After a round of bug fixes with Zeherah, simc produces the following values:

Spec	none	2-T14	4-T14	2pc delta		4piece delta	Totals
BM	105,097	108,051	109,350	+2,954	+2.8%	+1,299	+1.2%	4,523
SV	101,619	102,729	104,692	+1,110	+1.1%	+1,963	+1.9%	3,073
MM	96,385	97,793	100,609	+1,408	+1.5%	+2,816	+2.9%	4,224

So it looks slightly more balanced after those bug fixes. In particular, the SV bonus both has more value after the ICD bugfix for lock and load. However, I agree with Zeherah that the SV bonus is delicate because of it's sole dependency on ES damage and the cap of the LnL ICD.

Lynch Rush is an interesting ability. It currently sims as the highest dps of it's tier, but the simulation is missing one important element.

• Lynx Rush: pet melee dmg * 2 per attack (9 attacks total over 4 sec)
• The attack is yellow damage, and so does not glance
• Basic and auto-attack continue during the 4 seconds
• If you have multiple targets in range, it moves between them
• If you have no pet out, the spell does not fire
• The pet is moving around, so ~50% of that time, the pet is not behind the target. This also affects the basic and auto attacks during that time.

The last item means that at least 1-2 basic attacks, 2-3 auto-attacks, and 4 of the 9 strikes are subject to 7.5% parry and 7.5% chance to block. This will bring lynx rush's dps down some (and make it more comparable to other talents in it's tier). I'll update when this behavior is incorporated into simc.

Hello,

The 5.0.4 is supposed to go live next week if I'm not mistaken.
Do we have an idea what the dps impact is for each spec at lvl 85?
Is there somewhere I can simulate that?

Simc can simulate level 85 just fine, modulo bugs of course. When last I checked, we appear to lose a little bit of dps with the transition to 5.0.4.

Hopefully I not force you to recalc all your results, but the 2pc was nerfed for BM (e.g. Kill Command from 20% down to 15%) and not for MM. Think it is just a typo, but nevertheless...

Based on the changes in the latest beta build (16016), here are the latest formulas (note that I only updated abilities I had reason to believe had changed so it's possible other numbers have been stealth changed).

Spec Comparison:

Here are the current simulated damage numbers from my beta site in a T14H setup similar to what simc is using (and including whichever talent gave me the best numbers, which might not match simc), including the % damage drop from the previous patch mechanics. Note that I'm using an orc for these tests.

• BM- 107483.87 (no change), Talents: dire beast, lynx rush, glaive toss (this patch brought dire beast above fervor again)
• MM- 101229.26 (1.8% buff), Talents: dire beast, murder, glaive toss
• SV- 100213.02 (1.5% buff), Talents: dire beast, murder, glaive toss

Buff/Nerf Summary:

I did some raw calculations on a T14 heroic gear set. In this set I assumed 52k RAP and 15130.5 avg weapon dmg with a weapon speed of 3.0. With these stats here is a rough estimate of how much damage change there is to each modified ability:

• Aimed Shot: 9.2% buff
• Chimera Shot: 5.5% buff
• Explosive Shot: 7.8% buff

Current Formulas:

Hunter:

• Aimed Shot: (avg weapon dmg + RAP * .2) * 2.8 + 19541 (suppresses autoshot)
• Arcane Shot: (avg weapon dmg + RAP * .2) + 2306
• Barrage: (avg weapon dmg + RAP * .2) * 6.4 (total damage to the primary target over 16 ticks, suppresses autoshot)
• Black Arrow: RAP * 1.5 + 1870 (total damage, split over 10 ticks)
• Chimera Shot: (avg weapon dmg + RAP * .2) * 2.1 + 2617
• Cobra Shot: (avg weapon dmg + RAP * .2) * .7
• Explosive Shot: RAP * .234 + 291.5 (per tick, there are 3 ticks)
• Explosive Trap: RAP * .0546 + 246.767 (initial damage), RAP * .0546 + 32.4038 (per tick, there are 10 ticks)
• Glaive Toss: (RAP * .2 + 872) * 4 (to the primary target, 2 will hit)
• Improved Serpent Sting: serpent sting formula * 1.5 (does benefit from the 100% damage bonus)
• Kill Shot: (avg weapon dmg + RAP * .2) * 4.2
• Multi-Shot: (avg weapon dmg + RAP * .2) * .6
• Powershot: (avg weapon dmg + RAP * .2) * 8 (damage to the primary target, does not suppress autoshot)
• Serpent Sting: RAP * .08 + 1620.19 (per tick, ticks every 3 sec)
• Steady Shot: (avg weapon dmg + RAP * .2) * .5 + 997
• Wild Quiver: (avg weapon dmg + RAP * .2) * .80

Pet:

• Bite/Claw/Smack: (RAP * .168 + 142) * 1.5 (standard pet damage multiplier)
• Blink Strike: pet melee dmg * 6
• Lynx Rush: pet melee dmg * 2 per attack (9 attacks total over 4 sec)
• Kill Command: (RAP * .7 + 698) * 1.5 (standard pet damage multiplier)
• Melee: (RAP / 14 * 2 + 311.575) * 1.5 (standard pet damage multiplier)

Summoned Pets:

• Dire Beast: RAP * .2857 + 1246.3 per hit (attacks every 2 sec for 15 sec, attack speed scales with melee haste), now scales with BM mastery (not checked yet this patch)
• A Murder of Crows: RAP * .206 + 560.834 per hit (30 attacks) - does not scale with BM mastery (treated as damage coming from hunter)

P.S. Yes, Nerec, that was just a typo in my last post, since fixed, thanks for noting it.

Hello,

Guys, i would like to hear your thoughs about CD stacking for BM.
Do you think we should wait until Linyx Rush comes off CD and use it alongeside with Bestial Wrath and Rabid? (even though BW comes off CD 30 secs earlier?)
And, if this were to be the case wouldnt Kiroptyric Sigil be GREAT? it has the same CD as LR and it can also be stacked with BW and Rabid.

One more question; i am getting results like: Hit/exp.cap>Mastery>Crit>Haste for BM and SV. What about you guys?
Consequently which is the worst stat that i should start reforging to hit/exp when the patch comes out?

Of course i know about the haste plateaus, but, in general, which seems to be the worst stat?

With the addition of so many more instants (including glaive shot at level 90) for BM, haste has becomea "do not want" stat (also very much thanks to Focus Fire), mastery consistently beats out crit, but both are miles ahead of haste - both simmed and from an analytical standpoint.
The number one thing to notice with BM priority come 5.0.4, is that we may need to pay attention to maintaing SrS, as it is easy with BW and other instants to push to simply not get a CoS off in time. I'm seeing quite good results in the form of a steady "rotation" (its not, but it quickly feels like it) when making sure to add at least 1 CoS between each KC (so 2 CoS, 1-2 instants (depending on speed) if low on focus, and 1 CoS and 3 instants if high on focus (or with a TotH proc, though DB may (or may not) be better at level 85 - it is better at level 90) - it is a dps loss to delay KC more than by ~0.3 seconds, so we won't have 1 CoS+4 instants unless we have a haste CD up (CoS casttime + latency/response delay must be at 1.3 seconds or lower to have 4 instants be a gain)

Here are some observations I have on the hunter spec numbers from FD:

- BM is the spec that is most friendly to latency since much of the higher pet damage than that from the other specs is unaffected by latency. The FD numbers were all done using 150 ms latency. If the latency is reduced to 50 ms, the improvements for SV and MM are about 50% higher than that for BM, helping those specs to close a little of the gap. Although, they are still well behind, but now by only about 5% at level 90.

- SV is by far the most mobile of the hunter specs with doing the least amount of cast time shots. In the FD Sims, I see the following for each spec over a 7.5 min fight:

BM: 107 CoS at 1.708s cast unhasted and 1.314 during FF
MM: 149 SS at 1.298s cast unhasted and 21 AI casts all at under 1.5s or better
SV: 97 CoS at 1.623 unhasted

This means that SVs are more likely to be able to perform closer to their maximum DPS potential on movement fights than the other specs, helping them to close the gap some on BM on those type of fights. MMs are a lot more mobile than they used to be with doing less SSs and AI casts.

Of course, the higher damage from the pet for BMs while the hunter is moving but the pet is attacking as normal, helps BM's DPS while moving.

- With not including Stampede or DB, here is how much damage the sims show the pets doing for each spec:

BM: 49.68%
MM: 14.37%
SV: 14.60%

Hence, the same BM pet is doing about 3.6x as much damage as for MM or SV. This may be one place that Blizzard could buff MM and SV damage to help even things out a bit.

I have done more tests of the trade off of using AI or AS as the focus dump, and it does not look good for AI with the current numbers at level 90.

When comparing using AI and AS as from a purely damage point of view using the shot data from FD, the AI cast time that is the tipping point for when it starts to be better to use AI is given by the following formula:

AI + WQ = (x * AS) + auto + ( (1 + x) * WQ)

where x is the tipping point and with assuming 1 autoshot is suppressed during the AI cast. This formula accounts for the fractional portions of AS that can be performed during the AI cast.

The value for x is 1.614 outside the CA phase and x was higher than the maximum AI cast time during the CA phase.

From this information alone, one could assume that the guideline should be about 1.6s cast time outside the CA phase and always during the CA phase, but it is not that quite simple.

First, we ignored the focus cost. Even at a 2s AI cast time, AI costs 10 more focus than 2 ASs. At the 1.614s cast time, AI costs about 18 more focus than the corresponding ASs. That is the equivalent of about 66% of the focus generated by an SS cast. The damage difference between AS and the SSs need to make up the focus make AI actually still a much worse choice of DPS at that cast time. Using FD with doing a sim only during the Standard phase, I determined that a little below 1.4s AI cast time is the actual tipping point and is what I am currently using as my guideline.

This is low enough where it is not beneficial to cast AI during bloodlust unless you have 25.96% haste from gear or 15.88% haste from gear with the T14 4P. Even during RF, this requires 16.96% haste from gear or 7.61% haste from gear with the T14 4P.

Thus, outside the CA phase it is not beneficial to cast AI during BL at all with reasonable haste ratings in T14H gear. Even during RF, it is only worthwhile to use AI if you have the T14 4P.

It is still beneficial to use AI during the CA phase at all times though.

The moral of the story is that I think that AI needs to be buffed (at least outside the CA phase) so that it is at least beneficial to cast during large dynamic haste effects and so that it is not lower on the priority list than everything but SS and AS when not sufficiently hasted.

On BM, I've found it necessary to mix priority and rotation (and will be including that in my guide), namely to be able to continuously keep up SrS with CoS having a 6 second extension, you will need to cast at least 1 CoS every KC - this hasn't been a problem really (in regards to focuscapping), and means that you can safely go through BW without CoS as long as you do a CoS immediately after. It will be better to focus cap slightly from that CoS than having to spend focus (and a gcd) on reaplying SrS.
I should mention that it is easy enough to do with TotH, I haven't played extensively with DB or Fervor yet.