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: Paul Engel who wrote (75231)	3/3/1999 2:03:00 AM
From: Paul Engel	Read Replies (1) \| Respond to of 186894

Intel Investors - Here's an interesting Editorial on Merced - with an initial BLAST at the Pentium III. "When you have emulators that run 80 percent faster with EPIC than without, when there are companies working on massively parallel versions of IA-64-based architectures, when almost every hardware company has announced support and when you know that Intel is probably the only company that can produce these things in large quantities, then you know that IA-64 is the future of high-end computing. More interesting is that IA-64 will include some things for that elusive five-nine (as in 99.999 percent) reliability. For example, IA-64 will have elaborate features that help it recover from individual processor failure. " Paul {===============================} zdnet.com The Pentium III and the EPIC conquest By John Taschek Last week, Hewlett-Packard and Intel detailed the EPIC architecture--just six days after Intel launched the Pentium III. Perhaps Intel feels that not enough people are paying attention to its technology. Hmm. Intel is pushing the PIII as the chip that's going to change the way we use the Internet. It even showed a few ways in which the PIII boosted the speeds of speech technology and 3-D rendering of computer games. The chip maker is so enamored with the technical merits of the PIII that it's spending $300 million to promote it. However, maybe Intel is spending the big bucks because it realizes the chip is just a chip. The PIII is notable because of its use of SIMD extensions, which promise performance boosts for things such as TCP/IP and graphics rendering. Even if it's true that the processor will boost graphics rendering, most performance gains will still come from individual 3-D graphics cards, not the CPU. In any case, most people won't really be able to perceive a difference--on the Internet or anywhere--since the main performance limitation is bandwidth. The interesting performance gains from the PIII will come from the server side, in which the PIII will provide some much-needed performance boosts. This means that the PIII may put a little pep into a few queries, or it may pump up someone's noticeably slow Web farm and so on. But no one really thinks the PIII is anything more than evolutionary. The more interesting, but less immediately relevant, announcement was HP and Intel's release of further details of the EPIC (Explicitly Parallel Instruction Computing) chip's IA-64 architecture. HP and Intel did not debut anything that they didn't preview in 1997--they just went into more detail. What's really interesting is the way HP described how IA-64 would systematically and categorically destroy RISC. And this was from the technical people! It seems that HP and Intel are convinced that moving to the EPIC architecture will blow away everything else. They may be correct, since from the theoretical point of view, IA-64 is a much better architecture. When you have emulators that run 80 percent faster with EPIC than without, when there are companies working on massively parallel versions of IA-64-based architectures, when almost every hardware company has announced support and when you know that Intel is probably the only company that can produce these things in large quantities, then you know that IA-64 is the future of high-end computing. More interesting is that IA-64 will include some things for that elusive five-nine (as in 99.999 percent) reliability. For example, IA-64 will have elaborate features that help it recover from individual processor failure. But killing known quantities is hard. Sun's hardware team and the Compaq/Digital Alpha team have proven reliability. The writing seems to be on the wall, though. The last chance for the Alpha might be to lower the cost of the processor to nothing--make it open source. This would make the Alpha processor more ubiquitous and put a damper on Intel, which by law must make the chip. Which one is more important right now--the PIII, IA-64 or neither? John Taschek can be reached at john_taschek@zd.com.