You might reconsider this assumption.  I know people who
would beat the living tar out of you if they traced their
model's sudden lack of responsiveness back to your product..
;)  I'm not even that critical but I can tell the difference
in 50 to 60 Hz update rate at times, you must remember it's
multiplied hundreds of times sometimes before the final
results are apparent.  At least state clearly that it's not
for critical functions if you don't want people coming back
at you with things like 'why'd your inferior product crash
my model' etc.  There are even airplane people who complain
about things like this, and it's much more noticable on
helis..  Shouldn't really be much more hardware/programming
to get it into the 60 Hz and high res range anyway, and a
much superior end result..  You're right in that 90% of
flyers won't notice, but that last 10% can be pretty vocal.
There is a difference between following the curves input to
the sticks with the normal 1/4 second time lag and following
a more linearized version of that function due to lowered
update rates, regardless of the resolution of the endpoints
of the line segments..
Alan



I decided that even a realtime-controlled, model airplane
does not need a servo update of 50 times/sec.
    Thats means that every single pulse the servo receives
would be different!  I can do exponental/differential
    rates myself in software.  Simple calculus/physics is
all that takes. Resolution, however, is more