Serveur © IRCAM - CENTRE POMPIDOU 1996-2005. Tous droits réservés pour tous pays. All rights reserved.

Abstract

This paper will describe some of the problems encountered in the design and implementation of SYN4B, a programming language for control of the 4B digital synthesizer. These problems are considered from the point of view of a composer, not a computer scientist. It is my hope that other composers working with similarly sophisticated hardware and software may be able to more thoroughly utilize and control their individual computer music environments by gaining some insight into just what sort of decisions and compromises must necessarily be made in the design and implementation of software for a real time computer music system.

I. Introduction

While in the course of my description I will outline many of the features of SYN4B, I do not intend for this paper to be used as a manual for the use of the language, nor as a documentation of the language. Instead, I will focus upon the problems presented in designing a system which provides a maximum of flexibility and convenience for the composer, while at the same time allowing access to the synthesizer itself at the lowest level possible. The problems faced in the development of SYN4B grew out of the conflict between the desire for an ideal system and the necessity of implementing that system on a particular hardware configuration. The specific points of conflict in any such project depend necessarily upon the designer's concept of an ideal system and the specific hardware available. The problems involved in the design of a nonreal time system or a hybrid digital-analog system would, of course, be different from those described here. However, I believe that users of any system may benefit by seeing the kinds of problems which are encountered in the design and implementation of a language such as SYN4B. Other composers who find themselves in a similar position of responsibility for system development may be able to benefit from some of our solutions to specific problems which we encountered. At the same time, I would hope that any user of any system might gain some measure of understanding of the causes of the frustrations and annoyances he encounters by following the progress of the problems encountered in the design and implementation of a computer music system, and the many compromises necessary in order to reach workable solutions to those problems.