MAX 2005: Flash Game Development (2:45pm, 18 October)

This seminar was presented by Joshua Hirsch, Minister of Technology at Big Spaceship. Big Spaceship does a lot of work for movie sites, and Joshua showed off a couple of examples to start with, including a game for the movie Wimbledon (Someone made a movie about tennis? Man, I’ll see that after I see another movie about golf.)

Joshua stressed the importance of developing a working model of the game sans graphics, to ensure that gameplay is workable before all of the visual assets have been developed. He discussed the concept of the casual gamer — someone who likes to idly play a particular game — versus someone who comes to a game because they’re interested in a the subject matter (i.e. movie tie-in).

Joshua elaborated four basic categories that games fall into:

  • one-click: where the user basically just needs to steer the mouse around and press the button;
  • twitch/arcade: where time and coordination are major factors in gameplay;
  • puzzle: which relies on pattern-matching of one variety or another;
  • turn-based multi-player: non-real-time multiplayer interactive experiences.

Curiously, Joshue dismissed real-time multi-player as too problematic (for reasons of latency). Director users know that there are methods — or were….

He listed some personal favorites, including the HeliAttack series from Miniclip. (I see that their #1 game is the real-time Shockwave multiplayer game Robot Rage.)

Disadvantages of Flash for games: no 3D, no real-time. Advantages: yadda, yadda, yadda.

He considered several elements essential a game’s replay value:

  • it should be easy to learn;
  • it should be hard to master;
  • gameplay should be varied;
  • it should contain infinite possibilities;
  • it should have random incentives.

One simple way to increase the potential for creating a viral game is to allow users to email their high scores to friends for challenges.

He walked through the process of making a wireframed version of a games he called “Tank and Circles”, which was remarkably similar to “Asteroids.” Through the process he discussed breaking down the game into states, classes, and state transitions.

He brought up the issue of granularity in collision detection (where objects may pass through each other in-between collision checks) and the need to account for it. Havok!

Toward the end, he showed a couple of games Big Spaceship’s done for movies over the past few years, a pinball game for Starsky & Hutch, and a 2.5D streaking game for Old School.