Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

David vs Goliath?

IRCAM Computer Music

According to composer Tim Souster, the aisles were certainly full of noises at October's symposium on personal computers in music held at Paris' distinguished music research centre, IRCAM. Read his report.

Tim Souster, composer of everything from electronic string quartets to TV commercials, lays aside his Parisienne restaurant guide for long enough to report on a recent symposium on personal computers in music held at France's premier music technology centre, IRCAM.

Wired for sound: Michel Waisvisz steals the IRCAM show with a gloriously idiosyncratic demonstration of his MIDI sonar 'Hands' system.

Ah, Paris, city of a thousand odours, of exploding supermarkets and designer muggings; where the patisseries, and indeed many of the pavements, abound in a bewildering variety of colourful tarts... PLUS, "that'll be 15% service charge and a tip to you m'sieur, or I spill this Muscadet all over your MIDI specification".

For, yes, the weekend was not to be all pleasure and getting ripped off in restaurants. I persuaded the chaps at SOS that this symposium was to be a very serious business indeed, which the broad masses of UK synthesists would certainly part with folding money to read about.

"What symposium?" they countered, bleakly.

"Oh, the one about Systèmes Personnels et Informatique Musicale of course" I informed them nonchalantly.

"Ah" they gulped. "That symposium."

And off I flew.


But enough of this idle banter. This international get-together was organised in October by those awfully nice IRCAM people in order to allow the participants to get to grips (at length and in depth) with recent developments in the use of personal computers in musical contexts.

Many papers on all aspects of the subject were given by leading experts from all over the world. But before telling you how, in the very first paper, Mesias Maiguashca (an Ecuadorian of Indian descent, now living in Metz and Baden-Baden) described his transformations of Mandelbrot's mathematically-generated visual images into sound, I should like to summarise to you, dear readers, the situation as I see it.


It may not fully have been realised in this country, but the rest of the world is awash with Macintosh computers, now challenged of course by the Atari ST. (In Paris, the majority of systems involved MIDI control of TX816s by means of a Mac.) But the key to all this is that personal computer systems in general are becoming as fast and as powerful as the mainframe systems on which the most powerful music-programming languages have been developed until now. (And let us not forget that research and development work on the Frequency Modulation algorithms now enshrined in Yamaha's DX keyboards was carried out in years of toil on the hugely expensive mainframe system at Stanford University in California.)

In contrast to these multi-million dollar installations, the most basic attraction of the new PCs is sheer price: the hardware is now nearly a thousand times cheaper than 10 years ago!

Not so the software. The conversion of mainframe music-programming languages like LISP and C is a huge and laborious task. But it is being achieved, for example by the group of York-based composers including Trevor Wishart, Richard Orton and Tom Endrich who have already implemented C-music on the Atari ST with conversion via their own buffering system into a Sony PCM-F1. This is not real-time of course (and this remains the great divide in computer music terms), but given an open, software-oriented system, with the recent developments in hard disk storage, the composer can make a PC do an enormous number of different things.

But what sorts of things are these, I hear you cry. What kind of things can a mainframe system do, which are now coming into the range of the PC? Well, take the case of the American composer David Evan Jones who is currently working at IRCAM on the VAX mainframe computer. He uses the available program at IRCAM to recreate the resonances of the human voice one minute and then instantaneously to provide 300 filters simulating the resonances of various percussion instruments. The power of the mainframe programs is frightening. You want 300 oscillators, 200 harmonisers, 100 delay lines or an eight-fold randomly-modulated full frequency sampling of a Paris traffic jam?: then just dial it up, you only have to know the code!

Significantly enough though, David found some of the IRCAM programs just incapable of doing what he wanted. So he opted to make all his sounds in Paris and then take them back to Dartmouth, New Hampshire to put them together. Why? Well, the editing program on the current Synclavier is more flexible, powerful and quick than any of the mainframe-based editors available at IRCAM.

Even so, the Synclavier should not be confused with a PC-based system (Atari ST £700; Synclavier with 8-track hard disk storage $135,000!). But it's nevertheless important to evaluate the 'middle ground' dedicated systems like the Synclavier, the Fairlight Series III and the Kurzweil - of which more anon.


Back in Paris there seemed to be more Yamaha TX816s lying around than the whole Brit contingent had probably had hot breakfasts. For example, the Yamaha system used in just one piece consisted of three TX816s, two KX88 keyboards, one SPX90, two Macs and one MCS2!

But the use of the TX816s certainly made sense in some of the music heard in the concerts. In a piece called Ur, by the young Swedish composer Magnus Lindberg, the pianist in the ensemble also plays a DX7 which is not heard in itself but is used to trigger and control the real-time tempo of two Macs driving two TX816s, one of which is tuned a quarter-tone lower than the other by means of a MIDI System Exclusive command.

'Ur' pretty well summed up my reaction when I first saw the score of this piece too, but the music itself turned out to be very tight and powerful, even though the sheer energy of the instrumental writing seemed to mask a lot of what was going on in the electronics. Only the next day when the electronics were demonstrated on their own could one start to appreciate the potential of this system: the apparent ease with which it could generate huge bursts of very fast, dense chords which moved through time and space with immense vigour.

Another memorable demonstration gave rise to much less sound. David Wessel of IRCAM, who has been doing a lot of exciting work on MIDILISP for the Mac, had the misfortune of having to demonstrate a complex live interactive system with some substitute elements in it. Sod's law intervened, of course, and all that was heard were ever more frantically repeated diminished-seventh chords. (All the while the instrumental half of this duo, the jazz sax player Roscoe Mitchell, sat quietly looking on in total disbelief.) However, all worked beautifully at the concert and the system is certainly an ingenious one.

A Fairlight Voicetracker transmits MIDI codes derived from the live sax playing to the Mac. A KX76 keyboard is also interfaced with the Mac so that the player can 'catch and throw' material provided by the sax as he chooses. The Mac controls both a TX816 and an Akai S900 which produce synthesized or sampled sounds as required.

But the plot thickens. The KX76 is not used conventionally at all. It is split and programmed so that the lower half 'catches' the sax sounds and the upper half 'throws' back the 'recordings'. Furthermore, the white keys are programmed for long-term memory, and the black for short-term memory! There are even keys which give back not just transpositions of the original, but inversions and retrogrades. One can even remove, say, the odd or even notes in a phrase or even all the C sharps!

The Samuel Beckett-like demonstration ended with a memorable quote from David Wessel, which has to be understood against the background of much former controversy within IRCAM over the dominance of keyboards in electro-music: "After all the arguments, the keyboard's not too bad," he explained, "even if you don't want to play the notes."

You should also know that all this was delivered in David's own brand of French which it would be generous to call 'idiosyncratic'. In this area he is surpassed only by Chris Yavelow of Kurzweil to whom I would definitely hand 'The Golden Frog' award for services against the French language.

Which brings us to the current state of play in the world of the dedicated systems. There are two main trends here. One well-known ex-rock star has reported that he's selling both his Synclavier and his Fairlight (I should be so lucky!) and buying an Atari. At the same time, to cite examples I personally have come across very recently, Pete Townshend swears by his Fairlight as a compositional aid. Brian Gascoigne produced/arranged all his wonderfully evocative music and effects (the distinction's academic here) for the movie The Emerald Forest on the Series III Fairlight, while in Paris Chris Yavelow revealed the frightening power of the Kurzweil 250, particularly with the Mac as a front-end.

But even the Kurzweil on its own is pretty hair-raising. To demonstrate the total flexibility of the keyboard, Mr Y played a chromatic scale from the bottom of the instrument to the top. Touches button. Same thing, only the scale comes out upside down, ie. descending, while the fingers climb. Touches another button. Same thing and a Chopin Study (both hands!) blares out.

This demonstration was certainly more Rambo than Rameau. There was a whole lot of cringin' going on amongst the American fraternity in the room. Mr Y soon had a complete synthetic orchestra on the go and was editing material frenetically between orchestral sections, spewing out musical graphic readouts, calling up different order Markov chains to vary his material statistically [This is done by means of the KIT program developed by David Levitt and Burt Sloane of MIT] and generally create sampled mayhem. All this was achieved by Mr Yavelow courtesy of the Kurzweil and the Mark Of The Unicorn 'Performer' software for the Macintosh (after announcing that the Kurzweil 250 will be Sound Designer-compatible from November).

Le biscuit was well and truly taken by the Hook-up program from MIT whereby musical input is made to control the movement of animated graphics in real time. Just watch that shark shoot around your screen driven by transpositions of the Jaws cello motif (on sampled cellos of course)!

But the ultimate star of the weekend for me was one Michel Waisvisz, artistic director of the STEIM Institute in Amsterdam. Michel is as much an actor and performance artist as he is a composer, and one can sense his individuality just from a couple of extracts from his biography. Perhaps this one is not intentionally odd: "Together with singer/actress Moniek Toebosch he formed a duo that reached a wide audience with a show that matched relativating wit with heavy drama and the exhausting power of totally improvised performance. On rare occasions they still perform together." His new project is more obviously barmy: "Waisvisz is working on a piece that will be performed in the open air and makes use of explosives and electronics. A musical vocabulary is created by choice of the required chemical compound of the explosives (pitch, loudness, timbre) and by selection of the appropriate spatial location (distance and positioning under earth or water etc). The electronics are used as detonators as well as to create a suitable accompaniment. A special indoor version is considered after achieving a sufficient amount of control over the chemical instrumentalists."


In Paris, Michel introduced his 'Hands' system (see photo) which strikes me as the most brilliantly individual use of MIDI that I have ever come across.

In each hand there are three rows of four keys which provide pitch control within one octave. Further switches give more combinations providing an 8-octave pitch range. The left hand contains a sonar transmitter aimed at the right hand, which has a matching sonar receiver. The data derived from a sonar scan of the distance between the two hands is assigned to the key-velocity value. With proper scaling the performer has, compared with a velocity-sensitive keyboard, a much wider range of control of timbre, from very close to fully outstretched.

Further buttons and switches give control over MIDI channel and FM voice selection etc. The whole thing is connected to an analogue-to-MIDI converter interfaced with an Atari ST micro which drives the sounds in a Yamaha TX816 rack. And what sounds! They are, of course, more abstract in character than would be normally used in the pop field, but they certainly dispell the idea that Yamaha FM is limited in scope - well at least not on the TX816. (Another still top-secret but long overdue resource will apparently be introduced by Yamaha quite soon: keyboards with programmable microtonal tuning.) But the greatest achievement here is to get away from the "tyranny of the single note" which has been the most negative consequence of the dominance of keyboards and sequencers over the last decade. Touch Monkeys, as Michel Waisvisz's piece is called, contains wonderfully rich sustained sounds which he uses his own visually expressive gestures to modulate, mould and transmute internally. I hope that this remarkable performance will be witnessed over here as soon as possible...


An exciting weekend then, though slightly depressing when you compare it to what's going on in this country. Moreover the four concerts were packed with enthusiastic listeners, by no means all symposium participants either. The performances seemed immaculate and the technical set-up lavish to say the least. (More equipment per concert than exists for the whole of electro-acoustic music in this country, I shouldn't wonder, and so many monstrous JBL monitors hanging from the ceiling that I half expected The Who to walk in any moment.) The presentation of the papers was excellent too, with two CCTV cameras on the stage and seven video monitors over the listeners' heads, plus another comprehensive multi-channel sound system. A final credit must go to Jean-Baptiste Barriere, head of research at IRCAM, who did a brilliant job of organising the event and translating all the papers given in English.

Ho hum, back to... Luton Airport?!!! (I flew from Stansted actually.)

Ah, Luton, city of a thousand odours....

Previous Article in this issue

Hot Stuff

Next article in this issue

Ian Boddy: Phoenix

Sound On Sound - Copyright: SOS Publications Ltd.
The contents of this magazine are re-published here with the kind permission of SOS Publications Ltd.


Sound On Sound - Dec 1986

Show Report by Tim Souster

Previous article in this issue:

> Hot Stuff

Next article in this issue:

> Ian Boddy: Phoenix

Help Support The Things You Love

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!

Donations for July 2020
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £20.00

All donations and support are gratefully appreciated - thank you.

Please Contribute to mu:zines by supplying magazines, scanning or donating funds. Thanks!

Monetary donations go towards site running costs, and the occasional coffee for me if there's anything left over!

Small Print

Terms of usePrivacy