Re: Strangeness with cb_pcidas and 1602/16 board

On Fri, 9 Aug 2002, David Schleef wrote:

> On Fri, Aug 09, 2002 at 11:04:45AM -0400, Calin A. Culianu wrote:
> >
> > This behavior isn't exhibited with other drivers for other boards
> > (ni_pcimio, for one, doesn't have strange interactions between channels).
>
> Yes it does.  It's just at a lower level because the settling time
> is very short on NI boards.  Try a full speed 2-channel acquisition,
> channel 0 with a 10 Vpp sine wave, and channel 1 grounded, on the
> highest gain setting.

Yeah you're right.. I got interactions with extremes of gain/signal and
with fast acquisition.

You haven't replied to my last email, which is ok, I realize I rambled a
bit and basically agreed with you.  But how do you feel about, in
principle, alleviating some of the voodoo behind guessing the wait times
in the application by possibly building infrastructure in comedi that
allows a user to query the wait time for his multiplexer?  This way, (at
least ideally), as multiplexer settling times are determined for various
situations, the times can be put in the driver and so each application
that uses comedi doesn't have to 'reinvent the wheel' in determining
multiplexer settling times.  Right now, each application programmer that
wants to use more than one channel ultra-correctly needs to (at worst)
play a
guessing game on a per-board basis, or at best read the spec for his board
and write his application in a board-specific way...

I realize this isn't the first time in history applications have to be
hardware-aware, esp for performance reasons..


How does this sound?  Or am I beating a dead horse here?  :)


-Calin

Received on 2002-08-13Z14:04:12