Proving that code works in general does not seem to be a fun task
dtorresr
When building the first photorealistic renderers, were they only able to verify that it worked by attempting to render real world examples? I can image trying to render something like glass, have it look and "feel" right due to some visual warping, but in reality it is completely wrong. Like in our case with Pathtracer, I'm sure some of us ended up with renders that looked realistic, but were far off from the reference image.
Proving that code works in general does not seem to be a fun task
When building the first photorealistic renderers, were they only able to verify that it worked by attempting to render real world examples? I can image trying to render something like glass, have it look and "feel" right due to some visual warping, but in reality it is completely wrong. Like in our case with Pathtracer, I'm sure some of us ended up with renders that looked realistic, but were far off from the reference image.