[hpsdr] best audio sample size?

[john clark]

I know that we are all hopeful that in the future all of our digital modes will be directly decoded and the need for VAC and all the other sources of conversion errors will be bypassed, but until then I am trying to set my audio chains to the optimum values. I read that WSjt-x and HRD DM-780 prefer the audio samples delivered 48000 samples per second, my question is At what bit size?. 16, 24, 32?  What can PowerSDR VAC1 accept/deliver? 

>From WSJT-x I read   
"ReceivingWSJT-X acquires 16-bit integer samples from the audio input deviceat a 48000 Hz rate and immediately downsamples the stream to 12000 Hz.Spectra from overlapping segments are computed for the waterfalldisplay and saved at intervals of 0.188 s, half the JT9 symbol length."

When I set the VAC channel to 24 bits per sample WSjt-x complains. If I open up VAC I can see that wsjt-x pushes VAC  to 16 pits per sample on the input but I also see 24 pits per sample on the output VAC channel.
What can PowerSDR work with? How can I set the pits per sample on the audio channels?
John
Ti4/N0URE

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20150406/b3b8fb55/attachment-0001.htm>