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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: pgerassi who wrote (140432)	7/30/2001 1:13:43 PM
From: Mary Cluney	Read Replies (1) \| Respond to of 186894

Dear Pete: <<<It is interesting that both Mary and Tony think that most systems are built for computational loads and not database loads. Database loads are much more typical for servers. But, they keep pushing the computational configurations.>>> I only know you and Tony from what you post on this and the AMD threads. I think both of you have a lot of experience and I appreciate both of you sharing your knowledge with us. It is my opinion that you probably have broader experience than Tony. Tony's experience from what I can understand has been exclusively with very large companies where they have teams of people working in each narrow aspect of IT. You on the other hand have worked all over the lot, a man of many talents. Perforce, your clients are more demanding and cost conscious. Your ability to save your clients $50 here and $50 there by introducing cheaper but "just as good if not better hardware" all add up to much bigger initial savings. My guess though is that your clients for the most part do not appreciate all you do for them. They are paying Hyundai prices and want Maserati performance. Nontheless, I digress. My point however, is that your client base, is different than those Tony faces. Not only in the way they select hardware/software but also in what they do with their computer systems. My guess is that Tony's client do a lot of transaction processing - creating and using large databases, taking orders, billing, and applying payments, etc, but they also use these databases to do ERP type stuff where they do heavy computational, what if type departmental (marketing, logistical, budgeting, manufacturing, and planning) computation and regression analysis using databases created from transaction processing. But, in Tony's environment, wrt to something like a processor, they will pay top dollar for the same kind of equipment that their competitors are using. AMD has no visibility in that environment. No one is going to risk their careers by saving $50 on a processor when a system can go down for any of myriad reasons, and allow an adverary in the organization to raise the spector that it may have been caused by someone using hardware from a manufacturer without a terrific reputation. It will be a long time after AMD actually has products for this environment that it can earn sufficient respect to enter this environment. In a lot of ways, it may be tougher for AMD to break through this barrier, than it would be for women to break through the glass ceiling. To try to fool investors into thinking otherwise is something I object to very much. Mary

To: pgerassi who wrote (140432)	7/30/2001 1:17:02 PM
From: Tenchusatsu	Read Replies (1) \| Respond to of 186894

Pete, amazingly enough, I can't find anything in your recent post to disagree with. Except this: <Another curious fact is that 4 ways composed of two pairs of Xeons on two separate busses are much faster than four Xeons on a single bus. It shows both in database loads and web serving loads. This shows that the bus concept does not work well going from 2 to 4.> That statement is a little unfair. There's no way that a 700 MHz hefty-duty processor ISN'T going to be limited by a 100 MHz bus. (C'mon, a 7:1 processor-to-bus clock ratio?) Even huge caches can't serve as a total workaround. But hey, going from 2-way to 4-way on that 100 MHz bus still results in a substantial performance benefit. It might not be 100% linear, but it's good enough for now. <It appears that AMD and DEC were justified in their belief that P2P links were better than busses to connect CPUs to chipsets.> Obviously P2P links are better. They are also more expensive and harder to implement as well. We already saw this: AMD took their sweet little time getting to 2-way SMP, and they finally accomplished it with an expensive 900-pin north bridge. (Of course, the high cost is partially offset by AMD's willingness to sacrifice profits.) Tenchusatsu