Tedium and the Illusion of Lethality

[Igxfl]

Ever since getting half-decent at DCSS, I've been noticing it has a particular structural problem that encourages tedious play. It's probably too deep to ever get fixed, but I want to call attention to it anyway.

------------------------------------------------------------------------------------------------------------------------------------------------------

How False Lethality Encourages Tedious Play:

The crux of the issue is that almost every threat is all or nothing -- you die or you don't.
Most monsters ultimately just threaten HP loss, and HP can be fully recovered at an peaceful moment with no significant cost.
Thus, the only bad thing most monsters can do is kill you outright, anything less and you can "5" it all better at any calm moment.

Because of this, given decent play, fights tend to fall into one of three categories:
1. No significant chance of death. These fights just take up some time and give minor XP. Look up "grinding" in the manual's philosophy section.
2. No significant chance of death if consumables are used. These at least force actual decisions/tradeoffs -- either run or use up your stuff.
3. Significant chance of death even with consumables. These are the only truly threatening fights in the game.

Note that type 3 fights only really exist in the very early game.

A good player from Temple on can always avoid these fights, making them effectively type 1 (walk away) or type 2 (burn a ?blinking).
Doing this isn't much of an intellectual challenge either -- I suspect the vast majority of post-Temple deaths from sub-excellent players are just from boredom and lack of mental stamina. This seems like a serious problem. Playing the game isn't about making really good decisions -- it's about making half-decent decisions in an almost inhumanly consistent manner. Worse, since there's no real risk involved (with highly consistent proper play), a lot of that playtime is going to feel unexciting and extremely mechanical. Note that it's often the really good players who say the game is too long -- this is probably why.

-------------------------------------------------------------------------------------------------------------------------------------------------------

How Attrition Creates Player Interest:

Unfortunately, we can't solve this boredom by introducing more type 3 fights mid-to-late game, because encountering near-unavoidable chances of permadeath after a few hours of play would be unfair, even downright cruel. Worse, the length of the game means that even small chances of death throughout would make winning near-impossible, or at least excessively luck-based.

So, design-wise, we must fully admit that the vast majority of fights in a game the length of DCSS *should not* carry significant risk of death.
Currently, DCSS rides the fence on this issue and gets the worst of both worlds. The game has lots of fights that (with proper play) carry no real risk at all, but disproportionately punish brief moments of stupidity. As I've already said, it primarily becomes a game of mental stamina and resistance to boredom.

So, most fights in DCSS need to hit the obsolete features list:
Fights of type 1, which basically only serve to mentally wear down players until they don't notice that the next orc pack has a warlord in it.
Fights of type 3, which are too scary to use (outside of *maybe* Zot:5 and the like), and are really rare already.

Some fights, however, have some design potential. These are the fights that cause non-lethal but long-term damage:
Fights of type 2 are actually interesting, because they don't truly threaten death, but do cause damage of a sort by using up player resources. Unfortunately, they're a band-aid over a gunshot wound at this point.
Monsters that cause gimmicky longer-term effects like draining, rot, statrot, and item destruction would have the same effect as type 2 fights, except that their effects often don't matter, are implemented in an annoying way, or are, as with type 2 fights. a band-aid over a gunshot wound.

Whatever the problems with DCSS's particular implementations of it, the idea behind these non-lethal sorts of damage is sound -- it's something legitimately threatening that doesn't actually kill you. This offers a workaround to the problems inherent in having countless fights each with a tiny chance of killing players unless handled perfectly. Instead, when used well, these attrition-based fights cause minor damage if handled well, moderate damage when handled poorly, and almost never kill, *even if handled poorly*.

I use the word "attrition" very broadly here -- basically anything that hurts you in a nontrivial way without killing you.
Attrition-based fights can be consistently exciting by posing legitimate risk, without constantly threatening death in a way that a long permadeath-based game can't support.

Currently, players get away with poor play for a long while until they suddenly die at random. Good play is about constantly mitigating that .05% risk of death.
In a more attrition-based game, poor play would cause small but consistent losses in a clear and meaningful way, and good play would be about getting through most fights with minimal losses.

Currently, death comes from a random javelin toss when you stand in the wrong tile a turn too long. Even though it's probably your fault, this often seems arbitrary and unfair.
In an attrition-based game, death would come after a series of poor choices gradually weaken you. Losing happens in a more exciting and meaningful way, and the consequences of decisions are clearer.

Currently, taking a risk is almost always a bad decision, regardless of the reward, because "risk" means "risk of death" and you don't need the reward to win. This encourages players to avoid exciting situations and makes optimal decisions more simplistic.
In an attrition-based game, "risk" means "risk of resource loss". Risk/Reward situations create more meaningful player decisions with fewer clear best options. It will no longer be optimal for players avoid all danger always, resulting in more exciting gameplay.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Caveats:

It's important to emphasize that I don't advocate just taking DCSS as it is and piling on lots of corrosion, statrot, and malmutation.

As long as most fights are designed to potentially kill you outright but be entirely harmless provided you handle them optimally, most attrition mechanics will tend towards the gimmicky, annoying feel we all know and love. This is a very deeply embedded problem, and would probably need a major overhaul to remove or some very clever mechanics tweaks to partially mitigate.

Now, if HP/MP healed with XP gain instead of through mashing "5", and the dungeon was rebalanced around that mechanic, we might be talking.
I know it won't happen and that's fine, but the core problem is there, and it may be an important thing to design around in the future.

[TeshiAlair]

Really, really good post.

Background: I've been a bad player since 2007, and every time a new version comes out, I play for a while, then get bored and stop. My recent burst of playing has lasted longer than any other due to 1. Better spell balancing. 2. More interesting gods and races. 3. More interesting monster additions/changes, particularly yaks/ugly things.

As a result, 99% of my suggestions are for crazy things, typically hyper situational or high risk-reward.

Three main comments

Regarding type 3 fights: I think there should be more of these, of a specific type: I find the most satisfying fights in the game to be fights where the question is "Do I have the resources/positioning to win this fight?" Roxanne is one of my favorite examples- Avoidable if you can't afford to fight, a (mostly) interesting fight if you can. Player ghosts are similar for me. I think crawl in general would do well being a more stealth-oriented game in some regards- rather than having the binary of "stealth-stabber" or "fight everything," make it more viable to avoid fights. I'd also encourage more room-like layouts with glass windows, to make there more of a sense of scouting out the dungeon rather than o-Tab-o all the time.

Regarding attrition: More types of status ailment might be interesting, not necessarily tedium-inducing ones but strategic ones. For instance, Drain only affecting your highest skill but to a stronger degree, forcing you to adopt a different strategy, or temporary -1 LoS --Stealth. What about "Fear of <foo>", which causes a turn of fear any time a <foo> comes into LoS, until you've killed a certain number of that genus.

More interesting fight dynamics: Some of the best fights in Crawl for me are mixed-type pack fights. There's been a general increase in these, but I think they have a lot of design space. Orc packs are the classic, but I like gnoll packs with shamans, Pikel, and ESPECIALLY the unique Vaults guards.

What about a Ugly Thing Coordinator, that mutates nearby ugly things to types that you have no resistances to? Or a yak herder that summons back to them wounded yaks?

[TheDefiniteArticle]

it primarily becomes a game of mental stamina

Congratulations, you've figured out what Crawl is really about. Now you can go play a different game, one that you enjoy.

-Monster damage is very swingy (ie, no matter how good your defenses, an ancient lich can still hit for a hundred damage in one turn and there is nothing you can do about it)
-Player healing is extremely limited and very small in virtually all cases (Borg and Ely are both ludicrously powerful for precisely this reason)
-If HP reaches 0 even a single time it is impossible to recover and a new character must be created.

The natural consequence of these is that players are overwhelmingly encouraged to minimize risk in all situations regardless of reward. Now realize that all three of those things are deliberate design decisions, and that in reality you don't like Crawl.

Whatever the problems with DCSS's particular implementations of it, the idea behind these non-lethal sorts of damage is sound -- it's something legitimately threatening that doesn't actually kill you.

The thing is that it's not legitimately threatening. Unless it's something utterly crippling like Blurry Vision, it's just annoyance. When the whole game is built around being solidly winnable even for a very weak character with very limited gear, giving the player setbacks makes them feel bad, but doesn't have a real effect on the outcome of the game.

In an attrition-based game, death would come after a series of poor choices gradually weaken you. Losing happens in a more exciting and meaningful way, and the consequences of decisions are clearer.

Deep Dwarf of not-Makhleb is the most unpleasant possible character to play in Crawl. Death is not exciting or meaningful, and consequences of decisions are vague at best. Simply playing a Deep Dwarf is a good counter-argument to most of your post. The ability to "just 5 it away" is currently the only thing counterbalancing the "arbitrary and unfair" damage spikes.

[ackack]

Note that type 3 fights only really exist in the very early game.

A good player from Temple on can always avoid these fights . . . . Note that it's often the really good players who say the game is too long -- this is probably why.

I think it's important to note that these are comments on what skilled players can achieve. Most of the player base won't view the game this way. So this is a problem largely about retaining players who have already gotten very good at the game, or maybe differentiating between players who are merely very good vs. those who are excellent. That might be a problem worth solving but it's somewhat more narrow in scope.

As I've stated a bunch of times elsewhere, I feel like Crawl has tried to move more toward a model of "every fight is meaningful" recently with the Lair branch rebalancing and I find it quite a bit more tedious. It is fun to occasionally have stretches where you just slaughter, at least for me.

[WalkerBoh]

The danger of a fight in crawl is not fixed, but rather flexible based on your actions. As you correctly hint at, the goal is to make all fights into type 1 or type 2 fights. By making mistakes, even non-threatening type 1 encounters can turn into type 2 or type 3. So your issue is really that the encouraged methods of making fights winnable are all tedious (kiting, mashing 5, luring enemies out one by one, etc).

I don't think you need to (nor is it necessarily possible to) add more type 2 fights, i.e. You can't make all fights meaningful for all characters. Instead, the tactics available to make life-threatening situations not as threatening should be more "fun".

Now, it is easy to say, but in practice this is hard. I think crawl does a good job in general, obviously or I wouldn't have played it this much.

[Igxfl]

it primarily becomes a game of mental stamina

Congratulations, you've figured out what Crawl is really about. Now you can go play a different game, one that you enjoy.

-Monster damage is very swingy (ie, no matter how good your defenses, an ancient lich can still hit for a hundred damage in one turn and there is nothing you can do about it)
-Player healing is extremely limited and very small in virtually all cases (Borg and Ely are both ludicrously powerful for precisely this reason)
-If HP reaches 0 even a single time it is impossible to recover and a new character must be created.

The natural consequence of these is that players are overwhelmingly encouraged to minimize risk in all situations regardless of reward. Now realize that all three of those things are deliberate design decisions, and that in reality you don't like Crawl.

Do you specifically prefer to play a game based on performing a particular limited set of relatively simple risk-minimizing maneuvers over and over?
I don't know for sure about the rest of the fanbase, but you might be in one heck of a minority there. Note that just because the game happens to be designed a certain way doesn't mean the game should be designed that way. Otherwise we'd do best to call off the whole devteam.

As for the three design decisions you mention, probably only permadeath is really intentional.

I strongly suspect that Linely unthinkingly copied the swingy ultra-lethal style of monsters from other roguelikes back in the early 2000's and no one's fixed it yet because it would be a ton of work to do.

Also, note that healing is extremely widely available -- just find a safe little corner and hit "5". That's exactly the problem I'm talking about.
What you mean to say that extremely rapid healing is limited, and that only matters as much as it does because of said swingy ultra-lethal monsters.

----------------------------------------------------------------------------------------------------------------------------------

The thing is that it's not legitimately threatening. Unless it's something utterly crippling like Blurry Vision, it's just annoyance. When the whole game is built around being solidly winnable even for a very weak character with very limited gear, giving the player setbacks makes them feel bad, but doesn't have a real effect on the outcome of the game.

Deep Dwarf of not-Makhleb is the most unpleasant possible character to play in Crawl. Death is not exciting or meaningful, and consequences of decisions are vague at best. Simply playing a Deep Dwarf is a good counter-argument to most of your post. The ability to "just 5 it away" is currently the only thing counterbalancing the "arbitrary and unfair" damage spikes.

I talked about this.

It's important to emphasize that I don't advocate just taking DCSS as it is and piling on lots of corrosion, statrot, and malmutation. As long as most fights are designed to potentially kill you outright but be entirely harmless provided you handle them optimally, most attrition mechanics will tend towards the gimmicky, annoying feel we all know and love.

The occasional attrition-based fight won't matter as long as most monsters are still shooting to kill. You aren't worn down by drain, for example, because you probably won't encounter another draining monster before you've already worked off all the drain you have. Most current attrition is too brief to matter and too rare to accumulate. Temporary weakness also doesn't matte only because most fights carry zero risk anyway. If every monster could wear you down a little bit, but very few could outright kill you, then those minor setbacks would actually build up and become a serious concern.

As for Deep Dwarf of not-Makhleb, I don't think it would be unpleasant at all with a monster set tailored to fit lack of natural regen.
You're right that constant "5"ing is the only thing counterbalancing crazy damage spikes, but that's a boring solution strapped over an obnoxious problem.

[Wahaha]

This is the price that comes with "almost every death is avoidable". It's hard to make the game harder without also increasing the percentage of unavoidable deaths because of its randomly generated nature. I agree with most of the post. However I don't know if it's a bad thing, because this only really applies to good players. Increasing the difficulty of the game might be bad for the players who haven't reached consistently good play yet, and there are probably a lot more of them than good players. Players who aren't good have a much larger margin of fights where they can die, thus the game consistently challenges them appropriately.
Personally I wouldn't mind if the game was harder, like something closer to Sil, but again it's the whole player base that needs to be considered.

[TeshiAlair]

Another point- Changing the fight difficulty is tricky due to the diverse playerbase.

Recent development has been much more about playerside/playerchoice difficulty, namely, gods, races, and going back a bit portal vaults, which is SO INCREDIBLY SMART and one of the main reasons I've been so enamored with crawl lately <3

[damiac]

There's no single player game that's going to retain a challenge for people who have played it for years and can beat it in their sleep, while still being playable and fun for less experienced people. I want crawl to be a game where if I always make the best decision, I"m going to win. I want crawl to be a game that provides me enough information to make those decisions correctly. Crawl currently is that game, maybe with the exception of pre-temple.

I think TDA was right in saying that looking at your list of problems with crawl, the fact is that crawl doesn't seem like the type of game you like. What you seem to be proposing is changing crawl into a different game entirely.

[Zicher]

Another point- Changing the fight difficulty is tricky due to the diverse playerbase.

This. This and this yet again.

I'm sorry if this post brings nothing new to the game, but please, be aware that if the game gets harder and harder to satisfy veteran players, it may as well have serious problems in attracting new ones.

[Igxfl]

The danger of a fight in crawl is not fixed, but rather flexible based on your actions. As you correctly hint at, the goal is to make all fights into type 1 or type 2 fights. By making mistakes, even non-threatening type 1 encounters can turn into type 2 or type 3. So your issue is really that the encouraged methods of making fights winnable are all tedious (kiting, mashing 5, luring enemies out one by one, etc).

I don't think you need to (nor is it necessarily possible to) add more type 2 fights, i.e. You can't make all fights meaningful for all characters. Instead, the tactics available to make life-threatening situations not as threatening should be more "fun".

Now, it is easy to say, but in practice this is hard. I think crawl does a good job in general, obviously or I wouldn't have played it this much.

I actually have a huge problem with type 1 encounters that can suddenly become lethal. It takes very little brainpower to handle these easy fights safely, so mistakes will usually only come from inattention or boredom. The long sequences of easy fights, then, serve two purposes. First, they offer some cheap XP at no real risk (again, go reread "Grinding" in the philosophy section). Second,they wear down players mentally until they space out for a brief moment and die despite knowing better. This is about as fair as Xom dropping an anvil on you, and much less fun.

I guess this does tie into the tediousness of kiting/luring/etc, though, since that's the exact sort of repetitive behavior that players can get disproportionately punished for slipping up on. I don't think all our positioning mechanics are especially awful in themselves, it's just that since they're so excessively useful in every fight and have no cost at all, it's optimal to do them in every fight. Having to do the same little dance for every orc and jackal for fear of the 0.01% chance of bizarre instadeath is going to get really stale really fast regardless of what the actual dance in question is. I think the reason a lot of players like Cheibriados so much despite him being suboptimal is that they can give up a lot of the positioning dance and get rewarded for doing so.

I think it's important to note that these are comments on what skilled players can achieve. Most of the player base won't view the game this way. So this is a problem largely about retaining players who have already gotten very good at the game, or maybe differentiating between players who are merely very good vs. those who are excellent. That might be a problem worth solving but it's somewhat more narrow in scope.

As I've stated a bunch of times elsewhere, I feel like Crawl has tried to move more toward a model of "every fight is meaningful" recently with the Lair branch rebalancing and I find it quite a bit more tedious. It is fun to occasionally have stretches where you just slaughter, at least for me.

My problem with the skill curve of Crawl is that you get from newbie to decent by figuring out good gameplay, and then get from decent to excellent by taking essentially the same set of tactics a decent player would use, but forcing yourself through the tedium of applying them perfectly by rote on every hallway of every floor of every branch for several hours straight.

Consider:
Not many players get all the way to Vaults and then die because they don't know about luring monsters around corners.
Many, many players get all the way to Vaults and then die because they get so sick of yaktaur packs that they just hold down Tab.


I get where you're coming from with the Lair branches, but I think the rebalancing hasn't made fights more meaningful so much as made the tedious, repetitive dances we have to do to risklessly kill monsters slightly more complicated.

[Arrhythmia]

There's no single player game that's going to retain a challenge for people who have played it for years and can beat it in their sleep, while still being playable and fun for less experienced people.

Counterpoint: Tetris.

[duvessa]

That OP started off reasonably but went pretty far off the deep end when it started talking about "attrition", I think. Making consumables into non-regenerating HP is not really a graceful solution to lair+ Crawl being trivial compared to earlier Crawl. In fact I would say that getting rid of consumables is actually a better option.

[ackack]

I get where you're coming from with the Lair branches, but I think the rebalancing hasn't made fights more meaningful so much as made the tedious, repetitive dances we have to do to risklessly kill monsters slightly more complicated.

It's certainly upped my attrition style usage of consumables.

I think your picture of how skills develop is okay, but I'd say the real difference between decent players and good ones is generally more of an intuitive feel for when situations are actually dangerous enough to merit attention. But yes, Crawl's difficulty is weird. If you've the right sort of personality for it it's not very hard at all, but that personality appears to be sufficiently rare enough that most people find the game difficult.

[zardo]

I think I'm going to try to kind of agree with both Igxfl AND TDA here. I think attrition-oriented characters like DDHe or even more susceptible DD with no god healing at all can actually be pretty fun. But that' s partly because they represent a change from the status quo. It's also apparent to me that Crawl is not really meant to play like that in general. Really explicit attrition aspects like limited HP/MP regeneration give a very different feel to the game in that you know where you stand on a scale from winning to losing and you can only move slowly one direction or the other. This isn't necessarily more interesting - it often becomes at least as much about eking out small advantages in a different way. The attrition aspect is a *part* of Crawl as a result of consumables being... consumable but it plays out as an increase in the ongoing minor risk of death and a corresponding raising of the bar for the level of careful play that is required to succeed. It's apparent to me that while some roguelikes could be/are primarily about long-term resource management, DCSS is primarily about on in-the moment risk management, and both are fine with me but Crawl is certainly fine with me, or I wouldn't be playing it. I think if there's an actual problem it's with the details of how that risk management works and if there's a way to make it more interesting it's to make the risk evaluation deeper somehow, so that it is more often that taking an apparent chance of death in one situation actually spares you other risks, if that makes any sense. So that running at the first sign of danger is less obviously the right choice, but it's still about figuring out what the right choice *is* because that's exciting.

[crate]

If every monster could wear you down a little bit, but very few could outright kill you, then those minor setbacks would actually build up and become a serious concern.

This is true, but this doesn't sound like a very fun game to play, at least not with crawl's current mechanics. The only fun sort of attrition to me is using consumables ... there are many problems with attempting to make all fights use consumables:
1) you don't get very many compared to the number of fights in crawl (a typical win kills on the order of 3k to 4k monsters; if we are really generous and assume that an "encounter" is 5 monsters, then we're still talking six hundred consumables used per game.
2) characters vary wildly in power level; if the goal is to make the game winnable with bad characters (which it seems to be the case) then good ones will not have to use consumables in the same encounters! This is also a problem because then strong characters have more consumables for later and they're already stronger to start with, so it's kind of slippery-slope difficulty (this is already the case in crawl, yes)
3) how are you going to generate the correct consumables? which ones are helpful to specific characters varies a lot: obviously my berserker has no use for potions of brilliance, while my fire elementalist really loves them but doesn't care about !might so much...

The other attrition mechanics in crawl (drain, mutations, statloss, item destruction ... but this is tied in with the above) are just plain not things that are enjoyable for me if they happen often and build up.

[zardo]

Also, Crawl already has a scoring system that rewards more aggressive play. So if you think the playstyle required to maximize win rate/streak is dull, maybe you should be trying to set high scores/minimize turncount/even realtime speedrun instead

[Igxfl]

@crate

It doesn't work at all with Crawl's current mechanics, true. Attrition mechanics may not be the best solution.

I just think parts of Crawl's mechanics are deeply and seriously borked. I want to call attention to that and to the fact that there are some potential alternatives.

Currently, a significant portion of deaths by players who aren't very good or very bad are from inattention or boredom. That's a serious red flag.

[Igxfl]

Also, Crawl already has a scoring system that rewards more aggressive play. So if you think the playstyle required to maximize win rate/streak is dull, maybe you should be trying to set high scores/minimize turncount/even realtime speedrun instead

Another point- Changing the fight difficulty is tricky due to the diverse playerbase.

This. This and this yet again.

I'm sorry if this post brings nothing new to the game, but please, be aware that if the game gets harder and harder to satisfy veteran players, it may as well have serious problems in attracting new ones.

Disclaimer:

I am not a very good player.

I don't even want to make the game harder.

I just want the game to be based more on challenging player decisions than on sheer mental stamina.

[ackack]

I think TheDefiniteArticle had it right: probably what you really want is to play something other than Crawl, because there isn't much hope of a change comprehensive enough to fulfill that goal fully. (And if you did make such a change, you'd probably upset a lot of people who already like Crawl just fine.) I think a lot of these complaints apply to roguelikes more generally, although there are tons out there these days. You might try games like Sil or DoomRL.

[Igxfl]

I've played Sil and DoomRL. They're pretty good.

I don't post this expecting a complete DCSS overhaul, and I do think the game has some serious strong points, which is why I actually care enough to point out what I see as problems. It's be a shame to not bring them up.

[zardo]

Attrition mechanics may not be the best solution.

Actually I'm going to go so far as say they definintely are not. Having considered it more I think attrition mechanics encourage extremely cautious play more than the current paradigm, for psychological reasons. Players hate giving up power or resources. Just look at how people feel about item destruction, but also how new players constantly have to be taught not to horde consumables and that Revivification is good even though you lose max HP (some players would rather die than see their character lose power!) and so on. Whereas many are happy to gamble with their virtual lives in all kinds of ways, not just making combat decisions but straight up drinking mutation potions and such. This arguably indicates that what is optimal behavior to win Crawl does in fact diverge from what's fun for many players, but that's not a problem as long as both the winners and non-winners are enjoying the game.

I do agree that there are tedious aspects to (some definitions of) good play, and that it would be desirable to minimize them when possible but I think this would be done either by making situational evaluation deeper or more ambiguous, making the execution of optimal tactics more interesting somehow, or using a different definition of good play.

[twelwe]

There's no single player game that's going to retain a challenge for people who have played it for years and can beat it in their sleep, while still being playable and fun for less experienced people.

Spelunky, 2013 GOTY

This is the guy who can play it in his sleep, doing something no one has ever done before him solo (there is a co-op mode) https://www.youtube.com/watch?v=LGrPeu5NWk0

[khalil]

There's no single player game that's going to retain a challenge for people who have played it for years and can beat it in their sleep, while still being playable and fun for less experienced people.

Counterpoint: the Metroid games.
The Metroid games encourage speedrunning, to the point of showing how long it took you to win once you beat the game. The thing is, speedrunning the Metroid games requires skipping lots of items, thus making later fights more difficult. This means that speedrunners end up playing a more challenging game than other players. Example: the Screw Attack in Super Metroid is a "I win" button for any non-boss encounter. However, it's not required to beat the game, so speedrunners skip it and get a more challenging game as a result.

Speedrunning can add challenge for experienced players in almost every game, because while I may be able to beat Half Life with my eyes closed, I certainly can't do this: https://www.youtube.com/watch?v=VtI5HM7GVGY

[WalkerBoh]

My problem with the skill curve of Crawl is that you get from newbie to decent by figuring out good gameplay, and then get from decent to excellent by taking essentially the same set of tactics a decent player would use, but forcing yourself through the tedium of applying them perfectly by rote on every hallway of every floor of every branch for several hours.

Very well said, I see your point now.

My counterpoint is just that if the tactics aren't tedious, then applying them over and over does not get boring as quickly. I see no problem with a skill curve of "get decent by learning good tactics" -> "get excellent by applying them consistently". I see your issue as being that the tactics aren't fun, which causes reapplying them 2k times per game to be a drag. And I don't disagree with that. But I would rather have measures taken to address the problem of encouraging tedious tactics than radically altering crawl's difficulty and/or monsters.

[Igxfl]

That's probably true. Maybe I picked out the wrong culprit.
Reforming Crawl's tactical positioning minigame doesn't sound like a walk in the park either, but it's more manageable.

[tasonir]

My problem with the skill curve of Crawl is that you get from newbie to decent by figuring out good gameplay, and then get from decent to excellent by taking essentially the same set of tactics a decent player would use, but forcing yourself through the tedium of applying them perfectly by rote on every hallway of every floor of every branch for several hours.

Very well said, I see your point now.

My counterpoint is just that if the tactics aren't tedious, then applying them over and over does not get boring as quickly. I see no problem with a skill curve of "get decent by learning good tactics" -> "get excellent by applying them consistently". I see your issue as being that the tactics aren't fun, which causes reapplying them 2k times per game to be a drag. And I don't disagree with that. But I would rather have measures taken to address the problem of encouraging tedious tactics than radically altering crawl's difficulty and/or monsters.

This is one of the main reasons why Chei is my favorite god. I know it's important to retreat from every monster you see and fight them in cleared areas, but that's tedious as hell and I have no desire to do it. Chei instead gives you more power for not retreating, which is already what I want to do, because retreating is boring.

[TeshiAlair]

I've tried to make a few spell/god suggestions to discourage hallway fighting / encourage standing your ground, but its really difficult actually.

[sanka]

My experience is that the part of the game after Lair/Orc is extremly slow compared to the first part. There are some good moments, for example I liked the old last Vault level, and I usualy enjoy the last level of Zot. But getting to the interesting moments is slow and boring. Crawl's tactic is good, but it's not good and deep enough that you want to repeat it trough 25 floors when your character is already strong enough to win the game.

I'm not sure, but my feeling is that part of the problem is that character development almost halts after Lair/Orc. You will have a well established character at this point, and few things will change it. Aquiring one of those few things takes ages.

A crawl I think that I would enjoy would be significantly shorter. I would change all "normal" Lair rune branches to two levels, Depths and Zot also to two levels each. So if you completed Lair/Dungeon/Orc, you need to go trough 10 more levels.

If somebody thinks that compensation is needed for XP/items, than just give more xp for the monsters in these branches, and generate more loot in the lair rune vaults.

If somebody likes longer games, create more "optional" branches, although I think there are more than enough.

[TeshiAlair]

This idea is going to be insanely unpopular due to the dev time it would require, but I think a way to spice things up later is to have more fixedarts/more interesting randart properties. Even having a weapon thats not 100% optimal but fun to use and has some situational benefits would be cool to play with (I think the new rods are a great example of something that you never NEED to use but are hella fun).

I want a sword that shoots lightning, or a shield that can be wielded as a mace.

[Kismet]

My experience is that the part of the game after Lair/Orc is extremly slow compared to the first part. There are some good moments, for example I liked the old last Vault level, and I usualy enjoy the last level of Zot. But getting to the interesting moments is slow and boring. Crawl's tactic is good, but it's not good and deep enough that you want to repeat it trough 25 floors when your character is already strong enough to win the game.

I'm not sure, but my feeling is that part of the problem is that character development almost halts after Lair/Orc. You will have a well established character at this point, and few things will change it. Aquiring one of those few things takes ages.

A crawl I think that I would enjoy would be significantly shorter. I would change all "normal" Lair rune branches to two levels, Depths and Zot also to two levels each. So if you completed Lair/Dungeon/Orc, you need to go trough 10 more levels.

If somebody thinks that compensation is needed for XP/items, than just give more xp for the monsters in these branches, and generate more loot in the lair rune vaults.

If somebody likes longer games, create more "optional" branches, although I think there are more than enough.

I agree 100%. Climbing the power curve up to Lair 8 is fast and exhilarating and then things slow to a crawl.

[damiac]

There's no single player game that's going to retain a challenge for people who have played it for years and can beat it in their sleep, while still being playable and fun for less experienced people.

Counterpoint: Tetris.

Well, you can't really beat tetris... depending on which version you're playing at least. But the real tetris doesn't end, it just gets harder and harder. But I do see your point.
Other people pointed out self imposed challenges in other games, like metroid. I guess going for high scores in crawl is kinda comparable to that, but I don't think that's what top players do. I've never played Spelunky, maybe I should give it a shot.

But my point was, if we make crawl hard enough that it's challenging for duvessa and crate to win, how the hell do less experienced players win? And is that game any fun? Because my understanding of how you win crawl easily is to do exactly what this thread complains about, act like every popcorn encounter has a horde of pan lords behind it, and tediously lure them away 1 by 1 to safely slaughter in a corner. If you make that behavior necessary to win at all, nobody is going to want to play crawl.

If you make the game a battle of attrition, it becomes even worse. Now even if I know I'm just facing popcorn, I have to do what I used to have to do only for tough encounters. I have to do everything I can to minimize the damage I take, and the resources I consume to win the fight. Against every last orc, newt, and rat. Right now I can just charge in stupidly, take a bunch of extra hits due to horrible tactics, and rest it off, because the encounter was meaningless anyway.

I guess you're correct to say the real difficulty of crawl is NOT tabbing into that pack of green rats the one time there's a dangerous unique hiding around the corner. It's assuming that ogre will hit you for max damage 3 times in a row, even though it's highly unlikely, so you take an extra minute of real time to carefully kill it safely, rather than tab tab tab. Crawl sometimes asks you to trade boredom for survival, you can run around that pillar for 5 minutes and be safe, or you can roll the dice and see if you can kill the thing before it can kill you. One's more fun, and one's more likely to lead to a win.

I don't know how you fix that, without somehow scaling the difficulty to the character, which would go against crawl design philosophy in a lot of ways. What a mummy ice elementalist finds challenging is probably not so challenging for a minotaur berserker. In a game of random stuff, there are going to be random spikes in difficulty. I think the dev team does an excellent job balancing the game as it is now.

Also, just for the record, I enjoy the post lair game MUCH more than the pre lair game. That's when I feel like I actually start to be able to shape the character into what I want. I know some people enjoy adapting to what the randomness gives them, so I can see why others feel the opposite way.

[tasonir]

I'm not entirely sure if this is on topic or not, but I think it is: I play crawl primarily for the strategic aspects of the game. I am fond of designing a character that is incredibly strong - many of my wins have absurd stats, like a 37ac 54ev tengu berserker, or my 82 ac 46 ev gargoyle. I make sacrifices in the early game in order to stay 'pure' to my end game goal. I want to be so powerful that I don't have to bother with proper tactics most of the time. This does lead to a slight increase in errors when I'm discarding tactics for so long I fail to recognize there's a fight where I actually still have to use them, but most of the time I am able to get away with it.

I agree that I prefer the post lair game to the pre-lair & lair game. I am waiting for my strategy to come online before that, and then afterwards, I get to enjoy the fruit of my labor. I get frustrated when I haven't yet found a resistance I need (mostly MR) and might have to actually come up with a solution that isn't tab, because tab is the best solution to everything. I realize this philosophy goes against a lot of crawl philosophy, and I make compromises when I have to. I've gotten a streak of 3, but overall I'm not that reliable of a player and probably won't be, I don't have the mental endurance to apply correct tactics on every fight for hours, and tab myself into corners a fair amount of time. Hence why I love huge ac/ev characters.

[tabstorm]

I think the reason the game becomes boring once you reach some moderate threshold of skill is the way the character's power growth curve is structured.

Before you reach the Lair, as you said, there are some fights which will not be dangerous, some which will be unless you use consumables or make an error, some in which you will likely die and must avoid the fight.

In the early game I think there is a decent mix of these categories, and the player is also limited by lack of escape consumables. However, as the game goes on, encounters tend towards 1., with an occasional dose of 2. (that decreases as your skill does) and (very rarely) 3., as the player becomes much more powerful relative to enemies. Even then, Crawl dosen't have that much tactical depth, so it becomes a pain to go through 4-5 hours of gametime between Lair and Zot:5 with mostly trivial encounters - in addition the game encourages you to make encounters trivial by splitting packs and so on, because at this point on the curve you can defeat basically enemy that isn't super OOD in a 1v1 fight. It is unfortunate that by the time your character "develops" and has a variety of tools available, you have already reached the uninteresting part of the game.

I'm not sure how exactly this would be fixed aside from greatly buffing many enemies, or greatly nerfing XP.

Fortunately I don't really care about optimal tactical play, I just like killing enemies and getting loot, and building my character.

edit: Not to tell people to play the game a certain way, but I think that if you approach Crawl purely from the viewpoint of a tactical puzzle in which you optimize your win probability, you will not have a fun time after you reach a certain skill threshold, because there is just not that much depth there in that regard.

[TeshiAlair]

Agree with enjoying post Lair more, particularly with the variety of branches. I think D/Depths is currently the weakest part of the game, since nothing really sets it apart or changes over time. I really do hope theres a push for more, shorter branches and/or more portal vaults that aren't as annoying as Ice Caves or Volcanoes.

[Kismet]

I'm not sure how exactly this would be fixed aside from greatly buffing many enemies, or greatly nerfing XP.

Or greatly shortening the game.

[tabstorm]

I'm not sure how exactly this would be fixed aside from greatly buffing many enemies, or greatly nerfing XP.

Or greatly shortening the game.

I'm basically treating that as "nerfing XP", I should have been more clear.

[Igxfl]

I agree that I prefer the post lair game to the pre-lair & lair game. I am waiting for my strategy to come online before that, and then afterwards, I get to enjoy the fruit of my labor. I get frustrated when I haven't yet found a resistance I need (mostly MR) and might have to actually come up with a solution that isn't tab, because tab is the best solution to everything. I realize this philosophy goes against a lot of crawl philosophy, and I make compromises when I have to. I've gotten a streak of 3, but overall I'm not that reliable of a player and probably won't be, I don't have the mental endurance to apply correct tactics on every fight for hours, and tab myself into corners a fair amount of time. Hence why I love huge ac/ev characters.

This is pretty close to where I'm coming from.
Some tactical challenges in difficult fights is fine, but I greatly prefer the strategic elements of the game, like managing skill training and adapting to equipment.
Unfortunately, these aspects of Crawl seem to only matter after they no longer really matter.

I'm not sure how exactly this would be fixed aside from greatly buffing many enemies, or greatly nerfing XP.

Or greatly shortening the game.

I'd be behind this. Not quite to the point of Sprint, but even down to half the current setup.

[tabstorm]

Some tactical challenges in difficult fights is fine, but I greatly prefer the strategic elements of the game, like managing skill training and adapting to equipment.
Unfortunately, these aspects of Crawl seem to only matter after they no longer really matter.

I think this sums it up well - even if you want to "adapt" to floor drops, by the time at which you have the XP to do so, there is no real reason to do it outside of "interest" except for no-brainer choices like training charms and evo.

[Sprucery]

Am I the only one who actually enjoys playing Crawl as it is? If you want shorter games, play sprints. Instead of making Crawl shorter, make longer sprints.

[WalkerBoh]

While I agree with most of the above points, I think you all are exaggerating a bit when it comes to difficulty in the post-Lair to pre-Zot phase of the game. There are many tactically challenging situations that arise in lair branches, especially ends, and the Vaults-specific enemies make that whole branch quite a challenge sometimes as well.

However, I also am a big believer in the following philosophy: If the game is getting boring, hit > until it isn't any more. I find that diving through uninteresting parts of the game makes it much less tedious without really sacrificing survivability too badly (as long as you don't dive too fast ). This is why so-called speed running is fun to me, because the whole game becomes a constant balance of "how far can I push the limits of my char without going over the edge".

In the end, the game is as fun as you want to make it. This isn't to say that options for reducing tedium shouldn't be explored, but just something to keep in mind.

[TheDefiniteArticle]

The problem with Crawl's length is not some big conundrum about individual fights being interesting or not and tactics vs. strategy blah blah blah. The game just keeps going long after you've won, simple as that. It's like, you go to see a 90 minute action movie, and you get the cool plot and the action scenes and everything important happens and the characters and story are wrapped up nicely; and then the movie keeps going for another fucking hour just because Nolan has a boner for Harvey Dent.

Partly this is because Depths forces you to be strong enough to win before you even enter, and partly it's because the difficulty curve is completely fucked (PROTIP: good games don't get easier the further you go in them), and partly it's because fully 50% of the XP in a 3-rune game is not acquired until after you have 3 runes!

get from decent to excellent by taking essentially the same set of tactics a decent player would use, but forcing yourself through the tedium of applying them perfectly by rote

You've just accurately described every skill that I have ever learned, so I don't know what the problem is supposed to be.

If the game is getting boring, hit > until it isn't any more.

Well unfortunately the least interesting parts of the game (Depths) are also the hardest to dive, while the parts that are easy to dive are the best parts of the game by a country mile (this is not a coincidence). Trying to dive makes much of the game much worse.

Unless you're some dumb combo like HOBe. Then everything is easy to dive, but you quit playing mid-way through because making spreadsheets is more engaging than playing HOBe.

[WalkerBoh]

Well unfortunately the least interesting parts of the game (Depths) are also the hardest to dive, while the parts that are easy to dive are the best parts of the game by a country mile (this is not a coincidence). Trying to dive makes much of the game much worse.

Not sure I agree. I think it's easier to dive the later parts of the game when your char is stronger. Most chars (excepting things like HOBe and every DD/Tr) are not strong enough to dive very early on or even through Lair, which I think we both agree are interesting parts.

But I dive lair branches and Zot every game, and in speedruns I dive depths too if I feel I am prepared for extended/Zot. I personally think Depths is somewhat interesting though, so I don't mind exploring large amounts of it in normal games.

[tabstorm]

The problem with Crawl's length is not some big conundrum about individual fights being interesting or not and tactics vs. strategy blah blah blah. The game just keeps going long after you've won, simple as that. It's like, you go to see a 90 minute action movie, and you get the cool plot and the action scenes and everything important happens and the characters and story are wrapped up nicely; and then the movie keeps going for another fucking hour just because Nolan has a boner for Harvey Dent.

Partly this is because Depths forces you to be strong enough to win before you even enter, and partly it's because the difficulty curve is completely fucked (PROTIP: good games don't get easier the further you go in them), and partly it's because fully 50% of the XP in a 3-rune game is not acquired until after you have 3 runes!

get from decent to excellent by taking essentially the same set of tactics a decent player would use, but forcing yourself through the tedium of applying them perfectly by rote

You've just accurately described every skill that I have ever learned, so I don't know what the problem is supposed to be.

If the game is getting boring, hit > until it isn't any more.

Well unfortunately the least interesting parts of the game (Depths) are also the hardest to dive, while the parts that are easy to dive are the best parts of the game by a country mile (this is not a coincidence). Trying to dive makes much of the game much worse.

Unless you're some dumb combo like HOBe. Then everything is easy to dive, but you quit playing mid-way through because making spreadsheets is more engaging than playing HOBe.

Regarding when to enter depths, I believe people nowadays say you should do V:5 after completing depths, so I don't think it's necessarily true to say that you get 50% of your XP after having 3 runes. Though, if you mean to say "50% of XP is contained in depths and Zot", then of course this is true.

I'm not sure I agree about the difficulty of diving depths, either, it's one of the easier parts of the game to me, so I'm unconvinced that it is hard to dive. Again, it depends what you mean by dive - If you mean to say that after acquiring the silver rune at XL22 or so, you go down the first staircase you see until you find the orb, OK, maybe then you won't have the safest orbrun in the world. If you mean "not autoexploring every inch of the level", I don't agree about the difficulty of diving depths - It's one of the easiest parts of the game and goes pretty quickly (usually takes about 45mins-1hr for me autoexploring everything on a typical melee character that isn't some atrocious race that must rest constantly) so I don't see why it would be so bad to dive it.

With that said, your general idea that Crawl is not really a good game (if you play it with the intent of playing optimally and treating it as a tactical exercise) is one I agree with (for reasons that have been elaborated by many in this thread).

[Igxfl]

With that said, your general idea that Crawl is not really a good game (if you play it with the intent of playing optimally and treating it as a tactical exercise) is one I agree with (for reasons that have been elaborated by many in this thread).

That's an illuminating way to put it.

Often while playing I feel forced to make tradeoffs between out-of-game enjoyment and in-game advantage.

[all before]

Unfortunately, we can't solve this boredom by introducing more type 3 fights mid-to-late game, because encountering near-unavoidable chances of permadeath after a few hours of play would be unfair, even downright cruel.

I disagree with this statement, but I think it gets to the crux of some of the disagreements going on here. If you are a tactics-loving player, then the game becomes less interesting once most fights don't pose a serious threat of death. So the post-Lair (or so) game gets boring because you have enough options available to solve most fights without thinking too much. If you are a strategy-loving player, then as you "invest" more in building a character, you expect chances of danger to go down to compensate for the time/planning invested. You might agree that the game gets boring, but you also expect the game to become more boring on some level.

I don't think tactics and strategy are opposed, but the latter sentiment tends toiward RPG-like play, and results in a grind-y late game.

So, like I said, I disagree--if the post-Lair game needs to be reformed, the aim should be to dole out more type 3 encounters, as these are the encounters that really make you stop and think about what you're doing. That might seem cruel, but I guess in crawl cruelty and fun bear a close relation.

[crate]

the real problem with crawl's length is the game is very nearly exactly the same on turn 1 as on turn 100000

crawl's tactics are good; they're not good enough to support a four hour game without changing at all

[Patashu]

the real problem with crawl's length is the game is very nearly exactly the same on turn 1 as on turn 100000

crawl's tactics are good; they're not good enough to support a four hour game without changing at all

What would you give as an example of a roguelike where statement 1 is not true?

[crate]

I don't have any examples.

You can instead go the other way: DoomRL games take me about an hour, which is an acceptable length.

[zardo]

Honestly given the number of times the high-skill players in this thread are documented to have completed this game I think the fact that it got dull for them is one of those problems that's not really a problem. I mean I on the other hand have relatively recently reached a skill threshold where I want to play the game *more* because I am likely to get any given combo going pretty well in a handful of tries if not on the first. Which again is not to say there aren't real issues aired here (the more games I win the more I do find the last third or so a slog not to mention extended) but you know, that's still a lot of replay value.