seminarca

k, so the optimal melee group debate has been going on for some time. It's widely believed that Enh/Warrior/3xRogue is the most optimal use of an Enhancement Shaman, so as to not waste the Windfury totem, and without a doubt it's a very strong group. Recently, some discussion started in the Feral Druid megathread about whether substituting one of the Rogues for a Feral is worth it, and so I decided to look into it a bit more deeply.

Throughout all this, I hope to remain as impartial as possible, and will test a number of alternate scenarios. The goal is to figure out the group setup that achieves the highest possible cumulative DPS. This will be at the highend T6 progression level.

Our melee group will be as follows:
Enhancement Shaman
Fury Warrior
Sword Rogue
Sword Rogue
(empty)
This group will have imp Battle Shout + Saphire, improved SoE & WF totems, Unleashed Rage (no twisting).

Aside from our test melee group, a tank group with the following makeup will be used to slot the 6th person:
Prot Tank
Prot Tank
Feral Tank
Resto Shaman (WF Totem)
(empty)
This group will have base Battle Shout, LotP and base SoE & WF totems.

Additionally, an alternate scenario without a Resto Shaman will be tested, e.g. for Alliance guilds that may not have access to many:
Prot Tank
Prot Tank
Feral Tank
[Bum slot, e.g. Warlock for Imp, Holy Paladin or Resto Druid]
(empty)
This group will have base Battle Shout and LotP.

Other buffs assumed:
Blessing of Might (base)
Blessing of Kings
Imp Mark of the Wild
Appropriate food/flask/elixir buffs
No drums, Bloodlust or potions (e.g. Haste Potions) considered to keep it relatively simpler.
Debuffs assumed:
5 x Sunder
Faerie Fire (not improved)
Curse of Recklessness
Caster debuffs (for Shaman spells and Rogue poisons)

Tests are conducted as follows:

1) DPS is calculated for the Melee subgroup consisting of the 4 "permanent" players + an additional Rogue. The Feral is added to our two alternate groups. Total DPS is calculated for each scenario.

2) DPS is calculated for the Melee subgroup consisting of the 4 "permanent" players + the Feral. The remaining Rogue is added to our two alternate groups. Total DPS is calculated for each scenario.

I'll be using the various tools that each class megathread has produced. As far as I can tell, they are the most definitive simulations as widely accepted on these boards. In brief:

Enhancement Shaman:
Yo!'s sim
The thread: Enhance Shaman: The Collected Works of Theorycraft, Vol I
Represented by: Sebudai from Doomhammer

Fury Warrior:
DPS Spreadsheet
The thread: [Warrior] DPS Spreadsheet 2.3 and beyond
Represented by: Loderunner from Mal'Ganis

Rogue:
DPS Spreadsheet
The thread: http://elitistjerks.com/f31/t19926-r...s_spreadsheet/
Represented by: Furi from Illidan .. only I'll replace his Warglaives with the Savagery/Infamy combo to keep things sane.

Feral Druid:
Rawr
The thread: http://elitistjerks.com/f31/t16902-f...id_megathread/
Represented by: Arkadu from Doomhammer

All three Rogues will be duplicates of our representative. I have chosen people who (as far as I know) have been farming the T6 content for a while, and so are expectedly geared optimally. More importantly, they are logged out in what (as far as I can tell) is their PvE DPS gear.

Potential Flaw:
* Accuracy of sims, some might be underestimating, whilst others could be overestimating.

Anyway, the results:

Enh / Fury / Rogue / Rogue / Rogue

Enh: 1450
Fury: 1621
Rogue: 1733
Rogue: 1733
Rogue: 1733

*Feral: 1303 (Tank group, Resto Shaman)
*Feral: 1265 (Tank group, No Shaman)

DPS when tank group gets a Resto Shaman: 9573
DPS when tank group gets NO Shaman: 9535

Enh / Fury / Rogue / Rogue / Feral
Enh: 1511
Fury: 1721
Rogue: 1803
Rogue: 1803
Feral: 1412

*Rogue: 1637 (Tank group, Resto Shaman)
*Rogue: 1548 (Tank group, No Shaman)

DPS when tank group gets a Resto Shaman: 9887
DPS when tank group gets NO Shaman: 9798

RESULTS ARE ONLY ACCURATE AS THE LINKED SIMULATIONS

So it seems:
* Replacing a Rogue with a Feral results in higher DPS, regardless of whether the replaced Rogue gets WF totem or not (Rogue used MH IP in this case).
* Warriors seem to benefit a hell lot more from LotP than Enh or Rogues. This could have something to do with the gearing style of each particular representantive (e.g. Sebudai has a fairly high crit chance already, whilst Loderunner has a lot of passive APen).
* Enhancement Shamans are huge for Ferals, 147 dps over having no Shaman, 109 dps for UR + Imp SoE alone. This is without twisting btw, no GoA was used at all.

Another option used by many guilds is to put the orphan Feral into a 3 x Hunter + SPriest/RShaman group. This will likely make it much closer, I might track down some Hunter spreadsheets and give that a shot later.

Any thoughts? Have I made some glaring omissions? Silly assumptions?

edit: Typos

 

• constantius

Have you considered trying the following combinations:

DPS Group
Fury/Arms Warrior
Sword Rogue x2
Enh Shaman
Ret Paladin

Tank Group w/
Enh Shaman
Warrior (no Solarian's)
Feral Druid

as compared to

DPS Group
Fury/Arms Warrior
Sword Rogue x2
Enh Shaman
Feral Druid

Tank Group w/
Enh Shaman
Warrior (no Solarian's)
Ret Paladin

You can model a ret paladin fairly easily as a flat amount of dps and 2 raid buffs, without worrying too much about the variation in their dps due to losing 70 AP. We run with 1-2 ferals, 1 ret, 2-3 rogues, and 2 enh all the time, and the optimal setup for that is of interest to me.

(nice work so far, btw)

Edit: including a ret paladin gives Imp BoM for the raid, which can also be factored in if you're comparing a with ret vs without ret case. The raid buffs, for easy reference, are JotC (3% crit for all dps) and 2% dmg (aura, for party only).

Also, what happens if you try:

DPS Group
Feral Druid
Ret Paladin
Enh Shaman
Rogue x2

Tank Group
Enh Shaman
Warrior (no solarian's)

which was the case for us the other night. I'm curious how much is *lost* by not putting the warrior (i.e. Battle Shout) into the dps group, vs the 5% crit buff.

Anyone who cannot cope with mathematics is not fully human. At best he is a tolerable subhuman who has learned to wear shoes, bathe, and not make messes in the house. - R.A. Heinlein

 

Threeshee

I was wondering if you could do the same sim with the shaman dropping GoA.

 

dr_AllCOM3

The fury warriors DPS seems to be too high (and rogue too low).

 

• Vulajin

Keep in mind that the DPS spreadsheets typically model sustained DPS, which usually averages out to less over infinite time than it will in a 5-minute fight (as are the ones commonly used as benchmarks).

 

seminarca

2 Enhancement Shaman situations are ideal, everyone gets what they want =p

Enh / Fury / Rogue / Rogue / Rogue
WINDFURY
Enh: 1450
Fury: 1621
Rogue: 1733
Rogue: 1733
Rogue: 1733
Total: 8270
Grace of Air
Enh: 1492
Fury: 1546
Rogue: 1732
Rogue: 1732
Rogue: 1732
Total: 8234
I don't know why switching WF to GoA+IP results in such a tiny drop for the Rogues.

Enh / Fury / Rogue / Rogue / Feral
WINDFURY
Enh: 1511
Fury: 1721
Rogue: 1803
Rogue: 1803
Feral: 1412
Total: 8250
Grace of Air
Enh: 1554
Fury: 1637
Rogue: 1799
Rogue: 1799
Feral: 1472
Total: 8261

Again, WF to GoA+IP results in a very tiny drop for the Rogues. The only person who really suffers from lack of WF is the Warrior. Funnily enough, in the melee+Feral group, the circumstances of our test show GoA being slightly better. Probably due to the lack of taking things like Bloodlust and Haste Potions into account? So I wouldn't read much into the immediate results, aside from that this one particular scenario might warrant further investigation.

Of course, twisting is the best option if your Shaman is into that sort of thing.

I just realised I forgot to remove the passive AP on Berzerker's Call from the Warrior's AP (since he'll be using Saphire in that slot) but that wouldn't throw it off by that much will it? He was buffed in armory, so I subtracted 39 from his Strength (19, Imp MotW and 20, food), 78 from his AP (due to the same 2 things), and a further 90 AP (Elixir). Anything else I could have missed? I used the following inputs:
Talents 17/44 DW
Strength 594
Agility 192
Attack Power 2149 (should be 2059)
Crit Rating 384
Hit Rating 180
Haste Rating 31
Ignore Armor 917
Expertise Rating 0
Meta Gem: Crit 1
Meta Gem: Haste 0
Hyjal Ring (Exalted) 0
Trinkets:
Dragonspine Trophy
Solarian's Sapphire
Set-Boni: (2) (4)
T4 0 0
T5 0 0
T6 1 0
The Twin Blades of Azzinoth 0
Mainhand:
Speed 2.60
Minimum Damage 208
Maximum Damage 313
Additional Damage 0
Offhand:
Speed 2.60
Minimum Damage 187
Maximum Damage 349
Additional Damage 0

 

Pitbuller

How much dps will prot tank and prot warrior(dps mode) lose without lotp. I know this isnt huge factor but still have to take in calculations.

Slow, slower, shaman weapon.

 

Prost

Actually, the second enhancement shaman would probably go in the tank group so as not to waste a UR because they don't stack as well as increase the aggro cap of the entire raid. To "get what everyone wants" would require one of two things:

1) Get an enhancement shaman to twist WF and GoA, which is tough to model (I would think) because of cooldown management.
2) Help rogues and warriors to understand that WF is still considered the best weapon buff in WoW with most gear setups.

 

seminarca

Tank group has it's own LotP (2 Ferals in the raid), look at the groups.

I didn't mean two Enh in one group, a 2nd Enh in the tank group means you can put a Feral there without him losing UR. The only thing he loses there is Imp BS + Saphire, which is a lot less of a loss.

 

Pstar

We run with a group as suggested in #2, with a ret pally and a 33/28 MS warrior, two rogues and an enhancement shammy... sometimes we get a feral druid instead of the ret pally. We also have (in the raid) a SV hunter with 5/5 imp mark, a moonkin with imp FF, and curse of reck always up, all important buffs/debuffs to throw in there imo. It's probably worth noting that judgement of the crusader and blood frenzy buff the other 20 people in the raid, too, in one way or another, but that's obviously more difficult to quantify than the 5 group we're talking about.

#6 may have neglected to discount the AP bonus from berserker stance, perhaps? Also 910 is a very high amount of armor penetration; it could just be that this warrior outgears the rest of the players you sampled.

 

mad-doc

If you tracked down a hunter spreadsheet, i would like to see:

BM Hunter
Enh Shaman
Warrior
Feral Druid
Ret Pala or Rogue


@ twisting
Just add GoA to the whole group.
The WF uptime should be close to 100% if the shaman does it right.

 

• Daboran

Seeing as accepted wisdom for optimal melee dps is to have your first DPS Warrior as MS/Arms then all subsequent Warriors as Fury, where do we then put the MS Warrior?

I'd usually make sure he is in the Windfury group, but then it's a pretty tough choice between taking yet another Rogue out of the "optimal" group.

I'm assuming that if you only have a Resto Shaman in the melee group rather than Enhancement (which is still not that common a spec) then the accepted wisdom of Warrior, Shaman, 3 Rogues applies due to no UR/Imp SoE

 

Eledorian

Assuming your optimal group setup for melee would be:

Enh Shaman
Feral Druid
Warrior
Rogue
Rogue

You could replace the feral druid with a warrior and move the feral to the hunter group. We run that setup somtimes when there's 2 DPS warriors in the raid (1 arms, 1 fury).

 

khel

It would be nice to see these calculations done with a resto shaman instead of an enhancement shaman, as many guilds do not use an enhancement shaman. I could get a rough estimation removing the enhancement shaman from both calculations, but the 10% attack power bonus works so nicely for feral druids, as well as improved totems, that it should really be modeled separately.

 

dr_AllCOM3

Sure it's too high . You have to input his stats in Battle Stance. The armory is almost always in Berzerker Stance. You'll have to substract about 70DPS.

 

Unaz

Enhancement shaman over resto is about a 350 ap difference on rogues (if the rogue has ~3k AP before UR), and goes up with more AP. Not counting the AP bonus on imp wf totem (133 AP or 191 AP with UR). In addition to issues with a resto shaman having to be near the melee, and also wasting all his caster totems like tide, and not being able to use Wrath of Air.

You're also buffing a 4 person melee group instead of 5, so the ranged dps you bring has to make up for the AP difference in buffs + what the enhancement shaman would have brought in personal dps. Which at 1500+ dps base (and that's on sims, on actual fights, I hit 1600 dps in my current gear with 2x drums in the group, and Sebudai outgears me by quite a bit), and 300+ ap on the other 4 melee will be pretty near impossible. And your casters will be slightly worse for wear from missing wrath of air/mana totems.

 

tobiah

Using the Hunters in my guild as a reference and Cheeky's Spreadsheet I get the following (full raid buffs = flasked, kings and might, food buff, MotW, Faerie Fire (Not Improved), 5 Sunders, CoR) -- 7700 base armor on target, lvl 73 boss mob

The Druid numbers are from the druid simulation at this link: The Druid Wiki � Rawr

Totems are Grace of Air (Not Enhanced), Strength of Earth (Not Enhanced)

A group of the following classes: 2 Beast Hunters, 1 Survial Hunter, 1 Resto Shaman, 1 Feral Druid

The First Number is DPS without a shaman's totems, the 2nd number is with totems.

Me (Best Spec): 1891.85 DPS -> 2018.94 (127.09 DPS increase)
2nd Beast Hunter 1721.74 DPS -> 1846.26 (124.52 DPS increase)
Survival Hunter: 1519.02 DPS -> 1607.36 (88.34 DPS increase) -> Expose Weakness goes from 207.56AP to 233.04AP
Feral Druid: 1012.92 DPS -> 1089.93 (77.01 DPS increase)


*edit: If improved totems are used and rage from the shaman (rage effects Hunter pets (kill command) and feral druid):
Me (Best Spec): 1891.85 DPS -> 2063.77 (171.92 DPS increase)
2nd Beast Hunter 1721.74 DPS -> 1887.98 (166.24 DPS increase)
Survival Hunter: 1519.02 DPS -> 1664.40 (145.38 DPS increase) -> Expose Weakness goes from 207.56AP to 235.01AP
Feral Druid: 1012.92 DPS -> 1214.92 (202 DPS increase)

These stats do not include the bonus 6% damage from having 2 beast spec hunters in the group.
Hopefully these numbers might be of some use.

 

Krennick

Without being too terribly helpful I thought I'd throw in another idea:

We sometimes end up running a melee dps group with the following:
Fury warrior
Sword rogue
Feral druid
Enhancement shaman
Retribution paladin

With the tank group containing:
Prot warrior
Prot warrior
Feral Druid
Resto shaman
Holy Paladin

I can kind of see where you're coming from including the tank group as raid melee dps, but since some people in the tank group are tank support, some are tanking in some situations and dpsing in other situations and others (resto/enh shaman and potentially holy/retri paladin) will sometimes be heal and sometimes dps spec, I would not personally have chosen to look at total melee dps for both tank and melee dps groups. There are a few too many variables to me.

The difficult and intangible benefit here is the retribution paladin. Sure his group gets 2% extra dmg to anything (and 10% extra to holy which noone else does) - he benefits immensely from WF totem, and his presence in the raid (rather than absence - not comparing to putting him in the tank group as you were doing) gives everyone 3% crit chance when hitting the judged mob (including the other melees) and somewhat importantly - probably ensures 100% uptime on judgements of light and wisdom. Light is not a dps increase, but wisdom is - for pretty much everyone else who is dpsing who is not in the melee group.

Now, if you wanted to include a retribution paladin for the judgement giving everyone 3% crit rating you could include him in the tank group and still have your second rogue in the melee dps group and still get the judgement benefit without the 2% group dps increase and the retribution paladin probably still gets Windfury totem and maybe, meybe not gets Unleashed Rage from the shaman in the tank group.

 

seminarca

Re: Raid makeup in general, like MS Warriors and Ret Paladins. All good points, and exactly one MS Warrior is a nice boost in rDPS for sure, so is a Ret Paladin. But those are more macro level raid comp issues as opposed to micro level group comp issues. And in the case of Warrior, we can run the numbers again replacing the Fury Warrior with MS Warrior. I originally went with Fury because I didn't realise the Warrior DPS Spreadsheet had been updated to work with Arms builds (last looked at it ages ago). Unless there's a wildly different valuation for crit chance (LotP) between Arms and Fury, then it shouldn't affect the relative numbers by *that* much (except for the WF vs GoA debate, Arms will take a larger hit, but that's a complete digression and not one I want to explore just yet). That might be the case due to factors like Flurry and Blood Frenzy uptime, but at the T6+ level it should be safe to assume near 100% uptime anyway (?).

Re: asking for numbers on specific group comps. There are sooo many different ways to structure a group, that running sims for each specific one really is beyond the level of effort I'm willing to put into this =p However! What I'm doing isn't rocket science. I've linked all the tools I used, all found on these forums. For any missing classes, I believe tools exist for them as well. I would suggest gathering all the necessary info, searching armory for suitable T6 candidates (or even use the specific people you have in mind for higher relevance) and run them through the spreadsheets and sims. All you need to do is be very aware of which raid and group buffs you turn on or off for the particular case you're trying to test.

Pstar, I posted the armory links of all the people used for this, you can check and verify that they are mostly equivalently geared. And you're both absolutely right, I forgot about the Berzerker Stance AP bonus, which will bring down the Warrior's personal numbers accordingly.

Well you still have the base SoE totem and Battle Shout from the Warrior, both of which are not negligible.

If melee has to make do with a Resto Shaman it's assumed there is a shortage of the class, and so the tank group has a Bum slot character, 2 Prots and a Bear. Cat or Rogue takes the remaining slot.

Resto Shaman / Fury / Rogue / Rogue / Rogue

Resto: 0
Fury: 1498
Rogue: 1626
Rogue: 1626
Rogue: 1626

*Feral: 1265 (Tank group, No Shaman)

Total: 7641

Resto Shaman / Fury / Rogue / Rogue / Feral
Resto: 0
Fury: 1591
Rogue: 1691
Rogue: 1691
Feral: 1319

*Rogue: 1548 (Tank group, No Shaman)

Total: 7840

Taking this a step further. If all your optimized groups are bulging full, Hunter group doesn't have a spot, Tank group doesn't have a spot, and you end up having to lump a guy in with 4 healers (of the non Shaman variety), you get:

Enh / Fury / Rogue / Rogue / Rogue

Enh: 1450
Fury: 1621
Rogue: 1733
Rogue: 1733
Rogue: 1733

*Feral: 1205 (Healer group, no group based buffs)

Total: 9475

Enh / Fury / Rogue / Rogue / Feral
Enh: 1511
Fury: 1721
Rogue: 1803
Rogue: 1803
Feral: 1412

*Rogue: 1398 (Healer group, no group based buffs)

Total: 9648

What does this tell us?

LotP for an Enh/Warrior/Rogue/Rogue group + the increase to the Feral's personal DPS > the loss of personal DPS for the Rogue that was subbed out, regardless of where you end up having to put him.

Out of curiosity, do Hunter spreadsheets model pet DPS? Does enabling stuff like SoE totem increase your DPS (via AP to pet) as it theoretically would?

Precisely! The point of this wasn't to say "do x in every given situation". Rather just to open people's mind to the possibility that blanket statements such as "Enh/War/Rogue/Rogue/Rogue or die" might not be the holy grail.

I would encourage people to reevaluate their perceptions, utilize the available tools for their particular scenario and figure out what's best for their raid.

 

panny

Nice work verifying some of the things we've long suspected. And this doesn't even factor in the advantage of greater threat from the tanks.

Fix Spirit Wolves not responding to commands.
DK/Rogue

 

Krennick

The way I'm seeing it you're not really exploring the damage potential of a single group. You're somewhere between group and raid dps.

You are assuming that the feral you are not putting into your group is still in the raid doing damage - and you are assuming that the third rogue you are not fitting into the group is still in a different group doing damage.

In a similar vein a retribution paladin or second enhancement shaman could be examined, and it seems to be a bit arbitrary to decide to include the out of group rogue or the out of group feral druid, but not include the out of group enhancement shaman or retribution paladin. You've made an assumption that the raid will have one feral druid (and one only), 0-1 enhancement shamans (and not two), 3 rogues (and not less) etc.

All this said you are very correct to point out that it's silly to have a firm belief that only one specific melee dps group configuration is optimal. What is optimal will always be determined by what you have in your raid. Sometimes your invite priorities will allow you to get both your best dps players and the best combination of classes, but most of the time what you have is a trade-off between good players, a good balance of classes, and what fits into the group synergies you manage to build into your raids.

I think your numbers show that the enh/fury/rogue/rogue/rogue group is probably the best dps output if we look at the melee group in isolation - but I agree with you that looking at the melee group in strict isolation is probably very wrong. Typical for me is two groups in which you find melee characters (tank and dps both) - if we were looking at the optimal dps output of these we could imagine something like the following:

Prot warr, enh shaman, feral druid, retri paladin, arms warr
Enh shaman, fury warr, rogue, rogue, rogue

If we look at it with the eyes of optimising melee dps (and tank threat at the same time) this might look like a much better starting point than the single melee dps group and we can get into all kinds of discussions whether one of the classes in the list should be replaced with a different class, or whether someone from one group should be swapped into the other, and we'd be ignoring the huge big issue here which is healers. Is it fair to make a raid so melee heavy that you're not fitting any healers into the tank group? If you do, you're reducing the slots you have available for ranged dps - it means you won't have a full complement of curses or as many shadow priests you would like etc. Every time we take a limited view of a raid and optimising for it we're reducing the potential of the rest of the raid to compete in our area of focus - in this example when talking about dps because we foist all the healers on the range dps groups.

I'm not saying that deciding that there will be one additional melee dps person in the raid who is not in the melee dps group is a bad assumption, but it does move the discussion on from the alleged purpose which is one-group melee dps optimization (which as I said I think is not particularly helpful to investigate).

 

galzohar

I think you should seriously consider the options of simply removing all melee dps that is outside of the melee group, and bringing others that would simply do more DPS, especially considering the tank group could just have tanks+tank buffers and no dps classes so the casters/hunters you'd bring instead would actually gain proper party buffs.

Do you want to maximize dps with a given raid, or adjust the raid to maximize dps? This decision alone can make a big difference on your group composition choices. For example your feral druid might never see catform in a raid (except when his tanking target died...) if those simulations/spreadsheets are correct. I mean a T5 warlock on lieuler's spreadsheet specced affliction and sitting in the tank group can do 1450 dps based on lieuler (www.lieuler.com 1470/1345 shadow/fire dmg, 22.16 crit 11.69 crit 49.2 mp5 with full buffs and consumeables (but no potions which could add another 100 mp5)). Of course I'd verify how accurate the spreadsheets/simulations are when it comes to actual raid dps - discounting stuff like armor debuffs, hunter's mark etc could greatly lower the expected melee dps. Then again if a ranged DPSer can do the same dps as the melee, you should probably take the ranged as he'll have easier positioning, take less damage and have higher DPS time on many fights in the game.

 

Kukulkan

First i want to thank the thread starter since this is really good subject to discuss, and for his effort toward making this discussion as detail as it can.

I kinda disagree to some extent to this though, i think rogues are undervalued in those simulations. The following are some recent WWS from teron, which have mostly the warrior/druid/enh shamman/dual rogue setup.

DnT 2700dps rogues (almost 1k dps over the feral with them)
Drow 3000dps rogue (almost 1k dps over the feral with them)
Disbanded 2600dps rogue (almost 800dps over the feral with them)
Abananax 2500dps rogue (almost 700dps over the feral with them)

Given that they wouldnt achive this personal numbers without the druid and the fight is very short and cd matter a lot, the differece in the "optimal setup" between rogue and the feral isnt 400dps as your simulation shows, in some cases over double of it, given that we dont know what the feral is wearing for that fight, i would consider 400 being too low.

The difference between triple rogue vrs the dual rogue+feral setup its marginally less than 200, meaning that if the 3 rogue can put 70dps more over their current value, they will outdo the feral+dual rogue, which by those wws shouldnt be that outside of a rogue could achive according to those wws.

 

• Latito

As one of the 4 rogues listed in the post above, I'll make a few comments.

First off, I had a Enhancement Sham, Feral, Fury warr, Ret Pally group (Yibn, Deep, Bakatora, Zurm if you care to look). Also, I was on imp Expose Armor duty (note the lack of Rupture dps in the parse. Thus, my dps *should* have been around the 2700+ mark. Nevermind that no one realized we didn't have imp BoM and Faerie Fire got knocked off after 14 seconds and never went back on. But those are.. separate issues.

The biggest difference between a 2:13 Teron kill and an indefinite dps simulator / spreadsheet is CD's. Instead of 1 Bloodlust like I got, Teron was basically 4.5 Bloodlusts due to the length of the fight. 1 Adrenaline Rush becomes 2.25 AR's. Blade Flurry, Blood Fury, Haste Pot, Drums, etc. What CD's do cats have? Basically just powershift until they go oom for free energy?

High-end Teron kills such as the 4 you posted are more about raid composition, group stacking and blowing CD's. Given that raid makeup, Deep could have done nearly 1800 dps indefinitely - maybe 1600 or so. Could I have maintained 2400? No. 2k.. sure. Rogues will scale a lot more in a short fight.

Long story short, I wouldn't recommend using Teron parses as justification for group makeup superiority. Or anything really.. other than how to kill Teron quickly.

 

dukes

Taking this a step further - is a sustained DPS situation really the correct one to model for working out raid dps gains? By analysing fight duration against cooldown lengths, we should be able to get a better approximation of where particular classes will get major benefits.

The below is a summary of fight length taken from looking at the average of the bottom of the first page of WWS parses and then taking a minute time bracket around it (generally with the wws parses at the lower end)

Hyjal
Rage Winterchill : 2 - 3 minutes
Anetheron : 2 - 3 minutes
Kaz'rogal : 2 - 3 minutes
Azgalor : 4 - 5 minutes
Archimonde : 4 - 5 minutes

Black Temple
High Warlord Naj'entus : 3 - 4 minutes
Supremus : 4 - 5 minutes
Akama : About 1-3 minutes from my recollection.
Teron Gorefiend : 2.5 - 3.5 minutes
Gurtogg Bloodboil : 5.5 - 6.5 minutes
Reliquary of Souls : 1.5 + 1.5 + 1.5 (something like that anyway)
Mother Shahraz : 4.5 - 5.5 minutes
Illidari Council : 7 - 8 minutes
Illidan : 15 - 20 minutes

The fights which allow sustained DPS from a pure timeframe perspective often break up that timeframe with other events - Gurtogg has Rage/threat issues, Council is often gimped slightly because of things breaking up under AoE's and stuff happening, and Illidan is funky with phases. Cooldowns impact a lot of things - like Kukulkan posted, actual DPS for those classes that can use cooldowns and do use them correctly is a lot higher than that of classes stuck with no unique ones (for example, trinkets are a cooldown, but everyone gets them)


In essence, is sustained DPS really the correct way to look at this for the current state of the game?