Singapore Straits Times: World chess champion wins $1.3m in draw with computer

Mr Kramnik's draw with the machine exceeded the performance of his predecessor as world champion, Mr Gary Kasparov, who in 1997 was defeated by supercomputer Deep Blue in New York.

The final game was the shortest of the contest, with the 21 moves taking just under two hours.

After two wins each and four draws, the final tally was 4-4

Mr Kramnik said he was amazed by the strong performance of Deep Fritz during the tournament, which had been 'incredibly much stronger' than chess programmes he had played even a year ago.

'It is stronger in a positional way,' he said.

'It is not just strong in terms of calculations, which is to be expected, but in terms of positional moves, it plays like a very strong human, these are human moves.'

The Deep Fritz team said they learnt a lot from the eight-match tournament with Mr Kramnik. They vowed to further refine and improve their creation.

Wired News: Chess: Man vs. Machine Plays Out

Haifa University held a two-day symposium to address one question: What's the point of all this computer chess?

Fernand Gobet, a psychologist from the University of Nottingham who specializes in intelligent systems, spoke in a videoconference about the chunking theory. It postulates that a grandmaster sees the chessboard as patterns or "chunks" of pieces, then as groups of these chunks -- not as individual pieces.

Such research may help software developers who are progressing from brute force search to something more sophisticated.

"Now we have two more matches ... and we will get new data to see whether the machine is better than the man... If you want to understand intelligence, the game of Go is much more demanding," [Jonathan Schaeffer of the University of Alberta] said. "It doesn't have the silver bullet: deep search. Chess has somewhat outlived its usefulness. It turned out to be easier than we thought."