Fiery Weapon is a damage proc. It hits the enemy for 40 damage, that's it. There are no "gains" or "losses."

edit: err, nevermind, I just realized you were talking to the person above me.

Ahh right, I knew that. Temporary lapse in memory there. It is still interesting to say the least, that white damage procs in my test acted in a very different way than the other.


This brings up a new question: if procs on Yellow hits remain constant, then would a hunter operating under haste see a direct increase in proc rate because our yellow hits directly correlate to our white hits? Other classes are limited on their Yellow Hits based on energy/rage and the Global Cooldown, while hunters are limited only by shot time (until they get it low enough to reach the GCD).

Example: at 2s a BM sees 5 white and 5 yellow hits in 10 seconds. If he hasted to 1.65s, he would see 6 auto, and 6 yellow. With the white hit PPM remaining constant, while yellow attacks have a percentage based proc rate, he would see a 20% increase in yellow hit procs or a 10% total increase in procs.

A MM generally has 6 specials/4 autos per 10 seconds, so would a marksman see even more procs than your average BM?

This is going to need more testing.

Some Live testing - looks like the behavior is the same as on the PTR. I'm going to update the original post with a summary of the tests done so far.

I disagree. I definitely remember developers using ppm in discussing effects on the forums. Distinctly, I remember them using ppm terminology explaining paladin changes.

What % in haste rating does your ammo pouch give you?

I came to the conclution that your ppm went down with approxx 15% with your ammo puch equipped. I aint brilliant with my calculator that i shall admit, but
100-((1,1/1,3)*100)= 15,39.

It is long since i went to school :P so if my calculations are all wrong pls tell so, and i will remove the post next time i checks.

I dont think theres any need for me to do this test again on the PTR, it clearly shows from everyone elses tests that my live test is just broken. Sorry to have wasted anyones time.

Ammo pouch is 15%, the Quick Shots (improved Aspect of the Hawk) haste is also 15% and lasts for either 10 or 12 seconds (I forget off the top of my head, but the uptime is around 30-40% if I remember correctly).

I would not say that, it gives us a counter point to analyze and figure out why your test showed an increase in procs while others are not. It might tell us something we were unaware of, or it could have to do with how Slice'n'Dice works perhaps.

This is something I was considering as well, but I had Quick Shots proccing which is a temporary haste similar to the way the DST or Slice'n'Dice procs work. Even if my quiver was reducing the proc rate per shot to maintain a constant PPM, Quick Shots would have increased it if your theory is correct. Perhaps Slice'n'Dice alone works in a different way? It is the Rogue FAQ that people referenced when they said hastes increase proc rates.

Also, since Bloodlust reduces the Global Cooldown, if you have enough energy/rage you might be able to get more procs from having more yellow hits. This would require the proc percentage off yellow hits to be unaffected by haste.

Maybe so, but you did it, which is more than people before you had done. You showed your data, and obviously your data existed. You hit a fluke in testing which had odd results, but that happens. I'd suggest you never apologize for testing and putting in the effort. We know how much it sucks to sit around whacking on a mob with no real gain to it, and appreciate when people do that. And yeah, I'll say "we" and assume that other people feel the same way I do on that topic

I just want to add that there is no guarantee that a proc from classic (fiery) uses the same mechanics as a BC proc (mongoose, hunting rifle). Also, there could be a different mechanic between procs with a duration and procs with an instant effect; or there could be an individual flag for each proc "adjust (white?) PPM according to haste" for all we know. Not saying that there is, but when testing one has to consider all possible variables. I would not discard the fiery test as a fluke yet unless there are more test with fiery and different results that I somehow missed.

No matter if PPM is a player coined term or not. It is only a concept. An idea.
It is not implementable in any viable form (unless you want to go heuristics). Hence the flat proc chance on hit.

PPM was just meant to express that this flat proc chance on hit did depend on weapon speed (base was assumed back in the day).

I don't think your test was broken (unless you want to call testing an outdated proc "broken").
It does show contrary results however.

It seems to me the best way to test PPM mechanics is to test something with a very high proc rate. Seal of Vengeance, the alliance only Paladin seal is 20 ppm. On live it procs every swing with a 3.0 speed or slower weapon. So, get a shaman, bloodlust with a 3.0+ speed weapon on live, hit some stuff, repeat on PTR.

Edit: Went and gave that a shot, used bloodlust several times and even with a hasted weapon speed of 2.5 Seal of Vengeance always applied. Be nice to get a bigger sample size, but I'm sleepy, maybe Seal of Vengeance is not affected by haste, maybe some procs are changed to not check the current weapon speed and some are not?

Unbridled Wrath used to be a fixed percent chance (40% if I remember correctly), but it was changed to become weapon speed dependent. That might explain the "heavily weapon speed dependant" comment you recall. Its Flurried behavior was the same on the PTR as my healing trinket procs, and it had nearly the same PPM for weapon speeds ranging from 1.3 to 3.7, so I'd expect it to follow standard PPM mechanics. However, when I get home tonight I will go and look at the data for my healing trinkets from my Live test.

Just a question for rogues.

Sword spec use PPM mechanics or his 5% are regular (haste rating & SnD) ?

Sword spec is a % mechanic, % is not ppm. If it tooltips with a % then it isn't ppm.

If course it is easily implementable. All the test so far show that the proc percentage is adjusted by your hasted weapon speeds. Special attacks don't appear to be included in this so the calculations could be very straightforward.

PPM * (Weapon Speed / 60) = Proc Chance

It is looking like weapon speed is after hastes are applied, rather than before as previously thought. So proc chance changes over time as your weapon speed does.

We already know that some effects are PPM based and some effects (like Combat Potency) are percent based. I think what the "Fiery" experiment showed was that "Fiery" is still a percent based proc. It should remind us that there may still be such percent effects left in the game and we should be taking that into consideration.

Yes. This would cover the autoattacks.
Special attacks are what makes PPM in its strict form difficult to implement.

To add something wothwhile to the discussion:

I did some autoattack tests in Blasted lands simliliar to the ones Baldycenarius did, and here are the results (in textual form; I can provide screenshots to back them up):

Test A) I did not skill into flurry. Addditional haste Rating equipped on char: 50
Test Length 41.75 Minutes
#Hits 729
#Mongoose procs 44
#heroism card procs 81

Test B) 5 points in flurry. Additional haste rating equipped on char: 88
Test Length 44 Minutes
#Hits 923
#Mongoose procs 47
#heroism card procs 84


As one can deduct from the data, I was hasted by 20% in Test B) relative to Test A)
The number of procs stayed nearly the same. Though the data quantity is to low to conclude this evidently, it is way more probable that PPM in both cases is based on current weapon speed for both procs (Mongoose and Heroism Card).
The ratio hits/proc went up by ca. 20%, in line with the value that my haste increased.

This was on live.

Thats strange, when I was leveling my pally I noticed it to be PPM based. I didnt run proc meters or anything, but after using around 10-15 blue weapons all enchanted with Fiery, it was pretty obvious between, say, a 2.4 speed mace and a 3.7 speed sword. The slower weapon had a much, MUCH higher proc % on hit. This was especially evident with SoC too (fiery procs off it).

From the same run as the data from post 78, on the Live servers. Testing [Darkmoon Card: Heroism] (WoW 1.0 item) and [Mark of Conquest] (WoW 2.0 item).

I'd be interested what your swings/second come out to Rezarel.

Basically, flurried and unflurried weapon speeds.