The State of Play

 


Pictured is a book I thought lost until this afternoon. Purchased in 1988, The Complete Chess Addict is a wonderfully entertaining miscellany of chess lore and trivia, and one of those great dipping books that you keep coming back to for more random facts, figures and anecdotes. It turns out that it had randomly been deposited in a storage box in the spare room with a lot of unrelated [non]bookish] stuff: the last place I would have thought of looking.

Anyhow, thumbing through the thing, I came to the last section, entitled The End? which surveys the then state of computer chess play and what it might portend for the future of the game. The roots of computer chess began truly in earnest with a program created for the IBM 704 mainframe back in 1957, which although taking several minutes per move, was capable of playing a full game of chess, unlike previous engines.

By 1988, standalone chess engines were being marketed that could give all but the strongest players a good game, with even cheap machines offering lower strength players useful opposition. Soon after, human supremacy over the chessboard was being challenged, and by 1997, a computer chess engine, IBM's Deep Blue, became the first ever program to beat a reigning world chess champion: Gary Kasparov. Today, there isn't a human that can best the top chess engines; all are rated in the stratosphere, and are so strong that one playing the other, without any form of handicapping in the opening phases of the game would always arrive at a draw.

One might think that that would be that for the game of chess, but far from it. Today's top players, and many more in the game at less exalted levels of play, make use of the engines to analyse games and positions in depth to gain a better understanding of the game and insights into possible strategies and tactics that they can then employ either over the board or online. Cheating does occur, but engines don't play the game the way humans do: they can think so deeply into a game - within a few moves in, they are already in the winning position against anyone human - that they apparently make the oddest choices of move, calculating deep into the game for the ramifications of apparently random move choices to be felt.

A human regularly playing in that way would quickly be suspected - many have - of using an engine to cheat. However, the interesting thing is that the very strongest players in the world are now learning from their software counterparts and taking human chess to entirely new levels in all forms of the game. Which while demonstrating the supremacy of computers over humans in the game, also serves to emphasise the adaptability and ingenuity of the human mind: we developed the hardware and algorithms to play the game at such stratospheric levels in the first place, and now we are using them to inform our own chess-playing. The game is safely in our hands.

Comments

Popular posts from this blog

Of Feedback & Wobbles

A Time of Connection

Sister Ray