The new Scourge Strike "double dips" into crit - that is, both parts of the attack can crit independently of each other as far as I'm aware.
This drastically improves the effectiveness of crit - especially a flat out 5% increase to *everything* that can crit (including Blood Plague assuming 4pT9 in 3.3).
As Scourge Strike will make up the bulk of our damage, the increase this 5% crit gives is far larger than an increase to Death Coil damage (especially considering the halving of UB damage) whilst Necrosis has simply been dropping in relative value since the start of WotLK.
At least that's my interpretation - at work so couldn't run a sim with Morbidity/Necrosis.
Ok that definatley sounds reasonable. Unfortunatley I'm completly fail with sim stuff else I would try it out myself.
Yes, I am above melee hit cap, but not spell hit cap so that additional hit is not wasted as you seemed to imply.
In fact you are. You need 17% spell hit over all. I suppose you have a shadowpriest/moonkin/whatever in your raid to provide an increase of everyones chance to hit with spells by 3%. You're also specced into virulence, that's another 3 % and on top of that you're likely to have a draenei in your group providing yet another percent of increased spellhitchance. That's 7% through talents/synergy. 17% - 7% = 10% required spellhitchance on your part. You currently have 12,58 %, so it's save to reduce your spellhitchance by 2,58%.
Sup everyone. I ran a sim today using Kahorie's sim and found something interesting.
Average for DryRun | 8528
Average for AttackPower | 8591
Average for Strength | 8626
Average for Agility | 8573
Average for CritRating | 8598
Average for HasteRating | 8635
Average for ArmorPenetrationRating | 8565
Average for ExpertiseRating | 8487
Average for HitRating | 8441
Average for SpellHitRating | 8550
Average for WeaponDPS | 8564
Average for WeaponSpeed | 8546
These were made using my current stats on the armory, the 17/0/54 spec w/o BS, blood, RotFC, 150 hours and the unholy priority rotation. So with haste being at 3.4 and str at 3.11, is the sim suggesting to gem for it or was my sim flawed? I understand that str is superior is AoE situations, I'm just wondering if gemming for haste would be better according to the sim.
These were made using my current stats on the armory, the 17/0/54 spec w/o BS, blood, RotFC, 150 hours and the unholy priority rotation. So with haste being at 3.4 and str at 3.11, is the sim suggesting to gem for it or was my sim flawed? I understand that str is superior is AoE situations, I'm just wondering if gemming for haste would be better according to the sim.
I'm thinking you have something wrong in your sim. Haste has never had that high of value for any Death Knight spec, and there really isn't anything in the upcoming patch to increase it's value to the point you are seeing.
Haste rating is unlikely to be that good. Because it changes when spells occur and also when procs occur it's actually quite hard to get an accurate value for the value of haste rating. Additionally the simulator when doing an EP run isn't spell hit capped which can have an effect on some attacks as you can see in the damage breakdown.
If you want to get a better handle on how haste scales compared to strength (or crit), I suggest you use the stat scaling option with your base gear minus say 100 strength and 100 haste. Doing that you'll find that the dps varies with haste rather randomly with sometimes adding haste actually lowering dps. Strength is a lot more consistent, and for reference it is pretty much a consistent 3.0-3.1 AP (with the GotG).
I'm not sure if this is a question for the Rawr thread or here, but anyway.
There was desire for an unholy DK in my guild so I re-specced to fill it, and I was checking out optimizing for the Chaotic Skyflare when I noticed Rawr shows the "thundering skyflare diamond" a chance on hit haste proc at higher dps rating, and it can be activated using one nightmare tear.
Is there any reason this gem is not used/not popular?
I can see that the crit bonus on Chaotic might be a lot better for AoE and the current state of 4pT9, and more static, thus prefered.
It's not used/not popular because it's subpar. I can't comment on what Rawr is doing wrong, but 3% crit damage + 21 crit rating > 48 haste rating (which is what Thundering comes down to, more or less).
If, say, half of your damage can crit (a low estimate, with 4p t9) and you are doing, say, 8k dps (again, a low estimate, at present gear levels), then 3% crit damage is 120 dps. 120 dps (higher, in reality, since there were all of those low estimates) is easily better than 48 haste, especially when you factor in the 21 crit rating and/or AoE scenarios.
I would advise using Kahorie's Sim for stat weights and then Zerack's Optimizer for gearing options. I would not rely on Rawr in this area.
I just wanted to add that if you use a simulator like Rawr for live (3.2) AEP and gearing choices that it can be a bit misleading sometime.
One of the reasons is that the current 10 second rotation with GoIT is a bit on the tight side, and haste makes that rotation fit right (lowering GCD on IT and DC).
So the reason haste might be overvalued in AEP or suggested on things like meta gems, is because you are somewhat low on haste. Which is a good posibility if you're respeccing from frost or blood (it took me ages to get my haste up to about 200).
Still wouldn't quite explain the meta gem (as it is "burst haste"; a lot of haste for a short period of time), but maybe rawr averages the haste and bases its conclusions on the over time estimate gain.
Does anyone know the difference currently of using a 16/0/55 spec vs a 17/0/54 spec when 1 pt is shifted from necrosis to max out dark conviction (or bladed armor if you went that route for your 16)
Does anyone know the difference currently of using a 16/0/55 spec vs a 17/0/54 spec when 1 pt is shifted from necrosis to max out dark conviction (or bladed armor if you went that route for your 16)
The former is currently (as far as i know) better on single target fights, the latter is definitely better for multi-target fights like Anub.
[*]The AoE rotation will revert to the old:
PS > IT > Pest > DnD > DC > HoW // SS > BB > BB > SS > DC > (DC)
Is there a reason for the AoE-Rotation to exclude BS for Desolation? I have just checked it with some numbers from our Anub25H kill and as I've found out a rotation of IT-PS-BS-SS-PE-DC-HoW-DD-BB-SS-DC-DC outperforms the proposed rotation for any positive number of targets. I did not even include things like Necrosis, BCB, etc. in my calculation, which would further strengthen the argument, since desolation would yield a net gain for them. Of course, the numbers are with current gear and spec and leave a certain margin of error, however, they are robust enough even in face of this fact. If people want to have Death and Decay in the first rune set for whatever reason they could always just Blood Tap as it won't be needed (as frequently) for Bone Shield in 3.3.
The calculations are below, as stated I haven't included all single-target damage components, since they would only strengthen the point. All values have been divided by 1.05 to exclude the effect of Desolation.
BB 2778
BP 1951
DD 1435
WP 1902
FF 1822
Me 4081
DC 5509
BS 3108
IT 2197
PS 3381
SS 6274
EDIT: I just realized, a drop of 4T9 would actually tip the balance in favor of not using Blood Strike. To account for that I divided WP, FF and BP damage by 1.3 and solved the equation:
n * 2458.6 + 6628.5 < n * 2778
6628.5 < 319.4 * n
n >= 21
So, without 4T9, a Desolation-less rotation would be better for 21 or more targets, which would need further calculation due to the AoE cap. 21 targets is definitely a lower bound, as I've not taken into account all single target abilities and BB is affected by the AoE cap, while diseases are not (so all of the right side is affected, while not all of the left side of the equation). Even though one Anub'arak parse is by far not representative, the numbers certainly still stand in favor of a AoE-rotation including Desolation. Unless I made a mistake, which is always possible.
You didn't have to include all of that math. I know one needs to keep up Desolation when AoEing. I simply forgot. Such happens ^^.
Besides, if it did need proving, however, Anub'arak would be a poor fight to use to do it.
Anyways, with the new version of the sim, I'm getting "normal" haste values (~1.53) for 3.3. 1.5 is about where it was pre-3.2.2, and as such seems much more logical than the previous ~1.0 the sim was offering for the new patch. It still makes haste worse than agility and everything else, but doesn't make it quite so distasteful.
All the other stats are pretty much unchanged (aside from spell hit being higher than hit, but I'm thinking that's a sim error).
Edit: Maybe nevermind. It's giving me the old value of 1.0-1.1 when I turn off tier 9 bonuses (which should have no impact on the value of haste), yet 1.5-1.6 when I have them on.
Undocumented Patch Note (just tested and confirmed on PTR): AotD's cast time is down to 4 seconds.
Currently, AotD does between 80k and 90k damage over its duration. Next patch, that will be 40k to 45k. With a 4 second cast time, you'll have to be doing less than 10k to 11.25k dps for AotD to still be optimal to use even mid-fight on encounters without phase transitions or downtime.
Unfortunately, between ICC gear and the patch changes, we'll probably surpass that number (or come very close). Still, a buff is a buff.
hmmm interesting. The dps cost isn't necessarily that much. You're really only looking at 4 seconds of autoattack/BCB damage, a scourge strike, a blood strike and what 10 runic power? You'll still be doing your disease damage and pet damage which is roughly 2K out of that dps. Now, do they get any benefit from heroism/blood lust? If so then it probably makes sense to save it for then.
Now, do they get any benefit from heroism/blood lust? If so then it probably makes sense to save it for then.
Yes, like the Garg, AotD takes a snapshot of your haste when you use it. In addition, Lust should make you cast Army faster.
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
There hasn't been much discussion on Glyph of Scourge Strike here (atleast I haven't noticed it). I was playing around with glyphs on the PTR and I noticed that I got overall higher dps on dummies by using GoSS instead of Glyph of Frost Fever.
The rotation with GoSS felt a lot smoother, diseases dropped at the same millisecond when my runes came off cooldown; therefore no clipping on any ticks and EPB + BP uptime was yet 100%.
Then again, when I used GoFF instead of GoSS, my runes were ready a second before the diseases were dropping.
And that point is a bit problematic.
If I refresh them immediately: I "lose" the damage of the last ticks.
If I wait for the perfect spot: It's almost one global, but not quite (so can't use HoW or DC or w/e) before the diseases drop.
If I just wait, it's a "waste" of almost-one-global, If I don't wait and use some ability, I lose disease uptime+some possible debuffs.
So is there really some mathematical, practical or whatsoever proof, that 20% extra damage for Frost Fever outweights the damage difference of SS vs. PS+IT (in a single rotation) and the slight "problems" with the rotation?
EDIT: And yes, obviously I'm talking about single target DPS. For AoE heavy fights; the glyph choices are easy.
I noticed that I got overall higher dps on dummies.
Using the test dummies for anything other than testing a rotation or a mod will give inaccurate results. Anyway, if you don't trust the Simulator then use the glyphs you wish.
To answer a question, you would refresh diseases between time 0-2.9 seconds left.
Millions of words are written annually purporting to tell how to beat the races, whereas the best possible advice on the subject is found in the three monosyllables: 'Do not try.'
Using the test dummies for anything other than testing a rotation or a mod will give inaccurate results. Anyway, if you don't trust the Simulator then use the glyphs you wish.
To answer a question, you would refresh diseases between time 0-2.9 seconds left.
Thanks for your input.
Had another thing in my mind, too. What about Necrosis vs. BCB? Is there a realistic point of crit rating that turns the table in favor of Necrosis instead of BCB for point-per-point? I mean, BCB doesn't scale with crit, Necrosis does. They both scale with haste and ArP.
It's an interesting question. I played around with this a bit in the simulator at one point, it's currently undervaluing GoSS with most configurations because it renews the diseases the instant the last tick is supposed to go off and you don't infact get the tick damage (although you do get the 10% RoR applying to the new blood plague). When I did some approximate calculations to compare glyphs, the damage from the extra ticks (10 instead of 9 every 30 seconds, if it was managed perfectly) is actually more beneficial than the extra dps you get from replacing a third of your IT and PS with a SS.
However before we can actually get a complete handle on it, we need to work out how latency, communication and the server processing actually effect the reapplication. A simple model would probably be that 50% of the time you reapply before the tick happens and so miss out on the extra tick, and 50% of the time you got the extra damage (but if its a plague strike then you miss out on RoR), but due to the complexity of the process i wouldn't bet on it.
It would be good to test out exactly what happens without any gear that provides procs, to be able to model what goes on more accurately.
That said, the disease glyph is probably better for doing that, and provides the opportunity to roll diseases. I'll post some simulations to that end when i get a chance.
Consider, I may be missing something obvious, but why BS instead of BB for the AoE rotation? Typo or it's a choice tied to Reaping (I can't see the logic in that case)?
I believe this was covered just on the last page, but it's to keep up the 5% damage buff from desecration (along with 2pc t9, if you still have it)
11/10/09, 7:00 AM
