I mentioned in Theory of Fun that humans kinda suck as odds estimation and that this is one reason why games of chance persist. But maybe there’s more!
It turns out that the reward system that lights up when we get a near-miss in a game of skill (which makes sense) gives us the same reward when we have a near-miss in a game of chance (but only if we get to make a choice in the game, such as picking our lotto numbers, even though this has no influence over the outcome of the game).
Those of you who have read Theory of Fun for Game Design may recall this passage:
Let’s picture a game wherein there is a gas chamber shaped like a well. You the player are dropping innocent Jews down into the gas chamber, and they come in all shapes and sizes. There are old ones and young ones, fat ones and tall ones. As they fall to the bottom, they grab onto each other and try to form human pyramids to get to the top of the well. Should they manage to get out, the game is over and you lose. But if you pack them in tightly enough, the ones on the bottom succumb to the gas and die.
I do not want to play this game. Do you? Yet it is Tetris. You could have well-proven, stellar game design mechanics applied towards a quite repugnant premise.
We don’t need to wonder anymore. A comment in the last thread by the felicitously named Raphael Aleixo (my brother’s name is Alex!) tells us that the Brazilian game design club Loodo has made it, with a slight tweak to the theme: I give you Calabouço Tétrico. Read on for my thoughts!
Several folks have emailed me to let me know they got shipping notices from Amazon for the huge backlog of orders of Theory of Fun for Game Design, and now I got one too!
What’s more, it shows as available (sales rank 6225, and #2 in the game programming category!), so if you have been waiting, go grab it now…!
It could use fresh reviews too, so once you get it, please do post reviews. 🙂
Developing behaviors via genetic algorithms of various sorts has been around a long time now. You come up with a basic environment and ruleset, then you let loose millions of generations of simple AIs to keep trying to surivive. You then have the AIs tweak themselves based on what survived well, attempting to evolve the best survivor.
This can be used for lots of purposes — and now it’s being applied to game design. Starting with a simple Pac-Man like environment, researchers are generating zillions of procedural games, and then testing to see which is most fun. But how to measure the fun?
It should be pretty straightforward to see how game rules can be represented to be evolved: just encode them as e.g. an array of integers, and define some sensible mutation and possibly recombination operators. (In this particular case, we use a simple generational EA without crossover.) For other rule spaces, some rules might be more like parameters, and could be represented as real numbers.
What’s the much trickier question is the fitness function. How do you evaluate the fitness of a particular set of game rules? …
Our solution is to use learnability as a predictor of fun. A good game is one that is not winnable by a novice player, but which the player can learn to play better and better over time, and eventually win; it has a smooth learning curve.
The “Theory of Fun” website is down, caught between two webhosts, and has been for a while (with the book out of print, it hasn’t been a priority to sort out). But I keep getting requests for the materials that were hosted there and not here, so here they are.
