HOME | DD

Description
A funny variation on the cornell box scene. Lighting is from an environment map leaking in through the hole in the top of the box and the open from (where the camera is); the strongest light is actually behind the box, so mostly blocked.

The walls are quite noisy even with a rather large number of samples because the environment map's PDF almost completely fails to match the actual illumination (most of the environment map is blocked by the walls). BRDF sampling comes to the rescue of the glossy dragon, but the lambertian walls look rather sad... I'm not really sure how to handle this sort of situation well.

snogray -s1275 -e2 -a3 -Rlight-samples=225 -Rbrdf-samples=225 test:mesh407:+dragon.msh -b ../envmaps/debevec/pisa.exr ../images/dbox2.jpg

Description
A funny variation on the cornell box scene. Lighting is from an environment map leaking in through the hole in the top of the box and the open from (where the camera is); the strongest light is actually behind the box, so mostly blocked.

The walls are quite noisy even with a rather large number of samples because the environment map's PDF almost completely fails to match the actual illumination (most of the environment map is blocked by the walls). BRDF sampling comes to the rescue of the glossy dragon, but the lambertian walls look rather sad... I'm not really sure how to handle this sort of situation well.

snogray -s1275 -e2 -a3 -Rlight-samples=225 -Rbrdf-samples=225 test:mesh407:+dragon.msh -b ../envmaps/debevec/pisa.exr ../images/dbox2.jpg