'Fundamental new architecture' promises major speed boost.
'Fundamental new architecture' promises major speed boost.
Graphics processor giant nVidia has introduced what it claims to be a fundamentally new architecture for computing.
The company's GeForce 8800 graphics card is the first to use Cuda technology, which enhances traditional GPU performance by enabling processor cores to communicate, synchronise and share data.
Cuda uses hundreds of on-chip processor cores which simultaneously communicate and cooperate to solve complex computing problems up to 100 times faster than traditional approaches, nVidia said.
This architecture is complemented by nVidia's first C-compiler for the GPU.
The development environment is particularly suited to solving new problems in compute-intensive applications such as product design, data analysis, technical computing and game physics.
Cuda-enabled GPUs offer dedicated features for computing, including Parallel Data Cache which allows 128 1.35GHz processor cores in the newest generation.
Developers access these features through a separate computing driver that communicates with DirectX and OpenGL, and the C compiler for the GPU.
Nicolas Chavannes, director of software at Schmid and Partner Engineering, believes that the level of computing performance now achievable with Cuda-enabled GPUs will positively impact its customers' bottom lines.
"Our customers, including every cellphone manufacturer in the world, see the value in using nVidia GPUs with Acceleware's GPU-accelerated solver to speed up their time to market," he said.