On Sun, 4 Apr 1999, Rob Butera wrote:
> On Sun, 4 Apr 1999, Donald Gaffney wrote:
>
> > I'm impressed that you can run a coherent closed loop at 70kHz. At this
> > rate it seems like the jitter would, at least in the limit, be on par with
> > the task period.
>
> ... averaged over 2000 cycles, on a given timing trial, worse case jitter
> was never worse than 2 microseconds (and typically around 1 us) (this is
Oops. Donald is correct. The jitter I was quoting above was for the
cycle duration, not the period. I don't remember the actual numbers,
but they _are_ 10s of microseconds, and _are_ approaching the task
period. I would check to actual numbers right now, but I powered
down the machine that all this work is done on before leaving
the lab today :(
Luckily, the jitter is a problem that is tolerable. The reason is this:
in our particular application, although speed is important, precision
is less important. This is due to the inherent variability on our
measurements to begin with. So a few slipped cycles are tolerable.
It does not manifest itself as a serious problem, since we have tested
our system with know input signals at 1 kHz, 10 kHz and 70 kHz, and
obtain similar results -- the nature of our computations is that
if a significant number of cycles were being skipped, the results
would become noticeable.
But Donald's points do make me want to go back and quantify this
to a greater level -- thanks!
--
Rob Butera, Postdoctoral Fellow http://mrb.niddk.nih.gov/butera/
Laboratory for Neural Control, NINDS
National Institutes of Health, Bethesda, MD USA
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
