AMD grew sales 50 per cent year on year

Spansion factor 'hides' expansion between 2004 and 2005

Letters Last week, AMD appeared to have slipped in the world chip rankings, according to numbers provided by market watcher iSuppli. Not so, said AMD. With the inclusion of figures from the spun-out Flash memory division Spansion - absent, we should say, through an error of ours, not iSuppli's - the chip maker's ranking would have been higher. But there's another question raised by the story: is like really being compared with like?

Regarding your story on the new Semi rankings from iSuppli, you're comparing apples and oranges for AMD and, in the process, reporting something that is misleading. Your headline reads, "AMD falls in world chip maker rankings," based on an iSuppli press release that 1) reports that AMD took nearly three percentage points of market share from Intel in 2005 and that 2) doesn't provide an accurate, Apples-to-Apples comparison of our numbers year-over-year. Something here doesn't quite add up. If Intel had spun off half of its business, would you adjust your reporting to consider that? Or would you compare its pre-spinoff numbers from one year to its post spinoff numbers from the following year? Obviously, if you went with the second option and reported that Intel's ranking had fallen without adjusting the numbers for the spinoff, you would be misleading readers. When reporters cover congressional budget issues, they routinely adjust the dollar figures from past years to accommodate the rate of inflation so they can provide a much more meaningful comparison of budget growth over a period of time (apples-to-apples). This is precisely the same thing.

For the record, iSuppli's ranking for AMD in 2004 includes Spansion. Their number for AMD in 2005 doesn't. And while you acknowledge this, you don't provide any meaningful context beyond saying that AMD would have been number eight with Spansion. If you take Spansion out for both years - which would give a much more meaningful, apples-to-apples comparison - you would see that AMD has grown its business by a whopping 50 per cent. In 2004, our revenue without Spansion was $2.6bn. In 2005 without Spansion it was $3.9bn. But the iSuppli numbers compare our $5bn revenue figure from 2004, which includes Spansion, to the 2005 figure of $3.9bn that doesn't include Spansion. Seems to me you should at least fill this hole in for your readers. Even iSuppli's report helps to explain it by giving a little more context. And yet, your story praises Intel for having grown their business by 13 per cent. You might want to at least provide some more balance to your article. At least add in the figures I've cited here to give readers an unobstructed view of all the details, not just some of them, so they can come to their own conclusions.

Eric J DeRitis,

Sponsored: HPC and HPDA for the Cognitive Journey with OpenPOWER