WF proc from SS = 0.36 + ((1-0.36)*0.36) = 0.5904 vs WWS 0.5751.
I deduct mh wf proc number from total oh attacks becouse those are under shared CD. I wanted just see proc% from both hands invidially.
Can we now forget this subject?


Updated lootrank link for 3.02 at 70lvl raiding. Database still bugs. You need lower str items ep value by self. Assuming spell hit cap gemmed etc if not there.

That's just it - it is low. Because MH gets prioritized, OH WF procs from SS will be far lower. You can't say that the OH proc chance on SS is near 36% when MH gets first shot at it with it's 36% proc chance. You can say OH WF chance is 36% if MH didn't proc WF first, but that is only looking at the remaining 64% of SS attacks.

And as a mind exercise:

Actual proc rate from the WWS: (120/546)*100 = 21.99%
Realistic proc chance of OH WF proc from SS: 0.64*0.36*100 = 23.04%

That's not was this discussion was about. It was about any given attack of a single weapon outside the Windfury cooldown. And that's apparently 36%, even for the offhand.

SS is an atomic action that includes both MH and OH so talking about OH WF proc chance without taking the MH's proc chance into account makes no sense. That is my point. I'm being a pedantic here, but we're talking about the common rule-set we use in enhance theory-crafting so pedantic is good.

You're talking about all attacks having a 36% chance of proccing WF(outside of the 3sec rule). That's not right. The 36% applies to SS AFAIK. I did some quick mashup of a generic WWS report(twins, muru and some attempts on KJ):

Melee hit, crit, SS hit, crit:
1780+1278+421+248 = 3702 number of attacks
Excluding misses and parry
(Of course you automatically lose some potential WF procs due to SS, but you can't link the WFs from where they procced and every attack is a potential WF proc)

WF hit, crit:
827+475 = 1302. Procs is half that: 651

So my proc rate for the entire evening was ((1302/2)/3702)*100 = 17.59%. Now this is also a small dataset(too small in my book), but it's larger than the one we've been discussing in the SS debate.

What you're describing is overall proc chance. That is at ~17,59% overall including attacks made during the cooldown, the chance has to be significantly higher for attacks outside the windfury cooldown.

Use this formula to calculate proc chance.
(WF procs) / [(battle time - WF cd time) * (avarage swing per second)]

WF procs = (hits + crits) / 2
Battle time = combat time
WF cd time = WF procs * 3s
Avarage swing per second = (SS hits + SS crits + normal hits + normal crits) / battle time

Every Brutallus report I have parsed show > 0.3.
Formula uses avarages but at the end that not matter really much.

Seems like it. The whole point of the discussion - at least to my understanding - was to make sure that the simulator actually simulates windfury correctly. And in order to do that it needs to know how big the proc chance for windfury is for a single attack outside the 3 second cooldown. Which seems to be 36%, not 20% like some people repeatedly suggested in this thread. The actual proc rate may be 18% oder 20% or whatever, but that's not what the simulator cares about.

Punched my WWS entry into Pitbull's formula:

Attacks: 3702
WF procs: 651
Battle time: 3:48, 29% presence: (3*60*60+48*60)*0.29 = 3967.20

>>> 651/((3967 - (3*651)) * (3702/3967))
0.34637561381512555

~ 34.64%

The formula looks good for calculating an average, I have to say. I was thinking of how to get a good model for finding the 3sec silence, but this seems to do the job. I'm convinced. Thanks for the clear-up, Pitbull. It's probably the most solid work I've seen on WF mechanics in a long while.

tukez - the bootstrapping of weapon buffs in enhsimconfig.cpp is quite messy. It the first if-else paragraph in both functions. The else should always be false, no?

I don't actually know which lines do you mean, maybe these?

That works correctly. Actually the else statement is stupid, as the proc rate is already that one.

Yeah. Since WINDFURY_DUALWIELD_PROC_CHANCE applies when dual-buffed with WF, the else statement can never be correct. It's just nitpicking anyway.

Currently you can see the wf procs/hits and wf procs/swings per hand in the end of sim, just scroll up a bit. You can also see miss/dodge/hit/crit amounts for every ability there. BUT as I did the test I noticed that the WF% didn't take into account SS nor LL. If you test it, keep that in mind. I fixed that for the next version though.

I did these with default config, just took out LL from rotation and 0/2 improved stormstrike.

Anyway, with 36% proc chance outside cd:
And with 20% proc chance outside cd:

This could be a very big point of contention going forward.

If lava lash is a melee attack and can proc offhand windfury then won't it increase the incidence of offhand windfuries when using WF/WF?

I don't think there will be any way that anyone could track the Windfury cooldown and all the other things and delay hitting Lava Lash to attempt to maximise Mainhand Windfuries, the rotation is just too tight and constant. It will be hard enough just to utilize a priority list properly.

If Lava Lash does proc offhand windfuries is it still a dps increase if you use wf/wf with what we have just established as how the proc rates change when dual wielding wf/wf?

Can your sim determine the answer to that?

0.9.3
-Windfury statistics corrected to take into account SS and LL.
-armor_ignore changed to armor_penetration. Notice, that it is % now. Also ep_armor_ignore changed to ep_armor_penetration_rating.

enhsim - Google Code

Yes, you can try it yourself now.

The elemental t6 shoulders come out well ahead of the enhancement t6 shoulders on that list. If that's at all accurate, we need to be making a lot of noise.

0.9.5
-Fixed Windfury bonus ap to 475.
-Fixed Call of Flame to affect whole Searing Totem damage, not just base damage.
-Glyphs added: Lightning Bolt, Lightning Shield, Earth Shock, Windfury Weapon, Flametongue Weapon, Stormstrike.

enhsim - Google Code

You can use same glyph in multiple slots currently. Again, I don't know correct mechanics, so this is what I did with glyphs:
Lightning Shield stacks, multiplicative.
Lightning Bolt stacks, additive.
Earth Shock doesn't stack.
Windfury Weapon stacks, multiplicative.
Flametongue Weapon stacks per glyph, no extra bonus with dual flametongue, additive.
Stormstrike stacks, additive.

Windfury ap was 445. It is a bit wierd that Thottbot and Wowhead both show 445 bonus ap, but I logged in with my shaman and noticed it is 475.

From the Deathfrost thread, this is the post that seemed to summarize everything the best.

http://elitistjerks.com/805104-post128.html

In the coming days Ill go try and whack away at a dummy on the ptr to see if I can add to the information at all for proc rates.

I scanned my project folder and the zip file, but no detections. Now I'm scanning my hard drives, nothing so far. I'm using AntiVir.

I checked what's that file filtering means. So you recomprossed the sim and sent it to your e-mail? Then Antigen warned when you opened the email. And then a picture with settings...one being .exe file. So most probably your Antigen is configured to detect and warn about .exe files in attachements. Just to clarify, that it doesn't mean that it's a virus.

I was looking at Elixirs and noted that it may be useful to include [Elixir of Mastery] in the sim as it gives +15 all stats. For us that means +15 str, +15 Agi and +15 Int = +45 AP and some melee crit from Agi and some spell crit from Int. Which makes its look a decent elixir.

I was also wondering if it wasn't a good idea to include a separate section for Guardian Elixirs as whilst there seems to be only one good one at present that gives us a dps boost, it is very likely more will come out as they do final top level passes for all the professions and add in new top level Alchemy recipes.

The current Guardian Elixir of interest is [Elixir of Draenic Wisdom] that gives +30 Int = +30 AP and some spell crit.


Of course in addition to this is the elixirs already announced for WotLK but unavailable in patch 3.0.2 eg:

Elixir of Mighty Strength, Guru's Elixir and Elixir of Mighty Agility.

Edit: Replaced [ item] tags with URL tags as item tags point at Wowhead: OMG IT'S ALMOST HERE not http://woltk.wowhead.com

Nod32 doesn't report anything in the current EnhSim ... and didn't do so with all the versions i used before that one.

I realise it doesn't, but it's the work Antigen, something I don't even see (it's mailserver stationed), so I just wanted to make sure that it was just my side seeing a problem. Seeing as that is likely the case, there's no problem. Thanks for checking though.

Minor niggle with sim talent options.

There is no option to choose stormstrike - I guess because its kinda assumed as a given, same with Dual wield. However there is a choice of 3/3 Elemental weapons, which as a pre-requisite of Stormstrike obviously must be a given too?? So is it necessary to give Elemental Weapons as an option if you are assuming stormstrike?

BTW Lava Lash are you intending to include that as a talent choice? Or is it a case if its ticked in the priorities you assume they have it?

Actually thinking about it I'd be interested in assisting with the GUI perhaps adding options to import a tree from wowhead/blizzard/wowdb etc. Possibly also looking at the feasibility of importing stats from Armory. Although not sure armory gives all the stats I couldn't see ArP given for instance - certainly not as a percentage.

Only problem is my C++ skills are extremely rusty. I'd be happier coding in Java, which would have the benefit of being platform independent and helping the Mac & Linux guys.

I'd be willing to look at a Java port of the sim too if you were at all interested in heading in that direction? As I say the major plus is platform independence.

Are you sure the tooltip isn't just showing the current value as modified by talents or equipment effects? What value does it give when you have no talents and naked? I'm at work so can't test this on beta at present.