Tim Goodyer talks to the man who's programmed Peter Gabriel's synths, recorded a host of electronic albums under the name Synergy, and started up his own record label for instrumental music.
Larry Fast is known both as an electronic composer in his own right and as the provider of Peter Gabriel's synth textures. He's in love with the synthesiser and the orchestra, but sampling is another matter.
"Post-modern architecture uses composite plastics and high-efficiency glasses in the same way that I've used electronics and digital recording technology to express earlier harmonic structures."
When the time was right to record Metropolitan Suite, Sony offered Larry Fast free use of their PCM 3324 24-track digital recorder and PCM F1 two-track mastering machine. In fact, Fast's studio turns out to be a good destination for such technological experimenting, since the Sony affair represents just another line on a list that already includes development work on the Polymoog and work with Bell Laboratories' multi-million dollar mainframe computer. How did this technical adventure start?
"As well as my teenage obsession with classical and rock music, I was fascinated by electronics. At the point where my electronic tinkering and my musical obsession converged, I thought there was a way to make art out of technology. I was still at college at the time and I managed to use the music I was doing as a senior composition project in 20th Century composition. And that material became the basis of the first Synergy album a couple of years later.
"From that point I more or less followed the popular history of synthesisers: the modulars gave way to the MiniMoog, that evolved into the first polyphonic synthesiser, which then gave way to polyphonic synthesisers like the Prophet 5, then came samplers...
"It was around 1976 that I got to do some consulting work with Bell Laboratories on the earliest laboratory-based digital synthesisers. These required mainframe computers to run and they were wonderful; they showed me then what was going to be happening to synthesisers in the future. We were able to run software emulations of FM synthesis based on the papers that were being published at Stanford around that time. The early versions of those were like the Yamaha FM series of synthesisers is now, but with a lot of additive synthesis which hasn't really caught on yet. We even had a very sophisticated sampling system which was the first opportunity I'd had to grab a sound, throw it into the computer and play around with it. It was also my first exposure to the C programming language which has since become a standard — in fact C was developed at Bell Labs.
"The system was a combination of the in-house mainframe running a large series of wire-wrapped boards which were performing specific functions required by the digital synthesiser, and a series of linking programs in C. That way, the hardware could be reconfigured to work in different ways within the mainframe. You could turn the system from an additive, 256-oscillator, variable-partial synthesiser to, say, an 18-operator FM synthesiser by rewriting the linking programs. Fortunately I didn't have to worry too much about that because I had the help of a brilliant young engineer called Greg Sims.
"Looking back, the thing that surprised me more than anything else was the cost. The mainframe was in the $2-3m range, but by 1983 the DX7 had appeared costing only $2000 — I remember thinking that someday synthesisers with these facilities would be available a lot more cheaply; I just never envisaged it moving that quickly. Now, with the MkII DX7 available, we're going to start seeing DXs for under $1000. I'm sure that advances in digital recording and other related digital technologies will move even more quickly. And there will be other surprises all along the way — pleasant ones."
Given fast's considerable experience with synthesiser technology and his evident enthusiasm for the subject, it would be easy to think that technology could do no wrong by him. Not so...
"I've always had mixed feelings about sampling", he reveals. "I don't categorically reject it — it can be a very useful tool — but I have reservations about the wholesale lifting of sounds. There's even a market in trading sampled sounds now: how can that give a composer an individual identity? I'd rather retain the control over sounds that pure synthesis gives you so you can tailor them to fit your music; that's very hard to do with straight sampling."
"I don't categorically reject sampling, but I have reservations about the wholesale lifting of sounds. There's a market in trading sampled sounds: how can that give a composer an individual identity?"
A quick listen to Metropolitan Suite confirms the absence of any obvious samples, but Fast's reservations go beyond the odd stolen cello or snare drum.
"Grabbing a whole four-bar section off somebody else's record or co-opting somebody else's drum sound is as bad. I think it devalues what they've done and what you're doing, it takes some of your individual creativity away. It has good points as well as bad points, but I think sampling is fraught with dangers for the uncreative mind."
Fast's smile returns with the mention of resynthesis, a process that involves the sampling and subsequent recreation of the sampled sound.
"A sample is taken in by the sampling part of the instrument and analysed, rather than just spat out again as a sound that gets more Mickey Mouse-sounding as it gets higher and Gregorian as it gets lower. A sophisticated computer then sets the dials on the synthesiser part of the instrument so that it's as close to the original patch as possible. Once it's in there you can work on any element as if it were the original sound.
"It's only a starting point, but the hardest part of duplicating a sound is if you're dealing with 256 or 512 oscillators working as harmonic partials. Then you're talking weeks of manual work to create something you may not want quite like that anyway, so it's a good jumping-off point. That's where I see the future of sampling.
"When I deal with sampling myself I tend to use it as a waystation to resynthesis. If I want a string section I won't lift it wholesale, what I'll do is try to isolate the bow-scrape from the front of a sample and synthesise the remainder. I could do it all on a synthesiser but it'd take me a long time, so I look at a snippet of a sample as being the way to getting a complex waveform, and then use that in conjunction with other synthesised elements to create my sound.
"I don't deny there's a laziness factor involved, but sometimes it's more important to be working on a composition than spending three days on a little bow-scrape at the start of a sound."
Fast's involvement with technology has taken him through several generations of synthesisers. Some, like the Bell computer, he never actually owned, but many others involved the investment of his hard-won cash. And if you've ever wondered what happens to the gear that keyboard "wizards" put on the backs of their LP sleeves...
"I do have a lot of obsolete equipment, little Micromoogs and string synthesisers and so on. A lot of them aren't up to much any more: they're not very flexible, they don't integrate into the MIDI system, or they're very noisy — so they end up in storage because they're not really worth anything any more.
"The marketing penetration of the music video is closer to the way other consumer items are brought into the living room and sold. It worked for toasters, candy, food, cigarettes, and now it's working for music."
"But I'm lucky in that it's my business, so my accountant can write off the tax depreciation on them. Other things, like some parts of old modular systems, still sound good and I still use them now. They can't have much of a resale value any more, but that doesn't really enter into the equation because they're still a good way of getting an idea onto tape.
"If an older instrument has proved its musical value, then it can become a fabulous secondhand bargain. On the other hand, if it's, say, a two-generation old sampler and sampling isn't even what the composer's goal is, then its musical and monetary value might as well be zero.
"The point is that the flashiest new blinking-light equipment isn't important unless it justifies itself as a musical tool. I've always subscribed to the view that the composer should view instruments as tools to achieve his ideas. I've always been completely opposed to instruments driving compositions.
"The trouble is there's always been an element of the sports car collector in electronic music, just like there is with guitars. If there's a fabulous blues player whose '57 Strat with a maple neck makes a difference to him then more power to him, or even to a pure collector who doesn't pretend to have a musical interest in the things he's collecting — I can respect that. It's thinking that the equipment makes the music that's a fallacy - musicians make the music. Good music is hard to kill, it may sound 15% better on the latest equipment, but the music is what ultimately counts — as you can tell from the charts at the moment."
AND WITH THE top of the pop charts dominated by re-released "classics" and cover versions of old songs, it would be easy to agree. But couldn't it simply be acute nostalgia that's got everyone wearing Levi's and playing Ben E King's 'Stand By Me'?
"I think 'Stand by Me' and 'When a Man Loves a Woman' are now so old that the majority of the record-buying public are too young to remember them from the first time around.
"I'm more inclined to look at the American model which is: whatever the media push at any given time, whether it's commercials or movie scores or whatever, that is what's foremost in the public eye. Of course, good records are good records so they tend to be a bit timeless, but something has to push them back into circulation. 'Stand by Me' was a hit in the States a while back because of the Levi's ad - just like here, but not to quite the same extent.
"The power that film and TV have in mass-marketing is so much stronger than what's traditionally been available to the radio and record promotion markets, that anything getting into that film and TV pipeline tends to overwhelm the conventional marketing that exists for records released through the usual channels. When MTV arrived in the States it provided a single national forum for music videos, and it immediately became the prime dictator of what was going to become a hit and what wasn't going to become a hit. It completely overwhelmed 20 years of album-oriented rock FM radio overnight. It's tapered off a little bit now because competition has come in from other cable video sources, and I suppose a little bit of boredom has set in anyway. But I'm a firm believer in the power of the visual image as a marketing tool.
"There's a whole tradition of film and TV marketing, and the marketing penetration of the music video is closer to the way other consumer items are brought into the living room and sold. It worked for toasters, candy, food, cigarettes, and now it's working for music. Radio might have seemed pretty heavy-handed over the last 25-30 years of rock 'n' roll, but it's still mild in comparison to anything with a visual push, and especially now that Hollywood has linked filmscores to any kind of pop record that can be piggybacked onto them. It's almost got to the stage where there has to be something that can be pulled from a film as a music video, whether it's appropriate to the film or not. And that's led to some ridiculous juxtapositions of music and images in films that never really needed it."
Very true. Our conversation draws to a close when the call comes for Fast to return to pre-production duties with the Dream Academy.
"We've got the greatest drum sound going down there at the moment", he enthuses, "but it'll be a while before the serious electronic treatments begin".
And with a last smile, Larry Fast is gone. He seems to have enjoyed our conversation, but I think he'll enjoy the remainder of the day in the studio a lot more.
Interview by Tim Goodyer
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!