Real-time lighting design: a pioneer, a work, and a collaboration

ABSTRACT

This article proposes three stories that exemplify different metaphors, mechanics, and practices of real-time lighting design. Starting with the example of cybernetician Gordon Pask’s Musicolour, the article proposes a cybernetic-based model of control that favors principles of human–machine co-operation. Then, the analysis of the collaboration between a lighting designer and a multimedia artist illustrates the need for and challenges of radical interdisciplinarity during the production of a stage performance. Finally, the description of a contemporary multimedia installation demonstrates the principles of an aesthetic of lighting grounded in machine-based behaviors. Through those three stories, this article aims to initiate a discussion on the hybridization of lighting with real-time computing, a process that significantly transforms the practice of artists and lighting designers but remains understudied.

Acknowledgements

I thank Annie Leuridan and Cyrille Henry for sharing their story and insights with me, and Marcus J. Carney, who oversees the Pask Archive at the University of Vienna, for his help with archival research.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 For a detailed biography of Gordon Pask and discussion of his cybernetic practice, see Andrew Pickering’s The Cybernetic Brain: Sketches of Another Future (2010).

2 DMX512 is the main communication protocol currently in use for the control of lighting equipment, including control desks, light sources, dimmers, and various stage effects. It was developed in 1986 by the Engineering Commission of the United States Institute for Theater Technology (USITT).

3 The maker movement refers to a contemporary subculture that develops a do-it-yourself practice of technologies, particularly in the areas of electronics engineering and software programming. It promotes the democratization and appropriation of technology and encourages individuals to be creators and innovators (see Anderson 2012).

4 More precisely, while there are some light-based artistic practices that explore interactive, autonomous, and emergent machine-based behaviors (e.g. Memo Akten, Rafael Lozano-Hemmer, Gordon Pask, Nicolas Schöffer), there is little theoretical and historical discussion around these principles (see e.g. Daniels and Naumann 2010; Malina 1974; Popper 1968, 2006; Reichardt 1969; Weibel 2006). Most of the literature on light in the theater and the visual arts rarely touches on the issue of machine-based artistic expression, while literature interested in the new media arts has essentially ignored the medium of light (see e.g. Salter 2010; Broeckmann 2016; Penny 2017).

5 For a description and discussion of those machines, see Pickering (2010). For a study of Beer’s Cybersyn, see also Medina (2011).

6 The following section is based on an online semi-structured discussion with Leuridan and Henry. They were asked about the origin of their ongoing collaboration, and the challenges they faced in articulating stage lighting and interactive programming within the constraints of professional theater production.

7 Pure Data is an open-source visual programming language for interactive music and multimedia creation. It was created by Miller Puckett in 1996. See https://puredata.info/

8 See the work of psychologist Daniel Stern on ‘temporal contours’ or ‘time profiles’ that characterize the temporal evolution and perception of all actions and events (Stern 2010). Stern analyzes the role of such temporal dynamics in the arts (73–97).

9 SenseFactory premiered in September 2019 at the Muffathalle in Munich, Germany. It was created in collaboration with graphic designer Erik Adigard, machine learning and media artist Sofian Audry, music composer FM Einheit, producer Dietmar Lupfer, media artist and theorist Chris Salter, architect Alex Schweder, and olfactory artist Sissel Tolas.

10 This effect is similar to the light tests realized by Henri-George Cluzot for his movie L’Enfer. Extracts can be found online; see https://www.youtube.com/watch?v=aZ38CcJ-2vw

11 All AI aspects of the work were programmed by Sofian Audry.

12 Python is a general-purpose programming language regularly used in scientific, machine learning, and AI applications. See https://www.python.org/

13 The integrate-and-fire algorithm is a mathematical model that simulates the evolution of the membrane potential of biological neurons as they receive synaptic stimulation. With every stimulation the electrical potential of the membrane augments until it reaches a threshold and fires a ‘spike’ that propagates to other neurons. If no stimulation is received, the potential of the membrane decreases. Here, the integrate-and-fire algorithm receives the signals from the crossing sensors and outputs the continuous value of the simulated neuron membrane. The details of the algorithms are superfluous to the present discussion.

14 A finite-state machine is a computing model that transitions between a finite number of predefined states in response to its input data. Here, the signals from the sensors are analyzed and used to select which state should run. The different states of the machine correspond to different signal-processing technics that produce distinct lighting behaviors. It is important to stress that the term ‘state’ in a finite state machine refers to a real-time dynamic process, rather than a static set of pre-recorded lighting parameters, as commonly used in theater lighting control.

15 Max/MSP (Max Signal Processing) is a visually based data flow programming language for music and multimedia creation created by Miller Puckett in the 1980s and named in honor of computer music pioneer Max Mathews. It is currently maintained by Cycling ’74. See https://cycling74.com/products/max