![]() "And we're very impressed with the prospect of double-precision floating point, especially compared to our other GPU or FPGA options." "Three gigaflops-per-watt is pretty impressive," said Eric Aske, director of HPC integration at Source Code. So we wanted to let people in the HPC community program in something that felt like 'C'," Harrell said.įor companies that build HPC cluster solutions, like Waltham, Mass.-based system builder Source Code, the FireStream 9170 processor and SDK looks pretty good on paper. "The problem with 'Close to the Metal' was that it was really difficult for HPC programmers to understand the GPU. The first such language supported by AMD is called Brook+, based on a GPU programming language developed at Stanford University. The trick for getting HPC benefits from GPU technology is using "C" extensions to give HPC developers a familiar, high-level language for working with the processors, Harrell said. In addition to the processor itself, AMD will make available a software development stack that includes compilers and libraries, as well as third-party tools being built by partners such as Rapidmind. The silicon for the new processor is manufactured with the 55nm process. The features and specs AMD is claiming for the FireStream 9170 include the double-precision floating point, two gigabytes of on-board memory, asynchronous DMA, 320 stream cores (or more than twice as many as Nvidia has on a rival stream computing product), and up to 500 gigaflops of compute power with performance coming in at three gigaflops-per-watt on a processor that chews up less than 150 watts. Harrell said the FireStream 9170 and its associated development tools were built for the HPC community, but that AMD believed the technology would eventually make its way to the commercial and consumer spaces as well. ![]() Over the past few years, stream computing technologists have adapted GPUs to accelerate critical computational algorithms in mathematically demanding work such as computational chemistry, seismic processing and medical imaging. The new stream processor, which will ship in the first quarter of 2008 for $1,999-per-unit, is the first to have a double-precision floating point, she said. List of supported graphics and general-purpose computing APIs, including their specific versions.The FireStream 9170 was developed with AMD's ATI graphics technology and in conjunction with GPU application work conducted in the academic research community, said Patricia Harrell, director of Stream Computing at the Sunnyvale, Calif.-based chipmaker. OEM manufacturers may change the number and type of output ports, while for notebook cards availability of certain video outputs ports depends on the laptop model rather than on the card itself. As a rule, data in this section is precise only for desktop reference ones (so-called Founders Edition for NVIDIA chips). ![]() Types and number of video connectors present on the reviewed GPUs. Integrated GPUs have no dedicated VRAM and use a shared part of system RAM. Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. ![]() For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility). Useful when choosing a future computer configuration or upgrading an existing one. ![]() Information on compatibility with other computer components. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked. These parameters indirectly speak of performance, but for precise assessment you have to consider their benchmark and gaming test results. General performance parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |