What is chaos? Why would it be important to games?
Chaos isn’t just a character alignment or a final boss. There is a theory of chaos that has been investigated by mathematicians.
Wikipedia calls it “dynamical systems that are sensitive to initial conditions”, or in other terms, “deterministic but not necessarily predictable”.
This makes it different from randomness. Random systems are predictable in certain ways, but they don’t have deterministic properties. Randomness has no pattern to it. 1
There are many ways in which a chaotic system can appear, but most games contain a certain amount of it simply by having many different systems interacting with each other.
Examples of chaotic systems that have been studied include:
Fractals, such as the Mandelbrot set.
Cellular automata, such as Conway’s Game of Life.
The Lorenz attractor.
One of the fastest ways to ensure that a game provides interesting, systems-driven choices, is to intentionally build chaotic properties into it.
Timers and delays are a very common mechanism for chaos because they disconnect the moment of causation from its effect, introducing opportunities to change or cancel the effect.
Likewise, models of physics introduce properties like bouncing and rolling that make systems harder to predict. 2
Adding different states and modes that are implied by the context are an efficient way to create chaotic elements without making the choices explicit.
Existing chaotic systems like cellular automata can be reworked into a facsimile of an extremely complex system in the real world. This is how the first Simcity games worked.
Artificial intelligence algorithms often exhibit extremely chaotic properties. They are so chaotic, in fact, that they tend to be incomprehensible when observed by humans.
There is a possibility of having too much chaos, making the system resemble a random system.
This is a common occurrence when features like AI are leaned on to support the experience: players who play against AI opponents typically expect them to behave in ways that they can predict and exploit as part of a story experience.
However, if the AIs are too aware of the nature of the game mechanics, they start exploiting those mechanics as well, causing the entire game to fall into an uncanny valley where everything hinges on minor programming details that the player is unlikely to take notice of.
Similar cases of unmanagable chaos exist for games that burden the player with many layers of indirection. Immediate gratification and deep strategy are often in conflict with each other. 3
In the battle royale genre, the initial drop and collection of loot uses a mixture of random and chaotic behaviors.
The loot in a location is random, but player behaviors are often chaotic.
Different players given the same options for loot may make different choices about their strategy.
In contrast, the convention of many previous first person shooters has been to allow players to choose their favorite gear, or to design the map around fixed locations for all items, with no explicit requirement to adapt to the conditions.
By the time you reach the top 10 of a battle royale game, though, you have probably had multiple chances to build a kit of your own design, and so have the other players. Winning depends a great deal more on positioning and skill than on having a critical item.
And so loot acts as a way of adding some chaos and variation in the early stages, as the players with limited gear plot ways of defeating better equipped adversaries.
In perspective: Chaos and belief
A game that people can believe in has more opportunities to explain away its chaotic elements.
When we experience storytelling, in one sense, anything could happen, because the teller said so.
But a story that is believable is one that can justify itself, with the same kind of logic that justifies rules and systems in games.
Both games and stories occupy a kind of cathedral of the mind: we want to believe them, and we want to use them to reflect upon our beliefs, and to say, no, none of this was just a random event. It had a reason for being.
And maybe it is a chaotic and not easily understood reason - but even that is preferable to no reason at all.
- The pseudo-random number generation used in most video games is accomplished through systems that are chaotic but apparently random, since they can be observed and predicted. [return]
- Boulder Dash is an example of a cellular automata system emulating a physics system. The game tracks everything inside a single tilemap and updates the game by updating the map. [return]
- I once decided I wanted to try playing one of these complex wargames, so I got Totaler Krieg and my brother and I attempted to set it up… and it took an entire day. We never touched the game again. [return]