Building systems capable of replicating global illumination models within limited time-frames has long been one of the toughest conundrums plaguing computer graphics researchers.
A new and innovative method proposed by Cyril Crassin et al. in 2011 appears to provide a well-disposed balance between performance and physical fidelity.
Termed voxel cone tracing, the method utilizes seemingly uncorrelated techniques well established in the realm of computer graphics to achieve hardware accelerated indirect illumination.
This thesis explores the differences between the underlying mathematical model and an actual software-based implementation of this algorithm.