Killars

Ah I love how Hanos posts, because I tend to post nearly the same thing and believe he is right on the money.

PvE and PvP do not coinside, well not optimized anyway. I'm not going to go in depth because it is really that simple, and people seriously need to start getting that.

Best quote in pages and pages of discussion...

Anyway this is barely a Hemo thread anymore, so I am done here. The spec is completely mediocre and meant to fool around with or BG with. Combat once again destroys anything else.

/farewell to people I consistantly spoke to.

 

• Vulajin

Sorry for being unclear; build number 8 used for my test was actually 11/21/29, not 11/20/30. You see the results above.

If that 1608 figure incorporates the damage contributed by the Hemorrhage debuff, you can compare it to my list and you know exactly why it's not considered viable. Using the exact same gear and weapons you use for that 0/40/21 spec to get 1608 DPS, you could instead switch to 11/28/22 and do superior DPS while contributing the exact same debuff damage.

And to reinforce Hanos' point, rogue DPS is not just about having a number that is big. It's also not just about being number one on the meter, because almost any spec can accomplish those two things. Even once you're at the top spot, though, extra DPS still contributes to the raid. There's no reason to intentionally gimp yourself.

 

◊ Kaubel

Good. Because judging by the number of reports we've gotten, I'm pretty sure everyone hates you.

 

KasumiRevy

I'm mixed on the new changes I'm hearing about. On one hand, I was a veteran of sub spec for so long it's nice to hear some motivation to go back to a deeper point value. On the other, I really was enjoying my pve hemo tri spec, and the change is a direct nerf to my current playstyle.

You're not going to find it doing spreadsheets. Shadowstep mobility, cheat death, are dps buffs, but require skill and understanding of when/how to use it.

What I seem to be going over in my head is:

Losing Relentless Strikes <- problem

The choice between shadowstep and bladeflurry for 21/40 or 20/41 pve damage.

 

Ashere

With the Combat tree and the Assassination tree basically staying the same, I dare safely say that combat won't be the main support tree for Shadowstep in PvE. If you miss Bladeflurry (I already miss it too in my playstyle), then you might consider dropping Shadowstep for it, but for most people it goes at the cost of the major reachable DPS talents, which are mainly in Assassination.
I ran the numbers through the spreadsheet a number of times, to make different combinations, but I kept ending up with 11/9/41 every time. Of course, to maximize DPS, dropping Shadowstep and taking another point in Precision would be better. Now we just have to wait and see how the new Sinister Calling works out in the equation, and how it would compare to the old tri-specs in various gear.
My bet is that on lower grade equipment, the old hybrid wins, but that on end-game level shadowstep will beat it because the spec scales way better. The question then will be: will hemo still be strong enough to compete with combat swords?

 

• Vulajin

11/9/41 is definitely the best Shadowstep build of the bunch, but it's still terrible, at least 10% behind combat. Reference post 1461 on the previous page.

 

songster

I'd question the above quoted assertion. If the rogue in question doesn't have swords, clearly he can't use combat swords either! The fair comparison would be between 11/20/30 (or 11/21/29) trispec maces and 20/41 combat maces. Or the equivalent claw specs, of course. You don't include combat mace or combat fist builds, so we can't draw any conclusions.

Of course, the real conclusion is "For maximum PvE DPS, you need to go swords", but that begs a bunch of questions as to whether you happen to have gotten lucky with drops. For example, we've had one Talon of Azshara drop in total. The best weapon I have access to therefore is Dragonstrike + Arena season 2 mace. For me, the spreadsheet does indeed say that trispec is worth it in comparison to combat maces, but not by much last time I looked.

 

• Vulajin

You've got me here. That, of course, is what I get for posting yesterday on ~2 hours of sleep. However, I'm pretty sure the point about not wanting more than one Hemo rogue is accurate.

 

Left

Ah. Sorry, I didn't see those for some reason...

So it looks like if you are running with something other than swords and want to use a Hemo spec, 11/21/29 is the route to go. It's good to see that it can still come within a few % of the top end, even without Sword Spec.

The reason I like 11/21/29 is that (a) it won't require a respec if/when I get a weapon upgrade that isn't a sword, and (b) it allows me to gem for agility > hit, which is good for if I want to temporarily respec to fool around with a Mutilate spec or 11/9/41. I definitely miss Weapon Expertise, though. I notice the dodges a lot.

 

◊ Hanos

I would venture that the S2 Swords MH + OH, or even better, the S3 Swords MH + OH would beat Dragonstrike and the S2 OH Mace. I had the S1 Swords until S2, then I had the S2 swords until Hyjal. Once you get to T6 melee weapons tend to be plentiful, until then you just have to plan to use Arena weapons.

 

harmisajedi

i've lost count of the posts stating that trispec is dead come pve in 2.32. i don't understand how that claim can be made unless trispec is already dead in 2.3. here's why:

going from 125% weapon damage to 110% weapon damage is a 12% nerf.

assuming 2k ap & talon mh, the hemo nerf will be:

IN 2.3: [(2000/14)*2.4] + 260(1.25) ==> rounding, 675
IN 2.32: [(2000/14)*2.4] + 260(1.10) ==> rounding, 629

net negative, not including crits: 46.
if we assume 30% crit, the net loss will be roughly 60 (rounding up.)

ignoring armor mitigation, which would further diminish the amount of the nerf, 60 out 675 represents a true nerf of 8% to hemo damage.

as recapped through a week's worth of bt/hyjal runs, a trispec rogue's damage output averages around 25% in hemo damage. therefore, the reduction in weapon damage from 2.3 to 2.32 appears to be, with the stated assumptions and some slight rounding, a 2% baseline personal dps nerf.

under the most fortuitous circumstances, optimal dps w/ current gear peaks around 1500dps. a 2% baseline personal dps nerf represents a 30dps net decrease.

however, 2.32 will also bring about +36 to +42 damage change, as well as (way more importantly) the bugfix which will remedy magic damage eating hemo charges. if we assume 1 hemo strike per 4 seconds (which is extremely conservative) & about half of hemo charges being currently consumed by magic damage (which, again, is extremely conservative considering dots + channeled spells) then we have:

IN 2.3: [(36*10)]/4 multiplied by 0.5 (ineffective charges) ==> around 45 raid-wide dps increase via hemo debuff
IN 2.32: [(42*10)]/4 ==> around 105 raid-wide dps increase via hemo debuff

to recap: going from 2.3 to 2.32, we are looking at a personal dps nerf that at its possible worst is around 30dps, while we see a raid-wide dps increase of about 55. this indicates that in terms of net pve effectiveness, trispec will be more effective in 2.32 than 2.3.

where, if anywhere, am i going wrong here?

edit: corrected mistaken assumption (2.3 hemo debuff = +36 bonus rather than +39. oops. note that the edit does not change the substance of the post but rather enforces it.)

 

songster

The loss of DPS due to magic spells eating Hemo charges was never included in the spreadsheet. That means that trispec is currently underperforming on live, and is in all probabilty no better than combat. Possibly even worse, depending on your raid makeup. Even assuming that all Hemo charges get eaten by physical attacks, the improvement over combat is not huge, even on live.

Leaving the bugfix for magic effects out of the equation, the actual changes to Hemo in 2.3.2 are:

1) Going from 125% to 110%. This is a loss of 12% of your current direct Hemo DPS. If Hemo is 25% of your total, this comes out to 12% * 25% = 3% of total DPS. A 45 DPS nerf if we assume baseline DPS of 1500. 30 DPS nerf if your baseline DPS is 1000.

2) Going from 36 to 42 per charge. 10 charges per Hemo. That's a buff of 60 damage every 3.5 seconds, multiplied up by the average raid crit% (say 25%), mitigated by armor. So the debuff portion of Hemo has been buffed by a total of about ~15 DPS.


Thus, at lower gear levels (i.e a baseline of 1000 DPS), this is a net nerf of ~15 DPS. At higher gear levels (i.e. a baseline of 1500 DPS), it's a net nerf of 30 DPS. Not huge, but the margin tri-spec has over combat is not large, even if all charges get consumed by physical attacks.

It's likely that it still confers a very small advantage over combat builds, particularly if you don't use swords, and your gear level is relatively low (T4 to early T5). However that comes at the expense of a debuff slot, longer DPS cycles, little controllable burst damage and increased reliance on bleeds. Cumulatively, these mean that the very small remaining benefit to tri-spec just isn't worth it in context.

Ultimately, it boils down to the weighting you give to dirty deeds. If you think that DD performs better than a simplistic time-based figure (because you do more damage when you need it most), you will weight tri-spec higher. If you think DD performs worse than the simplistic figure (because the last few percent are faster due to executes, molten fury etc.), then you will go for the more reliable Combat damage.

 

harmisajedi

my original post did not rely on any spreadsheet but rather hard math using broad assumptions & observed averages as points of departure. you can challenge the assumptions & averages, but please address the merits of the argument i've presented rather than pointing to other methodologies which may or may not pertain.

please not that i'm disparaging the usefulness of spreadsheets here, but rather am asking for direct feedback to a very precise argument. answers to the tune of "plug your stats into the damn spreadsheet, nub" aren't terribly relevant.

this is erroneous. please consult my previous post for correct math on the extent of hemo debuff, or else explain to me exactly what equation you're using to calculate the change.

i apologize for mistakenly referring to the 2.3 hemo debuff as +39--it's actually +36. i've edited the original post to correct the mistake.

i note that the substance of my original post, which was that in terms of raid-wide dps, 2.32 will be a net increase for trispec relative to 2.3, was not addressed.

thoughts, anybody else?

 

• Vulajin

Almost all of the "increase" you suggest Hemo will receive in 2.3.2 comes from the bug fix where Hemo charges are no longer consumed by spells. If you take the changes separately from the bug fix (which you should, because bug fixes are independent of balance fixes), the changes are a reduction in the theoretical damage of the build. In 2.3, PvE Hemo was considered viable because the theoretical DPS (assuming charges didn't get eaten by spells) was competitive with (and in some cases, even superior to) combat. The actual performance in game was different from theory because of the bug, but it was still considered viable because a bug that obvious was eventually going to be fixed.

Hemo is now considered to be good enough to run with one or none because the theoretical DPS output is pretty much only even with combat even when the debuff is counted. It's not "dead" by any stretch of the imagination, but it's certainly a lot less viable.

Please consider paging back through this thread and spending more time reading rather than dumping a bunch of poorly thought-out math with no capitalization on us. At this point you are not being useful, as all the conclusions have already been reached.

 

hannigaholic

Just a quick question regarding your spreadsheet. How exactly does it model shadowstep? Is it using the Combo Point added to replace a 5th Hemorrhage or is it simply adding the shadowstep energy and damage boost after 5 CPs have already been generated? Or have I completely glossed over a PTR change removing the combo point from shadowstep altogether? Also is your spreadsheet using Shadowstep for finishers or for instants? It's significantly more effective when used for finishers.

It's not a huge difference but for me (mostly T4 with a few T5 pieces - Hemo average hit for around 1k including crits, rupture for 2k) 4s/4+1r cycles (where the +1 comes from Shadowstep) do work out about 3.5% more energy efficient than either the 5s/5+1r or the 5s/4+1r cycles and that's ignoring Relentless Strikes, which becomes more beneficial the shorter the cycle. That's about 1-1.5% more overall dps (since yellow damage only makes up about 30-35% of total dps)

Also I think you were overly harsh when saying "Finally, Shadowstep builds are still universally terrible." They may not be at quite the same level of sustained raid dps of other builds but I think you're overlooking the mobility aspect on certain fights. I can see it being very beneficial at Vashj to Shadowstep 25 yards to an elemental, for example, not to mention the little extra burst that it brings to adds generally. Also remember that you really should be gemming differently for Shadowstep builds. Agility becomes worth far more compared to other builds. I already have an AP-centred gem makeup but simply by regemming for more agility and less hit the rogue gear spreadsheet has me increasing my overall dps by 0.5%.

If your spreadsheet is modelling Shadowstep poorly (not saying it is, just saying *if*) then that's another 1-1.5% overall dps. All of a sudden 11/9/41 is within 2-3% of the pure Combat and the Combat Hemo trispec builds for sustained dps. I think that's more than respectable when you consider the increased mobility, burst and survivability offered by the build. Even at the 4% less sustained dps that you quote I think the versatility is good enough to at least warrant a description other than "universally terrible"

 

◊ s4dfish

It's now +70% run speed after SS, not 1 combo point.

edit: I thought this was changed, but from reading other threads I'm less than 100% certain. Don't want to spread misinformation and if I'm wrong I'll edit my post :P

 

• Vulajin

Who said anything about "4% less" sustained DPS? All three Shadowstep builds came out more than 10% short of pure combat.

My "model" for Shadowstep was simply to tack on an additional 30% to all Ruptures. This was obviously erroneous and, in fact, overvalued the talent. Although Shadowstep allows you to replace one Hemo per cycle with a Shadowstep (assuming a 5s/5r cycle runs just about as long as the Shadowstep cooldown), you lose the damage you would have gotten on that CP by using a Hemo instead. Compare the two theoretical cycles:

5s/5r w/o Shadowstep:
8.8 Hemorrhages for 8.8*H damage
1 Rupture for R damage
Total energy spent: 8.8*35 = 308 (30.8s)
Hemo DPS: 0.286*H
Rupture DPS: 0.0325*R

5s/5r w/ Shadowstep:
7.8 Hemorrhages for 7.8*H damage
1 Rupture for 1.2*R damage
Total energy spent: 7.8*35 + 10 = 283 (28.3s)
Hemo DPS: 0.276*H
Rupture DPS: 0.0424*R

You'll gain DPS equivalent to roughly 1% of your average Rupture damage, minus roughly 1% of your average Hemo damage. A ballpark estimate says this effect will contribute ~1-2% additional DPS. Now account for the fact that my previous model overvalued the damage bonus from Shadowstep and you lose another ~1% DPS. So the overall effect is a very slight gain, if not a wash. Shadowstep builds are still roughly 10-12% behind combat. I think it's fair to call that "universally terrible."

(edit) Forgot to note that this entire post sets aside the Hemo debuff DPS contribution, which is, for all intents and purposes, a fixed amount of raid DPS depending on the overall gear of the entire raid, and not so much your own gear. Shadowstep is roughly 6-8% behind other builds that provide the Hemo debuff, whether you count it before or after the debuff.

 

Buanna

You should just call it 10-12% worse than combat. This forum preaches objectiveness.

 

Ashere

Just put my gear and spec (11/9/41) into DMM's new spreadsheet, and found I'm ahead of combat sword by 0.47%. I was expecting to be about 2% below, so this is a very nice surprise. But it does prove one point: hemo builds are extremely gear sensitive. A piece of equipment that could buff a combat rogue by 0.5% could lower my DPS by the same amount. But at the same time, another piece that might buff a combat rogue by 05.%, might buff me by 1% instead. But it seems the second type of items are rather uncommon. -.-

 

Alacrity

This thread is dedicated to bettering yourself in pve and providing knowledge on how hemo can potentially compete with a combat build, if you can be doing 10-12% more dps by changing your spec.. then your spec is "terrible". Im sorry but there is no other way to put it.

Rogues are brought to raids for dps, I would force a rogue to respec or gkick him if he came to raids as shadowstep.

http://www.exo-guild.com

 

◊ Hanos

You have the wrong options selected. There is no way Shadowstep is going to be ahead of combat swords. Turn off Hemo in both the talents tab and in the buffs tab since both of the over estimate the effect (the first assumes all the charges are used), the second assumes you use a charge on every hit (not possible without several Hemo rogues). Now check your DPS as that will be closer to what you see.

 

Ashere

Note how Valujin conveniently ignores the increased raid DPS from the debuff when he says 10% worse. Depending heavily on gear, it ranges from -5 to +5%.


1) The new talentsheet has coded this so you can select only either one of the options. So ticking "include hemo debuff estimate" makes that hemo in the raidbuffs section is ignored. So you can't overuse it anymore.
2) Every hemocharge will *definitely* be used. Not activating "Include hemo debuff estimate" is plain stupid, since hemo as an attack basically IS about that debuff. If you want to know what a hemo rogue brings to the raid, then you have to tick that box, and then compare it with a combat-sword rogue with similar equipment, but who definitely doesn't have a single hemo thing activated (that includes the hemo raid buff in the gear tab)
3) What I see in Personal DPS doesn't mean a thing for raiding. It's the total raid damage that matters, and the increase for bringing a hemo rogue is NOT just his personal DPS, but his personal DPS combined with an increased overall raid DPS. That would mean there won't be a place for paladins or druids anymore either, since both can be replaced (pallies by priests, druids by warriors), since according to you their raid buffs are worthless, because they don't show on the damagemeters.

 

◊ Hanos

See my post a page or 2 back regarding the definition of "Raid Viable". Enhancement Shamans, DPS Warriors, Shadow Priests are all raid viable because they bring more to the raid then just their personal DPS. With the Hemo Debuff selected on the talent page I get a 4.3% decrease from combat, with both selected I get a 1.26% decrease, with neither I get a 13.73% decrease. I guess part of it is the gear gap as the damage from Hemo is more valuable with Kara/Crafted Gear then with T6 Gear, since the debuff doesn't scale, and SS is going to scale faster then Hemo with better weapons etc. Are you basing this off RogueDPS_2_3_2_4.xls (Rogue Gear 0.9 shows it as closer to 20%)

So even if you assume that you get charges for every swing, and you don't use any of your own charges, you still end up behind. If you just assume all the charges get used, then you end up losing 4.3%, if you ever manage to Hemo again before all the charges are used, it will be even lower. Not to mention that the spreadsheet doesn't account for stacking cooldowns, aka popping AR, BF, Drums, Haste Pot and Heroism all at the same time, which benefits Combat more).

If you can show a set of gear where Shadowstep beats Combat please do, but right now you just sound like you are talking out your...

 

Buanna

I never disagreed with that, yet you argue like I am. If someone can't see why it would be "terrible" after seeing that it's 10-12% lower DPS, then you telling them so isn't going to help either.

Boni and Bonii are not words.
Bonuses is.

 

• Vulajin

10-12% before the debuff, 4-6% after the debuff. (These numbers were given in post 1461.)

Considering that you could do trispec Hemo, still grant the debuff, and come in at +1-2%, I don't see why people are in such a huff about me calling Shadowstep terrible.