06-13-2006 10:06 AM
06-13-2006 12:27 PM
Absolutely. You can try this out for static DIO by simply running two of the shipping examples simultaneously -- 1 each for Input and Output.
It's also possible for correlated DIO. I have an app that uses 8 bits of the 32-bit port for hw-timed output while simultaneously using the other 24 bits of the port for hw-timed input. In my app, both are driven by the same hw clock but in general, they wouldn't have to be.
-Kevin P.
06-13-2006 01:19 PM
Great !!
But semi-continuous transfer from app code space to and from the DIO?
i.e. App buffer space (malloc) with ~1.2 meg of waveform transferred out @ ~18 kHz (66 sec of data)
AND
waveform (data) transferred in @ 18 kHz to an App buffer of ~ 1.2 meg?
Thanks for the info,
K.O. Smith
06-13-2006 02:06 PM
I can't speak from direct experience, but I would expect you can do 18 kHz for a minute. My app regenerates a small output buffer and the I/O rate is 10 kHz for 30-45 seconds. A bit less demanding than yours, but same order of magnitude. Don't recall CPU load offhand, but I often launch Word or Excel while running data acq without any difficulty.
Specific details & syntax: I probably can't help. My app is entirely LabVIEW. Based on your reference to malloc(), I figure you must be working out of CVI or directly from C.
-Kevin P.