Simplify/Rework Shroud of Golubria

[tasonir]

The current Shroud of Golubria spell is fairly complicated, with the wiki page including a handy graph to try to explain how your chances work. Reference to current version: http://crawl.chaosforge.org/Shroud_of_Golubria. As a reminder, it only applies to melee attacks - it won't block (or expire from) magic or ranged attacks.

I'm not going to get into it, but it's unintuitive, and of somewhat limited power due to being level 2 (although it's strong for level 2). My proposal to simplify it is to change it to trigger on 100% of attacks above a damage threshold. It then fully negates that attack, and consumes a 'charge'. When the spell is out of charges, it expires. The shroud does not expire based on time passed. The minimum damage threshold is to prevent it from 'wasting' charges on attacks that would only deal 1-2 damage. The threshold could either be static (say, 10 damage), or based on the player's max hp (10% of max hp). The number of charges should be low - I'm thinking 1 or 2.

That said, my dream version would probably bump this up to level 3 or 4 and give it 3 charges, but that might be a bit too powerful. Alternatively, the amount of charges you get could be based on spellpower; always at least 1, 2 at medium spellpower, and require significant investment for 3. If it does give 3, even conditionally on spellpower, you might want to hit the player with a -Shroud status for a few turns to prevent people from just recasting the spell immediately and keeping charges up at all times.

It would be a very helpful spell for high EV characters to avoid damage spikes, and might make light armor melee more viable; it would be less useful for heavy armor characters, but even so more damage reduction is always a good thing. By going for it not timing out I'm trying to avoid the "keep recasting it out of combat" charms problem, but you do still have the "cast it once out of combat" charms problem. Still I think this version is less tedious then the current version which has a timed duration and unpredictable removal upon taking damage. I think it gets rarely used because it's not obvious how good it is, and due to the tediousness of using it; I'm hoping this rework would fix both of those things.

[gameguard]

that seems too strong and even more tedious than the current version though

[chequers]

Rework suggestion: Remove 1d(spellpower) damage from the next attack. Spell power cap of 25. Times out after a while if not triggered.
Example: you have 15 spell power and cast shroud. You get a "Shroud" buff which can reduces the next attack's damage by up to 1d15= (say) 8. If you are dealt 6 damage, it's reduced to 0 and the buff expires. If you are dealt 10 damage, it's reduced to 2 and the buff expires. The buff has an expiry of 5 turns.
Knobs: maximum damage blocked

Rework suggestion #2: Total damage to be blocked is 1d(spellpower). Can survive multiple hits until its total damage cap is reached. Times out after a while.
Example: you have 15 spell power and cast shroud. You get a "Shroud" buff which has 1d15= (say) 8 points of health. If you are dealt 6 damage, it's reduced to 0 and the buff has 2 points remaining. If you are then dealt another 4 damage, it's reduced to 2 and the buff expires. The buff has an expiry of 5 turns.
Knobs: maximum damage blocked, maximum duration

[VeryAngryFelid]

Remove the spell.

Currently there is no monsters generation after initial level creation and if you don't have a god with decaying piety there is no downside in casting the spell before opening every door, turning around every corner or using every downstairs. It is worse than Phase Shift because many characters can get Shroud of Golubria for almost free (including zero spell hunger).

[bhauth]

Remove the spell.

Replace with Ozo's Armour in the Skald spellbook?

[Tumalu]

Make Shroud stick around more like Deflect Missiles (the shining example of a buff that isn't a PITA to use) but have it drain mp when it reduces damage, and spellpower impacts it in a meaningful way? So brutes with low spellpower can't get much out of it even though they can get it castable (exception:Cheibro or high-int races in non-caster builds), and casters can't just keep it up for low cost. It'd still be pretty useful on a Warper start which is pretty just melee with a few utility spells, they wouldn't mind burning some mp.

Still some specifics to sort out since I only spent 10 seconds on the concept, but I think this would make the spell workable. Instead of flat-out deflecting damage, probably just have it soak part of it. Maybe the chance/amount of damage reduced is spell-power dependant, or impacting harder hits requires higher spellpower so that a melee brute later in the game would have too weak of a shroud to actually help.

Alternatively, it could be shushed away from melee brutes by bumping it up to lv4~5. Then the spellpower wouldn't be forced to heavy-hand balance it like the old lv2 Repel Missiles that everyone and their grandma wants to cast.

[duvessa]

Make Shroud stick around more like Deflect Missiles (the shining example of a buff that isn't a PITA to use) but have it drain mp when it reduces damage

So, guardian spirit?

[Tumalu]

...point taken

[Shtopit]

I'm in favour of making the current version work like Deflect Missiles (in that you cast it once, then it stays there). I don't have any strong feelings about increasing efficacy & level. The description should be made more clear: e.g.

"Bends space around the caster's body. As a result, some blows that would have otherwise struck it are forced to miss. Ranged attacks cannot be diverted in this way. The shroud is highly unstable, and may fall apart under stress."

[tasonir]

Remove the spell.

Currently there is no monsters generation after initial level creation and if you don't have a god with decaying piety there is no downside in casting the spell before opening every door, turning around every corner or using every downstairs. It is worse than Phase Shift because many characters can get Shroud of Golubria for almost free (including zero spell hunger).

All of your complaints are solved by "Make the spell not time out." as well.

That being said, honestly, I wouldn't really be all that bothered by removing it as well, I hardly use it as it's just generally weak in its current form. Useful for a few early floors, but if you don't start the game with it, you'll almost never find it early enough to still need it.

[Sprucery]

Useful for a few early floors, but if you don't start the game with it, you'll almost never find it early enough to still need it.

I think you are wrong. The spell is useful throughout the game.

[VeryAngryFelid]

As far as I remember it is equivalent of +10 HP, always useful.
If it never times out, it is very similar to Fighting.

[Midn8]

It's an average of +10 HP, but against enemies that hit hard (early game example being ogres) it will often do nothing.

In its current form it blocks all the damage with (10/(10+damage)) probability and falls apart if that check fails (and it only checks this 2/3rds of the time). This means that the ogre or whatever will almost always nuke it but will rarely be deflected.

For example, if the monster is dealing 20 damage, it will only work 1/3 of the time.

[Siegurt]

It's an average of +10 HP, but against enemies that hit hard (early game example being ogres) it will often do nothing.

In its current form it blocks all the damage with (10/(10+damage)) probability and falls apart if that check fails (and it only checks this 2/3rds of the time). This means that the ogre or whatever will almost always nuke it but will rarely be deflected.

For example, if the monster is dealing 20 damage, it will only work 1/3 of the time.

Note that a 33% chance of avoiding a 20 damage attack is pretty useful (To think of it another way, it would require a fairly large bonus to evasion to get that much reduction of incoming damage)

It *is* odd that it's less useful the more damaging the attack is, But even a 10% chance of dodging a 90 damage attack is not terrible.

Calling it "+10 HP" has always actually seemed little weird to me, I'm not sure how the two are really equivalent.