Remove evokable invisibility (was remove "End invisibility")

[VeryAngryFelid]

Invisibility spell requires investment in Hexes, has miscast chance with very severe consequences and cannot be ended prematurely (ignoring potion of cancellation, quicksilver dragon breath etc. here).
Evocable invisibility does not require any investment, even with 0 Evocations it can be used reliably before opening a runed door or entering a new level and even if you have 1% Invisibility spell it is still better to use evocable one because it can be ended prematurely and thus you will get less contamination.

[ontoclasm]

That sounds like an argument for removing evocable invisibility instead

[VeryAngryFelid]

Yes, that would be the best decision. Ring of shadows might keep it though.

[duvessa]

fr: replace +Inv on ring of shadows with an accuracy bonus

[ydeve]

FR: remove evocable invis, replace with stealth on unrands.

[duvessa]

it already gives you an assload of stealth from the umbra

[4Hooves2Appendages]

[...] even if you have 1% Invisibility spell it is still better to use evocable one because it can be ended prematurely and thus you will get less contamination.

Wait a second! Is the contamination related to the length of the effect? Is this true for other things too, like haste?

[duvessa]

yep

[VeryAngryFelid]

http://crawl.chaosforge.org/Contamination#Accumulation

[Quazifuji]

I'd say that should be changed to a flat amount, regardless of what happens to evokable invisibility. It seems pretty unclear in the game - or at least I, like 4Hooves2Appendages, had no idea that was how it worked - but it also feels awkward to have the drawbacks of haste and invisibility be increased as your spellpower increased. Sure, you might still get more haste or invis per contam point, it's not like haste or invis becomes worse with spellpower, but there are still situations where a low-power haste or invis could be worth than a high-power one, which seems really undesirable to me.

You can make the contam amount closer to a high-power than a low-power haste/invis, if you want, which also works as a way to nerf haste for characters who invest the bare minimum into charms to get it working but doesn't hurt characters who invest more into casting it, which I think is also a good thing.

[dpeg]

No more evocable invisibility and flat amount of contamination for the spells sounds good to me, too.

[HardboiledGargoyle]

it already gives you an assload of stealth from the umbra

Robe of Night doesn't give you an umbra.

fr: replace +Inv on ring of shadows with an accuracy bonus

If I understand what you're getting at, it would be better to simply restrict the ring's umbra's size to a single tile.

[Wahaha]

I do not understand how "remove evocable invisibility" as a response to the topic makes any sense, in any way whatsoever.

If it can be used reliably at 0 evocations and this is a problem, the solution would be to increase the required evocations skill to use it. I recommend a failure % curve that stays at 100% failure until for example 10 evocations, and only then starts going down. This would prevent spamming the ring with, for example, 2 evocations at 95% fail until it works.

If there is a problem with inconsistency* between the spell and the evocation where one can be canceled and the other cannot, the solution is to make them consistent. Crazy, right.

*Inconsistency is not actually a problem by default. There SEEMS to be a problem because the ring is easier to use AND it can be canceled. But if the ring of invis required 20 evocations to use, it could be acceptable to give it the bonus of being able to be canceled to make up for the higher skill requirement (or the other way around).
Also, there is no problem by default with the same effect having different perks depending on the source. (If you're going to reply that it's confusing, don't, because it's really not)

Edit: the first post was edited so the first line of this post doesn't make sense but the rest still does.

[dracos369]

I have a feeling that, with a "BTW" sentence in my other poroposal, I sparked this, and that makes me sad.
Don't take away my cloaks of invisibility, please, they're the coolest cloaks around!

[duvessa]

Take away dracos369's cloaks of invisibility, please

[dracos369]

:'(

[VeryAngryFelid]

Removing evocable invisibility is a good solution because it duplicates a level 6 spell and can be used in heavy armour. Training evocations even to 10 is not a big deal because many (most?) characters do it anyway.

Spoiler: show

Yes, I would remove wand of hasting for similar reason

[Bart]

If it can be used reliably at 0 evocations and this is a problem, the solution would be to increase the required evocations skill to use it. I recommend a failure % curve that stays at 100% failure until for example 10 evocations, and only then starts going down. This would prevent spamming the ring with, for example, 2 evocations at 95% fail until it works.

That's not a solution at all, because the skill will still be spammed at 95% fail, except at 11 Evo instead of 2. The problem is that failing the evocation does nothing. Compare that to the severity of miscasts at 95% failure rate for a level 6 spell.

As for me, outright removal of +Inv would be the best way to go. There's really no good reason to duplicate the effect of the spell as an evocation.

Spoiler: show

Yes, duplicate sources of haste could also be removed, however here it's less problematic, as non-spell sources are not spammable.

[MrRokkomies]

Take away dracos369's cloaks of invisibility, please

Let's remove cloaks of invisibility and create a fixed art "Draco's cloak of invisibility" with fragile, so that you can not take it away.

[andreas]

That's not a solution at all, because the skill will still be spammed at 95% fail, except at 11 Evo instead of 2. The problem is that failing the evocation does nothing. Compare that to the severity of miscasts at 95% failure rate for a level 6 spell.

This seems quite right to me, but still solvable without wholesale removal of evocable invis. Evoking at high fail rates (in general, or just for invis, which is a special case already) could be given some drawback like miscast effects or just stacking contam.

Of course if you're already convinced that evo-invis should be removed because it duplicates a spell then you'll see this as a needlessly inelegant solution, but I'm not sure what's so bad about having a handful of evocables duplicating spell effects.

[VeryAngryFelid]

what's so bad about having a handful of evocables duplicating spell effects.

They can be used in heavy armour and with tiny Int, don't require spell slots, don't require investment in spellcasting and potentially multiple magic schools. Evocations is already OP, no magic school alone gives access to so many different things at once (summoning, conjurations, hexes, mana restoration, clouds, killing multiple things while silenced) and you don't need Evo higher than 10-15 for most.

[andreas]

I think I agree that those are all together a problem with evocations, I'm just not convinced that this is the right way to balance the skill. (I'm not saying I'm sure that it's not the right way, either.) For example evo invis could be made decisively worse than the spell, harder to reliably trigger than it is now, etc. There are a lot of variables in addition to removal that can be tweaked to adjust the power of evocations.

As for weakening the school by outright removal, I would rather see the elemental evocables removed; yet another way to directly deal damage seems less interesting to me than utility effects that are difficult to get otherwise.

[Rast]

The elemental evokers have amazing utility effects:

Water: push back enemy, put him in bad terrain (if he's non-flying).
Air: Push away a lot of enemies
Earth (RIP): Dig random holes in the terrain, sometimes shaft enemy down to another level.

[andreas]

Yeah, you're definitely right. I still don't like them very much, but I'll have to think more about why that is.

[VeryAngryFelid]

https://github.com/crawl/crawl/pull/290

[green]

I just made an account to complain a bit. I know the guideline everyone has been following lately to update the game has been "remove this" and "add this god". And I know I have never added content to the game.
Even then, why don't you add the following feature instead of killing the contents that someone added before?: When you fail an evocation on an item, like blink or invisibility, add any kind of contamination. Maybe add a chance to remove the blink ability from the item or even destroy the item. Would something like I just described be possible to add to the game, or it is already hardcoded?

[dpeg]

green: This is not really about throwing away the work of others, in my opinion. We have three sources of invisibility (potion, spell and cloak/ego items). The problem is that +Inv is very cheap to use, that you can spam it at low success rates, that it takes away from the spell, and is connected to a skill doing a lot of other strong things.

I like invisibility, it is great defensively and for stabbing. But going with potion and spell alone improves the situation, in my opinion: more choices, less abuse.

[Tressol]

As it stands, finding a +Invis item early and using it liberally (resting to burn off the contamination) can basically ensure that you make it through Lair (and probably Orc) with very few real problems. Invis is a very strong combat buff in addition to it's stealth benefits.

Simply adding Contam on a failed evocations check wouldn't solve too much - you could still get by with a pretty high failure chance. Generally, a player would come around a corner/move up on a threatening monster and, assuming it doesn't wake up on the first turn, just step back and use Invis. If you end up contaminated, you just wait it out. Food is generally such a non-issue in Crawl at the moment that the turns lost cost the player very little (unless playing for score, I suppose).

I think a better penalty for failing the evocation roll on +Invis would be to "unstealth" the player, to a degree dependent on the degree of failure (or, just to a random degree whenever the roll is failed). A near/lucky failure might do nothing, as it does now. A standard failure would generate a loud noise. A bad failure would place a Corona on the player as well as making the noise. The point is that if you spam it with bad Evocations, you can easily end up in a bad spot. Invis is cancelled by Corona, so it's the perfect penalty - you can't keep spamming the +Invis item to get out of badly failing to use the +Invis item.

Presuming that, for fluff purposes, we can say that (evoked) invisibility is bending light around the character, failure messages could be something like...

"Weird colors dance around you for a moment." (Failure, nothing happens)
"Strange colors build around your body before bursting with a crash!" (Failure, Noise)
"Your (+Invis Item) emits a high-pitched whine as it pulls the dungeon's light in around your body!" (Failure, Noise, Corona)

This would solve a lot of problems with +Invis, imo. It's still a bit weird that the character ostensibly has less "control" over the version of the spell that they cast themselves (which also involves a lot more investment...), but it would be strange if removing that ring didn't end the effect. One step toward a solution there would be to allow spell invisibility to be dismissed as well (potion still can't be - either way it's 2 one way and 1 the other).

[VeryAngryFelid]

Now compare that to Hexes miscasts:

1: corona or random uselessness
2: Curse one item, or Slowing, or Silence
3: Magic vulnerability effect on self, or Confusion, or 0-18 glow, or fall asleep for 5 turns

http://crawl.chaosforge.org/Miscast#Hexes

I have just tested HuFi WITHOUT body armour, with Hexes 15, spellcasting 10 and starting Int 8. It has 24% miscast (red).
With hexes 17 it becomes light red 12%. With hexes 19 it becomes grey 4%. With starting scale mail it becomes light red again (17%). With plate armour, Str 16 and Armour 15 it becomes red 83% (!) despite Hexes 19! With Hexes 24 it is still light red 14%...

TL/DR: Better leave spells to spellcasters.

[green]

Still, Invi items are rather rare. The solution is adding Corona and Noise miscasts on the evocation item, and buffing the spell Invisibility: Lvl. 6 for Invisibility is too much the effect it has. The trouble is that buffing Invisibility would make SpEn really hard to hit.
Maybe, as removing an amulet of harm has huge effect (Drain), misevoking an Invi item should get you the Mark effect or some strong effect.
Just stop resorting to removing stuff from the game and try to fix it

[VeryAngryFelid]

Also let's not forget about heavy contamination. Both noise and corona are irrelevant when you have cleared current floor and are going to open a runed door or enter a new level.

Edit. Frail or berserkitis can be a price for casters who are casting Invisibility too early. Would you use evocable invisibility if penalties were similar?