A re-post from the Preserving Games blog, February 12, 2010.
Montfort, N., & Bogost, I. (2009). Racing the Beam: The Atari Video Computer System. Platform Studies. Cambridge, Massachusetts: MIT Press.
Just want to give a brief rundown on a really great read I’ve come across. MIT has started a “Platform Studies” series of books where the idea is to examine a platform and its technologies to understand how this informs creative work done on the platform. Platforms could range from gaming consoles, to a programming language, to an operating system, or even the Web itself if this is the platform upon which creative work is being made. The platform in this case is the Atari Video Computer System, the first Atari home system, later referred to as the Atari 2600 in the wake of the newer Atari 5200.
The authors examine the Atari VCS as a computing system, and take care to elaborate the unique (really exceptionally odd) constraints found there. Six games are investigated in chronological order, giving the reader a sense of the programming community’s advancing skill and knowledge of the system: Combat (1977), Adventure (1980), Yar’s Revenge (1981), Pac-Man (1982), Pitfall! (1982), and Star Wars: The Empire Strikes Back (1982).
The most prominent technical details are explained in first few chapters, and they illuminate each game’s construction as an exceptional act of engineering and ingenuity. Just to give an idea of the unique affordances of the Atari VCS, here are a few of the most characteristic details:
- The custom sound and graphics chip, the Television Interface Adapter (TIA), is specifically designed to work with a TV’s CRT ray. The ray itself sprays the electrons onto the inside of a TV screen, left to right, one horizontal scan line at a time, taking a brief break at the end of each line (a “horizontal blank”) and a longer break at the bottom line, before resetting to the top and starting over again (a “vertical blank”). A programmer only has those tiny breaks to send any instructions to the TIA, and really only the vertical break provided enough time to send any game logic to the system.
- It was imperative that game logic be sent at these breaks because the Atari VCS had no room for a video buffer. This meant there was no way to store an image of the next frame of the game, all graphic instructions are written in real time (sound instructions had to be dropped in on one of the breaks). A designer or programmer could choose to restrict the visual field of the game in exchange for more time to send game logic instructions. Pitfall! is an example of this.
- This means there are no pixels on the Atari VCS. Pixels require horizontal and vertical planes, but for the Atari VCS, there is only horizontal scan lines. There is no logical vertical division at all for the computational system. As the beam goes across the screen, a programmer can send a signal to one of the TIA’s register to change the color. Thus, the “pixels” are really a measure of time (the clock counts of the processor) and not space.
- Sprites, such as they existed for the Atari VCS, were hard-coded into the ROM of the system. Programmers had five: two player sprites, two missiles, and one ball. Reworking that setup (clearly designed for Pong and the like) into something like Adventure, Pitfall!, or even the Pac-Man port is an amazing feet.
The book doesn’t refrain from the technical. I could have used even more elaboration than what is presented in the book, but after a certain point the book would turn into an academic or technical tome (not that there’s anything wrong with that), so I appreciate the fine line walked here. The authors succeed at illuminating technical constraints enough for the general reader to understand the quality of the engineering solutions being described. Moreover, the authors leave room to discuss the cultural significance of the platform, and to reflect on how the mechanics and aesthetics of these Atari titles have informed genres and gameplay presently.