May 24, 2012 -- Today's graphics processors are highly programmable, massively parallel compute engines. In this role, they are commonly called general purpose graphics processing units, or "GPGPUs." You can program them with the open and standard based OpenCL framework, distributing compute chores to CPUs, GPUs, and DSPs to optimize a system's overall performance.

What makes GPGPU computing so enticing is the availability of extreme floating-point performance in cost effective GPUs.

By Peter Mandl and Udeepta Bordoloi. (Mandl is Senior Product Marketing Manager, Embedded Client Business, Advanced Micro Devices, Inc.; and Bordoloi is Senior Member of the Technical Staff at Advanced Micro Devices.)