Synthetic Surfaces

Exploring the visual manifestation of synthetic encoded surfaces with randomly generated values..

All digital images are encoded. What looks like an organic complicated whole, is “atomized” in the computer … made up of a very large collection of very simple elements. They are a product of calculations which are part of the algorithmic and coded structure of their surfaces. The JPEG algorithm was created to compress photographic images. It is made up of constants, ranging from the value of 0 to the value of 255. What would happen if I decoded the JPEG, and rebuilt it with randomly generated values? 

The idea was to generate “live” random RGB values (255,255,255) on one screen which would then “build” a live JPEG on another screen which would be totally lacking in any form of representation. This was achieved by programming in Python and then running the program on a computer that was connected via HDMI into 2 monitors. Through the code I could control the size of the “pixel blocks” to fill. To generate a full HD screen (1920 x 1080) with randomised coloured pixels took over 2 hours and did not have the effect I was after. So I was able to increase the pixel sizes through the code which made for a far more pleasing result.

What I found interesting was the amount of data it took to generate one full HD frame with colour (an image). Understanding this, one can begin to understand the amount of data that we all use on a daily basis 

The resulting images, void of any representation, gave me good insight into how an image is made. Synthetic, constructed purely through computations and void of any representation, they become an algorithmic fractal and serve, perhaps, as a counter-information form of resistance.