A three-dimensional sequence is generated from raw data. The rendered sequence is projected onto the original scene in the program, rendered again, and projected once again as a shadow map. In this final step, the rendered image is also applied as a texture on the objects in the scene. This repetition produces superimpositions of light, shadow, structure, and perspective, which leads to doubling, deviations, errors, and erasures at various levels, which in turn, feed into the process as aesthetic information. The result is a clouding of the three-dimensionality, an imaginary, unstable, fleeting space, which through continual change steadily constructs, and simultaneously deconstructs new definitions of its components.