We are on the cusp of a rendering renaissance. Although GPUs have traditionally been used to drive hardware rasterization pipelines, and GPU computing has ventured into many different fields as a general computing platform, it is not to be forgotten that computer graphics stands to greatly benefit from these new programming models.

As GPUs continue to become more powerful and flexible, we will see an explosion of hybrid rendering techniques running on these processors: graphics pipelines that do not rely on a single rendering algorithm to produce their final images, but rather use the algorithms that are best suited to the needs of the desired images. As we are more and more able to use the right rendering tool for the job, the results will be spectacular, and GPUs will take us there.

We have begun to see the components of these pipelines being created, several of which are detailed in this section, and as new algorithms are developed, we will continue to experience increases in rendering quality and speed, enabling a new generation of fantastic imagery.

Chapter 25 , written by Geist and Westall, describes a method using Lattice Boltzmann methods for simulating light transport. The work formulates the scattering of photons through a scene as a diffusion process that can be efficiently solved in parallel on the GPU.

In Chapter 26 , Novak et al. describe a system for efficiently implementing randomized walks on the wide execution units of the GPU. They combat the main problem of these randomized walks, low processor utilization, by regenerating paths on the fly in otherwise terminated threads and apply the result to path tracing.

Bickel and Lang, in Chapter 27 , describe an animation system that combines sparse motion-capture data with fine wrinkle models to produce realistic facial animation. By parallelizing the operations required to construct a facial animation and running on the GPU, real-time rates are achieved.

Chapter 28 , by Huang et al. describes a rasterization pipe-line implemented in CUDA. This pipeline, similar to traditional hardware rasterization pipelines, renders triangles at real-time rates and is specifically optimized for handling order-independent transparency.