Radiosity: Textured or Not?
I''m fairly deep into a recursive lighting algorithm that bears a striking resemblance to radiosity, but I''m unsure as to to relevance of texture maps during the calculations. My theory is that having the materials (stone, metal, etc) present while testing will have an effect on the solution, as the reflectance of a surface changes with respect to the colour of that surface, as set by the texture mapped to it. Is this correct, or should I be performing my calculations based on plain surfaces with a predefined reflectance? I''ve tried both methods, and they give different results. I''m happier with the results generated by the mapped system, although both systems are not physically accurate at the moment, so need adjusting. Or should I just follow this to the end and use the one I prefer? My aim is lighting that''s physically realistic if possible.
In radiosity, you generally compute the lighting using all texture maps applied. As you said, the spectral surface reflectance (or more general: its BRDF) is not guaranteed to be constant over the entire face. In fact, texture mapping is nothing but a mathod to change this reflectance without introducing additional geometry. Thus, in order to get a realistic lighting solution, you should definitely take those maps into account.
in tsome radiosity applications as final render plugin for 3d max and mental ray plugin for 3d max you are able to control the emmisivitty of each object to control the appearance of each body according to its material type.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement