A Call to Arms - MoP Mechanics Testing

[daishan]

Thx for the info.
I'm not sure how to calculate in the 3 sec cd on Colossus but considering our 3.6 speed data worked out to 3.6-speed 2H axe: 15.27% +/- 0.69%, ~2.55 PPM, the logs that I did for that had no haste so couldn't of dropped under the 3 sec cd. Just doesn't look to be 3PPM from our data.
Elemental Force at 5PPM, unless they've changed it in the last month that's just not even close to the 3PPM that ppl here have reported.

[theckhd]

I suspect they have changed in the past month; we know that Windsong did at least (it went from 3PPM to 1PPM).

I'll have to take a look at our data and see if it agrees with those descriptions. I share daishan's skepticism, but it's also possible that they changed very recently.

[mythor]

I am trying to match the paladin avoidance values from this thread with prot warriors, and they all fit pretty good. Thanks for all the work, that makes me feel more confident about my own values.
The only thing that is different is block, for warriors there is a hefty reduction factor.
On the upside, critblock chance = mastery

Naked 85 prot warrior:

Code: Select all

block DR = baseBlock + BoD + 1 / (1 / 149.1 + 0.885 / (mastery * 1 / 4.75 ))
block DR = 3 + 10 + 1 / (1 / 149.1 + 0.885 / (17.6 * 1 / 4.75 ))
block DR = 17.07


[xstratax]

I am trying to match the paladin avoidance values from this thread with prot warriors, and they all fit pretty good. Thanks for all the work, that makes me feel more confident about my own values.
The only thing that is different is block, for warriors there is a hefty reduction factor.
On the upside, critblock chance = mastery

What is the Mastery rating conversion for Prot Warriors for Block, like for Paladins its 600 Rating per 1% Mastery (and Mastery = Block as well), knowing that would help me make my add-on much more accurate. (Your code looks like Warrior Mastery -> Block is Mastery%/4.75 or Mastery% * 0.2105, which matches my ~0.2 modifier)

[theckhd]

I am trying to match the paladin avoidance values from this thread with prot warriors, and they all fit pretty good. Thanks for all the work, that makes me feel more confident about my own values.
The only thing that is different is block, for warriors there is a hefty reduction factor.

Your DR equation is exactly the same (postDR=1/(1/149.1+0.885/preDR)). You just have a much worse mastery->block conversion factor than we do.

[benebarba]

Still want parry/dodge rating/percentage info or are the DR coefficients pretty well sorted now?

[theckhd]

Still want parry/dodge rating/percentage info or are the DR coefficients pretty well sorted now?

I was hoping to collect more data to refine the coefficient values, in particular the Parry one (Dodge seems identical to Cata, and we know that fairly exactly).

[mythor]

It shows how sleepy I was yesterday, I used dodge/parry data gathered from my paladin instead of warrior. Now that I do use my warrior data I see that both parry and dodge grow slower compared to the paladin data.

https://docs.google.com/spreadsheet/ccc ... ZOVE#gid=0

[benebarba]

Still want parry/dodge rating/percentage info or are the DR coefficients pretty well sorted now?

I was hoping to collect more data to refine the coefficient values, in particular the Parry one (Dodge seems identical to Cata, and we know that fairly exactly).

to the beta!

Here's what I have at the moment, I'll get some more focused on only different values of parry rating:

All are for a level 85 human protection paladin with no buffs. % shown is the character sheet number (unless a '+' is shown, then it is from the tooltip for the rating conversion)

Naked:
Strength: 164
Mastery: 0 (+0%)
Stam: 317
Dodge: 0 rating, 5.01 %
Parry: 0 rating, 3.67 %
Block: 21.52%

Ret Geared (strength plate and 2-hander, no dodge/parry):
Strength: 6287
Mastery: 1988 (+11.09%)
Stam: 9034
Dodge: 0 rating, 5.01%
Parry: 0 rating, 29.01%
Block: 31.84%

Prot Geared(hit/exp soft cap, stam/mastery gemmed):
Strength: 4077
Mastery: 1980 (+11.04%)
Stam: 12,240
Dodge: 1785 rating, 11.83%
Parry: 1469 rating, 25.78%
Block: 31.81%

Prot Geared (all gear reforged to have parry, stam/mastery gemmed):
Strength: 4077
Mastery: 2449 (+13.66%)
Stam: 12,240
Dodge: 1219 rating, 9.82%
Parry: 2588 rating, 29.63%
Block: 34.03%

Above gear, removing pieces to get each of the following:
Strength: 3129
Stam: 9944
Dodge: 913 rating, 8.68%
Parry: 2016 rating, 24.07%

Strength: 2061
Stam: 6572
Dodge: 666 rating, 7.73%
Parry: 1465 rating, 17.80%

Strength: 963
Stam: 3535
Dodge: 445 rating, 6.85%
Parry: 609 rating, 9.81%


Let me know if that helps, or if I need to do it at a constant strength value.

[theckhd]

Unfortunately, I was hoping for L90 values... :/

I can use the L85 ones, but I don't know the rating->pct conversion off the top of my head. I want to say it's 256 @ L85, but I honestly don't remember.

<edit> it's 265. Updated the spreadsheet, now there's an 85 version and a 90 version:
85: https://docs.google.com/spreadsheet/ccc ... vSlE#gid=3
90: https://docs.google.com/spreadsheet/ccc ... vSlE#gid=2

[benebarba]

Unfortunately, I was hoping for L90 values... :/

I can use the L85 ones, but I don't know the rating->pct conversion off the top of my head. I want to say it's 256 @ L85, but I honestly don't remember.

<edit> it's 265. Updated the spreadsheet, now there's an 85 version and a 90 version:
85: https://docs.google.com/spreadsheet/ccc ... vSlE#gid=3
90: https://docs.google.com/spreadsheet/ccc ... vSlE#gid=2

Sorry about that - I'd have figured out the conversion and posted it, but I just caught your edit.

[mythor]

Playing around with matlab files provided by Theck I find the following for 85 prot war to make the results fit:

Code: Select all

k = 0.956
Cp = 238.7
Cd = 91


k still seems to be the cataclysm value. Cp is not that far off but Cd seems to be pretty different. I think I need more data. Is there more warrior data available somewhere?

[theckhd]

Mythor, where is that data from? L85 on beta?

[mythor]

yes

[theckhd]

Using the L85 dodge data for paladins:

Code: Select all

General model:
     f(x) = 1/(1/C+0.885/x)
Coefficients (with 95% confidence bounds):
       C =       65.45  (65.17, 65.72)

Goodness of fit:
  SSE: 0.0006356
  R-square: 1
  Adjusted R-square: 1
  RMSE: 0.00389


So dodge cap is unchanged, and the formula definitely uses k=0.885.

Parry data is giving me:

Code: Select all

General model:
     f(x,y) = 1/(1/C+0.885/(x/a+y))
Coefficients (with 95% confidence bounds):
       C =       237.2  (234, 240.4)
       a =       243.8  (243.3, 244.3)

Goodness of fit:
  SSE: 0.0546
  R-square: 1
  Adjusted R-square: 1
  RMSE: 0.03649


That doesn't really narrow down C much better than our earlier data sets though. If we knew a exactly that would help - didn't GC give us that number?