The easiest way to explain this is with an example
Test 1:
waypoint - 120:00
Zerglings - 10
Melee +1
Run, Clock how long it takes to get < 6 minutes
Test 2:
Waypoint - 6:00
Zerglings - 10
Melee +1
Run, Clock how long it takes to get < 6 minutes
Hypothesis:
If it reaches a successful build it bases off that and modifies it rather than
starting over from scratch
It seems efficiency could be greatly improved by providing an option to
continue on a failed waypoint (possibly only until the waypoint is met / up to
a specified time) to assist in establishing a starting point to improve, and
possibly having failed waypoints increase the mutation rate?
Original issue reported on code.google.com by rudd...@gmail.com on 16 Jul 2011 at 3:07
Original issue reported on code.google.com by
rudd...@gmail.com
on 16 Jul 2011 at 3:07