Tojara

Im happy for these changes and everything, but am growing a little concerned with what might now be a hemo build for maximizing dps. This is great for PvP and all as sub really needed some help to make it viable. I think it's fair to make a hemo rogue viable with the debuff they provide, but a little over the top when a hemo rogues personal DPS exceeds that of a fully combat specced rogue, regardless of weapon spec.

Serrated Blades really needs to be made much more accessible to heavy combat rogues in my opinion.

 

• songster

Actually, that has nothing to do with the subtlety tree and everything to do with the bottom half of the assassination tree. That region is just so powerful (particularly Relentless Strikes) that it is preferable to the top half of combat and also the top half of Subtlety. The so-called Hemo build that's getting attention now isn't really a subtlety build at all, it's almost pure tri-spec.

As long as combat has both Imp S'n'D and Dual Wield specialisation, there will not be a single raiding build that doesn't put at least 20 points in the combat tree. Likewise as long as Assassination has Relentless Strikes, there will not be a single raiding build that doesn't put 11 points into Assassination.

 

• Playered

Is it really 'right' that a versatile cross-tree spec beats a full combat build however?

Having Hemo do 'close enough' damage but having the debuff provide more of a raid benefit should be enough reason to bring atleast 1 (out of 2-3) per raid, and without any real 'loss' for having it so.

If they intend to use Hemo to boost the performance of a full sub spec, then having it at say 110% damage and having a 35 point talent which also increase its damage modifier the extra 15% would maintain things better while allowing combat to still be the best.

 

• songster

Considering that deep combat is also one of the preferred PvP specs right now, why not? Look at it this way: you have enough talent points to invest in three half-trees. Right now, that's the bottom half of Assassination and both halves of combat. After 2.3, you can pick 3 out of 4 half-trees (bottom of Assassin, both halves of combat, and bottom of Sub).

It's nice to have a choice for a change.

 

Ruqas

Who cares about what spec is for what? Some people are almost talking like they have some kind of loyalty to combat specs. Others believe that hemo shouldn't be better at PVP and PVE. Why the hell not? Hunters have Beast Mastery; warlocks have Demonology and Affliction; priests have Shadow and Holy/Disc (especially after 2.3 buffs and more to come); paladins have Holy (and maybe soon Ret); druids have Resto and Feral; mages have Frost (because it doesn't really fall that far behind fire). All those specs can excel in both PVP and PVE. Some aren't optimal in both cases on one hand, but on the other, some of them are.

Also, rogues have never had a buff to give to the raid. Now they do. Every other class has buffs to give to the raid that don't sacrifice optimal DPS or HPS with the exception of hunter's Survival tree. Why not us?

Simply put, Hemo is a change. For me, it is a welcome one. I hate combat daggers with a passion and don't have the balls to go combat maces because I'd lose DPS (or at least I think I would).

 

• songster

Note also that the PvE Hemo spec is very different from a PvP Hemo spec. "PvE Hemo" won't be any better at PvP than a combat mace spec, for example.

 

amafi

I copied my rogue over last night, and I took a trip to Mulgore this morning, and bought two lvl 3 weapons.
One 1.9 speed 1-3 damage mace and a 2.6 speed 2-4 damage sword.
I equipped exactly 1400 AP to make it all easier, and just went to town on the level 1 plainstriders.

Mace (1.9) hemo hit for 302-303 and sword (2.6) hit for 303-305, so it's seems they did normalize it.
If it wasn't normalized there should be a ~70 damage difference, unless I'm completely mistaken.

 

Ruqas

Note also that PVP cannot be evaluated the same way PVE can. I can say spec "A" does more damage than spec "B". However, PVP has many more variables than just damage, such as survivability, and stuns (for rogues). PVP is based on opinion and style of play. One may be better with one spec than another, but the other may be better with his spec than the first. They may also have the same rating.

Even if none of what I said counted, most people believe that 0/31/30 is the new best PVP rogue spec. Also known as AR/prep to those that already assume the use of BF and Hemo.

I also said that PVE Hemo is only "good" for PVP, meaning better than combat swords, fists, and certainly daggers. It is also, in my opinion, better than mutilate and combat maces. I say this only because I tried both on the PTR and liked PVE Hemo much better.

Edit: Amafi's information is also orgasmically interesting. Anyone else have more information to back this up?

 

• Cluey

Yes, this needs to be emphasised.
If you had read all the thread you would have noted the posts made by people who tried it in PvP and found it worse than combat builds.
It was more specifically in arena play as it doesn't have any burst, these latest changes might help it there but we will have to see how it plays out.

 

Ashere

This may not be the right forums to post, since these are theorycrafting, and not the wow-suggestionforums, but here goes what's happening now:

Trispec hemo is getting very close to be being a preferred build right now. The hemo buffs are not boosting Subtlety viability at all, since there are still the must-haves from other trees to make it work, so becoming a subtlety rogue is still NOT viable at least for PvE.
There are basically 2 viable hemo raid builds when 2.3 goes live:
11/27/23 and 11/21/29
The key difference between these two builds is: Weapon Expertise or Dirty Deeds.

And yes, that is a mere 2 talent points, so what happened to the rest of the talentpoints?
Simple: Deadliness turns out to be a more valuable stat per talentpoint spent then Sword Mastery in terms of DPS. So someone going for Dirty Deeds would be rather silly for not dropping Sword Spec and replacing it with Deadliness. The remaining 21st point in Combat would be Blade Flurry. Also not as strong as Deadliness per talentpoint spent, but it gives a burst that would certainly make a difference in the shorter fights, with trashmobs for example.

Now why doesn't this change to hemo matter for subtlety?
That's because the trees do not ballance out like that. Take the strongest talent from each tree, and try to put them all into one single spec.
You'll find you're ending up with 11/20/30: trispec hemo. Here you have Relentless Strikes, Dual Wield and Deadliness in one single build. Each talent is just in reach of one another. The changes to hemo only strengthened an already viable spec, they didn't help subtlety in any way.

Now imagine going deeper into one tree:
-get 41 points into Assassination: For this you will drop the subtlety tree, since those 30 points there are still the weakest points out of the trispec-hemo build. But, deep-assassination gladly provides other forms of DPS to compensate.
-get 41 points into Combat: again, you'll drop the subtlety talents here, since these are the weakest talents again. Gladly, the deep-combat tree offers various forms of DPS increase to compensate for the loss of deadliness, so it's not that big a problem.
-get 41 points into Subtlety: Now you will notice a problem. There is no DPS in subtlety, apart from Sinister Calling, which is basically lacking. But to go deeper into Subtlety, you will have to drop either Relentless Strikes, or Dual Wielding. But there isn't a single talent that deep in subtlety which compensates for that loss.

My bet is that hemo won't go live with the current 125% modifier. It really is quite a bit over powered. The 125% modifier would actually make a hemo rogue preferrable over a combat rogue for 10-men raids. And no, that's not a joke, 125% damage and then those 10 hemo charges that WILL be used, will put hemo far ahead over combat for smaller parties. But to make subtlety as a tree worth it, something has to happen deep in the Sub tree, requiring 35 or more talentpoints in the tree already, that would be as valuable as Dual Wield or Relentless Strikes. Not taking those two talents is currently unviable for any raid build, and Sub basically has the bad luck that it has neither one in it's tree.

 

◊ s4dfish

I'm not 100% convinced on the claims that 11/21/29 and the ilk beat out a combat build. When you figure in the +36 damage to weapon damage then 11/21/29 does beat the combat build, but if you allow the combat build to assume the hemo debuff it exceeds the base damage of the Hemo build.

At least at my gear level (pre-Kara) this is true, may not hold true at higher gear levels

For completeness (Vindicator's/Latro's):
11/26/24 w/ Hemo on personal damage: 634 DPS
11/26/24 w/o Hemo on personal damage: 597 DPS
19/42/0 w/ Hemo on personal damage: 666 DPS
19/42/0 w/o Hemo on personal damage: 625 DPS

 

PSGarak

It used to be the case that combat was the PvE tree, and subtlety was a PvP tree, with assassination being the daggers tree that was at least somewhat viable in both cases. People are concerned about hemo hybrid builds outscaling pure combat in PvE, but they're assuming the old combat = PvE paradigm still holds, which I'm not so sure about anymore. If these changes go live (hell, even if they don't) it's time to re-evaluate what the place of each of the various builds.

Combat maces seems to be the arena spec of choice nowadays, and very well could end up being the PvP damage spec with dps cooldowns like blade flurry and AR. Deep-subtlety is the ganking spec, with heavy openers and escape tactics. Ganking as a whole got nerfed in TBC with stamina and resilience, but that's still the focus of deep-sub. PvE damage is this tri-spec hemo hybrid. And assassination is still the daggers build that is reasonably competent anywhere. I suppose that combat-swords is also now a multi-purpose (raid + arena) build rather than a build specifically optimized for one area.

Am I completely off my rocker? It's completely possible, since I don't play a rogue myself except on PTRs. I'm just trying to apply my experience as a warlock with the scope and situations of the various trees swinging back and forth between what was best. You have to evaluate the game as it is, which often means forgetting history and looking at the talents tabula rasa.

http://armory.mmo-champion.com.nyud....88847noszh.png

 

• songster

Simpler thing to do is put Relentless at the 16 point mark instead of 11. That gives you three strong raiding talents (Hemo, Relentless, Dual Wield), of which you can pick any two. You get a choice between specs, but you don't have an overpowered trispec that has all of them.

 

◊ s4dfish

Sounds like a reasonable assumption at this point, though I'm still not sure 11/2x/2x out damages 19/42/0 when the combat build is allowed to assume eating up a Hemo debuff.

 

◊ Vulajin

I did the exact same test with the 1.9 speed mace on Servants in the Blasted Lands and had the opposite results. My average Sinister Strike hit should have been ~215 damage, but came out as ~170, indicating roughly 21% armor reduction. My average Hemorrhage hit should have been ~153 if non-normalized and ~183 if normalized, and I experienced ~115 damage, which is only plausible if it's still not normalized (~24% armor if non-normalized, versus ~37% if normalized).

I'm not clear how to account for your results. I'll go back shortly and repeat the test using two different weapons myself. Are you sure you were only counting hits?

 

Ruqas

0/30/31 reigns supreme in PVP now, according to many.

But yeah, you're hitting pretty close to home there.

s4dfish, it doesn't hold true at higher gear levels. The whole point of Hemo's new buff is to allow it to scale (and do more damage overall). Besides, hemo sword spec is suppose to be 11/28/22. Though I am not sure after this buff.

Also, could someone please expand on amafi's information. It seems very vital at this point, even if it is doubtful that Blizzard would implement such a change and not notify us of it.

 

amafi

Yeah, only counted hits, crits panned out the same, 604-606 vs 604-608 if I remember correctly.
I think I'll head to orgrimmar and kill the piglets by the gates, they should have no armor and they respawn fairly quick so I can get more hits in without having to run all over mulgore.

Is it possible that US and EU are running different builds?

Edit:
Installed recount and hit the swine in the pens outside orgrimmar gates a bit. (just until 10 hits, should be enough since it doesn't vary much?)
Here's a pic of the damage breakdown with the mace:

And here's the one with the sword:


I apologize for the size of the pics, but I just recently reinstalled windows and haven't gotten the gimp installed yet.

 

Ashere

That is a very good list you have there, and it explains what's going on:
-As hemo, you *always* have the hemo-debuff up.
-With the change that reduces the amount of charges to only 10, all charges *will* be used up every time.
-Generally speaking, Kara raids take only 1 rogue at a time. (yes, sometimes 2, but generally, it's only 1..)

So the difference in 10-men raids is: hemo rogue with debuff, or non-hemo rogue without debuff. 634 DPS (including buffed raid DPS) or 625 DPS (no buffed raid DPS applicable). In case you bring 2 rogues, the choice is simple: one with, one without hemo.

 

◊ s4dfish

Are you sure? Filling in the spreadsheet with gear form Hyjal/BT I come up with:

unbuffed (2017 AP, +24.15% hit, 30.84% crit):
11/28/22: 996 w/ hemo damage, 944 w/o hemo damage
11/26/24: 1030 w/ hemo damage, 978 w/o hemo damage
19/42/00: 1057 w/ hemo damage, 1002 w/o hemo damage

buffed (3783 AP, +27% hit, 43.87% crit):
11/28/22: 2309 w/ hemo damage, 2224 w/o hemo damage
11/26/24: 2380 w/ hemo damage, 2294 w/o hemo damage
19/42/00: 2389 w/ hemo damage, 2300 w/o hemo damage

If I'm missing something, let me know. I'm coming to the conclusion that Hemo is on par with combat, but not replacing it.

edit: hemo's normalized, my numbers are wrong.

 

◊ Vulajin

Repeated the test once more on Servants of Razelikh using [Arcane Forged Mace] and [Shortsword] (note: identical damage ranges, the only difference is speed). I made sure not to pick up Murder or Dirty Deeds, so neither talent affected these numbers.

First note that all tests are run at 1601 AP. Thus, if I'm using a weapon with average damage of 3, I would expect an average normalized Hemo to hit for 382.82, an average non-normalized Hemo with [Arcane Forged Mace] to hit for 311.35, and an average non-normalized Hemo with [Shortsword] to hit for 411.41.

Over the course of 20 Hemorrhage hits landed using [Arcane Forged Mace], I dealt 5479 damage, for an average of 273.95 per hit. If we assume that Hemo is normalized, this indicates an armor reduction of roughly 28.44%. If we assume that it is not normalized, this indicates an armor reduction of roughly 12.01%.

Over the course of 20 Hemorrhage hits landed using [Shortsword], I dealt 5477 damage, for an average of 273.85 per hit. If we assume that Hemo is normalized, this indicates an armor reduction of roughly 28.47%. If we assume that it is not normalized, this indicates an armor reduction of roughly 33.44%.

Indeed, it seems that Hemo has been normalized. In my opinion, this was necessary, because Hemo with a 2.7 speed weapon and 125% modifier probably would've been a little nuts. I'll patch this into my spreadsheet and come back in a bit with some comparisons.

(Also, thanks, amafi, for introducing me to the proper testing method.)

 

Azaziel

Well what enchants and gems you have will matter also. Agility is better than +hit for 11/21/29 build but not for 19/42/0 build. I know with my present gear the 11/21/29 build would outdo the combat build even before the 125% change. And there is no way a combat rogue would get all 10 charges of a hemo rogue while the hemo rogue should always get atleast some of his own charges.

 

• Vandemar

I just did a bit of testing with [Shortsword] and [Club] on a Blasted Lands servant, and saw Hemo hits of 320-322 with the 1.9 mace, and 321-323 with the 2.6 sword, all with 1793AP, which indicates that Hemo's now normalized.

This was on the current PTR build, 0.3.0(7521).