Intel Corp (INTC)

[alan81]

Ah, I think you have it!
Failure is almost always neccessary on the way to success. I am not sure if you play bridge, but there is a rule that if you do not "go set" (fail to make your bid) about a third of the time you are not being aggressive enough. I think the same thing applies to successful companies; they should fail about a third of the time if they are being aggressive enough. Of course, it is important not to fail on the really important things. There are some companies, and some times with every company, that they play "bet your company" on a single program, and in these cases you must not fail. If every program is funded to guarantee a lack of failure, you are over funding.

Kate's original point was that AMD could not fail at these things because they did not even bother to try.

I do think there is a public perception thing at work, which makes it important to keep programs that are more likely to fail hidden away from view, which allows the companies to maintain an appearance of a higher success ratio than they really have. In terms of your penchant for just counting failures, keep in mind that Intel has about 10X the revenue, programs, projects, etc, that AMD does so you would expect around 10X more failures if both companies have about the same operating point and tolerance for failure.
--Alan