Computer music is not a new idea. IBM 1620 programmers would put an AM radio next to the computer, then write programs to produce "musical" bits of electrical noise. Later efforts used tape drives and/or printers for percussion. Some machines, like the CDC 3150, even had console speakers. Although this was all great fun, no concert-quality music was produced.

Fortunately, many computers now have reasonably good audio technology. It is not uncommon to find workstations capable of producing CD-quality audio. A few hundred dollars will add a CD-quality sound card to any PC. Not surprisingly, some pretty interesting freeware packages are emerging to take advantage of this capability.

Before we look at sources of music freeware, however, some mild words of warning may be in order. Music-capable workstations are relatively new to the scene. There are lots of standards to choose from, usually tied to particular types of machines. Finally, much of the code is written in C++ or Common Lisp. Unless you have the machines and languages the code requires, you are unlikely to get much joy from a package.

CCRMA

Stanford's Center for Computer Research in Music and Acoustics (CCRMA) brings in researchers from all over the world. Their FTP archive ftp://ccrma-ftp.stanford.edu/" is a gold mine of audio and music software. Their hardware environment is based on NeXT and Macintosh computers, and their software tends to be based on Common Lisp, Objective-C, and Smalltalk.

The pub directory contains dozens of compressed archives, as well as topical directories containing still more material. Pick up and print out CCRMA.README and CCRMAReport.ps.Z, read them, then start grabbing goodies. To tweak your interest, here is a precis from the latter document:

The Stanford Center for Computer Research in Music and Acoustics (CCRMA) is an interdisciplinary facility where composers and researchers work together using computer-based technology as a new musical and artistic medium, and as a research tool.

Areas of ongoing interest at CCRMA include: Applications Hardware, Applications Software, synthesis Techniques and Algorithms, Real-Time Controllers, Signal Processing, Digital Recording and Editing, Psychoacoustics and Musical Acoustics, Applied Pattern Recognition and Artificial Intelligence, Music Manuscripting by Computer, Composition, and Real-Time Applications with Small Systems.

MAEstro

MAEstro is a "Multimedia Authoring Environment" under development at Stanford University. It is written in C and thoroughly documented. Unfortunately, the current version has gone commercial and the old version no longer appears to be available via FTP. There is hope, however, that some site will pick it up. The old distribution was described as follows in the ReadMe file:

MAEstro is a distributed multimedia environment built to focus on student authorship of multimedia documents. Authorship is distributed among a number of "media editors" -- applications responsible for a single computer-controlled medium (e.g., CD audio, text, etc.). MAEstro was designed as a distributed environment to allow for quick integration of new media as they become available for the workstation platform. MAEstro is built around an inter-application communications protocol that allows programmers to add new capabilities to the MAEstro environment without disturbing any of the existing components.

Princeton

The music directory on Princeton University's FTP archive ftp://princeton.edu/pub/music/ contains over 14 MB of compressed archives. Once again, there is a heavy NeXT influence. Here are some sample package descriptions, ranging from the esoteric to the bizarre:

Audio Challenger randomly generates ascending and descending melodic and harmonic musical intervals which can be used in assisting music students in trying to improve their ability to aurally identify musical intervals. Audio Challenger features real-time synthesis on the DSP (digital signal processing) chip of the NeXT computer which gives it the advantage of a more natural and "lively" musical timbre than ear-training programs that currently exist on other platforms. Audio Challenger is released as FREEware to the internet archives by the researchers and students of DREAMS: Digital Research (in) Electro-Acoustic Music (at) Skidmore College.

MadonnaMorph.snd is a sound produced with a signal interpolating processor that I have created called: "turpentine". "turpentine" performs interpolated transitions or "morphs" between sounds. This particular sound was created by performing a linearly rising sweep replacement of frequency bands between an abstract formant collage and a short Madonna sample.

Hyperupic is an image to sound transducer. That's right, with Hyperupic you might be able actually to hear Whistler's mother. I think that he did all the whistling, actually. Meanwhile, you need to know that this implementation of Hyperupic only accepts 24-bit RGB TIFF images as transduction sources. Now don't turn blue. You can still scribble with Icon, save the masterwork as a 24-bit alpha-free image (even with JPEG compression if you like!) and Hyperupic will transform the image into a soaring melody. ...

Hyperupic is free. Give it to your friends. Show it to your mother. I won't make you feel guilty (yet) for using my software. You can even claim that you wrote Hyperupic yourself! If you do this though, you might never have sex again. You are a glue gun. Hyperupic has on-line infotainment, including documentation. Hyperupic is also filled with other pomo gimmicks and non-sequiturs that make specialized programmers tense and feisty. Relax, you'll retire early.

UCSD

MIDI enthusiasts should check out Brian Kantor's collection (/midi) on the University of California San Diego FTP archive (ftp://ucsd.edu/midi/). The collection includes software for Amiga, Atari, Commodore 64, MS-DOS, Macintosh, NeXT, and generic UNIX. It also contains MIDI scores for several pieces of music. Brian's README gives a pretty good rundown on his efforts, ending with this gem:

A further note on SHAREWARE: if the program is functional and the author is hoping that people will send him some bucks out of gratitude, or to get updates, etc, I don't have a problem with that. On the other hand, if the shareware version is a crippled version of the program so that it's really a thinly-disguised advertisement for the real thing that costs real money, well, we're not in the advertising business and I really don't want that kind of hucksterism on this system, so don't send those kinds of programs here. For example, when someone gives away an early version of a program with a note that all future versions will cost money, that ticks me off. It's one thing if he decides later to sell his brainchild, but using the old versions as come-ons is slimy.