PTR Devastate testing
#1
I got on the PTR and was able to test a "DPS Devastate" build and compare it with a DPS Fury / Arms build in identical gearing.
For those unaware, Devastate on the PTR adds an extra attack for the off-hand, so PTR Devastate DPS is considerably higher than on the live build... I wanted to test how much higher.

Builds:
Prot: http://www.wowhead.com/?talent=TZVV0VxZcytoI0z0xst
Fury: http://www.wowhead.com/?talent=LVMcdhbZVV0VxxoVuVoM

Gearing:
Drakefist MH (BS crafted epic mace)
Vindicator's Brand OH (Aldor exalted epic sword)
both of these are slow weapons that should be very good for Devastate

stats unbuffed in Battle (same gearing for both builds just different talents)
Prot: 9465 HP / 1632 AP / 23.02% Crit / 167 HR (+10.59% hit)
Fury: 9244 HP / 1534 AP / 23.02% Crit / 167 HR (+10.59 + 3 precision = +13.59% hit)

Both used two trinkets that proc health (Mark of Conquest and Tsunami Talisman) Tsunami Talisman was part of the Rogue T5 handout kit I think, either that or the Hunter. I checked those kits for jewelry... the warrior stuff was all defensive items, except the actual T5 'foo' items where you could turn them in for your choice of DPS or Defensive gear (you can get 2 sets of foo and get both complete T5 sets... I did).

--------------------------------
"Grind" testing at legion hold
--------------------------------
Prot: Charge and devastate repeatedly in Battle stance. I forgot to Victory rush so numbers are a little lower than they should be. When extra rage available, Heroic Strike. Keep battle shout up (between mobs, no wasted cooldowns)
Fury: Charge or intercept, switch to zerker, then BT, Victory Rush, WW, etc... Keep up rampage and battle shout, Heroic when extra rage is available.

Prot DPS via SWstats and ~40ish mobs: 487
Fury DPS via SWstats and ~60ish mobs: ~530

Fury had more fluctuation, so it took longer to settle on an average DPS. Even then there was some movement. Prot stayed in a pretty narrow range and it took many fewer mobs to get a devent average.

Fury kept up completely with damage taken with the proc trinkets healing and BT. Prot slowly lost health. For the testing I didn't need to stop, but at the end of testing I was ~25% health with Prot and pretty much 100% with Fury.

--------------------------------
"ideal case" testing in blasted lands
--------------------------------

In both cases I stacked 5 sunders before resetting SWstats on a "servant" mob. armor was very low but not zero (hamstring hit for 65 in prot, where zero armor would be 63 * 1.1 = 69.3).

Prot I basically just spammed Devastate and used heroic strike when I had rage (usually waited until haste proc-ed from the mace to waste less rage if I could do so and not max out my rage).
Fury I did the same like legion hold, BT + WW + Rampage + sunder every 30 seconds + keep up rampage.

I did each approximately 7 minutes (4 applications of Battle shout + some overlap)

Prot DPS via SWstats: 1073
Fury DPS via SWstats: ~970

Even with Fury having enrage from time to time in the "ideal case" it can't catch up to prot DPS with 5 sunders on the mob.

Conclusions (pertaining to PTR 2.1 only):
1) prot DPS is potentially slightly better than Fury DPS on a single target with 5 sunders when there is no concern for threat
2) Prot DPS is pretty close, but slightly behind Fury DPS in a grinding scenario. Fury had superior healing for reduced downtime.
3) If you're wearing Tsunami Talisman for a health proc and not Mark of Conqest... surprisingly, the 62 required blue trinket gave me significantly more health according to SWstats than the epic Tsunami Talisman:wacko:
4) The BT healing in Fury may be a primary reason to choose it over prot if grinding is your thing. DPS is close enough that the "extras" become a bigger decision point than raw DPS.
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#2
How much higher do you think your DPS would have been if you were DPSing with Prot spec in Zerker stance to use Whirlwinds as your rage dumps instead of Heroic Strike?

Also, did you use Deathwish whenever possible with the Fury build?

Either way, this is pretty much why Dev's won't be adding this change into the live build - who needs Fury when you can do comparable DPS as Prot?
Reply
#3
Quote:--------------------------------
"ideal case" testing in blasted lands
--------------------------------

In both cases I stacked 5 sunders before resetting SWstats on a "servant" mob. armor was very low but not zero (hamstring hit for 65 in prot, where zero armor would be 63 * 1.1 = 69.3).

Prot I basically just spammed Devastate and used heroic strike when I had rage (usually waited until haste proc-ed from the mace to waste less rage if I could do so and not max out my rage).
Fury I did the same like legion hold, BT + WW + Rampage + sunder every 30 seconds + keep up rampage.

I did each approximately 7 minutes (4 applications of Battle shout + some overlap)

Prot DPS via SWstats: 1073
Fury DPS via SWstats: ~970

Even with Fury having enrage from time to time in the "ideal case" it can't catch up to prot DPS with 5 sunders on the mob.

Conclusions (pertaining to PTR 2.1 only):
1) prot DPS is potentially slightly better than Fury DPS on a single target with 5 sunders when there is no concern for threat
2) Prot DPS is pretty close, but slightly behind Fury DPS in a grinding scenario. Fury had superior healing for reduced downtime.
3) If you're wearing Tsunami Talisman for a health proc and not Mark of Conqest... surprisingly, the 62 required blue trinket gave me significantly more health according to SWstats than the epic Tsunami Talisman:wacko:
4) The BT healing in Fury may be a primary reason to choose it over prot if grinding is your thing. DPS is close enough that the "extras" become a bigger decision point than raw DPS.

There is a slight problem with the ideal. The ideal situation would not have the fury warrior refreshing the sunder stack. A fury warrior in an ideal situation has a prot tank either keeping the sunder stack fresh via devestate or the reapplication of sunder. So there is 15 rage every 30 seconds that the fury build could turn into more DPS. I can't say exactly what kind of boost that would be but it would give the Fury build more DPS than the test limited you to.
---
It's all just zeroes and ones and duct tape in the end.
Reply
#4
Quote:How much higher do you think your DPS would have been if you were DPSing with Prot spec in Zerker stance to use Whirlwinds as your rage dumps instead of Heroic Strike?

Also, did you use Deathwish whenever possible with the Fury build?

Either way, this is pretty much why Dev's won't be adding this change into the live build - who needs Fury when you can do comparable DPS as Prot?

1) Whirlwind is barely more damage than devastate for use of a cooldown. DPS would be slightly higher in zerker just because it's zerker, but not sure that WW would be more than HS since it replaces a devastate.

2) Fury didn't use Deathwish at all, but was getting enraged from time to time

3) This was my original point in the 2.1.0 patch notes thread. I think this puts prot DPS too close to Fury. We'll see how it ends up, but I think this build is likely to go live given they've removed, tweaked, and re-introduced the devastate change. I think if this does go live, then in the future it will be nerfed or Fury DPS tweaked to be more effective without synergizing with MS for PvPers (meaning they'd have to add the DPS to the 31 point and beyond talents.)
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#5
Quote:1) Whirlwind is WORSE at both damage to rage and damage per cooldown than Devastate with both hands.

2) Fury didn't use Deathwish at all, but was getting enraged from time to time

3) This was my original point in the 2.1.0 patch notes thread. I think this puts prot DPS too close to Fury. We'll see how it ends up, but I think this build is likely to go live given they've removed, tweaked, and re-introduced the devastate change. I think if this does go live, then in the future it will be nerfed or Fury DPS tweaked to be more effective without synergizing with MS for PvPers (meaning they'd have to add the DPS to the 31 point and beyond talents.)

I really don't see what the problem is with the dual-wield devastate. It's less DPS than fury, still, and as you said, fury is still better for grinding. Under other than 'perfect' conditions, fury still did more DPS.

I watched fury warriors crow about being #1 on the damage meters for 2 years, and also watched them tank all but a few encounters just as well as a prot warrior could. Is there a problem now, if a prot warrior potentially could do 90%ish of their DPS? I don't see one. We're not *over* their DPS, except under 'perfect' conditions, which aren't likely to happen, as you said.
--Mav
Reply
#6
Quote:There is a slight problem with the ideal. The ideal situation would not have the fury warrior refreshing the sunder stack. A fury warrior in an ideal situation has a prot tank either keeping the sunder stack fresh via devestate or the reapplication of sunder. So there is 15 rage every 30 seconds that the fury build could turn into more DPS. I can't say exactly what kind of boost that would be but it would give the Fury build more DPS than the test limited you to.


Yeah. The Fury warrior had plenty of cooldown time to refresh, it was essentially 0.5 - 1 RPS that he didn't have to use with heroic strike. I don't think it would have been HUGE. But in a raid scenario he wouldn't be taking much damage either, or getting enrage. So it probably balances out (or even ends up balancing to the point where it's positive DPS).
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#7
Quote:Yeah. The Fury warrior had plenty of cooldown time to refresh, it was essentially 0.5 - 1 RPS that he didn't have to use with heroic strike. I don't think it would have been HUGE. But in a raid scenario he wouldn't be taking much damage either, or getting enrage. So it probably balances out (or even ends up balancing to the point where it's positive DPS).

Ah yeah I hadn't thought about the no rage from damage in the raids. And yeah I didn't think it a huge factor, just something that occured to me. Use of deathwish would probably be a bigger factor anyway.
---
It's all just zeroes and ones and duct tape in the end.
Reply
#8
Quote:Yeah. The Fury warrior had plenty of cooldown time to refresh, it was essentially 0.5 - 1 RPS that he didn't have to use with heroic strike. I don't think it would have been HUGE. But in a raid scenario he wouldn't be taking much damage either, or getting enrage. So it probably balances out (or even ends up balancing to the point where it's positive DPS).
It's pretty easy to test--grab a Protection warrior and have him tank the Servant mob in Blasted Lands. That way you can check how much threat you're pulling with the Protection warrior and see how much more DPS you can squeeze from Fury.

The Prot warrior wouldn't get rage from damage either, though he also uses less rage.
Earthen Ring-EU:
Taelas -- 60 Human Protection Warrior; Shaleen -- 52 Human Retribution Paladin; Raethal -- 51 Worgen Guardian Druid; Szar -- 50 Human Fire Mage; Caethan -- 60 Human Blood Death Knight; Danee -- 41 Human Outlaw Rogue; Ainsleigh -- 52 Dark Iron Dwarf Fury Warrior; Mihena -- 44 Void Elf Affliction Warlock; Chiyan -- 41 Pandaren Brewmaster Monk; Threkk -- 40 Orc Fury Warrior; Alliera -- 41 Night Elf Havoc Demon Hunter;
Darkmoon Faire-EU:
Sieon -- 45 Blood Elf Retribution Paladin; Kuaryo -- 51 Pandaren Brewmaster Monk
Reply
#9
Quote:1) Whirlwind is barely more damage than devastate for use of a cooldown. DPS would be slightly higher in zerker just because it's zerker, but not sure that WW would be more than HS since it replaces a devastate.

Well, Whirlwind + Normal attack will do about as much damage as Heroic Strike + Devastate, for less rage. I don't know how often you had to use Heroic Strike to dump rage, so I can't comment if rage efficiency is important at that point or not.

However, my bigger concern was the extra threat from Heroic Strike - you are already generating more threat than the Fury Warrior thanks to Devastate bonus, and extra threat from Heroic Strike will only make it worse.

I also would like to point out that on almost every fight, Warrior will be taking some damage in one way or another. In those cases, being in Zerker stance and popping Zerker Rage at the right time can increase your rage generation quite a bit and help you DPS.

But yes, I still agree with the overall feel that DPS is probably too high this time with this change if we are to change nothing about Fury. Perhaps giving Fury tools to deal with Threat would make it better than Prot for PvE DPS while the actual DPS remains close between these two specs. For example, make it so that when you are under the effects of Rampage, you generate 20% less threat. Or every time you activate Rampage, you lose X points of threat. Something along those lines.
Reply
#10
Quote:However, my bigger concern was the extra threat from Heroic Strike - you are already generating more threat than the Fury Warrior thanks to Devastate bonus, and extra threat from Heroic Strike will only make it worse.

But yes, I still agree with the overall feel that DPS is probably too high this time with this change if we are to change nothing about Fury. Perhaps giving Fury tools to deal with Threat would make it better than Prot for PvE DPS while the actual DPS remains close between these two specs. For example, make it so that when you are under the effects of Rampage, you generate 20% less threat. Or every time you activate Rampage, you lose X points of threat. Something along those lines.

Yeah, there is that aspect of WW. Though when you get to the point where someone else is tanking (and thus sunders on the mob), WW ends up being less damage than devastate... for more rage. If you're at the threat ceiling it would help, but othrwise devastate is a better use of GCD.

I really like that idea for Fury warriors. Don't necessarily make it easier for them to do damage, but make it easier for them to handle threat.
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#11
Quote:--------------------------------
"ideal case" testing in blasted lands
--------------------------------
I'm pretty sure testing on low level mobs will skew things a bit. For example, the extra hit from precision will probably be wasted. You'll crit more (and thus have flurry up more often).
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)