10

Were the synth parts actually played during recording (in general) or were they sequenced? I am thinking particularly of early electronic music e.g. John Foxx's Metamatic, as obviously sequencers were very common in late 80s.

40

'In general' is pretty hard to nail down.

There is so much to say about this era I could write a book - perhaps I should ;-)


Late edit - Let me add a quick TL:DR
Hardware sequencers were such a pain to programme is was simpler just to play it most of the time. One good bit of diddley-diddley sequence could make the whole track sound "electronic". It didn't all have to be sequenced.

Good example - Japan - Quiet Life One diddley-diddley sequence [probably actually an arpeggiator] a decent drummer & bassist & you have "electronic music", which you can play live.

Another one - & I don't know this for certain, but it sounds to me like 'played live over drum machine' - even the bassline. That kind of bassline you just learned to play in time, even if it took two hands. New Order - Blue Monday


So, some brief, rather rushed highlights - mainly just from memory. This was my era - I knew some of these people, I worked with some, I borrowed & lent gear with some.

  • Kraftwerk sequenced for sure, ARP & Moog Modular systems, but they also played over those sequences.
  • Eurhythmics famously had an early Movement MCS Percussion Computer but also played over them.
  • Human League, certainly up to Travelogue was mainly played, sometimes over simple drum machines. From Dare, of course, they had the Linn & a Roland MC8
  • OMD's early attempts were quite badly-played live drums & bass over a repeated 'hi-hat' synth on eternal repeat. See Electricity on youtube, the hat/shaker comes in with the full riff & the tracks sways in & out of time with it after that.
  • Gary Numan, like John Foxx, was pretty much played live, over real drums & bass, same as early OMD.
  • Ultravox, again, live over drum machines mainly, with the occasional sequence.
  • Depeche Mode... & many more - Live over drum machine.

Sequencers were a bugger ...erm.. difficult to program back then. I once spent the entire night - while everybody else went out to a pub, restaurant & to see a band - just programming up a bassline on some press a note, press a duration, press a space... nasty box [It might have been a TB303 or similar...can't remember now what it was, never wanted to do it again]

There was also the downside that these damn things used FSK time-code which meant you had to start the tape from the top every time, or they didn't know where they were. Sometimes the code reader would glitch & you'd have to start over. It was vital that you checked your drum machine track would repeatedly sync to it before you moved on.

It was only when Notator & Pro-24 came along on the Atari [& for the first time we had accessible SMTE code hardware which meant we could start the tape anywhere & the machines could catch up] that we finally had proper tools to work with. You could either click it in with a mouse, or play it & fix the bum notes afterwards. It was absolute heaven by comparison.

Even in 83/84 it was still mainly played. We had the Linn [with its famous 'turn the tape over & delay the timecode' method of getting the snare to play in time] , the Drumulator etc, but Midi was in its infancy & none of the manufacturers had ever really agreed on the CV/Gate protocol that was its precursor - that was something most people didn't have time to mess about with & even when you got it all hooked up it would drift over time.
The first DX7 I used in the studio didn't even have Midi. Even when stuff did have midi, sequencers were still nasty boxes that took too long to program, so most times it was just easier to play it.

One thing about the 80's & drum machines - by heck you learned to be able to play in time, or you didn't last long. they were totally unforgiving.
There was a lot of 'bounce it against the echo' in those days, where you & the echo would go simultaneously to tape - that's unforgiving squared.

Further waffling, just cos I can & in case anyone is interested...
In the 'early days' - for me 79-82 [I wasn't so early compared to some] - the entire studio would be a nightmare of bad filing.
Every synth sound you need to to recall for later was scribbled on a cheat sheet... either a hand-drawn representation of all the knobs & dials & where they were pointing, or later some photocopies of one carefully-made 'master sheet' with pencil marks. It would have the song title & a cue as to which noise it was making. Each song would have a stack of these, for each sound, for each synth, filed with first the demo master & then later again with the 24-track.

The Rhodes Chroma was my first ever synth with recallable presets; it was a good few years before they all did. Some of them you couldn't even get the same sound back again if you set up exactly as the paper scribbles seemed to imply. They just never sounded the same twice.

Sometimes you didn't have time to note on paper.
I recall one session; we found the sound we needed for a line, we had one track available to put it on. We had 4 un-connected synths to make up the sound.
The final record [a #1 hit, btw] was two of us, both hands, 4 synths, playing the same line over & over down the length of the track, in approximately the right places, just so we had more on tape than we needed, in case the 12" edit needed it later.
We then just broke it all down & on to the next bit. I doubt anyone could reassemble that 4-synth patch ever again. Frankly, I couldn't even tell you what the 4 synths were any more - probably Jupiter8, Mini Moog, Prophet5... & who knows.

  • 1
    Are you SURE you had a DX7 without MIDI? – Laurence Payne Aug 19 '17 at 12:49
  • 5
    @LaurencePayne - Yes, it was a prototype. Friend of mine was a "friend of the company"; they used to hand out precious pieces of equipment to people who might go on to say good things about them/use them on Top of the Pops etc. I was at the time very well-placed for blagging gear off people; we used to gear swap... my Chroma for your PPG/Emulator etc. Nobody owned them all, we lent them around. – Tetsujin Aug 19 '17 at 14:20
  • 8
    Please, write this book! Your first order is here. – bishop Aug 19 '17 at 18:56
  • This is a fascinating answer! – Matt Jones Aug 21 '17 at 9:52
4

Vince Clark famously hated the 'sloshy' timing of early MIDI and although he used to compose sequences using the UMI sequencer running on a BBC-B micro computer, when the sequences were complete, he'd download them into a hardware sequencer and use that to sequence the sounds for the actual recording. I read that he did a trick with the time code where he could nudge it backwards and forward, using an oscilloscope to make the instrument attacks occur simultaneously.

I've also read interviews with Human League members where they said one major problem they had in the studio was time syncing non-MIDI equipment (Synclavier and Linn Drum, I think) - I seem to remember them mentioning the Garfield Time Commander.

Also, I remember reading that the A-Ha hit 'The Sun always shines on TV' was sequenced on a BBC micro.

And I was surprised to learn that the Soft Cell recording of 'Tainted Love' despite being written on low budget synths was 'helped' by a Synclavier in the studio.

  • I never did get to use a Synclavier, or a Fairlight :( Only the big boys had those. I once saw Mike Oldfield's rig, but never got to use it. Re: Midi timing, it never did improve. I could write a whole new chapter on how even in the 90's we were getting around that... & probably another on the myths of 'which hardware devices would play it tighter' [which were all, of course, complete tosh ;) – Tetsujin Aug 20 '17 at 6:32
  • I wonder if there would have been any difficulty adding meta-events for "defer update" and "apply update", such that if a synth saw a defer-update command, it would attempt to buffer incoming data until it saw an apply-update command or gave up waiting? That should have been simple to implement, and could have cleaned up timings a lot. – supercat Aug 20 '17 at 19:08
  • These things were written on washing-machine chips; they didn't have the room to be smart. It was a straight serial protocol, no time-shifting allowed, just handle it as fast as you can. When we were all sequencing at 96ppq that didn't leave much room to manoeuvre; by the time we got to 480ppq, you could time-shift to allow the protocol to appear to handle it all in time, by moving data early/late - & of course, moving any sysex & non beat-essential data away from significant points... all by hand. – Tetsujin Aug 20 '17 at 19:30
  • To add a little background to that sweeping generalisation, by 1992 I'd left the 'music biz' & was instead working for the R&D Department at Yamaha. I'm not a coder, but my colleagues were - they were actually coding the very chips of which we speak. – Tetsujin Aug 20 '17 at 19:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.