I try to generate low frequency sine with a PXI 5422 (AWG min samplerate 5MHz) using the standard sine mode.
From the KB I knew that the buffer must be increased. Problem: I can't get lower than 180mHz (100mHz was the target).
My card has 256MS memory, available is still > 126M but when I try to increase the standard buffer to values >28M I get an Error.
Another 'funny' error come up (see 2nd Pic) when setting just the right buffer size. Seems that the driver try to set an invalid samplerate (rounding problem??)
Testprogram attached. LV2009
The first error has to do with running out of memory on the computer, not the board. How much RAM do you have on the computer? What OS are you using?
The second does look like a floating point rounding error. The NI-FGEN team should look into it. Perhaps you can attach a VI that reproduces it? That'd be very useful.
In the meantime, you can work around the issue by using Arbitrary Waveform mode. Sorry for the inconvenience.
Thank you for the hint with the memory (System is XP with 2GB I think). So the driver has no problems allocating the memory on the card, but can't write the waveform at once due to small OS memory??
That explain why this error show up sometimes at different values. Is there a way to pre allocate and reuse that memory at the beginning of the program (call of the driver?).I'm quite used to handle big data arrays and limited memory . Looks like I have to go the AWG way (is there an example for loading the AWG in chunks?). There is no cheat to run the sample clock (intern or extern) with less that 5MHz? What will happen if the external sample clock drops below the 5MHz? (Ups, I'm getting old and serious, I'm asking before trying 😄 )
The vi attached was used to produce the error . For the input values see picture. CAR? ( The workaround is a bigger buffer than needed I suppose)
Found the FGEN Incremental Writes example 🙂