re: [Intel cost > AMD's]
Since you pointed out an error in my calculations, let's redo them with the right data.
1. Intel's current ASP. I said it could not possibly be lower than $150. In reality, a very conservative (i.e., low) estimate is calcualted as follows: (IAG revenues) x 0.87 / 30M CPUs. I'm allowing 13% of IAG revenue for chipsets.
20M chipsets @$30 = $600M
30M CPUs @??????? =$5233M
----------------------------------
TOTAL IAG REVENUE= $5833
minus IAG Profit= $1843
EQUALS IAG COSTS of $3,990.
The ASP that falls out from the above is $174, closer to the high end of my $150 to $190 range.
Now, I'm going to be very generous and assume that, magically, Intel is able to jettison all of their non-CPU business that collectively had a loss of $567M.
Their IAG group had a profit of $1843M. So, if ASP dropped to $110, they would have CPU revenues of $3,300M, total IAG revenues of $3,900M, and a LOSS of $90M.
OK, how would AMD do with an ASP of $110 in Q2? Is that too low to make a profit? That would be an ASP increase of ($110-$62), or $48. On CPU sales of 6.5M, it would result in an additional $312M of revenue. AMD would easily overcome the losses in its other divisions and have a profit of ($312M - Q2 loss of $140M), a profit of $172M.
Of course, by jettisoning its losing divisions, Intel would have to instantly recognize all of the acquisition costs it has been ignoring the past 10 years. It would have billions of writeoffs. In addtion, AMD's flash sales would double as Intel drops a business they have never made money in. (Loss of $120M on sales of $465M last quarter.)
Now, isn't it obvious, that if company A can make money selling a product at $110 and company B cannot, company A, in this case AMD, has the lower cost of production?
Petz
