Music V: Back to the Future


In 1969 Max Mathews published The Technology of Computer Music in which he provides a  primer to digital sound and synthesis, describes the function of a computer music program, and provides a manual to the language he was describing, namely MUSIC V. It was to be the last MUSIC-N language he would write, and was the culmination of audio programming innovation that Mathews began in 1957 when he created the series of programs that formed MUSIC.

A Brief History

1957 could be seen as a sort of “Big Bang” moment in computer music.  The MUSIC series of programs would see constant improvements and added  functionality over the following decade that would influence computer music programs to the present time. As an educational and historical document The Technology of Computer Music is invaluable as it gives a first-hand account of what was required to create computer music in the first decade of its existence.  Thus, The Technology of Computer Music shines a light on the vast distance since traveled by computer technology in general, and specifically computer music technology. To overstate Max Mathews’ contribution to computer music would be difficult. Undoubtedly, programmers and musicians in the 20th century would have created computer music programs something like those we have now, but without Max Mathews the landscape would look vastly different and the timeline would almost certainly be longer. (The complete The Technology of Computer Music is linked to at the end of this post. )

While interaction with MUSIC V and the computers on which it ran was crude and time consuming by today’s standards, the MUSIC N series had great influence on the computer music programs written in the decades following, including programs like MUSIC360, Csound and CMix would borrow elements of linguistic syntax and functional design. The basic function of the earlier MUSIC programs was described by Mathews in 1963 in the journal Science.  Some of the material from that article was revised and included in the first chapter of The Technology of Computer Music.  Here, Mathews describes the basic workflow of creating music with an IBM 7090 computer in the Bell Laboratories.

In the first part the composer specifies, in computer language, the characteristics of a set of musical instruments. The program unit that represents the instrument (or the “instrument unit”) may be as simple or as complex as he desires. He then prepares a score consisting of a list of notes to be played on the instrument-units he has created. The samples of sound wave are generated by putting the score, in a form the machine can read, into the computer, together with the instrument units, and turning on the computer. The numerical output is recorded on a digital magnetic tape for subsequent conversion to acoustic form.1

Mathews then goes on to describe what would arguably be the most enduring component of his language, the unit generator.

The interconnected blocks of program which make up the instrument unit are called unit generators. There are a number of different types of unit generators, each of which has a specific function.2

By stringing together a few unit generators one could achieve complex and interesting sounds with a relatively small amount of code.  The instruments that Mathews wrote for the program along with additional instruments written later by Jean-Claud Risset (the “Risset Catalog”) provided a wealth of sonic options for composers who wanted to compose music with the computer but were not able to program at advanced levels. The result for composers was that they were able to create scores with less effort and less code that were easier to read.

IBM 7094 pic7094-ibm

The IBM 7094 Mainframe

MUSIC V was different from the programs Mathews wrote before it. Computers of the 1950s and 1960s occupied entire rooms at places like Bell Labs and computer time, because of the huge cost of computers, was extremely valuable and limited.  Prior to MUSIC V, the MUSIC programs were written in different assembly languages and were tied to the systems on which they were developed.  The new MUSIC V was almost exclusively written in Fortran which meant that it was easily ported to other systems.  One result was that MUSIC V was exported to numerous computer music centers around the world in the early 1970s.3 Another was that working with music was easier for anyone familiar with programming as Fortran was a common programming language by the late 1960s.

Soon after writing MUSIC V, Mathews began work on a real-time music system he would call GROOVE. This would mark the end of Mathew’s work on the MUSIC series. As mentioned above, however, the spirit was carried on in a number of programs and is is alive today, albeit in a highly modified and easier-to-use form.  Fortunately for us, one can still experience something of the wonder and frustration of working with those early programs.

MUSIC V : Back to the Future

Let’s quickly place MUSIC V in a perspective timeline.  It was developed 15 years before MIDI, 27 years before the mp3, 33 years before the iPod, and 45 years before the posting of this article. In short, it’s a fair guess to say that the reader does not use any piece of audio equipment or technology that is older than this program with the possible exception of LP records. Fret not, however, MUSIC V is still alive… (sort of…)

In 2010 The Audio Programming Book4 was published by MIT Press and was released with a DVD of additional materials.  On this DVD is an updated version of MUSIC V capable of running on modern operating systems.  Here are a couple renderings from the included scores score3 and score513.

I put together a folder with both MUSIC V and a copy of sox you can use to listen to / convert the raw audio produced by MUSIC V.  You can get it here.

Once you have the folder you will need to grab gfortran from  here.

If you are using Mac OS, choose the appropriate Binary from this page.

If you have a Linux distro, grab the appropriate gfortran compiler for your distro using your package management system.  You can also download it from the above gfortran site.

Building MusicV

‘cd’ into the MusicV folder (with the Makefile) and execute:

$ make


This will create pass1, pass2 and pass3 in the same directory.

To run an included example score through music V execute the following in a Terminal:

$ ./ score513


or Make a score

The included docs folder has chapters 2 and 3 from The Technology of Computer Music where instruments are defined and instructions for use provided.

1 INS 0 1 ;
2 OSC P5 P6 B2 F2 P30 ;
3 OUT B2 B1 ;
4 END ;
5 GEN 0 1 2 0 0 .999 50 .999 205 -.999 306 -.999 461 0 511 ;
6 NOT 0 1 .50 125 8.45 ;
7 NOT .75 1 .17 250 8.45 ;
8 NOT 1.00 1 .50 500 8.45 ;
9 NOT 1.75 1 .17 1000 8.93 ;
10 NOT 2.00 1 .95 2000 10.04 ;
11 NOT 3.00 1 .95 1000 8.45 ; 
12 NOT 4.00 1 .50 500 8.93 ;
13 NOT 4.75 1 .17 500 8.93 ;
14 NOT 5.00 1 .50 700 8.93 ;
15 NOT 5.75 1 .17 1000 13.39 ;
16 NOT 6.00 1 1.95 2000 12.65 ;
17 TER 8.00 ;

Running the script on whatever score file you choose will write a file called snd.raw to the same directory you run the script from.

Play it

Once you have your snd.raw audio file you can hear it one of two ways:

1. Play the sound file using ‘play’ (included with sox) by navigating to the sox folder and running the following:

$ play -r 44.1k -e signed -b 32 -c 1 ../MusicV/MusicV/snd.raw


To convert the raw file to a wave file:


sox -r 44.1k -e signed -b 32 -c 1 ../MusicV/MusicV/snd.raw outputname.wav


2. Import the raw audio into Audacity

  1. open Audacity and choose File->Import->Raw Data…
  2. Navigate to the folder with your audio file and select the file
  3. Make your settings as below (at least on Apple OS)


You should then be able to both see and hear the glory that is Music V.  You can export your efforts to another format using File->Export and choosing your desired format.

Something of a Conclusion (from practical experience)

While not nearly as  laborious as working with MUSIC must have been in the 1960s, creating music with this program is hard.  It’s hard to work with,  you actually have to know and remember things (both technical and artistic), some things don’t work, and the program does almost nothing for you short of provide the pieces and the glue. However, I find the process of editing the score file, running it through MUSIC, getting errors, trying again, getting crappy results and so on delightfully refreshing. The enjoyment is in the trying, and in trying to make music with MUSIC V you will almost certainly gain in appreciation for those early pioneers who toiled endlessly to create new paradigms and the resulting programs and methods of interaction we enjoy today, to say nothing of the fantastic music that simultaneously resulted.  So my exhortation to you: get compiling!


1. Science, New Series, Vol. 142, No. 3592 (Nov. 1, 1963), pp. 554
2. ""
3. Roads, Curtis. The computer music tutorial. Cambridge, Mass.: MIT Press, 1996.
4. The Audio Programming Book

Mathews: The Technology of Computer Music (complete)
[scribd id=140253767 key=key-jdc42ni3dmv3e1szy11 mode=scroll]