I was just doing my everyday browse through the web and nearly fell asleep until I found this pretty nice dissertation by two graduates at Detmold University over here in Germany which actually confirmed what I have always been thinking or at least been very suspicious about the industry's hype! Unfortunately, the dissertation is in German, but for those who speak and understand it, here is the link: http://www.hfm-detmold.de/texts/de/hfm/eti/index.html It basically says that there was no big difference between 24/48 and 24/96 compared to the original analog signal. The two 48kHz converters they tested even bet the two 96KHz ones. Besides technical measurement, there was an audition. Most people chose the 48 signal for being the closest to the original. A couple of people even considered the 48 being the original. The two graduates claim that the quality of converters depends on a high SNR and a good analog circuit design. Note that this is just a very short and hence unprecise extract from the graduates' dissertation.