RE: mmap vs cmd

Hi! Yes, thanks, the increased buffer size did the trick for
the demo/cmd, I can now sample at full 200kS.

I still have the (academic, by now-) question of why demo/mmap
doesn't work unsolved, but at least I don't have to worry about
it. The next thing to figure out is how to do real streaming
input for a looong time, the counter that cmd uses is obviously
just 24bits on the pci-6023e...

> -----Original Message-----
> From: Tim Holy [mailto:holy_at_pcg.wustl.edu]
> Sent: 3. september 2002 17:56
> To: Örnólfur E. Rögnvaldsson; 'comedi_at_comedi.org'
> Subject: Re: mmap vs cmd
> 
> 
> Hi,
> 
> > [..] The reason for
> > wanting to use mmap is that I thought it would be faster, and
> > thus to be able to sample at full 200ks (with cmd, I get
> > "read: Input/output error" once I crank up frequency and number
> > of samples). [..]
> 
> Possible workaround: to use cmd, you may wish to increase the 
> size of the read 
> buffer. I use the following line in my /etc/rc.local file:
> 
> /usr/sbin/comedi_config /dev/comedi0 ni_pcimio --read-buffer 4096
> 
> This may (or may not) solve the problem you're having with cmd.
> --Tim Holy
>

Received on 2002-09-04Z06:47:08