I have always wanted to build my own raytracer. When I took my first steps as a primary school kid in programming I (like many others at my age) wanted to create computer games. This started me off to computer graphics after which I've done some amount of games and graphics programming. After graduating I drifted off from graphics programming and have been doing various other sort of coding ever since. However, the thirst to write my own raytracer still exists unquenched.

First thing is to get an image rendered. I selected TARGA since it seems like an easy format and supports uncompressed, raw binary. Pixel content is just a gradient from black to full blue-green in a whopping 32 x 32 resolution.

I decided to follow this tutorial on the first steps of implementing the ray tracer.

First thing to do was to render one simple object on the image by casting rays.

In order to make the red blob in the previous phase look like a sphere a rudimentary lighting model is in order.

As mentioned in previous phase the ray-sphere intersection function (calculateSphereIntersections) returned a list of intersection points even though it is enough to be aware of the closest intersection point only. Also, as pointed out by Boojum in the related reddit thread dynamic memory allocation in ray tracer inner loop hinders performance unnecessarily.

Until now there has only been one sphere in the scene at a time. In this phase second sphere is introduced. The solution scales to any amount of spheres though.

This phase introduces reflections.

In this phase we introduce ambient light and specular highlights using Blinn-Phong algorithm.