If you still have DP applied (I belive it is auto applied in the default gearset on the Beta 4 sheet) haste should show small diminishing returns to scale as each additional point in haste adds less and less DP uptime. At least I think that's what's happening.

Edit - Wait, I'm an idiot. EP values are the value of the stat relative to AP and as haste goes up the value of AP goes up while the value of haste remains constant which means that the EP value of haste will show decreasing returns to scale.

It's pretty simple, really. When you gain a point of haste, you now gain more damage from all the other stats; hence, the relative EP value of haste goes down a bit. Similarly, when you gain a point of crit, you gain more damage from each of the other stats, so the relative EP value of crit goes down a bit. Hence, when you replace a Deadly gem with a Deft one, the EP value of crit goes up, and that of haste goes down.

So, in your example: initially, haste is a bit ahead of crit, so Deft is recommended over Deadly. As you replace successive gems, the values of the two stats converge, until after some reasonable number of them, crit catches and then passes haste in value, at which point the recommendation switches to deadly and swapping further gems results in a DPS loss. Hence, optimal DPS in this case is a mix of the two gems.

Using the beta 4 spreadsheet, running a mjolnir + axes setup, with use 3.2.2 set to yes, armor pen gems are outperforming agi to as high as 738 armor pen. Even this high over the soft cap, switching out mjolnir for deaths verdict (Normal) is a dps loss. This doesn't happen with use 3.2.2 set to no. Is there some reasoning for why this is so or is it perhaps a bug?

In 3.2.2 ArPen is being nerfed - it takes more ArPen to get the same amount of armor reduction. The new softcap is 734, which is why you're seeing the behavior you describe.

ok. I thought the effectiveness was being nerfed rather than the rating to % conversion. Thanks for the clarification.

Although I am just 2 pieces of upgrades from going to eviscerate only cycle, I am still just very curious how or if anyone's able to have a 90+ percent rupture uptime without screwing up your 100 percent slice and dice uptime. It seems to me no matter how you manage your combo points and finishers, you will always arrive at a point or another where the duration of rupture and slice and dice collide and you are forced to choose which is to drop first.

If anyone has some tips or even some sort of parses that you can show me, I would really appreciate it.

I switched to Eviscerate only cycle even tho it is a net DPS loss of about 150 dps according to the latest spreadsheet (when not using Evisc. Glyph). I simply did it because ToC emphasises short burst phases (twin shields), switching to short living targets (jaraxxus portals, infernal volcanos, possibly mistresses, adds on Anub'arak) that steady sustained dps seems secondary. Also you can (and should) glyph TotT in that spec which is another great short phase damage boost for someone else. In addition the Eviscerate only rotation feels alot less "clunky" (at least to me) then the high rupture one, allowing you to focus better on all the stuff going on arround you.

Tofuu I found that adjusting my cycle to a lower combo point SnD helped greatly reduce the number of instances where SnD and Rupture fell off together. Previously I was running 4/5/5 and I found that if I did a 4 point SnD and then followed up with my next 5 points in Rupture, that SnD and Rupture would both fall off together. Thus I would fall into a vicious cycle of repeating that over and over.

By switching to a 3/5/5 I found it staggered it enough that I rarely found myself stuck in that position. I also felt like I had to pool energy a lot less often.

There was a great post by Aldriana ages ago on how the finisher "cycle" was supposed to work, it's worth digging for if you're having issues. To paraphrase -- if you look up and discover you have ~5 seconds left on both rupture and snd, you've already messed up and there's not much you can do but let rupture drop. What you want to be doing is planning ahead several seconds more than that. If you're watching and both snd and rupture have 12-14 seconds left, build as many cp as you can until there's ~10s left on rupture, refresh slice and dice early (even though it seems wasteful) and then build CP for rupture so that by the time it drops you're ready to refresh it. The trap is trying to squeeze in an eviscerate every cycle -- sometimes you have to use energy on other things (like blade flurry, kick, etc) or the RNG just doesn't go your way and you won't have the energy to pull it off without letting rupture drop for significant periods of time.

You may need to play around with the exact timings a bit, but that's the general idea of the "high rupture" cycle.

I was assuming he simply was following his spreadsheets optimals cycle, and with 4/5/5 ive also seen the same problems unless i tweak my rotation at times like you just states.

Great work so far, Aldriana, and that you very much for taking the time to do it.

I'd just like to make two small suggestions before the sheet leaves Beta. The first is just to add some additional user friendliness to the sheet. Vulajin's sheet had a "current dps" cell where you could type in the sheet's dps estimate for a particular gear set up. (Or any other number I suppose) This let you muck about with your spec/gear/gems/enchants without having to record the sheet's previous estimate in a random empty cell or a piece of paper or whatever. The sheet is obviously excellent without it, but it's just something I found particularly convenient about Vulajin's sheet.

My second suggestion is to add a note somewhere within the sheet itself about what "high rupture," "low rupture," and "eviscerate only" cycles constitute. I know the information is on the boards, but I'm sure once the sheet goes live, there will be people asking what the difference is so putting it in the sheet in an obvious place might save you some moderation time.

Anyway, great work and thanks again!

Is there any chance of having the Relentless Weapons (at least the T2) ones put in the next beta version or official release? I seem to have had terrible luck with getting a good upgrade for my main hand above 178 dps of any kind, and while the relentless weapons probably don't have ideal stats, the T2 ones are probably still a reasonable upgrade.

Thanks.

Edit:

I entered in my current gear into the spreadsheet, swapped around my weapons to try Spinebreaker + Steel Bladebreaker (Heroic) version & CQC off course, and I started to get AP gem recommendations around the ArP softcap. Not sure if that's a bug, or if it's got something to do with having too high a crit rate with dark matter equipped

It's almost certainly crit-capping issues if Dark Matter and a CQC spec is involved.

With the possibility of a racial change(Faction Change back and forth). Am i assuming right, that the race tab in the sheet does only change base stats and not simlulate racial abilities.
My first calculation looks like a change UD to ORC could be worth around 110 EP in my gear.

From what I have heard if you faction swap to Horde and then back to Alliance (or vice versa) you can only be what you were to start with. So even if you go to Alliance and then back to Horde you can only be an Undead.

I'm fairly certain they removed that restriction. It kinda makes sense -- they'd have to keep track of two sets of data for your character if it "remembers" what your original race was.

With an eviscerate only cycle, the tier8 4/5 is still worth 75 EP to me. The 2/5 tier9 is correctly worth 0 EP. Bug?

Not a bug. If you look at the calcs sheet, row 470 determines if a cycle is Eviscerate Only or not. Column "B" is your gear exactly as is. The remaining columns determine the value of certain changes - Column "AC" is for Tier 8 4-piece. 470 is probably 0 for all but AC, where it is 1. That's telling you that if you got the 4/5 bonus without changing a single other stat (i.e, impossible), you'd go back to a Rupture-based cycle and gain some DPS.

On the other hand T9 2/5 still does not boost Rupture enough to give more DPS than Eviscerate, thus it is no gain at all.

Thanks Aldriana for all the work you do.

Just reporting an error: Knightbane Carapace on 1.3beta4 is set to 74haste and not 74arp which it should be.

i wonder why your'e talking about dual WP again.
savage combat i still bugged on live. (@ WP only)
and the selfbuff fades everytime WP is refreshed on your target.
doesn't happen with WP/DP .

did i miss something? ; )

The graphical buff fades, but testing has shown the buff is still there.

It doesn't happen with WP/WP if you're offhand swapping DP as well. In this case, Savage Combat stays on the target as long as you have a stack of Deadly Poison ticking. So it has nothing to do with using double wound as much as having another poison on the target.

Edit: Apparently only the animation is bugged.