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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Elmer who wrote (45510)	1/11/1999 11:55:00 PM
From: Petz	Read Replies (2) \| Respond to of 1572014

EP, "$10 [packaging cost incl. depreciation], is waaaaaay too high. Probably more like $1-$2." A year or so ago, a fellow called "Patient Engineer" estimated AMD's packaging/testing costs as $5 per chip for direct costs. So $1 to $2 for Intel's "test only" cost is probably about right. But total packaging costs per CPU are certainly above $10 for the Slot 1 package, much more if there's L2 cache involved. No analysts try to separate depreciation into multiple categories like you have been suggesting. (Fab building depreciation, fab equipment depreciation, obsolete equipment depreciation, test equipment depreciation, etc.) Securities analysts don't care -- they use the total depreciation estimate given to them by the company. However, the FTC might be interested in trying to assign depreciation costs to product categories for the purpose of determining if a product is being subsidized by profits from a monopoly market. I used the simple approach of saying each chip requires roughly the same type and amount of facilities to produce it, and therefore there is roughly $22.50 of depreciation per CPU chip, assuming Intel derives 75% of its revenues from the sale of CPU's. (I vaguely remember the 75% number from Pauly.) Intel would like to say that depreciation, R&D and Overhead (MG&A) should be apportioned to products based on their revenue contribution. Then they can claim that the Pentium II has $40 of depreciation per chip, but the Celeron only $10. That makes sense until you realize that Intel can make more Pentium II CPU's in a Fab than they can make Celerons, due to its larger die size. Similarly, take marketing expense -- right now I see a lot more Celeron ads than Pentium II ads and there's practically no advertising cost from Intel for the Xeon chips. R&D? The PPGA 370 is a clever design, then there was the original Celeron and the short time fuse on these projects pushed up the costs, no doubt. I'll concede though that big chunks of R&D like Katmai and Merced are totally unrelated to the $70 Celeron chips. Petz