NVIDIA nTerresting News: 03 December 2010 |
Written by NVIDIA - Brian Burke | ||
Saturday, 04 December 2010 | ||
NVIDIA nTerresting News: 03 December 2010In this issue:
Great Time to Be a Gamer Holidays are here and gaming is where it is at. You got the fastest GPU in the world with DX11 done right, Call of Duty: Black Ops is breaking records and World of Warcraft: Cataclysm is looming with two new races. "World of Warcraft: Cataclysm's two new races, Worgen for the Alliance and Goblin for the Horde, weren't exactly surprises thanks to rampant speculation and leaked info." More than ever, games are taking advantage of 3D. "But, with the way 3D gaming is shaping up, some of these games are looking better than ever before and really do thrust you into the action in a unique and exhilarating way. The Holidays are right around the corner and the line-up of games is staggering. "These games have great single-player experiences, but many of them also have excellent social sharing and multiplayer experiences too, a factor that's becoming increasingly important in this market." It is a great time to be a gamer. Image Quality Under a Microscope Several technology websites have discovered that AMD has reduced their default texture filtering quality for some products with the introduction of the Catalyst driver version 10.10. We have verified the new Cat 10.11 driver has the same problems. "Getting directly to the point, major German Tech Websites ComputerBase and PC Games Hardware (PCGH) both report that they must use the "High" Catalyst AI texture filtering setting for AMD 6000 series GPUs instead of the default "Quality" setting in order to provide image quality that comes close to NVIDIA's default texture filtering setting. 3DCenter.org has a similar story, as does TweakPC. The behavior was verified in many game scenarios. AMD obtains up to a 10% performance advantage by lowering their default texture filtering quality according to ComputerBase." As a result, several websites have already decided to test the affected AMD cards going forward with higher than default texture quality settings, to come closer to NVIDIA's default quality. Super and Green NVIDIA made big news when it was announced that NVIDIA-powered supercomputers took 3 out of the top 5 spots on the Top500 list for the world's fastest supercomputers. Well a new 500 list is out, the "Green 500", a list of the world's most energy-efficient supercomputers. Guess what? "..the only petaflop systems in the top 10 are powered by NVIDIA Tesla GPUs. Of these GPU-powered petaflop systems, Tsubame 2.0, from Tokyo Institute of Technology (Tokyo Tech), was ranked number two; and Tianhe-1A, the world's fastest supercomputer from National Supercomputer Center in Tianjin, was ranked number 10. Two other Tesla GPU-based systems made the Top 10 and Tesla GPU-based systems installed at CSIRO in Australia and National Supercomputer Center in Shenzhen were also ranked 11 and 12 respectively." GPU-powered systems are in spots 2, 3, 9, 10,11,12,13 and 17 on the Green500 List. Watch the video to find out why. Fastest and Fasterest It is a given, the GeForce GTX 580 is fast, really fast. So how fast are two of them? "As you could see from the performance benchmarks the GTX 580 SLI setup was pretty much untouchable and was more than impressive. Gaming doesn't get much better than this and to have all the eye candy turned on and cranked up at 1920x1200 and still being able to reach 60FPS in applications like Heaven 2.1 was awe inspiring." If the fastest GPU in the world is not fast enough, use NVIDIA SLI technology to double up on the GPU horsepower. Crank that 5#!T up! Cloudy with a Chance of GPUs Fancy yourself as a scientist? A bold inventor with a revolutionary product idea? If you are, it's highly probable that you will need lots of computational power to run simulations, experiments or develop prototypes to realize your discoveries. Until recently, you would have to shell out some green to buy a supercomputer, or at least buy time on someone else's, but this has changed with the latest announcement from Amazon Web Services, that it has now added Tesla GPUs to its Cloud. Amazon's Elastic Compute Cloud (EC2) GPU offering will bring the power of GPU Computing to businesses of all sizes, allowing them to run heavy duty simulations and more with a simple on-demand pricing structure, and no large up front capital investment. GigaOm Pro had this to say about the announcement: "Performance (of Amazon's Cluster Compute Instances) was high already, and the addition of GPUs just ups the octane level. According to a benchmark test by HPC cloud-resource middleman Cycle Computing, GPU Instances outperform in-house GPU clusters in certain cases". In a field obsessed with speed, GPUs can seriously accelerate performance for massively parallel, multi-threaded workloads. GPUs provided the brunt of the processing power for three of the top four systems on the Top500, and 11 in total. Dell, IBM and HP (as well as many HPC-focused vendors) are all rolling out GPU-based servers and systems. Amazon's CTO, Werner Vogels published an interesting blog on the topic here, and one of Amazon's technical specialists, Jeff Barr, gave a great technical overview of the new service here. Related Articles:
|