PPM formula testing

[Kai Moon]

Sources disagree on PPM area factor formula. Old dev posts on Paragonwiki say one way, SCoRE source code and @Bopper's spreadsheets say another way, an i24 change according to Bopper. Same sources disagree whether the final chance is clamped to a minimum nonzero floor.

 

So I ran two hour-long experiments with a level 50 ice/ice dominator on HC test server to confirm which sources are correct for HC.

 

Experiment 1 - Unenhanced

 

Frostbite with no recharge enhancements and a Trap of the Hunter proc.

Observed: 240 procs in 1673 hits

i23 Predicted: 5.5 area factor, 10.68% proc frequency

i24 Predicted: 4.375 area factor, 13.43% proc frequency

 

Binomial test on observed frequency gives z=4.81 compared to i23, z=1.06 compared to i24. 5.5 area factor is definitely wrong, 4.375 area factor could be right.

 

Experimental 2 - Enhanced

 

Frostbite with way too many recharge enhancements, a spiritual core paragon alpha, and a Trap of the Hunter proc.

Observed: 305 procs in 3161 hits

i23 No Clamp Predicted: 5.56% proc frequency

i24 No Clamp Predicted: 6.99% proc frequency

Clamp Predicted: 3.5 PPM, minimum 10.25% proc frequency

 

Binomial test on observed frequency gives z=9,99 compared to i23 no clamp, z=5.83 compared to i24 no clamp, z=-1.09 compared to clamp. No clamp is definitely wrong, 10.25% clamp could be right.

enhanced-log.txt.zip unenhanced-log.txt.zip