U (2-0-7)
Fall 2002
Thursdays 3-5pm
E15-335 (MIT Media Lab, 20 Ames St )
Jon Orwant (orwant@media.mit.edu)
Bruce Blumberg (bruce@media.mit.edu)
Office hours: by appointment
What makes a game entertaining? How hard should a game be? How can a game guide players through a path from novice to expert? How should a game balance risk vs. reward? How can a game appeal to players with different reasons for playing? What behaviors should computer opponents exhibit? What common design flaws do novice game designers make?
This course will show you how to design games: puzzles, board games, text adventures, role-playing games, strategy games, first-person shooters, gambling games, sports games, simulations, and other video games. We’ll design variations on classic games and occasionally design entirely new types of games. While video/arcade/console games will be covered, this class will not show you how to create your own Grand Theft Auto III or Metal Gear Solid singlehandedly – such games require millions of dollars and dozens of man-years. Game design is not game development; just as 6.001 teaches you how to think about programming so that you can learn any programming language, this class will show you how to think about games, concentrating on the timeless principles underlying successful games, rather than teaching details of game programming environments that will be obsolete in a few years.
You'll be performing critical analyses of games – both currently existing games and games your classmates design. Statements such as “This game bores me”, “This game is too easy”, and “I hate fighting games” are opinions, not analysis. Who does the game appeal to, why, and how could it be improved?
There is also a heavy synthesis component to this class: you’ll be designing several games throughout the course of the class, and completely implementing two games. Special emphasis will be placed on designing games that exhibit complex behaviors or model sophisticated topics.
18.01, 8.01, and programming experience. There will also be some linear algebra involved (simple matrix math) and some probability theory, but background will be provided in class. Programming experience is the most important prerequisite – you should be comfortable creating medium-scale graphical programs with a minimum of supervision.
Game design relies upon mathematics, psychology, physics, art, music, AI, user interface, graphics, creative writing, and a good dose of programming skill. You don’t have to be an expert in all of these areas to succeed in this course, but you do have to be competent in a few of them and undaunted by most of them. I’m here to help.
Most lecture notes will be available as hardcopy at the beginning of class, so you won’t have to take diligent notes. Focus on listening and participating in class.
Handouts will be available on the course web site, http://orwant.com/mas490. If handouts aren't available online, they'll be available outside Bruce Blumberg’s office in One Cambridge Center (on Main St near the Kendall Sq. T stop) during business hours.
There are no quizzes, midterm, or final. Your grade will be based on a combination of problem sets, two programming projects, and participation in class. If I had to attach percentages to these components, they’d probably be 45%, 45%, and 10%, respectively. But the actual algorithm will be more complex: if you’re the quiet sort and don’t want to speak up in class, do a good job on your problem sets and projects and you’ll do fine. If you’re not a great programmer, plan on being thoughtful and vocal in class. If you ignore the problem sets, your games had better be mind-blowing.
Halfway through the semester, I’ll tell you what your grade would be if the class immediately ended, and why. There is no curve. I’ll be delighted if everyone earns an A, and mournful if everyone gets a D.
Submit your problem sets as plain text email to orwant at media.mit.edu. (There may be a few problem sets where I ask you to draw something; in those cases you're welcome to hand draw on paper and turn that in.)
There will be two programming projects. For the first, you’ll create your own text adventure game. For the second (and larger) project, you’ll create a video game, strategy game, or simulation.
Deadlines are critical in game development, and I intend to be a stickler about them. You can, however, choose one problem set to turn in one week late. (This cannot be applied to the final project deadlines!) If a delay is unavoidable due to an emergency, let me know as soon as possible and we'll work something out.
You may work together with other people, so long as you attribute your collaboration — in which case the standards will be higher, since two heads are better than one. That said, don’t plagiarise. Don’t even think about it. I’ve written software that successfully identifies plagiarists of both prose and code. I've kept people from getting jobs and gotten publishers to take books out of print due to plagiarism. I'll be using your problem sets and code as training data for my software, and won't hesitate to notify the MIT Committee on Discipline if it finds that you've duplicated someone else's work without attribution.