> Giovanni Racciu wrote:
>
> It is a matter of fact that heavy DMA can affect (increasing)
> latencies and jitter. In our actual solution the software is running
> on a dedicated PC104 board, which is not corrupting the latency. But
> as a future develpment we would like to put it on a standard PCboard
> and the entire Linux OS plus the applications will run on it (using
> all the hardware features in a massive way).
> Is there anybody who has an idea on how big (worst case) could be the
> increasing latency and gitter ?
>
> Thank you.
Both NMT-RTL and DIAPM-RTAI have a latency calibration module. I always
got figures that have been confermed by using a scope with a module
generating a square wave, by toglling bits on the parallel port.
Use them with a quiet PC and with it loaded by a lot of disk activity
and network activity and you get what you need, first hand. No chance of
being very precise except by measuring on the very same hardware. So I
suggest you'll do it yourself.
Ciao, Paolo.
-- [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/
