I feel frost is going to scale incredably well in the long run. The amount of damage that sigil adds to frost strike is amazing. And that was with a 203 dps weapon, it should be much higher with Earthshaper or Voldrethar.
If you want to exclude AoE Damage, that might conflict with how their guild handles the fight. They might be required to deal with AoE so they can't spend those particular runes on the boss/heart. This is where the problem comes in. You'd have to try and figure a way to exclude the AoE Damage while compensating for the potential damage from those runes that could be done on the boss.
The parse with 7300 was actually including Friendly Fire.
If you want to use XT as a source for this, i recomend making one of the guidelines that you cannot deal more then "x" amount of damage to friendly players.
I feel frost is going to scale incredably well in the long run. The amount of damage that sigil adds to frost strike is amazing. And that was with a 203 dps weapon, it should be much higher with Earthshaper or Voldrethar.
Actually Blood seems to be the best scaling build because ArP scales exponentially and is already very strong at low levels or ArP, unlike all stats but Strength for Unholy and Frost.
If you want to exclude AoE Damage, that might conflict with how their guild handles the fight. They might be required to deal with AoE so they can't spend those particular runes on the boss/heart. This is where the problem comes in. You'd have to try and figure a way to exclude the AoE Damage while compensating for the potential damage from those runes that could be done on the boss.
No, i dont think that those persons did switch on adds. They simply aoed the adds that were tanked near the boss. And thats simply pushing your dps by lets say 1-2k.
The best way to measure DPS, in my opinion, in an no movement fight for melee is Ignis when ignoring the adds and just straight burning him down in attempt at the hard mode.
Should be even easier now to attempt the burn with recent nerfs. The only variable may be slag pot, but that is about it.
I don't really understand why people still think that the only "legit" parses are the ones with the least AoE. Look at how many boss fights in Ulduar that involves add control.
Razorscale: Add control for nearly half the fight
XT-002: Billions of bots
Auriaya: Guardian Swarm
Freya: The entire fight is based on adds, killing the boss is like a secondary objective.
Thorim: LOL @ the amount of adds
Mimiron: Most of phase 3 is add control, granted you're mostly on the assault bots and the ACU but AoE would take pressure off your scrap bot tanks and make transition into phase 4 much smoother.
Yogg-saron: Adds are minimal, but AoE damage is still beneficial.
When more than half of the bosses (Flame Leviathan doesn't count as a boss) in Ulduar requires some form of add control, isn't it kind of silly to benchmark DPS performances based on single targets? If your objective is no longer just "kill the boss and only the boss", why not account for a build's AoE capabilities?
No, i dont think that those persons did switch on adds. They simply aoed the adds that were tanked near the boss. And thats simply pushing your dps by lets say 1-2k.
I'm saying in general, Their guild might require them to AoE the adds. If you want to remove their AoE damage, They are still out the Damage from the AoE that was done to the boss/The damage that could've been done solely to the boss using those runes.
@Lorenzop16: For Friendly Fire, That you can easily remove all of the damage done and find the DPS without it, But Doing regular AoE, Aka Pest/BB/DnD/HB, You have to think of what is also damaging the Boss and if you want to remove all of that damage, what those runes could've done damage wise against the boss.
I don't really understand why people still think that the only "legit" parses are the ones with the least AoE. Look at how many boss fights in Ulduar that involves add control.
Razorscale: Add control for nearly half the fight
XT-002: Billions of bots
Auriaya: Guardian Swarm
Freya: The entire fight is based on adds, killing the boss is like a secondary objective.
Thorim: LOL @ the amount of adds
Mimiron: Most of phase 3 is add control, granted you're mostly on the assault bots and the ACU but AoE would take pressure off your scrap bot tanks and make transition into phase 4 much smoother.
Yogg-saron: Adds are minimal, but AoE damage is still beneficial.
When more than half of the bosses (Flame Leviathan doesn't count as a boss) in Ulduar requires some form of add control, isn't it kind of silly to benchmark DPS performances based on single targets? If your objective is no longer just "kill the boss and only the boss", why not account for a build's AoE capabilities?
Presumably the reason why single-target DPS is being considered more of a clean measure for build superiority, is because with adds there are too many variables that come into play in a raid. Add health, debuffs, positioning.. it's all too random. Adds can come from any direction and your individual DPS on them may be gimped if they come from the opposite end of the room to where you currently are after DPSing the last batch.
Bosses in Ulduar are not overly good for benchmarking DPS at all, but if focusing solely on a boss, the number of variables is significantly reduced. There's still variables, and some parses will have a few adds mixed in, but an overall picture is visible when doing boss analytics with a significantly smaller sample size needed. It can then more easily be said, "51/0/20 performs better overall than 51/2/18 by 5% on average with a small margin of error" or "blood appears to be scaling better than frost". If considering adds, it requires a much larger sample size to get an average without a ridiculous potential for error due to bad luck or positioning, or variable raid comp.
-No- boss in ulduar is a good benchmark for raw "how much dps can this build output on one target" dps, patchwerk is still the best benchmark for this overall despite the relative shortness of the fight, ignis being a relatively decent second but again, variables (slag pot, adds, flame jets) can screw things up there.
If considering fights where there are the most variables (adds, movement, positioning, etc) Freya would be the best benchmark to use for benchmarking DPS capabilities of a build on both single and multiple targets. The fight has every possible problem with it: many adds, knockbacks, silences, constant movement, repositioning, high incoming damage, target switches, "you are the bomb" debuff.. there is no fight I can think of with more variables, so if trying to measure the overall raid dps capabilities of the class across new-style fights, that would be the fight to use - but a very large sample size would be needed to be able to start seeing trends.
I had a long post written up, but it got destroyed by a browser error. I'll sum it up quickly.
1. The AoE capabilities are not too close for the 3 trees (UH and Frost destroy Blood on 3/4+ targets), making it so that if you ever need to AoE, it is unlikely you will consider Blood (and UH/Frost each have a type of AoE niche for themselves).
2. The single target DPS from the 3 trees seems to be comparable, as well, the fights which are a DPS race in Ulduar seem to be single target fights (though I am not 100% sure on this last part).
And where are you going to find a single target DPS fight that doesn't involve outside raid buffs or things that take away from your damage? Do you have a good answer? Because after looking there IS none. That's the issue. This is the best given the type of fights in the instance to gauge your DPS on.
Which brings us back to the real point. We're trying to figure out the best single target dps specc for fights that don't exist ingame.
What is the real point of the thread, finding how our dps speccs perform on content that we are facing at the moment or figuring out the best specc to go back to Patchwerk and epeen some?
If there is 1 or 2 fights over 15 where single target dps matters, then the best dps specc is the aoe one. It's quite simple. We should stop searching for the Patchwerk dps speccs, start parsing all fights and look for the Ulduar dps speccs.
Bosses in Ulduar are not overly good for benchmarking DPS at all, but if focusing solely on a boss, the number of variables is significantly reduced. There's still variables, and some parses will have a few adds mixed in, but an overall picture is visible when doing boss analytics with a significantly smaller sample size needed. It can then more easily be said, "51/0/20 performs better overall than 51/2/18 by 5% on average with a small margin of error" or "blood appears to be scaling better than frost". If considering adds, it requires a much larger sample size to get an average without a ridiculous potential for error due to bad luck or positioning, or variable raid comp.
-No- boss in ulduar is a good benchmark for raw "how much dps can this build output on one target" dps, patchwerk is still the best benchmark for this overall despite the relative shortness of the fight, ignis being a relatively decent second but again, variables (slag pot, adds, flame jets) can screw things up there.
Why is no boss a "good benchmark for dps?" DPS doesn't consist in hitting on one guy and killing him. It consists on hitting on whatever number of guys you can hit on and get them down. Single target dps is a thing of the past. These days friggin' rogues are an aoe class - everything is an aoe class, now.
Someone suggested stopping to watch this topic if you disagree with the premises. Sadly, it's what I'm doing at the moment, but I still think the initiative is worthy if slightly tweaked. I don't see keeping parses from all bosses as a major change in workload for the TO.
Keeping an top 10 for all bosses (and if you wanna make it visual, you can color the parse link green red or blue according to spec) could be the first step, and from there you can work on making averages or evaluate specc representation.
I think that XT-002 is a poor choice for many reasons, many of them have been exposed before.
Moreover, I think the dps meter will scale *very* poorly during the T8 life.
The main reason is that the top dps will be reached when the raid dps is enough to bring the heart just beyond explosion in a time as close as possible from the exposed heart time.
When your raid dps will go upper than that you will have 2 options :
1) stopping dps before the end of the heart exposure, which sounds quite dumb for a boss use as a staple for dpsmeters.
2) breaking the heart and modifying the mode, making the dpsmeter totally inconsistent with other dps estimations because of the huge difference between the 2 modes (and changing the fight duration of course)
The only thing in favor of XT-002 is the hard mode itself. But kills of XT-002 in hard mode will not be plenty before some time.
There are still adds, but nothing that can be aoed.
The movement component is slightly altered but your best dps will always be when you take almost no bomb. It could take time to achieve, but it's no a different random factor than others like crit.
For the time being I really think that Ignis is the closest thing that can be used a dpsmeter, almost no movement, slag pot is not an advantage nor a disadvantage.
This thread needs to be revamped, with more accurate data analysis on bosses that aren't so RNG. You got guys in that list that got the Light Bomb six times, of course they're going to do 6-8k DPS. Come on.
Imo the only boss that should be used as a guage outside of Patchwerk, is Ignis.
This thread needs to be revamped, with more accurate data analysis on bosses that aren't so RNG. You got guys in that list that got the Light Bomb six times, of course they're going to do 6-8k DPS. Come on.
Imo the only boss that should be used as a guage outside of Patchwerk, is Ignis.
Have to agree.
To emphasize, what the hell is the point of including parses in the OP that have obvious errors in them such as the Light Bomb issue?
Why is no boss a "good benchmark for dps?" DPS doesn't consist in hitting on one guy and killing him. It consists on hitting on whatever number of guys you can hit on and get them down. Single target dps is a thing of the past. These days friggin' rogues are an aoe class - everything is an aoe class, now.
I say no boss in Ulduar is a good benchmark because benchmarks are controlled and scientific in nature. Ulduar's bosses all contain too many random factors which you cannot control, such as adds. Freya's one wave of adds have no aggro table and run around to random players, XT's adds may spawn from any corner of the room, etc. The top 10 parses on a fight may be from people in raids using a certain strategy for the fight that allows more AOE on adds by melee due to positioning, thus not showing us any useful trends other than "doing the fight this way makes melee dps higher". It tells us nothing about where certain variant builds in our respective trees do well or don't do well because the parse is tainted by the inconsistent raid strategies used. Whereas on a boss like Patchwerk, there really is only one way you can do the boss which is "stand there and dps him, don't let tanks die".
Keeping an top 10 for all bosses (and if you wanna make it visual, you can color the parse link green red or blue according to spec) could be the first step, and from there you can work on making averages or evaluate specc representation
WMO already does this. While some results are questionable, trends are visible such as Unholy topping the charts on Kologarn or Frost having higher representation on Hodir.
I say no boss in Ulduar is a good benchmark because benchmarks are controlled and scientific in nature. Ulduar's bosses all contain too many random factors which you cannot control, such as adds. Freya's one wave of adds have no aggro table and run around to random players, XT's adds may spawn from any corner of the room, etc. The top 10 parses on a fight may be from people in raids using a certain strategy for the fight that allows more AOE on adds by melee due to positioning, thus not showing us any useful trends other than "doing the fight this way makes melee dps higher". It tells us nothing about where certain variant builds in our respective trees do well or don't do well because the parse is tainted by the inconsistent raid strategies used. Whereas on a boss like Patchwerk, there really is only one way you can do the boss which is "stand there and dps him, don't let tanks die".
WMO already does this. While some results are questionable, trends are visible such as Unholy topping the charts on Kologarn or Frost having higher representation on Hodir.
Your right, All bosses in Uldaur are poor representation of ones single target DPS. However, XT-002 Is the only boss where you have the option of doing uninterrupted DPS(Unless you get Unlucky Gravity Bombs etc...Have to run out). Ignis has 10(I think) seconds of downtime for Slag victims as well as Flame Jets. I'm not sure how other guilds do XT but normally we don't have Melee on Adds at all for the most part, unless they get out of controll. Melee got free reign to just do straight DPS consistantly.
Your right, All bosses in Uldaur are poor representation of ones single target DPS. However, XT-002 Is the only boss where you have the option of doing uninterrupted DPS(Unless you get Unlucky Gravity Bombs etc...Have to run out). Ignis has 10(I think) seconds of downtime for Slag victims as well as Flame Jets. I'm not sure how other guilds do XT but normally we don't have Melee on Adds at all for the most part, unless they get out of controll. Melee got free reign to just do straight DPS consistantly.
I wouldn't exactly call melee DPS "free reign" on that fight given that if we accidently burn the heart too hard XT immediately goes into a second heart phase after coming out of the first one and we end up with twice the amount of adds which can very easily overwhelm the shortage of ranged we have. When Ulduar first came out and XT's enrage timer was still strict we ran the "ranged on adds melee on boss" strat just so we have enough DPS to kill the boss. Nowadays it's more about effectively controlling the adds and not have any scrapbots get through. And people have mentioned this a million times before, Light Bombs artificially increases your single target DPS.
If you want pure uninterrupted single target DPS, Ignis on which ever difficulty is the best bet. Generally speaking, the 150% haste granted after the slag pot in no way makes up for the DPS lost during the slag pot for melee so there's no issue of inflation, and there's really no other aspect of the fight that makes it any different than Patchwerk. If you're standing close enough, not even a Flame Jet can put you out of commission for that 3 seconds where you're in the air. Other than that, the fight is basically Patchwerk 2.0. The most ideal DPS parse for Ulduar bosses would be an Ignis fight with 0 Slag Pots, and if you do get slag potted, note how many times.
@rh8452, I don't really buy into the whole "build superiority means higher DPS on Patchwerk-like bosses" anymore. Build Superiority should account for its ability to adapt to different situations effectively and not excel in tank-and-spank fights while failing the rest. The days of simple fights are over, at least until the next expansion. With boss fight dynamics becoming increasingly complex again, the variables you describe become a new problem to tackle when people are dreaming up new builds. Whether it may be the build's AoE capability, mobility, margin for error, these are all factors to consider when choosing a build. Adds are now a fundamental problem in raids, and if your DPS is going to be gimped compared to others because of it, you need to be able to adapt.
So apparently they nerfed the 4 piece; and my spec, according to some, will suck because of it. But I kicked some ass tonight. I could have done better on a few bosses, razor I was in wrong gear, and a couple others I was in UnP, but other then that I thought I did great.
So really goofing around in ulduar 10 and playing with hard mode stuff, pulled 7063 DPS in 17/51/3 frost. 4pc T7 + frost strike sigil & HB glyph. Did not get light or dark bombs at all.
Big thing here is the fact that it really depends on how much you are going to burn the heart and if you get light bomb. Burn the heart to 5%, then let her go, then burn it down again you get more of a max dps situation. Kinda skews things. I would say general is a better one as well since you can be pretty constant on dps as long as you arent on kick duty. Ignis is fun, but if you get potted, you are kinda screwed there. Hodir can be skewed by having the shammy buff delivered to the melee pack each time. So I'd say the general and XT are the best for now.
The only problem is that we're about to get supernerfed and forced to drop UP with T8 gear. Now I wish they would actually stop screwing with the damn DK class. Changing rotation, points / etc every 6 weeks is getting rather old. They have traditionally forced us to use BP pretty much across the board until the latest patch to us where UP T7.5 allowed us to do some good stuff than the old 0/44/27 or 0/32/39
As much as this list is great for the next DPS count.. Reality is that this will change again when we get forced into T8 and it really depends on what they give us in the set. Right now we're looking at 8%, but the loss of RP from the 7.5 set means less FS per rotation and I'm eeking out every bit i can to ensure i get max FS per set.
So today's top dps will most likely be tomorrows nerfdps and a new thing will arise again (until they switch up our class yet again!)
It just kinda sucks to have a full 4 piece T8.5 and have it sit in the bank doing nothing. I'll go to BP when they nerf my current set, but I would love to use UP and get the same or better DPS with 8.5 to be honest. Has anyone even seen a decent UP T8.5 rotation yet that can hold up to the 4pc T7.5?
Why aren't we using Vezax as the benchmark. Aside from the ones that are interrupting which in my case (unholy build) it usually cost me between 1-2 Deathcoil. If you're not on interrupt duties I believe Vezax is the prime example of a dps benchmark (no movement, no adds). Any thoughts?
I don"t know about your guild, but we move Vezax. No matter, I get your point. I'm not sure if he has a ton of armor or not. Other then that my guess would be the people able to access him currently. There are guilds out there that can't get to him yet, despite all the Ulduar nerfs.
@rh8452, I don't really buy into the whole "build superiority means higher DPS on Patchwerk-like bosses" anymore. Build Superiority should account for its ability to adapt to different situations effectively and not excel in tank-and-spank fights while failing the rest. The days of simple fights are over, at least until the next expansion. With boss fight dynamics becoming increasingly complex again, the variables you describe become a new problem to tackle when people are dreaming up new builds. Whether it may be the build's AoE capability, mobility, margin for error, these are all factors to consider when choosing a build. Adds are now a fundamental problem in raids, and if your DPS is going to be gimped compared to others because of it, you need to be able to adapt.
I agree, and what I mean is that there's no boss fight which works well as a benchmark for any specific build because of the complexity and variability of a fight. At this point, the amount of variables mean that build superiority is going to come down to subjectivity and raid strat/comp, rather than being able to create a build which is scientifically known better than other builds for the fights in Ulduar.
This may end up being a good thing as the days of cookie cutter builds are also likely over, meaning unskilled players won't be able to faceroll 5k dps like they could in naxx.
Thats why you have to reduce the total damage by the damage done to adds. Its still very easy, simply take make
/damage done to boss\
|--------------------------| * (TOTAL DPS) and you get boss dps that cannot be inflated (at leats not so easily).
\ total damage done /
05/02/09, 8:17 AM
Similar Threads
