Formerly About Advanced Micro Devices Message Board

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.

Politics : Formerly About Advanced Micro Devices

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Elmer who wrote (95704)	2/28/2000 7:34:00 PM
From: Tenchusatsu	Read Replies (3) of 1574261

Elmer, <I believe Foster will have that bandwidth to the processor this year.> Foster's bus will be the same as Willamette, 100 MHz quad-pumped, 3.2 GB/sec bandwidth. As for LDT, here's a problem with the quoted bandwidth. I think Ace's Hardware is getting confused with the bandwidth figures. They said, "The LDT bus ... boosts no less than 6.4 GB/sec on a 16 bit upstream, and a 16 bit downstream dataflow." This is kind of misleading. For two 16-bit unidirectional paths, LDT will have 3.2 GB/sec bandwidth each way, which will probably be quoted at 6.4 GB/sec total because data can move in both directions simultaneously. Sure, 6.4 GB/sec is a LOT of bandwidth, but it has to be looked at in the proper context. Speaking of which, in the specific LDT "foil" depicting a possible 8-way Athlon system, I think 6.4 GB/sec will be overkill for each LDT link. Even HotRail's HRC link (which is 95% similar to LDT) will each have a bi-directional bandwidth of only 3.2 GB/sec. So unless AMD is in the business of building big-iron mainframes, I think 6.4 GB/sec is a little too much per LDT link. Tenchusatsu

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next