Gloss Units from specularity and roughness

Does anyone have experience or hints on how to convert Gloss Units to the specular and roughness terms in Radiance, or vice versa?

I was able to find a correlation between Gloss Units and the Cooke-Torrance m parameter, but I’m not sure how to get to that from the Radiance specular and roughness terms either.


There’s a great paper by Jim Ferwerda, Fabio Pellacini, and Don Greenberg from 2001 that looks at gloss measurements and correlates them to specularity and roughness:

There is no direct correlation, as perceptual gloss is an amalgam of specularity and roughness, and therefore ambiguous by nature. (“Give me ambiguity or give me something else!”)