Nvidia has recently released a new version of its CUDA programming tools, this time for developers using Mac OS X. CUDA tools will help them use the graphics processing units, or GPUs, on newer Nvidia graphics hardware as parallel processing engines, according to Yahoo News.

CUDA basically lets programmers offload data processing to the graphics processing hardware found on Nvidia's high-end GeForce graphics hardware, using a dedicated driver, written using C language subroutines. Going deeper into the process, programmers using the new software will be able to use the cards to process other data than just graphics, all this without having to learn OpenGL or how to talk with the card specifically, according to the same source.

"Transcoding video is a good example," said Keane. Transcoding high-definition video from one format to another is an extremely processor-intensive task, and it's one that scales very well when you can distribute it to multiple processors. With 128 processors on an Nvidia GPU working in tandem, that task can be sped up by orders of magnitude rather than forcing the CPU, even a multiple core CPU, to handle it alone.

Supported hardware includes GeForce 8800GT and Quadro graphics cards (available as configure-to-order options on the Mac Pro) and GeForce 8600M graphics chip (Nvidia's latest) available with Apple's updated MacBook Pro.

The Mac OS X version of CUDA is specific to Nvidia graphics systems, which means that ATI or Intel integrated graphics hardware (found on other Mac models) are out of the question.

As CUDA doesn't make all software on an Nvidia-equipped Mac run faster just like that, programmers need to hit just the right spots and optimize to process data in parallel. However, as Andy Keane, general manager of Nvidia's GPU computing business, puts it, programming support for CUDA can make a big difference nonetheless.

CUDA beta tools are already available for free download from Nvidia's website. The toolkit and the Software Development Kit (SDK) are working with Mac OS X v10.5.2.