Results on Gyroscope max games + Visual Games.

Game 3: Maximize the gyroscope value in the same axis as the accelerometer above.
Game 3a, as above for a different axis.
Game 3b, as above for a different axis.

Poor performance in these games with the current method of evaluation and motor atom types. The fitness variance is quite high in the population at a given generation. I do not explicitly measure the fitness variance for an individual lineage through that population however I think if I did I think most of the fitness variance in a generation would be due to individuals with high fitness variance between generations. If individuals have high fitness variance between evaluations this is bad. It is high fitness variance longitudinally that is a good thing. 

I think I need to check that the networks are having significant effects. Probably its a good idea to visualise some of the Izhikevich atom activities. 

Visual Games

I think we should perhaps move onto looking at games that use the results of vision. This may give us a range of "hand-eye" coordination tasks that make sense.