This spreadsheet is awesome, really friendly-user, so thank you very much for this.
One question however : is there any possibility to add few low level items by ourself (like rings or trinkets) ?
One more time, thanks for the work !

You can't. And you probably won't ever be able to. Here's everything you need to know about using DP MH:

1) If your weapons are the same speed, you should swap the hands on poison - DP MH, IP OH.
2) The damage increase is quite small - of order of 10 EP.
3) It doesn't change the value of or ordering of stats in any meaningful way.

I figured some data on Heartpierce would help you model it in the next Spreadsheet update, so here's a world of logs for Saurfang. Heartpierce

Buff is Invigoration and the proc is 2 energy every 2s over 10s.

World of Warcraft - English (NA) Forums -> Impending 3.3 hotfixes

HfB has been nerfed to increase damage by 10%, under 15%. I imagine this will affect spreadsheet numbers, at least when it comes to comparing Combat and Mutilate dps.

Enchant Weapon - Black Magic
Could you add this enchant?It increase haste rating by 250 now.
Thanks alot~

It will, since I really have no idea how to edit the numbers myself not sure how big of a drop yet (if its just a flat 5% or not).

It should put mutilate about 1k ahead of combat, a much more reasonable range to cover for the utility benefits of combat.

If you really want to see it for yourself, edit Calcs row 531 in the Mutilate 1.2 sheet. The .15 factor in that row is HfB, and you can change that to .1 instead. As Sanosu pointed out, it's a 4.24% dps decrease.

Any idea of when [Deathbringer's Will] and Ashen Band of Endless Vengeance (exalted ring) will be added to the spreadsheets? I've got them both if you need to do testing with them or anything :o

Confirming the proc rate / icd of the ring would be a good helpful start. Assuming it's like trinket procs it would be somewhere in the area of 45 sec icd with ~30% proc rate. It is very possible that it is an exact mimic of the Hyjal rings.

I'm not sure if this has been touched on yet, but I spent some time with [Whispering Fanged Skull] and Proculas. Looks like it follows the standard 45 second ICD and WotLK trinket proc rate. (30%?)

Got this last night, Ashen Band of Endless Vengeance (yeah we have no lives), and tracked the proc with Procodile as we rep grinded. I'm fairly confident the ICD on this is 60 sec. As for the proc chance, I'm not certain of how to figure this out. If someone wants to let me know the best way to test this, please let me know.

I got this trinket last night and ran it on the dummy for 30 mins. Because the trinket buff lasts for 30secs as expected Blizz has given it a longer ICD compared to other trinkets. It seems to be in the realm of a 90sec CD. It proc'd AP and Agi quite frequently but I did get ArPen as well approx every 4th proc. The latter is somewhat useless for a mut rogue. The trinket is supposed to give haste as well but was not fortunate enough to see it.

I also allowed an enhancement shaman, marksmen hunter and fury warrior to try it out before the trade-able feature wore off. The hunter, rogue and shammy did proc agility and AP while only the warrior got strength. So on a safe note, respective classes will reiceve their optimal buff. Plate usually get strength, Mail/Leather will get agility. The rest seem to be random.

Possibly the best spreadsheet i have come accross. Its perfect. everything is right there in front of you and couldnt be more clear. Now then, your comment was if ur gear isnt in the it sucks. reasonable assumption, but do you expect everyone to be full icc or to25hc geared already? Could anyone give a little explanation on where i can find the end of the formula's cells so that i can change some of the items stats to ones that i have. dont care about changing names as will no doubt cause problems i cba to fix, just values. Thnks for the help and thanks for the time spent on making us rogues a very nice spreadsheet

sorted, hidden worksheets! very devious :P

What I think you're looking for Setholic is Format > Sheet > Unhide and then you can access the calculations that Archaon is pointing out.

Im unsure of the proc rate, it seems to be random each time i raid or dummy run... I have world of logs from about 4 hours of raiding with it, my procs were 14x600arp (precision of the iron dwarf), 10x1200ap (strength of the taunka), 9x600agi (agility of the vrykul).

Its confirmed that the buff is a random chance of agi/ap/arp for combat rogues, and there might be a chance it switches the arp to haste for mutilate... Nobody can confirm that because a mutilate rogue would not take the chance at picking it up if it's wrong. However people have had mixed results depending on spec (other classes like DK).

It's believe the internal cd on the proc is roughly 1:30 to 1:45, definately not 2:00. People are still testing this, i would but i do not know how. Also, it is class / talent spec specific and has nothing to do with your top 3 stats.

Also about whispering fang skull trinket, couldnt the previous model of mirror of truth be modified to fit the increase of passive crit, ap and duration for the current spreadsheet?

I ended up picking up this trinket Tuesday, and last night I spent a few hours auto-attacking a dummy with Procodile on. It made it slightly difficult to track, being that the trinket gives 3 distinct buffs, and I had to track all 3. By the end it was showing all 3 buffs as having a 105 sec cooldown. Because the trinket seems to proc the buffs at random, Procodile would have to wait until I got 2 of the same buffs in a row before deciding on what it's ICD was, but because of the length of time I spent on the dummy, I feel relatively confident that it got enough instances of the back-to-back procs so as to gather correct data. Also, because all of the procs showed a 105s ICD, that seems to further support the theory that it is correct, and indeed on a 105s ICD.

Whenever one of the 105s periods was up, the trinket procced almost right away, leading me to believe that it has a fairly high proc chance. Again, because there are 3 different procs to the trinket, Procodile wasn't able to track the correct proc chance or uptime gained, so more testing is needed to gather this data.

You were also wondering if it changes to a haste proc for Mutilate. It does not. As both Mutilate and Combat specs, the three procs remained AP, arp, and Agi.

I don't have any of this data saved to a sheet anywhere, but if someone wanted me to gather more and post it, I would probably be willing to do so.

Honestly, what I'd be most interested in seeing is a raw combat log. Figuring out proc rates through WMO/WoL/etc. tends to be tricky, but with direct log access it's actually relatively straightforward.

Heartpierce Log

Item: Heartpierce, unenchanted, equipped mh
Procname: Invigoration
Stats at time of test: 405 hit, 213 expertise, 40.37% crit, 430 haste (13.11% no Lightning Reflexes).
No trinkets, poisons or offhand equipped.
Time: Around 2 hours till the dagger broke. (PS: If you haven't already, watch the movie Gran Torino)

So, something sort of wierd is happening with Heartpierce. The data set just posted by MassMan shows 163 procs in 4608 swings, which would be a proc rate of 3.53%, with a 95% confidence interval of 3.00% to 4.07%. However, a data set sent to me via email had 135 procs in 4815 attacks, for a 95% confidence interval of 2.34% to 3.27%.

Now, this, by itself, isn't a problem. We could safely conclude the proc rate is just over 3%, and work from there. Now, the tricky part is that another log I received via Email showed 4317 attacks and 104 procs, for a 95% confidence interval of 1.95% to 2.87%. And the fun part is: that was done while dual-wielding, with Heartpierce and a 1.4 speed OH. So why is this a problem?

Well, if we assume the proc rate is on Heartpierce only, we'd expect only about 57 procs; seeing 104 would be more consistent with a 5% proc rate. Thus - particularly in light of the unconventional wording relative to most procs - makes it likely that it procs off both weapons. But if we assume that, now the proc rate is too low. As I see it, there are a couple of possibilities. First, it's possible that the proc is somehow affected by haste or some other stat, which is causing the proc rate to actually be different on different tests.

The second - which my gut tells me is going to be the answer, though we'll need more testing to prove it - is that the proc rate is 1 PPM on both weapons. That is: Heartpierce by itself procs at 3% (=1.8/60 = 1PPM), but your OH procs at 1 PPM based on it's weapon speed - that is, with a 1.4 speed OH, it will only proc 1.4/60 = 2.33% of the time. Under this assumption, the expected number of procs in 4317 attacks with a 1.8 and 1.4 speed weapon is about 113 - a much better fit with our data.

So, steps for testing: first, we need a couple thousand swings with Heartpierce alone at two vastly differing amounts of haste - say, one test with zero haste, and one with several hundred haste rating and lightning reflexes. And second, I'd be interested to see additional testing while dual wielding with a second weapon of vastly dissimilar speed - either significantly faster (1.4) or significantly slower (2.6).

Aldriana, do you still need data for Needle-Encrusted Scorpion, since it wasn't specifically mentioned in the sticky one way or another? I have a few hours of combat logs I can send if they are needed. Sorry to say it isn't looking especially competitive at this point, though.

I did forget to mention it; some testing wouldn't hurt, but realistically I'm feeling pretty comfortable with the proc rate on it; the brief test I did indicated a 50 sec ICD, and wowhead lists its proc rate as 10%, which, conveniently is exactly the proc rate of Mirror of Truth and the other proc-on-trinkets we've seen so far. So a little more testing wouldn't hurt, but after Whispering Fanged Skull this is the one I'm feeling second most confident about.

Note that, if true, this would make the uptime a couple percent lower than Mjolnir, which is going to make it relatively weaker than Mjolnir on the whole, It *probably* still beats Grim Toll, but not necessarily by as much as you might expect.

So, I got a chance to look at this, and there's a couple of important points. First, I note that the problem you're having with is the high-rupture cycle, and I haven't been able to reproduce it with the other cycles. This is significant for a couple of reasons.

First, the high-rupture cycle is and always has been the most suspect of the models in this sheet; I've never been as confident in it as some of the others. So I'd be willing to believe there is some manner of bug with it even if everything else is valid.

Second, high-rupture is largely dead; with the ease of getting triumph badges at this point, very few people will be using 4/5 T8 for much longer. And for those of you that are... sorry, but I'm not sure I care.

Third, the result is actually... not totally unreasonable. I mean, okay, it's clearly wrong, but it's not as wrong as you might think. Consider the factors at play:

• We've moved away from Relentless Strikes. This makes finishers more "expensive" and thus combo points relatively less "valuable". Thus, Glyph of SS in general grants less damage than it did when we were running 15/51/5.
• High rupture is about running a very tight rupture-based cycle; the unpredictable CP generation of Glyph of SS thus going to work against you to some extent.

So, basically, there are two competing factors at play - the increased damage you gain from extra CP gen, and the damage you lose from unpredictability of cycle. And with the move to 20/51, the former has been reduced in efficacy. so it's plausible that it could get *so* low that you actually lose more than you gain.

Admittedly, it still shouldn't *cost* you DPS as you can always just ignore that you have the glyph and burn the extra combo points... but the sheet isn't smart enough to do that. It always *tries* to make use of the extra CP gen, whether it turns out to make sense or not.

So... it's sort of a bug. But there's a grain of truth to it as well. And since it's a little-used and tough-to-model cycle anyway, I'm not inclined to worry about it too much.