Chess is an abstract strategy board game for two players which involves no hidden information and no elements of chance. It is played on a square game board called a chessboard containing 64 squares arranged in an 8×8 grid. The players, referred to as “White” and “Black”, each control sixteen pieces: one king, one queen, two rooks, two bishops, two knights, and eight pawns; each type of piece has a different pattern of movement. An enemy piece may be “captured” (removed from the board) by moving one’s own piece onto the square it occupies; the object of the game is to “checkmate” (threaten with inescapable capture) the enemy king. There are also several ways a game can end in a draw.

The recorded history of chess goes back at least to the emergence of chaturanga—also thought to be an ancestor to similar games like xiangqi and shogi—in seventh-century India. After its introduction in Persia, it spread to the Arab world and then to Europe. The modern rules of chess emerged in Europe at the end of the 15th century, with standardization and universal acceptance by the end of the 19th century. Today, chess is one of the world’s most popular games, with millions of players worldwide.

Organized chess arose in the 19th century. Chess competition today is governed internationally by FIDE (Fédération Internationale des Échecs; the International Chess Federation). The first universally recognized World Chess Champion, Wilhelm Steinitz, claimed his title in 1886; Gukesh Dommaraju is the current World Champion, having won the title in 2024.

A huge body of chess theory has developed since the game’s inception. Aspects of art are found in chess composition, and chess in its turn influenced Western culture and the arts, and has connections with other fields such as mathematics, computer science, and psychology. One of the goals of early computer scientists was to create a chess-playing machine. In 1997, Deep Blue became the first computer to beat a reigning World Champion in a match when it defeated Garry Kasparov. Today’s chess engines are significantly stronger than the best human players and have deeply influenced the development of chess theory; however, chess is not a solved game.

The rules of chess are published by FIDE (Fédération Internationale des Échecs; “International Chess Federation”), chess’s world governing body, in its Handbook.^[2] Rules published by national governing bodies, or by unaffiliated chess organizations, commercial publishers, etc., may differ in some details. FIDE’s rules were most recently revised in 2023.

Chess sets come in a wide variety of styles. The Staunton pattern is the most common, and is usually required for competition.^[3] Chess sets come with pieces in two colors, referred to as white and black, regardless of their actual color; the players controlling the color sets are referred to as White and Black, respectively. Each set comes with at least the following 16 pieces in both colors: one king, one queen, two rooks, two bishops, two knights, and eight pawns.^[2]

The game is played on a square board of eight rows (called ranks) and eight columns (called files). Although it does not affect gameplay, by convention the 64 squares alternate in color and are referred to as light and dark squares.^[2] Common colors for wooden chessboards are light and dark brown, while vinyl chessboards are commonly buff and green.^{[citation needed]}

To start the game, White’s pieces are placed on the first rank in the following order, from left to right: rook, knight, bishop, queen, king, bishop, knight, rook. Pawns are placed on each square of the second rank. Black’s position mirrors White’s, with equivalent pieces on every file.^[2] The board is oriented so that the right-hand corner nearest each player is a light square; as a result the white queen always starts on a light square, while the black queen starts on a dark square. This may be remembered by the phrases “white on the right” and “queen on her color”.^[4]

In formal competition, the piece colors for every matchup are allocated to players by the organizers. In informal games, colors may be decided either by mutual agreement, or randomly, for example by a coin toss, or by one player concealing a white pawn in one hand and a black pawn in the other and having the opponent choose.^{[citation needed]}

White moves first, after which players alternate turns. One piece is moved per turn (except when castling, during which two pieces are moved). In the diagrams, dots mark the squares to which each type of piece can move if unoccupied by friendly pieces and there are no intervening piece(s) of either color (except the knight, which leaps over any intervening pieces). With the sole exception of en passant, a piece captures an enemy piece by moving to the square it occupies, removing it from play and taking its place. The pawn is the only piece that does not capture the way it moves, and it is the only piece that moves and captures in only one direction (forwards from the player’s perspective). A piece is said to control empty squares on which it could capture, attack squares with enemy pieces it could capture, and defend squares with pieces of the same color on which it could recapture. Moving is compulsory; a player may not skip a turn, even when having to move is detrimental.

• The king moves one square in any direction. There is also a special move called castling which moves the king and a rook. The king is the most valuable piece—it is illegal to play any move that puts one’s king under attack by an opponent piece. A move that attacks the king must be parried immediately; if this cannot be done, the game is lost. (See § Check and checkmate.)
• A rook can move any number of squares along a rank or file. A rook is involved in the king’s castling move.
• A bishop can move any number of squares diagonally.
• A queen combines the power of a rook and bishop and can move any number of squares along a rank, file, or diagonal.
• A knight moves to any of the closest squares that are not on the same rank, file, or diagonal. (Thus the move forms an “L”-shape: two squares vertically and one square horizontally, or two squares horizontally and one square vertically.) The knight is the only piece that can leap over other pieces.
• A pawn can move forward to the unoccupied square immediately in front of it on the same file, or on its first move it can optionally advance two squares along the same file, provided both squares are unoccupied (diagram dots). A pawn can capture an opponent’s piece on a square diagonally in front of it by moving to that square (diagram crosses). It cannot capture a piece while advancing along the same file, nor can it move to either square diagonally in front without capturing. Pawns have two special moves: the en passant capture and promotion.

When a king is under immediate attack, it is in check. A move in response to a check is legal only if it results in a position in which the king is no longer in check. There are three ways to counter a check:

• Capture the checking piece.
• Interpose a piece between the checking piece and the king (possible only if the attacking piece is a queen, rook, or bishop and there is a square between it and the king).
• Move the king to a square where it is not under attack.

The object of the game is to checkmate the opponent; this occurs when the opponent’s king is in check, and there is no legal way to get it out of check. In casual games, it is common to announce “check” when putting the opponent’s king in check, but this is not required by the rules of chess and is usually not done in tournaments.^[5]

Kings can castle once per game. Castling consists of moving the king two squares toward either rook of the same color, and then placing the rook on the square that the king crossed.

Castling is possible only if the following conditions are met:^[2]

• Neither the king nor the rook has previously moved during the game.
• There are no pieces between the king and the rook.
• The king is not in check and does not pass through or finish on a square controlled by an enemy piece.

Castling is still permitted if the rook is under attack, or if the rook crosses an attacked square.

Pawns have two special moves:

• En passant: when a pawn makes a two-square advance to the same rank as an opponent’s pawn on an adjacent file, that pawn can capture it en passant (“in passing”), moving to one square behind the captured pawn. A pawn can only be captured en passant on the turn after it makes a two-square advance. In the animated diagram, the black pawn advances two squares from g7 to g5, and the white pawn on f5 takes it en passant, landing on g6.
• Promotion: when a pawn advances to its last rank, it is promoted and replaced with the player’s choice of a queen, rook, bishop, or knight. Usually, pawns are promoted to queens; choosing another piece is called underpromotion. In the animated diagram, the c7-pawn is advanced to c8 and promoted to a queen. If the required piece is not available (e.g. a second queen), an inverted rook is sometimes used as a substitute, but this is not recognized in FIDE-sanctioned games.

A game can be won in the following ways:

• Checkmate: The opposing king is in check and the opponent has no legal move. (See § Check and checkmate.)
• Resignation: A player may resign, conceding the game to the opponent.^[6] If, however, the opponent has no way of checkmating the resigned player, this is a draw under FIDE Laws.^[2] Most tournament players consider it good etiquette to resign in a hopeless position.^[7][8]
• Win on time: In games with a time control, a player wins if the opponent runs out of time, even if the opponent has a superior position, as long as the player has a theoretical possibility to checkmate the opponent were the game to continue.
• Forfeit: A player who cheats, violates the rules, or violates the rules of conduct specified for the particular tournament can be forfeited. Occasionally, both players are forfeited.^[2]

There are several ways a game can end in a draw:

• Stalemate: If the player to move has no legal move, but is not in check, the position is a stalemate, and the game is drawn.
• Dead position: If neither player is able to checkmate the other by any legal sequence of moves, the game is drawn. For example, if only the kings are on the board, all other pieces having been captured, checkmate is impossible, and the game is drawn by this rule. On the other hand, if both players still have a knight, there is a highly unlikely yet theoretical possibility of checkmate, so this rule does not apply. The dead position rule supersedes an older rule which referred to “insufficient material”, extending it to include other positions where checkmate is impossible, such as blocked pawn endings where the pawns cannot be attacked.
• Draw by agreement: In tournament chess, draws are most commonly reached by mutual agreement between the players. The correct procedure is to verbally offer the draw, make a move, then start the opponent’s clock. Traditionally, players have been allowed to agree to a draw at any point in the game, occasionally even without playing a move. More recently efforts have been made to discourage early draws, for example by forbidding draw offers before a certain point.
• Threefold repetition: This most commonly occurs when neither side is able to avoid repeating moves without incurring a disadvantage. The three occurrences of the position need not occur on consecutive moves for a claim to be valid. The addition of the fivefold repetition rule in 2014 requires the arbiter to intervene immediately and declare the game a draw after five occurrences of the same position, consecutive or otherwise, without requiring a claim by either player. FIDE rules make no mention of perpetual check; this is merely a specific type of draw by threefold repetition.
• Fifty-move rule: If during the previous 50 moves no pawn has been moved and no capture has been made, either player can claim a draw. The addition of the seventy-five-move rule in 2014 requires the arbiter to intervene and immediately declare the game drawn after 75 moves without a pawn move or capture, without requiring a claim by either player. There are several known endgames where it is possible to force a mate but it requires more than 50 moves before a pawn move or capture is made; examples include some endgames with two knights against a pawn and some pawnless endgames such as queen against two bishops. Historically, FIDE has sometimes revised the fifty-move rule to make exceptions for these endgames, but these have since been repealed. Some correspondence chess organizations do not enforce the fifty-move rule.^{[note 1]}
• Draw on time: In games with a time control, the game is drawn if a player is out of time and no sequence of legal moves would allow the opponent to checkmate the player.^[2]
• Draw by resignation: Under FIDE Laws, a game is drawn if a player resigns and no sequence of legal moves would allow the opponent to checkmate that player.^[2]

In competition, chess games are played with a time control. Methods and bounds vary, but time controls are generally divided into categories based on the amount of time given to each player (FIDE adds sixty turns worth of increment to the starting time when measuring), which range from classical time controls, which allot an hour or more to each player and which can take upwards of seven hours (even longer if adjournments are permitted), to bullet chess, in which players receive less than three minutes each. Between these are rapid chess (ten to sixty minutes per player), popular in amateur tournaments, and blitz chess (three to ten minutes), popular online. Non-classical chess is sometimes referred to as fast chess.

Time is controlled using a chess clock with two displays, one for each player’s remaining time. Analog chess clocks have been largely replaced by digital clocks, which allow for time controls with increments.

There are some aspects unique to online chess. A premove allows a player to submit a move on the opponent’s turn, which gets played automatically if possible using little to no time. Premoves, alongside the relative ease of digital inputs, make faster time controls feasible online.

Time controls are also enforced in correspondence chess competitions. A typical time control is 50 days for every 10 moves. Time is usually alloted per move in online correspondence chess.

Historically, many different notation systems have been used to record chess moves; the standard system today is short-form algebraic notation.^[10] In this system, files are labeled a through h and ranks are labeled 1 through 8. Squares are identified by the file and rank they occur on; g3 is the square on the g file and the third rank. In English, the piece notations are: K (king), Q (queen), R (rook), B (bishop), and N (knight; N is used to avoid confusion with king). Different initials are used in other languages. Moves are recorded as follows:

notation of piece moved – destination square

For example, Qg5 means “queen moves to g5”. No letter initial is used for pawns, so e4 means “pawn moves to e4”. When multiple moves could be rendered the same way, the file or rank from which the piece moved is added to resolve ambiguity (e.g. Ngf3 means “knight from the g-file moves to the square f3”; R1e2 means “rook on the first rank moves to e2”). If a move may be disambiguated by rank or file, it is done by file, and in the rare case that both are needed, squares are listed normally (e.g Qh4xe1).

If the move is a capture, “x” is usually inserted before the destination square, thus Bxf3 means “bishop captures on f3”. When a pawn makes a capture, the file from which the pawn departed is often listed even when no disambiguation is necessary; for example, exd5.

If a pawn moves to its last rank, achieving promotion, the piece chosen is indicated after the move (for example, e1=Q or e1Q). Castling is indicated by the special notations 0-0 for kingside castling and 0-0-0 for queenside castling. A move that places the opponent’s king in check usually has the notation “+” suffixed. Checkmate can be indicated by suffixing “#“. At the end of the game, “1–0” means White won, “0–1” means Black won, and “½–½” indicates a draw.^[2] Chess moves can be annotated with punctuation marks and other symbols. For example: “!” indicates a good move; “!!” a excellent move; “?” a mistake; “??” a blunder; “!?” an interesting move that may not be best; or “?!” a dubious move not easily refuted.^[11]

Moves are written as white/black pairs, preceded by the move number and a period. Individual white moves are also recorded this way, while black moves are rendered with an ellipsis after the move number. For example, one variation of a simple trap known as the Scholar’s mate (see animated diagram) can be recorded:

 1. e4 e5 2. Qh5?! Nc6 3. Bc4 Nf6?? 4. Qxf7#

The move 3… Nf6?? is recorded as a blunder, as it allows 4. Qxf7# checkmate.

Games or sequences may be recorded in Portable Game Notation (PGN), a text-based file format with support for annotative symbols, commentary, and background information, such as player names. It is based on short form English algebraic notation incorporating markup language. PGN transcripts, stored digitally as PGN (.pgn) files can be processed by most chess software and are easily readable by humans.

Variants of algebraic notation include long algebraic, in which both the departure and destination square are indicated; abbreviated algebraic, in which capture signs, check signs, and ranks of pawn captures may be omitted; and figurine algebraic notation, used in chess books and magazines, which uses graphic symbols instead of initials to indicate pieces for readability regardless of language.

Until about 1980, the majority of English language chess publications used descriptive notation, in which files are identified by the initial letter of the piece that occupies the first rank at the beginning of the game. In descriptive notation, the common opening move 1.e4 is rendered as “1.P-K4” (“pawn to king four”). Another system is ICCF numeric notation, recognized by the International Correspondence Chess Federation though its use is in decline.

In tournament games, players are normally required to keep a score (written record of the game). This is a requirement in all FIDE-sanctioned games played at classical time controls.^[2] For this purpose, only algebraic notation is recognized by FIDE, though variants such as long algebraic are acceptable; game scores recorded in a different notation system may not be used as evidence in the event of a dispute.

Chess has an extensive literature. In 1913, the chess historian H.J.R. Murray estimated the total number of books, magazines, and chess columns in newspapers to be about 5,000.^[12] B.H. Wood estimated the number, as of 1949, to be about 20,000.^[13] David Hooper and Kenneth Whyld write that, “Since then there has been a steady increase year by year of the number of new chess publications. No one knows how many have been printed.”^[14] Significant public chess libraries include the John G. White Chess and Checkers Collection at Cleveland Public Library, with over 32,000 chess books and over 6,000 bound volumes of chess periodicals;^[15] and the Chess & Draughts collection at the National Library of the Netherlands, with about 30,000 books.^[16]

Example of underlying pawn structure

a b c d e f g h 8 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 a b c d e f g h Position after 12…Re8 … Tarrasch vs. Euwe, Bad Pistyan (1922)[17] a b c d e f g h 8 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 a b c d e f g h … and its pawn structure, known as the Rauzer formation

Chess strategy is concerned with the evaluation of chess positions and with setting up goals and long-term plans for future play. During the evaluation, players must take into account numerous factors such as the value of the pieces on the board, control of the center and centralization, the pawn structure, king safety, and the control of key squares or groups of squares (for example, diagonals, open files, and dark or light squares).

The most basic step in evaluating a position is to count the total value of pieces of both sides.^[18] The point values used for this purpose are based on experience; usually, pawns are considered worth one point, knights and bishops about three points each, rooks about five points (the value difference between a rook and a bishop or knight being known as the exchange), and queens about nine points. The king is more valuable than all of the other pieces combined, since its checkmate loses the game, but is still capable as a fighting piece; in the endgame, the king is generally more powerful than a bishop or knight but less powerful than a rook.^[19] These basic values are then modified by other factors like position of the piece (e.g. advanced pawns are usually more valuable than those on their initial squares), coordination between pieces (e.g. a pair of bishops usually coordinate better than a bishop and a knight), or the type of position (e.g. knights are generally better in closed positions with many pawns while bishops are more powerful in open positions).^[20]

Another important factor in the evaluation of chess positions is pawn structure (sometimes known as the pawn skeleton): the configuration of pawns on the chessboard.^[21] Since pawns are the least mobile of the pieces, pawn structure is relatively static and largely determines the strategic nature of the position. Weaknesses in pawn structure include isolated, doubled, or backward pawns and holes; once created, they are often permanent. Care must therefore be taken to avoid these weaknesses unless they are compensated by another valuable asset (for example, by the possibility of developing an attack).^[22]

In chess, tactics generally refer to short-term maneuvers – so short-term that they can be calculated in advance by a human player. The possible depth of calculation depends on the player’s ability. In quiet positions with many possibilities on both sides, a deep calculation is more difficult and may not be practical, while in positions with a limited number of forced variations, strong players can calculate long sequences of moves.

Theoreticians describe many elementary tactical methods and typical maneuvers, for example: pins, forks, skewers, batteries, discovered attacks (especially discovered checks), zwischenzugs, deflections, decoys, sacrifices, underminings, overloadings, and interferences.^[23] Simple one-move or two-move tactical actions – threats, exchanges of material, and double attacks – can be combined into longer sequences of tactical maneuvers that are often forced from the point of view of one or both players. A forced variation that involves a sacrifice and usually results in a tangible gain is called a combination.^[24] Brilliant combinations – such as those in the Immortal Game – are considered beautiful and are admired by chess lovers.

A common type of chess exercise, aimed at developing players’ tactical skills, is a position where a combination is available and the challenge is to find it. Such positions are usually taken from actual games or from analysis of actual games. Solutions usually result in checkmate, decisive advantage, or successful defense. Tactical exercises are commonly found in instructional books, chess magazines, newspaper chess columns, and internet chess sites.^[25]

Chess theory divides chess games into three phases with different sets of strategies: the opening, the middlegame, and lastly the endgame. There is no universally accepted way to delineate the three phases of the game; the middlegame is typically considered to have begun after 10–20 moves, and the endgame when only a few pieces remain.

Competitive players typically learn, memorize, and play well-documented sequences of opening moves. The most common starting moves for White are 1.e4 and 1.d4, which usually lead to substantially different types of positions, and Black has multiple viable responses to both.

Sequences of opening moves are referred to as openings and are catalogued in reference works, such as the Encyclopaedia of Chess Openings. There are thousands of openings, though only a small fraction of them are commonly played; variations of openings may also be given names. Openings vary widely in character from quiet positional play (for example, the Réti Opening) to sharp aggressive play (like the Latvian Gambit). In some opening lines, the exact sequence considered best for both sides has been worked out to more than 30 moves.^[26]

Amateur players generally have little knowledge of opening theory, and may use heuristic strategies, such as playing conservatively to minimize the possibility of losing early to a prepared trap, though not all do so. Rigorous study is necessary for professional players, who often employ assistants called seconds to aid them in preparation. While top players typically play reputed variations of openings, they may also study dubious variations in the hopes that their opponent will overlook them in their own studies. The role of memorization in openings is sometimes controversial for delaying the onset of deliberation and creativity in chess games.

The fundamental strategic aims of most openings are similar:^[27]

• Development: moving pieces (particularly bishops and knights) forward to squares on which they are useful (defending, attacking, and controlling important squares) and/or have the potential to take part in future plans and ideas.
• Control of the center: control of the central squares allows pieces to be moved to any part of the board relatively easily, and can inhibit the mobility of the opponent’s pieces.
• King safety: typically secured by castling; incorrectly timed castling can be wasteful or even harmful, however.
• Pawn structure: players strive to avoid the creation of pawn weaknesses such as isolated, doubled, or backward pawns – and to force such weaknesses in the opponent’s position.

Most players and theoreticians consider that White, by virtue of the initiative granted from moving first, begins the game with a small advantage.^[28] Black usually strives to neutralize White’s advantage and achieve equality, or to develop dynamic counterplay in an unbalanced position.

The middlegame is the part of the game that starts after the opening. Because the opening theory has ended, players have to form plans based on the features of the position, and at the same time take into account the tactical possibilities of the position.^[29] The middlegame is the phase in which most combinations occur. Combinations are a series of tactical moves executed to achieve some gain. Middlegame combinations are often connected with an attack against the opponent’s king. Some typical patterns have their own names; for example, the Boden’s Mate or the Lasker–Bauer combination.^[30]

Specific plans or strategic themes will often arise from particular groups of openings that result in a specific type of pawn structure. An example is the minority attack, which is the attack of queenside pawns against an opponent who has more pawns on the queenside. The study of openings is therefore connected to the preparation of plans that are typical of the resulting middlegames.^[31]

Another important strategic question in the middlegame is whether and how to reduce material and transition into an endgame (i.e. simplify). Minor material advantages can generally be transformed into victory only in an endgame, and therefore the stronger side must choose an appropriate way to achieve an ending. Not every reduction of material is good for this purpose; for example, if one side keeps a light-squared bishop and the opponent has a dark-squared one, the transformation into a bishops and pawns ending is usually advantageous for the weaker side only, because an endgame with bishops on opposite colors is likely to be a draw, even with an advantage of a pawn, or sometimes even with a two-pawn advantage.^[32]

The endgame (also end game or ending) is the stage of the game when there are few pieces left on the board. There are three main strategic differences between earlier stages of the game and the endgame:^[33]

• Pawns become more important. Endgames often revolve around endeavors to promote a pawn by advancing it to the furthest rank.
• The king, which requires safeguarding from attack during the middlegame, emerges as a strong piece in the endgame. It is often used to protect its own pawns, attack enemy pawns, and hinder moves of the opponent’s king.
• Zugzwang, a situation in which the player who is to move is forced to incur a disadvantage, is often a factor in endgames but rarely in other stages of the game. In the example diagram, either side having the move is in zugzwang: Black to move must play 1…Kb7 allowing White to promote the pawn after 2.Kd7; White to move must permit a draw, either by 1.Kc6 stalemate or by losing the pawn after any other legal move.

Endgames can be classified according to the type of pieces remaining on the board. Basic checkmates are positions in which one side has only a king and the other side has one or two pieces and can checkmate the opposing king, with the pieces working together with their king. For example, king and pawn endgames involve only kings and pawns on one or both sides, and the task of the stronger side is to promote one of the pawns. Other more complicated endings are classified according to pieces on the board other than kings, such as “rook and pawn versus rook” endgames.

Chess problems (also called chess compositions) are composed positions, usually created for artistic effect rather than practical application. The creator is known as a chess composer.^[35]

There are many types of chess problems, the most common being directmates, in which White is required to move and checkmate Black within a specified number of moves, usually two or three, against any defense.

These are commonly referred to as “two-movers”, “three-movers”, or “more-movers”. “Many-movers” (also known as “long-range problems”) of over 100 moves have been composed, the current record standing at over 200; these usually require repetitions of the same manoeuvre in order to produce a repeated zugzwang and force detrimental pawn advances.^[36][37]

Directmates usually consist of positions unlikely to occur in an actual game, and are intended to illustrate a particular theme, usually requiring a surprising or counterintuitive key move. Themes associated with chess problems occasionally appear in actual games, when they are referred to as “problem-like” moves.^[38]

Other common types of problems include:

• Helpmates, in which Black moves first and cooperates with White to get Black’s king checkmated
• Selfmates, in which White moves first and forces Black to checkmate White
• Retrograde analysis problems, in which the solver is required to work out what has previously occurred in the game, for example to prove that castling is illegal in the current position

The above type of problems are usually considered orthodox, in the sense that the standard rules of chess are observed.

Fairy chess problems, also called heterodox problems, involve altered rules, such as the use of unconventional pieces or boards, or stipulations that contradict the standard rules of chess such as reflexmates or seriesmovers.

Studies are usually considered distinct from problems, although there is some overlap. In a study, the stipulation is that White to play must win or draw, without specifying any particular number of moves. The majority of studies are endgame positions, with varying degrees of realism or practical application.^[39]

Tournaments for composition and solving of chess problems and studies are organized by the World Federation for Chess Composition (WFCC), which works cooperatively with but independent of FIDE. The WFCC awards titles for composing and solving chess problems.^[40]

Chess is often played in public spaces such as parks and town squares. Although the nature of these games is often casual, the chess hustling scene has seen growth in urban areas such as New York City.^[41]

Contemporary chess is an organized sport with structured international and national leagues, tournaments, and congresses. Thousands of chess tournaments, matches, and festivals are held around the world every year catering to players of all levels.

Tournaments with a small number of players may use the round-robin format, in which every player plays one game against every other player. For a large number of players, the Swiss system may be used, in which each player is paired against an opponent who has the same (or as similar as possible) score in each round. In either case, a player’s score is usually calculated as 1 point for each game won and one-half point for each game drawn. Variations such as “football scoring” (3 points for a win, 1 point for a draw) may be used by tournament organizers, but ratings are always calculated on the basis of standard scoring. A player’s score may be reported as total score out of games played (e.g. 5½/8), points for versus points against (e.g. 5½–2½), or by number of wins, losses and draws (e.g. +4−1=3).

The term “match” refers not to an individual game, but to either a series of games between two players, or a team competition in which each player of one team plays one game against a player of the other team.

Chess’s international governing body is usually known by its French acronym FIDE (pronounced FEE-day) (French: Fédération Internationale des Échecs), or International Chess Federation. FIDE’s membership consists of the national chess organizations of over 180 countries; there are also several associate members, including various supra-national organizations, the International Braille Chess Association (IBCA), International Chess Committee of the Deaf (ICCD), and the International Physically Disabled Chess Association (IPCA).^[42] FIDE is recognized as a sports governing body by the International Olympic Committee,^[43] but chess has never been part of the Olympic Games.

FIDE’s most visible activity is organizing the World Chess Championship, a role it assumed in 1948. The current World Champion is Gukesh Dommaraju of India.^[44][45] The reigning Women’s World Champion is Ju Wenjun from China.^[46]

Other competitions for individuals include the World Junior Chess Championship, the European Individual Chess Championship, the tournaments for the World Championship qualification cycle, and the various national championships. Invitation-only tournaments regularly attract the world’s strongest players. Examples include Spain’s Linares event, Monte Carlo’s Melody Amber tournament, the Dortmund Sparkassen meeting, Sofia’s M-tel Masters, and Wijk aan Zee‘s Tata Steel tournament.

Regular team chess events include the Chess Olympiad and the European Team Chess Championship.

The World Chess Solving Championship and World Correspondence Chess Championship include both team and individual events. These are held independently of FIDE by, respectively, the World Federation for Chess Composition (WFCC), and the International Correspondence Chess Federation (ICCF).

In order to rank players, FIDE, ICCF, and most national chess organizations use the Elo rating system developed by Arpad Elo. An average club player has a rating of about 1500; the highest FIDE rating of all time, 2882, was achieved by Magnus Carlsen on the March 2014 FIDE rating list.^[47]

Players may be awarded lifetime titles by FIDE:^[49]

• Grandmaster (GM) is the highest title a chess player can attain. For the GM title, a player must have had an Elo rating of 2500 or more at least once and must achieve three results of a prescribed standard (called norms) in tournaments involving other grandmasters, including some from countries other than the applicant’s. There are other milestones that can substitute for norms, such as winning the World Junior Championship.
• International Master (IM). The conditions are similar to GM, but less demanding. The minimum rating for the IM title is 2400.
• FIDE Master (FM). The usual way for a player to qualify for the FIDE Master title is by achieving a FIDE rating of 2300 or more.
• Candidate Master (CM). Similar to FM, but with a FIDE rating of at least 2200.

The above titles are known as “open” titles, obtainable by both men and women. There are also separate women-only titles; Woman Grandmaster (WGM), Woman International Master (WIM), Woman FIDE Master (WFM) and Woman Candidate Master (WCM). These require a performance level approximately 200 rating points below their respective open titles, and their continued existence has sometimes been controversial.^{[citation needed]} Beginning with Nona Gaprindashvili in 1978, a number of women have earned the open GM title: 40 as of July 2023.^{[note 2]}

FIDE also awards titles for arbiters and trainers.^[50][51] International titles are also awarded to composers and solvers of chess problems and to correspondence chess players (by the International Correspondence Chess Federation). National chess organizations may also award titles.

Texts referring to the origins of chess date from the beginning of the seventh century. Three are written in Pahlavi (Middle Persian)^[54] and one, the Harshacharita, is in Sanskrit.^[55] One of these texts, the Chatrang-namak, represents one of the earliest written accounts of chess. The narrator Bozorgmehr explains that Chatrang, “Chess” in Pahlavi, was introduced to Persia by ‘Dewasarm, a great ruler of India’ during the reign of Khosrow I:^[56]

Dewasarm has fashioned this chatrang after the likeness of a battle, and in its likeness are two supreme rulers after the likeness of Kings (shah), with the essentials of rooks (rukh) to right and to left, with Counsellor (farzin) in the likeness of a commander of the champions, with the Elephant (pil) in the likeness of the commander of the rearguard, with Horse (asp) in the likeness of the commander of the cavalry, with the Footsoldier (piyadak) in the likeness of so many infantry in the vanguard of the battle.

— Translation by Murray, 1913.^[57]

The oldest known chess manual was in Arabic and dates to about 840, written by al-Adli ar-Rumi (800–870), a renowned Arab chess player, titled Kitab ash-shatranj (The Book of Chess). This is a lost manuscript, but is referenced in later works.^[58] Here also, al-Adli attributes the origins of Persian chess to India, along with the eighth-century collection of fables Kalīla wa-Dimna.^[59] By the 20th century, a substantial consensus^[60][61] developed regarding chess’s origins in northwest India in the early seventh century.^[62] More recently, this consensus has been the subject of further scrutiny.^[63]

The early forms of chess in India were known as chaturaṅga (Sanskrit: चतुरङ्ग), literally “four divisions” [of the military] – infantry, cavalry, elephants, and chariotry – represented by pieces that would later evolve into the modern pawn, knight, bishop, and rook, respectively. Chaturanga was played on an 8×8 uncheckered board, called ashtāpada.^[64] Thence it spread eastward and westward along the Silk Road. The earliest evidence of chess is found in nearby Sasanian Persia around 600 A.D., where the game came to be known by the name chatrang (Persian: چترنگ).^[65] Chatrang was taken up by the Muslim world after the Islamic conquest of Persia (633–51), where it was then named shatranj (Arabic: شطرنج; Persian: شترنج), with the pieces largely retaining their Persian names. In Spanish, “shatranj” was rendered as ajedrez (“al-shatranj”), in Portuguese as xadrez, and in Greek as ζατρίκιον (zatrikion, which comes directly from the Persian chatrang),^[66] but in the rest of Europe it was replaced by versions of the Persian shāh (“king”), from which the English words “check” and “chess” descend.^{[note 3]} The word “checkmate” is derived from the Persian shāh māt (“the king is dead”).^[67]

Xiangqi is the form of chess best known in China. The eastern migration of chess, into China and Southeast Asia, has even less documentation than its migration west, making it largely conjectured. The word xiàngqí (象棋) was used in China to refer to a game from 569 A.D. at the latest, but it has not been proven that this game was directly related to chess.^[68][69] The first reference to Chinese chess appears in a book entitled Xuánguaì Lù (玄怪錄; “Record of the Mysterious and Strange”), dating to about 800. A minority view holds that Western chess arose from xiàngqí or one of its predecessors.^[70][71] Chess historians Jean-Louis Cazaux and Rick Knowlton contend that xiangqi’s intrinsic characteristics make it easier to construct an evolutionary path from China to India/Persia than the opposite direction.^[72]

The oldest archaeological chess artifacts – ivory pieces – were excavated in ancient Afrasiab, today’s Samarkand, in Uzbekistan, Central Asia, and date to about 760, with some of them possibly being older. Remarkably, almost all findings of the oldest pieces come from along the Silk Road, from the former regions of the Tarim Basin (today’s Xinjiang in China), Transoxiana, Sogdiana, Bactria, Gandhara, to Iran on one end and to India through Kashmir on the other.^[73]

The game reached Western Europe and Russia via at least three routes, the earliest being in the ninth century. By the year 1000, it had spread throughout both the Muslim Iberia and Latin Europe.^[75] A Latin poem called Versus de scachis (“Verses on Chess”) dated to the late 10th century, has been preserved at Einsiedeln Abbey in Switzerland.

The game of chess was then played and known in all European countries. A famous 13th-century Spanish manuscript covering chess, backgammon, and dice is known as the Libro de los juegos, which is the earliest European treatise on chess as well as being the oldest document on European tables games.^[76] The rules were fundamentally similar to those of the Arabic shatranj. The differences were mostly in the use of a checkered board instead of a plain monochrome board used by Arabs and the habit of allowing some or all pawns to make an initial double step. In some regions, the queen, which had replaced the wazir, or the king could also make an initial two-square leap under some conditions.^[77]

Around 1200, the rules of shatranj started to be modified in Europe, culminating, several major changes later, in the emergence of modern chess practically as it is known today.^[78] A major change was the modern piece movement rules, which began to appear in intellectual circles in Valencia, Spain, around 1475,^{[note 4]} which established the foundations and brought it very close to current chess. These new rules then were quickly adopted in Italy and Southern France before diffusing into the rest of Europe.^[81][82] Pawns gained the ability to advance two squares on their first move, while bishops and queens acquired their modern movement powers. The queen replaced the earlier vizier chess piece toward the end of the 10th century and by the 15th century had become the most powerful piece;^[83] in light of that, modern chess was often referred to at the time as “Queen’s Chess” or “Mad Queen Chess”.^[84] Castling, derived from the “king’s leap”, usually in combination with a pawn or rook move to bring the king to safety, was introduced. These new rules quickly spread throughout Western Europe.

Writings about chess theory began to appear in the late 15th century. An anonymous treatise on chess of 1490 with the first part containing some openings and the second 30 endgames is deposited in the library of the University of Göttingen.^[85] The book El Libro dels jochs partitis dels schachs en nombre de 100 was written by Francesc Vicent in Segorbe in 1495, but no copy of this work has survived.^[85] The Repetición de Amores y Arte de Ajedrez (Repetition of Love and the Art of Playing Chess) by Spanish churchman Luis Ramírez de Lucena was published in Salamanca in 1497.^[82] Lucena and later masters like Portuguese Pedro Damiano, Italians Giovanni Leonardo Di Bona, Giulio Cesare Polerio and Gioachino Greco, and Spanish bishop Ruy López de Segura developed elements of opening theory and started to analyze simple endgames.

In the 18th century, the center of European chess life moved from Southern Europe to mainland France.^{[citation needed]} The two most important French masters were François-André Danican Philidor, a musician by profession, who discovered the importance of pawns for chess strategy, and later Louis-Charles Mahé de La Bourdonnais, who won a famous series of matches against Irish master Alexander McDonnell in 1834.^[86] Centers of chess activity in this period were coffee houses in major European cities like Café de la Régence in Paris and Simpson’s Divan in London.^[87][88]

At the same time, the Romantic intellectual movement had had a far-reaching impact on chess, with aesthetics and tactical beauty being held in higher regard than objective soundness and strategic planning. As a result, virtually all games began with the Open Game, and it was considered unsportsmanlike to decline gambits that invited tactical play such as the King’s Gambit and the Evans Gambit.^[89] This chess philosophy is known as Romantic chess, and its sharp, tactical style of play was predominant until the late 19th century.^[90]

The rules concerning stalemate were finalized in the early 19th century. Also in the 19th century, the convention that White moves first was established (formerly either White or Black could move first). Finally, the rules around castling and en passant captures were standardized – variations in these rules persisted in Italy until the late 19th century. The resulting standard game is sometimes referred to as Western chess^[91] or international chess,^[92] particularly in Asia where other games of the chess family such as xiangqi are prevalent. Since the 19th century, the only rule changes, such as the establishment of the correct procedure for claiming a draw by repetition, have been technical in nature.

As the 19th century progressed, chess organization developed quickly. Many chess clubs, chess books, and chess journals appeared. There were correspondence matches between cities; for example, the London Chess Club played against the Edinburgh Chess Club in 1824.^[93] Chess problems became a regular part of 19th-century newspapers; Bernhard Horwitz, Josef Kling, and Samuel Loyd composed some of the most influential problems. In 1843, von der Lasa published his and Bilguer’s Handbuch des Schachspiels (Handbook of Chess), the first comprehensive manual of chess theory.

The first modern chess tournament was organized by Howard Staunton, a leading English chess player, and was held in London in 1851. It was won by the German Adolf Anderssen, who was hailed as the leading chess master. His brilliant, energetic attacking style was typical for the time.^[94][95] Sparkling games like Anderssen’s Immortal Game and Evergreen Game or Morphy’s “Opera Game” were regarded as the highest possible summit of the art of chess.^[96]

Deeper insight into the nature of chess came with the American Paul Morphy, an extraordinary chess prodigy. Morphy won against all important competitors (except Staunton, who refused to play), including Anderssen, during his short chess career between 1857 and 1863. Morphy’s success stemmed from a combination of brilliant attacks and sound strategy; he intuitively knew how to prepare attacks.^[97]

Prague-born Wilhelm Steinitz laid the foundations for a scientific approach to the game, the art of breaking a position down into components^[99] and preparing correct plans.^[100] In addition to his theoretical achievements, Steinitz founded an important tradition: his triumph over the leading German master Johannes Zukertort in 1886 is regarded as the first official World Chess Championship. This win marked a stylistic transition at the highest levels of chess from an attacking, tactical style predominant in the Romantic era to a more positional, strategic style introduced to the chess world by Steinitz. Steinitz lost his crown in 1894 to a much younger player, the German mathematician Emanuel Lasker, who maintained this title for 27 years, the longest tenure of any world champion.^[101]

After the end of the 19th century, the number of master tournaments and matches held annually quickly grew. The first Olympiad was held in Paris in 1924, and FIDE was founded initially for the purpose of organizing that event. In 1927, the Women’s World Chess Championship was established; the first to hold the title was Czech-English master Vera Menchik.^[102]

A prodigy from Cuba, José Raúl Capablanca, known for his skill in endgames, won the World Championship from Lasker in 1921. Capablanca was undefeated in tournament play for eight years, from 1916 to 1924. His successor (1927) was the Russian-French Alexander Alekhine, a strong attacking player who died as the world champion in 1946. Alekhine briefly lost the title to Dutch player Max Euwe in 1935 and regained it two years later.^[103]

In the interwar period, chess was revolutionized by the new theoretical school of so-called hypermodernists like Aron Nimzowitsch and Richard Réti. They advocated controlling the center of the board with distant pieces rather than with pawns, thus inviting opponents to occupy the center with pawns, which become objects of attack.^[104] Among the innovations popularized by hypermodernists was the fianchetto: the development of bishops away from, rather than towards, the center, onto the b- and g-files.

After the death of Alekhine, a new World Champion was sought. FIDE, which has controlled the title since then, ran a tournament of elite players. The winner of the 1948 tournament was Russian Mikhail Botvinnik. In 1950, FIDE established a system of titles, conferring the title of Grandmaster on 27 players. (Some sources state that, in 1914, the title of chess Grandmaster was first formally conferred by Tsar Nicholas II of Russia to Lasker, Capablanca, Alekhine, Tarrasch, and Marshall, but this is a disputed claim.^{[note 5]})

Botvinnik started an era of Soviet dominance in the chess world, which mainly through the Soviet government’s politically inspired efforts to demonstrate intellectual superiority over the West^[105][106] stood almost uninterrupted for more than a half-century. Until the dissolution of the Soviet Union, there was only one non-Soviet champion, American Bobby Fischer (champion 1972–1975).^[107] Botvinnik also revolutionized opening theory. Previously, Black strove for equality, attempting to neutralize White’s first-move advantage. As Black, Botvinnik strove for the initiative from the beginning.^[108] In the previous informal system of World Championships, the current champion decided which challenger he would play for the title and the challenger was forced to seek sponsors for the match. FIDE set up a new system of qualifying tournaments and matches. The world’s strongest players were seeded into Interzonal tournaments, where they were joined by players who had qualified from Zonal tournaments. The leading finishers in these Interzonals would go through the “Candidates” stage, which was initially a tournament, and later a series of knockout matches. The winner of the Candidates would then play the reigning champion for the title. A champion defeated in a match had a right to play a rematch a year later. This system operated on a three-year cycle. Botvinnik participated in championship matches over a period of fifteen years. He won the world championship tournament in 1948 and retained the title in tied matches in 1951 and 1954. In 1957, he lost to Vasily Smyslov, but regained the title in a rematch in 1958. In 1960, he lost the title to the 23-year-old Latvian prodigy Mikhail Tal, an accomplished tactician and attacking player who is widely regarded as one of the most creative players ever,^[109] hence his nickname “the magician from Riga”. Botvinnik again regained the title in a rematch in 1961.

Following the 1961 event, FIDE abolished the automatic right of a deposed champion to a rematch, and the next champion, Armenian Tigran Petrosian, a player renowned for his defensive and positional skills, held the title for two cycles, 1963–1969. His successor, Boris Spassky from Russia (champion 1969–1972), won games in both positional and sharp tactical style.^[110] The next championship, the so-called Match of the Century, saw the first non-Soviet challenger since World War II, American Bobby Fischer. Fischer defeated his opponents in the Candidates matches by unheard-of margins, and convincingly defeated Spassky for the world championship. The match was followed closely by news media of the day, leading to a surge in popularity for chess; it also held significant political importance at the height of the Cold War, with the match being seen by both sides as a microcosm of the conflict between East and West.^[111] In 1975, however, Fischer refused to defend his title against Soviet Anatoly Karpov when he was unable to reach agreement on conditions with FIDE, and Karpov obtained the title by default.^[112] Fischer modernized many aspects of chess, especially by extensively preparing openings.^[113]

Karpov defended his title twice against Viktor Korchnoi and dominated the 1970s and early 1980s with a string of tournament successes.^[114] In the 1984 World Chess Championship, Karpov faced his toughest challenge to date, the young Garry Kasparov from Baku, Soviet Azerbaijan. The match was aborted in controversial circumstances after 5 months and 48 games with Karpov leading by 5 wins to 3, but evidently exhausted; many commentators believed Kasparov, who had won the last two games, would have won the match had it continued. Kasparov won the 1985 rematch. Kasparov and Karpov contested three further closely fought matches in 1986, 1987 and 1990, Kasparov winning them all.^[115] Kasparov became the dominant figure of world chess from the mid-1980s until his retirement from competition in 2005.

Chess-playing computer programs (later known as chess engines) began to appear in the 1960s. In 1970, the first major computer chess tournament, the North American Computer Chess Championship, was held, followed in 1974 by the first World Computer Chess Championship. In the late 1970s, dedicated home chess computers such as Fidelity Electronics’ Chess Challenger became commercially available, as well as software to run on home computers. The overall standard of computer chess was low, however, until the 1990s.

The first endgame tablebases, which provided perfect play for relatively simple endgames such as king and rook versus king and bishop, appeared in the late 1970s. This set a precedent to the complete six- and seven-piece tablebases that became available in the 2000s and 2010s respectively.^[116]

The first commercial chess database, a collection of chess games searchable by move and position, was introduced by the German company ChessBase in 1987.^[117] Databases containing millions of chess games have since had a profound effect on opening theory and other areas of chess research.

Digital chess clocks were invented in 1973, though they did not become commonplace until the 1990s. Digital clocks allow for time controls involving increments and delays.

The Internet enabled online chess as a new medium of playing, with chess servers allowing users to play other people from different parts of the world in real time. The first such server, known as Internet Chess Server (ICS), was developed at the University of Utah in 1992. ICS formed the basis for the first commercial chess server, the Internet Chess Club, which was launched in 1995, and for other early chess servers such as Free Internet Chess Server (FICS). Since then, many other platforms have appeared, and online chess began to rival over-the-board chess in popularity.^[118][119] During the 2020 COVID-19 pandemic, the isolation ensuing from quarantines imposed in many places around the world, combined with the success of the popular Netflix show The Queen’s Gambit and other factors such as the popularity of online tournaments (notably PogChamps) and chess Twitch streamers, resulted in a surge of popularity not only for online chess, but for the game of chess in general; this phenomenon has been referred to in the media as the 2020 online chess boom.^[120][121]

Computer chess has also seen major advances. By the 1990s, chess engines could consistently defeat most amateurs, and in 1997 Deep Blue defeated World Champion Garry Kasparov in a six-game match, starting an era of computer dominance at the highest level of chess. In the 2010s, engines significantly stronger than even the best human players became accessible for free on a number of PC and mobile platforms, and free engine analysis became a commonplace feature on internet chess servers. An adverse effect of the easy availability of engine analysis on hand-held devices and personal computers has been the rise of computer cheating, which has grown to be a major concern in both over-the-board and online chess.^[122] In 2017, AlphaZero – a neural network also capable of playing shogi and Go – was introduced. Since then, many chess engines based on neural network evaluation have been written, the best of which have surpassed the traditional “brute-force” engines. AlphaZero also introduced many novel ideas and ways of playing the game, which affected the style of play at the top level.^[123]

As endgame tablebases developed, they began to provide perfect play in endgame positions in which the game-theoretical outcome was previously unknown, such as positions with king, queen and pawn against king and queen. In 1991, Lewis Stiller published a tablebase for select six-piece endgames,^[124][125] and by 2005, following the publication of Nalimov tablebases, all six-piece endgame positions were solved. In 2012, Lomonosov tablebases were published which solved all seven-piece endgame positions.^[126] Use of tablebases enhances the performance of chess engines by providing definitive results in some branches of analysis.

Technological progress made in the 1990s and the 21st century has influenced the way that chess is studied at all levels, as well as the state of chess as a spectator sport.

Previously, preparation at the professional level required an extensive chess library and several subscriptions to publications such as Chess Informant to keep up with opening developments and study opponents’ games. Today, preparation at the professional level involves the use of databases containing millions of games, and engines to analyze different opening variations and prepare novelties.^[127] A number of online learning resources are also available for players of all levels, such as online courses, tactics trainers, and video lessons.^[128]

Since the late 1990s, it has been possible to follow major international chess events online, the players’ moves being relayed in real time. Sensory boards have been developed to enable automatic transmission of moves. Chess players will frequently run engines while watching these games, allowing them to quickly identify mistakes by the players and spot tactical opportunities. While in the past the moves have been relayed live, today chess organizers will often impose a half-hour delay as an anti-cheating measure. In the mid-to-late 2010s – and especially following the 2020 online boom – it became commonplace for supergrandmasters, such as Hikaru Nakamura and Magnus Carlsen, to livestream chess content on platforms such as Twitch.^[129][130] Also following the boom, online chess started being viewed as an esport, with esport teams signing chess players for the first time in 2020.^[131] The number of esport teams signing chess players rose considerably in 2025, after chess was added to Saudi Arabia‘s Esports World Cup.^[132][133]

Organized chess even for young children has become common. FIDE holds world championships for age levels down to 8 years old. The largest tournaments, in number of players, are those held for children.^[134]

The number of grandmasters and other chess professionals has also grown in the modern era. Kenneth Regan and Guy Haworth conducted research involving comparison of move choices by players of different levels and from different periods with the analysis of strong chess engines. They concluded that the increase in the number of grandmasters and higher Elo ratings of the top players reflect an actual increase in the average standard of play, rather than “rating inflation” or “title inflation”.^[135]

In 1993, Garry Kasparov and Nigel Short broke ties with FIDE to organize their own match for the World Championship and formed a competing Professional Chess Association (PCA). From then until 2006, there were two simultaneous World Championships and respective World Champions: the PCA or “classical” champions extending the Steinitzian tradition in which the current champion plays a challenger in a series of games, and the other following FIDE’s new format of many players competing in a large knockout tournament to determine the champion. Kasparov lost his PCA title in 2000 to Vladimir Kramnik of Russia.^[136] Due to the complicated state of world chess politics and difficulties obtaining commercial sponsorships, Kasparov was never able to challenge for the title again. Despite this, he continued to dominate in top level tournaments and remained the world’s highest rated player until his retirement from competitive chess in 2005.

The World Chess Championship 2006, in which Kramnik beat the FIDE World Champion Veselin Topalov, reunified the titles and made Kramnik the undisputed World Chess Champion.^[137] In September 2007, he lost the title to Viswanathan Anand of India. Anand defended his title in the revenge match of 2008,^[138] 2010 and 2012. Magnus Carlsen defeated Anand in 2013, defending his title in 2014, 2016, 2018, and 2021, whereafter he announced that he would not defend his title a fifth time. The 2023 championship was played between the winner and runner-up of the Candidates Tournament 2022: Ian Nepomniachtchi of Russia and Ding Liren of China. Ding beat Nepomniachtchi, making him the world champion.^[45] In 2024, Indian Gukesh Dommaraju beat Ding.

In the Middle Ages and during the Renaissance, chess was a part of noble culture; it was used to teach war strategy and was dubbed the “King’s Game“.^[139] Gentlemen are “to be meanly seene in the play at Chestes”, says the overview at the beginning of Baldassare Castiglione‘s The Book of the Courtier (1528, English 1561 by Sir Thomas Hoby), but chess should not be a gentleman’s main passion. Castiglione explains it further:

And what say you to the game at chestes? It is truely an honest kynde of enterteynmente and wittie, quoth Syr Friderick. But me think it hath a fault, whiche is, that a man may be to couning at it, for who ever will be excellent in the playe of chestes, I beleave he must beestowe much tyme about it, and applie it with so much study, that a man may assoone learne some noble scyence, or compase any other matter of importaunce, and yet in the ende in beestowing all that laboure, he knoweth no more but a game. Therfore in this I beleave there happeneth a very rare thing, namely, that the meane is more commendable, then the excellency.^[140]

Some of the elaborate chess sets used by the aristocracy at least partially survive, such as the Lewis chessmen.

Chess was often used as a basis of sermons on morality. An example is Liber de moribus hominum et officiis nobilium sive super ludo scacchorum (‘Book of the customs of men and the duties of nobles or the Book of Chess’), written by an Italian Dominican friar Jacobus de Cessolis c. 1300. This book was one of the most popular of the Middle Ages.^[141] The work was translated into many other languages (the first printed edition was published at Utrecht in 1473) and was the basis for William Caxton‘s The Game and Playe of the Chesse (1474), one of the first books printed in English.^[142] Different chess pieces were used as metaphors for different classes of people, and human duties were derived from the rules of the game or from visual properties of the chess pieces:^[143]

The knyght ought to be made alle armed upon an hors in suche wyse that he haue an helme on his heed and a spere in his ryght hande/ and coueryd wyth his sheld/ a swerde and a mace on his lyft syde/ Cladd wyth an hawberk and plates to fore his breste/ legge harnoys on his legges/ Spores on his heelis on his handes his gauntelettes/ his hors well broken and taught and apte to bataylle and couerid with his armes/ whan the knyghtes ben maad they ben bayned or bathed/ that is the signe that they shold lede a newe lyf and newe maners/ also they wake alle the nyght in prayers and orysons vnto god that he wylle gyue hem grace that they may gete that thynge that they may not gete by nature/ The kynge or prynce gyrdeth a boute them a swerde in signe/ that they shold abyde and kepe hym of whom they take theyr dispenses and dignyte.^[144]

Known in the circles of clerics, students, and merchants, chess entered into the popular culture of the Middle Ages. An example is the 209th song of Carmina Burana from the 13th century, which starts with the names of chess pieces, Roch, pedites, regina…^[145] The game of chess, at times, has been discouraged by various religious authorities in Middle Ages: Jewish,^[146] Catholic and Orthodox.^[147] Some Muslim authorities prohibited it even recently, for example Ruhollah Khomeini in 1979 and Abdul-Aziz ash-Sheikh even later.^[148]

During the Age of Enlightenment, chess was viewed as a means of self-improvement. Benjamin Franklin, in his article “The Morals of Chess” (1786), wrote:

The Game of Chess is not merely an idle amusement; several very valuable qualities of the mind, useful in the course of human life, are to be acquired and strengthened by it, so as to become habits ready on all occasions; for life is a kind of Chess, in which we have often points to gain, and competitors or adversaries to contend with, and in which there is a vast variety of good and ill events, that are, in some degree, the effect of prudence, or the want of it. By playing at Chess then, we may learn:

I. Foresight, which looks a little into futurity, and considers the consequences that may attend an action …

II. Circumspection, which surveys the whole Chess-board, or scene of action: – the relation of the several Pieces, and their situations …

III. Caution, not to make our moves too hastily …^[149]

Chess was occasionally criticized in the 19th century as a waste of time.^[150][151]

Chess is taught to children in schools around the world today. Many schools host chess clubs, and there are many scholastic tournaments specifically for children. Tournaments are held regularly in many countries, hosted by organizations such as the United States Chess Federation and the National Scholastic Chess Foundation.^[152]

Chess is many times depicted in the arts; significant works where chess plays a key role range from Thomas Middleton’s A Game at Chess to Through the Looking-Glass by Lewis Carroll, to Vladimir Nabokov’s The Defense, to The Royal Game by Stefan Zweig. Chess has also featured in film classics such as Ingmar Bergman‘s The Seventh Seal, Satyajit Ray‘s The Chess Players, and Powell and Pressburger‘s A Matter of Life and Death.

Chess is also present in contemporary popular culture. For example, the characters in Star Trek play a futuristic version of the game called “Federation Tri-Dimensional Chess“,^[153] and “Wizard’s Chess” is played in J.K. Rowling’s Harry Potter.^[154]

The game structure and nature of chess are related to several branches of mathematics. Many combinatorical and topological problems connected to chess, such as the knight’s tour and the eight queens puzzle, have been known for hundreds of years.

The number of legal positions in chess is estimated to be (4.59 ± 0.38) × 10⁴⁴ with a 95% confidence level,^[155] with a game-tree complexity of approximately 10¹²³. The game-tree complexity of chess was first calculated by Claude Shannon as 10¹²⁰, a number known as the Shannon number.^[156] An average position typically has thirty to forty possible moves, but there may be as few as zero (in the case of checkmate or stalemate) or (in a constructed position) as many as 218.^[157]

In 1913, Ernst Zermelo used chess as a basis for his theory of game strategies, which is considered one of the predecessors of game theory.^[158] Zermelo’s theorem states that it is possible to solve chess, i.e. to determine with certainty the outcome of a perfectly played game (either White can force a win, or Black can force a win, or both sides can force at least a draw).^[159] With 10⁴³ legal positions in chess, however, it will take an impossibly long time to compute a perfect strategy with any feasible technology.^[160]

A novel methodology in steganography explores the use of chess-based covers (such as puzzles, chess problems, game reports, training documents, news articles, etc.) for concealing data within a selection of moves, each hiding some bits.^[161][162] Several proof-of-concept projects have been developed that convert text or files into binary code, which is then converted into a series of legal chess moves, that can then be decrypted and downloaded.^[163]

Correspondence chess has been historically suspected of being a potential steganographic medium. Melville Davisson Post documented a chess problem that was used to create a pictorial cipher during World War I.^[164][165] During World War II, extensive postal censorship was imposed on military personnel from the United States and Canada that made playing correspondence chess impossible, arising from suspicion that chess could be used to send secret messages to the enemies.^[166]

There is an extensive scientific literature on chess psychology.^{[note 6][168][169][170][171]} Alfred Binet and others showed that knowledge and verbal, rather than visuospatial, ability lies at the core of expertise.^[172][173] In his doctoral thesis, Adriaan de Groot showed that chess masters can rapidly perceive the key features of a position.^[174] According to de Groot, this perception, made possible by years of practice and study, is more important than the sheer ability to anticipate moves. De Groot showed that chess masters can memorize positions shown for a few seconds almost perfectly. The ability to memorize does not alone account for chess-playing skill, since masters and novices, when faced with random arrangements of chess pieces, had equivalent recall (about six positions in each case). Rather, it is the ability to recognize patterns, which are then memorized, which distinguished the skilled players from the novices. When the positions of the pieces were taken from an actual game, the masters had almost total positional recall.^[175]

More recent research has focused on chess as mental training; the respective roles of knowledge and look-ahead search; brain imaging studies of chess masters and novices; blindfold chess; the role of personality and intelligence in chess skill; gender differences; and computational models of chess expertise. The role of practice and talent in the development of chess and other domains of expertise has led to much empirical investigation. Ericsson and colleagues have argued that deliberate practice is sufficient for reaching high levels of expertise in chess.^[176] Recent research, however, fails to replicate their results and indicates that factors other than practice are also important.^[177][178] For example, Fernand Gobet and colleagues have shown that stronger players started playing chess at a young age and that experts born in the Northern Hemisphere are more likely to have been born in late winter and early spring. Compared to the general population, chess players are more likely to be non-right-handed, though they found no correlation between handedness and skill.^[178]

A relationship between chess skill and intelligence has long been discussed in scientific literature as well as in popular culture. Academic studies that investigate the relationship date back at least to 1927.^[179] Although one meta-analysis and most children studies find a positive correlation between general cognitive ability and chess skill, adult studies show mixed results.^[180][181]

Online chess is chess played over the internet. This is done through the use of internet chess platforms, which use Elo ratings or similar systems to pair up individual players. Online chess saw a spike in growth during the quarantines of the COVID-19 pandemic.^[182][183] This can be attributed to both isolation and the popularity of Netflix miniseries The Queen’s Gambit, which was released in October 2020.^[182][183] Chess app downloads on the App Store and Google Play Store rose by 63% after the show debuted.^[184] Chess.com saw more than twice as many account registrations in November as it had in previous months, and the number of games played monthly on Lichess doubled as well. There was also a demographic shift in players, with female registration on Chess.com shifting from 22% to 27% of new players.^[185] GM Maurice Ashley said “A boom is taking place in chess like we have never seen maybe since the Bobby Fischer days”, attributing the growth to an increased desire to do something constructive during the pandemic.^[186] USCF Women’s Program Director Jennifer Shahade stated that chess works well on the internet, since pieces do not need to be reset and matchmaking is virtually instant.^[187]

The idea of creating a chess-playing machine dates to the 18th century; around 1769, the chess-playing automaton called The Turk became famous before being exposed as a hoax.^[188] Serious trials based on automata, such as El Ajedrecista, were too complex and limited to be useful. Since the advent of the digital computer in the 1950s, chess enthusiasts, computer engineers, and computer scientists have built, with increasing degrees of seriousness and success, chess-playing machines and computer programs.^[189] The groundbreaking paper on computer chess, “Programming a Computer for Playing Chess”, was published in 1950 by Claude Shannon.^{[note 7]} He wrote:

The chess machine is an ideal one to start with, since: (1) the problem is sharply defined both in allowed operations (the moves) and in the ultimate goal (checkmate); (2) it is neither so simple as to be trivial nor too difficult for satisfactory solution; (3) chess is generally considered to require “thinking” for skillful play; a solution of this problem will force us either to admit the possibility of a mechanized thinking or to further restrict our concept of “thinking”; (4) the discrete structure of chess fits well into the digital nature of modern computers.^[191]

The Association for Computing Machinery (ACM) held the first major chess tournament for computers, the North American Computer Chess Championship, in September 1970. CHESS 3.0, a chess program from Northwestern University, won the championship. The first World Computer Chess Championship, held in 1974, was won by the Soviet program Kaissa. At first considered only a curiosity, the best chess playing programs have become extremely strong. In 1997, a computer won a chess match using classical time controls against a reigning World Champion for the first time: IBM’s Deep Blue beat Garry Kasparov 3½–2½ (it scored two wins, one loss, and three draws).^[192][193] There was some controversy over the match,^[194] and human–computer matches were relatively close over the next few years, until convincing computer victories in 2005 and in 2006.

In 2009, a mobile phone won a category 6 tournament with a performance rating of 2898: chess engine Hiarcs 13 running on the mobile phone HTC Touch HD won the Copa Mercosur tournament with nine wins and one draw.^[195] The best chess programs are now able to consistently beat the strongest human players, to the extent that human–computer matches no longer attract interest from chess players or the media.^[196] While the World Computer Chess Championship still exists, the Top Chess Engine Championship (TCEC) is widely regarded as the unofficial world championship for chess engines.^{[197][198][199]} The current champion is Stockfish.

With huge databases of past games and high analytical ability, computers can help players to learn chess and prepare for matches. Internet Chess Servers allow people to find and play opponents worldwide. The presence of computers and modern communication tools have raised concerns regarding cheating during games.^[200]

Related games include:

• direct predecessors of chess, such as chaturanga and shatranj;
• traditional national or regional games that share common ancestors with Western chess such as xiangqi (Chinese chess), shogi (Japanese chess), janggi (Korean chess), ouk chatrang (Cambodian chess), makruk (Thai chess), sittuyin (Burmese chess), and shatar (Mongolian chess);

In the comparison of chess with games often referred to as national forms of chess, chess may be referred to as Western chess or international chess.^[201][202]

There are more than two thousand published chess variants, games with similar but different rules.^[203] Most of them are of relatively recent origin.^[204] They include modern variations employing different rules (e.g. losing chess and Chess960^{[note 8]}), different forces (e.g. Dunsany’s chess), non-standard pieces (e.g. Grand Chess), and different board geometries (e.g. hexagonal chess and infinite chess); In the context of chess variants, chess is commonly referred to as orthodox chess, orthochess, and classic chess.^[201][202]

• Glossary of chess
• Glossary of chess problems
• List of abstract strategy games
• List of chess players
• List of World Chess Championships
• Outline of chess
• Women in chess

• Bird, Henry (2008) [First published 1893]. Chess History and Reminiscences. Forgotten Books. ISBN 978-1-60620-897-7.
• Dunnington, Angus (2003). Chess Psychology: Approaching the Psychological Battle Both on and Off the Board. Everyman Chess. ISBN 978-1-85744-326-4.
• Fine, Reuben (1983). The World’s Great Chess Games. Courier Dover Publications. ISBN 978-0-486-24512-6. OCLC 9394460.
• Hale, Benjamin (2008). Philosophy Looks at Chess. Open Court Publishing Company. ISBN 978-0-8126-9633-2.
• Ilko, Krisztina (2024). “Chess and Race in the Global Middle Ages”. Speculum. 99 (2): 480–540. doi:10.1086/729294.
• Kotov, Alexander (1971). Think Like a Grandmaster. B.T. Batsford Ltd. ISBN 978-0-7134-3160-5.
• Lasker, Emanuel (1960). Lasker’s Manual of Chess. Dover. ISBN 978-0-486-20640-0.
• Leibs, Andrew (2004). Sports and Games of the Renaissance. Westport, CT: Greenwood Publishing Group. ISBN 978-0-313-32772-8.
• Mason, James (1947). The Art of Chess. Dover Publications. ISBN 978-0-486-20463-5. OCLC 45271009. (see the included supplement, “How Do You Play Chess”)
• Pachman, Ludek (1971). Modern Chess Strategy. Dover. ISBN 978-0-486-20290-7.
• Réti, Richard (1960). Modern Ideas in Chess. Dover. ISBN 978-0-486-20638-7.
• Rizzitano, James (2004). Understanding Your Chess. Gambit Publications. ISBN 978-1-904600-07-7. OCLC 55205602.
• Steinitz, William; Landsberger, Kurt (2002). The Steinitz Papers: Letters and Documents of the First World Chess Champion. Jefferson, NC: McFarland & Company. ISBN 978-0-7864-1193-1. OCLC 48550929.
• Verwer, Renzo (2010). Bobby Fischer for Beginners. Alkmaar: New in Chess. ISBN 978-90-5691-315-1.
• Wilkinson, Charles K. (May 1943). “Chessmen and Chess”. The Metropolitan Museum of Art Bulletin. New Series 1 (9): 271–79. doi:10.2307/3257111. JSTOR 3257111.

International organizations

• FIDE – World Chess Federation
• ICCF – International Correspondence Chess Federation

News

• Chessbase news
• The Week in Chess

History

• Chesshistory.com

Xiangqi (/ˈʃɑːŋtʃi/; Chinese: 象棋; pinyin: xiàngqí), commonly known as Chinese chess or elephant chess, is a strategy board game for two players. It is the most popular board game in China. Xiangqi is in the same family of games as shogi, janggi, Western chess, chaturanga, and Indian chess. Besides China and areas with significant ethnic Chinese communities, this game is also a popular pastime in Vietnam, where it is known as cờ tướng, literally ‘General’s chess’, in contrast with Western chess or cờ vua, literally ‘King’s chess’.

The game represents a battle between two armies, with the primary object being to checkmate the enemy’s general (king). Distinctive features of xiangqi include the cannon (pao), which must jump to capture; a rule prohibiting the generals from facing each other directly; areas on the board called the river and palace, which restrict the movement of some pieces but enhance that of others; and the placement of the pieces on the intersections of the board lines, rather than within the squares.

Xiangqi is played on a board nine lines wide and ten lines long. As in the game Go (围棋; or Wéi qí 圍棋), the pieces are placed on the intersections, which are known as points. The vertical lines are known as files (Chinese: 路; pinyin: lù; lit. ‘road’), and the horizontal lines are known as ranks (Chinese: 線/綫; pinyin: xiàn; lit. ‘line’).

Centred at the first to third and eighth to tenth ranks of the board are two zones, each three points by three points, demarcated by two diagonal lines connecting opposite corners and intersecting at the centre point. Each of these areas is known as 宮 gōng^ⓘ, a palace.

Dividing the two opposing sides, between the fifth and sixth ranks, is 河 hé, the “river”. The river is usually marked with the phrases 楚河 chǔ hé^ⓘ, meaning “River of the Chu“, and 漢界 hàn jiè^ⓘ, meaning “Border of the Han“, a reference to the Chu–Han War. Although the river (or Hanchu boundary) provides a visual division between the two sides, only two pieces are affected by its presence: soldiers have an enhanced move after crossing the river, and elephants cannot cross it. The starting points of the soldiers and cannons are usually, but not always, marked with small crosses.

The pieces start in the position shown in the diagram above. Which player moves first has varied throughout history and from one part of China to another. Different xiangqi books advise either that the black or red side moves first.^{[citation needed]} Some books refer to the two sides as north and south; which direction corresponds to which colour also varies from source to source. Generally, Red moves first in most modern tournaments.^[1]

Each player in turn moves one piece from the point it occupies, to another point. Pieces are generally not permitted to move through points occupied by other pieces, the exception being the cannon’s capturing move. A piece can be moved onto a point occupied by an enemy piece, in which case the enemy piece is captured and removed from the board. A player cannot capture one of their own pieces. Pieces are never promoted (converted into other pieces), although the soldier gains the ability to move sideways after it crosses the river. Almost all pieces capture using their normal moves, while the cannon has a special capture move described below.

The game ends when one player checkmates the other’s general. When the general is in danger of being captured by the enemy player on their next move, the enemy player has “delivered a check” (照將/將軍, abbreviated as 將 jiāng^ⓘ), and the general is “in check”. A check should be announced. If the general’s player can make no move to prevent the general’s capture, the situation is called “checkmate” (將死). Unlike in chess, in which stalemate is a draw, in xiangqi, it is a loss for the stalemated player.

In xiangqi, a player—often with a material or positional disadvantage—may attempt to check or chase pieces in a way such that the moves fall in a cycle, preventing the opponent from winning. While this is accepted in Western chess, in xiangqi, the following special rules are used to make it harder to draw the game by endless checking or chasing, regardless of whether the positions of the pieces are repeated or not:

• Perpetual checks with one piece or several pieces are not allowed; doing so results in a loss.
• Perpetual chases of any one unprotected piece with one or more pieces, excluding generals and soldiers, are similarly prohibited.^[2]
• If one side perpetually checks and the other side perpetually chases, the checking side has to stop or be ruled to have lost.
• When neither side violates the rules and both persist in not making an alternate move, the game can be ruled as a draw.
• When both sides violate the same rule at the same time (note that unlike in western chess, a mutual perpetual check is possible) and both persist in not making an alternate move, the game can be ruled as a draw.

Mutual perpetual check A mutual perpetual check occurs with (see #Notation below for notation) 1.Hd7+ Hf4+ 2.Ae2+ Hh3+ 3.Af3+ Hf4+ 4.Ae2+ etc.

Different sets of rules set different limits on what is considered perpetual. For example, club xiangqi rules allow a player to check or chase six consecutive times using one piece, twelve times using two pieces, and eighteen times using three pieces before considering the action perpetual.^[2]

The above rules to prevent perpetual checking and chasing, while popular, are not the only ones; there are numerous end game situations.^[3]

Each player controls an army of 16 pieces; the armies are usually coloured red and black.^[4] Pieces are flat circular disks labelled or engraved with a Chinese character identifying the piece type, and in a colour indicating which player has ownership. The black pieces are marked with somewhat different characters from the corresponding red pieces.

Even in mainland China, most sets still use traditional Chinese characters (as opposed to simplified Chinese characters). Modern pieces are usually plastic, though some sets are wooden, and more expensive sets may use jade. In more ancient times, many sets were simple unpainted woodcarvings; thus, to distinguish between pieces of the two sides, most corresponding pieces used characters that were similar but varied slightly.^[4] This practice may have originated in situations where there was only one material available to make the pieces from and no colouring material available to distinguish the opposing armies. The oldest xiangqi piece found to date is a 俥 (chariot) piece. It is kept in the Three Gorges Museum.^[5][6]

Generals (or kings) are labelled 將 (trad.) / 将 (simp.) jiàng^ⓘ (“general”) on the black side and 帥 (trad.) / 帅 (simp.) shuài^ⓘ (“marshal”) on the red side.

The general starts the game at the midpoint of the back edge, within the palace. The general may move and capture one point orthogonally and may not leave the palace, with the following exception.

If the two generals face each other along the same file with no intervening pieces, the 飛將 (“flying general”) move may be executed, in which the general to move crosses the board to capture the enemy general. In practice, this rule means that creating this situation in the first place means moving into check, and is therefore not allowed.^[7][8]

The Indian name king for this piece was changed to general because of Chinese naming taboos; China’s rulers objected to their royal titles being given to game pieces.^{[9][dubious – discuss]} Despite this, the general is sometimes called the “king” by English-speaking players, due to their similar functions as royal pieces.^[10]

Advisors (also known as guards and less commonly as assistants, mandarins, ministers or warriors) are labelled 士 shì^ⓘ (“scholar”, “gentleman”, “officer“, “guardian”) for Black and 仕 shì^ⓘ (“scholar”, “official”, “guardian”) for Red. Rarely, sets use the character 士 for both colours.

The advisors start on either side of the general. They move and capture one point diagonally and may not leave the palace, which confines them to five points on the board.^[8] The advisor is probably derived from the mantri in chaturanga, like the queen in Western chess.

There is some controversy about whether “士” really is intended to mean “scholar”, “gentleman” which would be “士人”, or “guard”, “guardian” which would be “衛士” (simplified Chinese: 卫士). One argument for the latter is that their functionality seems to be to guard/protect the general. The common Western translation “advisor” does not reflect this layer of meaning.

Elephants (or bishops) are labeled 象 xiàng (“elephant”) for Black and 相 xiàng (“minister”) for Red. They are located next to the advisors. These pieces move and capture exactly two points diagonally and may not jump over intervening pieces; the move is described as being like the character 田 Tián (“field”), in reference to the board’s squares.^[8] Blocking an elephant with a diagonally adjacent piece is known as “blocking the elephant’s eye” (塞象眼).^[11]

Elephants may not cross the river to attack the enemy general, and serve as defensive pieces. Because an elephant’s movement is restricted to just seven board positions, it can be easily trapped or threatened. The two elephants are often used to defend each other.

The Chinese characters for “minister” and “elephant” are homophones in Mandarin (Listen^ⓘ) and both have alternative meanings as “appearance” or “image”. However, in English, both are referred to as elephants, and less commonly as “bishops”, due to their similar movements.^[10]

Horses (or knights) are labelled 馬 mǎ^ⓘ for Black and 傌 mǎ for Red in sets marked with Traditional Chinese characters and 马 mǎ for both Black and Red in sets marked with Simplified Chinese characters. Some sets use 馬 for both colours. Horses begin the game next to the elephants, on their outside flanks. A horse moves and captures one point orthogonally and then one point diagonally away from its former position, a move which is traditionally described as being like the character 日 Rì.^[8] The horse does not jump as the knight does in Western chess, and can be blocked by a piece of either colour located one point horizontally or vertically adjacent to it. Blocking a horse is called “hobbling the horse’s leg” (蹩馬腿). The diagram on the right illustrates the horse’s movement.

Since horses can be blocked, and the path of a horse from one point to another is not the same as the reverse move, it is possible for one player’s horse to have an asymmetric attack advantage if an opponent’s horse is blocked, as seen in the diagram on the right.

The horse is sometimes called the “knight” by English-speaking players, due to their similar movements.^[10]

Chariots (or rooks or cars) are labelled 車 jū^ⓘ for Black and 俥 jū^ⓘ for Red in sets marked with Traditional Chinese characters and 车 for both Black and Red in sets marked with Simplified Chinese characters. Some traditional sets use 車 for both colours. In the context of xiangqi, all of these characters are pronounced as jū^ⓘ (instead of the common pronunciation chē). The chariot moves and captures any distance orthogonally, but may not jump over intervening pieces. The chariots begin the game on the points at the corners of the board. The chariot is often considered to be the strongest piece in the game due to its freedom of movement and lack of restrictions.^[8]

The chariot is sometimes called the “rook” by English-speaking players, since it moves identically to the rook in Western chess.^[10] Chinese players (and others) often call this piece a car, since that is one modern meaning of the character 車.

Cannons are labelled 砲 pào^ⓘ (“catapult“) for Black and 炮 pào (“cannon”) for Red. The names are homophones, though sometimes 炮 is used for both Red and Black. The 石 shí radical of 砲 means “stone”, and the 火 huǒ radical of 炮 means “fire”. Both colours’ pieces are normally referred to as cannons in English. The black piece is sometimes labelled 包 bāo.

Each player has two cannons, which start on the row behind the soldiers, two points in front of the horses. Cannons move like chariots, any distance orthogonally without jumping, but can only capture by jumping a single piece of either colour along the path of attack. The piece over which the cannon jumps is called the 炮臺 (trad.) / 炮台 (simp.) pào tái (“cannon platform” or “screen”). Any number of unoccupied spaces, including none, may exist between the cannon, screen, and the piece to be captured. Cannons can be exchanged for horses immediately from their starting positions.^[8]

Soldiers (or pawns) are labelled 卒 zú^ⓘ (“pawn” or “private”) for Black and 兵 bīng^ⓘ (“soldier”) for Red. Each side starts with five soldiers. Soldiers begin the game located on every other point one row back from the edge of the river. They move and capture by advancing one point. Once they have crossed the river, they may also move and capture one point horizontally. Soldiers cannot move backward, and therefore cannot retreat; after advancing to the last rank of the board, however, a soldier may still move sideways at the enemy’s edge.^[8] The soldier is sometimes called the “pawn” by English-speaking players, due to the pieces’ similar movements.^[10]

Piece	Points
Soldier before crossing the river	1
Soldier after crossing the river	2
Advisor	2
Elephant	2
Horse	4
Cannon	4½
Chariot	9

These approximate values^[12] do not take into account the position of the piece in question (except the soldier in a general sense), the positions of other pieces on the board, or the number of pieces remaining. In what follows, “minor piece” will refer to horses and cannons, and “defensive piece”, unless otherwise specified, will refer to the non-royal pieces that cannot cross the river, namely advisors and elephants.

Other common rules of assessment:

• A horse plus a cannon is generally better than two horses or two cannons.
• The chariot is not only the strongest piece, but it is also generally stronger than any combination of two minor pieces. When the relative values of both sides’ pieces are approximately even, the side with more chariots generally has the advantage, especially when one side has a chariot and one side does not (Chinese: 有車壓無車). However, the chariot is not particularly strong in basic endgames: for example, chariot versus four defensive pieces is generally a draw, while if the offensive side instead has two horses or even three unadvanced soldiers it is a win.
• In the earlier stages, the cannon is stronger than the horse, because platforms are plentiful and the horse is often blocked by the multitude of pieces on the board. In the endgame, the horse is stronger as an attacking piece, not needing any platforms, but the cannon generally has better defensive abilities.
• The values of soldiers vary in different stages of the game. In the opening and the middlegame, the initiative and mobility of pieces often require sacrificing soldiers. In these stages, soldiers closer to the middle file are generally more valuable, since they can effectively join the offence. With few attacking pieces on the board, soldiers have more power and can cross the river more easily. In this stage, advanced soldiers are generally less powerful, since soldiers cannot move backward. In basic endgames, three soldiers starting on the seventh rank are approximately equal to a chariot: they can force a win against four defensive pieces or a horse/cannon plus two elephants, while instead a chariot cannot, and a chariot cannot force a win against three soldiers on the seventh rank when well-defended. At the other extreme, even five soldiers all on the tenth rank cannot on their own force checkmate or stalemate on a bare general due to insufficient material.

There are several types of notation used to record xiangqi games. In each case the moves are numbered and written with the same general pattern.

1. (first move) (first response)
2. (second move) (second response)
3. …

It is clearer but not required to write each move pair on a separate line.

The book The Chess of China describes a move notation method in which the ranks of the board are numbered 1 to 10 from closest to farthest away, followed by a digit 1 to 9 for files from right to left.^[13] Both values are relative to the moving player. Moves are then indicated as follows:

[piece name] ([former rank][former file])-[new rank][new file]

Thus, the most common opening in the game would be written as:

1. 炮 (32)–35   馬 (18)–37

Name	Abbr.	Pieces
Advisor	A
Cannon	C
Chariot	R*
Elephant	E
General	G
Horse	H
Soldier	S
* using C would conflict with the letter for Cannon

A notation system partially described in A Manual of Chinese Chess^[14] and used by several computer software implementations describes moves in relative terms as follows:

[single-letter piece abbreviation][former file][operator indicating direction of movement][new file, or in the case of purely vertical movement, number of ranks traversed]

The file numbers are counted from each player’s right to each player’s left.

In case there are two identical pieces in one file, symbols + (front) and – (rear) are used instead of former file number. Direction of movement is indicated via an operator symbol. A plus sign is used to indicate forward movement. A minus sign is used to indicate backward movement. A dot or period or equals sign is used to indicate horizontal or lateral movement. For a piece that moves diagonally (such as the horse or elephant), the plus or minus sign is used rather than the period.

Thus, the most common opening in the game would be written as:

1. C2.5 H8+7

According to World Xiangqi Federation (WXF), in the case of tripled, quadrupled, or quintupled soldiers (pawns), the S for soldier or P for pawn is omitted. Instead, the soldiers are numbered starting from the frontmost soldier, and this number replaces the usual piece abbreviation. The file number is given immediately after as usual.

Thus the notation to move the middle of a set of tripled soldiers on the 5th file to the 4th file would be:

1. 25=4

In older books written in Chinese the system is the same, except that: the names of the pieces are written in Chinese; the name for the cannon on both sides is 炮; the name for the horse on both sides is 馬; forward motion is indicated with 進 (pronounced jìn); backward motion is indicated with 退 (tuì); sideways motion is indicated with 平 (píng); and numbers are written in Chinese either for both players or for just Black.

Thus, the most common opening in the game might be written as:

1. 炮二平五 馬8進7

This system is unofficial and principally used by Western players.^{[citation needed]} It is similar to algebraic notation for Western chess. Letters are used for files and numbers for ranks. File “a” is on Red’s left and rank “1” is nearest to Red. A point’s designation does not depend on which player moves; for both sides “a1” is the lowest left point from Red’s side.

[single-letter piece abbreviation][former position][capture indication][new position][check indication][analysis]

Pieces are abbreviated as in notation system 2, except that no letter is used for the soldier.

Former position is only indicated if necessary to distinguish between two identical pieces that could have made the move. If they share the same file, indicate which rank moves; if they share the same rank, indicate which file moves. If they share neither rank nor file, then the file is indicated.

Capture is indicated by “x”. No symbol is used to indicate a non-capturing move.

Check is indicated by “+”, double check by “++”, triple check by “+++”, and quadruple check by “++++”. Checkmate is indicated by “#”.

For analysis purposes, bad moves are indicated by “?” and good moves by “!”. These can be combined if the analysis is uncertain (“!?” might be either but is probably good; “?!” is probably bad) or repeated for emphasis (“??” is a disaster).

Thus, the most common opening in the game would be written as:

1. Che3 Hg8

For example, the following game is tied with several others as the shortest possible xiangqi game (analogously to Fool’s Mate), written in each of the three notations:

Black is mated and therefore loses. Red’s cannon cannot be captured, and the advisor on e9 blocks both Black’s general from escaping there and the advisor on d10 from unbecoming a platform for the cannon.

Because of the size of the board and the low number of long-range pieces, there is a tendency for the battle to focus on a particular area of the board.^{[clarification needed]}

Xiangqi involves several tactics common to games in the chess family. Some common ones are briefly discussed here.

• In a fork, one piece attacks two or more enemy pieces at once.
• A piece is pinned when it cannot move without exposing a more important piece to capture. Every piece except soldiers and advisors can pin, but only chariot pins exactly resemble pins in western chess; pins by other pieces in xiangqi take on many unique forms: Cannons can pin two pieces at once on one file or rank, horses can pin because they can be blocked, and generals can pin because of the “flying general” move rule. In pins by horses and elephants, the pinning piece never attacks the pinned piece, while in a pin by a cannon, only one of the pieces is directly attacked by the cannon. A general can only pin pieces to the enemy general, and the pinning general can never capture the pinned piece, since that would place it in check from the enemy general.
• A piece is skewered when it is attacked and, by moving, exposes a less important piece to be captured. In contrast to pins, only cannons and chariots can skewer.

Fork	Pin	Skewer
Red’s horse (傌) at d5 forks black’s soldier (卒) at c7 and cannon (砲) at e7.	Red’s advisor (仕 ) and horse (傌) on d2 are both relatively pinned by Black’s cannon (砲) at d4 to Red’s chariot (俥) on d1. Black’s horse (馬) on e9 is absolutely pinned by Red’s horse at e8. It is illegal for the black horse to move, but it is safe from attack, while Black’s cannon actively attacks the red horse and forces the red chariot to defend it.	Red’s cannon (炮) at e1 is skewering black’s general (將) at e8 and chariot (車) at e10. When the black soldier (卒) or general moves laterally to remove the check, the other will act as a platform for the red cannon to capture the black chariot.

• A discovered check occurs when an attacking piece moves so that it unblocks a line for a chariot, cannon, and/or horse to check the enemy general.
• A double check occurs when two pieces simultaneously threaten the enemy general. Unlike a Western chess double check, a double check in xiangqi may be blockable or, in one case, possibly met with a capture by a piece other than the general. The only blockable cases are either a chariot and cannon on the same file as the general, with the chariot acting as a screen for the cannon, two horses giving discovered check after another piece unblocks the attack from both, or a cannon using an enemy piece as a platform uncovered by a horse (see below). Double checks delivered by other means are not blockable. In one exceptional case, if a horse moves to give a double check by uncovering a cannon, and the cannon’s platform is an enemy chariot or defensive piece (advisor or elephant), the enemy chariot or defensive piece might be able to capture the horse, which removes the cannon’s platform at the same time. Otherwise, capturing either checking piece is insufficient to remove the threat, unless the general makes the capture.

Blockable double check	Double check met by capture	Double check that compels the general to move
Red’s horse has moved to d7, creating a double check against Black’s general, but Black can answer by moving the cannon to d8, blocking the horse and removing the cannon’s platform	Red’s horse has moved from e4 to d6, unveiling a double check with the cannon, but Black can reply by simply capturing the horse with the chariot.	Red’s horse has moved from e6 to f8 and placed the black general in double check. Since, as in western chess, there are two independent lines of attack, Black cannot respond with any move other than moving the general to f10, whereupon Red wins Black’s horse for free.

• Unique to xiangqi is a triple check, which arises in four combinations. In the first case of a cannon, a chariot or soldier, and a horse, the horse moves to give check, uncovering a double check from the chariot and the cannon. In the second, rarer case of a chariot or soldier and two horses, the chariot moves to give check, uncovering a double check from the two horses. In the third case of two cannons and two horses, one cannon may uncover a double check from the horses and act as a screen for the other cannon. Finally, a chariot or soldier can move to give check, uncovering a check from a horse while acting as a platform for a cannon to give another check. Quadruple check is also possible, arising with two horses, a chariot, and a cannon. Triple and quadruple check cannot be blocked or met by captures (again, unless the general makes the capture).

Triple check	Quadruple check	Triple check, alternate position
Red’s horse (傌) has moved from e5 to d7, giving check and exposing a double check from the chariot (俥) at e3 and the cannon (炮) at e2.	Red’s chariot (俥) has moved from f9 to e9, giving check and exposing a triple check from the cannon (炮) at e7 and both horses (傌) at f8 and g9. Replacing the chariot with a cannon or removing a horse produces a triple check.	Red’s chariot (俥) has moved from f9 to e9, discovering two checks from both horses (傌) at f8 and g9 and gives check itself.

In contrast to the ubiquity of pawn chains in western chess, soldiers typically do not support each other until the endgame, because from the initial position it takes a minimum of five moves of a soldier to allow mutual protection between two of them, and they are often prone to capture by other pieces.

Soldiers, horses, cannons and chariots can form up formations that protect each other. However, lining up chariots must be done with caution, as this risks losing one chariot to an enemy’s inferior piece. Horses that support each other are called Linked Horses (Chinese: 連環馬), which is a relatively safe formation of the horses, though it can still be threatened with a soldier, a chariot plus another minor piece, or a piece blocking one of the horses thus making the protection one-sided.

It is common to use cannons independently to control particular ranks and files. Using a cannon to control the middle file is often considered vital strategy, because it pins pieces such as the advisors and elephants to the general, which in turn restrict their general’s movement. The two files adjacent to the middle file are also considered important and horses and chariots can be used to push for checkmate there.

Since the general is usually safest in its original position before the endgame phase, attacking the general commonly involves forcing the general out of its original position with check or with threats. Thus, specific points and formations are very important in xiangqi.

For an attacking (Red) horse, the most fatal points are c9 and g9 (Chinese: 臥槽馬), especially since without proper defence a quick mate can follow with an extra chariot or cannon.

Liang vs. Zhao, 1982 Due to the pin of two pieces by the red cannon, Black’s centroid horse has become a liability rather than an asset.

For a cannon, one of the most fatal formations is the exposed cannon (Chinese: 空心炮), where the cannon directly controls the middle file with no other pieces between the cannon and the general. This formation is particularly dangerous since the defensive side cannot move any piece in front of the cannon; while with an extra cannon joining the attack, mate can follow on the spot, and with an extra rook, the offensive side can mount a double check (with the rook in front of the cannon) followed by a windmill (as when the check is blocked, the rook moving laterally discovers a new check from the cannon), often winning at least a piece afterwards. If the defensive side cannot chase the cannon away or capture it, it must move the general forward to avoid these threats, leaving the general vulnerable to attacks.

Another fatal formation, called the “cannon-controlled centroid horse” (Chinese: 炮鎮窩心馬, diagram at right), also requires particularly bad coordination of the enemy pieces. In the diagram, Black’s “centroid horse” occupies the centre of the palace, blocking Black’s own general and advisors, and being pinned to the general by the red cannon, cannot move. Black’s cannon at e8 is also pinned to its own general; it too is unable to move and restricts the movement of Black’s two elephants, making them unable to protect each other. Such a formation in the middlegame often produces deadly threats of smothered mates, while in the endgame, as in the diagram, Red’s cannon cannot be chased away, rendering Black’s general, advisors, cannon on e8, and horse all permanently immobilized. Even though Black is up a minor piece, Red has a clear win: The game concluded 41.Hg7 (forking the elephant and pinned cannon and creating a mating threat) Eg10 42.Hh9 Ci9 43.Hf8+ Cf9 (if not for the other black cannon, it is instant mate) 44.Hxg6, and Black resigned: Black’s only active piece (the cannon on f9) is absolutely helpless to stop Red’s horse and soldiers, which will soon invade the palace.

A common defensive configuration is to leave the general at its starting position, deploy one advisor and one elephant on the two points directly in front of the general, and to leave the other advisor and elephant in their starting positions, to the side of the general. In this setup, the advisor and elephant pairs support each other, and the general is immune from attacks by cannons. Losing any defensive pieces makes the general vulnerable to cannon attack, and the setup may need to be abandoned. The defender may move defensive pieces away from the general, or even sacrifice them intentionally, to ward off attack by a cannon.

Red mates in 11 Solution: 1.Rh10++ 1. …Eig10 gets mated faster: 2.Rgxg10+ Af10 (Exg10 3.Rxg10#) 3.Rxf10+ Ge9 4.Rh9#. 1. …Eeg10 gets mated immediately with 2.Rhxg10# 1. …Af10 2.Rh9+ Afe9 3.Rg10++ Af10 4.Rg9+ Afe9 5.Rh10++ Again both 5. …Eig10 and 5. …Eeg10 lead to faster mates. 5. …Af10 6.Re9+!! A brilliant smothered-check inducing move. If 6. …Gxe9 7. Rh9# 6. …Adxe9 7.Rh9+ Eeg10 8.Rxe9+ (the chariot is untouchable with legal moves) Gd10 9.Re10+ Gd9 10.d8+ Gxd8 11.Rd10# Note that if Red plays 4.Re9+? instead, Red cannot force a mate: 4. …Adxe9 5.Rg9+ Eig10 and Red cannot play Rxe9+ on the next move because the chariot is then not supported by the general, and black can simply play Gxe9. The purpose of using the g-chariot to give check is to place the h-chariot on the h9 point, blocking Black’s Eig10.

Long sequences of checks leading to mate or gain of material are common both in chess compositions and in actual play. A skilled xiangqi player would often have to calculate several steps, or even tens of steps ahead for a forced sequence. In the diagram on the right, Black has an immediate mating threat which cannot be parried, forcing Red to check Black on every move. Although it requires 11 moves to mate, its general idea is clear: Induce a smothered check by sacrificing a chariot at the centre of the palace (e9), then force Black to open the centre file, enabling the Red general to assist the attack, and finally mate by facing generals.

The most common opening pair of moves

Since the left and right flanks of the starting setup are symmetrical, it is customary to make the first move on the right flank. Starting on the left flank is considered needlessly confusing.^{[citation needed]}

The most common opening is to move the cannon to the central column, an opening known as 當頭炮 (trad.) / 当头炮 (simp.) dāng tóu pào or “Central Cannon”. The most common reply is to advance the horse on the same flank. Together, this move-and-response is known by the rhyme 當頭炮，馬來跳 (trad.) / 当头炮，马来跳 (simp.) dāng tóu pào, mǎ lái tiào^ⓘ. The notation for this is “1. 炮 (32)–35, 馬 (18)–37”, “1. C2.5 H8+7”, or “1. Che3 Hg8” (diagram at right). After Black’s 1. …H8+7 (Hg8) response, the game can develop into a variety of openings, the most common being the 屏風馬 (trad.) / 屏风马 (simp.) or “Screen Horses (Defence)” in which Black develops the other horse to further protect their middle pawn (…H2+3 or …Hc8) either immediately on their second move, or later when Black transposes the game into this opening.

Alternative common first moves by Black are developing either cannons (1. …C8.5/1. …Che8, or 1. …C2.5/1. …Cbe8); note that after either of these moves, taking the central soldier with the cannon (2. C5+4 or 2. Cxe7+) is a beginner’s trap that impedes development and coordination of Red’s pieces if Black plays correctly (for example, 1. Che3 Che8 2. Cxe7+?? Ade9 3. Hg3 Hg8 4. Ce5 Rh10 when Black develops the rook first, and the loss of Black’s middle pawn actually enabled Black’s horses to occupy the centre on the next moves).

Other common first moves by Red include moving an elephant to the central column (1. Ege3), advancing the soldier on the third or seventh file (1. c5), moving a horse forward (1. Hg3), and moving either cannon to the 4th or 6th (d- or f-) file (1. Chd3 or 1. Chf3). Compared to the Central Cannon openings, these openings are generally less restricted by theory.

General advice for the opening includes rapid development of at least one chariot and putting it on open files and ranks, as it is the most powerful piece with a long attack range. There is a saying that only a poor player does not move a chariot in the first three moves (Chinese: 三步不出車，必定要輸棋); however this is not to be taken literally, and is in fact often violated in modern Xiangqi games.^{[citation needed]} Attacking and defending the centre, especially the central soldiers / central pawns, are common themes in the opening, hence the Central Cannon openings. Usually, at least one horse should be moved to the middle in order to defend the central soldier; however undefended central soldiers can also become “poisoned pawns” in the early moves, especially if the attacking side does not have an immediate follow-up to retain the pressure on the central file.

Xiangqi strategy shares common themes with chess, but has some differences:

• Occupying the centre is relatively less important in xiangqi, but controlling and attacking the middle file is still one of the vital themes. Since the middle file is often well defended, players would then seek to mount an offense on either of the flanks on the enemy side, especially when the defense of one flank is neglected.
• The significance of pawn formation in xiangqi and chess are different. In xiangqi, soldiers (pawns) are often pushed to avoid blocking their own horses, and it is uncommon for them to defend each other (in contrast with a Western chess pawn chain). Successfully getting a soldier to cross a river as an attacking force can often tilt the scales of the middlegame by a large margin.
• In high-level play, the initiative is highly important, and a minor mistake can doom a game.
• Sacrifices are common in xiangqi, however they are more often tactical rather than positional. Usually, at most a minor piece is sacrificed for positional advantages, or a semi-tactical attack.

Like in chess, xiangqi piece values depend highly on the position on the board. The following study from Volume 42 of the Elegant Pastime Manual, dating from the Ming Dynasty, illustrates this dramatically. It is Red to play and win.^[15]

In this position, Red is up two soldiers and a cannon but Black threatens seemingly unstoppable mate with …Rf1#, since 1.Ec5? Ad8! renews the mate threat. Note that the red chariot is nearly useless, having only two legal moves, in stark contrast to the very active black chariot. However, Red averts the checkmate by sacrificing both the cannon and chariot: 1.Ca10+!! Hxa10 2.Ea3! Rxa1 (or the chariot is lost, since the general protects the elephant on e3, which in turn guards g1, and the red horse guards e2) 3.Eec1:

Despite the substantial sacrifice of material by Red, Black’s chariot has now become useless as it is permanently immobilized by the red elephants and horse; the red general prevents the black soldier on g2 from moving laterally to free the chariot (for example 3…g1 4.Gf2). Black’s horse similarly has no safe move due to the red soldier on c8, which also, along with the soldier on f9, prevents the black general from attacking the soldier on f9 from the behind. In addition, the black soldier on g6 is undefended and has no safe move, so Red can win it by pushing the soldier on c4 to c6 and moving it laterally to the g-file, after which the position is effectively an endgame of three soldiers against two advisers, an easy win for Red (see below) despite being down a chariot for three soldiers.

Though xiangqi endgames require remarkable skill to be played well, there are a number of widely known book wins and book draws. Without a counterpart to pawn promotion, xiangqi endgames instead focus more directly on forcing checkmate or stalemate, and in this regard resemble pawnless chess endgames. Since stalemate is a loss for the stalemated player instead of a draw, most book draws in xiangqi are due to fortresses, with a few draws due to insufficient material.

A general rule in xiangqi endgames for the advantageous side is that, when there’s less material on the board, do not trade pieces easily, as with fewer attacking pieces on the board, defending is easier (in contrast to Western chess, where it is almost always advantageous to trade pieces when up on material). Hence, if a certain type of endgame can transpose, by trading pieces, into another type of endgame which is a book win, then this endgame itself is a book win.

Red wins with either side to move.

Inducing zugzwang is a crucial theme in winning simple endgames, and almost exclusively in simple endgames. In the general + soldier vs general endgame shown on the right, Red’s first main goal is to occupy the middle file. Red wins with 1. Gd1, a waiting move, and Black is in zugzwang. Black must proceed with 1. …Ge8, as 1. …Ge10 instantly loses after 2. f9#. After 1. …Ge8 2. f9 Gf8 3. e9 Ge8 4. d9 Gf8 5. Ge1, Red’s general successfully occupies the middle file. The game would conclude with 5. …Gf9 6. e9+, and regardless of Black’s reply, 7. Ge2# (stale)mates Black.

Reciprocal zugzwang: Whoever moves first loses.

Reciprocal zugzwang is possible, but very rare and usually seen in endgame compositions. In this endgame shown on the right, whoever moves loses, since when either of the two generals moves to an open d- or f- file, it threatens unstoppable mate, while the player to move only helps the enemy general occupy one of the files. For instance, Red can only move their two soldiers if he is to move. Moving the f-(or d-)soldier allows the enemy general to occupy the f-file(d-file). Even if 1. fe9+ Gf10 2. d10, when Red threatens mate in 1, Black still mates immediately with either 2. …fe2# or 2. …f1#.

Red to play wins; Black to play draws.

• A soldier, as long as it does not reach the opposite rank, wins against a bare general easily. With any extra defensive piece on the defensive side, it is a draw; however, soldier vs advisor requires skill to play well. Any number of soldiers on the opposing back rank are also a draw due to insufficient material.
• Two unadvanced (i.e., on the 6th or 7th ranks) soldiers win against the following combinations: Two advisors, two elephants, a bare horse/cannon. Generally a draw against one advisor plus one elephant, or a horse/cannon plus a defending piece.
• Three unadvanced soldiers win against the following combinations: All 4 defensive pieces (2 advisors plus 2 elephants, Chinese: 士象全), a horse plus two advisors/two elephants, a cannon plus two elephants.

Red to play wins with 1. Hd7, preventing the elephant from getting to the opposite flank, and thus placing Black in zugzwang (1…Gd9 loses to the fork 2.Hb8+ while after 1…Ea8, 2.Hb8 puts Black in zugzwang again and the elephant is lost on the following move). Black to play draws with 1. …Ee8.

• A bare horse wins against a bare advisor, but not a bare elephant.
• A horse plus an unadvanced soldier wins against both combinations of 3 defensive pieces, or any combination of a minor piece plus a defensive piece except horse + elephant. This combination draws against all 4 defensive pieces.
• A horse plus an advanced soldier(on the 8th or 9th rank) draws against either combination of 3 defensive pieces, but defending requires precise positions.
• A horse plus a soldier on the 10th rank wins against two advisors, or one advisor plus one elephant. This combination draws against 2 elephants.
• A horse plus two soldiers can win against one minor piece + one advisor + two elephants. With an extra advisor on the defensive side, it is a book draw.
• Two horses win against all 4 defensive pieces, or any combination of a minor piece plus 2 defensive pieces except cannon + 2 elephants.

After 1. Ge3, Black must lose an advisor, since 1…Gd9 is met by 2.Ae2#. Note that after 1. …Ge10 2. Cxd8, taking the cannon with Axd8 is illegal as the advisor is pinned to its general (the generals may not face each other on the same file).

• A bare cannon, or a cannon with elephants, cannot win against a bare general due to insufficient material. Cannons without other offensive pieces need defensive pieces to act as platforms, especially the advisor, since the advisor can act as a platform on any of the three central files, while an elephant can only act as a platform on the centermost file.
• A cannon needs only one advisor to win against two advisors, or a single elephant. Meanwhile, even with all 4 defensive pieces, it is a book draw against two elephants, one advisor + one elephant, one soldier + one advisor, or any minor piece.
• A cannon with all 4 defensive pieces needs at least an extra soldier to win against 4 defensive pieces. A bare cannon with a soldier on the 6th rank wins against any combination of 2 defensive pieces.
• A cannon + 4 defensive pieces + 2 unadvanced soldiers generally draw against one minor piece + 4 defensive pieces. But if the defensive side lacks a single piece, it is a book win.

This type of endgame is considered one of the more complex endgames. Commonly known book wins and book draws are:

• Horse + Cannon + 4 defensive pieces vs a minor piece vs 4 defensive pieces: A win if the minor piece is a horse (the attacking side does not need all 4 defensive pieces to win), a draw if it is a cannon.
• With the same combination of the two minor pieces and all 4 defensive pieces on both sides, one needs two extra soldiers for a book win.
• If both sides have 2 minor pieces and 4 defensive pieces, and the advantegeous side only has one extra soldier, then regardless of the combination of the two minor pieces, it is a book draw.

The drawing fortress for chariot vs horse + 2 elephants. Other defensive positions, aside from this position’s symmetrical variation, generally lose: If, instead, the elephant on g6 had been on g10, then Red to play wins, starting with 1. Rb7.

Single chariot endgames:

• A single chariot generally cannot win against 4 defensive pieces, but with 3 or fewer defensive pieces, it is a forced win.
• Chariot vs one minor piece plus 2 defensive pieces: A win if the 2 defensive pieces are not the same, or if the combination is horse + two advisors. If the defensive side has horse + two elephants, a specific fortress is required to draw.
• Chariot vs one minor piece plus 3 defensive pieces: A draw.
• Chariot vs two minor pieces with no defensive pieces: A draw, but requires good defensive positions.

Chariot + soldiers (unadvanced):

• Chariot + soldier, with sufficient defensive pieces on their own side, wins against a chariot plus an advisor, a chariot plus two elephants, or a chariot plus a soldier.
• Chariot + soldier wins against any 2 minor pieces + 2 advisors. This combination also wins against horse + 4 defensive pieces, but not cannon + 4 defensive pieces.
• Chariot + soldier vs 2 unadvanced soldiers + 4 defensive pieces: If the offensive side has no defensive piece, it is a draw since the 2 enemy soldiers can still be a formidable force. If the offensive side has one advisor, it is a win.
• Chariot + 2 soldiers cannot force a win against chariot + 4 defensive pieces. In this endgame, both attacking and defending require great skill.

Chariot + horse:

• A chariot plus a horse needs one advisor on their own side to win against a chariot plus two advisors.
• Chariot + horse vs chariot + two elephants: With enough defensive pieces for the attacking side, it is generally a win if the move limit is not taken into consideration.

Chariot + cannon:

• A chariot plus a cannon cannot win against a bare chariot, as long as the defending chariot occupies the middle file. However, with any extra defensive piece on the attacking side, it is a win.
• Chariot + cannon + 2 advisors would win against chariot + two elephants.
• Chariot + cannon + 4 defensive pieces vs chariot + 4 defensive pieces: Draw.

Two chariots:

• Two chariots vs chariot + 4 defensive pieces: A draw with good defensive positions.
• Two chariots vs chariot + minor piece + 2 defensive pieces: The only drawing combination is chariot + cannon + 2 advisors.
• Two chariots vs 2 minor pieces + 4 defensive pieces: A win if the 2 minor pieces are 2 horses.

A game called xiangqi was mentioned as dating to the Warring States period; according to the first-century-BC text Shuo Yuan (說苑/说苑), it was one of Lord Mengchang of Qi‘s interests. However, the rules of that game are not described, and it was not necessarily related to the present-day game.^[16] Emperor Wu of Northern Zhou wrote a book in AD 569 called Xiang Jing. It described the rules of an astronomically themed game called xiangxi (象戲). Ancient reports state that “the pieces … were called after the sun, the moon, the planets and the star-houses”, in contrast with modern xiangqi. In addition, there are mentions of shuffling of pieces, which does not occur in xiangqi either.^[17] For these reasons, Murray argued that “in China [chess] took over the board and name of a game called 象棋 in the sense of ‘Astronomical Game’, which represented the apparent movements of naked-eye-visible astronomical objects in the night sky, and that the earliest Chinese references to 象棋 meant the Astronomical Game and not Chinese chess”. Previous games called xiàngqí may have been based on the movements of sky objects. In contrast, present-day xiangqi developed from Indian chaturanga according to Murray;^{[18]: 119–120} this is also the predominant opinion among historians.^{[19][20]: 334}

Some Sinologists and Chinese historians prefer an alternative hypothesis, according to which xiangqi originated in China and then spread westwards, giving rise to Indian and Persian chess.^{[19][20]: 351–352} Specifically, it has been claimed that Xiangqi arose during the Warring States period and was patterned after the array of troops at the time. David H. Li, for example, argues that the game was developed by Han Xin in the winter of 204–203 BC to prepare for an upcoming battle.^[21] His theories, however, have been questioned by other chess researchers.^[22] The earliest description of the game’s rules appears in the story “Cén Shùn” (岑順) in the collection Xuanguai lu (玄怪錄), written by Niu Sengru in the middle part of the Tang dynasty.

Xiangqi is the same as it is today from Southern Song dynasty.

Janggi of the Korean Peninsula originated from xiangqi.

With the popularization of xiangqi, many different schools of circles and players came into prominence, many books and manuals on the techniques of playing the game were also published, they played an important role in popularizing xiangqi and improving the techniques of play in modern times. With the economic and cultural development during the Qing dynasty, xiangqi entered a new stage. A Western-style Encyclopedia of Chinese Chess Openings was written in 2004.^[23]

Although xiangqi has its origin in Asia, there are xiangqi leagues and clubs all over the world. Each European nation generally has its own governing league; for example, in Britain, xiangqi is regulated by the United Kingdom Chinese Chess Association. Asian countries also have nationwide leagues, such as the Malaysia Chinese Chess Association.^{[citation needed]}

In addition, there are several international federations and tournaments. The Chinese Xiangqi Association hosts several tournaments every year, including the Yin Li and Ram Cup Tournaments.^[24] Other organizations include the Asian Xiangqi Federation^[25] and a World Xiangqi Federation,^[26] which hosts tournaments and competitions bi-annually, with most limited to players from member nations.

There are Europeanized versions of boards (10 × 9) and figures of xiangqi.^[27][28][29]

The Asian Xiangqi Federation (AXF) and its corresponding member associations rank players in a format similar to the Elo rating system of chess. According to the XiangQi DataBase, the top-ranking female and male players in China, as of June 2012, were Tang Dan and Jiang Chuan, with ratings of 2529 and 2667, respectively.^[30][31] Other strong players include Zhao GuanFang (female), Xu Yinchuan (male), Lu Qin (male), and Wang LinNa (female).^{[citation needed]}

The Asian Xiangqi Federation also bestows the title of grandmaster to select individuals around the world who have excelled at xiangqi or made special contributions to the game. There are no specific criteria for becoming a grandmaster and there are only approximately 100 grandmasters as of 2020.^[32] The titles of grandmaster is bestowed by bodies such as the AXF and the Chinese Xiangqi Association (CXA).^[33]

The game-tree complexity of xiangqi is approximately 10¹⁵⁰; in 2004 it was projected that a human top player would be defeated before 2010.^[34] Xiangqi is one of the more popular computer-versus-computer competitions at the Computer Olympiads.^[35]

Computer programs for playing xiangqi show the same development trend as has occurred for international chess: they are usually console applications (called engines) which communicate their moves in text form through some standard protocol. For displaying the board graphically, they then rely on a separate graphical user interface (GUI). Through such standardization, many different engines can be used through the same GUI, which can also be used for automated play of different engines against each other. Popular protocols are UCI (Universal Chess Interface), UCCI (Universal Chinese Chess Interface), Qianhong (QH) protocol, and WinBoard/XBoard (WB) protocol (the latter two named after the GUIs that implemented them). There now exist many dozens of xiangqi engines supporting one or more of these protocols, including some commercial engines.^{[citation needed]}

Blitz chess
Each player only has around 5–10 minutes each.

Manchu chess
Invented during the Manchu-led Qing dynasty. Red horses, cannons, and one of the chariots are absent, but the remaining chariot can be played as horses and cannons as well.

Supply chess
Similar to the Western chess variant Bughouse chess, this variant features the ability to re-deploy captured pieces, similar to a rule in shogi. Four players play as two-person teams in two side-by-side games. One teammate plays Black and other plays Red. Any piece obtained by capturing the opponent’s piece is given to the teammate for use in the other game. These pieces can be deployed by the teammate to give him an advantage over the other player, so long as the piece starts on the player’s own side of the board and does not cause the opponent to be in check.

Formation
Similar to Fischer Random Chess, one player’s pieces are placed randomly on one side of the river, except for the generals and advisors, which must be at their usual positions, and the elephants, which must start at two of the seven points they can normally reach. The other player’s pieces are set up to mirror the first’s. All other rules are the same.

Banqi
This variation is more well known in Hong Kong than in mainland China. It uses the xiangqi pieces and board, but does not follow any of its rules, bearing more of a resemblance to the Western game Stratego as well as the Chinese game Luzhanqi.

There are many versions of three-player xiangqi, or san xiangqui, all played on special boards.

San Guo Qi
“Game of the Three Kingdoms” is played on a special hexagonal board with three xiangqi armies (red, blue, and green) vying for dominance. A Y-shaped river divides the board into three gem-shaped territories, each containing the grid found on one side of a xiangqi board, but distorted to make the game playable by three people. Each player has eighteen pieces: the sixteen of regular xiangqi, plus two new ones that stand on the same rank as the cannons. The new pieces have different names depending on their side: huo (“fire”) for Red, qi (“flag”) for Blue, and feng (“wind”) for Green. They move two spaces orthogonally, then one space diagonally. The generals each bear the name of a historical Chinese kingdom—Shu for Red, Wei for Blue, and Wu for Green—from China’s Three Kingdoms period.^[36] It is likely that San Guo Qi first appeared under the Southern Song dynasty (1127–1279).^[37]

San You Qi
“Three Friends Chess” was invented by Zheng Jinde from Shexian in the Anhui province during the reign of the Kangxi Emperor of the Qing dynasty (1661–1722). It is played on a Y-shaped board with a full army of xiangqi pieces set up at the end of each of the board’s three wide radii. In the centre of the board sits a triangular zone with certain features, such as ocean, mountain, or city walls, each of which is impassable by certain pieces. Two of an army’s five soldiers are replaced by new pieces called huo (“fire”) pieces, which move one space diagonally forward. Two qi (“flag”) pieces are positioned on the front corners of the palace; they move two spaces forward inside their own camp, and then one space in any direction inside an enemy camp.^[37]

Sanrenqi
“Three Men Chess” is a riverless, commercial variant played on a cross-shaped board with some special rules, including a fourth, neutral country called Han. Han has three Chariots, one Cannon, and one General named “Emperor Xian of Han”, but these pieces do not move and do not belong to any of the players until a certain point in the game when two players team up against the third player. At that point the third player gets to also control Han.^[37]

Si Guo Qi
“Four Kingdoms Chess” is also played on a riverless, cross-shaped board, but with four players. Because there are no rivers, elephants may move about the board freely.^[37]

Qi Guo Xiang Qi
“Game of the Seven Kingdoms” is based symbolically on the Warring States Period.

Xiangqi pieces were added to Unicode version 11.0 in June 2018. They are assigned to the codepoints U+1FA60–U+1FA6D in the Chess Symbols block. For legibility, the red pieces are filled white, while black pieces are filled black.

• Lau, H. T. (1985). Chinese Chess. Tuttle Publishing. ISBN 0-8048-3508-X.
• Leventhal, Dennis A. The Chess of China. Taipei, Taiwan: Mei Ya, 1978. (out-of-print but can be partly downloaded)
• Li, David H. The Genealogy of Chess. Premier Publishing, Bethesda, Maryland, 1998. ISBN 0-9637852-2-2.
• Murray, H. J. R. (1913). A History of Chess (Reissued ed.). Oxford University Press. ISBN 0-19-827403-3.
• Wilkes, Charles Fred. A Manual of Chinese Chess. 1952.

• Li, David H. First Syllabus on Xiangqi: Chinese Chess 1. Premier Publishing, Bethesda, Maryland, 1996. ISBN 0-9637852-5-7.
• Li, David H. Xiangqi Syllabus on Cannon: Chinese Chess 2. Premier Publishing, Bethesda, Maryland, 1998. ISBN 0-9637852-7-3.
• Li, David H. Xiangqi Syllabus on Elephant: Chinese Chess 3. Premier Publishing, Bethesda, Maryland, 2000. ISBN 0-9637852-0-6.
• Li, David H. Xiangqi Syllabus on Pawn: Chinese Chess 4. Premier Publishing, Bethesda, Maryland, 2002. ISBN 0-9711690-1-2.
• Li, David H. Xiangqi Syllabus on Horse: Chinese Chess 5. Premier Publishing, Bethesda, Maryland, 2004. ISBN 0-9711690-2-0.

• Xiangqi.com Play Xiangqi for free
• Xiangqi Championships
• Learn Chinese Chess in English Rules, openings, strategy, ancient manuals
• An Introduction to Xiangqi for Chess Players
• Xiangqi, Chinese Chess Presentation, rules, history and variants, by Jean-Louis Cazaux
• Xiangqi (象棋): Chinese Chess by Hans Bodlaender, ed. Fergus Duniho, The Chess Variant Pages