Quote Originally Posted by Kurokikaze View Post
Tell that to the same people that gave you your damage formulas and weapon skill modifiers in FFXI. Dozens of man hours were spent on those bunnies in Ronfaure.
But to test which variables? I'm not saying data from level 1 mobs isn't valuable or necessary, I'm saying they're a poor control group for testing the variable in question. A handful of fights doesn't give you a statistically significant dataset for level 1 mobs. Dozens of man-hours does. That's a datapoint, not a control group for dLevel. Testing against the rabbits makes sense when you're isolating your maximum damage output, but not for finding dLevel.

Quote Originally Posted by Kurokikaze View Post
Theorycrafting damage formulas are done specifically for the purpose of performing well in End-Game. Testing on NMs and mobs that are at the floor are the most important monsters that can be tested on. No one cares about your damage at any other level but the cap.
This may be true, but to efficiently deduce the formulae, tests need to be conducted changing only one variable at a time. Since monster stats change from level to level, the only way to isolate dLevel is to test with characters of different levels against identical level mobs of the same type, equipping gear and assigning character stats to keep all other character variables the same ( as you can't change a mob's stats ).

If you recall the english enmity-testing exercise, he conducted a series of experiments with the intent to disprove the japanese claim that there were two enmity values, one static and one decaying. He ended up proving it. What was "obvious" through observation was found to be untrue through rigorous testing.