Reversi differs from Othello in two respects. In Reversi, the board starts empty, and in each of the first two turns each player plays a stone of his own colour in one of the centre four squares. In Othello, this is done for you, eliminating one of the starting configurations of Reversi.
If you are unable to play at Reversi, you miss your turn. However, once someone has used up all 32 of his discs, the person with discs left gets to play all of the last moves.
The modern rules of Othello were invented by Goro Hasegawa in 1971.
Thank you: Warren Smith and Michael Keller.
The following are rule variances that I know about. If anyone has additions to make to this list, please let me know and I'll add them.
2.1: Touch Move
The equivalent of "touch move" in chess normally applies. If a disc touches a square, and the move is legal, it must be played there.
2.2: Rules Regarding Time
Chess clocks are normally used to time tournament games, with the standard time limit being 25 minutes for all moves in USOA tournaments. In case you haven't seen a chess clock, there is a little piece of red plastic, called the flag, that is pushed up as the minute hand on the clock approaches 12. Once the minute hand reaches 12, the flag falls, indicating that a time default has occurred. Thus, in a standard 25 minute game, the clock is set to 5:35 (meaning that the flag falls when the clock reaches 6:00)
In the USA when a player's opponent must pass, it is customary for such a player not to hit his opponent's clock. He says something like "you don't have a move, do you?" and on an affirmative nod continues play. Of course there can be honest mistakes, and a player could be very short of time, and then he might nevertheless hit the clock. On the other hand a player can legally hit the clock after his move, and wait however long it takes for his opponent to realize that he has no move, and hits the clock. This may even be the custom in certain places, but does not appear to be the practice in the USA.
2.3: The USOA Time Default Rule
Let us take a game between two players (Al and Bob) to illustrate the USOA time default rule. Al's flag falls, and Bob informs Al that he has defaulted on time. In non-USOA events, this would be scored as a 64-0 win for Bob.
In USOA tournaments, the time default rule is more complicated. We go back to the original example: Al's flag falls, and Bob informs Al that he has defaulted on time, and calls the tournament director. Then the tournament director comes, and stops both clocks while he explains the procedure to both players (if necessary).
Al has lost the game, but the game is not immediately scored as a 64-0 win for Bob. The game will be continued, with Bob being awarded the board score if he wins. If he does not win on the board, then he is awarded the atificial winning score of 32-31. Al is given an additional two minutes on his clock to finish the game.
What happens if Al or Bob default after Al has originally defaulted? If Al's flag falls a second time before the end of the game, Bob wins 64-0. If Bob's flag falls, he is given an additional two minutes to finish the game. If Bob's flag falls a second time, Bob wins 32-31.
This guarantees that, under any circumstance, Bob will win if Al's flag was the first to fall.
2.4: Disc Flipping Etiquette and Errors
In a tournament game, you are supposed to use only one hand to flip the discs. The same hand that flips the discs must press the button on the clock to indicate that the move is completed.
You are compelled to fix any disc flipping errors on your own time. If the opponent notices that you have erred in flipping discs, he presses his button (restarting your clock), and should inform you of your error.
If your opponent starts to make his move (by placing his disc on the board) before noticing the error, the disc-flipping error stands.
2.5: Final Score Of A Game
The final score of the game is important for determining tie breaks in tournaments. In European tournaments, when there are empty squares on the board, they are given to the winner of the game. In US tournaments, a score of 64-0 is awarded when the winner takes all of the opponent's discs, but is not awarded the empty squares if the losing side has any discs left. In Japan, empty squares are split evenly between the two players.
2.6: A Japanese System For Removing Colour Bias
Although the system described here is not used in all Japanese tournaments, it is used in the Student Championship and the Student Meijin.
Before each game, one of the players takes a disc, puts it on the board, and covers it with his hand. The other player says either "black" or "white". If the second player correctly guesses the colour of the disc, he will choose the game handicap; otherwise, the first player will choose the handicap.
The player choosing the handicap decides what odds he's willing to give with each colour. This is done by choosing a number which will be added (or subtracted) from black's final score. For example, if I slightly prefer black I might choose a handicap of -0.5. Then when the game is over, rather than looking at the final game score to determine the handicap, 0.5 is subtracted from black's score and this adjusted score determines the winner. In this case, a 32-32 draw over the board would be scored as a win for white, whereas a 33-31 win for black would be scored 32.5-31, a narrow win for black. After one player has chosen the handicap, the other player chooses the colour he prefers, and the game proceeds. Handicaps must always be chosen with a fractional part of 0.5 to ensure no draws will occur.
Thank you: Clarence Hewlett, Tetsuya Nakajima and Colin Springer.
The top computer program in the world is Logistello, a program by Michael Buro. The construction of the program was the subject of his Ph.D. thesis research in Paderborn, Germany.
The current rating lists on the other page will give you an idea of who The other top players in the world are. It is the author's opinion that computers can beat any human in a prolonged match; however, this superiority has yet to be proven over a match of any length.
Year Location Winner Top Team Runner-Up ---- -------- ---------------------- -------- ---------------------- 1977 Tokyo Hiroshi Inoue (JPN) N/A Thomas Heiberg (NOR) 1978 New York Hideshi Maruoka (JPN) N/A Carol Jacobs (USA) 1979 Rome Hiroshi Inoue (JPN) N/A Jonathan Cerf (USA) 1980 London Jonathan Cerf (USA) N/A Takuya Mimura (JPN) 1981 Brussels Hideshi Maruoka (JPN) N/A Brian Rose (USA) 1982 Stockholm Kunihiko Tanida (JPN) N/A David Shaman (USA) 1983 Paris Ken'Ichi Ishii (JPN) N/A Imre Leader (GBR) 1984 Melbourne Paul Ralle (FRA) N/A Ryoichi Taniguchi (JPN) 1985 Athens Masaki Takizawa (JPN) N/A Paolo Ghirardato (ITA) 1986 Tokyo Hideshi Tamenori (JPN) N/A Paul Ralle (FRA) 1987 Milan Ken'Ichi Ishii (JPN) USA Paul Ralle (FRA) 1988 Paris Hideshi Tamenori (JPN) GBR Graham Brightwell (GBR) 1989 Warsaw Hideshi Tamenori (JPN) GBR Graham Brightwell (GBR) 1990 Stockholm Hideshi Tamenori (JPN) FRA Didier Piau (FRA) 1991 New York Shigeru Kaneda (JPN) USA Paul Ralle (FRA) 1992 Barcelona Marc Tastet (FRA) GBR David Shaman (GBR) 1993 London David Shaman (USA) USA Emmanuel Caspard (FRA) 1994 Paris Masaki Takizawa (JPN) FRA Karsten Feldborg (DEN) 1995 Melbourne Hideshi Tamenori (JPN) USA David Shaman (USA) 1996 Tokyo -------------------NOT PLAYED YET-----------------------The team championship started in 1987, and is awarded to the team of 3 players that gets the highest number of points in the 13-round Swiss.
Thank You: Emmanuel Lazard
The World Championship is normally run as a thirteen round Swiss System tournament which is played over two days.
On the third day, the top four players from the Swiss are seeded 1 through 4. #1 plays #4 in a best out of three game match, while #2 plays #3. The choice of colour for the third game is given to the player who is seeded higher. The winners of these two semi-final matches meet in the final; another best two out of three affair.
The 1996 World Championships are to be held in the Hotel Nikko in Tokyo, Japan, according to my most recent information. Please see the Upcoming Tournaments list for more information, as it becomes available.
In the English language, the best manual I have seen (to date) is the "Othello: Brief & Basic" manual written by Ted Landau. It is available through the British Othello Federation and the United States Othello Association. A new book, "Othello Strategy and Tactics" , is available from the editor, Hugo Calendar.
In French, there is a very reasonable manual given to first-time subscribers to FFORUM, "A la decouverte d'Othello" by Emmanuel Lazard. Rumour has it that there are a number of French players in the process of writing a large book on the game of Othello. I will be sure to put the information here when I have it.
In Italian, there is a complete book on the game: "Othello" by Augusto Brusca, Alessandro Maccheroni e Luigi Puzzo.
If you do not understand game tree search (minimax, alphabeta or NegaScout), there are a number of sources for information on game tree search, such as Judea Pearl's book "Heuristics" (for senior undergrads) or the series of books by David Levy, starting with "Computers and Chess" (for more basic information).
These are the four papers that I would personally recommend to interested Artificial Intelligence students who have already been introduced to game tree search.
"Artificial Intelligence" is a computer science journal that can be found in most university libraries throughout North America. The book on the First Computer Olympiad can also be found in many university libraries.
I've also found a couple of Java applet Othello programs:
The University of Alberta contains a mirror of the Paderborn site. Please use this site ONLY when you can't get through to Paderborn.
comp.ai.games is of interest for programmers of Othello games. It has already held some discussions on game-tree searching, evaluation function learning, et cetera. However, as the title suggests, the newsgroup does contain information about all games, not just two-player zero-sum games with perfect information.
comp.ai has occasionally contained some information about game-tree searching in the past (before the inception of comp.ai.games).
rec.games.othello has gone to the RFD phase a couple of times, but it is not likely to pass the voting phase, due to the limited traffic Othello has seen on rec.games.abstract, and the size of the Othello community currently on the Internet.
As a final note, Hugo has told me that he plans a more dynamic mailing list system in the future. If you want an othello mailing list for a specific geographic region or topic that isn't already covered, please send Hugo some e-mail.