Frost DPS is... wow.

[Prexie]

BiS Frost DW/2H is anywhere from 2-5k under BiS Unholy on live (check WoL rankings). I way outgear our frost DK but while his dps has risen since patch, it's not close to the boost I've seen as unholy. I'm pushing 20k on fights like Saurfang/Fester/Rot HM. Our other unholy is either right above me or below me depending on the fight. A lot of it comes down to effective pet and cooldown usage for unholy while frost has seemingly lost its niche as the proc-watching spec (compared to 3.3.3).

AoE dps on trash isn't really a strong measure of dps capabilities of a class. A caveat to that could be improved dps on Heroic Halion.

HM25 Saurfang: http://worldoflogs.com/reports/sklsl...?s=2087&e=2247
HM25 Rotface: http://worldoflogs.com/reports/sklsl...?s=2810&e=2981

[uglybbtoo]

Yeah prexie I am certain unholy will out DPS frost in raid absolutely no arguement.

However as you already noted I am in fairly average gear and yet I can out dps even highly highly geared toons in heroics because the burst dps off frost is massive. With frost the rotation is also very simple where it's alot more complex with unholy.

So if you like the groups frost is extremely attractive too is those ungeared and levelling, tanks who only dps as a second spec (hence why its on this site) and those who want to have some absolute laughs in heroics.

It actually makes heroics fun again .. here is a log for H-DtK from a week ago to give you an idea ... I have improved my DPS even more since then.
http://www.worldoflogs.com/reports/l...um/damageDone/

[penademuerte]

So am i understanding correctly then that you should reforge crit to be at hit and exp caps and then reforge crit to mastery.....what i am not understanding....is crit not that important for frost dk's now?? What is the break point that you do not want to go below on your crit in order to get those (exp, hit, and mastery) up??

[uglybbtoo]

Thats my work in progress atm I will post some more detailed numbers when I get time to run analysis tonight and hence why I started logging my runs.

The original work says:
Hit (until 8% ) > Expertise (until 200 rating) > Strength > Mastery > Haste > Crit > Agility > Spell Hit

EJ's latest stats tables shows:
Hit 3.37 Expertise 3.27 Strength 3.11 Mastery 3.14 Haste 3.23 Crit 2.30 Agility 1.43 Spell Hit 0.91 Weapon DPS 11.25 Weapon Speed 27.27

Notably there values are for a BIS setup and I am not sure how they are getting there number (I have a compaint with EJ stuff there is some vague reference to how they got them I am used to work like theck et al where everything is open) http://elitistjerks.com/f72/t105701-...ysm_countdown/ and I am assuming they are raid buffed and ICC.

I am coming up with very very diff numbers .. here is what I have noticed so far

There appears to be a some sort of haste cap I am at 689 haste on current gear and I am finding haste value is diminishing rapidly ... yet to do reading on why. I am running in unholy presence so need to look up stuff on this.

Mastery is much more important than any other stat as it increases frost damage as a flat percentage. For example 20 strength produces very little increase in damage 20 mastery produces massive increase in damage so I am yet to fathom how they can have mastery and strength as equal in value.

This may be a distortion of the sorts of fight but weapon speed is of little value. On values I am seeing only its damage matters because it is woven into frost strike http://www.wowhead.com/spell=49143. Our other main attack howling blast http://www.wowhead.com/spell=49184 simply uses AP in its calc. Most of the time my actual melee damage is tiny so not sure how they get weapon speed as important.

So from where I am ATM what I am seeing is

Hit (until cap) > Expertise (cap) > Mastery > Strength >>> Crit (if haste is over 600ish) > Haste

I should get a chance on weekend hopefully to fully publish numbers and logs etc so you can see how I derived the numbers.

[Prexie]

EJ's latest stats tables shows:
Hit 3.37 Expertise 3.27 Strength 3.11 Mastery 3.14 Haste 3.23 Crit 2.30 Agility 1.43 Spell Hit 0.91 Weapon DPS 11.25 Weapon Speed 27.27

Notably there values are for a BIS setup and I am not sure how they are getting there number (I have a compaint with EJ stuff there is some vague reference to how they got them I am used to work like theck et al where everything is open) http://elitistjerks.com/f72/t105701-...ysm_countdown/ and I am assuming they are raid buffed and ICC.

I am coming up with very very diff numbers .. here is what I have noticed so far

There appears to be a some sort of haste cap I am at 689 haste on current gear and I am finding haste value is diminishing rapidly ... yet to do reading on why. I am running in unholy presence so need to look up stuff on this.

Mastery is much more important than any other stat as it increases frost damage as a flat percentage. For example 20 strength produces very little increase in damage 20 mastery produces massive increase in damage so I am yet to fathom how they can have mastery and strength as equal in value.

This may be a distortion of the sorts of fight but weapon speed is of little value. On values I am seeing only its damage matters because it is woven into frost strike http://www.wowhead.com/spell=49143. Our other main attack howling blast http://www.wowhead.com/spell=49184 simply uses AP in its calc. Most of the time my actual melee damage is tiny so not sure how they get weapon speed as important.

So from where I am ATM what I am seeing is

Hit (until cap) > Expertise (cap) > Mastery > Strength >>> Crit (if haste is over 600ish) > Haste

I should get a chance on weekend hopefully to fully publish numbers and logs etc so you can see how I derived the numbers.

The guys on EJ are running Kahories' sim in BiS gear, not just glancing at values from a raid log.

Obliterate (your other main damage attack besides Frost Strike) doesn't deal any frost damage so mastery doesn't benefit it. You only utilize howling blast on a single target fight when you initially apply frost fever or proc rime.

Either way, in live there are no mastery gems and you cannot reforge to strength so their relative itemization values to each other are kinda moot.

The stat priority differences are explained fairly well in Consider's main post. What has you confused about it?

[uglybbtoo]

Cause just reading it ATM it has been reported to them in there own discussion people noticed it:

QUOTE: "with the priority changed from mastery>haste to haste>mastey as a 2h Frost, I wanted to ask weather there is some sort of softcap for haste. With all crit or haste reforged to mastery, I still have about 650 Haste. When i changed the priority to haste>mastery, it lowered my Dummy-DPS about 1.5k... is this just a matter of training or did I reach some sort of softcap?"

Yet it's sort of ignored in there conclusions

QUOTE:
"Haste is ridiculously strong these days, now that it affects the rate at which runes regenerate. There is a softcap to the stat, but the exact percent amount has yet to be determined (and would vary fight by fight, anyways, pending such factors as AMS soaking, Dark Simulacrum use, and so forth). Rough math would say you only start to get in real trouble when you are heading north of 30%, but it hardly matters; even past the soft-cap, haste rating is still above every other secondary stat except crit."


And I am programmer by trade and build "sims" as you call them and to create the "sim" you need data. I am parsing the data thru a full analysis process as they would have done not merely glancing at logs as you suggest.

There are a few comments that there sim is out from real worl and raid results.

Thats all I am saying is the result is out pure and simple will get to the bottom of it when I have time.

EDIT: I found this an interesting comment "You only utilize howling blast on a single target fight when you initially apply frost fever or proc rime" if you looked at the numbers I am seeing you might want to reconsider that. Again at BiS maybee the numbers reverse and given I don't have access to a toon with BiS I will leave it alone for now.

EDIT: I should also say this is probably getting outside the scope of tankspot discussion being as its a DPS discussion .. Kaz and you moderators please advise?

Assuming we aren't going on I would sort of suggest if you are a Frost DPS keep an eye on EJ's frost forum because there current sim numbers are slightly out and I suspect it is capable of preforming slightly better than they are showing with stat adjustments.

[Rampart]

EDIT: I should also say this is probably getting outside the scope of tankspot discussion being as its a DPS discussion .. Kaz and you moderators please advise?

Assuming we aren't going on I would sort of suggest if you are a Frost DPS keep an eye on EJ's frost forum because there current sim numbers are slightly out and I suspect it is capable of preforming slightly better than they are showing with stat adjustments.

This is a dps forum, so I don't know why there'd be a problem.

I do know that Frost does insane damage on AoE trash packs. Tough as nails to maintain threat on everything when there is a Frost DK, though still doable with tab-targetting and such. But there single target dps is definitely not as high as Unholy's, which is where it really matters due to bosses.

[Prexie]

Cause just reading it ATM it has been reported to them in there own discussion people noticed it:

QUOTE: "with the priority changed from mastery>haste to haste>mastey as a 2h Frost, I wanted to ask weather there is some sort of softcap for haste. With all crit or haste reforged to mastery, I still have about 650 Haste. When i changed the priority to haste>mastery, it lowered my Dummy-DPS about 1.5k... is this just a matter of training or did I reach some sort of softcap?"

First, this post received an infraction. Second, he states that he tested on a dummy (inaccurate measure of applicable DPS) and posted no logs, sim data or support for his anecdotes of drastic dps decreases. If there's a question as to why this hurts his post's chance of receiving any sort of serious response, read this: http://considerit1.wordpress.com/201...ght/#more-1449

The people posting about Kahorie's sim giving them greater or lesser EP values for crit and haste depending on whether or not the ICC buff is present is more relevant to the discussion, IMO.

EDIT: I found this an interesting comment "You only utilize howling blast on a single target fight when you initially apply frost fever or proc rime" if you looked at the numbers I am seeing you might want to reconsider that. Again at BiS maybee the numbers reverse and given I don't have access to a toon with BiS I will leave it alone for now.

The screenshots (a recount meter is not a combat log) you posted are not showing individual strike values but rather total damage done from those abilities over the course of a heroic. A 17 second heroic HoS fight is not a large enough sample size to make determining factors on what is a higher damage per rune/gcd investment.

If you're looking at DW Frost, make sure you're combining Obliterate's MH and OH hits when comparing to howling blast. There seems to be a differing opinion on "rune availability for OB vs. blowing F/D runes on HB in the meantime" but our frost DK's logs show average hit and crit for MH/OH being higher on Obliterate than on Howling Blast. This should be the case for every single-target raid boss.
http://worldoflogs.com/reports/sklsl...?s=2087&e=2247

For 2Her, the comparison should be easy enough. It could be possible that your 2Her is a very low ilvl item and your howling blast could potentially do more damage than a 2 disease OB. It would be interesting to see a log where that's the case. I just can't see it happening with a ToC or better weapon, though.

EDIT: I should also say this is probably getting outside the scope of tankspot discussion being as its a DPS discussion .. Kaz and you moderators please advise?

Assuming we aren't going on I would sort of suggest if you are a Frost DPS keep an eye on EJ's frost forum because there current sim numbers are slightly out and I suspect it is capable of preforming slightly better than they are showing with stat adjustments.

As stated above, this is in the DPS discussion section of Tankspot so it's fair game.

[uglybbtoo]

Okay bear with me because I am not the guru on DPS I am much more comfortable around tanks and tank sims. I have organize a run thru ICC on saturday so I should have alot better data which I will publish.

However I will ask you a generic question if I may because you are obviously a DPS'er which I sort of picked up on your answer for single target and I am NOT implying you are right or wrong so please bear with me I am trying to understand where to extend my sim work. Why are they loading up on mastery which only increases frost damage if on single target they only use HB to keep frost fever on and frost strike? At BiS does frost fever become really significant or does frost strike damage grow hugely?

As you say I have 15-30 sec real life combat logs or long periods on heroic dummy to go off .. which i readily acknowledge.

Second one is why are they ignoring the haste cap effect or does the actual melee damage come up to compensate for the drop off?

[Prexie]

This shows the secondary stat scaling for BiS DW Frost currently, http://considerit1.files.wordpress.c...tdwscaling.png

Values could be different for 2Her but I'm not positive on that and have not personally run my own simulations to chart relative values. I only have Consider's values but considering I can't think of a time where my coefficient tests on Beta or Kahorie's sims have run counter to his data. You'll notice that strength and mastery behave in a linear fashion while haste has minor plateaus and such, eventually scaling equal or worse to crit around 800 rating.

I didn't say JUST when FF needs refreshing. Rime procs are used for HB as well.

Mastery affects your Frost Strike, Howling Blast and Frost Fever damage (Icy Touch as well but that's rarely used by frost post 4.0.1). Frost DPS owes very little of their total damage output to the actual disease damage (FF accounts for ~4-5% of our DWer's dps). The major thing is the significance of having both diseases up to maximize obliterate damage. Mastery is currently scaling better than haste and crit for DW while 2H is seeing a better result with haste over mastery due to haste's interaction with unholy presence and more RP gains from MotFW procs.

Obliterate and Frost Strike are Frost's highest portion of damage done on single target fights. They'll trade places depending on mastery, but they're pretty even for both specs currently.

Side note: I don't take offense to differing opinions or anything of that sort so don't worry if you're implying whether I'm right or wrong . It's just a process with testing and hypotheses. If I'm making a claim where it's speculation rather than a statistically significant statement or postulate, I try to make sure that's clear. I don't believe anyone here is trying to be malicious, but rather just gather the correct information.

Dummies serve a good purpose for learning new rotations, checking spell coefficients, and things of that nature. It's just good practice to avoid using them to make theories about raid dps, relative class differences and such which is a problem commonly seen in trade chat, official forums and random posts on EJ that eventually end up in the dung heap.

[uglybbtoo]

If you read the all the above posts I am not seriously recommending anything for raid DPS'ers my data is not scaled up high enough or accurate enough at the moment for that.

At the moment all my data is good for is setting up histrionics thru heroics. So for peeps who don't have access to BiS gear because of raid situations as is the case with my DK alt the data I am gathering may be of some use.

As for the actual data my 1H data looks almost identical to the graphs you show but my 2H data has a marked plateau at 600 haste and it varies between Frost and Unholy presence ... I will get my web server back online on weekend and post some data an links.

At the moment having some fun and fine tuning stuff and its working basically lifted my heroic dummy DPS to 7.8K which would be VERY VERY VERY low by your standards but for my crap gear it's not bad :-) http://www.worldoflogs.com/reports/0...um/damageDone/

It is however getting a little rough for most tanks in heroics because I can get to that level very fast. As we aren't raiding now till cata I have organized to do a few ICC runs and will see where this all goes I am uncertain if my interest will keep up.

Thanks for all your help been most insightful and very very helpful as I said I really have never done DPS for a long long time.

[Sterbefall]

You're skewing your HB numbers by using dummies. If you look at what the HBs hit, you'll see that you're hitting a second dummy. Your DPS could very well be higher than Oblit, but it's misleading because your single target DPS is lowered.

[Krenian]

I wouldn't be worried about discussing DPS stuff in the DPS section. It is, what the forum's title is, being used for what is meant to be used: Discuss DPS.

Keep trudging.

[uglybbtoo]

You're skewing your HB numbers by using dummies. If you look at what the HBs hit, you'll see that you're hitting a second dummy. Your DPS could very well be higher than Oblit, but it's misleading because your single target DPS is lowered.

Yeah it's hard to get an angle the HB doesn't hit another on heroics but my parser filters all that out and you can sort of filter it on WOTL just choose the target actor

Goto here http://www.worldoflogs.com/reports/0...upb/details/3/
Now select damage by actor .. choose the heroic dummy and you will end here
http://www.worldoflogs.com/reports/0...upb/details/6/

You will see it doesn't change HB average damage per hit 6461.7 .. obliterate 6162.8
The crits come up very clsoe because of the glyph etc for obliterate
HB crit average 13856.8 obliterate average 13835.5

EDIT: Again I should say this only applies to my current gear level for those more geared stick with EJ's because I have no numbers on HB versus Obliterate at those levels and the scalings may take Obliterate past HB. However if you are in lower gear like me it may be worth looking on heroic dummy to see if HB or obiterate does more damage.

I have raid data from heroics bosses and VOA 25 I did saturday which shows the same but looking forward to this afternoons ICC run to get some real data. My VOA damage for torovan was 10348 dps which I was very happy with even though I lost rotation a few times.

[Sterbefall]

A couple things. First, I'm pretty sure WoL doesn't properly show ToT (making Obliterate look much weaker than it is). Second, Rime means that Obliterate gives you the damage of Obliterate + 45% of the damage of a HB over the long haul. It is not better to use HB over Obliterate.

HB is superior for multiple targets, Obliterate is better for single.

[uglybbtoo]

Two problems which I have dealt with above:

First: WoL is doing the same as I am parsing the entries in the combat the log so unless blizzard doesn't record the sources correctly that simply doesn't add up.

Secondly: Practical logs both on heroic dummy and actual runs were showing higher dps using HB.


Now and most importantly I say were because I very recently changed weapons (http://www.wowarmory.com/character-s...ng&cn=Ubbdk) I had enough honour points to buy the PVP 2H sword and initial testing with it is showing very different results and I am see why EJ's gets different numbers now http://www.wowhead.com/spell=49020 is based alot on weapon damage.


So I feel comfortable in saying that it's still a test thing that DPS should do. In BiS gear with a big arse weapon yes Obliterate is better to use than HB but it is not always the case in lesser gear. The reason is simply without really high weapon damage it wont scale like HB does.

EDIT: I guess what people need to learn most from all this is data is always skewed and not always tested across full ranges even from very reputable sources like EJ's. It also means advice is only valid for the range it was tested in a point I have conceeded all the way thru with my data which was showing much different to EJ's.

EDIT: My concern for DPS community is there seems to be a blind faith in EJ's and don't get me wrong there work is first class BUT the data isn't opened up for scrutiny and range limits of testing isn't given. What I have found in this journey is alot of advice is been dished out on ranges of gear that haven't been tested and if you aren't geared towards BiS you need to take all advice understanding that.

EDIT: In no way is this an attack on EJ's I feel they do a fanatastic job unfortunately they also spawn some blind zealots.

[Sterbefall]

The combat log shows Obliterates when DWing as two separate entries. You're now using a 2H weapon so this doesn't happen as you don't use ToT.

You've shown one log and that one had you hitting a second enemy with HB. Your actual runs are also mostly trash. Currently, HB gets priority in multiple target situations. For strictly single target, Obliterate is better.

I still don't buy the idea that HB is a better choice than Obliterate in single target situations as you still haven't addressed my above point about Rime modifying the value of Obliterate assuming the numbers listed in your log were correct (which I'm still not convinced of because my own WoL reports using near BiS DW weapons show HB average doing more damage than Obliterate which I know isn't true for ST).

[uglybbtoo]

It's relatively simple to understand and infact I was kicking myself for being so stupid.

Howling blast (http://www.wowhead.com/spell=49184) uses only AP in it's calc
Obliterate (http://www.wowhead.com/spell=49020) uses only weapon damage

What happens as you are gearing up is it's easier to get top end armor gear so you AP rapidly increases and you are reforging frost damage so it scales very fast. Weapons on the other hand you really don't get access to until running ICC or like in my case get the PVP one.

It's only when you get access to better weapons that the numbers start to reverse and I imagine by really high end it's quite a margin.

Assuming you have around 5500AP which is sort of all the 264/251 badge gear level this is sort of a simplification of what I am seeing.

ilvl 232 weapon (all 232.7DPS) HB will be higher by along way
ilvl 251 weapon (all 267.3 DPS) Very close HB marginally ahead
ilvl 264 weapon (all 294.8 DPS) Obliterate comfortably ahead

Within an iLvl of weapon there is only small shifts in number based on top end/bottom end damage and the stats on weapon so in my defense in my initial work I didn't think about weapon being crucial to the effect I was seeing.

[Sterbefall]

I checked and WoL doesn't show Obliterate damage correctly. I had 32k Oblit crits last night and WoL only shows about 20k.

Obliterate is not strictly modified by weapon damage. It is normalized.
http://www.wowpedia.org/Normalization

[Yves]

Is anybody using Frost DPS spec 3/31/2 with 2H weapon?
I am currently using 2x1Handed weapon, doing pretty fine.

But was wondering if 2H was better or at least equal to 2x1Handed.

Had a battered hilt drop, and i am undecided if i should sell it or trade it in for the 2H wep.