This guild is an attempt to gather all those who are interested in learning how to write scientific applications on nVidia's Compute Unified Device Architecture.
GPUs are essentially mass produced supercomputers in the sense that the multiprocessors in the cards have many thread processors, and by exploiting the cards' shader pipelines, we can perform a whole bunch of calculations in parallel for very little cost.
The trouble is CUDA code is new and can be difficult. I know almost nothing about the code myself. This guild could potentially (hopefully) serve as a community of people who would learn from eachother.
