The Tavern

Code:: 38228 | Snake:3 | Noticed Louise 38228 | Snake:3 | Cast into the Abyss (Louise)

Sure I only have MR++, but that's not fair !

I think things like this should be prevented... Although it's rare for that to happen (at least in my experience), there is absolutely no way for most characters to avoid something like that unless you just sit around and never explore new areas.

It is, alas, a random game, and things like this will happen. It's a design decision:

http://crawl.develz.org/other/manual.html#balance

And being cast from Snake:3 to Abyss is far, far, far from the worst thing that can happen in Crawl. It could have been Nikola

njvack wrote:It is, alas, a random game, and things like this will happen. It's a design decision:

http://crawl.develz.org/other/manual.html#balance

And being cast from Snake:3 to Abyss is far, far, far from the worst thing that can happen in Crawl. It could have been Nikola

I think you might have missed the point here.
The complaint wasn't about the abyssing of a character but the fact that it happened simultaneously with the sighting of the monster doing the abyssing. IE: There was literally no good play to avoid it. That isn't just random but wrong.

The complaint wasn't about the abyssing of a character but the fact that it happened simultaneously with the sighting of the monster doing the abyssing. IE: There was literally no good play to avoid it. That isn't just random but wrong.

Well monsters get the first move roughly 50% of the time you see them (since you move into their los, as opposed to them moving into your los), and there are no restrictions on when monsters can do actions that aren't escape spells, so this is working as intended. I don't see a reason to change this, other than my dislike of having abyss be a thing at all.

Note that I wasn't complaining about this being a bug... My post was rather ironical !

Btw, escaping things in Abyss made me enter 2 portals and reach Abyss:3, almost instantly discover the rune, take another portal to Abyss:4, escape, kill Louise... and die to a Great Naga in Snake:5...

Code:: 7739 | D:8 | Noticed Erolcha 7739 | D:8 | Cast into the Abyss (Erolcha)

Code:: 7986 | Abyss:1 | Noticed a green death 7990 | Abyss:1 | Killed from afar by a green death

Well at least Louise cannot steal your game this early.
This one just made me think, what the fuck could I have done. Last poison arrow doing 51 DMG.

Unpreventable? Well you decide. Nothing in my inventory provided MR or rPois, so it was a true RNG splat.

Not here to request changes but jesuschrist this game is sometimes unfair!

Sphara wrote:
Code:
7739 | D:8 | Noticed Erolcha 7739 | D:8 | Cast into the Abyss (Erolcha)

Code:
7986 | Abyss:1 | Noticed a green death 7990 | Abyss:1 | Killed from afar by a green death

Well at least Louise cannot steal your game this early.
This one just made me think, what the fuck could I have done. Last poison arrow doing 51 DMG.

Unpreventable? Well you decide. Nothing in my inventory provided MR or rPois, so it was a true RNG splat.

Not here to request changes but jesuschrist this game is sometimes unfair!

I think your mistake was entering the dungeon.

I think your mistake was entering the dungeon.

I'm still learning and you should too. For me, the other alternative was playing ADOM.

I'd rather die on DCSS.

I wouldn't mind if they special cased monster AI not to abyss you on the same turn you spot them for the first time, but then, it's not like things like this happen too often.

damiac wrote:I wouldn't mind if they special cased monster AI not to abyss you on the same turn you spot them for the first time, but then, it's not like things like this happen too often.

I'd rather see Abyss:1 less lethal, and perhaps have more exit portals.

Yeah, a while back I suggested making abyss 1 easier, and make more powerful banishments send you deeper into the abyss, as to scale down the danger of early abyss trips. Most people were generally opposed to the idea, and for some reason most players seem to be under the impression that it's trivial for a character of any level to escape the abyss.

Well if you want to bring up problems with banishment (this is the wrong forum so I didn't mention any of this earlier) the problem is really twofold:

1) Banishment has absolutely no interaction with positioning other than the fact it requires line-of-fire. This is not the case with any other hex in the game (edit: actually this isnt quite true since I forgot about Tele other and maybe there are some other mr-resistible player-blinking effects); paralysis from a monster at the edge of los is usually not that harmful, but from a monster adjacent to you is often deadly. Because of this it turns out that monsters paralysing you the turn they move into los is actually usually not a problem, whereas this is definitely not the case with banishment.

2) Abyss itself has lots of things that are either problematic or are at least so wildly different from the rest of crawl that it feels like a different game.

---

I don't really agree with restricting what actions a monster can do the turn it notices you. I would rather fix the problem by making the actual actions not problematic (from a design sense), instead of artificially fixing it by restricting them.

crate wrote:I don't really agree with restricting what actions a monster can do the turn it notices you. I would rather fix the problem by making the actual actions not problematic (from a design sense), instead of artificially fixing it by restricting them.

When I say the following, understand that I don't know how long a banishment action takes in aut or in comparison to other actions.

What if a monster's banishment action took slightly longer than, say, a move action?

And with the addition of varied Abyss depth, I also agree that Abyss:1 could be toned down; if necessary, tone up the lower levels to compensate.

What if banishment just had a range of like 6?

Currently spell ranges on monsters are completely invisible so unless that changes there's a spoiler cost to adding more of them.

Probably all spell ranges (Monster and player) should be listed in ?/S anyway.

Siegurt wrote:Probably all spell ranges (Monster and player) should be listed in ?/S anyway.

This this this. For the longest time I didn't know Iron Shot/LCS couldn't be shot at me from across LoS since I hadn't played a character that had memorized the spells yet.

I would like that more if I could actually tell how far a range of e.g. "4" could go (thanks circlelos!!!)

crate wrote:I don't really agree with restricting what actions a monster can do the turn it notices you. I would rather fix the problem by making the actual actions not problematic (from a design sense), instead of artificially fixing it by restricting them.

Well, it's not like the idea is unprecedented, after all monsters can't do anything the first time you move onto a floor, for pretty much the same reason, to prevent unavoidable screw overs(like being banished by a monster you had no way of knowing about).

However, I agree a nice elegant solution that's consistent would be better. But special cases are better than just getting killed when you move onto a new floor sometimes.

damiac wrote:Well, it's not like the idea is unprecedented, after all monsters can't do anything the first time you move onto a floor, for pretty much the same reason, to prevent unavoidable screw overs(like being banished by a monster you had no way of knowing about).

My understanding is that players getting the first move on a new floor is a coincidence of how level generation works rather than an intentional design choice.

It's a pretty cool coincidence, though.

Could you imagine all the 'YASD: I went downstairs' threads?

I'd be in favor of removing the "player always gets first action on a new floor" thing but yeah it's apparently difficult to actually do so. Yes it would increase unavoidable deaths (though honestly not by a whole lot) but at least that wouldn't be spoilery and unintuitive.

crate wrote:I'd be in favor of removing the "player always gets first action on a new floor" thing but yeah it's apparently difficult to actually do so. Yes it would increase unavoidable deaths (though honestly not by a whole lot) but at least that wouldn't be spoilery and unintuitive.

It's not difficult.
There's a piece of code that makes 2/3rds of a turn pass if you never visited this floor before and 4/3rds of a turn if you did visit this floor before.
If you removed the 2/3rds of a turn case, then enemies would always get the first turn (unless you were hasted when coming down the stairs)

Patashu wrote:
crate wrote:I'd be in favor of removing the "player always gets first action on a new floor" thing but yeah it's apparently difficult to actually do so. Yes it would increase unavoidable deaths (though honestly not by a whole lot) but at least that wouldn't be spoilery and unintuitive.

It's not difficult.
There's a piece of code that makes 2/3rds of a turn pass if you never visited this floor before and 4/3rds of a turn if you did visit this floor before.
If you removed the 2/3rds of a turn case, then enemies would always get the first turn (unless you were hasted when coming down the stairs)

That doesn't really remove it. What you'd want to do is give monsters a random amount of energy the first time you enter a new floor (since in practice they have random amounts of energy when you encounter them the first time, at least if they were not sleeping). All you're doing with this suggestion is replacing it with an identical thing (this would be equally spoilery) except it's not in the player's favor.

crate wrote:
Patashu wrote:
crate wrote:I'd be in favor of removing the "player always gets first action on a new floor" thing but yeah it's apparently difficult to actually do so. Yes it would increase unavoidable deaths (though honestly not by a whole lot) but at least that wouldn't be spoilery and unintuitive.

It's not difficult.
There's a piece of code that makes 2/3rds of a turn pass if you never visited this floor before and 4/3rds of a turn if you did visit this floor before.
If you removed the 2/3rds of a turn case, then enemies would always get the first turn (unless you were hasted when coming down the stairs)

That doesn't really remove it. What you'd want to do is give monsters a random amount of energy the first time you enter a new floor (since in practice they have random amounts of energy when you encounter them the first time, at least if they were not sleeping). All you're doing with this suggestion is replacing it with an identical thing (this would be equally spoilery) except it's not in the player's favor.

Oh yeah, you've brought up the 'random energy when spawned' thing before and I want to see that too.

Yeah, let's add a source of unfair, unpreventable deaths to the game on purpose. Because it's so unintuitive and spoilery that I don't just randomly die when going down stairs, when I have no way at all of knowing what's down there!

Hell, while we're at it, how about sometimes there's deep water or lava at the bottom of the shafts or downward hatches! It's just special case that it doesn't happen now, it's unintuitive and spoilery level design that it doesn't let that happen on level generation.

FR: Downward escape hatch and stair mimics that only ambush you *after* you've gone down them. And produce shatter-level noise, to draw attention to you. And they make the shatter-level noise by casting shatter at you.

IMO it's good to fix bugs, even if doing so makes the game infinitesimally more unfair.

99% Kill rate death trap next to start area. As long as we are aiming for lets see how stupidly lethal a dungeon can be with no good outs.

damiac wrote:Yeah, let's add a source of unfair, unpreventable deaths to the game on purpose.

Like generating monsters? I hope you agree that a crawl with monsters is better than a crawl with no monsters at all. Pretty much every single unavoidable death in crawl is caused by these monsters. (There are also a very small number that are caused by bugs that are not directly related to monsters.)

There's nothing wrong with unpreventable deaths. Crawl's design philosophy explicitly says this. It also says (or should, I haven't read it recently) that one of the goals of crawl is to be accessible without spoilers. Giving monsters a random amount of energy instead of no energy at all when a new floor is generated fits better with both of these design goals than the status quo.

Doing something like "sometimes make hatches instakill you" doesn't actually solve any design problems, so it's not an improvement.

crate wrote:There's nothing wrong with unpreventable deaths. Crawl's design philosophy explicitly says this.

Well, if you actually want to be explicit:

the manual wrote:The notions of balance, or being imbalanced, are extremely vague. Here is our definition: Crawl is designed to be a challenging game, and is also renowned for its randomness. However, this does not mean that wins are an arbitrary matter of luck: the skill of players will have the largest impact. So, yes, there may be situations where you are doomed - no action could have saved your life. But then, from the midgame on, most deaths are not of this type: By this stage, almost all casualties can be traced back to actual mistakes; if not tactical ones, then of a strategical type, like wrong skilling (too broad or too narrow), unwise use of resources (too conservative or too liberal), or wrong decisions about branch/god/gear.

The possibility of unavoidable deaths is a larger topic in computer games. Ideally, a game like this would be really challenging and have both random layout and random course of action, yet still be winnable with perfect play. This goal seems out of reach. Thus, computer games can be soft in the sense that optimal play ensures a win. Apart from puzzles, though, this means that the game is solved from the outset; this is where the lack of a human game-master is obvious. Alternatively, they can be hard in the sense that unavoidable deaths can occur. We feel that the latter choice provides much more fun in the long run.

... which is quite different from saying there's nothing wrong with unavoidable deaths.

Getting the first move on a new level has always struck me as a pretty reasonable feature. I don't even think it's spoliery. From my perspective as a player, it just means "hey look, I don't think I've ever instantly died when entering a new level."

Exactly. If it was going to take a bunch of programming and special cases to make it work like it does currently, I could see the devs saying "Nope, not a big deal".

But you'd have to be insane to say "Well, it works well and is fair right now, but let's change it to be less fair, with additional code, so occasionally players get killed by performing the absolutely necessary action of going down stairs".

Since monster energy is completely invisible to the player, nothing that happens with monster energy can be unintuitive. What would be unintuitive is just dying sometimes when you go down the stairs.

The game would be... not a game without the monsters, so uh... you might as well say "starting crawl.exe is the #1 cause of player deaths, so lets just delete dungeon crawl" Getting killed while going downstairs is not a source of fun for anyone, and frankly, if you think the game should do it, I would prefer that you never participate in GDD again...

Also, here is the most important quote from that design document people love to quote:

Ideally, a game like this would be really challenging and have both random layout and random course of action, yet still be winnable with perfect play.

We know that's just not possible, but it is the ideal, and thus, it completely goes against the design philosophy to purposely add a new source of unavoidable deaths.

Plus, could you imagine your first descent to zot5, then "You go downstairs, the curse toe torments you, the electric golem hits you, the electric golem hits you, you die"

I certainly would never play the game again...

I'm just going to say that I have seen many many players not know that the player gets the first move on a floor and only the very first time you enter said floor.

edit: another solution would be to always give the player a free move after using stairs, which I would be ok with, though I would prefer my original suggestion

damiac wrote:stuff

You complain about crawl having spoilery rules in half your posts, and now someone makes an effort specifically to make crawl less spoilery and weird, and nothing else, and you go berserk? What do you even want? Do you only post here because you want to antagonize people?

But on the offchance you aren't, you do realize that you can't come down a staircase on D:1, right? You have to find a staircase on D:1 first, and then go down it, for this rule to have an effect. The chance of something happening after so many turns actually causing an unavoidable death is so vanishingly small that it has probably never actually happened, and the chance of this specific rule change doing so is further orders of magnitude smaller.

Spoilery rules are bad but unavoidable deaths are even worse.

duvessa wrote:
damiac wrote:stuff

things

Do you only post here because you want to antagonize people?

things

BADUM TA-DADADA TSHINNNNGGG!!!

ROFLMFAO :lol:

duvessa wrote:But on the offchance you aren't, you do realize that you can't come down a staircase on D:1, right? You have to find a staircase on D:1 first, and then go down it, for this rule to have an effect. The chance of something happening after so many turns actually causing an unavoidable death is so vanishingly small that it has probably never actually happened

It's never happened because it's currently not possible. The player cannot be killed the first time going down the stair, because he gets a free move, so the chance of instant death when first going down to a new floor is 0%.

duvessa wrote:and the chance of this specific rule change doing so is further orders of magnitude smaller.

Nope, this change would make the current 0% chance to up to some number > 0%. So, it would be an infinitely larger chance.

I guess my question is, what percentage of player deaths due to going downstairs to an unexplored floor is acceptable to you? Why is 0% not acceptable? What exactly is confusing and spoilery about the current system?

It's actually entirely possible to already take a staircase and die without any real recourse, though this generally would involve something like gnolls with throwing nets. (It takes more than 1 turn to take the staircase, so you can get netted before you can go upward. You could also get paralysed before you can go back up the stairs, though getting an unavoidable death after like d:3 is really hard so...).

And of course there are also situations where you take a staircase, correctly decide that it is dangerous, return to the floor above, and then take a second staircase and whoops now you don't get your free move so you can die immediately.

So the current chance of dying via staircase is absolutely not 0%.

The chance of dying on the turn of going down a staircase onto a new floor is 0%. I never said anything about it being impossible to die on your next action.

Now, maybe you're right in saying you may then be in a situation from which there is no escape, although at least then you've got options to choose from, like using a blink, fog, teleport, heal potion, etc.. Which is much better than dying before you even see the gnolls waiting at the bottom with their nets and paralyze wands.

It's not 'dying via staircase' just because you took a staircase at some point before you died. It's 'dying via staircase' when you push '>', and before you can do another thing, you are dead. That cannot currently happen when first traveling to a new floor. You have proposed that it would be somehow better if sometimes after pushing '>' to go to a new, unexplored floor, you died. I am asking for your rationale behind that proposal.

edit: another solution would be to always give the player a free move after using stairs, which I would be ok with, though I would prefer my original suggestion

since you seem to have ignored this

anyway the rationale is simple, it's less spoilery and more consistent if going to a new floor via a staircase always acts exactly the same way ... it should not matter if you have visited the floor before
this is a design concern that, according to crawl's own design philosophy, trumps the possible "instadeath" concerns

crate wrote:And of course there are also situations where you take a staircase, correctly decide that it is dangerous, return to the floor above, and then take a second staircase and whoops now you don't get your free move so you can die immediately.

It's funny. I know this is spoilery and probably objectively Bad Design, but somehow this behavior feels correct and fair to me.

I know, in my head, that having stairs always work the same way is Better and Cleaner and more Mathematically Pure, but the way things work now seems subjectively right to me. Can't put my finger on why; it's a surprise to me because I generally value consistency. I guess I like consistency except when I don't. Kinda like the old "car door handles always open to the driver's side for consistency's sake" joke.

I suspect this discussion is going to come down, essentially, to people who feel like the current behavior is good and those who feel like it's bad, and there's no actual right answer to it. The philosophy section can be read in support of either side.

Just a hunch.

crate wrote:Kinda like the old "car door handles always open to the driver's side for consistency's sake" joke.

What's the punchline?

crate wrote:
edit: another solution would be to always give the player a free move after using stairs, which I would be ok with, though I would prefer my original suggestion

since you seem to have ignored this

anyway the rationale is simple, it's less spoilery and more consistent if going to a new floor via a staircase always acts exactly the same way ... it should not matter if you have visited the floor before
this is a design concern that, according to crawl's own design philosophy, trumps the possible "instadeath" concerns

The free move down any stairs (or even any new stairs) seems good at first look, but it opens up the possibility of abuse, whereas the free move down only the first set of stairs cannot be abused, since you can't know what's down there. For example, if I go down one staircase, and see a centaur standing by another staircase, I could go back upstairs, and go down another set of stairs so I end up next to the centaur without him getting a hit on me. Currently the centaur would get to hit me as I came down the other staircase.

I see how the first staircase acting differently from the rest is inconsistent, which is undesirable. I haven't seen the section of the design doc where one thing trumps another though.

So NVJack is probably right in saying it comes down to what you feel is worse, a minor inconsistency or a minor chance for unpreventable deaths. I feel the chance of unpreventable deaths is far worse than the inconsistency, especially given that it's the way it already works. At least with the way it works, you always get the chance to try at least one move after taking a staircase into a bad situation. The first time I went downstairs and died, without getting to make a move, I'd be pretty annoyed and feel cheated. I don't think anyone feels cheated that they always start a new floor with the same HP that they left the last floor with.

And yeah, like you say, it's possible you go downstairs into a terrible situation, and there's absolutely nothing you can do to save yourself. But your proposal only makes that even more likely.

stairs are already the most broken thing in crawl, might as well embrace it

damiac wrote:
duvessa wrote:But on the offchance you aren't, you do realize that you can't come down a staircase on D:1, right? You have to find a staircase on D:1 first, and then go down it, for this rule to have an effect. The chance of something happening after so many turns actually causing an unavoidable death is so vanishingly small that it has probably never actually happened

It's never happened because it's currently not possible. The player cannot be killed the first time going down the stair, because he gets a free move, so the chance of instant death when first going down to a new floor is 0%.

duvessa wrote:and the chance of this specific rule change doing so is further orders of magnitude smaller.

Nope, this change would make the current 0% chance to up to some number > 0%. So, it would be an infinitely larger chance.

I said "the chance of something happening after so many turns". I was referring to all unavoidable deaths occurring after so many turns.

damiac wrote:I guess my question is, what percentage of player deaths due to going downstairs to an unexplored floor is acceptable to you?

Any.

damiac wrote:Why is 0% not acceptable?

It is acceptable. But stupid, spoilery mechanics are not acceptable, and if a stupid spoilery mechanic just happens to have the side effect of making the game very slightly easier, that does not make it any less unacceptable.

damiac wrote:What exactly is confusing and spoilery about the current system?

That entering a level for the first time works completely differently from entering the level any other time.

Is an argument about nothing still an argument? There started off the philosophy section of this thread.

Thanks for explaining your reasoning, I think we just fundamentally disagree on what's more important, I prefer fairness over consistency, you prefer consistency over fairness. Perhaps I'm overblowing the amount of unfair deaths it would cause, although I also think you're overblowing the difference between the first (and usually only) descent onto a new floor vs subsequent descents to that same floor.

Now, if you could yell at a downstairs to get all the enemies on the floor below to come up, then it'd be fine with me if you didn't get any free moves. But since you have no way of affecting what's waiting at the bottom of those stairs the first time you go down to that floor, I think it's unreasonable to kill the player for taking the stairs.

The Tavern

Louise is cheating

Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating

Re: Louise is cheating