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: Joe NYC who wrote (111999)	10/1/2000 12:39:40 PM
From: steve harris	Respond to of 186894

Joe, re:But even AMD can't hope for more than 20% total market share, and that share is destined to decrease as Intel continues building more manufacturing capacity. That sounds like a Kumar analysis. Anyway, Intel currently has capacity to supply > 300% of the market, if Intel decided to build 486 chips instead of P3 / P4. But Intel has less than 1% capacity to supply market with 1 GHz CPU, if all people were willing to buy is 1 GHz chips. I agree. No product, bad management, what good are FABs? Some people still trying to sell the mantra: "who needs that much power anyway" Whatever happened to Kumar trying to sell that "sweet spot is 550Mhz" crap. And whatever happen to DELL trying to dump dual P3-500 systems to the consumers? A flood of P3-700s aint going to hurt AMD, just blow INTC's "investment portfolio". Even going to .13 and copper next year, Intel will still be in AMD's rear view mirror. They can sell cameras and video games I guess. :o) steve

To: Joe NYC who wrote (111999)	10/1/2000 2:04:15 PM
From: Paul Engel	Read Replies (3) \| Respond to of 186894

Joe - Re: ", but if all Intel has is a single core CPU with traditional architecture vs. dual core Sledgehammer with onboard Northbridge, do you think it is 100% certain that Sledgehammer will lose?" If you assumed that Intel did not have dual core x86 32 bit and IA64 bit cores underway, you would be very naive - and wrong. Oh - one other thing. The dual core IA64 bit project is based on FUNCTIONAL and PROVEN existing single core IA64 ITanium. You might ask yourself - What is the dual core Sludgehumper based on ? Paul

To: Joe NYC who wrote (111999)	10/1/2000 7:43:11 PM
From: Tenchusatsu	Read Replies (1) \| Respond to of 186894

Joe, <How is it that Intel will be able to enter the market with a chip that is totally incompatible with anything else, but AMD will not be able to do the same with a chip that is compatible with installed base.> IA-64 is indeed compatible with x86. Let me explain things a little more. Take a look at AMD's x86-64 architecture spec, page 2 (page 6 of 112 in the PDF doc): amd.com In it, you'll see three modes of operation. The first mode, called Legacy Mode, is the only mode you can run if you are using a regular shrink-wrapped 32-bit OS like Windows ME or the current version of Windows 2000. But in that mode, you cannot take advantage of the x86-64 extensions like the extra registers or the 64-bit addressing. The second and third modes, called Long Mode, requires a 64-bit OS (of which none have been announced as of yet). In Long Mode, 32-bit apps will run under Compatibility Mode, and those apps still cannot take advantage of the x86-64 extensions. The only way you can take advantage of the extensions is by running in 64-bit Mode, but that requires specific software support, just like IA-64. Now, take that table, chop off Legacy Mode, and keep only the two Long Modes. That's IA-64. The two things IA-64 loses will be lack of legacy OS support, and slower performance in Compatibility Mode compared to 32-bit processors. But IA-64 performance in 64-bit Mode is expected to be even better than that of x86-64. (Maybe not with Merced, but by the time Sledgehammer is out, McKinley will also be here, so the real comparison is with McKinley.) Intel is banking on the hope that IA-32 compatibility is enough for IA-64 customers, and that existing IA-32 applications that really need the performance will be the first ones to be recompiled for IA-64. AMD is banking on the hope that customers of 64-bit platforms not only value 32-bit compatibility, but also 32-bit performance. Of course, they'll sacrifice potential 64-bit performance relative to the competition. Nevertheless, AMD sees an opportunity that Intel isn't taking. Then again, maybe Intel did the right thing in passing on the opportunity? We shall see, but in my opinion, Intel's decision is the better one. (How Intel executes to that decision is another story, though, and I need not comment on that right now.) <It seems that you are assigning probability of zero for this to happen. [Sledgehammer outperforming P4 and P4 Xeon.]> Actually, I'm not sure at all. Your opinion is as good as mine. But I AM sure about this: Future performance will depend on specific software support. There is no such thing anymore as designing a processor that doesn't require specific software support. SSE2, x86-64, IA-64, thread-level parallelism, etc., all require specific support from software. Athlon and Pentium III might be the last processors to not really require it, and even those processors benefit from compiler optimizations. Tenchusatsu