Designing the Game

Once I figured out the basics of VCS programming, I set to work. Since programming the VCS was so difficult, management didn't place any constraints on a designer's early work; you just developed any idea that came into your head, and only after you'd gotten something working would management take a look at it and offer suggestions. Much later, when the game was taking clear shape, the marketing people would come into the picture.

The central notion I wanted to develop in my design was the idea known elsewhere as fog of war: You don't know where the enemy is, but he's coming to get you. All other games at the time kept you fully informed as to the location and movements of your enemy. This, it seemed to me, deprived the game of important elements of suspense. So I decided to build a game in which you must fight an enemy who can hide behind walls.

This required a carefully designed playfield: It had to provide lots of walls to hide behind, but also permit occasional long lines of sight. I spent a lot of time tuning the playfield to get the right balance. My final result looks like what appears in Figure 15.2.

15.2. The final result.

Note that this is a reflected playfield: The two horizontal halves are reflected images. This saved me 58 bytes of ROM space, a significant savings in a 2K ROM. Did I mention that the entire program, display tables, sound tables, and so forth had to fit inside a 2K ROM?

It didn't take a design genius to figure out that there would be two players, one for the human player, and one for the computer. I gave them asymmetric properties to sharpen the conflict (and to lessen the difficulties of computer AI on a 2K cartridge). The computer player henceforth the monster would simply pursue the human, shooting at him whenever he got a clear line of sight (LOS). The human would run around, evading the monster and getting off shots as best he could.

Since I didn't have the resources to build an AI smart enough to plot paths through the maze, the monster needed the ability to move through walls, but I saw no reason to give the human the same ability; this would be an advantage of the monster. Thus, the monster could, in effect, see through the walls, but the human couldn't. But this created a problem: It was mightily disconcerting to have the monster suddenly pop out from a wall when you least expected it. I'm quite proud of my solution to this problem: I equipped the monster with a characteristic sound that grew louder as it drew closer. To really make the player's hair stand on end, I needed a creepy sound. This was none too easy to do: The TIA's sound capabilities were optimized for gunshots, explosions, and so forth. After much experimentation, I fashioned a believable heartbeat sound. This turned out to be one of the finest elements of the game. When the monster was out there, steadily bearing down on you, the rising sound of its heartbeat raised the tension to unbearable levels.