roth wrote:

Dev-Ngreth wrote:

Rasper_Helpdesk wrote:

Dranatos wrote:

from what i read - with weapon affinity aaxp you can proc at 3 times/minute on average. and innate proc has priority roll over augged proc and you cant proc both on same swing.

so with chaotic strike proc weapon augged with damage aug - you will proc chaotic strike 3 times/minute.

with chaotic strike weapon augged with enraging blow - will you also proc combination of CS and EB for total of 3/minute? (example 2 CS, 1 EB)

OR you will proc CS and EB total for more than 3 times a minute - ( example 3 CS, 2-3 EB).
*disclaimer : this is to the best of my knowledge and not scripture

The way procs work is it tallies up how many total procs per minute you should get from all main hand sources, scales that based on your swing speed and checks once to see if you indeed proc. If you do it then rolls a weighted average over potential procs to see which fires.

Example, you have max weapon affinity, a CS weapon, an EB aug, Scorpion's Agony poison, and Lynx buff. You are swinging 60 times per minute (thats a 23 delay weapon with 100% haste + 30% bard overhaste). The CS should be 3 / min (2 base + 1 from WA), the EB should be 3 / min (2 base + 1 from WA), the poison is 2 / min (base), the Lynx should be 5 / min (1 + 4 from the proc mod, WA doesn't affect buffs) for a total of 13 procs per min. Each swing therefor has a 13/60 = 21.7% chance to proc something so the system rolls 1-1000 and if its 1-217, you proc. If you proc, it rolls 1-13, 1-3 you get CS, 4-6 you get EB, 7-8 you get poison, and 9-13 you get Lynx.

Main and off hand are independant.
This is about as close as to how it works as can be explained without dumping the code While the numbers may not be precisely what you get, they are within the ballpark.
I hate to dispute a Dev, but no. The now impossible "Shakerpage" illustrates why not.

The chance to proc, per round, is some rather small number which is multiplied by the weapon's delay, this allows for a greater chance to proc with a slower weapon, and a relatively consistant average number of procs per minute. Since the calculation uses the effective delay (arrived at after haste and slow effects are applied) certain effects can be used to adjust the chance to proc per round. Weapon Affinity, Combat Effects, etc. affect that chance to proc on any given round, increasing the average number of procs per minute.

Shakerpaging was the act of a warrior or berserker getting shaman slowed (75% slow, much greater chance per round to proc) and then gathering up as large a train as possible, gathering them all up into a small area, hitting a riposte disc for protection, and then using Rampage to generate one combat round on each mob. With the large number of mobs, and one combat round on each, the idea was that the AE proc on the Earthshaker (or Coral Hilted Tulwar) would fire off on enough of them to kill everything.

The point is that if procs really were measured per minute, then Shakerpaging would never have worked, no matter how many mobs one pulled. As it was, they had to nerf it by making it impossible to get a proc on a mob not currently targetted (nerfing Paladins and their healing defensive proc line in the process) in order to shut Shakerpaging down.

As I recall, the portion about how it takes the different procs and adds the chances of each one and then uses the random number to determine which procs was info given by a dev some time ago, but to where and when I do not recall (illustration : roll a 1 or a 2 to get a CS proc, a 3 or a 4 to get an EB proc from the aug, a 5, 6, 7, 8, 9, or 10 and get no proc). Also, since weapon and buffed procs can both fire in the same combat round, the second roll would not be to see whether it was CS or Lynx that fired, but if Lynx or the poison proc, and just as with the other, which range of values the die roll falls in determines which, if either, procs.

The way Rasper describes it, getting a Lynx (or Puma, or whatever) proc buff would have the impact of increasing the odds of a weapon's aug proccing, by virtue of increasing the total number of procs per minute you can get. Yet, this is not what happens.
Sorta.

A step was left out.

There is a multiplier in there for your actual weapon speed swing. So yes, the chance to proc is modified by the swing speed, to "normalize" it over slow weapons....

The Earshaker messed up the calculations be giving you an instant amount of extra attacks the breaks the "weaponspeed" curve.

the general point I wanted to be understood is that there is some math in that that puts together all of the procs you can get into one roll per swing. It is very rare that you will be even near 100% chance per "swing" so people do not "miss" procs, and unless a proc has an extremely high proc rate, multiple procs will not affect the chance of the other proc happening.

I did say that his numbers are not actual numbers.



More...