> .... aside from the fact that what you describe *is* cosine weighting (unless
> your approach would be wrong, that is) :)

Yea, I just meant that now we don't have to take the cosine factor into
account anymore when figuring radiance (or whatever) contribution as the
ray distribution handles it by itself. Both give the exact same result.

> doesn't sound too bad (however, performance should be compared with taking two
> random numbers and transforming them to circle distribution using trigonometry
> instead of the monte-carlo approach).

In my Monte-Carlo path tracing renderer the above method was faster than
those using polar coordinates (random phi and theta describing the
angles). It was very clearly faster. I have Core2Duo so things might be
different on other platforms / with other compilers. I use GCC.

Severi

Post a reply to this message