Hacker News Reader

does GA and GP behave somewhat predictable then? Couldn't it, if you are unlucky, take a very long time to converge?

Going by the javascript version he was inspired by, it takes a long time to converge if you are lucky (I kept it running for several days on and off). If you're unlucky, it doesn't converge before you get bored.

OK but then how is this "working"? I mean it is very cool and it might work if someone puts effort into it. But is there any reasonable way of using this right now?

It's actually not so bad. It's just JavaScript and maybe my code (I wrote JS version).

Roger Alsing (who was the original author and inspiration) did later an optimized compiled parallel version that was able to get Mona Lisa in 1 minute 34 seconds (on a 64 bit 4 core machine).

http://rogeralsing.com/2009/01/04/scaling-clustere-evolution...

Also there are supposed to be some efforts to run it on GPU, which should be even faster.

Reply on news.ycombinator.com