Peter Popov wrote:
> 
> Marc SCHEFFER wrote in message <361BDC68.F842C77A@wanadoo.fr>...
> >You know, you can find some tips on how radiosity works in PoV's doc.
> Roughly
> >(please, feel free to correct me if I do any mistake due to loss of
> memory...)
> >It's a way to achieve a better result than ambient light, by
> calculating
> first a
> >"normaliz (s ?)ed" ambient value for each of a lot of subdividings in
> the
> final
> >image, and then using it to render with a correct look and feel.
> Quite time
> >consuming, but really worth if you have a powerful enough cpu.
> >
> >And again, sorry if some of my explanations here were wrong, I
> haven't the
> doc
> >here...
> 
> Actually, my question was a little bit different. Yes, I know how
> radiosity
> is implemented in POV, but it's still raytracing, since it relies on
> shooting rays, intersecting with (mostly polynomial) objects etc. I
> think
> Radiance uses a completely different approach. For example, in the
> c.g.r.r.
> faq it is said that rendering speed in Radiance is not affected by the
> number of objects in the scene (?!?!?). I am just curios, that's all.
> 
> Peter

Different programs calculate global illumination differently.  Radiance
uses Monte Carlo raytracing to determine the indirect lighting
correctly.  It calculates additional rays in much the same way as
PovRay's radiosity part.  One difference is that it is not limited to
one level of indirection.  A ray can be bounced off many objects before
it reaches the light source.  Another difference is that Radiance uses
the gradient of the surface in its interpolation calculation.  Because
of all the work that has gone into Radiance and testing it, it is the
benchmark which most global illumination models are compared against.

Radiance: http://radsite.lbl.gov/radiance/HOME.html
Radiance lighting model: http://radsite.lbl.gov/radiance/refer/long.html

-Eric Brown

Post a reply to this message