Ken wrote:
> 
> No, I don't. I have never seen any situation in real life that this
> would represent.

How can you compare then? 

I have seen several photometric images of physical setups and
corresponding rad simulations in IEEE and other publications
which clearly show the validity of radiosity. There are differences
in implementations, like noise in the ray based radiosity, but the
basic concept is solid.

The problem with all radiosity implementations is that there are too
many cryptic parameters to tweak so users in general use non optimal
setup which ruins the simulation. There is an EU project ARCADE which
tries to automate the parameter tweak phase among other things so that
you don't have to be "Certified Radiosity Technician" to produce good
results. http://www-imagis.imag.fr/ARCADE/

This particular scene is a poor choice for meaasuring how realistic
the simulation is for two reasons. The "sky" emits the same amount of
"energy" as the "sun". Nobody has ever seen this situation so it's a
bit difficult to compare. The scene is also uniform in all directions.


There is something funny going on with background{} and radiosity.
I rendered the same scene enclosed in a sphere and the anomaly at
the bottom of the sphere disappeared.
http://hammer.prohosting.com/~kkivisal/sphere.jpg


______________________________________________________________________
Kari Kivisalo                                  http://www.kivisalo.net

Post a reply to this message