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: bacchus_ii who wrote (57563)	10/7/2001 6:45:29 AM
From: fyodor_	Respond to of 275872

Gottfried: I also think it's the maximum optimal number. 128 general purpose (64-bit), plus 128 floating-point (64-bit), 64 predicates (one-bit) is plain stupid unless you don't have to run more than few programs (or treads) per CPU available. Context switching (from tread to tread) would required saving and restoring 256 64-bits registers. Also, high level language procedure call and return would have to do that saving-restoring... The whole EPIC design seems to be geared at running single threads stupendously fast. -fyo

To: bacchus_ii who wrote (57563)	10/7/2001 12:35:05 PM
From: wanna_bmw	Respond to of 275872

Gottfried, Re: "128 general purpose (64-bit), plus 128 floating-point (64-bit), 64 predicates (one-bit) is plain stupid unless you don't have to run more than few programs (or treads) per CPU available." That's exactly the point. IA-64 was an architecture that was to be designed for the future of microprocessing, not just the present. And, if you listened to Paul Otellini's keynote at IDF, you'd know that Intel has plans for a multithreaded Itanium processor in the future. There are also plans for a multi-core Itanium, as well. wanna_bmw