The increasingly fierce war of words between Intel and Nvidia saw no sign of abating yesterday, as Intel’s CEO Paul Otellini took a swipe at its new tech nemesis at a tech conference in San Francisco. Rather than focusing on Nvidia’s Ion platform, which has been the recent target of Intel’s recent ire
, Otellini instead focused on Nvidia’s GPU technology.
According to CNet
, Otellini said that Nvidia wasn’t in a strong position to compete, saying that "If you don't have a microprocessor, what else do you have to sell?” This counters Nvidia’s recent claims that GPUs will play a big part in the future of computing with technologies such as CUDA.
Reacting to Intel’s decision to sue Nvidia, the graphics company’s president and CEO, Jen-Hsun Huang, claimed in a statement that "the CPU has run its course and the soul of the PC is shifting quickly to the GPU. This is clearly an attempt to stifle innovation to protect a decaying CPU business."
However, Otellini argued that “the graphics subsystem for most machines will be subsumed into the microprocessor. So what Nvidia is doing is making an argument to defend the status quo.]” Two weeks ago, Intel revealed its first CPU with an integrated graphics core
, which offers an improved equivalent of the integrated graphics found in Intel’s current G45 chipset. Intel recently claimed
that its current integrated graphics systems account for 50 percent of the graphics market at the moment.
Otellini also added that people who wanted faster graphics performance could still buy a separate graphics card. "You can buy it from them or you can buy it from us," said Otellini, referring to Intel’s forthcoming Larrabee graphics chip, which performs 3D acceleration on x86 processors rather than the huge banks of simple, scalar stream processors found in today’s traditional GPUs.
Is Intel right to dismiss GPGPU computing, and does Nvidia really need its own CPU technology to compete in this industry? Let us know your thoughts in the forums