Remove Shafts

[Laraso]

I believe that shafts really shouldn't exist in a game like this. Unlike banishment, it happens completely randomly and is entirely outside of the players control, and unlike teleport traps it puts you in completely unexplored + unfamiliar + unsafe territory with no sense of where you should retreat to. There is nothing a player can do to avoid falling in one, other than being able to fly or just getting lucky and not triggering one. They are one of the very few sources of unavoidable death, and definitely one of the most consistent. They are nothing but a source of frustration, and offer nothing interesting to the game, unless you find being shafted three floors and winding up adjacent to Nessos and Frances with no real means of escape and nowhere to run but unexplored hostile territory to be interesting.

At the very least, let them only drop you a single floor at most. Three is really pushing it.

[johlstei]

I think shafts are fun and escaping after getting shafted from say, D:5 to D:8 is really satisfying.

[duvessa]

D:1 shafts are so rare that I doubt shafts have ever caused an unavoidable death. I've certainly never seen it or heard of it happening.

If you just said "remove traps" then I'd agree, but instead you've made it sound like you not only don't understand why traps are bad design, but don't even care and just want an easier game. It's like supporting gay marriage legalization because you think it would give you more lesbian porn to watch.

[Laraso]

Would you care to elaborate on what I'm missing here? Because you don't seem to think traps are good design, so what makes them "bad design" other than what I already said? My argument against it is "it happens completely randomly and is entirely outside of the players control", which in a game like this sounds to me like bad design, but apparently that's not a good enough reason?

[duvessa]

Traps are tactically irrelevant and their existence makes it useful to track which spaces have been stepped on, and lure monsters over non-stepped-on spaces before you do so yourself (this is most visible in zot:5).

[milski]

D:1 shafts are so rare that I doubt shafts have ever caused an unavoidable death. I've certainly never seen it or heard of it happening.

If you just said "remove traps" then I'd agree, but instead you've made it sound like you not only don't understand why traps are bad design, but don't even care and just want an easier game. It's like supporting gay marriage legalization because you think it would give you more lesbian porn to watch.

Incredibly shitty metaphor and tired argument about how things besides jackals are dangerous aside, I've certainly gotten shafted three floors on D:4, and then shafted another three floors ro D:10 next to a centaur. While I probably could have avoided that death in some fashion, it was certainly so far beyond anything in Crawl that I'm pretty comfortable saying it wasn't reasonably avoidable.

[milski]

Would you care to elaborate on what I'm missing here? Because you don't seem to think traps are good design, so what makes them "bad design" other than what I already said? My argument against it is "it happens completely randomly and is entirely outside of the players control", which in a game like this sounds to me like bad design, but apparently that's not a good enough reason?

It's helpful to note that Duvessa believes the entirety of Crawl is trivial aside from possibly Jackal packs, so any arguments will be framed around robotic optimal play rather than e.g. difficulty or level of annoyance/convenience for suboptimal play.

[chequers]

Getting shafted is exciting! Double shafts are less so but they're relatively rare.

[tedric]

When they work well, shafts force the player into a situation where survival depends on slowing down, thinking carefully, and using absolutely every option at their disposal. The reactions to this situation provide an illuminating distillation of (what I see as) the fundamental philosophical divide among the Crawl playerbase.

For some players -- the ones who advocate a shorter game, decry the glut of XP-laden meatbags, or perpetuate the "no true Crawler should ever die past D:1" meme* -- getting shafted is everything they love best about Crawl, and more-or-less the way they think the game should play all the time. (When you think about it, getting shafted is really not that different from speedrunning...)

For others -- to whom Crawl is a supposedly "casual" game that should forgive small mistakes and make room for idiosyncratic, whimsical, and "sub-optimal" playstyles -- getting shafted is the RNG forcing them to play in a way that they'd really rather not, thank you very much.

Whether shafts are good, bad, or just holes in the ground is largely dependent on which side you identify with more. And they both have their merits! But it's pretty hard to have a coherent conversation when the people who care the most are talking right past each other from the beginning.

* Note the phrasing: It's not that nothing past D:1 is dangerous; it's that any Crawler worth her starting kit should be able to deal with the danger.

[Laraso]

Personally, I'm a strong believer that any and every game should be built around optimal play. Good game design is when you focus around the highest level of play, not the lowest common denominator. I don't particularly like it when games go out of their way to make sure that the player can't screw up or make mistakes, and I greatly respect games that manage to create a high skill-ceiling for its players while also keeping the game interesting and fun. Do I consider myself a good player? Not in the slightest, I can't be bothered to play optimally and I've splatted so many winning characters it's not even funny. That's not to say I want Crawl to be a casual experience, in fact it's quite the opposite, I would absolutely hate it if this game was dumbed down and casualized.

That being said, I think shafts are bad difficulty. The idea that you could randomly, quite spontaneously, be thrown into a very bad situation that you had no control over avoiding is very unappealing to me. It honestly feels like a Xom effect, and if I wanted to play a game with Xom effects (I sometimes do), I would just pick Xom. I greatly dislike teleport traps for that same reason, but I tolerate them because they aren't too bad, you at least have some of the map explored and have a general idea of where to run to, and the threat doesn't last very long. But getting randomly shafted down three entire floors, most often into an extremely undesirable situation, and trying to make your way back up is one of the least fun and least interesting challenges in this game, in my opinion. Good challenges are ones that you as a player have to put yourself into and overcome, such as Vaults:5, Elf:3, certain wizlabs, etc. Vaults:5 especially is probably my single favorite area of the game, it's really intense and very dangerous but the feeling of overcoming it is great and very satisfying. You go into it knowing that it's going to be dangerous and you have at least some idea of what to expect, even if things never play out the same way every time. In contrast, getting thrown down three floors lower than you're supposed to be is a completely unexpected and unintentional experience, and if you had any degree of control over it you wouldn't be dead center in the middle of Saint Roka and his band of orcs right now. Getting out of that kind of situation is way more stressful than satisfying, and getting back to your original position before you got shafted doesn't feel rewarding in the slightest.

Maybe I'm the only one who has this opinion about shafts, but please don't think that I'm just wanting the game to be easier, because I don't.

[tabstorm]

Would you care to elaborate on what I'm missing here? Because you don't seem to think traps are good design, so what makes them "bad design" other than what I already said? My argument against it is "it happens completely randomly and is entirely outside of the players control", which in a game like this sounds to me like bad design, but apparently that's not a good enough reason?

I encourage you to try playing Brogue. And seriously try to win it and play "optimally", don't just die on D:5 repeatedly and say "Wow, this game is so difficult and pretty and well-designed and...". If that dosen't make you think traps are bad I don't know what to tell you. Suggestion: Make sure to hit 's' to search for traps.

Searching for discoverable floor traps isn't overly enjoyable. Trying to track safe squares that have been walked on by enemies already is a pain. I don't think the "don't let enemies walk into zot traps!!!" is a good enough justification for them to be in Crawl. Seriously, just put everyone under the equivalent of weak Xom wrath all game if you want lolrandom things to happen to players.

Personally, I'm a strong believer that any and every game should be built around optimal play.

This is basically futile for Crawl. If you had to play close to optimally in Crawl hardly anyone would play it because it would be too painful.

[swoonis]

I think shafts are fun and escaping after getting shafted from say, D:5 to D:8 is really satisfying.

In a nutshell I agree with this. It's fun to overcome unforeseen obstacles.

[Laraso]

I encourage you to try playing Brogue. And seriously try to win it and play "optimally", don't just die on D:5 repeatedly and say "Wow, this game is so difficult and pretty and well-designed and...". If that dosen't make you think traps are bad I don't know what to tell you. Suggestion: Make sure to hit 's' to search for traps.

I'm not sure what point you're trying to make here, I think traps are terrible design (albeit for different reasons than duvessa, although I do agree that they are also bad design for the reasons duvessa suggested as well) and I thought I was implying that in my posts.

Personally, I'm a strong believer that any and every game should be built around optimal play.

This is basically futile for Crawl. If you had to play close to optimally in Crawl hardly anyone would play it because it would be too painful.

Yes, and that's why I don't play optimally. But I'm still able to play and enjoy Crawl despite that. Optimal play doesn't always have to mean "boring" or "tedious", even if that might often be the case in this game or other games. And removing boring and tedious "optimal play" is always a good thing. I'm not sure if you thought I was implying that Crawl is the pinnacle of game design, because I wasn't trying to, there are many things that I would change about this game if I was given the chance, shafts being one of them. Maybe "optimal play" wasn't the best choice of words, should I have said "high-level play" instead? My point is, balancing the game around playing the best as possible is better than balancing the game around casually screwing around and not having to care too much about what decisions you make, as long as you're not sacrificing fun.

[tabstorm]

I was trying to make the point that traps aren't bad because of unavoidable deaths, but because they don't necessarily induce unavoidable deaths (but instead induce unfun gameplay if you want to avoid the bad effects of traps)

[Sprucery]

I still think that as long as traps exist, we should bring back Traps skill so that the player could influence how much traps affect them.

In the meantime, how about if shafts would try to drop the character in a spot where no monsters are in LOS? (Wouldn't apply to Fo self-shafting.)

[RBrandon]

How about just remove traps since there are already a set of challenges that actually matter known as monsters.

[dpeg]

[Shafting] happens completely randomly and is entirely outside of the players control, and unlike teleport traps it puts you in completely unexplored + unfamiliar + unsafe territory with no sense of where you should retreat to. There is nothing a player can do to avoid falling in one, other than being able to fly or just getting lucky and not triggering one.

All of these are exactly as intended: ideally, a shaft forces you to adapt and increases chances of death. This means that game depth goes up, because of more chances to fail.

In my opinion, shafts are the best traps Crawl has, and they should certainly stay.

Exercise for the grumbling reader: devise a system that keeps shafts and does away with tracking squares. (It can obviously be done, just was never considered necessary to implement.)

Final word: these traps are one of the few things that Crawl has borrowed from Nethack. Gotta give credit where credit is due!

[Sprucery]

Exercise for the grumbling reader: devise a system that keeps shafts and does away with tracking squares. (It can obviously be done, just was never considered necessary to implement.)

The basic requirement would then be that stepping on a square does not affect the probability of a shaft triggering. So something like this, for example: in a given dungeon level there is a 1/n chance per turn of a shaft triggering on the spot where the PC (or monster?) is.

Then n can be varied to adjust shafting density.

[dpeg]

Sprucery: 20/20

Optionally, just make this shaft chance only apply to the player. Can be rationalised by saying that the local monsters know where to step etc.

[adozu]

make Shafts have a % chanche to trigger

X = chanche to trigger
Y = chanche of any given tile to be a shaft (if eligible)

chanche of the next never stepped on tile to shaft you = XY

chanche that a tile that didn't shaft you was, in fact, not a shaft = ((1-X)Y)/(1-XY)

given that if X(((1-X)Y)/(1-XY)) < XY it is still optimal to step on previously explored tiles
and that if X(((1-X)Y)/(1-XY)) > XY it is optimal to always step on new tiles!

therefore every time a shaft fails to trigger its activation rate is adjusted into Z such that:

Z((1-X)Y)/(1-XY) = XY which we can arrange as Z = (XY(1-XY))/((1-X)Y)

not sure if it is feasible computationally wise, but the concept is sound


ps: this is superior to simply rolling for a chanche of any step you take shafting you because eventually that would make every single tile into a shaft.

pps: general solutions to the problem also include X = 50%, X = 0 (makes sense but is an undesired solution), so i guess also making shafts have a 50% chanche to trigger solves it (but the general formula allows to adjust that % as desired). incidentally this applies to every kind of traps unless i made some gross mistakes (which i might have)

[johlstei]

Sure, shafts are a way to unavoidably have something bad happen to you - you can't use skill to evade them. (well, unless you track stepped-on tiles, as duvessa said). However, there's plenty of other things like that. You could roll snake instead of spider. You could spawn a particularly scary vault on D:2. There could be 3 oofs in sight of the orb. That alone doesn't seem like a good reason to remove it. I think it creates profoundly interesting situations that are entirely unlike the rest of crawl outside D:1. Changing the tradeoffs between running through unseen territory and staying and fighting is certainly welcome to me. Additionally, this is silly, but I feel much more mentally attached to a character that has survived a bad shaft than one who plods along. Suddenly this guy and I have seen some shit together and I want to make it through with him. It's a good feeling.

[partial]

D:1 shafts are so rare that I doubt shafts have ever caused an unavoidable death. I've certainly never seen it or heard of it happening.

If you just said "remove traps" then I'd agree, but instead you've made it sound like you not only don't understand why traps are bad design, but don't even care and just want an easier game. It's like supporting gay marriage legalization because you think it would give you more lesbian porn to watch.

i love u

[Sprucery]

ps: this is superior to simply rolling for a chanche of any step you take shafting you because eventually that would make every single tile into a shaft.

Shafts should be temporary in this case.

[adozu]

it still does mean that you could fall into a shaft on every single tile in the level if you walk back and forth long enough (hyperbolically). you can even say that you can find an empty 2x2 room with a closed door and just walk in circle until you happen to roll for being shafted if you wanted to skip a certain level.

i'm sure duvessa could find any reason why it would be optimal to just walk circles to scum for shafts, i'm not creative enough.

you'd then have to put in any sort of exotic restrictions like limit the max number of shafts per level, or "time-out" tiles so they can't turn into a shaft until X turns have passed since you last tried them or whatever, in any case it would potentially be scummable.

shafts being like they are now but simply without the activation being certain (when still undetected) is a lot less abusable (especially if the numbers are tweaked correctly)

[reaver]

Guys the obvious solution to keeping the Shafting effect around if current traps are bad is to turn it into a monster spell (For example replacing Confuse in Orc Wizard's list)

[adozu]

but the premise isn't that traps are bad, its that they induce bad gameplay by making previously-stepped-on tiles superior to never-before-stepped-on tiles.

if every unseen trap had a 50% chanche to not trigger/reveal when stepped on it would make that issue mostly solved.

[Sar]

wouldn't that require stepping on each square 4-5 times

[byrel]

In essence, fixing random traps effectively makes them Xom-lite in a sense; you (very rarely) have a negative effect happen to you which mechanically is not bound to any floor tile.

Flavor-wise this can still be a trap/shaft/whatever of course.

Traps can still actually be usefully applied in vaults I think.

[adozu]

@sar: only if you are actively looking to trigger a trap on you, which you wouldn't be normally (we could argue if such a trap existed it would be failing at what traps are supposed to be). i can see the point being made for some other trap types (so you don't trigger them at bad times) but it really shouldn't apply to shafts.

i mean sure, you could step on each tile 100 times just to be sure it was not a shaft, but what is your gain if it turns out it was?

unless you are planning to skip levels by shafting yourself but it would still take longer than just looking for stairs (the difference to the "xom-lite" version is that you could just walk back and forth the same 2 tiles indefinitely and eventually a shaft would happen, in my model if there is no shaft there is no shaft)

[Pollen_Golem]

Statues that cast Shaft Other on sight, and self-disintegrate, would be as unavoidable as current shafting, without the tile tracking.

and that if X(((1-X)Y)/(1-XY)) > XY it is optimal to always step on new tiles!

I haven't checked the other maths but this simplifies to Y>1, which is impossible.... it's always optimal to re-tread tiles even if there is a chance of the trap failing. If you go further, and come up with a system where the chance of triggering falls every time you step on a trap and it fails to trigger, you can, if you wish, degenerate to having a random chance of getting shafted every time you move.

[adozu]

you are right, there isnt a solution for that to reasonably happen whitout an adjustment, also i suck at math but that was just to show an edge case.

i agree that you can degenerate it to the point where it is the same as randomly rolling for being shafted on any tile (that's exactly the idea: so there are no more favored tiles) the difference is that which tile are traps or aren't is predetermined beforehand which prevents some scum opportunities.

[Pollen_Golem]

you are right, there isnt a solution for that to reasonably happen whitout an adjustment, also i suck at math but that was just to show an edge case.

i agree that you can degenerate it to the point where it is the same as randomly rolling for being shafted on any tile (that's exactly the idea: so there are no more favored tiles) the difference is that which tile are traps or aren't is predetermined beforehand which prevents some scum opportunities.

you didn't mention a mechanism where the chance for a trap to activate got lower every time it failed to activate, so in your system, there would always be these favored tiles.

If two things are the same, there is no difference between them.

[adozu]

i did actually, i called it Z and it repleaces X as chanche to activate after each failed activation. Z actually goes up, not down, after each unsuccesful activation.

and yes, there is difference, if every new step can shaft you (xom-model) walking back and forth two tiles will eventually shaft you. if traps are pre-determined but might just fail to activate that doesn't occurr. it's not the same at all.

[byrel]

Actually I think adozu is right... by changing the chance you get shafted each time you step on a shaft tile, you can make it have the same expected value whether you're walking on new squares or old ones. It would still be optimal to run monsters over every square a bunch of times, but that's easily fixed by making monsters unshaftable.

Note that this doesn't only work for randomly distributed shafts. If you have a shafty vault, you would modify the value of Y to compensate for the higher density of shafts. It would still not be optimal to walk across qazlal-shaft-central or whatever the vault is called.

[Pollen_Golem]

Actually I think adozu is right... by changing the chance you get shafted each time you step on a shaft tile, you can make it have the same expected value whether you're walking on new squares or old ones.

This is my point, not adozu's.

It would still be optimal to run monsters over every square a bunch of times

If monsters are affected by shafts in the same way as the player, it would not matter whether you run monsters over old tiles or new tiles.

[MainiacJoe]

Am I the only one who leads monsters over known shafts to try to get rid of them? I gladly trade the possibility of falling a few floors every once in a while to an undetected trap to get such tactically useful terrain. They're better than t-traps for this purpose (whether you jump in the trap yourself or lead a monster into it) because it is 100% guaranteed that you won't meet that monster on this level again. So the tweaks above that eliminate shafts as traps with a fixed location also eliminate being able to use shafts against monsters, changing a feature with positives and negatives into a feature with only negatives (though I am also one of those who thinks trying to survive after a shaft is fun and satisfying).

[byrel]

Actually I think adozu is right... by changing the chance you get shafted each time you step on a shaft tile, you can make it have the same expected value whether you're walking on new squares or old ones.

This is my point, not adozu's.

No.

given that if X(((1-X)Y)/(1-XY)) < XY it is still optimal to step on previously explored tiles
and that if X(((1-X)Y)/(1-XY)) > XY it is optimal to always step on new tiles! ...

therefore every time a shaft fails to trigger its activation rate is adjusted into Z such that Z((1-X)Y)/(1-XY) = XY

He said it first, so I'll give him credit.

[Pollen_Golem]

He said it first, so I'll give him credit.

I don't know man, he edited that post a few times and I remember not understanding what Z was supposed to be.

ps: this is superior to simply rolling for a chanche of any step you take shafting you because eventually that would make every single tile into a shaft.

It is not superior, it is the same. This system is such that you'll have as much luck shafting yourself by walking back and forth across 2 tiles 1000 times as you will by walking around an unexplored level for 1000 turns. Assuming player chars can't detect traps, of course.

If the increase in failure rate every time you step DOES work correctly, then it does not matter how many times you have stepped on a tile - 0, 1, or 1000.

If the increase in failure rate every time you step DOES NOT work correctly, then, well, you made a mistake because it does not work the way you said it's supposed to work.

The difference between the two is extremely marginal: namely, the Xomlite shaft lets you get shafted from the exact same square twice, and the degrading shaft doesn't.

You want treaded tiles to be as dangerous as untreaded tiles, but you also want treaded tiles to be declarable shaft-less by walking over them a bunch of times, which is a contradiction.

[adozu]

actually, i mean for the chanche to fail to go DOWN, not UP, after each failed shafting, that compensates for explored tiles being less likely to shaft you if the rate stood constant.

also not hunting for credit i'm just intrigued by the problem there's no need to debate that

(besides, my last edit is previous then your post, forum has time-stamps that sounded quite silly )

"It is not superior, it is the same. This system is such that you'll have as much luck shafting yourself by walking back and forth across 2 tiles 1000 times as you will by walking around an unexplored level for 1000 turns. Assuming player chars can't detect traps, of course."

the thing is, if you have to walk all around the level to find a shaft then you might as well look for stairs while you wouldn't have to if any tile could become a shaft if sufficiently stepped over.

[byrel]

Random side-note: it's spelled chance, not chanche

[Quazifuji]

I don't particularly like it when games go out of their way to make sure that the player can't screw up or make mistakes, and I greatly respect games that manage to create a high skill-ceiling for its players while also keeping the game interesting and fun.

In many cases, preventing the player from screwing up and making mistakes has no effect whatsoever on the game's skill ceiling, all it does is increase the skill floor, which many would consider a bad thing. For example, warning you when you are about to anger your god. That has no effect on the skill ceiling, because any player anywhere near the skill ceiling can always just keep track of every effect that could anger their god or looking it up if they're unsure anyway. All it does is increase the number of ways new player could accidentally mess up because they don't know all the rules.

Good challenges are ones that you as a player have to put yourself into and overcome

Why? I'm not being facetious, I honestly want to hear why you believe this. What's wrong with being unexpectedly thrust into a challenging situation and force to figure out how to escape? DCSS is all about adapting. It's one of the most core, fundamental elements of gameplay: that you have to adapt to the challenge each game throws. Why should the challenges be limited to events like going down a staircase or opening a door? I mean, heck, if we take your argument to the extreme, you could say the entire game should consist of nothing but runed door vaults. But it doesn't, because that would suck, because sometimes challenges should find you. The challenges like branch ends where you know roughly what's coming and prepare for it are fun, but the ones that are unexpected that force you to think on your feet are what make up the majority of the game.

[Laraso]

In many cases, preventing the player from screwing up and making mistakes has no effect whatsoever on the game's skill ceiling, all it does is increase the skill floor, which many would consider a bad thing. For example, warning you when you are about to anger your god. That has no effect on the skill ceiling, because any player anywhere near the skill ceiling can always just keep track of every effect that could anger their god or looking it up if they're unsure anyway. All it does is increase the number of ways new player could accidentally mess up because they don't know all the rules.

Well my problem with this is actually really simple, basically, why have bad effects for taking certain actions if you're just going to prevent or dissuade the player from ever taking those actions in the first place? It kind of defeats the point of bad things happening if the bad things will never happen.

That said, I do think there are certain instances where preventing things from happening makes sense if they're not normally possible in the first place, such as drowning by walking into deep water while confused (you can't ordinarily move into deep water with most races).

Good challenges are ones that you as a player have to put yourself into and overcome

Why? I'm not being facetious, I honestly want to hear why you believe this. What's wrong with being unexpectedly thrust into a challenging situation and force to figure out how to escape? DCSS is all about adapting. It's one of the most core, fundamental elements of gameplay: that you have to adapt to the challenge each game throws. Why should the challenges be limited to events like going down a staircase or opening a door? I mean, heck, if we take your argument to the extreme, you could say the entire game should consist of nothing but runed door vaults. But it doesn't, because that would suck, because sometimes challenges should find you. The challenges like branch ends where you know roughly what's coming and prepare for it are fun, but the ones that are unexpected that force you to think on your feet are what make up the majority of the game.

I don't necessarily find unexpected challenges bad in all cases, and I agree that challenge should indeed sometimes find you. Something like getting teleported by a monster, and then ending up in a bad situation as a result is interesting because you can reasonably expect a monster that has the spell to cast it at some point, and there are ways to play around it (MR, Cancellation, Stasis, etc). The snake branch is a good example of this, and naga mages are one of my favorite enemies because of it. You could argue that this challenge is a matter of a simple gear/race check (Do I have stasis? Am I a Formicid?), but you can also argue the same about shafts (Do I have boots of flying? Am I a Tengu/Gargoyle/Draconian?). You could argue that it's an "expected" challenge as it's expected that the monster will probably (not always) cast teleport on you, but then you could argue the same against shafts (shafts simply exist, you can reasonably expect to fall down at least one during a game). --- If you're in a reasonably tough fight, and have spent some resources and lost a fair amount of health, and then something really scary shows up during the middle of it (fighting an Ancient Lich and then an Orb of Fire shows up around the corner behind you, is maybe not the best but it is one example) is also quite interesting, and makes for some exciting moments. Stuff like that.

But as for shafts, I just don't feel like shafts are the way to do it. It doesn't feel like a legitimate challenge, it feels more like the game just decides to randomly flip you off and say "fuck you" for no particular reason. In the monster-teleport and suddenly appearing dangerous enemy examples, there is some element of RNG (does the monster decide to cast teleport this turn or not, does this monster decide to randomly wander in this direction or not), but in both cases there is also some amount of control the player can use to minimize the chances of those situations from happening. Shafts are entirely and completely in the hands of RNG, there is absolutely nothing the player can do about it (unless you want to waste your own time by dragging monsters over every tile of every floor), and the shafting wildly varies from either not being a challenge at all and simply being annoying (getting shafted one floor into an empty room with upstairs in view) or being extremely dangerous (getting shafted three floors down into a couple of dangerous uniques, teleporting away and ending up in a pack of dangerous monsters, etc.). You can go an entire game without ever falling down a shaft, or if you're particularly unlucky you can get shafted 3 or 4 times in a game (sometimes even double-shafted, down 6 floors). Like I said, shafts to me feel very much like a Xom effect. Sometimes I play Xom just for those kinds of things, but most of the time when I'm not playing with Xom I'm not really looking forward to that kind of challenge, at least not in the way that shafts present them. Would you consider the possibility of randomly being beset with hostile berserk trolls whenever you eat a bread ration to be interesting? Would you consider the chance that at any moment your weapon could jump out of your hand, come to life, and attack you to be fun? Would you like those things to be omni-present, able to happen at any time anywhere you are? Because that's how shafts feel to me.

I understand that I'm in the minority here, it seems that most players are in favor of shafts. Maybe the way I'm explaining it doesn't properly make sense out of how I like unexpected challenges while simultaneously disliking shafts in general, and after you guys have started questioning it I'm beginning to realize that it doesn't entirely make 100% sense to myself either, but it's just the way I feel. That being said, can you mention an example of any other "unexpected challenges", or "challenges that find you" in this game that function similarly to shafts? Because the only ones I can think of are Zot traps, teleport traps, and Xom in general. Random, unexpected challenges that go to the same extent as shafts are very rare in this game.

[hannobal]

I also agree shafts are fun and i'd miss them if they were removed.
Getting shafted is usually not that bad if you have scroll of magic mapping (so you can find staircase up fast).

Also, sometimes you can use them to your tactical advantage if you discover them without falling.
You can lure monsters to shafts.

[jason0320]

Please stop urging the devs to remove features, making DCSS a roguelite.

[archaeo]

I'm not 100% sure where the line is between roguelike and roguelite (and figuring it out is a topic for CYC, not this thread), but I get the feeling that Crawl is still really far from getting downgraded. Feature removal is a healthy thing in a game that's been around longer than some players have been alive!

It's extremely off-topic either way, though, in large part because I'm afraid you're right, Laraso; you do seem to be in the minority here. Either traps are basically fine, in which case I think you have to agree that shaft traps are among the better ones, or all traps are terrible for the reasons duvessa really clearly lays out every time this comes up.

I think there's an argument to be made that tabstorm's right, and Crawl would seriously be better with random Xom/Hell effects across the whole game via the trap-triggers-when-it-comes-in-LOS approach, but I tend to lean toward simply removing traps and giving those effects to monsters instead. Laraso, would you have a problem if, say, some kind of enemy Formicid had an AF_SHAFT attack or MR-gated spell? e: note that this satisfies, I think, dpeg's request for a non-trap way to do shafting, as well as basically all the other traps.

e2: I also didn't really notice that I was far from the first person to suggest this in the thread, so all credit goes to the geniuses up above this post.

[Croases]

Speaking of that "would you like it if your weapon suddenly became animate" comment, I'd like to see a mid-game Tukima's trap that animates your weapon as a hostile (and sets the trap as its guard point so you don't go looking all over the level for it later on). It's an interesting decision: Do you try to kill it and get your prized weapon back, possibly using consumables to do so, or do you flee the level and settle for an inferior version until you're strong enough to get it back?

It would encourage carrying a back-up weapon at all times and that's kind of bad, but the cost for not doing so is one teleport scroll and some annoyance so I guess it's not a spoilery "know this or die" thing.

[Laraso]

Laraso, would you have a problem if, say, some kind of enemy Formicid had an AF_SHAFT attack or MR-gated spell? e: note that this satisfies, I think, dpeg's request for a non-trap way to do shafting, as well as basically all the other traps.

I would be 100% OK with that, in fact I would quite enjoy it for the same reason I enjoy getting teleported by naga mages. I think shafting as a concept is interesting, just not the way they are implemented currently as a trap feature that happens at complete random. It would feel like there is actually a cause for the effect, that there is actually a reason for it happening, rather than the game simply deciding to deal you out a bad hand for no particular reason.

[johlstei]

Can you just pretend that shafts are like, monsters that self-destruct in order to shaft you as soon as you step into LOS? Isn't that basically what you described?

[Laraso]

What archaeo described is to implement shafts as either a spell that can be resisted, or an attack modifier. In either case the shafting wouldn't happen instantly at complete random, it would be a monster effect and the monster wouldn't just randomly disintegrate after shafting you. When you return to your original floor you would still have to either deal with the monster and risk getting shafted again, or work to avoid drawing their attention to avoid getting shafted. This would make shafts vastly more interesting, providing interesting decisions and choices to the player. Do I fight and risk getting shafted, or do I back off and stay safe? Is the monster guarding something valuable or important (vaults with this kind of monster would be extremely interesting I think) to make fighting them worth the risk? This would all go a long way to, in my opinion, make it feel like a legitimate and enjoyable game mechanic.

[byrel]

I think there's an argument to be made that tabstorm's right, and Crawl would seriously be better with random Xom/Hell effects across the whole game via the trap-triggers-when-it-comes-in-LOS approach, but I tend to lean toward simply removing traps and giving those effects to monsters instead. Laraso, would you have a problem if, say, some kind of enemy Formicid had an AF_SHAFT attack or MR-gated spell? e: note that this satisfies, I think, dpeg's request for a non-trap way to do shafting, as well as basically all the other traps.

I strongly prefer adozu's solution to sprucery's. Don't get me wrong, they're both an improvement over current behavior, but adozu's more smoothly handles shafts in vaults, and doesn't result in the newbie confusion (and following silly play) that trying to explain that 'shafts are traps, but not really placed, just things that randomly happen to you' will cause. Adozu's solution makes the actual mechanic line up with the expectation the flavor created: that a shaft is placed somewhere and camouflaged such that you aren't guaranteed to trip it. The fact that the probability of triggering varies as you fail to trigger it essentially can be explained the same way as chunks not benefiting from waiting to butcher: yes, the dev team thought of that and made it unabusable. Like most of the mechanics in crawl.