At
this year’s Computex, Nvidia is not only planning to give the world a preview
of its upcoming ION products, but has also set out to discuss some major plans
it has been working on for the past five years or so. The topic at hand will be
focused primarily on bringing the era of GPU Computing to the mainstream level.
Ujesh Desai, Vice President of
Marketing at Nvidia recently sat down with Digitimes for a Q&A session to
discuss the history of GPU compute development at Nvidia. In his central focus
point, Desai speaks of the importance of heterogeneous processing between the
Central Processing Unit and the Graphics Processing Unit and the future
expansion focus of the GPU.
“We'll be the first to admit
that the CPU is by far one of the most important processors ever developed in
the computing industry. But as people are starting to realize, the GPU is just
as important. For a really good modern computing device you need a GPU as a
dedicated co-processor. You can do a lot more with this combination of a good
CPU and a very good GPU, and this year at Computex that's the key message
you're going to hear from us, and it's the key thing you're going to see from
our partners as they demonstrate their new products”
The discussion can be found here.
Published in
Graphics
Nvidia to reveal its five-year GPU Compute plans
Building solutions around the GPU