Leading graphics chipmaker Nvidia on Monday introduced its new lineup of GeForce GTX 200 graphics processors. They are expected to be available in graphics cards later this month.

The GTX 200 GPU, according to Nvidia, “is the most advanced GPU ever built.” Instead of requiring more than five hours to convert a video for an iPod, the new GPU needs only about half an hour.

High Resolution

GPUs provide the graphics horsepower for some of the most processing-intensive tasks, such as rendering 3-D images in real time, encoding and playing high-definition videos, editing photos, or other such tasks.

The GTX 280 and 260 feature 50 percent more gaming performance, compared with the previous Nvidia GeForce 8800 Ultra GPU. The new GPU’s 240 enhanced processor cores offer resolutions as high as 2560 x 1600.

Three-way SLI technology provides alternate frame rendering for what the company called “the world’s fastest gaming solution under Windows Vista,” and the CUDA technology enables programmers to offload processing tasks from the CPU to the GPUs. The company also said its chips are the only ones to support PhysX technology, a physics engine that is already being used in more than 140 games.

Nvidia cited Elemental Technologies’ BadaBOOM Media Converter, which utilizes the new chips and is expected to be released in August. “By taking advantage of the massively parallel, general-purpose computing architecture of a GeForce GPU,” said Elemental CEO Sam Blackman, “we are able to transcode high-quality video 18 times faster than with CPU-only implementations.”

Faster Protein-Folding

Nvidia also noted that the GPUs can be utilized to increase the speed of one of the most processing-intensive tasks, simulating protein folding for research into treatments for diseases. The Folding@Home project from Stanford University utilizes distributed processing grids, with processing from many home-based personal computers and game machines.

Vijay Pande, an associate…