Many visitors to SIGGRAPH 94 would agree that chaos is a fitting term to describe the mayhem associated with the week-long event. In addition to the chaos of SIGGRAPH, visitors lucky enough to get into the CAVE to see and, more importantly, hear the virtual reality model Sounds from Chaos in Chua's Circuit, will experience mathematical chaos in the virtual environment.
Innovators at NCSA have created a graphical interface that allows users to hear the chaos created in Chua's circuit and alter the sounds it produces. This opens the door for two seemingly unrelated disciplines to come together--music and science.
"We are using sound to enhance scientists' experience by inventing ways that data can be turned into sound," says Robin Bargar, composer and NCSA visiting research programmer. "Not only have composers gained from our work, it has passed along to the scientific community."
"Virtual reality could learn a lot from adopting interface principles developed in musical instruments," says Bargar, who directs audio research at NCSA. "Musical instruments are some of the most sensitive real-time interactive tools made by humans."
The interface presents a digital representation of a ball that can be moved across a textured plane. Changing the peaks and valleys on the plane or moving the ball to a different area of the plane produces different sounds. A graphic representation of the chaotic trajectory floats above the plane in the CAVE so users can see how the sound evolves.
This interface can produce sounds from almost any numerical data, which makes it valuable to scientists and composers.
Chua, creator of the chaos circuit, is a leading researcher in chaos and nonlinear dynamics and editor of the Journal of Bifurcations and Chaos. He earned his Ph.D. in Electrical Engineering from UIUC in 1964.
"Many scientists show research interests in terms of methods for simultaneous control of multiple parameters in a high-dimensional system," she says. "The manifold interface, developed for navigating multiple control parameters according to acoustic states related to chaotic states, provides scientists with a unique representation of the problems and potential results of multidimensional control."
Choi is a guest composer with the Audio Development Group at NCSA. She is also a professional composer of new music, concentrating on electro-acoustic and computer music.
"We are interested in using sounds of chaos to produce music and in using audio to make discoveries about chaos," says Bargar. "Historically, composers did not make the sounds they used to create music, now we can."
Bargar oversees audio research including sonification, auditory display, and the integration of audio and visual information--specifically to explore paradigms of graphic and sound computation. This includes human-computer interface development, as well as the use of sound, with traditional computer graphics and video formats.
"I am impressed and excited by all aspects of the sonification project," says Chua. "The work Robin's group is doing represents very significant pioneering research of immense commercial potential. It requires a great deal of creativity and ingenuity, as well as a very broad background, to combine high-tech hardware, programming skills, nonlinear dynamics, and music into a unique synergetic system with wonderful capabilities. It represents cutting-edge research of the first magnitude."
"In addition to providing tools to science, we have to build the environment that we are going to work in," he adds. "That is both exciting and challenging."
As we went to press with this issue, members of the Audio Development Group were putting the finishing touches on a software audio pipeline of their own design. Their pipeline uses a client-server architecture that allows the SGI Indigo to perform as a real-time interactive sound synthesis engine. This system provides audio to all 43 VROOM applications at SIGGRAPH 94.
In addition to Bargar and Choi, Ulrike Axen, Kelly Fitz, Camille Goudeseune (NCSA/UIUC), and Sumit Das (EVL) form the Audio Development Group.
"We know sound enhances our lives; the spoken word tends to be more convincing than the written," says Bargar. "Numerical data converted directly into sound becomes qualitative. We are not improvising or making up the data. We postulate that there are ways of interpreting data that are still unknown."
