Posted on 11/16/2006 12:08:57 PM PST by Ernest_at_the_Beach
AMD has given us some more information about Fusion, albeit still a little vague on specifics as to exactly how a stream processing on the GPU would be leveraged. But if you thought CPU and GPU integration onto one die was just about low end computers, AMD is painting a different picture. CPUs with high powered stream processors onboard that utilize GPU power is the future that AMD sees.
Still though, power savings and efficiency are on the forefront of implementation. Certainly the mobile market will be the focal point of this Fusion technology.
And just for you rumor mongers. AMD states they are in NO way interested in leaving the discrete high end graphics space.
Here are slides from this morning.
(Excerpt) Read more at hardocp.com ...
Stream Computing
Torrenza
*****************************************
AMD Spins Up 65nm Production in 2006
Following AMD’s Fusion remarks this morning, AMD focused on their changing process announcing that they would certainly transition to 65nm in 2006 with full product conversion in 2007. AMD highlighted that they, as usual, would bring their next-gen process into existing architecture as that would allow them to more efficiently transition to Barcelona cores next year. As noted on the slides below, AMD plans to transition 45nm process into production in 2008 with 32nm in production in early 2010.
It's going to confuse buyers.
What's gonna confuse them?
Better stuff for less money. That always gets 'em.
Around here for sure.....LOL!
I hope they don't change the socket one more time....
Just skimming the graphic, it looked like multi-chip lines, perhaps on top of the others. Maybe I assumed too much.
See link at post #18.
AMD unleashes 65nm technology on 5 Dec
And we have this:
The World First GPGPU: AMD Stream Processor = R580?
**********************************************
At the Supercomputing 2006 show today, AMD introduced the world’s first dedicated stream processor designed to meet the specific hardware demands of high-performance computing (HPC) applications. Engineered specifically for compute-only systems such as workstations and servers, the AMD Stream Processor makes use of AMD’s new thin hardware interface called Close To Metal (CTM) to drive substantial performance gains in areas such as financial analysis, seismic migration analysis, and life sciences research, among others.
Stream computing leverages sophisticated massively parallel processors generally used to calculate and render millions of pixels onto computer monitors, hundreds of times each second for 3D graphics applications. Stream computing harnesses that tremendous processing power for a wide range of scientific, business and consumer computing applications, providing organizations the ability to process incredible amounts of information in significantly less time. The AMD Stream Processor is the first hardware of its kind dedicated to addressing stream computing problems in the enterprise space.
Basically, the AMD Stream Processor is the same as ATI Radeon X1900XTX besides the concept of General-Purpose Computation on Graphics Processing Unit (GPGPU). Through the 48 4-float vector ALU units in R580, Stream Computing is achieved. Like most of the graphic cards available in the market, this is a PCI Express-based add-in board designed, but its memory densities is higher than any consumer graphics card – a full 1 GB of GDDR3 memory for stream computing applications. The processor also features HPC-optimized memory performance designed to best handle unique stream computing problem sets. In the previous interview to Stan Ossias, ATi’s Senior Product Manager of Desktop Discrete Graphics, he stated R580, in fact, has been optimized for GPGPU, supporting 512 threads running simultaneous. That provides substantial performance gains in areas such as financial analysis, seismic migration analysis, and life sciences research, among others.
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.
Leading hardware platform providers, including GraphStream Inc., PANTA Systems, and Rackable Systems, also introduced new stream servers to the market today. Stream servers make use of the new AMD Stream Processor to deliver enterprise-class performance for the HPC and enterprise markets in industry standard form factors. The AMD Stream Processor also includes broad operating system support for server implementations; full-service, enterprise-class technical support; and a limited three-year warranty. The AMD Stream Processor is available now from distributors.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.