That isn't a good argument, considering this game is nothing but a math problem. If a spreadsheet calculates Ashtongue to be better, then it most likely is; especially since it doesn't model the effects of "energy pooling" - which puts the Ashtongue even further ahead.

Because I'm tired of people contesting the given explanation of sword spec functionality and providing no evidence or shoddy evidence to support their claims, I went to Blasted Lands this morning and did some solid testing.

The methodology was to first confirm that sword spec does function on a per-weapon basis; that is, your main hand can get sword spec procs only if your main hand is a sword, and your off hand can get sword spec procs only if your off hand is a sword. Once this was confirmed, I performed a substantial about of regular melee auto attacking using Slice and Dice (generating CP only with the special that would NOT proc sword spec) to confirm the combat log pattern upon a sword spec proc.

To perform all of these tests, I used two sets of weapons: [Vanir's Right Fist of Brutality] with [Blade of Savagery] for the non-sword MH/sword OH tests, and [Talon of Azshara] with [Tracker's Blade] for the sword MH/non-sword OH tests.

First test: confirm that non-sword MH/sword OH can only get sword spec procs from the off hand. To do this, I engaged a Servant of Razelikh and angled myself relative to the mob such that, with auto attack turned on, I still could not perform regular melee attacks, but I could perform special attacks. Then, I performed Shiv once and immediately had a sword spec proc:

After this, I performed Sinister Strike 100 times and experienced 0 sword spec procs. If Sinister Strike with a fist weapon were able to cause a sword spec proc, the chance of experiencing 0 procs in 100 swings is 0.59%. Thus, we can say with a high degree of certainty that with non-sword MH/sword OH can only get sword spec procs from the off hand.

Second test: confirm that sword MH/non-sword OH can only get sword spec procs from the main hand. I used the same positioning described in the previous test and performed Sinister Strike 8 times before having a sword spec proc:

I also performed one Eviscerate and one Rupture each, both of which caused sword spec procs:

I then performed 100 Shivs and experienced 0 sword spec procs. Again, we can say with a high degree of certainty that, with sword MH/non-sword OH, only the main hand may cause sword spec procs.

Third test: I now returned to the case of non-sword MH/sword OH and performed 1001 melee swings over the course of about 10 minutes. Note that my main hand damage range is roughly twice that of my off hand using either of the weapon combinations I identified earlier in this post. Thus, a main hand hit or crit should be readily distinguishable from an off hand hit or crit, provided both take place while under the effect of the same buffs. Note that my main hand speed with SnD up is roughly 1.92, and my OH speed roughly 1.08.

Here is an example sword spec proc event from this test (there are no damage events from the Servant of Razelikh in this snippet because he was in his "banished" state at the time):

What you see above, paraphrased:

-1.083 Vulajin's swing crits Servant of Razelikh for 1089. (MH swing)
-0.966 Vulajin's swing crits Servant of Razelikh for 491. (OH swing)
-0.197 Band of the Eternal Champion fades from Vulajin. (irrelevant)
+0.000 Vulajin gains 1 extra attack through Sword Specialization.
+0.001 Vulajin's swing crits Servant of Razelikh for 402. (OH swing, 0.967s since the last OH)
+0.351 Vulajin's swing crits Servant of Razelikh for 1168. (MH swing, 1.434s since the last MH)
+0.680 Vulajin's swing hits Servant of Razelikh for 472. (MH swing, 0.329s since the last MH, 1.763s since the prior one)
+0.947 Vulajin's swing hits Servant of Razelikh for 217. (OH swing, 0.946s since the last OH)

Looking at this chain of events, it is possible that either the OH swing at -0.966 or the OH swing at +0.001 caused the sword spec proc. If the swing at -0.966 caused the proc, then the swing at +0.001 must have been the extra swing. However, if this was the case, the two MH swings within 0.329 seconds of one another are completely impossible to explain. If, on the other hand, the OH swing at +0.001 caused the proc and the MH swing at +0.351 was the extra swing, then the MH swing at +0.680 was on the normal MH timer (within reasonable tolerance of my actual MH speed) and is perfectly explainable.

Here's another example:

-1.018 Vulajin's swing hits Servant of Razelikh for 184. (OH swing)
-0.435 Vulajin's swing hits Servant of Razelikh for 464. (MH swing)
+0.000 Vulajin gains 1 extra attack through Sword Specialization.
+0.000 Vulajin's swing hits Servant of Razelikh for 202. (OH swing, 1.018s since the last OH)
+0.115 Vulajin performs Slice and Dice. (note that there's no aura gain event, I was just refreshing the buff, so my speed is constant)
+0.117 Vulajin's swing crits Servant of Razelikh for 896. (MH swing, 0.552s since the last MH)
+0.982 Vulajin's swing crits Servant of Razelikh for 433. (OH swing, 0.982s since the last OH)
+1.348 Vulajin's swing hits Servant of Razelikh for 436. (MH swing, 1.231s since the last MH, 1.783s since the prior one)

Only the OH swings at -1.018 or +0.000 could have caused the proc. Again, if the -1.018 OH caused the proc and the +0.000 OH were the extra attack, then the MH at +0.117 is completely impossible. On the other hand, if the +0.000 OH caused the proc and the +0.117 is the extra swing, then the MH at +1.348 is reasonable. I won't waste space with more examples, there are plenty in the combat log, which I will attach. Suffice it to say that the pattern is consistent: the sword spec proc event always immediately precedes the swing that caused the proc, which is followed by the extra swing, which is performed by the MH even if the OH caused the proc.

Fourth test: I returned to the case of sword MH/non-sword OH and performed 756 melee swings over the course of about 7 minutes. Again, in this test my main hand damage range was roughly twice that of my off hand. In the interest of saving time, I will not go in-depth into examples for this test; I have provided the combat log below and invite any skeptics to open it themselves and verify that sword spec procs from the MH follow the pattern I have identified in the previous test.

Attachments: In the attached log files, "mhsspec.txt" contains the test of sword MH/non-sword OH using white attacks (listed above as "fourth test"). "mhsspec2.txt" contains the text of sword MH/non-sword OH using Sinister Strike, Shiv, Eviscerate, and Rupture (listed above as "second test"). "ohsspec.txt" contains the test of non-sword MH/sword OH using white attacks (listed above as "third test"). "ohsspec2.txt" contains the test of non-sword MH/sword OH using Sinister Strike and Shiv (listed above as "first test").

Please keep any questions to this thread.

I read all of that post, Vulajin, and it was very informative. My only complaint about it was that it lacked the cool little graphics for BUSTED, CONFIRMED, or even PLAUSIBLE in some cases that they use on Mythbusters. I think all future extensive reports of testing should include them!

Seriously, though, thanks for the testing! I hope everyone can now put to rest their doubts about the way sword spec works.

Typically speaking I would think being top DPS would be a viable enough reason to be in the raid.

Reading a bit more I'll expand upon that a bit more: I disagree that being a buff sponge isn't enough of a reason to bring a rogue as we still top the DPS charts. Consistently on boss fights I see rogues as 1-2-3, and in my guild I see saw with another rogue, all depends who was a bit faster onthe click for that fight.

If you'd like to see a bit more "reason" to bring a rogue I would suggest looking at mob debuffs as Aldriana suggested. I remember when expose armor was supposed to be % based and stack ontop of sunders.

Turns out that's a bit overpowered, but I would suggest something along that line. I'd also like to see some multi use poisions-especially since feral druids get to poision inthe expac.

Well, your melee group already needs a Warrior and a Shaman. Now you have Ret paladin, DK, and Rogues fighting over the last three slots. Ret offers more to the raid than a rogue, so bring one of those, and depending on what DKs can bring, they may take another slot. That leaves 1 raid slot in a 25 man raid for a rogue. In order to be worth that slot, we need to offer something really significant. A minor increase in dps is not equivalant to a raid-wide or party buff.

The issue, as I see it, comes down to the rogues ability to wipe aggro.

I had the same fears and concerns when TBC was being discussed and talents were previewed-I was disgusted with the lack of tweaking on rogues (still am) and appalled by the fact that other classes got bigger, cooler, more versatile moves/buffs. When they announced some of the changes to DPS warriors I threw a holy fit and almost quit.

Maybe I'm wrong-it's possible, afterall 99% of the posters here seems to be much more intelligent and congnizant than I am about the game's mechanics (<3 you guys btw) -but the fact is fury warriors, enhancement shaman even kitty druids tend to have a 'threat cap' versus their DPS, meaning there is a point (and not a super high one in my (limited) experience) where their DPS has to be scaled back to avoid dying.

At this point in time I'm not super concerned about it.

You realize the changes to fury warriors were massive nerfs don't you?

What do you guys consider imp.exposed armor?

It looks pretty good..But i'm not a rogue so.

Seems like that brings something to melee =O

It's very useful, but we can't use it if a warrior is tanking, since it prevents the warrior from being able to use Sunder/Devastate.

At this point I am not super concerned about your opinion, because you don't have any appreciation for the issues that rogues currently face.

Let me be blunt here, you don't have a clue. Your mid-level T5 raiding experience does not give you an accurate view of what current class issues are. You could kill any boss you have faced to this point with any raid make-up that meets some basic X Tanks, Y Healers, Z DPS requirements, synergy and group stacking is something you can do for fun, but it isn't required on any level. Also, you apparently have no idea about how group synergy works, especially when you have to start picking people to drop.

Right now the best possible melee group is:
Arms Warrior
Enhancement Shaman
Ret Paladin (if Horde)
Rogue
Feral Druid/Rogue/BM Hunter

Now, most guilds don't run 2 full melee groups, however they run some tank/hunter group of some variety.

A rogue is only viable in that group, so let's say a Death Knight is also only viable in that group. Who do you drop? You can't drop the Shaman, because that hurts everyone due to Unleash Rage and Totems, you can't drop the warrior because everyone loses Battle Shout, and right now the contribution of a BM Hunter is probably great then that of a rogue if you have the Feral in there and the Shaman is Twisting so the hunter get's GoA, because on top of their personal damage the hunter is tossing up Hunter's Mark and giving 3% more damage.

The blunt truth is that rogues are not actually the highest DPS, we are the highest on the meters, there is a HUGE difference. Try taking the buff values and subtracting them from rogues and applying them to the buffers, same with debuffs, you quickly see that we are very, very easily marginalized, if a Death Knight brings any sort of Buff/Debuff, odds are they will out do rogues, thus we need one to compete.

Right now a rogue is essential in no way shape or form. Threat isn't a material issue on any fight in Sunwell to this point, and never really was in T6 either... but you experience through mid-T5 wouldn't tell you that, so you still think our Threat Dump is king... oh yeah, did you forget Hunters have one too... that they can use a hell of a lot more? Our threat dump is nice, but it is basically an added bonus at this point.

What we need is something to make us essential, and something that doesn't penalize stacking. Right now there is no disadvantage to bringing 2 BM hunters in addition to 1 Survival, the stack Mark faster and they all buff each other, hell a 3 Hunter, Feral, Shaman group probably comes close to the DPS of the melee group and is lower maintenance.

We need a group buff, something equivalent to Battle Shout/Unleash Rage/Ferocious Inspiration/Leader of the Pack/Imp Sanctity Aura/etc. all of these are group only buffs, that require a spot in the group to reach the ideal, rogues need one that stacks so that we can hold onto at least 2 raid slots. A boss debuff would be nice, but it isn't enough to get us in the melee group, we need a buff that everyone else wants and needs.

The bottomline is we have gone from benefiting from buffs to requiring them to stay competitive. If we don't have an Enhancement Shaman and a DPS Warrior, we might as well not bring rogues to raids, we need to be able to provide synergy as well as benefit from it, we bring nothing else to the raid, and at this point the marginal difference isn't enough to make us required, especially if Death Knights can provide something more.


EA is Great, when you can use it, which is when you don't have a Warrior tank... which isn't very often.

This thread is slowly starting to give me visions of raiding returning to the early days of Kara, where a Rogue would get brought in for Shade, and kicked right out for another Warlock right after.

Our interrupt might be (arguably) 'the best' but if it's only important on one gimmick fight per raid instance, it does not justify our raidslot.

To be entirely honest, I don't neccesarily want a group buff. Some additional utility poisons might be nice, but I actually, the more I reflect on it, really kind of like the position we've been in for most of TBC raiding: If you don't bring anything to the raid but damage, you had better bring the most. I think it IS possible to do that without giving us additional synergies, and I think it can be done by creating something along the lines mentioned earlier - a stacking self buff that rewards time on target. Make it the new 51 combat talent, call it "Savage Concentration" or somesuch. Make it slow to stack up but easy to refresh - but only if you can stay on target, thus making it useless in pvp/trash scenarios - and at high stacks (say, on average possible after 3 minutes on target), have it be enough of a damage increase to make rogues unquestionably way ahead on damage.