I realize you're not convinced by the other thread, which I suppose is understandable. So, let me give you a little more data from the sets I used there to show why something's wrong with assuming PPM mechanics use base weapon speed to calculate the proc chance.

With 0/5 flurry and two 2.7 speed weapons, I observed 15650 Unbridled Wrath procs in 22927 landed attacks.
With 5/5 flurry, I observed 16123 procs in 25554 landed attacks.

I might be messing up some of the statistics, but here goes:

0/5 Flurry proc chance = 68.26 +/- 0.31 %
5/5 Flurry proc chance = 63.09 +/- 0.30 %

To get a quantitative measurement of how different these are, subtract them and compare to 0%.

(0/5 Flurry proc chance) - (5/5 Flurry proc chance) = 5.17 +/- 0.43 %

That's 12 standard deviations away from 0. The probability that this occurred as a random fluctuation is incredibly small, and so the only conclusion I could make was that flurry was affecting the proc chance. When I split the second data set into flurried and unflurried hits, I observed that the flurried hits had a lower proc chance. So, it makes sense to conclude that when converting from a PPM to a proc chance, you use the currently hasted weapon speed.

Incidentally, way back when the rage formula changed in 2.0 someone tested rage gen and found that it did in fact drop with flurry - post. I guess that got changed at some point.

The biggest thing that sealed it for me was the paladin ability that would proc every hit with no haste, but with haste rating did not proc every hit. Its not conclusive for all abilities, but its not realistic to test them all and reasonable to think they work the same until evidence shows otherwise. Its much easier to disprove than prove in this case.

Edit: Remember Flurry is not the same as haste, and so probably needs a separate test.

I don't have access to spell data offhand, but isn't flurry a different case than haste when talking about its effect on PPM?

Flurry was mentioned in blue post to reduce rage generation per swing, so the talent "remains in line" and only increase swing damage, without touching rage generation. Thats not the same with haste possibly (im not yet 100% convinced). Still additional testing i guess would be needed for other +attack speed% effects. Maybe someone can do rage testing under heroism/bloodlust?

And Machinator, I guess we just differ in this small point. I reach 50+% crit raid buffed on brutallus, and counting hunter buffs my ap is around 4600. In my eyes that IS inflated enough so that haste starts to pull ahead.

Grim13's sheet uses average weapon speed instead of actual weapon speed.

If the formula uses hasted weapon speed
rage = (1423 * 7.5 / 274.7 + 7 * 3.6 / (1.25 * (1 + 325 / 1577))) / 2 = 27.8

If the formula uses base weapon speed
rage = (1423 * 7.5 / 274.7 + 7 * 3.6) / 2 = 32.0

Just looking for a straight forward ranked list of trinkets for a Fury Warrior. Assuming that I am not ArP capped, haste capped, or expertise capped.

Shard of Contempt, DST, MotB, Tsunami, Berserker's Call. Where do they rank?

Okay, so we are bringing up the possibility that to balance flurry, it was changed to where it (and only it) would tone down other parts of our output to keep it in line with other talents. This could explain the disproportional change in certain data.

This could be our answer but might be very difficult to prove.

There really isn't much straightforward about it, I'd advise you to put your gear in the spreadsheet.

Its the sheets that seem to indicate it doesnt. Im just saying that in your eyes or mine doesnt matter, just that either it doesnt pull ahead there or the sheets arent accurate to what we think.


"Grim13's sheet uses average weapon speed instead of actual weapon speed."
Doesn't look like it does to me, looks like it references the base weapon speed.

On testing rage generation per hit with flurry I did a little experiment

I took a holy priest into zangarmarsh where he gathered up a dozen or so mobs and tanked with with healing. I had my regular gear on, and only my 2.6 speed one hander. I had battle shout and rampage going (for the most part) to bloat the rage number as much as I could.

Controlled situations: I would only hit the mob in between anger management ticks, not taking any damage, one for each of a crit without flurry, a hit without flurry, a crit during flurry, and a hit during flurry.

Here are the results:

611 dam, 13 rage, hit no flurry
1162 dam, 25 rage, crit no flurry
1461 dam, 29 rage, crit with flurry
541 dam, 12 rage, hit with flurry

If we plug these numbers into the rage gen forumla using weapon speed instead of attack speed we get the following:

.01365 * (611) + 3.5 * 2.6 / 2 = 12.89 -> 13
.01365 * (1162) + 7 * 2.6 / 2 = 24.96 -> 25
.01365 * (1461) + 7 * 2.6 / 2 = 29.04 -> 29 *pre-flurried
.01365 * (541) + 3.5 * 2.6 / 2 = 11.54 -> 12 *pre-flurried

From what I know, then from these controlled tests and conclusions, that rage is rounded to a whole number and also the rage per hit formula uses weapon speed, and is not modified by any factor in any way if flurry is active or not to give less rage per hit.

I dont understand where the confusion from "flurry lowering rage per hit to keep it on par with other talents" comes from. Can someone find the blue post?

I agree, I'm pretty sure that Grim13 never changed anything in the rage generation calculations when he started modding drALLCOM3's sheet, and that sheet definitely uses base weapon speed.

About finding the dodge chance for bosses, why the need for a "controlled environment"? Dodge is not as complicated to test as miss chance was. The dodge chance is affected by a lot less factors than miss chance so using the wws from many boss kills with the same amount of expertise should be enough. The only thing I can think of that might influence things is the idea that casting mobs can't dodge (has that been properly tested? it seems to be that way to me if I look at some Aran logs from some time back, I get very few dodges vs him).

I've never included a rage loss through flurry.

Found something rather strange with your Spreadsheet and weapons.

Anyhow. I have Talon of the Pheonix(MH) and Blade of Infamy(OH) and wanted to check if the badge fist were any better then the one I have.

When I equip Vanir's Left Fist of Brutality and Vanir's Right Fist of Brutality, and keep equipping them over and over again, my Overall DPS rises aswell. Even though I am equipping the same weapons.

Read the directions on the first page. Drop down lists that have several pieces of data that update into other boxes usually take a few re-entries for the result to be correct.

Anyways, in the new release there will be elixir of demonslaying v demon, a new selectable armor type (9200 brutallus? havent faught him so its a stones throw), deathwish option, and drums of war rotation option.

Brute force FTW.

I know this question will make me a super noob but i cant get the 2.4 spreadsheet to download... any advice to help me dl it? my comp wont read the file

Not sure what's your problem, I can download it just fine, just save the file and open it. You need OpenOffice to use it though, Excel will not work as far as I know.

I'm very sorry, as I'm sure this has already been covered a multitude of times and everyone who has been theorycrafting with spreadsheets for any time at all already knows this, but I have a question. When I select an item in the spreadhseet, say, Dragonstrike as my mainhand, and then select a different one, say, Vanir's Right Fist of Brutality, the DPS numbers change. That's reasonable, I understand that's the whole point. But when I select Vanir's Right First again, the numbers change further. And maybe even further if I select it again! Am I doing something wrong?

EDIT: I read above, I'm a tard, please ignore the post.

I'm a little lost on what you did between this spreadsheet and the previous update that you had, but I was testing it to see if [Vanir's Right Fist of Brutality] was overall more DPS then my [Dragonstrike].

In the previous sheet, [Dragonstrike] had about 1 less DPS on the weapon then [Vanir's Right Fist of Brutality] and the Fist added about 10 DPS to my total DPS.

Looking again today, and downloading the update on April 1st, I'm seeing the opposite, [Dragonstrike] has about 2 more DPS overall then [Vanir's Right Fist of Brutality] does.

I was hoping you could clear up what you changed so that I can really have the answer to my quest, which is, which weapon in the main-hand will generate more overall DPS.

Did you happen to read the post above yours? Same weapons, could be the same problem.

I did read above about the re-equiping, and I do that to both weapons. That just simply refreshes the numbers and gives you the actual total. That happens with all pieces of gear in the spreadsheet for me, so I just refresh it a few times and the problem is solved, but what doesn't explain why [Dragonstrike] was rated lower last sheet, and higher on the current one.

I'm just wondering if something changed, and if something did change, was it accurate? Is [Dragonstrike] better then pretty much all other Main Hand weapons?

it seems to scale better then anything but warglaive basically. That and lack of good MH weapons without agility or hit on them other then S3.

Cyclical behavior of [Dragonstrike] may be a normal occurence. The ups and downs of Haste SEP and its relative proc / uptime with different levels of gear can cause this to happen. For exmaple when I put in my gear, in which I cleaned up and got 5 upgrades this week /cheer!, and I switch in [Dragonstrike] from [Vengeful Gladiator's Slicer], the sheet grants me an additional 4.75 DPS. I'm not suprised because what this means is that your current gear is what is defining the power of its proc.

Since I have no passive haste gear, the SEP rating of haste was .937 with [Vengeful Gladiator's Slicer].

A cool thing that happens is when I equip [Dragonstrike], the SEP of haste goes down, and the SEP of all the stats that [Vengeful Gladiator's Slicer] had went up.