Took a quick second listen to the nine original files--with fresh ears on a slightly different setup. I ended up with a very different reaction and set of preferences. I can see why the original testers did so poorly on the 18 A/B/X tests. With so many variables, the more I listened, the more indiscernible and mushy it all became. "Psychoacoustics," as Arif Mardin would say.
With just the three files, it's somewhat easier to discern differences and retain those distinctions, and I'm having pretty good success blindly identifying each one. I'll guess that #12 is the analog, #10 is RADAR, #11 is Pro Tools SYNC. Directly comparing each of these three to the original nine...I think #12 is actually #6, #10 is probably #2, and #11 is #8. A far cry from my initial perceptions.
This was far from a definitive test, given that it is a rough mix generated from a dubious console, printed through yet another set of A/Ds. But it hasn't been a total waste of time, I don't think. The results are inconclusive, except to say there is no evidence to support the claim that the 192 categorically loses low end. This should be considered good news. There are refined differences between the different converters when paired with different clocks, and preference of one over the other is likely to change with every situation, programme, converter and environment. Any one of them could be made to sound "good", it's probably not a make or break decision, at least not in this instance. At this point no result would truly surprise me. I'm sold for now and ready to move on.
Done here.
-Steve