NVIDIA was on hand yesterday to speak at the Morgan Stanley Technology conference in San Francisco. The last 00:01:40 of the hour segment was easily the most interesting. The discussion started with an overview of NVIDIA’s position on Intel’s Larrabee GPU, moved to the Tegra, and ended with an admission that Big Green would like to move into x86 SoC CPUs. Here’s the relevant excerpt transcribed from the recording:
Question: “Can you provide a framework for us to think about whether, uh, if or when NVIDIA would need to or want to get into the general purpose microprocessor business, either as a separate component or integrated with a GPU?”
Answer: “Well I think some time down the road, um, it makes sense to take the same level of integration that we’ve done with Tegra, because Tegra is by (unclear) definition a complete computer on a chip, and the requirements of that market are such that you have to be very low power, very small, but highly efficient. So in that particular space it made a lot of sense to take that approach. And some day it’s going to make sense to take the same approach in the, uh, x86 markets as well. Not for every SKU, though. If you look at the high end of the PC market, I think it’s going to stay fairly discrete. Cause it needs to be the best of all worlds. I think when you get into the low cost, you know (unclear) on the MID or netbook type space, um, a highly integrated system on chip is going to make sense. Um, so the question is when. Not so much, I think, um, you know, um, if. I think the question is when.”
Articles RSS