Between The Idea And Work - Course Of Musical Data Processing

If musical data processing is essential us so muchprograms.
today, it is that it gradually created tools whichThe second effect of the arrival of the
are radically modifying the manner of thinking themicrocomputer was the design of the "mixed
music. However, its history is short. It mergessynthesis", synthesizers numerical, genuine
with the development of numerical technologies:computers adapted to the calculation of the sound
computers, initially, accompanied by creation bywave in real times, placed under the control of a
the languages symbolic intended for thecomputer. From second half of the Seventies
programming, then with a whole troop ofappear several achievements of this type; we will
inventions in numerical technology. Enough early inretain work of James Beauchamp,
its history, data processing will be shownJean-François Allouis, William Buxton, inter
sufficiently ripe to accommodate concerns of allalia, like those of Peter Samson (synthesizer of
kinds, energy of accountancy to scientificSystems Concept, conceived for the research
research, while passing naturally by what interestscenter -- CCRMA -- university of Stanford),
us, artistic creation.Synclavier de New England DIGITAL Corporation,
And it is undoubtedly there that it is necessary toconceived by Syd Alonso and Cameron Jones
distinguish what arises from data processing itself,under the impulse of the type-setter Jon
and what belongs rather to the broader world ofAppleton, design, under the impulse of Luciano
numerical technology. The music amply drawsBerio, of a bench of oscillators in Naples by
from these two fields its new resources.Giuseppe di Giugno, who continued his work in
However, since the field of the sound isIrcam (series 4A, 4B, 4C and 4X) under the
converted today into audio numerical, thedirection of Pierre Boulez; more recently, Fly 30 of
distinction is essential. Musical data processing isthe Center of recherché musical of Rome.
born from the meeting of the musical concernsLet us note that with the 4X of Ircam (1980),
and the environment resulting from numericalthe term of synthesizer disappears, replaced by
technologies and the specificity of the computer,that of numerical processor of signal, which
on the one hand, and of the scientific fields whichundoubtedly moves the accent on the general
clarify its research topics. If the musicalinformation of the machine.
composition appears there in good place,The industry of the electronic instrument does
practically all the other activities of the music arenot take a long time to adapt to these new
found there. And musical research partly coversdevelopments. The first stage consisted in
the ground cleared by data processing, acoustics,introducing microprocessors inside analogical
the treatment of the signal, even cognitivesynthesizers (Prophet synthesizers of the firm
psychology: thus musical data processing is at theSequential Circuits), charged to control the
center of several musical, scientific and technicalmodules ordered in tension; it is still of the "hybrid
fields.synthesis". The second stage follows soon: it
But it is the recourse to the specific contributionsconsists in designing genuine entirely numerical
of the data processing which characterizes itsmusical instruments. It is the noticed arrival of
step. New conceptual tools are unceasinglySynclavier II, then of Fairlight.
provided by the artificial intelligence, which areThe industrial field is today initially made up by the
concretized by languages such as Lisp or Prolog.market of the synthesizers and the processors
They are put at once at the musicologistof the sound, and by the software which makes
assistance or abstracting service to theit possible to exploit them. Today, all the
composition. Research in systems real time andsynthesizers are numerical, and necessarily meet
on the interfaces interactive makes it possible tothe Midi standard. The field of the synthesizers is
conceive new connections between thedouble: on the one hand apparatuses, often
instrumentalist and the electronic universe.provided with a keyboard, which propose a choice
Great stages of musical data processingof preprogrammed sounds which one can vary
With the origin of musical data processing, onecertain parameters by an elementary process of
finds two types of activities, independent one ofprogramming; in addition, the machines which are
the other. If these activities prudent today, it is inintended to reproduce sounds beforehand
another manner that the original vision whichrecorded and memorized, or stored on mass
caused their birth could enable to foresee. Thesememory: samplers, or "samplers".
two types of activities are: the musicalIt should be noted that all these technologies
composition, and production of the sound. In bothbecome accessible to the private musician, within
cases, the manufacture of the desired result isthe framework of what is called commonly the
ensured by the computer. These two types of"personal studio" (home studio).
activities are appreciably contemporary. The firstBut these machines, and a fortiori the personal
serious tests of musical composition per computerstudio do not function without adapted software:
go back to 1956: it is on this date that Lejarenthe sequencers control the execution of a piece
Hiller calculated a partition using rules encodes indirectly starting from a computer, the editors of
the form of algorithms on the computer Illiac I ofsound are intended for the treatment, the
the university of Illinois. It is about Illiac Suite forassembly and the mixing of sound sequences.
String Quartet, whose three movements arePrograms make it possible to write a partition,
carried out this year by the string quartet WQXR.which from now on is usually employed by the
In a famous structure, published in 1959 and whichmusical edition. Lastly, the machines can also be
supports the title of "Experimentalplaced under the control of supplementary
Music-Composition with year Electronic Computer",programmes to the composition.
Lejaren Hiller explains in detail the procedures thatThe most original character of the contemporary
it applied to the Illiac computer in order todata-processing stringed-instrument trade is the
produce the partition of his string quartet."workstation". To conceive a workstation consists
To locate this period, it is as into 1956 as Johnin gathering programs of various nature, intended
McCarthy forged the term of artificial intelligence.for the analysis or the synthesis of the sound, the
One year later, max Mathews, researcher at thecontrol of the sound or the composition. These
laboratories of the Bell Telephone, in the Newprograms are integrated within data-processing a
Jersey, a first numerical programmer of synthesis"environment" organized around a computer and
of the sound for the computer IBM 704 writes.of its peripherals, intended for the treatment of
Known today under the name of Music I, it is thethe sound on line. It is the case of the plug-in
first of a great family of acoustic compilers; acharts which, coupled to a software, make it
psychologist, Newman Guttman, generates first apossible to read "files of sound" stored on a disc,
one 15 seconds duration study, In the Silver Scale.in exchange of an order, coming for example
It is as into 1957 as the four movements of thefrom a Midi source. This system, if new that it did
Continuation Illiac for String Quartet of Lejarennot find yet a name true, is generally indicated like
Hiller are published; the same year is born the"hard" or "direct-to-disk" disk.
primitive version from the famous language ofThe musical representation
FORTRAN program (FORmula TRANslator). Let usSince the computer, unlike the electronic music of
note that during the creation of the work of Hillerstudio, claims a specification of the data, and thus
by the string quartet WQXR, it is max Mathewsa writing, the question of the musical
which organized a recording, which gave place,representation is a constant concern of the field.
thereafter, with the publication of this recording inWe will see two answers. The first illustrates a
a disc carried out in 1960 by the Bell Laboratories,step a priori compositionnelle: that of Xenakis. The
and entitled Music from Mathematics: even if thesecond, more general, is the Midi standard.
ways traced by these two inventors areIannis Xenakis innovates with the design of the
independent, it is not known as that they did notUPIC (Unit Polyagogique Informatique of CEMAMu).
cross...Conceived in the middle of the Seventies, this
From these two almost contemporary events,system rises naturally from the approach of the
the development will continue, gradually, in thesynthesis of the sound by this type-setter: within
traced directions: the composition and productionthe team which it had joined together, baptized
of the sound. We will see low the courses ofinitially Emamu (Team of Mathematics and Musical
them. But a third way is not long in appearing: it isAutomatic, 1966), and with the financing of the
born from the same observation that had madeFoundation Gulbenkian, Xenakis had made build a
Hiller: the computer is above all, at that time, adigital-to-analog converter of high quality. The
formidable calculating machine. Moreover, theUPIC represents a complete environment of
English term of selected computer indicated,composition with, in result, the sound synthesis of
before the appearance of these machines, thethe page of made up music. Become in 1971
employees charged to operate calculations. But atCEMAMu (Center of Mathematics and Musical
the same time, with a bit of fear, one spokeAutomatic) because of the creation of a place
readily at the time of electronic brains. An artistintended to shelter its research, the team joined
could not approach the computer not without atogether around Xenakis conceives a system
certain emotion, which explains without any doubtmaking it possible to the type-setter to draw on
the attraction sometimes terrifying that dataa broad table of architect of the "arcs
processing will exert on the artists of the followingtime-height", by choosing for each arc a temporal
decades. But they are two scientists who are attrajectory, a form of wave, a nuance. The music
the origin of these experiments: Hiller practicedthus is initially represented in graphic form. The
chemistry, while Mathews was an already famousprograms of the first UPIC are written for a
researcher. It is undoubtedly what explainsminicomputer Solar 16/65, connected to two
remarkable methodologies that they reflect inbodies of magnetic tapes to store the programs
place, each one on their side, and with completelyand the samples, a digital-to-analog converter, a
independent aims.cathode ray tube making it possible to post the
With the Bell laboratories, max Mathews, on hisforms of waves, but also to draw these waves
side, written in 1957 a first numerical programmedusing a graphic pencil. To hear the page which it
of synthesis of the sound for the computer IBMhas just drawn, the type-setter must wait until
704, equipped with 4096 words of memory.the computer finished to calculate all the samples;
Known today under the name of Music I, it is thethe generation of the sound is ensured by a
first of a great family. The program Music IIIdigital-to-analog converter of high quality. More
(1960) introduced the concept of instrumentrecently, the UPIC was redrawn for
modular. The model imagined by max Mathews ismicrocomputer, and functions without delay.
inspired more than one equipment of laboratoryTo represent the sound in the form of a
or an electronic studio of music that by anmodifiable image, it is the goal of the Phonograms
acoustic stringed-instrument trade. The programprogram, designed at the university Paris 8 per
offers a range of independent modules (UnitVincent Lesbros. With the manner of a
Generators), in charge each one of an elementarysonograms, the program posts the spectral
function: oscillator with form of programmableanalysis like a drawing, which can be modified; the
wave, adder of signals, multiplier, generator ofnew representation can then be synthesized,
envelopes and random signals, etc... The musicianeither by Midi, or in the shape of a file of sound,
builds a "instrument" by connecting a selection ofor even transformed into Midi file.
modules between them. The signals produced byThe reproach often today is heard that the
the oscillators or the generators are led towardsgeneration of young musicians who approach
other modules there to be modified or mixed.technology through the environment created
Several instruments can be joined together withinaround the Midi standard are not a satisfactory
a "orchestra", each instrument having its ownaware of not passed of musical data processing
identity. Contrary to what occurs in the materialand its problems. But is to forget that, in a
universe, it there not of limit to the number ofdirection, the birth of the Midi standard was done
modules usable simultaneously, except perhapswithout true filiation with the preceding stages of
the memory of the computer. The result of thethe field which one will name musical data
placement of the instrument is the progressiveprocessing. The phenomenon which represents
calculation of the sound in the form of a sequenceMidi is not at all a misadventure of this field.
of numbers which, put end to end, represent aThe Midi standard was developed in 1983 to allow
complex sound wave. These numbers are calledthe piloting of several synthesizers starting from
"samples". Today, the number of samplesone only keyboard; the messages are transmitted
representing one second of its was established toin numerical format, according to a well defined
44 100 per channel for the applications generalprotocol. With the origin, Midi is thus well based on
public, and to 48 000 for the professional field.instrumental gesturer control: it is a method to
Because of relative slowness of the machines andrepresent not the sound, but the gesture of the
design weight to be carried out, time put tomusician who plays of a Midi instrument. In 1983
generate the sound wave is quite higher than theleaves the first synthesizer to have an interface
duration of the sounds; the operation of theseMidi, Prophet 600 of Sequential Circuits. What had
programs is known as "in differed time". With thenot been defined, on the other hand, it is the
origin, the sound waves calculated in numericalsuccess which this standard was going quickly to
form were stored on a numerical tapegain, which today is used to inter-connect all the
progressively proceeding end of an arithmetic unitmachines of a studio of electronic music, and
of samples. This mode of production of the soundeven the sets of lights of a scene.
is called "direct synthesis". Thus creates for itselfMusical research
a "file of sound"; once completed, the musicianThe work undertaken since 1956 by Lejaren Hiller
calls upon a second program, charged reading thefor the composition of Illiac Suite for String
file of sound in real time and with sending theQuartet marks at the same time the true birth of
samples towards a digital-to-analog converter,musical data processing and the anchoring of this
which is connected to an amplifier andfield in the research, applied in this case to the
loudspeakers.automatic composition. The computer appeared
To activate the orchestra, the musician mustthen as a machine making it possible to treat the
write a "partition", in which all the parameterscomplex continuations of operations which
claimed by the modules of the instrument arecharacterize the composition of ambitious musical
specified. This partition is presented in the form ofworks. This way was going to be reinforced since
a list of numbers or telegraphy codes, each "note"1958 by the French type-setter Pierre Barbaud,
or a each event being the subject of a particularwho melts in Paris the Algorithmic Group in
list. These lists are ordered in time.connection with the company Bull-General Electric
But to specify each parameter is a difficult task,and begins his research of automatic composition;
more especially as the musicians are not trainedas of the following year, the first algorithmic work
to give values measured to sound dimensionsof Barbaud was made up:
which they handle. To fight against this obstacle,Unforeseeable innovations (Algorithm 1), with the
of the languages of assistance to the writing ofcollaboration of Pierre Blanchard. The program
partitions were conceived; most known is theMusicomp de Lejaren Hiller and Robert Baker, of
Score program of Leland Smith (1972). Score isthe same time, conceived for the Illiac computer
not an automatic program of composition: itafter the composition of the Illiac Continuation,
makes it possible to specify the parameters usingmaking university of Illinois one of the centers of
terms resulting from the musical practice (heights,musical data processing at that time. And when in
nuances, durations), automatically to calculate1962, Iannis Xenakis creates ST/10, 080262,
changes of tempo or nuances, and even towork written thanks to the stochastic program
supplement sections with notes corresponding toST developed since 1958 on a computer IBM
a trajectory given by the type-setter.7090, the composition using the computer enters
The model instrument-partition was firmlyits golden age. In the Netherlands, Gottfried
established with the arrival of Music IV (1962).Michael Koenig writes in 1964 the program of
This program were born from many alternatives,composition Project I (1964), followed soon of
of which some exist indeed still today. AmongProject II (1970). The composition
these misadventures, let us quote Music 4BFcomputer-assisted rests on mathematics and the
(1966-67), there is nowadays a version forstochastic one, drawing largely from the
Macintosh (Music 4C, 1989), and especially Musicresources of the processes of Markov (Hiller,
360 of Barry Vercoe (1968); this descendant ofBarbaud, Xenakis, Chadabe, Manoury).
Music IV has as a characteristic to be presentedWith the arrival of the micro-computers a new
in the form of a true programming language,tendency develops: assistance with the
which undoubtedly explains why it became todaycomposition, then computer-aided design of
with C-Music the acoustic compiler more used. Itcomposition (CAD). The program demiurge, able
was initially adapted to minicomputer PDP-11 fromto generate a whole composition the model of an
DIGITAL in 1973, then, entirely rewritten inenvironment of data-processing tools charged
language C in 1985, it took the name of C-Sound,succeeds to deal with precise musical problems.
and was quickly adapted to all kinds ofLet us quote among the principal ones: HMSL
data-processing platforms, including the(Hierarchical Music Language Specification, 1985)
micro-computers like Atari, Macintosh and IBM. Inwith Mills College in California, Forms, created by
1969 Music V appears, a program which isXavier Rodet, Draft and Patchwork, developed in
conceived to facilitate the musical programming ofIrcam under the impulse of Jean-baptiste Barrier,
the instruments and the partitions; nowadays,Experiment in Musical Intelligence of David Cope,
Music V is still largely employed, generally in theat the university of Santa Cruz in California. These
form of the adaptation that in made Richardprograms are open: they dialogue with the
Moore, C-Music (1980).type-setter in an interactive way, and are
The computer was also an unquestionable successconnected to the universe of the Midi devices.
in a strongly speculative field, the musicologistsExcept for M and Jam Factory of Joel Chadabe
analysis. With the eyes of the public interested atand David Zicarelli, they are structured by the use
the beginning of the Sixties, data processing, stillof languages nonnumerical, resulting from the field
rather mysterious and inaccessible, showed theof the artificial intelligence, such as Forth, and
possibility for strange musical work; inespecially Lisp, which explains why they rest not
composition, in musicology and finally, limited toon mathematics, as it was the case for the first
the laboratories Bell, production of sound. A greatgeneration of composition computer-assisted, but
musical upheaval of this decade was to comeon the formal languages and the generative
from the world of electronics, with thegrammars.
appearance into 1964 of the synthesizersReal time: computer and instrumental universe
modular, known as "analogical" since they do notThe Eighties see developing the use of the
contain numerical electronics. Conceivedcomputer in situation in concert; thanks to the
independently by Paolo Ketoff (Rome), Robertarrival of the numerical synthesizers in real time,
Moog and Donald Buchla (the United States), theor, more generally, numerical processors of sound,
synthesizers bring the response to theand languages of control real time, the conditions
technological aspirations of many musicians,are ripe to revisit this old surface of the music of
especially after the popular success of the discthe XXe century: electronic music on line (live
Switched one Bach of Walter Carlos who trulyelectronic music). In the majority of the cases, it
made know these instruments near a largeis initially a question of imagining a means of
audience. During this time, the program ofconnecting the computer and its computing power
Mathews knows adaptations on other sites, suchto devices of synthesis or treatment of sound,
as the universities of New York, Princeton orwith, if, possible, the interaction of musicians.
Stanford.Répons (1981), of Pierre Swell, by the
Another application of the computer appears withintegration of the procedures of treatment the
the piloting of analogical instruments. The machinewriting itself, showed in what the computer
generates signals with slow variation which modifybecame an instrument, integrated perfectly into
the adjustments of devices of studio: frequencythe orchestra. Following this work work appears
of oscillators, profit of amplifiers, cut-offon the follow-up, by the computer, of the play of
frequencies of filters. The first example of thisthe instrumentalist, operation known under the
system which one names "hybrid synthesis" wasname of "follow-up of partition". Let us quote the
established in 1970 in Elektron Musik Studio ofcontributions of Roger Dannenberg in the
Stockholm, foundation independent since 1969,automatic accompaniment and the languages
financed by the Royal Academy of Music, andoffering the conditions of the communication
placed under the direction of Knut Wiggen. Acomputer-instrument, those of max Mathews,
computer PDP 15/40 controlled twenty-fourinitially with the Groove system, then more
generators of frequency there, a generator ofrecently with its work on "Radio operator Drum"
white vibration, two third filters of octave,and the simulation of the rod of the chief
modulators: out of ring, of amplitude andorchestrates itself, Miller Puckette with the Max.
reverberations. The originality of the system ofprogram.
Stockholm lay in an extremely ergonomicThis is why one was in addition interested to give
operator console, with which the type-setter couldto instruments of orchestra this capacity, by
specify the parameters of synthesis by sweepingproviding them with sensors, allowing the
a panel of figures using a metal brochette.computer to follow the execution (flute,
Another studio is to be quoted: that of Petervibraphone, etc...). All musical industry is concerned
Zinovieff in London (1969), placed under thewith this tendency, although the process to be
control of a minicomputer DIGITAL PDP 8 forused is not decided yet: it will be electromechanical
which Peter Grogono wrote the language of(material sensors placed at strategic places of the
Musys piloting.instrument, conducting membranes, etc...), or will it
Another remarkable realization is the systembe necessary to have recourse to the analysis
Groove (Generated Realtime Operations Onefor stolen of the sounds themselves to know of
Equipment Voltage-controlled, Ca 1969) conceivedit the height, the spectral structure and the mode
at the Bell laboratories by max Mathews andof play?
Richard Moore. Groove is an instrument intendedThe community organizes itself
for the control of parameters of interpretation ofThe ripening of musical data processing was
a device of synthesis. In this direction, it placesaccompanied by an assumption of responsibility by
the musician more close to the position of athe community by the musicians and researchers
leader that to a type-setter or an instrumentalist,by themselves. Gradually, the conscience of
although one can consider that the electronicmembership of a field is done day. The
type-setter of music must often place himself ininternational congresses appear, followed later local
position of chief, by directly interpreting the musicconferences. The communications which are
which is made up.marked there are published in collections available
It is the middle of the Seventies which marks thefor all the community. These meetings also give
transition towards an inexorable widening fromplace to the presentation in concerts, which tends
now on from the life of musical data processing,more strongly to weld between them the
with the appearance of the microprocessor. Aconscience of a new field, with the scientific and
data-processing stringed-instrument trade willartistic components. It is the beginning of the
become gradually possible with the design of"International Computer Music Conferences"
complete computers on an integrated circuit:(ICMC). In 1978 is born an organization intended to
microprocessors. It will also be necessary that thefacilitate the communication and to help the
interface with the user improves, and that thebehavior of the congresses, the "Computer Music
punch cards by a more interactive mode of inputsAssociation", which becomes in 1991 "International
are replaced: it will be the keyboard and theComputer Music Association" (ICMA). The
cathode ray tube which will carry it.organizers seek to hold the congress one year in
The principle of the hybrid synthesis continued toNorth America, and the following year on another
be applied throughout the Seventies, before beingcontinent. These congresses see the ICMA taking
supplanted definitively by the numericala role growing in the assistance brought to the
synthesizers at the dawn of the Eighties. Thelocal organizers, like in the diffusion of the
American company Intel markets since 1971 thepublications resulting from these meetings, going
first microprocessor, the circuit 4004, which allowuntil placing from the orders of works which will
the design of genuine miniature computers, thebe carried out during ICMC (ICMA Commission
microcomputers: Intellec 8 (conceived startingAwards, 1991).
from microprocessor 8008 of 1972), Apple I,Another vehicle which welds the conscience of
Altair (1975), gathered soon under the name ofmembership of a common field is the Computer
micro-computers.Music Journal. Appeared in California in 1977, it is
The experiment musical of the Art Group andtaken again by MIT Press starting from volume 4
Data processing of Vincennes (GAIV) illustrate this(1980). The newspaper is essential like the
time of transition well. This team, founded to thereference as regards scientific publications of the
university of Paris 8 by Patrick Greussay and afield. Association ICMA publishes a bulletin, Array,
team of artists and architects, known for thewhich became a body very appreciated of
publication of a bulletin diffusing the research tasksinformation and discussion on the current topics of
in most recent art and data processing, entrustedmusical data processing. The review Interface
to the type-setter Giuseppe Englert the musicalDutchwoman, which becomes since 1994
coordination of her activities. It is Intellec 8,Newspaper of New Music Research, regularly
microcomputer with words of eight bits, orderedpublishes articles on musical data processing.
by a paper tape and a keyboard, which was usedCanada, Musicworks, guided by Gayle Young,
with the compositionally activities and as researchensures information on a broad range of concerns
on musical formalization; English synthesizersof the new musics. In France, publications of
EMS-VCS3 were controlled by theIrcam, InHarmonique, then the books of Ircam
microcomputer, via digital-to-analog convertersopen their columns with the aesthetic, theoretical
charged to provide power of order in exchangeand critical considerations which accompany
of the binary data calculated by interactivesudden appearance by new technologies in arts.