Cellular Synthesis
an author's technique of musical composition

Cellular Synthesis is - which explains the name - an author's technique of musical composition based on a specially designed cellular automaton and an additive synthesiser. The aim of the project is to enable the creation of complex and rich sound structures, controlling a large number of parameters and using the opportunities available only through computer programming - all these conditions are fulfilled while maintaining a very simple and intuitive way, susceptible to modifications in real time and using various types of equipment as controllers.

In other words, Cellular Synthesis is a concept (software, and a composer's approach) use of finite and small (and thus enabling easy to learn and use) set of variables and rules to obtain complex musical structures.

Due to the fact, that the Cellular Synthesis operates on both generative composition (score) and audio layer, shaping macro and micro leyers of the piece with the same set of rules, this concept can be seen as some kind of development of the mechanisms with which Iannis Xenakis experimented (for example, UPIC enabled graphical shaping of both the visual score itself and the waveforms of the sound oscillators - which is some analogy to some of the Cellular Synthesis features).

Also important is that the nature of cellular automaton itself allows to program it using "graphical" method (by setting initial status manually, eg. by filling selected cells) - it means that cellular automaton is an computational system which may be programmed even by persons who can not read or write "traditional" texts. From musical perspective Cellular Synthesis is a musical instrument and notation fully programmable (in the sense in which it works with programming languages), but does not require writing code (or any graph-based eqivalent of the code, as in case of the dataflow-based languages widely used by composers and new media artists). Cellular Synthesis is a "graphically coded" system for creating music.

The project is inspired by the research and publications by Stephen Wolfram. Wolfram studied the behavior of cellular automata (mainly one-dimensional) formulating a series of observations and indicating potential applications. Although my own machines were based on the most commonly on two- and three-dimensional matrices, thanks to Wolfram I was able to confirm various personal observations and solve some problems that I experienced before (one of the fundamental observations of Wolfram says that the nature of cellular automata, makes, the results achieved with their help, not create more complex structures, when we complicate the automat control rules - in other words: a set of simple rules can lead to complex patterns). In addition, Wolfram, what also seemed interesting to me, treated the cellular automata not as a mathematical curiosity, but as a full-fledged computing system that can be used for specific purposes - in practice, this means that a cellular automaton can be a programming language with a completely graphical data and rules encoding that does not require even literacy to create programs.

Since about 1998 I have been studying the possibility of using, for my own purposes, the algorithms describing cellular automata. There was probably nothing unusual about this - the cellular automata for a long time were laying, like applications fractal theory, in the area of interests of media art and many artists have tried to use them in some way in their works (moreover the cellular automata are the computing systems researched, e.g., by computer sciences, and they are sometimes used as simulation mechanisms in economics, anthropology, and many other fields of science). A cellular automaton is built based on an n-dimensional matrix of cells that can take a finite number of states. For each cell a set of cells known as adjacent is defined, and each cell changes its state depending on the state of their neighbors. The history of research on cellular automata is quite interesting and dates back to the 40s of the twentieth century, and the works of Stanislaw Ulam and John von Neumann at Los Alamos National Laboratory (thus, in which the first atomic bomb was constructed). Another important name is John Conway, author of probably the best-known two-dimensional cellular automaton game of life. In the 80s of the twentieth century Stephen Wolfram (Wolfram is an active researcher in the field of broadly defined computer sciences, inventor, and author of a number of concepts and technologies related to machine understanding of language up to cryptography, and also the creator of Mathematica program, Wolframalpha service, etc.).

I returned to work with cellular automata and sound - additive synthesis (involving, in short, the creation of a sound based on the manipulation of parameters of its basic components [e.g. Sinusoidal components] - generally the additive synthesis is difficult in practical use, because it requires the simultaneous handling of a giant number of synthesis track parameters, which is unintuitive and “manually” difficult to be executed, e.g. when playing live). The artist's technique, developed by me - cellular synthesis - is something between the systensys technology and concept / tool for composers (premiere performance of tracks based on cellular synthesis took place during Vox Electronica festival in 2014 in Lviv). It - in simplest terms - involves the use of cellular automaton, under properly established rules, for dynamic control of additive synthesis. It enables intuitive control and modification - also in real time - of a large number of parameters of the additive synthesis track. The cellular synthesis, is quite difficult to be assigned to any traditionally understood forms of artistic expression - it is, at the same time, a task in itself, a description of the creative process and an algorithm and its complementation in the form of a specific technology.