This little video shows the work of two scripts I made in Grasshopper and Rhino 5. You see two solvers that reassemble a puzzle of the shuffled pieces of a distorted dodecahedron.
The first part of the video shows the worksflow with completely random picks. The second part shows what happens, when the script "memorizes" which pieces it already used.
The scripts were elaborated within the scope of a class on parametric design at the chair of Knowledge Architecture at Dresden University of Technology.
The work is inspired by Marvin Minsky's theories on artificial intelligence and the subdivision of relatively complex tasks into simpler sub tasks handled by agencies. Also the whole simplification of scripting in Grasshopper (Rhino's scripting GUI) is based on achieving complex goals by combining super simple bricks of code.
Furthermore I wanted to show the improvement in efficiency of the script's work when a memory is applied to help pick the right pieces.
Have fun watching
also check out the classes blog: