SI
SI
discoversearch

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 ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Dan3 who wrote (82851)12/14/1999 1:28:00 AM
From: Tenchusatsu   of 1575354
 
Dan, <it's years ahead in its chipset/bus - because it has licensed the Alpha EV6 technology from Compaq/DIGITAL. ... And Itanium will hit the market on a version 1.0 bus architecture>

Wrong. The Merced bus is a shared multiprocessor bus which is pretty similar to the P6 bus. Gee, doesn't this run contrary to AMD's claims of "superior EV6 technology"?

My guess is that the architects looked at P2P buses and decided they weren't worth it, at least for now. Or it could be just simple leveraging of the proven P6 bus into a new architecture, and that was deemed a better strategy than going P2P. Time will tell.

<although, since Merced is several years late, there has been lots of time to develop its chipset - in simulation, at least>

Wrong again, at least on the "simulation" part. 460GX taped out, booted up, and ran on several operating systems even before Merced taped out. Among other things, that allowed Intel to demonstrate Itanium-based systems at the August IDF in Palm Springs. And that was just a couple of weeks after first silicon came back.

<And AMD has a perfect solution for the traditional X86 server market - while Itanium has been targeting the high end, $100,000 and above market.>

Wrong again. No 460GX-based system is going to sell at price points above $100K. The predecesor of the 460GX was the 450NX (Xeon 4-way chipset), and no 450NX system sold for more than $50K or $60K. Charging any more for such servers would go contrary to Intel's SHV strategy (Standard High Volume).

Sure, other companies like IBM/Sequent, NEC, Bull, etc., are going to create custom Itanium-based solutions that target the >$100K range. But none of them will come close to the cookie-cutter volumes that 460GX will enable.

<Sledgehammer's competition is Coppermine on single, dual, and Profusion based servers, not Itanium.>

Wrong again. By the time Sledgehammer is out, Foster should be well-entrenched in the 4-way and 8-way x86 server space. Profusion will be a fading memory by then.

And if Sledgehammer is only a K7 with 64-bit extensions and a few extra floating-point enhancements, it will fall severely behind Foster. Dual core Sledgehammer? Bah. (And there are several reasons for the "Bah," but you don't need to know what they are.)

Tenchusatsu
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext