Poison effect

[Sar]

Awesome! That could lead to interesting tactics like Chei players quaffing strong poison to strengthen their Slouch. Would be a nerf to Chei Nagas though since they're rPois - unless they learn PArrow, that is.

[dpeg]

Sandman: That's a great idea! The advantage over any damage-based model (like the one I suggested) is that we'll have a much easier time making poison different from the rest.

So I assume that "poison damage" would mean: (a) some damage, (b) a status effect. Different levels of poison ("red poison" etc.) would translate to more damage and worse status effect -- this implies that the effect has to be scalable. Obviously, we want something where the connection to real life poison makes at least some sense. Slow movement fits already, I also wondered whether we could sneak it worse casting abilities.

roctavian: Yes, you got it correctly but I withdraw my proposal in order to explore Sandman's idea.

[Sar]

To try and be a bit less useless: I don't really like that idea because suddenly being slow is Very Bad and will probably break the balance - especially early game balance - a lot, as poison is really common. Also scalable slow, so you got hit with some poison and now you're slow, how slow? Can you run to those stairs? I guess you can... whoops you can't.
How about trying to make poison relevant by being a threat at killing you directly? There are already enemies that can slow you.

[tompliss]

A scalable slow could be nice, and if it needs to deal some damage to slow you, it will be different from the "magical" slows.

[dck]

It's really bad to add random things like "slow but not normal slow" that don't let you know at what speed you're going to move or to have poison increasingly screw with your failure rates so you have to keep checking them when you're in... well, when you're basically anywhere since poisonous monsters exist everywhere.

[Sandman25]

It's really bad to add random things like "slow but not normal slow" that don't let you know at what speed you're going to move or to have poison increasingly screw with your failure rates so you have to keep checking them when you're in... well, when you're basically anywhere since poisonous monsters exist everywhere.

Well, it could increase movement delay by constant value: 0.1 for yellow (anti-boots of running), 0.2 for red (anti-Swiftness).

[Lasty]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

[dpeg]

Sandman: Yes, like this. It will be sufficient for players to know that they cannot walk away as usual. Bells and whistles can be added on top of this, but the fact that plain poisoning affects your tactical situation like this will matter at once. There'll be complaints about how unfair this is soon enough, but I'd rather be forced to stand my ground from new-poison than to mash-5 and hope the HP bar stops at 1 under old-poison.

[crate]

Poison slowing sounds (in a vacuum, i'm not considering how crawl monsters actually use poison here) like a much better idea than poison dealing damage, to me. Just-dealing-damage poison is something that if I were making crawl I would just remove entirely since I don't see a point to it at all; slowing poison at least does something that is potentially worth having in the game.

[dpeg]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

No, it means that GDD is a decent place to get ideas and a horrible place to polish them. Because posters like you will always be around --and plenty-- to whine in advance.

[rebthor]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

No, it means that GDD is a decent place to get ideas and a horrible place to polish them. Because posters like you will always be around --and plenty-- to whine in advance.

It's always nice when you give condescending answers to posters.

[earLOBe]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

The chance of poison being fatal at the moment is also mostly random. I'd rather have that randomness be a matter of tactical constraints instead of invisible numbers. Making poison take away escape options also synergizes interestingly with the enemies in swamp (amphibious hydrae) and snake (constrictors), and I guess to a lesser extent orb spiders, to the point where that may require some rebalancing.

It would require some rethinking of how poison works with Formicids, though - perhaps speeding/slowing effects should be differentiated into magical ones, which are blocked by stasis, and natural ones, which aren't (I struggle to think of a natural source of speed, unless we allow Fo to berserk which there seems to be a consensus against). It'd either make their poison problems meaningless or be a buff to Fo of Chei (which can't run away anyway).

[dpeg]

rebthor: Alright, let's digress for a moment and talk about communication rather than poisoning. We have a poster who arrived with a one-line message, which contained a genuinely new idea, relevant to the thread:

Perhaps poison should decrease movement speed, then it will be different enough from both rot and direct damage.

After that we get a sarcastic one-liner like this:

Awesome! That could lead to interesting tactics like Chei players quaffing strong poison to strengthen their Slouch. Would be a nerf to Chei Nagas though since they're rPois - unless they learn PArrow, that is.

This adds nothing to the discussion. It's a joke argument worthy of CYC and helps to make GDD worse. That poster then tried to rectify and gave at least some reasoning for the negative attitude. That's good because a number of posters stick with cynical one-liners throughout. It's a strawman argument, though: Sar complains that "it will be broken, because poison is common early game" -- sure, that's why we are trying to improve it. What about all the current early game poison deaths? His other point is that players get random movement speeds which make assumptions about the proposal that are completely arbitrary.
After that we have dck saying something but I don't know what, however he is supporting Sar. And then we get this marvel:

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

This is preemptive whining in all its glory: it takes two unpleasant things about the current situation, slaps them onto the proposal, and tries to kill off a new idea by being snarky. Needless to say, no reasons are given. I have reacted drastically, and you may disagree but I've seen stuff like this very, very often and that's why my reply to every

It's always nice when you give condescending answers to posters.

is clear: You say "condescending", I say plonk. Perhaps I miss good ideas from that, but I'm pretty sure that not. There is a huge gap between those who let off idiosyncratic one-liners and those who can articulate feedback (especially in-depth negative feedback).

[dpeg]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

The chance of poison being fatal at the moment is also mostly random. I'd rather have that randomness be a matter of tactical constraints instead of invisible numbers. Making poison take away escape options also synergizes interestingly with the enemies in swamp (amphibious hydrae) and snake (constrictors), and I guess to a lesser extent orb spiders, to the point where that may require some rebalancing.

It would require some rethinking of how poison works with Formicids, though - perhaps speeding/slowing effects should be differentiated into magical ones, which are blocked by stasis, and natural ones, which aren't (I struggle to think of a natural source of speed, unless we allow Fo to berserk which there seems to be a consensus against). It'd either make their poison problems meaningless or be a buff to Fo of Chei (which can't run away anyway).

Indeed, I briefly wondered about Formicids -- they're in a funny position with respect to Sandman's proposal, what with their rP- coupled with Stasis. However, that's the minor issue (because a general game mechanic vs a single species) and I am confident a natural (i.e. not artificial) solution can be found -- such a solution might boil down to a half-clause in their description explaining why poison works on them the way it does (the way we choose).

[Lasty]

Doesn't the slowness idea just change poison from "ubiquitous and annoying" to "ubiquitous, annoying, and also randomly fatal"?

No, it means that GDD is a decent place to get ideas and a horrible place to polish them. Because posters like you will always be around --and plenty-- to whine in advance.

Man, dpeg, you go from zero to angry so fast. I'm not generally inclined to throw out ideas before trying them out, but as written the idea doesn't appear to address the issues that the thread was trying to address, and that's the point I was trying to raise -- and trying to raise it genuinely, not as a snotty dismissal, though I can see how it would come off as the latter.

From some other posts, I gather that some people are under the impression that poison causing slowing would totally replace poison doing damage, and if so, then I can see how it would address the issues in the thread. However, that wasn't included in the original proposal or your restatement of it, which were the posts I was responding to. Is that what you had in mind? If not, is there something else in addition to slow which you had in mind that would address the issues that spawned this thread?

As a side note, is the proposal that poison would slow move speed or both move and action speed?

[Sandman25]

Lasty,

I didn't explicitly mention other effects of poison because I was trying to improve only one poison issue - being relevant tactically i.e. in a fight. We still need to choose a solution for "damage per turn never ends" problem but it can have little to do with poison in a fight.

[rebthor]

See, your above post gave reasons for your answer. I may not agree with your reasons but at least we can have a discussion about them. When you say something like

No, it means that GDD is a decent place to get ideas and a horrible place to polish them. Because posters like you will always be around --and plenty-- to whine in advance.

it makes me and other posters wonder why we bother to participate in discussions and frankly why you do as well.

I thanked Lasty's post because it is completely correct in my mind. Being slow is one of the most deadly things in crawl which is why curare is so strong, why Chei is so weak compared to the other gods and why centaurs, spriggans, haste and swiftness are so amazing. Giving a status effect at almost any point in the game that slows you down is very strong. Taking that effect and conflating it with one of the most common effects in the game is going to make it tough. Giving that status to monsters that appear on D:1 and 2 is going to be unbelievably deadly.

Now if you had said, "That is a interesting idea Sandman25, perhaps certain levels poison can also slow you. There is already a precedent for this from curare." you may still have gotten snarky responses. I can't say for certain. But I can say that the idea would be taken seriously instead of viewed as "are you serious?"

[galehar]

randomly fatal

It's a roguelike. Everything is random and potentially fatal.

Anyway, we already have poison that slows you. It's curare. Maybe we can try to give it to some monsters, although I always thought it was a good effect, but also good that it was rare.

[dpeg]

Lasty: I do, indeed. You had though luck that your reply rolled in after a few not-so-useful ones. A hint intended as to come across non-condescending: if you are unsure about what is meant, don't drop the negative comment (especially if it can be interpreted as a snotty dismissal) but rather ask what the mechanic is supposed to be.

It's true that Sandman didn't give a lot of context. The way I interpret his proposal is this:

• poison is a type of damage -- if it gets through, you lose HP and get the status effect
• the Pois status effect comes with a duration
• as long as it is around, your *movement* speed reduced (poison level informs by how much)

What do we get:

• no more 5-bashing, as poison is just a type of damage with a qualifier: you rest it off, but like you do Slow (single resting)
• get rid of the extremely high poison damage variance (which was the topic of the first page of this thread)
• a non-crippling tactical effect which is relevant throughout the game.

Some more analysis: the typical early game poisoner is a snake -- that monster moves faster than (almost all) player characters anyway. Against this monster, the change is a player buff. Mid and late game poisoning could actually mean more to the player than now because currently you can often ignore the damage until after the battle, whereas reduced movement speed will matter more often, I think. On the other hand, I don't think it is crippling: anti-running is a far cry from Slow, Confuse or Paralysis, and tools like !curing and rPois do exist.

The major obstacle to this proposal that I see is poison in the hands of players. The problem is the kiting potential.

[Sandman25]

I am not sure why you think being "slow" from poison is bad on D1. It is not slow, you still can kill adjacent monsters in the same number of turns. If poison damage is nerfed (it really should be if poison gets extra effect of slowing movement), survival rate for characters can increase even vs kobolds with blowguns and I hope nobody except Sp was trying to run from Adder while poisoned any way.

[crate]

oh I seem to have misread, I assumed that the proposed change was to eliminate poison damage entirely and replace it with a movement-slowing effect

this is the proposal I would have been supporting.

I do not like poison status actually doing damage from a gameplay perspective at all.

(By the way I don't think movement-slowing poison would actually change early D much at all ... if anything it might make it easier, since adders would do less damage, and nothing poisons you from range except the rare kobold-with-a-blowgun. However as part of changing poison I would probably remove poison needles entirely since I would have removed the damage-over-time component of poison so the kobold thing would go away.)

(I also recognize that unfortunately removing damage-over-time entirely from poison has basically zero chance of happening but I can hope, right?)

[dpeg]

Anyway, we already have poison that slows you. It's curare. Maybe we can try to give it to some monsters, although I always thought it was a good effect, but also good that it was rare.

I think that poison->slower movement actually makes sense in the context of curare. Curare would be the big brother of poison, as it should be (full slow, as opposed to reduced movement speed).

I've grown so unhappy with the 5-bashing required in Crawl (after every fight, essentially) that I think it's very much worth exploring ways how to circumvent it. Poison is an important but not extremely crucial mechanic, so it might be a good opportunity. (A crucial mechanic would be: "How to regain HP/MP?")

[dpeg]

I assumed that the proposed change was to eliminate poison damage entirely and replace it with a movement-slowing effect

This is exactly my interpretation. You get damage once, up front. After that, there's only the status effect left which doesn't conduct any further damage.

[Lasty]

@crate, I believe dpeg's fleshed out proposal is that poison status does not do damage over time, but rather that poison attacks do extra poison-elemental damage on hit, and then inflict the poison-slow non-damaging status for a while. So, this would allow poison attacks to do damage (or extra damage, as appropriate), but remove the problematic damage-over-time.

Edit: ninja'd.

[earLOBe]

The change to mephitic cloud would also turn venom mages basically into hybrids/debuffers, which could be interesting and even provide material for bringing back stalkers as hex/poisoners. However, something should be done to avoid riskless sting + polearm kiting, which would be horrible. I'm aware sting kiting already exists, but there's at least some risk when the monster can hit you when you stop to cast.

It'd also make -taurs with poison ammo much more deadly if found in the open without rpois - a melee character would basically be a pincushion and a squishy character would be in horrible danger. In that respect, it would make rpois more relevant in mid/late D (well, mid D/U) and vaults. I don't even want to think of what tengu reavers could do if their AI got lucky.

This change would have a ton of knock-on effects that we'll probably only exhaust when some numbers come up for the severity of the slowness (perhaps a new term?) and some games get played.

[crate]

You get shot an additional 1 time for every 10 steps you move in los of a yaktaur if you are poisoned as opposed to not poisoned (assuming the numbers used are the ones suggested earlier). This is obviously not absolutely nothing but it's not nearly as big a deal as the post above tries to make it sound.

Meph doesn't actually poison anything so it wouldn't change.

Personally I would suggest "fixing" player poison magic by removing it entirely (the spells I can think of that I think are worth keeping are: Meph, Summon Scorpions, and Spider form; these can all be moved to hexes or something if necessary) along with removing poison needles and possibly venom weapons (I think venom weapons as enemy weapons are ok, but letting the player use them is something I would not like and I don't really see a clean way to prevent that except eliminating them).... I do not find things that slow enemies to be interesting or worthwhile additions to crawl for the most part (see also: my most hated spell in all of crawl is swiftness and it's not even close). I don't expect this to actually happen though.

[dpeg]

crate: the larger the bundle of would-be changes becomes, the less realistic is any change Modifying poisoning in some form is realistic, I'd say. Getting rid of Poison Magic is not completely out of the question, but that plays in a completely different league. Heck, it might even be more contentious than the rune lock (because the lock is mechanically a simple change, easy to undo, whereas culling a spell school gets rid of actual content). I agree with the basic sentiment -- for example, kiting in spell form might be fun sometimes, but then we always have the awesome Sticky Flame for these needs.

However, in the interest of actually getting some changes into the game, I'll refrain from talking about the spell school. As I said above, kiting is the main problem with simply applying Sandman's proposal to player-based poison (on the other hand, kiting is a rather large issue in all of Crawl), but there could be simple ad-hoc solutions. For example, we could accept the glaring asymmetry for the time being (although there are developers who'd loathe to go there).

[Igxfl]

What if the damage over time effect of poison was replaced with a multiplier on subsequent damage taken?

If a player/monster took, say, +10% to all incoming damage per stack of poison, then poison would become a serious tactical threat with zero out-of-combat annoyance factor.

As a bonus, poison as used by the player would no longer reward obnoxious kiting, but would be highly effective at softening up monsters before engaging in some other form of combat.

[galehar]

I've grown so unhappy with the 5-bashing required in Crawl (after every fight, essentially)

Maybe it would help if you had to press 5 only once after a fight. I think this can be achieve by tweaking a few options. How about making it default?

[dpeg]

galehar: Always! The good interface should never been hidden behind options. Defaults should change, even if it drives some veterans to the hills.

(Regardless of that, I still think that slow moving has a lot of potential. But that idea is easy to remember, even if nothing happens for a few releases.)

[bisonbisonbison]

So:

AS-IS: Instant elemental damage + Variable DoT
DET-DOT: Instant elemental damage + Predictable DoT
STATUS: Instant elemental damage + equivalent of Catoblepas calcifying attacks?

[TwilightPhoenix]

I dislike the "poison causes slow" suggestions and this is not a "whining and complaining in advance" post.

First of all, we already have Curare, which does... Poison + Slow. Curare is a neat thing, perhaps a bit too rare, but it's good that it's not too common. Before delving into the rest of the post, I'd say that I'd rather Curare be more common than all poison slowing.

Second, while all poison inducing slow does make for more tactically interesting situations more frequently, it does not address the actual problem being discussed here: mashing 5. Poison slows you now? Whoop-de-do, just mash 5 once you're safe. In other, less sarcastic words, it does absolutely nothing to change that other than making it harder to get safer or making you reach for Curing more often.

Third, Slow as an independent effect exists. This effect is nerfed heavily in this proposal as Poison will effectively outclass it due to A) being more common and B) not needing to check MR as often as Slow does (especially player Slow). And if Poison's damage component is removed to compensate, then it becomes an incredibly redundant effect.

Fourth, building on the above, spells. If Poison inflicts slow, damage or not, then Sting becomes epic. Venom-tipped reaching weapons become "weapons-of-kite-everything". Olgreb's becomes a powerful AoE slow that can be cancelled for the player by a simple application of Cure Poison. Poison Cloud becomes "Cloud of Slowing". Poisonous Arrow becomes "Arrow of slows stuff that'd normally be resistant or immune to slow". Meanwhile, Slow and Mass Slow spells become... crap. Same for Chei's Bend Time. Spider Form becomes the "poke the foe and run away easily" form even more than it is now (and loses its offensive niche if damage is removed). And Poison Magic becomes either ridiculously good until everything is immune to poison or, if the damage is done, then just outclasses Hexes for many purposes.

Fifth. Again, Spears of Guaranteed Kiting. No. Just no.

Sixth: Lair Branches probably tend to being less dangerous if the damage component is removed. "Wait, but you get slowed!" Yeah, but once you're slowed, getting your movement speed reduced even more usually doesn't matter since you can no longer run away either way. (Yes, there'll be situations where it'll matter, but they are far, far less common than Poison is). Whereas with damaging poison, getting more and more poisoned can actually matter a lot. I mean, who cares if I'm getting hit by 80 non-damaging splashes of poison a turn once I've already been slowed? I can still Blink away from the Nagas just fine and then just mash 5 to get my speed back.

Seventh, player Green Draconian + slowing Poison Breath. No. Just no. And hell no if they're using a pole or a ranged weapon.

Eighth: especially if damage is removed, then Chei abuse. I mean yeah, Chei getting buffed would be nice and all, but it'd be dumb as hell to encourage players, directly or indirectly, unintentional or otherwise, to deliberately poison themselves to gain an advantage. Do we really want Chei followers to do stuff like cart around Poison Potions, Poisonous Chunks, repeatedly hitting themselves with Sting/Blowguns/Daggers of Venom, or drag an Adder with them everywhere they go?

There was one or two more, but I forgot them while typing this up. But, either way, no, I don't like it. And I also think it'd take a heck of a lot of work to implement just to adjust all the areas of gameplay and balance the change impacts while not addressing the actual problem we're trying to address. There are much simpler ways to A) actually address the problem and B) make poison more tactically relevant.

Also, and yes, removing the damage would most certainly mean removing the Poison Magic school, which would suck because that's one of my favorite schools (and VM is one of my favorite backgrounds, so yes I do have some bias on this side-topic). Not to mention that Poison Magic's only real fundamental problems are it's lack of high level spells and it's questionable utility in the later portions of the game. Oh, and it "encourages kiting", but you're not going to fix kiting completely without doing something dumb such as removing monster melee attacks. But meanwhile, Poison Magic offers a few interesting spells that don't behave like other spells, offers some unique synergies (Ignite Poison anyone?), and Venom Mages don't play like other Elementalists.


But, if you really want more Slow + Poison, I would be totally okay with more Curare-using threats.

[Sandman25]

1) It is not Slow, it is closer to mutation "Slow movement 1" (1.2 delay)
2) Mashing 5 has nothing to do with Slow effect. Nobody says the only change to poison is extra effect.
3,4,5 I stopped reading here) Player cannot slow monsters by poisoning them.
Please read whole thread before posting.

Ps. Please don't tell me I should read whole post before replying

[TwilightPhoenix]

I might have messed up on the Slow effect. Still though, there's too much overlap for my taste (and again, we already have Curare).

Also, mashing 5 has everything to do with it. Mashing 5 to remove Poison also clears the affiliated slow effect. Additionally, the proposal in this topic was to adjust poison to remove said 5-mashing. Therefore, adding slow does not address the problem this topic is attempting to find solutions to.

And I know monster/player asymmetry, but try explaining to a new player why all monster poison slows but only Curare Needles for the players do so? Also, some people are suggesting that player-poison gains slow or for all poison to lose damage and only slow. I'm addressing all of that as well, not just your post.

Ps. Please don't leave sarcastic, smart-ass postal-scripts. They don't contribute to healthy design discussion and are often the sort of factor that gets GDD discussions either locked or booted to Crazy Yuif's corner.

Sorry mods for backseating there, but I couldn't reply in any other fashion without being a dick in response.

[Sandman25]

I am sorry, I actually edited my post to write that PS. I should have used "Edit:" indeed.

[shock]

I'll reiterate what I said earlier because there is some support for the idea of poison slowing. Slowing isn't the only temporary and tactically relevant effect we can consider.

Another approach would be to give poison temporary and tactically relevant non-damage effects. Effects like stat damage, MP loss, slow, confuse, pass out+exhaustion, and paralyze could all fit from a flavor perspective. Poison could even function like contamination/glow where poison level is tied to severity of probabilistic effects.

[roctavian]

I agree with several previous posters -- I don't think poison-as-slowing is ideal at all. I do think that poison-as-debuff is preferable to simple damage over time, however.

A multiplier on subsequent damage is also a bit sketchy, because damage multipliers in any system quickly become central to player strategies. Games which include a damage-increasing debuff have to account for this (by constraining access, limiting the situations where the debuff can be applied, or bestowing some disadvantage to the player or an advantage to the opponent when the debuff is active) lest it become a no-brainer-gotta-use-it. This doesn't mean such a debuff has no place in Crawl, it just would have to be accounted for.

I still favor deterministic Brogue-like poison because it's simple and it addresses the 5-mashing problem, and maintains player-monster parity. But, I am with shock -- I think we can consider other effects and I especially like making poison similar to contamination/glow.

[TwilightPhoenix]

But, I am with shock -- I think we can consider other effects and I especially like making poison similar to contamination/glow.

I like this idea. Could have a variety of effects increasing based on severity and it'd never quite be the same thing. (And I'd be okay with a chance slow here in light or dark red poison since it's not "poison=slow".) I'd work it quite similar to contam in how you get secondary effects. Also, I'd say Yellow pois = just damage. Light red = minor to moderate debilitating effects such as temporary blurry vision, fumbling an attack (miss or light blow only), pausing to vomit (lose a turn), miscasting, stumbling (basically slow movement on that turn), and difficulty breathing (trigger Breath). Dark red = moderate to severe could include slowing, hallucinations (mislead), passing out, confusion, and other nastiness.

Yeah, most players aren't going to get into light red or dark red very often, but they'll definitely notice when it happens. Meanwhile, players aren't having to put up with annoying debuffs every time an Adder or Killer Bee pokes them (unless they let it poke them too much, but then the player deserves that).

It'd also fit well with how poison actually works in reality, so it works great for flavor. I know game =/= reality, but the flavor works here. But anyway, if you get poisoned, you're not going to get all the nasty side effects right away. It takes time or getting poisoned some more. Unless you get hit by some rather nasty poison, but that's what Curare, Poison Arrow, and other nasty poison attacks so that's all account for too.

I'd also suggest applying these types of effects to player poison. It'd give it a niche between straight-up damage spells/brands/whathaveyou (instant damage vs. damage over time) and also between Hexes (chance for random effect if the spell works vs. guaranteed effect if the spell works). Poison Magic already has some pseudo-control options (Mephitic), so the in-game precedent is there too.


Combine with deterministic poison and being able to rest for about 20 turns to clear it and I think that'd put poison in a pretty good spot.

[Mumcon]

So I suggest that poison duration is shorter but the average damage remains the same. Would be good to simplify it too, the current code is quite messy. Which is why this proposal doesn't contain actual formulas and numbers, I don't have time to source dive and make a complete proposal. Any help with that appreciated.

Regarding source diving for a complete proposal: based on your wording I can't quite tell if you're aware of this, but monster poison status is different from player poison status. The damage formula and chance of being damaged each turn is separate for players and monsters, but they're both quite clearly laid out on the wiki. For players, the status duration is based on a simple 1/8 chance of losing a poison level each turn, but for monsters, it's apparently much more complicated.

Months ago, I considered posting a thread suggesting changes to the poison formula too. However, I had wanted to make a relatively "complete" proposal with poison's effectiveness being based not on resistance/weakness, but rather on MR, in such a way that it would be scaled down drastically but still be relatively strong against powerful monsters in extended. I have always thought that poison could use some kind of reform, because I believe the "problems" (I realize that those who accept them just call them features) associated with poison--extremely good early on, really bad against things that resist it, no room for a good high-level poison spell--stem from design choices that are forced upon developers by the properties of the core poison mechanic. I spent hours looking at the code, and edited the wiki to have better info on monster poison damage (the player poison damage formula is simpler, and was already correct), but I couldn't figure out monster poison status duration. And judging the overall "effectiveness" of poison requires being able to estimate duration, which is complicated because multi-level poison is constantly dropping to a lower level after a somewhat randomized sub-duration based on (I think??) HD. So I never posted the thread.

Anyway, if the major concern right now is player formicids, it should be not too hard to write a script that simulates each turn before the poison wears off and gives you play-by-plays/average damage/average duration/etc based on this specific formula/pseudocode, without needing to dive into the code. (Much harder to do for monster poison status.) I'm late coming to this thread, so if this has already been discussed thoroughly in ##crawl-dev I expect there was no point in me mentioning it.

My ideal is to have poison
- work the same for monsters and players
- have long duration (doing damage over time is its real fundamental feature)
- start the damage rate (damage per turn average) very low for low power, and scale up the damage rate slowly for higher power
- have the damage amount be more regular (closer to the current player formula than the monster one--monster vulnerable damage ranges from 2 to poison_level*3+1, e.g. 2 damage one turn and 16 damage the next)
- have narrower duration ranges, rather than a flat 1/8 chance of losing a single level like with the player or a confusing rapid decay of high-level poison like with monsters
- perhaps have effects, if they are not too severe and mainly serve to reduce the poisoned party's effectiveness without directly leading to that party's death
I haven't particularly thought this through--this is just a rough statement of what I think the conceptual principles behind poison "should be" if poison is to remain in Crawl as a thematically consistent and strategically differentiated mechanic.

[galehar]

poison's effectiveness being based not on resistance/weakness, but rather on MR

Not all sources of poison are magical, so that doesn't work.

I have always thought that poison could use some kind of reform, because I believe the "problems" (I realize that those who accept them just call them features) associated with poison--extremely good early on, really bad against things that resist it, no room for a good high-level poison spell--stem from design choices that are forced upon developers by the properties of the core poison mechanic.

Well, personally I don't see any of that as a problem and this isn't the goal of the thread to address that. There have been countless discussions about the poison magic school and high level poison spells, I'd rather avoid derailing the thread with yet another one.

Anyway, if the major concern right now is player formicids

Not really. I mentioned them because they make the existing issues with poison more apparent, but I'd like to address them globally.

My ideal is to have poison
- work the same for monsters and players

This isn't a design goal. And since monsters and player have very different HP/AC/EV stats, damage formula are often different too. This issue at hand is player effect. What's wrong with monster effect? Is there really a gameplay issue with how monsters take poison damage?

- have long duration (doing damage over time is its real fundamental feature)

I have the opposite goal. More duration means more resting (for player) or more kiting (for monsters). What's the point?

[galehar]

I've tried to put together a spreadsheet of poison damage depending on poison level.

Note that currently, the damage is done per player turn, duration of turns has no effect. If it's changed to a normal duration, you'll take more damage for longer actions (total damage would still be the same). I'm not 100% percent sure of the accuracy of the spreadsheet, but it's probably close enough. If you notice any mistake, let me know.

Now, to find a good formula for poison duration and damage.

[Sandman25]

I've tried to put together a spreadsheet of poison damage depending on poison level.

What's "Total Avg damage"?

[BrewFall]

What if it completely stopped regen but dealt no damage.

[dpeg]

BrewFall: that is Sickness. Which (a) already exists, for example conducted by komodo dragon bites and is (b) too often not tactically relevant, in my opinion.

[spudwalt]

Twilight Phoenix's idea sounds interesting.

[galehar]

I've tried to put together a spreadsheet of poison damage depending on poison level.

What's "Total Avg damage"?

It's the average amount of total damage. I've renamed it to "Avg total damage", hope it's more clear. I've also fixed (hopefully) the formula which was completely wrong

[Sandman25]

I've tried to put together a spreadsheet of poison damage depending on poison level.

What's "Total Avg damage"?

It's the average amount of total damage. I've renamed it to "Avg total damage", hope it's more clear. I've also fixed (hopefully) the formula which was completely wrong

Thank you, now it makes sense. Previously it had 0.2 Total Avg damage for 0.2 per turn damage during 8 turns.

[johlstei]

I've tried to put together a spreadsheet of poison damage depending on poison level.

Note that currently, the damage is done per player turn, duration of turns has no effect. If it's changed to a normal duration, you'll take more damage for longer actions (total damage would still be the same). I'm not 100% percent sure of the accuracy of the spreadsheet, but it's probably close enough. If you notice any mistake, let me know.

Now, to find a good formula for poison duration and damage.

Does this mean that force attacking with a slow weapon (like darkmaul) will potentially allow you to heal more times in between ticks of poison damage? I suppose healing might also work this way also making this not true, but yeah, that's pretty tedious/spoilery.

[galehar]

Indeed yes. Regeneration depends on absolute time (aut) whereas poison depends on player turns. Which means you get a better healing/poison ratio by doing slow actions. Swinging a slow weapon to not die of poison might actually work, but it's quite marginal. A more realistic scenario is with a Chei worshiper. He'll lose less HP when poisoned by moving around than standing still.

[crate]

If you swing a weapon at empty space it takes 1.0 turns regardless of what weapon you swing. You can however use dark maul against plants or something.