"MessyBlob" <nomail@nomail> wrote:
> Yes, you did, and yes, we like it!

I do like to hear that ;)

> About the 'dimming' at near-tangents, you might want to consider a 'conserve
> energy'-type idea to manage the transition between diffuse and SSLT lighting on
> the incident rays (if you haven't already!)

I'm not sure what you're talking about.

There is no "transition between diffuse and SSLT lighting", because SSLT
completely replaces the classic diffuse lighting model.

Both the "coupling in" and "coupling out" of light at various angles of
incidence, based on the material's IOR, is already taken into account as part
of the SSLT formulae. There's a whole host of fresnel terms in there ;)

Maybe what you mean is a "conserve energy" type of thing regarding specular
reflections on the material? Such a thing is still miles away on my agenda.

First thing to do next is make sure SSLT learns a bit more about light sources
and shadows. And then there's radiosity to be integrated. Then it'll be time to
think about patterns and such, and where to ultimately "anchor" the subsurface{}
statement (for now, it just happens to be placed in the finish{} block because
it was easiest to code that way). And caches implemented for optimization. And
the RNG noise issue to be solved (unless the caching already does away with
it). Not to forget photon mapping. Sorting out geometry issues. And... well,
you'll get the point I guess ;)

Post a reply to this message