Are you implying that the scaling of SS is not superior to the aggregate independent scaling of IT and PS?

It is fairly trivial to prove that as AP and weapon damage approach some outward bound that SS damage will increase above that of IT+PS.

If we have to model the inequality that

(SS-(IT+PS))/30 > BS/20 and we accept that SS =n(IT+PS) then we can rewrite this as

((n-1)(IT+PS))/30 > BS/20

There is obviously going to be some link between the scaling of SS, IT, PS and BS where we can accurately solve for the lowest value of AP and Weapon damage where that inequality is true, provided that it is ever true. If accurate data exist for these values (co-efficients, modifiers, etc) I'd be more than happy to play around with it.

It isn't that we're trying to make it work against all reason, it's just that we're trying to figure out whether it does or doesn't actually work, when everyone else here seems to be determined to say "it clearly won't work, you're all dumb".

EDIT:

We're trying to get data samples to see if the loss of a BS actually outweighs trading a SS for IT+PS. I have a data sample which indicates that it does, but not by very much. Since I used aggregate data from all boss attempts, I was somewhat surprised to see that my SS damage was so low, which to me was suggestive of not doing something right. Other people have parses which show that the SS gain is actually above that of the BS loss. The reason that we're so focused on this is because, in addition to the other elements (consistency, increased aoe damage, etc) which are very difficult to appropriately value, we are trying to model the direct DPS difference. If it turns out that glyph of disease is a slight DPS loss but a RP generation, utility and rotation gain, then we can start debate on whether those gains outweigh the DPS loss. If it's a DPS gain (intuitively, this does not seem unreasonable), then the discussion is moot.

Challenge, I might have found something that will be keeping us from actually being able to put the Disease glyph into use in the real world, however, I need a bit more information to know for sure.


First off, the difference in your sample and my sample, which show differing data is probably because of our specs.
I am the frost/unholy build and you are the Blood/Unholy build, This could leave a slight difference in the data, as my SS has a 20% harder hit due to Black Ice.

Secondly, and most importantly, after playing with the disease glyph, (all of my idea came from straight thinking and working numbers around without the disease glyph) it seems that your rotation has to be PERFECT in order to throw it in for the first BS in your rotation and be able to keep the disease up.

Factors that could change this that I haven't been able to find a significant answer to:

Are our Rune cool downs effected by haste? (this has been a question thats been debated since beta, and everyone seems to have a different answer)
If so, much like a mage's arcane spec haste could make or break this glyph.
If not, proceed to question 2

and after my modeled data showing that it is in fact a dps increase, which I will assume until someone points out differently using numbers as I tried to do, would it still be a dps increase if i were in Unholy presence rather than blood presence? (Obviously to account for the quicker cool down of runes.)

This second question is much easier to answer as all it would take is a quick parse of data at the same time frame in each presence to show.

I'm 12/0/59 atm, so that might be the cause for the discrepancy.

I'm not sure whether rune CDs are affected by haste. If yes, then it could be entirely possible that this would change the way that the glyph works with our rotation. Haste would change GCDs, which would at least allow us to get faster ability usage in to tighten the rotation, though I'm not sure what change, if any, that would make.

I doubt that it would be a DPS increase in unholy, since 15% flat damage is likely better then 10% more rune abilities, though if the difference is that you have a tighter rotation and can use a more favorable glyph, then that might push it towards using unholy for DPS.

Currently using the Unholy rotation

PS, IT, BS, Pest, SS
SS, SS, SS

You have about a tenth of a second to hit that Pest on the first blood cool down before your diseases fall off, so without some way to speed up your runes, I don't see it being plausible in a real world scheme.

There already are calculations in this thread, look back.

Also, I want a recount for your damage. No one is pulling 8k on XT002 over a 6 minute period.

Well that would mean you'd have to change the rotation to:

BS>BS>PS>IT>SS
SS>SS>SS

Then you can afterwards shift to a:

PT>BS>SS>SS
SS>SS>SS
rotation

The problem with dummy tests is that SS is ignoring armor, and both BS and PS arent, which puts the favor into the PT glyph quite a bit.

Also there are certain situations in ulduar where using pestilence is a pain due to CC (mostly trash).

I think it's good to work on this though, since we discarded it fairly easy at first mainly due to PT not being affected by reaping before.

[edit] NOTE: PT stands for pestilence, not sure if its the standard abbreviation since most use "pest" but since we short everything to 2 letters i use PT. Also i call it glyph of PT, but i mean of course Glyph of Diseases.

Ok getting some boss data from a random unholy DK, this one i came across on the DPS topic here:
http://elitistjerks.com/f72/t53319-a..._cycle_2_data/

And the WWS here: Wow Web Stats

SS damage (averaged incl crit):
8655 dmg

PS damage:
3525 dmg

IT damage:
2277 dmg

BS damage:
3096 dmg

DC damage:
4985 damage

Lets assess what we really lose and gain.

We already know that with the SS glyph you lose 1 SS and gain 1PS+1IT. With the PT glyph you lose 1 BS

On top of that there a few more things. Ill handle the other 2 later, but one of them is the Tier 7 4-set bonus.

Because of 4xT7:
SS + PT = 30 + 10 RP = 40 RP
PS + IT + BS = 15 + 10 + 10 = 35 RP

So you'd get 5 RP more, resulting in 1/8th of a Deathcoil (doesnt seem much but that is another 623 damage)

So the formula should be:

SS - (PS + IT)
vs
BS - (DC / 8)

Now lets see who wins out:

8655 - (3525 + 2277)
vs
3096 - (623)

SS glyph: 2853 damage lost
vs
PT glyph: 2473 damage lost

-------

Ok so in this case you're looking at a situation where PT glyph is better than the SS glyph.
However there are 2 more factors

First off are GCDs:
With the PT glyph you always have 1 cycle with 4 GCD from runes, and 1 cycle with 3 GCD from runes, per rotation. (BS>PT>SS>SS // SS>SS>SS), with the SS glyph you have 4/3 without refresh and either 5/3 or 4/4 when you have to refresh diseases. So you get 1 extra GCD if you refresh diseases.

However, and this is a personal opinion, I never really struggle with GCD shortage as unholy. More free GCD is always useful as it gives a better oversight of the battle. But i'm not really concerned that it would provide me a huge advantage.

Second and this is where the coin flips in favor of the SS glyph:
The SS glyph makes it so that you often dont have to refresh each rotation.
The 2853 damage lost might only happen once or twice a minute, whereas you lose the 2473 damage every single rotation.

Since the SS glyph is hard to model, i can't really say how often you lose that damage, but probably the most accurate way would be to do this:

Per rotation you have 5 SS's, the chance of refreshing diseases in those is:
1 - (chance its not refreshed on any of the 5 SS procs) = 1 - (0.75 ^ 5) = 0.76 = 76%.

Basically what it roughtly comes down to (please correct me if I'm wrong, its late) is:

You only take the 2853 damage loss 24% of all rotations, and you take the 2473 damage loss 100% of all rotations:

End result:

SS glyph:
677 damage lost per rotation /or/ 34 dps

Pestilence glyph:
2473 damage lost per rotation /or/ 124 dps

I doubt that any benefit of an extra GCD or convenience while AoE'ing box up against that.

[edit]:
I said 1.5 rotations before but that was just a stab in the dark.

I'm not sure if this is much more accurate, i doubt it because i feel i refresh my diseases at leact once per minute, but even if its 50% of all rotations, then i'd still only be 1427 damage lost for SS (71 dps), which is still a fair chunk lower than glyph of diseases.

Tried out the 7/10/54 spec with 2pt necrosis and managed to pull off 7,9k dps on xt (no pot, no aotd, no ae damage) World of Log.

Had Death Bite on that try, World Carver dropped later on during that raid. Naxx BiS with 1 or 2 Ulduar items.

Downed 7 Bossed tonight, so I could test the spec a little. Things i liked were the extra 30 runic power which allow for much more flexibility while 2H Unholy allows you to actually use that extra RP because you're not GCD capped all the time. You can also often intentionally use AMS to fill your RP bar quickly.

Tried 0/10/61 before too, but I really do like the Aggro Reduction, especially on add fights like Thorim Arena I missed the reduced threat.

All in all I liked the build a lot, gonna try it a bit more.

That's probably why your plague strikes and blood strikes are so anemic. Full armor debuffs and savage combat help close that gap in your equations.

2/2 Epidemic's 21 second diseases can occasionally lead to 6 SS before the diseases wear off, which can further increase the chance for the SS glyph to proc. Looking only at 2 sets of runes is a bit short sighted when it comes to the glyph of SS vs glyph of pestilence and I appreciate the rough math done to show how little is actually lost to the glyph failing to proc.

Tried out the 7/10/54 spec with 2pt necrosis and managed to pull off 7,9k dps on xt (no pot, no aotd, no ae damage) World of Log.


Tsubbi I would like to know your rotation if at all possible.

I actually had that originally.

But I'm actually doubting if we can really abstract the math in such a way. With 6 SS it came down to about 80% chance to refresh, which means once in 5 rotations on average that you have to reapply, and I'm quite certain i have to refresh them more than that (could just be bad observation on my side though).

In the end I decided to stick with looking purely at a rotation base refreshing, so I went with 5 SS's. But in fact you indeed have 6 SS's to play with if you have 2/2 epidemic. So the damage lost should be even lower.

Another downside for Glyph of Diseases, is that you're way more likely to mess things up. You have 20 seconds to get in your next pestilence with diseases lasting 21 seconds. If you get a miss somewhere, you CAN'T redo the attack or you'll miss your window. A window that is only 1 second, so VERY susceptible for latency.

hi, long time reader but first time poster.

Regarding the calculations of [Sigil of Awareness] against [Sigil of the Vengeful Heart], I wonder if we are overcomplicating it.

Looking at the 10/61 spec, we see that there are many modifiers affecting both SS and DC but I wonder if it is valid to simply eliminate the common factors and see how much damage is added by each sigil per SS/DC.

Factors that affect only SS and not DC : Vicious Strikes (+30% crit dmg bonus, 6% crit), Outbreak (+20% SS dmg), SS (11% dmg increase per disease, total 33%)

extra dmg (not counting crit for simplicity purpose): 189 * 1.33 * 1.2 = 302

Factors that affect only DC and not SS : Glyph of Dark Death (+15% dmg), Morbidity (+15% dmg)

extra dmg : 380 * 1.15 * 1.15 = 503

a typical rotation for 10/61 assuming diseases up with 4pt7 would be
SS > BS > BS > SS > UB > DC
SS > SS > SS > DC > DC

that would be 5 SS against 3 DC.
extra dmg for 5 SS : 302 * 5 = 1510
extra dmg for 3 DC : 503 * 3 = 1509

Given that our melee crit is higher than spell crit (at least 6% from vicious strikes) and the critical strike damage bonus modifier for SS is higher than DC (230% against 200%), Awareness should pull ahead of Vengeful Heart. I know the calculations presented here are far from precise but if there are any glaring mistakes, please let me know so I can amend them.

If you are doing 7.9k with that spec, why change it up?

In an abstract sense, which would be more valuable as Unholy, 3/5 Bladed Armor and 2/2 2h spec, or 5/5 Bladed Armor?

Basically, it comes down to 4% weapon damage vs ~170 AP for an average DK in 15k armor.

4% applies to Melee, SS, PS, BS, and BCB, while the AP obviously catches everything. What do you think?

Are you people even did XT? Three times in that fight you beat shit out of hearth that take 100% more damage - that is why you can get 8k or other unusual numbers.

This sigil is sidegrade for Unholy - it would be bad to steal this from Frost or Blood DK just like it would be bad to steal T8 when it is sidegrade to T7.

2h spec is better per point than BA.

XT DPS is inflated of course, and not represantative for all bosses. In the log you can browse through other fights if you want to. Unfortunately there's no Patchwerk-type boss atm for true single target dps without distraction.

One thing that i have never seen mentioned about rotations is the best use of Blood Tap. Pre 3.1 you could "bug" a blood rune to become a permanend deathrune. This allowed for 2 extra SS for every following Blood Tap. With 3.1 the bug was removed but you can still get 1 extra SS by doing the following:

Rotation:

1st: PS, IT, BS, BS, SS, Runedump

than:

2nd: SS (with UF), SS (with Deathrunes), immediately Blood Tap -> Bloodstrike, SS (with UF), Runedump

This allows you to use SS three times in the following 3rd rotation.


Also try to use the 2 second delay on runes as much as possible to never get RP capped.

Sounds to me like you should take 2 pts out of Bladed armor and put 2 into Necrosis.

Math: (might be off, doing it quickly...)

14,000 armor = 388 ap w/ 5/5 bladed armor

3/5 bladed armor = 233 ap = 155 ap less = 11 dps lost

------

1 pt Necrosis = .7% dps

3/5 Necrosis = 1.4% more dps

------

Assume 6k original dps.... 6,000*1.014 (Necrosis gain) - 11 (BA loss) = 6073 dps

EDIT: Thinking about this again, my math is much too simple... didn't take into consideration the effect that 155 ap has on spells, diseases, etc.

If somebody wants to take the calculation of that 155 ap a little further that'd be great, but I doubt that in the end it will result in another 73 dps lost.

I haven't seen anyone post this spec yet, so I will. This is the "my guild has another unholy DK" spec and allots the points as 10/10/51.

With another Unholy DK in the raid, Ebon Plague will be on all targets already, so you don't need it. The only thing you need is a 3rd disease, which you get with a single point in Crypt Fever, freeing up 5 talent points. You lose the crit also, but essentially its exactly like giving up Dark Conviction, which no one seems to have a problem with.

Toughness is more DPS than Improved Icy Touch as a filler, if you have Bladed Armor (even 3/5 as I do). Icy Touch comes out to 2% or less of my damage on most fights, so increasing it by 15% is pointless. On many fights you can use Pestilence to keep diseases active without losing a SS (arms on Kolagarn, Feral Defender on Auriaya, Flash Freezes on Hodir, etc) for much or all of the fight, meaning you cast very few ITs. 9% more armor contibutes to BA and raises your AP enough to easily compensate.

Random aside: The bug that gives you a Death Rune when BS is dodged/parried can be used to cast Ghoul Frenzy, in addition to once per minute with Blood Tap. You should be able to keep GF up a fair amount of the time with it on any fight attacking from the front (Auri, Kolo) or with lots of movement (Hodir).

Questions:

BA vs. 2h Spec vs Black Ice
As you can see, you have the choice between removing 2 points in either BA, 2h spec, or Black Ice. I am speculating that you want 2h spec and Black Ice and should sacrifice 2 points in BA, but if someone disagrees, I'm interested.

RPM vs. Toughness
Since Toughness is contributing to BA, it provides a tangible benefit to your DPS, while RPM only provides a convenience by allowing you to cap your RP at a more useful amount. I like hitting the cap RP at the end of a rotation and dropping 3 RP moves, but maybe I need to tighten up and not cap without it.

Dirge vs Necrosis
Its hard for me to weigh the benefit of Dirge. Necrosis with 5 points averages around 3% of my DPS, which is pretty low, but I only have 31 Armor Pen and 160ish Haste, so as my ArP/Haste increase, Necrosis damage will too, obviously (since it is a % of white damage dealt). We're only talking 2/5 Necrosis vs 2/2 Dirge. I am guessing that Dirge is better, but I am curious if this is always true or if Necrosis might get better with enough ArP/Haste.

Give me your thoughts.

I may have missed this, I know that unholy blight is worth a point, its definitely worth using with more then one target, but is it worth using on purely a single target or is it best to use all RP for DCs on a single target?

When I see people say this I suspect if the best raid spec will be to pull points out of necrosis one by one and fill in subversion. The result will be specs like:

1/10/60 or 2/10/59 or 3/10/58

As DKs approach BIS they will be forced to move from left to right.

There are so many movement fights in Ulduar where you can't stick on a boss for the entire duration of Unholy Blight, so it doesnt make sense to spend RP for it over a single coil. For fights where you can stick on the boss I would use Unholy Blight with coils.

Methods did a calculation of Dirge on the old thread, and it was pretty large contribution to DPS. He said that you should never be with out it.
When I have a little more time Ill dig it up for you and post it here.