pedro

27 Jan 2015

01lr

My idea for the pebble assignment was to work with subdivision patterns to represent time. Initially, I investigated some fractal division such as an incremental cracking. But once the resolution of the display was limited I would not be able to represent small divisions, so I decided to use a voronoi diagram as the logic behind the division.

I used processing to simulate the display of pebble watch in order to evaluate how the gradients of a grayscale brute-force voronoi could be simplified for a black and white screen.  These simulations show that the logic itself of the work had to be changed to fit the constraints. Firstly, I tried to use probability to simulate gray areas, but it resulted in a lost of legibility of the diagram. Then, I tried to simplify the diagram, painting only areas of equal distance for two centres. It resulted in an interesting visual effect that resembled the idea of a cracking display.


After defining the visual pattern, I tried to evaluate what would happen if each of these cells were agents that interact with the neighbours and the frame of the pebble, with forces of repulsion. As the time passes and the cells subdivide, each agent try to find a appropriate space in the area, establishing a bottom-up balance of the visual composition.

On one hand, with the dynamic behaviour of the cells, the original idea of subdivision was taken even further, but on the other hand, the implementation in the pebble watch was a difficult process, due to the brute forces algorithms underlying the diagram. The constraints involved in the implementation were related to the limited memory of the device, what imposed several limitations for the design of the Voronoi diagram. After several memory crashes, the resolution, the number of cells and the movement were reduces, what took the design one step back.

Finally, I consider that the Voronoi diagram is a very interesting device for the visualization of time. The obstacles faced in the implementation demand a careful optimization of the algorithms in order to incorporate the full potential of the design.