![]() Street prices for NVIDIA Tesla C1060 and AMD FireStream 9270 are roughly $1,200 to $1,500. Perhaps more impressive, the AMD FireStream 9270 GPU can deliver up to 1.2 teraflops, according to its makers. That’s less then a hundred gigaflops shy of one teraflop. With 240 processing cores, NVIDIA’s Tesla C1060 GPU, now available in the market, can deliver up to 933 gigaflops, according to the product spec sheet. Teraflop computing is not a tantalizing dream but a reality with certain GPUs. In essence, it was to become a chip powerful enough to reclaim some of the markets GPU leader NVIDIA has wrestled away from Intel. It was meant to be a hybrid GPU-CPU with many cores, and many more threads. The chip was expected to debut in 2009-2010. Intel first unveiled its plan to develop Larrabee in SIGGRAPH 2008. Mark Priscaro from NVIDIA public relations office said, “The fact that a company with Intel’s technical prowess and financial resources has struggled so hard to succeed with parallel computing shows just how exceptionally difficult a challenge this is.” As a result, our first Larrabee product will not be launched as a standalone discrete graphics product” (” Intel: Initial Larrabee Graphics Chip Cancelled,” December 4, 2009). In a recent CNET article, he was quoted as saying, “Larrabee silicon and software development are behind where we hoped to be at this point in the project. Last Sunday, Intel spokesperson Nick Knupffer revealed Larrabee had been put on hiatus. “We were getting tantalizingly close to one teraflop ,” he later concluded. During his keynote, holding a prototype unit of Larrabee, Intel’s CTO Justin Rattner asked, “Would you like to see it do its thing?” Then he let the test unit run, with a SGEMM performance meter attached to it (to measure single-precision general matrix multiply subroutine). ![]() It has already announced plans to integrate graphics chips onto PC chips as part of its Fusion project, but it hasn't identified a timeframe for putting its powerful stream computing technology on a PC chip.Last month, Intel‘s Larrabee was the star of SC09 (Super Computing 2009, Portland, Oragon November 14-20). It will cost $1,999, which might seem like a lot, but this is something you should be able to add into an existing workstation or server for a performance boost when you need it, rather than buying a fancy server for just a few lines of code.Įventually, AMD wants to integrate this type of technology directly onto a PC or server processor. The 9170 isn't going to be out until the first quarter of next year, as AMD's graphics priorities for the holiday season are discrete graphics chips for PCs that all of us can use. "You're not worried about changing code for something that gives you an order of magnitude increase (in performance)," she said. They have a particular algorithm that (the researcher) knows will run well on a GPU," said Patricia Harrell, director of stream computing for AMD. "You don't have a researcher that's trying to port over thousands of lines of legacy code. You'll need a software developer's kit, and you'll probably only want to port limited amounts of your code to run on the 9170. It comes with 2GBs of memory, compared with 512MBs of memory on the most powerful ATI graphics chip.īut the programming is still a little tricky. The 9170 is essentially one of ATI's high-end discrete graphics chips that has been tricked out with more memory and double-precision floating point units, which apparently is better than single precision. Think of it as a high-end graphics chip with a lot more memory than usually ships with those products, said Robert Feldstein, vice president of engineering for AMD. Certain types of customers in labs and research facilities would love to be able to tap into that kind of processing power, but GPUs require special programming techniques.ĪMD is trying to bridge the gap between PC processors that are easy to program and graphics chips that offer great performance with the FireStream 9170. The high-performance discrete graphics chips from companies like Nvidia and AMD's ATI division have been designed with parallel performance in mind for a very long time. Graphics processing units (GPUs) have been doing this for years. This allows the chip to run at slower speeds, and therefore cooler temperatures. ![]() Instead of trying to crunch all the data through a single path moving as fast as possible, the cool kids are now adding paths so data can flow down multiple outlets. The big trend in chip design over the past few years has been parallelism.
0 Comments
Leave a Reply. |