Go (game)

from Wikipedia, the free encyclopedia
A Japanese Go table, called
Goban ( Japanese 碁 盤 , ご ば ん )

Go ( Chinese  圍棋  /  围棋 , Pinyin wéiqí , Jyutping wai 4 kei 4 * 2 ; Japanese 囲 碁 igo ; Korean 바둑 baduk ; literally " encirclement game ") is a strategic board game for two players. The game originally comes from ancient China and has taken a special form in Japan , Korea and Taiwan over the course of history . It was not until the 20th century that Go began to spread outside of East Asia . According to information from Mind Sports Online , the number of Go players worldwide was around 27 million in 2000, of which 22 million lived in East Asia alone. According to this calculation, Germany was in first place among the Western European countries. The International Go Federation (IGF) put the number of Go players worldwide at around 40 million in 2011. The British Go Association (BGA) put the number of global players at 60 million in 2013.


Game board

Two players try to conquer territories by alternately placing lenticular pieces on the intersection of the playing field lines, to secure them and, if possible, to "capture" the opponent's pieces. By laying suitable formations, secure positions can gradually be created. At the end of the game, the size of the territories conquered is compared and the number of game pieces caught is added. So the goal is not to completely destroy the enemy, but to get more points from conquered territories and prisoners. On the one hand, Go is a complex and deep game, which means that a player can work on refining his style and skill level throughout his life. The number of playable variants even exceeds that of chess by orders of magnitude. Second, the basic rules (there are only four) are so simple that they can be learned quickly.

The appeal of the game is that players should keep an eye on both local situations and the bigger picture at all times. A locally lost situation can still play an important role at a later point in time. Each piece that is placed often has several functions, from strengthening its own group of pieces to creating a connection with a second group to attacking territory dominated by the enemy. Therefore, depending on their disposition, players can concentrate more on creating as large territories as possible, or on preventing this with their opponents.

Weaker players often strive for safe territory at an early stage, while strong players often only make their territories safe territory in a late phase of the game.

Different levels of play can be balanced by using up to nine handicap stones . As a result, even a weaker player has a chance of victory and a sense of achievement, while the stronger player still faces the challenge of winning despite the handicap.

The strategy is an important, but not the only aspect of the game: Go should stimulate meditation and convey fundamental insights into the laws of nature, pose challenges to the mind and offer some players a mirror of their own personality. In addition, a game played well by both players can be perceived as a work of art.

The name Go

The game Go has had various names in China throughout history.

The common name for Go in China today is " Weiqi " ( 圍棋  /  围棋 ). There is also the rarer term “ Yiqi ” ( 弈棋 ) or “ Yi ” ( ) for short . In ancient China or in classical Chinese texts, the game is often simply referred to as “ Qi ” ( ), pronounced “ Go ” ( Japanese / / , kana ) after the Japanese pronunciation . The name “ Go ”, which is now established in the western world, has its roots in the classic name of this game after the Japanese pronunciation.

In ancient China, for example, B. preoccupation with playing one of the four classical arts , also called “ Qin , Qi , Shu , Hua ” ( 琴棋 書畫  /  琴棋 书画 ) in Chinese , which an educated person should be able to master.


Go player of the Song era ( 重 屏 會 棋 圖 , Chóngpíng Huìqí Tú ) - Chin. Art , Zhou Wenju 942-961

The origins of the game are largely in the dark. According to some authors, passages in the Zuozhuan annals (4th century BC) already refer to the game known in China as Weiqi . Reliably assigned references and archaeological finds date from shortly after the turn of the century . That is why Go, along with backgammon and mill, is one of the oldest known strategy games in the world. During the Han period , Weiqi spread rapidly among the population and became an accepted pastime among the civil servants. Weiqi experienced its first heyday during the Tang Dynasty , so it was played extensively at the imperial court. The Tang period was a particularly significant period in Chinese history , when the culture experienced a climax. The imperial bureaucracy required innumerable officials, which resulted in a well-trained class interested in the Weiqi game.

The board game was to retain its great appeal even under later dynasties. The Song Emperor Huizong is said to have been an enthusiastic Weiqi player, as was the first Ming Emperor Hongwu , who lost a famous game against his General Xu Da and then had to give him his garden villa in Nanjing . The game was also very popular at the imperial court of the Qing Dynasty . With the fall of the Empire in 1911, Weiqi sank into a crisis, as the cultivated upper class of China largely broke away. Only after the Cultural Revolution did Weiqi experience its renaissance among the Chinese people.

Legend has it that Kibi no Makibi (695–775) brought the game 735 to Japan . He was appointed envoy to the Chinese capital Chang'an , the political and cultural center of the world at that time. There he was supposed to study sciences and arts at the court of Tang Emperor Xuanzong . He stayed in China's capital from 717 to 735. On his return trip he is said to have taken a Weiqi game with him, which he then made known in his homeland under the name Go . It may actually be thanks to Kibi no Makibi that this game was introduced to the Japanese aristocracy, as the refined culture of the Tang rulers was considered exemplary for the Japanese. Nevertheless, the Japanese character for Go ( ) can already be found in Kojiki from the year 712, which suggests that the game was already known in Japan before Kibi.

Edo Period Go Players - Jap. Art , Kikukawa Eizan circa 1811

With the beginning of the Edo period in the early 17th century, the political situation in Japan changed fundamentally. The new Shogun from the Tokugawa house was very fond of Go and promoted it strongly: He occupied the post of Godokoro ("Go Minister") and left the o-shiro-go , a Go ceremony in the presence of the Shogun which the strongest Go player has been determined. Scholarships were given to the strongest players of the four big Go schools that came into being around this time: the Honinbo School, the Inoue School, the Yasui School and the Hayashi School.

There was great rivalry among these four schools, which helped the game of Go to a level never reached before. Among other things, a ranking system was introduced during this time, which was based on that of the martial arts. The best player of the Edo period , Shusaku Kuwahara , developed, among other things, a new opening, the Shusaku opening named after him, which was played well into the 20th century. Shusaku won the annual o-shiro-go 19 times in a row before dying of cholera during an epidemic at the age of 33.

The government ended support for the Go schools in 1868 after the fall of the Tokugawa shogunate. Over time, however, daily newspapers took on the role of Go sponsors, so that the high level of Japanese Go was maintained. In honor of the Honinbo School, from which most of the strongest players in premodern Japan came from, one of the most prestigious Japanese championships still bears the name Honinbo . The heyday that Japanese Go experienced through its early promotion may be one reason why the game is better known in Western languages ​​by its Japanese name than its original Chinese name.

For a long time Go was a male domain, similar to chess . However, the opening of tournaments and the rise of strong female players, primarily Rui Naiwei , have increasingly demonstrated the competence and skill level of female players. Since the late 1980s there has been a real go-boom in China and especially in Korea , which has led to Japan losing its former supremacy in international tournaments.

There are an estimated ten million Go players in Japan. Since 1998, the Japanese manga and anime series Hikaru no Go , the story of which is about Go players, has greatly increased the popularity of Go among children and young people. All over the world, the number of Go clubs, Go AGs and young Go players has increased significantly since then.

Go became known in Europe in the 1880s through a series of articles by Oskar Korschelt : The Japanese-Chinese game "Go". A chess concurrent . A German Go newspaper appeared as early as 1909, but it was not until the 1950s that the game of Go slowly spread. Clubs emerged and the first regular tournaments took place. The German Go Association has over 2000 members today. The number of all Go players in Germany should be around 20,000.


Ming Age Go Player - Jap. Art, Kanō Eitoku (1543–1590)

There are various legends about the creation of the game that illustrate the philosophical ideas and cultural values ​​behind the Go. According to tradition, the game was designed by the mythical great emperor Yao as a teaching tool for his son Shun to teach him discipline, concentration and mental balance. After all, the son became the first great player and also a good emperor with a balanced human nature. Another suspected genesis of the game indicates that in ancient times, Chinese warlords and generals used pieces of a stone to depict positions on the battlefield. These legends reflect the two fundamental ideas of Go: the development of one's character and the illustration of the competition between two elements. Often one refers to the elements yin and yang , which are rooted in Daoism and which act as driving forces on the go board.

Many claim that the game cannot be understood through logic alone, but that its complexity and depth can only be mastered through intuition.


In a nutshell: The players take turns placing their own stone on the intersection of the lines of the board. You can hit opposing stones and groups of stones by enclosing them all around. In the end, the player who controls the greater part of the board wins.

The basic rules have remained unchanged since the game was created. A Japanese version of the rules that is popular in Germany is presented here. Other rules (the Chinese rules or the Ing rules) differ in details. For example, the counting is done differently at the end of the game, but this almost always leads to the same winner.

Game components and sequence of moves

The playing field consists of 19 horizontal and 19 vertical lines that form a grid of 19 × 19 = 361 intersections. The stones are placed on these points. Both players, called black and white , have in principle an unlimited supply of stones of the same type in the respective color. Traditionally there are 181 black and 180 white stones, which is almost always sufficient. For shorter games and especially (but not only) for beginners, smaller game boards are suitable, mostly in the size 13 × 13 or 9 × 9. The rules of the game are the same for all board sizes.

The board is empty at the beginning, unless the weaker player is allowed a handicap. The players take turns, Black begins. The player whose turn it is may place a stone from his supply on any empty spot. In contrast to chess , however, there is no compulsory move, i.e. a player may also forego his move ( pass ) . The game usually ends when both players pass one after the other. They do this when they realize that further betting would not mean gaining or even losing points. Here, too, the “soft” philosophy of the game is expressed in that no player is forced to make a move that is unfavorable for him.

Stones that have been placed are no longer moved in the rest of the game. (This is why some players, especially following a tradition in the GDR , do not speak of “moves” but of “sets”.) However, stones can be struck under certain conditions . H. removed from the board.

Hitting stones

A single rock is beaten (one says also captured or killed ) and removed from the board when his final freedom has been occupied by an opponent's stone. Freedoms are the vacant points adjacent to a stone. Points are adjacent if they are right next to each other and connected by a line on the game board. This means that horizontally and vertically adjacent, but not diagonally opposite points are adjacent. A point in the middle has four, one on the edge three and one in the corner only two neighboring points. The picture below shows five individual black stones, four of which only have one freedom (indicated by a square).

The same applies to chains of stones. Several stones of the same color that are connected by being adjacent to the next form a chain. In the picture there is a chain of three black stones on the right edge, above a chain of two white stones. A stone in a chain does not have to have a free neighboring point, only the entire chain is considered. The freedoms of a chain are the vacant points that are adjacent to one of its stones. The black chain in the picture only has one freedom (square). If the opponent occupies the last freedom of a chain, he beats all the stones in the chain. You can only beat a chain as a whole, not just part of a chain.

A single stone can also be understood as a chain that consists of only one stone. From now on, “chain” should also refer to a single stone. A train can take the last freedom of several chains at the same time. In any case, all opposing chains that no longer have freedom are defeated.

If a chain only has one freedom, then they say it is in the Atari . The opponent threatens to beat her on the next move. To prevent this, it can be useful to give her additional freedom by adding a stone. In the picture, black could therefore rely on one of the freedoms of a stone in order to save it (at least temporarily). In the case of the chain of three, however, this would be of no use here, because afterwards it would only have one freedom (below the square) and could be beaten immediately.

Struck stones are removed from the board and kept as “prisoners”. Each prisoner counts one point and is added to the own area points in the final account.


It is not allowed to place a stone in such a way that the chain to which it belongs has no freedom after the move. The complete execution of a move also includes the removal of stones. Therefore it is not suicide if a train simultaneously takes the last freedom of a chain of its own and that of an opponent. Because the own chain is given freedom again by removing the opposing chain.

There are also rule variants that allow suicide. Then the following applies: If a move does not hit any opposing stones and the chain with the stone placed has no freedom, then this chain is hit itself and its stones count as prisoners for the opponent (in these rule systems, however, the prisoners usually do not leave at all in the result, only the area counts). In practical play, however, there is hardly any difference, because it rarely makes sense to hit your own stones.

A Kō situation

The immediate repulsion of a single stone that has just hit a single stone is prohibited. In other words: A stone may not be captured if the same arrangement of stones would arise afterwards as after the previous move. Such a situation is called (derived from Japanese , pronounced koh , in German eternity). The purpose of the Kō rule is to prevent the position from being repeated endlessly.

At the same time, an interesting tactical element comes into play in the form of the Kō fight . If player A has beaten in Kō, player B plays a "threat" ( Kō threat ) as an intermediate move , to which A must respond. This prevents A from occupying and securing the Kō point. Then B may hit again in the Kō. This can be repeated any number of times until a player no longer has a Kō threat or a threat is too small so that the opponent does not answer it. Players must weigh the value of the threat against the value of the Kō win. In addition, before the beginning of a valuable Kō, the totality of all threats that are available to both players must be considered in order to decide whether to get involved in the Kō fight.

This "simple" Kō rule does not prevent all possible repetitions of the position. If there are about three different Kō situations on the board, you can always strike back in at least one of them. In such a situation, if no player wants to deviate from the replay, the game according to Japanese rules ends with no result and is replayed. But such a case is extremely rare.

As an alternative, some rule systems use a global Kō rule , also called the Superkō rule . There are slightly different variants. For example, a Superkō rule forbids placing a piece so that the resulting arrangement of the pieces on the board matches any previous arrangement and it is the same player to move and the difference in pieces taken is the same (i.e. with the moves between them both players have captured the same number of stones). An endless cycle, from which no player should deviate in self-interest, can no longer occur.


An end situation, prisoners are already removed. White has 15 and Black 20 area points. If White has caught 6 stones and Black 4, then Black wins with points opposite if no Komi has been identified

The game is over when both players pass one after the other. Passing at the end of the game is in the interests of the respective player. Otherwise he would reduce his own area or give the enemy prisoner stones unnecessarily. A player's score is the sum of the free points (area) enclosed by stones of their own color and the stones caught (of the opponent's color). The player with the higher score wins the game.

If at the end there are still stones on the board that can be struck, i.e. are dead, then they are considered prisoners. They are removed from the board prior to the area count and counted along with the stones hit. You agree on the status of these stones with your opponent after the game is stopped.

This agreement is unproblematic among experienced players, because most of the time it is obvious which stones are dead and thus trapped. If there is any disagreement, then the situation must be played out: In this case, the game continues, and whoever has claimed that opposing stones are dead must prove it by hitting them. If he fails to do this, they are considered alive. The stones placed when playing may not influence the count. You must either restore the situation before the lead or compensate for the stones placed in your own or the opponent's area in a suitable way.

If both players have the same score, the game is a tie, which is called "Jigo". Both players also have the option to abandon the game if the situation on the board seems hopeless. The opponent has then "won by giving up".

It is sometimes difficult for beginners to tell when the game is over. In the example on the right, the borders where black and white stones touch are completely played out, so that there are no longer any free intersections between stones of different colors. This is a good indication that the game is over. According to the rules, it is in principle possible for the game to continue "one-sided", namely if one player is still betting because he believes he can make worthwhile moves, while the other player does not share this assessment and therefore does not answer moves. Since one ultimately gives the opponent trapped stones through hopeless attack moves, it would not be beneficial for him to react in every case. He would give up this point gain by making counter moves on his own territory that was already safe.


At the start of the game there is a slight disadvantage for White, since Black has the advantage of the first move. This disadvantage is usually compensated for by a "compensation" in the form of additional points to the white player. These points are called komi ( コ ミ ) and vary depending on the rules or agreement between players. To avoid a tie, a komi with half a point is usually chosen; common values ​​are 5½ or 6½. Sometimes only ½ point is given when it is important to avoid jigo . The fair value is still the subject of debate, and so the disadvantage of playing white is compensated for by up to 8½ points in some tournaments. The problem can be solved through a kind of Komi auction or an exchange rule, for example by one player specifying the Komi and the other then choosing a color. But that has hardly caught on. The komi can also be used to replace or add to handicap stones (so-called return comi , when black gets komi ).

Concepts of strategy and tactics

life and death

An example of a group with a "fake eye". White can capture a black stone and black can put atari.
An example of a Seki situation; no player can add a stone without endangering his group.

Victory points for an enclosed area are only awarded when the game is over; As long as a player encloses free points with his stones in the course of the game, this area is initially only claimed by him. Two possibilities are conceivable as to how the opponent can still dispute it: Firstly, if the opponent succeeds in permanently settling with his stones inside the claimed area without being hit. This is easier the larger the area claimed. Second, when the groups claiming territory can in turn be surrounded and defeated by the enemy. Both scenarios lead to the realization that the survival of the claimed area depends on whether the groups of stones that are decisive for it can still be cut or not. A group that can no longer be beaten under any circumstances is said to be alive . Similarly, a group is dead if there is no way to save it from being beaten.

The reason why a group can be (live) unbeatable is as follows: If a group includes a single free intersection (which is called internal freedom) and is completely surrounded by opposing stones (i.e. has no external freedom), the opponent can put a stone on this last freedom of the group and hit them with what is called blasting . However, if the group also encloses a second free intersection that is not adjacent to the first intersection, the opponent cannot bet on either of the two intersections, since there is still a freedom - and two stones cannot be placed at the same time. Therefore, the following sentence also applies: A group lives when the area that surrounds it is divided into two separate areas or can be unconditionally divided in this way.

These areas are called eyes. Eyes can contain a single intersection, but also several neighboring intersections. In addition, prisoners may also be in one eye. One difficulty is that there are also “false eyes”. These are surrounded by stones of one color, but not by a continuous chain . This means that part of the surrounding stones can be placed separately in "Atari" by a series of moves by the opponent. Thereafter, the other player could close the supposed eye to prevent hitting the partial chain, or accept hitting. In both cases the eye is destroyed. In general, the following applies: only a group with two “real” eyes lives unconditionally.

Another way to live is the seki : this is a kind of local stalemate in which neither player can occupy the freedoms of the opposing group without depriving his own group of vital freedoms. In a situation where the player who bets the first move loses his group, neither player will bet. For the final settlement, these groups will remain on the board even if they do not enclose an area. In this way, permanent neutral points can arise on the game board, i.e. free points that still do not represent an area.

Life and death is the most fundamental and important element of strategy in the game of Go, and is critical to the course and outcome of a game. If a group is dead, it is also trapped and at the end counts points for the opponent, even without the situation having to be played out until the final capture. Often the life and death of a group depends on who makes the next move because, depending on who's it is, they can often be killed in one move or brought to life. Because of the great importance of life and death to the game of Go, players should be aware of the life and death of all groups at all times during the game. Because adding stones to an already dead group is just as pointless as securing groups that are already alive. On the other hand, trains that threaten a living group or trains that could bring a dead group to life are classic co-threats (see above). Hence, practicing life-and-death problems is essential for anyone looking to improve their skills.


Opening of a game of go
Game in the Fuseki phase

The opening of a Go game is roughly the first 30 to 40 moves. Since the board is empty at the beginning, there are theoretically an immeasurable number of playable variants for the first moves. Still, certain moves have proven to be particularly good. Almost every game starts with a move near a corner. Only after all four corners have been filled with one or two stones each, the sides are filled. Then the expansion of the positions into the center begins.

With the first stones that are placed on the board, you try to achieve the most perfect balance possible. This means that the stones should neither be too close together nor too far apart and neither too high nor too low, and also that you can react flexibly to the actions of the opponent with the stones placed. This again shows that Go is a game of balance in many ways (see Philosophy section ).

For advanced players, the opening game is characterized by the use of full-board patterns ( Fuseki ) and fixed corner game sequences ( Jōseki ). Fuseki and Jōseki are the most variable elements of the game of Go and are constantly being developed. The number of different openings in go exceeds that of openings in chess many times over. Very experimental openings are also occasionally played.

Territory and Influence

Territory and influence are strategic concepts of the Go. An area- oriented style of play places special emphasis on fixed, safe positions in the corners and on the edge of the board (this is where it is easiest to make an area because you no longer have to surround it on the edge of the board). This has the advantage that you mark out safe areas at a relatively early stage of the game and thus collect safe points. Later on it is all the more important to reduce the size of the opponent's territories as much as possible. A suitable means of doing this is invasion (building up a living group in the enemy's sphere of influence). Area-oriented play therefore sometimes also requires risky tactical maneuvers.

On the other hand, it is possible to play in an influence-oriented manner. In a certain way, this represents the counterpart to the area-oriented game. Above all, one tries to build up strong positions that often look like “walls” and are directed towards the center. This does not initially create an area, but rather has an influence on the surrounding parts of the board. Influence-oriented players anticipate battles in their area of ​​influence, i.e. in situations that are advantageous for them. Fixed territory only emerges as a result of these battles.

Attack and defense

A go game in the middlegame (the last move is marked).

In the middle game, which begins after the last opening moves, fights often arise. Among other things, the following tactical and strategic means are used:

  • It is often beneficial to separate opposing stones from one another . The reason is that separate groups are on their own and then have to establish a living position independently of one another. Instead of making territory, the affected player has to make many moves in a confined space in order to secure the two eyes of his groups. But if his groups were connected , it would be much easier for them to keep the necessary space for eyes. Conversely, it is of course just as important to connect your groups with one another as much as possible .
  • The catch or killing of a group means that the challenged group enclosed by the opposing pieces and not enough internal freedoms has to two eyes to make (s. O.).
  • A group that has no eyes and is in danger of being caught can try to escape , that is, to spread in any direction until a connection with another group or two eyes can be established. A certain willingness to make sacrifices is very important here. Instead of trying to save every single stone, one should play moves that develop the position quickly and are flexible. The loss of part of the group may have to be accepted in order to secure at least the other part. This is called "easy" play.
  • So-called good form is necessary for successful fighting. Many stone patterns have proven to be "good" because they have positive properties in combat, such as the greatest possible number of freedom, the smallest possible number of superfluous stones or good developability. "Bad shape" is the term used to describe chains that are clumped together in a narrow space and that are easy to beat due to the lack of freedom. "Good forms" are tried and tested standard patterns, but do not necessarily represent the best move in every game situation.
  • Unusual moves of particular efficiency for certain tactical maneuvers (saving or catching stones, gaining the forehand , breaking out of an encirclement), or quite simply the best move in a tactical standard situation is called Tesuji . Tesujis can e.g. B. consist of sacrificing individual stones in order to beat opposing stones in exchange or to gain another advantage.

Forehand and afterhand

The middlegame moves on to the endgame, which is mainly about pinpointing the boundaries between the areas. As a rule, in this phase of the game it is already clear which groups are alive and which are dead. The goal is then to reduce the opponent's areas as much as possible and to enlarge your own.

Here is another strategic aspect plays a major role and that is the use of forehand ( sente , 先手 ) and hind ( gote , 後手 ). Forehand means that every move you make requires a reaction from your opponent. A forehand sequence can consist of any number of moves, as long as it ends with only one safety move by the opponent. After each sentence sequence , the first player retains the initiative and can continue playing at another point. Gote (afterhand) means exactly the opposite, namely to have to make the last move at the end of a sequence of moves. Then the opponent takes the initiative. Maintaining the Sente (forehand) often brings game-deciding points in the endgame. Also in the middle game and in the opening certain moves can be called Sente if they have to be answered locally in order to avoid a bigger loss of points. Out of consideration for potential knockout threats (see above), such forehand sequences should not be played out too early.

Tradition of the game of Go

Traditional game material

Although you can of course play Go on a piece of cardboard and with a bag of plastic stones, the Japanese Go culture places particular emphasis on high-quality play sets.

In China, people traditionally play on flat wooden boards that are up to about 5 cm thick. Today, as then, you mostly sit on chairs at a table. In Japan, on the other hand, Go is ideally played on the floor, with the players sitting on flat cushions ( zabuton 座 布 団 ). The traditional go board ( goban 碁 盤 ), which is located in front of the players on the ground, is also made of solid wood, but about 15 cm to 20 cm thick and stands on short legs. The most valuable boards are made from the rare, golden yellow wood of the kaya tree ( Torreya nucifera ), some from the wood of trees over 700 years old. The grid lines that represent the playing field are still occasionally scratched into the surface of the wood by their own professionals with a sword ( katana ) and then traced with varnish.

The Japanese go board is not perfectly square. The playing field traditionally measures 1 Shaku and 5 Sun in length and 1 Shaku and 4 Sun in width (455 mm × 424 mm), whereby some space must be left at the edges so that playing on the edge lines and corner points is possible . These dimensions describe a ratio of 15:14. The extended length is used to compensate for the optical distortion (shortened perspective) that arises from the fact that the players do not look straight at the board, but at an angle from above. Another reason given is the Japanese aesthetic, which avoids perfectly symmetrical structures and thus also a perfect square.

Wooden cans with go stones

The game pieces ( go-ishi 碁 石 ) are preferably made of white shells or black slate, ground ellipsoidally and are kept in wooden boxes ( goke 碁 笥 ). Since the corresponding resources are limited (mussels and kaya trees take a long time to reach the required size and are now very rare), traditionally manufactured game material can often only be purchased at exorbitant prices.

The containers for the stones are simply shaped like an ellipsoid with a flattened bottom. The loosely fitting lid is turned over during the game and serves as a container for captured opposing stones. The containers are usually made of turned wood; small woven bamboo baskets are also common in China.

These traditional Japanese sets are usually not in use at go clubs and championships where a large amount of sets are maintained (and also purchased). Most of the time, games are played on western tables and armchairs. For such situations, 2 cm to 5 cm thick table boards without legs are used. The stones are mostly made of glass, the boxes made of plastic. Table boards and glass stones are also the most common in Europe. Although cheap plastic stones are also in circulation, these are rejected by many players due to their low weight and the correspondingly unsatisfactory haptic and acoustic experience when placing the stone.

Holding technique of the go-stones

Experienced players distinguish themselves in the entire Go world by a special way of placing Go stones on the board: The stone is held between the middle finger and forefinger in order to hit the board firmly, with a full "click" being heard . Ideally, the stone does not wobble after letting go. The quality of the game material can of course influence the acoustics of the move. The pyramid-shaped hollow on the underside of a traditional Japanese go board is sometimes explained as improving the sound. A game board is also considered to be more noble if slight traces of stones are visible that have slipped over it over the decades - or centuries.

Behavior on the go board

The etiquette of Go is considered by many players as important and followed. Accordingly, one should always show the opponent the necessary respect so that he does not perceive the game played as unpleasant. First of all, it is fundamental what attitude you have towards the game. You can play to relax, have fun, study and much more. In any case, one should appreciate the attitude of one's opponent. A one-sided fixation solely on winning the game contradicts the philosophy of the game, which is anchored in East Asian culture. Thus, bragging about a victory, ridiculing about a defeat and the like clearly violate the good morals of the game of Go.

More manners
  • Usually, players greet each other before the start of a game.
  • When playing at the table, it is considered polite if the player's first move takes place in the top right corner. The reasons for this are first that the other can comfortably place his first stone (and the box is usually to the right of the board, so the front right corner is made attractive for white), second, that there is a uniformity in the notation, and third, showing respect for your opponent by “bowing” to him to set the stone.
  • It is perceived as very annoying to distract the opponent with noises (stir with your hand in the can). The concentration on the game should not be impaired if possible.
  • Doing other things at the same time (looking at other boards, listening to music) makes the other person feel bored, which can be perceived as pejorative.

On Go servers on the Internet (see web links ), the usual game situation in which you sit across from one another at the table is shifted to a chat room. Of course, some of the above rules are invalid here. But here, too, there are norms, for example that you greet each other briefly at the start of the game and that you do not leave the game without a message. Players who regularly abandon games in this way when they are about to lose are called escapers. Most Go servers have mechanisms that ensure that escapers do not take advantage of their abortion. The public denunciation of escappers ("xxx is an escaper!") Can be observed again and again, but is not part of good behavior on the Go server either.

Classification and ranking systems

Go players who play in clubs and tournaments usually wear a rank that may include a. serves as a guide when choosing a game partner.

  • Master ranks, which are referred to as Dan , theoretically range from 1st to 9th Dan. The 1st Dan is the lowest master level, a 7th Dan for amateurs (in Japan rarely also the 8th Dan) the highest.
  • Student ranks, called Kyū , are graded from 30th to 1st Kyū, with 1st Kyū being the highest rank. Beginners are usually classified as 20th to 30th Kyū.

The three leading Go nations Korea, China and Japan each have their own ranking systems for professional players, which also range from 1st Dan to 9th Dan. Professional ranks are awarded by the associations on the basis of tournament results or, in exceptional cases, on an honorary basis. In the amateur field, with a few exceptions, it is a system of self-classification. For example, in the GDR, high school ranks and all master ranks were awarded on the basis of tournament results according to fixed rules. In Japan, amateurs' level certification fees are an important source of income for Nihon Kiin , the largest organization of professional Go players.

The ranking systems in America, Europe and Asia are slightly shifted from one another, but the difference in skill level between the respective ranks is always the same for the amateurs. It is measured according to a fixed system of handicap stones to compensate for the difference in skill level. A 1st professional dan in Japan is roughly equivalent to a 7th dan for amateurs.

If two Go players of different ranks meet, a handicap is determined from the difference in rank : A 1st Dan receives a handicap of 4 stones against a 5th Dan. This means that the weaker player plays with the black pieces and is allowed to place 4 pieces on the board before his opponent makes the first move. In Japan and also in Europe, the default stones are placed on the intersection points that are drawn a little thicker on the Go board. These nine points, which are arranged axially and point-symmetrically, are called hoshi ( "star"). In China, on the other hand, it is common for the weaker player to choose where he wants to place his handicap stones.

If there is a difference of only one rank, the weaker player begins without placing handicap stones. In the case of equally strong players (tie game), the player who follows (White) receives a few points ( called Komi ) in advance , which compensate for the advantage that Black has from the first move. The standard komi in Japan and Europe are 6 or 6.5 points and in China 7 or 7.5 points. Non-integer komi is used when trying to rule out a tie ( jigo , 持 碁 ). The height of the Komi is generally freely selectable (by tournament organizers).

In the case of professionals, roughly three ranks correspond to a stone requirement and thus the difference from an amateur rank.

Time systems

Tournaments are usually played with a certain time limit. The basic playing time is measured by means of a chess clock during the thinking time of each player. It can range from ten minutes (blitz tournament) to an hour (average national tournament) up to eight hours (Japanese title fights). Often the players have additional time after the end of the basic game time, which is called Byōyomi ( 秒 読 み ). There are two types of Byōyomi:

  • In the classic Byōyomi, each player has a certain number of Byōyomi periods with a specific time (often 30 seconds). If he manages to make his move within this time, the period is considered not to have been touched and starts all over again with the next move. However, if it takes longer for the move, one period is used up, and he has one less period for the rest of the game. If all periods are used up, he loses the game.
  • In the Canadian Byōyomi the player has to place a certain number of stones in a certain time (15 stones in 5 minutes). If he succeeds, the period begins again and he must again place the specified number of stones (with progressive Byōyomi even more and more stones) in the given time. If he fails to do this, he loses the game.
  • With the progressive Byōyomi the player has to place more stones in each period (typically 15 stones in the first 5 minutes, then 20 stones in 5 minutes, then 25 stones in 5 minutes, ...).

Since classic chess clocks are overwhelmed by these time systems because the remaining time has to be reset too often, there are also special (electronic) Go clocks that can cope with the comparatively complicated time rules of Go.

Before such clocks existed, time had to be measured manually, i.e. by a person. In the classic Byōyomi, a timekeeper had a clock and informed the players by announcement how many seconds they still had for the train. This led to an increased noise level, especially at tournaments.

Professional go

Professional Go has mainly developed in Japan , Korea , Taiwan and China . In Japan, the game has been subsidized by the state since the 17th century. This funding was limited to a few families, but it laid the foundation for the modern professional system that was subsequently established in the other East Asian countries. Go professionals enjoy a high status and can make a living just by teaching the game. Top professionals also take part in tournaments that are mostly sponsored by daily newspapers or other companies and are endowed with prize money of up to 300,000 euros. However, the Korean and Taiwanese tournaments are still paid a little lower.

Until the second half of the 20th century, Japan produced the most and strongest professional players. However, the Chinese professional go in the eighties reached at least as high a level, while in Korea since the nineties a new generation of go (baduk) players has been pushing to the top of the world. Today the top players from these three countries are of roughly comparable strength; Korean professional players are currently doing best in international tournaments.

There is no world championship for professional Go players. Instead, there are a number of highly regarded titles. The most important Japanese titles include Kisei ( 棋聖 , "game saint", more precisely: Go grandmaster / Go luminary), Meijin ( 名人 , "master"), Hon'inbō ( 本因坊 , name of an earlier Go school), Ōza ( 王座 , “royal throne”), Jūdan ( 十 段 , “10th Dan”), Tengen ( 天元 , “middle of heaven”) and Gosei ( 碁 聖 , “Go saint”, more precisely: outstanding Go master).

Applicants for professional status usually have to prove their skill at a qualifying tournament. The professional organizations in the countries named hold a tournament, usually once a year. The top-ranked players will then be promoted to professionals. Only a handful of professionals are appointed per year per organization.

Go pros usually start playing in their childhood. Each student has an experienced player as a teacher ( sensei , 先生 ). A candidate for a professional is called insei ( 院 生 ) in Japan . Only every third insider makes it to a professional. There are currently around 470 professionals in Japan.

Shigeno Yuki (General Secretary of the International Go Federation IGF 2006–2014, lived in Italy for a long time) and Guo Juan (lives in Amsterdam ) came from Asia to Europe as the first professional players, teachers and experts for playing Go with children .

Well-known professional players

Go Seigen 1952
Eio Sakata 1961
Lee Sedol 2016
Michael Redmond 2008


  • Go Seigen (1914–2014, chin. 吴清源 ), b. as Wú Qīngyuán, trained as a professional in Japan and is one of the candidates for the title “best Go player of all time”.
  • Rui Naiwei (born 1963, chin. 芮 迺 伟 ) is the first woman with the 9th Dan and at the same time the first woman to win a male professional tournament (2000 in Kuksu, Korea).
  • Gu Li (born 1982, chin. 古力 ) is one of the top 20 in the world rankings. In addition to numerous Chinese titles such as Mingren and Tianyuan , he also won the 10th LG Cup in 2006.
  • Fan Tingyu (born 1996, chin. 范廷 钰 ) youngest 9th Dan in history since 2013.
  • Ke Jie (born 1997, chin. 柯 潔 ) currently (May 2017) at number 1 in the world rankings with 3620 BayesElo .
  • Yu Zhiying (born 1997, chin. 於 之 莹 ) 5th Dan, since 2012 strongest go-player in the world.


  • Hon'inbō Shūsaku (1829–1862, Japanese 本因坊 秀 策 ) was probably the strongest player in the Edo period . With the black stones, he was considered invincible.
  • Kitani Minoru (1909–1975, Japanese 木谷 実 ) set up a new opening theory together with Go Seigen and was the teacher of many professional players.
  • Sakata Eio (1920–2010, Japanese 坂 田 栄 男 ) and Fujisawa Shukō (1925–2009) dominated the golden age of Japanese Go in the 1960s and 1970s.
  • Iyama Yūta (born 1989, Japanese 井 山 裕 太 ): youngest Japanese 9th Dan and currently the strongest player in Japan (as of 2015). Represented in the top 5 of the world rankings. By winning the Jūdan on June 30, 2016, he became the first professional player to hold all seven Japanese Go titles at the same time.

South Korea

  • Chō Chikun (also Cho Chi-hun, born 1956, kor. 조치훈 , Hanj . 趙治 勲 ) trained as a professional in Japan by Kitani Minoru, dominated the Japanese scene in the 1980s and 1990s. He was able to defend all important titles (Kisei, Honinbo, Meijin) against his challengers for several years.
  • Lee Chang-ho (born 1975, kor. 이창호 , Hanj. 李昌鎬 ) was considered the strongest player in the world from 1991 to 2006. His teacher Cho Hun-hyeon (born 1953, kor. 조훈현 , Hanj. 曺 薰 鉉 ) was one of the strongest players in the world in the 1980s and 1990s.
  • Lee Sedol (born 1983, Korean 이세돌 , Hanj. 李世 乭 ) was considered the strongest player in the world from 2007 to 2011.
  • Park Junghwan (born 1993, Korean 박정환 , Hanj. 朴廷桓 ) youngest Korean 9th Dan, who has been one of the strongest players in the world since 2012. Currently (November 2015) in second place in the world rankings.


  • Manfred Wimmer (1944–1995, Austria). In 1978 he was the first western player to receive a Japanese professional diploma.
  • Michael Redmond (born 1963, USA). First western player to reach the 9th professional dan. Active in Japan.
  • Hans Pietsch (1968-2003, Germany). So far the only German professional Go player, trained in Japan. He was murdered in an armed robbery on January 16, 2003 while on a go promotion tour in Guatemala . He was awarded the 6th Dan posthumously.
  • Catalin Taranu (born 1973, Romania), 5th professional dan, trained in Japan
  • Alexandre Dinerchtein (born 1980, Russia), 3rd professional Dan, trained in Korea, multiple European champion.
  • Svetlana Shikshina (born 1980, Russia), 3rd professional dan, trained in Korea, first professional go player in Europe.

Go in German-speaking countries


The German-speaking area is playing a pioneering role in the spread of the game of Go in the West. Until the late 19th century, Go was only known by name in Europe. Not until the German chemist Oskar Korschelt , who worked in Japan from 1875 to 1886, and in 1880 the series of articles The Japanese-Chinese game “Go”. A chess competitor published, interested parties in Europe gained the opportunity to learn the game. Shortly before, in 1877, the famous English sinologist Herbert Giles published a description of the game under the title Weichi or the Chinese Game of War . Apparently, however, Korschelt's work was given a wider public. This work is still available today in its English translation, The Theory and Practice of Go . After his stay in Japan Korschelt moved to Leipzig, since then Go has been played there.

In 1905 a small group of chess players formed in Berlin who practiced Go under the guidance of a Japanese student. Emanuel Lasker , who was reigning world chess champion from 1894 to 1921 , joined this group in 1907 . Another member of this Go circle was Eduard (Edward) Lasker (only indirectly related to the world chess champion), who soon emigrated to the United States and co-founded the American Go Association there . After the First World War, in 1919, the first German Go Club was founded in Berlin.

In 1909, the Austrian physicist Leopold Pfaundler published the first German-language Go publication in Graz. During the First World War , the largest go-group in Europe was established in the Austrian naval base in Pula , in Istria .

From 1920 the German Go-Zeitung was published again by Bruno Rüger from Dresden and quickly developed into an important communication medium for Go players in German-speaking countries. At that time, Felix Dueball , whose skill level corresponded to about a 1st Dan degree, was considered the best player in Germany. The notation of a game against the aforementioned Emanuel Lasker has been preserved from a tournament in Berlin in 1930. Lasker won the game against Dueball. In 1930 Dueball and his wife were invited to Japan by the Japanese multimillionaire Baron Okura for 12 months, where he studied the game of Go intensively and took part in a number of tournaments. A game of Dueballs against the most prominent player in Japan, Honinbō Shūsai, has gone down in Go history . In 1936 Dueball played a long distance game of Go against the former Japanese Minister of Culture, Ichiro Hatoyama , for advertising purposes . The current game was published move by move both in the Völkischer Beobachter and in the Japanese newspaper Nichi-Nichi . Hatoyama, who was classified with the 2nd Dan, won the game. Felix Dueball is also mentioned by name in the novel Meijin , a key novel from the Go scene at the time, by the Japanese Nobel Prize winner Yasunari Kawabata .

In 1978 the Austrian Manfred Wimmer was the first non-Asian to receive a Japanese professional rank, just a few months later the American James Kervin received the same honor. In addition to the USA, the leading non-Asian players today come mainly from Eastern Europe, especially from Russia and Romania, where the game has spread particularly strongly since the political opening (1989).


In Germany, Austria and Switzerland there are now around 2500 Go players organized in clubs and associations, the largest by far is the German Go-Bund (DGoB). In recent years the number of seminars, training courses and simultaneous games with strong amateurs (7th Dan) and professional players from China, Japan and Korea, the go centers of the world, has increased significantly. Yoon Young-Sun , originally from South Korea, is the first professional player to move to Germany. She teaches Go in Hamburg. In the last few years professional players from Japan have found temporary domicile in Vienna. The reigning German champion is Benjamin Teuber (6th Dan amateur).

The work of the growing number of Schul-Go-AGs is strongly promoted by the German Schul-Go championship ( Hans Pietsch Memorial) that has been taking place since 2003 . This has made it possible to move the typical Go entry from university to school. In Germany, a number of Go publishers have been set up since 2002, such as Hebsacker Verlag and Brett und Stein Verlag , through which Go material can also be ordered on the Internet.

In every major European city there are Go-Meetings and game evenings. In metropolises like Hamburg, Berlin or Vienna you can play Go every evening in a game club. Regular tournaments take place in many cities. A pan-European rating list (European Go Database) is maintained for over 5000 active European tournament players.

The amateur world championships are held annually in Japan. 2008, were first World Mind Sports Games (World Mind Sports Games) in the Olympic city of Beijing held. Many countries send their representatives to these international events.

KPIs and game theory

In game theory , Go is assigned to the finite zero-sum games with perfect information . Theoretically, depending on the Komi , you could determine whether black or white wins or the game has to end in a draw if both sides play perfectly. According to the current state of knowledge, however, it seems impossible that this question can be answered by fully calculating the search tree , since the complexity of the game even exceeds other unsolved games such as chess by far.

3 N Share of
1 × 1 1 3 33.3% 1
2 × 2 4th 81 70.4% 57
3 × 3 9 19,683 64.4% 12,675
4 × 4 16 43,046,721
= 4.30 × 10 007
56.5% 24,318,165
= 2.43 × 10 007
5 × 5 25th 8.47 × 10 011 49 , 0 % 4.1 0× 10 011
9 × 9 81 4.4 0x 10 038 23.4% 1.04 x 10 038
13 × 13 169 4.3 0× 10 080 08.7% 3.72 x 10 079
19 × 19 361 1.74 x 10 172 01.2% 2.08 x 10 170

The number of valid 19 × 19 positions was determined in 2016 by Tromp et al. calculated exactly (about 2 × 10 170 ), this number has (in the usual decimal system) 171 digits, shown here in nine lines of 19 digits each:

6620397247064840935773 6620397247064840935773 6620397247064840935773

Computer go

5 × 5 solved

In 2002, a computer program called MIGOS (MIni GO Solver), written by Erik van der Werf from the “Computer Games Group” at Maastricht University , calculated all the game possibilities for Go on a 5 × 5 board and solved the game completely: With optimal game the starting player, Black, wins the entire playing field, regardless of what White does.

19 × 19

Developing go-playing computer programs turned out to be considerably more difficult than in the case of the game of chess: until 2015, there was none that could compete with a strong amateur on the 19 × 19 board. It was relatively early to start writing such programs (for example Gobang for the Commodore VC20 1982, GO for the Commodore 64 1983 or the Atari around 1987). It was not until August 2008 that a special program on the Huygens supercomputer won against a go professional, but only with a handicap of nine stones. In October 2015, the AlphaGo program developed by Google DeepMind won against multiple European champion Fan Hui (2P) and in March 2016 four out of five times against Lee Sedol , who is considered one of the world's best players.

Strong Go programs are / were: The Many Faces of Go , MoGo , MyGoFriend , Leela , Crazy Stone and Zen . With Hikarunix there was also a live CD that contained various free Go programs and clients. Beginners are usually advised to play their first games against human opponents. Otherwise there is a risk that the beginner will over-adapt to the specific weaknesses of an individual computer opponent.

DeepMind now has four versions of AlphaGo. They are all based on a combination of neural networks and tree search technology. While the neural networks of the first three versions were trained with millions of positions from games between strong human players, the now revealed version AlphaGo Zero learned the game from scratch within 36 hours, just because of the rules of the game and playing against yourself In internal tests, Alpha Go Zero outperformed the “Master” version of AlphaGo by far. Alpha Go Zero surpassed the human Go knowledge of 1000 years in just 36 hours. During the learning process, many as yet unknown game tactics were discovered. She no longer has to compete against people.

Goprogramming uses different techniques than most other two-player games without chance and with complete information. In chess, a medium level of skill can be achieved by combining an error-free implementation of the chess rules, the alpha-beta algorithm with a quiet search, and a relatively simple evaluation function. In Go this seems to fail at first glance because of the greater number of variants (the unimaginably high number of different positions that are possible on a 19 × 19 board is around 2.08 × 10 170 , in chess "only" around 10 43 ). In comparison: The number of all atoms in the entire universe is approximately 10 80 . That means that if the universe had re-created as often as it has atoms today, there would still be more Go positions than atoms in all these universes combined. The real reason, however, goes deeper than that: it is more difficult than in chess to write a good, fast scoring tool for use with an alpha-beta like search.

There are programs, such as GoTools , that limit themselves to solving idealized partial positions. For certain types of positions, this program can far outperform human analysis. For the goal of the game-strong Go program, however, almost nothing is gained, since these idealized and self-contained positions play a relatively small role in practice. The same applies to the results that can be achieved for some late endgame positions with the help of combinatorial game theory .

Monte-Carlo Tree Search

The evaluation of the best Go programs on the KGS server from 2007 onwards. Since 2006 the best programs have been using Monte Carlo tree search methods.

A different approach is therefore used in the Go, known as the Monte-Carlo Tree Search . In these programs, the move selection is based on the statistical evaluation of the results of a large number of games that have been completely played, starting from the root position. Since with such a procedure the evaluation of the end positions of the random games can be derived directly from the Goregeln, these programs only need Gowissen for the search. Considerable progress has been made on the 9 × 9 board since the end of 2006 through the use of Monte Carlo methods for search and job evaluation. The performance of the best 9 × 9 programs in mid-2007 was probably equivalent to the skill level of a European 3-Dan with average experience with the peculiarities of the small board.

Feng-hsiung Hsu , who became known as the programmer of Deep Blue , thought it possible in 2007 to come up with a Go program by 2017 that would defeat the best human gamers. By then, he believes that hardware will be available that can calculate more than 100 trillion positions per second.

In August 2008, the supercomputer Huygens succeeded for the first time in an official competition against a person with a handicap of nine stones at the 24th Annual Go Game Congress in Portland , Oregon . The Korean 8-dan professional Kim Myungwan was defeated after 255 moves with 1.5 points. The joint press release by the University of Maastricht , NCF, NWO Physical Sciences and SARA says that this is the first win of a computer against a go pro. The MoGo Titan application developed together with INRIA France runs on Huygens , which is located at SARA in Amsterdam.

It is difficult to assign ranks to computer programs because, on the one hand, the skill level of modern Goprograms is heavily dependent on the performance of the underlying hardware and the time used to think about it, and on the other hand, human players usually quickly find typical errors in the programs and exploit them. Often you only evaluate the first game of a person against a computer program for classification. If you score even more games, the perceived playing strength of these programs drops considerably in the opinion of many go-players. The strongest programs have achieved stable ranks in the region of 6 Dan on the KGS Go server (Zen, Crazystone - as of 2015). A more recent approach to rating is the Bayesian Elo rating.

Development since 2015

In March 2015, the Crazy Stone program won the 8th UEC Cup of the Japanese University of Electro-Communications in Chōfu , in which various Go programs competed against each other.

In October 2015, the AlphaGo computer program from Google DeepMind defeated European champion Fan Hui (skill level October 2015: 2nd professional dan, BayesElo : 2908) in five consecutive games. It was the first equal victory of a computer over a person of professional strength.

The program first plays through the possible development of the game in a network. With the help of millions of archived games, as well as the analysis of games it played against itself, the program achieved a prediction model for the next human move in this match that, with a correct prediction of 57%, is well above the previous result of 44 , 4%. Based on this model, the program builds a decision on the best move in a second network by predicting the winner based on each position. Several systems were tested as hardware. The AlphaGo configuration had 48 CPUs with 8 GPUs and achieved with max. 5 seconds to think about a BayesElo playing strength of 2890. The distributed AlphaGo configuration achieved 3140 BayesElo with 1202 CPUs and 176 GPUs. With 1920 CPUs and 280 GPUs, 3168 BayesElo were achieved. In order to provide the massive computing power required during the learning phase, the Google Cloud Platform and TensorFlow Processing Units (TPUs for short, ASICs for the TensorFlow software collection ) were used.

In March 2016 there was a match against the multiple title winner Lee Sedol (level January 2016: 9th professional dan, BayesElo: 3515). Here AlphaGo won four of a total of five games by giving up, Lee only won the fourth game.

Variants and varieties

Different but common board sizes are 13 × 13 and 9 × 9. In addition, there are variants that involve changes or additions to the strategy or the rules of the game.

When going on a circular game board with circular segments as lines (round go) there are no corners and therefore no corner jōseki with the same rules of the game. Considerations about going on a cylinder jacket lead to the same effect. When going on a torus , the edges are also omitted. Each point is therefore equal at the beginning of the game.

The normal game material is used for the Keima-Go . However, every player puts two stones in his train Rösselsprung spacing.

In addition to the normal game material, Poker-Go uses a shared or two player's own deck of cards with the same content, from which the players alternately draw cards with instructions to be carried out. These can consist of setting certain stone formations, moving or removing one's own or opposing stones.

Atari-Go is played by beginners as a preliminary stage to the actual Go game. The rules stay the same. However, the winner is whoever catches a stone first.

See also: Govariants , Gobang , Ninuki Renju and Five in a Row


Non-fiction books (selection to get started)

  • Gunnar Dickfeld: Go for beginners . Brett and Stein Verlag, Frankfurt 2016, ISBN 978-3-940563-40-8
  • Gunnar Dickfeld: Life and Death. Textbooks of the Go . Brett and Stein Verlag, Frankfurt 2013, ISBN 978-3-940563-42-2
  • Gunnar Dickfeld: Black on the move. The Go exercise book . Brett and Stein Verlag, Frankfurt 2013, ISBN 978-3-940563-31-6
  • Jörg Digulla u. a .: The Go game. An introduction to the Asian board game . Hebsacker Verlag, 3rd, corr. Ed., Hamburg 2008, ISBN 978-3-937499-04-8
  • Michael Koulen: Go. The middle of the sky. History, rules of the game, master games . Hebsacker Verlag, 5th edition, Hamburg 2006, ISBN 978-3-937499-02-4
  • William S. Cobb: The Empty Board. Considerations on the Go Game . Brett and Stein Verlag, Frankfurt 2012, ISBN 978-3-940563-01-9
  • Richard Bozulich: Tactics and strategies of the game of Go. What you need to know after learning the rules . Hebsacker Verlag, Hamburg 2009. ISBN 978-3-937499-05-5
  • Isamu Haruyama: Basic techniques of Go . Ishi Press, Tokyo 1984.
  • Thomas Hillebrand: Textbooks of Go. Elementary techniques . Brett and Stein Verlag, Frankfurt 2013, ISBN 978-3-940563-41-5
  • Toshirō Kageyama : Lessons in the basics of Go . Brett and Stein Verlag, Frankfurt 2009, ISBN 978-3-940563-05-7
  • Karl-Friedrich Lenz: Elementary basics of the game of Go . Tokyo 2004. (pdf; 857 kB)
  • Siegmar Steffens: Go. Learning to play the oldest board game in the world . Rittel Verlag, Hamburg 2005, ISBN 3-936443-03-3

Go story, legends and backgrounds

  • Richard Bozulich (Ed.): The Go Player's Almanac 2001 . Kiseido Publishing, Tokyo 2001, ISBN 4-906574-40-8
  • Günter Cießow (Ed.): The board game from Japan . Exhibition catalog. Ethnological Museum, Berlin 2000.
  • Noriyuki Nakayama : The treasure chest. Nakayamas Go-Stories and Riddles , Brett and Stein Verlag, 2008, ISBN 978-3-940563-02-6
  • John Power (Ed.): Invincible. The Games of Shusaku . Kiseido Publishing, Tokyo 1982, ISBN 4-87187-101-0
  • Franco Pratesi: Eurogo (3 vols.), Shaak en Go Winkel, Amsterdam 2004–2006:
  • Ti-lun Luo: Weigi. From the sound of the black and white stones; History and Philosophy of the Chinese Board Game . Lang, Frankfurt / M. 2002, ISBN 3-631-36504-7
  • Zhang Ni: The classic of Go in thirteen chapters . Translated by Dr. Martin Boedicker. CreateSpace Independent Publishing Platform 2014, ISBN 978-1-5027-0459-7


Manga (pop culture)

Web links

Commons : Go (game)  - album with pictures, videos and audio files
Wikiquote: Go  - Quotes
Wikibooks: Go  - learning and teaching materials

Rules of the game for beginners

Go for children and teenagers

  • Euro Go Kids - Information and links for children, young people, parents and teachers on the subject of Go

Go server

  • KGS Go Server (KGS) - well-attended Go server with the possibility to play even without a program download
  • IGS Pandanet - the very first Go server on the Internet (English); Playable via the GoPanda2 client for Windows, Linux & OSX with German language option and the PANDANET (Go) app for Android & iOS.
  • Dragon Go Server - turn-based Go server
  • Online Go Server (OGS) - browser-based go server with German language option, offers many features such as tsumegos (exercises), different board sizes etc.
  • Tygem Go Server - Korean Go Server with mainly Asian audience, playable via the Tygem Client (English, available as Windows or iOS app)

Go lessons on YouTube

  • Nick Sibicky on YouTube - Go lessons for beginners (Live Go, reviews, filmed lessons from the Seattle Go Center) by Nick Sibicky, American 4D player (English)
  • Haylee's World of Go / Baduk on YouTube - Go lessons for advanced learners (mostly live Go) by Lee Hajin, Korean 4P player (English)
  • Dwyrin - Go lessons (English) for intermediate beginners and better players. The channel has several hundred videos on a wide variety of Go topics (reviews, tactics, exercises, etc.)
  • Sunday Go Lessons - Channel by Jonathan Hop (4D player from the USA). There are numerous instructional videos aimed at newbies, as well as reviews of professional games and original Go competitions from Japan with English subtitles.

Tournaments, games, problems and other things

National Go Associations

International Go Associations

Go story

  • Go History - Collection of essays on the asiat. Go story on gobase.org (eng.)
  • Pok's Go Space - Essays on the history of Go in Austria, Europe and Asia

See also

Individual evidence

  1. Mind Sports Online: World Go population is 27 million ( July 18, 2008 memento in the Internet Archive ) , accessed May 16, 2019.
  2. Super User: About the IGF. In: intergofed.org. June 14, 2010, archived from the original on April 15, 2015 ; accessed on May 16, 2019 (English).
  3. Frequently Asked Questions about Go - British Go Association. In: britgo.org. April 12, 2013, accessed December 25, 2015 .
  4. ^ Term "Weiqi (圍棋 / 围棋)" - Chinese: zdic.net - Retrieved on May 27, 2017 - zdic.net - Online
  5. Term "Yiqi (弈棋)" - Chinese: zdic.net - Retrieved on May 27, 2017 - zdic.net - Online
  6. Term "Yi (弈)" - Chinese: zdic.net - Retrieved on May 27, 2017 - zdic.net - Online
  7. Term "Qi (棋)" - Chinese: zdic.net - Accessed on May 27, 2017 - zdic.net - Online
  8. ^ Term "Go (碁)" - English / Japanese: tangorin.com - Retrieved on May 27, 2017 - tangorin.com - Online
  9. Term "Qin Qi Shu Hua (琴棋 書畫 / 琴棋 书画)" - Chinese: zdic.net - Retrieved on May 27, 2017 - zdic.net - Online
  10. Tobias Berben: German Go Association eV In: www.dgob.de. Retrieved September 27, 2018 .
  11. For a more precise definition of the term, see J. Fairbairn in Tesuji , Suji , technique on Senseis
  12. ^ Matthias Reimann: A contribution to the history of Go in the GDR. February 11, 2000, archived from the original on October 20, 2016 ; accessed on June 21, 2017 .
  13. a b c d e f g Go Ratings. In: goratings.org. Retrieved November 3, 2015 .
  14. Fan Tingyu at Sensei's Library. In: senseis.xmp.net. September 28, 2014, accessed October 25, 2015 .
  15. ^ Professional Records at Sensei's Library. In: senseis.xmp.net. Retrieved October 25, 2015 .
  16. Ke Jie at Sensei's Library. In: senseis.xmp.net. July 24, 2015, accessed November 3, 2015 .
  17. ^ History of Go Ratings Ladies. In: goratings.org. Retrieved November 3, 2015 .
  18. Yu Zhiying at Sensei's Library. In: senseis.xmp.net. September 19, 2015, accessed November 3, 2015 .
  19. Iyama Yuta at Sensei's Library. In: senseis.xmp.net. Retrieved October 25, 2015 .
  20. Cho Chikun at Sensei's Library. In: senseis.xmp.net. June 30, 2015, accessed October 25, 2015 .
  21. Yi Ch'ang-ho at Sensei's Library. In: senseis.xmp.net. Retrieved October 25, 2015 .
  22. Yi Setol at Sensei's Library. In: senseis.xmp.net. April 27, 2010, accessed October 25, 2015 .
  23. Jung Hwan Park at Sensei's Library. In: senseis.xmp.net. December 21, 2010, accessed October 25, 2015 .
  24. ^ Obituary for Manfred Wimmer
  25. Michael H. Koulen: Go. The middle of the sky. Hebsacker Verlag, 5th edition, 2006, p. 61
  26. ^ Tromp 2005
  27. sequence A094777 in OEIS
  28. a b Counting Legal Positions in Go. In: tromp.github.io. January 20, 2016, accessed February 2, 2016 .
  29. 5x5 Go is solved . Retrieved January 28, 2016.
  30. a b c Harald Bögeholz: Google AI beats human professional players in the Go - heise online. In: heise.de. January 28, 2016. Retrieved January 28, 2016 .
  31. a b c David Silver, Aja Huang u. a .: Mastering the game of Go with deep neural networks and tree search ( Memento from January 28, 2016 in the Internet Archive ) In: Nature. 529, 2016, p. 484, doi: 10.1038 / nature16961 , PDF file; 1.5MB.
  32. Supercomputer beats go pro. www.20minuten.ch, accessed on September 27, 2018 .
  33. a b egc2015.cz (PDF file; 72kB), accessed on May 16, 2019
  34. How Many Atoms Are There in the Universe? . thoughtco.com.
  35. Elwyn Berlekamp , Introductory overview of Mathematical Go endgames , in: Richard K. Guy , Combinatorial games , Proceedings of Symposia in Applied Mathematics, Volume 43, 1991, ISBN 0-8218-0166-X , pp. 73-100. Elwyn Berlekamp, ​​David Wolfe: Mathematical Go , 1994, ISBN 1-56881-032-6 . An overview is given in: Jörg Bewersdorff : Luck, Logic and Bluff: Mathematics in Play - Methods, Results and Limits , 2010, ISBN 3-8348-0775-3 , pp. 147-168 as well as Go and Mathematics (PDF; 133 kB )
  36. Sensei's Library: KGSBotRatings. Retrieved May 3, 2012 .
  37. IEEE Spectrum - October 2007: Feng-hsiung Hsu: Cracking Go ( Memento from January 15, 2008 in the Internet Archive )
  38. Deutsche Go-Zeitung, issue 4/2008, p. 45 f.
  39. Dutch national supercomputer Huygens wins against Go pro players. Retrieved September 27, 2018 .
  40. Huygens supercomputer at SARA ( Memento from June 10, 2012 in the Internet Archive )
  41. Supercomputer Huygens beats Go professional: no one is safe. Retrieved October 1, 2018 .
  42. KGS Go Server Rank Graph for CrazyStone
  43. KGS-Go-Server-Rankgraph for Zen19S
  44. ^ Rémi Coulom: Bayesian Elo rating. In: remi-coulom.fr. February 6, 2008, accessed January 28, 2016 .
  45. ^ UEC Cup at Sensei's Library. In: senseis.xmp.net. Retrieved September 27, 2018 (English).
  46. ^ The Eighth UEC Cup Computer Go
  47. ^ Rémi Coulom: Crazy Stone. In: remi.coulom.free.fr. Retrieved September 27, 2018 (English).
  48. heise online: Google AI beats human professional players in the Go. In: heise online . Retrieved January 28, 2016 .
  49. Christof Windeck: Google I / O 2016: "Tensor processors" helped win the Go - heise online. In: heise.de. May 19, 2016. Retrieved November 23, 2016 .
  50. Martin Holland: Google-AI AlphaGo wins in the first game against Lee Sedol. In: heise.de. March 9, 2016, accessed March 9, 2016 .
  51. Match 2 - Google DeepMind Challenge Match: Lee Sedol vs AlphaGo on YouTube
  52. bähr / dpa: Go-Genie loses against the computer. In: FAZ.net . March 10, 2016, accessed March 10, 2016 .
  53. Harald Bögeholz: Man versus machine: AlphaGo loses a game. In: heise online . March 13, 2016, accessed March 13, 2016 .
  54. ^ OGS at Sensei's Library. In: senseis.xmp.net. October 19, 2013, accessed November 8, 2015 .
  55. ^ Tygem Baduk at Sensei's Library. In: senseis.xmp.net. December 27, 2011, accessed November 8, 2015 .
  56. Nicksib at Sensei's Library. In: senseis.xmp.net. Retrieved November 8, 2015 .
  57. Lee Hajin at Sensei's Library. In: senseis.xmp.net. Retrieved November 8, 2015 .
This article was added to the list of excellent articles on August 5, 2004 in this version .