Don Green, despite holding the impressive title of Professor of Political Science and Director of the Institution for Social and Policy Studies at Yale University, may be best-known for something much more elegant (and for most of us more enjoyable) than politics: He's the designer of the award-winning abstract game Octi.
Octi is played on a 9x9 grid of squares, six of which (three per player) serve as bases. The winner is the first player to occupy all three enemy bases at the same time.
Simple so far -- but this is where things get clever. Each player controls seven octagon-shaped pieces called "pods" and 25 pegs called "prongs." The game starts with an empty pod on each of the bases. On a given turn, players may: (1) add a prong to a pod (the prongs are placed in holes on the sides of the pods), giving it the ability to move in a particular direction; (2) move a pod; or (3) add a new pod to the board.
"Octi is a game akin to directional checkers in which players must decide, on each turn, whether to move a piece, add a new piece to the board, or improve a piece currently on the board," Green said. "In effect, they must choose among quantity, quality, and tempo. Because the pieces jump (over both friendly and opposing pieces) in any of 8 possible directions, the action is fast and furious."
Octi has been compared favorably to ancient games like Chess and Go. Unlike those games, however, it has proven -- so far, at least -- to be "computer-resistant." That is, no genuinely effective artificial intelligence engine has yet been developed for Octi.
Green was kind enough to answer some questions about Octi, Games Magazine's Best Abstract Strategy Game for 2000.
Is Octi your first completed game design?
Octi was the first, and before that, I had never considered designing a game. Octi came to me one day as I was playing with K'nex and Tinker Toys with my children. That was Memorial Day of 1998. It took about 2 days to formulate the whole thing.
Then, it took off fast, as my colleagues and students got caught up in the novelty of playtesting. To my surprise, the core idea -- building your pieces as you play -- held up under playtesting by many curmudgeony people. Small refinements occurred after that (e.g., the four-player game, the superprong variant), but for the most part it was born fully-formed. Odd how that happens sometimes with ideas.
Once you began to seriously develop Octi, how long was it before you decided to look for a publisher?
As a longtime game junkie, I knew that I had something special and unique in Octi, but I had absolutely no idea about prototypes, marketing, the game biz. Nothing.
So I did some silly things like send opaque sketches to my favorite companies, which of course got me nowhere. Then, I got more serious and created some nice prototypes, contacted a local game company (Great American Trading Co.), and they decided to go with it in November of 1998. Octi made its debut in retail stores in June of 1999.
What response did you get from publishers when you submitted the game?
GATCO [The Great American Trading Co.] was enthusiastic right from the start. The headaches initially had to do with production issues (e.g., how to make pieces in wood, what sort of material to use for the board), but those were eventually sorted out.
I like working with them and, more generally, working with a family-owned business such as theirs. They actually enjoy games. Many others in the game business may as well be selling shoes or cars.
What, if anything, surprised you most in the process of designing Octi?
The most remarkable thing was the extent to which the hobby of game inventing changes one as a person. Before Octi, I never did anything with my hands, and I never aspired to make anything tangible. Now, I am an avid woodworker, and I have come to have a new appreciation for craftsmanship and design. I walk through each day with a totally different view of my surroundings.
The following comments from Green explain Octi's background in some detail.
Octi is rooted in my academic work on game theory in politics, in that it illustrates how a game with simple rules can nevertheless have an "equilibrium" that is so deep that it is very difficult to anticipate how players (or computers) will attack the problem or formulate strategies. In this respect, Octi is similar to Go.
The other objective was to formulate a game that would give humans an edge over computers. At the risk of oversimplifying, this tends to occur when a game emphasizes long-range strategic planning over short-range tactics.
While there is nothing new under the sun, Octi expands the role of "promotion" in chess or checkers: pieces are continually being promoted or added to the board. Moreover, the pieces are synergistic -- two weak pieces working in tandem are typically stronger than one powerful piece working alone. Thus, players must create their own pieces with an eye toward their capacity to work together down the road.
The serious details of how Octi evolved are many, but the short story is this. At first, my tendency was to revel in the recalcitrant complexity of the game. The board was 11x11 with 9 pieces per side. GATCO (correctly) perceived that this was more game than the typical person would want to play, and so the current incarnation is 9x9 with 7 pieces per side.
Moreover, they urged me to formulate a "fast game," which I did. Next, they wanted a "family" version that could be played with 4 players. That was easily done, and it turns out to be the most intriguing version of all.
Meanwhile, my colleagues weighed in with their own tastes, which were for more complexity, and so I introduced a toroidal board option and a "superprong" option, both of which appeal to people who revel in complexity.
This feature was originally posted on May 23, 2000.
Green also designed Royal Carpet, a free game that can be played on a standard 8x8 board with game pieces you probably already have.
