RocketBowl, a 2004 IGF Finalist.

RocketBowl was developed by Large Animal Games, a small, independent developer based in New York City. Large Animal was founded in January 2001 and has been developing web-based promotional games and downloadable games for the casual game market for the past four years. We currently have a full-time staff of eight and a number of regular contractors.

RocketBowl was Large Animal's fourth original self-funded title and by far our most ambitious. Production started in June of 2003 and the game launched in November of 2004. The core production team consisted of five people, including one who came on board five months before the game shipped. I was the lead game designer and producer on RocketBowl, and the following is my perspective on how things went.

1. Clear Vision for the Game. The RocketBowl concept is a simple one: bowling over contoured terrain. Having a clear and compact core mechanic gave us a point of focus even as great ideas for game features proliferated and threatened to overwhelm the project. We were pretty aggressive about cutting or limiting features, such as moving platforms, multiplayer, and more intricate level designs. We did this not only to contain the scope of the project, but also out of sensitivity to our audience - a broad demographic consisting of 25-50 year-old casual players who don't necessarily have a lot of gaming experience. For each feature, we would ask ourselves the hard questions: Is this critical? Will our players miss it? Would they "get it?" It was a constant struggle to separate our desires as experienced players of traditional console and PC games from our instincts as creators of games for more casual players. We knew we needed to stay focused on the core mechanic, and keep the game simple and easy to learn.

2. Torque and ODE. After investigating a number of 3D engine possibilities, we settled on Garage Game's Torque engine. This option was attractive for a number of reasons: Garage Games was very vocal in its support for independent game development, the price was within reach ($500 for a commercial license), and it had fostered an active community of developers around their engine. This community proved to be an incredible resource for us.

Early in the development process, we made the decision to integrate the Open Dynamics Engine, an open-source physics engine, into Torque. Since the whole mechanic of the game was based around objects rolling and colliding in interesting ways, a realistic physics simulation was crucial. Though others had previously had difficulties integrated ODE with Torque, we managed to avoid the problems they had due to the fact that RocketBowl was a single-player game. We went with ODE because it offers a feature set similar to more expensive physics middleware packages, but at an open-source cost.

3. The Independent Games Festival. The IGF was positive for RocketBowl in a number of ways. First off, it gave us a concrete external deadline to meet. Actually it did this twice, since we submitted versions of the game to the competition in both 2004 and 2005. The additional year's worth of development paid off: the game was selected as a finalist in the 2005 IGF and went on to win the award for Technical Excellence. Being an IGF finalist resulted in some great exposure for RocketBowl. Media outlets picked up on the IGF story, and we were invited to show off the game at the IGF pavilion at this year's GDC. We try to take the whole team out to GDC every year and those free conference passes are important for a small shop like ours.

	RocketBowl

4. Eighteen Months of Development. As a small, self-funded studio, there were many times when most of the production team had to step away from the development of RocketBowl and work on a client game in order to pay the bills. Though there was at least one full-time programmer who was able to keep working on the game during those periods, the net result was nonetheless a long and drawn out production schedule. However, such an arrangement gave us plenty of opportunities to come back to development with fresh eyes, which helped tremendously when iterating on the gameplay, levels, assets, interface, and sound. Furthermore, we had plenty of time to tune the physics and tweak difficulty settings. If a publisher had funded the game, we wouldn't have had the luxury of spending as much time polishing as we did.

5. A Great Team. Through RocketBowl, we found a fantastic programmer and an outstanding associate producer and level designer. Both started out as interns, contributed to the game in a huge way, evolved into contractors, and are now full-time employees. Yossi Horowitz started working with us in the Summer of 2003. Having a programmer dedicated to this project was absolutely critical, given our other obligations and the fact that we were working with technology that was new to us. Yossi climbed the learning curve steadily and made sure that the rest of us we were at least thinking about RocketBowl, even when we didn't have time to actually work on it.

About five months before we launched the game, we brought in Coray Seifert as an associate producer and level designer. Coray earned our respect right away with his impressive Halo skills. He also quickly got his head around the Torque level design tools, dove into writing the remaining game copy, and helped playtest and tune the physics. Most importantly, Coray brought enthusiasm and a fresh perspective, which really helped to re-energize the project in the final stretch.

I think we may take it for granted sometimes, but our team worked (and continues to work!) together extraordinarily well. Everyone on the team was committed to making a great game and impacted RocketBowl in a fundamental way. As a result of everyone's hard work, we managed to maintain a quality of life that, by all accounts, defies the industry standard. We only crunched for a handful of days and even when we did, it felt more like a barn-raising than a death march. I feel incredibly lucky to work with such a smart, creative, and dedicated bunch of people.