Valley 1:Beta Series 3 Release Notes

Beta 0.902

(Note: this prerelease is not available yet, we're still working on it)

• Fixed an exception that could occur mousing around the inventory.
  • Thanks to Minotaar and kof91 for reporting.

• Put in some attempted preventative handling where if a music or ambient-sounds file does not exist, the game will tell you about it and not try to load it, rather than erroring out. We're not sure if this is even the problem, or if it's something internal to the unity engine itself, but at least this eliminates one possitiblity for the occasional "GameMusic+<LoadMusicAsyncInner>d__0.MoveNext ()" error.
  • And actually, now put in yet another round of preventative handling here that should be even more reliable. If anyone sees this again, please let us know.
  • Thanks to jerith providing a semi-reproducible case of this.

• Put in several fixes to water/lava seeding in interiors, so that when automated water is seeded it will work better and not have strange gaps next to stairs, etc.

• Fixed a bug with stairs being possible to hang in an invalid position over no solid ground, leading to wonky physics.
  • Thanks to Smiling Spectre for reporting.

• Made it so that in the oceans and ocean shallows, interior boss rooms are now filled with water -- that way the bosses and enemies from that kind of region can actually seed!
  • Previously, these rooms were just completely devoid of bosses, making them automatically won. Whoops!
  • Thanks to jerith for reporting.

• Fixed a number of issues with monster spawner spawning, especially in oceans, of late. This was often leading to far too few monster spawners in something like the ocean or the ocean shallows boss chambers, and no spawners at all in some boss rooms of regions of those sorts.

• Previously, monster spawners were accidentally being removed when bosses were destroyed. Fixed.
  • Thanks to Smiling Spectre for reporting.

• Previously, the boss chunks in the intro mission were not changing into former boss rooms when their bosses were killed, nor were they having their boss music stop once the bosses were gone. Fixed.
  • Thanks to GrimerX and others for reporting.

• Updated our spritefont engine so that it now has some mappings of some unsupported unicode characters ("fancy" apostrophes and double quotes, for instance) and now maps those to their prosaic ASCII cousins. This keeps those sorts of characters from drawing invisibly and making it look like we can't tell I'd from Id when someone typed up some dialogue in MS Word.
  • Thanks to jerith for reminding us about this.

• Updated our spritefont engine so that any invalid characters it finds will be rendered as underscores rather than just being rendered invisibly and causing glitchy problems. That way the offending textual definition in whatever xml file can be found and corrected, versus the game itself just appearing glitchy.

• Completely redid the premise of how "ongoing conditions" are applied to players and enemies. This basically refers to enchants, and effects that get applied to you like being set on fire, running from storm dash, etc.
  • Previously, most stuff was being handled on addition and on removal of the effect. So if your damage was to be doubled from storm dash, it would set those flags on your character when storm dash got turned on, and would remove those flags from your character when storm dash ended.
  • That approach had the advantage of being very CPU-friendly, doing the least possible amount of calculating while a condition was applied to you or an enemy. The thing is, there really don't tend to be that many enemies or characters in a chunk in the first place to where this would become a CPU issue (it would take thousands at once to really become a problem), and it had the drawback of being rather brittle from a code standpoint: if small code errors were made, then you could wind up with ever-spiraling more damage incoming while you storm dash, or immunity to fire not going away after fire shield was taken off, or whatever else.
  • Those drawbacks weren't a big deal when we had only a handful of ongoing conditions, but now there are several hundred thanks to enchants, and this was now a major issue. And in fact, the ever-spiraling-upwards damage from storm dash was happening in recent versions, which is what prompted this change in the first place.
  • The new methodology wipes all of the from-ongoing-condition flags back to their default states every frame, and then loops over all the ongoing conditions that are currently applied to you to get the final stats, and then uses those stats for the rest of the frame. That means a bit of extra processing per frame, but any CPU from the last 10 years will laugh at the difference, and it keeps us programmers from introducing accidental bugs that are a real pain to hunt down.
  • The _downside_ of this change is that it's a fairly substantial change, and in the short term we may have just introduced a raft of new bugs of the sort that we're trying to prevent. If you find any such things, please let us know and we'll fix them. Past a few days from now, this is something that will drastically reduce bug counts from this sort of thing, though, and it's possible that even in the next few days you won't find anything amiss from this (we checked ourselves pretty carefully, but that only goes so far).

• Fixed a bug where things that modified your incoming damage -- such as enchants reducing that damage, or miniature and storm dash increasing that damage, were actually having their modifiers added TWICE. This made a 2x boost in incoming damage actually a 3x boost in damage, and so on.
  • Thanks to GrimerX for reporting this one a month or so ago -- it's a bit of an older bug.

• Improved the way that double-tap handling is calculated. If any other key is pressed in between a double-tap, it no longer counts as a double tap.
  • In other words, previously hitting Left Right Left would cause storm dash to the left if you hit it fast enough, same as just Left Left would. Now Left Right Left doesn't count as a double tap, and thus no storm dash gets activated.
  • Thanks to khadgar and Toll for reporting.

• Put in a couple of sanity checks to prevent players and NPCs from ever having health or mana that are below a certain pitiful amount (previously, in super rare cases the max mana could actually be negative on a character, which was completely glitchy).
  • Thanks to Minotaar for reporting.

Beta 0.901 I May Or May Not Want To Be That Guy

(Released March 2nd, 2012)

• Fixed a bug that was causing tombstones to seed in stash rooms instead of enchant containers!
  • Thanks to Moo for reporting.

• Fixed a bug from a couple versions ago where a monster would always only drop one "item drop" entity on death, and packed everything into that one drop. So bosses dropped a single "shard" (that was actually all of them) instead of a cloud of them, etc.
  • Thanks to BobTheJanitor for the report.

• Battlegrounds:
  • Now work in MP. Previously ally minions would just ignore enemy minions and run right by them. Enemy minions generally did the same. Now they actually target each other and fight, etc. This should also help NPCs in MP NPC-rescue missions actually shoot and be shot at.
    • Thanks to Underfot for the report.
  • Minions now close up the range a bit tighter before standing still.
  • Along with that, they now try to spread out a bit better so that your allies aren't all standing on one spot (and generally taking aoe damage from even simple projectiles).
  • Fixed a bug that was effectively creating a 0.5 second "global" cooldown on ally shots in the sense that at most 1 ally could shoot per 0.5 seconds. This was causing a lot of strangeness, and some other bugs elsewhere in the game.

• Since the tombstones-in-stashes was so interesting, now tombstones will tend to be seeded in stashes in general in addition to their other stuff.

• Lava escape missions can no longer be seeded on the world map as missions. You'll now only find them as secret missions.
  • These are meant to be very punishing for players who like that sort of thing, and as such having them relegated to secret-only status makes good sense.
  • Thanks to many players for complaining about this, including rchaneberg, Terraziel, and BobTheJanitor.

• Fixed a bug where the open-region-map and open-dungeon-map keybinds couldn't fire while a popup (tutorial message or otherwise) was showing. Particularly amusing when the tutorial popup is telling you to open the dungeon/region map.
  • Thanks to BobTheJanitor for the report.

• Fixed a bug where health drops could heal again after being flagged for removal after the first heal performed by them, causing them to be much stronger in multiplayer due to latency.
  • Thanks to TyberZann for the report.

• Fixed a bug where when a continent was generated that had no valid location for a port region it would just dump the new port region on world coordinate 0,0, complain about it in the chat log, and call it a day.
  • Now if there's no valid location, and there's already a port region it just doesn't seed any more for that continent and done. If there are literally no port regions on the continent (and no places to put one) it just ditches the newly generated continent and makes a new one.

• Tooltip for Miniature no longer claims that double-tapping down will activate it.
  • Thanks to MaxAstro for reporting.

• Fixed an error with a particular boss room map that could lead to that room being flooded with lava.
  • Thanks to Terraziel for reporting.

• A sound effect now plays when "do not worry" messages appear.

• An explosion sound effect now plays when urban predator missiles explode.

• The urban predator plasma shot now uses a different sound effect.

• Casting a light snake now uses a different sound effect, as do teleport, greater teleport, plasma bolt, and energy pulse.

• New spell that can be crafted: Storm Fist:
  • Mid-power touch-range spell that knocks you and your target back when used. A little bit useful for harvesting, but actually works better as a minor way of knocking yourself up to ledges that are just out of reach, along the ground while ducking, or similar. Just don't knock yourself into any passages from which you can't escape!
  • Thanks to KDR_11k for suggesting.

• Whenever NPCs move between chunks, their dialogue is reset. This should prevent them from complaining about wanting to be rescued after they are already at the settlement, but let us know if that doesn't work.
  • Thanks to BobTheJanitor and Ixiohm for reporting.

• Mana costs should now reflect their true values on spell descriptions if those costs have been reduced by enchants.
  • Thanks to Terraziel for reporting that they previously did not.

• The cooldown entries are now shown in a greatly condensed format in spell descriptions; rather than taking up one line per type of cooldown, they now show on a single line in a much less wordy fashion, sorted by length of cooldown time.

• Damage Per Second, Damage Per Mana, and Damage Per Second Per Mana are now all shown in the description of offensive spells.
  • And yes, these numbers take into account your current tier, any enchants you're wearing, your own innate stats, and so on. Just like all the other spell stats.
  • Thanks to BobTheJanitor and Terraziel for suggesting.

• Fixed a small typo in the description of wooden crates.
  • Thanks to Hyfrydle for pointing this out.

• Fixed a typo referencing "Guaridan Ilari".
  • Thanks to terraziel for pointing it out.

• More dialogue work has gone into a lot of the NPCs from Marisa.

Major Improvements To Enchant Acquisition And Design

• Enchants are now given every 200 enchant-points (for reference, each container gives from 8 to 12 points), rather than a sliding scale based on how many enchant points you've "spent" in the past.
  • In fact, there is no longer really a concept of "spending" the points on item effects, etc. The only purpose of the points now is to space enchants out temporally.
  • Incidentally this means that the progress indicators are now always right, because the question "how many points will the next enchant cost?" is now answerable instead of merely guessable.

• Each enchant type (that can be given from containers, not the stash-room-ones) now internally has 5 "quality tiers".
  • The first quality tier is what the first enchant of that type generated for you in that world will always be; just the base effect which is always the same.
  • The second quality tier is what the second enchant of that type generated for you in that would will be; the same base effect, but with one optional effect.
    • Additionally, there is a low chance that a second-tier enchant will be an "Uncommon", and the optional effect will be chosen from a more powerful set.
  • The third quality tier is what is generated after the game has already generated a certain number of enchants of that type for you (i.e. after you've had a few second-tier ones). Enchants of this tier will have the base effect and two optional effects.
    • There's still the chance of an "Uncommon" with one extra-powerful effect, but also a lower chance to be a "Rare" with both effects being pulled from the more powerful set.
  • And so on. Right now there are only two more tiers (3 and 4 extra effects, respectively), but it is likely that more will be added later if this system works well.
  • These quality tiers, it is important to note, are not shown but can kind of be inferred. Since multiple bonuses to the same stack condense into one single stat (three damage boosts of varying numbers just combine into a single damage boost showing their total, but this is still quality tier 3 rather than quality tier 1).
    • So in some senses, we really shouldn't even be telling you about quality tiers, because they are not visibly manifest in the game. But since there's been a lot of discussion about the enchants progression of late, we felt we'd share at least a few of the details here.
    • The net effect to players is that the first time they get a kind of enchant (say, acid gills) it just does the base effect. Then for a while after that, it has the base effect plus a little better, unless they find an uncommon version of it (yay that player, in that case!). Then after a while, they're getting even better versions, and so on. From our testing the progression feels really natural and should be pretty familiar if you've played other games with procedural loot.
  • One other key thing to note is that, internally, these quality tiers are tracked per base enchant. This is important for several reasons.
    • First of all, if you use something like the fire seeking enchant in order to better your chances of getting fire-related enchants, that means you'll actually get to the higher quality tiers of fire-related stuff somewhat faster. But a lot of your other colors will lag behind.
    • Secondly, whenever we add a new enchant base type, you start at the bottom. Right now we have double and triple jump enchants. If we later add a quadruple jump enchant, for instance, and you are on continent 15 and have put in 500 hours into the game, you're still going to start with quad jump quality tier 1 and have to work your way up from there. It will be significantly underpowered compared to the other enchants you're getting for a while.
      • This is actually good in a lot of senses, because it doesn't mean you need to restart your whole world every time we add a new base enchant type, just so that you can go through the progression in a satisfying way. Our goal with all added content for the game (now that the base mechanics of the game itself have settled down) is that no matter how far along you are in the game, you'll get a fun and interesting progression with that new content. That's really the only way to keep long-term interest in a given world, rather than having players periodically starting over new worlds every time a bunch of new content gets added.

• The number of optional effects available to each base enchant type has been _dramatically_ increased, and that's without the other whole new kinds of enchant effects (mana-regen-buff, etc) planned for 1.0.
  • Accordingly, the game no longer tries for any kind of brute force duplicate-prevention. You only get one of the first tier for each type, so no duplicates there. It's possible at the second tier but not terribly likely since you only get a few of those before moving on to tier 3 (it will happen, make no mistake, but we're talking a very occasional pair of duplicate enchants, not a string of 17 identical things).
    • At tier 3 the game would have to roll the same for both optional effects from a fairly large set, given not-very-many chances before progressing to tier 4, and so on. So the chances of getting fully duplicate enchants goes down almost geometrically as the quality tiers rise. For that matter, so does the variety (though sometimes the variety is just a percentage point or two of some stat, it's still variety even in those cases -- we're not sure how many unique enchant combinations there are now in the game, but tens of thousands would probably not be a bad guess).
    • If you're playing a world started before this version you'll likely get some duplicates of the first-tier stuff since you already have them, but it should be a pretty shortlived time that's happening to you. And now that you're getting new enchants on a more predictable basis, rather than on ever-increasing intervals, that won't be as annoying in the first place.

Difficulty Split: Combat And Platforming

• The game has had its "action difficulty" split into two separate difficulties due to a growing divide in players over the platforming aspects of the game. Instead of one "action difficulty" that covers everything, there is now a "combat difficulty" which covers everything combat-related, and a "platforming difficulty" which covers everything platforming-related.
  • Thus players who are not as good at combat, but love hardcore platforming, can get the experience they want.
  • And likewise, those players who hate platforming but want as stiff a fight as the game can give them, can also get the experience they want.
  • And of course everything in between.
  • Thanks to a number of players, but BobTheJanitor and zebramatt most of all, for making it clear to us that this change was needed.

• The new platforming difficulty levels are:
  • 1 I Have No Desire To Be The Guy
  • 2 I Am Afraid Of Heights
  • 3 I Can Jump, Thank You (Default)
  • 4 I Get Mistaken For A Certain Plumber
  • 5 I Am Not The Guy, But I Am Close (This is what Chris likes)
  • 6 I Am Already The Guy

• The general rate of lava rise in the lava escape missions has been toned down some. What was difficulty 4 on the older "action difficulty" scale is now difficulty 5 on the new platforming scale. And consequently, difficulty 6 has been toned down a bit because it was probably impossible for anyone. Difficulty 4 has thus also become much tamer, and difficulty 3 slightly tamer. Difficulties 1 and 2 are unchanged, as they were already incredibly tame.

• Tooltips in the game wherever you can change the combat difficulty now make it clear what the combat difficulty actually affects:
  • Monster health, speed, telegraphing time, cooldown time.
  • Monster spell speed and damage dealt.
  • Accuracy of certain monster spells like circle of fire.
  • Damage to players from combat effects such as being set on fire.

• Tooltips in the game wherever you can change the platforming difficulty now make it clear what the platforming difficulty actually affects (so far just this):
  • Amount Of Damage Taken From Falling
  • Rate of lava rise

• The method for calculating falling damage has been completely changed.
  • Old Method (based on number of pixels fallen):
    • >= 4000: 100% base health
    • >= 3400: 80% base health
    • >= 2800: 60% base health
    • >= 2200: 40% base health
    • >= 1600: 20% base health
    • > 1000: 10% base health
  • New method:
    • Must be >= 1000 damage
    • Base multiplier calculated as: distance / 10000.
      • Note: this part of the calculation in particular could stand to be more interesting; we welcome feedback on it, but decided to start simple.
    • Base multiplier further multiplied by difficulty:
      • I Have No Desire To Be The Guy: 10%
      • I Am Afraid Of Heights: 50%
      • I Can Jump, Thank You: No Change
      • I Get Mistaken For A Certain Plumber: 150%
      • I Am Not The Guy, But I Am Close: 200%
      • I Am Already The Guy: 300%
    • The fully adjusted multiplier is then multiplied against the base health of the character, and that is the final damage dealt.
  • Thanks to zebramatt for getting us looking at this in general.

Major Improvements To NPC Rescue Missions

• Completely did the pathing work for the NPCs being rescued:
  • They can navigate themselves around whatever obstacles, like the shadow bats do. They don't have to have line of sight to you in order to work their way around complex geometry, and you can leave them behind with confidence that they will catch up (presuming that monsters don't kill them while you ditch them).
  • They are now perfectly capable of following you through any terrain that is at least one tile high or wide. Their hitbox in general is now less than one tile tall, which is needed in order for the pathing noted above to work. It's also needed in order
  • They are now able to fire while moving, so they will continue to help you but you won't be forced to wait for them to fire at something they have no hope of hitting.
  • They no longer try to keep a "safe distance" away from you, although since they move a bit slower they will tend to lag a slight bit behind which serves the same purpose. Thus there are no longer cases where you can stand on the door to the exit of the area and they can't escape with you (talk about frustrating to get to the end and then have them not go through!).

• Fixed an issue in prior versions where if you left a room with an NPC being rescued, the NPC would revert back to their original position in that room (While still being vulnerable).
  • Though this is difficult to test with the above fixes, so if you see this recur please let us know.

• Previously, NPC stats were not being increased with the continent tier. Meaning that rescue missions on tiers greater than one had vastly higher difficulty thanks to their too-low health.
  • Now any NPCs who are not in a settlement get the same tier buff from the continent that ally minions get.

• Fixed a bug where sometimes NPCs who were rescued wouldn't actually make it to the settlement alive, but rather would die when you left the mission.

Previous Release Notes

AVWW - Beta Series 2 Release Notes