I've always been frustrated by the fact that two dimensional projected patterns deform as they map over a given shape rather than respond intelligently to the geometry of the shape. What I wanted to do with Cloudform was to make a script that would map a geometrically consistent pattern on the surface of a complex form. It is simple in concept, harder in execution. After a bit of thought I realized the problem could be solved in a way similar to how faces are mapped in animation. By projecting a two dimensional pattern onto a topologically identical but flattened version of the form, then reassembling the form, the pattern actually folds with the form rather than distorts around it.
I've been playing around with an updated version of Fractalizer, a MEL script we wrote at Bureau V that produces a
field of fractals on any given geometry. This image shows a plane with four levels of fractal division both in line and solid form.
This is a series of tests using our Koala MEL Script in conjunction with the new randomization selection tool in Maya.
We made these color + pattern + variation studies earlier today while testing Koala, a MEL Script we have been developing for some time.
Bloom is a MEL script that produces variable, three dimensional topologies. By setting these variables to make random selections within a certain range, a sort of snowflake logic arises with intricate micro-scale variation within a coherent macro-scale formal language.
This is an example of a shape made using the bloom MEL script that I wrote recently.
I’m working on the design of a parametric structural mesh that uses engineering calculations to drive the design of a surface and then uses the resultant form to re-engineer the structural members. Its a pretty simple concept, when there is a need for more structure, the surface compresses producing a tighter mesh, when there isn’t much force, the mesh becomes looser and the surface expands. I got the idea from looking in the window of Wolford on my way to work and seeing the mesh of some stockings stretching and getting tighter are they went over the mannequin. The trick is getting the surface to realize exactly how much it needs to deform versus how much it can deform. The analysis creates a feedback loop the cycles through the structural data and the surface until the form settles on the best balance between structural and design performance.
This is a chandelier that we are designing at Asymptote for lighting manufacturer Zumtobel. The idea is to produce a a single minimal surface component that can expand infinitely via seamless edge to edge connections. The form is generated by a scripting process I’m developing that evolves the topology of a primitive towards particular goals, in this instance heightened refraction and a lack of undercut surfaces.