14 DAY TRIAL // The competition between Intel and Nvidia becomes stronger nowadays, and it is spread over a few areas in the computing industry. The chipset manufacturing is the first and probably most important battle field between the two companies, but the latest months have brought to us rather the fighting on the integrated graphics market, where Intel's devices are leading.
Another important competing arena is that of data processing, where Nvidia has recently introduced the CUDA programming language. Pat Gelsinger, co-general manager of Intel's Digital Enterprise Group, said in an interview with Custom PC that the annals of computing history would register NVIDIA's CUDA programming model only as an interesting footnote.
Gelsinger said that programmers would need a large amount of time to learn how to program the new architectures, including CUDA, and that is out of the question. "The problem that we've seen over and over and over again in the computing industry is that there's a cool new idea, and it promises a 10x or 20x performance improvements, but you've just got to go through this little orifice called a new programming model. Those orifices have always been insurmountable as long as the general purpose computing models evolve into the future," he explained.
Gelsinger also added that the Sony Cell is a good example to illustrate his sayings. According to him, the performance gains promised by the Cell architecture were huge in comparison to normal ones, but a lot of developers still do not support the Cell.
According to Gelsinger, Intel's Larrabee graphics chip will be widely adopted by developers, as it uses entirely the x86 cores Intel Architecture. This means that developers do not have to learn a new language to program for the graphics processor, and that will make the whole difference. Also, Larrabee will feature full support for APIs like DX and OpenGL.
NVIDIA's CUDA architecture brings the innovation of allowing complex physics calculations to be processed on the GPU, by transferring the PhysX from the CPU on the graphics processor.