I would expect the issue with computer activity to be driver-related (?)
I too would like to see a 16-bit trial at a couple of sample rates. I wonder if the SRC is messing it up... Could also be how they are handling the 24-to-16-bit conversion but unless something is really hosed that shouldn't corrupt the bits that are there. And yah seems like there is no buffering; with this DAC that would have to be added externally (I think some DAC chips include a FIFO on-chip but have not looked at audio DACs much). And maybe there's simply no clock buffering/isolation/resynch so you get GIGO.
Interesting that your previous test of their $400 DAC also had issues. Strange, 2 kHz (small k, capital H, I'll nag you 'til you get it right ) and 19 kHz or so are rather odd spur frequencies for an 11 kHz tone sampled at 44.1 kS/s. Almost like harmonics around 40 kS/s instead of 44.1 kS/s. I've put in 90+ hours working this week so maybe brain-fade is causing me to miss something obvious...
I'd be really surprised if they were interleaving the two channels to create a balanced output; for balanced output the usual solution is just one more op-amp or inverting buffer. Often just one quad op-amp to do it all (inverting and non-inverting outputs for each channel). Much easier than ping-ponging DACs.
I too would like to see a 16-bit trial at a couple of sample rates. I wonder if the SRC is messing it up... Could also be how they are handling the 24-to-16-bit conversion but unless something is really hosed that shouldn't corrupt the bits that are there. And yah seems like there is no buffering; with this DAC that would have to be added externally (I think some DAC chips include a FIFO on-chip but have not looked at audio DACs much). And maybe there's simply no clock buffering/isolation/resynch so you get GIGO.
Interesting that your previous test of their $400 DAC also had issues. Strange, 2 kHz (small k, capital H, I'll nag you 'til you get it right ) and 19 kHz or so are rather odd spur frequencies for an 11 kHz tone sampled at 44.1 kS/s. Almost like harmonics around 40 kS/s instead of 44.1 kS/s. I've put in 90+ hours working this week so maybe brain-fade is causing me to miss something obvious...
I'd be really surprised if they were interleaving the two channels to create a balanced output; for balanced output the usual solution is just one more op-amp or inverting buffer. Often just one quad op-amp to do it all (inverting and non-inverting outputs for each channel). Much easier than ping-ponging DACs.