Always glad to help!

I came up with this system whilst trying to create "Treasure", in which
I've made my first attempt at subsurface-scattering for realistic hands
and a woman (see my website's gallery), and fumbling with density,
absorption and scattering I found the combination these can make. Its
become very useful for me since then, I don't know of any other approach
which is quiet as flexible and (for me at least) intuitive.

I might put this on my "Hints" section, cause subsurface-scattering
seems to pop up now and then on the newsgroups...


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

> Tim,
> Thanks much for this information! I have never been able to get scattering media
> to work well for me in other than one unit containers - yes, I divided the
> density by the scale. The trouble is always that the media gets far too bright
> at settings I 'think' should be perfect. When I adjust the scattering to get the
> brightness right, the media looked simply awful. I never thought to try
> adjusting the absorbtion to dim the media! I'll have to play with this later
> today.
> Regards, Bill P.
>
> >
> > It is different!
> > If the volume is denser, this should have different effects, like being
> > absorbed faster, but since we've got more "particles" to light, it also
> > gets brighter. Hence I use absorption to compensate for that.
> > Its quiet difficult to explain, actually, and you should experiment with
> > it. Here's a code snippet for you to start with (its just the media),
> > you might want to place it in a sphere on a plane... :-)
> > And then just fumble with the settings, and modify the code.
> >
> > //SubSurfaceScattering
> > #local SSS=2;
> > #local SSS_Color=<1,0,0>;
> > #local Extra_Absorb=1.5;
> >
> > #local SSS_Media=
> >   media{
> >    scattering{2,vnormalize(SSS_Color)*SSS}
> >    absorption (vnormalize(<1,1,1>-SSS_Color)*SSS+Extra_Absorb)
> >    emission Extra_Absorb*vnormalize(SSS_Color)
> >    density{rgb SSS_Cloud}
> >    method 3 intervals 1 samples 10,30
> >    }
> >

Post a reply to this message