The 3.1 update is on the opening poster. Information on Armor Penetration is there. The build on the opening poster is a work in progress. I have to take into account not only simulator information but other considerations such as movement and AoE capability. Unholy Blight is a slight loss but worth it for the AoE aspect. Improved Unholy Aura is good even in Blood Presence because movement speed increases DPS uptime on movement fights.

Orlgin are you still actually using obliterate or did you forget to update that information on the first post? If you are using the DC spam rotation then the stat info and presence info needs to be updated as well. The DC spam rotation uses unholy presence and haste is very low on the stat priority (just above agility). If you're using the oblit rotation still I'd love to know your priority so that I can run the numbers.

Edit:

Priority
<BloodTap></BloodTap>
<UnholyBlight></UnholyBlight>
<DeathCoil></DeathCoil>
<FrostFever></FrostFever>
<BloodPlague></BloodPlague>
<Obliterate></Obliterate>
<BloodStrike></BloodStrike>

After adding an additional 200 ArP and Haste free on top of my current stats, using the above listed priority sheet with the build on the front post it almost breaks even with 0/17/54 DC spam in unholy presence. If I've got the priorities wrong please let me know. For reference these are the stats I used and again the haste and ArP are 200 over my current values. If you drop 200 crit and all my expertise you could just about get the haste and ArP used but it doesn't look pretty from there.

<Strenght>1301</Strenght>
<Agility>246</Agility>
<Intel>49</Intel>
<Armor>13868</Armor>
<AttackPower>641</AttackPower>
<HitRating>304</HitRating>
<CritRating>810</CritRating>
<HasteRating>404</HasteRating>
<ArmorPenetrationRating>281</ArmorPenetrationRating>
<ExpertiseRating>124</ExpertiseRating>

Edit2:

I did another test with the ulduar gear set listed on the front post for the 0/18/53 oblit build. It sims equal to 0/17/54 in my current gear.

Which Simulator are you all using?

I picked up Kahories', but I must be inputting something wrong. With my unbuffed stats plus Glyph of DarkDeath and Sigil of the VengefulHeart I'm putting out 5534 DPS over a simulated 24 hrs with 100ms latency. Does this simulator assume full raid buffs including HoW and BnS? Is that a reasonable DPS and I'm just out of the loop, or did I indeed input something incorrectly?

I'm using a 0/19/52 build, Angry Dread/Hatestrike, FC/CG.

Priority is set to:

<BloodTap></BloodTap>
<DeathCoil></DeathCoil>
<FrostFever></FrostFever>
<BloodPlague></BloodPlague>
<IcyTouch></IcyTouch>
<PlagueStrike></PlagueStrike>
<BloodStrike></BloodStrike>
<GhoulFrenzy></GhoulFrenzy>

I'm in Unholy Presence


Edit:
I discovered at least part of why the DPS was so high through a little reading. Changed the AP to what values it should be. DPS still seems a bit high though. 4567 over a simulated 24 hours with 100ms latency.

I'm using Kahories Sim. Try changing your talents up just a little bit to: 0/17/54

Change runeforge to FC/FC.

Move Ghoul Frenzy above Icy touch in your priority sheet and plague strike below blood strike. Ghoul Frenzy is never going to get used if it's below plague strike. Plague strike below blood strike is a small DPS increase (credit Kyr).

Let me know if that's an increase or decrease compared to your current build. If your build is simming better link it.

While running a few tests my Simulator had a critical error. The numbers after re-opening the Sim were generally lower, but the numbers should still relate the same.

The 0/19/52 build runs ~90 DPS below the 0/17/54 build.

3975 vs 3888

Edit: Due to my having to leave the house in a bit of a rush I was unable to finish this post.

Adding 200 ArP rating increases DPS by ~100 to either spec. Just a fun fact.


Also, what are the stats that all the BiS gear from Ulduar generates? I'd like to use them for testing in scalability.

Please link the build that you are using.

I'm not sure I understand this recent fascination with simulations. While they may provide the basis for a build, they don't compare to actual testing. I have tried every 3.1 build proposed in this thread and I have yet to break 2600 DPS on the boss dummy on the PTR. Meanwhile I can switch to 2H Frost and easily pull 3400. While I am a hoping the DW spec will scale better with raid buffs, that is a lot of ground to make up.

I prefer sims to boss dummies because some builds scale better with raid buffs. A boss dummy is just going to tell you how much damage you'd do against a boss all by yourself. Neither method is perfect but in the case of DW builds those raid buffs make a large difference in their comparitive dps. In the end the best testing will be live against real bosses but we can't do that yet.

I don't trust my sheet either, in regards to DW. It shows DW Unholy as the best by far, but this is not the case on dummies. E.g. Blood does more damage (I didn't even use Hysteria) and it doesn't self-debuff like Unholy with EP.

I can see how this makes sense from an observational standpoint, but we need to be realistic too. DW builds were not trailing far behind 2h counterparts in dummy testing when the DW builds were crushing the DMs, I know when I was testing builds for pve use the DW builds always came out on top on the dummy as well - the only real sleeper was blood heavy builds because of it's reliance on physical damage.

The problem is the dummy testing is considerably off for the DW builds, and I do not see raid buffs causing gains to DW builds that will not impact their 2h counterpart builds comparitively - and the 2h builds, at least for me, are easily doing 1k+ better in repeated tests. Even if just for simplicity we give the DW builds 1k additional dps from raids buffs, and only give 2h builds 50% of that... the 2h builds are still coming out ahead considerably.

Actually normally DW build benefit much more from raid buffs than 2H builds do. This was the case prepatch and I am sure it will be the same after the patch. This is one thing that has always made DW builds scale so much better than 2H builds.

DW dosen't scale that much better for DK's. You get a little bit more white damage, but that's it.

In very general terms, DW white damage (including necrosis and BCB obviously) scales ~35-40% better with AP than 2H damage, depending on how much hit you have and whether you have NoCS/2H spec, while 2H strikes scale about 40% better with AP than DW strikes, assuming 2H spec. Of course, neither setup has an obvious advantage for spells/diseases (though of course relative contributions from these will vary). DW will have higher relative white damage making it scale better with haste and, to a lesser extent, ArP for builds using elemental strikes. I think the latter is why people usually claim that DW specs "scale better with raid buffs"--windfury/sunder are some of the most significant raid buffs and they provide higher relative benefit to DW.

It's important to distinguish "raid buffs provide more relative benefit to DW than to 2H," which is the statement that I just argued to be true, from "raid buffs make DW scale better with gear than 2H," "raid buffs make DW do more damage than 2H," or even "DW scales better than 2H." I'm not at all arguing the voracity of any of the latter three statements; they're a lot harder to evaluate in an accurate and meaningful way.

well, isn't what makes DW better in a raid environment, that its less affected by lag more than anything else? I doubt the term is used this way, but it scales with scenarios of latency, poor client performance, and user error better than 2h.

Getting Unholy to scale better has been what I've been working on. Here's my thoughts on the matter:

1) Superior white scaling. Duel wielding naturally scales better. Necrosis and Blood Caked Blade increase this advantage. Nerves of Cold Steel increases this further.

2) Death Coil. It scales really well now. It is what Kithus is using to achieve his results. He basically uses the frost ability to generate massive amounts of runic power and power out as many as he can. Using Unholy Presence, he can fire Icy Touchs (the most efficient method of producing runic power) faster than any Blood Presence build could hope to match.

3) Strike scaling. As you would guess, this favors the 2H builds. I have been looking at Obliterate. It generates a whopping 25 runic power fully talented which is the same as Icy Touch with the glyph. With annihilation, this won't reset disease ticks and scales better than any spell not named Death Coil. It uses physical damage which is an important consideration if you get any armor penetration.

4) Pet scaling. The ghoul gets benefits from our Strength, Haste, and Hit. These stats are worth more as a result. Buffs like Horn of Winter double dip in their effectiveness. First we get the strength which increases the amount the ghoul gets. Then the ghoul gets the buff and increases that amount. So the ghoul benefits twice from such effects. The ghoul only gets Haste rating so things like Improved Icy Talons don't benefit the ghoul twice.

Unholy Presence - +15% haste for melee is nice. The 1.0 sec GCD means you are comparing 115% attacks at 1.6 sec vs 100% attacks in 1.1 seconds assuming that the opposing build doesn't utilize haste. Prior to Improved Unholy Presence, you needed to produce nine quality attacks in ten seconds which is hard. Now you need eight quality moves in nine seconds which is much easier.

Blood Presence - +15% to all damage is very nice. Haste rating lowers the GCD of spells which is important for any comparison to Unholy Presence. The more haste you have, the better Blood Presence looks compared to Unholy Presence. The key to Blood Presence is quality of attack. Strikes are heavily penalized as they get no haste benefit at all.

Improved Unholy Presence is the one to beat by default since DW Unholy can easily fill 8 GCDs.

Strength is obviously the top stat. Hit to the spell hit cap should be automatic. Everything else must be chosen:

A) Expertise - This effects (1) and (3). This has no effect at all on your spells or pets.
B) Haste - In Blood Presence, this effects (1), (2), and (4). In Unholy Presence, it's just (1) and (4).
C) Armor Penetration - This effects (1) and (3). This has no effect at all on your spells or pets.
D) Critical Rating - This effects (1), (2), and (3). This has no effect at all on your pets.

Haste is the best choice for a secondary stat for Blood Presence builds using mostly spells. It just raises white and pet damage for Unholy Presence which means it's much poorer.

Armor Penetration is really cheap. It's effectiveness is directly tied to how much attack power you have and how much of your damage is physical based. Blood Presence w/ Obliterate has a higher physical damage % and thus benefits greater from Armor Penetration. Unholy Presence has a smaller physical damage % which means armor penetration isn't worth it at their attack power level.

Expertise is an okay stat for either build. It's rate of return isn't high but Unholy Presence doesn't have much options here. The worst stat for Blood Presence is the second best stat for Unholy Presence.

Critical Rating is expensive but it has a wide reach as most of your damage is directly linked to it. It gets a small boost from the metagem that grants 3% increased critical damage. The third worst stat for Blood Presence is the best stat for Unholy Presence.

Thus, you see a pattern. Blood Presence scales better with the stats that are available. If you have enough of these stats, Blood Presence eventually catches and surpasses Unholy Presence.

However, I see to have underestimated how much of these stats are required before that level is reached judging by Kithus's sim results. I need to recheck my numbers and look for errors as I had 0/18/53 as just about equal but with better scaling for future gearing.

I appreciate the work that you all have done here. I'll update the OP from time to time. For now, assume that Unholy 0/17/54 is the best build and work from there. If I made any mistakes with the above, I need to know so I can go back and look for the error that led to the mistake. Thanks.

Just to add to what Orlgin said, I did some testing with 0/17/54, both UP and BP. I used the best gear I had for the stats for that specific build. The BP build had the following stats:
I know, a little below the spell hit cap but it appears like no spells missed, or very few. A Priority of DC>BloodPlague>FF>IT>BS>PS and was doing about 4975dps with a 300ms over 100hours. This is how the EP turned out.
Now for the UP build. It had the following stats:
It had a Priority of DC>BloodPlague>FF>IT>GS>BS>PS and was doing about 5109dps with a 300ms over 1000hours as well. This is how the EP turned out:
So to summerize, there was only a ~135dps difference with the builds in UPs favor, but the BP build scales better than the UP build. The UP build doesn't gain a lot with Haste like BP. I believe that at first, it might be best to be in UP, but as you start to get Ulduar epics, you might want to start thinking about switching to BP.

I have been trying to think of a way to fit GF into the Priority for BP but it would run too long...

Because that is a priority system not a rotation. The priority is on maintaining blood plague on the target (obviously with the use of PS) but if blood plague is already up, during its duration, PS is at the bottom of the priority order and you would use any other ability instead.

I'll go ahead and make an idiot of myself as well- GS is what exactly? I'm assuming it's summon gargoyle, but not positive on that. Also, thanks for this thread all, I wasn't quite ready to bury my DW DK yet.

Well said Orlgin! There is just one thing I would change. You understated the value of expertise for unholy presence builds. It is important to remember that even though a dodged yellow attack causes the rune to refresh much faster it still wastes a GCD. Unholy builds using a proper rotation need almost every GCD they can get and almost every time you lose one you are losing DPS.

Edit: Kyr in your post you mentioned doing a 300ms latency EP test and at one point you said over 1000 hours and another over 100 hours. I assume you meant 1000 because the sim will yell at you for anything under 500. Also could I interest you in running those numbers again with 200ms latency? I plan to run my numbers today with both 200 and 300 but I suspect unholy presence builds suffer worse from higher latency.

Dual Wield Unholy with Scourge Strike

With the changes to haste rating, scourge strike, as well as black ice, I began to think about builds that could be done in blood presence deep in unholy.

A strike heavy build is usually frowned upon for dual wield because of the lack of weapon damage, and it makes sense in a vacuum. If most of your attacks are strikes, using a 2H will be favorable, but that is without taking other talents into consideration. I began to wonder if the loss in dps due to using Scourge Strike, Blood Strike and Plague Strike with a 1H versus 2H weapon could be evened out by other talents/abilities that may favor dual wield (the more common ones being white damage, necrosis and BCB). First testing on the ptr, scourge strike hits for about 20% less with a 1H (156.6 dps MH 2.5 speed versus 222.9, 3.4), which is much less than the 30%+ disparity I find on live. Since blood strike and plague strike are such low factors in overall dps anyway (1% or less for both dual wield and 2H) the difference is minute to negligible.

So, I decided to run some simulations on kahorie's simulator (Version 0.8.2a) to see what I could come up with. Here are the most polished results.

Using:
-Talent build: 0/13/58
-glyphs: Unholy Blight, Ghoul and Scourge Strike
-runeforge: FC/FC (I tried FC/CG and CG/FC but FC/FC still comes out slightly ahead, which surprised me)
-Sigil:Awareness (Vengeful Heart was a dps loss)
-4 piece t7.5
-200 ms latency
-priority, classic unholy:

With the following stats (current character stats):

I ran 20 simulations, 100 hours each. The dps on average was: 5440.6

I then used the unholy 0/10/61 spec to compare with a 2H dps spec (as it has the most in-common talents). I simply adjusted the weapon portion to be 1 weapon, with 222.9 dps and 3.4 speed. I did not change any other stats or conditions, this is to reflect the fact that many people assume with these specs that simply swapping to a 2H weapon will provide an increase in dps, so I wanted to test that.

Again, I ran 20 simulations, 100 hours each. The dps on average was: 5457.8

The results were actually quite similar. On average using a 2H weapon over dual wielding yielded only a 17 dps increase. Certainly not the 3 or 400 dps some were predicting. Which makes the spec a viable option for live raid testing.

A criticism I have of the build is that it is not very movement friendly (without improved unholy pres), and requires a lot of attention without the ability to take epidemic. Also, not being able to have ghoul frenzy may not be desirable as well. On movement heavy fights (like Iron Council or Hodir for example), I will likely drop the points in desecration (which would be wasted to an extent) for these talents. Another criticism is that, like the 2H unholy builds, t8 is not very desirable and neither is the new sigil.

As a note my stats and EP test for the 0/17/54 spec. Test was done with the same conditions as Kyr, 300 ms latency and 1000 hours.

pintor: Why do you take the UB Glyph instead of the Dark Death Glyph? Dark Death should be a DPS gain over UB. I'll quote what I said in another thread of why this is.
Now let's cut that to 3 DCs per 20 secs. That's 9 DCs per minute. That's still equal to 1.35 DCs which should be better than the UB Glyph.

RickTheDestroyer: I meant GF.

Kithus: I did mean 1000hours, silly mistake.

EDIT: pintor, also try taking 2 points out of Desecration and putting them in Epidemic, it should be about a 20-30 dps increase.