Paul Wiffen made no secret of his fondness for the OSCar in September's issue, and now reminisces about the other great keyboard love of his life — the Elka Synthex, a synth whose reputation and influence far outweighs the number that were ever sold.
Throughout the '70s, Italian manufacturers acquired a reputation for string synths often combined with electronic pianos or brass sounds. These rejoiced in names like the Crumar Multiman — the first keyboard I ever owned — and the Siel Orchestra. Perhaps the most well‑known was the Elka Rhapsody, famously pictured as the only identifiable keyboard in the mass of spaghetti and towering modulars which was the Tangerine Dream onstage rig.
Up until the beginning of the '80s, professional polyphonic synthesizers were chiefly an American phenomemon, with Sequential Circuits' Prophets and Oberheim's OB series ruling the roost (Moog never really made the transition to true polyphonics until the Memorymoog in the final days of the company). But at the beginning of the '80s an Italian company, Elka, released a polyphonic synth whose reputation has survived and grown with the passage of time.
The Synthex was the brainchild of an independent Italian designer called Mario Maggi, who did much of the development work on his own. However, back in those days, polyphonic synths cost three or four thousand pounds and had enormous development costs, so Mario took his design to the town of Castelfidardo, where most of the Italian musical‑instrument manufacturers were based, to find financial backing. Elka were at the peak of their success, with a successful range of home organs and accordions, and wanted to build on the share of the professional keyboard market which the Rhapsody had recently given them, so the already well‑advanced Synthex with its truly programmable synthesizer approach presented them with a fast track to this goal.
I was fortunate enough to get a look at, and more importantly, a listen to the Synthex at its first trade preview in Frankfurt. At this stage, Mario had been working on it for well over a year, but the owners of Elka‑Orla UK Ltd, Nando Fabi and his son Fausto, were a bit nonplussed about the synth's appearance in the Elka range and wanted to get as many views on it as possible. I had little expectation of anything earth‑shattering, but then they took me into their back room... and I fell in love! An Italian home‑organ manufacturer was the last company I had expected to make something better than a Sequential Prophet 10 or an Oberheim OBXa. "Blasphemy!" I hear you cry, "Everyone knows that Prophets and Oberheims were the best synths of their day." That's right, everyone knows — everyone who's never heard the Elka Synthex, that is.
From power‑on, the first advantage of the Synthex hit you straight away: whilst American synths would still be going through the time‑consuming routine of tuning their analogue oscillators, the Synthex was in perfect tune, ready to be played — and it never ever went out! Like the OSCar, the Synthex benefited from digital oscillators implemented with TTL circuitry, which was a major innovation back in the early '80s. In 1999, it is difficult to remember what a royal pain the tuning of analogue synths was, even with the benefit of auto‑tune functions. Half way through a gig, or a take, the temperature would change and the oscillators would go out of tune (did you stop to tune, which could take 40 or 50 seconds, or just plough on?). One night or on one take, the synth would sound great because the oscillators were just the right amount out of tune, and the next, it just didn't sound so good. Nowadays, this unpredictable element of synths has been all but completely eradicated, but back then it was a fact of life. The Synthex was the first fully featured polysynth I had seen with digital oscillators, which made using it in the most unstable environments a pleasure.
An Italian home‑organ manufacturer was the last company I had expected to make something better than a Sequential Prophet 10 or an Oberheim OBXa.
But this is a practical advantage, not an aesthetic one, and it would have been of little value if, as the experts of the time proclaimed, all digital oscillators sounded thin and weedy. The problem with the first digital oscillators was that the engineers concentrated on producing perfect representations of the various commonly used analogue waveshapes such as square waves and sawtooths. However, it only took a quick glimpse at the output of an analogue oscillator on an oscilloscope to see that what they actually produced was a very rough approximation to the nominal waveshape. The best early digital synths, then, used the accuracy of digital to recreate the waveshapes analogue machines actually produced, rather than the theoretical versions. This is why the oscillators on the Synthex sounded as good as their analogue counterparts — I was talking recently with a famous Synthex owner who had never realised that they weren't analogue. "How do you think they stayed in tune?" I asked. "I just thought they must be very good analogue oscillators," came the reply!
But it wasn't just the sound of the oscillators which made the Synthex fantastic, it was what you could do with them. In addition to the standard waveforms, you could modulate the pulse width for both oscillators independently, sync each to the other (both at the same time if you were into really weird stuff), ring‑mod each or cross‑mod them, all without tying up the equivalent of the poly‑mod section on the Prophet. For all the options mentioned, there was a simple button on the Synthex's oscillators which you could switch on or off. This made programming a breeze as you didn't need to spend ages changing the routing to see what PWM, sync or cross‑mod would add to a sound. The sync settings in particular gave the Synthex an edge with which you could have cut tempered steel, allowing its bass and lead sounds to carve their way through the fullest of backing tracks. I was once asked by a supergroup's producer to tone down the edge in a Synthex patch, because he was worried he would never get the guitars to speak over the top of it!
But it wasn't just the oscillators of the Synthex which outperformed those of its analogue contemporaries. The 4‑pole filters (which are of course analogue, as digital filtering was still over 10 years away) went much further than any of the American polysynths, with not only low‑pass operation, but high‑pass and two different widths of band‑pass configuration as well. As a result, there were thousands of timbres it could produce which the transatlantic polys couldn't even come close to. I never understood this reluctance on the part of the Americans to put high‑pass and band‑pass filtering on polysynths. All the best monosynths had a multi‑mode filter (even some of the very cheap ones), but when you graduated to a polysynth you had to manage without them for some reason. I have heard the occasional apologist for the transantlantics claim that high‑pass patches sound stupid when used polyphonically, but I have 30 or 40 Synthex patches which prove them wrong.
But even if you restricted yourself to low‑pass operation (which the majority of all synth patches use), the filter on the Synthex was magnificent. The bottom end was solid enough to base an entire track on, especially when accentuated by a resonant sweep. In normal operation, it could always be relied on to bring out the meat of the sound. As a result, a single Synthex pad or bed could form a solid backdrop for a track, saving on overdubbing.
The sound was further thickened by the addition of a chorus unit with three fixed presets. Of course, in these days of built‑in DSP multi‑effects this seems a little on the light side; at the time, however, it was not only revolutionary, but it was also dismissed by many traditional synthesists and manufacturers as 'cheating'! Those of you who followed my 'Synth Effects' series in SOS earlier this year will remember that I credited the Synthex with opening my ears to the possibilities of onboard effects. With its chorus facility, it was the only polysynth which could do as good a job on strings as a dedicated string synthesizer. This, more than anything, gave a sense of its pedigree: I still maintain that what passed for strings on Prophets and Oberheims was pathetic even by contemporary standards, and it was all down to their manufacturers' refusal to specify an onboard chorus unit
When I came to demonstrate the Synthex at Frankfurt the following year, you wouldn't believe the number of people who asked me to take the chorus off the string patches and would then say, "See, it doesn't sound any better than the strings on the Prophet". Not until you put the chorus back on, no! There really seemed to be an idea that Elka were not playing the game properly by putting chorus on the machine.
I have never found three settings on any other chorus unit which so perfectly captured the different uses to which chorus can be put. It really didn't matter that you couldn't edit them. The first setting was so subtle you really wouldn't characterise it as chorus — it just gave a certain ambience to the sound (remember that digital reverb was years in the future) and worked perfectly on sounds like church organ to place them in the kind of sonic space you would expect them to occupy.
The second setting moved into the area of real chorusing, though it was still fairly subtle. Almost any sound could benefit from this, giving it a thickness and richness without screaming 'effect' at you. The third setting was full‑on — undeniably the richest, creamiest, most over‑the‑top chorus they could get away with — and it was this which turned the usual thin, bland synthesizer string patch into a full‑ensemble program.
Of course if, like me, you believe in the maxim "too much is never enough", then you would not be satisfied with a single Synthex timbre, however fat and rich. Fortunately, for less than the price of a Prophet 5, you got the Prophet 10's ability to have two different sounds simultaneously, but because the Synthex didn't feature a poser's dual manual, you had the choice of layering the two sounds together or placing them either side of a user‑definable keyboard split. I found some fantastic combinations using the layering. The brass sounds on the Synthex were alright for a polysynth of the day, but if you layered a chorused string patch behind them, they suddenly turned into the brass on ELP's 'Fanfare For The Common Man'. It was as if you had put the brass patch through a very expensive reverb unit. When I took the Elka Synthex to show to Keith Emerson, this was the first thing I showed him. It was around the time his hideously expensive Yamaha GX1 (on which he had recorded 'Fanfare') was starting to fall apart, so he ended up using the Synthex to perform the song live. In general, layering sounds allowed you to have something punchy and in‑yer‑face doubled with a more ambient, washy patch to create a much more complex sound — the sort of thing which is very common in the Combi or Multi modes of today's synths, but which was almost unheard of back then.
Splitting the keyboard (by simply holding down the Split button and playing the then‑silent key which you wanted to be the lowest note of the upper patch) was more of a performance option, especially as performance controls could be separately assigned to either the upper or lower side of the split or both. This meant you could have movement of the Synthex's joystick in one direction doing the pitch‑bend for a lead or bass sound, while moving it in the other direction introduced vibrato or tremolo (from the second LFO) to the other timbre.
Normally, I am not a big fan of joysticks, preferring the dual‑wheel method of performance control. But the joystick implementation in split mode on the Synthex really made it a great performance control, which could do things unachievable using two wheels. It was a little tricky to control, as it was placed on the main programming panel rather than directly at the left‑hand side of the keyboard. But you got used to it, and it freed the left‑hand end of the keyboard for the Synthex's secret weapon, the four‑track sequencer (see 'Step‑time Heaven' box).
In a nutshell, the Elka Synthex was significantly better than the Prophet 10 and the OBXa, because it stayed in tune, had far more flexibility in its parameters and routings, had a much better sequencer and most of all, sounded better. Oh yes, and it weighed and cost less! Of course, the comparison is a little unfair as the Sequential and Oberheim machines came out earlier, so the Synthex was designed after they were already on the market.
Although most people who listened to the Synthex at the BMF and Frankfurt seemed really impressed, I was horrified to discover that this did not translate into orders in the UK. Of course, it took them a while to tell me this (I was only the demonstrator after all), but once I found out I took it personally and went on my own crusade to rectify the situation. I started with Argent's, which was the only real specialist keyboard store at the time (the London Synthesizer Centre having gone the way of all flesh the year before). I remember an extensive session with Argent's manager Mickey Taylor, where he put the Synthex through all kinds of tests which concluded with him being quite positive about the machine's capabilities and promising to think about ordering one. This never happened; nor were other major stores prepared to shell out. The only Synthexes sold through stores in this country went through organ shops (who already had accounts with Elka‑Orla) and it began to dawn that the Elka name was the problem. There was even a half‑hearted discussion at one point about trying to get Elka to produce it under the anagram Lake, but I was told that Italian pride would never allow this.
Undaunted, in 1984 I started taking it direct to end users, despite the fact that by now I had landed a job as UK rep/demonstrator for Sequential Circuits. Theoretically, my continuing Elka promotion should have put this job on the line, though there was little actual danger of them finding out, as all Sequential's other European employees were based in Holland. Sent to see Geoff Downes (formerly of Buggles and Yes, but in supergroup Asia at the time) to show him Sequential's wonderful Drumtraks and the contrastingly horrendous Sixtrak (he bought both — he was never known for restraint when it came to buying keyboards), I politely enquired if another day he would like to see a couple of other interesting synths. He agreed, and so back I went a few days later to an Asia rehearsal with the Synthex and the OSCar. Geoff bought them both, and singer John Wetton also bought a Synthex. A year later, Geoff asked me to come and program for him during the recording of the third Asia album, and we ended up MIDIing the Synthex into almost everything.
At about the same time Geoff bought the Synthex, we heard that the French distributor had got a Synthex to Jean‑Michel Jarre, and although he was quite impressed he was asking for more patches. We sent over the tape of 100 programs I had done for the Synthex and heard nothing more. A year later I was sunbathing in a friend's garden and heard one of my Synthex patches coming from her living room. I went indoors to discover the new Jean‑Michel Jarre album Rendezvous, with Synthex all over it (even credited on the sleeve for every track). The track I had heard from outside (the section usually referred to as the Laser Harp, from his triggering of this sound in live concerts) featured Synthex almost exclusively, and one of my sounds was featured in splendid isolation for almost a minute. The Rendezvous concert on the TV a few weeks later featured the Synthex prominently. At last someone appreciated it and was prepared to be seen in public with it! To hear the Synthex at its best, listen to this album. Years later, I finally met Jarre and even ended up lending my Synthexes to him for the Docklands concert, from which they returned full of water — though on drying out they worked perfectly.
I have never found three settings on any other chorus unit which so perfectly captured the different uses to which chorus can be put.
Unfortunately, by the time my guerilla marketing started to pay off — getting the Synthex on TV with players like Peter Oxendale in Bonnie Tyler's 'Total Eclipse of the Heart', Mark Stanway of Magnum and Phil Lynott in Grand Slam — Elka had lost faith in the instrument and were selling it off directly for £500 through the music technology magazine Electronics and Music Maker. Some people got a magnificent bargain, considering the original retail price was over three grand. As payment for programming work on some hideous FM‑based unit that Elka came out with (my mind has deliberately blanked out the model number), I got Elka‑Orla to give me two of the last Synthexes and to ship one of them to Los Angeles, where I was moving. Called in to try to make sense of Stevie Wonder's new Sequential Studio 440 drum machine (into which my Prophet 2000 samples would thankfully load), I was soon up to my old tricks, asking if I could bring in other synths. Next thing we were MIDIing the OSCar and Synthex together for the killer bass sound on his 1987 single 'Skeletons' and I was in there. After the album's release, I toured the world with Stevie as MIDI programmer, ensconced under the stage with the Synclavier guy. Unfortunately, because of their limited MIDI implementation, neither the Synthex nor the OSCar went on the road (except recorded into the Synclavier), but Elka‑Orla cobbled together one last Synthex for Stevie from several ones that had been cannibalised for parts (because I refused to sell him mine).
Ironically, since its demise as a production unit, the Synthex's reputation has grown. They are certainly changing hands now for more than the £500 for which they were eventually all sold off and, whilst visiting a dealer earlier this year, I overheard a phone call from someone desperate to get hold of one at any price.
The final Synthex anecdote I would like to tell is from around the time of the opening of Martin Newcomb's synthesizer museum. I went over a couple of days before at the request of a friend, as apparently there was a broken OSCar only making bloops and gloops that I might be able to fix. When I got there, I discovered that it had merely lost its memory and we didn't have any way to reload it in time. However, I looked in another corner and there was a Synthex. "Oh that's completely useless, it lights up but won't make any noise at all," I was told. A vague memory stirred in my mind and I asked if I could take a look. Permission was given and I switched it on. Sure enough, the lights all came on and I reached over to the back panel and flipped a switch. Instantly the massive sound of the first preset came to life. "It's a miracle!" was the astonished reaction from those around me, and I ended up playing it in the video that was made at the opening (Analogue Heaven). The moral of the story is, if you come across an old Synthex which is reputedly beyond repair, try switching the Cassette Interface off!
Thanks to Martin Newcomb of the Museum of Synthesizer Technology and Andy Horrell of the EMIS Synth Museum (+44 (0)117 956185) for their assistance in sourcing photographs for this article.
Those of you who read my OSCar retro article in September's SOS will know that it and the Synthex were, roughly speaking, contemporaries. As such, both were shipping before MIDI became available as an open standard, and sales of both suffered from the lack of the now‑familiar interface. There was actually a D‑connector computer interface on the back of the Synthex (though no software was ever written for it, as far as I'm aware) and it was this socket which was sacrificed for a retrofit MIDI interface. The original software for this was really basic and didn't allow the channel to be changed. Eventually this facility was added, but I don't think the Synthex ever supported Program Change or SysEx dumps. The tape interface was the only way to save or change the program set and sequences.
The sequencer on the Synthex had four monophonic tracks which could be programmed in step or real time. Although real time was perfectly usable (and had a sort of auto‑quantise effect if you recorded with the clock set fairly slow, because of insufficient resolution), my favourite input method was step time. This was because the controls were set out in such a way as to make it really fast to program. For the shortest notes in your sequence (which you set to the step interval), you simply played in the required notes (unlike most other sequencers of the time which required the confirmatory press of a button). The Beats/Rest button allowed you to increase the length of a note to a multiple of your step interval (by pressing it any number of times whilst still holding a note) as well as insert spaces of any length (by pressing it whilst not holding a note). As a result, step‑time programming became almost as fast as real‑time input, but with greater accuracy.
What was really innovative was that if you made the sequences different lengths and then looped them, you could get a constantly‑changing interaction between the sequences. For example, you could make track one four steps long, track two five steps, track three six steps and track four seven steps. This would result in a pattern of notes which would only repeat every 420 steps. Of course, you had to program the sequences so that the notes didn't clash (by sticking to pentatonic scales or at least keeping the pitch ranges separate).
But this wasn't the end of the sequencer's innovation: a fairly early software update allowed you to assign the tracks to either the lower or upper synth sound. So you could, for example, set one track to a bass or lead sound and the others to a three‑part chord sequence. All previous built‑in sequencers (or external ones) that I am aware of could only trigger one timbre. Does this make the Synthex the first multitimbral sequencer?
- Jean‑Michel Jarre Rendezvous
- Asia 'Suspicion' and 'Voice of America' from Astra
- Stevie Wonder 'Skeletons' (Characters album)
- Magnum 'How Far Jerusalem' (Progressive Classics)