Recite and weird spoileryness

[Siegurt]

So Zin's recite has some odd rules, and it results in some oddities, and a little information leak:

New rules (Well since last november, not that new any longer, really) are If something could be recited against in more than one category, it's only able to be effected by one type of recite, in the order of Chaotic, Impure, Unholy, Heretic. This is roughly in order of effectiveness, so in theory you should only be able to hit something with the strongest recitation available, which seems like it would simplify the whole recitation business (It used to be you'd have to choose every time, and you'd just kinda have to know that chaotic recitation was stronger than impure recitation)

Now however, that means you can have several of the *same* creature in LOS and only be able to effect a subset of them.

For example, if I have 3 liches in LOS, and I want to recite against them, but one has the spell set with banishment in it, that lich can *only* be hit with chaotic recitation, and can't be hit with 'impure', and the other two can *only* be hit with impure.

That's very awkward, I have three of the same critters, and they're in separate recite-groups because one happens to have 'banishment' in it's spell list.

The second problem is that now I can *know* immediately with no actions (I can cancel the recite) that that a lich does, or does not have banishment (or an ogre mage for that matter), without a single spell being cast or action taken, which strikes me as information leak that I shouldn't have (Or if I should have it, it should just be announced on things coming into LOS, like was done for shapeshifters)

The third problem I have is with the actual effects chosen, I don't particularly have a problem with them being random, or with the more powerful effects being reserved for higher invocation levels, It's that some of the "better" effects are worse, and some of the effects which would be most useful, aren't applicable in situations where they'd be useful.

For example, ignite chaos *can be* devistating, if something is very chaotic, in the case of our lich above, he's only a little chaotic, so ignite chaos, a top-end chaotic effect, actually does less damage than smite, which is the *bottom* end effect (Smite isn't effected by the degree of chaotic/unholy/impure/heretical you are), in this case I actually get a less damaging effect by training more invocation, and am forced to use the less damaging type of recitation (holy word would do more damage with the same invocation to a lich) because anti-chaos effects are more powerful in general (if not in this particular instance) -- This amounts to a complaint and I'm not *terribly* worried about it, I just don't like it when "training more skill results in less effect" which can happen in odd cases with recitation.

Also currently the effects are *mostly* deterministic (there's a range of 5 for each tier, and a random factor of 6, so there's some chance of you getting a different effect, but mostly at a given invocation, with a given piety you're going to be doing one of two effects (mostly it's a coinflip between the two effects at a given tier for a given invocation))

I suggest the following:

1. Zin's prayer is usable against the thing you are *most* degree of, rather than the thing that comes first in that list (That order is good for ties though) This will eliminate some of the spoilery stuff (Since e.x. liches are more impure than they are chaotic even if they can cast banish) It will make it more likely that the effect will be applicable to the creature it's being applied to, and more likely that you'll have the recitation you expect to have available for a creature.
2. Zin takes into account things that can actually effect the critters when selecting the effects, so he's more likely to corona an invisible thing, he's more likely to antimagic a spellcaster (and isn't likely to select those effects when they aren't appilcable) I'd also like it if the antimagic and mute effects were available to more than just 'heretics' (Since a lot of undead, chaotic and demonic things are nasty spellcasters as well)
3. At high power recitation could be able to stack multiple "poor" effects instead of one super nasty effect (So as an example at really high invocations, you might saltify (a Single very high effect) a critter, or with the same power effect you might Smite, blind, and confuse him (3 middle/lows), or holy word and bleed him (one fairly high and one low).)
4. Effects are selected randomly with it being very unlikely, but possible, to get a high powered effect with "just enough power to possibly effect a creature" and it also being possible to select a lower-powered effect with high power (And if there's enough leftover power, it gets to pick again.)


This would result in recitation being a more useable, but not really more powerful ability, and it'd reduce some of both it's spoilery-ness and it's deterministic-ness.

This would be fairly easy to implement (I thought about writing a patch for it, but I thought I should get some feedback first)

[and into]

That all sounds good.

My only (slight) worry would be that "situational appropriateness" isn't given *too* much weight. If a lich goes invisible before you begin reciting, it would kind of suck if the recite were too strongly weighted toward doing nothing but hitting the lich with corona. I'm sure all this can be worked out with proper balancing, of course, it is just a question of getting the numbers right.

The only addition I'd make is perhaps giving some kind of visual feedback about how effective/powerful your recite is. For a lot of other invocations your failure chance gives a pretty good indication (at least ballpark) of how strong the ability is, as well. Not so with recite. Having a high degree of randomness is its thing, but maybe it wouldn't be too much detail if on the a screen it listed a bit of info, such as the strongest effects you can produce with recite given your current level of invocations. I don't know the numbers behind recite so maybe that is an inappropriate sort of info to give, but some feedback would be helpful and make Zin a lot more user friendly, IMO.

[TheDefiniteArticle]

Your post is pretty unclear to me, so possibly I am misunderstanding something somewhere, but this:

If something could be recited against in more than one category, it's only able to be effected by one type of recite, in the order of Chaotic, Impure, Unholy, Heretic.

Sounds crazy to me. If something fits into more than one category, why can't I pick the one I want to use?

[Leafsnail]

Would it be possible to have a single unified "recite" ability? It could still affect monsters as it does now (impure monsters are hit with the current Ablutions effects, unholy are hit by current Anathema effects, etc), only there wouldn't have to be a weird behind-the-scenes choice about which set of effects should be used. I don't think this would represent a major buff to the ability (especially if you combine it with proposal 1 here) and it would remove the unintuitive and spoilery aspects.

[dpeg]

Leafsnail: I think you are correct. However, it shouldn't be forgotten how the current system came around: players had a choice between the (up to four) types of incantation. Having those types made a lot of sense back then; it does less so (or perhaps not all) now.

[Kate]

Patch for affecting things based on whatever category they are the highest degree of sounds good to me. There is already some limited situational effectiveness (non-spellcasters can't be hit by antimagic, non-bleeding/rotting monsters can't be hit by bleeding/rotting, etc). This should probably remain simple, restricting mute to priests and actual_spellcasters and maybe weighting ignite chaos depending on chaotic degree sounds okay. Less deterministic effect choice would also be good. I don't think multiple effects per monster is necessary, nor allowing the stronger heretic effects to work on demons (it's intended to be weaker against them compared to the other categories, and is already pretty great at high skill generally).

Removing book choice entirely and just affecting each visible monster with whichever category has the highest degree also sounds okay, I considered that when making the last set of changes but it's more effort to implement.

[reaver]

For example, if I have 3 liches in LOS, and I want to recite against them, but one has the spell set with banishment in it, that lich can *only* be hit with chaotic recitation, and can't be hit with 'impure', and the other two can *only* be hit with impure.

Source code comments indicate this behavior is actually a bug, because which categories of Zin Recite a monster falls under should be consistent no matter what spell set is chosen. This is done by making sure all spellsets ping the same for the purposes of Zin Recite, but this has never worked correctly for a few monsters, normally those with Banishment. This bug has just become more dramatic with the recent recite changes.

[Viashino_wizard]

Removing book choice entirely and just affecting each visible monster with whichever category has the highest degree also sounds okay, I considered that when making the last set of changes but it's more effort to implement.

This seems like a good way to go. Recite already has a fairly lengthy cooldown, and not having to choose between books makes the ability less fiddly in general. Which book actually gets recited (as in the flavor text passages that your character quotes) could be determined randomly or, if someone felt like putting in the time, be based on the most numerous/most powerful category of targets in sight.

[Siegurt]

Patch for affecting things based on whatever category they are the highest degree of sounds good to me. There is already some limited situational effectiveness (non-spellcasters can't be hit by antimagic, non-bleeding/rotting monsters can't be hit by bleeding/rotting, etc). This should probably remain simple, restricting mute to priests and actual_spellcasters and maybe weighting ignite chaos depending on chaotic degree sounds okay. Less deterministic effect choice would also be good. I don't think multiple effects per monster is necessary, nor allowing the stronger heretic effects to work on demons (it's intended to be weaker against them compared to the other categories, and is already pretty great at high skill generally).

Removing book choice entirely and just affecting each visible monster with whichever category has the highest degree also sounds okay, I considered that when making the last set of changes but it's more effort to implement.

It's not currently really significantly weaker against demons than any other category:

  Code:

Against:              Recitation:      0+           5+             10+             15+
Intelligent           Heretic(Minor)  Daze         Daze/Confuse   Confuse         Confuse/Para
Priests & Evil        Heretic(Major)  Bleed/Smite  Blind(33%)/    Blind/Para/Mute Mad/Dumb
                                                    (-mag|Corona)
Corp Undead & Necro   Impure          Bleed/Smite  Corona/(rot)   Corona/Holy     Saltify
Incorp undead&demon   Unholy          Daze/Confuse Corona/Confuse Corona/Holy     Saltify
Chaotic               Chaotic         Bleed/Smite  Corona/Smite   Corona/Ignite   Saltify



In particular it's nearly the same as corporeal undead, with the exception of not being able to do damage at the low end (of course I'd take confuse over minor damage at the low end myself most days)

The only real "deficit" against demons is that there's only one 'degree' of susceptibility (Which lowers the spellpower at which holy word is cast vs corporeal undead who might have several 'degrees', or possibly only 1) All the other effects don't care about degree at all.

With regards to being able to reach the higher-end effects, Currently to get to a 10+ effect you need to beat HD+d(6) on your power by at least 10, where power is 1-15 for invocation, and 0-15 for spellpower (So 1-30).

So against a HD19 creature (A lich), 2:3 times you'll be getting a 5+ effect(Corona for a lich, they can't rot), and 1:3 times you'll get a 10+ effect (Corona or Holy word)

Which means, At 27 invocation and 200 piety, against a lich, 1 in 6 times you'll get holy word against them, the other 5 times you'll get corona. That's a rather large amount of XP to have a 5 in 6 chance of using a level 1 spell's effect against a HD19 creature (for admittedly no MP and just the 'breath' cost) (Not to mention the flat 1:8 chance of it just doing nothing at all)

Obviously, it being a low level ability, it *shouldn't* be super effective against high level creatures, but it's such an odd distribution with weird breakpoints, I think we could make it more interesting. At the simplest just reducing the frequency of corona's use in the table would make it more interesting.

Another thing I thought was that the duration could be based on your invocation. It has a 50% chance of working each turn on each creature, (Which is why it has a 1:8 chance of doing nothing to a given creature) I think if it started at 2 turns (1:4 chance of doing nothing for each creature) and could extend up to say, 5 turns at max invocations It'd give it an interesting ramp up.

Maybe we need to get creative and come up with some additional middle-level effects (How about slow as a 10+ effect?)

[Siegurt]

Your post is pretty unclear to me, so possibly I am misunderstanding something somewhere, but this:

If something could be recited against in more than one category, it's only able to be effected by one type of recite, in the order of Chaotic, Impure, Unholy, Heretic.

Sounds crazy to me. If something fits into more than one category, why can't I pick the one I want to use?

This used to be the case, It's now not:

commit ffde107490c85f74278fcaaf4153d12eeaba2056
Author: Chris Campbell <chriscampbell89@gmail.com>
Date: Wed Nov 6 08:54:52 2013 +0000

Make individual monsters only ever be affected by one recite type

Only use the first type that they are eligible for, in order: chaos,
impure, unholy, heretic. Also reword the prompt to be slightly clearer.

Rather than return to the old "Pick which category you'll recite" method (Which was actually *really* terrible at the time, because you couldn't even tell without spoiler-info what things in your LOS would be affected by which recitation) I'm suggesting a compromise where rather than the "first category in an arbitrary list" it's the "category into which it fits the most strongly"

Further suggestions in the thread have been "You actually get to recite against everything, even if they fit into different categories" which would be a minor bump in power level (Currently you have to pick which subset of things in LOS can be recited against, based on which category you choose to recite)

[Siegurt]

Hm, another Zin-recite effect idea:

Abjuration (in particularly of chaotic or demonic summons) (Only applicable critters which would otherwise be recited against and, of course, which are summoned)

[Leafsnail]

As I said I don't think that would be a massive boost over the old system (which I did play before). In most cases you're just reciting against one type of enemy anyway, since enemies usually come in themed packs/branches. I can understand not doing it if it's harder to implement though.

[Kate]

Not sure about scaling duration on top of all the existing scaling, Recite could do with being less complicated generally. This is why just letting it recite to everything in LOS as appropriate would probably be good even if it increases the ability's power slightly.

[Siegurt]

So I've written a patch which implements the changes discussed here;

https://crawl.develz.org/mantis/view.php?id=8490