NVIDIA’s Developer site contains the NVIDIA Cg SDK and plenty of demos, white papers and presentations to help you program NVIDIA GPUs.

The MSDN DirectX page is a good starting point for DirectX documentation and tutorials.

Scott Larsen and David McAllister of UNC Chapel Hill describe the use of GPUs to perform large matrix-matrix multiplies. (Fast Matrix Multiplies using Graphics Hardware. E. Scott Larsen,
David K. McAllister. Supercomputing 2001 (Denver, CO) November, 2001.)

The 3DLabs Developer Support Site has some documentation on OpenGL (including OpenGL 2.0). Their main developer site requires registration.

Matrox’s developer site has information and SDKs for programming their GPUs.

Harris et al. at UNC Chapel Hill use programmable graphics hardware for real-time visual simulation of diverse dynamic phenomena. The simulations use an extension of cellular automata known as coupled map lattice (CML). (Follow this link for more info, papers, videos, and demos.)

ATI’s Developer site contains lots of reference material, demos, and utilities to aid in programming ATI GPUs.

Strzodka and Rumpf at the University of Duisburg use GPUs to solve PDEs. Applications include nonlinear diffusion and segmentation in image processing. Also of interest is a method for doing 16-bit computations with 8-bit RGBA textures. Follow this link for an overview and several papers.

Chris Trendall and James Stewart of the University of Toronto explore the mathematical functionality available in graphics hardware by using it to perform a non-trivial calculation. (General calculations using graphics hardware, with application to interactive caustics. Chris Trendall and James Stewart. In Proceedings of Eurographics Workshop on Rendering. 2000.)

OpenGL.org is the place to look for OpenGL news, specs, documentation, and other details about OpenGL programming.