I don’t play a lot of video games, but I’m quite fond of strategy, and have recently become besotted with Starcraft 2. Something that struck me while looking through the online strategy community was how ruthlessly empirical that community was.
It shouldn’t be too surprising. Players are provided with an environment governed by the immutable laws of the game engine, and the only objective is to win. You can accomplish this however you like: efficient micromanagement, complementary unit selection, economic superiority, stealth tactics, mind games, aggressive map dominance, off-the-wall strategy your opponent can’t plan for...however you manage it, provided you’re the one left standing at the end, you win.
As a result, players continually test the behaviour of the environment, see what works and throw away what doesn’t. This often involves setting up contrived scenarios to explicitly test certain theories. The result is a massive body of knowledge on how to effectively win the game.
I would say that it’s kind of heartening to find that when given proper incentive, even people with (presumably) no formal scientific training can apply systematic methods for testing the behaviour of their environment, but I don’t know what kind of crossover exists between the SC/scientific community.
It very thoroughly data-mines a strategy WoW forum and observes how their discussion about the hidden variables of the game world mirrors quite clearly the scientific study of the natural world.
A: “I suggest that the Octopus Lord is vulnerable to fire.”
B: “Good idea! I have a fire sword and an identical ice sword, I’m off to try.”
C: “Wait! Maybe he’s just resistant to ice! We need to design a better test.”
It’s actually a lot more complex than that (there’s a HUGE spreadsheet quoted in the paper), but you get the idea.
Reverse engineering isn’t quite the same as science (because you know from the start that all natural laws must be traceable back to a short piece of human-written code), but they are definitely kin.
I don’t play a lot of video games, but I’m quite fond of strategy, and have recently become besotted with Starcraft 2. Something that struck me while looking through the online strategy community was how ruthlessly empirical that community was.
It shouldn’t be too surprising. Players are provided with an environment governed by the immutable laws of the game engine, and the only objective is to win. You can accomplish this however you like: efficient micromanagement, complementary unit selection, economic superiority, stealth tactics, mind games, aggressive map dominance, off-the-wall strategy your opponent can’t plan for...however you manage it, provided you’re the one left standing at the end, you win.
As a result, players continually test the behaviour of the environment, see what works and throw away what doesn’t. This often involves setting up contrived scenarios to explicitly test certain theories. The result is a massive body of knowledge on how to effectively win the game.
I would say that it’s kind of heartening to find that when given proper incentive, even people with (presumably) no formal scientific training can apply systematic methods for testing the behaviour of their environment, but I don’t know what kind of crossover exists between the SC/scientific community.
Here’s a full research paper on the subject.
It very thoroughly data-mines a strategy WoW forum and observes how their discussion about the hidden variables of the game world mirrors quite clearly the scientific study of the natural world.
A: “I suggest that the Octopus Lord is vulnerable to fire.” B: “Good idea! I have a fire sword and an identical ice sword, I’m off to try.” C: “Wait! Maybe he’s just resistant to ice! We need to design a better test.”
It’s actually a lot more complex than that (there’s a HUGE spreadsheet quoted in the paper), but you get the idea.
Reverse engineering isn’t quite the same as science (because you know from the start that all natural laws must be traceable back to a short piece of human-written code), but they are definitely kin.