> There are no particles, media only considers total density over a ray.
> The closest thing to a "particle" is a single media sample. Intervals
> and sampling methods simply control number and location of samples.

Thats what I thought. Probably misunderstood what you were saying
about samples.

> Multiple medias add together, multiple densities multiply. That part of
> the code is quite simple and clear, and the documentation backs it up.

No argueing about that one. I was thinking if it doesn't change in effect
if I have e.g.
scattering{1,1} emission 1 and density{rgb 5} versus
scattering{1,5} emission 5 and density{rgb 1}.

> But the only places those values are used are as multipliers for the
> density, the only place the density is used is where it is multiplied by
> those values, and I am certain that was what was intended. If the
> results are not the same (you implied you had tried it and got different
> results), it most likely is a bug caused by something I'm not seeing.

Hm, so you ARE the guy with insight, eh? :-)
I'm not sure if there was something I actually felt as "difference in
look" or if it was more like "my idea of an intuitive approach to
subsurface-sampling" that became my combination of scattering/absorption/
emission/density.
I guess I should make some quick tests for that...


--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde

Post a reply to this message