I wrote my Raytracer in MS Visual C++ 7.0. The only function used to draw
something on the screen is SetPixel. There's no OpenGL or DirectX
performing the calculations...all algorithms were coded into the program,
which then would read in the scene description (sample in the upper-left
picture). The program crunched on the scene description for a while
(sometimes a very long while) and would finally spit out the image. The
image is also converted into a DIB section in case the user wishes to export
the graphic to a file.
This program was written to learn:
Ambient, diffuse, and Phong illumination with point light sources
String parsing (just on the side, for fun)
Included are three sample images the raytracer can produce (currently it can
only draw spheres, cubes, and cylinders). The top image is 4 spheres inside
a very large cube with reflections cranked up really high. This image, with
2x AA, 10-deep reflections, at 1600x1200 resolution took just a couple
minutes over 2 hours to render. The middle image was just a fun one to
demonstrate different material properties and the Phong (shininess)
difference between then. The bottom picture was the first one I did to
demonstrate all 3 types of objects interacting, as well as multiple light
sources with shadowing and reflection.
Coming up with the project, I will be adding:
And maybe a cel-shaded implementation
Let me know how you like it!