The geometry of the grottos consist of 260 million and 1.35 billion individual faces respectively. This large amount of data cannot be adequately processed by existing CAD Software. Customized algorithms were therefore developed to calculate the construction details and to convert the form into printable data.
The articulated mesh is a self-intersecting surface that does not enclose a volume as it is not a differentiable, orientable manifold. In order to extract a volume with a uniquely definied inside and outside from this geometry, a strategy analog to casting is implemented: The folded surface is filled with water. Every part that stays dry in this simulation is counted as the inside of the wall construction. As a result, a clean volume emerges.
To calculate this volumetrization and to insert construction details, the entire mesh is voxelized at resolution of 1mm (0.8mm for grotto 2), leading to 30 billion spatial data points. Such an enormous amount of information cannot be processed as a single entity in the computer: the geometry is loaded only where needed, streamed layer per layer.