GAME JOBS
Latest Jobs
spacer View All     Post a Job     RSS spacer
 
June 7, 2013
 
Hammer & Chisel
Senior Gameplay Developer
 
International Game Technology
Lead Artist
 
Gameloft - New York
Senior Programmer
 
LeapFrog
Game Designer
 
YAGER Development
Senior Game Systems Designer (f/m)
 
RealTime Immersive, Inc.
Animation Software Engineer
spacer
Blogs

  The Undoing of Canabalt
by Rob Solomon on 06/29/10 03:00:00 pm   Featured Blogs
6 comments Share on Twitter Share on Facebook RSS
 
 
The following blog was, unless otherwise noted, independently written by a member of Gamasutra's game development community. The thoughts and opinions expressed here are not necessarily those of Gamasutra or its parent company.

Want to write your own blog post on Gamasutra? It's easy! Click here to get started. Your post could be featured on Gamasutra's home page, right alongside our award-winning articles and news stories.
 

(Before reading, if you haven't played Canabalt, please head over to http://www.canabalt.com and give it a spin. It will only take a few minutes. If you haven't played it in a while, it might be helpful to play it again as well. Thanks!) 

There aren't many simple games left to be made that can purport to be perfect, but Canabalt is one of them. The concept is so simple, it almost has certainly been done before, but never this well.

How? Many point to the art style, and indeed it perfectly illustrates the power of abstraction in today's graphically saturated age. It is doubtful we would identify as much with our nameless protagonist if we could see any sort of facial expression, and we readily accept the limitations of the color palette and resolution. This allows our imaginations to fill in the details, much in the same way poetry does. Indeed, the fluttering birds are all that is needed to evoke a rooftop.

However, for me personally, it is the audio design that really shines for this game. Despite the bombast of the excellent music, it is an exercise in the power of a subtle approach. You have to listen carefully to hear the sound that signals an oncoming drop of machinery on the next platform you jump to. You can only hear the footsteps of the protagonist on the metal railings when the music is quiet. The sound of glass breaking is perfect, and the follow up sounds as the glass particles hit the ground in front of you evokes the feeling of jumping through glass in a way that the greatest graphical representation can't touch.

A beautiful consequence of this subtle sound landscape is that it forces the player to pay closer attention to the game itself to hear everything it has to offer. Immersion happens almost immediately, and while the platforms vary little besides the odd window and the occasional drop of debris from the sky, it scales beautifully in difficulty as the speed of the player increases.

But why do we pay such close attention? We wouldn't care about the sound design or the visuals if there wasn't some unifying factor linking them together. Indeed, the story is non-existent. We can only glean from the visuals and the explosions that there has been some sort of alien invasion, and the city our hero is in is being destroyed, so he is running for his life... making his "daring escape". Of course, every game ends with the hero dying. It manages a strange melancholy.

Most survival games make light of the hero's plight, making death a humorous experience, or making the enemies cartoons so that we don't feel bad about blowing them away. Yet Canabalt does neither, and despite its reduced visual resolution it does not have cheesy 8-bit sounds to match. Instead, the sounds are very realistic, and for a casual game, Canabalt is extremely serious. All of the elements of the game work together to this goal, down to the game over screens which usually end with "hitting a wall and tumbling to your death.", or the more haunting variant "before turning into a fine white mist."

All of these elements work together perfectly, with nothing superfluous. Like the core gameplay of Pac-Man, there is little to be added that will enhance the gameplay.

The original iPhone port of Canabalt was a straight conversion of the original game, only with a slightly different resolution. Despite the $2.99 price tag, I felt it was certainly worth it to support such a game, at least until the first update for the game came through.

In the update, the fundamental gameplay was changed with the addition of billboard platforms, platforms that would crumble more easily and an additional obstacle where a huge piece of debris would fall from the sky and take out most of the platform you were about to jump on. There were also more varied death messages depending on how you died.

Normally added gameplay elements are welcomed, but in this case, they not only did these changes break the gameplay, they violated the fragile diegesis that held the game world together.

First, the billboards. Certainly, the billboards do not change any of the fundamental gameplay mechanics, but what they do manage is to obscure the teeming life that surrounds the rest of the game and makes it a living world. The buildings you crash into via the windows already serve this purpose, which makes this addition pointless, not to mention ugly.

Second, the crumbling buildings now occur on very long stretches, as opposed to before where they were usually very short. This in and of itself is not a problem, but in this addition the designers made it possible for jumps that normally would have succeeded to now fail. Several of my own playthroughs resulted in my failure not through my lack of skill as a player, but because the randomly generated backgrounds were aligned against me. I'm fairly certain that algorithmically that these jumps are indeed possible if you hit them perfectly (as evidenced by the high scores posted), but hitting perfect jumps was never what this game excelled at. It was the experience of the player surrounding those jumps.

This brings me to the most egregious of gameplay additions, which is the platform that vaporizes as you are about to land on it, save for a small section of the platform. Again, if my jump did not happen to land perfectly with a perfect jump following it, I would almost always die.

Now, we could perhaps take these random events in the same way we accept that the window is sometimes positioned in a way to make a leap into it extremely difficult, and reflect on these events as some sort of meditation on how death can occur to us at any moment, and that we all need to rely on luck. Well, I won't.

Take Pac-Man, a game that has notoriously suffered due to added "features". Despite its difficulty, it is above all things exceedingly fair. If you die, it is due to your own lack of knowledge about how to traverse the maze. Before this update Canabalt was fair. After the update, it wasn't. Pure and simple.

Despite these misguided additions to gameplay, they at the very least do not break the mood of the game. This purpose is instead served by the new death sentences, which obliterate immersion with such phrases as "somehow hitting the edge of a billboard."

The absolute worst thing you can do within a game is point out that it is a game. The player is already aware of this, but there is a willing suspension of disbelief that happens much in the same way that we are able to enjoy the most ridiculous movie. We accept the created world for what it is, as long as the parameters of the world are not violated.

With these simple, seemingly harmless phrases, the world of Canabalt is destroyed. I am completely aware that it is a game. I am also aware that the game is mocking me for managing to die in an unlikely way. I'm already frustrated as a player for having failed, but it adds nothing to have it followed up by a message saying "before missing another window." Before, I had tumbled to my death in a daring escape. Now, I'm just a crappy video game player.

 

How to mock the player who doesn't understand your game they just paid $3 for. Good job guys. 

 

In a comedic game (or film), we'll sometimes welcome these asides, but Canabalt is not a comedy, except maybe to the programmers who can't believe that something coded so quickly could be so successful. It's honestly a shame.

(P.S. To the devs, you'd make me and I'm guessing a lot of people happy if you added a "classic mode" to the iPhone port, because these days I'm only playing the Flash version.)

 

 

 
 
Comments

Chad Wagner
profile image
This immediately made me think of Tim Burton. I watched "Sleepy Hollow," and loved the cool visual appearance, the great atmosphere, and the uniformly 'heightened reality.'



Then I saw the extra features on the DVD. And listening to Mr. Burton, and company, deconstruct the film and laughing at their own goofiness, really changed my perception of the movie. He specifically talked how he was attempting to produce the hilarious, overwrought urgency of the Hammer films of old (with a wry grin). Seeing this behind the scenes perspective seriously added a "nod & wink" to the film for me...and as a consequence removed a bit of the fantasy magic.



I certainly know that Tim Burton is this kind of person -- and that he and his team have fun making movies this way, and don't begrudge it. However, perhaps in the editing room, he manages to produce films that cleverly don't call attention to their construction, and really weave a fine tapestry of fantasy that you can accept on its own grounds. I was a little disappointed to see him disrespecting his own vision this time.



It sounds like you felt a little the same with Canabalt.

Robert Marney
profile image
The crashing ships issue could be averted with an earlier audio cue that allows you to jump to the center of the next building, instead of hoping it doesn't disappear on you. I see no problems with billboards, as they break up the monotony of buildings + background. The near-impossible crumbling building jumps are a source of frustration, but again this could easily be tweaked with a slightly longer delay before crumbling.

Ernest Adams
profile image
I'm willing to accept the argument that Canabalt is the perfect, clean platform game, stripped to its bare essentials while still retaining an appealing visual and auditory style. Kudos to the designers for finding the soul of platforming.



Unfortunately, it uses the learn-by-dying progression model. That's outdated. We should have abandoned that model forever once we got rid of coin-op. (The main reason for learn-by-dying in the first place was to generate "coin drop," as the arcade people call it.)



Also, I loathe platformers. I hate jumping the way Mario hates turtles. So while I can respect the game's achievements as a designer, I won't be playing it as a player.



Now, if someone can make a platform game that doesn't involve learn-by-dying and makes jumping fun rather than an opportunity to fail, then we'll be talking.

Chan Chun Phang
profile image
@Ernest Adams



1) I disagree that "learn by dying" was to generate coin drop, though it was a useful convenience. Learn-by-dying has been present from the very first game.

2) Learn-by-dying is not in itself flawed. There are games which plays with the whole dying concept to make it not as negative as one might assume. One of the better executed examples is Braid (using a time factor). Also, the Thomas Edison quote: "I have not failed, I've just found 10,000 ways that won't work."

3) Making a game with no failure condition is not exactly simple. Even if there is no implied failure condition, the player may still interpret it as one, and is likely to do so regardless.



But if you insist: search flyguy. Granted it's not jumping, but flying is similar.

Lorenzo Gatti
profile image
Learning by dying, in such a simple game, is learning to die less by jumping better: perfectly aligned with the purpose of the game (survival) and its core mechanic (jumping).



Since there is no progression, only longer randomized games rewarding increasing skill, the annoyance of traditional instances of learning by dying in a fixed content setting (e.g. not opening instant death doors or prescient dodging of sudden attacks) is avoided: compare learning to jump in Canabalt with the difficult memorization in the equally simple Dragon's Lair.

Lorenzo Gatti
profile image
A complex platform game like the Super Mario Bros. family, with puzzling maps, planning, tactics, fancy moves, etc. is quite different from a pure action game like Canabalt.



The only thing in common is platform jumping, which is not enough to define a genre.


none
 
Comment:
 




 
UBM Tech