Get the latest Education e-news
 
  • Ambition And Chaos: A Snowfield Postmortem

    [09.20.12]
    - Matthew Weise
  • The Snowfield is an experimental project in emergent narrative set during World War I. It was made by a team of 10 student interns from Singapore and the U.S. in 2011 as part of the eight-week GAMBIT summer development program at MIT. In this article, originally published in Game Developer magazine's 2012 Career Guide, project owner Matthew Weise walks us through what went right and what went wrong with this storytelling experiment.

    I was the product owner for The Snowfield, as it was based on theories of narrative design I had developed over several years as GAMBIT's game design director. Our goal was to take these theories, turn them into a clear method for development, and test that method with a game. The Snowfield was a response to the idea, in both industry and academia, that there are principles of good storytelling that we need to systematize in order to marry video games with narrative effectively. This has led to ideas like "drama management," which assumes that emergent narratives require an A.I. storyteller who controls events to ensure the emerging story is satisfying.

    Such notions are built on an assumption that "good storytelling" is somehow an objective, with rules we can apply to achieve consistent results. Storytelling, as any storyteller who isn't a Hollywood huckster will tell you, is not science.

    I wanted The Snowfield to illustrate how we could build an emergent, emotionally rich, player-driven narrative experience from the ground up by relying solely on art, audio, and character A.I. (refined via player testing), instead of trying to identify formulaic notions of "good" storytelling and work them into the game beforehand. I wanted the team to develop an open simulation with strong emotional hooks and minimal dissonance and see if the user's imagination would do the rest.


    What Went Right

    1. Ambition

    One of the best things we did with The Snowfield was to make it a fully 3D game with a semi-traditional over-the-shoulder camera. This greatly contributed to the atmosphere and general emotional impact of the game, which ended up being our saving graces when other aspects did not turn out as planned.

    Early in development this was considered a risky move. Traditionally, the GAMBIT summer program advises interns to avoid 3D whenever possible, since it adds an extra layer of complexity for things like collision, modeling, texturing, and lighting. Using 3D also can put interns in the frame of mind that they might achieve results similar to a triple-A game in terms of detail, which means they're likely to spend more time wrangling technology than actually cutting their teeth on game design.

    We talked a lot about whether to go 2D or 3D during the first few weeks of brainstorming and paper prototyping. As the product owner, I was against 3D. I felt our narrative goals might only be achievable in the eight-week time frame if we used a highly abstract art style, like pixel art or stick figures. I thought this would allow us time to generate enough assets to sufficiently populate the large number of narrative events and situations we were planning.

    In retrospect, I may have been right-our decision to go full 3D did have a major impact on our ability to implement a large number of narrative content. However, I'm not convinced the game would have gotten the recognition it did had it been a highly complex pixel art game. The immediacy of the horrific audiovisual design of a WWI battlefield was largely cited as the defining feature of the game, and the main source of its emotional impact.

    Also, by making The Snowfield a 3D game with an over-the-shoulder perspective, it became a sort of commentary on traditional triple-A war games like Gears of War, which sport similar conventions. Since we wanted to depict war in a way that counterpoints most commercial games, speaking their language had certain advantages; The Snowfield could play upon the player's expectations to drive additional narrative impact.

    2. Unity

    Choosing Unity was absolutely the right move for this project. Even though it had certain drawbacks and caused certain problems, without it we would not have gotten the results we did in 3D within the same time frame.

    Unity's easy-to-use interface and readily available tutorials allowed even people who'd never used it before to produce on-screen tests and prototypes quickly. It is very easy to create a 3D space, place objects with behaviors, set lights and collision, and go. In fact, one of the reasons the team decided to go 3D was because Unity made it so easy to take our prototypes and move the camera around to see how it would look from a variety of angles.

    Special kudos have to be given to Unity for making our sound design pipeline a snap. Because it is very easy to move sound sources around in Unity, our sound designer easily filled the world with an emotionally rich, haunting soundscape without fighting the tech or taking time away from programmers, which played a large part in making The Snowfield as effective as it was.

Comments

comments powered by Disqus