Tonight I spent 4 solid hours listening to low generation mastertape copies, on a state of the art music playback system at the Newport Beach T.H.E. show. Never have I been more convinced that analog tapes like these (15ips, 2 track, current state of the art mods to Tascam deck by United Home Audio) have more musical information than any kind of digital recordings, and especially DSD recordings. They sounded wonderful. The playback system other than the tape deck was all MBL's best, including MBL's best omnidirectional speakers, and floating bias Class A/AB amplifiers. I would say it was generally the best reproduced sound I have ever heard. A solid group of dedicated audiophiles like me listened to most or all of the presentation.
I continue to believe that we need more than 16 bits at 48kHz as some objectophiles argue would follow from the Shannon-Hawkins corollary to Shannon's Information Theory, and tonight's listening was further personal confirmation. I was not the only true believer...all the chairs were filled for the 4 hours I was there.
I have never heard any form of digital recording sound this good. I think that either Shannon-Hawkins is wrong, misapplied to human musical perception, or misinterpreted.
Today I also heard the Audio Note exhibit which features a pure PCM DAC, with no oversampling or upsampling or filterning. It sounded good generally, though I like my home system (using Onkyo RDV-1 as DAC) better. Although this demonstration of an Audio Note system was far better than the one I heard in 2009 (which I could barely stand due to harshness and distortion) and I thought their current DAC was more revealing in some ways like mine than currently typical sigma delta DACs, there still was a whiff of harshness. I think my DAC, which uses PCM 1704 DAC and 8x oversampling, is better--it has no harshness at all.
I think that PCM dacs sound better than sigma delta DAC, including all forms of DSD. But it is not necessary to remove overampling reconstruction filters (used in the very first PCM products, including the first CD player the SCD-1). In fact, digital reconstruction filters are perfectly fine and likely better than analog equivalents (though we might be better off with higher rate digital--at least 48kHz--so as to make the reconstruction filtering easier). The fly in the ointment, I believe, is with the sigma delta DACs, and the theory which says they are equivalent to PCM dacs. Is that Shannon-Hawkins again? So far, they don't sound equivalent to me. Sigma Delta dacs have a characteristic smoothness which seems to be information lossy. DSD has this same property, but adds just enough noise to make the resulting presentation about equivalent in brightness to pure PCM, but with a slightly fake quality.
Now, maybe an all MBL system like the one I heard tonight would sound equally good with high rez digital like 24/96? I don't know, but the last times I heard sound equally good--was at previous mastertape demonstrations at the 2009 CES. That guy had first generation safety masters which he himself had recorded of musicians like Frank Sinatra. Digital, in all the various forms I have heard it, doesn't reach that level, but high bit rate PCM comes closer than DSD or sigma delta digital.
Another odd thing...I thought the analog tape recording of Kind of Blue sounded much more like the LP record than the remasterings done by Sony after they developed DSD. CD masterings in the DSD era have sounded much brighter and slightly harsh. The analog tape sounded neutral but not dark, only slightly brighter than the 1985 CD version but way more transparent than any version I've ever heard. Stuff which doesn't seem to make sense in other recordings sounded real on tape.
I continue to believe that we need more than 16 bits at 48kHz as some objectophiles argue would follow from the Shannon-Hawkins corollary to Shannon's Information Theory, and tonight's listening was further personal confirmation. I was not the only true believer...all the chairs were filled for the 4 hours I was there.
I have never heard any form of digital recording sound this good. I think that either Shannon-Hawkins is wrong, misapplied to human musical perception, or misinterpreted.
Today I also heard the Audio Note exhibit which features a pure PCM DAC, with no oversampling or upsampling or filterning. It sounded good generally, though I like my home system (using Onkyo RDV-1 as DAC) better. Although this demonstration of an Audio Note system was far better than the one I heard in 2009 (which I could barely stand due to harshness and distortion) and I thought their current DAC was more revealing in some ways like mine than currently typical sigma delta DACs, there still was a whiff of harshness. I think my DAC, which uses PCM 1704 DAC and 8x oversampling, is better--it has no harshness at all.
I think that PCM dacs sound better than sigma delta DAC, including all forms of DSD. But it is not necessary to remove overampling reconstruction filters (used in the very first PCM products, including the first CD player the SCD-1). In fact, digital reconstruction filters are perfectly fine and likely better than analog equivalents (though we might be better off with higher rate digital--at least 48kHz--so as to make the reconstruction filtering easier). The fly in the ointment, I believe, is with the sigma delta DACs, and the theory which says they are equivalent to PCM dacs. Is that Shannon-Hawkins again? So far, they don't sound equivalent to me. Sigma Delta dacs have a characteristic smoothness which seems to be information lossy. DSD has this same property, but adds just enough noise to make the resulting presentation about equivalent in brightness to pure PCM, but with a slightly fake quality.
Now, maybe an all MBL system like the one I heard tonight would sound equally good with high rez digital like 24/96? I don't know, but the last times I heard sound equally good--was at previous mastertape demonstrations at the 2009 CES. That guy had first generation safety masters which he himself had recorded of musicians like Frank Sinatra. Digital, in all the various forms I have heard it, doesn't reach that level, but high bit rate PCM comes closer than DSD or sigma delta digital.
Another odd thing...I thought the analog tape recording of Kind of Blue sounded much more like the LP record than the remasterings done by Sony after they developed DSD. CD masterings in the DSD era have sounded much brighter and slightly harsh. The analog tape sounded neutral but not dark, only slightly brighter than the 1985 CD version but way more transparent than any version I've ever heard. Stuff which doesn't seem to make sense in other recordings sounded real on tape.
No comments:
Post a Comment