Lasie

Well, even if it is only 10%, that's still a pretty significant itemization gap, and a very odd one. Gearing for 19% hit and then no more is weird, plate doesn't have enough, leather has more, so you'd have to have a mix. Having a reduction of hit in TG only farther compounds the existing Plate vs. Leather itemization disparity.

At least TG may now be a slight DPS increase instead of a decrease.

I still don't understand their reasoning behind keeping a "counterbalance" to TG. The un-nerfed version wasn't making Fury DPS any more overpowered then other classes on beta, if they need some counter to PvP burst, then they need to make the counterbalance not affect PvE DPS.

 

Graul

That's not actually true at all. I've been able to piece together a random plate set that gives 23% hit, capped Expertise and 12% haste. Not sure how "good" the AP is in comparison, but the crit rate was abyssmal at 11%. Thing is, it was unsocketed, unenchanted and missing 2x rings, 1x weapon, 1x trinket and a ranged weapon. Now the argument about leather being better in general is still true, but now it's not as clearcut.

 

JTLJudoMan

About the mace 2 hander question.

I couldn't put my Stormherald into my offhand so I would assume that it is broken for maces as well.

Kinda frustrating as I wanted to get a bit of killin in before bed. =-P

 

Gruntle

Your conclusions are probably correct, I didn't quite realise how fast the weapons would swing in a full raid setting. Thanks for redoing the calculations with more realistic stats.

As to using the flurried speed for white dps normalization, I thought that was how it worked, but I'm probably wrong (seems the spreadsheets use the unflurried speed for normalization). What I do know is that for rage gain the hasted speed is used in the "hit term".

 

Mia

Seems to be happening with every melee attack. I am missing quite a lot with my devastate build on lvl 69-70 mobs, and I have move than 14% hit and am lvl 71.

Edit: Seems to be a bug on beta right now, where +hit rating actually decreasing your +hit

 

Darmon

I managed to do some test on protection after last PTR patch. Talent build is 5/3/53, for raid MT role. Fighitng BL immortal mobs. No glyphs, the 10% block value is not on PTR. The other that exists i don't think it can influence the results. Still no rage from dodge/parry.

T6+ quality pure tank gear (lots of avoidance), 1.5k AP, 822 BV, 4.75% crit, 82 hit rating. In def stance, deva + sslam spam, selfbuffed with AP flas, str food and BS. Rage was not an issue, i could use my abilities abilities without any rage problem, even drop a HS here and there.
Sustained dps 893 over 360s. SSlam 36%, Dev 28%, Melee 26% of total dps.
Until further threat analysis, for single target TPS i don't think anything can beat deva + sslam spam, backed by S&D procs. SW and CB seems to be pure situational, with TC on debuff duty as before. There something wrong with hit/miss, i have 3% misses(not dodge or parries) and there should be none. Same gear/buffs before last patch gave around 900 sustained DPS also, so not a big difference.
I think it is still a solid dps output for tank, compaired to live.

T6+ quality DPS gear, 2.7k AP, 26% crit, 212 hit rating. In bers stance, deva spam, using WW/CB/SW on cooldowns, HS when above 60 rage. Rage generation dificult, because of chain miss.
Sustained dps 1298 over 360s. Deva 32%, melee 31%, HS 14%, CB/SS/WW each 6-7%.
Most important thing to notice was missing. Deva 10% with from 160 counts, Melee 31% from 255 counts, HS/WW/CB/SW 10-15%. That is kind of big miss rate vs lvl57 mobs.
Second thing was a drop from over 1700 sustained DPS from last build, due to Deva change and miss rate.
On live, in raid conditions, without group staking with shaman/feral/hunters, as prot dps with just raid buffs and self buffs/pots/flasks i can sustaine 1k dps on most bosses, with peaks up to 1.5k+ in fights like Teron. Group staking means a huge increase for prot dps, so i think in new build we still got the solo dps buffed. Deva was nerfed, but it seems ok.

T6+ quality dps gear, 2.85k AP, 26% crit, 191 hit rating, 742 BV. In bers stance with badge fist + illidan shield, deva+ss spam, CB/SW on cooldowns.
HS when above 60 rage. Rage generation was good, no problem using skill whenever i wanted.
Sustained dps 1281 over 382s. Melee 31%. Deva 27%, SS 25%, HS/CB/SW around 5%. Miss rate around 10%, with SS out of patterns 1 miss out of 59 counts.

PVP gear. 2s4,1s3,1s2, full guardian where possible (neck,belt,wrist,boot, ring, vindicator second ring), no pve armors, badge fist + shield+ badge xbow. Def stance, 2.8k AP( soalrian saphire+bersk call, no pvp trinket because human racial), 26 crit, 98 hit rating(MH shield with +hit gem), 711 BV. Using deva + SS spam
Sustained dps 1262over 370s. deva 32%, melee 26%, SS 22%. Missing over 5-7% across all skills.

Last thoughts.
Rage generation/miss rate looks to be very broken. 1H+shield has no problem to generate rage to use any skills you want. When switching to DW, the rage generation feels so unreliable that makes sustained dps cycles harder to mantain.
DPS balance between play style/gear seems so good right now that i can't belive it.
Def stance dmg penalty is offset by enrage.
For sustained dps/tps when in def stance you want to skip using CB/SW, because you are using shield block to give a huge boost SS, while in zerk stance you want to use those 2 each time they are up.
Here Blizzard managed to balance it very carefully. 20s+30s cooldown vs 2x big SS on 40s cooldown. It is a bit difficult, but with proper timing+stance dance you can use all the long cooldowns.
The balance between AP/BV is very important. It can favor one playstyle or another. High AP (dps gear) means melee dmg, deva, CB and SW give a lot of bang. High BV, but low AP (tank gear) makes SS and S&B the core of skill use, with deva spammed to proc S&B,

I didn't tested Amrs/Fury much, i'm a prot warrior from day one, but i can say this is a good state for prot warriors.It does feel very balanced in all protection playstyles, and that must mean something. There are still problems, like Vigilance and Imp TC beeing very low on talent priorities, but this affects just threat generation, without other DPS tweeks. If there are further changes on dps output of prot tree, it is very easy to break the current balance. Another concern is about gear. I have almost top PvE gear, and that is means what i consider balanced stands for 1% of warrior population. With that in mind i did the PvP gear test, which any warrior can obtain.

Bottom line. Fix Vigilance to be something that matters, a mittigation buff of some kind to the prot warrior, not to someone else. It is a talent on par with MS/BT, and now is a joke compaired with those 2. Except that, protection tree is done and balanced in my view. Everything else is a matter of gear, and fixing the damn miss formula. As personal oppinions, Arms/Fury DPS would be balanced at 1.6-1.8k on same gear quality and test, and most important as long as other classes are on same dps range.

 

Warfield

Attack power's contribution to white damage is not normalized. This mechanic only applies to instant abilities.

 

Shha

Well another math problem I considered is 2h Arms Rotation. With imp MS at 5 sec cooldown the question would be - whether to prioritize GCD use or rather give instants priority.

You can go with 2 sorts of rotation I guess

MS,WW,Slam,MS,Slam,Slam - 10 seconds prioritizing instants

or MS,WW,Slam,Slam,MS,Slam,Slam,Slam,WW,MS,Slam,Slam,Slam,MS,Slam,WW,Slam,MS,Slam,Slam,S lam

31.5 seconds rotation consisting of 5xMS/3xWW/13xSlam.

Lowest "common time" would be 630 seconds when

* 1st rotation repeats itself 63 times for 126xMS,63xWW,189xSlam
* 2nd rotation repeats itself 20 times for 100xMS,60xWW,260xSlam

Rotation difference in skills =

* 1st rotation 26xMS 3xWW
* 2nd rotation 71 Slams - White damage reduced by 35.5 seconds.


Lets see the damage difference numbers as function of speed of weapon. Assuming 50% crit with axe spec, and 5600ap. 200 dps weapon , 10% haste.

S=Weapon speed.
s=Modified weapon speed = S*(1/1.06 *BF*)*(1/1.2 *WF*)*(1/1.10 *haste*)*(1/1.03 *aura haste)=0.69S

Slam base damage = 600*S+250
MS base damage = [200*S+400*3.4+380]*1.1(Imp MS)*1.1(glyph)= 242*S+2105
WW base damage = 200*S+400*3.4= 200*S+ 1360

Since all those attacks have same chance to crit and same crit multiplier, we can compare it before applying those modifiers.

* 1st rotation = S*[26*242+3*200]+(2105*26+3*1360)= 6892*S+58810
* 2nd rotation = S*600*71+ 250*71 = 42600*S+11750


Difference - 2nd rotation wins obviously (since we just compare the skills damage for now), 35708*S-47060. (a)

Unless your 2hander is faster then around 1.4 its obvious 2nd rotation provides more SKILL damage. However now we need to see what kind of dps we lose by slamming 71 times more.

White swing damage = 600*S*(0.25*0.76+0.4*2.06+0.35)= 818.4*S
White dps = swing damage/s. = 818.4 *S/0.69*S = 1186
31.5 seconds of white dps = 37360.

However before adding it to equation we need to convert our values from (a) to ones affected by crit

Modified difference = [35708*S-47060]*(0.4*2.27+0.6)=53848*S-70967

Now adjust it by -37360 from lost white dps and we get the final equation

53848*S-101327

The minimum S is equal = 101327/53848=1.88.

With any weapon slow then 1.88 - which is obviously true for any real 2h out there (I think the fastest ever was brain hacker with 2.1 speed - correct me if I'm wrong), its more beneficial to use up any gcd then to keep MS/WW on cooldown.

Thats something that felt pretty obvious - but still had enough doubts to follow up with this math. However this leads to one suggestion.


Since we wont be using MS on cd - it makes the talent somewhat lackluster. Even for 3 points, MS is much lower % of damage now then before Wrath. My idea would be to change MS glyph from increasing damage by 10% and reducing healing debuff, to decreasing cooldown by 0.5 sec - something with about same effectiveness - 10% increase on damage of skill, and less useful for pvp. It would however actually increase usefullness of our talents.

 

Liar

Sorry to interrupt your DPS theorycrafting but I'd like to go back to the idea of having avoidance on Diminishing Returns. Anyone think they overdid it here? From what I know, they were already going to itemize less avoidance on gear since logically Strength and BV would eat up some of the itembudget on the items (if you look at the SW tanking plate you will see they are completely devoid of Strength and mostly BV which is why they got so much avoidance). On top of that they added the DRs. Wow. Way to force every tank to stack Stamina and nothing else. I will have to agree with Gruntle when he said (albeit in a different context) that the only way Blizzard seems to make other options viable/desired is by breaking the ones that work.

Really, what is so bad about gear diversity? If someone wants to stack Stamina, let him. If someone wants to stack Avoidance, let him. I understand the concerns of creating unhittable tanks but they got 2 other levers to pull here by itemizing properly and by making avoidance cap. Back in the day of WoW classic people assumed avoidance had a 75% cap when it didn't. Now that we came full circle, maybe it's time to introduce the cap and leave things be? Clicky Avoidance trinkets work for a reason after all and messing it up with DRs will completely break that system.

Unexpected TankPoints error Clearly someone doesn't understand how EH works. That, or upgrading from T8 to T9 is optional in beating Arthas. Clearly.

 

Shha

Frankly, I dont like how avoidance went in BC. I think the diminishing returns could be a good thing, although frankly I can see better ones.

Avoidance right now in BC is obviously overpowered. In terms of survivability, avoidance basically always wins there is a very reasonable argument for that.

They cant make bosses hit TOO hard, or some rng would completely own players. A boss in BC seems to take at least 4 swings to kill a player (yea even brutallus). Speaking about average - fitting for the content gear.

Four is a number when statistics start to win. If it takes 4 hits to kill someone , then avoidance mechanics work pretty well - if you have lets say 75% avoidance and can survive 4 hits, vs 70% avoidance and 5 hits - the chance of you dying from BURST is lower with avoidance. Thats somewhat excessive. Stamina stops serving its purpose which is preventing bursts, and starts just being a hindrance for healers - since you take more damage and more healing.

This resulted in following actions by blizzard:

a) Increase raid damage. Notice how tank healing is somewhat secondary now in Sunwell? Well except brutallus but thats later. This creates healing imbalances, and takes away from the game. Some fights with tank being far on healer priorities are fine - a lot of them its too much.

b) Introduce Sunwell Radiance. Thats the only reason fight like brutallus is even possible. I can get 60% avoidance with SR. Without it I would have 85%. That means that Brutallus would do 2.66 times more damage. Now imagine that ... Stomp hits hitting for 20k to produce same amount of dps... that would be a completely random game from there. RNG for the loss.

c) Introduce multitank fights, with limited healing and raid damage. See a) , it was covered there. M'uru is a good example. I tank Void Sentinels with 90% avoidance almost. While all cool and dandy, it does give you an unhealthy amount of RNG there. Could you tank then with stamina? Not reall - you simply cant have enough healing diverted to Sentinel tank in that encounter to let him live with stamina stacked gear.

In short - current avoidance model FORCES you to go with avoidance after certain point. No itemization will change the fact that avoidance scales with itself, while other tanks stats do not. Diminishing returns are one way around it.


Another option though, one that I like more - would be making Sunwell Radiance like buffs on bosses - except this time more visible and open.

Imagine a boss that for example hits pretty hard, rather slow, and completely ignores avoidance (aka always hits). Stam stacking is obviously king here.

Imagine a boss hitting 10 times per second for a small amount of damage , and take full avoidance - bam avoidance is best.

Somewhere between those 2 will always be some golden point where the intended avoidance amount shines. Why instead of capping or diminishing avoidance, they cant give all the bosses VISIBLE buffs with X expertise?


Finally though - mind one thing. Avoidance as such isnt a huge concern to me - not because Im a dps warrior - in fact I tank more. But because it affects all the tanks the same - and as we know its encounters that are adjusted to tanks, not other way around. With dps , its important to see what kind of dps we do - to see how we compare to other classes. With tanking , avoidance , and such - as long as all classes are affected - we will just see bosses that are doable with our gear and mechanics.

 

Liar

I am not worried that Warriors won't be able to tank boss X anymore or something along that lines. I am more worried about the fact that they killed the strategic part of gearing for either avoidance or stamina. Come to think of it, they probably killed gearing for threat as well and are throwing out so much expertise that you might as well wear your one tank set that is the best everywhere.
Really, is noone else seeing the problem in this super gear homonization? It's not fun at all.

Unexpected TankPoints error Clearly someone doesn't understand how EH works. That, or upgrading from T8 to T9 is optional in beating Arthas. Clearly.

 

Graul

You can just as easily swap "gear" with "class" and "skills" and yes it's a major concern for a lot of us. The reason given was because it would make it "eaier to balance classes" with, but it just seems more like a way to take less time having to design items. And, when they come up with a new idea they don't have to come up with a new idea for all of the classes when they can simply take that one idea and give it to two or three classes and the only difference is the name and icon.

 

Moomore

Hello everyone.

I was wondering if anyone has thought about the following spec yet.

An arms spec while DW 1 Handers.
I have seen a video of someone testing this on a dummy. Aswell as testing a Fury TG build and in the end the Arms DW build/rotation seamed to put out significantly higher DPS then the Fury TG build.
This spec would go up to wrecking crew in Arms and up to Imp execute in the fury tree.

The build would be based on Taste for blood, sudden death and wrecking crew.
There are 2 rotations. The first is a rotation in battle stance. Using overpower on every taste of blood proc and execute on every sudden death proc aswell as always beeing able to use enraged assault on every CD due to the enrage effect of wrecking crew being up almost all the time. This rotation requires one to keep up Rend ofcourse.
Or stay in zerk stance and instead of the Rend/Overpower abbilities use WW instead. Altho this would been that the uptime of Blood Frenzy might be less.

The Berserk stance rotation seemed to slightly outdps the Battle stance rotation altho that doesn't take in consideration the effect of Blood Frenzy.
The fury TG build did alot less DPS. This however was tested on a mob dummy and not a boss dummy.

I was just wondering if this has been tested further or not? Or am I just stating something now that is really old?

Cheers

 

hellord

Wasn't Enraged Assault removed? And anyway way would you use it to consume a WC effect?
There were some posts about 2 fast 1handers and they can pull out nice dps at high level of crit but i'm not sure how reliable is going for 2x1h vs slam since in the latest builds the changes in Slam make it useful over 1.8-2.0 speed compared to Live where it sucks under 2.5s (mostly due to gcd).
Anyway 6% from 2h spec is going to scale much better and overpower is very dependant from weapon damage (the higher damage the better increases so slow weapons wins).

I didint test both specs yet, but seems that in the last build there will be no reason to not go for a 2hander, at least in the earlier stage of the expansion.

 

Hypatia

From the numbers I'm currently seeing, they seriously screwed something up. Like "misplaced a decimal point" screwed something up. Unfortunate that they did it on a weekend. :l

 

Danin

This whole patch seems incredibly rushed (especially with the whole "two builds in three days" thing). I'm guessing they didn't have time to test a lot of the changes they put in this build, which would explain why everything is bugged to hell.

 

Darkrenown

I can't log in to confirm it, but people on the beta forum are reporting hit and crit ratings are actually lowering your hit/crit chance instead of raising it.

 

Fellwraith

It depends on what their target avoidance cap is. At some point, I'd imagine every avoidance stat caps out, just like armor did (that would also prevent rogues tanking Gruul, which isn't exactly intended). My avoidance is only slightly higher than it was in T5 gear, which was when all the avoidance items started showing up.

The DR appears to be before we put points into talents like parry and anticipation. I went and respec'd and got 1 point in each avoidance stat. If the DR was after talents, I'd expect to see less than 1% increase per talent point since it would affect my existing gear.

Aside from just stamina and armor, I think this means a warrior will also want to be stacking block value and block rating. Block rating doesn't appear to have a DR, and it gets some benefit from talents like critical block and shield spec.

 

Qed

In a perfect world, avoidance stats would be tweaked so that at all levels of gear there is a useful choice between avoidance and stamina. At low levels of gear in TBC, stamina was the only sensible thing to put in item sockets. Even with the crazy avoidance scaling in sunwell gear, I tend to socked by color rather than just for avoidance. To me this means that I am getting reasonable returns from either stat. That said, the best thing to do would be to set the avoidance normalization so that you gain it faster than in BC at low percentages and slower than in BC at high percentages. (Unfortunately, it looks like the current plan is just to nerf it at all levels: The tooltip gives the "Amount before diminishing returns" as old avoidance number)

If block value and rating does become a viable way to gear out your warrior (say, if they added gems for those stats) I'd be really excited. I'm not holding my breath though.

 

moxy

Seems like DR on avoidance was a necessary change. What other stat in the game scales exponentially? I think we just need to see exactly how the scaling works, and I'm pretty certain that it's far from decided right now.

 

Lithose

The problem is pretty simple..Avoidance has always been better than armor in regards to long term damage, however, tanks are concerned with very very small sample sets, usually in the realm of 3-4 seconds. Whats the worst that can happen in those 3-4 seconds. The only reason avoidance got used in sunwell is because stamina/armor could not be stacked to the point where it was a 100% survival chance for those 3-4 seconds, so people stacked avoidance to "lower" the odd of that "worst case" because they couldn't avoid it all together.

I hope they keep that in mind when putting a DR on it..The DR shouldn't be as stringent as armor, it should allow a higher rate before it starts to take effect (Even if the curve is large.) because quite simply, its not as good under those important "worst case" small data scenarios, so it needs a benefit elsewhere..

Pros of armor and stamina=Can take more damage before death+Consistency.

Pros of avoidance=Can take more damage before death (Over a long period.)+exponential gains.

I agree that exponential gains was "too large" an advantage once we reached "end game"...but it still needs something else to be comparable with stamina/armor, though what that mechanic would be is beyond me (Perhaps something like parry? Where the tank gains something cool for avoiding? A small bonus that would be there to make up for the loss in consistency from armor/stamina, like when he dodges he gets a small %bonus to armor? or when he is missed he gets a +hit buff for XX seconds?)

 

• Morsexy

I think it's fair to be concerned. Even though I love how warriors work, I've always been in favor of completely scrapping us for a more rogue like system at least for DPS. Then when we went to defensive stance we could be like a druid that switches to rage.

Instead they are taking inherently different energy systems, Kintetic and Potential, and trying to balance them. I just find this type of thinking doomed to fail from the start, especially in the hands of the hardcore community that always min\maxes.

I say all this because you clearly see them approaching the tanking questions with the same mindset, with 4 classes that work extremely differently and trying to make them act roughly the same, in the same situations, or otherwords having little variation in performance.

So if currently a Paladin is a 10\10 for AoE tanking, and a warrior is a 3\10, they want to make it so the paladin is 9\10 and the warrior 7\10, with the deathknight a 7.5\10 and the druid an 8.5\10 or however you'd like to break it down. My issue stems from the fact that since we all work on such disperate systems, without crippling what makes us unique this task would be impossible. This is the current system we live in.

In the end I'm not surprised that they're approaching gear with diminishing returns, because stacking stamina is one way we know ( with predictable leather x bear multiplier ) that X amount of plate stam gear = 30,000 health and can balance encounters around that, not around unique and perhaps fubar sets ( 2500 BV, or 80% avoidance etc ). I know I would prefer if they just went ahead and had 12 bosses, and they made them so that each class was either an optimal or secondary for each boss.

This way if you had any one of two tanks ( out of 4 ), you could tank each boss and prepare accordingly. So you take either a Warrior or a Bear, and your other tank is a Dk or a Pally. Why is that a bad system? Homogenization is going to lead to boredom and perhaps still result in a min\max best tank thus nullifying all this work.

We can play Starcraft 2 for balanced gameplay, we play WoW for that character customization that they seem to be trying to do away with.

 

Qed

To nitpick: Avoidance scaling is not exponential. It might be singular, in that there comes a point where our effective life becomes undefined or infinite, but 'very fast' doesn't mean 'exponential'. Exponential has a specific meaning: the rate of growth is proportional to the amount you have, which doesn't fit what we're taking about here.

 

Grymm

I think it's important to note that TG is not a flat dps reducing talent. It increases your weapon damage contribution on white attacks and your normalization/weapon damage on slam/WW. It also synergizes with 2 hand spec. What the counterbalance does is up the point where you want to have your base hit at. If you work it out, it's about a 33.5*15 stat balance tax on the talent at level 80. However, this doesn't mean much. In decent fury gear at 70, it's almost impossible to have less than maybe 14% total hit. I usually raided at 17%-18% hit. Do draeni +hit auras go raidwide now? The TG tax on me would then be 6-7% hit. If blizzard intends to keep a counterbalance on TG, I certainly hope it stays a static penalty and not something that scales. Also, I do get some increased damage from that additional hit that I would have gotten without a penalty so it's not like the extra hit is merely lost stats.

What is wrong with TG is not really the talent its self. It's that the fury tree as a whole is not matching other classes on DPS output. If the TG talent is providing a 10% dps bonus, I don't think blizzard is going to make it a 25% talent just to balance fury warriors. They'll have to add it into the tree spread over all the talents. I hope a couple talents decrease in cost from 5 to 3 and that some talents (flurry/UF/rampage) get buffed to balance us out. A similar approach will have to be made with arms. Blizz has just started seeing results from 25 man raiding on the beta so hopefully we'll start seeing some total dps buffs. I wouldn't blame TG for the overall underperformance of the fury tree.

 

moxy

Depends how you measure its value. If you measure in terms of expected burst damage over a period of n boss swings (an extremely useful way of thinking) it scales by the exponent n. Even if you only measure in terms of average incoming damage, it scales by its own reciprocal, which as you point out, allows it to approach infinity at 100% (i.e. 1/(1-1)). I agree that's not quite exponential in the strict mathematical sense. Suffice it to say it's super-linear--the more you have the more each new gain is worth, until it approaches infinite value. Can any other stat make that claim?