In the client spell data that Simcraft parses the constant term is listed as ranging from 776-866. That would suggest that the RNG term is affected by the weapon damage multiplier as well since 2*90 = 180 matches your pre-hotfix testing. I believe it's safe to conclude that Aimed shot is implemented with the weapon damage multiplier actually being a total multiplier:

160%*(weapon_damage + 0.724*RAP + random(776,866))
(where weapon_damage includes the RAP contribution to dps normalized to 2.8 speed)

Since it will still be a while before I have my complete analysis updated and in the OP, I wanted to give you all a heads up on a new benefitial rotation option for some of you.

The AI nerf has made AI damage outside the CA phase just a little better than CS when just comparing them alone. However, when factoring in the shorter CS cast time with a partial SS to take up the difference, SrS refresh damage, autoshots not being locked out, and more WQ opportunities, CS is the big winner.

For those of you with the T11 4-set, I had previously provided two options - a high haste option with 2 AIs in the cycle with CS refreshing SrS late and a low haste option with one AI in a 10s CS cycle. Neither of these were great options, The first was very haste demanding and unforgiving of non-optimal execution with SrS falling off. The second didn't utilize natural haste from gear well.

With the nerf to AI, I believe that the first option is not a good alternative, although I have not completed the analysis. The second option is still okay, but what appears as a good option is using the CS glyph for a 9s CS cycle with 1 AI.

With this option I would like to recommend, although I have not finished the analysis, is going with the CS glyph instead of the AI glyph with about a reasonable 8.5%-9% haste from gear and 1 point in Termination. This would result in a sustainable cycle of CS-AI-SSx5 (order flexible) during the Standard phase with the MMM AI proc replacing the 5th SS and a similar cycle during the KS phase but with only casting KS to replace the 5th SS on cycles where the MMM AI proc does not (so about every 18 to 19s).

I do not quite yet have the T11 4-set myself to test this out. Could someone that does please do so and report the results.

Read below if you want more details on how I came up with this recommendation as well as some caveats.

So let’s look at the haste requirements for casting CS off the 9s CD. With the constraint of two SS pairs over the rotation to maintain ISS uptime and to try to balance focus costs of the special. The default rotation to look at is the chosen one for without the T11 4-set:

CS-SS-SS-AI-SS-SS-SS or
CS-AI-SS-SS-SS-SS-SS

For this to fit in a 9s CS CD, the amount of haste needed from gear can be determined from the following (note that the 1.30295 factor is the total haste from the raid buff, ISS, and 3/3 Pathing):

9s – 1s CS GCD = 5 * 1.8 / 1.30295 / x + 2.9 / 1.30295 / x
8x = 6.9074 + 2.2257 = 9.1131
x = 1.1391

Hence, about 13.91% haste from gear is needed for this to be a tight rotation. The focus regen requirement in this case with assuming no AI glyph is 94 – 5 * 9 = 49 focus or 5.44 FPS or less than 4.5% haste from gear. Thus, this rotation is focus positive. To support replacing the fifth SS (or third) in the cycle with the MMM AI proc when it occurs, requires about 23.7% haste from gear. Considering that the MMM AI proc should occur every other cycle, then the haste requirement for a focus neutral rotation is 14.1%. With the AI glyph, the requirement is 9.8% haste from gear.

Now, I am sure your first reaction is, if you couldn't support the previous 10.8% or better, what makes me think you can support 13.9%? Well, I don't. The good news is that you need the 13.9% only if you want to take full advantage of the CS glyph, but that you do not have to take full advantage of it if you do not have the haste to support it.

The table below shows the haste requirements for rotation and focus neutral depending on how much of the CS glyph you want to support.

CS Cycle Time	Tight Cycle*	Focus w/o AI glyph^	Focus w/ AI glyph^
10.0	1.26%	2.65%	-1.19%
9.9	2.39%	3.69%	-0.19%
9.8	3.56%	4.75%	0.83%
9.7	4.75%	5.83%	1.87%
9.6	5.97%	6.93%	2.93%
9.5	7.21%	8.05%	4.01%
9.4	8.49%	9.20%	5.12%
9.3	9.80%	10.38%	6.25%
9.2	11.14%	11.58%	7.41%
9.1	12.51%	12.80%	8.59%
9.0	13.91%	14.06%	9.79%

*Haste from gear for a tight cycle.
^Haste from gear to meet having a focus neutral cycle.

Thus, you can see that at a reasonable amount of haste from gear, about 8.5%, you can replace the AI glyph with the CS glyph and gain about 0.5 the benefit of the CS glyph. The advantage with this approach over the low haste approach, it that you take better advantage of the haste you do have naturally and your DPS will increase as your gear scales since you should have plenty of room to add more haste while keeping crit and mastery at decent levels before you get to the about 14% where you have reached the haste cap for the glyph and this rotation.

Another factor is that on every other cycle, your MMM AI proc replaces the 5th SS. Since the MMM AI cast is the GCD while the SS is not, on those cycle you save a little time. This time savings allows you to have tighter CS cycles every other cycle at the lower haste values. The cross over point is at roughly about 8.5% to 9% haste where your MMM AI proc saves about 0.4s and the standard CS cycle without the MMM AI proc is about 0.4s too long so that you have a tight rotation on cycles with the MMM AI proc. Thus, if you go with the CS glyph option, I suggest aiming for about 8.5 to 9% haste from gear. A little more haste from gear is still fine to account for inefficiences. Still more haste from gear up to the cycle cap is still fine, but has some decreased value since it will result in some deadtime in your rotation some cycles, although it will tighten up the cycle on others. If you want to split the difference so that you have about equal deadtime on MMM AI proc cycles as you have CS cast off CD delays on the 5th SSs cycles, then about 11.5% haste from gear is recommended.

Buyer beware though, this optional rotation only applies in the Standard phase when not under dynamic haste effects. It may not be optimal in the following situations:
- During the CA phase since still want to cast more AIs and not want to necessarily cast CS off CD. Still need to do the analysis on whether CS is worthwhile to cast or not in CA phase. I believe that it will now be. If so, then this rotation will still be good over this phase when unhasted since you may be do an extra AI than with the slower 10s rotation.
- When hasted, it may stil be beneficial to add a second AI cast inside the CS cycle. Still need to analyze this.
- When in the KS phase and add it into the cycle. Adding the KS into the cycle is difficult since you will have only 4 SSs already, the minimum to maintain ISS uptime, on cycles with the MMM AI proc. Hence, on those cycle, you want to use the MMM AI proc instead of KS since it does so much more damage and only use KS in place of the 5th SS on cycles without the MMM AI proc. Thus, you can only use KS once every 18s, meaning that you do not want the KS glyph with the CS glyph.

Here is the same table as above but for the situation where a KS is cast every other CS cycle:

CS Cycle Time	Tight Cycle*	Focus w/o AI glyph^	Focus w/ AI glyph^	Focus w/o AI glyph and 1/2 Term^
10.0	-3.10%	11.29%	7.45%	-11.74%
9.9	-1.88%	12.41%	8.53%	-10.85%
9.8	-0.62%	13.56%	9.64%	-9.94%
9.7	0.67%	14.73%	10.77%	-9.01%
9.6	2.00%	15.92%	11.93%	-8.06%
9.5	3.36%	17.14%	13.10%	-7.09%
9.4	4.75%	18.39%	14.31%	-6.11%
9.3	6.19%	19.66%	15.54%	-5.10%
9.2	7.66%	20.96%	16.79%	-4.06%
9.1	9.18%	22.29%	18.08%	-3.01%
9.0	10.74%	23.65%	19.39%	-1.93%

*Haste from gear for a tight cycle.
^Haste from gear to meet having a focus neutral cycle.

As can be seen, the haste requirement for a tight cycle is lower in the KS phase than the standard phase. Hence, the recommended 8.5-9% haste from gear for the Standard phase that resulted in a tight CS cycle on cycles with a MMM AI Proc will result in a tight cycle every CS cycle in the KS phase.

The problem is the haste requirement to be focus neutral is much higher and not supportable, but the good news is that 1 point in Term makes it so that you do not have to have any haste on gear to have enough focus to support this rotation in the KS phase.

I might be mistaken...but if you're doing CS->SSx2->AiS->SSx3 (or any variation there of) in any given rotation...when you enter the Termination phase, your SSs will double in focus regen...so wouldn't we be able to replace 2 of the SSs with the 2 KSs (if KS glyphed) or 1 otherwise also making the rotation tighter in Termination range instead of firing it every 18 seconds?

Standard non-CA/non-dynamic haste in your setup being...CS->SSx2->AiS->SSx3 (AiS focus bleed cast whenever encounter appropriate in the CS CD window)

Termination being...CS->KSx2(or KS->SS w/o KS glyph)->AiS->SSx3? Replacing one of the SSs w/ your proc AiS almost every 3rd cycle. (RNG permitting or lacking on your 3 SSs)

Also factoring in the increased focus from RF coming backup for a 2nd or even 3rd time w/ Readiness eases any focus regen issues sub 20% on most fights. Especially if you're doing normals where the bosses die fairly quick under 20%.

Using this rotation in a PUG Argaloth w/ a few missing buffs (4% physical dmg, flask, dbl pots) w/ a 359 raid gear iLvl I posted just a hair over 23k. Nothing about this parse shouted to me as being RNG friendly, just a statistically average one as fara s I could tell. I didn't use CS during the CA phase even though after reviewing my Recount I think it would be worthwhile to do so. I found my CSs and AiSs doing nearly the same dmg per hit/crit avg overall. Wish I had saved the log few some analysis but slipped my mind.

I had thought of what you suggested but have not mathed it out since it I figured it would be a loss of the following reasons (although I may analyze it - its on the plate):
- SS focus regen does not double. It goes from 9 to 12 with 1/2 Term and to 15 with 2/2 Term.
- Talenting Term costs you other talents, so taking Term better be a big increase during the Term phase
- Remember that every other cycle one of the SSs is already being replaced with the MMM AI proc; hence, there is only 1 extra SS to replace every 2 cycles
- You cannot do a KS every cycle since KS has a 10s CD, but this is a 9 to 9.5s cycle. If you delay the CS cycle to fit in a KS every cycle, then you defeated the benefit of the CS glyph during this phase.
- Even if you could do a KS every cycle, doing so would cause ISS uptime to briefly drop off every other cycle and only be about 90-94% uptime on average depending on your haste
- Also having the KS glyph for back to back KS would cost you the RF glyph, which is really awesome for increasing autoshot, WQ, and AI DPS during RFs
- With the KS glyph, you can't fire 2 KSs per cycle since it is a 9-9.5s CS cycle with the KS cycle being 11-11.5s. So if you fire KS everytime roughly off CD, they will need to float in your rotation. This can cost the SS pair before your AI cast and can result in a slower AI cast. To avoid that, you can only replace the SSs after the AI cast, which means you cannot do the KSs every cycle hurting the benefit of the glyph. Furthermore, the more frequent SS replacements with KSs, results in even lower ISS uptime
- The replacements of SS with KSs, lowers your MMM AI proc frequency, which is a loss of DPS, especially considering that AI damage is about 2.25 times KS damage with 359 gear and factoring in PS.
- When under dynamic haste effects, your SS and AI casts are shorter and you have more room on the rotation for more shots, but you still cannot do a KS every cycle since it has a 10s CD with this being a 9-9.5s cycle
- During RFs, you are generating so much extra focus that 1/2 Term is not needed, much less 2/2 Term

Thus, with all those factors, I feel it is best if you use the CS glyph to:
1) Not also use the KS glyph
2) And since you can't fire KS every cycle anyway and still benefit from the CS glyph, use it every other cycle as I suggested to maintain ISS uptime and maximize AI MMM procs.

There may be options better than what I suggest since I know I haven't actually tested the suggestion (have asked someone with the 4-set to do so) and I definitely have not analyzed all options, but until I see hard analysis showing that one is better than the other, then I will stick with my recommendations. However, recommendations are never one size fits all, that is why I try to provide several options for people to try to see which works best for their situation and playstyle. A player needs to do whatever works for them with this post just trying to provide suggestions for hopefully improving game play.

If you reply was not to my post and your suggestion was not with using the CS glyph, then I agree with much of your suggestions. If you are doing a 10s CS cycle, there is little reason not to do a KS every cycle. I have not yet updated the OP for the KS phase post-4.0.6.

Glyph of Silencing Shot will generate focus only when a spell is interrupted by silencing shot while casting. Nothing happens otherwise

I have trouble seeing why itt would need to be a big increase. With druids giving kings, paladins are free to give might. Might is better than TSA. For my raids that means TSA is entirely optional, and usually wasted. Silencing shot is certainly a valid choice, but it's also far from mandatory. Unless your raid is starved for might, rapid killing is the only talent that competes with Termination for DPS. I suppose it's possible that your raid assigns you to killing all adds and you often get the killing blow to take advantage of rapid killing, but it seems unlikely.

Termination on the other hand is guaranteed to give you better focus regen sub 25% on every fight. Therefore I argue that if the focus regen from Termination does in fact allow for a higher dps rotation sub 25% then it is a mandatory talent outside of raids that require both silencing shot and TSA. Even then, 1 point in termination seems better than 1 point in rapid killing.

How many rapid killing procs would you need per fight to gain more DPS than a single point in termination? That math is a bit beyond my attention span at the moment, but that's what it would all boil down to I suppose.

It might also be worth mentioning that after the recent changes, bombardment is less appealing as well. This is due to the fact that survival DPS is going to be more competitive and perhaps even surpass marks on heavy AoE fights. ie you might be best served by speccing survival for Cho'gall.

I have 1497haste from gear (11.69%) with raid buffs i cast SS=1.274 AiS=2.0525
So i chose variant with CS glyph.

My rotation is CS-SS-SS-AiS-SS-SS-SS(!AiS) = 9.4225 (9.1485)

Whitefyst - what do you thing about that rotation in KS phase with KS glyph?

CS-SS-SS-KS-KS-SS-SS-ArcS(!AiS) dont cast AiS because you can cast 2xKS in that same time, not delayed AutoShot. KS dont cost focus and you have 1sec window and focus for cast extra 1x Arcane Shot.
CS-SS-SS-SS-SS-KS-KS-ArcS(!AiS) There you delay cast KS, when you cast 4x SS and you are waiting for Cd on KS.
CS-SS-SS-AiS-SS-SS-ArcS(!AiS) Standart rotation with AiS, because you dont have CD on KS.

Sorry for my bad english.

I've been curious about the viability of using Arcane Shot as a focus dump post-hotfix since it would help improve mobility somewhat and had the time to run some sims today. Detailed results are available at Simulationcraft Results .

The all Aimed Shot rotation is still simming on top but I've found one interesting viable competitor. Swapping the Aimed Shot glyph for the Arcane Shot glyph and only hardcasting Aimed Shot during Careful Aim or dynamic haste effects is within ~80 dps in the 372 gear set, and actually sims ~120 dps better than always using Aimed Shot in my own ~353 ilvl gear.

The big advantage of this rotations is that more of your dps comes from auto-shot, Wild Quiver, and Arcane Shot, so movement should have less of an impact on your damage output. Since missing out on a single Aimed Shot due to movement is ~100 dps lost over a 5 minute fight (assuming 40k DPE) I suspect that the benefits of hardcasting Aimed Shot less frequently will be greater in practice.

Darn, I forgot to factor in the AS glyph (which would replace the AI glyph since doing less AIs and since two ASs are cheaper than 1 AI) in the AI vs AS comparison, the updated results are:

Assuming that the AI cast lasts as long as 2 AS casts when under no dynamic haste effects (which is almost true at about 10% haste on gear), then for the worse cast scenario of losing two full autoshots (which should never happen):

with the 12% bonus from the glyph, AS dmg is 12823 (rest of shot numbers from the Ai vs AS section in the OP)

AI + 1 * WQ vs. 2 * AS + 2 * auto + 4 * WQ
47211 > 48538

In the best case scenario of losing only 1 autoshot (which can happen exactly with a 2.9 speed weapon):

AI + 1 * WQ vs. 2 * AS + auto + 3 * WQ
47211 >> 38916

How much of the second auto needs to lost for the two options to be equal DPS-wise?

47211 = 25646 + 7798 + 5472 + x * (7798 + 1824)
8295 = 9622x
x = 0.86

I really doubt that on average that we will lose 1.86 auto shots for every AI cast considering that if you have a 2.9 speed ranged weapon or slower that your auto frequency is your AI cast time or slower. On average, you should expect to lose about 1.5 autos. Hence, DPS-wise on a standstill fight, I still think that the AI rotation should be higher.

Focus-wise the AS rotation will cost on average 4 focus less (with accounting for the AI glyph in the AI case). For a 10s CS cycle, that is the equivalent of 0.4 fps or the focus regen from about 10% haste. This can be significant and allow a 3rd AS on some cycles instead of a 5th SS. This factor is what is probably putting AS ahead in the simulation.

Note though that even if we do go with an AS focus dump rotation, you would still want to use AI as the focus dump during:
- the CA phase
- anytime under large dynamic haste effects - especially RF

I will update the OP tomorrow.

Wouldn't it be appropriate to include piercing shots into the equation?

Regarding the Termination value, why is it that everyone seems to be willing to drop Silencing/Trueshot to get it, and no one seems to mention Marked for Death?
For me, MoD is the most obvious candidate for getting rid of. The talent is not only absolutely worthless on most boss fights (as you can simply apply the mark just before the fight) but even in those fights where it is useful, it will at best save you a few seconds of casting. And that's assuming you don't need to move during the switch, which would force you to interrupt your rotation anyway.
On top of that, since we hardly use Arcanes atm (especially against full health targets, which they typically are when you switch to them) and Chimera has a long cooldown, relying on MoD might mean you'll actually lack the mark on the new target for several seconds until it gets applied. Considering all this, I find MoD barely worth 2 talent points. I'd rather take Termination and "worry" how to spend the extra focus (it's really not that hard given how encounters have a nasty habit of breaking up those nice and tidy rotations leaving us focus starved). Situations where the need to mark a new target leads to a serious dps loss are extremely rare. Situations where lack of focus results in dps loss are very common.