Upsampling and Stereophile


Last summer there were several threads here on Audiogon about digital upsampling and over sampling (a couple long and heated), but the opinion(s) of the audio press were conspicuously absent. In the Dec. 2000 issue of Stereophile, page 3, John Atkinson, has an editorial explaining their position on this in his "As We See It" column, page 3. I encourage everyone interested in the subject to look up the article. I have excerpted the following quotes that I think sums up Stereophile's opinion, and that may pique your interst: (1) "....the audio industry has settled on an 8X-oversampling ratio, the 44.1 kHZ CD data being converted to a 352.8 kHZ datastream before D/A conversion." (2) "no matter how good these upsampling products can sound-- and the dCS, Bel Canto, and MSB products indeed sound excellent-- there is no conceptual difference between them and traditional CD playback systems. I am now convinced that the sonic differences we have heard and reported on are merely due to the different choices in digital filters made by the designers of these products." (3) "In the meantime don't buy a digital product because it has "24/96" emblazoned on its front panel. Buy it because it makes your CDs sound great". Cheers. Craig.
garfish
Garfish: Add the Marantz DV18 to the list of DVD players that output a 24/96 digital signal--I use that one for that purpose. I think their less expensive ones in that line also can output a 24/96 digital signal too, but I'm not sure. You have to use a video monitor to program the digital out that way.
My integrated CD/DVD payer (Muse 9) definitely passes a 24/96 signal through the decoder. The sound of 24/96 discs is better than all except the most astounding regular CDs. Upsampling on the other hand does not have theoretical advantages. I recommend the white paper in the Muse site or the recent Stereophile article. In any event, the quality of the analog stage is more important than the above considerations. I believe a good CD player with good parts where they count, will in most cases sound better even with regular CDs, than a cheap DVD player at 24/96. Likewise, the sound of an upsampling DAC or CD player is more likely to be affected by its analog stage than by the upsampling itself.
Joe: I certainly agree with you about the quality CD player vs. the cheap DVD player, as the power supplies and analog stages have an awful lot to do with sound quality. With respect to upsampling, I'm not sure, and actually doubt, if it is upsampling alone that makes the Purcell improve the sound of an already excellent transport/dac combo, but the unit certainly makes an improvement to my ears in my system. I have also heard the MSB unit with and without the upsampling, in an A/B test (you can switch it on the back), and felt there was a slight improvement with the upsampling engaged. Could it be that the chip used for the upsampling module in the MSB and Bel Canto units has an effect apart from its upsampling claims (much like the Crystal chip used in HDCD players was thought to provide at least some of the improvements apart from the HDCD encoding)?
Sol322; Thanks for the references re: DVD players that output a 24/96 digital signal. I'll check them out. Craig.
Rcprince, thank you, interesting thoughts. My own sense is that the analog domain (in a digital player), and the recording and production have a much bigger effect in sound quality than the word length and the sampling rate (once you are above redbook CD). From listening to 24/96 material in my Muse, I do get a sense that a more quiet background can be achieved (i.e. dynamic range) than in CDs, but the differences are smaller between a good regular CD and a 24/96 DAD than between a good CD and a badly recorded/produced CD. My opinion on vynil, by the way, is that unless you spend $30K or more you will be hearing the groove, the wow and the flutter, etc, and that it is definitiely not worth it.