CROWD SIMULATION MINIDUMP

Ahead of my residency at humainTropHumain in Montpellier, where I'll be making a moving image installation using mocap and crowd simulation software, I've been learning how to use a package called Golaem Crowd. It uses logical operators (AND/NOT/IF etc) to build complex behaviours for digital bodies. A few weeks in, and I've broken the software numerous times, reported quite a few bugs to the developers, and have finally managed to find a delicate stability in the process. I've realised that frustration and actual red hot anger are very much part of this learning process for me. These feelings only go away once I know which buttons to avoid clicking at which time.

This is software: learning the route of least resistance. Too many crashes, failures, errors, lost files and lost time means you're constantly optimising your technique. You are funneled through easily at points, and you are kettled away at others. You map the behaviour of the software, or maybe the software is mapping yours.  Which makes this whole process very meta - I'm designing behaviours for digital bodies, and the software I'm using is designing behaviours for me. I'm being conditioned and conditioning at the same time. 

Anyway, here's some test images and tweets with happy accidents.