- From: Daniel Nilsson <comedi-list_at_dnil.se>
- Date: Fri, 17 Feb 2006 19:33:53 +0100
Hi All, On Mon, Feb 13, 2006 at 03:03:10PM -0700, Jim Benson wrote: > On Mon, 13 Feb 2006, Daniel Nilsson wrote: > > > > > Should I expect to be able to sample at 10kHz using this card? How can > > > > I figure out why the FIFO is not emptied? > > > > > > > > > > I don't know...though a few years ago I was running a > > > NI AT-MIO-16E-1 (yup the 16E-1) with a 450 MHz machine, > > > with RTAI. My scan rate was 200 kHz with 4 channels. The > > > 16E-1 however has a 8196 byte fifo. When I switched to > > > the PCI version of the card, which does only have > > > a 512 byte fifo, my 3 GHz machine with RTAI > > > was not fast enough, until I went to DMA. > > > > Hmm, discouraging... I need a faster machine then, or sample slower > > (which is actually a possibility for me). New HW is a little too > > expensive though for this little project of mine... > > Please wait for advice from the real comedi experts before > getting discouraged. If you haven't tried already, > you might try slowing it down and/or try sampling > fewer channels and see if the problem goes away. So I made a few more attempts to resolve this problem... I upgraded to the latest kernel (2.6.15.4), the latest comedi drivers (from CVS as of yesterday). That didn't make any difference that I could notice. I also lowered the sampling rate to 1000Hz, but I still get the same error from the scope application: scope: read: Broken pipe dmesg says: ni_mio_common: ai error a_status=e4b3 I'm now running the card at 1/100 of it's capacity (1kHz versus 100kHz) sampling rate. I must be missing something, because if that still is too fast I don't know how to get to a stable situation. The FIFO buffer should now be full every 250ms which is a rather long time ever for my old Pentium Pro 200 CPU. The scope application also seems to be failing after about the same amount of time (a few minutes) regardless of the sampling rate so I don't really think it has to do with the kernel getting to many interupts per second. In fact, if I sample with 1kHz the counter in /proc/interrupts isn't even increasing while the card is running (it is though at 10kHz). Maybe it's hard to give solid advice on the cause, but maybe someone could suggest a method of debugging this issue? Thanks! -- Daniel Nilsson
Received on 2006-02-17Z18:33:53