Haste outweighting str 1:1? Even upping the cloak proc, seems strange really strange...
Haste outweighting str 1:1? Even upping the cloak proc, seems strange really strange...
Last edited by mmoc38db56fadf; 2013-11-01 at 05:33 PM.
Well strength isn't exactly all that great for dps !:
What I find odd about that is crit only being about twice as strong as haste o0
It isn't that bad either. True that if that's on ~400k vengeance it sounds about right, but haste should be around... 0-1. As befits that horrid stat. Seems to be off.
So avoidance tanks trying to max out damage are using
Phase Fingers and River's Song now?
Anyone able to determine if getting symbiosis Savage Defense is a big deal? 30% dodge = 15% crit?
Phase fingers is what I would suggest yes and definitely River's Song
Symbiosis on the other hand will not work as its only rating based, so flat percentages like Savage Defense sadly don't contribute to riposte
For some clarification, the numbers may seem a bit odd due to using the Hardest hitting boss preset so vengeance is at a pretty much fixed value of 500k but as I'm sure some of you are familiar with this isn't an unreasonable number for the harder hitting bosses such as Thok, Malkorok and Garrosh and if you are solo tanking some encounters
For most other bosses I would say in 25HC a 300k Vengeance average would be more reasonable in which case you can expect that STR would become more valuable than Haste point per point
These were just designed to give you a relative weight of Dodge / Parry / Crit, which would not be affected relative to each other if you were changing the amount of vengeance you had
Still don't get how haste is so high compared to crit when white hits are only like 10% of total damage dealt.
Also 500k vengeance is still only for a limited window on thok (late p1) and Malkorok (uhm Bloodrage only?)
I'm also surprised to see Haste "that high compared to crit" and "that low compared to mastery".
Mastery provide more crit block so more rage so more dps. Not a huge deal, but I would have guessed it would be comparable to haste.
Also, I totally forgot about Symbiosis but with our current build, it puts us at 80%-90% avoidance for 6sec evevry 1min. The 1min-cd is quite a pain but it looks like it can be a good survival tool now.
Well... if autoattacks are 9%, Flurry is 6%, that's a 15% of our damage. 425 haste is 1%. So 1 point of haste = 0.000352%. 1 point of crit, 0.00166%. It should be 1/5 with napkin math, and if anything, skull banner, ultimatum, enrage will make crit even better in the real deal. If a simulation is almost an order of magnitude off of napkin math, the simulation must be wrong. Simcraft keeps getting better, but I guess it still needs more umphh for prot DPS.
Last edited by mmoc38db56fadf; 2013-11-01 at 07:59 PM.
Well, the reasons I can possibly see for haste being high are;
Proc rate on Clock / Helm if using them
White damage increase, albeit small
*Possible* I am not entirely sure but for Dragon Roar and Thunder Clap their GCD might be reduced by haste given that they function like spells and as you know how high our damage per global is, this would account for a high stat weight for haste
Sure you will get more rage off Mastery but by using a high avoidance build you are pushing the percentage of blocks off the attack table, reducing mastery's value quite dramatically with the 65% avoidance you will be playing with in BiS gear with stam trinkets, +6% if using avoidance ones
As for its effects on Enrage, with an avoidance build the crit you gain from it you will have pretty much the same, if not higher, enrage uptime due to crits relative to if you were using a Mastery build and relying off getting crit blocks to keep your Enrage up. Also the moment the boss starts to cast anything or delay its swing timer due to abilities this further pushes the advantage to an avoidance build utilizing crit
Eddy, out of curiosity, how many iterations did you run? And what was the margin of error on Strength, Haste and Crit?
- - - Updated - - -
For amusement I ran a baseline sim, then three more, one for each of an additional 5000 haste, strength and crit.
5000 haste = 1.75% increase apparently all from additional melee swings and River's Song procs
5000 strength = 2.82% increase
5000 crit = 6.68% increase
My guess is the stat weights Eddy published earlier line up with these figures within the margin of error.
For grins I ran another with 5000 extra attack power and got a 1.00% increase. Obviously, our dps numbers are going to depend the most on how well we can maintain our vengeance levels. Something we all knew, I know, but never hurts to have a reminder.
On this topic, Theck posted a weak aura based on Rigg's that shows when a vengeance gain is to be had by taunting. It's a little finicky requiring some custom LUA modifications to make it track the tank you want. I've been meaning to attempt to make it more resilient by keying off Main Tank raid status or the Tank role. If anyone wants to fiddle with it, please do.
Hi all, my name is Alexander of Nemesis -EU, Alliance Side.
I'm an italian mathematician and i'd like to share what i've found trying to maximize the warrior avoidance. I'm sorry for my bad english so i hope u can forgive me if i won't be perfect!
First of all you'll need to know what is Diminishing Return (DR). You can read what it is on wowwiki or on wowpedia but i prefered sacredduty.net.
What i've done is to calculate the exact amount of parry/dodge rating you have to get at a determinated total rating. I can say that with a good gear u can get up to 1% of avoidandance simlply changing your gems: the typical condition i've considered is that we have a lot of gems that we can switch between dodge and parry. What's best? Parry or dodge? Not always parry is the best solution, this because parry scales in a different way.
I recap Theck's formulas (sacredduty.net):
Parry = 3 + baseStr/Q_s + (1/C_p + k/((Str-baseStr)/Q_s+pre_Parry))^(-1)
Dodge = 5 + baseAgi/Q_a + (1/C_d + k/((Agi-baseAgi)/Q_a+pre_Dodge))^(-1)
dove
C_p = 237.1860 parry cap
C_d = 90.6425 dodge cap
k = 0.956 scale factor
Q_s = 951.158596 conversion factor str-parry.
Q_a = 10000 conversion factor agi-dodge.
pre_Parry = parry percent before DR
pre_Dodge = dodge percent before DR
Now to make an accurate calculus we keep Stamina, Strength, Agility, Hit ed Expertise fixed and let only dodge/parry mastery change. (This because we only change socket/reforge to maximize the avoidance).
To make the optimal parry rating value i've made the sum of parry and dodge and then i've derived respect parry rating (dodge rating = total rating - parry rating). This value is given by this equation :
I've used MATLAB for a better accuracy... The dodge rating optimal value is equal to Total_Rating minus the new Parry Rating.(1435567988408320*baseStr)/5580180707304480 - (7693268565897510*baseAgi)/120150415888788000 + (7693268565897510*Agi)/120150415888788000 - (1435567988408320*Str)/5580180707304480 + (6519719123641950*Total_Rating)/9011281191659110
During some test i've found that askmrrobot try to equalize dodge and parry value and this is wrong!
I've also made a macro for a fast in-game control: u have to download Super Duper Macro (Wowinterface) because this is a very long macro! U have to use the macro only if u are without any buff: if u use it with buff u'll get wrong stats. The equation that i've made don't contains buff-adjustment but this macro is made to be in raid fullbuffed condition. If u prefer to use str flask (with mixology or not) and/or str food u can change the StrAdj value inside the macro (default is 0 but usually i use 1300 : flask + food). You also need to change baseStr and baseAgi depending on your race.
You can find the full (italian) discussion on battlecraft.it that is the biggest italian fan site of WoW./script baseStr = 202; baseAgi = 137; local base, Strpg, posBuff, negBuff = UnitStat("player",1); StrAdj=0; Str=(Strpg+StrAdj)*1.05; local base, Agipg, posBuff, negBuff = UnitStat("player",2); Agi=Agipg*1.05;Total_Rating = GetCombatRatingBonus(CR_PARRY)*885+GetCombatRatingBonus(CR_DODGE)*885; d=(1435567988408320*baseStr)/5580180707304480 - (7693268565897510*baseAgi)/120150415888788000 + (7693268565897510*Agi)/120150415888788000 - (1435567988408320*Str)/5580180707304480 + (6519719123641950*Total_Rating)/9011281191659110; print("Parry Rating: ") print(d); print("Dodge Rating: "); print(Total_Rating-d);
I hope u enjoyed this discussion! Ty all
I have checked your mathematics and it is perfectly valid, however I feel that you have overlooked the fact that maximizing avoidance isn't really what people are after. It is the DPS that is what people are after and realistically (after some quick napkin maths) you will never get to the point where Dodge will give you more DPS compared to Parry
You also have to consider that when talking about a 1% difference in avoidance from using your formula's the only gain you will be getting is a slightly higher TotalDamageReduction, and the size of this is going to be extremely negligible and RNG will be a much more deterministic factor than that 1% avoidance
You also have to consider from a TDR standpoint that by going further into dodge you reduce your DPS, albeit by a small amount, meaning that the boss might live longer to give that one extra melee swing that totally negates any TDR you have got by getting the 1% extra dodge
THat post should be copied and pasted into peoples minds. It's valid for so many scenarios.
I made a review:
the difference trying to maximize avoidance and dps is that talking about dps we have to consider the glyph of hold line.
without it the best solution is to maximize avoidance: the swing reduction of the parry is sightly worse than a crit increase, if i crit 1 more shield slam i make 600k more dmg and 1 more revenge crit is 300k more dmg, etc!
with that glyph, it's sure that parry is better, but the new question is how much? 3-4-5-10-20 % ? i'll investigate to take the exact value in this week
Hm.... what is better for the shredders on siegecrafter heroic?
DR + SB (obviously precasting SB so it lands shortly after the shredder has landed again) or DR + BB?
In terms of using Shield Barrier vs. Shield block, I read somewhere recently that when using an avoidance based build it is typically better to focus on shield barrier and use excess rage in shield block. Is this theory correct? Thanks in advance.
I use the two abilities reactionary.
If a single big hit is incoming, be it physical or magic; I use shield barrier. If a steady intake of blockable damage is incoming then I use shield block. At this state in the game I am swimming in rage so I tend to use as many shield blocks as the cooldown allows. And use shield barrier to prevent myself from capping. Unless there is a big hit incoming then I want a full barrier for that obviously.
What I aim for is to have as much shield block uptime as possibly to benefit the most of the Heavy Repercussions glyph. Not only do I aim for a high uptime I also pay attention to cooldown of sheild block + Shield Slam to make sure that when I use Shield Slam it will benefit from the glyph.
Basicly what you want to do as a tank is to smooth out the damage intake. With high levels of avoidance its really hard to predict any kind of dmg intake except for magic dots and big magic hits. As an example you could theoretically pop a full 60 rage barrier and see yourself dodge/parry several attacsk in a row and competely see it go to waste. A shield block would have been more useful in that case due to some extra damage from shield slam even though it didnt really block anything.
Therefor I personally tend to use alot more Shield blocks than Shield barriers.
guys, is it worth to use dps back and gem?
10m normal raiding atm 562 ilvl