NVIDIA unleashes CUDA offensive

As AMD is rolling out the Radeon HD 4870 X2 today NVIDIA has prepared a CUDA and PhysX PR offensive in an attempt to spoil the party. NVIDIA shows off the power of its more than 80 million CUDA-enabled GeForce 8, 9 and 200-series GPUs with Folding@Home and the first trial version of the Badaboom video transcoder software from Elemental Technologies.

All of the 80 million plus GeForce 8 Series and higher GPUs in the field are CUDA-enabled, the largest installed base of general-purpose, parallel-computing processors ever created. The same GPU architecture that delivers stunning onscreen computer graphics in video games is also ideal for many other types of applications. The latest generation of NVIDIA GeForce GPUs offer up to 240 processor cores, compared to a maximum of the four cores found on the highest-end CPU. Any process that can be divided into multiple elements and run in parallel can be programmed to take advantage of the massive processing potential of the GPU.

NVIDIA first released its CUDA programming technology in 2007, providing software developers a programming environment based on the industry-standard C language for easy creation of applications running on NVIDIA GPUs. Numerous commercial and scientific applications have adopted CUDA technology and now consumer applications are starting to emerge that take advantage of the technology.

"CUDA has the potential be a disruptive force in both the GPU and CPU industries," says Anand Shimpi, CEO and editor-in-chief of AnandTech.com. "Apps like Badaboom, that solve significant problems for the home PC user, could give NVIDIA hardware a significant advantage over other GPUs and it points to the need for consumers to optimize their PCs so they have both decent CPU and GPU power."

Elemental Technologies' Badaboom is a video transcoding program that converts video files into other formats. For example, the program can convert an MPEG file to play on an iPod or other portable device. Video transcoding can be one of the most time-consuming tasks in home computing. Converting a two-hour movie, for instance, can take six or more hours when using the computer's CPU. However, with Badaboom on the GPU, the conversion process can be up to 18 times faster than traditional methods, getting the job done in a few minutes and, in the meantime, also freeing the CPU to handle other tasks like email and Web browsing.

Tackling the intense processing demands of ongoing medical research, Stanford University's Folding@home distributed computing program, gives consumers the opportunity to share their computer processing power in an effort to help find a cure for disease. Running up to 140 times faster on an NVIDIA GPU over a CPU, Folding@home makes use of idle computer cycles to perform scientific calculations. Folding@home studies protein folding, where proteins in our bodies assemble themselves. Biologists simulate protein folding in order to understand how proteins fold so quickly and reliably, and to discover what happens if they do not fold correctly. Diseases such as Alzheimer's, cystic fibrosis, BSE (Mad Cow disease), an inherited form of emphysema, and many cancers are believed to result from protein misfolding. The Folding@home client is a free program that runs in the background of the PC, allowing ordinary people to have a real impact in the search for a cure of these diseases.

You can download a 30-day trial version of Badaboom video transcoder at The Force Within.