Archimage Fights Diabetes with Serious Games

Some developers take games seriously. Take Archimage for example. Archimage brings over two decades of design and project management skills to the emerging genre of Serious Video Games. In 2006, Archimage started development on a pair of sci-fi adventure video games aimed at preventing obesity and type II diabetes in children. With a $9M grant from the National Institute of Diabetes and Digestive and Kidney Diseases (a division of the National Institutes of Health) and an experienced team of behavioral scientists, nutritionists, and physical activity experts at the Children’s Nutrition Research Center (CNRC) of Baylor College of Medicine, they embarked on a two-year journey to create Escape from Diab and Nanoswarm.

Escape from Diab is a third person story-driven game that centers around Deejay, an athletic youth that accidentally falls into the nightmare city Diab. After being rescued from the city’s guards by five children, Deejay learns of the evil King Etes who withholds exercise opportunities and nutritious food from his subjects as a means of social control. Deejay must help his new friends escape the evil King Etes by coaching them to better health and escape to the magical Golden City.




The follow-up title, Nanoswarm: Invasion from Inner Space, is a first-person role-playing game where the player, nicknamed Wings, is training with four other teenagers at a high-tech laboratory called the M.E.C.H.S. Lab. The story of Nanoswarm is set in the year 2030 - when there is no war, no crime, and no hunger due to a new technology called nanobots. The lead scientist, Dr. Gunderson, is on the verge of eradicating disease altogether using this amazing nanobot technology, but something goes awry. People all over the world start getting sick, including fellow team member Fred, and no one knows why. The player must save Fred and discover the origin of the disease, all while training to become a member of the world’s leading research team.




Melanie Lazarus, Director of Marketing, took time to poke her developers and answer my questions about the challenges of developing serious games. Enjoy!

How do your games stand apart from traditional video games?

The ultimate goal of Escape from Diab and Nanoswarm: Invasion from Inner Space is to change diet and physical activity behaviors in the player. We utilized the compelling aspects of video games to grab and hold children’s attention. During gameplay, the player learns skills and gains knowledge about nutrition and physical activity. Expert content is stealthily weaved into entertaining gameplay to model and teach real life. Players directly participate in video game story situations, interact with other characters, make decisions, and learn how their own results affect the characters in the game. Since children learn through planning and decision-making, they witness firsthand cause and effect relationships. The player can experiment with concepts such as meal balancing and watch role models face their own struggles.

How do you use the games to encourage real world behavior?

After the completion of each episode, the player must set a diet or physical activity goal that he/she must achieve in the “real world.” The player is locked out of the game for a specific period between episodes in order to give him/her time to achieve that goal. Once the game is unlocked, the player reports back to the game how successful they were in meeting their real-world goals. Progression and the ultimate outcome of the entire game is dependent upon how many nutrition and physical activity goals the player meets in real life.

What was your development process like?

Our experience was a bit unusual because we worked with researchers at the Children’s Nutrition Research Center (CNRC) during the entire development process. A story had already been written for both games, and the animated cut scenes and blue screen video were complete by the time our game designers stepped in. Nanoswarm contained blue screen live acting versus Diab, which was all animation. Thus, we had character restrictions during the development of Nanoswarm that made it slightly more cumbersome. We basically had to fit gameplay into the existing storyline. Usually it’s the other way around.

There were a lot of standards we had to meet because these are serious games, not just entertainment titles. Not only did the programming, gamplay, and art assets need to be up to par, but we had to make sure our expert content was integrated correctly and in a way that kids would understand. Weaving a health message into a large game is not easy. Everything was first created in-house as a dirty prototype, tested to make sure it worked, and then sent to Baylor College of Medicine for focus group testing to make sure we accomplished what was needed from a behavioral standpoint. We had a lot of iterations of each game component.

In terms of how Archimage made prototypes of the game, each iteration was mapped out by our Lead Artist/Designer. The map was distributed to the entire development team to think about for about a week. Everyone looked for inspiration and ideas that they felt might be a good fit. A weekly meeting was held in which an initial idea was presented, and then the floor was opened to everyone in the office to give their opinions. The leadership team made the final decision on implementation.

The final games were part of a large clinical trial that was completed this past spring. Data from those trials is being analyzed by researchers as we speak.



Food, exercise, and other lifestyle choices players make in the real world affect the ultimate outcome in the games.

Can you walk us through this complex decision making process?

Let’s take the example of the Escape from Diab mini-game: “What’s A Fruit.” In episode 1, the player watches a cut scene that ends in our hero, Deejay, sayings he’s going to enter a warehouse. The next cut scene picks up with a fight inside the warehouse. Since these cut scenes were predetermined before we stepped in as game designers, we knew the game environment had to be in a warehouse. We also knew that the researchers at Baylor College of Medicine wanted the player to learn about fruit in episode 1 - the difference between foods that count as a serving of fruit (ex: grapes, strawberries, dried tropical fruit) versus those that may contain a small amount of fruit or taste like fruit but don’t actually count as a serving of fruit (ex: grape jelly, strawberry toaster pastry, pineapple upside-down cake).

Because of the environment and subject, it seemed logical to have a free-roaming warehouse in which Deejay has to choose between boxes of food. Some boxes contained fruit, some did not. The robot enemies that impede Deejay from his main objective were already existing in the plot, and we used bottles of water as energy boosters for our player, which was in line with our existing healthy message. And that's how we created a prototype for this mini-game.

How long did it take you to develop both games?

Our development timeline was also different from traditional game design. The core team consisted of 10 people, but when you include everyone that was involved, from the filmmakers for Nanoswarm to the composers that created custom soundtracks, it took over 100 people to make these games happen. Actual game development for both games took a little over two years, but the process of testing and getting these games ready for research at Baylor College of Medicine took a total of four years.

What software/tools did you use to create both games?

Programming was accomplished with MS Visual Studio C++, Torsion, and Torque Game Engine Advanced. We originally began work on Diab and Nanoswarm using TGE. Although lacking some of the more advanced features, we found it to be very stable and easy to work with. About six months into development, however, we made the switch to TGEA because we knew we had another year and a half to go in the development process with additional time allotted for research. We didn’t want to be behind after research was completed and the games were ready for the public to play.

Art and animation was accomplished with 3D Studio Max and Photoshop. We chose 3D Studio Max over something like Blender because Archimage was the first architecture firm to use 3D Studio. Our principals were even alpha and beta testers for its first release. This made 3D Studio Max a logical choice.



The core development team consisted of 10 people, but when you include everyone that was involved, from the filmmakers for Nanoswarm to the composers that created custom soundtracks, it took over 100 people to make Escape from Diab and Nanoswarm happen.

What technical hurdles did you overcome in creating these games?

The biggest challenge for us was getting 60 minutes of commercial quality animation to compress so it would play well. Our target hardware specs were not great, so we couldn’t take up a lot of CPU. Our programmers used Theora and wrote initial support for TGE. Once we upgraded to TGEA, we got it almost 100% working, but there is still a slight timing issue that only we can detect.

Our other major hurdle to overcome was the massive amounts of game mechanics that were integrated into Escape from Diab and Nanoswarm. Each episode was full of tiny mini-games, each with their own mechanic that had to be programmed from scratch, about 20-30 per game. Fortunately the scripting language in Torque helped us make the best out of a hard situation.

If you had to do it all again, what would you change?

If we could do it all over again, we would have let our game development team be much more involved during the early stages of the project. We would have made the gameplay components first and then engineered the cut scenes to fit within the existing gameplay. This would have made Nanoswarm and Escape from Diab more fluid products.

What can we expect to see next from Archimage?

We have taken what we have learned from these two large projects and moved towards development for the Web. We’ve created an online health games site called Playnormous where kids, their parents, and teachers can learn about health in a fun way. We chose to move towards Flash-based games in order to reach a larger audience and make distribution easier.



Each episode was full of tiny mini-games, each with their own mechanic that had to be programmed from scratch, about 20-30 per game.

Thanks Melanie for your time. I hope you can post results of the study to this site so we can follow your progress.

For more stories like this, check out GarageGames' Developer Interview series.