11/06, 4:40pm
Metal-air ionic liquid could extend batteries
A new development by partly government-backed Fluidic Energy could potentially extend the battery life of notebooks, cars and other devices well beyond existing lithium-ion cells. Known as Metal-Air Ionic Liquid (MAIL), it would improve energy storage beyond relatively efficient zinc-air batteries by using an ionic liquid salt to conduct electricity that is much more stable and isn't prone to drying out either by accident or by eventual decay. The move would let battery makers use metals denser than zinc and therefore hold a much larger charge in a given volume.
more
09/30, 5:35pm
NVIDIA Fermi
NVIDIA this evening provided an early look at the next generation of its graphics processors. Nicknamed Fermi, the architecture for future GeForce, Quadro and Tesla chipsets will jump from 240 cores to a much larger 512 and should be much faster in each core courtesy of some industry-first techniques. Fermi chips will be the first GPUs to have a real cache hierarchy, with Level 1 caches to keep specific information on hand and a single, shared Level 2 cache for larger tasks; they will also have a new GigaThread engine that can transfer data in both directions at once and handle "thousands" of tasks at once.
more
11/18, 1:05pm
NVIDIA Tesla supercomputer
NVIDIA on Tuesday announced the availability of its GPU-based Tesla architecture Personal Supercomputer, which, the company promises, has as much as 250 times the computing performance of a PC thanks to NVIDIA CUDA parallel computing architecture and up to 960 parallel processing cores. Along with its partners, buyers can now opt for desktop PCs with the sort of computing power previously reserved for supercomputing clusters.
more
08/22, 12:50pm
NVIDIA CUDA 2.0 Final
NVIDIA today formally released the finished version of CUDA 2.0. The second generation of the company's general-purpose programming language for its video chipsets supports 64-bit versions of Mac OS X and Windows Vista and adds support for instructions that can help offload more specific tasks from the main processor to the video card, such as 3D textures and hardware-accelerated interpolation of information.
more
07/23, 9:45am
DirectX 11 with GPGPU Tech
Microsoft's next version of DirectX will have its own alternative to the OpenCL standard proposed by Apple, the company revealed yesterday at its GamesFest conference. DirectX 11 will have support for "compute shader technology" that allows modern, more generalized video cards' effects processors to perform tasks other than rendering video, including physics calculations and other chores that would normally be handled by the main system processor.
more
06/17, 11:40am
OpenCL Standard
The Khronos Group late yesterday established a new alliance between vendors that could see standards for high performance computing such as OpenCL gain a foothold across many operating systems and hardware platforms. Called the Computer Working Group, the team includes graphics rivals 3DLabs, AMD, and NVIDIA, processor makers such as ARM, Freescale, Intel, and Qualcomm, and end product manufacturers such as Motorola and Nokia, all of whom hope to create and maintain genuinely open and royalty-free standards for using newer graphics hardware to process very demanding compute tasks.
more
06/16, 4:30pm
NVIDIA Tesla T10P
NVIDIA on Monday used the occasion of its GTX 200 introduction to quietly update its Tesla line of workstation processor cards. The T10P chipset added to the cards is virtually identical to the 240-core GTX but is her spun entirely towards accelerating high performance computing tasks such as medical research and very high-level math. The very specialized nature of the chip lets it calculate as much as 900 gigaflops by itself, or 73 percent more than the earlier card it replaces.
more
