Get in the Game: Agile Lessons From Video Game Developers

More than twelve years ago I left a defense industry software engineer job to become a video game programmer. The higher level of technology and quicker pace of product development in games—about ten times greater, in terms of lines of code—shocked me. The game industry attracted some of the best, most highly motivated programmers, and the latest software engineering metaphors were discussed with a passion. Yet for all the enthusiasm, there was a nearly complete lack of—and a high level of disdain for—process in project management.

In just three decades, video games have gone from a niche toy industry making Pong boxes to a branch of the entertainment industry larger than Hollywood’s box office. The change has increased development teams and budgets exponentially.

Both of these factors have led to massive industry problems. One is the sometimes egregious abuse of people who work in games; crunch times lasting a year are not uncommon and have resulted in lawsuits for overtime pay. Project delays and cost overruns are also common.

In response to these challenges, we at High Moon Studios have taken lessons from other industries and adopted agile methodologies to stay competitive and deliver on time and within budget. We've had to make changes to account for unique development environment issues, but they haven't broken the vision of what it means to be agile. We focus on communication and iterating on an emerging product from the start, and we can react to changes from many different directions to make the product better.

Background of Video Game Development
When I entered the video game industry, it was less than twenty years old. In the late 1970s and early 1980s, it had experienced significant mainstream growth. Coin-op arcades—and later consoles that re-created the experience at home—were raking in huge amounts of money.

During this time, a team of one to three people could create a game in about three months. The hardware consisted of 8-bit processors running at a few hundred kHz, 16 K of memory, and low-resolution (256 color) 2-D graphics. Individual heroics were common because the consumer demanded that the hardware be pushed to its limits, though the resources were limited. Often the game art was created by the programmers! The main obstacle for development was that mass producing arcade cabinets—or cartridges for home consoles—was far more expensive than creating the games. As a result, development was iterative and incremental. Each game had to show value improvement at frequent intervals or risk cancellation. For each game released, a dozen games might be cancelled, yet the rewards for making a hit game were substantial.

The high profits and exploding demand created problems, namely a relaxing of quality control to allow more game releases. The result in the mid-1980s was the first and to date largest crash for the game industry. Companies like Atari lost hundreds of millions of dollars, and landfills received tens of thousands of cartridges that no one wanted to buy.

Innovation in home consoles helped the market recover quickly, and sales continued to grow with more control by the console hardware manufacturers. Nintendo was a major catalyst in the industry’s recovery, introducing its hugely successful home console and ensuring stringent quality controls with a seal of quality on all software released for the system.

The major pressure for game development soon became the growth of hardware power. Within a decade after the crash, consoles had processors running at more than 20 MHz , 4 MB of RAM, and NTSC-resolution 3-D graphics capability. Skilled artists were hired to create highly realistic art, and teams of programmers took on specialized