AMD's first Stream Computing chip is called the FireStream 9170 and it initially aimed at high-performance computing, such as workstations used for scientific modeling and heavy-duty number-crunching.

"With a broad range of customer engagements underway, notably customers in the oil and gas, financial and engineering analysis industries, AMD is delivering on its vision of accelerated computing with breakthrough benefits for our enterprise customers," said Rick Bergman, senior vice president and general manager for graphics products.

Bergman said the new processor combines ATI's "immense graphics-processing capabilities" and AMD's high-performance expertise. "We are developing strong relationships with system vendors and the supporting technology ecosystem to deliver processing innovation through an open platforms approach," he said in a statement.

From Games to Simulations

The FireStream 9170 features double-precision floating point technology fine-tuned for scientific and engineering calculations, AMD said. While the chip is expensive, priced at $1,999, it boasts 500 gigaflops of computing power for critical algorithms. Built with a 55 nanometer technology, it consumes little power, which fits in with AMD's push to tighten energy consumption in the data center.

"People want interaction in games to be more realistic, so a lot of the underlying physics turns out to be similar to what you want to do with real-world simulation," Phil Hester, AMD's chief technology officer, told Reuters.

"It's a very good opportunity to exploit the power and price performance advantage the GPU (graphics processing unit) gives you," Hester said.

Looking to Fusion

The FireStream 9170 chipset includes 660 million transistors and 320 processing units and will be made by Taiwanese chipmaker TSMC.

The chip is a step on the way to AMD's Fusion project, which the company says will combine a graphics processor and general processor on the same piece of silicon. AMD hopes to release Fusion in early 2009.

The Stream Computing breakthrough is in handling central processing tasks the way graphics processors do -- in parallel. By breaking down massive amounts of data into smaller chunks to be handled in parallel, AMD should be able to deliver better performance for data-intensive tasks.

The goal of Fusion is to apply these techniques to lower-end desktop and laptop machines. AMD is also offering a software development kit for programming to the new chip.