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 (69365)	1/29/2002 8:38:42 AM
From: Dan3	Read Replies (1) \| Respond to of 275872

Re: Gee, all it took for Intel to go to 8-way was a two-FSB bridge And all they had to do for an MTH was an RDRAM to SDRAM bridge. What happened there?

To: Tenchusatsu who wrote (69365)	1/29/2002 1:19:00 PM
From: pgerassi	Read Replies (1) \| Respond to of 275872

Dear Tench: Do you know that someone (Microway) is selling a 64 way (actually 66 way) SMP Athlon ccNUMA system? They use multiple Myranet 24 port 100mb ethernet switchs. The interconnect does not need to be traces on a PCB. Pete

To: Tenchusatsu who wrote (69365)	1/30/2002 11:22:23 AM
From: fyodor_	Read Replies (2) \| Respond to of 275872

Tench: Because of the P2P interface, AMD needed a two-FSB bridge just to go to 2-way. And it took them quite a while just to get there. Think about how long it will take them to get to 8-way. With HyperTransport, won't this be a heck of a lot easier [than EV6]? From what I've read, HT components can be daisy-chained, thus relieving the need for an 8-way switch. The downside would be a performance penalty stemming from the higher latency of going from one end of a daisy-chain to the other - as opposed to just having all the connections switched. -fyo