Remove hunger costs from casting

[and into]

This is possibly a radical proposal (maybe?) but I think the game would be improved by it. While it probably makes things less likely, I think that removing hunger from spell casting could be the occasion for a few serious changes.

1.) Casting spells does not incur any hunger cost.
2.) Even "minor" miscasts should do things to your character more consistently. Not extreme things, but, you know, something.
3.) Even with miscast protection (from gods), miscasts should potentially cause quite a bit more glow than they currently do.

The idea is that you do not need to worry about hunger, but the spell success rate is made somewhat more meaningful. It shouldn't insta-kill you (the way some really unlucky miscasts of poorly prepared, high level spells can). But stuff like draining 3d4 mana or so could be quite worrisome. I've always felt that miscasts were largely irrelevant, so long as you weren't trying to cast very high-level spells at ridiculous failure rates. If the miscasts are actually going to become interesting, it probably shouldn't be the case that divine miscast protection completely negates the issue. The amount of glow from miscasts while under divine protection could be scaled up.


A few concerns. It is possible that making miscasts into an actual thing that happens to characters more than once in a blue moon will mean that casting spells in armor is too strongly penalized, whereas I think the balance here is fairly good in current Crawl, for the most part. Also, swapping in wizardry might be too strongly incentivized (and it is already annoying, as it is). So, the following adjustments might be a good idea:

4.) Armor skill has a more significant impact on improving your spell casting success.

Obviously we shouldn't go back to how it was back in the 0.4/0.5 days, but letting armor negate the spell casting penalties more than is currently the case could, with miscasts mattering more, keep heavy armor casting at roughly the same level of appeal as it is currently.

5.) Wizardry items need an "adjustment" period before they start improving your spell casting, a la gourmand.

"Your wizardry items are not yet attuned to your magic." "Your wizardry items are attuned to your magic and are now improving your spell success."

6.) Amount of glow is not well indicated. It should be better indicated.

Regardless of the other changes this is probably just a good idea. How much glow-inducing stuff you can do before you start actually glowing is a bit spoilery, and since it can permanently (or semi-permanently) affect your character, it should probably get a more nuanced display. You don't need to numerize it, but something like glow, glow+, glow++ of gray/yellow/red colors (giving 9 gradations), perhaps with a final dark red glow! (10th), would be sufficient.

[savageorange]

I think it is well worth trying. A few points:

* How will spriggans fit into this (and to a lesser extent, high hunger species like Troll and Centaur)? Hunger management is definitely a distinguishing factor in playing a Spriggan caster currently. Are there ways to maintain their distinctness? This would leave only Berserking as a hunger pressure for Sp.
* Working out the details of 2) and 3) is fairly non trivial IMO and should be done mainly via the dev wiki.
* Would a 'gradient' display (using grey, yellow, red in different proportions between characters) be okay for glow? I'd prefer that over a system where, AFAICS, it would cycle grey: glow, glow+, glow++, yellow: glow, glow+, glow++, red: glow, glow+, glow++, because the 'glow pips go up, then changes color and go down' is a little odd.

[bel]

I like (5) for wizardry. This can be done independently of the main point of the proposal, why not?

Slightly off-topic but related. I have never understood why failing evocation does not count as a "miscast". It should have glow too. Can remove hunger from evocations as well.

[and into]

I think it is well worth trying. A few points:

* How will spriggans fit into this (and to a lesser extent, high hunger species like Troll and Centaur)? Hunger management is definitely a distinguishing factor in playing a Spriggan caster currently. Are there ways to maintain their distinctness? This would leave only Berserking as a hunger pressure for Sp.
* Working out the details of 2) and 3) is fairly non trivial IMO and should be done mainly via the dev wiki.
* Would a 'gradient' display (using grey, yellow, red in different proportions between characters) be okay for glow? I'd prefer that over a system where, AFAICS, it would cycle grey: glow, glow+, glow++, yellow: glow, glow+, glow++, red: glow, glow+, glow++, because the 'glow pips go up, then changes color and go down' is a little odd.

Some good points. Yes, it is easy to say, "make miscasts matter," but actually doing that in a balanced way is difficult. I'd probably start by keeping the miscasts that are spell-school specific more or less as they are, but then add a few "general miscast effects" that are less extreme, but much more likely to happen. Mana drain came to mind immediately; that plus glow (which is already in effect a generalized miscast effect across spell schools) might be enough.

As for the glow indicator—yeah, that's a good point. But it might be hard to have 10 different colors, or 3 colors with sufficiently different intensities. Perhaps display as glow(1) to glow(10), with grey-->yellow-->orange-->red colors, as that would be the most unambiguously clear.

Honestly, I hope that food is mostly phased out of Crawl over time. In all my years of playing, I died from lack of food on a spriggan exactly once*, and this was long before guaranteed food vaults (and I also think food spawn rate was somewhat lower/rarer overall, IIRC).

Granted, spriggan isn't my most-played species (that prob goes to demonspawn), but I've played a lot of them. "Train a little extra spell casting than you normally would, maybe" isn't a very interesting conduct for Sp book starts, particularly since Sp has a really good spell casting aptitude and super-slow metabolism anyway. I mean, there is currently an item in the game (staff of energy) that lets you completely bypass spell casting hunger, and it was so underpowered (even for Sp) compared to enhancer staves, that it was repeatedly buffed over subsequent versions, such that it also now lets you channel while being wielded. Spriggans have enough stuff going on that is actually interesting that you could entirely remove their weird food restrictions (not just spell casting hunger) and they would still 100% have their own niche in the game and be a distinct species.

*Note: I actually died from enemies, because I was diving super fast and just running around, trying to cover as much space as I could, in order to maximize my chances of finding a ration. Eventually got cut off and surrounded, with no consumables left, and got owned. So I'm not just counting actual starving deaths, but *any* deaths that were caused by food shortage.

[linear]

If food were removed from the game, then ramping up miscast effects would be a good element to balance spellcasting. As long as food is in the game, hunger is an important cost of spells. IMHO the staff of energy has far too dramatic of an effect and should be eliminated.

[Siegurt]

Well, but hunger costs and miscasts aren't similar enough in how they effect your character.

Hunger puts the breaks on casting spells successfully, miscasts puts the breaks on trying to cast spells with a low failure rate too often.

Unless we added a very small amount of glow for successful spellcasting, which was negated by spellcasting+int, the two systems aren't really comprable, they don't do the same thing.

However then we'd still need some sort of 'glow reducing' long term resource that we could use to negate that, which I suppose could just be adding more cancellation to the game, but then we have a weird situation where a tactical resource becomes strategic, of course I rarely use cancellation tactically (mostly because it's so rare and the situations in which it's useful are also rare)

However I fail to see how using a plentiful cancellation to reduce your glow, so you can keep casting high level spells is an improvement over doing the same with food.

So then why do we need breaks on repeated casting of spells (particularly ones that are high level relative to you)? I'm not sure exactly, I guess it's because mana is too readily available? It doesn't seem that way to me, although I train in a balanced way does anyone train in an unbalanced way where not having some sort of drawback (other than MP use) to casting lots of high level spells would be some sort of min-maxing advantage?

I guess the thing that I'd want to examine is what does having a food cost for spells actually do for the game currently?

[Berder]

I think removing hunger in exchange for more severe miscast effects would be a good exchange. But I don't think miscasts should cause more glow, too. Glow and bad mutations are just too much of a pain in the neck. It would encourage leaving the fight after every miscast, to rest it off so you don't get mutated.

[Rast]

Reminder that there already is an effect that applies to every miscast in addition to the formal "miscast effects":

You waste your turn and some mana.

[duvessa]

I don't think removing spell hunger needs any compensation at all, it has so little effect in the first place. Yeah you eat permafood more often but you're still not going to run out.

[Quazifuji]

I hope this isn't derailing the thread too much, but it's related to the topic of spell hunger and has been bugging me for a while: the general consensus is that spell hunger is largely insignificant at the strategic level, and mostly matters as a tactical limitation. But isn't mana already there as a tactical limitation on using spells? What design purpose is spell hunger currently serving that mana isn't?

Both primarily limit the number of spells you can cast in a single fight, both are much bigger issues for characters who cast a lot of spells but pretty insignificant for characters who just cast a few charms or translocations occasionally, etc. The numbers involved in each are different, but they seem to both be serving the same purpose right now. Why do we need both?

[Sandman25]

Without hunger it's possible to rush to level 9 spell and spam it (especially with Vehumer and wizardry). With hunger you get near starving after 2-4 casts (difference between satiated and near starving is only 1078 satiation points, level 9 spell costs 1000).

[dpeg]

and into: I don't think that spell hunger does nothing. While I read those claim from time to time, it certainly affects my games.

Coming up with better (minor) miscasts is a good idea, but that's quite a bit of work.

I would have liked to go in the other direction: make spell hunger more relevant by making food matter. The most extreme solution is to remove chunks (only food is permafood). I think this can be done, but then the onus is on balancing, of course.... perpetual tweaking.

I never liked the staff of energy (neither in its old nor its current form) because it is a tool to override a gameplay barrier (spell hunger before, now also MP cap -- I am not a fan of channeling in general). So in my world, the existence of that staff should not be hold as an argument against spell hunger.

Interesting write-up, as usual. Many thanks!

[Sprucery]

I would have liked to go in the other direction: make spell hunger more relevant by making food matter. The most extreme solution is to remove chunks (only food is permafood). I think this can be done, but then the onus is on balancing, of course.... perpetual tweaking.

What if chunks rotted away really fast, say a dozen turns? That way you could still make the decision to either eat some permafood or try to kill something for chunks, but you would have to eat permafood much more frequently than now.

[duvessa]

Without hunger it's possible to rush to level 9 spell and spam it (especially with Vehumer and wizardry). With hunger you get near starving after 2-4 casts (difference between satiated and near starving is only 1078 satiation points, level 9 spell costs 1000).

Why do you need to cast a level 9 spell 2-4 times in one encounter? Spell hunger doesn't do nothing, I just don't think it does very much, either.

The motivation for removing spell and ability hunger is that as long as hunger costs try to be a tactical impact, Crawl won't have a working food clock. Which is a big deal because Crawl doesn't have another working clock (piety decay is closest but it has the major problem of never being able to kill you). I don't think it needs to be replaced by anything because it's not a strategic cost, and the alleged tactical role of spell hunger - preventing you from casting too many spells close together - is already the entire purpose of MP. If MP isn't accomplishing that, surely it would be better to fix MP than have two mechanics try to do the same thing.

[Sandman25]

Why do you need to cast a level 9 spell 2-4 times in one encounter?

Do you use my version of crawl? Otherwise I don't see how you can one-shot all monsters with level 9 spell.

[Berder]

I'm opposed to anything that is a net nerf to magic. I'm also opposed to anything that makes food a real concern in a normal game, because dying due to running out of consumables is simply the worst way to die, tactically inevitable, like running out of healing on a DD. If anything would make you just want to ctrl-q yes in disgust, running out of food would be it. So, if you only want to remove food from magic as a preamble to tightening food for everybody, that would not be good.

I think replacing food costs with more miscast effects would be good. Miscast effects are rarely a concern at all, and I think they could be potentially interesting.

Why do you need to cast a level 9 spell 2-4 times in one encounter?

Lots of individual enemies won't die to a single level 9 cast, you may be attacked from multiple directions, or the noise may attract things you have to keep killing.

[duvessa]

well I assume that if you have a level 9 spell you also have some not-level-9 spells that you can use

[Sandman25]

Going for level 9 spell is suboptimal for most characters, going for level 9 spell to cast it just once in every encounter is stupid for all characters.

[Berder]

well I assume that if you have a level 9 spell you also have some not-level-9 spells that you can use

The level 9 is generally a lot more efficient with MP than your lower level damage spells. An exception might be made for certain single target spells like OOD/LCS/Iron shot, which can match or exceed level 9 efficiency against a single target, but are much less good against multiple enemies.

[tabstorm]

I honestly think the game would be better if the food clock was just a clock and did not interact with abilities or spells in any way. In all the games I have played, spell hunger has not done anything to me even when I play 0-melee conjurers, aside from making me eat the entire dungeon, which really disrupts game flow. The same goes for invocation hunger, rod hunger, and so on. You could say "You can just edit your .rc to autoeat and autobutcher!" but I think this is a little unsatisfying. You could just nerf invocations, conjurations, etc. by removing chunks and not touching hunger costs, or reducing their hunger cost. Still, I think investing XP towards the biggest hunger-inducers like high level spells is generally agreed to be suboptimal, so I do not think there is really an issue with them becoming too powerful from removing spell hunger.

Spell hunger aside, there are some other obstacles to having food not interact with abilities, like Elyvilon and Sif (in theory), and undead races' awkward food interactions.

[duvessa]

ely doesn't really have the food problem now that pacification costs piety instead of giving it. channeling, sure

[tabstorm]

I think Sif/energy channel is more of a long-term "pressure" on your food supply. I've never had an issue with being too starving to channel mid-battle and needing to do it.

[Sprucery]

Rod hunger has affected some of my games, when I've found an early rod.

[Lasty]

The original proposal here is a nerf to magic in general -- it's harder and more dangerous to cast spells -- but I don't see anything in the thread to suggest that magic needs a nerf. Removing casting hunger has an effect on magic, but it is not a buff in a similar axis or to a similar degree as the nerf in the original proposal.

[mps]

Indeed, the OP takes an unnecessarily equivocal position. There's no need to reach some kind of compromise re: spell hunger by nerfing magic in another direction. Along the magic vs. heavy armor axis, the balance currently heavily favors the armor side. The existence of magic is little more than a nod to tradition from the perspective of optimal play.

At the end of the day, the food system has a slight impact tactically yet it has a large cost in terms of complexity, UI burden, and mental burden (I don't know about anyone else, but I don't like being constantly told that I'm hungry in a game when I'm hungry irl... ya, I know, you have an rc for that...). I think the near unanimity of experienced players on this point says all there is to say about it. Spell hunger is soup, people.

[and into]

Well, but hunger costs and miscasts aren't similar enough in how they effect your character.

Hunger puts the breaks on casting spells successfully, miscasts puts the breaks on trying to cast spells with a low failure rate too often.

Unless we added a very small amount of glow for successful spellcasting, which was negated by spellcasting+int, the two systems aren't really comprable, they don't do the same thing.

For spell casting, the biggest issue with food costs is, "making sure this spell doesn't put me from somewhere just above hungry to Very Hungry / Near Starving, in only a few casts," because sometimes bolts miss or do low damage, and you don't want to be needing to eat stuff in the situations where you want to be casting Bolt of Fire. How do you defend against this possibility? By training more spell casting and raising intelligence. How would you defend against the possibility of miscast effects that actually matter? By training more of the skills that lower failure chance (including spell power), and by raising intelligence. Strategically, in practice, "spell success" and "hunger" are quite similar—more similar than your argument implies. I think these proposed changes would only very slightly affect ~~optimal~~ (whatever that is) skill allocation for book backgrounds.

Because the skills that increase spell success also increase spell power, I think these changes would only bring greater clarity and directness to strategic decisions. When you want more spells, you train spell casting for the slots. When you want power/success, you train the relevant magic skills. This is already what people do, except for the occasional odd case where you throw a couple of extra levels into spell casting because the hunger cost for one of your often used spells is too high. I don't think that training spell casting for hunger-related reasons very often presents an interesting decision, so why not remove that aspect and retain the aspects of these strategic decisions that tend to matter the most?

The original proposal here is a nerf to magic in general -- it's harder and more dangerous to cast spells -- but I don't see anything in the thread to suggest that magic needs a nerf. Removing casting hunger has an effect on magic, but it is not a buff in a similar axis or to a similar degree as the nerf in the original proposal.

(Also somewhat in reply to dpeg & mps):

But would somewhat—there is a lot of flexibility in "somewhat" obviously—harsher miscast penalties make magic more interesting? If this is a nerf to magic that is also interesting in terms of game play, perhaps it can be a door to other changes that would end some longstanding, balance-related compromises that aren't interesting, nor intuitive—such as the fact that it is rarely a good idea to train magic schools very far beyond the point where success rate is acceptable for your spells (due to the steepness of diminishing returns for spell power), the various limits and caps on max MP values, etc.?

[Lasty]

I already removed the limits and caps on MP.

But to your broader point, I'm not sure whether spell hunger serves any real purpose, but if it doesn't it's probably best to just remove it and see how that goes.

[Blobbo]

Getting to starving during a difficult encounter when you have to cast multiple level 9 conjurations or LCS is a tough gotcha for new players. You simply don't feel the effect in simple encounters, and when you do it's in a difficult situation and there's no easy way to resolve it without completely retreating (consuming the big foods takes 30 aut which is lethal in a high danger situation - again this time cost is not clear to new players either FWIW - not really clear that pizza/jerky have tactical applications).

[Siegurt]

A secondary point is that removing the need to eat in combat (The only other application I can think of is berserk, which is very similar to spells in the way it uses hunger) would allow the removal of tactical-food items (there's no need for a 'fast eating food' if you don't every need to eat in combat)

[duvessa]

A secondary point is that removing the need to eat in combat (The only other application I can think of is berserk, which is very similar to spells in the way it uses hunger) would allow the removal of tactical-food items (there's no need for a 'fast eating food' if you don't every need to eat in combat)

Even if you didn't remove that need, the fast foods are so common that you won't really ever run out...so you could make all foods fast and have basically the same situation.

[Sandman25]

If you want to make hunger relevant, change all food to 10 turns (not auts) and make it interruptible. No more eating during combat.

[Sprucery]

Or you could just make the fast foods much less common. If you only found a few pieces of fast food during the game, you would have to think when to use them.

[Quazifuji]

Without hunger it's possible to rush to level 9 spell and spam it (especially with Vehumer and wizardry). With hunger you get near starving after 2-4 casts (difference between satiated and near starving is only 1078 satiation points, level 9 spell costs 1000).

So the only difference (on the tactical scale) between spell hunger and MP is that spellhunger more strongly rewards training the spellcasting skill for characters who cast a lot of high level spells.

Is this the only significance of spell hunger right now? That seems to be the general leaning I'm gathering from this thread. The strategic cost seems to be universally agreed to be insignificant for non-spriggans, and considered insignificant even for spriggans by some. The tactical cost is largely redundant with mana, except for the way it scales. So is that it? Spell hunger's sole purpose in the game right now is to add an additional incentive to train spellcasting? If that's the case, I think there are two main questions that need to be answered:

1. Is the extra incentive to train spellcasting necessary? Are spell levels and a small universal spell failure decrease not enough?

2. If spellcasting does need the extra incentive, can we add a different incentive that's more interesting and flavorful and less annoying than spell hunger to do so? Heck, maybe there's even a way to do something with mana costs to pull it off, mana already serves the same purpose in every other way.

[Blobbo]

Seems like the tactical solution is to reduce the base amount of MP per level and increase the bonus MP per spellcasting. So you'd need more spellcasting if you want to spam more spells, rather than a larger stomach.

[savageorange]

^ If so, initial levels of Spellcasting training would also need to be increased. TBH, a lot of caster starts have fairly crappy MP already.

[Sar]

I think that was already done.

[savageorange]

I think that was already done.

?

Whether increasing initial Spellcasting training levels was already done or not, I don't see why that would impact whether it needed to be (re) done if we were to

reduce the base amount of MP per level and increase the bonus MP per spellcasting

Or were you referring to blobbo's suggestion itself (reducing base amount of MP and increasing bonus mp per SPc level?) when you say that was already done?

[johlstei]

and into: I don't think that spell hunger does nothing. While I read those claim from time to time, it certainly affects my games.

Coming up with better (minor) miscasts is a good idea, but that's quite a bit of work.

I would have liked to go in the other direction: make spell hunger more relevant by making food matter. The most extreme solution is to remove chunks (only food is permafood). I think this can be done, but then the onus is on balancing, of course.... perpetual tweaking.

I never liked the staff of energy (neither in its old nor its current form) because it is a tool to override a gameplay barrier (spell hunger before, now also MP cap -- I am not a fan of channeling in general). So in my world, the existence of that staff should not be hold as an argument against spell hunger.

Interesting write-up, as usual. Many thanks!

The concern I have is that food tries to be both a strategic and tactical limit, but in order to enable the "caster" playstyle, there has to be enough food that you can generally cast spells throughout the game, which vastly limits how strong the strategic pressure is on melee characters. If there's enough food to support a casting playstyle, primarily-melee characters are going to have a lot of extra food. If there's so little food that melee characters feel pressured by it, then casting as one's primary offense becomes pretty tenuous. Maybe this is good, no opinion from me, but it's not how things work now, and if it happens it should at least be an intended change rather than a side effect of something else. It would be very hard to adjust the amount of food in a way that has the same effect on both types of characters.

Basically, I don't like that "tactical limitation on spells/abilities" and "strategic clock pushing one forward throughout the game" are the same mechanic, even if both type of limitations are good for the game. It's a shame they can't really be tuned independent of each other. There's another universe where crawl has "hunger" and "thirst", for strategic and tactical limitations, with food being "goldified" and water limiting spells and letting you burst with perma-drinks. That's pretty silly as stated but I think it'd be nice if the two mechanics were split in some way, so that the devs can have two knobs to turn as they please rather than one.

[Sar]

@savageorange: I mean increasing the influence of Spc/Invo on MP and reducing MP per level bonuses was already done, I think. Have you played trunk recently?

[Quazifuji]

Seems like the tactical solution is to reduce the base amount of MP per level and increase the bonus MP per spellcasting. So you'd need more spellcasting if you want to spam more spells, rather than a larger stomach.

This has the side effect of allowing evocations or invocations to serve as a substitute for spellcasting if you don't need the spell levels. I'm not sure if this is necessarily a bad thing (maybe it's even a good thing), but it seems a significant enough side effect that it's worth discussing.

[savageorange]

@savageorange: I mean increasing the influence of Spc/Invo on MP and reducing MP per level bonuses was already done, I think. Have you played trunk recently?

I always play trunk. Does this mean I should actually train spellcasting more, for a change ?

(still, it doesn't effect my point AFAICs, since starting characters haven't had any choice in their skilling yet.)

[Sar]

Maybe. I just remember having a bit less MP than I would have normally with similar Spellcasting training, maybe ten less in midgame.

[archaeo]

I generally agree that removing hunger costs from casting is a fantastic idea, better still if it isn't accompanied by a new limit, even though I think there are certainly devs in favor of such a limit.

Basically, I don't like that "tactical limitation on spells/abilities" and "strategic clock pushing one forward throughout the game" are the same mechanic, even if both type of limitations are good for the game. It's a shame they can't really be tuned independent of each other. There's another universe where crawl has "hunger" and "thirst", for strategic and tactical limitations, with food being "goldified" and water limiting spells and letting you burst with perma-drinks. That's pretty silly as stated but I think it'd be nice if the two mechanics were split in some way, so that the devs can have two knobs to turn as they please rather than one.

Are both types of limitations good for the game?

First, we have to accept that both types of hunger are more or less useless right now, insofar as one honestly needs both terrible play and terrible luck for hunger to ever matter. If you want them to matter, I think separating them into "hunger" and "thirst," or however you want to conceptualize these limits, is one of the few ways to go.

However, I don't think the game derives any benefit from either. Tactical limitations all being linked to a single meter is a blunt instrument, when said limitations could be better tied to the mechanic they are limiting (for example, glow with spellcasting, or exh for Trog, though I think the former is unnecessary and the latter would have to be changed without tactical hunger, maybe an xp timer??). Meanwhile, the idea that you could "tune" the strategic meter seems unlikely to me. What's the right level? Should good players die, occasionally, because there was no way to survive the Dungeon with the amount of food in the game? Should the game dribble out food just enough so that you're constantly near starving? At what point does it become significantly more suboptimal to autoexplore?

edit: I feel like I should add that I know this has been divisive for some, and I don't mean to be; I'm honestly curious about my questions, and they're not just rhetorical. I'd be interested to hear a hearty defense of hunger in Crawl, for both tactical and strategic purposes. I see them as vestigial, and if they're not, I'd love to be corrected.

[dowan]

EDIT: Warning! Post contains toxic levels of snark! Open spoiler at your own risk!

Spoiler: show

If only there were some other way to limit spellcasting beside hunger and glow. Some kind of recharging resource, so as to be tactically limiting... I guess I'd call it... magic points. It could just be abbreviated to MP. I'd put a display for it right under the HP display.

Summary for the easily offended and humorless: Spellcasting is already limited by MP, so there's no need for a separate limit.

[johlstei]

I think the strategic one is almost certainly good for the game, even if it never affects anyone playing the game in any way that could be considered "normal". Squashing optimal and tedious behavior is a lot easier when you don't have to consider the space of strategies that assume unbounded turncount. Compare DoomRL - it is, in some sense, optimal to stand in one place for thousands of turns, waiting for each of the monsters to walk into LOS.

That said, I think it's extremely unfortunate that it's tied together with the tactical limit for spells, and that (strategic) eating requires manual player intervention when it really never requires any thinking for a non-spellcaster.

As far as the tactical limit on spells/abilities/rods goes, I'm not a huge fan myself. There is something there in that you can decide to subside on chunks, or decide to use a permanent resource instead to be able to "burst" for longer. This might well be an interesting decision in some cases. I don't think it is great at the moment but that's probably something that could be tweaked and made more impactful, if it were independent of the strategic clock.

[archaeo]

I think the strategic one is probably good for the game, even if it never affects anyone playing the game in any way that could be considered "normal". Squashing optimal and tedious behavior is a lot easier when you don't have to consider the space of strategies that assume unbounded turncount. Compare DoomRL - it is in some sense optimal to stand in one place for thousands of turns, waiting for each of the monsters to walk into LOS.

It seems like it should be possible to tweak the AI to avoid this. For example, the longer you spend on a given level, the more likely the monsters in it become aware of you. Maybe the smarter ones would set up ambushes out of your LOS? Or what if waiting in one place for a period longer than it takes to heal increases the likelihood that an ambush spawns?

Suffice it to say that I think that are more interesting solutions than hunger timers, with the proviso that I honestly don't think I'm as clever as any of the devs, and respect the fact that retaining a loose hunger clock is a hell of a lot easier than excising it and looking for a bunch of different solutions for each problem that arises.

As for tactical hunger, it seems likely to me that if you remove the hunger clock, you could re-conceptualize chunks and permafood as an alternate "energy source," should it be desirable to have a non-MP source of "energy."

[dowan]

EDIT: Toxic sarcasm and snark contained within! User was warned for this post!

Spoiler: show

You could go the ranged route, and have some kind of ammo for spells that never will run out in practice. Then you can get your warm and fuzzy feeling that you've (for some reason) limited the player from casting infinite spells over the course of the game, just like you can't shoot infinite arrows. That way we can all keep pretending that these things are strategically limited, while taking some annoying and unclear mechanics out of the game.

[wheals]

You could go the ranged route, and have some kind of ammo for spells that never will run out in practice. Then you can get your warm and fuzzy feeling that you've (for some reason) limited the player from casting infinite spells over the course of the game, just like you can't shoot infinite arrows. That way we can all keep pretending that these things are strategically limited, while taking some annoying and unclear mechanics out of the game.

This obviously isn't serious, but I'd just like to point out that the joke you're making is exactly the problem with casting hunger cost.

[Tenaya]

If only there were some other way to limit spellcasting beside hunger and glow. Some kind of recharging resource, so as to be tactically limiting... I guess I'd call it... magic points. It could just be abbreviated to MP. I'd put a display for it right under the HP display.

How about MP regenerates with experience or with exploration?

[Siegurt]

The problem IMHO isn't the natural regeneration of MP, it's the sources of channeling combined with the ability of players to min-max when they have such a source.

Easy solution: Remove hunger cost *and* all sources of channeling (Of course this sacrifices a god, and some fun items, and possibly the ability to reliably do zigs with spells)
Harder solution: Remove hunger cost, make sources of channeling limited in some other fashion which makes it non-optimal to min-max (Perhaps limit channeling to a percentage of max MP or be proportional to spellcasting+int or some such?)
Easiest solution: Do nothing (Just for the sake of completeness)