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: fyodor_ who wrote (69501)	1/30/2002 11:29:25 AM
From: Joe NYC	Read Replies (1) \| Respond to of 275872

Fyo, If you look at the Hammer presentation, page 25, it appears that Sledgehammer will have a built-in 4 way switch (XBar or CrossBar). If this is in fact a swich (which I think it is), not some sort of hub or a daisychaining device, the performance penalty / latency would be less than daisychaining, since the number of hops is reduced. platformconference.com Joe

To: fyodor_ who wrote (69501)	1/30/2002 1:02:55 PM
From: Tenchusatsu	Read Replies (2) \| Respond to of 275872

Fyo, <With HyperTransport, won't this be a heck of a lot easier [than EV6]?> Actually, it will be harder. The biggest problem in a 2-way Clawhammer system will be keeping the two memory interfaces coherent. Compare this to the 760MP, where there is only one memory interface, and one single northbridge collecting and controlling all of the memory accesses. With 2-way Hammer, the memory accesses have to be split between two interfaces, and that's not easy. Once you get it to work, of course, you'll get some incredible scalability. Tenchusatsu