Charms change (once again)

[hannobal]

I propose following change to charms:
- They will never time out (except Swiftness, cTele, dDoor and Haste).
- You can have only one charm active. If you cast a new one, it will cancel the previous one.
- If you cast the same charm again, it's cancelled. (costing 0 MP, 100% success, confirmation prompt)

This would mean that you'd have to (almost) never to cast charms outside combat (e.g. keep always shroud up), but you'd have to switch between different charms depending on the situation, which is more interesting gameplay.

- Melee enemies in sight? Cast Shroud
- Centaur appears? Cast rMsl, Shroud ends
- Fight ends, you're hurt - cast Regen, rMsl ends (Regen would stay on, spellpower would affect how much your regen is enhanced instead of duration. Hunger cost should be per regenerated HP)
- After healing, cast rMsl again in case you meet more centaurs and explore onwards
- Hard enemy appears, cast haste, ending rmsl
- Etc

- If this feels too powerful, have charms reserve also max MP equal to their cost while active
- If this feels too weak, make it so that instead of just one spell you can have X spells levels total of charms active based on charms skill.
But i don't think these extra adjustments are needed.

And yes it's mostly a nerf, because you couldn't have dMsl and haste at same time.
Phase Shift could possibly be made a charm/translocation to obey same rule, but that's a minor detail.

Same treatment should possibly affect also transmutations (it would affect just stoneskin and appendage, since you can have just one form active anyway). They should be on separate bin, so you could have 1 charm + 1 transmutation active.

I'm sure details need adjusting and some individual spells would have to be nerfed/powered up because of this, but i'd like to hear thoughts on the main concept?

[kroki]

i dont like it, dont fix whats not broken

[kuniqs]

If you can have only 1 charm at a time, then there's little difference in having them permament.

[dpeg]

Having only one charm makes removes the point of spell selection, I am afraid.

Perhaps we just have to bite it and accept that charms don't belong on spells. It's not as clearcut as it was with divinations, but there is some parallel evidence: (a) the meaning of spell for charms is sometimes not entirely clear; (b) charms can (like divinations and unlike, say, hexes) be cast out of battle.

[reaver]

Perhaps we just have to bite it and accept that charms don't belong on spells.

Seconding this.

I've said before that I like most Charms effects but they do not belong on spells. Most decent perma-buff proposals change Charms to be so different they're only spells in name.

Charms that actually "work" (new Swiftness, Death's Door, etc.) could still be spells, just not "Charms" spells.

We could even make a "Charms god" for the dislocated effects like Ash is the "divination god" (since having a piety cost on Charms would make them not-spammable).

[bel]

Another idea mentioned elsewhere is to make casting a charm reduce your max MP, and let the charm be active always. Casting it again toggles status.

[mps]

Another idea mentioned elsewhere is to make casting a charm reduce your max MP, and let the charm be active always. Casting it again toggles status.

Indeed, this is a pretty standard way to manage buff spells across RPGs of all kinds and it's known to work.

I don't know what to make of the opinion that charms don't make sense as spells, but I would say that removing charms and particularly haste as a spell would probably not be well-received by the existing player base.

[hannobal]

Reasoning behind this idea was that if you can have just one charm active, you have to cast them during battle because another charm you have memorized might be better for the current situation. Which leads to that charms would be cast mostly during battle, like other spells and so be worthy of preserving their spell status.

There was also the alternative that instead of just one charm, you would have some amount of total spell levels that can be active at same time, but it feels complicated so that's why my main proposal is just one charm active at same time (no limits on memorization of course)

[njvack]

I don't know what to make of the opinion that charms don't make sense as spells, but I would say that removing charms and particularly haste as a spell would probably not be well-received by the existing player base.

That doesn't mean it isn't a good idea. Removing Divinations as a spell school caused quite the hue and cry (WHAT DO YOU MEAN NO MORE DETECT MONSTERS?!?) but I think Crawl is better off without.

There are already *two* reliable sources of non-spell Haste.

I think Crawl would be better off without the whole class of spells you want to keep active the whole time*. Making those effects limited in availability (as consumables, through divine sources, or occasionally as item properties) seems like a good choice.

Not all god abilities would need to be active; passive divine buffs would be OK as switching gods is nontrivial.

* OK, so I want Death's Door active all the time but you know what I mean.

[Siegurt]

Ugh, I feel like I've said this about a dozen times now.

So the solution for "Most charms pretty much become a permanant buff if you have the patience for recasting them all the time" isn't to make them actually permanant, and it isn't to remove them (When used not-patience taxingly, they give you interesting choices about what to train for, and how to effectively augment your character in a way that doesn't tax the already-burdened inventory system

How to fix charms is to fix them one buff at a time, and fix them by making them *not optimal to have on all the time* Some charms already have this; haste, swiftness, DDoor.

So four tactics that can be employed to make tedious buff-spamming not a thing:

1. Give it a long term cost or risk of it for repeated use (haste, invis) this means you only want to use the spell when you WANT to use the spell
2. Give it an inherent drawback (swiftness, DDoor, berserk) spamming these types of spells is bad because you are as likely to be caught in the "weak" phase as you are in the "strong" phase, although this is actually the least-effective form of protection (Since you can rest off the drawback, theoretically you could spam DDoor by casting it, exploring a few steps, retreating, resting off the damage, moving up to the next area to explore, casting Door, exploring a few steps, etc. etc. this really is only impractical because of the OOD timer and wandering monsters)
3. Tie the buff to a location, Lightning spire is an example of a "tied to a location" spell, it doesn't move, so spamming it with no creatures in your LOS is just dumb.
4. Making it only useful against creatures in your LOS when the spell is cast, While technically not a buff, metabolic englaciation is an example of "works on creatures in your LOS", inherently it does nothing if there's no creatures to apply it to.

I'd personally like to see most charms be given the 3rd and some the 4th treatment, since I consider those the best types for interestingness.


A generic, but effective implementation of the 3rd tactic would be "When you cast this spell, it creates an aura on all squares in your LOS, as long as you are inside that aura you get X effect, the aura expires after a the spell's duration, and doesn't move" This could be done to most charms spells in a generic and simple fashion. Note that this doesn't prevent you from casting them outside of combat, but they have to be cast *just outside* of combat to be effective, (Giving scrying and antennae their proper benefits) casting them in combat would be just about as effective (perhaps more tactically challenging if you have to run away) and spamming them would be useless

Another implementation is "The effect itself is removed if you take a step" This would be the weaker version of above, I'd place this on a spell that you want to weaken slightly at the same time you want to make it not-spammable (I feel like OA is perhaps a candadate, if it's got the no-move thing, and stoneskin has the 'los' effect, it might create an interesting diversification)

A slightly different example would be for example battlespheres could be placed and immobile a-la lightning spires, and not be mobile. I would consider this an easy and extremely effective way to make them not optimal to spam out of combat. Yet another is the proposed replacement for repel missiles, the "wind wall" spell (Which spams short-lived clouds around your current location which decrease ranged accuracy when fired through)

The fourth tactic: "Effects creatures in LOS when the spell is cast" does put them more into the realm of hexes rather than charms, but I'm not so tied to the thematic division between the two, that I would be opposed to say, making phase shift a "gives all creatures in your LOS a penalty to hit, regardless of their MR or HD" spell but leaving it called a 'charm' Although perhaps calling it a "Vision bending aura that only you can clearly see through" and giving it the first 'LOS aura' treatment would be better for making people not complain about it in the forums.

In any case, I personally think ditching the entire concept of spells that improve your combat abilities is bad, it adds a dimension to play, without which the game becomes more and more single dimensional and choiceless.

Also please note that not all buffs are charms, and not all charms are buffs, and not all charms or buffs exhibit the problem in question (Spammability)

[mps]

Well, people who still play the current version think removing divinations was a good change, mostly. In terms of discoverability of key mechanics, like noise, it was probably a move in the wrong direction. In terms of having traps be a thing in the game, it was a move in the wrong direction. Detection spells are often overpowered in roguelike games, but if the alternative is a situation where optimal play is luring monsters to an extreme degree unheard of in other games, maybe offering the player more information makes sense.

As far as removing haste as a spell, it would reduce the pace of the game in extended, which frankly extended cannot afford. Being able to cast haste 50 times + consumables instead of just consumables makes extended much more bearable. Reasonable alternatives to get that tempo back are what? Trog? Zig-scumming (lol)? Remove a pan rune or two? If you're just talking about removing charms as a spell classification, e.g. going back to enchantments or moving charms to other schools, that sounds great. If you're talking moving a lot of charms effects to consumable/god ability only, that sounds terrible. There's just not a problem there to solve.

[Also, I'd second some of what siegurt is saying, except that there's not actually a buff spamming problem as it stands anyway. The idea of tying spells to an aura/AOE or artifact, say a "totem" or sth, that provides the effect in LOS and can be destroyed makes sense to me, although it's not as good as just keeping the current situation.]

[njvack]

The biggest problem I have with charms, actually, is that they're often fairly low-level spells that your character is better with than without. Learning Stoneskin if you find it is a pretty easy choice -- it's easy to cast, and even at low power, a couple AC is better than no AC. Similarly OA and Phase Shift (well, it's level 5 so not so easy maybe) and Shroud and rMsl. But casting it a lot (before every nontrivial fight is a good idea) is tedious. Having a pre-fight dance for modestly challenging encounters is not great.

I don't mind using consumables or piety before an occasional really tough fight (a unique or branch end) because I simply can't do it every time I see a pack of yellow monsters, and then I need to balance the benefit of getting more use out of my consumable with the risk of getting killed, and that's nice.

Yes, removing haste would change the balance in extended pretty significantly. Stuff changes the balance of parts of crawl all the time. It would ultimately be OK.

[Siegurt]

Well, the reason this thread exists (and the dozen or so like it) is because there is a perception that buff spamming is a problem, the op's unsaid take is that you *can* have buffs up all the time, it is just tedious, so why not simply just have them be permanent. The other side of the coin is that is isn't very interesting to have "I found a spell in a spellbook, and invested some xp (and many charms don't even require much of that to be effective) and now have a permanant buff up" be the case, and having the primary limitation on that be tedium is even worse.

So you either remove the limit, make the limit not tedious, or remove the buffs altogether (personally I think this is throwing the baby out with the bathwater, but that is just my opinion)

[mps]

Well, if "ultimately okay" includes Trog becoming even more powerful relative to other gods, sure. I mean, haste really helps make interesting character types work. If Trog characters and characters in CPA or GDA have the same access to haste as a character in a robe, you've just made the most basic kind of character build even stronger relative to ones that actually use a diverse cross-section of available skills and work well in current crawl. The best advice for book backgrounds these days is go Trog and burn your book.

If what you're actually worried about is "spamming" or "dancing" with charms and you think making their effects permanent is overpowered, then the obvious solution is upthread: make the cost of those charms that don't have a cost like contamination or other crazy stuff like death's door come out of max mp for their duration and put an ability in the a-menu to deactivate them. Notice that this is a straight nerf to current charms, except maybe from a scoring perspective. If what you're actually worried about is spamming spells, then this solves the problem.

Anyway, it's not necessarily true that changing the balance of extended has to be okay. It could always be not okay. I don't see any argument that making extended more of a pain in the ass is going to make things better.

[Berder]

If it's such a huge hassle to buff up before a fight, let's throw away transmutations too! And summonings for that matter - summoning before a fight is too much like buffing before a fight! Hexes too, because is there really that much difference between buffing yourself and debuffing an enemy? Heck let's just remove all spells entirely, since clearly they are too complicated for everybody whose idea of fun is holding tab.

[Sar]

  Code:

For this message the author Berder has received thanks:
duvessa


[njvack]

The other side of the coin is that is isn't very interesting to have "I found a spell in a spellbook, and invested some xp (and many charms don't even require much of that to be effective) and now have a permanant buff up" be the case, and having the primary limitation on that be tedium is even worse.

So you either remove the limit, make the limit not tedious, or remove the buffs altogether (personally I think this is throwing the baby out with the bathwater, but that is just my opinion)

And as you suggested, the right solution will probably be different for different spells. There really aren't that many of these effects.

[Rast]

I understand that maxmp is being overhauled. Perhaps this is a good time to look at making charms cost a certain amount of maxmp to keep active all the time.

The amount does not have to be based on spell level; it should be customized per-charm.

[qwesdf]

I mean, haste really helps make interesting character types work.

What sorts of characters require a 50% buff to their action speed to be viable?

[mps]

In extended, essentially all characters use haste extensively, unless they can't (e.g. formicids). I'll put it this way: If you don't have a haste spell available, it makes lot less sense to run melee characters in light armor with spells for support. There aren't that many other higher level spells that justify the skill investment when you have solid melee and heavy armor is, like, pretty good if you don't want to cast high level spells.

[dpeg]

Perhaps we just have to bite it and accept that charms don't belong on spells.

Seconding this.

I've said before that I like most Charms effects but they do not belong on spells. Most decent perma-buff proposals change Charms to be so different they're only spells in name. Charms that actually "work" (new Swiftness, Death's Door, etc.) could still be spells, just not "Charms" spells. We could even make a "Charms god" for the dislocated effects like Ash is the "divination god" (since having a piety cost on Charms would make them not-spammable).

Sounds all good to me! I should get going with the slow monsters proposal, but after that I'll float a trial balloon on the principal idea.

Berder: Sarcasm doesn't help your point in any way. There's a huge difference between Charms, Summonings and Hexes. Won't explain.

mps: This has nothing to do with Trog. This has nothing to do with balance for extended.

nvjack: Yes, the low level of many utility spells is a problem, that's a good point. (And obviously, a spell school needs low level entries.)

Ugh, I feel like I've said this about a dozen times now.

I am sure it wasn't intended, but that's a very strange way to start a reply.
I'd think that many of the posters in this thread have been around for long enough to witness that the topic has been scrutinised thoroughly. It's not as if the idea to sacrifice the school is a sudden inspiration.

I don't feel like spending too much time here, but some comments on your reply:
(a) Removal is always an option. (A better solution may be possible, but it's a non-trivial cost, and removal will in this case improves the status quo, so it'd be a gain. Note what reaverb wrote, so it's a selective notion of "removal".)
(b) I challenge that the Haste spell has been addressed properly by previous nerfs (of which there are many).
(c) On your 1: Glow does not really impede repeated use of Haste or Invisibility. It does prevent their combined use, but like the OP's suggestion, that only shifts the player's burden to spotting the most relevant spell. (Hint: Haste generally wins.)
(d) On your 2-4: As reaverb said, some spells work (Swiftness, Spire, Englaciation etc.). There's no reason to remove those, and it hasn't been proposed.

I don't know what to make of the opinion that charms don't make sense as spells, but I would say that removing charms and particularly haste as a spell would probably not be well-received by the existing player base.

This is a non-argument as far as Crawl development is concerned. Any removal alienates part of the player base (Hive, Dwarf, Divinations, anything) but the goal is not to make all existing players happy, but to improve the game. That's a subjective notion but roughly, if we think that a new player has a better experience with a change, there is zero reason to shy away from it on grounds of how veterans receive it. Many players actually understand and support this approach, so there is no problem whatsoever in this regard. (It's one of the luxuries of developing a long-running, free game: no obligations.)


Experience has shown that it is very hard to create interesting choices with buff spells beyond what/when to learn them and skill allocation (standard choices for spells, nothing special about Charms here). The question of using them is much more pointed when they're a finite resource. Conveniently, we already have an item category for buffs. (Self-buffing wands are odd in this regard, but that's a separate topic.)

[roctavian]

One thing I've mentioned elsewhere is that if it seems ideal to do something like, "spell stays active permanently with a max MP cost while it's on," isn't that just functionally the same as putting on equipment? It would add more choice to the equipment game if some charms became egos. Cloaks of Rmsl that goes inactive for a few turns when you get hit, gloves of blocking that behave like shroud of golubria, and so on.

[Berder]

There's a huge difference between Charms, Summonings and Hexes. Won't explain.

Don't forget Transmutations. Charms, Transmutations, Summonings, and some Hexes (particularly hexes like Metabolic Englaciation, Mass Confusion, or Silence). They have differences, but they all have one big similarity: they're things you do prior to actually fighting, part of the pre-fight dance. I would include in this category also spells like Freezing Cloud or Poisonous Cloud, which you normally cast once at the start of the fight, and then continue fighting with other weapons/spells. Death Channel, too.

So my point there was, if you want to kill charms because they involve a pre-fight dance, all those other spells fall into the same category. I don't want anything to be removed.

Experience has shown that it is very hard to create interesting choices with buff spells beyond what/when to learn them and skill allocation (standard choices for spells, nothing special about Charms here).

If it's a question of creating interesting choices - having the option of support spells may not be amazingly interesting, but it is more interesting than not having any options. If you want to cast support spells without massive xp investment then you need to be in medium or light armor; that's the choice they create, whether you will take that plate or that crystal plate, or that ring mail, chain mail, or FDA.

If support spells like that did not exist, nearly every melee character would just grab some crystal plate and forget about spells. And that's a reduction of choice, making the game less interesting. That's what mps was talking about.

[basil]

I usually don't want to have Blade Hands and Silence up while exploring. However, summons have that problem to a degree so maybe they should time out quicker with nothing in LOS (except that you'd also need to address "lure a quokka around for summon extension.")

[KoboldLord]

I don't see a problem in principle with spellcasting outside of combat. Charms can be the strategic magic skill, which you invest in for passive bonuses. It would function like training stealth or dodging, which don't give you an ability to activate in combat but rather give you a passive bonus that is always on. Stealth and dodging are generally recognized to be Not a Problem, I believe, so it should at least be possible for passive charms to be acceptable as well.

The really bad problem with charms as the status quo is that you have to keep re-casting them. There's no reason to object to the practice of having them up at all times, because this is just a really obvious choice that players have already made by choosing to invest in charms in the first place. The problem is in constantly re-casting them, because this is a constant trivial chore to ratify a decision that has already been made. If stealth worked this way, your character would break stealth every 100 steps, and you'd have to enter a special command to re-enter sneak mode. Which you would obviously enter every 100 steps, because if you didn't want to be sneaking you wouldn't have trained stealth.

I realize some participants in this discussion hold as an aesthetic ideal that spells should always be activated abilities that are only ever used in combat, but I don't understand why they hold that ideal. Why should spells not be active or passive depending on how best to implement the effect in question? Why straightjacket the design team for what is at most a flavor consideration? Just make the de facto passive spells truly passive, rather than obnoxious-passive.

Passive charms won't create any additional power issues that don't already exist, because almost all charms are already functionally passive. Even if a particular charm is overpowered, the best way to rein it in is to weaken it, not make it more annoying to use. If charms had been implemented as passive magic from the start, would anybody here seriously argue that they should be balanced by having them poke the player with an annoying query during routine travel?

[dpeg]

KL: 1. Spellcasting out of combat has some principal problems. For example, casting success or MP cost matter less than for combat spells (potentially not at all). This is crucial: your hexing or summoning or damage spells crucially depend on casting chance, and running out of MP in battle is a serious issue.

2. Note that Stealth and Dodging are experience sinks achieving a single effect. (Not literally true, as Stealth also informs stabbing damage.) Compare to Charms, which does a load of things.

3. I believe that "strategic" and "magic" does not combine well (in Crawl, that is), a lesson learned from divinations, Alter Self, Fulsome Destillation.

4. This is not really about an aesthetic ideal, it is about repurposing spells to what spells do best in the game: every single effect (damage, buff, enchanting etc.) could in principle be on a consumable, a spell, a god, a card etc. Which effects fit best where is a non-trivial question. Some of us claim that buffs shouldn't be spells, because the standard forces that keep spells in check don't apply.


Berder: The usual disclaimer about removals holds: the complaints about loss of decisions nonewithstanding, it actually often happens that a removal *enables* decisions. I'd reckon that would hold for Charms. Again, bear in mind that interesting effects would survive in other, limited (i.e. better at choice-generating) shapes.
Also, I don't think "don't want anything removed" is a good mindset for design. On top of that, it is not about removing effects, only removing spells. That's a difference (last time I'll point that out.)

[Berder]

Berder: The usual disclaimer about removals holds: the complaints about loss of decisions nonewithstanding, it actually often happens that a removal *enables* decisions. I'd reckon that would hold for Charms. Again, bear in mind that interesting effects would survive in other, limited (i.e. better at choice-generating) shapes.

If you mean there might be a potion of stoneskin or something like that, then first of all, that's another thing everybody's going to have to lug around in already crowded inventories. Inventory management and deciding what to drop is just not fun. Second, every character could use a potion of stoneskin including characters in CPA, so it's removing the strategic decision of whether to wear medium/light armor and train spells. So it's less interesting from that perspective.

Also, I don't think "don't want anything removed" is a good mindset for design.

I meant, I don't want any charms or similar spells removed.

On top of that, it is not about removing effects, only removing spells. That's a difference (last time I'll point that out.)

I didn't say anything about removing effects in my previous post so I don't know why you're saying that like I need to be reminded.


My main question for you: what do you have to say about the fact that removing Charms and similar spells will mean nearly every melee-oriented character would just use heavy armor and ignore spells?


Also, I am concerned for the plight of octopodes and felids. One of their major strong suits is the fact they have no armor penalty to spellcasting, letting them really stock up on support spells. Stoneskin is really essential for a statue form octopode or felid. Without that option and others like it, they would be far weaker, and that would be a problem since they're already weak.

[Wahaha]

Lowering max mp to sustain a charm is a bad idea because lower max mp means nothing to characters that use a melee weapon.

Mp drain while a charm is active could easily fix pretty much all issues associated with charms, but what I gathered from the mp drain thread is that since charms would be weaker due to not being practically free, some people are opposed to this, ignoring the fact that their power could be increased if necessary to compensate for being actually limited in duration. It would even remove the need for secondary drawbacks on spells such as contam from haste and invis, and swiftness could go back to being a simple spell.

In any case it's probably better to redesign charm spells to be used tactically instead of removing them.

By the way, if swiftness was changed to prevent running away from speed 10 monsters at will, it still doesn't "work".

[XuaXua]

I think the "active charms reduce maximum MP" works well, but only in conjunction with "total MP amount consumed by active charms reduces MP regeneration rate accordingly".

[mps]

mps: This has nothing to do with Trog. This has nothing to do with balance for extended.

Oh, I see. It has nothing to do with Trog or extended endgame balance. I guess I was confused. Now that I know it has nothing to do with Trog and nothing to do with extended, this whole thing makes a lot more sense.

Lowering max mp to sustain a charm is a bad idea because lower max mp means nothing to characters that use a melee weapon.

Untrue, but anyway there are various ways to do max mp lowering. It would seem to me that the way to go would be to make the mp reduction depend on the relevant skills and ER in the way that fail rates do now. If the reduction is large enough with low skill and/or heavy armor, you get scarcity of resources which creates choice.

I don't know what to make of the opinion that charms don't make sense as spells, but I would say that removing charms and particularly haste as a spell would probably not be well-received by the existing player base.

This is a non-argument as far as Crawl development is concerned. Any removal alienates part of the player base (Hive, Dwarf, Divinations, anything) but the goal is not to make all existing players happy, but to improve the game. That's a subjective notion but roughly, if we think that a new player has a better experience with a change, there is zero reason to shy away from it on grounds of how veterans receive it. Many players actually understand and support this approach, so there is no problem whatsoever in this regard. (It's one of the luxuries of developing a long-running, free game: no obligations.)

I don't think there's any reasonable measure of what improves the game other than the opinion of experienced players.

[acvar]

It is not that charms are bad in crawl. It is that charms a bad in any system. They fall into the everything is better with me catagory. If you really want to be good at anything you need charms. Who is better at dodging arrows a characther who sinks everthing into dodging or a character who diverts some fo their experience into repel missles? Who is better at sneaking past monsters a character who sinks all their experience into stealth or somebody that sinks a sizable portion into charms for invisibility (hell even most of your experince). Anything you do if you want to be really good at it you need charms. People who only train stealth should be better at stealth then people that train stealth and charms. People that train only dodging should be better at dodging then people that train dodging and charms. People that train to cast fireballs only should be better at casting fireballs then people that train casting fireballs and charms (haste).

[KoboldLord]

KL: 1. Spellcasting out of combat has some principal problems. For example, casting success or MP cost matter less than for combat spells (potentially not at all). This is crucial: your hexing or summoning or damage spells crucially depend on casting chance, and running out of MP in battle is a serious issue.

That really depends on the character. Only direct damage builds, summoning builds, and guardian spirit builds care about actually running out of mp. For the vast majority of spell use in the majority of winning games, mp constraints are no more a serious limit than hunger costs. Possibly they were originally intended to be, but in the actual game we have most spellcasters are well-advised to pick up a weapon so they can ignore that constraint completely.

2. Note that Stealth and Dodging are experience sinks achieving a single effect. (Not literally true, as Stealth also informs stabbing damage.) Compare to Charms, which does a load of things.

Almost all charms improve the ratio of damage inflicted to damage received. Then there's Flight and Control Teleport. Sometimes those charms improve the ratio to a broken degree by applying a ridiculous multiplier or even functional invincibility, but that's a problem with those charms, not necessarily the concept of charms.

4. This is not really about an aesthetic ideal, it is about repurposing spells to what spells do best in the game: every single effect (damage, buff, enchanting etc.) could in principle be on a consumable, a spell, a god, a card etc. Which effects fit best where is a non-trivial question. Some of us claim that buffs shouldn't be spells, because the standard forces that keep spells in check don't apply.

Spells don't seem to particularly need to be held more in check. Zero-magic builds seem to do fine. A crossbow is at least competitive with and is arguably superior to conjurations. If melee builds seem to go for charms an awful lot, perhaps it happens because there are only a couple spell skills that are actually good enough to ever be worth picking up without substantial pre-existing skill investment? There is certainly no plausible in-game scenario where you would be advised to pick up conjurations on a character who didn't start with a book of them.

[hannobal]

In my proposal there is some analogy to conjurations/other spells which i'll try to enlighten with couple of examples:

Imagine a conjurer with OOD, Iron Shot, Freezing Cloud and Bolt of Fire.
If it meets a couple draconians, freezing cloud is a good solution.
Ice Dragon? Bolt of Fire.
Lich? Probably OOD or Iron Shot depending on range.
Orb Guardian? Freezing cloud/OOD/Iron shot depending on range and other factors.

Then we have the new charms/melee dude who has dMsl, Regen, Shroud and Haste.
Player would have to make a choice vs. each enemy which charm is best, like our conjurer dude above.

While exploring, i'd keep dMsl on.
While draconians are at range, i'd cast dMsl (if it wasn't on already), especially if there was an annihilator. I'd switch to Haste if fight starts to look hard, but i'd lose dMsl in the process so it would not be such an easy choice.
Ice Dragon? dMsl.
Lich? Haste, but losing simultaneous dMsl could be nasty because of possible crystal spears.
Single Orb Guardian @Zot5? Probably shroud if i was confident melee'ing it. I'd haste if something else would come in sight. Shroud would be gone, but it would be worth it anyway.
When haste would run out after fight, i could cast regen to heal up, and then continue exploring with dMsl or Shroud.

So you can see player would have to make choices to depending on the situation, and the choices might not always be easy.
Player would not usually have to spam same charm over and over again when it times out, but juggle between different ones.

[The Ferret]

KL: 1. Spellcasting out of combat has some principal problems. For example, casting success or MP cost matter less than for combat spells (potentially not at all). This is crucial: your hexing or summoning or damage spells crucially depend on casting chance, and running out of MP in battle is a serious issue.

You could always just make those things matter. For example:

• In addition to ordinary miscast effects, Charms have a fail% chance of ending in combat. Optional additional nerf: (some?) Charms take multiple turns to cast. (This would require re-examining and possibly tweaking Charms that can already be ended early - Shroud, Ozocubu's Armour, Repel Missiles. That doesn't seem to be a serious obstacle in principle; probably just tweak their inherent end-early probability downwards. Also, "in combat" is subject to interpretation - enemy visible? You take damage?)
• Or, in addition to ordinary miscast effects, Charms maxMP cost scales with spell failure chance, reaching 100% maxMP cost at some arbitrary failure chance.
• Hunger cost: your metabolism is increased proportional to the spell hunger cost as normally calculated.

In theory, even if Charms' maxMP cost is equal to their spell level, it's a nerf when it comes to MP cost - you can't recover that MP after casting them, say by quaffing a !magic. MaxMP cost should probably be greater than spell level for many charms, maybe.

Another issue: how to handle Summoning? It's not broken per se and I think it's reasonable to say that summoning during the pre-combat dance is a fine mechanic, but it's also reasonable to think that Summoning is like Charms in that you always want summons active for perfect play. It's less clear how to handle Summoning under the latter point of view, but here are some ideas:

• Summons act kind of like the proposed Charms change - you get an "aura of summon ice beast" or whatever, and your ally is automatically summoned when you see an enemy. Not desireable for many reasons (stealth and stabs, how do you summon more than one thing?).
• Tie all summons to a specific area, as with Siegurt's earlier list. When summoning, you are given a targetable area of effect, and the summoned ally cannot move outside that AOE. ("its astral connection is weak"). Optional: AOE scales with spellpower. Optional: this is clearly a nerf to summons; to re-buff and create tactically interesting choices, summons are smite-targeted. Optional of the optional: summons automatically receive an attack order (as with [t][a]) for an enemy within their LOS, chosen by... some method. I'm a fan of the latter two.)

[dpeg]

hannobal: That's a good example but you have similar decisions when everything's back to consumables and, as you can see, the decision isn't too hard: you could script it! Like you say, in the lich case, Haste probably trumps Deflect, unless you opt for safety and take Haste from another source, so you can Deflect per spell.

I'd also like to underline acvar's posting. He makes a really good point. I believe the problem he singles out goes away when moving buffs to consumables (or rather, then it shifts to item generation, which is much easier to balance).

mps: So you're wrong again, a second time, and *another* time, but that's no surprise from the Berlin Interpretation Conspiracy Man.

[lessens]

If Charms are on the chopping block I'd suggest just removing Deflect Missiles and leaving Repel Missiles (and the rest of Charms) mostly unchanged, because the former is much more of a problem than the latter. Repel Missiles costs very little in terms of xp to get the failure% to a reasonable level even in GDA, and at zero skill it will repel an average of 2 missiles before failing so it offers a minimal benefit to players who cast it at zero skill, so there's very little incentive for players to remove armour to cast it. This does not apply to Deflect Missiles, which requires much more xp to cast (especially in heavy armour) and gives players a strong incentive to remove their armour. If rMsl remained in the game you could even duplicate most of the effects of the higher-level version by increasing the spellpower cap and maybe even tying spellpower to the repel rate, the latter of which would add a further disincentive for players to cast it at zero skill. Full disclosure, I want to remove Deflect Missiles anyway for reason unrelated to the topic.

[lessens]

People who only train stealth should be better at stealth then people that train stealth and charms. People that train only dodging should be better at dodging then people that train dodging and charms. People that train to cast fireballs only should be better at casting fireballs then people that train casting fireballs and charms (haste).

I don't think this is self-evident at all. There's no reason per se that a given mechanic should only be governed by a single skill or that there should be no interaction between skills. It's also worth noting that what you're saying doesn't just apply to Charms, it applies to any activated ability that comes with an xp cost, namely any evocation activated from the 'a' screen and a number of god abilities that supplement various roles and require invoc.

[Sandman25]

I don't see why it is bad to have a reason to stay in light armour. DMsl is fine IMHO, it helps EV-based characters not be strictly inferior to AC-based characters. I don't mind if Haste as spell will be removed, as side effect it will make DMsl more expensive, currently its cost is not that high because you train Charms for Haste anyway.

[Hirsch I]

If Charms are on the chopping block I'd suggest just removing Deflect Missiles and leaving Repel Missiles (and the rest of Charms) mostly unchanged, because the former is much more of a problem than the latter. Repel Missiles costs very little in terms of xp to get the failure% to a reasonable level even in GDA, and at zero skill it will repel an average of 2 missiles before failing so it offers a minimal benefit to players who cast it at zero skill, so there's very little incentive for players to remove armour to cast it. This does not apply to Deflect Missiles, which requires much more xp to cast (especially in heavy armour) and gives players a strong incentive to remove their armour. If rMsl remained in the game you could even duplicate most of the effects of the higher-level version by increasing the spellpower cap and maybe even tying spellpower to the repel rate, the latter of which would add a further disincentive for players to cast it at zero skill. Full disclosure, I want to remove Deflect Missiles anyway for reason unrelated to the topic.

the great problem with repel missiles is that it is useful for everyone, all the time, can be casted by everyone, all the time, costs virtually nothing, except making you want to ctrl-q-yes after an hour.

[lessens]

I don't think "useful to everyone" is a good criticism of a spell. Blink, swiftness and flight are also not problems just because you reflexively learn them if you find them.

That said I wouldn't be bothered at all by raising rMsl to level 4 Charms/Air and removing dMsl, it's just that people are very, very opposed to people potentially taking off armour to cast it being higher level would encourage that.

[KoboldLord]

the great problem with repel missiles is that it is useful for everyone, all the time, can be casted by everyone, all the time, costs virtually nothing, except making you want to ctrl-q-yes after an hour.

I would argue that the problem with Repel Missiles is that it grants a preposterously huge bonus against the majority of really dangerous attacks with almost zero investment. It doesn't even scale meaningfully with further charms/air magic investment. If it granted zero bonus at zero investment and scaled up in effectiveness with spell power like the majority of spells that are not buffs, then it would not be such a problem.

[mps]

mps: So you're wrong again, a second time, and *another* time, but that's no surprise from the Berlin Interpretation Conspiracy Man.

Your methods of argument are childish. When people criticize your designs, your responses range from non sequiturs to grasping at any defense other people raise. What arguments you bother to make consistently rely on mistaken claims of fallacious reasoning, dismissal of opinions that aren't your own, usually in a tone of patronizing condescension (which is unjustifiable in view of the way you talk about these issues publicly), and outright personal attacks if the first two don't work. If you're looking for substantive engagement or even to defend your design choices, I don't see how posting this way helps your cause.

[Hirsch I]

can this please not become another boring festival of personal attacks, kids? this is actually a very good discussion, that can help improve the game in a meaningful way.
lets not have our egos ruin it again. relax for a little while.

[bcadren]

I would argue that the problem with Repel Missiles is that it grants a preposterously huge bonus against the majority of really dangerous attacks with almost zero investment. It doesn't even scale meaningfully with further charms/air magic investment. If it granted zero bonus at zero investment and scaled up in effectiveness with spell power like the majority of spells that are not buffs, then it would not be such a problem.

I believe you've pointed out a problem with -a lot- of the buff spells. They don't scale with spellpower as well as Conjurations or Hexes. I mean; I'd say all spells that don't have much use for spellpower are flawed in that way; be it Shadow Creatures or Haste. You can cast it; you're done, no need to train further. The marriage of success rate; MP cost and Spell Power relative to total spellpower created by the Spell Level mechanic was balanced for attack spells; more than buffs; but we can make it work for buffs; we just need to make the buffs have a portion that is -meaningfully- affected by the difference. Ozocubu's Armour is a great example of a buff spell that uses spellpower well; giving an extra point of AC for every 12.5 levels of ice/charms spellpower up all the way to the max. It has obvious breakpoints to the player; but not that disruptive of them.

I will say though; I'd greatly approve of going the 'Repel Missiles' route on all charms. Make Ozo's Armour last 18 + 2d(Spellpower) HITS instead of that many aut. Having the buffs just stay on instead of being cast before every (dangerous) battle would cut out a lot of tedium; especially on the extended characters that branch into buffs late for Regen to survive hell, Stoneskin for a little more AC on top of the CPA they trained armour for; etc. etc.

[edgefigaro]

I like charms, they cost a turn, super unstealthy chars can't always retreat around corners when they encounter a serious threat while stealthies can. I dislike gripes centering around optimal play. The game does not encourage you to play optimally. Most encounters are not serious threats and don't need to be geared up for like they are. Generally, charms behaving more like invis/haste where contamination discourages constant use is nice, as is the swiftness penalty. More charms could be retooled to provide some kind of tradeoff/clock/I shouldn't want to keep this always active but it is great to have around.

[johlstei]

I just think it would be really easy to code a change to charms right now to make them permanent. That's what they are, just now there are annoying prompts, as koboldlord said with the stealth key. That's an easy improvement. Other, proper solutions are Difficult and require time and effort to think up and code, and that's fine, they should be thought out. It's still a good idea to, right now, fix the interface hassles of the current system.

I'm pretty sure it could be lua scripted even but I'm not at all that tempted to put in the time to do it, it'd be much easier to do it in the game itself.

[dpeg]

johlstein: That is absolutely true, and doing that would fix an actual problem in very simple fashion, so it has my sympathies. However, It would not address another problem that those spells have (everybody can and should benefit from them, disregarding the usual investment) and, in fact, your solution would rather stress this problem more.

At this point there is no clearcut solution: one can choose between (1) the simple interface solution, or (2) the simple design solution (do away with the spells), or (3) keep status quo while trying to look for a better solution than (1) and (2).

So in the end it's a judgement call between different weightings of perceived problems (annoying interface, balance).

[kuniqs]

If charms are permament and contaminate you then they still have a duration - but it is your problem of keeping count now.

[Berder]

johlstein: That is absolutely true, and doing that would fix an actual problem in very simple fashion, so it has my sympathies. However, It would not address another problem that those spells have (everybody can and should benefit from them, disregarding the usual investment) and, in fact, your solution would rather stress this problem more.

But that's not true. If you're in crystal plate armor, you have quite a hard time getting even basic charms online. If you're in plate armor with low int, you can do it, but the xp investment is high. If you are with Trog, you can't get any charms. There are tradeoffs if you want charms - no Trog, light/medium armor or high xp investment. Not every character gets charms. Fact: a high proportion of wins don't use any spells at all, even non-Trog wins.

[dpeg]

Berder: Yes, and that's why the situation is not as urgent as it was with Detect Creatures, say. If I understand johlstei correctly, there are no provision against changing dress (and if you do, which is the prominent permanent-recasting suggestion, then things become complicated again).

Nobody is saying that Charms win the game for you, but there quite easy pickings, so any properly casting character will want to put some Charms on top, and even otherwise non-casting character may use some. There are some spells like this (I had non-casting characters who still learned Apportation, Flight, cTele although I didn't use those very often) and it doesn't have to be a problem. My claim is just that another really simple solution improves balance and creates more chances than it removes. (For example, I don't think that if Crawl had no Charms in the first place, then body armour choice would be trivial.)