Game Designer
Mark Steere has been designing games for nearly 30 years. He has created a substantial collection of games, several of which have become well known in the world of abstract games. Many of his games are highly original, even counter-intuitive in the way that he reinterprets and recombines ideas. The designer's games embody a pure philosophy of game design. Mark Steere, himself, is an original—he brings to the field a unique perspective that deserves to be understood and appreciated.
I caught up with Mark in Mongolia, where he has been living since 2019, trading time between an apartment in Ulaanbaatar and his cattle ranch in northern Mongolia. Mark returned to game design earlier this year, with a plethora of new games. We struck up an email conversation about his new games Zola and Gopher, and I asked him about 10x10 Mongolian Chess, Hiashatar. Mark writes,
"I played Zola and Gopher with my lawyer. When he won, he was smiling. When he lost, he wasn't. He said he likes Settlers of Catan. I played my driver at Zola and Gopher and he looked pained. He said I should design a good game, using Mongolian tiles. I've never heard of Mongolian Chess. I've only seen International Chess in the stores."
My interest in Hiashatar remains, and I hope eventually our investigations may bear fruit in Abstract Games.
For now, my intention is to discuss the designer's games and his thoughts about game design more generally. The collection of games I have chosen to highlight constitutes a very selective retrospective of Mark Steere games. These games jumped out because they are interesting in their design or interesting because of their strategy and tactics. Of course, other games by Mark Steere may also have desirable features, and the selection is not intended to be a list of his best games. Mark's comments interspersed throughout are lightly edited extracts from our email communication.
I caught up with Mark in Mongolia, where he has been living since 2019, trading time between an apartment in Ulaanbaatar and his cattle ranch in northern Mongolia. Mark returned to game design earlier this year, with a plethora of new games. We struck up an email conversation about his new games Zola and Gopher, and I asked him about 10x10 Mongolian Chess, Hiashatar. Mark writes,
"I played Zola and Gopher with my lawyer. When he won, he was smiling. When he lost, he wasn't. He said he likes Settlers of Catan. I played my driver at Zola and Gopher and he looked pained. He said I should design a good game, using Mongolian tiles. I've never heard of Mongolian Chess. I've only seen International Chess in the stores."
My interest in Hiashatar remains, and I hope eventually our investigations may bear fruit in Abstract Games.
For now, my intention is to discuss the designer's games and his thoughts about game design more generally. The collection of games I have chosen to highlight constitutes a very selective retrospective of Mark Steere games. These games jumped out because they are interesting in their design or interesting because of their strategy and tactics. Of course, other games by Mark Steere may also have desirable features, and the selection is not intended to be a list of his best games. Mark's comments interspersed throughout are lightly edited extracts from our email communication.
The oeuvre
Mark Steere lists more than 50 games on his website, which I have re-ordered below according to their year of design. The brief descriptions are mostly taken from the website, but I have supplemented the descriptions here and there.
Mark Steere lists more than 50 games on his website, which I have re-ordered below according to their year of design. The brief descriptions are mostly taken from the website, but I have supplemented the descriptions here and there.
Mark Steere Games
Quadrature (1992) (Reversi-like squaring mechanism)
Tanbo [orignally Rootbound] (1993) (Go set, game of roots, adjacency rules)
........
Impasse (2003) (checkers set, armies march through each other)
Byte (2005) (checkers set, passive stacks, move distance, merging)
Cephalopod (2006) (territory, dice as pieces)
Copolymer (2006) (multiple cells claimed per turn, adjacency rules)
Diffusion (2006) (standard mancala set)
Nested Y (2006) (connection game, nested boards)
Scribe (2006) (pattern making)
Crossway (2007) (Go set, square connection game)
Dipole (2007) (checkers set, merging, stacking, attacking game, annihilation)
Oust (2007) (board starts empty and ends with only one colour, adjacency rules)
Palisade (2007) (Go set, territory)
Rush (2007) (Go set, multiple stones added each turn, adjacency rules, Go -like)
Anchor (2008) (triangular board with two triangular holes, connection game)
Anchor 3D (2008) (3D connection game)
Atoll (2008) (a game of perimeter islands, connect islands of your colour, connection game)
Begird (2008) (generalized form of Y, connection game)
Blood Diamonds (2008) (online play)
Lariat (2008) (3D connection game, world’s simplest connection game)
Super Lariat (2008) (3D connection game, highly unusual geometry)
Loophole (2008) (rhombus-shaped board with two rhombus-shaped holes, connection game)
Loophole 3D (2008) (3D connection game)
Mobius (2008) (connection game, Moebius strip board)
Variable Trump Tute (2008) (card game)
Basic (2009) (checkers set, mixed stacks, stacking)
Mosaic (2009) (tile game, territory, adjacency rules)
Grand Hex (2009) (Hex with an added complication, connection game)
Hex KB (2009) (connection game, Klein bottle board)
X (2009) (connection game for three)
Cage (2010) (checkers, centre move distance)
Colonnade (2010) (stacking, one dimensional)
Fractal (2010) (beautiful board designed to minimize first move advantage, connection game)
Flume (2010) (territory, dots and boxes)
Jostle (2010) (combinatorial, adjacency rules)
Mad Bishops (2010) (combinatorial, annihilation, adjacency rules)
Mad Rooks (2010) (combinatorial, annihilation, adjacency rules)
Rive (2010) (high churn rate, adjacency group rules like Oust)
Monkey Queen (2011) (stacking, chess variant)
Marvin (2012) (stacking, territory)
Redstone (2012) (Go set, Go-like, invulnerable red stones with captures)
Trivor (2012) (connection game, trivalent, variable orientation cell division)
Tripen (2013) (connection game, trivalent, variable orientation cell division)
Gyre (2015) (surround the centre, connection game)
........
Anaash (2021) (stacking, annihilation, move distance)
Bamboo (2021) (clumps)
Dodo (2021) (simple rules, no moves possible for the goal)
Gopher (2021) (simple rules, adjacency rules)
Inchworm (2021) (stacking, merging)
Kobudai (2021) (checkers set)
Manhattan (2021) (consolidation, Manhattan distance)
Marmot (2021) (territory, Gopher variant)
Pathway (2021) (combinatorial, like Gopher, adjacency rules)
Zola (2021) (distance from centre, simple, robust, strategic)
Quadrature (1992) (Reversi-like squaring mechanism)
Tanbo [orignally Rootbound] (1993) (Go set, game of roots, adjacency rules)
........
Impasse (2003) (checkers set, armies march through each other)
Byte (2005) (checkers set, passive stacks, move distance, merging)
Cephalopod (2006) (territory, dice as pieces)
Copolymer (2006) (multiple cells claimed per turn, adjacency rules)
Diffusion (2006) (standard mancala set)
Nested Y (2006) (connection game, nested boards)
Scribe (2006) (pattern making)
Crossway (2007) (Go set, square connection game)
Dipole (2007) (checkers set, merging, stacking, attacking game, annihilation)
Oust (2007) (board starts empty and ends with only one colour, adjacency rules)
Palisade (2007) (Go set, territory)
Rush (2007) (Go set, multiple stones added each turn, adjacency rules, Go -like)
Anchor (2008) (triangular board with two triangular holes, connection game)
Anchor 3D (2008) (3D connection game)
Atoll (2008) (a game of perimeter islands, connect islands of your colour, connection game)
Begird (2008) (generalized form of Y, connection game)
Blood Diamonds (2008) (online play)
Lariat (2008) (3D connection game, world’s simplest connection game)
Super Lariat (2008) (3D connection game, highly unusual geometry)
Loophole (2008) (rhombus-shaped board with two rhombus-shaped holes, connection game)
Loophole 3D (2008) (3D connection game)
Mobius (2008) (connection game, Moebius strip board)
Variable Trump Tute (2008) (card game)
Basic (2009) (checkers set, mixed stacks, stacking)
Mosaic (2009) (tile game, territory, adjacency rules)
Grand Hex (2009) (Hex with an added complication, connection game)
Hex KB (2009) (connection game, Klein bottle board)
X (2009) (connection game for three)
Cage (2010) (checkers, centre move distance)
Colonnade (2010) (stacking, one dimensional)
Fractal (2010) (beautiful board designed to minimize first move advantage, connection game)
Flume (2010) (territory, dots and boxes)
Jostle (2010) (combinatorial, adjacency rules)
Mad Bishops (2010) (combinatorial, annihilation, adjacency rules)
Mad Rooks (2010) (combinatorial, annihilation, adjacency rules)
Rive (2010) (high churn rate, adjacency group rules like Oust)
Monkey Queen (2011) (stacking, chess variant)
Marvin (2012) (stacking, territory)
Redstone (2012) (Go set, Go-like, invulnerable red stones with captures)
Trivor (2012) (connection game, trivalent, variable orientation cell division)
Tripen (2013) (connection game, trivalent, variable orientation cell division)
Gyre (2015) (surround the centre, connection game)
........
Anaash (2021) (stacking, annihilation, move distance)
Bamboo (2021) (clumps)
Dodo (2021) (simple rules, no moves possible for the goal)
Gopher (2021) (simple rules, adjacency rules)
Inchworm (2021) (stacking, merging)
Kobudai (2021) (checkers set)
Manhattan (2021) (consolidation, Manhattan distance)
Marmot (2021) (territory, Gopher variant)
Pathway (2021) (combinatorial, like Gopher, adjacency rules)
Zola (2021) (distance from centre, simple, robust, strategic)
The first of Mark Steere's games, Quadrature is about pattern-forming to capture opposing pieces. Mark discovered the game could be drawn, which he perceives as a flaw. All his games since Quadrature have been designed to be finite and drawless. Tanbo, originally called Rootbound and played with a Go set, followed in 1993.
After the start with Quadrature and Tanbo, Mark designed no new games for some years. Then, he entered a long and prolific period of game design from 2003 to around 2012. Cephalopod, from 2006, a game using dice as pieces, has keen players on Boîte à Jeux, and may be the second best known of Mark's games. Easily the most popular and influential of his games, Oust, came out in 2007.
After 2013, Mark's output of new games slowed. Aside from the connection game, Gyre (2015), Mark produced no new games until 2021. According to the designer,
"Michael Amundsen contacted me about Cage to see if his understanding of the rules was correct. Reading the Cage rule sheet didn't refresh my memory. It was totally gone. But our discussions inspired me to end my eight-year hiatus from game design. I cranked out Zola and seven other games in about as many weeks."
The games in this selective retrospective are given in chronological order below, his earlier games first. As I mentioned above, the selection does not imply a particular ranking or that these are necessarily the very best of the designer's games. They stood out for one reason or another and the choice is subjective. Nevertheless, I have played these games, and I can attest that they all have interesting features illustrating various aspects of Mark Steere's philosophy of game design.
Byte
One of Mark's design principles is that his games should use generic equipment. He wants his games to be as accessible as possible. You can see from the ludography above that many of the games are playable with a checkers set or a Go set. Even those games that need something more than one of these basic pieces of equipment usually rely on the simplest materials that most gamers have—hexagonal or squared boards of various sizes and pieces of one type, usually checkers.
Byte, one of the earliest Mark Steere games, uses nothing but a checkers set. Byte is a clever and unusual stacking game. Byte is still playable on SuperDuperGames, although I do not think it attracts much attention these days. Many other good games use nothing but a checkers set—Boom & Zoom from AG21, for example. Byte deserves to stand with the best of these games, in my opinion.
After the start with Quadrature and Tanbo, Mark designed no new games for some years. Then, he entered a long and prolific period of game design from 2003 to around 2012. Cephalopod, from 2006, a game using dice as pieces, has keen players on Boîte à Jeux, and may be the second best known of Mark's games. Easily the most popular and influential of his games, Oust, came out in 2007.
After 2013, Mark's output of new games slowed. Aside from the connection game, Gyre (2015), Mark produced no new games until 2021. According to the designer,
"Michael Amundsen contacted me about Cage to see if his understanding of the rules was correct. Reading the Cage rule sheet didn't refresh my memory. It was totally gone. But our discussions inspired me to end my eight-year hiatus from game design. I cranked out Zola and seven other games in about as many weeks."
The games in this selective retrospective are given in chronological order below, his earlier games first. As I mentioned above, the selection does not imply a particular ranking or that these are necessarily the very best of the designer's games. They stood out for one reason or another and the choice is subjective. Nevertheless, I have played these games, and I can attest that they all have interesting features illustrating various aspects of Mark Steere's philosophy of game design.
Byte
One of Mark's design principles is that his games should use generic equipment. He wants his games to be as accessible as possible. You can see from the ludography above that many of the games are playable with a checkers set or a Go set. Even those games that need something more than one of these basic pieces of equipment usually rely on the simplest materials that most gamers have—hexagonal or squared boards of various sizes and pieces of one type, usually checkers.
Byte, one of the earliest Mark Steere games, uses nothing but a checkers set. Byte is a clever and unusual stacking game. Byte is still playable on SuperDuperGames, although I do not think it attracts much attention these days. Many other good games use nothing but a checkers set—Boom & Zoom from AG21, for example. Byte deserves to stand with the best of these games, in my opinion.
Only the dark squares are utilized. Checkers can move onto other checkers of either colour, and stacks will form. No stack can have more than eight checkers. As soon as a stack of eight checkers is formed, it is captured by the player whose colour is on top. With 24 checkers, there will be three stacks. The objective is to win two of the three stacks.
Even before we go to the rest of the rules, it is worth noting some of the structure of Byte: three stacks of eight checkers; each stack must be won by one player or the other so the game is decisive; even games with one stack won by each player will come down to a sharp endgame as the final eight pieces unite to form a stack. I think Mark would agree that these features of Byte are good "architecture"—see Oust, below, for a bigger discussion of the meaning and importance of architecture in Mark's games. Three stacks of eight is a simple way to make a decisive game from a checkers set.
Here are the rest of the rules. One player takes the white checkers, the other the black checkers. White moves first, and thereafter the players take turns to move. If a move is available, you must move; if you have no move available, your opponent continues until you can make a move.
On a turn, a player can either slide a stack or merge a stack. (A single checker in a square is a stack of one.) Two stacks next to each other may merge. If you have a checker in one of the stacks, you may lift this checker along with any other checkers on top of it, and place this "sub-stack" on top of the stack it is adjacent to. There are two restrictions: firstly, no stack may be formed greater than eight checkers high; secondly, your checker at the bottom of the "sub-stack" you are moving must end up strictly higher than the position it started from. Merging is the only way stacks can be split.
Otherwise, you can slide a stack to the next space, provided one of your checkers is at the bottom of the stack. Stacks must be slid as a whole, without breaking them up. When sliding a stack, the stack must be moved in a direction that reduces the distance between the sliding stack and the closest other stack, measured by the number of sliding moves between the two stacks. If two stacks are of equal closest distance away, you may decide which one to move nearer to. If a stack is already adjacent to another stack, it may not be slid, because its distance to the closest stack is already minimal—its only option is to merge.
The distance mechanism to constrain move options is a theme that runs through many of Mark's games, whether distance between pieces, as here, or distance from a certain point on the board. We will review this point more thoroughly below in the discussion of Monkey Queen and Zola. I think it is fair to say that the distance mechanism is one of the primary ways in Mark Steere games to guarantee a decisive outcome.
As indicated above, as soon as a stack of eight checkers is formed, it is removed from the board and scores one point for the player with the checker on top. The first player to capture two stacks in this way is the winner.
There are some obvious first points to note about Byte strategy. When you merge stacks with the bottom checker of a stack, you lose a stack that you control. Obviously, you have no choice but to merge stacks in this way at the start of the game, but generally you want to keep as many stacks on the board as possible with your colour at the bottom, because it will help to keep your options open.
On the other hand, when merging stacks to reach eight high, the moving stack of checkers needs your checker on top for you to win. It is also important, therefore, to have stacks with your own checkers on top. However, you will quickly find that when two stacks face off, the stacks with checkers on top will not always win. The winner depends on the particular colour distributions in the two stacks, and you need to be sure to read out the outcome before moving these two stacks together. Overall, I think it is more important to have checkers at the bottom of stacks rather than the top.
Ralf Gering has some interesting analysis of Byte endgame strategy here in BoardGameGeek. He deals with the situations that can result when stacks face off to create the third and final eight-stack of the game, although his analysis works whenever two stacks face off earlier. Ralf's analysis is what opened me to the interest of Byte. It is a game that has excellent internal structure, perfectly utilizing a checkers set to create a decisive, unusual stacking game.
International Byte is a scaled up version of Byte, played with the 10x10 board and 40 checkers of International Checkers. The objective now is to be first to capture three of the possible five eight-checker stacks. Otherwise, the rules are the same.
Diffusion
Diffusion is Mark Steere's interpretation of the mancala games. Diffusion is not another version of an existing type of mancala game, but rather a game with unusual objective and manner of sowing the seeds around the board. The starting point for Diffusion is a 2x6 mancala set, with the typical outsize holes for captures at either end. As mentioned above for Byte, one of Mark's design criteria is that his games should use generic equipment. Diffusion is an alternative use of the most common and generic type of mancala equipment. Diffusion is playable on SuperDuperGames.
The board and starting position is shown below. The holes do not have to be differently coloured, and a regular mancala set will do just as well. The colours are useful to clarify some of the special characteristics of Diffusion.
Even before we go to the rest of the rules, it is worth noting some of the structure of Byte: three stacks of eight checkers; each stack must be won by one player or the other so the game is decisive; even games with one stack won by each player will come down to a sharp endgame as the final eight pieces unite to form a stack. I think Mark would agree that these features of Byte are good "architecture"—see Oust, below, for a bigger discussion of the meaning and importance of architecture in Mark's games. Three stacks of eight is a simple way to make a decisive game from a checkers set.
Here are the rest of the rules. One player takes the white checkers, the other the black checkers. White moves first, and thereafter the players take turns to move. If a move is available, you must move; if you have no move available, your opponent continues until you can make a move.
On a turn, a player can either slide a stack or merge a stack. (A single checker in a square is a stack of one.) Two stacks next to each other may merge. If you have a checker in one of the stacks, you may lift this checker along with any other checkers on top of it, and place this "sub-stack" on top of the stack it is adjacent to. There are two restrictions: firstly, no stack may be formed greater than eight checkers high; secondly, your checker at the bottom of the "sub-stack" you are moving must end up strictly higher than the position it started from. Merging is the only way stacks can be split.
Otherwise, you can slide a stack to the next space, provided one of your checkers is at the bottom of the stack. Stacks must be slid as a whole, without breaking them up. When sliding a stack, the stack must be moved in a direction that reduces the distance between the sliding stack and the closest other stack, measured by the number of sliding moves between the two stacks. If two stacks are of equal closest distance away, you may decide which one to move nearer to. If a stack is already adjacent to another stack, it may not be slid, because its distance to the closest stack is already minimal—its only option is to merge.
The distance mechanism to constrain move options is a theme that runs through many of Mark's games, whether distance between pieces, as here, or distance from a certain point on the board. We will review this point more thoroughly below in the discussion of Monkey Queen and Zola. I think it is fair to say that the distance mechanism is one of the primary ways in Mark Steere games to guarantee a decisive outcome.
As indicated above, as soon as a stack of eight checkers is formed, it is removed from the board and scores one point for the player with the checker on top. The first player to capture two stacks in this way is the winner.
There are some obvious first points to note about Byte strategy. When you merge stacks with the bottom checker of a stack, you lose a stack that you control. Obviously, you have no choice but to merge stacks in this way at the start of the game, but generally you want to keep as many stacks on the board as possible with your colour at the bottom, because it will help to keep your options open.
On the other hand, when merging stacks to reach eight high, the moving stack of checkers needs your checker on top for you to win. It is also important, therefore, to have stacks with your own checkers on top. However, you will quickly find that when two stacks face off, the stacks with checkers on top will not always win. The winner depends on the particular colour distributions in the two stacks, and you need to be sure to read out the outcome before moving these two stacks together. Overall, I think it is more important to have checkers at the bottom of stacks rather than the top.
Ralf Gering has some interesting analysis of Byte endgame strategy here in BoardGameGeek. He deals with the situations that can result when stacks face off to create the third and final eight-stack of the game, although his analysis works whenever two stacks face off earlier. Ralf's analysis is what opened me to the interest of Byte. It is a game that has excellent internal structure, perfectly utilizing a checkers set to create a decisive, unusual stacking game.
International Byte is a scaled up version of Byte, played with the 10x10 board and 40 checkers of International Checkers. The objective now is to be first to capture three of the possible five eight-checker stacks. Otherwise, the rules are the same.
Diffusion
Diffusion is Mark Steere's interpretation of the mancala games. Diffusion is not another version of an existing type of mancala game, but rather a game with unusual objective and manner of sowing the seeds around the board. The starting point for Diffusion is a 2x6 mancala set, with the typical outsize holes for captures at either end. As mentioned above for Byte, one of Mark's design criteria is that his games should use generic equipment. Diffusion is an alternative use of the most common and generic type of mancala equipment. Diffusion is playable on SuperDuperGames.
The board and starting position is shown below. The holes do not have to be differently coloured, and a regular mancala set will do just as well. The colours are useful to clarify some of the special characteristics of Diffusion.
Diffusion is a game for two players. One player owns the left side of the board (green above); the other player owns the right side of the board (yellow above). The objective is to empty of seeds the six round holes on your side of the board. As soon as you accomplish this, you win.
On a turn, a player picks any of the round holes, lifts all the seeds from this hole, and sows these seeds one by one in other holes, just as with regular mancala. However, the sowing starts in the hole immediately to the right of the hole emptied of seeds, and proceeds around the hole emptied of seeds in a counter-clockwise direction. Thus, the seeds from c1 at the start would be sown in d1, d2, c2, and b2. If there were 5 seeds in c1, the last seed would fall into b1. No hole (aside from the capturing hole at either end) can have more than 5 seeds. A player starting from c2 instead would sow into b2, b1, c1, and d1. And so on.
The capturing holes at either end count as two holes for sowing seeds. For example, a move starting from a2 would sow two seeds into the green capturing hole to the left, and then one each in a1 and b1. The capturing holes at either end accumulate seeds throughout the game. No move can start from the capturing holes.
When in the course of sowing, a seed would fall into a hole causing it to have 6 seeds, this seed instead is placed into a capturing hole (which capturing hole is irrelevant), and sowing continues from the next hole. To repeat, no hole can have more than 5 seeds.
Remember, you can move from any of the 12 round holes that contain seeds, not only those on your side. However, you win immediately by emptying your own 6 round holes of seeds.
Ever since he discovered the possibility of a repetitive position in Quadrature, Mark has striven to make sure his games are necessarily of finite duration and drawless. Perhaps the drawlessness of his games is his most consistent design motif. It is not immediately obvious that Diffusion is drawless. Is it possible for seeds to be recycled back and forth endlessly? The answer is, No. However many seeds are in a hole, when these seeds are lifted and sown, one seed must be moved to the hole to its immediate right. Every possible move has this quality. Eventually seeds must inevitably shuffle off into one capture hole or the other; eventually the game must cycle to a conclusion.
Assuming you are playing the green holes above. There are four ways in which you can cycle seeds out of your six holes. The first is to move them into the left capture hole from a2; the second is to move them into the opponent's holes from c1. Thirdly, with at least four seeds in a1, you can also cycle seeds into the left capture hole; lastly, with at least four seeds in c2, you can cycle seeds into your opponent's holes. Two seeds in a2, for example, will bury two seeds in the green capture hole; likewise, two seeds in c1 will enable you to move two seeds into your opponent's holes (and none into yours). Of course, your opponent is trying similar manoeuvres. The person who most efficiently cycles seeds out of his side of the board will win.
There are two possible strategies: aim to bury seeds in your capturing hole or aim to recycle seeds to the opponent's side of the board. Neither strategy alone is likely to succeed, and you will need to balance the two.
Each of the holes has an efficiency rating for you, depending on how many seeds it contains, and how well it can cycle the seeds out of play or onto the opponent's side. Suppose, for example, c1 contains 1 seed and d1 and d2 are empty. You can move from c1 to d1, placing a seed on the opponent's side, furthest from being moved out of the game by the opponent or recycled back to your side. On the other hand, if c1 contained 2 seeds, while d1 and d2 were still empty, a move from c1 would place seeds in d1 and d2—ready from d2 to be recycled straight back to your side! The first option is more efficient for you than the second option.
I do not pretend to understand the details of Diffusion strategy and tactics. However, I am sure that this kind of thinking is on the right track. Diffusion is a mancala game, but the recycling of pieces, and maximizing the efficiency of this recycling, is different from other mancala games. There is a rhythm to a game of Diffusion, with the seeds cycling back and forth. The endgame is sharp and interesting.
I already confirmed above the drawlessness of Diffusion, and mentioned Mark's major design criterion of ensuring his games are always finite and decisive. Aside from ensuring the games are drawless, however, we need them also to have interesting tactics and strategic options. Shogi and Go have interesting tactics and strategic options, although neither satisfies at least one of Mark's design criteria: they can be drawn. I asked Mark for his opinion of games like Shogi and Go that can be drawn, given his contention that drawlessness is a paramount quality of games. His answer deals also with a key point about the playability of his games:
"There's an assumption that, as a game designer, I probably have some appreciation of the classics. And I do, in a way. But I totally don't understand them. When I say I'm a way below average player of games, this isn't false modesty. I tried to learn Go a few times, because of its history and tradition and mystique—and because well meaning friends were always recommending it. But my attempted foray into Go only established that it's well beyond my grasp. In Chess and its variations, I get caught up in little local things, and I don't see the forest through the trees. Now, after playing Zola many times, I can see that there's something to it. I know I'll never be good at it, but I also know that skilled players can advance their strategy without bounds. If they ever wring out Zola on a size 6 board, it's scalable, so they can advance to a size 8. That reminds me. One more criteria for games of the highest form: scalability.
"But my strategy blindness is why my philosophy only relates to architectural interest with no consideration of ensuing tactics or strategy. I do have a deep appreciation of games, but it's like standing outside a car showroom and admiring the cars with no intention of actually driving them. If my game luckily turns out to be strategic, yay, I'm all for it. It's just not what drives me from the outset. Now, that being said, I think my odds of developing a strategic success are about the same as other designers who are trying to do it. I call Christian Freeling the 'game whisperer' because he seems to have the insight required to tweak variations of Chess and Checkers into strategic games. But other than Christian, I really don't know of anyone who can do it without a big helping of luck, just like me."
I think here Mark is referring to the fact that for him, game architecture is paramount. Shogi and Go may well be great games, although he will never be able to judge them properly. He has the humility to admit, in fact, that he will never have the skill to judge the quality of his games properly in terms of their actual playability. All he can judge adequately is a game's architecture, and the fact that a game is drawless by necessity is good architecture.
Oust
Oust is certainly Mark's most influential abstract game. I would guess that Oust has been studied and played more than all his other games put together. Oust can be played on a squared board, up to the size of a full Go set, or on a hexhex board with base-7 standard, although other sizes are possible. The sense in the abstract games community seems to be that Hex Oust is the better game, but I do not think there is certainty on this point. (See, for example, the discussion thread here on BoardGameGeek.)
You need a collection of black and white pieces, and of course Go stones are ideal. The board in Oust starts off empty. The players take turns to place stones in vacant board spaces, and it does not matter which colour moves first. The objective is for one player to capture all the opponent's stones. A collection of stones of the same colour that is connected (as in Go or Hex) is a group.
A non-capturing move is placement of a stone that does not connect to any friendly stones already placed on the board, and therefore does not extend any friendly groups. The non-capturing move can connect with one or more enemy stones.
A capturing move is placement of a stone that does connect to one or more friendly stones and does extend a friendly group or even unite two or more friendly groups. A capturing move is only possible provided that the friendly group so extended does thereby connect with one or more enemy groups and furthermore that these enemy groups are strictly smaller in size than the friendly group so extended. These one or more enemy groups connected to by the extended friendly group are captured and removed from the board.
After a capturing move, a player may make another capturing move in the same turn, and so on, until the player's turn ends with a non-capturing move. It is not required to make a capturing move when one is available. For example, below, Black can play A, uniting two groups and capturing the two isolated white stones next to them. Black can then play B, capturing the two-stone white group, and then C, capturing the three-stone white group. Black could not play C then B instead. Lastly, Black could play D, making a non-capturing move, even though it is adjacent to a white group. (This is an example for the rules only, and not a demonstration of good play!)
On a turn, a player picks any of the round holes, lifts all the seeds from this hole, and sows these seeds one by one in other holes, just as with regular mancala. However, the sowing starts in the hole immediately to the right of the hole emptied of seeds, and proceeds around the hole emptied of seeds in a counter-clockwise direction. Thus, the seeds from c1 at the start would be sown in d1, d2, c2, and b2. If there were 5 seeds in c1, the last seed would fall into b1. No hole (aside from the capturing hole at either end) can have more than 5 seeds. A player starting from c2 instead would sow into b2, b1, c1, and d1. And so on.
The capturing holes at either end count as two holes for sowing seeds. For example, a move starting from a2 would sow two seeds into the green capturing hole to the left, and then one each in a1 and b1. The capturing holes at either end accumulate seeds throughout the game. No move can start from the capturing holes.
When in the course of sowing, a seed would fall into a hole causing it to have 6 seeds, this seed instead is placed into a capturing hole (which capturing hole is irrelevant), and sowing continues from the next hole. To repeat, no hole can have more than 5 seeds.
Remember, you can move from any of the 12 round holes that contain seeds, not only those on your side. However, you win immediately by emptying your own 6 round holes of seeds.
Ever since he discovered the possibility of a repetitive position in Quadrature, Mark has striven to make sure his games are necessarily of finite duration and drawless. Perhaps the drawlessness of his games is his most consistent design motif. It is not immediately obvious that Diffusion is drawless. Is it possible for seeds to be recycled back and forth endlessly? The answer is, No. However many seeds are in a hole, when these seeds are lifted and sown, one seed must be moved to the hole to its immediate right. Every possible move has this quality. Eventually seeds must inevitably shuffle off into one capture hole or the other; eventually the game must cycle to a conclusion.
Assuming you are playing the green holes above. There are four ways in which you can cycle seeds out of your six holes. The first is to move them into the left capture hole from a2; the second is to move them into the opponent's holes from c1. Thirdly, with at least four seeds in a1, you can also cycle seeds into the left capture hole; lastly, with at least four seeds in c2, you can cycle seeds into your opponent's holes. Two seeds in a2, for example, will bury two seeds in the green capture hole; likewise, two seeds in c1 will enable you to move two seeds into your opponent's holes (and none into yours). Of course, your opponent is trying similar manoeuvres. The person who most efficiently cycles seeds out of his side of the board will win.
There are two possible strategies: aim to bury seeds in your capturing hole or aim to recycle seeds to the opponent's side of the board. Neither strategy alone is likely to succeed, and you will need to balance the two.
Each of the holes has an efficiency rating for you, depending on how many seeds it contains, and how well it can cycle the seeds out of play or onto the opponent's side. Suppose, for example, c1 contains 1 seed and d1 and d2 are empty. You can move from c1 to d1, placing a seed on the opponent's side, furthest from being moved out of the game by the opponent or recycled back to your side. On the other hand, if c1 contained 2 seeds, while d1 and d2 were still empty, a move from c1 would place seeds in d1 and d2—ready from d2 to be recycled straight back to your side! The first option is more efficient for you than the second option.
I do not pretend to understand the details of Diffusion strategy and tactics. However, I am sure that this kind of thinking is on the right track. Diffusion is a mancala game, but the recycling of pieces, and maximizing the efficiency of this recycling, is different from other mancala games. There is a rhythm to a game of Diffusion, with the seeds cycling back and forth. The endgame is sharp and interesting.
I already confirmed above the drawlessness of Diffusion, and mentioned Mark's major design criterion of ensuring his games are always finite and decisive. Aside from ensuring the games are drawless, however, we need them also to have interesting tactics and strategic options. Shogi and Go have interesting tactics and strategic options, although neither satisfies at least one of Mark's design criteria: they can be drawn. I asked Mark for his opinion of games like Shogi and Go that can be drawn, given his contention that drawlessness is a paramount quality of games. His answer deals also with a key point about the playability of his games:
"There's an assumption that, as a game designer, I probably have some appreciation of the classics. And I do, in a way. But I totally don't understand them. When I say I'm a way below average player of games, this isn't false modesty. I tried to learn Go a few times, because of its history and tradition and mystique—and because well meaning friends were always recommending it. But my attempted foray into Go only established that it's well beyond my grasp. In Chess and its variations, I get caught up in little local things, and I don't see the forest through the trees. Now, after playing Zola many times, I can see that there's something to it. I know I'll never be good at it, but I also know that skilled players can advance their strategy without bounds. If they ever wring out Zola on a size 6 board, it's scalable, so they can advance to a size 8. That reminds me. One more criteria for games of the highest form: scalability.
"But my strategy blindness is why my philosophy only relates to architectural interest with no consideration of ensuing tactics or strategy. I do have a deep appreciation of games, but it's like standing outside a car showroom and admiring the cars with no intention of actually driving them. If my game luckily turns out to be strategic, yay, I'm all for it. It's just not what drives me from the outset. Now, that being said, I think my odds of developing a strategic success are about the same as other designers who are trying to do it. I call Christian Freeling the 'game whisperer' because he seems to have the insight required to tweak variations of Chess and Checkers into strategic games. But other than Christian, I really don't know of anyone who can do it without a big helping of luck, just like me."
I think here Mark is referring to the fact that for him, game architecture is paramount. Shogi and Go may well be great games, although he will never be able to judge them properly. He has the humility to admit, in fact, that he will never have the skill to judge the quality of his games properly in terms of their actual playability. All he can judge adequately is a game's architecture, and the fact that a game is drawless by necessity is good architecture.
Oust
Oust is certainly Mark's most influential abstract game. I would guess that Oust has been studied and played more than all his other games put together. Oust can be played on a squared board, up to the size of a full Go set, or on a hexhex board with base-7 standard, although other sizes are possible. The sense in the abstract games community seems to be that Hex Oust is the better game, but I do not think there is certainty on this point. (See, for example, the discussion thread here on BoardGameGeek.)
You need a collection of black and white pieces, and of course Go stones are ideal. The board in Oust starts off empty. The players take turns to place stones in vacant board spaces, and it does not matter which colour moves first. The objective is for one player to capture all the opponent's stones. A collection of stones of the same colour that is connected (as in Go or Hex) is a group.
A non-capturing move is placement of a stone that does not connect to any friendly stones already placed on the board, and therefore does not extend any friendly groups. The non-capturing move can connect with one or more enemy stones.
A capturing move is placement of a stone that does connect to one or more friendly stones and does extend a friendly group or even unite two or more friendly groups. A capturing move is only possible provided that the friendly group so extended does thereby connect with one or more enemy groups and furthermore that these enemy groups are strictly smaller in size than the friendly group so extended. These one or more enemy groups connected to by the extended friendly group are captured and removed from the board.
After a capturing move, a player may make another capturing move in the same turn, and so on, until the player's turn ends with a non-capturing move. It is not required to make a capturing move when one is available. For example, below, Black can play A, uniting two groups and capturing the two isolated white stones next to them. Black can then play B, capturing the two-stone white group, and then C, capturing the three-stone white group. Black could not play C then B instead. Lastly, Black could play D, making a non-capturing move, even though it is adjacent to a white group. (This is an example for the rules only, and not a demonstration of good play!)
You can see, even from the simple example above, that a player can grow groups by capturing a series of smaller enemy groups. Indeed, it is possible in Oust even to be reduced to a single piece and come back to win, provided all the enemy groups are small and weak. You may even want to avoid captures, if it means creating small, weak groups.
Oust rules proscribe certain possibilities and limitations depending on adjacency rules. Thus, the last piece of a move must be adjacent to no friendly pieces; otherwise, if a move creates an adjacency between a larger group and a smaller enemy group, the smaller group is captured. Many of Mark's games depend on adjacency rules, which will be discussed more fully under Flume, below.
Oust's simple rules are in a way completely obvious. You create a new group on any vacant space, even a space next to an opponent's stone. The point is that your own groups can only grow by destroying opposing groups. If a group can expand through many captures it can become large enough that it is invulnerable and can crush any opposition. The game ends this way.
One of the key points of Oust opening strategy is to try to force your opponent to create many small groups. Sacrificing stones to force the creation of small opposing groups is a common opening tactic. Having said that, a second key point of Oust strategy is that you should only capture (and thereby expand your groups) when you have to. Keeping the board full of small enemy groups reduces the enemy's options until the enemy has no option but to make forced errors. Forced errors are moves allowing more stones to be captured.
If both players are following good Oust strategy, the opening will end with a large number of small groups of both colours. Then one or both players must make forced errors, and certain groups expand through capture to dominate sections of the board. Eventually, one behemoth will eliminate all opposition.
The flow of Oust is magical, from an empty board, to the creation of many weak groups, to the combat between these weak groups, to their gradual assimilation into larger and larger groups, and to final domination by the largest group. The character of the game is utterly unguessable from the rules themselves, and only emerges through play and subsequent reflection. If any Mark Steere games are played in 500 years, my guess is that Oust will be one of them.
An excellent resource is The Oust Strategy Guide, which demonstrates some of the richness of Oust, both tactically and strategically. A key part of Oust tactics is the manner in which you attack weak groups to grow your own groups most efficiently. The Guide has some suggestions in this regard, but I am certain that there is much more waiting to be discovered. Oust can be played on SuperDuperGames and Mindsports, and also in Ludii and Ai Ai.
Mark defines his highest form of game in the following terms, clearly including Oust in this company, and starting with his key measure of game quality, its architecture:
"Oust is an architectural goliath. Something I had never heard of was a game that starts with an empty board, is played with two colours, and finishes with one colour. So that became a puzzle for me—which I solved with Oust. My designs are driven solely by architectural interest. If they turn out to be fun, I got lucky. Maybe 5% to 10% of my games have quality gameplay. Oust is one of them.
"Architecture is just the 'wow factor' of a game's rule set as something to behold, with no consideration of the ensuing gameplay. The design should be clever, simple, unique... and beautiful. The game shouldn't just be based on a new mechanism or principle. It should itself be a new mechanism or principle. Zola is one of my best architected games. At first it isn't obvious why annihilation must happen. Then it hits you. 'Oh, I get it. The checkers end up in the corners, and from there it's a straight line of attack to the other corners.' Boom! Architecture.
"Fractal is high on my architecture list. Just one look at it and Boom! Architecture. Gyre has outstanding architecture. It's a pure geometric principle, like Hex or Y, but arguably even simpler.
"And of course Oust. Just the beauty of its mechanism. Starting with an empty board and guaranteeing annihilation. It's otherworldly. Incidentally, finite annihilation is the highest form of a game. By finite I mean two things. Naturally finite with no need of superko or the 50-move rule. And hard finite. You can't have a cycle even if you want to.
"Tripen seems to have interesting architecture. I don't even remember from 9 years ago why it works or how it works. Mad Rooks is nicely architected. Super simple, all out massacre. Other designers have since incorporated its principle in their games. Monkey Queen is pretty awesome. It has rightfully received a lot of compliments over the years. Redstone was well designed. For me, Redstone was nothing more than a solution to a problem—how to make Go naturally finite. Not to diminish Redstone. That's a lot. But as with most games, including my own, just knowing the rules is satisfying. Gopher's sheer simplicity is remarkable.
"For me, game architecture is closely analogous to building architecture. Like the MahaNakhon in Bangkok. Wow!"
Oust rules proscribe certain possibilities and limitations depending on adjacency rules. Thus, the last piece of a move must be adjacent to no friendly pieces; otherwise, if a move creates an adjacency between a larger group and a smaller enemy group, the smaller group is captured. Many of Mark's games depend on adjacency rules, which will be discussed more fully under Flume, below.
Oust's simple rules are in a way completely obvious. You create a new group on any vacant space, even a space next to an opponent's stone. The point is that your own groups can only grow by destroying opposing groups. If a group can expand through many captures it can become large enough that it is invulnerable and can crush any opposition. The game ends this way.
One of the key points of Oust opening strategy is to try to force your opponent to create many small groups. Sacrificing stones to force the creation of small opposing groups is a common opening tactic. Having said that, a second key point of Oust strategy is that you should only capture (and thereby expand your groups) when you have to. Keeping the board full of small enemy groups reduces the enemy's options until the enemy has no option but to make forced errors. Forced errors are moves allowing more stones to be captured.
If both players are following good Oust strategy, the opening will end with a large number of small groups of both colours. Then one or both players must make forced errors, and certain groups expand through capture to dominate sections of the board. Eventually, one behemoth will eliminate all opposition.
The flow of Oust is magical, from an empty board, to the creation of many weak groups, to the combat between these weak groups, to their gradual assimilation into larger and larger groups, and to final domination by the largest group. The character of the game is utterly unguessable from the rules themselves, and only emerges through play and subsequent reflection. If any Mark Steere games are played in 500 years, my guess is that Oust will be one of them.
An excellent resource is The Oust Strategy Guide, which demonstrates some of the richness of Oust, both tactically and strategically. A key part of Oust tactics is the manner in which you attack weak groups to grow your own groups most efficiently. The Guide has some suggestions in this regard, but I am certain that there is much more waiting to be discovered. Oust can be played on SuperDuperGames and Mindsports, and also in Ludii and Ai Ai.
Mark defines his highest form of game in the following terms, clearly including Oust in this company, and starting with his key measure of game quality, its architecture:
"Oust is an architectural goliath. Something I had never heard of was a game that starts with an empty board, is played with two colours, and finishes with one colour. So that became a puzzle for me—which I solved with Oust. My designs are driven solely by architectural interest. If they turn out to be fun, I got lucky. Maybe 5% to 10% of my games have quality gameplay. Oust is one of them.
"Architecture is just the 'wow factor' of a game's rule set as something to behold, with no consideration of the ensuing gameplay. The design should be clever, simple, unique... and beautiful. The game shouldn't just be based on a new mechanism or principle. It should itself be a new mechanism or principle. Zola is one of my best architected games. At first it isn't obvious why annihilation must happen. Then it hits you. 'Oh, I get it. The checkers end up in the corners, and from there it's a straight line of attack to the other corners.' Boom! Architecture.
"Fractal is high on my architecture list. Just one look at it and Boom! Architecture. Gyre has outstanding architecture. It's a pure geometric principle, like Hex or Y, but arguably even simpler.
"And of course Oust. Just the beauty of its mechanism. Starting with an empty board and guaranteeing annihilation. It's otherworldly. Incidentally, finite annihilation is the highest form of a game. By finite I mean two things. Naturally finite with no need of superko or the 50-move rule. And hard finite. You can't have a cycle even if you want to.
"Tripen seems to have interesting architecture. I don't even remember from 9 years ago why it works or how it works. Mad Rooks is nicely architected. Super simple, all out massacre. Other designers have since incorporated its principle in their games. Monkey Queen is pretty awesome. It has rightfully received a lot of compliments over the years. Redstone was well designed. For me, Redstone was nothing more than a solution to a problem—how to make Go naturally finite. Not to diminish Redstone. That's a lot. But as with most games, including my own, just knowing the rules is satisfying. Gopher's sheer simplicity is remarkable.
"For me, game architecture is closely analogous to building architecture. Like the MahaNakhon in Bangkok. Wow!"
Mark reiterates the necessity for drawlessness for good architecture in the strong sense that cyclical situations cannot be constructed even if both players want to. In addition, his ideal game is a combinatorial game of perfect information. Significantly, the objective of the ideal game is annihilation, as with Checkers, or with Oust. These are all aspects of a game's architecture. The "wow factor," however, and the comparison with physical building architecture implies that architecture has an artistic as well as a technical-utilitarian meaning. Of the games that Mark mentions specifically above, we will discuss Fractal, Monkey Queen, Redstone, and Zola below. Here first is Fractal.
Fractal
FractaI is high on the designer's list of games with great architecture. Fractal is a hex-like connection game designed to reduce the advantage of playing first, but without use of the pie rule. The board design, shown below, utilizes a fractal pattern. The reason for highlighting Fractal here, selected from among all of Mark's many connection games, is the brilliance of the board design.
The board starts empty, and the players take turns to occupy a space with a stone of their colour. The objective of Black is to connect outside spaces that are next to the dark border; the objective of White is to connect outside spaces that are next to the light border. (The border, by the way, illustrates rural scenes from Mark's adopted home, Mongolia.) Some of the spaces have connections to both light and dark borders, similarly to corner spaces in Hex.
We need no other rules—the rules are simply those of Hex.
Fractal
FractaI is high on the designer's list of games with great architecture. Fractal is a hex-like connection game designed to reduce the advantage of playing first, but without use of the pie rule. The board design, shown below, utilizes a fractal pattern. The reason for highlighting Fractal here, selected from among all of Mark's many connection games, is the brilliance of the board design.
The board starts empty, and the players take turns to occupy a space with a stone of their colour. The objective of Black is to connect outside spaces that are next to the dark border; the objective of White is to connect outside spaces that are next to the light border. (The border, by the way, illustrates rural scenes from Mark's adopted home, Mongolia.) Some of the spaces have connections to both light and dark borders, similarly to corner spaces in Hex.
We need no other rules—the rules are simply those of Hex.
Normally in Hex, the central spaces are more valuable than edge spaces, because central spaces offer more options for creating connections. This is not the case in Fractal: edge spaces are larger, and possibly more valuable that central spaces.
One of the key features of good games, according to Mark, is scalability. Oust, for example, is highly scalable, and International Byte is a scaled up version of the regular game. Diffusion, too, has a larger variant based on 2x8 holes, although I think the game would be somewhat distorted thereby, not worse, but different. I wonder, too, about the scalability of Fractal. How would it play with a fourth fractal level? I can easily imagine playing Fractal with three levels in physical form, but a fourth level, with tiny central spaces, would probably be unplayable in physical form.
Fractal is one of many connection games that Mark has devised that are inspired by Hex or are otherwise generalizations of Hex. I am not much of a Hex player, and I do not feel able to comment on these games. Nevertheless, as I mentioned, the design of Fractal is different and original, and hence its inclusion in this selective retrospective.
Before leaving the topic of connection games entirely, we should note that Mark has three three-dimensional connection games, Anchor 3D, Lariat, and Loophole 3D. Hex KB (where "KB" stands for "Klein bottle") is effectively a fourth multi-dimensional connection game. These original ideas deserve investigation. Proper study of these games ought not to be hampered by the difficulty of creating a physical set, provided adequate electronic representations are available.
Flume
Flume is a game like the pencil-and-paper classic Dots & Boxes. Flume has hexagonal and square versions. I do not know which is best, and so here is just the square game. Play on a board with an odd number of squares on each side. The outer ring of squares can be filled with neutral pieces, although you do not have to use neutral pieces. The neutral pieces simply help to indicate that a square on the edge or in a corner already has one or two or filled neighbours, respectively. Without the surrounding wall of neutral pieces, you just have to remember this fact—really no more difficult than remembering that a Go stone on the edge has three liberties and a Go stone in the corner has only two liberties—and your board then is two squares each way larger!
The players take turns to place a piece of their colour on an empty square. Before too long, some squares will acquire three or four orthogonally adjacent neighbours, being thereby "surrounded." The colour of the surrounding pieces, black, white, or grey, is irrelevant. If a player places a piece on a surrounded square, that player must now place another piece of his colour. The second piece, too, may be on a surrounded square, and so on. There is no obligation at any time to occupy surrounded squares, if they exist, but when you do, you must place another piece somewhere else, if you can. A player's move finishes with a placement that is not on a surrounded square. As soon as the board is completely filled, the game ends, and the player with most pieces on the board wins.
See the example below, where Black has just placed A. White can in turn take squares from B to J (or from J to B) and win the game 17 to 8.
One of the key features of good games, according to Mark, is scalability. Oust, for example, is highly scalable, and International Byte is a scaled up version of the regular game. Diffusion, too, has a larger variant based on 2x8 holes, although I think the game would be somewhat distorted thereby, not worse, but different. I wonder, too, about the scalability of Fractal. How would it play with a fourth fractal level? I can easily imagine playing Fractal with three levels in physical form, but a fourth level, with tiny central spaces, would probably be unplayable in physical form.
Fractal is one of many connection games that Mark has devised that are inspired by Hex or are otherwise generalizations of Hex. I am not much of a Hex player, and I do not feel able to comment on these games. Nevertheless, as I mentioned, the design of Fractal is different and original, and hence its inclusion in this selective retrospective.
Before leaving the topic of connection games entirely, we should note that Mark has three three-dimensional connection games, Anchor 3D, Lariat, and Loophole 3D. Hex KB (where "KB" stands for "Klein bottle") is effectively a fourth multi-dimensional connection game. These original ideas deserve investigation. Proper study of these games ought not to be hampered by the difficulty of creating a physical set, provided adequate electronic representations are available.
Flume
Flume is a game like the pencil-and-paper classic Dots & Boxes. Flume has hexagonal and square versions. I do not know which is best, and so here is just the square game. Play on a board with an odd number of squares on each side. The outer ring of squares can be filled with neutral pieces, although you do not have to use neutral pieces. The neutral pieces simply help to indicate that a square on the edge or in a corner already has one or two or filled neighbours, respectively. Without the surrounding wall of neutral pieces, you just have to remember this fact—really no more difficult than remembering that a Go stone on the edge has three liberties and a Go stone in the corner has only two liberties—and your board then is two squares each way larger!
The players take turns to place a piece of their colour on an empty square. Before too long, some squares will acquire three or four orthogonally adjacent neighbours, being thereby "surrounded." The colour of the surrounding pieces, black, white, or grey, is irrelevant. If a player places a piece on a surrounded square, that player must now place another piece of his colour. The second piece, too, may be on a surrounded square, and so on. There is no obligation at any time to occupy surrounded squares, if they exist, but when you do, you must place another piece somewhere else, if you can. A player's move finishes with a placement that is not on a surrounded square. As soon as the board is completely filled, the game ends, and the player with most pieces on the board wins.
See the example below, where Black has just placed A. White can in turn take squares from B to J (or from J to B) and win the game 17 to 8.
The moves in Dots & Boxes are to the sides of squares, whereas the squares themselves are captured. In Flume there is no distinction between making a move and capturing a square, either way a piece is moved to occupy an empty square. As simple as Dots & Boxes is, therefore, Flume is even simpler.
I mentioned above the fact that Oust depends on adjacency rules. Flume is absolutely a game of adjacency rules. For the design mechanism of adjacency rules, the actions of pieces, the possibilities for movement and capture or for occupation of spaces, are constrained by adjacency requirements, between stones on the same side or between opposing stones. Games like Copolymer, Rush, Jostle, Mosaic, Mad Bishops and Mad Rooks, and Gopher all make use of adjacency rules. Crossway prohibits a particular adjacency pattern.
I would guess that Flume and Dots & Boxes are closely related and that much of the deep theory about Dots & Boxes can be adjusted for Flume. Dots & Boxes, after all, does have deep theory, as can be attested to by the strong players on Little Golem, for example.
The author himself thinks Flume is one of his best games, certainly his best territorial game—which is the reason Flume is included in this selective retrospective. Surely Flume is as good as Dots & Boxes, although I have yet to penetrate the strategy of either.
Flume is playable on Ludii.
Monkey Queen
In Monkey Queen, the objective is to capture one key piece belonging to the opponent. In this respect, it shares the objective of Chess, and might therefore be classified as a chess variant. However, Monkey Queen is very much a stripped-down version of Chess, much like the game Chad by Christian Freeling, where only the most basic elements of a chess-like game are preserved.
In his original rules, Mark uses a 12x12 squared board and two stacks of checkers, 20 black and 20 white. I think the physical game is better played with two Queens and a collection of 19 checkers (or pawns) each, and with Mark's agreement, I have reformulated the rules in this respect. The starting position is shown below.
I mentioned above the fact that Oust depends on adjacency rules. Flume is absolutely a game of adjacency rules. For the design mechanism of adjacency rules, the actions of pieces, the possibilities for movement and capture or for occupation of spaces, are constrained by adjacency requirements, between stones on the same side or between opposing stones. Games like Copolymer, Rush, Jostle, Mosaic, Mad Bishops and Mad Rooks, and Gopher all make use of adjacency rules. Crossway prohibits a particular adjacency pattern.
I would guess that Flume and Dots & Boxes are closely related and that much of the deep theory about Dots & Boxes can be adjusted for Flume. Dots & Boxes, after all, does have deep theory, as can be attested to by the strong players on Little Golem, for example.
The author himself thinks Flume is one of his best games, certainly his best territorial game—which is the reason Flume is included in this selective retrospective. Surely Flume is as good as Dots & Boxes, although I have yet to penetrate the strategy of either.
Flume is playable on Ludii.
Monkey Queen
In Monkey Queen, the objective is to capture one key piece belonging to the opponent. In this respect, it shares the objective of Chess, and might therefore be classified as a chess variant. However, Monkey Queen is very much a stripped-down version of Chess, much like the game Chad by Christian Freeling, where only the most basic elements of a chess-like game are preserved.
In his original rules, Mark uses a 12x12 squared board and two stacks of checkers, 20 black and 20 white. I think the physical game is better played with two Queens and a collection of 19 checkers (or pawns) each, and with Mark's agreement, I have reformulated the rules in this respect. The starting position is shown below.
Each player starts with a Queen on the board and a stack of 19 checkers off the board. White moves first and thereafter Black and White take turns to move. The pie rule is used. After White's first turn, Black has the option of switching colours and taking the white pieces for the rest of the game.
The Queen (a.k.a., "The Monkey Queen") moves like a Chess Queen, any number of vacant squares orthogonally or diagonally. Every time the Queen moves without capturing, the Queen leaves a "Baby" on the square just vacated. A Baby consists of a checker of the player's colour, placed on the board from the stack off the board. As the Queens move around the board, each Queen will create an army of Babies for defense and attack. Each Baby, too, moves just like a Chess Queen.
The Queen also captures like a Chess Queen. The Queen captures an opposing Baby or Queen within its movement range by replacement. When capturing, the Queen does not deposit a Baby on the square it just vacated. A Baby is left only with a non-capturing move. If a player has no Babies left off the board in reserve, the Queen is not permitted to make a non-capturing move.
The Babies likewise move and capture just like Chess Queens. However, if a Baby makes a non-capturing move, it must move closer to the enemy Queen, where the distance between the Baby and the enemy Queen is measured as the straight-line distance.
There is never any requirement for a Queen or Baby to make a capturing move, except if no other moves are available.
The objective is to capture the enemy Queen. A player may not pass. If a player has no moves available, the player loses.
Monkey Queen is unlike many Mark Steere games in that there are immediately and obviously strategies available to try. The initiative is crucial. We started thinking about mating nets, where the Queen drops a series of Babies to surround the enemy Queen. Otherwise, the Babies can be used defensively, with the Queen “castled” behind a barrier of friendly Babies.
Note that Monkey Queen is another Mark Steere game that uses the notion of distance to restrain move options, as with Byte above and Zola below. Again, with Monkey Queen, the distance mechanism is used to guarantee that the game will end decisively after a finite number of moves.
We have already noted above Mark's claim that annihilation games are the highest form of game, with the cleanest most basic and visceral objective. Monkey Queen requires the capture of one special piece and is a chess variant in this respect. Other classes of games we have already mentioned are the connection game Fractal, the territory game Flume, the mancala game Diffusion, and so on. I was interested in whether Mark set out to create games belonging to each of the major traditional categories or whether there were themes he liked to return to over and over. He responded as follows:
"There's no particular class of games that stands out for me. They each shine in their own way. Probably the most nearly perfect are connection games, if you can look past the pie rule. They're the purest, being little more than a geometric concept. I don't think there's much left to discover in the way of ultra simple connection games after Gyre and Lariat, but you never know.
"The Checkers jump has been used so much that it's kind of a cliche. But I have used the jump, and elements of Chess. The King and Queen are generic enough that I don't mind using them. Monkey Queen is clearly Chess related, having checkmate, but I think that's the end of the Chess variant line for me.
"The alignment goal has also been used a lot. But... if I could find a way to make a decisive alignment game, I'd do it.
"Games can be semi-territorial. But if it's pure territory where you're adding up cells at the end to see who won, that could be non-decisive (unless the rules somehow preclude ties).
"Elimination, yes. I really like a game to be a fight. There's nothing like killing enemy pieces, especially when it leads to total annihilation.
"The only theme I can think of in my games is simplicity. Simple rules and simple equipment. I never use more than one type of playing pieces in a game, with the exception that I wouldn't mind having a game with some subset of kings, queens, and pawns. I didn't realize that I already had such a game in Monkey Queen until you mentioned that it could be played with Queens and Pawns. Nice to know.
"Games that have made a big impression on me include Reversi, Hex, Amazons, and Breakthrough. I never intended to fill out game categories."
Monkey Queen is not a standard Mark Steere game in that strategic choices are immediately apparent, rather than opaque or hidden. Of course, our ideas about mating nets, castling, and so on, may need to be altered or supplemented following more experience at Monkey Queen. Either way, it is a good game that will certainly reward experimentation.
Redstone
I wrote about Redstone in the print version of AG21. This article is reproduced here, and I will supplement it below with a few additional comments.
The first thing to note about Redstone is the emergent objective. On paper, the objective is to capture all opponent's stones. With a Go-like game, and the creation of eyes, we must therefore rely on the opponent having to fill in his own eyes before a "live" group can be captured. At first, you might think that a larger territory would be the goal to aim for. However, the emergent objective instead is to make as many one-point eyes you can make in live groups. It makes no difference whether an eye consists of one point or more, because either player can fill in the remaining points, and even if stones are captured thereby, it does not matter, because no points are awarded for captured stones.
When the number of eyes is significant, it matters how many eyes can be made from any particular territorial shape. First theories of Redstone should investigate which shapes provide two eyes, which ought to be somewhat the same as in Go, although the red stones do alter Go tactics. Then, which shapes can provide three eyes, four eyes, and so on?
For this reason, I doubt that 19x19 is the best size for Redstone, because the larger territories that develop on the larger board will still need to be divided into a specific number of eyes—it does not matter how many points of territory are contained therein. Perhaps 13x13 might be the best size, although we do not have sufficient experience with Redstone to make a good judgment in this respect.
Let me supplement the Redstone discussion with one last point for now. See the diagram below, which shows two minimally live corner positions in Redstone.
The Queen (a.k.a., "The Monkey Queen") moves like a Chess Queen, any number of vacant squares orthogonally or diagonally. Every time the Queen moves without capturing, the Queen leaves a "Baby" on the square just vacated. A Baby consists of a checker of the player's colour, placed on the board from the stack off the board. As the Queens move around the board, each Queen will create an army of Babies for defense and attack. Each Baby, too, moves just like a Chess Queen.
The Queen also captures like a Chess Queen. The Queen captures an opposing Baby or Queen within its movement range by replacement. When capturing, the Queen does not deposit a Baby on the square it just vacated. A Baby is left only with a non-capturing move. If a player has no Babies left off the board in reserve, the Queen is not permitted to make a non-capturing move.
The Babies likewise move and capture just like Chess Queens. However, if a Baby makes a non-capturing move, it must move closer to the enemy Queen, where the distance between the Baby and the enemy Queen is measured as the straight-line distance.
There is never any requirement for a Queen or Baby to make a capturing move, except if no other moves are available.
The objective is to capture the enemy Queen. A player may not pass. If a player has no moves available, the player loses.
Monkey Queen is unlike many Mark Steere games in that there are immediately and obviously strategies available to try. The initiative is crucial. We started thinking about mating nets, where the Queen drops a series of Babies to surround the enemy Queen. Otherwise, the Babies can be used defensively, with the Queen “castled” behind a barrier of friendly Babies.
Note that Monkey Queen is another Mark Steere game that uses the notion of distance to restrain move options, as with Byte above and Zola below. Again, with Monkey Queen, the distance mechanism is used to guarantee that the game will end decisively after a finite number of moves.
We have already noted above Mark's claim that annihilation games are the highest form of game, with the cleanest most basic and visceral objective. Monkey Queen requires the capture of one special piece and is a chess variant in this respect. Other classes of games we have already mentioned are the connection game Fractal, the territory game Flume, the mancala game Diffusion, and so on. I was interested in whether Mark set out to create games belonging to each of the major traditional categories or whether there were themes he liked to return to over and over. He responded as follows:
"There's no particular class of games that stands out for me. They each shine in their own way. Probably the most nearly perfect are connection games, if you can look past the pie rule. They're the purest, being little more than a geometric concept. I don't think there's much left to discover in the way of ultra simple connection games after Gyre and Lariat, but you never know.
"The Checkers jump has been used so much that it's kind of a cliche. But I have used the jump, and elements of Chess. The King and Queen are generic enough that I don't mind using them. Monkey Queen is clearly Chess related, having checkmate, but I think that's the end of the Chess variant line for me.
"The alignment goal has also been used a lot. But... if I could find a way to make a decisive alignment game, I'd do it.
"Games can be semi-territorial. But if it's pure territory where you're adding up cells at the end to see who won, that could be non-decisive (unless the rules somehow preclude ties).
"Elimination, yes. I really like a game to be a fight. There's nothing like killing enemy pieces, especially when it leads to total annihilation.
"The only theme I can think of in my games is simplicity. Simple rules and simple equipment. I never use more than one type of playing pieces in a game, with the exception that I wouldn't mind having a game with some subset of kings, queens, and pawns. I didn't realize that I already had such a game in Monkey Queen until you mentioned that it could be played with Queens and Pawns. Nice to know.
"Games that have made a big impression on me include Reversi, Hex, Amazons, and Breakthrough. I never intended to fill out game categories."
Monkey Queen is not a standard Mark Steere game in that strategic choices are immediately apparent, rather than opaque or hidden. Of course, our ideas about mating nets, castling, and so on, may need to be altered or supplemented following more experience at Monkey Queen. Either way, it is a good game that will certainly reward experimentation.
Redstone
I wrote about Redstone in the print version of AG21. This article is reproduced here, and I will supplement it below with a few additional comments.
The first thing to note about Redstone is the emergent objective. On paper, the objective is to capture all opponent's stones. With a Go-like game, and the creation of eyes, we must therefore rely on the opponent having to fill in his own eyes before a "live" group can be captured. At first, you might think that a larger territory would be the goal to aim for. However, the emergent objective instead is to make as many one-point eyes you can make in live groups. It makes no difference whether an eye consists of one point or more, because either player can fill in the remaining points, and even if stones are captured thereby, it does not matter, because no points are awarded for captured stones.
When the number of eyes is significant, it matters how many eyes can be made from any particular territorial shape. First theories of Redstone should investigate which shapes provide two eyes, which ought to be somewhat the same as in Go, although the red stones do alter Go tactics. Then, which shapes can provide three eyes, four eyes, and so on?
For this reason, I doubt that 19x19 is the best size for Redstone, because the larger territories that develop on the larger board will still need to be divided into a specific number of eyes—it does not matter how many points of territory are contained therein. Perhaps 13x13 might be the best size, although we do not have sufficient experience with Redstone to make a good judgment in this respect.
Let me supplement the Redstone discussion with one last point for now. See the diagram below, which shows two minimally live corner positions in Redstone.
These beautiful positions set the seal for me that Redstone is a game worth investigating. The physical play of the game itself has an appeal in addition to the already gorgeous Go aesthetic. A red stone brings to mind a drop of blood at the kill location. Perhaps this mini-narrative should not carry much weight, but I do wonder what it would be like to play Redstone with traditional glass stones in black and white—now supplemented with red glass stones.
Dodo
Dodo is one of the two recent games by Mark Steere that is covered in this retrospective. The other is Zola, below. Dodo is here because it is another example of an interesting Mark Steere game of extreme simplicity. Dodo can be played on BoardGameArena and Ai Ai. The starting position is shown below, which is a screenshot from the BoardGameArena implementation of Dodo. The yellow highlights behind some of the red pieces are present just to indicate which pieces can make the next move.
Dodo
Dodo is one of the two recent games by Mark Steere that is covered in this retrospective. The other is Zola, below. Dodo is here because it is another example of an interesting Mark Steere game of extreme simplicity. Dodo can be played on BoardGameArena and Ai Ai. The starting position is shown below, which is a screenshot from the BoardGameArena implementation of Dodo. The yellow highlights behind some of the red pieces are present just to indicate which pieces can make the next move.
One player takes the red pieces, the other the blue. Red moves first, and turns alternate. A piece in Dodo can move either directly or obliquely forward, never backwards. You have to make a move if you can. If on your turn to play you cannot make a legal move because your pieces are blocked, you win. And that's it! Dodo has extremely simple rules.
The strategy of Dodo is potentially interesting. You have to open up channels between your pieces to allow enemy pieces to move through them. You definitely do not want to block the movement of enemy pieces! One choice is to move your pieces to the sides and allow enemy pieces to flow through the centre. Another choice is to occupy the centre and send enemy pieces down the sides. I do not know which option, if either, is best.
I suggested to Mark the possibility of using a larger board, with more space and more strategic choices. Like many of his games, I think Dodo is scalable. Perhaps a larger version of Dodo would provide more strategic options.
Dodo looks almost childish, it is so simple. However, I think this appearance is deceiving, as you will see after playing just one or two games. Dodo often presents clear move choices, where it is by no means certain which is the best option. Some observers have suggested that there might be a simple winning algorithm for Dodo. I am not sure. Dodo may be more complex than it first seems.
One of the interesting things about Dodo is that the objective seems almost counter-intuitive. Normally, one would expect the objective to be to deprive your opponent of moves, rather than yourself. However, if you try the "misère" version, where the objective is to leave your opponent with no moves, then you see that the game is perhaps not as effective, because the pieces quickly get snarled up in a standoff. The objective of not blocking your opponent seems to lead to a richer gaming experience, which is curious.
Zola
The last of the games in this short and selective review of the Mark Steere games is Zola. Initially, Mark used a checkered board for the setup, but after a little experimentation we decided it might be better to colour the board differently. I will explain why below. The diagrams below show an empty Zola board to the left and the Zola board with pieces ready to begin to the right. The board is available here.
The strategy of Dodo is potentially interesting. You have to open up channels between your pieces to allow enemy pieces to move through them. You definitely do not want to block the movement of enemy pieces! One choice is to move your pieces to the sides and allow enemy pieces to flow through the centre. Another choice is to occupy the centre and send enemy pieces down the sides. I do not know which option, if either, is best.
I suggested to Mark the possibility of using a larger board, with more space and more strategic choices. Like many of his games, I think Dodo is scalable. Perhaps a larger version of Dodo would provide more strategic options.
Dodo looks almost childish, it is so simple. However, I think this appearance is deceiving, as you will see after playing just one or two games. Dodo often presents clear move choices, where it is by no means certain which is the best option. Some observers have suggested that there might be a simple winning algorithm for Dodo. I am not sure. Dodo may be more complex than it first seems.
One of the interesting things about Dodo is that the objective seems almost counter-intuitive. Normally, one would expect the objective to be to deprive your opponent of moves, rather than yourself. However, if you try the "misère" version, where the objective is to leave your opponent with no moves, then you see that the game is perhaps not as effective, because the pieces quickly get snarled up in a standoff. The objective of not blocking your opponent seems to lead to a richer gaming experience, which is curious.
Zola
The last of the games in this short and selective review of the Mark Steere games is Zola. Initially, Mark used a checkered board for the setup, but after a little experimentation we decided it might be better to colour the board differently. I will explain why below. The diagrams below show an empty Zola board to the left and the Zola board with pieces ready to begin to the right. The board is available here.
White moves first and then the move alternates, with the players taking turns to make a move. If you have a move you must move, otherwise your opponent moves until you do have a move.
We have already discussed distance as a mechanism for restraining the moves of pieces, used by Mark in his games over and over. Distance comes up in Byte and Monkey Queen, above, but also in Cage and Manhattan. In Zola, distance is measured in a straight line from the centre of the board, marked with a red dot in the diagrams above.
The colours on the board represent squares that are equidistant from the centre. Closest are red squares, followed by orange, yellow, green, blue, and lastly violet. I like to imagine the red dot in the centre is the summit of a mountain, and that the collection of squares with the same colour is a "contour line," much as you have contour lines on regular maps.
On a move you have two choices. You can choose to capture an enemy piece. To do this, your own pieces move and capture like Chess Queens, with the proviso that they must either move closer to the centre or at least stay the same distance from the centre. Captures, therefore, can take place by moving pieces closer to the summit or at least keeping them on the same contour line.
The other choice is to move a piece without capturing. In this case, the piece must be moved just one square, like a Chess King, with the proviso that it must move strictly further from the centre.
And that is it, the first player to capture the last of the opponent's pieces wins.
The first obvious point to note about Zola strategy is that pieces further from the centre have the largest number of free capturing moves, and as they move inwards towards the summit their options decrease. It would seem best to capture, therefore, within the same contour line, as that preserves the distance from the centre and the number of options. There will be exceptions, however, and preserving maximum distance form the centre is not always best.
Every non-capturing move must be made with careful thought, because it puts you one step behind in the race to capture all enemy pieces. Sometimes, of course, you have no option, but sometimes it is best to make a non-capturing move even when you do have a capture. These moves and their timing, I think, are key to Zola strategy and tactics.
In this regard, the violet corner squares are the most valuable locations. Firstly, they are impervious to capture, aside from capture by other corner pieces. Secondly, eventually any piece moving without capturing must end up in a corner square. If you have a piece already in a corner square, whichever corner the moving enemy piece ends up in, it can be captured immediately—provided, of course, the path for capture is not blocked by another piece. Sometimes a corner piece may be forced out of a corner to a non-optimal location if the player has no other moves available except to capture with the corner piece. Remember, you can only pass if no other move is available.
For Zola strategy, I think the contour lines might be important. lt is also important to pay close attention to the shapes that develop around corners. There is much to discover with Zola.
According to Mark,
"I stumbled onto a lovely mechanism for assured annihilation with Zola. Luckily, the simple design engenders quality gameplay. It doesn't always. As you've seen, the endgame is a distinct phase where you can draw your opponent out of a corner and win."
With its unusual board, Zola is entered in the Unequal Board Spaces Game Design Competition. Zola is playable on Ludii and Ai Ai.
Conclusion
I have played a fair number of Mark Steere's large collection of games, and written about those that stood out for me. To reiterate, the collection of games I have written about here is not meant to be a list of the best Mark Steere games. My choice is subjective, and many games I simply have not played yet. Gyre, Rive, and Manhattan, for example, all look intriguing, though I have yet to try them.
My investigation of Redstone, here and in AG21, was a bit of a revelation. I do not think anyone previously had given Redstone much thought, even the designer himself. The few old comments I could find about Redstone typically referred to its slowness. Of course, it will seem slow if you need to capture every enemy stone on the board, but the emergent objective means that all you need to do is count the number of eyes to decide the winner. The investigation of Redstone uncovered hidden depths and interesting and beautiful features of the game. How many other Mark Steere games are like Redstone in this respect? At least those games I have highlighted in this article are worthy of consideration, and certainly there are others.
Mark Steere's design philosophy is stark, pure, and honest. He produces games and judges the results of his efforts on their architecture alone. He makes no pretence that he knows how to make good moves in any of his games. His goal is to create games with perfect architecture. If they are also fun, perhaps with interesting strategy and tactics that others can pick up and develop, then that is a bonus. I hope I have made a good argument that Mark Steere has a unique perspective on game design that we can understand and appreciate. ◾️
We have already discussed distance as a mechanism for restraining the moves of pieces, used by Mark in his games over and over. Distance comes up in Byte and Monkey Queen, above, but also in Cage and Manhattan. In Zola, distance is measured in a straight line from the centre of the board, marked with a red dot in the diagrams above.
The colours on the board represent squares that are equidistant from the centre. Closest are red squares, followed by orange, yellow, green, blue, and lastly violet. I like to imagine the red dot in the centre is the summit of a mountain, and that the collection of squares with the same colour is a "contour line," much as you have contour lines on regular maps.
On a move you have two choices. You can choose to capture an enemy piece. To do this, your own pieces move and capture like Chess Queens, with the proviso that they must either move closer to the centre or at least stay the same distance from the centre. Captures, therefore, can take place by moving pieces closer to the summit or at least keeping them on the same contour line.
The other choice is to move a piece without capturing. In this case, the piece must be moved just one square, like a Chess King, with the proviso that it must move strictly further from the centre.
And that is it, the first player to capture the last of the opponent's pieces wins.
The first obvious point to note about Zola strategy is that pieces further from the centre have the largest number of free capturing moves, and as they move inwards towards the summit their options decrease. It would seem best to capture, therefore, within the same contour line, as that preserves the distance from the centre and the number of options. There will be exceptions, however, and preserving maximum distance form the centre is not always best.
Every non-capturing move must be made with careful thought, because it puts you one step behind in the race to capture all enemy pieces. Sometimes, of course, you have no option, but sometimes it is best to make a non-capturing move even when you do have a capture. These moves and their timing, I think, are key to Zola strategy and tactics.
In this regard, the violet corner squares are the most valuable locations. Firstly, they are impervious to capture, aside from capture by other corner pieces. Secondly, eventually any piece moving without capturing must end up in a corner square. If you have a piece already in a corner square, whichever corner the moving enemy piece ends up in, it can be captured immediately—provided, of course, the path for capture is not blocked by another piece. Sometimes a corner piece may be forced out of a corner to a non-optimal location if the player has no other moves available except to capture with the corner piece. Remember, you can only pass if no other move is available.
For Zola strategy, I think the contour lines might be important. lt is also important to pay close attention to the shapes that develop around corners. There is much to discover with Zola.
According to Mark,
"I stumbled onto a lovely mechanism for assured annihilation with Zola. Luckily, the simple design engenders quality gameplay. It doesn't always. As you've seen, the endgame is a distinct phase where you can draw your opponent out of a corner and win."
With its unusual board, Zola is entered in the Unequal Board Spaces Game Design Competition. Zola is playable on Ludii and Ai Ai.
Conclusion
I have played a fair number of Mark Steere's large collection of games, and written about those that stood out for me. To reiterate, the collection of games I have written about here is not meant to be a list of the best Mark Steere games. My choice is subjective, and many games I simply have not played yet. Gyre, Rive, and Manhattan, for example, all look intriguing, though I have yet to try them.
My investigation of Redstone, here and in AG21, was a bit of a revelation. I do not think anyone previously had given Redstone much thought, even the designer himself. The few old comments I could find about Redstone typically referred to its slowness. Of course, it will seem slow if you need to capture every enemy stone on the board, but the emergent objective means that all you need to do is count the number of eyes to decide the winner. The investigation of Redstone uncovered hidden depths and interesting and beautiful features of the game. How many other Mark Steere games are like Redstone in this respect? At least those games I have highlighted in this article are worthy of consideration, and certainly there are others.
Mark Steere's design philosophy is stark, pure, and honest. He produces games and judges the results of his efforts on their architecture alone. He makes no pretence that he knows how to make good moves in any of his games. His goal is to create games with perfect architecture. If they are also fun, perhaps with interesting strategy and tactics that others can pick up and develop, then that is a bonus. I hope I have made a good argument that Mark Steere has a unique perspective on game design that we can understand and appreciate. ◾️
Acknowledgements
- The image of the MahaNakhon skyscraper in Bangkok by Kyle Hasegawa was originally posted to Flickr here. It was reviewed on August 29, 2016 by FlickreviewR and was confirmed to be licensed under the terms of the cc-by-2.0.
- The Dodo starting postiion is a screenshot from BoardGameArena.