The last charms reform post ever.

[Hurricos]

To fix charms forever:

1) No charms automatically expire; all charms stay active until triggered to deactivate (a la DMsl / RMsl) or deactivated (UI: deactivate through spell I inerface).
> Infusion can trigger a power-dependent number of times before deactivating, stays active otherwise
> Shroud of Golubria is permanent until it unravels due to hit. Gains slight power dependence; 0 power -> 5 dmg avg for split break/block chance, 50 power -> 15 dmg avg for split break/block chance (curr. 10)
> Song of Slaying is permanent (causing noise 7 at your location every turn as though quietly chanting) until it accrues a bonus (and begins to become louder), at which timer starts and ends based on current power-dependent scale
> Swiftness is on until a monster notices you, at which point it begins ticking down - cannot be deactivated after timer has started, no indication that it has started so as to avoid letting players 'notice' invisible monsters
> Spectral weapon is permanent until it disappears or weapon is switched.
> Ozocubu's armour is permanent until melted or destroyed by damage.
> Regeneration is permanent, rate of healing (and thus hungering) scaling logarithmically with power. While HP is non-full, MP is non-empty and you are regenerating, remove 1 MP for every 5HP regenerated (div_rand_round).
> Excruciating Wounds works as Infusion. Replace on-cast weapon scream with whimper every time opponent writhes in pain.
> Iskenderun's works as Spectral weapon, consuming (dPower > d50 ? 1 : 0) MP extra per cast to fire its beam
> Ring of Flames is permanent.
> Death's Door moved out of charms to L8 Necromancy (seeing as Stasis does not exist any more for Borgy's, and Kiku gives no extra HP, this is not a buff compared to old DD)
> Move Cigotuvi's to Charms / Necromancy - continue current functionality, but armour will never automatically slough off unless it is greater than a power-related value or you take damage / block (or you are disenchanted by !canc)
> Move Portal Projectile to Charms / Trloc - continue current functionality. N.b. that accuracy boost depends on spellpower.

2) You can "take 20" to cast all charms when no monsters around you notice you; unmitigated encumbrance lowers spellpower as well as non-take-20 casting chance
> Removes scummy take-off-plate-armour-to-cast-RMsl behavior
> All interactions are based on current spellpower - if you want your current charms to be more reliable NOW, quaff !brill.

3) You can designate particular charms to be automatically take-20 recast when out of combat through I interface for spells
> Autorecasting occurs during auto-explore

4) Having a charm activated reduces your MaxMP by that charm's level. Casting charms does not cost MP, but rather lowers MaxMP and rescales current MP using div_rand_round. The end of a charm, by unravelling / canc / manual deactivation does not rescale MP.

All problems with having to recast constantly before battles disappear. All scummy behavior of removing armour to cast RMsl disappears. Spamming Regen - gone. Manually recasting spells - gone (as it's no longer inefficient to automatically recast spells, or necessary to manually recast them out of combat to achieve HOP). Going to a faraway place to cast Excruc to avoid noise - gone. Spamming battlesphere to charge - gone. Recasting Spectral - gone.

The problem was that in order to achieve good play you had to make boring grindy decisions outside of battle (spamming Regen, casting infus before all battles etc). Now you do them once outside of battle or as you need to inside battle (e.g. dropping PProj so you can penet, or dropping RoF when an ice fiend gets next to you).

[Sar]

Devs: also we don't want to make Charms into slotless equipment (even though we did with RMsl/DMsl)
Hurricos: *makes Charms into slotless equipment*

[Hurricos]

Devs: also we don't want to make Charms into slotless equipment (even though we did with RMsl/DMsl)
Hurricos: *makes Charms into slotless equipment*

What's the point of NOT making charms into slotless equipment if they're already just slotless equipment plus spammy functionality? What exactly is good about having to recast repeatedly outside of or right before combat? Isn't the entire point of charms reform to fix the fact that you have to recast things to the point at which it becomes stupid?

The way I suggest makes charms slotless equipment that is mildly contraindicated with usual equipment by lowering the power based on the failure rate. Instead of gambling repeatedly to get a charm to function you have to actually sacrifice the charm's effectiveness if you can't cast it.

This also completely fixes the RMsl / DMsl issue: before, a player would cast the spell and deflect missiles as they liked, but they needed to be able to cast the spell. Now, they can remove armour and cast the spell, leading to degenerate behaviour to achieve optimal play. After, this degenerate behaviour would contribute exactly nothing - the idea would be that at zero power RMsl would be almost immediately deactivated upon being tested - and the spell does not need to be recast manually.

[Hurricos]

Actually, you know what else is slotless equipment?

Mutations.

But in order to get mutations, you need to risk getting bad mutations, just like you need to (under this framework) spend (risk) skill points actually training charms. That, and in the endgame, mutation power falls off - just like charms' power fall off if not actually worked on.

Finally, as you've failed to notice, under a framework which I expressed previously of reduced MaxMP, charms are less 'slotless equipment' and more 'new items for existing slots that don't need to be spammily invoked every time you want to use them.'

[Sar]

I think what devs ideally want to do is to make Charms something you actually cast in combat, and even then only occasionally, so they give them drawbacks, like slowness for Ozo (which doesn't really change things.

Hm, I wonder if Ozo could use a small delay to actions too. Nothing drastic, maybe 0.1-0.2 auts to every action, not just movement.

Muts are slotless equipment, yes, but it's much harder to get a good set.

[Hurricos]

I think what devs ideally want to do is to make Charms something you actually cast in combat, and even then only occasionally, so they give them drawbacks, like slowness for Ozo (which doesn't really change things.)

N.b., w.r.t. when you should be casting charms, an equivalent to my solution would be to not be able to cast charms outside of combat. When charm deactivation triggers are hit, you can either choose to recast or do nothing. It's up to you whether they're worth anything.

The problem, then, is figuring out when 'outside of combat' is.

[dpeg]

What's the point of NOT making charms into slotless equipment if they're already just slotless equipment plus spammy functionality?

Assuming that description of Charm spells is true as written (it certainly holds for Repel/Deflect Missiles and for Shroud of Golubria), then your approach keeps the "slotless equipment" and does away with the "spammy functionality". That's one way to proceed, and development has actually done this for Repel/Deflect.

However, your underlying assumption seems to be that Charm spells have to be like that, or at least that they're fine as slotless equipment. I don't think there's concensus about this within the devteam (after all, the new mechanic for Repel Missiles got implemented), but I believe that the current trend is to not do it like this. I can really only speak for myself, so here's why I don't like Charm spells working like 0.19 Repel Missiles:

• In my opinion, spells are something to be cast frequently. For example, strategic spells of old have been purged: Alter Self, Portal, Selective Amnesia.
• Permanent effects should go on items (or on gods). I could definitely imagine a version of Crawl where all those buffs sit on armour pieces.

What exactly is good about having to recast repeatedly outside of or right before combat?

Nothing! We agree on that. The discrepancy is how we try to fix the problem. Your approach is to move Charms away from other spells (cast often, mostly in battle), towards items (change rarely, and out of battle). My goal is to modify Charms such that they work more like other spells, but that's the content of Psieye's thread.

All problems with having to recast constantly before battles disappear. All scummy behavior of removing armour to cast RMsl disappears. Spamming Regen - gone. Manually recasting spells - gone (as it's no longer inefficient to automatically recast spells, or necessary to manually recast them out of combat to achieve HOP). Going to a faraway place to cast Excruc to avoid noise - gone. Spamming battlesphere to charge - gone. Recasting Spectral - gone.

It's correct that your proposal achieves these goals.
These are why I do not like even so:

• Players can pile up buffs. This means that eventually, everyone will run around with many of these buffs active. You might argue that choice of body armour, stats (how much Int) etc. play a role, but I am not so sure: you will start with the low-level Charms you can cast right now, and increase the set as you go. I think this would uniformise characters, rather than differentiating them.
• The proposal is very much about streamlining the current system. I'd rather go for changing buff spells such that they're more relevant to tactical situations.

It definitely took a while to come up and write down this. I cannot really see this go into Crawl (you probably guessed so). Thanks for the contribution nonetheless!

[Hurricos]

What exactly is good about having to recast repeatedly outside of or right before combat?

Nothing! We agree on that. The discrepancy is how we try to fix the problem. Your approach is to move Charms away from other spells (cast often, mostly in battle), towards items (change rarely, and out of battle).

The idea is that if a charm is going to last any duration it's going to be advantageous to cast it before battle. This is why the current setup leads to garbage HOP, and any duration-based setup that doesn't vary based on some combat-related figure like tension is going to lead to garbage HOP. Thus, permanency.

Players can pile up buffs. This means that eventually, everyone will run around with many of these buffs active. You might argue that choice of body armour, stats (how much Int) etc. play a role, but I am not so sure: you will start with the low-level Charms you can cast right now, and increase the set as you go. I think this would uniformise characters, rather than differentiating them.

Except MaxMP and spell-slots (can be adjusted precisely to) limit the number of buffs you can memorize, having to recast during battle after triggers have broken them will continue to limit MP, thus redoubling the effects of aptitudes and spellpower (assuming instead of div_rand_round rescaling is replaced by just reducing current MP as well as MaxMP) - and of course, because of the casting-rate-to-power idea, which will greatly diversify how Charms are used.

The other thing that must be mentioned is that I believe that the idea that charms should be used often in combat stems evidently stems from the notion that combat actually changes significantly over the spans of time that charms have ever been made to work. Unfortunately, this is hardly true. If you're in a branch, there will be a certain average level of threat from ranged attacks, a certain average level of value of having a Pain branded weapon, a certain amount of utility in having Spectral weapon (worth shit if you're getting damned by some hellions). The particular movements of combat change too quickly for Charms to be pertinent in particular like spells are - like how you'll switch from Bolt of Fire to Bolt of Cold on a Balrog, for example; you'll Fireball the Enchantress, instead of Spearing her.

Charms should be instant spells that affect combat now, like Translocations - is this a correct reflection of your point of view?

[dpeg]

The idea is that if a charm is going to last any duration it's going to be advantageous to cast it before battle. This is why the current setup leads to garbage HOP, and any duration-based setup that doesn't vary based on some combat-related figure like tension is going to lead to garbage HOP. Thus, permanency.

I'm unhappy with the jargon ("garbage HOP"), but I challenge both claims. What you really mean: "I don't see a way to avoid this behaviour", and that's fine. I think some of us do.

[...]the idea that charms should be used often in combat stems evidently stems from the notion that combat actually changes significantly over the spans of time that charms have ever been made to work. Unfortunately, this is hardly true. If you're in a branch, there will be a certain average level of threat from ranged attacks, a certain average level of value of having a Pain branded weapon, a certain amount of utility in having Spectral weapon. The particular movements of combat change too quickly for Charms to be pertinent in particular like spells are.

Charms share with all other spells the fact that they come in spellbooks, have to memorised (sharing spell slots) and cast. If buffs really were that different to other spells, then I'd support scrapping the whole school, adding some of them as armour item egos.

Charms should be instant spells that affect combat now, like Translocations - is this a correct reflection of your point of view?

Not instant, but they affect combat. As I see it, Charms should provide buffs, i.e. some support for a time, with an attached drawback (buffs from consumables don't need the drawback, because they have built-in drawback of being limited). If we cannot come up with relevant drawbacks, then Charms (as I see them) make a poor spell school.

(The differentiation between spell schools is not very clear: Translocations/Air/Fire are thematic, whereas Charms/Summonings/Hexes are mechanic. I don't think this is a problem.)

[goodcoolguy]

The way to make charms "tactical" is not to force you to cast them in combat. This isn't actually what happens most of the time anyway. In the vast majority of cases, you have time or can make time to cast your duration spells before engagement -- this is true of charm and non-charm duration spells alike, including summons, death channel, etc. If you want to address the interface issue with charms, you need to look at all arbitrarily recastable duration spells.

Here's the right model guys: Arbitrarily recastable duration spells cost max mp, they are cancellable, they do not time out, but they can be broken by combat actions like rmsl and shroud. If they're broken, you cannot no longer cancel or recast them and their max mp cost is not refunded until they automatically recast after a cooldown.

As for appropriate combat breaking actions, dispelling breath attacks already do this. Other sources could be figured out. The tactical issue is not casting, which is not a real choice for existing charms, but the possibility of getting your spells broken. The decision is whether or not to disengage, which is more interesting than whether or not to cast/recast a charm.

[dpeg]

I won't go into details, but just one quick comment:

Here's the right model guys

No idea if you get away with this rhetorical style in real life, but it takes away from whatever merits your contributions have. This also holds if the style is meant tongue in cheek.

[duvessa]

I won't go into details, but just one quick comment:

Here's the right model guys

No idea if you get away with this rhetorical style in real life, but it takes away from whatever merits your contributions have. This also holds if the style is meant tongue in cheek.

I think he was mocking Hurricos, who said pretty much the same thing in the OP.

[gammafunk]

So someone fill me in, did this really end up being the last charms reform post ever?

[kimnosuk]

what is take 20?

[archaeo]

what is take 20?

I believe this is a term from Dungeons and Dragons, short for "take 20 minutes." The idea is that, if you are sufficiently skilled, you do not have to roll the dice to see if you can accomplish something, like disarming a trap, as long as you spend 20 minutes on the problem.

Hurricos means that you could cast any charm as long as no monsters around? I assume this is balanced by spellpower, but it is hard to tell.

[dpeg]

I won't go into details, but just one quick comment:[...]

I think he was mocking Hurricos, who said pretty much the same thing in the OP.

Thanks, I missed that. Sorry about my tone then.

[goodcoolguy]

Well, I guess it's time you laid those details on us then.

[dowan]

what is take 20?

I believe this is a term from Dungeons and Dragons, short for "take 20 minutes." The idea is that, if you are sufficiently skilled, you do not have to roll the dice to see if you can accomplish something, like disarming a trap, as long as you spend 20 minutes on the problem.

Hurricos means that you could cast any charm as long as no monsters around? I assume this is balanced by spellpower, but it is hard to tell.

Take 20 means you take a roll of 20, because there's no active danger. This means you don't have to roll the dice over and over to see if you finally picked the lock, in dungeons and dragons. At least, that's how it worked in the PC D&D games that had the mechanic. Maybe the Pen and paper version also included a time penalty (which makes sense).

[Hurricos]

The way to make charms "tactical" is not to force you to cast them in combat. This isn't actually what happens most of the time anyway. In the vast majority of cases, you have time or can make time to cast your duration spells before engagement -- this is true of charm and non-charm duration spells alike, including summons, death channel, etc. If you want to address the interface issue with charms, you need to look at all arbitrarily recastable duration spells.

Here's the right model guys: Arbitrarily recastable duration spells cost max mp, they are cancellable, they do not time out, but they can be broken by combat actions like rmsl and shroud. If they're broken, you cannot no longer cancel or recast them and their max mp cost is not refunded until they automatically recast after a cooldown.

As for appropriate combat breaking actions, dispelling breath attacks already do this. Other sources could be figured out. The tactical issue is not casting, which is not a real choice for existing charms, but the possibility of getting your spells broken. The decision is whether or not to disengage, which is more interesting than whether or not to cast/recast a charm.

This is a mix of mockery and concise summary. Not sure if I like it, but then I posted in CYC. Speaking of that, I'm both glad that I posted in CYC and mortified that devs are actually taking posts in CYC with anything more than an entire desolation of salt. But then maybe I'm lost in a sick joke because I'm in CYC.

Assuming the seriousness of this dialogue, lemme just try to rephrase what I meant in very short terms: Because charms last, people will use them before combat, so to get rid of most of the spam before combat, make you not have to actively use them before combat; when combat depletes them, make it a tactical decision to reapply them or leave. Basically, make charms MP. For most situations it was equivalent to making them instant without the interface spam. Just a proposal.

I'm unhappy with the jargon ("garbage HOP"), but I challenge both claims. What you really mean: "I don't see a way to avoid this behaviour", and that's fine. I think some of us do.

I get the sentiment; I just hate HOP more viscerally because I spend all of my time on Crawl purposefully avoiding it. I cite my ever-spiraling winrate for evidence.

Charms share with all other spells the fact that they come in spellbooks, have to memorised (sharing spell slots) and cast. If buffs really were that different to other spells, then I'd support scrapping the whole school, adding some of them as armour item egos.

Charms should be instant spells that affect combat now, like Translocations - is this a correct reflection of your point of view?

Not instant, but they affect combat. As I see it, Charms should provide buffs, i.e. some support for a time, with an attached drawback (buffs from consumables don't need the drawback, because they have built-in drawback of being limited). If we cannot come up with relevant drawbacks, then Charms (as I see them) make a poor spell school.

What the problematic charms don't seem to share is instantaneity. I can cast a buff NOW that affects my combat slightly in the next 50 turns, and that makes little sense to me from the perspective of someone who thinks a spell decision is between BoF, BoC and Fball depending on the particular allocation of enemies and their resistances. It's a more meaningful choice if it's made now, acts strongly now, and fades soon.

Some spells already have drawbacks that are just fine - Infusion and Portal Projectile have on-hit MP costs, which are basically the streamlined equivalent to casting a short-lived charm every turn, and which is why I'd argue they really oughtn't need to be renewed before combat (it's spammy). But Ozocubu's Armour threatens to become a HOP deathtrap as long as hydras exist.

[Psieye]

It might be time to bring your ideas to my thread in GDD, they wouldn't feel out of place with the other posts there.