DirectX 11 introduces DirectCompute as a way to access the computational capability of the GPU in a more flexible way. It opens the door to operations on more general data-structures beyond just arrays, and to new classes of algorithms as well. This talk covers the key features of DirectCompute’s compute shader, and how they can be used by developers in their applications. These features include: explicit thread dispatch, communication of data between threads, and a rich set of primitives for random access and streaming I/O operations. These features enable faster and simpler implementations of techniques already in use on the CPU.