Elitist Jerks

Elitist Jerks (http://elitistjerks.com/forums.php)
-   Theorycrafting Think Tank (http://elitistjerks.com/f47/)
-   -   Spell Penetration (http://elitistjerks.com/f47/t15548-spell_penetration/)

Vontre 08/27/07 3:57 PM

Spell Penetration
 

 

Contents

[top]Spell Penetration


While spell hit rating reduces the base chance for a mob to outright resist a spell, spell penetration only reduces the effective magic resistance stat of a target when calculated against the spell. In other words, if a target has 50 magic resist, and the caster has 20 spell penetration, only 30 of the target's magic resist will be counted against the spell.

Since the vast majority of raid bosses have no magic resistance at all, spell penetration is entirely useless for the vast majority of raid encounters. However, there are a few notable exceptions, such as Supremus in the Black Temple, who has a great deal of fire resistance.

[top]Full Resists


The game interface offers no distinction between a spell that is resisted due to hit chance (or a "miss") and a spell that is resisted due to magic resist. Both events are recorded as a resist in the combat log. Magic resist is a rare stat for mobs, so most resist messages seen are due to hit chance, and not the result of magic resistance.

[top]Partial Resists


A partial resist occurs when your spell only delivers a fraction of its total damage. Partial resists are either the result of magic resistance, or an unavoidable magic resist modifier for targets above the caster's level. Level-based magic resist is counted as 8 magic resist for every level the target is above the caster, meaning a raid boss has 24 magic resistance against all spells. This innate magic resist cannot be countered by any means; spell penetration does nothing against partial resists caused by level difference.

[top]Binary Spells


Certain spells are considered "binary" and cannot be partially resisted under any circumstance. These spells are usually tied closely to an essential debuff. Most spells are not binary, Frostbolt and Mind Flay being the notable exceptions. Binary spells are also exempt from the innate magic resist counted for targets of higher level than the caster.

Kalman 09/02/07 8:13 PM

Mind Flay is also binary, and in 2.0 I'd consider it a notable exception; whether it needs to be included, I leave up to Vontre.

DeeNogger 09/02/07 8:50 PM

Im not sure if this is the appropriate article but I heard that the Eye of Mag doesnt proc off of partial spell resists. Therefore, one could argue, its not so great in the hands of a fire mage who will get a lot of partials.

Maledict 09/14/07 4:48 AM

You say Gurtogg Bloodboil in Black Temple, don't you mean Supremus?

Docjowles 09/14/07 3:48 PM

Edit: Disregard, I completely failed to read the last sentence.

Vontre 09/17/07 1:32 AM

Quote:

Originally Posted by Maledict (Post 477313)
You say Gurtogg Bloodboil in Black Temple, don't you mean Supremus?

Thank you, haven't been yet. =D

masanbol 09/30/07 4:05 PM

Quote:

there are a few notable exceptions, most notably Supremus in the Black Temple
The Department of Redundancy Department wants you to take a look at this phrase.

Kavan 04/03/08 9:10 PM

Information about level-based partial resists is not accurate. See http://elitistjerks.com/f31/t23832-s...ances_studies/

kershner 05/14/08 5:21 PM

While I have not personally verified the information, it is my understanding that the two major raid bosses with known resistances are:

Supremus - 200 Fire Resistance
Rage Winterchill - 200 Frost Resistance

The two other things that might be worth noting in the article is that no amount of Spell Penetration will affect a mobs immunity. Second that Spell Penetration in small quantities is appropriate as a PvP stat to overcome the variety of Player buffs that provide resistances.

Binkenstein 06/10/08 7:50 AM

Question: where did the 8 resistance per level in excess of the casters come from?

Based on the WoWWiki entry for Resistance it gives 5 per level rather than the listed 8. 5 works out to be roughly 3% mitigation, which is close to my observed figures from raiding.

Spazzball 06/10/08 3:13 PM

Perhaps it was changed. I remember an old article in our Mage forums on my guild site that had a link to the WoW site and it talked about 8 added resistance. The link was: WoW Basics: Spells. That link I thought was the one that showed that there was 8 added resistances, but I skimmed and it didn't; it does hold this section though: WoW Basics: Resistances. The resistances article doesn't have anything stating added resistance either, so it should use some testing.

Esoth 06/12/08 5:45 PM

Quote:

Originally Posted by Binkenstein (Post 776273)
Question: where did the 8 resistance per level in excess of the casters come from?

Based on the WoWWiki entry for Resistance it gives 5 per level rather than the listed 8. 5 works out to be roughly 3% mitigation, which is close to my observed figures from raiding.

I think you might be misreading it. The wowwiki article I'm looking at says "Damage reduction = (effective resistance value / (caster level * 5)) * 0.75"

The "effective resistance value" is 8*(level difference between mob and player), or 24 (the magic resistance) for a boss. So if you're level 70 and the mob is level 73, 24/(70*5)*.75 ~= 5.14% mitigation. This is close to the 6% supposedly found from massive parsing; I'm not sure how much you can contribute to chance.

Also, this is the formula Cheeky uses for Lightning Breath partial resists - not that this proves anything, but perhaps he would know where this formula came from. I'm curious as well.

Arnath 06/19/08 12:57 PM

Is there some concrete data on whether the boss partial resists apply to binary spells (Frostbolt in particular)? Theoretically, 5-6% partial resists for a binary spell would lead to an extra 5-6% miss rate. However, as said in this post by Manly: http://elitistjerks.com/f31/t24323-m...21/#post745738, we don't see this in practice.

Binkenstein 06/22/08 5:46 PM

Quote:

Originally Posted by Esoth (Post 779594)
I think you might be misreading it. The wowwiki article I'm looking at says "Damage reduction = (effective resistance value / (caster level * 5)) * 0.75"

The "effective resistance value" is 8*(level difference between mob and player), or 24 (the magic resistance) for a boss. So if you're level 70 and the mob is level 73, 24/(70*5)*.75 ~= 5.14% mitigation. This is close to the 6% supposedly found from massive parsing; I'm not sure how much you can contribute to chance.

Also, this is the formula Cheeky uses for Lightning Breath partial resists - not that this proves anything, but perhaps he would know where this formula came from. I'm curious as well.

I'll have to do tally up some figures from my own WWS parses, but I'm seeing lower than 5.14% resistance, on average

Roywyn 06/22/08 6:48 PM

Quote:

Originally Posted by Binkenstein (Post 789770)
I'll have to do tally up some figures from my own WWS parses, but I'm seeing lower than 5.14% resistance, on average

5-6% damage loss due to partials is what it seems like on average.

The issue is that combat logs (and WWS) have a bad way of displaying partial resists.
Read http://elitistjerks.com/f31/t23832-s...ances_studies/ it explains it very well.

Short version:
Your Lightning hit for 2k on an undebuffed mob. It then crits for 4k (ignore CSD) * 1.05 (Misery) * 1.2 (Stormstrike) = 5040.
Now you get unlucky and get a 50% resist, the log will then produce the following.

"Your Lightning crits for 2520 (1000 resisted)."
It's a 50% resist (you lose 2520 damage), but the log only lists a 1000 damage loss (50% of the undebuffed base hit).
Your 50% resist only appears as a ~28% resist in the log.


Quote:

Originally Posted by Arnath (Post 786843)
Is there some concrete data on whether the boss partial resists apply to binary spells (Frostbolt in particular)? Theoretically, 5-6% partial resists for a binary spell would lead to an extra 5-6% miss rate. However, as said in this post by Manly: http://elitistjerks.com/f31/t24323-m...21/#post745738, we don't see this in practice.

The "unremovable resists" (partials) do not apply to binary spells at all.
Otherwise, you'd always have a ~6% resist chance on counterspells and purges, which would make Council/Maulgar/Reliquary a RNG fest.


All times are GMT -4. The time now is 8:30 PM.

Forum Infrastructure by vBulletin 3.6.12 ©2000-2007, Jelsoft Enterprises Ltd.