aroma

I think you're wrong here. Only WF-AP does not get the OH penalty, the regular AP from gear still gets the OH penalty (and if i understood you correctly, thats the AP you use in this example).
So the actual AP bonus on your OH is only half the amount you were using.
Sorry if i misunderstood your post, might as well be the case.

 

Lujaar

Well, if you only have windfury on a single one-handed weapon, it won't have a 20% proc rate either. The hidden cooldown will still render one swing after every WF proc unable to proc windfury, which will give you a 16.6% proc rate.

Mainhand Proc Rates: 2.6/2.6 WF/WF vs 2.6/whatever WF/FT:

Assume permanent flurry and no other haste buffs, so those Merciless Gladiator's Lolarenaweapons drop to 2.0 speed.


WF/WF: If each hand has a 36% proc rate, each hand will proc WF every 2.77 eligible swings. Figure in 1 mainhand swing locked out from a mainhand proc, and 1-2 mainhand swings locked out from an offhand proc, and you have your mainhand proccing windfury once every 4.77-5.77 swings.

WF/FT: WF procs every 5 eligible swings, and you lose 1 swing per proc to lockout. WF procs every 6 swings.


Theoretically, enchanting WF on a slow offhand should actually increase your mainhand proc rate, although anyone who can prove my math is bad is welcome to do so.

 

Rizelu

There are lots of pages here and lots of discussing going on so...
What are the values I should put into pawn? and what is intellect / stamina / spirit in this all?

Person #1: 2500HP
Person #2: 2800HP
Both get hit for 40-80 (average) and 90-120(crit)
Rest of the stats is all the same.

Won't the person with the most HP be able to deal more dps as he will most likely stand longer?

Intellect and spirit seem useless damage-wise unless you count in the amount of heals you can squeeze in, the longer you stand... the more damage you are able to produce.

If I am dead, I can't do dps .. simple lol.

 

Ujin

I think you are right in that WF/WF will be more procs of windfury in a given amount of time compared to single WF, but you are sill stealing MH procs every once in a while and lowering the average damage of WF by a lot, so its hard to judge without detailed theorycrafting if you can make up the lost damage by less WF procs with FT on offhand if you are increasing the average damage done by WF and adding in the damage done by the FT proc on your offhand.

 

aroma

In a fight where you get constant damage end don't get healed you might be right.
Stamina is a stat to consider when you're way below 10k hp fully buffed, but after that point it just isn't important anymore. Bosses will onehit you regardless of your HP, and anything else has to be healed anyway.
Your manapool works simmilarly. All you need is a manapool that is sufficient to keep up constant shocking and also big enough so that your manaregen with SR is never wasted. Beyond that point you don't have to care abaout intellect anymore.
Spirit as a stat is worthless for every shaman, if you need manaregen get items with mp5, but as an enhancement shaman you should always be able to reg enough mana with SR alone.

 

• Disquette

Let's go through point by point, shall we?

-How WF is effected by both avoidance and mitigation from armor assuming level 73 MoB with 5 Sunders

I assume a 23% damage reduction. That is more than fair to FT, considering most people are claiming an average of 20% damage reduction after sunders. I included dodges, so there goes the the avoidance issue.

-Flametongue ignoring armor

My sim does indeed use melee mitigation through armor, and allows flametongue not to be affected by armor.

-Flametongue with Nature's Guidance +3% spell hit

I modeled this with ZERO resists. I essentially gave natures guidance a 16% spell hit boost. If that's not more than fair to flame tongue, I don't know what is.

-Flametongue doesn't proc when you miss, possibly leading to hit rating gaining some importance in a WF/FT set up? I don't know

And guess what - the sim doesn't proc FT when the attack is avoided or missed. Pretty neat, eh? ;-)

-Flametongue benefits from any fire debuffs stacked on the MoB

Fair point, but I think the nature's guidance part more than covers this. However, to be complete, I just looked at the FT number, divided by 1.13 (16% resists, but 13% after natures guidance), multiplied by 109% (imp scorch), then multiplied by 110% (Curse of elements), then multiplied by 105% (crit rate), and the difference comes out to be 7dps. So, Those two numbers should be closer by 7dps due to resistances, nature's guidance, curse of elements.

-Flametongue always scales well with haste whereas too much can negatively effect WF

Another good point. I have no answer to this other than to say, it's a corner case most of the time, and the simulation I used took it into account. It'd be different for a person with lots of passive haste or item-based haste procs. My model never had the OH (with WF) dropping below 1.5 speed.

-Stormstrike DPS, higher with WF/WF Slow/Slow, but by how much?

The model shows takes this into account too. I use a 10.5 second delay between stormstrikes due to an assumed lag.

-Could using WF/FT versus WF/WF possibly land a net increase in MH WF procs, which could help scew the results in favor of FT? (I know you get a net loss in WF proc rate, but would you gain more overall MH procs that if the OH was "stealing" some of those procs, resulting in a bit more MH WF dps?)

This is all covered in the simulation, and was part of the testing. Here's the breakdown since you're interested:
WF Dps Slow/Fast: 124 WF MH Seconds per Proc: 7.81
WF Dps Slow/Slow: 128 WF MH Seconds per Proc: 8.06 (if you caught an early revision of this, i copy/pasted the wrong numbers from excel, hence the stealth edit)
So, even though the OH was stealing some, the increased flurry uptime, and the 20% to 36% jump was enough so that you had *more* MH WF in the same time period when OH'ing a WF weapon.

================

So, does this cover most of the concerns to the point where you consider it to be closer to reality than you originally thought?

 

• Disquette

Aroma, you are absolutely correct. I'm going to leave up my original post, but edit it at the end to correct my (rather colossal!) mistake. Luckily, this mistake is not in my simulator, so it doesn't affect the tests that I did earlier.

 

aroma

That's good to hear, because im rather fond of WF and would be almost sad if i had to switch to a different enchantment for maximum damage

 

Rescorla

Assuming those values above are correct, when the denominators are equalized the values are pretty close. The variable that needs to be determined is how many weapon swings are required to generate a proc. That number will vary based on OH weapon speed, haste, etc. Take that number and multiply it by 3.74, and then compare it to 22.33. For example, if it takes 6 autoattack swings to generate an OH WF proc, then you end up with FT having a value of 22.44 (6 * 3.74) compared to 22.33 for WF.

 

• Disquette

Yes, "Assuming those values above are correct" :-( Unfortunately I made a large mistake in it (see my post above). However, I don't think those numbers are anywhere near the be all / end all of the debate though, as weapon damage, crit scaling, flurry scaling, etc, are all in favor of windfury.

However, if you had your current stats, but had 10 000 AP instead of your current AP, FT would almost definitely win out.

I really wish I had my updated sim available though, because I don't remember what I changed in it between 2 months ago and today. I know the today version is more correct, but crushing my laptop with a truck prevents me from seeing how much of a difference it makes.

 

aroma

One more thing I noticed:
the AP bonus for the OH WF procc gives an increase of 11.16 DPS, not damage itself, so for a 2.6 OH you would actually get 29.016 extra damage per procc, which is most likely way ahead of the increase FT gets from the same amount of AP.

 

Rescorla

I just saw your edited values:

Assuming 6 autoattack swings required to generate an OH WF proc, equalizing the denominators results in FT adding 22.44 damage per 100 AP and WF5 only adding 11.16 (your corrected value). That is a fairly substantial differenct isn't it? At first glance it would seem to indicate that FT7 is clearly the way to go. Like you said however, weapon damage, crit scaling, flurry scaling, etc may allow WF to close the gap.

 

• Disquette

hah, aroma, you should be doing this instead of me!

This is one reason I wrote my sim - it's easy for me to get confused when I'm not building everything and checking it against in-game values. I make careless mistakes sometimes when I don't have a reference like in-game-testing upon which I can verify and correct.

 

aroma

No way, all I can do is some nitpicking, I better leave the brilliant theorycrafting to the pros.

 

Aramund

This, I think, is starting to get much closer to an accurate model. However, I have to ask if this takes into account the 100% proc chance on hit of FT versus 15-18% proc chance on hit (average) of WF? And there is still the question of lost DPS from Slow/Fast when it comes to Stormstrike, and the potential gain in DPS from more MH WF procs from OH FT.

Also, the (minimal) loss of Flurry and UR up time from losing the extra attacks that WF provides would almost certainly be more than made up for by the increased number of crits / unit of time being landed by the now (much) faster OH weapon.

 

Aramund

Forgive me if I missed of those items in your simulation. As I stated in my first post I don't claim to be an expert and its entirely possible I didn't fully understand the original model enough to notice some of those variables being factored in. I'm just trying to do my part to help.

 

Tornhoof

No time yet to put the FT/FB changes into my code, just some ideas...
1.) UR uptime isn't affected much at all, we're 95% uptime anyway
2.) Flurry uptime is a slightly more complicated case but not that much different
expected wf proc time while flurried is: wspeedflurry/0.2 (for mh only), for a 2.6 weapon this is 10s, so the wf cd percentage is actually 1- (10/13), which is 23%.
so we actually can assume that 2*(1-(1-0.2)^((10/13)*3)) = 0.80 swings are expected to be added to our flurry uptime calculation, for DW the value for two 2.6s weapons would be: 1.16
for 35% crit the difference in uptime is: 85% vs. 87%
3.) Stormstrike as a haste factor is a difference of maybe 10% between a fast weapon and a slow weapon

Anyway I can say more about this, as soon as I've modelled it into my code.

http://code.google.com/p/wowequipoptimizer/

 

Malevolencia

Does shadow priests Misery debuff (5% spell damage while dots are ticking) also effect FT? If so another 1.05 multiplier is needed.

 

bloodwitch

crusader is not better than potency in any way shape or form at level 70. Crusader is equivalent to a +15str permachant in terms of autoattacks. unlike a warrior, we cannot artificially pump up the proc rate by spamming instant attacks since we only have the one, and its on a cooldown. in all the testing and parsing i did (both real combat, and blasted lands servants), with ss spamming and WF buffs up, crusader only barely managed to match the constant +20str when it got ungodly proc strings in a row, and that happened only very rarely.

potency > crusader.

 

Malan

Sorry, but you're wrong on that one. The calculations for potency vs crusader were figured out ages ago. You're not suddenly discovering something we didn't know.

Shitting up every single thread on EJ since '06

 

Malevolencia

So as aroma said, it's 11.16 dps not dmg per proc.
Assuming optimal (unrealistic) 20% proc rate, that's 11.16/5 = 2.226dps added over time.
FT is 3.74 dmg * 1.05 (shadow priest misery) giving 3.927 dmg per swing.

Assuming a fast offhand of 1.4spd (I actually never looked at fast OHs, do they go faster than this? if so FT is better..) that's 3.927dmg every 1.4sec or 2.805 dps.

FT seems to scale significantly better then (at the 35% crit rate)?

Any other factors that have yet to be included?

EDIT: Also -target armor items would be useless with FT, while WF would scale nicely from them.

 

Freyalis

This holds true purely for white attacks untalented only. After taking into accounts extra attacks from Stormstrike and Windfury Attacks and also the semi constant 30% speed increase from flurry you come a lot closer to double the expected number of procs. Hence its a lot closer to +30 STR than +15 which is alot better than potency.

Like Malan said the math has been worked out before and you should be able to find it easily with a search.

 

Yavon

Currently using a Dragonstrike in MH working on Dragonspine. I want go for a haste type of build. Would I need to increase hit rating to max (24.5 I think) to max my dps output. Or does current scale for str crit hit rating hold true for shamans that want to value haste as well?

 

Hedin

full stack of improved scorch debuff gives 15% bonus :-)
And BTW for OH WF it's 11.16*speed dmg/proc

WF3sec Ace2 - Shows WindFury and Stormstrike cooldowns.

 

Lujaar

What I wrote out above covers mainhand windfury only, and should account the offhand "stealing" mainhand procs.

I'm not just saying you should get a windfury proc every 4.77-5.77 mainhand swings; I'm saying you should get a mainhand windfury proc every 4.77-5.77 mainhand swings. (You'll also get offhand procs at about the same rate, but that's not what I'm interested in.)

To put it another way, you will see more than twice as many total WF procs using slow/slow WF/WF compared to WF/FT, and half of those procs will be mainhand.