Digital Domain (April 1984)

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting





A BINARY BEGINNING


Frankly, this could turn out to be an unbelievably horrible mistake for me. You have no idea how much trepidation I have undergone since I received the invitation to write a new column for Audio. When I accepted, it was on the thin ice of impulse the question of ego is an enormous one for all writers. Tonight, as I stare into my microcomputer's green screen, I am convinced that the ice has broken, and I have fallen through, into very cold water indeed.

Don't misunderstand me. This should be a piece of cake. My education is solidly electrical engineering, I've spent years designing digital circuits, and I'm a recording engineer and a Music Engineering professor.

I've written articles on digital audio, and I'm even secretly working on a digital audio textbook. Qualifications, seemingly, no problem. Yet, I'm in a real jam, because of you, the readers.

You see, I can't relate to your interest: I have never cared for recorded music.

Yes, that's right. I feel better, having gotten that off my chest. Although this first in a series of articles might be the last, at least I can say I was honest.

Let me explain. My office adjoins a concert hall. My office, or lab, has al ways adjoined a concert hall. All day or all night long, it's music. Specifically, the creation of music. I am immersed in it, drowned in it.... As a result, I don't have much enthusiasm for re corded music. Frankly, it all sounds bad to my ears. Can you blame me? Is there really any comparison between the experience of live music and the pressure wave resulting from the LP on your turntable? No, I, at least, was never remotely seduced by hi-fi. After a day in the concert hall, I could never bring myself to listen to a stereo sys tem and try to believe there wasn't a tragedy being perpetrated.

Don't misunderstand. I do a lot of listening at home. My trusty AR turntable has tracked a million miles of vinyl groove, my faithful ReVox A-77 has had untold head transplants; I listen all the time-critically, appreciatively, thankfully. But I was never fooled, not one bit. Those were only recordings, generations and generations removed from their live birthright. There was not even much solace for me in the recording studio: The orchestra on the other side of the glass was only a cruel re minder of imperfect reproduction. In my chair behind the console, I could see that none of those controls would help; at the sound of the final chords of a symphony, the sound off the tape would never cause me to jump from my seat and applaud, spontaneously, instinctively. They were only recordings, in an analog medium which was intrinsically disappointing.

Analog was always a mistake, be cause medium limitations were inherently indistinguishable from signal.

Whether magnetic tape and its flux densities or a vinyl groove and its mechanical variations, the medium's method of storing information fused the analog data to the analog medium's noise. Even in the best analog systems, the noise and distortion accompanying the signal were audible-and that spells failure for an audio information-storage technique. If it sounds bad, the stored information is wrong.

Thus, for information storage, analog technology was doomed. I think every design engineer can see the reasoning behind that statement. Design is essentially a question of technology, but with music it is ultimately a question of hearing. The limitations of analog storage are simply too audible. In the past I could never satisfactorily listen to recordings, simply because they were analog.

Then there was digital. With digital, the nature of storage medium and its content have been divorced. Thanks to the digital computer, we can store in formation as pure data untainted by the fact of its storage. Binary data can be recorded as simple flux reversals on magnetic tape or dimples on a reflective disc, and the noise and distortion from the storage medium--and degradations like the rotational variations of the machinery--will not affect the quality of the data. Suddenly we are afforded the opportunity to record music with much greater accuracy, and thus greater fidelity. Of course, limitations still exist; specifically, the method of conversion from the analog acoustic waveform to the digital storage medium, and back again, presents a formidable engineering challenge. But our advantage is irrefutable--our digital data exist independently of the medium and enjoy all of the efficient methods of computation available only to a digital processor.

Moreover, with digital storage our information is irrevocably permanent. Fu ture systems of still higher fidelity will be devised, but our digital data will never be eclipsed-perhaps future players will merely interpolate that data to provide new information from its content, for with digital such things are possible. Analog music recording was time spent in waiting.

Now the wait is over. That is why I can at last write this column: Because digital music exists, because at last I can listen to music playback and hear so much information that I begin to feel as if I am there-right there-at .the performance. High fidelity will have to be redefined as higher fidelity.

It is a new beginning for the art and science of audio. Just as the wax cylinder gave way to the 78-rpm record, and the 78 rpm gave way to the LP, the LP will give way to the CD. Even if I'm wrong. even if the CD fails, digital will not. It's the first step in a new evolution.

Sure, mistakes will be made, and, sure, a complete rethinking of recording technique is required. Our sweeping conversion to digital audio will necessitate this. But the underlying impetus is clear: Because of digital technology, the amount of information is so much more vast and our methods of analyzing and processing it so much more efficient, and because the power of the available music information is much greater, we have begun a new era in the re-creation of acoustic events.

Today's equipment is just the beginning. Eventually the CD will be an antique. That's right-it's not too early to say "CD" and "antique" in the same breath. Sure, the CD has tremendous firepower--a channel bit rate of over 4 million bits per second, and its silvery face might pack 13 billion bits of information-but engineering teams are al ready working to perfect erasable and recordable CDs. Perhaps in the future, consumer storage formats will disappear entirely, to be supplanted by a commercial library of digital recordings residing in memory-accessed and auditioned at home through your personal computer. Imagine: Any re cording ever made, available any time, with a few keystrokes (for a small monthly charge, of course). But even as we attempt to comprehend that, we must try to imagine successive methods of music storage, methods which may someday render digital storage obsolete. Well, maybe I've said too much. Perhaps I've insulted the readers of this magazine who have actively supported the slow evolution of analog audio technology and thus implicitly pre pared the way for digital audio. Per haps I've confused people with some of my highly speculative arguments.

Well, in my opinion, provocation and confusion are the early stages of understanding. And our goal now is to achieve an understanding of digital audio, in terms both of bits and bytes and of the philosophical nature of this important technological development. It's an exciting time. I think we're lucky to share in the makings of a revolution.

Also see: Philips Oversampling System for Compact Disc Decoding (April 1984)

(adapted from Audio magazine, Apr. 1984; KEN POHLMANN )

= = = =

Prev. | Next

Top of Page    Home

Updated: Monday, 2018-01-08 9:32 PST