DPS Warrior Specs
#21
Quote:Thanks for all the work Conc. Let me know if there's anything I can do to assist!

Well I have the DW Fury sheet pretty much done, except my Flurry model is probably very sucky. Legedi's post suggesting a ~90% flurry factor at 30% crit and ignoring dodges is what threw me off... my current model has more like a ~75-80% factor at 30%.

I actually don't know a whole lot about Flurry (never had it) and have a couple questions:
1a) will it eat charges on a white swing that is a miss?
b) dodge or parry?
2a) will it eat charges on a yellow swing (not HS or cleave, something like Hamstring, BT, etc..)that is a hit?
b) miss?
c) dodge or parry?

I'll post what I have by tomorrow or Friday when I finish up an initial sheet so interested people can check for my mistakes.
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#22
Quote:Well I have the DW Fury sheet pretty much done, except my Flurry model is probably very sucky. Legedi's post suggesting a ~90% flurry factor at 30% crit and ignoring dodges is what threw me off... my current model has more like a ~75-80% factor at 30%.

I actually don't know a whole lot about Flurry (never had it) and have a couple questions:
1a ) will it eat charges on a white swing that is a miss?
b ) dodge or parry?
2a ) will it eat charges on a yellow swing (not HS or cleave, something like Hamstring, BT, etc..)that is a hit?
b ) miss?
c ) dodge or parry?

I'll post what I have by tomorrow or Friday when I finish up an initial sheet so interested people can check for my mistakes.

Flurry Mechanics

The only thing that seems off there is I'm 99% sure that "on next attack" abilities use a charge as they are a modified form of a normal swing.

Thus, answers to question 1 are yes, and question 2 are no.

Anything else?
~Frag B)
Hardcore Diablo 1/2/3/4 & Retail/Classic WoW adventurer.
Reply
#23
Quote:Well I have the DW Fury sheet pretty much done, except my Flurry model is probably very sucky. Legedi's post suggesting a ~90% flurry factor at 30% crit and ignoring dodges is what threw me off... my current model has more like a ~75-80% factor at 30%.


That's another reason flurry works so well with a 2H slam build. If you are always using slam after a white swing you are basically getting two attacks that could crit and refresh flurry in one swing timer (well, one sing time +0.5 seconds for slam). Gives you a much better chance of keeping flurry going.

And when you are in full AOE spam mode... oh, you'll have flurry up all the time :lol:.
[Image: 21740hrsxL.png]
Reply
#24
Quote:Well I have the DW Fury sheet pretty much done, except my Flurry model is probably very sucky. Legedi's post suggesting a ~90% flurry factor at 30% crit and ignoring dodges is what threw me off... my current model has more like a ~75-80% factor at 30%.
I've actually done a fair amount of thinking about how to model Flurry. My initial thought was to try and solve it using an infinite series, though I imagine that could get pretty crazy pretty quickly. I'll have to do some more thinking on how to actually come up with a good Flurry model...
-TheDragoon
Reply
#25
As Legedi pointed out, figuring the factor for chance to proc flurry within a flurry is pretty simple...

What I did was come up with a number for events per natural swing (even = white swing or yellow swing)
I already have yellow attacks per second average for overpower calculations, so this is pretty easy.... average time for a swing * yellow events per second.

then
1-(1-crit chance)^(events per natural swing)
gives percent chance for each attack. Looking at the full flurry you do another exponential with 3 swings 1-(1-that number)^3

when you drop out of flurry your time between swings increases, but your average yellow attacks per second doesn't change, so the chance per natural swing increases.

I'm not quite sure what to do with this number once I have it.

The DW case I'm looking at is relatively low gearing, with 17.59% crit chance in battle stance, and against 20% armor is doing 0.2405 yellow attacks per second on average with 2.3 speed MH and 2.1 speed offhand.

crit events per natural swing while flurried is 1 + 0.844 seconds * 0.2405 = 1.203
which is 20.76% chance per natural swing event or 50.25% chance per total flurry (3 swings).

crit events per natural swing while not flurried is 1 + 1.098 seconds * 0.2405 = 1.264 ==>21.69% chance per natural swing event

Total time under flurry in this case will be higher than 50.25%, but I'm not sure how to figure how much higher. I should know, but I'm blanking.

Here is what my worksheet looks like thus far:
http://64.81.59.71/forumpics/Conc%20Warr...20v0.5.xls

This is just the DW Fury version, I plan to use this as a semi template for making sheets for other options, like 2H MS, DW Devastate, 1H SS / Devastate, 2H BT and 2H Slam (sheets based upon the attack with the highest rage priority.

The current flurry chance is an old approximation of a single case. That part will change. Anyone with inclinations check out what mistakes you can find. If you want to figure out what is going into each formula, you can choose Tools --> Auditing --> Trace precedents and it will draw arrows from the other numbers used in the formula.

- User changable variables are tinted green.
- The "Quick Add" columns allow adding effects like gems and such for trying out different scenarios.
- Hamstring rage will adjust the rage for hamstring when under WF and allows determining if something like the ZG necklace or the PvP gloves make sense to use even if they have lower stats because you will have the rage to keep spamstringing. to proc WF
- Rampage value is the AP you want to use as Rampage adding if unused rage reports "Can't Keep Rampage" you won't have enough rage to keep it up (I'm assuming 20 rage per 25 seconds) for 100% of the time unless you use bloodrage (not in the model), or you delay a Bloodthirst.

I think most of the rest of the stuff is straightforward.

Most of the calculations happen over to the right hand side, with the left hand side more the useful output information. I use a 1280x1024 display, so it's most comfortable at that width at 100% zoom.
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#26
Quote:Assuming you are just standing and DPSing a target forever you will slam after everyone regular attack. Chances of flurry ever running out are about .7^6 ~ 10%. So if you just assume flurry is going all the time an attack cycle is (3.5/1.3+.5)=3.2 seconds. Every cycle does (800+940)*.7+(1600+2068)*.3=2318 damage. So 2318/3.2=725 DPS against zero armor. Even if the mob is at 30% reduction that is about 500 DPS. Remember, that is all unbuffed. When you start adding things like blessings and WF totem that number climbs very very fast.

The slam model is even more complex than I thought.

1) reaction time / lag will extend the cycle length.
2) if you do ANYTHING other than slam and it triggers a WF you cannot slam until the next cycle without losing damage, because the earliest you can possibly slam is +2 seconds after the white swing (1.5 seconds GCD + slam time). For now I'm going to model using no other yellow attacks
3) you need 15 rage every cycle, which means you need about 13-14 rage per hit depending on if you have improved berserker rage and assuming you'll get anger management.

Interestingly, it's looking as if improved berserker stance + rampage is more damage than axe spec... Actually it should be that much of a surprise... 5% crit is equivalent to about 110 crit rating, which is equivalent to 220 AP in the itemization formula, and for warriors (especially 2H warriors) AP > crit in the itemization formula. So rampage by itself (if it can be maintained at 5 charges) will usually trump axe spec. Rampage + Imp. Berserker stance combined should always be better than axe spec.

Points in impale look to always be inferior to points in 2H spec, so a spec like this:
http://www.wowhead.com/?talent=pVMcdiZE0zVzgRizVo
should be the ideal slam build for raid DPS... provided you can maintain Rampage.
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#27
Quote:http://www.wowhead.com/?talent=pVMcdiZE0zVzgRizVo
should be the ideal slam build for raid DPS... provided you can maintain Rampage.
Heh, fine I'll respec again.

One nit though, Improved Execute isn't a negotiable talent when speaking of raid dps. From that perspective (as amazingly useful and handy as it is) Imp. Intercept should be dropped for 2/2 Imp. Execute & 1/3 Precision. That's if we're *only* considering max pve dps, however.

Speaking of, I've done two runs on Tiga since going to the build I posted earlier: One Sethekk Halls (At 65) and Crypts (At 66), and I've been if not very competitive for dps, winning. On a pure stand still and use slam every cooldown fight earlier I did 404 dps on an outdoor raidmob earlier with Terrok's Quill.

Thanks for the research, Conc. One thing to note, I've found that if you "feel" out (need a stopwatch mod, really) your swing timer, you can start up Slam .5s before you are going to hit, and both land at the same time with no loss of swing timer / dps. Pulling that off is *very* hard to do consistantly though.

Cheers,
~Frag B)
Hardcore Diablo 1/2/3/4 & Retail/Classic WoW adventurer.
Reply
#28
Quote:Heh, fine I'll respec again.

One nit though, Improved Execute isn't a negotiable talent when speaking of raid dps. From that perspective (as amazingly useful and handy as it is) Imp. Intercept should be dropped for 2/2 Imp. Execute & 1/3 Precision. That's if we're *only* considering max pve dps, however.

Speaking of, I've done two runs on Tiga since going to the build I posted earlier: One Sethekk Halls (At 65) and Crypts (At 66), and I've been if not very competitive for dps, winning. On a pure stand still and use slam every cooldown fight earlier I did 404 dps on an outdoor raidmob earlier with Terrok's Quill.

Thanks for the research, Conc. One thing to note, I've found that if you "feel" out (need a stopwatch mod, really) your swing timer, you can start up Slam .5s before you are going to hit, and both land at the same time with no loss of swing timer / dps. Pulling that off is *very* hard to do consistantly though.

Cheers,
~Frag B)

I don't know if big trouble will work for a warrior swing time or not. Works for hunters. I had to break down and get it when I was using a gun because I just couldn't time the shots right without it and delaying an autoshot because of a badly timed steady does not make for maximal DPS.
---
It's all just zeroes and ones and duct tape in the end.
Reply
#29
Quote:Heh, fine I'll respec again.

I wouldn't respec right away. It takes pretty decent gear to generate enough rage to slam each cycle AND keep rampage up. Rampage is a little slow to build momentum too with a 2H build, so the ideal situation may or may not be better at your gearing
Conc / Concillian -- Vintage player of many games. Deadly leader of the All Pally Team (or was it Death leader?)
Terenas WoW player... while we waited for Diablo III.
And it came... and it went... and I played Hearthstone longer than Diablo III.
Reply
#30
Kind of old thread, but I decided to respec this weekend for some PvP fun. So I wanted to get MS. But at the same time I was intersted in a 2H slam/flurry raiding DPS. So I specced: http://www.wowhead.com/?talent=pVMhdNioM0zZVVzm0gRV. It picks up PvP and PvE talents. Now I only did one Gruul run with this build, so it will take a bit of practice. But with my average gear, and a good weapon I did quite well on damage. We actually had a lot of paladins, and an enhancement shaman. So I did have about the best buff I could get. But I was able to put out over 800 DPS on trash. For Maulgar I had to tank the first mob, then pulled aggro on Kiglar without salv on and died. On Gruul I was putting out 615 DPS with having to deal with shatters, and would have been near the top of DPS if the OT hadn't died and all melee DPS dieing shortly after.

The flurry nerf to 25% is off set by the reduced glancing blows, and actually makes weaving in MS and WW between slams and normal swings easier. I would only lose about 0.5 seconds or less on starting a slam because of global cooldowns with flurry going. Add in the fact that rage generation has been buffed I had a lot more rage to play with.

While a 2H build doesn't have the same potential as a DW build for DPS, I think it is much easier to get gear that makes the build good. And with a good amount of skill you can do a lot of DPS.

I also found a mod by mistake that helped out with slam. I downloaded a mod call Quarts for helping timing my mage's spells. I found that it also tracked my swing timer on my warrior. So I could start a slam right when a regular swing went off. It's pretty nice, because at times my warrior doesn't look like she attacked until well after a swing happens.

This build is also very fun in PvP. I don't really notice other warriors using slam in PvP. But let me tell you, it is amazing at times. I did a lot of PvP over the weekend. By the end I was getting the timing of slam better with the fast pace of PvP. It's amazing when you get three crits in a row on a mage in about three seconds for over 6000 damage total:).
[Image: 21740hrsxL.png]
Reply


Forum Jump:


Users browsing this thread: 3 Guest(s)