I'm not surprised either. This IS a complex project, Intel is trying something that has never been done before (A graphic card based on X86 cores), with hardware and software to develop for it.
I'm thinking that performance is not where they want it to be. Or, if you prefer, performance is not high enough to be competitive enough to sell Larrabee at a high enough price to cover costs.
Not me of course. Intel can't produce a decent graphics chip to soldier on their own chipset.
Not that it matters, given their market share. I don't think it's a matter of can vs. can't, I think it's a matter of can vs. need. Intel is the largest supplier of graphics chips in the world... Why lay out billions for a marginally better IGP?
My point is that Larrabee has been a hot topic for the geek hype machine. I understand when NVIDIA says, here is FERMI, hold on to your shorts, its going to be awesome, and they miss delivery and everyone is still buzzing about how incredible its going to be. NVIDIA has earned that. In Intel's case, Larrabee is not hype worthy. Intel has never produced a graphics product that is even okay much less any good, but yet, people talk about Larrabee as if its going to be a game changer just because Intel may have talked about ray tracing or something in a demo. I never understood the hype surrounding Larrabee. There comes a time where you need to either "put up or shut up"
Intel is and will remain a major player in the graphics sector as long as their processors (and therefore chipsets) remain a major player. Most office workstations have no need for extra dollars spent on discrete graphics setups, so by implementing an inexpensive, low-power IGP on their motherboards, Intel will automatically sell their graphics technology to a very large segment of the computing industry.
This being said, Larrabee did nothing to make my geek sense tingle. Ever since I heard that announcement, all I could think was "well, they got their work cut out for them".
primesuspect
Beepin n' BoopinDetroit, MI Icrontian
Comments
Personally, I didn't see it happening. A company with no discrete card experience suddenly jumping into the high end? Yeah right.
I'm thinking that performance is not where they want it to be. Or, if you prefer, performance is not high enough to be competitive enough to sell Larrabee at a high enough price to cover costs.
Either that, or their drivers suck a usual...
Lemme rephrase... a company with no recent discrete card experience.
Not me of course. Intel can't produce a decent graphics chip to soldier on their own chipset.
Not that it matters, given their market share.
This being said, Larrabee did nothing to make my geek sense tingle. Ever since I heard that announcement, all I could think was "well, they got their work cut out for them".