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: Tenchusatsu who wrote (239624)	8/30/2007 9:37:32 PM
From: Dan3	Read Replies (3) \| Respond to of 275872

Re: The funny thing is that Intel has been speculatively fetching data for quite some time now. It's called "memory disambiguation," Exactly his point - it's why 4 socket Intel systems are barely faster than 2 socket Intel systems and there's no point to an 8 socket Intel system since it's slower than a 4 socket Intel system (unless you add a heroic chipset bandage that only HP and IBM seem to be able to design. Intel's MP chipsets haven't been able to cut it.

To: Tenchusatsu who wrote (239624)	8/31/2007 1:45:05 AM
From: pgerassi	Respond to of 275872

Dear Tench: Memory disambiguation can be shown to yield correct results. Intel's implementation is like that with CSI though. They speculatively do it in a limited number of cases, but rewind, if it fails. The big difference is that n is small like the number of cycles to calculate an address, IIRC. That is much more limited in scope than a CSI "hop". AMD, OTOH, will implement it (MD) in a way that doesn't speculate and is easily shown to be always correct. That makes the risk adverse, happy. Pete