Alter Igo

    Keywords: Rules, Variant

Alter Igo is a game that is almost, but not quite, entirely unlike go. It is basically Bass's exercise in reducing the rule set of go to a minimum, while maintaining the concepts of surrounding territory, making strategic choices and making living groups. Still, it seems to make a playable game.

Complete rules of Alter Igo

Rule 1: Prerequisites.

 The game is played on a grid of lines. There are
 two players, one having black stones and the other having
 white stones.

Rule 2: Game play.

 The players alternatingly place their respective
 stones on empty intersections.

Rule 3: Strings.

 If a stone is adjacent to another stone of the same color,
 these stones are said to belong to the same string. A stone
 that is not adjacent to any same colored stone is a string by

Rule 4: Capture.

 Immediately after each stone is placed, every string that
 is not adjacent to at least one empty intersection, is
 removed from the board.

Rule 5: End of the game.

 The game ends in
 * a draw, if the same board position is repeated three times
 * opponent's win, if a player resigns
 * opponent's win, if a player places a stone so that it
   is the only stone to be immediately removed by rule 4.

Rule 6: There is no rule six.

I have played a couple of games on a 9x9 board using these rules, and have not yet run into any unexpected trouble, wiz. all the trouble that was ran into, was very much expected. Some analysis on the most obvious changes is available [ext] in Finnish.

If you ever try these rules, please be sure to report your experiences below.

By the way, the original name was "Epägo", which is Finnish for "non-go". The translation "Alter Igo" was suggested by Mikael Runonen, whom I no longer owe a beer for this favor :-)

Have fun,


DougRidgway In the capturing rule, capturing strings might also get removed. This is what is wanted, not the usual go capturing rule?

Eratos I don't understand the ending condition - Opponent wins if a player plays a stone that does not alter the board position - example please?

Pajaro About not altering the board, I suspect that if you play a stone inside an eye, then it's a suicide, it is removed, and the board doesn't change. Game over. I've also noticed that there is no pass rule, or counting of territories. I guess that when the game would be over (assuming real go), the players could start playin inside the territories, until one side is forced to commit suicide. The player with less territory might be the one forced to lose, but I'm not sure, according to the wat stones are removed.

Hi all, and thanks for your comments.

The capturing stones may also get removed, since preventing that would require an exception to the rule. I actually thought about adding the exception, but then the game would resemble actual go so much, that someone might mistakenly think I had just tried to define a new ruleset for the game of go ;-)

About the game end, if you actually play a game, you will notice that eventually one of the players cannot play anything but a single stone suicide. This will happen after all that player's stones have been removed, and all the other player's eyes have been reduced to one space. I wanted the rules to be complete without defining a "single stone suicide that does not remove any opponent's stones", that is the cause for the wording. I think I must reword that rule somehow, so that it is easier to understand.

The absence of passing and counting is very, very intentional. Since counting requires some kind of determination of living and dead stones, a complete ruleset and discussion of its flaws would fill many books. (see for example :-)

In Alter Igo the winner will be the player with most "spare eyes", that is, eyes that are not needed for living. Often the player with the most territory will also be the one with most spare eyes, but of course this is not always the case.


I have tried to put together some endgame positions for analysis on how exactly these rules work. There's even an endgame problem! --Bass, 2008-01-10

MrMormon: I'm surprised hardly anyone's shown interest in this variant, specifically the rule of simultaneous capture. Basic ko would work differently. Games would be longer. Superko would have a bigger role. Perhaps Go can be a deeper game?

willemien It is the first time i see this, so a first impression

It is no pass go with suicide (of more than one stone) allowed. because passing is not allowed (or is it just an ommission in the rules)

For the rst the rules look like the new zealand rules

  1. suicide of more than one stone is allowed.
  2. The superko rule here (5.first star) seems to be positional superko but can also be seen as situational superko

Bass: passing is not included because I thought it to be a non-essential part of go. Also, the capture rule is not exactly "suicide allowed", since it is possible for both coloured stones to get captured with the same move. willemien (to Bass) I moved your comment to here In your ruleset passing is not allowed (rule 2 says a move is placing a stone... this definitlyexcludes not placing as stone, and given rule 5.3 a one stone suicide is no pass but a loss.

Your second comment is interesting i put my remarks on Simultaneous Capture

ThorAvaTahr: This ruleset exhibits grouptax

MrMormon: Adding to that, there are disadvantages to traditional stone scoring: wasting time making useless moves, no jigo (debatable), and inability to compensate White.

Bass probably the variant is far enough removed from regular go, so that players are not instantly interested. Go players tend to think that anything that is different from their traditional rules is automatically worse. willemien please not this kind of comments, stone scoring is a much longer game than area or territory scoring. try and see. maybe it is better to have this discussion in the rules forum in life in 19 x 19 and here only to post the agreed conclussions (and it is more clear what came after what and so :) willemien Under these rules it is even worse than stone scoring , if a player doesn't want to resign you need to play on till every move of him becomes a one stone suicide. as long as he can play suicide with more stones the game doesn't end, Bass, you really need some scoring rule.

ThorAvaTahr: Regarding compensation for white: you could make a rule 7 that allows white 6 passes. (I am not sure why rule six should be skipped :P)

MrMormon: Do you mean free handicap placement? With only one stone, yet alone six, the game would be reversed and Black would need compensation, so that doesn't solve it.

ThorAvaTahr: No. Give white the right to pass 6 times, which white can use at any time he wants. (So using it in the beginning is whites worst use of this right)


@MrMormon, probably the variant is far enough removed from regular go, so that players are not instantly interested. Go players tend to think that anything that is different from their traditional rules is automatically worse.

@ThorAvaTahr, you are correct on both accounts, although there is no need for a komi system really, if chess players are to be believed at all. Which, it is possible to argue, they sometimes are. Also: [ext] :-)

MrMormon: In chess, the first player wins 52% of the time. Compare that with 53% in go with 5.5 komi. Imagine the margin without komi.

ThorAvaTahr: Even though this game is not Go, I think it is interesting, and perhaps worth the effort to balance, however, at the expense of additional rules :P.

Bass: Hmm, maybe. But adding a rule that does not really change the nature of the game very much would go against the purpose of this ruleset. Which is, of course, to see if the golden rule of game design ("the simpler the rules, the better the game") holds even in the notably simple-ruled game of go.

MrMormon: The nature is changed without compensation: everyone would want to be Black.

Bass: Well, then you get to play two games instead of one. This makes the system totally fair, as opposed to the komi system, which, as you pointed out earlier, is not an exact science. As the nature is changed to the better, I do not think it is totally fair to demand extra compensation too :-)

MrMormon: Yes, komi is occasionally changing to reflect statistics (its higher than 5.5 now, regardless of the ruleset). No, the nature is not better because White might not want a rematch (time and/or wanting other opponents). Less importantly, will Black always be noble enough to accept a losing game?

Bass: Well, I'll come half way and agree that a game that tries to be balanced is different from a game that does not. Whether it is better to be openly unbalanced, or to try to guess numerical values that might balance the game, that seems to be mostly a matter of taste.

MrMormon: Fair.

willemien: i disagree, having a game that draws if both players play optimal keeps them both scharp a mistake and you lose. having unbalanced games, you win one anyway and the other a you will lose it anyway

MrMormon: Not if the scores from both games are summed.

willemien you don't have score you will win, or lose depending if you are black or white, not bacause you are stronger or weaker.

MrMormon: brain failed. Forget my last two comments; komi is not a matter of taste.

Bass: Never mind, happens all the time. :-) As we do not know the correct value for komi, we won't have a fair game anyways, the only thing we know for certain is that the correct komi does not include a fractional part. Also, very annoyingly, if we _did_ know the correct value for komi, then that would mean that the game has been solved, thus diminishing the point in continuing to study it.

MrMormon: But do you want to make it more unfair?

Bass: To achieve perfect fairness, two games need to be played. This is true even when komi is used, since we can be sure that the komi we are using is unfair: if both sides play perfectly, the result will not be a draw because of the .5 part of the komi. Now it could happen that somebody actually thought that 5.5 is a fair komi. In this case it would be unlikely that he would take the unfairness into account when deciding on tournament systems and the like. If this should happen (as it has), then the tournaments would be unfair, though the unfairness would probably be hidden by "noise" from random factors. If the game is openly and blatantly unfair like, say, chess, then the unfairness is always taken into account in the tournament system, and the result may even be more fair than in the former case when the unfairness per game is smaller. As fairness is a matter of perception, you can probably get away with a small amount of unfairness anyway, so you can factor other preferences like "it's boring to play the same opponent for hours and hours" into the equation, so it is only in the extreme case of seeking perfect fairness where any of this becomes relevant.

This seems to be getting largely offtopic, maybe we should move it to "komi/discussion" if we want to continue?

MrMormon: So, you prefer to make Alter Igo fair with two games. Again, that requires a cumulative score. Also, I agree that fractional komi is unfair, but don't bash komi; bash the people who set it (not New Zealand!).

Bass: To win, you have to get at least a win and a draw out of the two games, and you're good. Or are you suggesting there should be various degrees of win? That might be difficult to implement, since there is no scoring system in Alter Igo.

MrMormon: Idea: the winner is the one who won as black in less moves (given two games)?

Alter Igo last edited by on May 12, 2014 - 14:24
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library