As for the dps comparison of rogues vs. hunters I believe we need to pick up the old discussion about glancing blows.

This is basically the problem of non-Combat specs, I agree.

Many combat rogues switching to a (badge) fist + (whatever) sword spec, dropping their 5 points into Vile Poisons without thinking twice should've woken up some heads at Blizz HQ.

The Mutilate build is rather dependant on poisons. The damage caused can be awesome, but it doesn't show the scaling up of Combat Potency.

But now for the hemo rant..
Hemorrhage was basically ruined by normalization. Sure, it works nice at lower gear, but from the day this change was announced, it was proven over and over how the "new" hemorrhage had rather backward scaling with the equipment available. Unnormalized had a potential to scale rather awesomely, but even with the badge and Sunwell gear available now, the personal DPS would've been more or less (slightly less) in line with a Combat build, while adding good raid support (wasn't there a blue post about turning Sub into more of a Support tree?), though still limited to the only 1 hemo rogue/raid.

At the same time -all- rogue itemization was simplified. Was that the reason? Assassination, Combat and Subtlety rogues now all find their performance increases in exactly the same gear. Before Mutilate existed, Assassination rogues favoured Crit by a fairly large margin. Before 2.3, Subtlety rogues favoured AP by a fairly large margin. Right now, ALL rogues are after roughly the same gear, which used to be the gear that especially Combat rogues preferred. Easymode itemization, one size fits all. Vanilla WoW wasn't pre-TBC, instead, the singular flavor game is what it is now.

This is very much not the case. The reason most rogues, reguardless of spec, are after the same gear is because the best-in-slot sunwell options for each slot are far and away the best choices reguardless of spec. There are dramatically different weights attached to hit and haste depending on the presence of combat potency, any build without it will favor those stats much less than one with it. These differences are, unfortuantly, marginalized by the lack of variety in DPS leather drops. The only slot in sunwell that actually has two viable options in sunwell is the chest or the neck and they each require a crafting profession to introduce a second option.

That said, I do agree with your point. Presently combat potency is the driving force behind the differences in stat weights and the differences scale with gear. It would be nice to see some changes in the trees that introduce similar mechanics to assassination and subtelty that favor different stats which would result in more dramatic differences between optimal gear sets. However, all that would be for naught if blizzard continues to itemize all DPS leather with pretty much the same stats.

Energy per 5pt SnD: 225
5 point SnD = 30.45 seconds
Cost per second of SnD uptime @ 5 points: ~7.4 energy

I'm assuming that every point of energy you spend is either spent on combo points or finishers, and every combo point is either spent on SnD or Rupture. I'm also acting as if all the energy spent on combo points is 'spent' on the finisher that consumes the combo points. This means that 5 sinister strikes + 5pt SnD 'costs' 225 energy, since all the combo points you generated are consumed by the finisher.

If SnD is 'free' the cost drops to 200 energy, or ~6.6 energy/sec for uptime. Note that this isn't a straight 'cost', since that energy also produces damage through sinister strike (I account for this later in my calculations).

This might be wrong, though. I'm not a math major.

Ok, I don't think this makes sense. You can't just assign the "cost" of SnD as 6.6 energy/sec, then say that refreshing it early loses you that energy. It doesn't. Either way, you have to do 5x Sinister strike + SnD, so no matter when you refresh SnD (early or late) the total amount of energy you spent remains perfectly constant.

I see now how you are trying to handle it, but it's an artificial and inaccurate way to do so. If you refresh SnD 1 second early, you aren't magically "losing" 6.6 (or 7.4) energy. Energy doesn't come in (and isn't spent) in a continuous fashion like that. You spent all the energy for your SnD uptime in the first X seconds of the cycle, where X is the amount of time it took to get to 5 CP. So after you built to 5 CP and used SnD it doesn't matter when you refresh it, you still spent the same amount of energy on achieving SnD.

As for the bit where you took that "loss" of energy and applied it as lost Sinister Strike damage, well, that makes even less sense to me. Unless you let energy cap out, you didn't "lose" any energy. And if you are refreshing SnD early, it is usually because you already have enough CP for your chosen rotation and you don't want to cap out.

Basically, what I am saying is that normalizing the cost of SnD over the entire cycle given the expected time of SnD is a fallacy. Refreshing early loses you nothing unless there is a different finisher you would have used instead. I can't explain it any better than that without launching into the whole explanation of rogue cycles and energy calculations, but I don't need to do that; Vulajin's article does that already.

Just clarification on this: Felmyst parses show lower DPS for hunters than rogues because if they DPS during flight phases they don't benefit from pet DPS/FI (and this difference seems to be amplified somewhat by the way WWSScoreboard calculates DPS compared to WWS).

Assuming the figure you're interested in is how much damage each class can do to Felmyst herself during the fight, open the WWS report, browse to the mob named Felmyst, look at her breakdown of incoming damage, and sum up each hunter with his pet (the breakdown does not automatically combine them for you) - the handful of parses I've done this for have put BM hunters far in excess of their rogue guildmates (predictably of course, the rogues don't get to do any damage during flight phases) - often in excess of warlocks/mages too, but I imagine those were cases where the casters focussed more on AoE.

Also, I don't know why you say rogues are topping the list of physcal DPS damage dealers, WWS Scoreboard has the top 8 hunters placing above the top 1 rogue for DPS (again, by their own definition of DPS).

If rogues and hunters did the fight exactly the same, with neither doing any dps during air phases, it would be exactly like brutallus, just with encapsulates instead of burns/slashes, and a non-demon boss.

edit to keep this post at least slightly on topic for the tread: it's always been my feeling that a troubling aspect of the rogue class is how much you need to gear for +hit - having not seriously played a raiding rogue much this is probably quite uninformed, but I can't help thinking that that much attention put into stacking one stat causes issues. On the hunter side, we mostly want to hit cap, but due to a much lower cap we can afford to gear for more interesting stats to stack, such as ArP or haste. Obviously blind stacking or agi or haste or ArP isn't what we try to do, but as an example this means that our end-sunwell gear-set features a very large amount of passive ArP, in the 1600-2200 range. What that amounts to is that if optimal sets for rogues don't feature similar amounts of ArP (because of increasing returns from +hit along the way), the absolute difference in DPS between rogues and hunters increases as target armour goes down.

We've already had the discussion about why mean rogue DPS can't be sensibly compared to that of classes with utility specs ^^;

Thank you for the explanation of the ArP point though, I completely neglected to consider that rogues have effects that proc ArP other than just 300 from Madness.

Pets are generally around 20% of a BM hunter's damage (bit more if you get BS + UR, bit less on fights like Felmyst).

Not on a constant DPS in time fight like brut, no, but on felmyst? If the flight phases put BM at a dissadvantage because they loose their pet DPS and FI uptime, doesn't this put survival (which i assume is the utility spec, correct me if i'm wrong here) on much better footing comparitatively since their losses would not be as great?

Still, even on brut the difference between affliction and destruction DPS is pronounced enough to actually appear on the class breakdown as a distinctly different mean. This is not the case for hunters whoes DPS curve appears to be a uniform bell curve around a single mean in both fights.

Well, of the top 15 parses listed, only 1 is a survival hunter, at #10. Losing 23% of your dps for half a fight isn't enough to close the gap between survival and BM all that much. And if the survival hunter happens to be grouped with the BM hunters, he loses a flat 3-9% damage during flight phases too.

The difference in histograms for hunters vs warlocks (and fwiw I don't see that much of trend to a secondary peak for warlocks, unlike say the massive irregularity of the warrior histogram) is I think because the top survival hunters already outdps a lot of BM hunters, so the difference between a mediocre BM hunter such as myself and a really good survival hunter could be as little as 100 dps. However the difference between me and a survival hunter stuck in the tank group without a DST is easily 1000-1300 DPS, and that's what screws with the averages, since it's not an uncommon setup. Hence looking at the averages for top 15-20 of each class is more representative of what a class can do (with the possible exception of legendary bows, since as you mention having them isn't really relevant to progression).

We should probably continue this argument in PMs though, instead of cluttering a rogue thread further.

I agree with Aldriana in that having some utility in a raid setting would be nice. Currently, rogues are hampered by having to bring along a pocket arms warrior and an enhancement shaman before the rogue can be competitive in dps. If your resident enhancement shammy was a no show, you'd cringe a little because now you'd be fighting to stay within the top 5. By giving a rogues a little utility, raid leaders can be a little bit more flexible with raid grouping and not worry about arms warrior and enhancement shammy no shows. The rogue will drop in dps but at least the rogue would be able to provide something else to the raid.

The hemorrhage idea was good before the nerf. I hope Blizzard learns that static debuffs are not the way to go in WOTLK. Another idea would be to have the vanish aggro drop affect all party members. I also wouldn't mind interchangeable/duplicate buffs/debuffs. That way you'd have a little bit more flexibility in composing a raid. Suppose rogues also had some sort of EW debuff at the expense of personal dps. Then a raid leader can still run an optimum raid if a SV hunter doesn't show up.

First off I would like to introduce myself, my name is Dillinja 70 alliance rogue on the Vek'Nilash - US server.

Now to my quote and question. Lately I been taking the time to start breaking down my class. I am a TBC raider I never had pre-tbc experience so I am learning a lot. I am a combat sword rogue, and you mention more hit is essential for us Sinister Strike rogues. Where should I be at hit? I currently have 299 and 219 with hit food. I sacrificed my hit for my T5 META helm, my other helm is the Grim grin faceguard.

Now my main focus question what should I have my hit at and how can i balance my Hit/Agi/AP/Haste/EXP/etc.?

Edit: I forgot to mention I have the T6 4 set, and I am being out dps'd by rogues with equivalent or slightly lower gear. I use Agi elixirs on farm content and Flasks on new content. Hit food is always on and DP on my OH for every fight except ROS and IC. Any suggestions for this to?

Here was last nights Hyjal WWS Wow Web Stats

I still don't really understand WWS.

For the hit whats the least amount should I have? From your recommendation, what would be the least amount you use.


EDIT: Also could you give any tips when i should use CDs, Drums, e rush, etc.

So your saying T4 is way better than ZA helm? Is T5 superior over T4 and ZA helm?

All these things are covered by the PvE DPS article and the first post in this thread. Please take the time to read them.

I think the quote is self-explanatory, and the second portion of your query can be addressed via the spreadsheets that are widely available. Please don't post if your only question is whether, when I wrote something, I really meant it.

Ok many of you have state that a rogue for wowlk will need some sort of synergy.. this being said i came up with an idea that may have been discused before or not i dont know nine pages of roughly the same thing is hard to memorize. But in general this would be a talent that would make the Rogue a vaulable part of the raid offering more than dps without affecting pvp or arena in the slightest.. Also before i go into it for those of you who are talking about unballincing the arena's, Blizzard could simply make that ability non viable in an arena or pvp, I.E like lay of hands is for a Paladin or etc. But regardless of this fact i propose instead of people trying to bring back other spec or builds to be more pertenant why not build apon an already combat/ assasination platform by adding an ability in the combat tree that offers a raid buff for spell and mellee haste. For instance lets call this ability "Unbridaled Speed" this ability has up to 3 talent points assigned to it each one boosting the affect by 2%.. The affect of this would be that a haste effect of 1% would be added to the raid per every 2.5 combo points roughly. So a rogue with only one talent in this tree will be able to take a 5 combo stack and use this to increase the entire raid spell haste and mellee haste by 2% for 15 seconds. But a rogue with 3 talent points in this area would be able to increase the Entire Raids Haste by 6% for 15 seconds with combat points. Not only would this benefit the rogue but would vastly improve raid dps, it would also make it a viable option to bring up to 3 rogues to one raid. For instance if a Rogue were to use a 3snd/5r rotation and you had three Rogues doing the same rotation they would be able to take on fo the rotations and instead of hitting Rapture hit Unbridaled Speed.. and have the three rogues switch out rotations to keep the 6% haste effect up the entire boss fight.. Also to make sure that you would need more than one rogue to do this you could give the ability a 30 second cd making it viable to bring three rogues for the fight to keep the spell haste bonus.. Plz note the idea was based of a rogues mane ability use in raids anyways and for one that would benefit the rogue as well as the raid at the same time.. also plz note that this ability would still stack with Drums of Battle and all other haste effect as it is now.. Even in single combat this could be very nice for a rogue increasing there base haste increase from 30% to 36% for the entire raid.. But alas its just an idea.

I'd expect to see the stack limit on Hemo increased or lifted in WotLK. With the introduction of yet another melee class, it needs it.

Even a talent-based/modified poison would be a nice addition; something like,
"Enfeebling Poison - Increases physical damage taken by the target by 1%. Stacks up to 5 times. 30 second duration."
Sure, not very original. You could design it to utilise some interesting new flavour/mechanic; but the concept is there. So just like Blood Frenzy; you sacrifice a little bit of personal DPS to boost the raid.

Alternatively, give us a reason to use Envenom. If Envenom applied such a debuff, it would really add a bit more versatility into our DPS cycles.

Also, with the increased bleed damage modifiers in WotLK, Rupture is looking even more attractive.

EDIT: I'm 100% correct in assuming the Ashtongue Talisman is complete garbage for a Combat Dagger Rogue, yes? The only way I could justify using it would be with a 5s 1r cycle or something ridiculous... and it's still doesn't appear worthwhile.

Posting this here because it is relevant to both rogue spreadsheets and the rogue community at large. The following WWS snippets are from my Brutallus parse tonight. I run with [Vanir's Right Fist of Brutality] and [Blade of Savagery] and 64 haste rating, so I have speeds of 1.78 and 1.00 under Slice and Dice, respectively.

The above, paraphrased more clearly:

-1.314 Vulajin's swing hits Brutallus for 574. (glancing) (MH swing)
-0.996 Vulajin's swing hits Brutallus for 387. (OH swing)
-0.653 Vulajin's Sinister Strike crits Brutallus for 2724. (irrelevant)
-0.316 Brutallus suffers 339 damage from Vulajin's Rupture. (irrelevant)
-0.220 Vulajin's swing hits Brutallus for 271. (glancing) (OH swing)
+0.000 Vulajin gains 1 extra attack through Windfury Attack.
+0.127 Vulajin's swing hits Brutallus for 788. (MH swing)
+0.143 Vulajin's swing crits Brutallus for 1979. (MH swing)
+0.462 Vulajin's swing hits Brutallus for 351. (glancing) (OH swing)

What do we learn from this log snippet? First, we know that WF can only proc from the main hand, so the extra attack must have been caused by the MH swing at +0.127, and the +0.143 MH swing must have been the actual Windfury attack. Thus, we see that Windfury extra attacks follow the same combat log pattern as sword spec extra attacks; the extra attack notification immediately precedes the attack causing the proc, and the extra attack itself immediately follows the attack causing the proc.

We now consider the following snippet:

I'm not going to paraphrase this entire thing. The two sections of interest are where sword spec and Windfury procs follow one another back-to-back. So the question is, which ones are associated with which extra swings? Well, first note the sequence at 51.532. Our existing knowledge tells us that the attack which causes an extra attack proc will be immediately preceded by the extra attack notification. Thus, the 1478-damage MH swing must have caused the Windfury proc. The resulting swing must have been an MH swing, based on the mechanics of Windfury; however, the immediately following attack (at 51.532) was not an MH attack, but rather an OH attack. Thus, it might not necessarily be the case that the extra swing always immediately follows the swing that caused the proc. We can obviously tell that the OH swing at 51.532 caused the sword spec proc (since only my off hand is a sword). Both extra swings seem to have come together at 52.009.

So, it seems the extra swings caused by the procs are delayed a short amount before all proccing together simultaneously. We can apply this knowledge above to the extra attack procs at 49.605 and 49.836. It's clear that, in that case as well, the extra attacks are procced by two separate attacks. This is not a case of proc-on-proc behavior. However, this knowledge enables me to search through the logs with a better idea of what I'm actually seeking. This yields:

Here you can see the OH swing proccing sword spec at 11.273, and the MH swing proccing WF at 11.399. So the question is which attacks are extra attacks. Well, the first MH swing in this snippet is at 9.942, and there are only three MH swings that could be the "next scheduled" MH swing (based on the normal swing timer) -- 11.399, 11.588, 11.835. At least one of those must be a regularly scheduled swing, not an extra attack. If the swing at 11.399 is the regularly scheduled swing, it comes only 1.457 seconds after the previous MH swing, which is unusually fast. Thus, either 11.588 or 11.835 must be the regularly scheduled swing. This means that the swing at 11.399, which caused the WF proc, was actually an unscheduled swing, or in other words, an extra attack. Well, it couldn't have been a WF attack, since it caused the WF proc, so it must be a sword spec proc. Conclusion: sword spec swings can proc Windfury swings.

Inspecting the rest of the log fairly thoroughly indicates zero instances of back-to-back Windfury procs; in other words, a Windfury proc can't proc a Windfury proc. As for whether sword spec procs can proc sword spec procs? Well, an inspection of my previous combat log on sword spec procs will show that no, they cannot. Can Windfury procs can cause sword spec procs? Since my main hand isn't a sword, I can't check this using myself; but I can use my fellow rogue, Bandire:

The MH attack at 16.440 causes a WF proc. Then there are MH swings at 16.574 and 16.909, both of which are too fast for his normal MH swing timer (he uses [Blade of Infamy], and DST was not active during this snippet). His MH attack at 16.574 caused a sword spec proc. The sword spec swing must have been the MH at 16.909, which leaves the MH at 16.574 as the WF proc. Conclusion: Windfury procs can cause sword spec procs.

Now the real question: can procs chain back and forth? In other words, can a swing proc Windfury, which then procs sword spec, which then procs Windfury again? Well, this is harder to check, but I did a pretty thorough search of Bandire's logs and couldn't find a single instance of more than two chained procs. That is, there were plenty of instances of a Windfury swing proccing sword spec, and plenty of a sword spec swing proccing Windfury, but none of a chain longer than that. I invite anyone to try and find one. Obviously this is a little bit of a pain, since the likelihood of seeing sword spec > WF > sword spec is only about 0.05% per MH or OH swing even if you're hit- and expertise-capped, and the likelihood of seeing WF > sword spec > WF is only about 0.2% per MH swing.

If I get a chance, I'll take a shaman friend out to Blasted Lands and beat on Servants for a long time. Still, reasoning from the above knowledge, it seems reasonable that extra attacks are programmed to be able to proc other types of special attacks (e.g. sword spec can proc Windfury and Windfury can proc sword spec), while not being able to proc the same type of special attack (e.g. sword spec can't proc sword spec). But then, this doesn't explain why I've yet to find an instance of an extra attack proccing an extra attack that procs another extra attack. I dunno, hate this complicated shit. >_<

The WWS is available here: Wow Web Stats

Comments and discussion are welcome.

Did you happen to count the number of times that a Windfury -> Sword Spec occurred in the logs you were examining? Naturally you would expect there to be a 20% chance for a second Windfury to proc given this event occurs, which suggests that if you can find a significant number of these events occurring without the extra Windfury proc occurring, it's probably reasonable to assume it can't happen.

You only need to produce 24 such occurrences with none resulting in the second Windfury proc to obtain a 99.5% certainty.

I'm referring to my previous test, several pages back in this thread, wherein I determined how sword spec procs appeared in the combat log using a combination of [Talon of Azshara] and [Tracker's Blade]. No sword spec chains occured. You could also confirm this using Bandire's log from the WWS I linked, but my test log would be much easier on your eyes.

I did not. Frankly, looking at WWS logs hurts my eyes. If I find time tomorrow, I'll take a look through it again and let you know. Otherwise, if someone would like to do it in the interim, please feel free.

I am looking through it right now, jotting down all WF/SS procs happening within roughly two seconds of eachother.

I doubt this will happen but I'd love to see an end to the need for specific groups for specific dps types. I can't see that it would be imbalanced if most buffs that were currently group-wide were changed to raidwide. Imagine a single Windfury Totem benefitting all melee, wherever they were, who could also receive Grace of Air, Tremor and Strength of Earth, Battle Shout and Commanding Shout, Ferocious Inspiration and so on. Shadow Priests' mana regen would also benefit Enhancement shamans and both Retribution and Protection paladins.

The main benefit would be that a raid could take whichever dps they wanted and not have to worry that somebody (or a few people) would be sat outside their group and thus doing significantly less dps. Boss encounters would simply be tuned accordingly, and it would make no difference at all in the arenas, so I really don't see a downside to it.

Other than the ability to over-stack buffs?
If every class got every available buffs (kinds of like when you could drink down a bunch of elixers and food buffs) then dps would be well over the needed amount for the encounters.
Some of the encounters are really amazingly balanced in that aspect now, and doing what you just pruposed would totally throw that off.

*deleted*