The Tavern

njvack wrote:Hi friends, I'll remind y'all that this is getting hugely offtopic. Is there anything to add to the "let's remove hunger" thing (I suspect that the devs are aware that hunger bugs folks) or can we move the "how to not starve with low spellcasting" discussion to Advice?

I honestly don't think there's anything more to add, njvack.

As you say, the devs are certainly aware that hunger bugs people. Those of us who think hunger is a bad mechanic have already put forward every reasonable argument against it, and piling up more examples isn't going to help. While it might be useful to see the devs respond to some of these arguments, I doubt they have counterarguments that haven't already been aired in this thread, making it an exercise in futility. We'll just repeat the same arguments using increasingly angry words.

The only argument that would actually matter is code. If my OP was a patch instead of a list of suggestions, we could have an experimental branch and actually hash out these issues based on real play experiences. I encourage someone who isn't as worthless as me to put together a foodless branch. Otherwise, I can't think of anything else anyone can really say, either for or against hunger, that isn't just repeating what's already been said.

archaeo wrote:That's certainly a correct assessment of why this player lost the game, but,

araganzar wrote:It would have been unfair to good players to NOT to kill this player.

Why is that? And, if it's true, why is hunger the best way to ensure that fairness?

It's unclear to me why Duke's play was so abusive that the game needed to end it rather than waiting for a monster to kill him. If Crawl had no hunger costs, Duke probably still would've lost with this character; with just 27 MP, no channeling, and terrible defenses, Duke would've exhausted his resources quickly using his high-level spells and would be left defenseless.

We've moved on quite a bit but you did ask so I'll answer briefly. I see no reason why this character could not have won with 20 earth, 10 Conj, OOD, and a likely gift of Shatter and/or LCS incoming. Presumably he buys defenses and SC after final gifts. His consumable use at the end of lair looked insane though and he had no healing left, so who knows.

Your question of "why we needed to kill this player" is not relevant because this was not good or even mediocre play that was unfairly pressured or punished by hunger. Any one of the other mechanics proposed in this thread to replace hunger would have performed the same function, perhaps better and earlier. It's therefore not a point for or against hunger mechanics.

I think the point it does illustrate is that hunger is doing little to curb high level casting. I have to agree with you that in this instance other mechanics would probably have done a better job of both applying pressure and giving the player information they can use to react.

Really the long and the short of it is that hunger does not act as a clock. All it does is act as a limiter for high level spells at the moment. If you need a mechanic to limit high level spell why the hell do you plague every charcter with it not just those casting high level spells. If you want to limit high level spells create a mechanic that does just that and does not force everybody including those not spamming high levels spells to do a bunch of tedious inventory juggleing, doesn't constalntly interupt autotravel, and all the other bullshit that hunger brings with it. Hunger is a bad mechanic that does not fulfill its original design and has been shoehorned into a new purpose that could have a better mechanic that is specifically designed for the intended purpose. It is bad design and bad programming plain and simple.

gammafunk suggested that I put my proposal in spreadsheet form. If you'd like, you can view and edit that spreadsheet here.

Sar wrote:IMO a good triple sword is a much better weapon than Firestorm outside of Ziggurats, but I imagine most people will disagree.

That's just kind of a weird way to compare them.

Looking over my few pure caster morgues, I found that I ate somewhere around 700-800 chunks in a typical Sif game (without a staff of energy), vs. around 110 for a melee character. Is this really necessary? This is part of why people don't like playing conjuration-only characters.

I don't think hunger is a major factor in (some) people not playing conjuration characters. Sif is an extreme case. With Vehumet, it wouldn't be anywhere near that. In this normal DEFE run, I ate 340 chunks before lvl 25.

tabstorm wrote:Looking over my few pure caster morgues, I found that I ate somewhere around 700-800 chunks in a typical Sif game (without a staff of energy), vs. around 110 for a melee character. Is this really necessary? This is part of why people don't like playing conjuration-only characters.

If this does not show that the system is broken and needs to be removed I don't know what does. Boring, tedious, bullshit thrust upon all to limit a few and it doesn't really put much of a limt on those few.

Hunger also creates a noob trap in that it makes you think you must train lots of spellcasting, or play a mummy, to avoid the horrible pressure of the food system while playing a conjurations focused character. Of course that pressure is mostly imaginary with decent play, but a new player doesn't know that, nor do they necessarily play decently.

Why can't we just let the monsters kill the bad spellcaster, just like how they kill the bad berserkers?

dowan wrote:Why can't we just let the monsters kill the bad spellcaster, just like how they kill the bad berserkers?

Sandman25 wrote:wait for a monster to kill the character who has Shatter and Vehumet on Lair 8, right? What can I say? Good luck to the Orb of Fire and/or pack of Draconians

Sandman25 wrote:
dowan wrote:Why can't we just let the monsters kill the bad spellcaster, just like how they kill the bad berserkers?

Sandman25 wrote:wait for a monster to kill the character who has Shatter and Vehumet on Lair 8, right? What can I say? Good luck to the Orb of Fire and/or pack of Draconians

There are only like 5 races that can realistically cast Shatter or Tornado on Lair:8 - Tengu, Gargoyle, High Elf, Deep Elf, Formicid. All of these races are frail early game, 3 of them are frail all game, and one can't teleport or haste. Do we really need to subject everyone else to an eating contest because a few races can suicidally rush a level 9 spell? I don't think "Gargoyle mages would be too good!" is a reason to leave spell hunger in...

Also, low power shatter isn't even that good. It takes like 3 shots to kill e.g. a greater naga without any enhancers fairly often. More and more enemies resist it as the game goes on, and casting it is basically as good as marking yourself.

I would be happy if we removed spell hunger. I was just saying that it's hard to kill a character who has high level spells, we cannot expect it to die to monsters reliably.
Though I am not sure that we should limit high level spells, they are limited by MP anyway.

Honestly that GrEE was a lot more frail than the average Gargoyle in a plate armour. I actually came close to dying a few times due to miscasting shatter 3x in a row at 10% or having 10 fliers onscreen, whereas with like a GrBe my chances of dying are basically 0%. Shatter is the most overrated spell in this game.

3 miscasts in a row at 10%? Sounds like clustering illusion and/or confirmation bias!

mps wrote:3 miscasts in a row at 10%? Sounds like clustering illusion and/or confirmation bias!

... not really. Clustering illusion is the belief that clusters signify nonrandomnesss. Confirmation bias is when you pay more attention to data supporting your beliefs than data that does not. I don't see how any of those are relevant here.

Maybe you are misunderstanding clustering illusion to mean seeing clusters where they don't exist? I can't even guess what resembles confirmation bias to you.

That spreadsheet is awesome. It might be cool to color-code it so the more marginal solutions are red and the no-brainer ones are green, especially if a dev agrees. Personally I think removing hunger from various actions piecemeal is a fine solution and will be much more palatable to everyone, even if I want to get to the same endgame that you do.

The only difference is that I'm fine with an overall clock pushing me forward, I just don't want to do anything to advance it other than "play crawl". Completely passive hunger that's unrelated to spells would be fine with me.

Not sure why formicid is mentioned, +2 to Earth? Then DD should be mentioned, +3 to earth, and DD is not frail.

First off, Hurkyl, you're getting meme'd on by mps, who is on record as being disdainful of the Crawl community's use of "clustering illusions" and "confirmation bias" in response to statistical oddities. I think that was just bone dry sarcasm.

johlstei wrote:That spreadsheet is awesome. It might be cool to color-code it so the more marginal solutions are red and the no-brainer ones are green, especially if a dev agrees.

Thanks. I think that, right now, it's hard to tell what's a "no brainer" in the context of this proposal. In the past 24 hours, I've heard that it doesn't go far enough, or that it's all an abomination against a system that just needs to be tightened up. The only dev who's given me much feedback is gammafunk, who mostly took issue with the changes I proposed to the monster spawner.

The only difference is that I'm fine with an overall clock pushing me forward, I just don't want to do anything to advance it other than "play crawl". Completely passive hunger that's unrelated to spells would be fine with me.

In response, I'd point out that tabstorm's 110 chunks for his melee bros sounds like many of my own characters. I think 1 second per chunk is likely a low estimate of the average time spent per chunk (once you factor in inventory juggling, the butchering menu, etc.), which is already nearly 2 minutes of game time. When you add in all the other ways that the hunger game costs you time, I feel like 5 minutes per 3-rune game is still a very conservative estimate, and that would mean that tabstorm has spent at least 9 hours of his life managing the hunger clock during winning games. It wouldn't surprise me if the real total, including unwon games, is over 24 hours.

However you want to slice it, that's not "passive," it's a lot of player time being wasted. If Crawl needs an "overall clock," I think we'd be better off with something truly passive.

I also think the spreadsheet is awesome, though I am not a dev.

I think the numbers in the int drain proposal instead of hunger for high level spells are too harsh. Could be toned down. Especially, as getting int drain progressively higher would screw you harder.

archaeo wrote:In response, I'd point out that tabstorm's 110 chunks for his melee bros sounds like many of my own characters. I think 1 second per chunk is likely a low estimate of the average time spent per chunk (once you factor in inventory juggling, the butchering menu, etc.), which is already nearly 2 minutes of game time.

And also, 120 one-second interruptions is way more disruptive than one 120-second interruption.

(and there is all of the associated stuff being left out too, like the extra work paying attention to your hunger level, looking for butcherable things, et cetera)

Thanks, bel.

I don't honestly think that spell hunger needs a replacement to begin with; I'm nearly positive that the best thing to do would be to remove it altogether and then rebalance MP caps/skill costs around that norm. However, not everybody agrees, and I think some form of int drain is probably the best compromise.

I arrived at those numbers by observing that, if you just divide all the spell hunger and hunger mitigation formula results by 100, you get numbers that scale nicely with stats. If they're too much, one could probably shift that entire list down by a spell level, which would result in only getting drained by one or two int points about a third of the time when casting level 9 spells at 30 int and 20 spellcasting (or 29:21, 31:19, etc.), which seems reasonable for 3 runes.

That said, I'm the least qualified person in this thread when it comes to manipulating numbers, and I bet these int drain costs still aren't great. I encourage people to edit the spreadsheet with their own suggestions, though it would be helpful if they would sign their contributions.

I've started coding a patch for this based on what's in the spreadsheet - hopefully I can finish it without getting too stuck.

That would be amazing, Bodrick.

FWIW, while you should do whatever you want to do, I think it might be more productive to have a branch that simply removes all hunger costs for now. Maybe that's as easy (relatively speaking) as setting the base hunger of all non-Vp/Gh species to 0 and then removing all other costs? My suggestions for replacements aren't universally beloved, and I think it'd be better to see what actually needs to be replaced first.

archaeo wrote:That would be amazing, Bodrick.

FWIW, while you should do whatever you want to do, I think it might be more productive to have a branch that simply removes all hunger costs for now. Maybe that's as easy (relatively speaking) as setting the base hunger of all non-Vp/Gh species to 0 and then removing all other costs? My suggestions for replacements aren't universally beloved, and I think it'd be better to see what actually needs to be replaced first.

The easiest way would be to simple change the equation that decraments nutrition over time to setting nutrition to a constant each time. Playtest that, and see if we can break it, then play a normal version in the same fashion and see if it is possible with hunger. My guess is in all but a very few corner cases any percieved abuse is allready possible.

i hope someone can rework vp to retain a satiation mechanic, esp since this chart just says "maybe remove vp", but... uh. please don't. if any dev likes vp enough to care, perhaps they will consider to, for example, give them a "blood meter" (like the sorts of passive food counters suggested in some of these threads), filled automatically by killing enemies, which they can use to access alive and undead abilities.

Ok, having poked around in the code a bit to get to grips with it, I just set hunger to HUNGER_MAXIMUM whenever it would normally be decremented for now. I've put a fork up at https://github.com/BodrickLight/crawl/tree/foodless, and you can download windows builds from https://www.dropbox.com/s/7xefy81wa059o ... 0.zip?dl=0. Please let me know how you get on!

I've only skimmed this topic, but in case it hasn't been pointed out: it is vitally important that there is some penalty for spending several thousand turns waiting in place. The goal of having a food clock is entirely to curb these super-abusive tactics. It is not supposed to affect players playing in ordinary ways. (It is likely that crawl's food system fails to achieve this goal, though.)

If you don't see why it is important, I will direct you to DoomRL. DoomRL is generally fun, but it has some design flaws. The largest, in my opinion (the fact that there could be a debate about this shows how serious some of the flaws with DoomRL are), is that the best way to play DoomRL is to stand in a safe place for thousands of turns so that you always get monsters moving toward you, instead of you moving toward monsters. Because of other things about DoomRL (all monsters generate at the start of the level, almost everything is ranged and does a lot of damage) the problem is obviously worse in DoomRL than it would be in crawl, but even in crawl there's a noticeable advantage to having monsters come to you instead of the other way around.

It's quite possible to come up with other ways to force the player to actually, well, play the game. In crawl (outside of hell), monster spawns could do that. In many parts of crawl, they even succeed (you don't want to wait around for ten thousand turns on d:3; out-of-depth spawns in D are pretty brutal). But this is not true in all parts of crawl; lair out-of-depth spawns are, well, just more lair dudes. It's possible that if food is removed entirely with no other incentive to play added, that standing around on the entrance to lair (on L:1) for fifty thousand turns would improve your chance of winning. It's important to prevent this, since that's exactly the kind of grinding that the design philosophy is against. Standing around for fifty thousand turns in every lair level could make you starve to death (but see below), so the food clock (maybe) does its job here.

(Yes, mummies have no such clock. It's entirely possible that mummies should already be playing this way, but I'm going to ignore them in this post. Technically vp could probably do this also.)

---

The biggest problem with food in crawl is that it is used in two completely separate fashions that should not be connected. There is the food clock, which is as I described above. You probably noticed that that has absolutely nothing to do with the actually-playing-the-game experience of having to eat in crawl. The main thing food is used for in practice (i.e. when not trying to do specifically abusive things like wait around for fifty thousand turns on every floor) is for limiting a range of tactical abilities in combat (god abilities and spells, primarily). Since these use the same resource as the clock, and since the clock is not supposed to affect players who actually play the game and make progress, the clock itself needs to be lengthened pretty substantially: one bolt of fire might take a hundred turns of nutrition (or three hundred for spriggans!). If a player casts bolt of fire five hundred times, that's fifty thousand turns of nutrition. Of course, this means that actually the clock probably fails to curb the behaviours it is supposed to block, since you need to give the player such a huge amount of leeway to not adversely affect normal games. So if you just don't spend lots of extra nutrition on things like spells, you can in fact spend a hundred thousand turns doing nothing, with no drawback. (I did this once, to abandon Beogh and wait out wrath in lair. Of course god abandonment has changed since, but it's precisely this sort of behaviour--spending literally a hundred thousand turns doing nothing--that is supposed to be prevented by the food clock.)

If there is going to be a resource other than just MP that limits the use of various abilities, it should not be food (or alternatively food should not be used as the incentive to actually make progress and something else should; it doesn't matter what name you attach to each function). Then a clock that properly does not affect normal play but does curb abusive behaviour could be created successfully.

Sorry but I have to disagree. If you wan't to sit around for 50,000 turns on L:1 that is your problem. Again we should not fuck over everybody because somewhere between a microscopic minority and nobody at all might do stupid things. Hell I don't care if most people do it. That is there problem. Please do not make it mine!

acvar wrote:Sorry but I have to disagree. If you wan't to sit around for 50,000 turns on L:1 that is your problem. Again we should not fuck over everybody because somewhere between a microscopic minority and nobody at all might do stupid things. Hell I don't care if most people do it. That is there problem. Please do not make it mine!

The issue isn't that they do it. The issue is we need to make sure (somehow) that it isn't actually the OPTIMAL thing to do, because many players enjoy the game less if they experience the cognitive dissonance involved in doing the suboptimal-but-not-boring-as-hell thing. And they certainly enjoy it less if they do the optimal-but-insanely-tedious thing.

It's important that grindy 'stupid things' are never the best way to win. It's OK if they're usually survivable; that's why Abyss summing is OK for the game (Or could be with sufficient tweaking; no point arguing if it's there right now.) They just can't be optimal.

Edit: I suppose I should plug my hunger replacement again: make the OoD monsters that spawn include the dangerous abyssal monsters with increasing severity and percentage as the scumming goes on. It's direct, immediately reveals itself to a scumming player, and doesn't directly kill them for scumming; it just makes it dangerous.

acvar wrote:Sorry but I have to disagree. If you wan't to sit around for 50,000 turns on L:1 that is your problem. Again we should not fuck over everybody because somewhere between a microscopic minority and nobody at all might do stupid things. Hell I don't care if most people do it. That is there problem. Please do not make it mine!

It's literally one of the games design tenets that grinding should be prevented. That is one of the axioms of crawl design; it's a starting point. If you want to go disagree with the design principles underlying crawl, you can do that, but that is another topic entirely.

As I said, the goal of the food clock is clearly to be completely unintrusive to players who are actually playing the game. The problem is that it doesn't work very well, largely because the same resource is used for something else.

crate wrote:snip

I agree with almost everything regarding DoomRL, but I think it's worth clarifying that in difficulties like Nightmare it's not about whether you apply "optimal" tactics as much as you apply the only few viable tactics. One of these is waiting for thousands of turns if needed. Obviously DoomRL does lots of things that would make crawl devs and/or players foam at the mouth, but I really like both games myself. I find that the "waiting for thousands of turns" part really only adds minutes to a nightmare run, which may be unacceptable for a so called "coffee-break" roguelike game, but I digress. You still have to be able to find an effective choke point to wait on which still requires you to make some tricky decisions.

There's really weird spoilery knowledge required to play on any difficulty (you can shoot around a corner with near impunity), even moreso on Nightmare (stuff like corpses can't spawn on doorways or water tiles, and pistols can start knocking back with enough damage upgrades), but I find that you have to do a careful mix of super-abusive tactics like waiting a ton of turns while actively seeking an effective choke point, which I find interesting despite the apparent "scumminess". DoomRL is a case where tactics that can be deemed as scummy or highly abusive just made their way into the "metagame", and they became increasingly embedded into the game instead of being removed. In any case, from a crawl point of view this is probably unacceptable, and that's fine.

As far as Mu/Vp scumming: I parked myself as a MuFi in a special level around XL11, dug some killholes and scummed with Xom until I had a reasonably good kit of items. This got nerfed by lantern no longer amusing xom endlessly and stat drain being tied to XP, but you can still do this with sustain attributes for a very long time. The risk is relatively low as long as you get a feel for Xom's most dangerous actions and summons and only commit to scumming when you know you're ready.

I did something similar as an experiment with a VpAs on FMD sprint to survive with the Orb until I'm XL27 (if you kill everything you'll be at around XL13). Even when bloodless Xom will set your HP to 50% when he mutates you, which heals you when you're below 50%. This means you don't need to heal. Vp aptitudes are better than a mummy's and they can also mutate, which means you can keep scumming until you have a good enough mutation set.

These examples have more to do with exploiting Xom of course, but having tried to exploit OoD monsters with these races before, I think it's much more effective to just exploit Xom than it is to wait for OoD monsters.

The crawl design philosophy is mostly spent at this point. It's just not relevant to current crawl the way it was six years ago. Maybe it's a victim of its own success, but the fact that it can be used to defend the food system is evidence that new thinking is required.

I know the design tenet. It is wrong. You can not design for the lowest common denominator or you will have a product that sucks. The object of the game is to have fun. If you can not figure that out and you choose to bore yourself to death that is your problem. At some point you just have to let stupid people be stupid and ignore them.

But all of that is irrelevent since chunks/mummies/Vp exist. You can allready do what you fear and the game has not devolved into the nightmare scenario you invision.

crate wrote:As I said, the goal of the food clock is clearly to be completely unintrusive to players who are actually playing the game. The problem is that it doesn't work very well, largely because the same resource is used for something else.

Well, and the fact that even if you untangle all of the current hunger costs from the food clock, you're left with a fairly intrusive mechanic that, as has been said, acts more like a parking meter than a clock. You can completely automate this to remove player annoyances, of course, but at that point, why bother referring to "food" or "hunger" at all?

I broadly agree with byrel, as I think Crawl can keep players from standing still with tweaks to the monster spawning mechanics, likely by spawning OOD monsters earlier. Ideally, I'd like the game to remove the monsters as well, to avoid the annoyance of killing popcorn when traveling from branch to branch, but that's probably a harder problem to solve. There are also fancier proposals, like mps' suggestion to remove spawning and reform monster AI to prevent scumming, but most of these are likely impractical.

Of course, the hungerless patch that Bodrick released earlier is completely broken in these terms, but it's a good playground for trying all this theory in practice.

acvar wrote:Sorry but I have to disagree. If you wan't to sit around for 50,000 turns on L:1 that is your problem. Again we should not **** over everybody because somewhere between a microscopic minority and nobody at all might do stupid things. Hell I don't care if most people do it. That is there problem. Please do not make it mine!

The problem is that I don't want to sit around for 50,000 turns on L:1. If I believe that to be the best play, then I will be unhappy whether I do it or not.

mps seems to be working backward. He hates food, so anything that requires food is bad and must go. The correct way to do so, (one option, which crate also mentions), is to make a clock (it does not have to be food). Nothing requires that the clock be called "food". Like Sil has "min-depth" clock, which of course works better in a linear game, instead of a game with lots of branches.

I would definitely not like a crawl where scumming is optimal, though of course the anti-scumming part doesn't work well in Lair now. I mostly stopped playing nethack after camping altars and sacrificing stuff which was mind-numbingly boring. (though the straw that broke the camel's back was interface related: performing cannibalism mistakenly by trying to eat a food item in slot "y").

acvar wrote:Sorry but I have to disagree. If you wan't to sit around for 50,000 turns on L:1 that is your problem. Again we should not fuck over everybody because somewhere between a microscopic minority and nobody at all might do stupid things. Hell I don't care if most people do it. That is there problem. Please do not make it mine!

When answering what they would do about pudding farmers, the Nethack devs answered: "The DevTeam has arranged an automatic and savage punishment for pudding farming. It's called pudding farming."

I mostly agree with this statement. Maybe Nethack's pudding farming is OP, but sitting around Lair for thousands of turns isn't THAT overpowered, plus obviously being horribly degenerate behavior. I don't see it becoming that much of a problem.

I see that I did not make myself clear enough in my previous post, because the most important thing I was trying to illustrate apparently got lost. This is probably my fault, because I constructed the previous post in a poor fashion. I will attempt to fix that here.

What I was trying (and failing) to say is this:

Crawl food, currently, attempts to regulate two completely different activities that should not be connected in this way. Because food is attempting to regulate both tactical actions (spellcasting etc.) and attempting to act as an incentive for active exploration, it automatically fails to regulate either of those things (see previous post for explanation). This situation is unfixable, except by disconnecting the two different types of "food". I admit I skimmed this topic, so it is possible someone else had already brought this up, but it is important and not something that I have seen brought up in the past in other topics. (Or, at least, not something I see brought up as a clear problem that needs to be fixed. Asking for food removal is not asking for one removal, but rather asking for two. This is important.) I should have said this first and more clearly in my previous post.

I also believe that the real complaints are with the implementation of food (which is very bad) rather than with the ideas behind regulating the two different things that food tries to regulate.

The rest is less important, but something I feel strongly about, so some more detail below:

---

The devs have clearly decided that both a tactical limiter other than MP is worth keeping (this is why spells still have hunger costs), and that some sort of exploration incentive is worth having (this is why you can starve to death). Well, or they haven't thought about this at all, but I'm pretty sure I've mentioned that food does two separate things to at least one dev in the past. So I think they're aware of the fact it does two things.

---

I have no real opinion on the former; I think it's entirely reasonable to have MP be the only limiting factor for spellcasting during a single fight, and I also think it's reasonable to have some other resource act as a limiter also. Probably I would lean toward eliminating this type of food entirely, but I'm pretty sure at least one dev would like to keep it around. So, no comment from me.

---

The latter, having some sort of exploration incentive so the player cannot just stand in one place, is something I do have a definite opinion about. The devteam has a firm stance on "grinding", which is that activities that are grindy are things that should be actively prevented by the game (debate on this deserves a separate topic, if you want to discuss it). Personally I agree with this stance, but really any discussion of it deserves a separate topic so I will not expand upon that here. Food attempts (and fails) to partially fill the role of exploration incentive. The reason to have something pushing the player to actually go around and explore levels is pretty clear, and if you don't see why then DoomRL is a great example of a game that has nothing pushing the player to explore and suffers very clear problems from it.

Here is where I run into terminology problems in my previous post. I use the terms "food" and "clock", without really explaining what I mean. I certainly don't believe that keeping food in the form of "item you find while exploring, and occasionally you press two keys to increase a somewhat-hidden value that kills your character if it reaches zero" is necessary or useful (this is what people complain about). I do think that keeping food in the sense of "thing that you find while exploring that gives you more time before your character dies, even if it's not supposed to do anything if you actively explore and play the game" is useful. Note that it doesn't actually have to be an item your character picks up; you could just automatically gain nutrition for revealing tiles, or something.

It's not technically necessary for an actual clock (i.e. number that ticks down with inactivity and kills you directly if it reaches zero) to exist to create a situation in which active exploration is definitively better than sitting around for hundreds of thousands of turns. Crawl attempts to do this in some other ways, one of the more obvious being out-of-depth spawns increasing in frequency with time spent on a particular level. However, it is my belief that the non-food clocks in crawl fail to properly incentivize exploration in all areas of crawl, so I think having an overall "food clock" is worthwhile, even if just to prevent the most extreme abuses. Perhaps eventually crawl will reach a point where any sort of food clock would be completely meaningless, as it is moving in that direction already, but it has not achieved that yet.

---

I will freely admit that it's very likely that boring-but-marginally-beneficial activities (i.e. grindy things that I don't want to do) existing decrease my enjoyment of a game to an unusually large degree. I never actually did any pudding farming, and rarely did any polypiling, but the fact that those are possible in Nethack still significantly decreased my personal enjoyment of the game. As far as I can tell, this is not a common thing, and certainly it bothers almost everyone else less than it bothers me.

---

I said this below, but it's important here too:
Okay I thought I said this but apparently I didn't say it in plain language: the act of having to press buttons to eat food is something that should be removed from crawl (at least, as far as the clock function of food is concerned. Having to consume items to deal with the tactical-limitation function is okay, in the same way that !magic and channeling in combat are okay). I believe there should be something that successfully performs the intended clock function that food tries to perform right now, though I understand why other people do not think that that is necessary and acknowledge that almost no one else would really care much (I said this last bit already).

Ok, than what do you crate think about the potential of the following proposal:

1. Remove chunks and all food cost for any spells, abilities, etc. Only keep perma food and be hungry by ellapsed game time. Food is only there to limit the available time.
2. Add contamination cost to abilities/spells etc. instead of food cost. Remove contamination by exploration or something instead of time. Change contamination mechanics a little bit to make it more comfortable to use (for example make it`s deterministic to allow warnings, make it more transparent, etc.).

The idea is to use existing mechanics in a better way.

I'm astonished at the way you manage to go on at such length without acknowledging that the supposed type of scumming you're talking about is not done by anyone, even though it easily could be (this is a problem that is not solved at all by the mechanics that are supposed to address it), meanwhile the problem people are actually talking about, viz. food, affects nearly every game. We have a link to a morgue file showing over 700 chunks eaten in a game -- this is absurd.

Crawl does not have an interesting food game. It doesn't have a good food game. If anything, it has an especially obnoxious food game. Food does not usefully address any real problem in crawl. It should be gone.

mps wrote:I'm astonished at the way you manage to go on at such length without acknowledging that the supposed type of scumming you're talking about is not done by anyone

crate wrote:I never actually did any pudding farming, and rarely did any polypiling, but the fact that those are possible in Nethack still significantly decreased my personal enjoyment of the game.

The fact that it's not done is irrelevant here; I am arguing from a design perspective. Pudding farming is also not done, from my perspective, yet hurts my enjoyment of Nethack. In the same way, knowing that I can grind for xp in several places in crawl does hurt my enjoyment of crawl.
I've not opposed having to eat things (i.e. pressing buttons, the thing that people actually complain about) being removed from crawl. If you think I have, then you are mistaken. I have acknowledged that the current food clock is not actually successful in its goal. So, as far as I can tell, you actually have not disagreed with anything I said, or if you are doing so you are purposely being obtuse and indirect for no reason other than to be argumentative. If you have a problem with a clock existing, just come out and say so please, instead of trying (and succeeding! I hope you're happy) to annoy me.

edit: Okay I thought I said this but apparently I didn't say it in plain language: the act of having to press buttons to eat food is something that should be removed from crawl (at least, as far as the clock function of food is concerned. Having to consume items to deal with the tactical-limitation function is okay, in the same way that !magic and channeling in combat are okay). I believe there should be something that successfully performs the intended clock function that food tries to perform right now, though I understand why other people do not think that that is necessary and acknowledge that almost no one else would really care much (I said this last bit already).

Whatever your stance on the matter of eating is, it's lost in playing up a problem that doesn't appear to be real in the first place.

edit after additions to crate's previous comment: Okay, good.

I bolded a section of my second large post and added two sentences. Hopefully there is now no more confusion. I'm willing to take the blame here for not being one hundred percent clear.

sanka wrote:Ok, than what do you crate think about the potential of the following proposal:

1. Remove chunks and all food cost for any spells, abilities, etc. Only keep perma food and be hungry by ellapsed game time. Food is only there to limit the available time.
2. Add contamination cost to abilities/spells etc. instead of food cost. Remove contamination by exploration or something instead of time. Change contamination mechanics a little bit to make it more comfortable to use (for example make it`s deterministic to allow warnings, make it more transparent, etc.).

The idea is to use existing mechanics in a better way.

1. Rogue tried that. It was such a hated mechanic that its decendents like nethack added eating corpses to detooth it. It is a bad mechanic.

2. I am not a big fan of contamination. I don't like the "permanent" punishment. I don't like the lack of immediacy. The punishment for casting high level spells (note I don't think there should be a punnishment for casting high level spells) should be temporary and immediate so that it is a tactical considertion not a stategic one.

crate wrote:The fact that it's not done is irrelevant here; I am arguing from a design perspective

My guess is that you have not done any actual design work. I have done some professional game design. The main goal of game design should be to create games that are fun to play. One thing that I learned was that it is impossible to create rules that can not be abused and are still fun because trying to plug all the holes makes for tedious rules. You learn that at some point you just have to say yep some dick is going to abuse this, but you just don't play with those dicks. Luckily crawl is a single player game and you never have to play with those dicks.

crate wrote:The reason to have something pushing the player to actually go around and explore levels is pretty clear, and if you don't see why then DoomRL is a great example of a game that has nothing pushing the player to explore and suffers very clear problems from it.

DoomRL's Nightmare difficulty addresses this to some extent since monsters have a chance to resurrect, and the chance per turn increases as time goes by. If you don't explore, you risk having those cacodemons respawn while other monsters join the fight. It's not perfect, but it's something. This still leaves every other difficulty of course, but I personally find N! the most fun once you get the hang of it.

My guess is that you have not done any actual design work. I have done some professional game design. The main goal of game design should be to create games that are fun to play.

Games rapidly become unfun once you realize good play is unfun; playing badly for the sake of avoiding tedium is almost as frustrating and irritating as engaging in the tedium itself.

Lots of professionally made games are designed poorly, so I'm not particularly impressed by the (claimed) credentials. (okay, to be fair, many may have been designed well, just with an unappealing design target -- e.g. focusing on trying to making it attractive at a superficial level without really caring about the deeper levels)

I believe piety decay is a good reason to move forward and it does not apply to Dg, Xom, Ru and Chei only. Crate described problems with Mu of Xom, for example. Is it optimal to farm Lair while worshiping other gods? If not, probably we don't need anything else, we just need to do something with these special gods and Dg. (If players chooses to stay godless to be able to farm, that's fine IMHO as it is not optimal).

Piety decay is not a good reason to move forward. Sticks are never good reasons to move forward. Your games should not contain sticks. Your games should contain carrots and obstacles. Treasure is a carrot. Treasure is a good reason to move forward. Goals like getting the orb are carrots. The orb is a good reason to move forward. If the game needs to punish me to keep moveing forward it is porly designed. If you must abuse me to play the game then why am I playing the game in the first place??? I should not be punished while playing a game. If you want to put in a goal of beating the clock that is fine, but make it an actual goal and not some in the background punishment for a percieved undesirable style of play that only comes into effect if you choose that style.

Sticks and carrots are often the same, once you think in terms of opportunity costs. The difference is mostly for psychological and PR reasons.

Also, you are missing the point. The game is not punishing you for playing the game: the anti-scumming measures are generally so light that people don't encounter huge OOD spawns in normal games.They are designed to punish people scumming a lot. You can still do so, but that is not normal players one should be concerned about. One should think of how normal players feel when they don't scum, but feel bad/annoyed that the game has this loophole which they are not exploiting.

Of course it is just a psychological difference. That is the point. And yes even if I don't scum that stick is still there. As you said it is the psychology of the matter. The point of the stick is not to hit the ass, but to put duress on the ass to pull the cart forward. There is just no need for sticks in games. You can do better.

The Tavern

But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger

Re: But seriously, remove hunger