Hexes and Magic Resistance

[jejorda2]

The problem is, anything powerful enough to be threatening usually has high enough MR to resist most hexes (which I think could stand to be balanced better).

(from here)

I've seen grumbling about this on the wiki and in the tavern. Here's a set of mechanics that might help out:

• Glowing reduces effective magic resistance. More glowing reduces effective MR more.
• Using magic resistance to defeat magical attacks causes glowing. The more powerful the attack, the more glowing increases.
• Higher base magic resistance causes glowing to decay or disappear faster. Items that increase MR and abilities like Trog's Hand would speed glow decay, but glowing would not change MR's effect on glow decay rate.

This way, if the first attempt at hexing an opponent doesn't work, the second is more likely to work. And it still might take five tries to confuse something, but at least you make some progress.

Perhaps cap the effect such that glowing can never cut MR beyond 50% of its base value if there is concern that this makes things too easy.

And we could take spells like paralyze and petrify and make them such that they only work on glowing targets, so players can't get one-shotted as soon as they see an Ogre Mage.

[galehar]

The idea that magic resistance gets easier to overcome after successive castings is good. But I don't think glowing is a good mechanic for that. It does seem thematic that resisted magical attacks cause magic glowing, but the gameplay effect isn't that great. First, monsters don't glow currently, so we would have to implement it. Then, gaining mutations from the glow because you resisted hexes isn't good. Allowing high MR to speed glow decay is outright bad. The point of glow is that you can't do anything about it, so you just have to try not getting any.

[jejorda2]

monsters don't glow currently,

Don't monsters glow when you cast corona at them?

My original idea (which I worked into this glow mechanic because I thought the original might be too complicated) was to have graduated states for each enchantment. Instead of "Confused" or "Not Confused" you could be "30% Confused." There would be different colors for the confusion indicator like glow has now. At 30% confused, you'd have a 70% chance of an action working normally and a 30% chance of working like confuse does now. There would be a 30% chance of getting a stab opportunity (at whatever the normal chances are) against a 30% confused enemy.

The second cast could move from 30% to 60% or 30% to 130% confused, based on spell power and resistance. Getting confuse beyond 100% just makes it last longer. The @ line or xv'ing an enemy would show some standardized line like "substantially confused."

We could distinguish between alistair's, mephitic cloud, the confuse spell, etc. by making the spell increase confusion level based on spell power checked against resistance, alistair's always sets to exactly 40%, and mephitic cloud never goes over 30% or some fraction based on HD.

That idea has to be thought out for each hexed state, and probably has to be stored and coded separately for each state on each spell on each monster. I figured glow would be simpler than graduated hexes, but maybe we just need a different universal resistance reduction mechanic. We could just have Current MR/Max MR like we already have Current HP/Max HP.

[acvar]

There is no need to get so complicated, or to force it into an existing "box" that does not fit. Just make resisted hexes temporarily reduce the targets MR by a random amount dependent on the level of the hex. This gets rid of the all or nothing aspect of hexes and makes it far easier to balance.

[galehar]

There is no need to get so complicated, or to force it into an existing "box" that does not fit. Just make resisted hexes temporarily reduce the targets MR by a random amount dependent on the level of the hex. This gets rid of the all or nothing aspect of hexes and makes it far easier to balance.

That seems sensible. However, it's going to need a significant amount of balancing and playtesting (even more if we also apply it to the player), so I guess this will have to wait for the 0.8 release.

[Tiber]

Your basic idea sounds interesting, though I do agree with galehar that it should probably not be tied with regular glowing. I would instead though make the amount your MR is reduced based on how easily the hex was resisted.

[TGW]

Partial success is the correct solution. Successful hexes work as now, unsuccessful ones can confer a lesser effect.

With regards to mephitic cloud, don't screw with schools to try to balance things. 1) Reduce the confusion duration. 2) Tie cloud duration more strongly to spell power. 3) Intelligent monsters don't move while confused and next to water. 4) Confused monsters don't hit themselves. 5) Revert the HD check nerf.

1 and 2 prevent people from confusing stuff forever with just one cloud, nerfing it in the early game and making the effect scalable. 3 and 4 are good balance and imitate the way confusion works against the player. 5 because the HD nerf only really hit Mephitic Cloud where it wasn't great. Spell noise plus increasing amounts of rPois plus the old HD check were reasonable deterrents from using it in late game.

[dolphin]

What do we do with Teleport Other[...]?

Splinching.

[jejorda2]

For some of the hairy ones like minmay listed, maybe when an enemy is 80% teleported, there is not an 80% chance it will teleport, but there's an 80/10=8% chance or 80/4=20% chance. Just pick a factor, and have a big jump in the success rate when you pump the effect up to 100%. And, given the P is the built up hex effect and F is the fudge factor applied when P<100, maybe for teleport, there is a P/F chance that it blinks instead of the full teleport, and a (100-P/F) chance that nothing happens.

For polymorph, there's a P/F chance that it changes to something else with the same glyph.

For banishment, maybe a P/F chance of teleporting on the same level.

Disintegration: small damage works, but I don't like it. Maybe small chance of minor Armor Class reduction?
Paralysis: P/F chance of not moving and being able to be stabbed.
Enscorcled Hibernation: P/F chance of not moving and being able to be stabbed.
Corona: Glow more by a tiny amount.

But it's probably simpler to just reduce magic resistance and not keep track of partial success results.

[TwilightPhoenix]

I prefer partial success to the other proposed solutions, but of course there's the problem of coming up with appropriate partial successes - what do we do with Teleport Other, polymorph, banishment?

The ogre mage mutters some strange words.
You partially resist.
Your torso is devoured by a tear in reality!
You die...

[TGW]

I prefer partial success to the other proposed solutions, but of course there's the problem of coming up with appropriate partial successes - what do we do with Teleport Other, polymorph, banishment?

tele other: longer duration
polymorph: partial as now, full success polies to something with lower HD
banishment: not a spell

[jejorda2]

banishment: not a spell

If magic resistance has an effect on Lugonu's ability, it might make sense to address banishment along with spells. I can't say if it does.

[acvar]

Partial effect is a horrible idea. It may seem neat at first glance, but it is all kinds of broken in actual play. What you get is irresistible spells. For example a common idea is to have Ensorcelled Hibernation have a slow effect if it fails. That makes it a 1 mana guaranteed slow which is better then the 3rd level must beat MR Slow spell. Guaranteed de-bufs are bad. Having MR decrease with successive attacks allows the caster to eventually succeed, but at the cost of both time and mana, and It is far easier to implement (you don't have to come up with tons of weaker effects).

It would require quite a bit of balancing, but I don't think it would be that hard. It does not have to work the same way for monsters attacking characters either which will save some time. In fact it makes no sense for it to work the same way with monsters since they don't have mana to worry about. Monster spamming high level hexes would make character MR all but useless. It would also create a need for a MR bar graph just like HP and Mana which is just way more then is needed. Let monster stay the way they are, raise monsters current MR (or lower the base chance for the character), and implement something simple like MR = MR - (spell level)d20, or whatever d?? suits your fancy.

[acvar]

Players have limited mana. They can't spam high level spells without fast mana regeneration at which point monster MR should become irrelevant. Monster do not have to worry about mana and as such can spam away.

Now to address your "solution". It does nothing, but lower their base MR and give you a possible lesser effect. If they are significantly above your level you still will never effect them no matter how many times you cast your spell at them, and that leaves hexes still useless at higher levels.

[TGW]

'Hexes aren't supposed to be good against everything' is a bad design decision, because nobody will ever consider getting 20 levels of a skill if it's useless against most endgame enemies.

[acvar]

What I propose does not make hexes good against everything. It simply makes it so the are not completely useless against most things past the mid game. There is a chasm of difference between good and not useless.

[acvar]

It simply makes it so the are not completely useless against most things past the mid game.

They aren't useless against most things past the midgame. Giants, yaktaur captains, Zot draconians...
Keep in mind that most hexes, when they work, make their victim completely trivial.

Yes but it does not kill them. Most non-hex spells outright kill the target. Yes it takes several castings often, but the target ends up dead. So how is taking several castings of a hex to make the victim "completely trivial" better then making the victim dead? Cast several icicles at the opponent and they are dead. Throw several confuses at the victim and they are still very much alive and still need to be dealt with. If anything hexes are still significantly less powerful then other spells.

In the end this has nothing to do with how powerful hexes should be. It has to do with how reliable they should be. The simple fact is that when one option is unreliable it is almost always ignored for a more reliable option unless the unreliable option is significantly more powerful. If there were a spell that missed all the time, but when it hit did massive damage would you use it (bolt of inaccuracy)? Most simply would not, and for good reason. This makes unreliable options hard/impossible to balance. We do not need to make hexes more powerful. We need to make hexes more reliable and less random.

[Kate]

Lower duration as a partial resistance effect for at least some hexes (cause fear, slow, enslave, control undead, corona, confuse/mass confusion) seems perfectly reasonable to me (agreed that things like making slow a partial resistance effect for EH is a bad idea, though).

Other resistible effects: pain/agony are fine, confusing touch uses HD instead of MR and is also fine. Poly other, tele other, petrify, banish, cigotuvi's degen, EH/englaciation, dispersal are all a bit more all-or-nothing and might work nicely with use acvar's system. I think either or a combination of both systems could work well but it'll still require rebalancing lots of hexes spell-by-spell.

[jejorda2]

Partial effect is a horrible idea. ... What you get is irresistible spells. For example a common idea is to have Ensorcelled Hibernation have a slow effect if it fails. That makes it a 1 mana guaranteed slow which is better then the 3rd level must beat MR Slow spell. Guaranteed de-bufs are bad.

Saying it's bad for magic resisting enemies to be slowed by failed Ensorcelled Hibernation is like saying it's bad for cold resisting enemies to get hit by stone arrow when throw frost fails. Of course that is a bad idea, it's much too powerful an effect. I suggested that there be something like a 10% chance of movement or attacks failing for a short period. That's much weaker than successfully putting an enemy to sleep, but it still gives you some chance to run away. It's weak, like weak damage received after resisting an elemental attack. It's not nothing, like the no-effect from resisting a poison attack. Non-effects are boring, but they are simpler to design, implement, and balance than interesting effects.

Having MR decrease with successive attacks allows the caster to eventually succeed, but at the cost of both time and mana, and It is far easier to implement (you don't have to come up with tons of weaker effects).

I agree that it is much simpler to implement and balance to simply lower magic resistance. It may be more interesting to find partial effects. Then it may make sense for players to invest a little in hexes to use them as escape routes, even if you don't invest enough to get full effects and therefore kills with stabs. It makes more interesting choices for players, which seems like a good thing.

[acvar]

The problem with partial effects is that some partial effects are obvious, others not so much, and other still have absolutely no good partial effects. This means partial implementation. This means hexes with no lesser effect and as such much lower success rates become useless for all practical intents and purposes. If the developers decide to not implement a change till every hex has a lesser effect that means it will either never get implemented or many hexes will get removed. It also means many possible hexes will never get implemented because there is no good/obvious lesser effect.

Lesser effects "might" be more interesting. I doubt it. I think it is far more interesting to wear my opponent down and eventually get the effect I was actually trying for. I think it would be cool to walk into elf 7 vault and cast a few mass confusions until some of the elves start wondering aimlessly and then finishing the job with a couple of enslavements on the ones that are not confused. That is tactical. All the elves being slowed right off the bat instead of confused is not tactical or interesting in my opinion.

[Cybermg]

I think partial effects and repeated casting both address different problems, and they aren't necessarily exclusive. Partial effects remove the current binaryness of hexes, and I suppose paves the way to buffing hexes against resistant enemies, since characters might now be able to land at least partial effects on them (depending on balance). Repeated casting mostly attacks the second problem (needing to buff hexes) - you can overcome high MR with repeated casting, an option not available before. I personally like repeated casting a lot, and I think it spices up the hexes game both offensively and defensively.

Regarding partial effects, I'd much prefer duration changes over weird partial versions of the spells, since the partial versions would be hard to balance and keep consistent. In any case, worries about the spells becoming overpowered aren't really justified since we can alter the range when partial effects take place arbitrarily, even to the point where they give spells absolutely no benefit over what they were before.

[SinsI]

Don't know about generic MR resistance, but, IMHO, anything that is vulnerable to Mephitic Cloud must be made at least as suspectable to Hexes's Confusion. Mephitic Cloud(via Evaporate) is a 2nd level Transmutation/Fire spell - so it is cheaper and much easier to cast, is an area spell and has several chances to affect its targets - it is so much more powerful it is not even funny. And it is practically required to fight groups of orcish priests with their broken ignoring-every-defense Smite.

[ieattime20]

Yes but it does not kill them. Most non-hex spells outright kill the target. Yes it takes several castings often, but the target ends up dead. So how is taking several castings of a hex to make the victim "completely trivial" better then making the victim dead?

Well, for me at least, it's nice to pick a god for a spellcaster besides Vehumet or Sif Muna. Hexes are good for low-MP races or low-health races that want to basically just have an edge to their combat game. If you have spells like Icicle, there's no reason to get into hexes really. However, if you have awesome melee aptitudes and sort of crappy Conjurations, there's every reason to go into hexes to make your melee more effective.

[LunarHarp]

On the idea to make MR in general be reduced when monsters resist. This could open up the practice of using cheap hexes to soften up enemies before you blast them with something more powerful/expensive. Say using confuse a bunch before spending the piety to try to banish a high MR enemy. The partial resists idea would take care of this, but such an overhaul seems to be a nightmare to balance.

Linking duration to power as a higher level buff looks like the easiest, but still has problems with some hexes. In particular if you are a stabber, you really only need EH to last one turn to get that stab in.

[smock]

On the idea to make MR in general be reduced when monsters resist. This could open up the practice of using cheap hexes to soften up enemies before you blast them with something more powerful/expensive. Say using confuse a bunch before spending the piety to try to banish a high MR enemy. The partial resists idea would take care of this, but such an overhaul seems to be a nightmare to balance.

Good point. I think this can be fixed by having MR reduced only until the next spell of the same type targets the monster. So, to banish a monster, you'd have to try banishing it several times. To confuse a monster, you'd have to try confusing it several time. Spamming corona would only help to have corona hit but wouldn't affect the likelihood of a successful banishment.

[dpeg]

I don't like the idea of splitting Hexes spells into each having a variety of effects. This is a lot of work, not just code but also design, test and balance. Using durations (where applicable) is acceptable.

Reducing MR when Hex spells are resisted is a good idea. The amount of reduction should clearly depend on the Hex spell (level and/or power). Second, we can indicate the reduced MR (both in the monster's description, and probably also as a flag, like "wandering", "distracted").

[LunarHarp]

Reducing MR when Hex spells are resisted is a good idea. The amount of reduction should clearly depend on the Hex spell (level and/or power). Second, we can indicate the reduced MR (both in the monster's description, and probably also as a flag, like "wandering", "distracted").

With this idea should the MR be reset after a successful hexing? If it stays at a reduced level then the monster could become rather easy after the MR is reduced enough to keep casting the effect. I also think the MR should regenerate back over time or else you could hit and run to soften them up, rest to get MP back, then continue hexing until they succumb. Not that this is optimal behavior, but it shouldn't be encouraged.

[SinsI]

It should not be an MR reduction effect, but a separate hexes-only curse effect, so that the resistance is only reduced for further hexes spells.

[bt]

Maybe the problem can be solved with a new spell (Hex obviously) that directly lowers MR?
It can work like this:
1. smite targeted
2. monster resists this spell with MR/4
3. if monster fails to resist, then it's MR lowered to MR/2
4. multiple casts are ok
5. spell level 4 or 5
Numbers need tweaking of course.