Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Magrathea who wrote (210431)	9/8/2006 1:13:22 PM
From: jspeed	Respond to of 275872

How come IBM went with an Operton-Cell solution instead of Power-Cell. Possibly because Los Alamos said, "we've got x86 Code, Make it work on the next Super." I agree. IBM is looking for a way to market the Cell as a server. Trying to migrate it's Power customers to Power-Cell probably doesn't make much sense from a business standpoint. On the other hand, if they can migrate x86 customers to x86/Cell they can potentially capture a nice chunk of the HPC market that they don't already have. I don't read it as a combined Opteron-Cell Blade. Poor choice of words on my part. I meant to say Opteron-Cell system.

To: Magrathea who wrote (210431)	9/8/2006 7:48:49 PM
From: seti	Respond to of 275872

Possibly because Los Alamos said, "we've got x86 Code, Make it work on the next Super." This sounds somewhat strange. Code is most likely written in Fortran. They are doing scientific computation, as opposed to device drivers. Normally the reason one would write x86 specific code would be to use SSE instructions, possibly in assembly. Those would have to be re-written anyway to use CELL. There is also Amdahl's law. If you speed up 90% of the computation by an infinite amount, the last 10% means your overall speedup is only 10. Does Opteron do better on the last 5%, 1%, or 0.1% than Power?