Skip to content

300 Word Review – Canabalt (21691m)

At the heart of Canabalt is a little piece of ludo/narrative dissonance.

In the game you play a character, whom I’ll call MJ (for his resemblance to the deceased pop star), that has the bad luck to be on a rooftop jog when his city is attacked by giant robots. Players must navigate MJ from rooftop to rooftop by deciding when he should and should not jump. This can get tricky because MJ gets faster the longer he runs without interruption, making it more difficult to see what’s coming and react before it’s too late. The robot apocalypse is a dangerous time and there are several gruesome ways MJ can perish.

What’s wonderful about Canabalt is that while it has only one interaction, jumping, it still manages to provide interesting decisions for the player. For instance, while players have direct control of MJ’s jump they also have indirect control over his speed. Scattered across the rooftops are boxes and office chairs that slow MJ down whenever he collides with them, and players can check MJ’s speed by purposely running him into these obstacles. However, since the levels are procedurally generated players must be careful about slowing down too much or MJ might not be able to leap the next gap.

The strange thing about all this is that if you were playing the game by putting yourself in MJ’s shoes you would never slow yourself down to the steady jog that gives you the best chance of survival. You certainly wouldn’t crash yourself into office furniture at regular intervals. Success in Canabalt then depends on the player acting in a way opposite of that which would make the most coherent narrative.

It would be interesting to know how many people role-played Canabalt and how many went for the high score.


  1. James L wrote:

    This might be an unpopular opinion, but I don’t think that roleplaying and going for the high score are mutually exclusive.

    I’d argue that if you’re “roleplaying” Canabalt by purposefully avoiding furniture and boxes you’re simply going for the high score in a different game – even people who play the game without crashing into furniture for narrative reasons are probably trying not to die, they’re simply giving themselves extra constraints; giving the game extra rules and changing it slightly. They’re playing a mod for Canabalt where furniture kills you instead of slowing you down – but they’re still trying to play it well, and run for as long as they can.

    Tuesday, January 5, 2010 at 5:58 pm | Permalink
  2. James L wrote:

    Quick qualification of that argument: there are probably people who really are only interested in the narrative they’re creating by playing the game and really don’t care about running far or jumping over things successfully. However, I think these people are few and far between – my guess is that most people who play Canabalt do care about doing well within the restrictions they set for themselves, even if they’re not looking for the global high score.

    Tuesday, January 5, 2010 at 6:03 pm | Permalink
  3. Charles J Pratt wrote:

    “I’d argue that if you’re “roleplaying” Canabalt by purposefully avoiding furniture and boxes you’re simply going for the high score in a different game”

    This is a really great point! You’re right that the question should probably be rephrased to – “It would be interesting to know how many people played Canabalt as an RPG and how many played for a high score.”

    Tuesday, January 5, 2010 at 9:27 pm | Permalink

Post a Comment

Your email is never published nor shared.

Comment spam protected by SpamBam