I think you forgot about Relentless Strikes. Consider: assuming we gain 1 CP/sec, even with Ruthlessness (assuming we have it, which we may not), it takes 4.4 combo points - and hence 4.4 seconds - to build up the CPs for a 5pt finisher. During this time we regain 4.4 * 10 = 44 energy; a 5pt Eviscerate costs (net) 10 energy, with Relentless Strikes. Thus, we gain 34 energy - almost a full Hemo - every Eviscerate. In reality, of course, if we perform a hemo we only need to wait 3.4 seconds, but I suspect it still works out that you can hemo between "most" finishers, yielding a cycle something along the lines of sHrHeHeHe (that is, SnD, Hemo, Rupture, Hemo, Evis, etc.), with only the SnD not getting a Hemo.
Plus, in reality, I think our current HAT CP regen is on the order of 2-2.5 a second, meaning we'll only get one every 1.4 to 1.5 seconds with the cooldown - which means we take over 6 seconds to build up enough energy for a 5pt finisher, which still only costs 10 energy, which gives almost enough energy for a Backstab - more than enough, if you take Slaughter from the Shadows (whether you do or not is yet to be determined). So I can imagine cycles of the sort sBrBeBe or HsHHrHeHHe being very possible as well. The exact pattern of Hemos and Finishers - and number of finishers per cycle - will need to be determined, of course. But I think there are very realistic cycles of weaving in CPGs that can now be performed with HAT, and while I'm sure they're weaker than the current spec, how much weaker will require some investigation.
When simulating a stacked party in a simple stand-and-fight scenario I couldn't really produce a HAT proc rate better than 1.28sec........ so 1.4 to 1.5 seconds in a real raid environment is a reasonable assumption.
Right now now I am getting about 1.4cps from party in raids picking best players from my raid based on TopHat, so cps rates of 2+ seem too high to me.
As for weaving in hemos and backstabs, I tried many combinations, tweaked talents going deaper into subtlety and picking puncturing wounds, going deep into combat. I also tweaked DPS rotation for most promising talent specs. Either way I looked at it, if cps including rogue's own is lowered to 1 per second, the spec is going to do 1.3k less than the 10k projected for it in 3.2BiS. I was hoping for backstab to be viable particularly because we can pick up several talents favoring that CPGM and could finally put the glyph of backstab to some use. However, hemo setups produced more dps. Highest I've seen from backstab was around 8.4k. So it's all very disappointing at least if this is the only change they are going to make to HaT.
What bothers me is that I did not think the nerf was warranted unless they never meant for HAT to be a pve spec and they suddenly realized that it was a viable PVE spec over the last 8 months. So if their intention is to make it non-viable in PVE, I am pretty sure changing cps to 1 from entire group will do it. I think they could have equally effectively taken out the talent altogether since going this deep in subtlety does not look like a dominant pvp build either and is going to be nerfed even more. Perhaps they will buff the rest of the tree for pvp if that's the intention.
If they in fact wanted to nerf the PVE aspect of the spec, then this is an overkill. As it was, HAT was projected to do slightly more than combat, but did not have many burst dps cooldowns and did not have many aoe capabilities, making its use only warranted on certain fights requiring mostly single target dps. Perhaps the simulation was inaccurate or the cps rates in 3.2 BiS are much higher, but hardcapping it at 1 per second is also hardcapping the dps of the spec at a non-PVE-viable level.
So when is 3.2.2 coming out and has there been any confirmation that the intended change is in fact to cap all cp income from HaT at 1 per second?
The damage is behind the other specs in PvE, and due to all the neat utility tools, Subtlety would immediately become the default spec in PvE if the damage were comparable.
Now, never mind the fact that it's a bit out of touch in terms of the relative damage outputs of the specs, the point here is that they clearly think that, due to the other advantages the spec gives (which is mostly in terms of survivability talents and some random utility), DPS output needs to be a bit lower than the other specs to keep them balanced. Now, I suspect this was written more with 5mans in mind where things like Dirty Tricks actually matter, but realistically speaking the survival talents of Subtlety make as much (if not more) difference on any number of fights as Combat's cooldowns do. So while I'm not sure it needs to be behind, I can certainly see the argument that it shouldn't be ahead.
Now, it's certainly possible that this might be too much of a nerf. It's entirely possible that we're going to crunch all the numbers, conclude it sucks, and all just head back to Combat and Mutilate. But I'm not sure we, as a community, are 100% confident in saying that at this point, as theorycraft for low-CP-regen HAT is somewhat in its infancy. And perhaps more relevantly: I'm not sure Blizzard knows with any confidence how much damage HAT is going to do after this nerf. So I suspect what we're seeing is Blizzard figuring out how much of a nerf it is. If after 3.2.2 goes live, there's a max exodus from the spec, and the few people that do keep it do abysmal damage - well, they'll have a better idea of how much behind it is than they do now. And at that point, if they feel it needs a buff to get it to where they want it to be, that's certainly something they'll think seriously about doing.
Note also that the nature of HAT doesn't give you a lot of granularity for adjusting its power; it's not a knob they can easily turn to get our DPS within 1%. So my suspicion would be that if they *do* decide to buff the spec a bit in some future patch (and/or expansion), they will do so by adjusting some of the other high-end subtlety abilities rather than trying to fiddle with HAT to get it exactly right that way.
Excelent points, but I don't agree with that last comment of yours about HAT not being a proper "knob". Sticking to the analogy the problem (as I see it) is the valve simply isn't very well designed for the job at hand.
So far they've been balancing HAT by adjusting the amount of abilities capable of proccing HAT while holding the cooldown constant. Combo point donations vary alot from class to class and this allows for some serious stacking in raid groups, as any one effect will always be superior to all others for HAT donations. Explosive Shot, Arcane Missiles and whatnot comes to mind as super HAT donors of the past. As a consequence they should be tuning HAT by adjusting the cooldown rather than the population size.
As Tinwhisker was hinting in the Incoming Rogue Changes-thread, this could be (and hopefully is) a move towards raid wide HAT donations. Allowing all 5 raid parties to proc HAT ones per X seconds would be an excelent knob not entirely different from HfB as within certain limits it merely adjust the frequency of Eviscerates.
edit: just to get it on record. PTR tests ruled out option c) from http://elitistjerks.com/f78/t38187-f...7/#post1350281 completely. Despite lag and a unusually incompent pug, VOA runs and a group of rogues FOK spamming the dummies simultaniously indicates option b) is correct and HAT is hardcapped to 1cp/sec at max.
I'm curious, what does TopHAT usually measure everyone's max cps as?
If the change is indeed 1 combo point per second from party members and unlimited from yourself, I'm not too sure it will cripple the spec. Usually in raids I'm capping anywhere between 1.3-1.7 cps, with my individual cps being around .45 I would land right in the middle of that.
For every fight I reset TopHAT and it put me just around 1.5 cps, with Ignis being around 1.7. The numbers aren't in the ridiculous realm, but they certainly are higher than what I'm typically used to. I first designed my spec around Firefighter as thats what we were learning at the time so I picked up shadow step and cheat death for utility. If the change is indeed just a minor shaving to those numbers, the utility alone would make it be worth keeping I think. If "your party" includes your own CPS, then I suppose the difference would be fairly substantial.
Thing is, that gets sort of silly. I mean, the damage difference between 1 CP/sec and 1 CP/1.5 sec is considerable; they can't just set the regen rate to some nice number and expect it to work out. And somehow I think that setting a cooldown to 1.22 seconds is not something that Blizzard would choose to do. Never mind the fact that that number would need to be adjusted for each new tier of content.
The second point, of course, is that HAT is already too much of a "1 point wonder" talent - its singlehandedly propping up the tree right now, and spreading the value around would be as beneficial for Subtlety as spreading the value of HFB would be for Mutilate.
I have to agree with Bural that the approach Blizz has been taking in balancing the spec is just not right, so failing at it does not mean that HaT is in principle a spec that can never be balanced. Look at it from a different perspective, while all other rogue specs scale only with the rogue's own gear, HaT scales not only with the rogue's own gear (and not too bad with agility and AP multipliers it has), but with the gear of 4 other people in his party. Without implementing a smart group or raid cooldown, it's hard to balance HaT in the beginning of a patch and still have it balanced relative to other rogue specs at the end of the patch due to scalability problems.
Taking into account dedmon's comment, since party members are capable of giving together more than 1 combo point per second, having one of them die will not impact HaT dps, so it's a nice step. However, I think if the cooldown is on the group level, it just has to be higher, something like 1.5 sec from entire group or 1 sec from the other 4 people in the group.
Also with regards to utility talents in subtlety tree, for PVE purposes, we value utility talents very low. So extra 200 agility and the resulting gain in dodge of mutilate spec is not worth a 50 dps loss to me. Same with extra stamina. The problem arises when subtlety talents significantly change the difficulty of the encounter, such as using Cheat Death on certain fights. I can see how it could be a problem, however instead of nerfing the damage to balance for that, I would change the encounter to fix it. For instance, COS does not work on overload, guardian spirit does not save a tank on Steelbreaker. Why not make it so that Cheat Death does not work on certain encounters. Or for instance, why not move Cheat Death lower down the tree so that it becomes very clear that by choosing those talents, you are seriously hurting your dps. I bet you vast majority of HaT users would forgo Cheat Death and just enjoy the single target game play of the spec. And those who decide to go deep Subtlety for pvp purposes will still have Cheat Death available to them.
I always thought the design of HaT talent took greater care of Arena formats problems with rogues than PvE balance : rogues were in BC too strong in 2s, not working in 5s. So I understand Blizzard designing HaT to give rogues a tool clearly better in 5s than 2s.
As a fan of the soloing talents in the subtlety tree for my alts (eg Setup / Ghostly Strike synergy, Evasion as a damage cooldown for soloing elites), I also like the gameplay of combo points coming outside of purely CPG attack. Yet I dislike the reports I have seen of "pure HaT play" being restricted to spamming Evis. It doesn't feel like how it feels when soloing, where you get free CP, but don't spam Evis.
Obviously, HaT in PvE deserved a nerf, seeing some parses, and I like that the talent is meant to be more in line with the potency of Setup when soloing. (Not so) low HaT is fun ! I hope a compensation will be given if needed for PvE viability, my preference being Slaughter from the Shadows recieving a boost, as I'd like Backstab to become viable again, and I prefer the 45 energy gameplay to the 60 energy. Low HaT still needs theorycrafting to know which one of 7/21/43, 11/9/51 or 0/20/51 is the best DPS, I hope SftS+SD wins in long term.
If you are referring to 0-Yogg parses, it was nerfed in 3.2 in that HaT rogues no longer get mass combo points while others are AOEing, which is a valid nerf.
As for low HaT, it's so not interesting because of how low damage it does, that I did not finish testing, but highest I've seen was 8/20/43 up to LR in combat and with BS in assassination. Can't drop ruthlessness. Going daggers is a drop in dps, even when picking puncturing wounds and a couple points in aggression. Going deeper into subtlety is a dps loss. Mace spec is a dps loss. Going into lethality is a dps loss. 23/5/43 is lower in dps. I think the only thing I did not try was maxing malice instead of picking LR, but I am sure it will be a dps decrease.
The problem arises when subtlety talents significantly change the difficulty of the encounter, such as using Cheat Death on certain fights. I can see how it could be a problem, however instead of nerfing the damage to balance for that, I would change the encounter to fix it. For instance, COS does not work on overload, guardian spirit does not save a tank on Steelbreaker. Why not make it so that Cheat Death does not work on certain encounters.
Although there are several cases where Blizzard makes something "not work" in a particular encounter, is this the kind of thing we want to encourage? Why should our tools simply not work because they make a certain element of the encounter easier? Isn't it better to address the root cause of the balancing problem with HAT?
Why not throttle the CP impact of HAT to a low level and put the choke point on the player, not his group? Then give more emphasis on using a new-and-improved Hemo as the primary CP generator, you'll be back in the "scales with your weapons" neighborhood. HAT then becomes kind of an equalizer by compensating for a weaker CP generating attack by providing free CP's elsewhere and enabling more finishers. It would still give the flavorful, unpredictable style of play that a lot of us crave. It also removes the need for group stacking
The reason HaT will be unplayable after the CP nerf is because Subtlety has no form of energy returns outside of Relentless Strikes. In its current state, HaT is competitive only because it is able to spam high cp finishers to proc a ridiculous amount of RS. If you look closely at parses, HaT absolutely destroys Combat and Mutilate in energy returns. Since the "fuel" to constantly proc RS is being reduced by 75%, it will undoubtedly kill the spec because energy returns translate into too much damage, and HaT is very efficient at dumping it.
The other problem with the direction they are taking HaT is that it lacks an efficient CP builder. Hemo and Backstab do absolutely awful damage for their energy cost, and they simply do not generate enough CPs to proc enough RS, even with the HaT talent. I suppose they could add Hemo and BS to Initiative to make them more viable as CP builders, but I really think they need a huge increase in damage. If Backstab was hitting anywhere close to Mutilate, it could make specing Slaughter from the Shadows and Puncturing Wounds actually worth it. Hemo is actually in a position where it could give HaT Rogues some very desirable raid utility. If they made the Hemo debuff a significant increase to a raid's physical output, it could possibly make up for the garbage damage it does on its own.
I just find it very discouraging that they are breaking the spec without any mention of ever fixing the tree. From the Rogue Q and A, the developers feel that Subtlety offers too much utility, and that if it puts out competitive numbers every Rogue will just spec HaT. We know this wouldn't be the case because even though it has an Execute and optional utility like Cheat Death and Shadowstep, Subtlety lacks damage cooldowns and many players simply dislike the playstyle. I really don't think it's asking too much to just make the tree viable, but seeing how long it took them to make changes to Arcane I doubt Subtlety is going to get some love anytime soon.
As requested, here's some quick numbers for the new finisher cycles to compare energy gains using this Algalon parse (WoW Meter Online - Combatlog Replay
During a combat time of 345 seconds, Relentless Strikes proced 152 times for 3,800 energy in HaT's current state. Now, let's say you used one of the cycles proposed such as HsHHrHeHHe with the best cycle time of 3.4 seconds for every 5 point finisher. In 3.22, under those conditions in the same fight, you are going to proc RS ~ 101 times, which will return 66% the amount of energy HaT would be getting currently. If we want to use more realistic numbers, such as the 4.4 second cycle, you are only going to be getting around 52% of the energy back from RS. As for the six second Backstab cycle? It's only 38% of current HaT, roughly 1444 energy.
So what exactly are the implications of this? Well, not only will the 3.22 HaT be generating less energy than a Combat Rogue (and half of its former self), but the nerf to CP generation also taps into HaT's supply of high CP finishers, which account for almost half of a 3.2 player's damage. When you trade those finishers for a CP builder such as Hemorrhage, you are now directly cutting into both your damage and energy pool. I'm not going to go in depth about Hemorrhage and finisher formulas since they are readily available, but even with talents and glyphs aside Hemo is barely going to do a 1/3 the damage of a high CP finisher while costing 2/3 more energy. Incorporating such a DPS loss into a cycle that now generates significantly less energy is such a huge plummet in damage that you don't need a Combat Simulator to visualize how bad things really are.
Now, never mind the fact that it's a bit out of touch in terms of the relative damage outputs of the specs, the point here is that they clearly think that, due to the other advantages the spec gives (which is mostly in terms of survivability talents and some random utility), DPS output needs to be a bit lower than the other specs to keep them balanced. Now, I suspect this was written more with 5mans in mind where things like Dirty Tricks actually matter, but realistically speaking the survival talents of Subtlety make as much (if not more) difference on any number of fights as Combat's cooldowns do. So while I'm not sure it needs to be behind, I can certainly see the argument that it shouldn't be ahead.
This is what I cannot figure out at all: The only real utility you're going to grab in PvE is Elusiveness, Prep, maybe Shadowstep and 2 misc points. Of those only elusiveness is really going to have any real effect on raiding. In 5 mans I can see it being of even less value, sure my sap now costs 30ish energy, but by the time I'm back on the burn down target, I've probably already regenerated it anyways. Not that I've been asked to sap a target in 8 months....
Prep really just seems overhyped for everything but PvP. I shouldn't be using evasion or vanish in any case (or at the very least I shouldn't need 2 in less than 5 minutes), sprint isn't going to get me to safety in any instance and you're not going to have cold blooded as a deep sub rogue. In PvP you're already going 21 deep in sub, so the extra 'utility/defense' argument weak. Given that HaT doesn't play well with resilience so I don't think PvP balance has even been a concern for it.
Unless I'm missing something (and please enlighten me if I am) I don't think the 'utility' is remotely make a difference in fights, and definitely not worth modifying HaT downwards with no apparent long term plan.
Cheat Death is often over-and-under rated. It gives you the ability to eat an instagib ability once every minute (though your DPS will likely suffer if it's AoE and the melee in your group are running away), and it means you don't die if your healers have a hard time keeping you alive or you don't move out of something nasty fast enough. The second case shouldn't happen, but it wouldn't be progress content if everybody stayed alive through every attempt. It's not the prettiest way of surviving ever, but with Cloak of Shadows and Vanish it adds up to a lot of survivability.
Preparation is almost decent, having two snare/root breaks and sprints can really help on encounters that involve being rooted often (Hodir) and a lot of movement that forces you off-target.
I'll agree that neither of these seems game-breaking at all, but Blizzard's problem with HaT seems to be their inability to fine-tune it to be either a little ahead, equal to, or a little behind Combat and Mutilate without a clunky change to HaT's cooldown. Changing which abilities can proc CPs didn't work and probably consumed a lot of their time and effort and making HaT a 1-second cooldown per player didn't work. Now they seem to want to remove the massive RNG factor so they can build and balance the spec around the CP generator they choose for it, if they choose to rebuild it for PvE at all.
Well was playing HaT since Wotlk and today after seeing the "imba" dmg i was doing in the new Onyxia raid, had convinced me to switch to assassination, was using daggers so combat is out of question. Well dont know where blizz wants to put sub spec but the small community of HaT players got even smaller, on my server there are 2 rogues who actually raided with this spec, well had really good time being a subtlety rogue and as for patch 3.2.2 this music i find appropriate for this sad day, YouTube - Titanic Soundtrack - Nearer my God to Thee. Cya and good luck my sub spec maybe we see us in some better times.
I also gave it a try today since I didn't test it on the PTR and the nerf is... devastating :|
Hoped that the impact of the change wouldn't be that hard, since I had an average of 1.2-1.3 cps prior patch. This new mechanic just negates the bursts of CPs that make this spec so interesting to play, but keeps the moments where you don't get any CP for 2-3sec.
Bottom line is, that the actual cps are at about 0.6-0.7.
Blizzard should have really done it better... if they wanted a hard cap at 1cps so that group stacking won't be imbalanced they should have done it in another way. For example 5 seperate cooldowns, each for 1 CP at the duration of 5 sec. That would result in an average and max 1cps but wouldn't cancel out the fun gameplay and bursty behaviour of extern crits...
Way back when, when HaT was bugged and absurdly good I had thought about how they could fix it. Go figure they go well beyond that and castrate it. I figured they could just do like all the other trees and move away from the adding cps idea and make it into a regen talent like FA or CP. So that you get X energy on party member crit, and then have the spec put points further down into the tree and having Subt be all about Backstab.
I really loved raiding as this spec but it doesn't seem like that will be possible anymore.
Mavanas, is it reasonable to assume if I plug a 13/7/51 spec into your simulator with the goal of using Backstab to build CPs and high Rupture uptime that the simulator would give me accurate feedback?
I am certainly not trying to imply anything negative about the simulator, my question is, was the simulator was designed with this type of build as a possibility?
There was some discussion earlier about a backstab oriented HaT build post-3.2.2. My thoughts were to get Slaughter from the Shadows with the Rupture oriented talents and see if I could make it work, but assuming .8 CPs/sec from HAT on the Group Setup tab, I can't get the spec within 20% of combat in similar gear.
I removed HAT support from my 3.2.2 version of the simsheet because it was at least 10-15% below in damage. Besides, I never modeled Shadow Dance before. So no you will not get an accurate answer.
08/17/09, 1:06 AM
Similar Threads
