A Call to Arms - MoP Mechanics Testing

Here you go, from 10K to 16.5K AP.

Str 9354

Parry 14.38%
Parry of 785 adds 0.76%

Parry 14.84
Parry of 1252 adds 1.21

Parry 15.10%
Parry of 1516 adds 1.46%

Parry 15.29
Parry of 1714 adds 1.66%

Parry 15.48
Parry of 1902 adds 1.84%

Parry 15.77
Parry of 2202 adds 2.13

Parry 15.98
Parry of 2420 adds 2.34%

Parry 16.28
Parry of 2724 adds 2.63%

Parry 16.65
Parry of 3106 adds 3.00%

This is the best I can do without changing my STR zilch

I'll incorporate that data tomorrow, but what I asked for was the opposite - vary strength while keeping parry rating constant.

theckhd wrote:I'll incorporate that data tomorrow, but what I asked for was the opposite - vary strength while keeping parry rating constant.

Blah, I'll think of something then.

does the amount of parry rating mater? or as long as its the same its ok?

Like... having 700 parry rating but varying the amount of str as much as I can...

Parry of 2178 adds 2.10% (Will remain the same)

Str - 3638
Parry chance of 9.50%

Str - 4242
Parry chance of 10.17%

Str - 4752
Parry chance of 10.74%

Str - 5477
Parry chance of 11.55%

Str - 6115
Parry chance of 12.25%

Str - 6606
Parry chance of 12.79%

Str - 7210
Parry chance of 13.44%

Str - 7935
Parry chance of 14.23%

Str - 8779
Parry chance of 15.13$


----

I'd have more data but I got no bagspace

Had to sell a bunch of my gear, I had no more bagspace... So my sets will be limited from now on.

Thanks Klaud, I'll take a look at the whole data set later today, hopefully that's enough spread of data points to nail down the parry details.

Coming out of the blue: my beta's running again, so I'll be able to log some time.

Any particular element a 87 pal can check that is needed rght now?

#28 and #29. As far as I can tell, spell coefficients are static, so any level character can do #28. Just be sure to report your level and the dummy level, along with all of the other info I requested.

From Klaud's data sets, the armor coefficient @93 seems to be roughly 25050. That's good enough for now, we'll check the exact formula and the level scaling later on.

Sagara wrote:Any particular element a 87 pal can check that is needed rght now?

If the level 90 dummies have finally been implemented, we'll definitely need a level 88 anything (i.e. a Paladin is mighty fine) to check the scaling of glancing blows. For now, lower-level Paladins can tackle the weapon enchants (see leading post), as those test are basically level-agnostic (tests #3, #4, #21).



@Theck : please edit test #29. We don't need the +3 scaling, but the +2 one (i.e. level 88 on level 90 dummies); also, increase the duration to 1h+ and specify that the gear must be crit-heavy, as I'd like to snipe data to see if block events are indeed one-roll in the case of AAs.
EDIT : Here's how it should look like.

• Level : 88 (exactly 88).
• Gear: Crit-heavy (Ret-oriented).
• Target: Any level 90 dummy (exactly 90, not 93), attacking from the front.
• Attack Sequence: just Auto-attacks.
• Goal: 60+ minutes worth of auto-attacks.
• Report: total physical crit (character sheet), upload the combat log.

Here are some attempts we did yesterday. I haven't had a chance to really delve into uptimes and what not but here you go.


http://www.worldoflogs.com/reports/rt-l ... 425&e=2598
http://www.worldoflogs.com/reports/rt-llbz58stty2jdenc/

tlitp wrote:@Theck : please edit test #29.

Done

I got a question in #28

How bad is it gonna affect that there are no lvl 90 dummies, seeing as there's only lvl 85 and 93 dummies out there, AND do we have to scrap the log if someone hits our dummies while we hit them, or puts a debuff on them? Just want to make sure.

It was not much of a big deal on tiltp's logs as he only required 1+ mins per AP set, and I did have to scrap some of them actually because of a lock just went around dotting everything...

@Theck :
1. Please update #29 again (sorry). As the lovely Dal.Sword got changed to 1.8 in MoP, the fastest weapon purchasable from vendors is now Tomahawk. Please add it as a requirement for this test suite, as we want to get as many swings as possible within a given time frame.
2. Remove #31, as it's been covered here.


@Klaud on #28 : use isolated low-level dummies for anything else but CS/HotR. Use boss-level dummies (again, isolated) for CS/HotR alone. In other words, you can avoid the crowded dummies if testing anything that doesn't deal physical damage (thus not being affected by armor).

Gotcha. I'll work on that as soon as I feel better from my migraine...

You need 30 mins of each at least, one for CS/HotR and one for the rest, right?

Should I be casting SotR on the CS/HotR log or not?

tlitp wrote:@Theck :
1. Please update #29 again (sorry). As the lovely Dal.Sword got changed to 1.8 in MoP, the fastest weapon purchasable from vendors is now Tomahawk. Please add it as a requirement for this test suite, as we want to get as many swings as possible within a given time frame.
2. Remove #31, as it's been covered here.

Done and done.

Klaud: IIRC, SotR does holy damage, so you only need to use it on the low-level dummy test.

The parry data is still giving me trouble. I'm fairly certain the STR component of parry isn't subject to DR (or if it is, the cap is something ludicrous like 200+). However, the modeling is giving me a very low cap for the rating component of parry (something like 15, which seems unlikely). I'm pretty sure this is just a case of not having enough data to get a solid surface fit, so I'll wait for more data points.

Edit: at this point, it looks like you get 0.0011%*strength as parry, so every ~909 strength gives you 1% parry.

Parry data, yet again

Parry rating of 1191 adds 1.15%

Str 1350
Parry chance 5.85%

Str 2236
Parry chance 6.87

Str 2870
Parry chance 7.60

Str 3631
Parry chance 8.46

Str 4602
Parry chance 9.56

Str 5236
Parry chance 10.27

Str 5870
Parry chance 10.98

Str 6386
Parry chance 11.55

Str 7056
Parry chance 12.28

Str 7817
Parry chance 13.12

Str 8578
Parry chance 13.94

Str 9548
Parry chance 14.99

Str 10309
Parry chance 15.80

I'll start working on #28

#28 -- first 3 sets (CS/HotR) against lvl 93 dummy. No Glyphs, no enchants, no procs, no debuffs, no buffs. There might be some white noise, there was a lvl 80 dummy next to the raiding lvl dummy and HotR was hitting it. Just so you know. It was the best I could do since I had to go to Undercity to find dummies not in use .

1st
Str 3804
AP 7858
SP 3929
Haste 0
Hit 1.73
Crit 6.16
Exp 1.08

2nd
Str 5192
AP 10634
SP 5397
Haste 1.07
Hit 1.73
Crit 6.16
Exp 3.31

3rd
Str 6720
AP 13690
SP 6720
Haste 3.38
Hit 1.73
Crit 6.97
Exp 3.31

Will start working on two more sets once i return from my jog.

4th set
Str 8170
AP 16590
SP 8295
Haste 3.38
Hit 2.87
Crit 8.14
Exp 3.31

5th set
Str 9819
AP 19885
SP 9944
Haste 4.51
Hit 2.87
Crit 9.60
Exp 3.31

There might be some extra noise in the last set, a mage and a lock decided to nuke the lvl 80 dummy i was cleaving with HotR -- I hope it doesnt affect too much.

if you need more granularity, I'll provide those later on.

As for the ones required the lvl 85 dummy -- I'm out of luck at the moment trying to find a dummy no one else is hitting...

#28 (J, HW, Cons, SotR, AS vs lvl 85 dummy)

1st
Str 3804
AP 7858
SP 3929
Haste 0
Hit 1.73
Crit 6.16
Exp 1.08

2nd
Str 5192
AP 10634
SP 5397
Haste 1.07
Hit 1.73
Crit 6.16
Exp 3.31

3rd
Str 6720
AP 13690
SP 6720
Haste 3.38
Hit 1.73
Crit 6.97
Exp 3.31

#28 (J, HW, Cons, SotR, AS vs lvl 85 dummy)


4th set
Str 8170
AP 16590
SP 8295
Haste 3.38
Hit 2.87
Crit 8.14
Exp 3.31

5th set
Str 9819
AP 19885
SP 9944
Haste 4.51
Hit 2.87
Crit 9.60
Exp 3.31

Klaud, quick question: For the parry data, are all of those values unbuffed? I.e. you didn't use Kings or other buffs to increase STR? Just purely gear?

<edit> also, what race was that character?
<edit> nevermind, I see it's BE. Can someone (anyone) tell me what the base (naked, unbuffed) dodge and parry values are for a level 90 BE on beta. Literally take off all gear and cancel all buffs, and just read off the base dodge/parry values from the character sheet.

i never had any buffs with me. i can get more parry sets data later today, once i get out of the office.

also, let me see if i can get some sets with zero parry rating and full str, including trinkets, would that help?

Klaudandus wrote:i never had any buffs with me. i can get more parry sets data later today, once i get out of the office.

also, let me see if i can get some sets with zero parry rating and full str, including trinkets, would that help?

I think at this point, I have it pretty much nailed down, so it's not necessary. I did some testing this morning with an 85 premade, and convinced myself that STR->parry is affected by diminishing returns (using the same test you just proposed - naked to full ret gear with 0 parry or dodge rating).

Using the data from that 85 premade, I figured out the dodge/parry formulas at 85. Armed with that knowledge, I went back to your data sets and figured out how things work at 90. The one missing piece of the puzzle that could confirm or refute the formula I have is the base amount of parry a naked level 90 BE paladin has. So if anyone can give me that value (guess: 3.18ish), I'd appreciate it.