August 1997

Retro this, buddy!
part 2

by Paul D. Lehrman

Are you salivating over all the new "retro" gear that’s flooding the market? Do you thing big knobs, burnished steel, VU meters, and glowing tubes are going to drag your facility out of the doldrums? Do you think Scully tape decks, Minimoogs, Pultec equalizers, and RCA consoles are magical and if you just had some of them in your studio you could make million-selling records? Well, if you read my column last month, you know that I think that for the most part, the movement back to the gear of yesteryear is pure horse hockey.

You see, I was there. And I’m not even so old. But I can tell you that dealing with this stuff was one big pain in the butt. I’ll show you–come with me now back in time, way back, to a classic electronic music studio, before the days of digital synths, heck, before the days of any synths. No programming, no sequencing, no computers, no digital readouts–just knobs and sliders, all of which do just what they’re supposed to do, in real time. And lots of tubes. Analog heaven, right?

Your sound sources are a shelf full of discrete, industrial-strength oscillators. They’ll produce sine, square, or triangle waves, at any pitch you want. Of course, you better not want it for every long, because as the tubes warm up, those babies will drift. You patch them through a gorgeous passive third-octave filter built for some European telephone company, based around moving ferrite cores. The cores are open to the room, and even if you tell people not to smoke, you still often get a nice crackling sound when you move a slider. So you either leave the filter alone during a take, or if you need to change the curve, you keep recording until you get a take in which you did the move right and it didn’t sound like a bowl of Rice Krispies.

Don’t forget that the input impedance on the filter is way too low for the signal from the oscillators (they weren’t made for this!), so you have to manually patch the signal through a couple of pads and transformers, lest you get (at best) a horribly distorted "blaaatch!" or (at worst) fry a core or two.

There’s plenty more: the ring modulators, frequency shifters, broadband equalizers, and envelope generators all need to be calibrated, level-matched (meters? we don’t need no steenking meters!), and dealt with quickly, before they drift off. Want to increase the RT60 on the reverb plate? Just pull the long rod attached to the damper. But you can’t do it while there’s sound going through it, because the horrendous roarrr it produces obliterates everything else.

Now you’ve been here about three hours, patching, tweaking, cajoling, programming, re-tweaking because of the drift, mixing, and monitoring along each stage, and you’ve created five seconds worth of brilliant sound, which you’d better put on tape pronto. If you want to use that sound in more than one place in your piece, you’d best record it a number of times; if you just do it once, and then later make copies, you run the risk of adding hiss and distortion, and if you try to duplicate the sound any other day, even if you’ve kept the best notes in the world, you’ll never be able to get it exactly right.

Are you drooling in envy yet? Me neither. But this studio, at the Columbia-Princeton Electronic Music Center, is where I got my first formal training in electronic music. Looking back on it now, it sounds like about as much fun as building up DNA molecules, one amino acid at a time, would be to a biochemist. As an educational experience, it was priceless, but as a way to actually create something, it sucked. In fact, after a couple of years working like that, I gave up electronic music completely for a solid decade, only coming back to it when systems became available that would allow me to produce more than one composition in my lifetime.

Retro, feh.

Okay, I got that out my system. Now I can admit that there were things I liked about working with unpredictable analog gear, and that some of those things really have been lost in the rush to digitization, computerization, and automation. And there are valuable lessons to be learned from the currently fashionable (if I may steal from Mao) Great Leap Backwards, as long as we don’t take them too literally.

To my mind, the most valuable lesson of the retro movement has to do with the human interface. Digital workstation users miss having all the faders and knobs right in front of them. Fortunately, manufacturers are finally getting hip to the idea that a mouse is not the best way to handle a 64-channel mix with 14 effects inserts, and alternative controllers with real sliders and buttons, some of which even move, are popping up all over the place. Musicians are getting tired of those tiny little parameter windows on their synths, and in response, new instruments are appearing loaded with knobs. In addition, alternative, fluid, controllers like Bob Moog’s Theremin, Don Buchla’s touch and space controllers, and Yamaha’s breath and wind controllers are making a comeback. These will mean that players’ real-time performance gestures are taking on added importance in sequences, which will help make their tracks more organic and convincing. But because they’re MIDI, the gestures are repeatable and editable, and that makes all the difference.

Another lesson can be called the "comfort" factor. For a lot of people, it’s not really the sound of discrete components that matters, it’s simply the fact that they’re there. Workstations, in which the sound is pumped in at one end, spun around as invisible numbers, and spat out at the other, have taken away much of the feel of working with audio: we don’t see tape reels spinning any more, we don’t physically set up the patch bays, we don’t see the processors’ flashing lights responding to the signal, and processor settings are buried in menus within menus. That loss of control, even if it’s illusory, can be unnerving. I’m not sure that putting a picture of a 40-year-old equalizer on the screen is the best solution to this, but certainly speeding up the displays, so that the level meters don’t look like they’re suffering from spastic paralysis; making the signal routings obvious with big, bright lines; and making all active processing controls instantly visible (not just one at a time, Digidesign!) can help. In the long run, however, we’re going to have to compromise, and learn to be as comfortable with virtual patch bays and processors as we are with the "real" things (and don’t forget, virtual patch bays don’t need cleaning!)

The third thing is serendipity–finding something in the behavior of a piece of equipment that the designers didn’t intend, and turning it to artistic advantage. That can be built into digital systems too, although it might not be easy. The Yamaha DX7, you may recall, would go into aliasing under all sorts of circumstances. It could be ugly, but there were plenty of programmers willing to exploit that little weirdness and make some fascinating sounds from it. Perhaps designers could look at ways to "overspec" digital gear, so that instead of hard clipping when you exceeded normal operating parameters, it did something completely different, and potentially useful. (Of course, today ugly is in, so the motivation isn’t there yet to do anything else, but hopefully that shall pass.)

And what about the sound of retro equipment? Can we really reproduce in a musical way the sound of analog tape saturation, of tube non-linearities, and of unstable oscillators? Well, from what I’ve seen in the aisles of trade shows and on my own system, I think we’re at that point where talented engineers can create digital models of just about anything you want, that are good enough to fool just about anybody. And these models can be made flexible enough so that we can call up not just, say, a 12AX7, but any tube or transistor or capacitor or choke ever made, or that ever could be made. With modeling technology getting more sophisticated all the time, I’m sure it won’t be long before we see a respectably convincing "tape saturation simulator" module, too.

Assuming the control interfaces are designed well, and assuming that there’s a way to build serendipity into them, they should let us do amazing things–not just re-create the sounds of yesteryear, but use our knowledge of them, and what makes them pleasing, to create new sounds. And we won’t have to worry about drift, noise, impedance mismatchs, and getting the same sound tomorrow, or next month.

I will leave you with this thought about looking backwards. Once upon a time, there were radio station disk jockeys who were allowed to play anything they wanted to. There were executives and producers at record companies who were encouraged to find new talent, not just sign acts who sounded exactly like the current faves, and develop that talent over time. There were record companies who were judged on the quality of their roster, not on how the stockholders felt about their last quarter’s numbers. There were acts who weren’t tossed out on the street when their first album failed to go Molybdenum. The Beatles were encouraged to get past Love Me Do; Simon and Garfunkel got a second shot at the Sounds of Silence; and Bruce Springsteen was allowed to hone his craft and streamline his lyrics through two muddled albums until he could come up with an absolute masterpiece of a third. As I doze off listening to yet another "adult contemporary" station, I realize that’s the "retro" part of the industry that we really need to revive. Because art takes time, and people need time to find their voices. And if it ain’t in front of the mic or in the fingers, no amount of distortion, no matter how cool and ‘50s/’60s/’70s it might be, is going to save it.

Paul D. Lehrman is a product of the ‘50s, but seems to still be operating within spec.

These materials copyright ©1997 by Paul D. Lehrman and Intertec Publishing