I know all of this, tilly. In fact, I think we've discussed part of this before (but I'm too lazy to go look up the node, assuming it wasn't by /msg).

What I'm doing is exploring to see if I can get a program to learn about shapes and pressure, then make decisions based on that. Part of the problem that the standard alpha-beta algorithms have had is that they used static board evaluations - evaluations of this position without any concept of where this position may or may not go. I want to try and use a more dynamic evaluation.

And, yes, I know this is exponentially more difficult to program than static evaluations. *grins* Therein lies the challenge, neh? :)

Once I get going into it and have my theory actually start to make some real moves, I'll probably throw a Meditation up there on the concepts I'm using and see what happens when people who actually have working braincells bang on it. Till then, I'm gonna be sorta mum.

------
We are the carpenters and bricklayers of the Information Age.

Vote paco for President!