Ok, here are my equations. They are probably a bit too idealistic, but I think they are still more accurate than the current equations.
So the average + spell damage for a trinket with Duration=D, Cooldown=C, +Spell Dmg=S, and Boss fight Time=T is as follows:
If MOD(T/C)<=D(If the Boss Fight stops in the middle, or right at the end of the trinket buff)
(DS*INT(T/C))+MOD(T/C)S)/T (The total damage increased+the damage increased in the last buff all divided by the length of the boss fight)
Else(If the Boss fight ends when the buff is not applied)
(DS*INT(T/C))+DS)/T(The total damage increased+The total damage increased in the last Cooldown cycle of the trinket all divided by length of boss fight)
When I say "the total damage increased" it is basically (the Spell damage given by the trinket)*(the duration of the effect)*(the total number of times that the trinket is activated). That value does not really manifest itself in game, but it is useful for calculations. I know this is a bit confusing, but let me know if I am dead wrong in my Theorycrafting, or if you don't understand it. Thanks! (Vontre I have these formulas in excel format too if you want them.)
Current theorycraft already takes those kings of factors into account plus much more such as cooldown stackings.
found a bug in the gear import engine:
The first column of the "Pants" isnt correctly imported - when you import for example "leggings of the chan. e." int becomes spirit, spirit becomes fire and so on
I was playing with the magegraf.com version trying to model the impact of a totem of wrath for mage vs. warlock (very nice program btw ). I did get a surprising result that I wanted to run by you to see whether or not its a bug.
If I add 67 crit rating to a shadow destro warlock, I get a 3.44% DPS increase (3% personal plus 0.4% due to ISB) which is exactly what I predicted from the math. When I do it for a mage though, increasing crit rating by 67 only gives a 2.15% DPS increase instead of the 3.3% (crit plus ignite damage) I was expecting to see. Is this a bug in the program or is it accurate? If it's accurate, would you mind explaining to me how you arrive at that value?
Also, on an unrelated note, may I ask what are the assumptions that you put into the ISB uptime calculation (ie number of spriests and aff locks vs number shadow destro)? My apologies if you've posted it somewhere else; please just shoot me a link if thats the case. Thanks for the nice tool, and thanks in advance for the help!
The old spreadsheet shows me at 1755 dps and the new beta one shows me at 1858 in the same gear with the same consumable/buff options. Also it shows that a +12 damage gem in t6 gloves is about 5 dps better than using a reckless pyrestone (6 dmg/ 5 haste) while the pyrestone nets you the +2 damage socket bonus. That doesn't seem right.