Wow, I didn't remebered the lineal "Entropic" AI to be so limited! I always tested it in the open field as I knew it didn't perform well in maze-like environments, but it looks like the google car racing against a F1.
So just have a look to the fractal "One Way" version of the last post solving the maze:
OK, quite nice, I played it a little "slow motion" so you could observe the "flames" used to decide more clearly. This time I downgraded it to 500 futures and 10 seconds, going beyond this is just a mere way to impress the audience, not any real advantage.
Now lets watch the clearly weaker lineal version of the "Entropic AI" (BTW, it didn't finish the maze, the app crashed before that... my "spagetti programming" habits to be blamed, I guess).
Nothing more to say...