The ancient, intellectually challenging game of chess originated in India and is played by two players on a checkerboard consisting of 64 squares, with 32 pieces divided equally by color. Each piece has its own role and ability to perform a unique function on the board. The game has enjoyed steady popularity and minor rule changes over centuries of time. With the development of computer technology, more and more players were able to improve their skills by playing against computers, using information stored in computer databases and even using computers as personal trainers.
How Chess Computers Work
There is no principal difference between a chess computer and a regular one. The computer programs evaluate chess positions on a virtual chess-board and select the best moves according to the specific criteria. Usually a computer calculates a numerical score for each position it evaluates, based on four variables. The highest score determines the move. These four variables include:
- Pieces. Each piece is assigned a value at the start of the game. For example, a pawn gets a value of 1; a queen 9. These values may also be modified later as the game progresses.
- Position. The computer observes all of the chess pieces on the chess-board and assigns them unit values based on how much territory they are able to attack.
- King safety. The computer evaluates both kings' positions and attempts to put its own king on a safer square and to force its opponent's king to a more vulnerable square.
- Tempo. The computer seeks to place its pieces on the intended squares to carry out its current plan faster than its opponent can counter.
The most popular opening chess move for the player of the white pieces is "e2-e4," which means the Whites move the pawn directly in front of the King 2 squares forward. This is only one of twenty possible first moves the Whites can make. Blacks' pieces have the same number of possible responses. So, there are 400 possible combinations for the first move alone. As the game continues, the number of possible positions keeps growing. All these possible moves may be presented in the form of a tree. (See Figure 1.)
It is impossible for any computer to calculate all possible moves by brute force. Therefore, a computer estimates positions for only several moves ahead. Today, computers are able to calculate up to 14 moves. Moreover, there are many well-developed artificial intelligence (AI) strategies, such as minimax algorithm or a technique called alpha beta pruning , that limit the search for the best move to an analysis of the most promising positions. In addition, computers use extensive databases of previous games for openings and endgames. For the endgames, when the number of pieces and thus the number of possible moves reduces significantly, computers are able to calculate all possible positions when five or fewer pieces are left on a chess-board.
First-class chess computers also require very powerful hardware. IBM's Deep Blue supercomputer uses a total of 512 microprocessors working in parallel to break down and analyze chess positions. Each processor simultaneously calculates and evaluates one possible sequence of moves, enabling the computer to examine 200 million positions a second!
The idea of a machine capable of playing chess was born long before computers. Historical annals mention Baron Wolfgang von Kempelen, who traveled around Europe in the eighteenth century with his Maezal Chess Automaton, nicknamed the Turk. The machine looked like a marionette that was able to make chess moves. The trick, of course, was that a tiny chess master was hidden inside the box.
The connection between modern computers and chess games began with a prophecy of British mathematician Alan Turing (1912–1954) that machines would compete in all purely intellectual fields including chess. He specified the first computer program for chess as early as 1947. Claude Shannon estimated the number of possible computer positions and proposed basic strategies for restricting the number of possibilities to be considered in a game of chess.
The first computer program for a full-fledged game of chess was written in 1957, and since 1966 computers have been playing in human chess tournaments, gradually increasing their rating and the level of tournaments over the years. Former world chess champion and scientist Mikhail Botvinnik (1911–1995) contributed to the development of computer algorithms for playing chess and to the creation of the KAISSA chess-playing computer program. KAISSA won the world computer chess championship in 1974.
As the result of a growing population of computer chess programmers the International Computer Chess Association (ICCA) was founded in 1977. The same year was marked with the first defeat of a grandmaster of chess by a computer. Since 1990 computers have occasionally beat the top players and even world champions Anatoly Karpov and Garry Kasparov. In 1997 the Deep Blue computer—coached by a research team of IBM scientists and engineers, and a grandmaster and U.S. chess champion Joel Benjamin— defeated the reigning world champion (Garry Kasparov) in a classic chess match against a computer for the first time. Kasparov had won a prior match in 1996. A number of computer programs now have a chess rating comparable with the top human players.
Recent Development: Internet and Chess
With the introduction of the Internet, millions of chess players and fans became a global chess community. Chess enthusiasts are able to play online with partners all around the world, browse through the latest chess news, download chess games from databases containing millions of games, and even play with the chess world champion. The "Kasparov versus the World" online chess tournament over the Internet lasted for more than four months from June to October 1999. More than 58,000 individuals from 75 countries submitted their move votes and were able to give a respectable fight to the world champion. The game lasted 62 moves and was observed by more than 3 million web site visitors. The quality of the chess played throughout the game has been recognized by chess experts around the world as one of the best public chess games ever documented.
Growing proliferation of computers and the Internet gave an opportunity to masses of chess devotees to improve their expertise and chess literacy and to widen horizons by providing easy access to chess databases and other information. Used by grandmasters and novices alike, computers prove to be exceptionally handy tools to enjoy the ancient and popular game of chess.
see also Artificial Intelligence; Games; IBM Corporation.
Hamilton, Scott, and Lee Garber. "Deep Blue's Hardware-Software Synergy." Computer 30, no.10 (1997): 29–35.
Krol, Marina. "Have We Witnessed a Real-Life Turing Test?" Computer 32, no.3 (1999): 27–30.
Lawrence, Al, and Lev Alburt. Playing Computer Chess: Getting the Most out of Your Game. New York: Sterling Publishing Company, 1998.
Levy, David N. L., and Monty Newborn. How Computers Play Chess. New York: W. H. Freeman Company, 1990.
Newborn, Monty, and Monroe Newborn. Kasparov Versus Deep Blue: Computer Chess Comes of Age. New York: Springer-Verlag, 1997.
Pecci, Ernest F., and Edward Sheppie. Chess: How to Beat Your Computer. Walnut Creek, CA: Pavior Publishing, 1999.