Dear Monica:
You forgot a few things, lets go back to Q4 2000, $21.15B and Q1 2001, $18.739B. That's a drop of $3.65B in 5 quarters. Check cash for the last 2 quarters, Q4 2001, $10.326B and Q1 2002, $9.231B, for a one quarter drop of $1.095B (it was $13.473B in Q4 2000). An average burn rate of $850M a quarter. At that rate, they will be out of cash in 11 quarters. At the rate last Q, they will be out Q2 2004, 2 quarters sooner.
YOY needs to compare Q1 2001 with Q1 2002, which you did not do (it was down $1.25B or down 6.7% YOY).
Second, revenue being down $200 million does not translate into a loss of $200 million as gross margin was above zero. Taking your method, you must argue that Intel is down $500 million in revenue, thus has a $500 million loss. Even I do not make such a fundemental error! And if Intel has $5B in revenue for a loss of $1.7B, I would not assume that they had that big a loss. More like $800-900M.
And do not start with it could not happen to Intel. Intel assumed a normal backend loaded June (June is 50% of Q2) when they put out the flat revenue guidence. If it was flat (June = April = May), a $1.7B hit to revenue is what would result.
Hammer was running at 1.6GHz prior to the decision to only hand out 800MHz clock locked CPUs to OEMs for testing and validation efforts. So they were not just capable of 800MHz (You must have missed the Intel Celeron 300A that typically could go 450MHz with ease. That wasn't as fast as Intel could make them go), but twice that and probably more. Probably just wishful thinking on your part. AMD's reasons are good ones.
Pete |
