SleepyEgg

NVIDIA CEO Jen-Hsuan Huang kicked off his company's GPU Technology Conference this afternoon with a keynote that unveiled NVIDIA's next-generation GPU architecture, codenamed Fermi. The fact that this architecture is named after a physicist, and that is shares this naming convention with NVIDIA's supercomputing-oriented Tesla line, tells you most of what you need to know about where NVIDIA sees its future.

In short, the GPU maker must walk a fine line between pleasing its shrinking core audience of hardcore PC gamers and courting its growing user base in the high-performance computing (HPC) realm. In times past, walking this line meant erring on the side of pleasing gamers, even if it meant making design decisions that were disadvantageous to HPC. But Fermi marks the point at which NVIDIA has officially begin making its discrete GPU tradeoffs favor the HPC market at the expense of gamers. This being the case, it now seems likely that NVIDIA's real future in the gaming market lies with the Tegra line.

<