February 6th, 2012 by Elliott Mitchell
Now and then, as a professional 3D technical artist and game designer, I find it’s helpful to step out of my usual routine and make a game over a weekend. Why? Because it keeps life fresh and exciting while providing a rare sense of instant gratification in the crazy world of video game development. Making a video game over a weekend isn’t easy for one person alone. For this, Global Game Jam was created.
This year’s Global Game Jam was held last January 27 – 29, 2012. I registered with was the Singapore-MIT GAMBIT Game Lab, in Cambridge, Massachusetts. Here is the lowdown of my experience.
The Global Game Jam (GGJ) is an annual International Game Developer Association (IGDA) game creation event. The event unites people from across the globe to make games in under 48 hours. Anyone is welcome to participate in the game jam. Jammers range from industry professionals to hobbyists and students. The primary framework is that under common constraints, each team completes a game, without preconceived ideas or preformed teams, in under 48 hours. This is intended to encourage creativity, experimentation and collaboration resulting in small but innovative games. To support this endeavor, schools, businesses and organizations volunteer to serve as official host sites. Several prominent sponsors such as Loot Drop, Autodesk, Microsoft and Brass Monkey also helped foot the bill.
HOW IT WENT DOWN
GGJ site facilitators kicked off the Jam with a pre-recorded video from the IGDA website titled How to Build A Game in Less Than 48 Hours. The speakers in the video were Gordon Bellamy, the Executive Director of the IGDA, John Romero (Quake) and Brenda Brathwaite (Wizardry) both co-founders of Loot Drop, Gonzalo Frasca (Ludology.org) the co-founder of Powerful Robot Games and Will Wright (The Simms) co-founder of Maxis. They speakers all gave excellent advice on creativity, leadership, scope and collaboration within a game jam.
Global Constraint -
Our primary constraint was revealed after the keynote video. It was an image of a snake eating it’s own tail. The snake represented Ouroboros, a Greek mythological immortal. Variations of the symbol span across time and space from the modern day back to antiquity. The snake, or dragon in some instances, while eating it’s own tail has made appearances in ancient Egypt, Greece, India, Mexico, West Africa, Europe, South America and elsewhere under a host of names. It’s meaning can be interpreted as opposites merging in an a unifying act of cyclical creation and destruction, immortal for eternity. To alchemists the Ouroboros symbolized the Philosopher’s Stone.
Group Brainstorming –
After the keynote game jammers arbitrarily split into 5 or 6 groups of 11 or so and went into different labs to brainstorm Ouroboros game pitches. After an amusing ricochet of thoughts, references, revisions, personalities and passions each room crafted 6 pitches which were mostly within the scope of the 48 hour Game Jam.
Pitch and Choose -
When the groups reassembled into the main room it was time to pitch.
- Pitches needed to be under a minute
- Title is 3 words or less
- Theme related to the Ouroboros
- The person pitching a game did not necessarily need to be on that potential team
There were about 30 or so pitches, after which each jammer had to choose a role on a game / team that appealed to them. Each Jammer had a single piece of colored coded paper with their name, skill level and intended role.
- Game Design
Games with too many team members were pruned and others lacking members for roles such as programmer were either augmented or eliminated. Eventually semi-balanced teams of 4-6 members were formed around the 11 most popular pitches.
My team decided to develop our game for the Commodore 64 computer using Ethan Fenn’s Comma8 framework. We thought the game narrative and technology married well.
Time to Jam -
Post team formation, clusters of lab space were claimed. Even though most of us also brought our personal laptops, the labs were stocked with sweet dual boot Windows 7 & OS X systems with cinema displays. The lab computers were pre-installed with industry standard software such as Unity3d, Maya, Photoshop… We were also provided peripherals such as stylus tablets and keyboards. Ironically, I was most excited by the real world prototyping materials like blocks and graph paper which were also provided by or host.
First Things First –
After claiming a lab with another awesome team we immediately setup:
- Version control (SVN)
- Installed custom tools for Comma8 (Python, Java, Spite Pad, Tiles and more)
- Confirmed the initial scope of the game
- Set up collaborative project management system with a team Google Group and Google Doc
Cut That Out –
We needed to refine the scope once we were all aware of all the technical limitations such as:
- Commodore 64 from 1982 is old
- 64 kb of RAM for system not much
- 8 bit
- Programed in Assembly Language
- 300 X 200 pixels
- 16 pre-determined crappy colors
- 3 Oscillators
- Rectangular pixels
- Screen Space
- Developing in emulation on a network
- Loading and testing a playable on legacy Commodore 64 hardware
- Less than 48 hours to get it all working
- Our scope was too big, too many levels
- Other factors causing us to consider limiting the scope further included:
- None of us had made games for C 64 before
- Comma8 is an experimental engine that was untested in a game jam situation and is currently in development by Ethan
- Tools such as Sprite Pad and Tiles are very archaic and limiting apps for art creation
- Build process would do strange things to art after build time which required constant iteration
Rapid Iterative Prototyping -
Physical prototyping was employed to reduce the scope before we went too far down any rabbit holes. We used the following materials to prototype:
- Glass white board
- Masking tape on the walls
- Paper notes tacked to the walls
- Graph paper
- Wooden blocks
Results of Physical Prototyping-
- Cut down scope from 9 levels to 5 levels as the minimum to carry the Ouroboros circular theme of our narrative far enough
- Nailed the key mechanics
- Refined the narrative
- Determined scale and placement of graphical elements
- Limited overall scope
Naturally we ran into design roadblocks and need to revise and adapt a few times. Physical prototyping once again sped up that process and move us along to completion.
We enlisted a few play testers on the second night and final hours of the game jam to help us gauge the following:
- Comprehension of the narrative
- Recognition of the lo-res art assets
- Overall player experiences
- Feelings about the game
We did wind up having to revise the art, level design and narrative slightly to reach a better balance and game after play testing.
1.5 hours before the game jam was to end it was pencilsdown. Time to upload to the IDGA Global Game Jam website, any other host servers and on to the site presentation computer. Out of the total 48 hours allotted to the game jam, we
only had about 25 working lab hours. Much time was spent on logistics like the keynote video, brainstorming, pitching, uploading and presenting. Our site also was only open from 9 am to midnight so there was not 24 hour access. With 25 hours of lab time all 11 games at my site were uploaded and ready for presentation.
The best part ever! The presentations were so exciting. Many of the jammers were so focused on their work they were not aware of what other teams were up to. One by one teams went up and presented their games in whatever the current game state was at the deadline.
Most were pretty innovative, experimental and funny. Titles such as The Ouroboros Hangover and Hoop Snake had the jammers in stitches. Fire farting dragons, Hoop Snakes, drunk Ouroboros and so on were big hits. Unity, HTML 5, Flash, Flex, XNA, Comma8 and Flixel were used to create the great games in under 48 hours.
Take Aways -
My teammates and I consider the game we made, Walking Backwards, to be a success. We accomplished our goals:
- Experimental game
- A compelling narrative
- Awesome audio composition
- Most functionality we wanted we achieved
- Runs on an original Commodore 64 with Joysticks
- Can be played with a Java emulator
- Got to work together under pressure and have a blast
Would have liked-
- Avatar to animate properly (we had bi-directional sprites made but not implemented)
- More audio for sound effects
The final take away I had, besides feeling simultaneously exhilarated and exhausted, is how essential networking at the game jam is for greater success. Beyond just meeting new people, networking at the jam made or broke some games. Some teams didn’t take time to walk around and talk to other teams. In one instance, a team didn’t figure out a essential ghost mechanic by the end of the jam. They realized at presentation time another team had implemented the same mechanic they failed to nail down in the same engine. Networking also provided mutual feedback, play testing, critique, advise, friendships and rounds of beer after the event ended. Many of the jammers now have a better sense of each other’s strengths and weaknesses, their performance under stress, their abilities to collaborate, lead and follow.
I, for one, will be a life long game jammer, ready to collaborate while pushing into both familiar and new territories of game development with various teams, themes and dreams.
Follow this link to see all the games created at my site hosted by the Singapore-MIT GAMBIT Game Labs
Technical Director- Infrared5
Twitter: @ Mrt3d