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: wanna_bmw who wrote (77251)	4/16/2002 2:28:46 PM
From: tcmay	Read Replies (1) \| Respond to of 275872

"Dan, Re: "RIP, Itanic." "HP to Provide U.S. Department of Energy Laboratory One of World's Fastest Supercomputers Agreement Results in World's Most Powerful Linux-based Supercomputer "Consisting of 1,400 of the next generation of Intel® Itanium(TM) Family Processors (code-named McKinley and Madison), the new HP supercomputer would have an expected total peak performance of more than 8.3 teraflops -- roughly 8,300 times faster than a current personal computer. " The death announcements for the IA-64 have been premature. Some game fans look at the Quakemarks and conclude "Hey, give me a souped-up Athlon 2300+ instead! Intel is dead!" Companies with long-range views, companies like H-P, IBM, Sun, Compaq, these companies have to look at the base of their computers for decades to come. And they all realized that older x86-based architectures cannot be simply souped-up with 64-bit registers bolted on and thereby become the core of their future systems. H-P pursued the PA-RISC line, IBM the Power4 line, Sun the Ultrasparc line, and Compaq/DEC the Alpha line. NONE of them, not one of them, was planning to build large "enterprise" systems around x86 or even souped-up x86 processors. So, the jump to a clean, 64-bit from the gitgo, architecture was made by all of these companies. And Intel saw the math on the wall. Some of those companies, notably H-P and Compaq, looked at the development costs and fabrication facilities required to support a state of the art 64-bit processor...and they sought partnership with Intel, the company with the proven best ability to make very large processor chips. You should all know this part of the story. The fact that Quake and maybe Microsoft Office have been slow in coming to the IA-64, or that Dell floated a machine and found, not surprisingly, not many sales to end users, doesn't change the basic calculation above: the building blocks for large computers in 2005 will almost certainly NOT be x86-based. AMD may well do pretty well with the Hammer, if they can resolve the support chip and process (SOI thread here) issues. But even they have to be looking out several years and realizing that the Hammer is basically a late 70s architecture with a bunch of kluges bolted on. Even they must be realizing that IBM is using an advanced 64-bit architecture, Sun is using one, HP _would_ have used one of their own had they not partnered with Intel, and so on. Intel is covering both bets. But announcements like this supercomputer one tell us that a lot of activity is happening that is not included in the "Dell discontinues prototype box...Quake doesn't run fast on Itanium" noise we see here. --Tim May

To: wanna_bmw who wrote (77251)	4/16/2002 3:37:51 PM
From: Petz	Read Replies (1) \| Respond to of 275872

Consisting of 1,400 of the next generation of Intel® Itanium(TM) Family Processors (code-named McKinley and Madison), the new HP supercomputer would have an expected total peak performance of more than 8.3 teraflops -- roughly 8,300 times faster than a current personal computer. Well, which is it? McKinley or Madison? These CPU's aren't even available yet and they only have 71% more flops than a $250 XP2100+. I find Madison's floating point performance underwhelming. That's a theoretical peak of 5.9 gigaflops per CPU. An Athlon XP2100+ at 1.73 GHz has a theoretical peak of 3.46 Gflops (64 bit floating point). Petz