|Art & Artificial Life International Competition |
| Home | Entry | Editions | References | Español |
|The Appearance Machine is an industrial entertainment complex for the continual transformation of the input of locally generated refuse to output as global media. It is an autonomous synaesthetic system that perpetually constructs a cinematic aural/visual space. |
The machine, permanently situated in New York City, physically transforms detritus into a live audio/video stream which is unicast directly to a viewing public.
Computesr act as translators. The machine´s images, created in realtime, are analyzed for movement and colour and perform the accompanying soundtrack. The soundtrack then directs the mechanical impulses that animate the set. Feedback loops between all aspects of the machine are the embodiment of its networked intelligence - the composition. The machine invents in continuous response to its self-create accidents.
The Appearance Machine technical information
The Appearance Machine is an elaborate network of devices all controlled by the MAX programming environment from a computer. The devices communicate via MIDI or serial control. A Basic Stamp network under serial control runs many of the custom mechanisms.
A compressor powers pneumatic valves which move the objects. Vibrators move the set. Stepper and servo motors activate the lights and other mechanical devices including the main turntable and thje systemwhich carries objects onto th table. Fans and shadow casters can be directed and controlled. Numerous lights can change colour or ramp in intensity as can an internally lit central dome which acts as background.
Video mixers switch between multiple cameras and can luminance key-in imdividual 'characters' created on an 'individuator' set. The cameras themselves pan, tilt, zoom and focus under serial control. The Very Nervous System and BigEye software perform image analysis. Multiple sample players, synthesizers and multi-effects unit create the soundtrack.
We ourselves create the programming rules and some data sets. Additionally though, in a form of contemporary resource extraction data is mined from used films and is input into the system as edit decision lists, controlling the duration of the camera's cuts and dissolves. Thus only the structure, and not the content is appropriated from the production factories of the dominant providers of mass appearances.
Sorenson Broadcaster is used to encode the resulting audio and video on a second computer. The esults are then streamed over a broadband connection live to the viewing public in quicktime format.