Masters Dissertation

An Exploration of Performance and Optimisation in Procedural Terrain Generation

Abstract

In my project I have focused on exploring different optimisations for improving the performance of procedural terrain generation. To best demonstrate and evaluate these optimisations, I have implemented them into my own procedural terrain program, I have then recorded the performance of the program after each change to view its impact. Using this data, I draw conclusions on which optimisations make the largest performance improvements, compared to the difficulty of implementing them into a given system, to determine how each optimisation is best used. I aim to make generalised statements commenting on the effectiveness of the optimisation, as well as specific remarks such as how different implementations change performance, or what kinds of procedural generation algorithms each optimisation works best with and why.

Github Source Code