Since I dropped the SS glyph and running with ghoul/UB/DD I basicly use this rotation:

SS > SS > IT > PS : dump
SS > SS > BS > BS : dump

Whenever something unexpected happen I "restart" rotation with PS > IT > SS > BS > BS, kinda idea to cut my losses and go from there. Each 20s I can squeeze out 3 rp attacks (rotation gives 125 rp as 0/10/61 with no 4p t7) and have 1 free gcd for any other stuff like gargoyle/HoW or catch up on the freak miss/parry/dodge/resist. Now 6 gcd in a 10s frame should be np, and starting out with a double SS each time then finish with BS or disease attacks depending if you have death runes should fit your a/b concern. About c, reason I prefer the above mentioned rotation:
- blood runes -> death runes at end. This avoids blood runes getting off cd late, because a frost/unholy rune was used instead of the death rune due a minor delay.
- each 10s has the same rythme: 2 rune attack, 2 rune attack, 1 rune attack, 1 rune attack. Would you not do this you get pushed behind 1 gcd where you wait on the 2nd 1 rune attack to drop a 2 rune attack in the other part.

@Sekke, even with 4p t7 the SS glyph is less then a 50 dps increase, favoring (multiple) other runes.

(Insert random "long time reader, first time poster" paragraph here)

I am not really sure if it is important at this point, but since I have done some work myself on the SS glyph in the past few days, I don't think it is a bad idea to post it here for everyone to see. Quite a few people have already posted estimations about its value, but the numbers seem to vary a bit. I think though that Amroo's simulation of a 3 min fight is quite accurate. I tried to approach the correct answer from a more theoretical point of view. However, my probability mathemactics are not that strong and I had trouble modelling efficiently SS glyph procs that are happening after one (or more) of the SSs in the first 20 second rotation proc. Therefore, I tried to use a more analytical approach (with mathematical sequences) and I think I found the solution.

Let's assume the following sequence of SSs:
PS+IT or SS glyph proc -> SS -> SS -> SS -> SS

Chance that 1st SS procs:
Chance that 2nd SS procs:
(Note: Chance that 1st SS doesn't proc multiplied by chance that 2nd SS procs)
Chance that 3rd SS procs:
Chance that 4th SS procs:
Chance that there are no procs:

We want to calculate the average amount of SSs that we can cast before the diseases fade from our target. To achieve that, first, we are going to make a very rough estimation based on the assumption that only the 4 SSs (let's call them 1st degree SSs from now on) described above can proc.

If we expect none of the SSs to proc, then we can only cast 4 of them before having to renew the diseases with PS and IT. If we expect the first SS to proc, then we can cast that one and 4 more, for a total of 5 SSs. In the same way, if assume that the second SS will proc, we can cast in total 6 SSs, 7 SSs when we expect the third one to proc and 8 SSs for the last one.



Obviously, the above number is far from accurate, because we didn't take into account the fact that any extra SSs casted because of previous glyph procs (let's call them 2nd degree SSs), have a chance to proc as well. However, we can use the previous result to get a better estimation, by taking into account the chance of the 2nd degree SSs to proc as well.

If we expect none of the SSs to proc, then, as before, we can cast only 4 of them. If we expect the first SS to proc, then we can cast that one and 5.653 more, for a total of 6.653 SSs, which is the average amount of SSs before the diseases run off that we just calculated. Therefore:



Again, that last number is not accurate, because we assumed that the 3rd degree SSs cannot proc. However, we can use that number to make a third estimation that is going to be even better and so on. Essentially, we have constructed the following sequence:



with

The Nth member of the above sequence is the average amount of SSs that we can cast before the diseases fade, while making the assumption that only the SSs up to Nth-degree can proc. It can also be written as a function of n:



Therefore, the limit of the sequence is the average amount of SSs that we can cast, assuming that all the SSs can proc and it should be the real benefit of the Scourge Strike glyph.



So, in an endless fight we can cast on average 8.649 SSs, before having to cast again PS+IT. A specific member of the sequence, e.g or for exceptionally long encounters, might seem better suited for practical use, because otherwise, by using the limit to infinity, we are taking into account SSs that we will be never casted, since combat in WoW lasts only a few minutes. Still though, the difference is quite small and I don't think it is an issue. Assuming that a 20 second rotation includes either 1 PS, 1 IT and 4 SSs or 5 SSs, we can conclude that by using the glyph we have to cast PS and IT once every 1.930 rotations, which is 51.8%.

On a 3 min fight, there are in total 9 20 second rotations, of which 51.8% will have 4 SSs and 42.8% will have 5 SSs. That is a total of 18.648 + 21.690 = 40.338 SSs. Without the glyph, we would had casted only 36 SSs, so the benefit was 4.338 SSs which is almost exactly the same result as the one from Amroo's simulation (4.3365).

I gave up SS glyph as soon as as t7 got nerfed and I never looked back. I love the set in stone rotation, and i believe that SS glyph is a dps loss.

Anyway here's my question my rotation is ps-it-bs-bs-ss-UHB-DC-horn-SS-DC-SS-SS-DC-DC

The problem is after my final DC my first unholy is about to come up but my diseases still have 1 tick left.

Do I gain more DPS from waiting 1 second to let my diseases fall completely off or do i start my rotation immediately??

I assume figuring this out wold be very complicated

No, not really. Clipping diseases would yield a damage loss of about 2500 (1000 per disease tick + WP chance roughly at 25%). Now, if you wait every time for 1 second, after 20 rotations (400 seconds or 6.5 minutes) you will have delayed an entire rotation consisting of 1 PS, 1 IT, 2 BS, 4 SS, 2 DC and 1 UB.

Taking the following average values:
SS: 6500
PS: 3100
IT: 2300
BS: 2500
DC: 4400
UB: 5300

This yields a loss of 50500 damage over 20 rotations or 2525 damage per rotation. This is already higher then the dps gain, but in addition to that, you deal 10% more damage to targets afflicted by BP, so if your BP runs out your next PS will deal 10% less damage, which is another 300 damage difference. So clipping the last tick is generally nothing that is advised against.

@Sekke: I didn't use 4T7 in the simulation, since 4T8 is better and most people will have switched soon (if the loot god smiles down on me, perhaps I, as well).

@Alexnick: The small difference will probably be due to the fact, that the simulation takes small variances in the FU (DD) set you use for SS into account, since not all of them are absolutely equal. But the difference is neligible, so what method said stands, that all our simulations / theoretical approaches give the SS glyph pretty much a zero value.

How are your PSes hitting on average for 3100, mine are hitting for 1100-1300 and critting for 2700-2900. Those are some magical values there.

I didn't look at IT nearly as extensively but 2300 still sounds like a high "Average".

SS sounds a bit on the money though (Maybe a little low? Mine hits I believe for 5-5.7k and crits for nearly 10k and if I'm lucky over 10k). I'm going to start doing some WWSes on my character but that is very suspicious there :P

Also I'm looking to do some updating to the first post mainly about the arguement for and against SS I don't really see too much to update though.

I checked the last several pages conscerning the SS glyph, but haven't seen anyone mention if this is going to affect our choice of Sigil or not. If we start using the UB glyph and DD glyph together, this is going to amount to stronger, more frequent DCs. Just through personal testing I find the two to be about the same actually, but does anyone have anything more conclusive?

With my stats running the simulator I get:

Averages only
Scourge Strike 6566
Plague Strike 2421
Icy Touch 1945
Blood Strike 2115
DC 3763
Frost Fever 800
Blood Plague 801

6566
4368 = 2198 damage gained from Scourge strike simply over IT and PS.

WIth same gear level say I lose an entire tick of my diseases that's 1600 (800 per tick on average from the same simulation)

so I still gain at LEAST 600 damage per hit, now say my diseases just ticked, I wouldn't lose that extra 1600. Worst case, I also lose a wandering plague tick, which would make my SS a loss of 200 damage with the extra 800 damage lost.

So include that it also frees up another GCD which you may or may not use, Scourge Strike isn't necessarily a loss. I don't think I'm forgetting anything here. So I'm just finding your #s to be a bit unrealistic and disproportionate. You can argue my #s off as I am arguing yours are off, but even if my PS is hitting harder that would most likely mean my Scourge strike is also hitting harder.

So I decided to do another test using the simulator, this time I bumped my strength from 1342 (I was actually facotoring in my Crafted boots) to 2000, these are the numbers I got for this simulation.

Scourge Strike 7435
Plague Strike 2814
Icy Touch 2413
Blood Strike 2543
Death Coil 4889
Frost Fever 1047
Blood Plague 1049

7435 - 5227 = 2208 gained over strictly IT and PS.

It does seem to me that the scaling is in favor of PS and IT though because here the two ticks from the diseases catches up if you blow an entire tick of the two diseases you'll lose just short of all the damage you gained. I still haven't gotten to test out what happens if you delay your strikes to move in synch with your disease ticks (I can't find a good addon to allow me to do that easily) though. So there is some debate opened there, but how many of us are pushing 2k strength?

Whether or not I'm arguing for SS Glyph here it's hard to say, but your numbers seem completely unrealistic.

And of course SS provides more freedom in your rotation via freeing up a GCD. Which is honestly sometimes just useful for running out and not losing as much DPS hah. But yeah we're not exactly GCD starved. Yet.

Unholy already has spare GCDs so you can use HoW every 20 sec regardless of using the SS Glyph or not.

The numbers in the simulation (you can check the sheet) are beyond realistic, they are, in fact, real, since I used the top 3 Vezax Hard parses on WMO (a completely unbiased fight for that matter since single target and no non-standard buffs, the anti-haste-aura does not affect results). For the caluclation of the cost of delaying the rotation I did use Amero's parse which had a high crit rate on PS, what explains the high PS number. But that doesn't really change the outcome of that particular calculation, so I disregarded it.

The numbers for the calculation on the SS glyph, however, very very real.

Edit: Since calculations assume that you don't have 4T8 you have to factor in the 5 RP lost or 1/8 of a DC, as well. And on average you clip about 1.4 ticks of each disease per extra SS you get. So your numbers start to converge to the ones I postet.

Was hoping to get a little clarification on expertise as unholy.

Currently, i'm sitting at 23 expertise in my gear set, however, i have the +15 expertise enchant on my bracers. The BiS thread OP states to use the 50 AP enchant. Now, the 15 expertise works out to about ~+1.83% expertise. Yet, at the T7 BiS subsection, it lists the expertise rating at 19. Is anything above 19 superfulous for unholy so they don't list the enchant? Or is it a simple matter of the benefits of +50 AP enchant outweight the expertise? I'm a little concerned about going into Ulduar and not being able to be consistent with my numbers. I was just hoping for a little clarification is all. I was under the assumption that i had to look to cap expertise at all cost (after hit, of course).

Even in blood presence Unholy has spare global cooldowns that can be used to make up for dodged strikes (Unlike Blood, whom are pretty well crammed for GCDs). Expertise isn't bad by any means, I'm expertise capped as Unholy because of Worldcarver and the orc racial, but not everyone will be like that. It's just not as good as strength.

I actually facotred in 4T8 into the stronger parse, however I now understand that you merged your disease ticks into PS and IT which makes a bit more sense.

And yeah I was pretty much asking forlinks to your posts so I could take a look at them. However since you're saying on average you cut 1.4 seconds off your disease ticks, you'd actually only apply 50% of the damage (Since it's just delaying them not completely removing the ticks) as damage lost (The real model would probably be more complex but that's as simple and still accurate as I can depict) And you say damage lost on PS because you don't have the debuff but you're supposed to reapply PS and IT before they drop off, not after otherwise you've already lost disease ticks. - This is just clarification because I realize you've already done that.

I'm not going to say SS glyph is dead just yet but I certainly wouldn't fault anyone for not doing it, however for me the playstyle is much more fun, and is worth the small DPS loss (As long as I'm still doing competitive DPS I probably wouldn't be picky) but for the sake of testing I'll probably spend the next few weeks doing parses of my raids with different glyphs applied and post results here. So for these tests and I urge anyone reading to try this as well, what glyph combinations do we want to see tried. Unfortunately my guild hasn't been as progressed as much as I'd like (Lack of healers kills :/ ) but first combo is SS Glyph, Death, and Ghoul. Second would be Death, Ghoul, Unholy blight. Do we need testing on Icy Touch glyph for the extra runic power? I would think Unholy blight blows that out of the water most the time.

Icy Touch is clearly subpar to Unholy Blight. It will provide 30 RP/Minute, while UB will provide 40 RP/Minute in addition to an extra GCD (which, as has been stated several times, isn't that valuable for unholy, but regardless, 40>30!).

Also, a (very) minor point to remember in calculations which some people seem to be ignoring; 4p t8 buffs the damage of blood strike as well, which will slightly increase the value of the set bonus. It's not a ton, but when you are trying to figure out what is most optimal, every bit counts!

You don't gain an extra GCD with the UB glyph, you just get to spend on of your UB GCDs and RP on a death coil. You lose a GCD every 80 seconds or so with the IT glyph as you're gaining RP but have no extra time in which to spend it.

Either way UB wins hands down anyway.

No, I didn't merge the disease ticks yet, but the "3100" for PS was from one of the three parses that had a high crit rate on PS. In that post where I used the 3100 value I just needed some benchmark value to calculate the cost of an entire rotation as somebody asked if it is worth it to delay rotations by 1 sec to give diseases their final tick.

For the other calculations, which we are referring to, I used three different parses, and the other two had PS values of about 2000+ I think. It was just to see how things work out when you have a a) above average crit rate on SS, b) above average crit rate on PS / IT, or c) average crit rates. Even in case (b) SS glyph was still slightly beneficial before disease-clipping, but in case (a) it had, of course, the highest value and in case (c) a value between (a) and (b).

As for the disease clipping, just look at the calculation I made earlier. SS-glyph yields on average 4.3 extra SS per three minutes and these SS actually decrease the number of how often you clip diseases (because otherwise you would have used PS-IT, which would have clipped diseases). However, during those 3 minutes you do 40.3 SS casts and each of them has a 25% chance to proc, so you have an expected value of procs of about 10 (one proc is not equal to an extra SS, it just delays the next PS-IT; it is equal to one PS-IT once you have delayed it by 20 seconds, which needs on average 2-3 procs). Those 10 procs are "additional clips", you subtract the saved clips and get about 6 "final additional clips". Each of those clips cuts on average 1.5 seconds off, so every other clip is an entire tick of both diseases, so 3 ticks over 3 minutes or 1 tick per minute or 2000 damage per minute + a potential Wandering Plague proc. Which is how I got the 44 dps lost. The most that was won by the SS glyph was 48 dps in the parse with an above-average crit rate on SS, so even there it is +/- 0.

----

It is easily possible that all the simulations miss something, perhaps we overestimate the cost of disease clipping or underestimate the value of auto-refreshing diseases. It really takes a lot of data before any conclusions can be made, and even then it's probably highly dependent on personal preference, so I agree that the SS glyph shouldn't be dismissed, yet.

I believe that Amroo's PS number that you are referring to is off by about 300-400. If I remember correctly, SS+DC+IT hit harder in BI specs (at the cost of auto), while PS+BP+FF remain roughly the same. 6.5k is almost certainly for 2HWS specs, so PS probably averages around 2.7k (instead of the listed 3.1k). Note that overall (with bigger SS/DC/IT), the SS glyph is generally worth more in BI specs than in 2HWS specs.

- - -

There are tactics you can use to maximize your disease time without an addon. The most basic of which is to space out your SS.

Consider this starting sequence...

PS>IT>SS>BS>BS
ub>SS>hw>SS>dc>SS

Paired with a repeating sequence...

??>SS>??>SS>BS>BS
??>SS>??>SS>??>SS

...where you sub ??>SS with PS>IT as needed.

Compared that to the standard rotation without a SS glyph:

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

Unless you have very high latency, you are actually gaining disease ticks with the SS glyph.

Remember that every time the standard rotation refreshes its diseases, it effectively loses 4-sec worth of diease damage (2 sec loss for 2 diseases) because its disease refreshes are two rune sets apart. Effectively, the standard rotation only has 18 sec disease uptime every 20 sec.

Now, the SS glyph rotation refreshes diseases roughly 1.6 times as often (if I remember correctly). However, almost all of the refreshes are either near-zero-loss or 2-sec loss; very few refreshes incur a 4-sec loss that the standard rotation refreshes incur. This results in a net gain in disease damage.

- - -

There are other advantages to the SS glyph, of course. Whenever you are forced to break melee contact with your diseases ticking, your "redundant" refreshes may suddenly become not so redundant, as you are significantly less likely to be at the tail end of your disease duration (the redundancy also helps when the boss is about to die).

I think that, given the frequency of breaks, I am not even sure that the UB glyph necessarily beats the Dark Death glyph overall (it obviously does in tank-n'-spank, of course). The SS glyph, on the other hand, I am fairly certain still beats the two RP glyphs, even though its value is diminished with the loss of 4pT7 and the advent of the VH sigil.

I think the rotation you proposed is actually very hard pull. I often switch between Frost and Unholy as DPS specs and once you have practiced a bit FS weaving is relatively easy, however, only due to the fact, that the rest of the rotation is more or less set in stone. What you propose is sort of a priority system mixed with rune dump weaving gone wild and, while the idea isn't bad, the efficiency losses it induces once you make slight mistakes (which is very very likely here) simply kills it.

The point about UB-glyph is valid, as well, and the reason why I'm still running with Ghoul/SS/DD. You have 385 RP per minute (saying you use HoW once), of which 25 are used up by Gargoyle and 120 by UB, leaving you with 240 or 6 DC. DD is then worth 0.9 Death Coils versus the one Death Coil from glyph of UB. Since you always get the benefit from the DC glyph and situationally you don't get the benefit from the UB glyph, they are probably equal or even favoring DD for reliability.

You hvae points in morbidity with 12-0-59?

Hm, I'm guessing my 2 floating points that could have gone to morbidity went to IUP instead. I'll stick with the run speed buff.

Hello all. I use 0/10/61, and I have a suggestion. It occurs to me that the 2 points in morbidity can be moved into Icy Talons, providing you with 8% haste to attack speed. I'm not sure if it would increase dps or not because I'm bad with theorycrafting, I normally read the frums and do my own trial and errors. If anyone can please do the math, or comment on this idea please do so. I will be trying it out tonight.

The reason these things aren't mentioned is because they are mentioned in the opening post where I say you can move points around.

This is a really interesting point, which I haven't really seen in the debate up to now.

Miracleknight just answered my question. The reason is that in 25 mans, if someone provides haste, you will not benefit from this talent; or something amoung those lines.