GDC 2011:Wed.05 – The Failure WorkshopPosted April 27th, 2011 by Warren Blyth
Chris Hecker (definition six, inc.), Kyle Gabler (2D Boy), Matthew Wegner (Flashbang Studios), Kyle Gray (Tomorrow Corporation), George Fan (PopCap Games) and Brad Wardell (Stardock) : 3pm -4pm
(official GDC brief) (note: I was 5 minutes late)
Each speaker described a game they’d made (or prototyped) which they considered a complete failure. They went into detail about why they exactly they thought it failed.
Some very interesting stories about their quick prototypes (much like the exercise I’d quickly whip up for various instructors at OSU’s Ecampus), and a strong theme around “get to prototype asap, and start testing it.”
1) Kyle gabler (known for: world of goo)
failed project: “Robot and the cities”
Robots are a metaphor for the destructive passage of time. So he made it a musical. … (w: uh oh, I smell hipster humor in this talk)… Game design document was a napkin sketch (“take a sip of this”). … the game was boring, underneath all the flash. Had to break it down to a silent rectangle prototype to discover this.
Lesson: No amount of “theming” will save a bad idea.
2) George fan (pop cap) (known for: plants vs. Zombies)
Shows drawings … Robot dragon … his failed project: “Cat mouse foosball”
The idea is to keep the cats away from the mice, as they progress/pour through a simple maze. He thought it would have the appeal of bomber man (staying away from dangerous things, easy on small scale. Hard to use peripheral vision)
But it turned out there was simply too much going on. Brain explodes. … Got depressed. … Had to recognize the difference between illustration and game design.
(Easy to come up with an idea and translate to paper. Games are more like complicated machine with moving parts. You can’t keep track of it all in your brain or paper.)
Lesson: You can only get into mindset of player by playing it yourself.
Now, he never writes more than a page of notes before diving into prototype.
4) Mathew Wegner (known for : Off-Road Velociraptor Safari)
He started “Blurst”, and learned that “putting out free game every 8 weeks was horrible idea.”
… Traffic between releases was the same as during release. There was no spike. Plus, 8 weeks was too small to make a WorldOfGoo level project, and too long for a Canabalt scale project.
Raptor safari still drives wayyyy more traffic than anything else. So they decided to pursue an HD version for xbla and steam (which is the failed project).
Using that name (“HD”) set expectations way too high for 3 man team.
They got stuck in “parking lot science” (looking for your keys under the street lights, where the best light is. He equates this to staying in your comfort zone)
They “…Confused comfort with happiness” … Solution was to “bracket” the problem between extremes. … End Lesson: Don’t underestimate “Gamifying games”?
(Don’t forget to design for players)
3) Brad Wardell (CEO of star dock?) (known for: galactic civilizations 2, which made 12 times cost to make)
After their success, they scaled up for bigger game, which was their failure: “Elemental: war of magic.” (the only failure here that was actually released as a product).
But they didn’t set up new management structure. No project manager. “So no one owned the game.”
Trying to jump in and cover problems with coding (like taking on a second job, 40 hours a week extra) – led to total lack of objectivity.
Lesson: What works for team of 6 to 7 falls apart when growing beyond that.
Need project manager. Single line to QA
4) Chris Hecker (known for: spy party)
Failed Project: Rock climbing game.
He was totally overwhelmed with distractions, for years, while trying to develop it.
Three specific problems: Technology ratholing, non game distractions, lack of ass-in-chair programming.
Address this by having playable game at start.
Joystiq’s Summary (with videos!)