I saw a captcha with dice thread on hacker news. The general idea
of the captcha was to find the picture of dice with the right sum on it.
I was wondering how hard it would be to generate several thousand of
these dice images, and how long it would take using POV-ray.
POV-ray is the Persistence of Vision Ray Tracer. It takes a text based
description of the scene built using various primitives like spheres and
boxes to produce a ray traced image. The POV-ray language is Turing
complete and allows some really advanced programming features, such as
macros, loops and conditionals, which I take full advantage of for this
scene. The dice are even transparent.