Bob Moog's name is synonymous with analogue synthesizers, but these days you're more likely to find him experimenting with MIDI theremins and alternative MIDI control systems. His company, Big Briar, manufactures both inexpensive traditional theremins and a far more sophisticated MIDI version housed in a splendid mahogany cabinet. On Bob's business card, his position with Big Briar is described as 'Grand Poobah', which makes a refreshing change from all that President and Chairman stuff! I caught up with Bob Moog at the opening of Turnkey's new in-house synthesizer museum and 'Loopstation' showroom and asked him about the past, present and future of electronic music.
When you set about designing the Minimoog to be both affordable and portable, did you agonise over which modular features you could include and which you'd have to leave off?
"The first trial feasibility prototype consisted of modules -- I think there were two oscillators, a filter, a VCA and one or two envelope generators. It also had a very small keyboard, which was maybe just a couple of octaves. Of course, that wasn't enough, so we began adding things back in, and by then we knew what people liked most -- what they had to have for a basic voice."
Is there anything from the modular system that you really feel it would have been nice to include if you could have at the time?
"It would have been nice to have had patching, of course, linear frequency modulation would have been interesting, perhaps synchronisation of oscillators and a fixed filter bank. When it comes to modular synthesis, there was no idea of how much was enough. The simplest modular systems were just one cabinet systems and were probably about as powerful as a Minimoog, but we always understood that we could add any number of extra modules. The flexibility was just down to how many modules you had."
Do you think synthesis lost anything when it went polyphonic? The controllers that you built into those early machines seemed applicable more to single note lines than they were to chordal work.
"Oh yes, of course. A polyphonic instrument with a keyboard, from the point of view of the player, is more like an organ. A synthesizer to my mind is more of a lead instrument or monophonic instrument where you put all your attention into getting a sound that changes in interesting w
We now have digital synthesizers claiming to replicate the sound of the original analogue synths with absolute tuning accuracy and stability, but do you think that the sound has lost something?
"I think the absolute accuracy and regularity of digital systems detracts from an aspect of the sound that is pleasing and comfortable to our ears. I like analogue sound better, but I'm sure that as time goes on and digital systems get better, you'll be able to simulate those pleasing imperfections and uncertainties. Right now I don't think that anyone really knows sufficient to make an accurate enough model so that an informed listener could not tell the difference. I don't think there's a digital system in existence that would fool anybody in a side by side comparison."
Which imperfections of analogue do you think are the most important in creating that pleasing sound?
"The things that really distinguish analogue instruments from digital instruments are right at the edge of perception -- we have trouble pinning them down and describing them accurately. A lot of it comes down to the fact that analogue instruments are not exactly accurate in the way that digital instruments are. Acoustic instruments are not exactly accurate either, and with an analogue synth, when there are two oscillators slightly apart, the rolling beating that you hear is not perfectly regular -- it has a random or drifting component to it that keeps our ears from getting fatigued."
Does physical modelling interest you as an area worth pursuing, or do you feel it might be a dead end?
"Oh no, I think it's probably the most fruitful digital synthesis area to explore right now. You can go every way from modelling a real physical system like a violin, all the way to modelling something fictitious that's never been heard before. If I were to work with a digital system for my own edification and enjoyment, it would be physical modelling."
Synthesis methods, such as physical modelling, give us more parameters we can control in real time, so what improvements do you think are possible in that area given that a typical keyboard player has got a couple of free feet, maybe the left hand, and perhaps the opportunity to use a breath controller?
"Ah, the optimum control system! I don't think there's such a thing as the optimum control system, but there are a lot of good ones and more will come. That, I think, is the big area for the near future. I believe it is a mistake to think in terms of one parameter for one finger, another parameter for another finger, but somehow you should be able to control many high level parameters that have to do with things that we hear. For example, if we had a sound with 40 overtones, I don't think anybody would want to be able to control the overtones individually, but we may want to be able to control a whole band of overtones at once. With MIDI now, you can build a controller and then use the output of that controller to vary any parameter you want, and I think that type of experimentation could shed some light on what aspects of sound musicians like to control."
As a guitar player, I've experimented with MIDI guitars, which are pretty imperfect things, but the one thing I do like is being able to apply different degrees of vibrato at different speeds to different notes. Is that something that could be translated into keyboard terms, for example by using a key with a position-sensitive surface?
"I've built a few keyboards like that and they were described in the Computer Music Journal maybe six or seven years ago. I began with a regular wooden keyboard and then put a touch sensitive coating on top of each key so that the key was position-sensitive to where you placed your finger in two dimensions. That's very interesting, but it's also expensive to do. There are various ways of approaching that problem in engineering terms, but the real difficulty is getting musicians to study it -- to put behind them the techniques they've learned and to try something new."
Surely this will always be a problem as long as the keyboard is the main player interface for electronic music?
"That's right, and that's why I'm more interested these days in alternative controllers. The multiple touch-sensitive keyboard that I developed was built around a conventional keyboard because th
|"I don't think there's such a thing as the optimum control system...that, I think, is the big area for the near future."|
Aside from theremins, what other forms of non-keyboard controller have you explored?
"One that everybody talks about is what you see in the Beach Boys Good Vibrations video, a ribbon controller that we made back in the mid '60s. I think that sort of thing would be interesting to explore further. Another approach I find interesting is something built by Wendy Carlos, which she calls a Circon. It's really a big pointer that rotates through 180 degrees, and you hold the end of the pointer in front of a scale that shows the notes being played. This gives you some kind of visual feedback, and though it controls pitch in the same way as a ribbon controller, it feels different. Wendy has tried the theremin and the ribbon controller, but she likes the Circon because of its feel, and that's very important."
Essentially though, it's basically a frequency pot with a pointer attached to it.
"Yes, but the size and the shape of the part you put your hand on, exactly how smooth it is, how fast the frequency changes when you move it -- all those things affect how it feels, and that's part of my craft, getting those things as close to right as we can."
Moving onto theremins, they used to be notoriously unstable -- one of the oscillators has to be free to drift so that you can use hand capacitance to change its frequency. How have you improved the stability -- the only way I can think of to make a MIDI theremin would be to do a quick autotune routine every time a MIDI Note Off occurred?
"Well, we could do something like that, but in fact we don't, though it would be useful for applications such as having a theremin set up in a museum where you can't have somebody tuning it up all day. Our instruments are fairly stable, but they're not 100% stable, they couldn't be, because the temperature and the moisture content of the air affect the tuning. In practice it doesn't matter too much because you quickly feel where the notes are and compensate for any changes in tuning -- you basically play the instrument by ear. You don't think in terms of putting your hand in exactly the same place to get a given note."
I guess that if you feed the output of a MIDI theremin into a sequencer, you're going to see oceans of pitch bend data and not much in the way of note information.
"That's right, there's almost no note information, though there are a couple of different MIDI modes you can choose to use. In one there is a Note On message when you begin a sound, then when you go back to silence, there's a Note Off. Everything in between is pitch bend, and that has a range of plus or minus four octaves."
So editing a performance could be tricky!
"Yes, and we're interested in working with sequencer manufacturers to get some good graphic editors, because there are some very interesting things you can do with graphic editing if you display the pitch bend contour as a function of time, and just move it around a bit. You can, in principle, correct notes, change the amount and speed of vibrato and things like t
|"A synthesizer to my mind is more of a lead instrument... where you put all your attention into getting a sound that changes in interesting ways."|
The theremin is an intriguing instrument, but is the old capacitance-controlled oscillator technology the best way to do it? For example, wouldn't it be more stable if you wore a small transponder on your finger, then used a radio system to continuously calculate the distance of the transponder from the sensor? Couldn't you use one of the passive transponders that are used in anti-shoplifting tags?
"The advantage of analogue theremin technology is that it is absolutely continuous, whereas having something like a supersonic transducer would provide the data in steps, which would take time to compute."
Unless you checked it at a high enough rate that it was effectively continuous. After all, what comes out over MIDI is essentially a series of steps.
"Yeah, we update the pitch information 200 times a second, though you can select a lower update rate if you want to. It turns out not to be a problem that this is an analogue device, and as an engineer, I'm not sure that any other transducer type that I know of would have all the musical advantages of the traditional theremin."
What do you think about some of the optical MIDI control systems that have been developed?
"I think they hold promise as controllers, but they're different from the theremin. With the theremin, you can get up to a distance of less than a centimetre and still be playing. This is ideal for the playing range of the arm, around 60 cm, whereas optical controllers work over a different range. Buchla's Lightning [infra-red light-based alternative MIDI controller] is interesting in that it is two-dimensional, so you can move your hand from side to side and control another parameter. They're two different types of controllers that are appropriate for two different types of gestures."
As Bob had brought a theremin with him, I had to give it a try! Because there's no physical feedback other than what you hear, it takes some getting used to, but even so, I could appreciate the appeal of the instrument. I asked Bob if he had considered extending the control capabilities into three dimensions.
I guess that if you were to develop an effective three-dimensional scanning system covering the space around the antennas, then you could have control over pitch and two other parameters from one movement.
"Yes that's right, whereas the theremin has two antennas, one for pitch and the other to generate a control signal, usually for amplitude. There is another MIDI mode where we can assign the antenna outputs to any MIDI controllers at all -- pitch bend, filter, modulation or even for turning lights on and off. There are all sorts of theatrical applications. There's also a footswitch and pedal input for further control."
Obviously you're always thinking of new ways to provide more performance control. Is there anything you can talk about without giving away too many secrets?
"Well, I can tell you what we've done so far, but there are also things I don't want to talk about, not because they're secret, but because they're less than vapourware and may not turn out to be good ideas. However, there are four MIDI modes on this instrument, one of which is a conventional theremin where the right hand antenna sends out pitch bend and the left hand antenna sends out one of a small selection of control change messages. The second is a more general continuous controller, where either antenna can be assigned to any MIDI controller. Third is a chromatic mode, where we use Note On messages to quantise the space around the antennas. There's a selection of 16 different scales you can play like air harp or air guitar. That's very different from playing with continuous pitch change. The fourth mode is called Smart Theremin mode, where we use a combination of Note On and Pitch Bend to add pitch bend to whatever MIDI note you happen to be playing. In normal theremin mode, the note is always middle C plus pitch bend. This last mode is appropriate for playing things like sample playback, because samples don't usually work well when they're being bent over four octaves, they're best used over a relatively narrow range."
Does your vision extend much beyond theremins at the moment -- is there something else you'd particularly like to explore?
"Well, I love thinking of new things, but I'm running a small business now, so we need to think of having new products that people will buy. We've been very gratified at the acceptance of our theremins, but that market is very small, so now we're looking at introducing some new products going back to analogue synthesizer-like things. There's a high awareness of the musical value of analogue instruments amongst musicians these days, so I expect that in around six months time we'll be introducing a set of effects that are organised more like analogue synth modules than the usual pedals. There'll be things
|"I think the absolute accuracy and regularity of digital systems detracts from an aspect of the sound that is pleasing and comfortable to our ears."|
Will that be a replica of the original design or will it be enhanced in some way?
"It'll have all the basic sound quality and function of the original model, but with a few additional things like synchronisation and different ways of controlling the filter. We'll use 1990s components instead of 1960s components -- you can't get 1960s components any more, and even if you could, I'm not sure we'd want to use them! I'm an engineer, and an engineer will build a design to meet a specification. The specification I've set myself is that it should sound like the original Minimoog, but it should also be stable and reliable."
Is that going to be a difficult task when, as you said earlier, some of the things that make the original sound distinctive are actually quite subtle? Is there a danger that you might 'throw the baby out with the bath water'?
"Between my associates and myself, we're pretty confident that if we design something that doesn't sound like a Minimoog, we'll know it. We won't sell anything that doesn't sound like a Minimoog. As an engineer I'm confident that if we spot something that isn't right with the design, I'll know how to fix it. At worst, we could go back to doing things as they were in the old instrument, but I don't think it will come to that. We'll be keeping the original ladder filter circuit, but using modern transistors that should be quieter and more reliable."
"Beyond that, I'm interested in all types of alternate controllers. Wendy [Carlos] and I have talked about us manufacturing a Circon-like device, and we certainly know how to build ribbon controllers. I can foresee a ribbon controller with MIDI output."
Would this be a new kind of instrument, like a device with ribbons replacing strings?
"I was once asked what I thought the future held in store for electronic music, and I predicted that musicians who played electronic instruments would own their own finely crafted controllers which would be very sophisticated and quite expensive, and for a performance, they'd plug this into a generic sound generator, the same way musicians today plug into a generic PA system. And I'm amazed today at how fast we're approaching that. I also predicted the advent of MIDI, and I missed it by only one year. This was back in 1976. Now you can plug a sound card into your computer and emulate fairly closely, not exactly, but fairly closely, just about anything. You get all that stuff on a card now, so if you look 10 or 20 years ahead, I think it's safe to say that sound production is going to be generic. With just a little bit of money, you'll be able to make any sound you want, and the real expense will be in the device that you use to play and control those sounds."
Presumably, different players will demand different things from their controllers, so does that mean the controllers themselves will end up being modular?
"You can already customise guitars or have a custom guitar made, so why not MIDI controllers? You're still a human being, you're still using your hands, and the same sort of things are going to continue to make you feel good."