malvasia bianca

The upcoming game that I’m most excited about is Spore. It’s by Will Wright, the guy who did Sim City and The Sims, and I’ve never seen anything like it. Or rather, I’ve seen tons of things like parts of it, but nothing that puts its all together the way that this game is trying to do.

He apparently got his inspiration from the old Powers of Ten movie, that starts at a microscopic scale, and pulls the camera back ten fold over and over again until it gets to a human scale, a planetary scale, a galactic scale. Spore is trying to use the same idea in its gameplay: you start off by playing a microscopic organism, that goes around eating stuff and laying eggs. Each time it lays an egg, it changes a bit, gradually evolving into a larger organism. Eventually it gets to the level of a fish swimming around in the ocean, or an animal walking on land, trying to eat, survive, kill other animals and reproduce.

Once your creature gets a big enough brain, it becomes sentient, so the gameplay switches into a of tribal level, where you’re discovering tools and beginning to accrue elemnts of civilization. Then, when your tribe gets big enough, it turns into a Sim City-style game, with Civilization-style conflicts with other cities on the planet. Once you develop further, you begin to terraform and colonize other worlds in your solar system; then you become able to interact with other nearby civilizations, and eventually expand out further through the galaxy.

Which is pretty amazing: lots of my favorite games, all tied together in a wonderful-sounding way.

Wright gave a presentation on it at the 2005 Game Developer’s conference; I recently learned (from Sun’s president’s blog, amusingly enough) that there’s a video of his presentation. (It’s free, but you have to register; I hope I don’t get too much e-mail from them, but it’s worth it, and lots of the other presentations look interesting, too.) It shows lots of gameplay snippets, and also provides some interesting game design insights.

One of the big issues in game design these days is that, as computers get more powerful and storage media gets bigger, games get bigger in scope, requiring huge amount of resources to flesh out. This is a problem: games are now, for example, trying to simulate entire cities fairly seriously, and it takes a huge number of person-hours to, for example, design all those buildings. So content creation eats up huge chunks of games’ budgets.

One thing that Wright noticed about The Sims was that people often really enjoy doing content creation themselves: it’s a lot of fun, it’s a way for people to put their own personal stamp on the game. Apparently people can create stuff in The Sims and upload it to a web site where other people have access to it, and there’s a vast amount of content available this way, of increasingly high quality.

Spore is taking this lesson to heart. In the first place, the evolution of your creatures isn’t random: you control it, with different (but related) editors for the different stages of the game. So the creatures that you end up taking over the galaxy with are unmistakably yours: you decided on every aspect of their development, and another player would have come up with a completely different design.

That’s great, as far as it goes: the game’s creators don’t have to worry about designing the perfect protagonist for the game, they leave that up to you. But what about the creatures you’re going to interact with? How are they going to be designed?

One conventional answer these days would be to simply make the game networked, so the other creatures were played by different people. And I can imagine that an online version of this could be a lot of fun; I wouldn’t be the one having the fun, however, because I don’t have time to regularly devote to online game playing.

But that’s not necessary to solve the problem at hand. Making this an online game would mean that other people both design and control the other creatures; what if we divorce these aspects? Controlling other creatures is tricky, but it’s a problem that we’re getting better at solving without require massive amounts of human resources. Doing large-scale interesting designs, however, is labor-intensive.

So the solution is that, whenever you design something (a creature, a building, whatever), it can get downloaded to the game’s servers. Then, whenever the game needs to populate the surrounding area, or give you choices for buildings to build, or whatever, it downloads new content from those servers. This is done asynchronously behind your back: you don’t have to worry about being logged on at specific times, you don’t play directly against other people, you just all work together to get a big content library.

And it tries to do this in a clever way. It choses creatures to fill up an ecosystem sensible; it makes building designs available that are somewhat similar to the buildings you’ve already got. It notices which designs are more popular and less popular, and adjusts their availability accordingly.

The game is still a work in progress: its current planned release date is the end of 2006, and release dates slip a lot more often than they get pushed up. If they can pull it off, though, it will be amazing.