Music preservedThe first kind of audio recording method was analogue recording. Every step in the process involved either vibrations or something analogous to vibrations. An orchestra plays. Sound waves strike the diaphragms of microphones. The microphones convert sound vibrations to their electrical analogue. The electrical signal magnifies a recording head, which arranges the pattern of metal oxide in yet another representation of the original sound waves.
The tape might then be used to master a phonograph record, in which the sound wave analogue is a pattern of slight irregularities in a groove of vinyl. And finally, either the tape magnetizes a playback head and produces an electrical signal, or the stylus of a phonograph vibrates according to the record groove and produces an electrical signal. That signal is amplified, and the strengthened signal causes a speaker to vibrate and reproduce, with greater or lesser fidelity, the original sound.
At each stage of an audio recording, there is something of the original sound wave intact. The electrical signal resembles the sound wave. The magnetic pattern resembles the electrical signal. The grooves in the record resemble the magnetic pattern. The electrical signal resembles the vibration of the grooves. The vibration of the speaker resembles the electrical signal. And finally, the sound waves produced by the speaker resemble the vibration—and every previous form of the wave back to the original. The trick in analogue audio recording is always to capture as much of the original wave as possible, losing as little as possible and adding as little as possible (distortion) in the process of recording and reproducing the sound.
It's remarkable what this technological descendent of Thomas Edison's invention—speaking into a cone, which caused a needle to vibrate and trace a pattern in a wax cylinder—can produce. The best analogue audio recording and reproduction equipment results in a sound that is remarkably faithful to the original, including its dynamic range, its frequency range, and its harmonics and overtones.
Of course, most of us can't afford the finest reproduction equipment—things like a super-quiet turntable, wooden tonearm, multiradial stylus, low-distortion amplifier, and speakers accurate from 20 to 20,000 Hertz.
Music encodedDigital recording does not produce analogues of the original sound wave. Instead, it encodes sound information into a stream of ones and zeros. What is remarkable is that this works at all and that a reasonable facsimile of what the New Pornographers sounded like in the studio comes out of our speakers or headphones—without any signal loss, hiss, hum, clicks, or other byproducts of analogue recording and reproduction. What is far less remarkable is the reality of how well this encoding actually works. Yes, you can hear the squeak in Ringo's bass drum pedal. But are you truly hearing what the Vienna Symphony Orchestra sounded like in the concert hall?
Sound information is incredibly complex. Take just one instrument, such as a cello. The sound it makes can be loud or soft or in between. It has a wide frequency range. And the frequency range includes not only the actual notes played but the harmonics and overtones that are also generated by bow and fingers on strings amplified by a wooden casing. There are also more subtleties, such as how each note starts and ends and whether a note sounds rich or thin.
In digital recording, all of this information has to be encoded in such a way that it can be reproduced faithfully. And it works, to an extent. For most of us, a compact disc sounds wonderfully clear and realistic. To an audiophile, CDs have a harshness that is not present in analogue recording, more apparent in orchestral music than in the rock and hip-hop and other popular music that many of us listen to. (I remember an all-digital recording of the solo piano version of Pictures at an Exhibition by Modest Mussorgsky that even to these non-audiophile ears was unlistenable.) Digital is an improvement in many ways over analogue, but in many other ways there is a loss of quality that is not dissimilar to the difference between reading from a Kindle and reading a book. It's a difference in quality more than in quantity.
The loss of quality is exacerbated in a file format such as MP3 versus CD. Encoding all that complexity takes one heck of a lot of ones and zeros. MP3s are usually encoded at lower sampling rate than is used for CDs, so that the files don't become even larger than they are. That means an MP3 lacks both dynamic and frequency range. It simply doesn't have enough information to do a great job of reproduction. We sacrifice quality for convenience, and most of us don't notice the difference.
The warmth of imperfectionAnalogue recording isn't simple, but there is an elegance to it. Waves becomes waves become waves and finally turn back into waves very much like the originals. Wax cylinders are still playable using very simple equipment. And in the words of a story I once heard, vinyl is final. It warps, is too easily scratched, and even breaks, but in many ways it is incredibly durable. You can still play great-grandpa's 78s. We now know that CDs aren't nearly as immutable as was once thought. When analogue media deteriorate, you can still make out a semblance of the original sound. The same is not true for encoded bits. They work or they don't.
Despite what I know, I have far more CDs than records. I have a turntable that is not set up. I listen to MP3s more often than anything else. Yet I understand the appeal of records and tapes. There was no mystery to it. You can understand how sound vibrations can be turned into something similar and then turned back into sound. I have no clue how music is digitally encoded.
I have some MP3s that I ripped from an old audio cassette that was recorded from a borrowed copy of Unknown Pleasures by Joy Division. Each pop is familiar. There is something oddly comforting about this imperfection.