One thing I have considered is that the value of total damage done can not be understated.

I look at Turyn on Arathor. He has a recorded 6599 dps, DPS time: 2:17, Total Damage 904,198 (6.93%) listed here WoW Meter Online

It took me 8 seconds longer to do 309 damage more damage than him. Yes, his total DPS number is higher because there was a shorter fighter but he also did more damage in less time therefore... if the fight had gone longer he would have still done more dps than myself.

So I really think total damage and time might be valuable entries to help determine relative dps.

6181 dps
32/39
normal 32/39 rotation
WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!
2min 30 seconds
a good 15 sec of lag during bloodlust where i thought i was going to disconnect.
Angry dead(crusader) Widow's fury (cinderglacier)
25 man.
Dropped split greathammer for me afterwards .
Curious to see what i can do next week without a lag spike and the new weapon.

3247 is extremely low for Patchwerk and cannot be blamed on your guild taking too long. You also only did 2189 in VoA in the WWS you linked. Look at top dpser's gear, look at Zurm's first post in the Unholy DPS thread, and then look at your own gear. Stop blindly socketing for socket bonuses, especially when they are things like stam and hit when you're already 1.67% over hit. Of all the socket bonuses to skip, the only one you skipped is the strength bonus, and you socketed an AP gem instead of a strength gem. You have str/haste socketed in your chest. These sorts of bizarre decisions betray a lack of understanding about the class and itemization, and suggest you'll gain a lot more from reading up and thinking about the class than seeking personal advice here.

I'm sorry but, I don't really get what you're getting at.

You want people to look at the damage done in the given time, rather than the dps. While those 2 are the exact same. You said: "he did more dps, but he also did more damage in less time", which is the definition of higher dps. Having a higher number before the divider (damage) than after the divider (time).

You could look purely at damage done, but this number is very dependent on your raid, if the rest of you group is doing kick ass dps as well your damage will be lower than if you're raiding with slackers.

The problem is that this game lacks a benchmark, patchwork is just zerged down too quickly. And any other fight has some kind of factor (AoE / Movement / Special phases), that make it impossible to get a clear answer. I hope ulduar is bringing us a good benchmark again.

Stoical,

You are posting nothing I don't already know. I have a great understanding of the class, and I understand the gems I currently have are not 100% ideal and that I am way above the hit cap w/ the shoulders I just gained last night. I do not see how a couple of gem changes will make up for over 1k DPS that I seem to be missing. Especially the strength/haste and attack power gems I got for free, one of which I am useing on the crap emblem belt I am hoping to replace soon. Anyway, thanks for the tips. Maybe my gem compositions will make up for 1k DPS.

Just regemming and keeping mostly the same slots everywhere would add 124 strength (assuming 3 JC strength gems, helm/shoulder/legs, and replacing the boot gems with 16 strength, and adding a belt buckle), at the cost of 30 AP (assuming you also enchant your bracers with +50 AP - although you could also pick up +10 stats to chest and +23 haste to cloak. AP to boots would also probably be better than Icewalker). That would be a net gain of 256 AP.

T7 legs will also get you a 4 piece bonus and let you shed a lot of excess hit. Darkmoon Card: Greatness would also be a boost, when you can get it in 2 weeks.

Unholy Blight did 148 ticks in 4:27, out of a potential 267.

Kazridan is also Unholy, so either one of you didn't use a ghoul, or you both got the same ghoul name.

You generated 1576 RP during combat but only used 1540 (I don't see a usage of Blood Tap, Bone Shield, or Horn of Winter for more RP). 4t7 would add 500 RP, or 38165 damage if used on DC, or 143 DPS if used on DC.

I don't really look at the dps numbers, the position relative to the rest of the raid is interesting.

51/0/20
5055 on patchwerk, ~5300 with AotD

Wow Web Stats

0/32/39 (Widow's Fury MH + Maexxna's Femur OH)
4496 on patchwerk, ~4800 with AotD

Wow Web Stats

I was fairly impressed with blood since pre-patch I was doing ~4k, but DW was a waste for me. I'm going to try 21/50/0 next week for more comparisons. =D

So far I've been 21/50/0 for a week.

WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!

10 man numbers.

I wish I could pump out more damage. I seriously see 25man DPS with DK's any spec doing 4.5k DPS. If it's buffs holding me back then nothing I can do.

21/50/0 (2xBS > 2xOB > FS Dump > 6xIT > FS Dump), Unholy Presence
Patchwerk, 2'34"
AotD at start of the pull, Ghoul at around 40%.
Buffs: Flask of Endless Rage, Fish Feast
The World of Warcraft Armory

5344 DPS

27/44/0

http://talent.mmo-champion.com/?deat...h=101213060504

Tried to fit in Abdominations Might into the standart 21/50/0 2H-Frost to provide AP Buff because no marksmen or enhancer present that evening.
Rotation: IT - BS - FS - OB - FS - OB - FS (-Rime/FS).

Patchwerk DPS 4382 at (AotD missing) WWS WWS Loading...
Recount showed 4780 (including AotD).

No great numbers after all, but still no Betrayer and no DM:Greatness.
Used Deaths Bite (Fallen Crusader), 4xT7 and the Sigil of Awareness.

10/10/51 Talent Calculator - World of Warcraft
Patchwerk 25 man 02:42 (DPS time 02:35)
DPS 5736 (Actual 5979)
WMO: WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!

Main hand is [Split Greathammer] with Fallen Crusader, offhand was [Fang of Truth] with Razorice, sigil was [Sigil of the Frozen Conscience]. New offhand, no other changes from my current armory.

Endless Rage, 40str food, 30str pet food. Used [Potion of Speed] + Hyperspeed Accelerators - Spell - World of Warcraft for AotD on the pull in Unholy presence, then switched to Blood and ran in.

My execution had some problems and my gear has some weak spots, but I felt it was a good showing nonetheless, and another data point for deep unholy DW. I'm pretty happy with the strength of this build. I may use this as a personal benchmark and compare it to 0/20/51 later.

Priority system is pretty simple, same as 0/20/51 but without Killing Machine, so no need to space out ITs. Keep up Blood Plague > cast Icy Touch > convert Blood runes to Death. RP dumps are Deathcoil and keeping up Unholy Blight. I'm currently experimenting with Blood Boil vs Blood Strike, mostly because my expertise is low and I hate losing a GCD to a dodge. Having an Unholy rune sit unused most of the time is normal. Oh, and Blood presence.

Glyphs are Glyph of Icy Touch, Glyph of the Ghoul, and Glyph of Bone Shield. I will probably change Bone Shield to Death and Decay.

I played 0/32/39 before this and I feel like I do better single target with this build, but the patch happened the day after I switched specs so I can't say for sure. I also thought I would do less damage on trash from losing the quick burst aoe of Howling Blast, but my damage output has been great. Here's overall WWS for that night in Naxx, if you can get it to load: Wow Web Stats

Unfortunately I don't have our most recent Sarth 3d attempts to link, and our older ones are missing big chunks of data. My single-target on drakes is very good, and I can easily pestilence diseases when the add tank is nearby for damage and debuffs. I also don't have to worry about Howling Blast grabbing flame adds anymore, which is nice.

edit: added info

I apologize if it wasn't clear. What I am getting at is that time spent on the encounter and their relatively contribution to the damage in the encounter gives a much large picture of their damage contribution overall then only a dps value. I think context is very important.

I give three scenarios to try to make my point more obvious:

1) A car travels at an average rate of 100 miles/hour for 10 seconds.
2) A car travels at an average rate of 100 miles/hour for 10 minutes.
3) A car travels at an average rate of 100 miles/hour for 10 days.

To say a car travels 100 miles/hour is not enough.

I think the best option is to work with what we have. Back to the original statement I believe people are dismissing dps numbers as worthless if they are below a certain time threshold. I don't believe that good at all. Wrestlers compete based on weight class. I think the dps numbers are valuable if you recognize the time frame represent different "weight classes" for measuring dps. While I was under a minute of dpsing Patchwerk my dps read 8000 dps but quickly fell as criticals began to normalize and temporary buffs began to drop. I completely agree we should categorize the dps based on time frame.

Your analogy still doesn't clarify what you're getting at to me, because damage over time, aka DPS, is not analogous to miles/hour over time. Damage over time is equivalent to miles over time, aka miles/hour. So your analogy is really more appropriate if you were saying "Person 1 does 4k dps for 10 seconds, person 2 does 4k dps for 10 minutes, and person 3 does 4k dps for 10 days." But that's not what you seem to be saying. Your first post simply referred to amount of damage done in a certain amount of time, which, as Foxx pointed out, is simply DPS.

Honestly, most of these problems just demonstrate why WWS logs and anecdotes have never been a good indicator of DPS, and why it's always better to rely on TCing and spreadsheeting (once you work on reliable test sheets/simulators/methods, which are still in the process of being created in the DK community), even though those will generally overpredict actual performance as they assume perfect variables. Even beyond the issues pointed out by many of these posts, such as proportion of the fight spent trinketed/bloodlusted/etc., roughly half of the data points in the first post note some sort of issue with the fight - new spec, screwed up a gargoyle pop, forgot to flask, etc. etc. When you combine those with different buffs, different skill levels, different amounts of gear, and RNG from single fight dps results, those numbers are pretty useless as comparisons even before you get into how long the kill took.

I shouldn't have used miles/hour, but I didn't want put theorical dps numbers in a thread based on recording actual data. What I was trying to reach and I still stand by is that the context of dps numbers can tell us quite a bit. I see recording 2 minute dps fights and 3 minutes dps fights as an argument over the value of burst dps vs sustained dps. I understand that some builds may gain more benefit out of bloodlust than others. I don't value either higher than the other and I think both are valid measurements.

At a very minium, I would think this data would be useful for either challenging or validating theories and spreadsheets. I know what I am looking for in this data is recognizing trends. It doesn't have to be perfect, but after a while the data should average out and a more complete picture should arrive.

Ah ok, I think I understand what you're getting at now.

I definitely agree that short patchwerk fights are not to be dismissed quickly. However if we are really gonna accept them we might as well take up dummy fights as well, since you're dealing with a similar problem.
Dummies are no good because they inflate certain damage types due to lack of buffs. So is blood on a dummy often subpar because of its reliance on armor debuffs (relative to the other 2 specs).
Patchwerk 2 minute is no different, certain abilities / specs are hugely inflated by the high uptime of trinkets / procs / bloodlust / AotD / Gargoyle. And the short fight does no good to RNG either, crits / KM procs are all out of balance on such short fights.

I somewhat agree that maybe we should divide records in certain classes like your wrestling analogy suggested. However, where do we draw a line ?
What is considered inflated and what isn't. I personally find that very hard.

The reason that I personally mind, is that these kind of 2 minute fight comparisons give out false information to a lot of people. I've seen heroic geared people wondering why they cant reach 7k dps on their 10 man patchwerk kill.

--------

Also we should stop this discussion in this thread. Maybe we can make a discussion topic about this Top DPS Raid thread, so that we can keep all WWS / WMO / Recount reports in this thread and all discussions in the other ?
Not sure if someone could create such a thread and a kind moderator could take the time to move the discussion posts over there.

Alright updated WWS

http://wowwebstats.com/d5pxpcymmiit3...441&a=x1c2061c

Patchwerk, not including army. Includes ghoul I believe.

Spec = 33/38

Used army after IT - PS - BS

Did they fix WWS with AotD and perm-ghouls? Your ghoul damage now says "Limbdrinker & co.", and looking at the contribution I'd say that is accounting for the AotD ghouls. That would be awesome.

Got a few upgrades and didn't screw up on my rotation midway into the fight this time.

Patchwerk
Combat time - 1:52
Weapon - [Betrayer of Humanity] w/ Fallen Crusader
Sigil - [Sigil of Awareness]
17/0/54 Unholy 2h

WoW Meter
DPS: 6058
Act. DPS: 6388

Army of the Dead and [Insane Strength Potion] prior to the pull. Forgot to pot a second time during the fight. Still no deck or Loatheb's Shadow.

Keep in mind that this Patchwerk is kill is fast. To be fair, I honestly think 2H Frost is more optimal for slower Patchwerk kills where Gargoyle / DRW / Hysteria doesn't have as much uptime due to the length of the fight. Of course, raid debuffs (Ebon Plaguebringer) take priority on what I decide to spec. It's nice having alternatives though!

http://img142.imageshack.us/img142/8...9205415wd1.jpg

Spec: 21/50/0

WoW Heroes - World of Warcraft PvE character info & ratings <--- Wow-Heroes profile

I don't have a WWS for it, but here's a screenshot of my Patch DPS tonight.

Weapon - [The Jawbone] (With FC)
Sigil - Frozen Conscience

Army of the dead about at 90% health.

Rotation: BS - BS - OB - OB - FS dump - ITx6 - FS dump, Repeat

My previous screenshot showed 6.7k dps from the warlock and 6k dps from me, on patchwerk. The WWS just hasn't been loading for me until yesterday ><. The WWS, i do not believe, included AoTD ghouls or the warlock's doomguard.

My 2 cents to the thread:
Having played 0/32/39 and 21/50/0 post patch, I can say that DW is a VERY bursty specc. it peaks at 12k ~ 15k DPS when gargoyle and army of undead are out, and it goes down to 3 ~ 4k dps after 2 minutes of the fight.

21/50/0 is a STEADY build. no burst, it is just the same damage over the time.

So, in any PW kill, any build with Gargoyle will obviously TOP any other build (as we can see in the numbers) but in fights with more than 2 minutes, the other speccs even out.

17/0/54 has lost a little of it's strength, but it is still TOP dps for PW, because or AotD and Gargoyle, even though i think unholy only reaches it's desired DPS with 4 pieces of T7.

I may be all wrong, but seeing the numbers, that's what i feel.