What Went Right

Now that you have a good idea of how the Viper engine was implemented, I’ll summarize the most prominent things that we did right and wrong. As is always the case at the end of a project, battles lost are always more prominent than battles won. If it were not for some key things that we did right, we might very well have failed.

1. Hardware acceleration support

Two years ago, publishers looked at 3D hardware accelerators with skepticism. Zombie made two key decisions at this point. First, we decided to fight tooth and nail to target a set of art specifically for hardware accelerators. Second, we made this art push the limits of the best accelerator on the market (at the time, the 3Dfx Voodoo 1). The result was that by the time the game shipped, there were many cards that were capable of running the game. In fact, even overshooting the graphics complexity by as much as we did, we weren't pushing the second generation of accelerators at all.

2. Using a third-person perspective

The third-person perspective effectively conveyed the sense of being a Ranger.

SpecOps was designed and under development for almost a year as a first-person game. When we set up a camera over the shoulder of the AI characters as a debugging tool, it was immediately clear that the game was meant to be third person. The sense of being a Ranger, the most important element of the game, was conveyed perfectly by seeing the character move through the environment as a Ranger. I thank Tomb Raider and Resident Evil for establishing this as a valid game interface. If we hadn’t been exposed to these titles, I know we would have thought it was too risky to change perspective that late in the project.

3. Outdoor environments

We wanted this title to kill the flood of dungeon crawlers that have monopolized the market for years. We also wanted to present beautiful, realistic environments that demonstrated the computer’s capabilities. Our levels were built to allow (and even encourage) the players to explore the world in which we placed them. Our first levels were more than two miles across, had nearly 100,000 trees, and required hours to traverse. We ended up scaling the levels back for playability purposes, but still managed to retain this concept. We also chose to create five totally different environments (forest, snow, jungle, desert, and city). This decision had two purposes: to create different tactical combat situations and to keep players’ visual interest as they progress through the game. Kudos to games such as Terra Nova for helping us break down the walls of the dungeons.

4. Complex mission objectives

Games are based on levers and keys because they’re easier to program. Most gamers just find this insulting to their puzzle-solving skills. As the lead programmer, part of me likes switches and keys. But as a game player, another part of me (the part that, admittedly, makes projects late) wants to constantly confront players with new and different challenges. The latter required painstakingly-created custom logic for every mission, with little reuse between levels. It also made debugging levels more painful than it might have been. Still, the end result was a big win for players. You will never quite get familiar with the game, because it will always throw something different at you. I hope more developers will pick up on this and stop building tools that spit out myriad games that all look and feel the same.

5. Realism

SpecOps Producer Sandra Smith on maneuvers with the U.S. Army Rangers.

Whenever you’re making a product that targets simulation fans, realism is key. The executive producer (a former Army Ranger) told us from the start that the product "was more like a movie than a game." With this as a premise, every detail in the game was researched and reproduced as perfectly as possible. We had Rangers come in for the motion capture sessions and photo shoots. We sampled sounds from the actual weapons used in the game. SOCOM (Special Operations Command) officers came in and reviewed our missions, and the game designers spent the better part of a year researching everything from environment characteristics to standard equipment carried by troops. Any one of these details might not have made much of a difference, but as a whole they brought the game up several notches. This attention to detail also created a very positive atmosphere for the entire team.