So yesterday I was thinking, "Just for kicks, let's see if I can come up with a game that would be simple to code and not require too much graphics." And naturally my thoughts went to board games. The advantages are clear: Small game world, fairly abstracted rules (in most cases), relatively easy graphics, and the AI doesn't need to be incredibly general. Indeed, it didn't take me too long to come up with a game based on combat between the four elements, though it suffered from the fatal design flaw that the only workable strategy was to gather all your pieces in one blob and march that across the board. It would have been pretty easy to code, though. Anyway, I got to thinking: Why don't we see more board games from indies? Sure, nobody will pay $40 for a board game on the computer, so you won't see the big companies making them, but they would seem ideal for indies. Yet I almost never see them mentioned hereabouts. As a take-off from that, what makes a board game on the computer enjoyable? (Which is what brings me to the design forum.) How can we make board games that take advantage of the computer's strength, while avoiding its weaknesses? Now, you can certainly add a lot of detail to your board game, and let the computer keep track of the numbers. But I'm not convinced this really adds anything except for the hard core of grognards. Most of the really popular board games actually have very simple mechanics - think of Settlers of Catan, for example. One dice roll, one trading mechanism, and some building, that's all! But I strongly suspect that it fails completely when moved to the computer, because the social element is lost. So, what kind of board games can reasonably be played by a single person?

One of the appealing aspects of boardgames is thier interactivity with the peices. Which is one thing that is not as much fun in the computer realm. To make a board game on the computer more compelling is to make it more interactive.

Well, it wasn't after weeks of work - I decided that since the kind of game I was thinking of was very simple, I would carefully consider the design first, and ended up spotting the flaw before writing a single line of code. Come to think of it, you might be able to fix the problem by putting in random setup instead of player-controlled; that should give you lots of little combats all over the place, instead of One Big Phalanx. Faster, too, with no waiting for people to put in pieces one by one. Perhaps I was a little hasty in discarding the idea.

Hmm. I must say I hadn't considered computer Civ as a board game, although I see the similarity now it's been pointed out. It seems to me that it's a slightly different kind of game, though - in what I think of as a 'board game', the pieces do only one thing, and remain constant in strength through the game, and you make decisions about where to send them. In Civ, you have to think about how to increase the strength of the pieces, and decide what they do. At some point, the difference in degree becomes a difference in kind. Then again, perhaps that's exactly the difference I'm looking for; board games are no fun on the computer unless you actually use the strength of the computer to compensate for the weakness. Possibly, yes.

Well, in some sense all games with a top-down view are 'board' games, but I think a distinction between games that could be played on an actual, physical board by humans with counters, and a game that needs a computer to track everything, is useful. In that sense, neither Starcraft nor any other RTS is a board game; their time slices are too small and the number of variables too large for humans to play them without a computer.

I wish you good luck in your project; one of my own is taking shape fairly rapidly, though admittedly the graphics are something only a programmer could love. But, unlike any prior game of mine, the AI actually does moderately sensible things! Since AI is where I always get stuck, I'm quite happy with this development; starting small and keeping the choices limited was apparently a good way to go. It's much easier to write AI when you can actually ask the game board for a list of your legal moves.

As for the one-big-blob problem, the forced-split solution did occur to me during the weekend; the problem was not so much my combat mechanics, which favoured big groups, as having only a single objective. If you make the goal of the game occupation of several different regions, why then that becomes an interesting tension, as large groups are powerful in combat, but can only occupy one victory spot. So I may yet rescue that game concept, but right now I'm making progress on a game about robber barons.