On Fri, 23 Apr 1999, Paolo Mantegazza wrote:
> Yorck von Collani wrote:
> >
> > Thanks for the answer.
> >
> > Yes, 10-20 us is the time I have measured, so I have to use an external timer.
> >
>
> Pure illusion the jitter will remain the same as that of any internal
> timer.
>
> Just send a square wave to a pin of the parallel port and grab it with
> an RT interrupt handler send it back to a scope by outb, wich adds just
> a couple of us uncertainty, and you'll see that you have the same jitter
> for the same computer load.
> The jitter is embedded in the hardware architecture not in
> external/internal chips.
>
Exactly! Which is why I suggested the hardware remedies I did in my
earlier posting. If a project requires strict periodicity (<~20 us
max variability), then a hardware design is the only viable approach.
-Don
--- [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/
