Posted by mike demler on November 5, 2007
It may seem odd for me to defend something digital here, but I just hate to see myths perpetuated. I also enjoy debunking them 🙂
I am an audiophile and an analog guy, but as an engineer I often think back to a quote attributed to Merrill Brooksby, one of my former managers at GE R&D Labs: “there are certain facts that are true”!
I ran across an article on CMP’s Audio Design Line titled “Audio myth: Vinyl better than CD?”. The article highlighted a little blog storm that apparently got kicked up over at Wired Magazine recently with the provocative post “Vinyl May Be Final Nail in CD’s Coffin”, where it was stated that a “reason for vinyl’s sonic superiority is that no matter how high a sampling rate is, it can never contain all of the data present in an analog groove, Nyquist’s theorem to the contrary”. The Wired blogger lists his qualifications on the subject as playing bass and riding a bicycle (I’m not sure if that means he does both at the same time), but regardless – he is mistaken in perpetuating this myth. The Nyquist theorem is a proven fact, and it is very true. As an audiophile I take no issue with anyone’s subjective preference, but having a preference doesn’t make the science wrong.
It’s pretty funny to me that you never see any of the vinyl lovers talking about the RIAA equalization curve that is a required part of the LP recording and playback process. The RIAA curve is very non-linear; describing specific frequencies at which gain or attenuation is applied to compensate for limitations in the vinyl medium. Any mismatch between recording and playback, which is inevitable, will result in distortion. Throw in the inherently non-flat frequency response of the phono cartridge, and the result is a not so high-fidelity medium. You would think that these analog audio “purists” would be calling for an elimination of the RIAA pre-emphasis and de-emphasis so they can get to the “real” sound! But the facts are that the RIAA curve requires playback attenuation of -20dB at 20 KHz, to cancel out a boost of +20dB that is applied during recording to overcome groove hiss. Low frequencies have the opposite problem, since they are de-emphasized during recording to avoid adjacent groove interference.
The goal for LPs, as it is for CDs, is flat 20Hz-to-20KHz reproduction. By applying Nyquist sampling and high-resolution A/D and D/A converters, CDs are capable of overcoming the physical limitations of analog vinyl recording. But I expect this myth to live on, because there’s just no convincing the vinyl lovers that less fidelity is not more.