Works well in Chrome, but gets bog slow in Safari (10 beta). Instruments shows that it spends an enormous amount of time in memcpy copying IOSurface data around; I wonder why it is happening on the CPU rather than the GPU. But it may not be entirely Safari's fault: it appears that every single organism first gets drawn to a canvas the size of the entire screen (as opposed to the size necessary for the organism), then that canvas gets drawImage'd to the main canvas. That's an enormous number of wasted pixels. Also, this seems to happen fresh every frame: I don't know why they're using an intermediate image in the first place, but it's not for caching.
Mount Improbable: Play With Evolution
11–14 of 14 posts
Re: Mount Improbable: Play With Evolution
#12The title was incomprehensible until reading the post. I was wondering what a "Climb Mount" was.
Re: Mount Improbable: Play With Evolution
#13The title was incomprehensible until reading the post. I was wondering what a "Climb Mount" was.
"Climbing Mount Improbable"[1] is the title of a book by
Richard Dawkins.
Haven't read it yet, but I can absolutely recommend "The Greatest Show on Earth"[2].
[1]https://en.wikipedia.org/wiki/Climbing_Mount_Improbable
[2]https://www.amazon.com/Greatest-Show-Earth-Evidence-Evolutio...
Re: Mount Improbable: Play With Evolution
#14I always thought this* was a much better demonstration of evolution in action because it's completely undirected by the user and quickly converges to fit solutions from obvious randomness. * http://rednuht.org/genetic_cars_2/
I had this running in the background for hours one day. They got pretty good. He's also got one for biped walkers that's fun to play with. http://rednuht.org/genetic_walkers/
Playing around with the wheel friction and letting it run for days was interesting: https://github.com/red42/HTML5_Genetic_Cars/blob/master/cawr...