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: tcmay who wrote (77255)	4/16/2002 7:43:07 PM
From: pgerassi	Respond to of 275872

Dear Tcmay: IA-64 is basically a design that was looked at and found to be wanting. It is ok if hand coded assembly is used, but no good at all if compiled with current technology compilers. They still do not have a compiler that can optimize well at all for a EPIC CPU. And that is a design decision needed for EPIC (essentially VLIW) to work. Hammer is a lot easier to compile for and optimize as its architecture is well understood and compilers are highly effective at getting good performance from them. As to the "Supercomputer", a massively parallel Athlon cluster of 1400 AMP 2000+s would be about 8.9 Teraflops peak capacity at far less money (unless Intel is supplying them gratis (well maybe not even then)) than the above. The article is also incorrect since a personal computer gets more like 3 to 10 Gigaflops now, not 1 Gigaflops like they claimed. And peak performance is not close to what the real performance would be. The real performance will be between 25% and 50% of the peak, if well done at that. Pete