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 : Compaq -- Ignore unavailable to you. Want to Upgrade?

To: Tomcat who wrote (87500)	12/6/2000 12:09:41 PM
From: John Koligman	Respond to of 97611

Tomcat, I'm certainly not saying 'unseat', the question is can they eat away enough market share to product pain. The other question will be how far they will go with Itanium. Time will tell. I've been pretty brutal knocking Mikey on the Dell thread this year as he has said lots of things early in the year that have made him look like a fool, so I'm in no way pumping Dell here. Just saying keep an eye out! Regards, John

To: Tomcat who wrote (87500)	12/6/2000 12:44:58 PM
From: rudedog	Respond to of 97611

Tomcat - excellent post. At the time of CPQ's announcement about OEMing the 32-way Unisys box - which was at the Windows2000 launch - Capellas went into some detail on CPQ's reasoning. He said that CPQ had planned to field a large Intel-based system in the McKinley timeframe, using the next generation of Wildfire technology as a base. This would allow CPQ to develop a single high end switch fabric architecture with common memory, storage and I/O subsystems, which could support both the Alpha base and also the Intel product line. This also explained the decision to drop NT support onm Alpha - the future large NT systems would be Intel based. Since that architecture is projected to deliver a 256 processor machine, it certainly solves the "big iron" positioning problem. But a small number of CPQ's customers were looking for a bigger box than the 8-way, primarily for server consolidation. The Unisys machine, which supports Xeon now and will support Merced, offered a quick solution for those customers - less than 100 according to Capellas - without taking CPQ down a dead-end engineering development. 32 bit systems can not effectively use the larger architecture, which is still limited to a 4GB physical address space, which means that they have to use PAE to address large memory (a paging scheme). So 32 bit machines are not suited to more than 4-way or 8-way consolidation even when the system has 32 processors. Merced has a memory scheme and an I/O scheme which is different than McKinley, essentially relegating it to being a test bed for IA64 architecture. I doubt that anyone will buy many of these machines, and I agree that Unisys will be the "leader" - if leading a 3 digit sales market is important. It does give CPQ a clean way to explore large system Microsoft-based solutions while they work the next generation development.