So, I sort of have to disagree with you on this point - and honestly, it's a fairly common misperception. Relentless strikes is not a burst talent. Relentless strikes is a sustained DPS talent. It's power comes from the ability to chain finisher after finisher after finisher. It doesn't really do anything until you've already been on the target for 5-10 seconds, and when you pull off the target for a switch, it's very possible to wind up wasting the terminal RS proc (given that if you're out for long enough to cap energy, the fact that your last finisher was discounted is irrelevant). Crit and Haste, on the other hand, work just fine on fights of any length. It means that whether or not the fight is stable enough for you to maintain ideal cycles, you still hit the mob a lot.

Opportunity, I'll grant you, helps your burst to some extent... but in total, it's not at all apparent to me that 51/13/7 is clearly "better" for interrupted fights than is 51/18/2. Similarly, when AoEing, Opportunity and Relentless Strikes do *nothing*, while 2% crit from CQC (and to some extent LR through FA) still increase your damage. Thus, your points in terms of considering the reality of fights are well taken, but I disagree with your conclusion that 51/13/7 is going to be relatively stronger due to this consideration.

All that said: I do agree that the damage difference is not large, and if for whatever reason you do find yourself able to perform better with 51/13/7 it's not an unreasonable option.

I'd also disagree that the merit of RS has any real bearing on the 4+ vs 3+ debate. It's true that 4+ finishers have a higher chance of proccing, but it's also true that you're going to be doing fewer finishers. In point of fact: RS regen is directly proportional to combo point consumption; thus, so long as you spend all the combo points you generate, it doesn't matter in the slightest in what pattern you spend them.

Of course, with 4+ cycles, you *don't* spend all the combo points you generate. When you get a SF proc at 3 CP, you waste one combo point, an issue 3+ cycles don't have. Hence, 3+ cycles actually get *higher* energy regen via Relentless Strikes than do 4+ cycles. So if you're going to talk about which cycle loses more damage to the loss of Relentless Strikes, the answer is not 4+, but 3+.

Finally, I should note that it's not really that hard to model the damage gain from each cycle type, and, as such, on that point at least we need not speculate. And I find that in BIS gear, all combinations of 51/13/7 vs 51/18/2 and 3+ vs 4+ are within a couple percent of each other. So I think the proper conclusion to draw is that if you want to pick one cycle to use in general, 51/18/2 with 4+ usually proves to be the correct choice; however, the other options are completely reasonable to use at times, provided you keep in mind the strengths and weaknesses of each and are using them in the proper time and place.

Interesting column there, I like you drew a linear conclusion for us "non read'n" folks. I was curious as to why you were still gemmed agility though considering that in most cases instant dmg out-preforms any other damage source, and as far as I can comprehend (I could be wrong) that the gains from gemming AP are greater than AGI in terms of IP dmg?

It could be, like you said that the gains from 51/18/2 scales better with T10 than t9.25, just a theory.

First of all let me say that just because i am quoting you Omni, that doesn't mean that my post refers just to you or has to do with you. It's some things i had in my mind for quite a long time and wanted to posted them. I think that they can help with clearing up some confusion regarding "which is the best assassination spec at the momment?", "why does the best spec doesn't work for me that well?" and other similar questions. Sorry for the long post in advance and excuse my mediocre skill in English.

Let me start by saying that only good things can come from further discussion, still i don't think that anyone who is suggesting a specific spec as being the best is trashing the other variations of this spec or misleading people into inferior specs. Those people that are spending hours of research, in order to report here their findings and make suggestions, will sometimes be wrong. Whoever doesn't understand that there's such a risk being involved, is naive if you ask me. Same with whoever doesn't understand that opinions and findings, that are being posted in such forums, are never the absoloute truth, but their degree of validity and accuracy depends on those characteristics that are unique to our raiding envinroment. Meaning that if a certain spec, gear choice, e.t.c. is the best or not, heavily depends on if we raid 10's or 25's, what our gear level and kind is like, what our gemming is like, which are the raid buffs that we usually have, which are the debuffs usually present on the bosses, which is the kind of boss fights that we want to be optimal for (when progress raiding is involved) and a lot more factors.

Those who want (not impyling you are one of them, just saying) to just profit from the extended research and testing of other people, without bothering to spend some time towards understanding mechanics and thus being able to adjust and adapt facts stated in here to their specific raid envinroment and case, deserve to end up with a non optimal spec in my opinion. Having said that, my point of view is that the one who trashes the other specs is each one of us who chooses to blindly follow (without testing for oneself), what some will suggest as the best spec. And additionally will further suggest it to other people in game as being the ultimate spec, without defining what gear requirements, raiding style, raiding envinroment, e.t.c. this spec needs in order to be optimal. I don't think that it's realistic to blame the community of rushing into conclusions and thus suggesting things. Further discussion is always welcome, but i don't think that lives depend on if 51/18/2 is the optimal spec at the moment, so even if those suggesting it are partially wrong, no big deal.

In my case, i have dual spec using a 20/51/0 and a 51/18/2. We run 10's and just because i will not be using my assassination spec on any boss that i can't stay on for the full (or almost full) duration anyway, 51/18/2 is the optimal spec for me. Also talking into account my level and kind of gear, the kind of raid buffs that are usually present in my raids, e.t.c. Finally, trying to find the ultimate assassination spec is not a one way road, we can instead partially focus on trying to explain the differences between the different assassination specs and point out in which cases/situations the one is better than the other, why, e.t.c.

What might some of the strengths and weaknesses be?

For example, I'd imagine it more beneficial to maximize envenom buff uptime during heroism. What about trinket procs? 4+ is probably better during death's choice procs, while ensuring envenom buff might be better during comet's trail.

Yes, the true power of RS is in sustaining burst, and the entire point was to demonstrate how changes to the percentage of possible RS procs have drastic repercussions that are not immediately apparent until actual practice. While it is true that it is possible to waste an RS proc on a target change under certain conditions, Mutilate DPS is somewhat of a sigmoidal curve until you level off into 4cp chains - and full RS just makes it faster to reach that peak, albeit only slightly. However, the main point is not really the difference in burst, because as noted it is minor but still observable. In an encounter like Marrowgar, Opportunity is going to increase overall burst damage by about 1/5 of a percent, which may be a small skew but it is still a skew in favor of 51/13/7.

When it comes to AOE, OP may do nothing, but RS is absolutely vital in that more energy available is more control of where your damage is allocated (in this case, as FOKs). If one has ever done Anub'arak as Mutilate, it is readily apparent the effect RS has on maintaining a sustained cycle while still managing FOKs under Envenom buffs. Without full RS, you are going to have a significantly smaller window to land a FOK with Envenom still up, and the impact on poison damage during AOE is huge when not timed properly. 2% crit is all well and good, but if you have say a 15 second window to AOE, then 2 energy returns from RS translates into a free FOK entirely.

You have a solid point on the 3+ cycle without RS, but the problem is that I've yet to see the 3vs4 cycles with RS accurately modeled in the entirety of this thread. If the modeling is as easy as you say then it should be no problem to post it (as I've requested it multiple times), because in practice my personal findings conflict with it, particularly with the rotation changes (Envenom). The loss of Opportunity is still a significant factor in the 4+ cycle (Mutilate vs Envenom), and in particular I want to see how the modeling reflects the poison output for each cycle.

It is true that the difference in specs is marginal, but that ultimately leads to deciding which is the most practical (or easier). I imagine that Envenom uptime is a nightmare with 51/18/2, but at the same time it's impossible for the spec to maintain it 100% so it is probably not worth stressing over its timing outside of the obvious (no overlapping, maintaing during Bloodlust, etc.) On the other hand, players specced into full RS should always be aiming for 100% uptime, which is a lot more involved cycle overall but it makes maintaining the buff a lot easier when it's really needed (i.e. cooldowns and lust). The parses point toward the latter edging out damage-wise, but if one is not solid at managing their Envenoms then I imagine they are better off going for the passive talents.

(@Zky I have not converted to AP yet out of convenience, not because Agi is better. Regemming to AP would be mostly neglible as far as the numbers observed in the parse are concerned.)

I don't get how you can claim that the parses are favored towards 51/13/7 at all. In fact, the only 25m parse you linked for comparison beat yours by 1k+, while you were on par with 10m parses..

Are you talking about DPS? Because of course they should be higher - I don't have BiS gear. The comparison isn't about how much damage they are doing per second, but how much the difference in talents are affecting the fractional damage of their output (such as what % is Melee, IP, Mutilate, etc.) You should also always be careful about comparing DPS between different parses because it is highly consequential of raid DPS and encounter duration (meaning DPS comparisons are usually only applicable between players in the same parse). The ones I linked should have been 25man, so you may need to toggle something to get them to show up. If not just disregard those and go here (World of Logs - Real Time Raid Analysis).

Aldriana, when you say the average Envenom is 3.61, I assume you are referring to the cp size and only toward a 3+ cycle? The problem with using averages is that in practice you should never be 'forced' into clipping unless in risk of energy capping during Bloodlust or possibly Overkill. At the same time though it is definitely possible to get enough bad RNG to be energy starved and have Envenom fall off during those same times - even more so with 4+ cycle. As you said, 100% uptime may be impossible from RNG, but from a cycle combining 3+ and 4+ the highest I've been capable of is mid 80s. I haven't played a strictly 3+ cycle, but if I'm reading your post correct in that 3+ cycles are going to have significant clipping, then I will definitely test how it works in practice.

Also, is there a way to show the hidden pages in your spreadsheet? I am trying to view the modeling for the different cycles but I can't figure out how to make them appear.

Gearing can also shift the balance between those percentages, Omniwank. For example, swapping out Armbands of Dark Determination for Bracers of Swift Death yields nearly identical damage for me in Aldriana's spreadsheet (~2-3 DPS), but raises physical damage by about a half percent and lowers spell damage by about a half percent. This is a fairly minor example, but if this small a change can shift the percentages by this much, the difference in proportional damage between your gear and the people with BiS should be far larger.

The point is not exactly how big the finishers are, or exactly what the attack spacing is; the point is that with high-end gear, running 51/13/7, you generate envenom uptime faster than you can spend it. If each envenom is launched the instant the previous one drops, so that you have 100% uptime with no overlap, you will, on average, steadily gain energy. If the fight goes on long enough there is nothing you can do about this. No matter how good your pooling is, no matter how precise your timing is, the simple fact is that you're generating Envenom uptime faster than you can possibly spend it; thus, clipping is inevitable. And the simple reality of the situation is that perfect envenom matchup of this sort is functionally impossible anyway. In reality, if you do a 3pt Envenom followed by no Ruthlessness proc and a noncrit Mutilate (leaving you at 2 CP), you're going to have to do 2 Mutilates and a finisher there, and no matter how well you pool there's simply not enough energy in the 4 seconds of envenom buff to do that - leading to a gap.

So: you're generating envenom uptime faster than you can spend it; and perfect expenditure is impossible anyway. As such, you're going to be clipping. And probably clipping quite a bit. One of the major advantages of 51/18/2 (and 4+) is that it minimizes the amount of this.

Note that this all assumes that you're doing "3+" or "4+". In reality, what to take away here is that 3+ finishers are a reasonable option to the extent that you can do them without envenom clipping. And if you do them *all* the time, you're going to clip envenom too much and lose DPS; but that's not to say that you can't weave them in if you know how to do so without clipping.

Additionally, this is a problem that doesn't exist to the same degree at current gear levels. While the principal that 18/51/2 clips less still stands, clipping is a significantly smaller issue as your energy and CP generation are significantly lower. It still happens, but you don't find yourself in the situation described where you can't keep up on regen even in theory. I still show 18/51/2 4+ ahead in most cases, but the specific comments around clipping aren't necessarily valid quite yet.

First of all i would like to thank both you Aldriana and also Mavanas for the extensive work that you do to help out the entire community i know it is a big timesinc on both your parts, so thank you.

My question is this: As long as you are not clipping envenom buffs with a 3+/4+ weaving cycle it is considered the best. But, during a heroism or bloodlust with an active speed pot does your haste go up high enought hat perhaps clipping is realistically your best option.

I know that this is a bit vague of an explanation but as long as you do not clip by a vast margin would it be more benefitial to concentrate more on keeping your envenom buff up as much as possible with 0 down time during a bloodlust/heroism due to the haste benefits of these buffs. I read through the posts and did not see a topic regarding this. Perhaps it is an intuitive thing that i have merely missed in my own mind but it seems to me that during these buffs it would be more of a benefit due to instant poison procs to keep the envenom buff up the entire time even if you are clipping by a margin of .3-.1 seconds but of course as always it is entirely possible that i am wrong about this. Any insight on this would be much appreciated. Thanks in advance.

Perhaps there is a point where stacking enough haste procs/buffs together causes clipping envenom buffs together (as little as possible but still clipping) to ensure 100% uptime may be the better way to go.

I would just like to quickly go back to Omniwank's original post regarding 51/13/7 vs 51/18/2. I am glad you posted that because I can use it as an illustration of why arguments like the one you made are misleading and your earnest effort to model dps that way does more bad than good.

I have put in your gear and spec into the simsheet. I also tracked which buffs you had that fight and which debuffs were on the boss. I even put in the values for wild growth and rejuvination such as to match the 62 energy you gained from Rapture and Arcane Torrent. With those values I ran 51/13/7 and 51/18/2 side by side in the simsheet. The first spec gave 11,194 and the second one gave me 11,222 dps. I am pretty comfortable with the results relative to your report given that the uptimes of buffs and debuffs are about 5% short for some key debuffs. The difference between the two specs was only statistically significant at 16% meaning they were so close that they were not statistically distinguishable in 300 iterations I did.

In those 300 tries, when I did pairwise comparison between the dps of 51/13/7 and 51/18/2, the latter was ahead in 156 cases. The dps of the two specs ranged between 10500 and 11900 dps. That is RNG at play, and in a real fight the RNG is even higher because there is also weapon range rng, there is always a human factor, latency and fps, and most importantly, there is RNG due to buff and debuff uptimes.

So I did my calculation 300 times, and you come here with 1 try as 51/13/7, take the damage numbers and show that 51/13/7 scores better. Next time it is likely that you can do the same analysis and find that 51/18/2 is ahead. The point is that we have these spreadsheets and simsheets to minimize the effect of RNG on our conclusions. If necessary I run the simsheet 10,000 times to find which spec is better and to bring RNG to very low levels. Of course the difference between the dps could be so huge that no RNG could ever explain it. But for two specs so close to each other, for you to come here with 1, 10 or even 100 combatlogs and make conclusions that are not statistically significant and then claim some findings on the matter that contradict the spreadsheets is just a disservice to the rogue community. They are misleading to say the least.

P.S. The simsheet has a way to model 3+ vs 4+ cycles, so the reason you have not seen 3v4 cycles properly modeled is because you have not looked around. Aldriana's sheet can do it as well. In the simsheet 51/13/7 and 51/18/2 give 11094 and 11146 dps respectively, that's taking into account all related factors, such as for instance, envenom uptime changes. So as you can see once again using an absolutely different approach to modeling, that 4+ cycles score slightly higher in dps.

Omni are you trying to maximize your envenom uptime or your dps, because these are two different optimization problems. Weaving in 3-combo envenoms is not guarranteed to increase your dps just because it increases your envenom uptime. In fact it's quite clear to me that in a 3 combo situation, you are supposed to get another mutilate in and do a 5-point envenom to maximize your dps.

If you are just maximizing your envenom uptime, then I misunderstood your whole point, and perhaps 3+ is the way to go for that.

Whether 3+ envenoms can be a damage increase over 4+ envenoms in the abstract is actually a hard question, and I think it depends on your current stats to some extent. I will say that at every gear level I've tested, 4+ wins; however, I'm not ready to assert the inherent inferiority of 3+ finishers in circumstances where they can be woven in without clipping envenoms.

I have a question about crit-cap and how I'm not understanding it. I'll be using Mavanas and his current gear set and spec as a model.

Assuming that it is an ideal situation with proper raid synergy he'll get:

5% from - Blessing of Kings/Mark of the Wild/Horn of Winter or Strength of Earth Totem

5% from - Rampage/Leader of the Pack

3% from - Heart of the Crusader/Totem of Wrath/Master Poisoner

So that's 13% from Raid buffs.

Mavanas's crit is 47% unbuffed. So Mavanas's total crit raid buffed is (47) + (13) = 60

Taking a gander over at World of Warcraft Rogue Crit Cap Calculator to determine his crit cap, is 73% vs a Boss. Without his trinket, its 69%. One can estimate that with Greatness proc, his crit would be 65% so hes under, no?

I just wanted to verify that I was correctly calculating crit-cap raid buffed. Thanks.

Mavanas, the way I currently run my rotation is with an emphasis on maximizing 4cp chains while constantly maintaining the Envenom buff. In essence it is running a primarily 4+ cycle with a 3+ cycle when the Envenom buff warrants it. In preparation for bad RNG, I usually pool energy during solid 4cp chains and before every Envenom, clipping it at less than .5 seconds for the extra IP proc. It's typically good to have an energy stock of around 10-20 in case you don't get returns from RS and Ruthlessness.

Keep in mind that while it is impossible to maintain 100% uptime, cooldowns can more than make up for bad RNG. I usually blow my cooldowns (i.e. Vanish) liberally, but Arcane Torrent and Thistle Tea are very clutch in maintaining Mutilate cycles. Energy drought? Just sip some tea. 0cp drought? Bust Vanish if its available. The best way to counter the BS RNG of Mutilate is to simply prepare and counteract it with everything at your disposal. On my Saurfang parse, I did have bad RNG in the second half of the fight, but the 3+ cycle absolutely saved it. It is still a dice roll regardless, but 3+ seems to be the proper choice because you can maintain the Envenom buff even when RNG is not in your favor.

That crit cap calculator is wrong. It uses the old formula 100 - Dodges - Misses - Glances + 4.8 which is incorrect. The correct crit cap formula is 100 - D - M - G, as demonstrated in http://elitistjerks.com/f31/t76785-c..._combat_table/.

Using your Mavanas example, his crit cap is therefore 68.8%. Or is 64.9% without his Shard of the Crystal Heart. I also don't see 47% crit chance unbuffed on his armory profile, I see 44% at this time. Finally, your 5% from Kings/GotW/HoW is a bit on the high side. It's closer to 4% for Mavanas with these 3 buffs.

Are you doing your calculations using 51/18/2 or 51/13/7? That needs to be specified first because the math is going to be different, and honestly with 51/18/2 there's really no point aiming for high Envenom uptime because it's impossible considering the lack of energy gains. The logic behind manipulating RNG for Envenom uptime can be derived by how close 3+ and 4+ cycles are coming out on the spreadsheets. The 4+ cycle is static but at the same time unforgiving when it comes to Envenom uptime, while the 3+ cycle - according to Aldriana's observations - is generating too much uptime for one to spend. A middle ground between these two that uses the best option for the situation, almost always depending on Envenom uptime and possible RNG, should come out higher in damage.

You are on the right track with your calculations but you need to take everything another step further in the cycle to compare the output of each option due to RNG. Here is where it gets complicated, but in your example what needs to be calculated is the damage difference with the worst possible RNG. For the 3+ Envenom, it is possible to have no procs from either RS or Ruthlessness, which leaves the player with low energy and no combo points. The player's option from there is to Mutilate again into a 3cp Envenom. The ability to maintain the buff will depend primarily on how much energy you have, and partly on your gear. If you assume only 35 energy, then it is possible to lose some uptime. This is where pooling your energy during stable 4cp chains is ideal, but to stay on track generally a player in a 25man raid should be able to 'save' the buff. From there it leads to another RNG dice roll, but if you calculate the worst possible scenario for a 0 RS and 0 Ruthlessness on a 3+ cycle you can just add that for equivalent parts of the 4+ cycle.

Now, from that same point (3cps with Envenom running out), we must calculate the worst possible RNG for the next step. If you Mutilate instead of Envenom, you gain 2 cps, but lose Envenom uptime, 20 energy, and damage ( comparison of 3cp Envenom vs Mutilate). At 5cps, you then Envenom, no Ruthlessness and depending on the spec, no RS. That leaves you with 0 cps, low energy, and 6 seconds to get back to a 5cp Envenom (the only option in 4+ cycle). Once all that is calculated, it then has to be synched with the same amount of time that would of spent on the 3+ option, and that's where the math is pretty involved. From field testing the 3+ cycle should deal with the worst RNG better, so you should find the same conclusion if you can beast those calculations.

The purpose of comparing worst possible RNG outcomes of 3v4 cycles is that if 3+ cycles are better with dealing bad RNG at the '3cp with Envenom wearing out' scenario, then it is worth it to take the risk with the Envenom because its positive RNG (an RS and Ruthlessness proc) will direct you right back into 4cp chains, which is the highest damage scenario anyway. And if it doesn't, you can go right back to another dice roll while minimizing the bad RNG via Envenom uptime.

Maybe I am misunderstanding the larger debate going on here, but what I have to say should generally apply. Simply from a statistics perspective you can't just analyze the worst possible RNG and leave out the best. If Mavanas calculates damage loss as a result of the worst RNG, he'd also need to calculate damage gain from the best possible RNG to offset the skewed result he'd be getting (if you always assume the worst possible RNG, you'll start clipping your envenom uptime once you run into "normal" RNG or even "good" RNG). You can't just work at one of of a distribution, thats not how RNG works.

Yes, but this needs to be taken one step at a time as the math is very complicated.

I analyzed the best-case scenario for 3-point envenoms to be viable versus choosing to do another mutilate and then a 5-point envenom. If it does not compete in that case, the gap between those two options will be only larger in other cases.

I had 51/13/7 in mind, although it never appears in the calculations. If you start getting into energy consideration, 3-point finishers suffer even more.

Please analyze carefully what I wrote. If you have more than 35 energy at the start, you lose less envenom uptime by choosing to go for 5 combo points. If your 3-point envenom does not proc RS, it will be an even bigger loss for going that way instead of choosing to use a 5-point envenom.

You just did not understand the whole setup. I am not comparing a 3-point envenom to a mutilate and start subtracting energy or adding combo points. I am comparing two cases:
A. Envenom early then go on with a mutilate
B. Reverse the order, mutilate first to get to 5 points, only then envenom
Under case A, you gain envenom uptime, so you get more IP damage from autoattacks and one extra mutilate under the effect of envenom. Under case B, I lose envenom uptime, but I gain a bigger envenom damage. I also may gain energy relative to using a 3-point envenom due to RS, but this argument is not even necessary to show that path B is preferable.

It is a very straightforward analysis and it shows that path B comes out ahead without heroism. Under heroism effect, the two are almost equal, but if you take into account energy differences, path B will still end up ahead.

I understand your analysis, and it is sound. The problem is that you are analyzing only a small section of the rotation that doesn't encompass the entire point of doing a 3+ cycle in the first place. Yes, part of the reason is to maximize Envenom uptime, but it is equally important to minimize bad RNG. In your analysis, your claim is that going the 5cp route is ~600 more damage total, but you completely ignore the fact that in the 3+ situation you are sitting on 3 (possibly 4) combo points, while in the 4+ situation you have the possibility of only 1 combo point. Those cps equate to damage currency (a lot more than 600 damage, by the way), so stopping the analysis at the point you did doesn't really conclude anything.

Now, as I've said the math from that point gets very complicated, but you absolutely have to analyze where each cycle leaves you in each possible RNG scenario. For the purpose of this discussion, however, the worst possible RNG is primarily important because the cycle I do only resorts to 3+ when RNG is bad. If a 3+ cycle sustains more damage in the worst case scenario, then it should be worth it - under those specific conditions - because the best scenario of both cycles is the same (RS and Ruthlessness return), with the 3+ cycle just having more chances of landing it, meaning it has the possibility of minimizing a bad RNG streak during a normal 4+ cycle.

First off I want to say thanks for all the work you guys are doing with the spreadsheet and simsheets.

To explain a little about my background real quick, my main is a Mage and hence that is my focus since it is my raiding toon. However I've been gearing up my Rogue a lot lately and have started to get into raids on this toon as well. The amount of theorycrafting involved regarding the Mutilate spec in 3.3 is absolutely amazing to me. I have a few friends in game who do some theorycrafting as well and I bounce ideas off of them quite a bit but I had a few that are still unanswered so I'd like to see what your opinions on these matters are. *If this is posted in the wrong section forgive me, however they all regard the Mutilate cycle so I thought this would be the right thread.


First question I have is that reading through all of your threads I see for mutilate you have one of two options for specs, a 51/13/7 vs 51/18/2. However I have a question regarding a possible 51/15/5 spec where you max out Relentless Strikes and put 0 points in Opportunity. I realize that losing the Lightning Reflexes is a DPS loss when swapping targets frequently in a fight but I want to focus on a tank and spank like Saurfang. Unfortunately I don't have 300 logs to browse through and I'm still new to WoL but in MY experience and that of talking with a few other rogues in game, the Mutilate damage in comparison with Instant Poison damage or melee damage is always 3rd or 4th on the total damage done list. With that being the case, would putting 2 points in Opportunity for a 20% damage increase be worth it for an abilitythat's coming in 3rd on damage versus a 2% total melee crit chance increase?
To relate this to the topics above regarding a very in depth debate regarding using a +3 combo point rotation followed with an envenom vs a +4 combo point rotation followed by an envenom and trying to determine which is better, possible clipping of the envenom buff with a +3cp rotation or having the envenom hit hardeer but have the buff run out in +4cp rotation. If you're running a 3+cp cycle and doing Mut, Envenom, Mut, Envenom...etc you'd have less overall Mutilates versus a 4+cp cycle where you Mut, Mut, Envenom, Mut, Mut, Envenom thus making the +damage to Mutilate from Opportunity being even less desirable for a 3+cp rotation. So I know its kind of a long way to get here but my question would be, why is a 51/15/5 spec not being considered in these calculations? Is there something I'm missing?


My second question is actually an idea that my friend brought up to me who has been playing Mutilate specs since the start of BC. Prior to 3.3 it was a no brainier to put Instant Poison on your Main Hand and Deadly Posion on your Off Hand weapons, but with the 3.3 Patch changes to Deadly Poison is this still the absolute truth?

"Deadly Poison: In addition to its existing effects, when a rogue applies Deadly Poison to a target which has already reached the maximum number of applications, this will also trigger the poison which the rogue is using on his or her other weapon."

My thought process (possibly flawed) is this: If you decided to make a swap and put DP on your MH and IP on your OH you would increase the ramp-up-time of getting a 5 stack of DP on the target with standard melee swings plus special attacks like mutilate or envenom/rupture. IF you get to the 5 stack of DP quicker, that would mean each time the DP hits after that point you are also hitting with IP. Now if i'm not mistaken IP is based off of a modifier from your AP and not the weapon speed or dps. So putting it on the lower DPS and quicker weapon of your offhand would do no actual damage reduction to the proc of Instant Poison. Instead it would do two things, increase the number of times IP hits the target with white attacks due to a quicker OH speed, but it would reduce the number of times IP hits the target on a special attack (should still hit on mut but not envenom/rupture). Now before 3.3 it was argued that having IP hit a few less times on white attacks (with a slower MH weapon speed) and more on special attacks was better. But if IP is hitting every time DP hits after a 5-stack, (white hits, plus all special attacks) would that added number of IP hits from DP be enough to make up for the loss of IP on all special attacks if it was on your OH weapon?

Now you mix that in with the theorcrafting from above, and if you run a 51/15/5 or 51/13/7 spec where you're really focusing on a 3+cp rotation so you have maximum envenom up-time, the envenom buff will give a 75% increased frequency of applying Instant Poison which is now on your offhand and hitting more quickly. Plus 15% increase to applying DP gives 15% more increase of applying IP too.



I want to remind everyone that I'm not telling people to switch to a diff spec or they should swap weapon poisons. I'm simply bringing up questions that my level of knowledge is insufficient to answer. I hope you can understand what I'm asking, I've read through it many times trying to make it as clear as possible. So I thank you for any responses in advance.