CS 5620/6620 - Fall 2025 - Rendering with Ray Tracing

Project 11 - Monte Carlo Global Illumination

<< Back

Jake Wagoner

Wed, Nov 12, 2025 @ 10:57 am

Caustics work by default, bounce limit set to 2 from the beginning makes caustics possible right away

Wed, Nov 12, 2025 @ 10:33 am

Uniform hemisphere rejection sampling

Wed, Nov 12, 2025 @ 9:57 am

3:52 render time, 32 samples per pixel for gi, 16 samples for shadows, cosine-weighted importance sampling

Tue, Nov 11, 2025 @ 11:54 am

Tweaked some scene values, 32 samples for gi, 16 shadow samples, all cosine-weighted importance sampling. 261s for render

Tue, Nov 11, 2025 @ 10:55 am

Emissive sphere added, looks great! 360p render in 60s Main thing I notice is that the reflection with the rear wall is brighter than it should be and Im not entirely why

Mon, Nov 10, 2025 @ 4:45 pm

Increasing attenuation to 4 provides a better looking scene. Still need to do the emissive surface and figure out why the reflective sphere is very dark.

Mon, Nov 10, 2025 @ 4:38 pm

Attenuation makes a difference, but the scene is still very washed out for some reason

Sat, Nov 8, 2025 @ 9:38 am

Gloss re-applied and changed the emissive sphere to blue temporarily. Looks like the blue is in-fact showing on the right sphere. Intensity of the light turned down to 1, since I havent done gamma correction yet.

Sat, Nov 8, 2025 @ 9:05 am

Updating the seeding and increasing sample count helped with the banding. Lowering the soft-shadow count decreased render time substantially, but the shadows look much worse. Needs gamma correction, emissive material property, and the light source render.

Sat, Nov 8, 2025 @ 8:52 am

An interesting first result. Looks to me like the secondary ray is being stopped prematurely. Beyond some other things that are clearly wrong with this set up.