Since it's a fairly frequent question, the formula for calculating the total crit damage done when you combine the Relentless Earthstorm Diamond (I'll refer to it as RED in the rest of the post) with talents. (With thanks to Roywyn and a number of other people over in the Relentless Earthstorm Diamond - melee only? thread).
First things first, the RED is multiplicative not additive; this means that it makes your crits do more than just 3% extra damage on top of what you would normally do. Here's a demonstration of the difference for a 1000 damage melee crit of the difference this makes:
This also means that the RED stacks very favorably with talents that increases the damage your critical strikes do. The way we need to calculate this is by figuring out the damage a crit does first with the RED, here's the formulas:
These values will at this point be multiplied with your crit talents to figure out the full multiplier for crits; however, you need to subtract your base damage before you do so due to how these talents work, and add it back in after you've multiplied the crit damage multipliers. As far as I know crit talents are (luckily) additive and thus fairly easy to calculate. I'll take a Frost Mage, an Arcane/Frost Mage and a Rogue with Ruthlessness as examples here.
The one exception to the above rules is the Feral Druid talent Predatory Instincts, which actually multiplies damage in much the same way the RED does (Which means it's actually a 20% increase on damage done by a crit); I'm unsure whether it stacks multiplicatively or additively with the RED. But here's what the two possibilities are for Feral Druids:
There really isn't math behind it as much as a general school of thought. The basic thought behind it is that you can have all the avoidance you want, at some point you get unlucky, healers will have to heal you with the worst case scenario in mind anyway, so going for higher stamina and lower avoidance means you're simply more likely to survive while also making use of the heals being cast anyway.
The only slightly math related part to it is the fact that Stamina is pretty cheap on the item budget in comparison to other stats (As you can see by the fact that blue quality stamina gems are +12, instead of +8).
Note however that this isn't universal, while in a lot of situations you're best off maximizing your stamina, there are a few specific situations in which you'd be better off going for other options as well. E.G. At Leotheras you might want to maximize your block value instead: This has two reasons, the first being that he doesn't hit all that hard so Block Value is actually a very effective way of reducing incoming damage, the second being it gives you stronger Shield Slams, which is good for dealing with Inner Demons.
But since there aren't any block value gems and there isn't *much* choice in actual tanking items this isn't really an issue regarding the "what should I use?" question ;p
Translation from Mathlish to English of my above post on the Relentless Earthstorm Diamond:
The Relentless Earthstorm Diamond multiplies the total damage done by your crits by 3%, and is applied before any talents are. For melee that means that since crits normally 200% of the total damage your white hits do they will do 206% of total damage with the Relentless Earthstorm Diamond. For spells crits are normally 150%, which means 154.5% with the Relentless Earthstorm Diamond.
"But Chicken," I hear you say "doesn't multiplying always give the same result regardless of the order of multiplication?"
Which is where things get interesting, most of the talents works by multiplying the extra damage your crits do, while the Relentless Earthstorm Diamond multiplies the total damage. What this basically means is that the value multiplied by your talents gets higher than if it was calculated the other way around. Let's take a look at it in practice:
A Destruction Warlock with the talent Ruin and the Relentless Earthstorm Diamond. We first start by taking a look at how much extra damage his crits would do without talents... 54.5%, Ruin adds an additional 100% to this, 100% of 54.5% being 54.5% means we now have crits doing 109% extra damage; or 209% of the total damage a white hit would do. Due to the fact that Ruin got added after the Relentless Earthstorm Diamond, this is just a little bit more damage than a non-talented melee crit with the Relentless Earthstorm Diamond would have been, despite the fact that without the Diamond both the melee crit and the spell crit with Ruin would deal the same amount of additional damage.
CoE and CoS are the reason that spell penetration is all but useless in PvE. (Elemental shaman and wrath-spamming moonkin would be the exceptions as there is no nature curse.) Correct me if I'm wrong but I believe no meaningful boss has yet been found whose resistances were not reduced to 0 by warlock curses.
(I suppose if you regularly had one or less warlocks in your raids then you'd be another exception.)
Did some searching, and couldn't seem to find a place where it mentions a damage reduction cap?
Not sure if it's a hard cap on reduction or just the tooltip(haven't really tested it, just noticed it on the tooltip while tanking), but I suppose there is a max cap. And from what I remember it's at ~31.7k armor(75.00%). Might be worth mentioning somewhere, for druids at least.
If this were the case, then why is it when CoE is not on a boss, the number of spells that get partial resists goes up?
Prime example is Supremus on how CoE+spell penetration affects partial resists.
Supremus is one of the very few exceptions on the matter; for the vast majority of PvE content Spell Penetration has no effect on the partial resists you get. However, even on Supremus at some amount of Spell Penetration you'd still get the same baseline partial resists you get everywhere else; he's just a rare example of a boss with Fire Resistance.
Correct me if I'm wrong but I believe no meaningful boss has yet been found whose resistances were not reduced to 0 by warlock curses.
Shade of Aran has heavy spell resistance that can be mitigated with spell penetration, while he is not meaningful to a lot of guilds, my guess would be that the number of guilds still stuck in Kara is probably greater than the number of guilds plowing through t5 and t6 content.
If your stuck on aran though, spell penetration is the least of your worries ;p
I never even noticed aran resisting exceptionally because I'm always watching stuff that make or break the raid instead such as blizzards, flame wreaths and spells that need to be interrupted ;p
Did some searching, and couldn't seem to find a place where it mentions a damage reduction cap?
Not sure if it's a hard cap on reduction or just the tooltip(haven't really tested it, just noticed it on the tooltip while tanking), but I suppose there is a max cap. And from what I remember it's at ~31.7k armor(75.00%). Might be worth mentioning somewhere, for druids at least.
DR is a function of Attacker Level and it's target's Armor.
For Attacker Level greater than 59: Armor/(Armor+400+85*(LvL+4.5*(LvL-59)))
For Attacker Level Less than or equal to 59: Armor/(Armor+400+85*LvL)
This puts the Armor cap vs level 73 at 35880, vs 72 is 34478, and vs 71 is 33075
Are spells on a 2 roll system where hit/crit/miss is calculated first then resist
or hit/miss then crit/resist?
If i did a test of say an elemental shaman using elemental mastery on a rogue with cloak of shadows, will it be able to give me the answer?
Depends on whether the spell is binary or not.
Basic Roll Order: (Non-binary, ie.Fireball)
#1 -- Land vs Miss
#2 -- Hit vs Crit
The resist roll is parallel to this, and unaffected by hit/miss/crit rates. Combat Log suggests the resist roll is done between roll #1 and #2 based on cues of how much damage was done and how much was resisted.
Binary: (ie.Frostbolt)
#1 -- Land vs (miss_rate + resist_rate)
#2 -- Hit vs Crit
The resist roll is rolled into the miss rate. Since spells either hit for full or not at all, the perceived miss rate goes through the roof against a high resistance target.
AFAIK, this is the believed/accepted two-roll system for spells.
Shade of Aran has heavy spell resistance that can be mitigated with spell penetration, while he is not meaningful to a lot of guilds, my guess would be that the number of guilds still stuck in Kara is probably greater than the number of guilds plowing through t5 and t6 content.
I'd like to see some verification of this. I've never expereinced anything like it, and a quick glance at WWS parses of guilds both with and without warlocks show no appreciable spell resistance for Shade. Since it is blatantly obvious in WWS parses that Supremus has fire resistance, and the mitigation from Aran is comparable to every other boss according to WWS, I'm thinking your mistaken here.
Aran drops shoulders with spell penetration, perhaps thats what confused you. But spell penetration is as useless against Shade of Aran as it is against any TBC raid boss that is not named Supremus.
I feel foolish but I never knew about the /stopcasting macro. As a warlock, would it also help speed up spellcasts on instants or only on spells that have a cast time (immo, sb, ua)
I feel foolish but I never knew about the /stopcasting macro. As a warlock, would it also help speed up spellcasts on instants or only on spells that have a cast time (immo, sb, ua)
I'd like to see some verification of this. I've never expereinced anything like it, and a quick glance at WWS parses of guilds both with and without warlocks show no appreciable spell resistance for Shade. Since it is blatantly obvious in WWS parses that Supremus has fire resistance, and the mitigation from Aran is comparable to every other boss according to WWS, I'm thinking your mistaken here.
Aran drops shoulders with spell penetration, perhaps thats what confused you. But spell penetration is as useless against Shade of Aran as it is against any TBC raid boss that is not named Supremus.
Actually I think he may only have spell resistance to arcane, my only experience with it so far was on my alt arcane mage where he resisted every other counterspell - which was unusual considering that I was hitcapped. Additionally since I was using arcane blast and missiles as primary nukes, I noticed that almost every single hit was getting partially resisted. Anyways, after figuring it out, I switched to scorch/fireball and didn't have the same problem. It is possible that I'm completely imagining this, and unfortunately I don't have any results logged as evidence. We don't do Kara regularly anymore so I can't easily get them either. If someone wants to get a mage or druid to test this though, I would be happy to see some actual data.
Even assuming Aran does have arcane resistance, this really only affects balance druids and one fairly uncommon mage spec on a trivial fight. If for some reason some guild has a mage heavy raid and are relying on those mages for interrupts, I suppose some spell penetration would make a difference. Overall, spell penetration is probably a lot more useful for warlocks/shadow priests in the arena (on a "backup" set of course).
I made sure I found a mage who used primarily arcane spells in a WWS when briefly checking into the claims of Aran's resistance. Didn't seem to represent anything out of the ordinary.
Unless you have more then the vague recollections of a single fight without any hard data or statistical significance, I'm inclined to beleive that Shade of Aran, just like every other TBC raid boss (excluding Supremus) has no resistance. Not to fire. Not to Frost. Not to arcane.
Empirical evidence would do wonders towards convincing me otherwise. If you can provide any, I'd love to see it.
After doing some research as well, I managed to find a WWS report of an arcane mage on Aran without a warlock applying CoS (had to sift through almost 300 reports... ugh), it didn't show anything out of the ordinary. There were a few partial resists, but nothing that couldn't be attributed to the level difference.
Either I just got an unlucky string of resistances on that one fight, and just made the unfounded assumption that it was because a mage boss might have some magic resistance, or he has been changed since then. This was pre 2.1, and they made modifications to a lot of fights in Kara (including Aran) in 2.1, so it is entirely possible that he did have resistance at one point and has since been "nerfed". In any case it doesn't really matter, because given the data that has been recorded recently, I think it is fairly safe to say that the live version of Aran doesn't have any resistances worth mentioning.
- "Once you reach your defense cap, stamina is generally the best stat to stack."
Whats the mats behind this?
The number of hits needed to kill tank is K=HP * (A + C)/ C * D, where
HP is tank hp
D is incoming damage
A is tank ac
C is a coefficient in armor mitigation formula
It can be easily seen that this number rises linearly with the increase of HP or AC. Only one random element here is the incoming damage.
Adding avoidance (a) to the formula gives: K=HP * (A+C)/C*D*(1-a). Note the formula gives an expected (average) result over sufficiently large sample.
It can also be easily shown that adding more avoidance will increase expected number of
hits like an inverse square which is much better (DK ~ 1/(1-a)^2 da) then linear increase.
Avoidance also decreases likelihood of getting killed by burst damage exponentially:
P(N=n) = (1-a)^n.
For example, tank A has 45% avoidance, can take 3 hits from boss, but 4th will be fatal.
Tank B can only take 2 hits, but has 57% avoidance.
Probability of tank A getting killed by burst: (1-0.45)^4 ~ 0.092.
Same for Tank B: (1-0.57)^3 ~ 0.078
The math seems to heavily favor avoidance once minimum effective hp (HP/AC) is reached
for the encounter. It also explains why getting more HP is relatively cheap in the item budget.
What you're missing is that no matter how much you reduce the chance for a bust to happen, there's still going to be a good chance for that burst to happen at least once in a fight, and if you don't have the HP/armor to handle it, you're dead. So while avoidance reduces the chance for it to happen, the only way to eliminate deaths altogether is stacking HP/armor.
On top of that, higher HP also increase the effectiveness of healing due to less overhealing, so it's not like it's totally useless for mana efficiency, although I agree for mana efficiency for healers avoidance is the #1 stat. Unfortunately healer mana efficiency is far from your top priority. Staying alive through a 5-10 minute fight being able to take an unlucky burst even if it only happens 0-1 times a fight on average is simply more important.
And another thing is that a short streak of avoided hits will leave you rage starved and hurt your TPS, it's not hard to calculate how much increasing dodge increases the chance for that to happen.
If the fight is hard, the healers will probably be proactively healing instead of reactively healing. To be specific, this healing will be tailored such that the tank will survive a worst case string of hits, and given the nature of hard TBC fights this means that the heals will be cast in the expectation of future damage, and will probably not be canceled if the damage doesn't materialize.
Since the heals will be landing regardless of whether the tank dodges (given lag, reaction time, and casting time, it's usually not a good idea to try and cancel a heal before it lands), and since the tank is guaranteed to survive regardless of whether the tank dodges, there really is no benefit gained from dodge. (And indeed, there is a minor negative since avoidance reduces rage generation.)
More generally, in a long fight you shouldn't be relying on luck to avoid wipes but healing, since you're very likely to get a "bad' string of hits in a 10 minute fight. AC and HP help healing much more than dodge does.
07/03/07, 11:58 AM
Similar Threads
