- From: Tomasz Motylewski <motyl_at_stan.chemie.unibas.ch>
- Date: Sat, 21 Dec 2002 00:20:28 +0100 (CET)
I am not sure whether this is just my experimental Comedi version (current CVS+CMDF_LOOP flag), but I feel that it would also apply to not modified CVS version. I am using NI PCI-MIO-16E-1 card. I first start AI (start_src = TRIG_NOW). Then I start AO (start_src = TRIG_INT). I get in my logs: before measure_setup() HERE AI STARTED before output_setup AO CMD: 4 after command_test AO AO CMD: 0 after command_test AO AO CMD: 0 after command_test AO AO CMD: 0 after command AO ni_mio_common: ai error a_status=a4b3 ni_mio_common: wait for dma drain timed out AO TRIG: 1 after INSN_INTTRIG If I do not do AO, AI works very well in continous mode. If I do AO, I can see only about 400 samples in AI buffer, after that it is stopped. But AO works OK after that in continous mode. I will use INSN_INTTRIG to start both AI and AO at the same time, but I have tried first this way. Best regards, -- Tomek P.S. I have got working comedi with my buffers (part of mbuff) and without overrun checking (loop mode) both AI and AO. Above error is the last remaining problem (OK, buffers not aligned on page boundary is "nice to have"). Should I post a patch showing this spoiled Comedi version?
Received on 2002-12-20Z23:20:28