GDX 2009, Ian Bogost, Bone of My Bones and Flesh of My Flesh: The Genesis of Ms Pac-Man

by Darius Kazemi on April 16, 2009

in conferences,design,gdx2009,hardware,transcript

Here are my raw session notes for Ian Bogost’s GDX talk, Bone of My Bones and Flesh of My Flesh: The Genesis of Ms Pac-Man. This is my best attempt at a transcription of what he said. Any mistakes or misinterpretations are mine and mine alone. My comments are in square brackets.

Platform studies: looking at the hardware systems that underlie software systems and how the hardware influences creativity. Been thinking about this era of ’77-’82 in general, and Ms Pac Man is a game I have been interested in for many years now and after my recent book came out I have time to focus on it now.
What i’m going to do is tell you a little about the game and how it came to be, some of this will be old hat, but there are some new observations I’d like to cover in the course of collecting old research before it gets lost to history. After I talk about the machine and the game I’m going to do some crazy shit.
This is Ms Pac Man, popular arcade game from 81-82, similar to Pac Man from 1980. Pac Man is a game whose popularity literally launched a hysteria (one might even say a fever) when it came out. If you approach this as a player might approach a cabinet, and it would be tempting to think that Ms Pac Man is sequel or followup to the immense success of the other. But it’s not exactly true.
It’s commonly said Ms Pac Man is a mod. But mod is an unfair characterization to apply to Ms Pac Man. It would be like calling CounterStrike a mod — it came out of one game, but it’s so different from what modding means today that it would be dangerous to call it that.
In order to understand how MPM came about we need to look at two factors: coin op platforms and arcade enhancement kits.
By 1980 there were several home game consoles: Atari VCS, Intellivision, Fairchild Channel F. These platforms were created to let the player buy one piece of hardware and play many different games on that hardware. This would exert constraint on the person making the game because they’d have to take into account the system’s design. I’m mentioning this to highlight that way before Pac Man people were thinking about consoles.
Coin ops were one-off design affairs. Machines that played one game. It’s a big piece of wood, metal, chips that played one game. But it’s not exactly right. Platform thinking in the coinop world began very early on. Pong was a game that was all TTL logic, no microcontrollers. Coin op is a very easy way to launder money, so they were subjected to regulations. Ownership regulations around businesses like arcades, so Nolan Bushnell decided he would spawn a fake competitor to Atari called Key Games (Nolan was on the board). They kicked over their best programmer who took the knowledge of what he did at Atari and made games at Key. Tank, which is like Combat on the Atari 2600, is in some ways from the TTL logic standpoint, similar to Pong. Bouncing balls and stuff.
So there was still reusable design thinking going on.
If you look at the design of a machine like pac Man it’s worth digging into its guts. Not just what’s inside but how it was built on other designs. Worth noting: Z80 chip, 16k ROM (huge ROM compared to Atari carts), 2K RAM, 16 colors paletized, and 8 16×16 sprites. From a visual perspective, the way the ROM and RAM was divided was RAM was just video memory, what we would call a tile map but they called characters. 244×228, 28×36 tile grid of 8 pixel tiles. It’s remarkable how the visual design hides how much of a grid the game really is. Already there was reuse of the hardware. 
Pac Man and Rally X are the same game in certain ways. They are both mazes, but they also use the exact same hardware architecture. What it suggests is the idea that there’s a porous underlying system that changes in small ways from machine to machine. In a somewhat ad hoc way, coin op boards were inching towards a platform (eventually we got straight up standardized coin ops like the Neo Geo).
A coinop game is a weird object. It’s a major investment, not just for the dev who makes the game, software, hardware, etc. You don’t sell the game to consumers, you sell it to operators. It costs thousands of dollars to buy one of these. But more importantly, they are huge and take up a lot of space. You can only have so many of them in your arcade, hard to haul out the failed games and replace.
One way to address this problem was through enhancement kits. They would attach to an arcade machine and give it some slightly new behavior, sometimes changes to graphics or game logic. Sometimes small changes like an update the scoring system.
Asteriods enhancement kits could extend the number of digits in the scores and save the scores after powering it down. In the original Asteroids, high scores were huge, so this feature refreshed Asteriods significantly without doing a lot of changes. The enhancement kits were simple to install, just yank out a chip, plug in the enchancement board.
Pac Man Plus from 1981 speeds up the play of Pac Man and changes the maze color. In addition the fruit and bonuses are changed, etc. You’d also get a new marquee for your cabinet. Cheap easy way to renew your existing game.
This would have been really useful if you only had a few machines in a bowling alley or tavern.
My favorite enhancement kit is for Dragon’s Lair called Super Don Quixote!
This leads back to Ms Pac Man. There were students at MIT who started messing around with coinop boards. They messed around with their own enhancement board, started a company General Computing Corporation to sell enhancement boards for existing games. What’s interesting about the boards is that they’re meant to physically draw in new players (new attract mode and sounds, etc). Companies used to publish the circuit boards for repair but that made them too easy to hack so they stopped doing it. By this time the games were running largely in software. The hackers would reverse engineer everything. The MIT guys would use microprocessor emulation systems to do tests and figure out how the existing code worked, and find a minimalist way to change the code.
Maybe they’d change 5%-10% of the code itself. There were a few ways to distribute the kits. One is to distribute ROMs, but people could copy the ROMs. There were interesting questions about selling a 10% modification on 90% code other people wrote, apparanltly it was legal. GCC did a lot of hardware extensions to the Atari consoles eventually and even designed the 7800.
Pac Man was a good target for enhancements. There was a game called Crazy Otto, sometimes called Pac Man With Legs, looked basically just like Ms Pac Man with a different protagonist. 
For major gameplay changes: four different mazes in four different colors. A feeling of progress. Different, more random monster behaviors (Pac Man had deterministic monsters so you could memorize a way to beat each level). Bonus objects such as fruits moved around the screen. And then narrative intermissions. Otto had Anna chasing him around on the screen. Crazy Otto was entirely Ms Pac Man.
Pac Man had 16K of ROM split into four ROMs of 4K each, ROM A B C D. The Auxilary Board would modify some base code, rom D was removed, a ROM E was added, a ROM F, and then 40 8-byte patches were added so if you hit a place in memory, it would do a jump to new code, and then jump back. So GCC went to Midway who distributed Pac Man. Midway was mad at Namco since Namco was not moving on making a sequel.
10/9/81: GCC shows Otto to Midway, almost entirely done at that point.
10/29/81: GCC signed contract, added Midway logo.
11/81: Midway calls GCC, and they settle on female pac man
Nov-Dec 81: Going back and forth to finish things up. Final title was established at this point. Initial title suggested: Miss Pac-Man. Then, Pac-Woman. Then Mrs. Pac-Man, because there’s concern over baby pac man in the intermissions. Eventually they settled on Ms. Pac Man.
01/82: Deliver the final code to Midway.
I promised you weird shit, though. The title of my talk is the genesis of Ms Pac-Man. Two weird things.
One is the notion of the genesis of the machine, the other is the concept of Ms.
When I say “genesis” I mean it, Biblically, the creation of man and woman.
Gen 1:26, let us make man in our image. Man in the image of God, man and woman in the image of God. There’s a lot of discourse about the language in hebrew Genesis. The word Selem “image” can refer to form or ideal, or idols false idols. Comes from the root of carving. Another word meaning more “likeness” or “appearance”. Makes man according to ideal of God and God’s appearance. Mankind is a representation of a higher form, and that likeness suggests that man appears like god as well.
Gen 2:21-3, man’s ribs used to create woman. Man: “This now at least is the bone of my bones, and flesh of my flesh.” Hebrew word Selam: means rib, but can also mean a side, a plank, a BOARD.  Yet simialr to the word for ideal or idol, this board.
This is demonstrating an equivalency of origin, woman and man are instances of mankind.
I’m using the analogy because it helps us create a metaphor for how Pac Man and Ms Pac Man are related. One can say without hyperbole that Ms Pac Man was created from the rib/board of Pac Man, but almost reversed: a chip removed, a daughterboard added. They are two instances of common underlying hardware structure. The “god” is the platform, the abstraction of the integrated circuits that makes this possible.
The second thing that fascinates me is where the Ms. came from. In the interstitials you see Act 1: PM and PW are being chased, the ghosts bump together, the two are joined. In act 2, both are chasing each other. In act 3, they are together at the bottom of the screen and a stork drops off baby. Common conflict, love affair, creation of child. But if you look at how Pac Man was advertised, Ms. Pac-Man was a vampy femme fatale! On the cabinet itself she’s got her legs all up. How do we reconcile this?
It’s in the concept of “MS.” Mrs and Miss are abbrevations from the 18th century “mistress”. We don’t use it anymore but it used to just mean the female form of Mister. It was neutral like Mr. In 1950s, we see it reappear as a convenience for writing business letters. By the 1960s it was used as a title of a woman who did not belong to a man (women’s liberation). By the 1970s it took off and it became Gloria Steinem’s magazine. By the 1980s, Ms. was a standard cultural practice to refer to a woman without refering her marital status.
Ms as a concept introduces ambiguity: decouples a woman’s professional life from her personal. A lot of the ambiguity is performed in the game. We have the vampy seductress in the marketing, but she’s the wife and the mother in the game movies. And then in game, she’s a working girl who just does what Pac-Man does, but better (it’s a harder game). For another part she’s a traditionalist, family woman, a mother. These world are mechanically separately in the game between the gameplay and the movie: the challenges of work itself bring the two together, the common struggle around the work brings her together with her family.
Another ambiguity is the circumstance around which the Ms Pac Man name was created. Was this iteratively arrived at in the halls of Midway?
Ms. Pac Man is not only a game with a particular game, but is a game that is performing this notion of “Ms.” if you look at the way that it’s a pop cultural icon, she’s still capable of performing these roles.
To finish up, tehre are two major cultural revelations thorugh Ms Pac Man: it shows us two takes on a common platform in which each take sheds a different light on the underlying hardware. In some weird way this is how the God of Genesis described how man and woman mirror equally the likeness of God. Similarly games that are built upon the same form invite us to envision each of them individually but also the common machinic form.
It’s also the apotheosis of the feminist video game, through and through. It’s a woman who triumphs over a man by playing his game better than he ever could, captivates by being more challenging, manages to balance the many sides of being female.


Nick April 17, 2009 at 3:27 am

I might not always completely agree with Mr. Bogost, but I always come away from his talks or his writings feeling I’ve learned something. He’s a joy to listen to.

Ian Schreiber April 23, 2009 at 8:35 pm

For the sake of accuracy, the early Atari spinoff/competitor was “Kee Games” (named after Bushnell’s next-door neighbor, Joe Keenan). The company logo did picture a key, though, so the homonym was intentional.

My understanding is that the main purpose behind this was that a lot of early arcade-game distributors required exclusivity agreements. Two companies making clones of each other’s games = double the profits. Sneaky but brilliant.

Comments on this entry are closed.

Previous post:

Next post: