Complexity and Depth

Leigh Alexander, over on her blog Sexy Videogameland, struck on something in her latest post about Spore that I thought was interesting, and is something that I’ve been considering for a little while now. Namely, the seemingly widespread (and I believe mistaken) impression that ‘complexity’ in games necessarily leads to ‘depth’.

Now, it’s important not to get wrapped up in arguments about definition, so just let me say what I mean by these two words, ‘complexity’ and ‘depth’.

By complexity I simply mean the number of elements that a player has to manipulate. In other words, is there a lot that the player has to/gets to do? This could be as simple as knowing if a certain mushroom will make you grow or kill you instantly, and as complicated as the handling of a Mitsubishi Evo.

Depth is a little more complicated a concept.

When people talk about ’emergence’ in games what they usually mean is that, unlike other artforms, games each have their own logical space of possibility. Most games have a finite number of ‘states’ they could ever possibly be in. This number of states could be incomprehensibly large or it could be inconsequentially small. These different states rise logically out of the interaction between the rules as carried out by players. Here I am defining ‘depth’ in a game to be the size of the game’s ‘possibility space’; The larger the space, the deeper the game.

So far I don’t think I’ve said anything too controversial, but here we come to the rub: You can easily imagine a game with a very large ruleset, but a very small possibility space.

In fact, I would say that this has become the dominant form of games in the past twenty years or so. The fact remains that designing game with a lot of depth is very, very hard. Most modern games have very small possibility spaces when compared to more traditional games. After all, there are only so many times that you can play through The Legend of Zelda before you memorize exactly where everything is going to be and when. Compare this to a game like Go, which is impossible to memorize because the game has more possible states than there are atoms in the universe.

Game designers even have a term for this idea: ‘elegance’. A game designer will refer to a game as ‘elegant’ when a very small ruleset gives rise to an enormous possibility space. Many of the folk games that we’ve inherited, such as Chess or Mancala, adhere to this ideal. Not to mention more physical games like Wrestling or Foot Racing.

(Keep in mind that these terms are relative, a game like Starcraft has a lot of rules on the face of it, but nothing compared to the enormous possibility space those rules evidently create!)

I got in trouble a little while back for saying that I was disappointed that Braid wasn’t anything more than a puzzle game. It’s not that puzzles are somehow illegitimate or un-entertaining. It’s just that puzzles aren’t great examples of the potential and beauty of games.

The problem with puzzles is that their possibility spaces are about as big as their rulesets. Think about a jigsaw puzzle: the rules governing the shape of a piece also exclusively govern the way that the piece fits in the whole. This is in contrast to a game like Itagaki’s Ninja Gaiden, where people were still exploiting new techniques and perfecting their run-throughs years after the game came out (which is why they kept releasing new versions).

So then, it seems clear that one can have games that are complex but not very deep, and games that are incredibly deep but with a very simple ruleset.

Whether or not this applies to Spore, I have no idea; I haven’t gotten a chance to pick up a copy. What I can say from following the development is that I’ve been given the impression that the aspect of the game that does have a large and interesting possibility space, the Creature Creator, has little to no effect on the rest of the game, which alone has barely any depth at all.