A few weeks ago we featured a video of the chess genius Garry Kasparov talking at Google. This is a man who spent his life studying the game of chess and became world champion, but was eventually beaten by a computer.
The reason the computer won, is due to the amount of processing power it has available to analyze every potential move possible and select the best course of action. It can do this fast enough for every single move, and is called a brute-force approach.
While that’s easy to understand, it’s much better to see it visualized. Thanks to website turbulence.org, we can now not only see that brute-force approach visualized, but actually play against a computer using it.
Each turn the computer takes is preceded by a decision time at which point we see the visualization of the future move choices being considered. This is explained as follows:
When the machine is thinking, a network of curves is overlaid on the board. The curves show potential movesâ€“often several turns in the futureâ€“considered by the computer. Orange curves are moves by black; green curves are ones by white. The brighter curves are thought by the program to be better for white.
If you decide to play against the computer, and have an average knowledge and skill at the game, chances are you will win. This is actually version 4 of the Thinking Machine, and it uses basic algorithms and no standard starting moves as expected by experts.
If you are wondering how the Thinking Machine was created to run in a browser, it uses a mix of Java for the underlying chess computer, and Processing for the graphics used in the visualization.
From playing with this briefly I appreciate seeing the decision stage visualized. In a way it makes the opponent seem a little more real than if the next piece on the board just moved.
Play the game at turbulence.org