Fellwraith

I still think the ideal weapon set-up for prot dps is slow MH, fast OH. You don't whirlwind enough to make the slow OH worthwhile and you don't have the same flurry charge concerns that a fury warrior has. You want to have enough rage to fuel an attack every global cooldown. In theory it should equal out between a fast or a slow weapon before you get to 20%. However, having 2 very slow weapons makes your rage gen very lumpy and reduces your execute efficiency (and probably the dps as well since you won't be executing on every cooldown).

Back of the envelop comparison:
Crit and hit are a constant in this example, I'd assume you'll be wearing the same gear regardless of whether or not your dump your rage into an execute. Your target will also have the same armor value regardless of what attack you choose to use.

Execute benefits from focused rage and 1H spec if I'm recalling correctly. Before armor, the base damage of execute is 1,018 with 1H spec (925*1.1) for 12 rage and 23 damage (21*1.1) per 1 rage thereafter. At 12 rage your execute does 85 DPR, at 100 rage you get 30.5 DPR. The base damage range is 1,018-3,050 for one cooldown. Realistically it will be a much lower number as the damage gets affected by the mob's AC.

Devastate does 35*5*1.1+(MH*0.5*1.1) for 9-15 rage. With 2500 AP and a decapitator (we'll use that for the sake of simplicity) devastate will do an average of 560 damage before armor. The most devastate will cost is 15 rage, which means the worst DPR you get in this example is 37, the best DPR is 62 when devastate only costs 9 rage. Even at 3500 AP with something like Siphon of the Nethrezim, you're only looking at maybe 75 DPR per devastate, which puts it in league with a 15 rage execute in terms of rage to damage efficiency (and you're only getting 676 damage on the devastate, which is far less for that cooldown than what you'd get out of execute).

This is a look at rage efficiency, not dps. I'd say a prot warrior should always be executing after 20% because it does considerably more damage per cooldown and you're going to want to be in the "sweetspot" of 12-20 rage as much as possible. In addition, if you're rage-starved to the point where you're looking at efficiency, you should be executing because it doesn't start getting worse returns than devastate until you're over 17-20 rage (the number will vary based on your AP and MH weapon). There may be special cases where devastating 5 times is a better call than a 50 rage execute, but if you dump your rage into the execute, then do several more efficient 15 rage executes after a couple autoattacks, that should make up for it.

 

rayijin

Yes, I need to get some good fast weapons for execute range. On the few fights I get to DPS on, or trash pulls, I find rage generation "smoothness" to rarely, if ever, be an issue.

Did gorefiend again last night and managed to hit 1260 without recklessness or my chestpiece up, what made the difference was watching my white swings closely to devastate then execute to be as efficient with it as possible. However, your math on execute is correct, but I still don't see a fast OH beating a slow one pre-20%, unless it has a significant DPS or stat edge (maybe a swiftsteel bludgeon?).

Wow Web Stats

 

Gruntle

I was testing some stuff in the spreadsheet (test version/9.1.6), using a fast MH weapon and comparing it to a slow one (cf. Akil'zon Talonblade and Fury, all other gear was the same). The output dps is remarkably similar which made me a bit confused, since the accepted consensus is that slow weapons are better (WW and flurry uptime). It seems that the rage generation with the fast weapon is higher than with a slow, giving a larger number of available HS hits in the sheet.

Looking in detail at it, it seems that it's UBW that's the culprit. The rage from that talent is still modeled as 40% chance per hit regardless of weapon speed, but this was changed in patch 2.0.10. From patch notes:
Has anyone actually done some testing to find out the exact normalization? Is it completely independent of speed?

I guess this error is still in the heavily scripted beta versions, not sure how to correct it until someone actually do some testing. Can you use Procmeter/Procwatch to check for UBW procs?

 

dr_AllCOM3

Also, Execute doesn't take into account a more steady rage generation. It's always on the same cooldown.

 

Wug

Yeah, it doesn't look like the doctor wants to do it, which seems fair enough. He's put more than his fair share into this project already.

I do think it would great if someone did, though. There has been a lot of discussion in the "Future of a DPS Warrior" thread about specs like 3/44/14, 17/41/3, even 0/31/30 to some extent. As a GM that relies on DPS warriors to offtank on fights like Mag, I have a keen interest on which of those specs to recommend, as well as how to gear for it. As for learning how to pwn dps as a tank, myself...well, who wouldn't be interested in that:-)

 

Olon97

3/44/14 is quite easy with the existing spreadsheet. In the custom area, simply put in 24 expertise rating, which adds the 6 expertise from defiance into the calculations.

31/30 you can also fudge by using 0/31/0 and looking at base DPS for mainhand. Add enough custom AP so that the mainhand DPS is 10% larger than the original value, and then the only thing to tweak beyond that is to tack on another 10% to the new execute DPS. Cycle DPS will take care of itself. Offhand dps may be a bit off if your OH has a significantly different DPS value than your mainhand, but it should be a small error at most.

 

Rezarel

I did a lot of testing way back on the 2.2 PTR and found 5/5 UBW to follow a proc per minute mechanic, with about 15 ppm. This held for weapon speeds between 1.3 and 3.7.

Lower ranks scaled linearly - 4/5 was about 12 ppm, etc.

 

Maddin

Can anyone upload the Spreadsheet again, because the link is broken.

 

dr_AllCOM3

Can anyone fix his computer, because the link works.

 

zaon

RapidShare: 1-Click Webhosting

I did modify the sheet for a 1WW/6Devastate rotation as well as a modified flurry calculation. There is a circular reference error in there and I am too lazy to fix it right now. But if you want to get an idea for how much dps you can do as prot go ahead. I usually load the 13/5/43 build in the talent sheet, cancel the circular reference box if/when Excel loads it and either toggle circular references on the Buffed DPS worksheet to FALSE or just click one of the "Save" or "Save Set" buttons to get it to update the dps.

The important protection talents are in the bottom of column B. I did not implement the expertise from defiance.

Typically 13/5/43 build is about 15-20% behind a standard 17/44 build and generates around 5-10% more aggro.

If you can't get my modified version of the sheet to work, sorry I am not providing any support

 

Natural

Teron Gorefiend last week with me 13/5/43 for 1556 DPS: Wow Web Stats

With 17/44 1800-1900 is relatively standard, so I would say that the 15-20% estimate posted above is fairly accurate.

 

Zalein

Has anyone been able to use this properly with Office 2007? If not, are there any plans to make it compatible? Sorry if this information has been posted previously, couldn't find anything in the last few pages.

 

Gruntle

Just wondering, how did you modify the flurry modeling? Was there anything wrong with the implementation in the test version sheet (that is the version I'm using, since it has better modeling of HS usage, and I can't get the new Beta version to work in OpenOffice)? The current Flurry modeling is based on the post by Caesar some pages back in this thread.

Also, saying "the important protection talents" but not implementing the Defiance expertise is like modeling Fury without support for Precision talent, why not include it? Maybe it's not that important for a 13/5/43 build but for looking at e.g. 3/44/14 it's quite important. Or did you just not want to include stuff that's not in live currently?

 

dr_AllCOM3

1.0beta4
I did something wrong here. Is it horribly slow for you with Circular References enabled? It is for me in Excel 2007. I just added some trinkets and moved cells around.

 

zaon

I changed it from a circular calcuation to a 5 celled incremental calculation.

It is the same model just without using circular references.

I didn't include because at the time I made the modifications it was assumed that this was the way it was supposed to work. The blue post about it being a bug and fixed in the next patch wasn't until recently.

46/15 or 47/14 is a stronger build dps-wise than 3/44/14.

 

Halle

Hi all, long time reader - first time poster.

Just like to add that the spreadsheet doesn't work at all for me in Excel 2007.

There is a formatting error on initial load and too many other errors to list during operation. The first of which is it doesnt seem to like saving talents.

[edit] Macros are enabled btw. Maybe I'm missing something else?

 

Gruntle

Sounds cool, the circular references for Flurry have always been a bit messy in the sheet, would be good to be able to do away with them.

True, it was just an example of builds that use the Defiance talent as a dps talent.

 

Crometheis

Hi all, quick quetion about a weapon choice. I am raiding as a 20/41 2h slam build right now (I rememeber reading on the wow europe forums that once you reach 3k ap raid buffed that this build outperforms 33/28). My question is since I no longer go deep enough into arms to get a weapon type spec (mace/swords/axe), does it matter what Season 2 weapon I grab? Ideally I would like to get the mace for arena, and since I no longer get the benefits of spec, and since all 3 season 2 weapons are basically the same, top end and stats, wont they all perform the same with my build?
Crom

 

Buanna

Unless you have a talent spec or ractial bonus to a weapon, weapon types are completely irrelevant.

 

Chirality

re:#368

2Hfury is just too much work, in practice. Keeping rampage up (@20 rage per 30 seconds when you already have 40-45 rage per 4.3ish seconds to spend already is very significant). I tried it last week, and though dps can spike extremely high it honestly takes too much concentration (compared to 33/28) to maximize your own dps.

You might as well get a sword so you're not locked into 20/41 and can spec 33/28 without losing 7-8% damage from SwordSpec vs. MaceSpec should BloodFrenzy become an attractive option for your raid setup.

 

rayijin

Assuming his raid takes two rogues, a hunter, and has at least one non-pally tank, I think it's safe to say that BF would only NOT be attractive if there was someone else using it.

In which case, wouldn't dual wield fury be best?

 

Crometheis

Thank you for your replies. I agree that its alot more work then 33/28. The reason I want to grab a mace over the sword is to help my 3v3 team hit 1850 easily, then i plan on using the points to get the S3 sword, and then yes I will be able to spec back to 33/28 if I chose and still have good damage.
I currently use Gorehowl, I have the points to grab a S2 weapon right now, I just wanted to grab the best one. I figured the mace would be the best to allow me to get my rating up to grab the S3 weapons. So any S2 weapon I get will be better dps then my current weapon, so I am not down grading at all unless I am missing something here.

 

Gruntle

Could you please try to keep discussions on sword vs mace for pvp or to get high arena rating from this thread. In this thread we discuss the dps warrior spreadsheet, i.e. possible modifications to it and bug reports. If you want to know which weapon is the best for you, you should post in another thread (Future of DPS warriors or End-game 2hand comes to mind as two better threads to post these kind of questions in) or just download the spreadsheet and use it to get an answer. If there is something you don't understand about the spreadsheet, don't hesitate to ask in here.

edit: hmm I see that you actually double posted your question in different threads. I would refrain from doing that if I were you, these forums are quite heavily modded.

 

Grim13

### 9.1.6 Modded

PROT DPS!


I saw a lot of people asking for prot DPS on the spreadsheet, and I was kinda interested myself, being a numbers junkie. Since It didn't seem like anyone was gonna have the time to do it, I went ahead and learned enough Excel from reverse engineering dr_Allcom's excellent work to be able to do it. This is the result; hopefully I didn't fuck it up.

Mods:
Prot DPS (Pretty sure I modeled it correctly.)
4x Slam + 2x Bloodthirst + WW cycle added (Either it wasn't there, or I accidentally deleted it...dunno)
Expertise added to SEP
Bloodfury added to Orcs as 35 AP
Heroic drop gems added (I know most of you don't use them, but I'm a scrubby casual raider who isn't in BT. If you don't want them there, delete them)

I know none of you know me, as I've only been a lurker here, this being my first post. I'm a casual PvE'er, though I've been serious from time to time, especially in PvP. I'm a total numbers whore, having made my own shitty spreadsheet previously. Once I saw this spreadsheet though, I immediately abandoned my project. I knew the math already, but didn't know excel. I've learned more about Excel from this spreadsheet than I thought was possible.

Anyway, enough of the "Hello, my name is..." shit. I hope people find my contribution useful, and if it is, dr_Allcom, in your next release, feel free to incorporate any of it you wish. Also feel free to let me know if I missed anything or if there is any other stuff that I can be of use for. I'm not nearly as pro with this shit, but I have a ton of down-time at work, and don't mind working on this stuff.

Edit:
I was in a hurry when I posted this; I had free beer calling my name. As such, I ended up using the first shitty free file hosting I came across, so hit the "Skip This Ad" button is in the upper right corner when it fades in. If anyone has a better suggestion for hosting, hollar.

P.S. I wanted to say thanks to dr_Allcom and everyone else who has contributed to this project.

 

seasonlight

Does anyone know for sure, in hyjal/bt. what would produce more dps.
Leguards of endless rage or leggings of devine retribution? Im so lost as to which to use!!!