Interview: David Cockerell

Left: David Cockerell in c1975 with Tim Orr (EMS Archive, used with permission)
Right: David Cockerell at home in 2016 ( James Gardner, used with permission)

David Cockerell is an electronics engineer and designer. He started working at Peter Zinovieff’s electronic music studio in Putney in 1966 and designed classic EMS synthesizers such as the Synthi VCS3, Synthi AKS and Synthi 100, and one of the first digital sequencers. From 1974 he worked for Electro-Harmonix in New York, designing their Small Stone phaser and Electric Mistress flanger among many other effects pedals. He also worked briefly at IRCAM in Paris and for Akai, where he was involved in the design of many of their samplers including the S612, S900 and S1000. At present he is working for Electro-Harmonix once more, and continues to design ground-breaking pedals such as the POG and the Crying Tone wah.

 

David Cockerell interviewed by James Gardner, 20 May 2010. Corrected and revised by James Gardner and David Cockerell, March 2012.

James Gardner: What are you working on at the moment?

David Cockerell: At moment I’m working for Electro-Harmonix again, making gadgets for guitar players.

How does it feel to be back there?

Well, guitar players...it’s a rather limited set of all musicians. But they have lots of interesting problems.

And you like to solve them.

Innovation: that’s the most interesting part of it all. Trying to make things that are new. And there are so many other engineers working on it too and it’s quite hard.

Is that because on the one hand there’s a demand for replicas of old pedals but also that people expect a lot from a new pedal?

It’s certainly true that there are a lot of vintage old things, and people have this golden view of the past in which nothing sounds like it used to. I’ve been breaking my head trying to reproduce some of the old sounds. One’s on a hiding to nothing when doing that, so I really prefer to do new things.

That’s been a thread all through your engineering career, hasn’t it?—trying to push the boundaries of what’s possible and come up with new things.

Well, actually the technology isn’t all that advanced. This is a small corner of the engineering universe, where one just picks up the crumbs that fall from the table of more advanced branches of engineering. Nowadays one can’t compete with huge markets in telephones and computers. You just have to make do with the little byproducts that are overlooked elsewhere.

How did you get to work with Peter Zinovieff?

I was a medical technician then, still studying for my engineering qualification, and a school friend of mine, Mark Dowson, was doing work for Peter. He found Peter too demanding and so he shoved him off on to me.

What sort of things was Peter asking for in those days?

Peter had built up his own studio using his own rather amateurish engineering skills. It had all sorts of uniselectors and bits and pieces from surplus stores and relays and that sort of thing. He really needed someone with a bit more technical competence—well, more patience—than he had, to make the sounds that he had in his head.

One of his goals was to eliminate tape splicing, wasn’t it?

Yes, yes. He wanted to automate things and have sequencing machines. That was simply not possible with Post Office relays. They just didn’t work well enough.

Were you aware of the kind of musical requirements that he had or did you view them from a purely engineering point of view?

I was never very sympathetic to modern intellectual music, but I always appreciated really musical people and he is certainly one such.

At the time, then, he was experimenting with electromechanical sequencing. What were the first things you built for him that took him closer to his goal?

Voltage-controlled oscillators. On most oscillators before then you’d had to turn a knob to change the pitch but Peter found some rather good American ones where the pitch could be changed by a voltage, which meant the pitch changes could be automated.

And from there you went on to build banks and banks of equipment in Zinovieff’s Putney Studio.

Yes, I soon left my job with the National Health Service and went to work full-time for him making the equipment for his studio.

There were hundreds of unlabelled knobs and switches on those racks: what were they for?

Well, very soon he bought this American computer.[1] They’d just started making computers that were affordable then, and his principal control was by Teletype machine, you know—an old-fashioned telex, and he soon found that very limiting, so he wanted, simply, to have lots of knobs to turn. Typing numbers in just isn’t very conducive to creativity.

So they were providing inputs to the computer—they weren’t actually banks of oscillators or things like that.

No, the oscillators were extra pieces of equipment that I built, and they were controlled by the computer. The computer was in turn controlled by a program or by gestures from the musician.

They were built specifically to work with the computer.

That’s right.

And at that time computers were very expensive...

Yes, a computer then cost £5000, which was about the price of a small house.

...with very little memory.

The memory was less than you have in the gadget that controls your washing machine nowadays.

Who had the idea of using a computer to control the sound devices?

That was Peter’s vision [2]. He was bold and adventurous, and he thought “let’s do it”, and he did it. And then I spent all my time getting it to work.

Did you have any experience with computers at that point?

Only in a sort of data processing way, in a clerk’s job that I once had.

So you didn’t come in knowing how it was going to be hooked up to produce sound.

No, but I’d been spending a lot of time fiddling around making—for my own amusement—a sort of synthesizer and a sequencer, having read a very early article by Bob Moog in a popular magazine.

Were you led to that by an interest in electronic music or by the technical challenge?

Well we had a band—I played bass guitar, very badly, but everyone’s ambition at that time was to be a pop idol.

What did you learn from Moog’s article?

He had the idea of exponential voltage control, which was to control oscillators and filters and all the sound-generating stuff from voltages so that you could patch them all together and make something that made a really complex sound.

But at the Putney studio we soon got past voltage control and got into this computer control of everything.

So the computer was directly controlling the analogue sound-producing devices.

Yes, that’s right.

What are your recollections of working at that studio? presumably a lot of your time was just getting the stuff to work.

The studio was in a wonderful place on the river at Putney. Peter had quite a lot of money available then and he really made it a nice place to work. He had a big bunker built at the bottom of his garden looking over the river, where he had his musical inspirations.

Did you get much time to appreciate the visitors and the music that was being made there?

Well, I worked about seven miles away, up here in Cricklewood and I went to the studio only to do work there, so I didn’t see many of his visitors. I was a sort of backroom boy, really. But sometimes there were interesting people like Ray Dolby and Bob Moog. Harrison Birtwistle came. Oh, and Stockhausen, whom I later met at another place. After we started selling synthesizers all sorts of pop groups would come.

At the time, you were doing things in the Putney studio that not many people were doing anywhere else.

Yes, those were the days when a transistor cost 7/6. At the National Health Service that’s what I got paid for two hours’ work...so it was really a rather expensive hobby.

The idea of building a device to sample sound into the computer—how did that come about?

At one point Peter asked me to make an analogue-to-digital convertor, that is, a thing that allows the computer to look at control-knob movements, but we found that it was fast enough to look at signals. So I built this analogue-to-digital convertor and as an experiment he just plugged the radio into it. The computer memory then was enough to sample about one second of live speech. That was in about ‘67, ’68, I think.

Very very early, then. Did you develop that or was the memory insufficient to make it practicable?

Well, the computer’s memory was so limited then that this one second didn’t do you much good. The bandwidth wasn’t that wonderful, either—rather worse than telephone speech.

By the time you get to something like Chronometer [3], as I understand it, the sounds of the clock mechanisms and all the rest of it were effectively sampled by an ADC, stored and manipulated by the computer and then spat out again. What was the breakthrough that got you from this one second of sound to being able to put together a piece like that?

Peter kept buying the latest computers that came out and of course the memory increased. Then I built him a hard disc recorder so that one could store some of the sounds on this hard disc. At that time a hard disc had 32,000 words [32K] of memory, as opposed to 32 gigabytes of memory like we have now.

And you devised a filter and oscillator bank to deal with the inputs and outputs of that. Is that right?

Yes. At that time, memory was still pretty short so in order to get more sound one had to transform it to the frequency domain, which means reducing the data so the computer could store more of it. That was what the filter bank was for.

It’s a kind of vocoding idea rather than a sampling idea as we’d think of it now.

Indeed, That’s exactly the principle of the vocoder.

It was a vocoder with very many channels.

Yes, exactly.

Do you recall how many?

64.

Is that what gives the music realised on this equipment its particular sort of ‘shimmering’ quality?

Yes. Yes, we found that one can reproduce a sound by imitating it on an organ. We found that if you play an organ fast enough, the pipes of the organ—even at their own particular frequencies—can make a sort of reproduction of almost any sound.

So it’s a chromatic resolution, a semitone resolution.

Yes, that’s exactly it.

In 1969, you, Peter and Tristram Cary set up the synthesizer company EMS in order to finance the studio. How did that come about?

Well, Peter’s demands on his funds were greater than he could afford...we sold the synthesizers to pay for Peter’s studio. That was the motivation for making a commercial company.

Before EMS’s first commercial product, the VCS3, there was a VCS1 that had been commissioned by the Australian composer Don Banks.[4]

Yes, that was our first experiment in making something that one could sell. I just put together various bits and pieces that I’d previously built for Peter and put them in a self-contained box, to see if musicians would find any use for it.

Was it called into being by Don Banks or by you and Peter?

Peter knew Don Banks—I’m not sure how—and he asked me to make something for him. It was all built by hand, before we actually made printed circuits. We only made a few of those [5] [VCS1s].

Were they actually used on any recordings?

I simply don’t know.[6]

How did the VCS1 evolve into the VCS3? How did it get its distinctive shape and how did you decide what features it should have?

Tristram actually built the first prototype box by hand. So really the shape of it was his doing. We’d had long meetings to discuss what features it should have, how many knobs, how many oscillators and so on.

A kind of ‘down the pub’ discussion.

Yes.

Literally?

Well, no, actually in Peter’s magnificent front room overlooking the river.

Did it surprise you that the VCS3 became so popular?

Yes. I never thought that many people would be interested in such outrageous sounds.

And it was popular despite its lack of keyboard.

Yes, well that was a bone of contention. Peter and Tristram weren’t really that much interested in well-tempered music. And I thought it should have had a keyboard from the start. We added one on as an afterthought.

While at EMS you were involved in the design of the VCS3, the Synthi A, the Synthi AKS and the Synthi 100. Was working on those synths interesting in itself or was the real interest the advanced one-off devices you were building for Peter’s studio?

Peter’s studio certainly had the more advanced stuff that we did. But trying to make a product has its own challenges and interests—trying to make it cheap and affordable. And I quite liked working on such little problems to make them optimum from an engineering point of view.

Did you get much feedback from musicians?

No, I was kept very much in the background as far as musicians were concerned.

That’s quite different from your experience at Electro-Harmonix.

Yes, we all worked in a jolly rock‘n’roll factory, where a musician would come into my room and tell me what was wrong.

And that didn’t happen at EMS.

No. Peter was my only musician. I did things exactly as he wanted.

Was the large Synthi 100 designed to a specification from anybody in particular?

Actually the first one was for Radio Belgrade in Yugoslavia. They’d asked us to make something and we didn’t really want to do it, so Peter told me to do it the easiest way I could, even if it was expensive, and then they said “yes”. Then we had to get down to work and make something that we’d be proud to sell.

Did Radio Belgrade come up with the specifications?

I think they gave Peter a rather vague specification and he just wrote some numbers down rather quickly: “make it equivalent to eight VCS3s” or something.

During your time at EMS, what pieces of music do you think used the studio to best advantage?

I didn’t like intellectual music, but I always appreciated Harrison Birtwistle, the things he made. They sounded pretty musical to me.

Chronometer, you mean?

Yes, that one in particular, I remember that.

Did you ever wish that someone would use the studio in a different way?

Yes—I was quite excited when I heard Pink Floyd’s album...I forget what it was called...

...The Dark Side Of The Moon...?

Yes, that’s right. I even bought the eight-track!

You found that personally a lot more interesting than a lot of what was going on at the studio.

Yes.

Did that precipitate your move to Electro-Harmonix?

No, I’m sorry to say it was economic. The first Korg keyboard synthesizer [7] came out and I could see that the writing was on the wall for us. Their engineering was just so good and it was tailored to what a working musician would actually want.

And they had the R&D budget...

Yes, they just engineered it beautifully, and it worked perfectly.

Did Electro-Harmonix make you an offer you couldn’t refuse?

Yes. In England back then there was a long miners’ strike and they turned off the television and the lights at 10 o’clock and I remember sitting in my dark room once, thinking “I've had enough of this”. So I went to visit some friends in New York and I just happened to speak to Mike Matthews, who made me an offer and I said “OK”.

So you entered what was, presumably, a very different culture from EMS.

Yes—the rock‘n’roll factory was a rather hippy kind of place. Mike hadn’t been in business for many years then, but it was all musicians who made really wonderful sounds that I liked.

When you got there, were you given a particular project to work on or did you have carte blanche to initiate things?

No—Mike wanted one particular thing that’s now called a phaser, which is a sort of chorus effect that applies to guitars. It’s something like the effect of shortwave radio fading in and out (sings), and he wanted that particular effect. That was something that I did at EMS. We had one brief try at making a gadget for guitar players.

The Synthi Hi-Fli...

(surprised) Yes! ...not many people know of that.

Mike Matthews wanted some of the devices that were in the Synthi Hi-Fli to be put into separate pedals?

Yes. He’d been noticing EMS at various trade shows that he’d been to and knew that our products were quite innovative.

In your first stint at Electro-Harmonix your designs included what are now considered classic pedals, like the Small Stone phaser and the Electric Mistress flanger.

Yes, they’re two of the gadgets I made that were well accepted.

Presumably you didn’t name them.

No! (laughs) Mike had his own feeling for the market and what was hip and had complete control of that.

Did you enjoy working on those devices?

Yes. It’s very satisfying to get immediate feedback from a musician who tells you when it’s good.

And that feedback would go into revised versions.

Oh yes. One spent hours slaving over what we call a breadboard—that is a thing where you can experiment with all the values and tweak them until you get it right.

So you worked closely with the musicians to give them the kind of sounds they were after.

Yes, they’d keep coming back to me saying “make it more milky”, “make it more toothy”.

Did you expect the pedals to become classics, or that 40 years later people would be obsessing over the differences between the various versions of them?

No—that’s astonished me. It was all rather hit-and-miss. Musicians are rather...I wouldn’t say superstitious, but they have this nostalgia for old things. Like most guitar amps now still use vacuum tubes.

What are your thoughts on the heated online debates on the pros and cons of specific components in vintage pedals?

Well I don’t have so wonderful an ear as real musicians have, so it could be that they hear things I simply can’t. And that my instruments can’t measure. Actually I’m rather sceptical about it.

You’re  more interested in working with new technology than going back to discrete old components?

Oh yes. One of the challenges of engineering now is to keep up-to-date. The pace of things changes so fast and so many new devices are available.

What are you finding particularly interesting at the moment?

We’re working on a new wah-wah pedal called the Crying Tone, which is amazing because it has no­ moving parts! It’s much more responsive than a regular pedal where you have this rack-and-pinion mechanism that turns a potentiometer. It’s all rather stiff, and there’s hysteresis. This new pedal gives the musician much finer and more immediate control.

After your first spell at Electro-Harmonix you went to work at IRCAM for a while. How did you find working there after the ‘funky pedal factory’?

It was good to work there...IRCAM had all sorts of wonderful engineers from MIT and Stanford. It was good to work with those great people.

What did you work on at IRCAM?

The French Ministry of Culture had decided to build the best computer studio that they could, so they had ample resources to get engineers from everywhere. And I was one small cog in that large machine.

Were you involved with the 4X?

Professor [Giuseppe] Di Giugnio built that and I only had a minor part. I did little extra gadgets—I connected a keyboard to it and a box with knobs and a few mundane things like that.

What was the atmosphere like at IRCAM?

Oh, it was a very nice culture, with all those bright people. And then there were jollies where we went on trips to Boston and other places for conferences. It was a whole different world from that of commercial production.

And then you went to work for Akai?

Yes. The business in New York had financial troubles and one of the things...I’d made a sampler [8] for Mike in New York at a time when it was only just possible—the memory chips were just becoming big enough to make it feasible. And so we showed it round to many Japanese manufacturers and found that Akai was bold enough to take it up. That was 1984, I believe.

So Fairlight technology was already around at that time.

Oh yes, Fairlight had done a pretty good job, but their machines were very expensive and not available to the average working musician

When you designed the Instant Replay, was that something Mike Matthews had requested?

Well, it started off as just an extension of our Memory Man, which was a sort of electronic echo machine. And instead of a tenth-of-a-second echo we found that could have two seconds or, with a bit more memory, ten seconds. And then with ten seconds you can do interesting things like a looping recorder.

And that was the start of the now-ubiquitous looping pedals...

Yes, yes. That was ‘84 or ‘85, just around the time when Electro-Harmonix faded and we went to Akai in Japan.

Had you been keeping an eye on the musical use as well as the technical aspects of sampling?

Yes. I’d always kept up with Computer Music Journal and quite a few academic papers.

Was the Akai S612 a deliberate attempt to make an affordable sampler?

Yes indeed. That was exactly it. Yes, the basic principle of a sampler is like the Mellotron where you record a second or two of each note in a little memory and you play it back by pressing a key. And if you play it back at a faster speed you change the pitch of it.

What was the breakthrough that allowed Akai to produce a cheap sampler?

Simply the falling costs of memory chips—and Akai’s engineering was really good. They had all sorts of engineers paid not very much money to do a really first class job on the nuts and bolts of it. And if you do things on a reasonable scale, as Akai was doing, you can buy components much more cheaply.

Were you given a particular brief by Akai or did you direct it?

Well, I designed the first product, the features and functions of the first product. But after that they had their product planning department, where they more or less took over that part of it. They told me exactly what knobs and buttons it should have. In fact my first sampler had 26 buttons, so the musician could summon one of 26 sounds at will. They soon squashed that idea and made it more rational.

At the time, did you foresee that using digital samples would become part and parcel of normal music-making?

Well, yes. It was certainly much better than…the previous synthesis method was what they call subtractive synthesis, where you have a buzzy tone and a filter and the filter removes the buzziness. And then there was [John] Chowning’s synthesis, which is frequency modulation as used in that classic Yamaha DX-7. That was algorithmic synthesis. But the idea of the sampler just took over. It was all so much easier. You simply recorded the sound that you wanted on this chip and replayed it at will. That just gave you much better fidelity.

Did you have a hand in the Akai MPC60?

Yes, but Roger Linn created that product and just told me to do the electronic circuit design, so my contribution was limited to that, and designing a silicon integrated circuit to do the tricky bits.

Is there anything you’d like to design or make if someone offered you the chance?

Yes. It’s always been my ambition to create a new musical instrument.

Do you know what it would be like?

It would be a miniature keyboard. But the trouble with a musical instrument is that a musician generally has to spend half a lifetime learning how to play it. So if someone were to invent something like a saxophone for the first time, before anyone had ever played it, it would be half a lifetime before anyone could get some good sounds from it.

And the marketing department isn’t going to like that too much…

Exactly. (laughs)

Looking back, what are your thoughts on the current state of electronic music technology? What are the remaining challenges?

The big challenge—actually the thing that ruined Akai—was the computer revolution. Now—with a personal computer or a hand-held computer—a chap, with just a bit of patience, can sit in his bedroom and write a program that can do things that were just unthinkable ten years ago. And it’s quite hard to make physical hardware products that can compete with that. So I think all sorts of people at home can make great innovations in music creation simply by writing a computer program.

Do you think there’s a need to change the way in which humans physically interact with software?

Yes. Computer programs are rather limited by a mouse and a keyboard. I think that new kinds of keyboards—or the interface between a musician's fingers and the sound apparatus—is still something that is just beyond the reach of a computer.

Is that the sort of thing you were alluding to when you mentioned a small keyboard instrument?

Yes.

And it would have a different method of interaction?

Yes. I can see they’re making good strides in touch screens except they’re not quite fast enough…and there’s also a sort of random delay of ten milliseconds, which is just enough to make it annoying—to limit how good it can possibly feel to a musician.