AMD ''Close to Metal''™ Technology Unleashes the Power of Stream Computing
New Open Interface Drives up to Eightfold Increase in High-Performance Computing Application Processing Speed1
TAMPA, Fla.--(BUSINESS WIRE)--
AMD (NYSE: AMD) today announced that software developers can put the ‘pedal to the metal’ for emerging stream computing applications, making use of a new thin hardware interface known as CTM™ (for “Close To Metal”) to increase processing application performance by as much as eightfold more than traditional 3D application programming interfaces (APIs).1
“CTM effectively unlocks the hardware for the RapidMind Development Platform, giving us open access to the massive computational power of stream processors, in turn allowing software developers to create their own stream applications quickly and effectively”
CTM gives developers unfettered access to the native instruction set and memory of the massively parallel computational elements in AMD Stream Processors™. Using CTM, stream processors effectively become powerful, programmable open architectures like today’s central processing units (CPUs). By opening up the architecture, CTM provides developers with the low-level, deterministic, and repeatable access to hardware that is necessary to develop essential tools such as compilers, debuggers, math libraries, and application platforms.
Through CTM, AMD intends to foster strong growth in the software industry for stream computing by enabling development of the best tools possible, unfettered from performance barriers, coding hurdles, and esoteric dependencies on drivers. Today more than 60 companies and research institutions are taking part in CTM trial programs. These organizations are bringing best-of-breed software to market that enable application developers to have a broader choice in how they develop and deploy their applications. This approach serves a wide range of markets, including high-performance computing and consumer software – two segments with significantly different development needs.
RapidMind Inc. is working to let developers easily take advantage of the parallel architecture of AMD stream processors for a variety of professional and consumer users. “CTM effectively unlocks the hardware for the RapidMind Development Platform, giving us open access to the massive computational power of stream processors, in turn allowing software developers to create their own stream applications quickly and effectively,” said Ray DePaul, president and CEO of RapidMind Inc.
“Using CTM today, AMD is working with a number of companies to deliver the tools ecosystem for stream computing,” said Marty Seyer, senior vice president, Computational Product Group, AMD. “As part of our Torrenza initiative and CTM, AMD is enabling companies to work with best-of-breed vendors that understand how to optimize their software across all processor architectures, whether in stream processors or high-performance CPUs. For these organizations, the development of highly capable, and efficient software is their business, not a sideline. Allowing open innovation to flourish will ultimately enable better software, with more features to come to market faster than any proprietary approach.”
CTM is available to developers to license today at no cost. For more information, developers should contact AMD developer relations or visit http://ati.amd.com/companyinfo/researcher/resources.html. Today AMD also announced its first stream processor offering. For more information on this announcement, please see the relevant press release.
Advanced Micro Devices (NYSE: AMD) is a leading global provider of innovative processing solutions in the computing, graphics and consumer electronics markets. AMD is dedicated to driving open innovation, choice and industry growth by delivering superior customer-centric solutions that empower consumers and businesses worldwide. For more information, visit www.amd.com.
This press release contains forward-looking statements which are made pursuant to the safe harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements are generally preceded by words such as “plans,” “expects,” “believes,” “anticipates” or “intends.” In particular, statements in this press release including but not limited to the performance, capabilities, market potential and the applications for stream computing and the close to metal interface, including statements relating to accelerated processing and reduced processing times, exponential performance gains, enhanced problem-solving and ease of programming, may be considered "forward-looking." Such forward-looking statements are based on current expectations and assumptions and, accordingly, entail various risks and uncertainties. Assumptions applied in making, and potential risks that could cause actual results to differ materially from such forward looking statements include, among others, the ability to drive commercial adoption of enterprise stream computing and to build a stream computing ecosystem, software and/or hardware-related issues and conflicts, the development of new products or technologies, overall system performance and continued market demand for accelerated processing. We therefore cannot provide any assurance that such forward-looking statements will materialize. We assume no obligation to update or revise any forward-looking statement, whether as a result of new information, future events or any other reason. Additional information concerning risks and uncertainties affecting our business and other factors that could cause actual results to differ materially from any forward-looking statement is contained in our filings with Canadian and U.S. securities regulatory authorities.
AMD, the AMD Arrow logo, AMD Opteron and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Other names are for informational purposes only and may be trademarks of their respective owners.
1. Based on tests using AMD ATI Radeon X1900 XTX comparing GPUBench OpenGL-based MatMult results to CTM-based MatMult results. The OpenGL-based MatMult performance is approx. 12 GFlops vs. the CTM-based MatMult performance of approx. 96 GFlops.