Open another instance in a second window and you'll quickly see the answer. Only one car would best fit the constraints (that's by definition). However, it's p

Open another instance in a second window and you'll quickly see the answer. Only one car would best fit the constraints (that's by definition). However, it's probably not the car you are seeing, because the algorithm tends to get stuck on local maxima. After about 2 hours of running, the thing has flatlined as each window I opened has 20 copies of pretty much the same car (although they differ between windows), and they are not getting that far.