Well, I got the Clarett+, and got some interesting results.
After registering some frustration with cables with this interface too - but having far more ability to move it around to find "quiet" chunks of space to position the interconnects - I decided to go back to the Lynx and try harder to work a suitable configuration with the cables. And, surprise, surprise, the Lynx wins hands down... Not even a contest.
Disregard the measurements from different channels. I didn't pay a lot of attention to which cable goes where, and relatively speaking, the best result is the likely actual performance.
A couple of conclusions:
- The Lynx, on 44.1k/65536 is at -116.5dB THD, which I think is pretty amazing. The only thing topping this would probably be an RME >$1.6k... If even that. I managed to see about -113dB on the Clarett - which is really good by itself and I believe exceeds its spec - but, not consistent (though probably very short cables, such as 1.5ft, may improve consistency and maybe even overall noise) and also, see below.
- The actual deal breaker may be that the Clarett+ is useless over about 45kHz. The Lynx maintains its noise floor all the way its sample rate ability can reach (encroaching into 100kHz).
Maybe all I needed is a trigger to re-examine my cables layout and clearance from EMI broadcasters and other such crap. I probably should still try to shorten the interconnects (3ft long now) with something as short as I can make it.
The funky thing is the worst offender seems to be the desktop computer (where the Lynx is!!), which points to the amazing engineering skill of the Lynx team in designing such an exceptionally performing device.