damnbot,

No one can give you the answers you seek... they can certainly point you toward them, or perhaps even explain them for you, but in the end, it will take work on your part - in the form of implementing what you have read - for all of it to sink in.

You''ve indicated that you''ve read widely on neural nets... that''s great... and certainly more than a lot of people do before asking for help in this forum.

My advice to you is to implement a simple feed-forward network. Nothing fancy... just a couple of input nodes and an output node or two. Play around with it and see what you can produce. From that basis, you can extend your code to a larger network... try different threshold functions, add hidden layers and implement different network concepts (like a Hopfield network, or a recurrent network).

There is only one place to start and that is at the beginning!

Cheers,

Timkin