From our front-page news:
We first learned about Larrabee, Intel's discrete GPU, at last spring's IDF in Beijing, and then saw it tackled again at this past fall's event. Although a launch date was never approximated, it was thought that we would be seeing product in the area of mid-2009, but that has now been debunked, courtesy of Intel CEO Paul Otellini at the fourth-quarter earnings conference.
"Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," ... "I still think we are on track for a product in late 2009, 2010 timeframe."
Could we have asked for a better answer? What makes this interesting is the timing. As discovered during Otellini's keynote in September, both Intel's processors and GPUs should cross in 2010 and both be based on 32nm, including Nehalem's successor, Westmere. Of course right now, everything boils down to support and whether or not game developers will exploit all of what Larrabee brings to the table. We will have to wait a few years to find that out, though...
<table align="center"><tbody><tr><td>
</td></tr></tbody></table>
With Larrabee not arriving until 2010, this gives GPU makers quite a bit of time to catch up to multicore x86 in terms of programmability. And by programmability, I specifically mean the kind of improved exception handling and process switching capabilities that GPUs will need if they're going to compete with a multicore part like Larrabee for the coprocessor spots in commodity-based supercomputing clusters.
Source: Ars Technica
"Larabee first silicon should be late this year in terms of samples and we'll start playing with it and sampling it to developers," ... "I still think we are on track for a product in late 2009, 2010 timeframe."
Could we have asked for a better answer? What makes this interesting is the timing. As discovered during Otellini's keynote in September, both Intel's processors and GPUs should cross in 2010 and both be based on 32nm, including Nehalem's successor, Westmere. Of course right now, everything boils down to support and whether or not game developers will exploit all of what Larrabee brings to the table. We will have to wait a few years to find that out, though...
<table align="center"><tbody><tr><td>
</td></tr></tbody></table>
With Larrabee not arriving until 2010, this gives GPU makers quite a bit of time to catch up to multicore x86 in terms of programmability. And by programmability, I specifically mean the kind of improved exception handling and process switching capabilities that GPUs will need if they're going to compete with a multicore part like Larrabee for the coprocessor spots in commodity-based supercomputing clusters.
Source: Ars Technica