While I suspect that all remaining 0's should be 1's in the abilities file, I personally tested the following:
["Death and Decay"] = 1,


Also do not forget that some moves gain ability to crit with tier 9 set bonuses:
["Blood Plague"] = 1,
["Frost Fever"] = 1,
["Serpent Sting"] = 1,

Is there going to be a Best in Group comp list like there has been in the past for combo point generators. I tried using 2 feral druids, ret pally, and enhancement shaman, and i seemed to be combo point starved and as well i was energy capping a lot. If I'm energy capping a lot, would it be best to hemo, to fill in and help with combo point gen as well?

I am not sure what the best composition for HaT is, as I haven't really tested it at all, but I'm pretty sure it's the general consensus to NEVER let your energy cap. And if you are feeling CP starved then yes by all means generate them yourself, whether it be by Hemo, Mutilate, or SS. By not doing so you are basically doing yourself and the rest of your group a disservice, unless you can offer some other utility to the group during the lulls (Interrupts, CC, etc...)

SimulationCraft r3080 should be up-to-date with the latest HAT findings. (Or at least my understand of those findings!)

It also can generate HAT "donor" reports:

Running the Raid_T8 config with no changes results in the HAT Rogues being in a "virtual" party and simply using an average interval between HAT procs of party members. When a Rogue is actually placed in a party, this avg interval is ignored and the actual crit occurrences of the party members are used.

EDIT: Deleted donor reports as I'm moving to a different format: Simple critical strike rate.

Awesome, thanks for your work! I'll go over the damage breakdowns once the new release is out to check for inconsistencies with the HAT report.

Now, I believe all we need to cover is ret paladins, scrubknights and of course healers.
As far as ret paladins go, TopHAT seems to be fairly reliable. I've gone through a few parses from Algalon and 0.33cps counting only damaging abilities seems to be a fairly representative and reliable number. Grao was kind enough to share a picture from his parse on WMO and the TopHAT report is resonably close. As for ret paladin heals I remain unsure which if any are capable of donating HAT procs.

When it comes to scrubknights we need to make sure Sudden Doom Death Coils can infact proc HAT or if it's a mechanic similar to Enhancement Static Shock. Same goes for Bloodworms and the Razorice Runeforge.

When it comes to healers in general there's little help to get from Simulationcraft, but judging by some anecdotal reports it seems it's definitely worth looking into paladins at the very least. Judgment of Light, Flash of Light Glyph and the redesigned beacon could be awesome HAT donors. Last but not least we need to look at HOTs as with the amount of rejuvenations resto druids are throwing around could -with the t9 bonus- prove to be atleast as awesome. However, we need to make sure if there isnt any condition such as overhealing where any of these secondary healing effect don't proc HAT.

The lack of a 258 ilvl fist from the ToC loot table is going to hurt the traditional HaT spec that drops points into CQC. It appears that the fist from the tribute chest is too far behind its contemporaries in raw damage to really be viable in the 25-man heroic content, which begs the question as to what will be BiS for HaT's mainhand.

If one were to simply go Axe/Dagger, the loss of the crit from the mainhand needs to be factored into the output from 10/18/43, and whether or not that crit will affect your Eviscerates. Also, the fact that 23/5/43 is able to function with any weapon type may give it an advantage with ToC itemization despite its other shortcomings.

I suppose it could be possible to go Dagger/Dagger as well, but we then need to calculate the loss of poison damage due to PPM. I'm not sure exactly how to model it, but from what I understand a 1.8 dagger has roughly a 30% less chance to proc Wound than a 2.6 weapon. This would not only hurt overall poison damage but also CP generation from Wound crits, but it is possible that the crit and raw DPS of the weapon will outweigh this loss.

It makes for some pretty complicated comparisons, and I'll have to get some of my own HaT parses before I can work out some of the math.

How much worse is Shadowstep than one point in Ruthlessness/Blood Spatter/Lightning Reflexes? I guess the better question is would the effect of Shadowstep make it a better talent on high movement fights, such as Yogg p2?

In the Pocket Guide section, Aldriana mentioned that once having t9 BiS gear, it would be optimal to basically drop rupture from the rotation for combat rogues, which outputs the most dps. I was wondering if that would be the same for HaT rogues, to drop rupture and to spam eviscerate while keeping SnD up.

Was testing out TopHAT some more last night, seemed to be fairly accurate, I would always get around 80% accuracy. In general, I would always get anywhere from 1.0 cps to 1.2 cps on most fights, which allowed me to top the meters in some cases. Some of the accuracy is going to be inherently impossible to get rid of since it's fudging numbers of estimating how much ruthlessness contributes to combo point gain. When I find time I'll update the abilities again some more.

Possibly, do note that Aldriana's numbers are based on Macespec, Which leaves HaT with very little "fillers" outside Sub.
3/25/43 Possibly.

Mathematically Shadowstep is around a 1% damage increase if you were to use it on cooldown with a high CP Eviscerate or Rupture. However, blowing it on every CD isn't very ideal on anything other than completely stationary fights. I would argue the utility of Shadowstep rather than its damage increase as justification for grabbing the talent. Obviously it has its usefulness at maximizing time on target after shock blasts, fears, and overloads, but there are many other applications to allow one to be creative (taking out trees effectively, fast tricks to tanks, stepping through cosmic crashes, etc). It's also desirable for Engineers that want to utilize Nitro Boosts to maximize stats with mobility since they will be lacking a % movement increase.

As for which talent to substitute for it, of the three listed I would suggest not dropping Ruthlessness. Part of the new appeal to HaT is a Rogue's ability to help generate his own combo points thanks to poison crits. With Ruthlessness maxed there is a solid chance on an Eviscerate crit to return 3 combo points from procs. Your highest damage output as HaT will be when GCD locked stringing Eviscerates, and since CPs as HaT only lead to more CPs, Ruthlessness is a no-brainer.

As for the other two, I'm not sure the math for Lightning Reflexes but Blood Spatter is pretty easy to do. If your Rupture is accounting for 10% of your damage, 1 point of Blood Spatter is a 1.5% damage increase. What Lightning Reflexes has going for it is a small contribution to CP generation from poison procs on top of the increase in damage to melee swings and poison. Personally, I would prefer the latter since Rupture uptime is sometimes lacking on an encounter with a lot of adds such as Yogg.

HaT will almost certainly phase out Rupture from its 'rotation' as soon as the 8.5 4-piece is broken and it no longer gains CPs from its crits. It's also due in part to the nature of Eviscerate crits procing HaT to give more CPs, which lead to more Eviscerates, and so on. However, the loss of 8.5 4-piece will also hurt the great CP generation we've seen from being grouped with other Rogues.

I'm combat but sometimes I go HaT for specific bosses (and for increased survivability on trash). I agree fully that the dps loss from picking CD over MoS is negible.

If you do go HaT, consider Cheat Death your HardHaT garuanteeing your safety

I swicthed from MoS to CD since we were on Patchwork trys long ago to prevent from being OS. I never switched back.
Dying in the middle of a fight is obviously a massive dps loss. Cheat Death saved my life so often that i wouldn't let it go especialy for a mere 10% dps for 6 sec.

Multiple times i saved the pull tanking on evasion (not for long :p) because MT has died. This could hardly be done without CD, because when the boss switch agro on you he would one shot you most of the time before you could trigger evasion.

Rule number 1 is number 1 because it's more important than any other rule. That why i think Cheat Death is a must have.

Does your most extreme perfect condition not include Tricks of the Trade, Bloodlust, Speed Pot, Trinket procs or Hysteria?

If I end up deciding to drop MOS I will most likely put the points in Enveloping Shadows. Consistant damage reduction seems better than 3 points in a talent that will never ever proc. If you are going to assume that player is bad then you are defeating the purpose behind min maxing. The player is bad enough that he can't hit evasion before dying but he is good enough to execute optimal dps rotations? riiiiiiiight.

It's not quite that simple. A mistake by the rogue isn't the only reason CD might be needed... like a healer falling asleep at the wheel, or someone else in the raid nailing you with a bomb or chain damage effect.

Clearly, the decision to grab CD over MoS or Enveloping Shadows is one that will depend on your individual raiding situation. If your healers are bringing their a-game you won't need ES either.

Personally I always use CD on those rare occasions I use HaT spec.

If you are to maximize the damage output from MoS, you need to be Vanishing on cd and blowing all your cds at the start of the fight. As I've mentioned, missing a single Vanish or Prep due to the RNG of encounter length can drop the damage bonus from MoS as low as a quarter of a percent. Since Tricks of the Trade is on a short cooldown, it's possible for you to wait for it to pop Vanish, but even if I was lenient and let your Tricks magically fall on top of your Vanish cd (while still under those perfect conditions), the increase in damage from timing them together (if you were sustaining 7k DPS) would be .057% - a fraction of a tenth of a percent.

Be reminded though that even achieving that number is next to impossible. You would have to assume your Rogue partner is tricking you in such a manner that they fall exactly on top of your Vanish cd and that again - the boss dies immediately after a MoS. Add into the situation latency and the fact both have the same duration and it's almost impossible to synchronize the two perfectly together. A second missed is a sizable chunk out of that "perfect" .057%.

As for the other buffs, if you were even doing 75% (!) more damage during a Bloodlust with all of those procs/cooldowns/trinkets, the total damage bonus from timing it together with MoS would only be .066% if you were sustaining 7k DPS the rest of the fight. The reason I'm using these extreme, impossible numbers that favor MoS is to prove my point - those two talent points do a completely neglible amount of damage in PVE. It might sound alright on paper, particularly because Subtlety is lacking a lot of damage talents low in the tree, but a Rogue Vanishing on CD with 3/3 MoS is not going to be doing any significant amount of damage over one with 1/3 MoS who Vanishes once or twice a fight when appropriate.

The great thing about HaT is that there is no true cookie cutter spec. Some of the talents are simply fillers while others come out very close in damage. CP generation from your group as well as efficient use of CPs will always affect your output far more than the different flavors HaT comes in.

From a min-max perspective though, it's really up to the player, and if you find yourself never dieing to anything, then yes you can squeeze out an extremely small amount of damage from MoS if you are adamant about abusing the Vanish cd. But when it comes down to the argument of MoS vs. CD, I don't think it's a matter of being pro or sucking. I have high survivability in raids thanks in part the nature of the Rogue class, yet I would still take 3/3 CD over 3/3 MoS for when RNG happens - like getting constricted just before entering portals or having a lasher detonate as a ground tremor hits. CD also gives a distinct advantage to HaT on progression, in that while it's more likely to lose a group member while learning an encounter, you yourself have a much higher chance of survival.

I can't argue that Cheat Death is more damage, because it's not. What I can argue is that MoS is such a small damage difference that the utility CD provides far outweighs it. If you find that the utility of Enveloping Shadows would better benefit your raiding than having a permanent Guardian Spirit, then you should by all means take it. But as for maxing MoS, it is going to require a risky and unrealistic amount of both RNG and precision to have any sort of significant change to your overall damage.

hello, all im reading this topic for over 3 months when i started playing hat, now in t9 content we got a chance to get close to cap with armor pen and reaching 1232 rating or even more, i provided a RAWR file, i know its not as good as spreadsheet for dps mesurment but its not used here for that, it was just easier to google around the gear and equipment. the items arent BIS and not all are ilvl258, i was just beeing curious what one has to get in his hands to reach the 1232 arpen rating.
unlike Mavanas i skipped the 4/5 t9 and went only for the 2xt9 since the bonus from 4/5 isnt worth that much imo. My poin is: maybe its more worth to skipp the 4xt9 and go with 2xt9 and full armor pen rating.

RapidShare: 1-CLICK Web hosting - Easy Filehosting <- Rawr file

3.2.2 patch notes were just released and it looks like HaT is being hard-capped at 1cp/sec.

Some amount of testing would be required again atleast. The HAT tooltips have been wrong or misleading in the past.

edit: asked for clarification on the ptr forum and used the following wording:
a) The 1 second cooldown that was always described in the tooltip, but never effective before 3.2, is now in effect. This note is simply to stress there's now a 1 second cooldown on each player in the party and it prevents them from donating the rogue more than 1 combo point per second. The rogue can gain 1cp/sec per member and 5cp/sec at max per party and this change simply documents the mechanic on live servers.

b) The 1 second cooldown is now party wide and the rogue can at max gain 1cp/sec per party.

c) The 1 second cooldown works on party members, preventing the rogue from gaining more than 1cp/sec from his 4 party members, but there's no cooldown on combo point generation through HAT for the rogue himself.

If anyone can test this it'd be awesome. I currently cant get on the ptr as there's a problem with some EU battle.net accounts linking to old ptr clients, mine inclusive.

Also, I think it's worth keeping in mind that this change may not be the direct end of the spec. It's certainly a nerf, but given that all our theorycraft to date has centered on optimizing straight Evis spamming; now we're going to have to investigate HAT builds that weave in conventional combo point generators - Hemo and Backstab being the two obvious cantidates - which will certainly be weaker, but it's unclear by how much. We'll have to take some time to analyze these options and see how much DPS we can squeeze out of the spec - my guess is that it will finish behind Combat and Mutilate, but depending on the margin it may remain a circumstantially useful spec due to its other advantageous properties.

SimulationCraft r3114 available for download
HAT-related changes include fix for DoT-crit and sharing the player-specific CD for both attacks and spells
Also includes patch 3.2.2 support which moves 1sec CD from the party member to the Rogue. Accessed via patch=3.2.2 on the cmd line.

EDIT: I'm rebuilding the T7, T8, PTR wiki pages now and will update them tonight.

I was curious about this, so I went and ran some nearly-napkin math using a very, VERY basic Excel spreadsheet. Essentially, I was running numbers to see just how bad HAT energy regen was, and at what point it would even have the energy to use Hemo/Backstab.

Due to the 3.2.2 changes to HAT, I made the assumption (best case) that we'd get 1 CP/sec. What I found was that, Tricks and everything BUT Evis and an occasional Hemo aside, you barely have enough energy generation for 1 Hemo every 3 "cycles," but doing so condenses your cycles from 5 seconds each down to about 4 (or 12.5ish for 3). And this was assuming we began at full energy, and thus could pop 2 Hemos in the first cycle just to get started.

Gah, and now looking at that, I completely forgot about Premed, which at the start of the fight would speed up the first cycle even more. Too many variables. But needless to say, HAT energy generation is really not in a good place if Backstab is supposed to be a viable CP builder, unless of course that Backstab is worth delaying the inevitable Evis (which would otherwise roll around every 5 seconds).

Tossing out Tricks every cooldown means you're losing another 15 energy every 30 seconds, which actually means you're using Hemo LESS, since you don't have the spare energy to waste on not-Eviscerate.

I'm just not even sure if Blizz knows how they want Sub to gain CPs now. HAT is seeming too difficult for them to balance.