Mists of Pandaria: All Specs - Page 7

Starfox · 07/30/12, 11:14 AM

I updated the warrior module in simulationcraft, if anyone, who is probably familiar with warrior than me, finds anything weird feel free to reply here or send me a private message.

Derkthar · 08/01/12, 2:54 PM

From my current evaluations Arms will be dominant over TG for at least the first tier unless something changes with the hit/haste/crit scaling of Fury TG, I have not had time to calculate SMF but I think it will be in better shape than TG.
I have also considered taking shockwave for the extra damage if the AP ratio stays high enough to be worth the GCD and 20 rage cost.
We shall see how things level out, any other opinions?

Ratak · 08/03/12, 5:15 AM

Originally Posted by Starfox

I updated the warrior module in simulationcraft, if anyone, who is probably familiar with warrior than me, finds anything weird feel free to reply here or send me a private message.

Hello,

I have downloaded build 504.1 from simc-504-1-win32.zip - simulationcraft - Patch 5.0.4 Release 1 *BETA* Win32+Source - World of Warcraft DPS Simulator - Google Project Hosting

I did not found warrior profiles. Anyway, it is good news that someone is willing to give our class some Simcraft Support.

Since Wow in Game tooltips are not accurate, I have done some extended testing in order to get the correct formulas for every single one of our spells one the Beta. Note that SC wasn't accurate either for Cataclysm, because IG formulas and tooltip were no more correct than they are currently on the Beta (but that is not the topic here..).

So the followings are the formulas for non crits damages of ARM SPEC. Please MP me if you can find that any of those is not correct :

Stats :
AP = Attack Power
BWS = Base Weapon Speed (does not take into account Haste Bonus)

Weapon :
WR = Weapon Range
NWD = Normalised Weapon Damage = WR + 3,3*AP/14
NNWD = Non-Normalised Weapond Damage = WR + BWS*AP/14

Please note the difference beetween the two last ones.

Modifiers :
DR = Armor Reduction through armor (i.e. DR = 0,6 means 40% of damage taken)
DT = 1 - DR
PD = Physical debuff = 1 (if ON) ; 0 (if OFF)
Mod = 1,2*(1+PD*0,04)

TFB = Number of Stacks of the buff Taste for Blood

Formulas for non crit Skills damage (ARM SPEC) :

Mortal Strike : = (1,5*NWD+1923,3333)*DT*Mod

Deep Wounds (5 tics) =(0,99968287*AP+913,507602)*Mod

Opportunity Strike = 0,55*NWD*DT*Mod

Overpower =(1*NNWD)*DT*Mod

Slam = (2*NWD+1870)*DT*Mod

Execution = (4,06201349*AP+8720,85404)*DT*Mod (Note : this one has not been tested)

Colossus Smash = (1,75*NWD+388,33)*DT*Mod

Heroic Strike =(499,1666+1*NWD)*(1+TFB)*DT*Mod

Victory Rush =(0,56*AP)*DT*Mod

Cleave =0,75*NNWD*DT*Mod

ThunderClap =(0,51*AP+376,14)*DT*Mod

Whirlwind =(0,85*NWD)*DT*Mod

DragonRoad = (AP*1,4+126,889)*Mod (Note : this one has not been tested since build 15851)

Blade Storm (per tic) =(1,2*NWD)*DT*Mod (Note : total = 7 tics in 6 secs)

Storm Bolt =(4*NNWD)*DT*Mod

Heroic Throw =(0,5*NNWD)*DT*Mod (Note : No longer resets Auto attack)

Heroic Leap = (0,5*AP)*DT*Mod

ShockWave = (1,2*AP)*DT*Mod

These formulas give non crit damages. Please MP me if you see any mistake.

Average damages must take into account crit chances, opportunity strikes chances, and dot clipping from DeepWounds. But that is quite easy to get from what is listed above.

If there is someone willing to update SC accordingly, I will do the same work for Fury Spec TG and SMF. If not, there is no point and since it takes me a lot of time i will just keep the results and my spreadsheet for me and my guild.

Sorry for my english also,
Cheers

Berthold · 08/03/12, 7:21 AM

IIRC simcraft is able to parse the scalefactors from the client data. So as long as there are no server-only hotfixes, we don't need those formuals. Still, thanks for the data, it helps to know whether the outcome is correct :-)

regarding simcraft binary.. Even compiling directly from source (mop branch) doesn't work for me. :-/

Ratak · 08/03/12, 8:57 AM

Originally Posted by Berthold

IIRC simcraft is able to parse the scalefactors from the client data.

Can we be sure that the scalefactors taken from the client data are the correct ones ?

For exemple, IG formula for Heroic Strike says Damage = 100% weapon damage + 599, where as real damages are 100% Weapon damage + 499.

Another exemple, Overpower IG formula says : 100% Weapon Damage, where as real damages use not-normalised weapon Damage (which is higher).

I don't know how simcrafts works or extracts its data, but I know for sure its forecasts were very unreliable for the Cataclysm Warrior. If it is the same for Mop then it will be difficult again to use simcraft....

There is a simple way to check if the datas from SC are correct : by making SC run a simulation with this test weapon Worn Battleaxe - Item - World of Warcraft equiped (we need this testing profile implemented). Unequip anything that can temporarily boost AP (trinkets with procs...) and don't use CS or Avatar.

Damages will be static and should match the formulas above if it is correct.

Thx in advance if someone can implement the profile. I will do the testing.
Cheers.

Starfox · 08/04/12, 7:37 PM

Originally Posted by Ratak

Can we be sure that the scalefactors taken from the client data are the correct ones ?

For exemple, IG formula for Heroic Strike says Damage = 100% weapon damage + 599, where as real damages are 100% Weapon damage + 499.

Another exemple, Overpower IG formula says : 100% Weapon Damage, where as real damages use not-normalised weapon Damage (which is higher).

I don't know how simcrafts works or extracts its data, but I know for sure its forecasts were very unreliable for the Cataclysm Warrior. If it is the same for Mop then it will be difficult again to use simcraft....

There is a simple way to check if the datas from SC are correct : by making SC run a simulation with this test weapon Worn Battleaxe - Item - World of Warcraft equiped (we need this testing profile implemented). Unequip anything that can temporarily boost AP (trinkets with procs...) and don't use CS or Avatar.

Damages will be static and should match the formulas above if it is correct.

Thx in advance if someone can implement the profile. I will do the testing.
Cheers.

I'll look into the normalized thing, our parsing does

  case E_NORMALIZED_WEAPON_DMG:
    normalize_weapon_speed = true;

Seems like the base value (0/1) can determine if it actually affects the spell.
Overpower and Cleave have the "Normalized Weapon Damage" effect, but with a base value of 0.
Heroic Throw, Storm Bolt do not have it at all.

Heroic Stike: (normalized weapondamage + 499)*100%

#1 (id=49)    : Normalized Weapon Damage (121)
                Base Value: 1 | Scaled Value: 498.519 (avg=0.4)
#2 (id=50)    : Weapon Damage% (31)
                Base Value: 100 | Scaled Value: 1 (avg=0)

Ratak · 08/05/12, 6:30 AM

Reply to Starfox :

- According to my datas, it seems like the 498,519 taken into account for Heroic Strike is correct (instead of the 599 given by the IG tooltip), which is good news.

- I am 100% sure about Overpower using non-normalised-weapon-damage. Same for Heroic Throw, Storm Bolt, Cleave. So if it is not the case in SC, then we will get unreliable forecasts.

- I don't understand the meaning of :

#1 (id=49)    : Normalized Weapon Damage (121)
                Base Value: 1 | Scaled Value: 498.519 (avg=0.4)
#2 (id=50)    : Weapon Damage% (31)
                Base Value: 100 | Scaled Value: 1 (avg=0)

- Also, I have made a spreadsheet with the correct formulas, from which I can predict non crit damage of every single spells on the beta. That is why if someone can implement the testing profile (with the appropriate weapon and stuff to get static output) explained above for MOP , then I would be able to do some testing and compare with my spreadsheet to tell which spells are correct, and which ones are not.

Starfox · 08/05/12, 12:25 PM

Originally Posted by Ratak

Reply to Starfox :

- According to my datas, it seems like the 498,519 taken into account for Heroic Strike is correct (instead of the 599 given by the IG tooltip), which is good news.

- I am 100% sure about Overpower using non-normalised-weapon-damage. Same for Heroic Throw, Storm Bolt, Cleave. So if it is not the case in SC, then we will get unreliable forecasts.

- I don't understand the meaning of :

#1 (id=49)    : Normalized Weapon Damage (121)
                Base Value: 1 | Scaled Value: 498.519 (avg=0.4)
#2 (id=50)    : Weapon Damage% (31)
                Base Value: 100 | Scaled Value: 1 (avg=0)

- Also, I have made a spreadsheet with the correct formulas, from which I can predict non crit damage of every single spells on the beta. That is why if someone can implement the testing profile (with the appropriate weapon and stuff to get static output) explained above for MOP , then I would be able to do some testing and compare with my spreadsheet to tell which spells are correct, and which ones are not.

It means ( NWD+498.519 [#1] ) * 100% [#2]
Whirlwind is not NNWD?
Overpower - Spell - World of Warcraft
Cleave - Spell - World of Warcraft
They both have the effect "Normalized Weapon Damage (121)", but don't deal normalized weapon damage
I thought maybe it's if base_value = 0 for that effect it is still NWD, but:
Whirlwind - Spell - World of Warcraft

#1 (id=512)   : Normalized Weapon Damage (121)
                Base Value: 0 | Scaled Value: 0 | Coefficient: 1.000

:c

Ratak · 08/05/12, 1:32 PM

Whirlwind uses normalised weapon damage

Starfox · 08/05/12, 10:15 PM

Originally Posted by Ratak

Whirlwind uses normalised weapon damage

Well then I have no clue how to get from spelldata that cleave and op use NNWD, so I hardcoded it.

Ratak · 08/06/12, 5:38 AM

Originally Posted by Starfox

Well then I have no clue how to get from spelldata that cleave and op use NNWD, so I hardcoded it.

Okay. Thank you for your time. I will do some testing as soon as possible and will give feedback. cheers

Berthold · 08/06/12, 10:00 AM

Unshackled Fury is currently bugged, but apart from that, simc is now looking good and gives reasonable results. As soon as the mastery is fixed, I will post some html for your further analysis.

Abbys · 08/06/12, 10:29 AM

Originally Posted by Starfox

Well then I have no clue how to get from spelldata that cleave and op use NNWD, so I hardcoded it.

Originally Posted by Ratak

Okay. Thank you for your time. I will do some testing as soon as possible and will give feedback. cheers

I have just downloaded the 504-2 version and still cant see the profile, any chance you could provide us with the link ? there was nothing on the official website as well.

Berthold · 08/06/12, 11:39 AM

while we are finalizing the next release, can somebody support us with some validated data on the current (=MoP) enrage behavior?
Especially what does happen when an enrage procs, while we are already enraged? Does it prolong enrage, does it reset enrage? Does it give a charge for the Raging Blow? And how does berserker Rage come into play.

Ratak · 08/06/12, 1:22 PM

Originally Posted by Berthold

while we are finalizing the next release, can somebody support us with some validated data on the current (=MoP) enrage behavior?
Especially what does happen when an enrage procs, while we are already enraged? Does it prolong enrage, does it reset enrage?

The buff Enraged can come from different sources : Mortal strike, CS crits (in Arms), or the use of Berserker Rage.

When we get another enrage while already enraged, It just resets the buff Enraged.

Does it give a charge for the Raging Blow?

Raging blow is only usable if the warrior has the buff Enraged. It is not a matter of charges. But i'm not very sure about that since I haven't been testing Fury on the beta for a while.

And how does berserker Rage come into play.

When berserker Rage is active, it also provides the buff Enraged