcsihGRSJB8m6yTbX8/edit?usp=sharing
  • This gives them much more aggressive attack and hull and shield boosts, and even breaks the normal rule and gives them shield radius boosts.
  • Until now, the AI has been nerfed pretty heavily so far in this rebalance in general, and the player has been getting substantial buffs left and right.
  • This is the first substantial buff to the AI, and it makes guardians actually relevant again -- at the topmost levels, the guardians are about twice as dangerous.
  • That said, the overall growth is pretty steady and linear, just higher, which makes sense for opposing forces. The curves for enemy growth and your growth are very different, as they are meant to serve different purposes.
  • It's possible that too few guardians are being seeded -- if that's the case, please let us know. They shouldn't be everywhere all the time, but they should be enough of a presence to be freaky and a problem when you see them.

Prior Release Notes

AI War 2: Interlude And Refinement