Pardon me, I use FPGAs all the time, and I don't think I'm
incompetent. Yet, if I'm going into mass production, the FPGAs
get taken out. You won't find one in your mouse, for instance.
(By mass procudtion, I mean more than a volume of 50,000 or
so.)
For small volumes, you can't beat FPGAs, just like you once
couldn't beat the popcorn logic that FPGAs replaced. The
question we really need to answer is:
What percentage of the popcorn market has already been
taken over by FPGAs? Cause if everybody is already using
them, we really can't expect big growth out of the industry
that supplies them. It is only when a market niche is in the
process of being created that the big growth occurs. After
the market is filled, your growth returns to the growth of the
industry as a whole, until some other competitor comes along
and replaces you, in which case you decline.
My (nearly) total guess is that a pretty good fraction is already
in the FPGA camp, and that is why the growth rate in the
business is slowing down even in the face of an excellent
economy.
Just thought of an unbiased way of determining just how much
FPGAs have infiltrated the engineering business. Take a look
at advertising for engineering positions. Compute the percentage
that mention FPGA experience as a requirement. Check to see
if this percentage is still increasing with time.
The above research can be performed by anybody with an
interest in the answer. I don't have any money riding on the
question. How's about one of you financially interested types
down in Si valley computing the quarterly figures for the last
3 years? I am a little curious, and I'd love to hear the results
of your research.
-- Carl |
