Okay, patch 3.3.3 dropped this week. The Revenge change? Pretty darn awesome. The change to Imp Revenge makes it really great for PvE. The change to Vitality was a nice bump in our health. But since I've covered tanking topics for weeks now, it's long past time I make good on my promise and talk about DPS. First off, they already hotfixed the 4pcT10 set bonus down from its near 100% up time back down to the 20% intended proc chance. That's a shame, because for one brief, shining moment, our 4 piece set bonus was almost as good as the rogue 2 piece. Of course, there was a nice duration buff to Trauma, a much less nice nerf to Bladestorm, and Rampage changed to a passive aura (meaning that now I have to use Bloodrage if I need to use Enraged Regeneration, but otherwise I'm happy).
We can talk about Rend Dancing. What, pray tell, is rend dancing? Well, it's fitting Rend into a fury warrior's DPS rotation. The reason this works (when it does) is because a fury warrior's DPS rotation has a lot of dead space between abilty and global cooldowns and waiting for procs to switch stances and put Rend on a target for more DPS. Generally speaking, a macro to switch you into Battle, apply rend, then switch you back to Berserker is very useful for Rend Dancing.
Rend Dancing is very subject to problems with latency, and it ends up as a net DPS loss on any fight with a lot of AoE. But on fights like Festergut or Saurfang, where you stand in one place and beat on one boss, it's better to use those empty spaces between Bloodthirst and Whirlwind cooldowns to actually generate some DPS. I've tested it out and seen some decent DPS from it (about 65 more DPS on 10 man Festergut) with the Glyph of Rending. And it's not like it's terribly onerous to switch over to Battle once every 20 seconds or so and slap a Rend up. (Even without the glyph, which I don't really like using because it costs me either Cleaving or Heroic Strike glyphs, 15 seconds is pretty doable.)
What the rend dance indicates to me is that the fury warrior rotation ends up with a lot of empty space. You're wholly at the mercy of Bloodsurge (and the 4pT10 set bonus) as without a Bloodsurge proc, you're basically just hitting Bloodthirst every four seconds and Whirlwind every 8 while using Heroic Strike to bleed off excess rage. In a situation where you have enough rage, you'll basically never stop hitting HS, which means your rotation looks like this: BT (HS) - BT (HS) - WW (HS HS). Using the Glyph of Rending to provide extra DPS is fine on fights where rage really isn't an issue, because you can switch stances to slap rend on and back within the four seconds between your WW and the next BT. What complicates the issue is, what else are you expected to do in that time?
Will you be expected to Sunder, to use Demo, to keep Commanding up? These all take up GCD's and keep you from using those dead zones in the fury rotation for a rend dance. This is where I have to salute the ingenuity and skill of the people who came up with this technique in the first place. But it still says to me that fury, at present, manages to be constrained by cooldowns in a way that's unappealing to me. Yes, you can get good DPS out of a fury spec. But the idea that people are actually getting more DPS on single target by putting in the Glyph of Rending and slapping Rend up (dumping a decent amount of rage and risking a 4pcT10 proc duing a period of low rage) amazes me. It's as if a feral druid felt it was worthwhile to switch from cat form to caster form and cast Wrath every 20 seconds or so.
Again, this isn't for every fight. You're going to want Cleaving on Arthas, I'd bet, for all those annoying little adds to smash with Cleave and Whirlwind, and to get Val'kyr's when they're all grouped up, and to maximize Deep Wounds for 2pcT10 procs. It's absolutely not something I'd use in a 5 man heroic, either, because most 5 man bosses don't really last long enough at this point to justify losing rage to stance dance just to put Rend up.
Frankly, I wish arms had some similar move they could use to eke out more DPS in raids. As much as I love fury, I'm really sad to see arms warriors relegated to debuff bots at best with their DPS lagging behind other hybrids (and even the other DPS spec of their own class) while they eat PvP nerfs that weren't really necessary or justified. A really well played arms warrior can still do some significant damage, of course, but it's a far cry from the days where you saw a warrior with a Cataclysm's Edge and you backed away slowly.
Basically, both warrior DPS specs feel sort of aimless and un-fun lately. Going from fury to prot, I always feel like someone took a set of manacles off, as with Sword and Board proccing free Shield Slams, Revenge now worth using, Devastate and HS abilities I can use pretty much constantly when nothing else is up, TC and Shockwave on longer cooldowns that can be fitted in-between other abilities, and Cleave available for trash you have a wider variety of abilities and more variety in when you use them. I'd love for Whirlwind to be broken up into a short cooldown instant and a longer duration true AoE move (frankly, the criminal under-utilization of Victory Rush in endgame raiding/instancing surprises me.... why isn't there a glyph or talent that makes VR useable as part of a fury or arms rotation?) - since Whirlwind suffers from the same AoE damage cap as spells with no target limit, why does it have one? (This affects Bladestorm as well, since basically Bladestorm is just seven whirlwinds cast consecutively.)
What I'd like to see is arms get a small but significant DPS boost via a new ability (possibly another proc, but with Taste for Blood and Sudden Death arms is proc reliant enough in my opinion) that filled in some cooldown gaps. Fury, I don't know that it needs more DPS, per se (well, okay, I always think fury needs more DPS) but I'd love to see a more fluid rotation with less dead space encouraging people to come up with weirdly effective stance dance moves to increase its damage. Heck, maybe something as simple as making Rend work in zerk.
Next week, I'll outline what I see as the state of the arms spec: where it's good, where it could use work.