Thanks for the heads up. I haven't noticed a significant increase/decrease between the two builds myself yet, but I'll keep an eye on it for sure.

So does that mean that there really isn't a point on speccing 51/7/13 anymore, even with the 4 piece?

According to Aldriana, the difference between the two isn't very large (it wasn't large to begin with). However, there are a few things to remember about the 51/7/13 spec. The 51/7/13 spec assumes you have a very high Rupture uptime (you should be striving for this anyway). If you cannot maintain the high Rupture uptime due to fight mechanics or your own play, then the spec will lose out because you're putting points into damage you don't do.

That said, proper use of the spec will allow you to continue to do significant damage through Rupture to your target when you are not actively on them. This is all do to timing and how well you are able to predict what is happening in the fight.

This is a case where you should be able to look at your own Rupture uptime and determine whether or not the spec is right for you. In any case, you can only lose DPS when compared to 51/13/7 and not gain it.

I suppose that may be why I gained DPS by swapping the points in Ruthlessness and Lethality into picking up Serrated Blade for over a 100 DPS increase as CQC. As long as Rupture remains up for a large portion of the fight as it should always be then this DPS gain is realistic and not an error on my end with the spreadsheet. I did notice though that the gap b/w these two specs got smaller when I selected MR as one of my trinkets and other items with ArP on them, so the ArP portion of Serrated Blades may be playing a larger part at the moment.

It has been extensively tested that as CQC combat, or any combat spec, speccing into serrated blades is invariably a DPS loss, this was shown even when the incorrect rupture damage modifier calculations were being used (although I'm not sure the incorrect formula was ever using the combat sheet), and is even more clear now.

So, now that 3.2 has been released and more of the new gear will begin to roll out, I'm curious which spec (between 18/51/2 and 5/15/5) you guys think will scale better for combat types? Currently I'm using the vile setup with CQC, and even in just shy of full BiS ulduar gear, for me at least, it pulls a bit ahead.

According to most of the posts I've read, 15/51/5 seems to be the final resting place for combat rogues in full BiS though, and I'm curious if people think that will remain true in 3.2?

The mutilate spreadsheet produced by Aldriana is still using the incorrect Rupture formula and as such, it is overestimating the benefit of Serrated Blades. There's been quite a lot of talk about this same thing around these forums over the past couple of weeks. I don't know the exact errors in the calculation off the top of my head, but I remember seeing it be said that 51/7/13 at best will give similar DPS to 51/13/7 and on average be an absolute DPS loss. Shame I can't comment myself on exactly what the difference would be.

EDIT: As it turns out it may actually be the way Serrated Blades' effect on ArP is being calculated that accounts for the differences. Not entirely sure myself right now. But the consensus, it seems, is that 51/7/13 is inferior and the current state of the Mut. spreadsheet isn't accurately representing this.

What is being miscalculated dealing with the armor penetration on SB? The tooltip looks fairly straight forward, not sure what could be acting up with it unless it has some strange effect on the value of arpen rating or something.

If I recall correctly, can't seem to find the post now, Serrated Blades actually lowers the amount of ArP needed to cap it. I can't personally see how this would effect 51/13/7 vs. 51/7/13 though now that I think about it, unless the ArP cap was a concern before switching to SB. My previous post was based off of what I heard being discussed here on the forums and I have not personally checked the formulas in Aldriana's mutilate sheet, but I don't think the mutilate sheet has been updated since the formula for calculating ArP was discovered. It is however implemented in Aldriana's combat sheet.

The bug was whether rupture bonuses apply additively or multiplicatively. My sheet erroneously has them as multiplicative, when they're supposed to be additive.

Losing the T8 rupture crit, will gemming for AP be better than agility for mutilate again?

I'm sure that he was speaking in terms of pure DPS, and not the added survivability that agility gives.

Without the T8 4 piece bonus, AP and agility will become very similar again to gem, as they were before the set bonus.

Regarding gemming in 3.2, is it worth to make the switch to ArP right away? In my current setup with T8, Mjolnir and mostly U10 hard mode gear, geming Agi still comes out about 5 dps ahead of gemming ArP/Agi-crit in Aldrianas sheet updated with the new gems. On the premise that I don't want to regam all my gear as soon as ArP tips the scale and becomes better, would it be better to consider it a worthwhile investment to gem everything ArP right away instead of having to switch as soon as the ToC loot starts coming in?

I'd argue that if ArPen was 50 dps better it would not be worth gemming for ArPen. Or for that matter 150dps. Some would argue the opposite. ArPen is usually to you when something is about to smash your face or when an add is about to die and the crit could be enough to push it over (while an extra 2% of ArPen won't).

All the "ArPen is better" logic is predicated on the very true fact that smashing something consistently with certain gearsets will result in higher steady-state DPS than gemming for agility.

For ArPen to actually >be< better, however, it needs to make up for the fact that agility has merit ArPen doesn't. And therefore sacrificing 1-2% of your maximum "steady state dps" might well be worth doing. Sacrificing 5% of that, perhaps not.

There are people who gemmed AP in Tier 7 when min-maxing for dps was a weird choice 95% of the time. I supposed they'd argue that the 5% is what matters. But until you know there is no chance you'll be hit on the fight that's important in Coliseum or that it won't rely on add killing, you might think twice about gemming ArPen. As always, your mileage may vary and if you are a slave to the spreadsheet, switch to ArPen when it tells you to. Because there are fights like XT and Patchwerk.

So is this basically the same for 3.2? I currently have jc and mining and finally decided to switch mining up for something more beneficial. I'm currently leveling engineering mainly because 340 haste for 12sec seems rather nice (among other things) but, I just wanted to make sure.

AFAIK that is updated for 3.2. In 3.1 All the values were in the 60-66 EP range and Engineering wasn't really competitive.

Disccussing agility as a survival stat, is like considering stamina to be a valuable stat, if we proceed that way there are probably a lot more situations where it would be worth matching a blue socket, and it would alter the EP values of gear greatly. However the community at large has never really considered this a valid idea,

The reason for this is that we are brought to raids to put out as much dps as possible and choosing stats for their survival value does not contribute to that, healers are brought to the raid to heal us, and everyone else, and as long as it is possible for them to do that, there is not very much to gain from these survival stats unless there were fights implemented in the game that required high stamina to avoid being killed instantly, pre nerf Freya +3 might have been close. Additionally as rogues we are singularly capably of reducing the damage we take from raid damaging effects through feint, evasion and cloak of shadows and as a result survival stats are even less useful for us.

I sort of have to disagree with this. It's true that agility does give some advantage in terms of survivability. And it's certainly true that that survivability is worth something. But I think you overestimate how much it's actually worth.

Resocketing full agi to full ArPen is typically a swing on the order of 300-400 agility, depending on your exact gear. This is, after dimininshing returns, something like 6% dodge. So how often do we need to take potentially lethal damage before 6% dodge is worth, say, 100 DPS?

Well, 100 DPS is about 1% of what a fully ToC geared rogue is going to be doing, and a death costs you around 50% of your damage for a fight; thus, you need to have your life saved by that dodge once every 50 fights or so to average the same amount of DPS in the end. And given that 6% dodge is only 6% likely to save you from that killing blow, that means you need to be in a situation where that dodge matters once every 3 pulls. Admittedly, dying can be bad for other reasons, but as many fights has significant DPS checks to them these days, doing significantly suboptimal damage can also be bad. And even if we suppose that dying is 3 times worse than would be indicated by the average case damage computation, we're still looking at needing to take potentially lethal damage every 9 pulls. And given that your dodge rate is probably only ~50% anyway, that means you're currently dying to direct physical damage once in every 20 pulls if this is to be worthwhile. And I have no reservations in saying that if you're dying that often, you're doing it wrong.

So: yes, agility does have survivability benefits. Yes, they are worth having. But I personally think the amount of DPS it's worth giving up for 1% dodge is in the small single digits (i.e., less than 5). Your may think differently, of course, and that's fine; but I would take arguments like the quoted one with a grain of salt.

On a related note: stamina also has value in terms of survivability, not so much in terms of absorbing usual raid damage, but in that some fights (Freya and Algalon come to mind) have "threshhold" HP numbers where if you have at least a certain amount of HP, you're significantly less likely to do die, as the main threat to the raid is certain random damage combinations that can do significant amounts of damage on unhealably short timetables; thus, while on a typical fight an extra 500 HP isn't going to matter; but on a few key fights, it can be the difference between dying with 300 overkill and living with 200 health.

That said: this is only true of a few fights, and the sta differences between otherwise comparable items are usually not large (with a few notable exceptions like Drape of the Faceless General), so it's *usually* safe to disregard this fact to some extent.

Regarding EP values for professions: those are about as correct as EP values can be, however, there are a few notable caveats. It is certainly the case that JC is one of the two professions you want to have; what the other one should be depends heavily on your individual circumstance.

For combat, there are two reasonable options in terms of optimal ToC gearing for Combat, and the fundamental issue is whether you want to wind up using the ilvl 258 mace or the ilvl 258 axe in your MH. Let me address each option in turn.

For the mace to be optimal, one gears almost entirely for ArPen. Your socketing winds up being one enchanted tear in your helm to activate your meta gem, and every other socket gets ArPen. You use Gloves of the Silver Assassin as your offspec piece, take mace spec to get more ArPen, and generally try to get as close to the hard cap as you possibly can - thus, your trinkets wind up being Death's Verdict and Comet's Trail. In this approach, the value of ArPen climbs to almost 3, which increases the value of Blacksmithing to almost 120, making it the clear winner. Note, however, that this approach doesn't become viable until very high gear levels.

The other option uses the axe; rather than going all-out ArPen, it gems it only to the Mjolnir softcap (566), and uses Mjolnir Runestone. This setup has Leggings of the Broken Beast as it's offspec piece, and gems a fair amount of agility to avoid overcapping ArPen; it also gets the set bonuses from its yellow sockets, usually be filling them with Deft gems (as haste actually catches up with crit in value in this setup). Under this approach, two extra sockets would just be filled with agility, and the profession is thus worth around 2.1 * 40 = 84 EP.

Meanwhile, tailoring - while only worth about 83 in the abstract - comes out almost 10 DPS ahead, due to the fact that its procs will often overlap your Mjolnir procs. It's EP value thus winds up in the low 90s, putting it ahead of Blacksmithing.

The third option is Engineering; if you use bombs every cooldown, accelerators every cooldown, and nitro boosts every time it makes sense to do so, you may come out a few EP ahead - even as high as 100. This would put it slightly ahead of Tailoring for the second option above. However, engineering has a few problems of it's own. The first is that you have to use bombs every cooldown, which not everyone may want to deal with. But I think the bigger issue is what happens if you want to use HAT as an offspec for those fights where it's advantageous (and they do exist, even if it's not every single fight right now). HAT benefits less from haste than does combat, and can't synergize accelerators with Blade Flurry; this lowers its value. And HAT may not be able to afford the GCDs to drop bombs, by the nature of the spec. As such, for HAT it's clearly an inferior option.

Thus: if you're full time combat and don't mind spamming the crap out your extra cooldowns, Engineering is the way to go. At high gear levels, if you want to go the all-ArPen method, Blacksmithing wins. But for most rogues, who aren't in either of the previous 2 categories, Tailoring probably holds a narrow edge on the competition.

Note that I haven't specifically analyzed Mutilate, but my guess would be that the all-in ArPen approach doesn't really work very well, which would put it in the same category as combat with Mjolnir's: engineering or tailoring, depending on your exact situation.

Early simsheet analysis indicates that partly due to itemization of offhands, the optimal setup for combat is Hack and Slash with two ilvl 258 axes. However instead of choosing MR and gemming arpen up to the soft cap, it is a considerable dps increase to fully gem the gear with Arpen and use CT instead of MR. Switching to MH mace and changing to mace spec is close but the lack of high level mace offhand makes it a worse option. While the EP value of ArPen with this strategy is probably slightly lower than 3 because you are not quite at the armor pen cap, it's still high enough to make Blacksmithing a profession of choice.

I find that using Steel Bladebreaker (Heroic) with Remorseless and 1/5 CQC beats using axes for heavy ArPen setups; it's true that using an offspec OH isn't ideal, but it's worth noting that the 258 OH Axe, at 1.6 speed, actually kind of sucks - it's little better than the 245 1.5 speed OH sword.

In terms of specifics, the highest DPS setup I've found so far is as follows (all items are the Heroic version if such exists):

T9.258 Helm, Shoulders, Chest, and Legs
Collar of Ceaseless Torment
Vereesa's Dexterity
Bracers of Dark Determination
Gloves of the Silver Assassin
Belt of the Merciless Killer
Treads of the Icewalker
Ring of Callous Aggression
Planestalker Signet
Death's Verdict
Comet's Trail
Remorseless
Steel Bladebreaker
Talonstrike
5/5 Mace Spec, 1/5 CQC
JC/Blacksmithing
All sockets ArPen except for helm, which has an Enchanted Tear.

Stat	EP
Agi	2.12
Crit	1.82
White Hit	1.49
Expertise	1.74
Haste	1.7
ArPen	3.03

Estimated DPS: 9700.2

If we switch the weapons to Stormpike Cleaver and Lionhead Slasher, this necessitates the following additional changes: offhand weapon enchant changes to Accuracy, and neck changes to Charge of the Demon Lord. The resulting DPS number for this setup is 9676.2 - close behind, but still behind.

The third option is to go axes and use Mjolnir; in this arrangement, I come up with the following setup:

T9.258 Helm, Shoulders, Chest, and Gloves
Charge of the Demon Lord
Vereesa's Dexterity
Bracers of Dark Determination
Belt of the Merciless Killer
Leggings of the Broken Beast
Treads of the Icewalker
Ring of Callous Aggression
Planestalker Signet
Death's Verdict
Mjolnir Runestone
Stormpike Cleaver
Lionhead Slasher
Talonstrike
JC/Tailoring
Yellow Sockets in both axes, boots, belt, and ring socketed with Deft Ametrine; Helm with Nightmare Tear; 1 Delicate Dragon's Eye, 4 Delicate Cardinal Ruby; all other sockets ArPen.

Stat	EP
Agi	2.07
Crit	1.75
White Hit	1.6
Expertise	1.85
Haste	1.76
ArPen	1.9

Note that up until the cap, ArPen is ~2.7; but as this setup is 3 points over the cap, the marginal value has dropped to 1.9.

So, the raw DPS estimate I get here is 9669.8; however, this doesn't utilize trinket synergies with Swordguard Embroidery, which I estimate as about 15 DPS; hence, I'd say in reality this will be close to 9685 DPS. Which is ahead of axes with straight ArPen, though behind Mace with full ArPen. However, this resurrects the recent discussion of the value of defensive stats, as this setup has an extra 200 agility (and thus 3+% dodge) and 77 stamina relative to the straight ArPen one. Thus, I consider either the first or the third option viable, but see no reason to go axes with straight ArPen, as you get neither the DPS advantage of the first nor the defensive stats from the third.

As a side note, it's worth noting that dropping rupture from cycles in the ArPen setup and going straight SnD/Evis appears to be only a minor DPS loss; I've been playing around to see if it ever comes out ahead, and while so far the answer is no, it's worth noting that it's only behind by about half a percent, and in practice there may be gains from having a simpler cycle.

As an additional aside, it's unclear to me as yet which of these transitions more easily to HAT; additional modeling will be needed to deduce EP for that spec.

Just as an aside, the Horde OH axe is 1.5 speed so I imagine the Alliance version will be changed to 1.5 as well in the near future.

With the OH axe as 1.5, I get 9741.3 for tailoring/Mjolnir and 9749.4 for full ArPen/Blacksmithing. Note that the previous points about unmodeled synergy and survivability still apply, so I think if that change should happen, I'd advocate axes/mjolnir as optimal.