INSTANT PLACES : CANADA CODE : 2009/2010
GENERATIVE PROGRAM :: INSTALLATION :: VIDEO SAMPLES
The installation computer generates widescreen images and accompanying five-point surround sound compositions. Each sequence begins in silence with the screen faded to black. The program chooses at random four images and a sound file from a database of images and sounds we recorded across Canada, and begins the generating process: sounds and images fade up, are transformed over a two-minute sequence, and fade back down to prepare for a new sequence.
As the sequences exist only as the output of processes that occur in the moment of generation the viewer experiences a series of audiovisual events that have never happened before and will never happen again
.

Video

The source for the video transforms is over 2000 still images we recorded at locations across Canada. The program chooses four images to transform over each two-minute sequence. It looks for areas of light and dark and specific colours within the images; these areas become windows that reveal underlying image layers as the sequence unfolds.

None of those four images will be chosen again until all 2000 have appeared in a sequence. In the continuously running program it takes about three days for a specific image to appear again.

Audio

Many of the source sounds are recordings of musical instruments we found at locations across Canada:
— a pennywhistle played in WW II concrete bunker in St Johns
— a piano in a living room Northwest River Labrador
— an antique zither found in Smithers BC

The location recordings were transformed during performances across at artist run centres, libraries, and schools across Canada, as well as a three-hour live to air performance on community FM radio in the Jean Talon neighbourhood in Montreal.

We recorded these performances and edited them into audio files around four minutes in length. The computer program chooses at random one of these sound files and a starting place within the file, and creates the surround sound score for a sequence by reworking about 14 seconds of the audio over the two-minute sequence.