Some technical differences are: Both the Scott Nixon and the Audio Mirror use the TDA1453 DAC chip, but the SN uses one while the AM uses eight in parallel. The 1453 chip definitely isn't state of the art by today's standards. For example, it's a 16 bit chip with at THD of .018 and a S/N ratio of 96, while a state of the art chip like the Burr Brown 1704 is a 24 bit chip with a THD of .003 and a 120 S/N ratio. But when you parallel eight of these you apparently get an averaging effect that reduces the error rate plus and improved S/N ratio. The SN uses a tube output stage while the AM uses OPA602 OP amps. The dAck! dac (should use the Aflac duck in their ads) uses the TDA1545 chip, claiming the 1453 output is "utterly anemic". The duckdack is battery powered too.

The differences between the three wasn't pronounced, but there was a marked difference between the three and the far more expensive modded Benchmark dac. This tends to comfirm my belief that "you don't always get what you paid for, but you never get what you didn't pay for". Someone described the difference, after we heard the Benchmark after the other three as "It's not broken like the other three". A bit of an exaggeration, but it did sound more natural and less fatiguing.

I believe the bottom line is that these very inexpensive non oversampling dacs are a meaningful step up from a mass market player, and if you're looking to upgrade a mid fi player like an older Marantz Cd-67 or such, with a good transport mechanism, any of them would be a practical alternative to buying a new $1,000+ player. But despite the hype their sellers use, they aren't the equal of a $2,000+ oversampling dac or a premium SACD player like the Exemplar modded Denon ($4,000+) or the Sony ($3,000). Not that this should be a surpise to anyone.