Geoff Downes, once one half of the Buggles, then keyboardist with Yes and Asia, and now a solo artist with a forthcoming album, owns one of the biggest synthesiser collections in the world. How has he acquired them, and does he still find a use for them all?

SO THERE WE WERE. An equipment-laden photographer and I, walking innocently into London's Townhouse Studios. Suddenly, we were surrounded by keyboards. They filled the control room as well as the studio floor. They were standing against the walls, on the floor, racked up in threes and fours.

We soon gave up trying to count them, and settled down to the serious business of trying to cover them all in our discussions and photographs. Bear in mind that what we have here is the pick of 8000 words of interview transcription and no fewer than 10 rolls of film...

I suppose, to be logical, we should talk about the first keyboard you ever owned.

Well, that's one of the few I don't still have. It was a Farfisa... No, before that there was a Vox Jaguar, which cost me £60 when I was 13. That was an absolute fortune in those days. But I soon graduated to a Farfisa Compact Duo and then eventually bought my first Hammond when I was 16, a J122. The first two are long since gone (they're the only two keyboards I've ever sold), but the Hammond still exists today. It still works but it's been cut down and modified by Bill Dunn, London's longest-established organ customiser. It's now just a top keyboard and a box with all the gubbins inside it, to make it more transportable.

So you started gigging on the organ, rather than the electric piano?

Well, there was actually one other keyboard I sold, a Hohner Pianet, which was the first electric piano I had. Of course, I'd been playing piano for years, but acoustic piano was a bit inconvenient for gigs. Having studied classical organ as well, the portability of electronic organs made them the ideal main instrument.

In those days, it didn't matter what keyboard you had; if you could get it to the gig, you were in. Having a Compact Duo was the equivalent these days to a rack of DX7s, and the Hammond was really upmarket, you know, like having a Fairlight or Synclavier now.

The electric piano started to emerge as a viable alternative to the organ in the late '60s, like on the Doors records or the Zombies or 'I Heard It Through the Grapevine'. That was the new sound - if you had that then you were really cool, which is what inspired me to get something that sounded vaguely like that. The organ was great for covering orchestral lines; you could get that big sound out of it, but you needed the percussive edge for those sort of cover versions.

So I got the Pianet, which was an absolutely dreadful machine. It worked by plucking the strings with sticky pads. Mine just wore out, because the pads started getting less and less sticky as the dust got in.

When the Pianet died, what did you move onto?

That's when I got the Fender Rhodes, which I still use occasionally. At the time that was the standard setup: Hammond and Fender Rhodes. Of course, the Moog systems were starting to come out - you know, the wall-to-wall telephone exchange systems built into wardrobes, but you couldn't use them live unless you were Emmo (Keith Emerson).

What was your first synth?

Well, I waited 'til the Minimoog came out to get into that, simply because of the convenience. A lot of bands started using that - Yes, ELP, and so on - so I figured I'd better get one. I was at college at the time and we were "studying synthesiser" on that early EMS thing, the VCS3, which Floyd and Roxy were using. But no-one at the college knew how to get a decent sound out of it. We were always experimenting with it, waggling the joystick, but all it made was horrible noises. It's difficult to believe that was only 12 years ago, but that was the leading edge of synthesiser technology.

You've now got two Minimoogs, plus numerous other Moogs. I notice that with them and the numerous Prophets, you tend to favour the American sound.

Yes, I've always been an American (or at least an English-speaking) synthesiser man. It's the warmth of the oscillators I like, which even today the Japanese can't seem to produce. The Prophets and Moogs, in particular, excel at this. Take the Moog Taurus pedals, for example. You still can't get that huge sound out of anything else.

The next thing I bought was a Polymoog, which, although pretty terrible by today's standards, was the first polyphonic machine. That really did open up a whole new set of possibilities, because then you had instant string sounds and all those other pad sounds. And although the Prophet was probably far closer to being a polyphonic Minimoog, the polyphony was still a major thing.

Of course, when the Prophet 5 came out that was a whole new thing. The sort of changes you could make to a Minimoog patch you could now do polyphonically. The Polymoog was really quite rigid in its tonal alterations, but the 5 was a dream machine in terms of flexibility. It was like five Minimoogs plus a programmable memory, which was really revolutionary.

"I've always been an American synthesiser man. It's the warmth of the oscillators I like, which even today the Japanese can't seem to produce."

Was there anything between the Polymoog and the Prophet?

Yes, the Yamaha CS80. That was a glorification of the Polymoog approach. It did have four memories you could set up (if you include the two under that lid with the fiddly little sliders), but it was still mainly presets.

The CS80 has lasted a lot longer than the Polymoog in terms of its sound. It's still being used, despite its limitations...

Yeah. I bought mine around 1980, to help out with the Yes thing, but I was never that into its sound. I suppose it's part of my prejudice against Japanese synths. They just don't have quite the right sound for me. I've still got it though, and I'll be using it somewhere on the new album I'm sure. I'm going to get everything on there somewhere.

So what was the next major purchase in the Geoff Downes arsenal?

The next thing must have been the Solina. I got that around 1978, and it was the string machine to beat all string machines. At that time, on sessions, I was using that, the Clavinet D6 (bear in mind that when I got that in the late '70s, funky disco music was in demand), the Hammond, the Rhodes and the Minimoog. That was like the state-of-the-art setup at the time. Theoretically, you could get any sound you would be asked for out of that lot.

It was when we were doing the Buggles stuff that I got the Polymoog, the CS80 and then the Prophet 5. Then, when I went out on the road with Yes, which was the first really major live stuff I'd done, I added a Novatron and a set of Taurus pedals.

It was just before the Yes tour that I bought the Fairlight. In fact, I used it on the album on a track called 'Man in a White Car'. I'd literally just got it and I was still experimenting with it. I'd never spent anything like 15 grand on a synth before.

There had been the Yamaha GX1 before that, which Emmo and Stevie Wonder had got, which was supposed to be worth over twice that, but I don't know if they ever paid for them or whether they got them on some sort of endorsement deal. It was funny, when we were in Japan a few years ago - we saw quite a few of them stuck in resorts and hotels and places like that.

But a lot of people couldn't understand it when I spent that much on a "synth". The thing was, the Fairlight was so revolutionary that nobody could understand what it was about.

I had one of the first Fairlights in the country, certainly the first that went out on sessions. When word got around that there was an instrument which could sound like timps one minute and strings the next, then people wanted to use it on records. They'd been dreaming of it for years. So my name came up on record sleeves in connection with the Fairlight.

What was next after the Series II?

Well, the Prophet 10 came along shortly before we formed Asia. I bought it in 1981. I'd always loved the 5, and when I heard the 10 it just seemed a devastating sound. You'd just put it on the track and instantly the track sounded big - you know, that sort of epic hugeness... Because you had that instant stereo output, it made a great tonal enhancement of whatever else was playing with it. If you had the whole rhythm section going and you had a statement of the harmonic structure from the 10, then it was a great base to build on. People often referred to my sound in Asia as being a wall of keyboards, and the 10 was really the foundation that was built on, the backdrop against which everything went.

Was the Memorymoog another instrument that came into that "big analogue sound" category?

"People referred to my sound in Asia as being a wall of keyboards, and the Prophet 10 was really the foundation that was built on, the backdrop against which everything went."

To a certain extent, but the 10 was more fundamental. I used the Memorymoog a lot monophonically, so it didn't fit into the "instant pad" thing the way the 10 did. But it still had the warmth of those Moog oscillators, so it was great for bass sounds and so on. It's one of those synths that'll cut through anything. The problem was that to start with, it kept breaking down. Fortunately, it's reasonably reliable now because I had the 'Plus' update - that gave it the sequencer (which I never use) and the MIDI (which I use all the time). It also seems to have helped the tuning, which was all over the place originally.

When did the PPG Wave 2.2 arrive?

I think that came along about '82. That was another one I got because it had a sound all of its own. It was away from the normal analogue synths, yet away from the samplers as well. It was such a good idea to have a machine with the variability and sharpness of digital sound, but still with the warmth of filters and envelopes. There wasn't a need to learn a completely new way of working, but you got a completely new sound. Even today, there's nothing that sounds quite like the PPG. You can hear its individual quality no matter how much other stuff you MIDI in with it.

Moving back to Sequential machines, you're still using the SixTrak, which a lot of people reckon was not one of the company's better ideas. What is it about it that you like?

It's great for basslines, using the Stack mode to get lots of different elements and then combining them to make one big sound. I know on its own each voice sounds a bit thin, but you can't argue with six oscillators all going at once. It's actually better for bass sounds than the Prophet 5, because Stack mode gives you different sounds combined together, whereas Unison mode - which is what the Prophet had - only gives you five identical voices on top of each other.

But basically I'm an American synthesiser freak. I'll buy anything that comes out of the States.

Yet at about the same time you got the Korg Poly 800. That seems a bit out of place among predominantly American gear.

Well, I've always quite liked Korg stuff. I've got a Polysix and a Mono/Poly which I use a bit, and then most recently a DW8000. They're the only ones I find suitably different-sounding from anything else. And that's the reason I buy machines, after all - because they don't sound like anything else.

The Polysix and the Mono/Poly in particular have a warmth you don't find in most Japanese stuff, although with the Poly 61 it was back to the cold Japanese digital sound. The DW8000 and the DVP1 are getting back onto the right track, and the DW does have some very nice things on it, like the aftertouch and so on. It has a nice, wide, encompassing sound.

The only thing I find lacking in Korg stuff is bite. As soon as you put something next to them, it tends to impoverish their sound a bit. But they are definitely my favourite among the Japanese makes. I don't own a DX7 and wouldn't want to.

The DX must have something going for it, though. It has sold more than any synthesiser in history, after all...

Well, I think it's really a case of "more is less" as opposed to "more is more". If you stack a load of Japanese stuff together, you end up sounding like Howard Jones. I don't mean that in a derogatory sense, but you can't actually differentiate between the different synths he's using. I remember hearing Dave Bristow doing a great demo with the TX816 controlled from the QX1, but the problem was it all started to sound the same. The horns started to sound like the cellos which started to sound like the saxes, simply because everything was being done by DX7 voicing.

I don't think you can get a wide range of sounds without a wide range of synths. Add eight DXs together and it doesn't sound like very much, which is what I meant when I said more is less. Add together eight different American and European synths, and suddenly you have more than the sum of the individual machines - and that's what I mean when I say more is more.

Would you say that's because each machine is creating frequencies differently, so they complement each other instead of cancelling each other out, as they would if they were all using one type of synthesis?

"I don't think the DX7 has a very interesting sound. Sure, it does good superficial impersonations of lots of different sounds, but they all have the same quality to them."

I think it's more than that. I don't think the DX7 in itself has a very interesting sound. Sure, it does good superficial impersonations of lots of different sounds, but they all have exactly the same quality to them...

After the Poly 800 came a couple of less well-known machines - at least in world terms - the Elka Synthex and the OSCar. What attracted you to them?

Well, when I said I was into American synths, I meant as opposed to Japanese. I'm really into European stuff as well. That dates back to the Solina string machine, I suppose.

The Synthex was just a great-sounding machine. Before I got the Solina I used to use an Elka Rhapsody on sessions to do the same job, so when the Synthex came along I didn't have a big prejudice against Elka because of the home organ thing. It's actually a very, very good synth. I think of it like a programmable Solina. It does those warm strings sounds that only the preset string synths used to do, but it has loads of other good sounds as well, all with the same warmth.

I've always thought of the Synthex as a sort of cross between the Prophet and the Solina. Something with the programmability of the Prophet but with an extra warmth and size that you only got with chorusing and a stereo output. Sequential got that by doubling two 5s to make the 10, but it was an expensive way to get the same effect. The Elka had some other great things too, like digital ring mod and cross pulse-width mod, which could give FM-type sounds but with warmth. You didn't have to go through a lot of complicated procedures to get a good sound.

The Elka was worth its weight in gold, partly because I got it just as the Emulator II was coming out.

I got the EII a day or two before we started work on the third Asia album and it didn't have a sequencer on it at the time, and the MIDI hadn't come through on either the Fairlight or the Synclavier (which I was on the point of buying), so the Synthex's MIDI sequencer was really useful for sequencing the EII.

Then there was the OSCar, a great monophonic synthesiser which was programmable. It came after the majority of the monosynths, and it was the only other monosynth that came anywhere near the original Minimoog in terms of sound. And the great thing was, you didn't have to have all the patch charts that I still have to use with my two Minimoogs. The OSCar has all the features of a classic monosynth, but because it's more recent it's got a lot more, like MIDI, a pretty nifty sequencer with program changes and so on. Then there are nice little refinements, like delayed vibrato and a single-triggering portamento which lets you keep one hand free for a polyphonic accompaniment on something else at the same time.

And on top of everything, its got such a ballsy sound. On lead stuff it'll cut through anything. On several occasions I've got everything but the kitchen sink MIDI'd up for a huge backdrop, and there's the OSCar soaring away on top, cutting through it all.

The Emulator II, which you mentioned earlier, has been a very important instrument for you, hasn't it?

Yes. The EII came out at just the right time, bridging a gap that existed between the low fidelity and short sampling time - but polyphony - of the Fairlight, and the high fidelity and long sampling times - but monophonic playback - of the Synclavier as it was then.

The original Emulator was little more than a poor man's Fairlight with short sample time, and I didn't think much of it. But when the EII came out they had made it a real instrument. And like the PPG, it had standard analogue controls for altering the sound, so it wasn't another whole new thing to learn. I know that now its quality has been surpassed by much cheaper machines, but in those two years I've built up such a library for it that it would take forever to transfer it to, say, a Prophet 2000.

I'm perfectly happy with its quality. On the sounds (orchestral and tuned percussion, mainly) that I use it for, it has a lovely warm quality, which probably comes from the fact that it's not 100% faithful.

Over the years I've built up enormous libraries of sounds for the Fairlight, the EII and most recently the Synclavier, and I'm not about to throw all that away just because something with higher recording quality comes along.

"In my setup I have practically unlimited polyphony; it's great for a keyboard player to just be able to go ahead, play the part, and have the machine take care of voice assignment."

That's another reason why my setup becomes bigger. I still need access to all those sounds. I know what they all are in my head, so if I remember something that works in a particular context, I can call it up, or get my roadie to dig it out of the archives.

You're now in possession of both a Fairlight and a Synclavier, which some people would consider as being over the top. Having been such a champion, of the Fairlight almost since day one, how come you're not updating to the Series III?

Well, when the Synclavier came out originally it wasn't that interesting, to me at least. It was pretty expensive just as a synth and then, when it was updated to sample, it was just a straightforward recording-type implementation. The great thing with the Fairlight back then was that it was a complete system. You could play samples polyphonically and mess about with them, get different samples at the same time, and sequence them, first with the real-time recorder and then with Page R.

It was only when the Synclavier's sampling became polyphonic, which was less than two years ago, that it became more appropriate for me. It was very frustrating, being able to play synthesiser sounds polyphonically, but not the samples. And although the quality and bandwidth were a whole lot better, it still wasn't of that much use. You couldn't walk up to the machine and say "right, here's a piano or strings sample, let's record this piano or strings part". We used it on the last Asia album for drum samples but little else.

The great thing on the Fairlight was that you could do all sorts of things with samples: merge, overdub, sequence and so on. But since then the Synclavier has come on in leaps and bounds, to the point where it's a complete recording environment with high-fidelity polyphonic sampling, multitrack recording, and MIDI sequencing. Plus, most of this was there a good year before there was anything but rumours of the Series III. So I got the full Synclavier package, because it was available first.

Now the Synclavier is so far down the road, I don't foresee the Series III ever catching up. In fact, the system as NED foresees it is going to keep expanding into areas that Fairlight haven't even looked at - tapeless multitrack recording and so on.

And on the Fairlight, you can still only record monophonically on each track like Page R has always been, whereas the Synclavier records whatever you play and assigns voices accordingly. This is particularly useful when you're controlling external machines via MIDI, as in my setup I have practically unlimited polyphony. It's great for a keyboard player to just be able to go ahead and play the part and have the machine take care of voice assignment. There's none of this "I played five notes at one point so I've got to have five tracks of my 16-track sequencer assigned to that part".

Then there's the facility to drop in and record exactly as you would on a multitrack tape machine with rewind and fast forward, which makes the whole Synclavier system so immediate. It makes life so much easier.

When you first got the Synclavier it was still at the monophonic stage, and you were using it primarily as a super high-fidelity drum machine. Yet presumably you now see the recording side of it as the main thing now, as a central MIDI workstation?

Yes, it's my main compositional tool. I play all my other MIDI keyboards from it and sequence them with it as the master unit all the time. It's great for arranging as well as for writing, because I can change the sounds around. I've actually got the best of both worlds, because I can get all my old sounds from the Prophets (which are all MIDI'd) onwards, but also have state-of-the-art sampling, recording and sequencing in one integrated system.

Do you use the synthesiser side of the Synclavier much?

I do have a few synth sounds on it, but I don't really regard that aspect as terribly important, as most of the sounds I want are easier to get from an analogue synthesiser. Making up sounds from scratch on it can really be quite time-consuming, which is a bit of a contrast to how quick it is for sampling and sequencing/recording. I still prefer ADSRs and filters to partials, FM amounts and harmonic addition. The synthesiser side of it is fairly old now, and I find it quite limited, really. I don't think it gives you the weight of sound you can get from, say, the Prophet 5 or a Memorymoog. And you can't tweak cutoffs or jump oscillator octaves as you go along.

It's the newer side of the machine which really interests me: using the sequencer to record parts played using internal samples (drum sounds and so on) and external synths for the most part. It's like a major compositional tool and orchestrator. I record the performances into the Synclavier sequencer with either justified or unjustified recording (more often referred to as auto-correct or quantization on other machines) as seems more appropriate to the part, and then review the sounds to be used in context in the recording studio, so that time is not wasted there getting the part right. I tend to use justification on the drum parts and other rhythmic stuff, and then do the solo parts freehand.

The other thing about the Synclavier sequencer is that you can slide tracks against each other.

Yes, that's great for compensating for the different MIDI delays in the various synths I'm triggering, or the fact that a slow attack synth patch or sample is speaking late because of its rise time... If a part is dragging against the rest of the track, then I just advance it until it's in the right place, until it feels right. Then I store that as part of the sequence and it's always right on.

If I decide to change the sound or even the synth(s) playing that part, then I'll slide the track again 'til it feels right. For example, if I record a part with a guitar sound, but then decide I prefer the part on a horn sound, I might advance the track 30 milliseconds to make up for the fact that the horn hasn't quite the attack of the guitar. It's much better than cutting off the front of the horn sample to make it speak faster, which just makes it sound less like a horn. It's the ultimate control, because you can fiddle until it feels right - and that's the sort of control I've always wanted from all this technology...