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: Windsock who wrote (165285)	5/21/2002 1:02:34 AM
From: Dan3	Read Replies (2) \| Respond to of 186894

Re: Where's Mine Jerry HOPES that AMD will be able to ANNOUNCE You're such a charmer. Meanwhile, judging by their stagnant or dropping availability in a slow market, Intel seems to be forgetting how to make P4s, what's up with that? My take is that Intel's problem isn't that they couldn't produce more chips if they wanted to, but that the market has moved in a direction they weren't expecting, and they're stuck in something of an economic corner. The trouble in paradise is partly due most corporations now having closets stuffed with last generation PCs that are too slow to be useful, but somewhat vexing to get rid of, since they're still being carried as assets on the books. It's making buyers think twice about paying that extra $500 for a box that's 10% faster, and will be adding to the trash problem in just a couple of years. With product cycles making PCs practically disposable items, buyers are refusing to pay more than disposable prices - at least, as long as it's just "more of the same, a little faster." Some of the magic (and willingness to pay high prices) has gone from the "new PC." And just as buyers are refusing to pay as much as they used to in the past, the cost to make these things is rising. Masks for the .13 generation of chips cost 3 times what they cost for .18. And while they can fit more (smaller) die on a (larger) wafer, additonal processing steps require more masks per wafer. Masks must be replaced periodically, and are adding significantly to variable costs. We all know how horribly expensive FABs are to build ($14 Billion in two years, for Intel) and that substantial portions have to be rebuilt or replaced within 18 months to two years as the next generation of chips is released. Although they do seem to be having trouble supplying the market right now, soon, Intel will be able to produce enough chips to support the entire market. But to take share away from AMD, Intel will have to price its chips nearly as low as AMD does. That would mean cutting their ASPs substantially, and they just can't do that with the steadily rising cost of making chips. The current issue of eeTimes has 3 articles, clustered together on page 1, that cast some light on what's going on: "Economics hold industry back from plunge into 90 nm" eet.com "TSMC's research chief mulls next-gen process knots" eet.com and a third, "Designers decry mask costs as CICC sounds scaling alarm" eetimes.com Intel has set itself up with a lot of very expensive FAB space to produce a lot of very expensive chips. If there were no AMD, Intel could restrict supply enough to keep prices at a profitable level. As it is, AMD (especially AMD's lower production costs) is providing plenty of high performance chips to the market at moderate prices. Which is making it more and more difficult for Intel to maintain the rising ASPs it needs to keep up with its rising costs of production. The market for CPU's is finite, and not very price elastic in the short or even medium term. As Intel pushes more production into the market to keep AMD from gaining share, Intel loses heavily in total revenue. For example, if the market's initial equilibrium is Intel 80% at $170 ASP and AMD 20% at $85 ASP, for Intel take back 1/2 of AMD's share it would have to cut its prices to something like $125 - which would grow the market some - say 10%. Intel's current (quarterly) CPU revenues are around $5 Billion, and their direct costs (including the costs of the FABs) to produce those CPUs are about $2.5 Billion. If they take back 1/2 of AMD's share, that's a factor of 1.1. If their lowered prices grow the market 10% that's a factor of 1.1. But to do this they have to lower ASPs by 25%, and raise cots by 20%. So, they wind up with 1.1 * 1.1 * .74 * $5 Billion or $4.5 Billion in CPU revenue and 1.2 * $2.5 Billion or $3 Billion in costs. Even if they don't have any additional marketing and administrative costs to sell those 20% additional CPUs (we're assuming cutting prices is enough), they still have $1 Billion less to cover their administrative, marketing, and "other" costs. At the same time, AMD has positioned itself to make money at lower ASPs by focusing very hard on designing small CPUs and keeping FAB costs low. So AMD may just cut its prices along with Intel, and stop Intel from gaining any share, and leave Intel with even less revenue (and a little lower costs). AMD also has the first really exciting product to arrive in this industry for a number of years - the chip that takes desktops and departmental servers, finally, to the 64-bit realm that has, until now, been reserved for expensive UNIX workstations and servers. Intel, with its high cost structure, is looking at the possibility that it will be on the other end of that $85/$170 ASP ratio as soon as late next year - just as their die size and capex have locked them into the highest cost structure they've ever had