Should You Extend at All?#
This chapter has watched a cartridge grow from a slab of read-only memory into a machine carrying its own ARM computer. Which leaves an honest question hanging over all of it: just because you can put a supercomputer in the slot, should you?
The clearest way to think about it is to follow one game through three lives.
Three Pac-Mans#
1982 — the original. Atari handed Pac-Man to Tod Frye — a capable programmer who’d actually wanted Defender and got Pac-Man by default — under enormous corporate pressure: ship the most-wanted arcade game of the year in time for the holidays, on the cheapest cartridge the company would pay for (a single 4 KB chip), against a deadline that left no room to refine it. The result became the best-selling game on the entire system and a byword for disappointment — the ghosts flickered relentlessly, the colors looked nothing like the arcade, the whole thing played like a rough sketch of the real game. Atari pressed something like twelve million copies — more carts than there were consoles to play them — and it turns up in nearly every story told about the 1983 crash. The lesson everyone drew at the time was simple, and wrong: the 2600 just can’t do Pac-Man. The fault was never the programmer’s talent; it was the calendar and the boardroom — the very same pressure that, that same year, gave Howard Scott Warshaw about five weeks to make E.T. and turned the designer of the brilliant Yars’ Revenge into the other name blamed for the 1983 crash.
Decades later — the same 4 KB, done right. Homebrewer Dennis Debro went back and built Pac-Man 4K — far more faithful, steadier, much closer to the arcade — in the very same 4 KB, no extra hardware at all. With no ship date and no boardroom, the constraint that supposedly doomed the original turned out to be plenty. The 4 KB could do it all along; the schedule couldn’t. The original’s flaws were a deadline, not a ceiling.
Then — extended. Debro’s later Pac-Man 8K reaches for bankswitching and twice the ROM to push closer still — more animation, fuller graphics, fidelity the bare 4 KB genuinely can’t reach. Same hand, same arcade target, one more tool on the cart. This is the cartridge doing exactly what this chapter has been about.
Two valid arts#
The temptation is to read those three as a ranking — bad, good, best — but that misses the point. The same craftsman built both the 4 KB version and the 8 KB one; they aren’t rivals but two different kinds of achievement, and neither is the lesser:
- Mastery within the constraint. Fitting a real Pac-Man into 4 KB is a feat of compression and timing, every byte and every cycle argued for. The constraint is the art — the whole reason these feats impress is that the box shouldn’t be able to do them. It’s the tradition of Combat’s twenty-seven games in 2 KB and Pitfall’s jungle conjured from a single byte.
- Ambition through extension. Adding hardware doesn’t so much cheat the machine as refuse to waste four decades of better tools. If the goal is the game in front of the player, a fuller, smoother, more faithful version is a real win — and the cartridge that reaches it is a legitimate instrument, not a crutch.
Extending was there from the start#
If extending felt like a betrayal of the “real” 2600, the platform’s own history would be the first thing to throw out. The cartridge was extended from the very beginning — Atari’s bankswitching in 1981 and its Superchip RAM by 1983, then Activision’s DPC coprocessor — David Crane’s own design — in 1984. Extra hardware on the cart isn’t a modern intrusion on a pure machine; it’s been part of how 2600 games were made for almost the whole commercial life of the system, by Atari and by the very best of its rivals. The line between “pure” and “extended” is one you draw for yourself — not one the machine draws for you.
So there’s no verdict here, only a choice. You can chase the small, perfect, fully-constrained game, or the ambitious one the modern cartridge makes possible. Both are real 2600 programming. They’re just aimed at different kinds of wonder — and the proof is that the same game, Pac-Man, can be a triumph either way.
Conclusion#
Step back far enough and that is the real marvel of this machine. The VCS was designed by Jay Miner and Joe Decuir — from a concept hatched at Atari’s Cyan Engineering by Steve Mayer and Ron Milner — to do one modest thing: play home versions of Pong and Tank. Two players, two missiles, a ball, a walled field, and barely anything more. That was the entire brief.
From that tiny, single-purpose engine grew everything this book has walked through — asymmetric playfields, multiplexed armies, procedurally generated jungles, bankswitched epics, coprocessor soundtracks, and finally cartridges with their own computers inside. Depths of play its designers, sketching a handful of chips in the mid-1970s, could not have imagined.
That arc is the story of the Atari VCS: a machine asked, again and again, to do what it was never built for — and finding a way every single time. Now you know how it’s done. Go ask it for something it was never meant to do.