Interfacing The Past (Part 1)
The missing sync?
For those who would indulge their passion for vintage synths, sequencers and drum machines, the journey 'back to the future' is fraught with difficulties - incompatibility, non-standardisation and plain, downright awkwardness. But help is at hand...
I've tried unsuccessfully to sync a Roland MC202 from the Trig Out of a Boss Dr Rhythm Graphic. What's the problem? And is there an interface I could buy to solve it?"
That cri-de-coeur from the December 1984 edition of MT's fore-runner E&MM is typical of many technical queries from that era. In many ways it was the hi-tech version of the kind of letters that appeared in the problem pages of the tabloids. For: Dear Marge, I'm having trouble with my boyfriend... read: Dear E&MM, I'm having trouble with my interfacing.
But if you were around in the early days of mass-market synthesiser production, you'll know that the biggest frustration of all was that there was just no affordable way of playing a synthesiser in the same way you would a piano or an organ - that is, with more than one finger at a time.
The second biggest frustration was that whenever you tried to overcome this limitation by hooking up a sequencer that ran several synths to produce polyphonic lines, no two pieces of gear from different manufacturers could ever be guaranteed to work together. In fact, you couldn't even guarantee compatibility amongst products from the same manufacturer if they were of different periods or product divisions (like Boss and Roland in the 1984 letter).
Then along came affordable polyphony, and, even more importantly, the widescale adoption of MIDI as an interfacing standard. Nowadays, with the move to General MIDI, things are set to become even more simple and problem-free. But just when manufacturers were expecting a pat on the back for really giving the punters what they wanted... what happens? A collection of awkward buggers start saying, "That's all very well, but...".
It's really convenient to be able to hook everything up 'just like that' and assigning similar types of patches to particular memory slots does prevent any sequence from ever sounding as ghastly as some of mine do (with old fashioned MIDI) when I've forgotten to save the patch numbers and been too lazy to write them down. But... doesn't it all sound a bit the same?
And of course there are the other awkward buggers who will still claim that dense MIDI sequences somehow lack the crispness of timing of a good old CV/gate sequencer. Some of this can be put down to technostalgia; an aberration that can easily be cured by an hour or two spent trying to find the leads to sync a Moog Prodigy and a CR78 to an SCI sequencer, and run the whole thing with an M8 micro-composer.
And yet... while Vince Clarke is obviously a bit of a maverick, going out gigging with his massed analogue sequencers and his cage of anachronistic goodies, if you look at almost anyone who's making waves in music today, they've all got a soft spot for the old gear; and in some cases the soft spot is the size of a mutant marshmallow.
Part of the reason for this is simply that old stuff (when it's up and running) makes you feel good. Just like there's no comparison between playing a tonewheel Hammond through a vintage Leslie and any (even the finest) imitation, there's no comparison between having a disk of great 808 sounds (or having samples of them in your drum machine) and having a real 808 under your fingertips, in all its splendid awfulness.
Whatever the reasons, let's assume that you might, at some time, hanker after something from a bygone age to slot into your sleek MIDI system - just to humanise it a bit, or simply because it's nice to occasionally buy something that doesn't depreciate faster than the pound against the Deutschmark. First off, let's consider sequencers and drum machines. What they need, as a bare minimum, is a synchronisation system. A drum machine can, of course, also be controlled by your MIDI sequencer in much more elaborate ways, but let's say that you're actually quite keen to work within the limitations of its original design - which isn't always as daft as it may sound.
Assuming you want to connect it up with the minimum of hassle and expense, let's look at the alternatives open to you (bearing in mind that the most straightforward things are not always the cheapest!). The simple solution is to go to a specialist firm like Kenton Electronics to get an internal MIDI interface fitted to each piece of gear you want to hook up. Problem solved. They seem to have MIDI'd up practically every electronic instrument under the sun at sometime or other. Examples of drum machines include Linndrums, Roland CR78's and 808's, and Simmons SDS5's, and the good news here is that (presumably because there's nothing like pitch-bend and modulation wheels to make interfacing complicated), it's a fair bit cheaper to fit MIDI to a drum machine than it is to a polyphonic synth.
Kenton's price for the 808 is £188 - and that includes a lot more than just the sync function we could actually get away with. For instance, programmable MIDI in and out channels, velocity response on MIDI in (or at just two levels if you want to be a purist and stick to the 808's original accent system), and all setups and MIDI note assignments memorised in non-volatile RAM. The bad news is that if you have a lot of stuff to sync up, the £188s start adding up to considerably more than the actual buying price of all but the most sought-after old drum machines and sequencers.
There's also the small point that in fitting a MIDI interface, no matter how professionally, you're still mucking about with a rare vintage machine. Remember that all electronic music equipment from before the DX7 is genuinely rare compared with the DX7, HR16, SPX90, M1 - and anything from the present era. Whether that bothers you or not depends on how long term a view you take, and on how precious an artefact you think your piece of gear actually is.
There are other ways of doing it, and they almost certainly work out cheaper. The trade-off, of course, is that they may well be much more of a fiddle to organise.
There isn't a great deal of current equipment which fits the bill; but Kenton themselves have bowed to public demand and produced a MIDI-to-CV converter which has a wealth of sync facilities as well as being able to send pitch control voltages to two separate monophonic synths. This little box (retailing at £176.25) has switchable gate and S-triggers on each channel, and DIN Sync 24 and arpeggio clocks out. (The Kenton internal mod on the 808 also includes the ability to run DIN sync from the 808 while you're slaving it to MIDI - very useful for the purposes envisaged in this article.) But for those puzzled by expressions such as DIN sync, gates and S-triggers let's go back to the beginning...
The earliest sequencers used DC pulses to control the note-on timing of the music they were sequencing. Should have been straightforward enough, but it wasn't. For one thing, there were disagreements about the level of signal required; for another (and this was much more crucial), some manufacturers opted for positive and some for negative pulses. Moog, as the originators of modern synthesis, might have expected other manufacturers to follow them with their negative S-trigger, but of the major manufacturers that emerged, only Korg did anything similar, and even their negative pulse was incompatible with Moog. Almost everyone else plumped for a positive pulse, which meant that a little box was called for if you wanted to sync up, say, a Multimoog to a Roland sequencer.
Responding to this, mags like Music Technology were always ready with a useful article and circuit diagram to help their readers overcome the problems. But as far as I know, Korg were the only commercial manufacturer at the time to put together a dedicated interface box so that Korgs could sync to Moogs and/or Rolands. That was the MS02, which used to cost a bit under £100 - but probably didn't sell well, even at that price.
Gradually, over the next few years, the Roland/Arp CV/gate standard became the accepted norm in mass-produced equipment and was adopted by the majority of manufacturers of synths with arpeggiators. Then Roland and Korg introduced a new system of synchronisation. This was DIN sync, which foreshadowed MIDI by its use of five-pin DIN plugs, and went some way towards being a reliable and effective synchronisation system - except that Korg chose 48 beats as their time-base, while Roland opted for 24!
In any case, most other manufacturers didn't go along with DIN sync, and stuck with positive trigger voltages for their sync signals. This wouldn't have been so bad had there not been even more variation in the time-bases that they chose. Drumulators and MemoryMoog Plus sequencers, for instance, used 24 clocks per beat, while Linn and Simmons used 48 and machines like the Fairlight used 384! In practice, you could sync a Fairlight to a Drumulator - but unfortunately, the Drumulator pattern would finish in a sixteenth of the time of the Fairlight's! Great for bizarre speed machine effects, but not much else.
It was obvious that something needed to be done, and in the early eighties a Californian firm called Garfield Electronics stepped bravely in. They produced a box which converted all these different time-bases, and DIN sync signals, so that everything could talk to everything else - and at the same speed, too!
Their first product was called Doctor Click. Buy this rather chunky piece of gear, and at a stroke you could solve practically all your sync problems, and could also set up some amazing cross-rhythms and syncopations. Even awkward customers like the Oberheim DSX, and Roland's MC4, CR68 and CR78 fell instantly into step - well, they did once you'd hunted through the manual to find which timebase to select.
There were two channels available, which was at least better than one; and although there was no DIN sync input, the potential was there for elaborate manipulation of trigger timing with things like a pulsewidth variable Gate Out and a Time Lag Out - as well as invert and mask facilities, and a whole lot more besides.
The trouble was, such sophistication didn't come cheap. In Britain, a price-tag of almost £1500 meant that only top studios and programmers could really consider them. So the next move was to strip out the more esoteric elements of the Doctor Click, re-package it in 19" rack size, and sell it for half the price, under the name, 'Mini Doc'.
The Mini Doc came out in about 1984, and proved another success; although you lost much of the amazing versatility and creative possibilities of the original, you actually gained a great deal in ease of operation, and most importantly, in the ability to run virtually any number of different sequencers, arpeggiators and drum-machines at the same time.
The Mini Doc had individual outputs for each of the time-bases, so you could keep all your gear permanently wired to it, and run it all simultaneously. As a bonus, you could still produce superb arpeggiator polyrhythms with the Invert switch and 11-position Rotary Clocking Rate control - triplets included.
But Garfield Electronics hadn't yet finished the process of miniaturisation and simplification and went on to produce a third member of the 'Click' family not long after. Somehow they missed out 'Micro' and went straight to Nano Click - the real baby of the bunch - with an external power supply, and only four output and four input sockets - each handling DIN sync, 24, 48, and 96 pulses per beat.
All three products are solidly made, well-conceived, and are currently going for a fraction of their original selling price - if you can find one. But, as the astute amongst you will have spotted, they have one little omission: no MIDI. You're going to need another box to make the crucial link from (or to) MIDI.
Candidates that spring to mind are the Kenton box already mentioned: the excellent Roland MPU101 MIDI/CV interface, which can be configured to send a trigger in time with MIDI (but then, unfortunately, loses all its brilliant CV capabilities); the new dBm EXCV, and Korg's more elderly KMS-30 MIDI synchroniser.
Just as in a previous era, Korg were the only major manufacturer to produce a much-needed interface; the MS02 had been unique in the late seventies, and in the mid eighties they did it again with the KMS-30. But before I get too lyrical about it, we need to look at the circumstances which led to its development and which certainly didn't merit quite so much praise.
About this time, Korg produced a series of mould-breaking products all within a few months of each other: the world's first genuinely cheap polyphonic synthesiser, and the world's first genuinely cheap digital drum machines - the DDM 110 and the DDM 220. I remember them well.
After a break from buying any new products because they seemed (and were!) so expensive, I splashed out on a Poly 800 and a DDM 220. Surely this was going to be it: a complete music system which would revolutionise my music without breaking the bank. It came as a bit of a shock to realise that the Poly 800's sequencer was MIDI, but the DDM 220 was DIN sync, and that they had no intention of talking to each other.
In fact, I was so pissed-off by this that even when the KMS-30 became available shortly after, I couldn't find it in me to spend the £140 to sync the two instruments up. A shame, really, since the KMS-30 is one of only a handful of products to be worth almost as much now as it was new. The reason is simple. If you want to synchronise MIDI to DIN sync or vice versa - the KMS-30 does it, no fuss, no bother. Two MIDI outs, and two SYNC outs, independently switchable between the Roland and Korg standards provide all the control that is needed. Of course, you need something like one of the Garfield boxes to talk to anything that doesn't use DIN sync, but the KMS-30 still provides the vital link.
There are other ways, though. There are just a few products from the early days of MIDI whose manufacturers were considerate enough to spare a thought for the past as well as the future and were prepared to face the extra expense of including pre-MIDI connections in their MIDI machines.
Prime example here is the legendary Oberheim Xpander, which could respond to CV/gate info as well as MIDI, and do it multi-timbrally - and simultaneously. Trouble is, as an expander rather than a controller, it wouldn't be much good for our purposes. The same applies (in part) to the Roland MC4 sequencer's OP8M interface, which did include a MIDI connection, but only MIDI out. This is fine if you want to have the MC4 as the heart of your system, but there can't be many Vince Clarkes in the world!
Better for our purposes would be the Roland MSQ700 sequencer, which was probably the first industry-standard MIDI sequencer which also had the decency to include DIN sync in and out. The bizarre fact here is that these sequencers are scarcely more expensive second hand than a little KMS-30. And they do pretty much the same job and give you a limited, but great, 8-track sequencer as a bonus.
They also sync up (and send note information) via the DCB interface that Roland were pioneering before MIDI wiped the floor with it. Not a great deal of use unless you happen to have a synth like a Jupiter 8 or Juno 60, but if you do... it certainly comes cheaper than having a MIDI retrofit.
Any other ways of synchronising on the cheap? Well, I see no reason why any drum machine with both MIDI and DIN sync ins and outs shouldn't at least do the basic job of getting MIDI to talk to the previous generation of equipment - and that list includes SCI's Drumtraks, Yamaha's RX11 and RX15 and Roland's TR909, TR707 and TR727 drum machines - though of course you may need another box to turn the DIN sync into the right trigger pulses to control a particular piece of gear. In the vintage sequencer locker, so do the Roland MSQ100, a stripped-down MSQ700 without DCB, and the French sequencer from 1985, the Poly-Midi 1.
Returning to dedicated sync boxes, the Roland SBX80 SMPTE/sync box has both MIDI and DIN sync connectors, and also a fully variable trigger out socket. And its little brother, the SBX10, may miss out on the SMPTE functions, and the totally variable trigger setting, but does everything else really well.
There are perhaps only two pieces of gear that can rival the original Doctor Click and Mini Doc for versatility and comprehensive facilities - and beat them by including MIDI. One is the Doctor Click 2, which came out in 1985 specifically to cater for MIDI. In its 2U rackmount form, it has very much the same implementation as the Mini-Doc (rather than the Doctor Click), with a fantastic range of input and output connectors. Most of the inputs are, fairly logically, on the back where they are most likely to be of use in permanent installations. The outputs, on the other hand, are on the front where they can be readily changed.
The other piece of gear is the Bokse US-8, which came out in 1986, and was apparently made in Britain despite its foreign-sounding name. This unit really justified its title of Universal Synchroniser; I remember it as an excellently designed 1U 19" rackmount, with an air of solidity, and five inputs and nine outputs mounted (for accessibility) on the front panel. This made it of particular use in the studio where it could be quickly and easily set up to provide a range of different sync requirements. Certainly, it ranks equal first with the Doctor Click 2 if you want to make use of pre-MIDI sequencers, arpeggiators and drum machines.
Myself, I've still got a lot of time for the original Doctor Click and Mini-Doc - maybe because if you are going to sync up that cuddly ol' analogue stuff it's nice to have something similar to do the job. That still leaves the MIDI connection to worry about, and although the KMS-30 is small, neat and efficient, I think I'd go for the MSQ700. Sometimes being interesting is as important as being efficient.
Feature by Peter Forrest
Previous article in this issue:
Next article in this issue: