Interesting article about design

[JHCC]

A friend of mine (an engineer at NASA's Jet Propulsion Lab, and probably the closest thing the real world has to an actual wizard) posted this article on Facebook, about how computer-generated structures (for example, an engine block) can mimic organic structures (such as the interior of a bird bone) when both put material only where the load is. While this has significant implications for 3D-printed machine parts, the forms so generated can be quite beautiful.

https://medium.com/intuitionmachine/the-alien-look-of-deep-learning-generative-design-5c5f871f7d10#.lxmv3ibov

[Anachronist58]

Fascinating.  Reminds me of how I am learning to tie various knots - the outcome is driven by end-use criteria rather than the steps it takes to build the knot. Don't laugh at me!

[Alan Evans]

I had a friend, Jamie McCulloch, who always introduced himself as ...a sculptor who builds bridges...

In the late 80s / early 1990s he developed a system for structural design and was given computer time at Stirling University to explore it. He would make up a physical wire frame structure and then dip it in washing up liquid. He then digitised the surface form of the bubble. The bubble obviously linked the frame in the most efficient way possible under tension and so was the optimum lightweight skin under compression.

It was a similar-ish concept to the sand bags and string system that Gaudi used to plot the angles of his tilting pillars in Sagrada Familia and the Colonia Guell church. He would attach both ends of the strings to various points on a board representing the pillar bases on plan of the building, tie sand bags of scaled weight to the appropriate place on the string length. Invert the board so that the strings dangled at the angle determined by the weight position and photographed it. He then plotted the angle of the pillars from the prints. It enabled him to use monolithic pillars at extraordinary angles rather than be limited to vertical masonry columns.

He also used catenary curves for his arches everywhere...and there are reports that these were plotted empirically. They actually hung a chain and plotted the curve on a board then inverted it to build the arch. I think he may have then made the monumental leap to the 3 dimensional system from that.

Alan