Go Variant
Redstone is a game that we discuss in the selective restrospective of Mark Steere's games in this issue. The investigation of Redstone below appeared in the print version of AG21, but not the web version. We are addressing this omission now.
Mark Steere is a game designer responsible for dozens of games since his first, Quadrature, in 1992. His games are often very clear conceptually, what he calls having good “architecture.” How to play them well is often much more opaque. Mark adheres to certain principles in his game design—foremost, games should always necessarily have a decisive outcome, no draws, in other words, and no repeating positions.
Go itself does not adhere to Mark’s principles. For example, repeating positions can occur, in ko. Of course, Go has supplementary rules for handling ko; and otherwise, Go can eliminate draws artificially by giving an extra half point of komi. Redstone is a game by Mark Steere that takes another approach to “perfecting” Go by removing the possibility of repeating positions and draws.
The board starts empty. The players move and capture, as in Go, except that whenever a capture is made, a red stone is used for the capturing move, instead of a white or black stone. If there is a move that reduces a group of stones to zero liberties, then that move must be made with a red stone. The red stones belong to neither player, and are completely invulnerable to capture. The red stones are effectively “board edges.” Placing a red stone on a point is equivalent to removing that point from the game entirely.
Unlike in Go, suicide moves are possible, killing one of your own groups by reducing the group to zero liberties. Of course the suicide move must be made with a red stone. The player may simultaneously reduce two or more groups to zero liberties with a single move. In this case, again unlike Go, all groups of whatever colour are captured and removed from the board simultaneously. In this respect, the Redstone rules are perhaps a little cleaner and more logical than those of Go.
There is no passing in Redstone. A player must always move. The objective in Redstone is to capture all enemy stones. Captured stones do not count for points.
Black moves first, and the pie rule is used to even out Black’s advantage. After Black has placed the first stone, White can decide to switch sides and play Black, or stick with White and play the second stone.
Some tactics come over from Go, but some are different in Redstone. There is no passing, so no seki. Of course, ko no longer exists, which was Mark’s main objective with the game. Ladders work in Redstone. At first thought, the standard kind of snapback position from Go does not carry over to Redstone. Below is a snapback-like position from an actual game.
Mark Steere is a game designer responsible for dozens of games since his first, Quadrature, in 1992. His games are often very clear conceptually, what he calls having good “architecture.” How to play them well is often much more opaque. Mark adheres to certain principles in his game design—foremost, games should always necessarily have a decisive outcome, no draws, in other words, and no repeating positions.
Go itself does not adhere to Mark’s principles. For example, repeating positions can occur, in ko. Of course, Go has supplementary rules for handling ko; and otherwise, Go can eliminate draws artificially by giving an extra half point of komi. Redstone is a game by Mark Steere that takes another approach to “perfecting” Go by removing the possibility of repeating positions and draws.
The board starts empty. The players move and capture, as in Go, except that whenever a capture is made, a red stone is used for the capturing move, instead of a white or black stone. If there is a move that reduces a group of stones to zero liberties, then that move must be made with a red stone. The red stones belong to neither player, and are completely invulnerable to capture. The red stones are effectively “board edges.” Placing a red stone on a point is equivalent to removing that point from the game entirely.
Unlike in Go, suicide moves are possible, killing one of your own groups by reducing the group to zero liberties. Of course the suicide move must be made with a red stone. The player may simultaneously reduce two or more groups to zero liberties with a single move. In this case, again unlike Go, all groups of whatever colour are captured and removed from the board simultaneously. In this respect, the Redstone rules are perhaps a little cleaner and more logical than those of Go.
There is no passing in Redstone. A player must always move. The objective in Redstone is to capture all enemy stones. Captured stones do not count for points.
Black moves first, and the pie rule is used to even out Black’s advantage. After Black has placed the first stone, White can decide to switch sides and play Black, or stick with White and play the second stone.
Some tactics come over from Go, but some are different in Redstone. There is no passing, so no seki. Of course, ko no longer exists, which was Mark’s main objective with the game. Ladders work in Redstone. At first thought, the standard kind of snapback position from Go does not carry over to Redstone. Below is a snapback-like position from an actual game.
The White move is the throw-in at E7. Then Black may place a red stone at E6, capturing the white stone at E7 and the two black stones at F7 and F6 simultaneously, resulting in the diagram below. In the actual game, Black responded differently to the White throw-in at E7, and White subsequently captured with E6.
For those used to Go, odd things can happen in Redstone close to the edges of the board. See the position below, later in the same game. White has just played C1. In regular Go, Black would not be able to prevent the stone on C1 connecting to the white stones on the right.
This tactic does not work in Redstone. Black can play D1 to disconnect the white stone. White can try D2, Black responds E2, and then White captures at E1, resulting in the diagram below.
The white stone on C1 is no longer connected to the white stones on the right, because the red stone is a barrier as impassable as the edge of the board itself!
On the other hand, red stones can form a wall against which a player can build a live group. Note that the concept of “live” groups having two eyes carries over from Go—you cannot capture a group by playing two red stones simultaneously! See the example below. The White eye on the right would be a false eye in regular Go, if the red stone were white, but the red stone makes that point like a corner, and the White group is alive.
On the other hand, red stones can form a wall against which a player can build a live group. Note that the concept of “live” groups having two eyes carries over from Go—you cannot capture a group by playing two red stones simultaneously! See the example below. The White eye on the right would be a false eye in regular Go, if the red stone were white, but the red stone makes that point like a corner, and the White group is alive.
Sacrificing stones to help create a wall is a very real possibility in Redstone. The red stones prove to be crucial bulwarks against which to build eyes. Remember, also, the object in Redstone is to eliminate opposing stones, not to score points. Your stones that the opponent captures do not count as points scored against you.
Of course, a stone cannot be placed in enemy territory if it will have zero liberties, even though suicide moves are allowed in Redstone—the capturing stone that reduces a group to zero liberties must be red!
The endgame in Redstone is quite different from that of regular Go. It comes down not to the largest territory, but rather which player has more eyes. When there are no other moves left, a player must fill in his own one-point eyes. Then, down to one one-point eye, he will be captured—by a red stone, of course.
At first, it may seem that following the creation of a sizeable vacant space by means of a large capture, that one or both players could start playing in it, triggering other captures, and so on. However, this process would be as meaningless as playing in secure enemy territory in Go—you need to be able to make two eyes for it to be worthwhile.
Redstone needs much more investigation, but we can already see that Redstone has tactics and strategy not present in Go. And at least the tricky ko rules in Go are needed no longer! ◾️
Acknowledgements
Thank you to Paul van Wamelen for helping to investigate Redstone.
Of course, a stone cannot be placed in enemy territory if it will have zero liberties, even though suicide moves are allowed in Redstone—the capturing stone that reduces a group to zero liberties must be red!
The endgame in Redstone is quite different from that of regular Go. It comes down not to the largest territory, but rather which player has more eyes. When there are no other moves left, a player must fill in his own one-point eyes. Then, down to one one-point eye, he will be captured—by a red stone, of course.
At first, it may seem that following the creation of a sizeable vacant space by means of a large capture, that one or both players could start playing in it, triggering other captures, and so on. However, this process would be as meaningless as playing in secure enemy territory in Go—you need to be able to make two eyes for it to be worthwhile.
Redstone needs much more investigation, but we can already see that Redstone has tactics and strategy not present in Go. And at least the tricky ko rules in Go are needed no longer! ◾️
Acknowledgements
Thank you to Paul van Wamelen for helping to investigate Redstone.