Page 1 of 1

Long term trend discussion: Stealth nerfs

PostPosted: Friday, 14th August 2020, 10:06
by le_nerd
Generate Gauntlet monsters awake
Let players explore around and see all the options without feeling
bad about waking monsters up. (You'll still make monsters notice
you, but there's less downside to that.)


TLDR: a long series of small incremental changes make being stealthy much less rewarding, and thats not fun.

Lets talk stealth. I thinks its an awesomely fun mechanic. No-spell support Stealth is great, very high risk, very high reward. Its defensive and offensive at the same time. Not being noticed by monsters is good defensively being able to oneshot sleeping monsters is awesomely offensive. At the same time, it rewards the things crawl is all about: proper positioning, item+spell use, taking chances. Its also an equalizer to characters that can't or won't use crawls best defense mechanic AC. If you can't use heavy armour, at least you get to be stealthy!

The above change takes away a real consideration for weak chars: it definitely was an option to just T1 pop invisibility and - without checking rewards - hopp into a teleporter.

There have been other long term stealth nerfs that made being stealthy less fun:

- evocable invis reduced to cloak wearing characters. Invis definitely could use the nerf, its stupidly strong, but why limit it to cloak wearing characters? Two of the weakest, most stealthiest species can't wear those. Octopodes+Felids.

- Generating Gauntlet Minotaur awake. Labyrinth Minotaur had a good chance to be asleep, making stealth kill possible. Now, with him guarantueed awake and having Javelins, it presents a very hard cap on the survivability and damage output of any character. And with !agility gone, there is only !invis left as literally only viable defensive potion against him. I think thats bad design, we want the gauntlet to make up a lot of the "can I do it" decision space, not the Minotaur. I think Minotaur also got MR buff - Lab Minotaur was at least mildly hexable, Gauntlet Minotaur is just not.

- Change of Slime vault to having glass interior. Previously you had the LOS blocking winded stone corridor, giving you a chance at not waking Royal Jelly up. Stabbing Royal Jelly was *hard* and on the stealthiest of chars maybe 50% successful, if you did it it was a great achievement. Now it just gets tripple the distance to spot you and wake up.

- Noise monkeys. Those are fine per se, but you can't avoid or silence them easily. Even if you paralyze/enslave/otherwise wand the monkeys, they still get to make Noise before being paralyzed! Thats just not at all what I'd expect from paralyzation, and player chars don't get a turn to read a scroll before paralyze kicking in. Monkeys also come in packs which is a decision that only hurts stealthy chars, since unstealthy chars will get Noise even from a single one. If they had windup time like orb spider at least.

- Bog ossuary for some unfathomable reason starts with awake monsters, which wake up *everything* on the level. It'd be a fine challenge otherwise, but that single fact makes it the hardest ossuary of them all.

- Dart stab removal. Sure, that was necessary to avoid darting+kiting 100% of monsters since it allowed silent monster wakeup. Still, there'd been other solutions to it, for example tying it to very high stealth skill.

- Mandatory alarm traps in timed portals.

- Alarm traps. I don't disagree with those and much has been written about them already. Its just that the Noise part of them disproportionally hurts stealthy chars. AC melee brutes have woken up enough monsters anyway.

I think I forgot some. None of these are backbreaking on their own, and their have been good changes for stealthy chars overall too. (small weapon fight noise+wake ups, non-stealth stabs, blowgun removal, unrandarts, stealth goddess) Its just the general direction of the slope. Many of these which could often be avoided without much down downside. I wish there was a "if it nerfs Stealth inadvertendly, find a solution that doesn't" step in the design process.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 14th August 2020, 20:06
by Majang
I stopped playing stabbers a long time ago, because it is quite clear that this game isn't really designed around cloak-and-dagger fun. It is a great way to start the game, but no matter how much experience you sink into your stealth and stabbing skills, you can't finish the game that way. As such, stabbing is a detour around the skills you need to win, and so I go those paths right away. Probably this is why nobody really cares anymore about nerfing stealth and stabbing considerations.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 15th August 2020, 06:45
by le_nerd
@Majang: Its entirely correct to play that way I think on normal apt characters. I play only octopodes and with +4 apt its still worth it. You can't finish the game but you get enough incidental value that you don't mind the XP sunk.

Eg. stealth is so bad that you need +4 apts to make it good. Luckily the species which need it the most do have good stealth apts.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 15th August 2020, 10:00
by sanka
Warning: I am not an expert on stealth, so do not take my words as definite source of information.

I feel that one benefit of a stealthy character is that enemies sometimes forget to chase you after you take a corner. This is not very useful in general since there are staircases that do the same job much better (and without require you to be stealthy).

So it could be that if you remove the ability to go upstairs stealth suddenly becomes much better. Of course, it still would not allow you to kill everything with stabs, but thats probably fine. Somebody who actually played Hellcrawl a lot maybe could comment on this assumption.

Also, you could reduce the number of very open levels, where stealth is often useless. (I feel that these levels are less frequent than they used to be in very old versions. If this is true, then not all changes were against stealth.) One reason I do not like stealth because it does not help you much in the hardest levels.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 15th August 2020, 20:25
by Majang
le_nerd wrote:I play only octopodes and with +4 apt its still worth it. You can't finish the game but you get enough incidental value that you don't mind the XP sunk.

Eg. stealth is so bad that you need +4 apts to make it good. Luckily the species which need it the most do have good stealth apts.


Stealth is not the only skill that matters here. Hexes share exactly the same fate: great in the beginning, and entirely useless in the end. So you boost your ninja with hexes, only to later find that you need to kill monsters in noisy and less subtle ways if you want to get the orb. And who knows, now that they've dunked the charms school, they may look around for the next candidate to apply the axe to, and that could very well be the hexes school.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 15th August 2020, 20:43
by sanka
Hexes are not useless lategame: they are very useful even in extended. What you probably mean is that you can not prepare to stab all monsters completely with hexes anymore.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Sunday, 16th August 2020, 15:29
by Majang
Fair enough. That's what I mean. Although "prepare to stab all monsters completely" sounds a like a very positive spin on things. The monsters you really want to stab are entirely non-hexable.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Monday, 17th August 2020, 06:43
by bel
I don't have any particular ideas of how stealth "should" work.

However, stealth without hexes (what the OP calls "no-spell support stealth") has always been bad. It's not stealth that's powerful, it's hexes that are powerful. Stealth is ok as a skill, and many characters can benefit from training some, but a "stealth-based" character type (without Hexes) doesn't really exist in Crawl.

I haven't played stealth-based characters often recently, so I can't say whether things have gotten worse in that regard.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Tuesday, 18th August 2020, 01:08
by andrew
If anything, I'd think the addition of Dithmenos made things better in that regard, since that greatly increases how many things one can stab naturally. But, yeah, hex-less stabbing is a rather large newbie trap IMO.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Tuesday, 18th August 2020, 14:32
by damerell
le_nerd wrote:I think Minotaur also got MR buff - Lab Minotaur was at least mildly hexable, Gauntlet Minotaur is just not.


FWIW, Lab minotaur got the buff one (I think) version before Lab removal.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 21st August 2020, 13:28
by petercordia
Hex-less stabbing is possible with throwing support (for darts), though passwall is also extremely valuable to sleep-stabbers.
However, even if hex-less stabbing is fun (I think it is) it is always tempting to switch to a brute-force style, because of perma-death. Dith doesn't support needles, so it's often better going Oka, but then you can quickly kill everything with javelins. If you keep passwall castable, it's tempting to dabble in other magics.
I haven't played a low-magic stabber in a long time now.

In Zot you can still sleep-stab half of the monsters, including OoFs if you're lucky (but you always need to kill some OoFs with brute force).

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 21st August 2020, 18:38
by andrew
I suppose I've never gotten good enough at using darts.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 4th September 2020, 14:16
by Vaamat
The current iteration of Stealth, as I understand it, is an auxiliary skill that lets you save resources by seeking out the easiest fights, chipping away at enemy groups before the "real fight" starts, and killing a certain percentage of monsters "for free." This leaves you with more resources for the tough fights, but your primary combat skills are a little weaker. I'd argue this is a fair trade in the early game and midgame alike. Not sure about the late game where resources are already plentiful, especially for strong characters and very-skilled players.

But it doesn't feel very much like a "stealth playstyle" because stealth doesn't play a direct role in the big, memorable fights, except insofar as you stumble into big fights by failing stealth checks. In fact it makes the big fights smaller and less interesting. Mobs are smaller, your character is weaker in combat, enemies are stupider and don't use their best abilities because of confusion.

Evidently a lot of people WANT there to be such a thing as a "stealth playstyle." I think offering an "Assassin" background gave those people some false hope, and changing the name was good in that regard. Still, if there were a way to offer such a thing, surely it would be worth considering (and would also be a much-appreciated buff to the current Enchanter). Of course that's easier said than done.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 4th September 2020, 21:49
by andrew
I certainly have found it fun to try to play a "pure" stabber, especially with spells. (Though I've never won that way.) So it would be sort of cool if it were a viable playstyle; but maybe not feasible.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Friday, 4th September 2020, 22:47
by Shtopit
I think a non-magical stealth character could be added, but it would need a different version of stealth. For example, stealth as an ability that can be activated and deactivated, which imposes solid advantages and limitations.

For example, it could activate and show vision ranges (or vision cones) for monsters and gradually decrease them. Outside these ranges, you would be always safe. You could have different range shapes for different kinds of perception, for example with canids who use smell. At the same time, stealth mode could activate super stabs.

The flip side could be that the skill requires light armour to work, allowing for heavier ones as it increases, or will be broken by loud/flashy spells (again, higher levels could allow you to dampen those).

It is a more tactical, maybe less exciting system than the simple roulette, but it offers more control. The question is what should happen once an enemy sees you: how permanent should the consequences be? Should everyone become aware? Should the stealth mode be interrupted? Should the monster simply get an "aware" status and, if you manage to run away and come back later, have a wider sight range?

Another problem with this idea is that Crawl has a fairly small sight range for showing monster sight ranges, but maybe it would work anyway.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 5th September 2020, 08:54
by petercordia
Shtopit wrote:The flip side could be that the skill requires light armour to work, allowing for heavier ones as it increases, or will be broken by loud/flashy spells (again, higher levels could allow you to dampen those).

This sounds a lot like passwall-aided sleep stabbing.
Needs light armour, needs more magic skill in heavy armour, doesn't work after you've made noise.

The chief problem with passwall-sleep-stab builds is that there are too many enemies you have to kill. If you passwall into a pack, it is not unlikely that something will wake up before you've killed them all, and then the whole pack wakes up. The benefit of having killed a few creatures 'for free' is outweighted by suddenly being surrounded.
In general awake creatures wake up other creatures and make plotting your stabs harder. Many levels are so full of monsters that this amplifies until the whole level is awake.
Awake monsters also make escape and auto-travel risky/annoying.
Maybe if there were fewer more valuable targets, and if the Big Bads dismissed their minions after being slain, it would be more worth it to go sneaking around assassinating Big Bad Guys.

I would be in favour of making enemies fewer but stronger, but I'm not sure everyone wants to take the game that way.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Sunday, 6th September 2020, 00:39
by andrew
With a "stealth mode", though, it could be set up that a monster's seeing you, without more, doesn't break stealth; so you're still safe from anyone outside of the reduced vision range. (Of course, if you're in the middle of a pack stabbing, there will likely be several monsters within that range; so it's still risky.)

Re: Long term trend discussion: Stealth nerfs

PostPosted: Tuesday, 8th September 2020, 18:55
by le_nerd
Upon further thought of your reception here, I feel like I've missed to make my point in the OP. Its not that stealth is bad, or that stealth should be good. Its that very weak species, namely Octopodes and Felids, are able to make use of stealth to an amazing degree. This offsets their weaknesses to an extent and is also incredibly fun. Reducing these fringe stealth benefits makes Octopodes and Felids weaker and less fun. These characters ride the margin so very much that a single good item from a portal can make or break them. Creeping stealths nerfs are thus a bad and unfun thing.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Wednesday, 9th September 2020, 04:57
by andrew
Well, to the limited extent that felids can use items. But I've never won one so I suppose you know better than I.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 2nd January 2021, 02:05
by snow
Stealth does not need a nerf. Crawl has the best stealth mechanics in any game ever, and nerfing stealth would eliminate one of the 3 "pillars" of gameplay: tank, wizard, and stabber.

Stealth isn't just about stabbing things. It's about enemies not noticing you while awake, and not following you around corners. Stealth is one of those things that seems like it sucks to a newbie but is something you love as an expert.

Also, late game stabbing is done with things like summoning. Want to stab that orb of fire? Summon some popcorn and stab it while it's distracted. It's a really deep, complex, and rewarding to master.

As for playing a "pure" stabber, that's like trying to play pure conjuration. It doesn't work because there's no silver bullet that takes out everything. Crawl is about building a toolbox that can handle every situation, not turning everything into a nail for your hammer.

Re: Long term trend discussion: Stealth nerfs

PostPosted: Saturday, 2nd January 2021, 04:05
by Hellmonk
Playing "pure" conjuration is significantly easier than playing "pure" stabber though. Anyway, stealth in crawl becomes much more useful when the player doesn't have easy access to stairs for 90% of the game.