N. T. wrote:
> Have you noticed I said "at less than 22 nm process and mass
> production"?

Yes, but I ignored it since it's irrelevant.  We are talking about a
relative difference between dedicated micros and soft cores.  You have to
assume both manufacturers have access to the same fab technology.  With the
same fab capability, a soft core is going to take more area and therefore
cost more than a dedicated micro.  There will always be cost-sensitive
applications where such a cost difference matters.

The two are apples and oranges already, and that will be more so as the
geometry gets smaller.  Some micros are deliberately at a larger feature
size today to get low quiescient current.  Very small FETs have to use lowe=
r
voltages and leak more, at least with today's technology.  Look up what
fraction of a modern x86 processor's power is due to leakage as apposed to
switching losses.  It may surprise you.  That level of leakage is
unacceptable in a number of micro applications.  Again, dedicated micros an=
d
soft cores are apples and oranges.  I don't think the two compete
substantially today, and I don't see that changing as feature sizes get
smaller.  We'll just see more stuff in the same package at the same price
for both cases.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.