January 2007 saw the end of the beta for a new library known as the AMD Performance Library (APL). This set of low-level software routines, covering everything from basic math to image and signal processing, dramatically increases parallelism in thousands of functions. When APL functions are called with large data sets (e.g. arrays with thousands of elements), they enhance your code by implementing the parallel threading techniques mentioned here, so that you don't have to worry about it. Since APL is built specifically for AMD's x86 processors, it will help you optimize for the AMD multicore platform in less time, with better results, than hand-coding your threads. It is optimized for non-AMD processors as well. Right now, it's available as a static library for Visual Studio 2005, in both 32- and 64-bit flavors, as well as GCC 4.0, and as a dynamic library for both Linux® and Windows®.

Read "Image Processing The Easy Way" by Brent Hollingsworth for an impressive side-by-side comparison of a zoom function using single-thread, multithread, and APL coding techniques. Also learn more about APL in general and keep up on the latest APL news.