Poll: What do you think about shaft effects?


If it doesn't fit anywhere else, it belongs here. Also, come here if you just need to get hammered.

What do you think about shaft effects?

Shaft effects are alright as they are
14
50%
Probability should be slightly reduced
4
14%
Probability should be reduced
2
7%
Probability should be strongly reduced
3
11%
Shaft effects should be removed
5
18%
 
Total votes : 28

Vaults Vanquisher

Posts: 461

Joined: Tuesday, 11th December 2018, 19:14

Post Tuesday, 3rd December 2019, 21:07

Re: Poll: What do you think about shaft effects?

duvessa wrote:But there are a ton of things you can do to minimize the risk of malmutation and paralysis before they happen...


True. You can even guarantee you'll never get teleported into Zot:5 lungs in every single game.

...if you're willing to endure sufficient tedium to make it so. You can *always* scum abyss for -tele or *always* switch to Ashenzari s a non-Dg, or even *always* gradually farm up to magic like D Door + revivification + shatter/tornado/firestorm with mana pots. Similar, but to less extremes, for blocking para/malmutes using summons (butterflies aren't reliable, but SLS is pretty reliable as are options like fog/cblink/etc).

I do dislike no-MR option games, though I've died to paralysis in a no-MR option game exactly once in 700+ games (which is why I remember it). That felt pretty bad, but it's the kind of fluke-odds thing even I'd not advocate putting significant time into addressing. Too many potential interactions at lotto ticket odds to care about them.

As for shafts specifically, they have the potential for extreme lethality/unfair outcomes on D:1 and quickly fall off after a couple floors. Kind of an annoying extra cheap death way to end streaks, in addition to the typical D:1 "jackal pack vs wizard" or "no pillar gnoll encounter" that decides your streak is over.

if shafts aren't dangerous past early game it's because nothing is dangerous past early game; that would be a problem with past-early-game, not with shafts


This is fairly disingenuous; in crawl many floors have a few situations where poor play by the player will take you from 100 health to 0 health in 5 turns or fewer. Sometimes as few as 1 turn. Good players recognize these threats for what they are, properly anticipate which actions will give them low-to-no risk, and choose those actions. But that doesn't mean the situation isn't/wasn't dangerous; for every good player that almost never dies in depths there are 10 more who are lucky to get past depths at all. Maybe more than 10 more.

What later areas don't share with starting areas is the number of options available to the player. Early on, you can lack resources due to RNG or even if you have them, they're not ID'd yet. Meliai are much more dangerous than bees. A pair or D:11 Meliai can smite a -hp% character to death 100-0 even if you read TP on-sight, if unlucky. But in practice they're much less dangerous than D:2 to D:4 bees, because by the time you encounter Meliai you have wands, god abilities, identified potions, and at least good odds of getting away with TP or blink. On D:3 you probably don't have any of those things, just a couple 20 speed monsters you can't 1v1 bearing down on you, so you blind use whatever resources you have and hope to get lucky.

I consider the Meliai encounter to be the better one, despite that it's less lethal to expert players. I died to them early as a player, learned the danger, and improved. Can't say the same for D:3 bees before identifying resources, or getting shafted next to an adder I can't 1v1 or escape.

~~~

TL/DR:

Crawl does not need more of the game to be like D:1.
User avatar

Blades Runner

Posts: 562

Joined: Wednesday, 5th March 2014, 03:52

Post Wednesday, 4th December 2019, 03:33

Re: Poll: What do you think about shaft effects?

I really like shafts, think they make the game far more interesting, and would like to see more of 'em.

Vaults Vanquisher

Posts: 461

Joined: Tuesday, 11th December 2018, 19:14

Post Wednesday, 4th December 2019, 17:24

Re: Poll: What do you think about shaft effects?

Reptisaurus wrote:I really like shafts, think they make the game far more interesting, and would like to see more of 'em.


Fundamentally, what do shafts do to "make the game more interesting" that you couldn't do by intentionally moving into bad situations?

People say what you say about shafts fairly often, enough that there's merit in unpacking why that is. What standards for "interest" apply to shafts, beyond that they occasionally force good players into situations otherwise only seen with poor play? There must be some other standards being used, but it's not clear to me what they are.

Zot Zealot

Posts: 952

Joined: Tuesday, 4th January 2011, 15:03

Post Wednesday, 4th December 2019, 18:24

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:Fundamentally, what do shafts do to "make the game more interesting" that you couldn't do by intentionally moving into bad situations?


Whenever someone proposes to remove a bad feature like a broken spell or god or having upstairs, somebody proposes: why don't you simply ignore the feature?

In game design I take it somewhat granted that the aim is a game where you do not need to come up with your own set of conducts that form your own minigame in order to make the game interesting.

For this message the author sanka has received thanks:
duvessa
User avatar

Shoals Surfer

Posts: 320

Joined: Thursday, 1st November 2018, 02:33

Post Wednesday, 4th December 2019, 19:08

Re: Poll: What do you think about shaft effects?

It's a common error in crawl talk to say that a given situation only occurs with bad play and therefore can be dismissed. In dcss, good play virtually guarantees nothing interesting ever happens and good play is not otherwise interesting or exciting -- the greater the player's capacity for boredom, the better they can play.

You have to look at the possibilities of the game, judge them carefully, select those that create interesting play, and make those things happen. The fact that the player has so much control over the game that they can essentially eliminate situations that everyone with sensible opinions on the matter agrees are some of the most interesting is a basic failure of design.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Elite Crawl Power Player -- Pro Advice, Superior Design Opinion

For this message the author tealizard has received thanks: 2
duvessa, Implojin

Spider Stomper

Posts: 223

Joined: Saturday, 29th October 2016, 17:41

Post Wednesday, 4th December 2019, 19:53

Re: Poll: What do you think about shaft effects?

Past say level D5 what is the effective difference from a gameplay perspective between having shafts enabled and wearing a cursed ring of teleportation?
User avatar

Shoals Surfer

Posts: 320

Joined: Thursday, 1st November 2018, 02:33

Post Wednesday, 4th December 2019, 20:38

Re: Poll: What do you think about shaft effects?

Without shafts, you can nearly always stairdance your way into a foothold on every level. Shafts sometimes but by no means always create a situation where the player has to make that foothold with no stairs to rely on or get through a chase scenario. Both of these are very different from the easy stairdance. They are also better than the stairdance. They force real engagement with randomly generated content by removing much of the player's control.

In current dcss stairs are a core mechanic. Exploiting the broken mechanics of stairs is a huge part of high winrate play. Shafts also play a major role in high winrate play in that they're one of the only ways reasonable players are going to lose. These are related issues. We should be looking to make entry to new levels closer to getting shafted than stairdancing.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Elite Crawl Power Player -- Pro Advice, Superior Design Opinion

For this message the author tealizard has received thanks:
Implojin

bel

Dis Charger

Posts: 2117

Joined: Tuesday, 3rd February 2015, 22:05

Post Wednesday, 4th December 2019, 20:39

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:
Reptisaurus wrote:I really like shafts, think they make the game far more interesting, and would like to see more of 'em.


Fundamentally, what do shafts do to "make the game more interesting" that you couldn't do by intentionally moving into bad situations?

People say what you say about shafts fairly often, enough that there's merit in unpacking why that is. What standards for "interest" apply to shafts, beyond that they occasionally force good players into situations otherwise only seen with poor play? There must be some other standards being used, but it's not clear to me what they are.

Shafts, like all traps, increase variance. In other words, you sometimes encounter them and sometimes don't. The situations can lead to very different outcomes and different decisions. Similar to other mechanics in Crawl like random monster generation, or creating artifacts randomly, or randomly generating starting items for Wanderers etc., more variance can mean more replayability and more avenues for players to make decisions.

Traps should be relatively "dangerous" because the player has a lot of resources and if you don't increase the danger by a significant amount, you're not increasing the variance by a meaningful amount. Certain types of variance can feel un-fun or unfair because the player doesn't have meaningful control over them or the variance is too high. These kinds of discussions occurred over traps as a whole when the "Traps and Doors" skill was removed.

Ziggurat Zagger

Posts: 6244

Joined: Tuesday, 30th October 2012, 19:06

Post Wednesday, 4th December 2019, 20:50

Re: Poll: What do you think about shaft effects?

severen wrote:Past say level D5 what is the effective difference from a gameplay perspective between having shafts enabled and wearing a cursed ring of teleportation?

Primarily that reading a scroll of teleportation isn't as likely to improve a bad situation after a bad shafting as it is after a bad teleport. (Also rings of teleportation are guaranteed to put you in LOS of at least one creature, and shafts are not)
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!
User avatar

Shoals Surfer

Posts: 320

Joined: Thursday, 1st November 2018, 02:33

Post Wednesday, 4th December 2019, 21:13

Re: Poll: What do you think about shaft effects?

You can increase the variance of something that's good by randomly adding things that aren't, right? Crawl doesn't have a problem of lacking replay value by virtue of too few theoretical possibilities. It has a problem of competent play reducing most of this possibility down to the same thing. It's fine to decrease some naive concept of "variance" if doing so increases the player's exposure to the better part of what remains or limits the ability of the player to make what remains all look the same.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Elite Crawl Power Player -- Pro Advice, Superior Design Opinion

Spider Stomper

Posts: 223

Joined: Saturday, 29th October 2016, 17:41

Post Wednesday, 4th December 2019, 21:45

Re: Poll: What do you think about shaft effects?

tealizard wrote:Without shafts, you can nearly always stairdance your way into a foothold on every level. Shafts sometimes but by no means always create a situation where the player has to make that foothold with no stairs to rely on or get through a chase scenario. Both of these are very different from the easy stairdance. They are also better than the stairdance. They force real engagement with randomly generated content by removing much of the player's control.

In current dcss stairs are a core mechanic. Exploiting the broken mechanics of stairs is a huge part of high winrate play. Shafts also play a major role in high winrate play in that they're one of the only ways reasonable players are going to lose. These are related issues. We should be looking to make entry to new levels closer to getting shafted than stairdancing.


Didn't Hellmonk already do this in Hellcrawl? I dunno what people's win rates would turn out to be in Hellcrawl but its definitely quite winnable (I have won it at least 10 times and there are much better players than me out there) and taking the downstairs is functionally the same as taking a one-way hatch. There is no stair dancing because there are no upstairs, in fact you can't even stay on a level for an extended period of time either (which sucks for my OpBe's when they are clearing the bottom of Vaults since they need to rest so much).

The implementation may bother some people since it fundamentally disables the ability to go anywhere in the world, but its affect on winnability/level gameplay is certainly concrete and can be studied. Especially since Hellcrawl is also harder than normal crawl. Additionally due to this implementation Hellmonk removed shafting since you can't go back up it would be stupidly punishing. The fact is half the things in this thread already have a ready made experiment with tons of playthroughs as evidence.

Vaults Vanquisher

Posts: 461

Joined: Tuesday, 11th December 2018, 19:14

Post Wednesday, 4th December 2019, 22:11

Re: Poll: What do you think about shaft effects?

sanka wrote:
TheMeInTeam wrote:Fundamentally, what do shafts do to "make the game more interesting" that you couldn't do by intentionally moving into bad situations?


Whenever someone proposes to remove a bad feature like a broken spell or god or having upstairs, somebody proposes: why don't you simply ignore the feature?

In game design I take it somewhat granted that the aim is a game where you do not need to come up with your own set of conducts that form your own minigame in order to make the game interesting.


In this case, however, we're stuck with a "feature" that cannot be ignored. Yet the question remains what it does to make the game interesting. I reject the notion of a perfect player, unless google's secretly been doing machine learning on crawl no player is close.

It's a common error in crawl talk to say that a given situation only occurs with bad play and therefore can be dismissed. In dcss, good play virtually guarantees nothing interesting ever happens and good play is not otherwise interesting or exciting -- the greater the player's capacity for boredom, the better they can play.


As such, I reject this statement as an *untrue* assertion, outright. Even the world record streak players have each encountered multiple dangerous scenarios of their own making, well into the games in question, during their streak. No human has the capacity to do what you assert. It's fantasy.

The fact that the player has so much control over the game that they can essentially eliminate situations that everyone with sensible opinions on the matter agrees are some of the most interesting is a basic failure of design.


"Everyone with sensible opinions" ---> drivel. Quoted is a poorly vield ad hominem.

If the game is actually boring when played well, that is indeed a failure of design. But that assertion doesn't seem consistent with the reality of crawl play + the crushing majority of players playing it, newbies + elite players alike.

Without shafts, you can nearly always stairdance your way into a foothold on every level.


Sometimes attempting this objectively lowers your expected winrate. Which is 1) dependent on build/species 2) consistent enough to happen in games as frequently as shafts and 3) is an interesting choice with significantly more agency.

You claim that "exploiting broken mechanic of stairs is a huge part of high winrate play", and yet the more experienced + higher winrate I've become, I stairdance less frequently, because there are multiple inherent risks to it that I have learned the hard way. I still do it when I evaluate it to give me the best odds of survival...which isn't even 1/3 of floors. Maybe not even 1/4.

You seem to consistently assert that removal of player agency = "interesting/engaging", yet the logical conclusions from that argument suggest that such preferences are better suited to another game entirely. Like slot machines. In crawl, the overwhelming majority of RNG interactions are intentionally designed with player agency in mind and interact well with player choices/control. It's not clear why the repeated assertion of an apparent intentional violation of this design is arbitrarily desirable only in specific contexts.

The game could force you to only use one spell school randomly. It could randomly instantly kill you on high rolls from any particular monster. It could make what skills you train random. Most of the arguments being made in favor of shafts would also apply to these examples...all of them "increase variance" for instance. In other words, dilute the outcome differential between good and poor play arbitrarily. I reject that as a desirable goal. If you don't, do you accept all these extra random bits tossed in without question? If not, why not? Let's see more coherence and less insult.

Traps should be relatively "dangerous" because the player has a lot of resources and if you don't increase the danger by a significant amount, you're not increasing the variance by a meaningful amount. Certain types of variance can feel un-fun or unfair because the player doesn't have meaningful control over them or the variance is too high. These kinds of discussions occurred over traps as a whole when the "Traps and Doors" skill was removed.


Yes, player agency is important. Critical, even. You should anticipate better outcomes when you make good decisions rather than bad decisions. Crawl's design mostly manages this, and part of the design philosophy mentions that the occasional unwinnable scenario is a byproduct of this design, but rare enough that it's not worth sacrificing the mechanics to get rid of it.

But (some) traps break that rule, intentionally, for what appears to be arbitrary reasons. They create variance independent from player decisions, and in some cases constrain player decisions rather than add to them (especially under "theoretical optimal play"). This isn't a case of a randomly generated combination of monsters showing up at the wrong time too early being near-unwinnable but so rare that it'd be silly to try to tweak against it. It's a case of straight up copying bad play and pasting it on good play because some people think that's fun.

If variance is fundamentally interesting, why are traps okay but random OHKOs against player + making skill progression completely random not okay? Those increase variance. They can "force player to re-evaluate and make different choices". We can even conveniently ignore previous/implied choices that now don't matter regarding them, just as we do for traps. Or are players in favor traps actually okay with such mechanics, carrying that preference set to its logical conclusion? If so I disagree, but at least it's a coherent, if strange in game context, position.

bel

Dis Charger

Posts: 2117

Joined: Tuesday, 3rd February 2015, 22:05

Post Wednesday, 4th December 2019, 22:43

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:
Traps should be relatively "dangerous" because the player has a lot of resources and if you don't increase the danger by a significant amount, you're not increasing the variance by a meaningful amount. Certain types of variance can feel un-fun or unfair because the player doesn't have meaningful control over them or the variance is too high. These kinds of discussions occurred over traps as a whole when the "Traps and Doors" skill was removed.


Yes, player agency is important. Critical, even. You should anticipate better outcomes when you make good decisions rather than bad decisions. Crawl's design mostly manages this, and part of the design philosophy mentions that the occasional unwinnable scenario is a byproduct of this design, but rare enough that it's not worth sacrificing the mechanics to get rid of it.

But (some) traps break that rule, intentionally, for what appears to be arbitrary reasons. They create variance independent from player decisions, and in some cases constrain player decisions rather than add to them (especially under "theoretical optimal play"). This isn't a case of a randomly generated combination of monsters showing up at the wrong time too early being near-unwinnable but so rare that it'd be silly to try to tweak against it. It's a case of straight up copying bad play and pasting it on good play because some people think that's fun.

If variance is fundamentally interesting, why are traps okay but random OHKOs against player + making skill progression completely random not okay? Those increase variance. They can "force player to re-evaluate and make different choices". We can even conveniently ignore previous/implied choices that now don't matter regarding them, just as we do for traps. Or are players in favor traps actually okay with such mechanics, carrying that preference set to its logical conclusion? If so I disagree, but at least it's a coherent, if strange in game context, position.

It's unclear to me whether you're agreeing with me, disagreeing with me, or partly agreeing with me.

I don't think "player agency" is necessarily important. At least it's not clear to me what you mean exactly. You don't have any control over whether Grinder generates or not, or what particular layout of Snake:$ you'll encounter. You make decisions after the fact, same as you do with a trap.

Your last paragraph doesn't make sense to me. Yes, variance is one of the factors which is good, but that doesn't mean it's the only factor. Nor am I saying that the variance should be so high that your situation after the fact should be completely independent of your situation before the fact. In fact, I say the opposite. If I'm talking about a glass of wine at dinner, that doesn't mean that I want you to drink alcohol till you puke.

As for OHKOs, Crawl combat is very swingy. Also, for instance, an early Ogre (or, say, an Ettin or Stone Giant in Orc, or a Hydra in Lair) can be a OHKO (or close enough) for many characters. I think, overall, it's a good thing to have swingy combat. Though, of course, one can debate the degree of swinginess and particular cases.

For this message the author bel has received thanks:
duvessa

Blades Runner

Posts: 626

Joined: Saturday, 12th December 2015, 23:54

Post Wednesday, 4th December 2019, 23:23

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:You should anticipate better outcomes when you make good decisions rather than bad decisions.

This thread of the discussion makes it obvious that shafts feel bad because they are perceived as the game arbitrarily and randomly "doing something" to the player. Seems clear that much of this bad gamefeel is removed if you simply remove upstairs and make "shafting" part of the dungeon structure. Psychologically, this would make shafting feel more like other aspects of dungeon generation, as part of the background game structure that one makes good decisions or bad decisions against. It goes without saying that good decisions made following a shafting are more likely to lead to good outcomes, even with crawl's current trap system. There is no evidence whatsoever for the assertion that shaftings "dilute the outcome differential between good and poor play arbitrarily"; highly dangerous situations (of which there are few in a typical crawl game) are precisely the situations where skill is most important and shafts, for all their faults, occasionally generate these.
Remove spell hunger.

For this message the author Hellmonk has received thanks: 2
duvessa, Implojin
User avatar

Vestibule Violator

Posts: 1613

Joined: Saturday, 18th June 2016, 13:57

Post Wednesday, 4th December 2019, 23:41

Re: Poll: What do you think about shaft effects?

I didn't read these last posts. Personally, I find shafts very marginal, especially while there exists an effect like Banishment, which is essentially Shafts on Steroids. Yes, they can mix up things a bit, which is nice. But I probably would handle them in a different way, with something like a "Shaft" status from a spell or trap that makes you fall down shafts for a certain period. It ends sooner if you can climb upstairs.

I understand how infuriating they must be for streakers, given how much important a game is for them, but I never will try to streak, so I don't see much of the problem for the common player.

But I agree that they are part of the problem of front-loaded difficulty, and could need tweaks because of this.
I Feel the Need--the Need for Beer
Spoiler: show
3DSBeTr 15DSFiRu 3DSMoNe 3FoHuGo 3TrArOk 3HOFEVe 3MfGlOk 4GrEEVe 3BaIEChei 3HuMoOka 3MiWnQaz 3VSFiAsh 3DrTmMakh 3DSCKXom 3OgMoOka 3NaFiOka 3FoFiOka 3MuFEVeh 3CeHuOka 3TrMoTSO 3DEFESif 3DSMoOka 3DSFiOka

Ziggurat Zagger

Posts: 4347

Joined: Friday, 8th May 2015, 17:51

Post Yesterday, 12:15

Re: Poll: What do you think about shaft effects?

sanka wrote:In game design I take it somewhat granted that the aim is a game where you do not need to come up with your own set of conducts that form your own minigame in order to make the game interesting.


This is a wrong premise in a game without explicit difficulty levels. Hard MiBe game for experienced player without own set of conducts, really???
I like how you use the word "minigame". Yes, shafts can be a minigame, you are instantly moved into a different mode where you don't have ability to retreat, teleport, stair-dance etc. It is a different game and should be optional.

Shafts are bad because they try to do too many things, even according to those who like them:
1) make boring game more interesting (it would be better to fix the root cause of boredom instead)
2) remove player control (I don't know word "agency") in a game which is all about control. Go play roulette if you like randomness, this game is closer to chess than to roulette. Even banishment is better as at least there are items that can decrease/remove the risk of banishment
3) regularly does nothing, players get excited and end with simple return to previous level
Underestimated: cleaving, Deep Elf, Formicid, Vehumet, EV
Overestimated: AC, GDS
Twin account of Sandman25
User avatar

Shoals Surfer

Posts: 320

Joined: Thursday, 1st November 2018, 02:33

Post Yesterday, 14:05

Re: Poll: What do you think about shaft effects?

@severen: There's no doubt that the most successful approach to shafts/stairs/level-entry taken in an actual playable version of crawl is the no-upstairs approach of hellcrawl and I tend to think this is in fact the best way forward for dcss. To my mind, with the proliferation of forks that basically take a "crawl classic" approach, dcss ought to adopt a more ambitious approach to solving the problems of crawl (and to some extent, that may be happening). That could mean looking for a middle ground between hellcrawl and recent dcss, like shaft-stairs or shafts to enter branches, or maybe a different direction that hasn't occurred to me at all.
This is where mechanical excellence and one-thousand four-hundred horsepower pays off.

Elite Crawl Power Player -- Pro Advice, Superior Design Opinion
User avatar

Vestibule Violator

Posts: 1613

Joined: Saturday, 18th June 2016, 13:57

Post Yesterday, 14:59

Re: Poll: What do you think about shaft effects?

VeryAngryFelid wrote:Go play roulette if you like randomness, this game is closer to chess than to roulette.

Actually, I think that this game is perfectly between roulette and chess. You take a chess-like decision (I want to hit that orc), and then you have to roll the roulette (get red to hit the orc).

I wonder how it would be if the roulette were before the chess part. It would be something like "roll to see if you can attack right now", and if you win you can use a non-randomised attack.
I Feel the Need--the Need for Beer
Spoiler: show
3DSBeTr 15DSFiRu 3DSMoNe 3FoHuGo 3TrArOk 3HOFEVe 3MfGlOk 4GrEEVe 3BaIEChei 3HuMoOka 3MiWnQaz 3VSFiAsh 3DrTmMakh 3DSCKXom 3OgMoOka 3NaFiOka 3FoFiOka 3MuFEVeh 3CeHuOka 3TrMoTSO 3DEFESif 3DSMoOka 3DSFiOka
User avatar

Zot Zealot

Posts: 947

Joined: Monday, 15th April 2013, 15:10

Location: Augsburg, Germany

Post Yesterday, 16:01

VeryAngryFelid wrote:Go play roulette if you like randomness, this game is closer to chess than to roulette.

Maybe.

Vaults Vanquisher

Posts: 461

Joined: Tuesday, 11th December 2018, 19:14

Post Yesterday, 22:01

Re: Poll: What do you think about shaft effects?

It's unclear to me whether you're agreeing with me, disagreeing with me, or partly agreeing with me.


Partly agreeing.

I don't think "player agency" is necessarily important. At least it's not clear to me what you mean exactly. You don't have any control over whether Grinder generates or not, or what particular layout of Snake:$ you'll encounter. You make decisions after the fact, same as you do with a trap.


Agency --> player's outcome is consistently determined by player's choices.

There are many interactions in crawl where player gets more variety at the expense of an extremely small amount of agency on average. Traps, on the other hand, have a tendency to significantly reduce agency (still small, but much larger than another arbitrarily selected mechanic on average) without adding commensurate variety...even very good players (the best in the world by any objective measure) encounter trap-like scenarios without traps.

Your last paragraph doesn't make sense to me. Yes, variance is one of the factors which is good, but that doesn't mean it's the only factor. Nor am I saying that the variance should be so high that your situation after the fact should be completely independent of your situation before the fact. In fact, I say the opposite. If I'm talking about a glass of wine at dinner, that doesn't mean that I want you to drink alcohol till you puke.


Yet that glass at wine at dinner has several orders of magnitude greater chance of causing you to puke than any other thing on the table. Maybe all other things on the table combined, after D:1 to D:4 or so. Though this is also where traps are at their most damaging.

OHKO example was that literally everything has a chance to just kill you regardless of HP or its damage. That would add variance. We could even make the chance small (1 in 100,000 hits, or even 1 in 1,000,000). That way people just randomly die, to add "interesting" variance :p.

This thread of the discussion makes it obvious that shafts feel bad because they are perceived as the game arbitrarily and randomly "doing something" to the player. Seems clear that much of this bad gamefeel is removed if you simply remove upstairs and make "shafting" part of the dungeon structure. Psychologically, this would make shafting feel more like other aspects of dungeon generation, as part of the background game structure that one makes good decisions or bad decisions against. It goes without saying that good decisions made following a shafting are more likely to lead to good outcomes, even with crawl's current trap system.


Yes, you could rework crawl significantly to the point where getting shafted has reduced chance of no-agency lethality, rework inventory/backtracking, and leave the game playing more like Tomb (perhaps with less torment). And then shafts wouldn't be meaningful. That's a lot of work to create a few extra scenarios that even the most elite players in the world still encounter without doing this, but you could do it.

There is no evidence whatsoever for the assertion that shaftings "dilute the outcome differential between good and poor play arbitrarily"


True or false: player movement choices have a significant impact on outcome differential?
True or false: there is a non-zero chance, however small, that players can be shafted/TP'd next to something they don't yet have resources to defeat or escape?
True or false: zero is less than any positive probability

We can argue "how much" it dilutes the outcome, or discuss different value judgments about how much dilution is acceptable. Pretty much everyone in this thread accepts at least some amount of dilution. But let's not pretend it doesn't exist.

highly dangerous situations (of which there are few in a typical crawl game) are precisely the situations where skill is most important and shafts, for all their faults, occasionally generate these.


It should be relatively uncontroversial to point out that frequency of "highly dangerous situations" is inversely proportional to how skillful somebody is at crawl.

I didn't read these last posts. Personally, I find shafts very marginal, especially while there exists an effect like Banishment, which is essentially Shafts on Steroids. Yes, they can mix up things a bit, which is nice. But I probably would handle them in a different way, with something like a "Shaft" status from a spell or trap that makes you fall down shafts for a certain period. It ends sooner if you can climb upstairs.


But banishment has significantly more counterplay than shafts, especially early shafts. You can't even get banished before mid-dungeon unless you're worshiping Xom (and he can move stairs away from you to mimic shafting too lol). While MR gear is RNG, it's at least an option. You can also identify and move away from banishers, or block their LoS. On top of all that, outside a few edge cases banishment is typically safer than D:1 through D:4 or so shafts!

If shafts had a similar degree of *consistent* counterplay to banishment when the player encountered them, I'd not be bothering to argue in this thread!

The more I discuss here the more I realize that I agree with you WRT front-loaded deaths. I'm not sure I'd call them "difficulty", which implies agency/better choices available. Often there are, sometimes there aren't...and that "sometimes" is *extremely* front-loaded in crawl.

Ziggurat Zagger

Posts: 6244

Joined: Tuesday, 30th October 2012, 19:06

Post Today, 00:41

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:
There is no evidence whatsoever for the assertion that shaftings "dilute the outcome differential between good and poor play arbitrarily"


True or false: player movement choices have a significant impact on outcome differential?
True or false: there is a non-zero chance, however small, that players can be shafted/TP'd next to something they don't yet have resources to defeat or escape?
True or false: zero is less than any positive probability

We can argue "how much" it dilutes the outcome, or discuss different value judgments about how much dilution is acceptable. Pretty much everyone in this thread accepts at least some amount of dilution. But let's not pretend it doesn't exist.

You're missing a part of the picture, yes, there's a non-zero chance that shafts will put you in a situation where no difference in skill will make a difference in the outcome, however there's *also* a non-zero chance that shafts will put you in a situation in which skill will make *more of a difference than would be made if you were never shafted*

We aren't comparing the chances that being shafted will produce a negative outcome to *nothing*. Which of these two possibilities is greater is a subject which there's not a reasonable way to quantify, it's not reasonable to say with any certainty whether shafts have a positive or negative impact on the degree to which skill influences a positive outcome in the game.
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!

For this message the author Siegurt has received thanks:
petercordia

Blades Runner

Posts: 626

Joined: Saturday, 12th December 2015, 23:54

Post Today, 03:48

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:Yes, you could rework crawl significantly to the point where getting shafted has reduced chance of no-agency lethality, rework inventory/backtracking, and leave the game playing more like Tomb (perhaps with less torment). And then shafts wouldn't be meaningful. That's a lot of work to create a few extra scenarios that even the most elite players in the world still encounter without doing this, but you could do it.

Indeed, I could (and did) make a fork of crawl with these mechanics. It was good, hence I am advocating the adoption of these mechanics for mainline crawl.

True or false: player movement choices have a significant impact on outcome differential?
True or false: there is a non-zero chance, however small, that players can be shafted/TP'd next to something they don't yet have resources to defeat or escape?
True or false: zero is less than any positive probability

We can argue "how much" it dilutes the outcome, or discuss different value judgments about how much dilution is acceptable. Pretty much everyone in this thread accepts at least some amount of dilution. But let's not pretend it doesn't exist.

"There is a non-zero chance, however small, that normal crawl monster generation can generate something in a position where it can unavoidably kill you, therefore generating monsters that can kill the player dilutes the outcome in player skill." Perhaps ponder this for a moment.

The mistake you're making here is forgetting that, in order for player choice to "have a significant impact on outcome differential," some of the player choices have to have bad outcomes. Situations that can kill the player are good at doing this. It's easy to make DCSS have no unavoidable deaths, by playing in wizard mode, for example (wizard mode also gives the player much more "agency," though in a shallow, meaningless sense). Wizmode DCSS has no meaningful skill differentiation at all. If shafts are more likely to generate a dangerous situation with few survival outcomes than to generate an unavoidable death then they may increase player skill differentiation. DCSS monster generation and not playing in wizmode likewise increase skill differentiation even though they sometimes lead to unavoidable deaths.

It should be relatively uncontroversial to point out that frequency of "highly dangerous situations" is inversely proportional to how skillful somebody is at crawl.

In a theoretical sense, a situation is "more dangerous" than another if a relatively smaller number of sequences of play survive it (one can add probability to this model, but the point should be clear enough). Expert players are better able to make the right moves and survive, thus differentiating them from players with less skill. Situations where you can do whatever you want and live do not evaluate player skill at all. Increasing the frequency of dangerous situations is good for differentiating between different levels of player skill.

With regard to skilled players playing themselves into dangerous situations, I will not speak for everyone, but in my experience this occurs largely because half of the game is fucking boring unless you are doing an explicit challenge run. I do not play very seriously for significant portions of the game even while on streak. I would expect the top players to play better (and for skill differentiation among the top players to improve) if DCSS generated dangerous situations with greater frequency. Unlike Siegurt I think it's reasonable to say qualitatively that shafts are more likely to lead to dangerous, interesting gameplay than to unavoidable deaths. Crawl's trap mechanic has bad gamefeel and should be removed, but not out of some misguided idea that it lowers the game's skill cap. DCSS should generate more dangerous situations than it currently does.
Remove spell hunger.

For this message the author Hellmonk has received thanks: 4
bel, duvessa, petercordia, tealizard

bel

Dis Charger

Posts: 2117

Joined: Tuesday, 3rd February 2015, 22:05

Post Today, 09:30

Re: Poll: What do you think about shaft effects?

TheMeInTeam wrote:Agency --> player's outcome is consistently determined by player's choices.

Let us unpack this statement.

First, "determined" is a strong condition. Crawl has a lot of RNG-influenced stuff, so it's hard to have deterministic outcomes. The game has inherent variance. Instead, a better formulation would be "player outcomes are correlated with player choices", that is, better choices will probably lead to better outcomes, but not always. Now, the degree of correlation can be argued.

To illustrate this, let's extend an analogy already made in this thread, chess and roulette. Instead of using roulette (which is just luck), I'll use poker, because it's a strategy game with inherent variance.

If I play chess against Magnus Carlsen, I'd lose pretty much 100% of the time. If I play poker against Phil Ivey, I have a decent chance of winning by simply going all-in every hand. Even the worst hand in poker (seven-deuce offsuit) has a 12% chance of winning against the best hand (pocket Aces). However, if we look past the inherent variance of poker, then the decisions you make are still correlated with outcomes; it's just not a one-to-one correspondence.

Coming back to Crawl: some people like to play a more chess-like game, suppressing variance as much as possible; other players are happy with the more swingy aspects of Crawl. My own preference is towards the latter.

Ziggurat Zagger

Posts: 6244

Joined: Tuesday, 30th October 2012, 19:06

Post 18 minutes ago

Re: Poll: What do you think about shaft effects?

Hellmonk wrote:Unlike Siegurt I think it's reasonable to say qualitatively that shafts are more likely to lead to dangerous, interesting gameplay than to unavoidable deaths.

I should clarify that I don't think it's reasonable to *prove* either assertion beyond all reasonable doubt, given the type of information available. I *strongly suspect* that shafts create more survivable player skill challenges than they create unavoidable deaths, but I don't see any way of collecting enough empirical evidence to prove it one way or the other, and arguing assertions which are unprovable seems a fruitless sort of endeavor to me. (Particularly when these kinds of decisions can, as you rightly point out, be made from other, more concrete assertions)
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!
Previous

Return to Crazy Yiuf's Corner

Who is online

Users browsing this forum: No registered users and 7 guests

Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software for PTF.