Computer Chess: Fundamentals and Applications

By Fouad Sabry

What Is Computer Chess

Hardware and software that are both able to play the game of chess are required for computer chess. The game of chess played on a computer offers players the chance to develop their skills even when they do not have human opponents to compete against. It also offers options for analysis, amusement, and training. Applications for computer chess that can play at a level equivalent to or higher than that of a chess master are available for hardware ranging from supercomputers to smart phones. There are other chess-playing machines that are available on their own. Free and open-source software like Stockfish, GNU Chess, and Fruit are all downloadable for a variety of computer operating systems.

How You Will Benefit

(I) Insights, and validations about the following topics:

Chapter 1: Computer Chess

Chapter 2: Artificial Intelligence Projects

Chapter 3: Deep Blue (Chess Computer)

Chapter 4: Chess Engine

Chapter 5: Evaluation Function

Chapter 6: Endgame Tablebase

Chapter 7: Human-Computer Chess Matches

Chapter 8: Stockfish (Chess)

Chapter 9: AlphaZero

Chapter 10: Leela Chess Zero

(II) Answering the public top questions about computer chess.

(III) Real world examples for the usage of computer chess in many fields.

Who This Book Is For

Professionals, undergraduate and graduate students, enthusiasts, hobbyists, and those who want to go beyond basic knowledge or information for any kind of computer chess.

What is Artificial Intelligence Series

The artificial intelligence book series provides comprehensive coverage in over 200 topics. Each ebook covers a specific Artificial Intelligence topic in depth, written by experts in the field. The series aims to give readers a thorough understanding of the concepts, techniques, history and applications of artificial intelligence. Topics covered include machine learning, deep learning, neural networks, computer vision, natural language processing, robotics, ethics and more. The ebooks are written for professionals, students, and anyone interested in learning about the latest developments in this rapidly advancing field.
The artificial intelligence book series provides an in-depth yet accessible exploration, from the fundamental concepts to the state-of-the-art research. With over 200 volumes, readers gain a thorough grounding in all aspects of Artificial Intelligence. The ebooks are designed to build knowledge systematically, with later volumes building on the foundations laid by earlier ones. This comprehensive series is an indispensable resource for anyone seeking to develop expertise in artificial intelligence.

• Language: English
• Publisher: One Billion Knowledgeable
• Release date: Jul 4, 2023

Book preview

Chapter 1: Computer chess

Hardware (special purpose computers) and software (applications designed to mimic human play) are both components of computer chess. Computer chess not only allows players to hone their skills without the presence of human opponents, but also serves as a tool for analysis, entertainment, and education. From supercomputers to smartphones, chess programs that can compete with human grandmasters are readily available. There are also chess-playing machines that stand on their own. Free and open source software like Stockfish, GNU Chess, and Fruit can be downloaded and used on many different computer systems.

Both hardware and software chess programs use heuristic methods to construct, search, and evaluate trees representing sequences of moves from the current position, with the goal of executing the best such sequence during play. Usually, the size of such trees ranges from the thousands to the millions of nodes. Extension and reduction heuristics that narrow the tree to mostly relevant nodes, combined with the computational speed of modern computers that can process tens of thousands to hundreds of thousands of nodes per second, make such an approach effective.

In the early days of digital computing, when vacuum tubes powered computers, software programs were the first chess machines (1950s). The earliest computer programs were so bad at the game that a complete novice could easily beat them. In 1997, chess engines running on supercomputers or specialized hardware were able to beat the best human players. This was only 40 years after the invention of the game. In 2006, the same functionality could be found in software for desktop PCs. Professor of Computer Science at McGill University Monty Newborn declared the science has been done in 2006. However, due to the vastness of the game's permutations, modern computers cannot yet solve chess.

Standalone chess machines (typically a microprocessor running a software chess program, but sometimes as a specialized hardware machine), chess software running on standard PCs, websites, and mobile apps are just some of the ways you can interact with chess machines and programs. The range of possible computing platforms includes both supercomputers and smartphones. The applications typically take up no more than a few megabytes of storage space on disk, utilize a few megabytes of memory (but can use much more, if it is available), and run smoothly on any processor 300Mhz or faster. However, having enough memory to store a large transposition table (up to several gigabytes or more) is more crucial to playing strength than having a fast processor.

Using multi-core and hyperthreaded CPU architectures, most commercially available chess programs and machines can play at super-grandmaster strength (Elo 2700 or higher). Stockfish and other top programs are now stronger than human world champions. Most chess programs, like Winboard and Chessbase, consist of a chess engine coupled with a graphical user interface. The user interface allows for tweaking a variety of performance-related settings, including volume, time limit, and others. In addition to setting up and editing positions, reversing moves, offering and accepting draws (and resigning), requesting and receiving move recommendations, and viewing the engine's analysis in real time, most GUIs also allow the player to set up and edit positions.

Sargon, IPPOLIT, Stockfish, Crafty, Fruit, Leela Chess Zero, and GNU Chess are just some of the chess engines whose source code can be obtained for free on the Internet.

Typical chess software consists of, well, programs that play chess. A human player makes a move on the board, the AI analyzes the situation and responds with its own move, and so on until the game is over. Sometimes, the GUI and the chess engine (the part that does the actual calculation) are two entirely different pieces of software. Connecting various engines to the user interface opens up the possibility of playing against a wide variety of different opponents. While many engines only provide a text command line interface, graphical user interfaces may allow for a wide selection of pieces, boards, and even animations. Because modern engines are so strong, they, or their graphical user interfaces (GUIs), may provide a way to handicap the engine's strength, increasing the human player's chances of victory. Using the uci limitstrength and uci elo parameters provided by the Universal Chess Interface (UCI), engines like Fritz and Rybka may have an in-built mechanism for lowering the engine's Elo rating. One can adjust the current engine's power, error rate, and play style in the Handicap and Fun modes available in some versions of Fritz. Fritz also includes a Friend Mode, where it will try to play at the same skill level as the player.

A chess database is a digital repository of games that can be used to learn opening strategies and tactics. Professional chess players often use the PC program Chessbase for this purpose, but there are other options, such as Shane's Chess Information Database (Scid)

Playing games against other online players is possible with software like Playchess.

Chess is taught in training programs. In Chessmaster, Grandmaster Larry Christiansen and International Master Josh Waitzkin provided playthrough tutorials. The Step chess books by Rob Brunia and Cor Van Wijgerden are the basis for Stefan Meyer-Shredder Kahlen's Chess Tutor. Play Magnus, founded by former world champion Magnus Carlsen, has released an app called Magnus Trainer for Android and iOS. For younger players, Chessbase offers both Fritz and Chesster. Using lessons from Grandmasters Alexander Kalinin and Maxim Blokh, Convekta offers a plethora of training apps like CT-ART and its Chess King line.

Computer programs exist specifically to solve chess puzzles.

It was predicted in 1967 that a computer would beat the world human champion after a team from Carnegie Mellon University discovered refutation screening, the application of alpha-beta pruning to optimizing move evaluation, in 1957. In a 1989 exhibition match, Deep Thought defeated Levy. However, Deep Thought was still a long way off from World Championship level, as the reigning world champion, Garry Kasparov, showed with two convincing wins in 1989. In Deep Blue versus Kasparov, 1996, game 1, Kasparov suffered his first loss to a computer playing under tournament time controls. As a matter of fact, this match marked the first time a reigning world champion had ever lost to a computer under standard time limits. Nonetheless, Kasparov rallied to win the next three games and tie the other two for a commanding victory.

In May 1997, an updated version of Deep Blue defeated Kasparov 3½–2½ in a return match.

In 2003, the confrontation was the primary focus of a documentary, Kasparov vs. Computer: Game Over.

Rising computational capacity and refined assessment tools, Commercially available workstations' chess programs eventually surpassed human grandmasters.

In 1998, Viswanathan Anand was bested by Rebel 10, who was the world's number two at the time, five games to three.

However, Those games were not typically played under regulation time constraints.

Among the eight contests,, Out of the four, two were blitz matches (five minutes plus five seconds Fischer delay for each move); The Rebels prevailed, 3-1.

Two were semi-blitz games (fifteen minutes for each side) that Rebel won as well (1½–½).

Finally, Regular tournament play (forty moves in two hours) was used for two of the games, (Death occurs within an hour); here it was Anand who won ½–1½.

In quick-paced competitions, In a game, computers outperformed humans, advantage was less obvious at classical time controls, however, which are used to determine a player's rating.

Junior and Fritz, two commercially available programs, drew matches against Garry Kasparov and Vladimir Kramnik, two former world champions, in the early 2000s.

The eight-game Brains in Bahrain match between Vladimir Kramnik and Deep Fritz in October 2002 ended in a draw. In games 2 and 3, Kramnik beat the computer by using conventional anti-computer strategies, in which he played cautiously in order to gain a long-term advantage that the program could not detect through its game tree search. However, after Kramnik made a monumental mistake in game 5, Fritz came out on top. The commentators at the tournament thought Game 6 was spectacular. Even though Kramnik had the upper hand in the early middlegame, he made the risky move of sacrificing a piece