This is a learning exercise and barebones approach to software raytracing that's built on spheres and casting rays from the camera to the objects in the scene until they hit a light source. This is the opposite approach as real life but it's faster this way because you only cast the rays from the camera instead of millions that come from the light source and miss the camera entirely.
- Depth of field
- Materials:
- Lambertian diffuse: takes in attenuation color and realistically diffuses light
- Metal: takes in attenuation color and roughness and reflects light
- Dielectric: takes in index of refraction and attenuation color and refracts light, also has total internal reflection
- Anti-aliasing (samples each pixel multiple times)
- Camera position movement
- Ray bouncing
- Motion blur
- Basic sphere animation (moving linearly)
- Command line arguments:
-Size 1200gives an image that is 1200 pixels wide, like the image below-FOV 90gives a 90 degree vertical field of view. The image below has an FOV of 20 degrees-Samples 50samples each pixel with 50 rays. The image below is with 250 rays per pixel-Depth 50allows each ray to bounce up to 50 times like the image below-CameraPos 1 1 1places the camera at (1, 1, 1), with the default looking at (0, 0, 0)
This will be updated as it progresses. Three materials currently exist: Lamberian diffuse, Dielectric, and Metal. These materials interact with the rays in different ways, allowing the raycasting to bounce around the scene and attenuate and scatter according to the material's properties. If the rays don't hit a material, they end up at the skybox and carry the color back to the camera.
