Technical Discussion February 5 2009

SFU Burnaby 1:30pm - 3:20pm

Warm Up Topic

How would you program Super Mario Bros.?

Super Mario Bros. was most likely programmed in assembly. Today we would write it in higher level languages such as C, C++ or Java. The game is quite simple to program because there are only a handful of systems that need to be programmed. These are assumptions we made as we do not have the source code and have not studied it in depth. The discussion below are not listed in any particular order.

  • Level graphics and collision detection are stored separately
  • Collision information is stored in an integer array.
  • Event/trigger system for warp pipes, coin blocks, hidden blocks etc.
  • Animation is done over tile data so all images on the screen sharing those tiles will animate the same way. This is not a bad thing as it adds unity to the game world.
  • AI
    • Only a few basic actions: move, attack (including homing), walk, jump, stand
    • Lakitu spikes walk in the direction the player was
  • One tileset per level and many levels reused tilesets. ie. cloud = bush
  • Only so many enemies allowed in a level design due to sprite limit considerations.
  • Coin blocks are background tiles until hit, then they become sprites to animate them.

Looking at triggers in depth

  • Triggers are persistent in the level, no new ones are added.
  • Used for secrets (mushrooms, coins blocks (must run out), blocks, warps, flag, axe etc.)
  • Once you hit a coin and it runs out of coins, change it to a different type of block. This means there are different type of collisions.
  • Since you can't move backwards, you only need to check for events on the current screen.
  • Must also check triggers when pressing down (warp pipes)

Topic 1: What do you think about the gaming industry adopting a development model like the film industry.

This topic stemmed from the articleOpinion: How We Can Reshape The Game Industry. The film industry is designed around a contract basis where people from all areas will come together to make a movie and then go their separate ways. If this was adopted in the gaming industry it would promote individual designers to hold onto IP and specify it on a per contract basis.

  • Unlike movies, games may need bug fixes after it is shipped so teams must stay together longer or a minimal team must remain.
  • Downloadable content (DLC) is becoming more popular and this means the original team must stay together longer to create it.

Topic 2: What about releasing episodic games similar to the manga industry in Japan?

Manga is usually serialized in magazines. Readers will read one chapter a week while the authors are writing/drawing it. In order to get published, a pilot chapter (similar to TV) is frequently published and if there is interest it is serialized. Can this model be adopted by the gaming industry and a new market created?

  • Different from downloadable content which is just additional content.
  • Break the game into episodes (chapters) that logically follow from each other.
  • Video games are all about choice and non-linear gameplay, however in order for episodic games to work they must be linear. Instead these games could have gating at each episode so you can make choices but they all lead to the same conclusion or multiple paths/scenarios that are developed.
  • Makes games cheaper for consumers by breaking it into parts
  • Consumers can stop buying new content when they lose interest in the game. This means developers will need to work hard at making the whole game interesting.
  • Must convince the consumer that it is worthwhile to buy.
  • What happens if a company goes bankrupt? It is difficult to find missing chapters.
  • Let the consumer jump into the series at any chapter.
  • Chapters could be as cheap as a few dollars.
  • Games would have to be story driven
  • Once a month may be a good development cycle
  • There will have to be a minimum maintenance crew for previous chapters to fix bugs.
  • Could give episode 0, the pilot, away for free.


  • Less development time to get a game on the market
  • Digital distribution is fast
  • Cheap to sample marketability of the game
  • How would you advertise it?
    • Internet, magazines — the usual ways
    • Disadvantage: hard to convince people of business model
  • Constant advertising because each
  • Promotes individual developers such as writers, programmers, designers and artists.


  • Need a game engine that is highly customizable
  • How much play time can you generate in a new chapter without people losing interest?
  • People may not want to wait for next chapters
  • Will people find the process of buying episodes confusing?

Games that already have episodic content

Members Present

  • Eric Raue
  • Colin Hume
  • Long Tran
  • Jon Derrick
Add a New Comment
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License