Game Design Sketchbook

Game Design Sketchbook: Testing the Limits of Single-Player

Jason Rohrer | 8 Aug 2008 17:00
Game Design Sketchbook - RSS 2.0

So it's two players that lets us wring limitless gameplay depth out of the incredibly simple mechanics of Go. But multiplayer is not just an ingredient that can be added to a game like yet another power-up. Instead, multiplayer is more like the fertile soil in which each gameplay system can grow and blossom into its unique, emergent potential. The rules are like the DNA, and the sprawling space of possible gameplay is like the resulting organism. Without multiplayer, it seems that even the most complicated systems of mechanics become stunted and never blossom.

So what do we, with our single-player videogames, have instead? Speed-based physical challenges, randomization, puzzles, character progression, and story advancement - lousy stand-ins for truly interesting gameplay.

Art games in particular carry on the single-player tradition, as do story-centric mainstream games, which also tend to have artistic aspirations. Are we using other mediums, which are fundamentally single-viewer/reader, as a model for how an artistically-meaningful work should be consumed? We can consume paintings, sculptures, films, books, plays, and music by ourselves. We tend to discuss works at length with our companions, but companions are not a necessary component of the art experience. You could still reap deep personal benefit from these works if you were alone on a desert island, though your Go set wouldn't do you much good.

Look around for single-player non-videogames, and you'll find that they are almost non-existent. On the one hand, you have the aforementioned Solitaire card games, which rely on randomness for depth. On the other hand, you have Peg Solitaire in its various forms; no randomness, but another NP-hard optimization problem, like Tetris with a known piece queue. If you ever win at Peg Solitaire, you have officially exhausted the game's depth, since you can then win every future game, so long as you can remember the move sequence that resulted in a win.

So is that what single-player games are doomed to be, at their intellectual best? Puzzles with solutions that are too long to memorize? Is there even such a thing as a single-player game? How is card Solitaire in a fundamentally different class than Rubik's Cube? Turning to videogames, how is Tetris, with the time challenge turned off, not a puzzle? Does placing a time limit on Rubik's Cube turn it into a game?

One videogame that I mentioned earlier is actually a special case. Starcraft, and other realtime and non-realtime strategy games, are not subject to these same depth limitations, even in single-player mode. But what does "single-player mode" for these games involve? AI opponents - additional players controlled by the computer. Though these games are famous for their endless depth when played against other humans, the mechanics can still blossom into interesting gameplay when you play against an AI. You can only exhaust the depth of the single-player variant if you find a weakness in the AI that you can exploit to guarantee a win. In that case, the problem is just that AI for such complicated games is not well developed.

The discussion of AI highlights that the human factor is not what allows simple game mechanics to blossom. It's not what humans bring to the game, but what two competing players - human or not - bring that allows the beautiful complexity and subtlety to emerge. Granted, there is nothing like playing a good game against a good friend, but that social enjoyment, as well as seeing your friend's personality expressed through his or her gameplay, is not a fundamental component of Go-like depth. Go's depth exists separate from the personalities that play it, like a property of the universe just waiting to be discovered whenever two entities sit down, in opposition, to explore it.

But modern single-player videogames, both fringe and mainstream, have almost completely abandoned AI as a game feature. Yes, enemies in most 3-D games are equipped with rudimentary path-finding, planning, and randomized behavior, but it's nothing like what an AI does for Chess. These mainstream AIs are essentially trying to mimic believable human behavior, not provide an opponent with which you can explore the depths of the game mechanics.

Comments on