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: maui_dude who wrote (170393)	8/30/2002 1:12:00 AM
From: NITT	Respond to of 186894

Re: "Not bad, if their prediction comes true. thats $31B of sales from the high margin chips. Add celerons, chipsets, flash and communication revenue and they must predict about $50B+ revenue in 2007!" I think that the author is referring to system revenue in his piece and not uP revenue. I don't have a lot of respect for these analysts; they missed the real PC ramp during the 90’s by a bunch. But yes even this forecast would mean growth for Intel in these segments. Let's hope Intel can see its vision of turning the high-end server into a commodity business that they control the heart of... IBM the service company would be lining up to support them and collect that very high margin consulting and support business. Nitt

To: maui_dude who wrote (170393)	8/30/2002 2:15:31 AM
From: wanna_bmw	Respond to of 186894

Maui, Re: "servers running UltraSparc, the competing chip family from Sun Microsystems, will account for $6.6 billion. Revenue from IBM machines using the Power processor will total $8.6 billion, Hewitt said." UltraSparc is all but dead. Sun is late, late, late in delivering their UltraSparc IIIi chips introduced at last year's microprocessor forum, and these were nothing more than simple shrinks of their current UltraSparc III processors with minor enhancements. What does this say for UltraSparc IV? V? Not to mention that Sun seems to be failing as a company. I just don't see the UltraSparc line being a dominant player in 2007, unless someone else powerful takes it over. And who would have any reason to do that? As for Power, that has a future for as long as IBM wishes to invest in it. It is competitive with Itanium 2 now, but I don't know how long IBM will want to continue investing funds into microprocessor development when they can get it from Intel for much cheaper. Right now, I think Power is just a hedge for IBM, while introduce the Itanium line to their customers. If their customers see the price/performance of the Itanium line over Power, IBM may decide to move Itanium into its high end product lines. IBM has been known to switch architectures to the whims of their clients before. It's the benefit of hedging, of course. wbmw

To: maui_dude who wrote (170393)	8/30/2002 10:09:05 AM
From: Dan3	Read Replies (1) \| Respond to of 186894

Re: By 2007, yearly revenue from ... I really think this guy is way, way, off. Which chip will do better in that time frame, I don't know. But long, long, before 2007, the notion of a 32-bit server or workstation will be ludicrous. I've got program managers now who were disappointed that they were basically limited to 3 gig of useable RAM on servers they ordered months ago. A year from now, when Intel is pushing Madison hard, or has released Yamhill to keep Hammer from taking too much share, even entry level servers will be marketed with the capability for 4 or more gig of RAM. By 2005, I'd expect $500 to $1,000 SNAP-type servers to offer models with more than 4-gig. For a database or web server to maintain responsiveness under load, it must be able to keep indexes and/or frequently hit pages in cache, and that means a lot of memory. 4 gig of RAM, which is more than can be used by a 32-bit CPU, is down to $1,150 on pricewatch (as 4, 1gig DIMMS). A year from now, it will almost certainly be less than $500. We have bought a couple of 32-bit servers with 4gig of RAM, but it was something of a waste, since the hardware peripherals (video card, disk controllers, network cards, etc.) reserve big chunks of the address space such that 1/3 to 1 gig (depending on the other peripherals selected for the box) is never used because it can't be accessed. The main point is that I had programs that wanted 4gig+ servers a year ago, and couldn't get them. As far as going to SUN, Alpha, or Itanium is concerned, those solutions are too expensive for these programs, and (with the exception of SUN) some of the applications that needed to be run weren't available. Remember the universally acknowledged wisdom of the statement "the 32-bit 386 will be limited to servers for the next 3 to 5 years" - and then Compaq came out with its Deskpro, to the befuddlement of many. IBM didn't bother to offer a 386 system for many months because they were convinced there was zero demand for such systems. If this guy were around then, that's the kind of stuff he would have been writing.