|
Originally Posted by dreadnor,April 28th, 2006 @ 7:45AM
I havn't seen Kalmans spreadsheet/model but I am yet to see an accurate model of str vs crit that factors in flurry correctly.
Most the math you see on the issue is flawed in some way. For example, most people equate crusader to 25 strength (100 x (15/60)), however it should be slightly less since you can get overlaps.
|
Actually, it should be more on the mainhand, since you proc it more frequently than once per minute. Depending on +hit and weaponspeed, for DW (assuming all attacks are Slice and Dice/Flurried) Crusader can scale pretty high; for 10% +hit and a 2.8 speed mainhand weapon, a rogue will proc Crus enough to keep it up around 40% of the time.
On the offhand, SnD/Flurry assumed constantly up and 10% +hit pretty much eliminates the penalty for overlaps.
All that data is derived from a Monte Carlo simulator I wrote to test the Crusader uptime; if you have access to MATLAB (or probably on scilab, which is free, but untested for compatibility) you can run it with your own rogue's numbers/gear to get the answer for you. Send me a PM, or an email (kalman@radiationnow.net) if you want the simulator code.