The Truth About Bender's Brain
David X. Cohen, of "Futurama," reveals how MOS Technology's 6502 processor ended up in the robot's head
This is part of IEEE Spectrum's Special Report: 25 Microchips That Shook the World.
On 14 November 1999, an episode of ”Futurama,” the animated sci-fi comedy series conceived by ”The Simpsons” creator Matt Groening, jolted computer geeks with a display of technological acumen absolutely unprecedented in prime-time entertainment. In the episode, ”Fry and the Slurm Factory,” a character named Professor Farnsworth points his F-ray at the head of the show’s famously ill-tempered robot, Bender. It reveals a little rectangle, apparently a chip, labeled ”6502.”
The 6502 was a beloved--at least by geeks--8-bit microprocessor created by MOS Technology in 1975. It was the chip that the scruffy-bearded, sandal-wearing Steve Wozniak used to build the Apple II in 1977--”The Machine That Changed Everything,” as PC World once put it. It was also used in the Commodore PET, the BBC Micro, and a host of other systems that fomented the personal computer revolution.
The chip’s cameo on ”Futurama” rocked the nerdosphere, prompting a burst of commentary in online forums (the nerd equivalent of cocktail party chatter). There was also a description on Wikipedia, in the site’s entry about the 6502. But we here at IEEE Spectrum had a few questions that somehow didn’t come up amid all the noise. For starters: Why the 6502? And also: Is it possible that Bender’s maker, Mom’s Friendly Robot Co., somehow obtained Bender’s design from a scruffy-bearded, sandal-wearing computer geek who lived in northern California circa the late 20th century?
To get to the truth, Associate Editor Erico Guizzo sought the brain behind Bender’s brain. That’d be David X. Cohen, executive producer and head writer of ”Futurama,” who, as it turns out, is quite a geek himself. Here’s Cohen’s response:
I spent a good percentage of my high school years programming the Apple II Plus in 6502 assembly language, so I have fond memories of long nights alone with this chip. My greatest 6502 achievement was a video game I called Zoid that was played heavily by me and my father and no one else. Incidentally, Zoid incorporated digitized speech (me saying the word ” Zoid ,” slowed down to make it mightier), which was pretty rare at the time. The digital audio for that single syllable used much more memory than the entire program. I tried to sell the game to Broderbund Software, but I knew I was in for bad news when the return letter they sent me started with a misspelling of my name.
From a programming perspective, a more impressive feat was the creation of an actual working computer language (”FLEET”) for the Apple II Plus that I developed with two high school friends, David Borden and David Schiminovich. We called ourselves ”The Glitchmasters.” This language was intended to make it easy to write high-speed graphics programs (that is, video games) for the Apple. None of us knew anything about compilers, yet using no references on the subject and working entirely in 6502 assembly, we somehow wrote a working compiler. It is even more impressive when you consider that virtually no comments appeared in the program--just page after page of assembly language.
In fact, the resulting compiler was extremely good: It was lightning fast, the language was easy to learn and program, and the compiled programs were comparable in speed to anything we would have written directly in assembly language. I believe it would have been an extremely useful product. However, our timing was astoundingly poor. The compiler was completed in 1984, just as the Apple II was fading forever into oblivion and we were heading off to college. Thus our fabulous compiler never really got used for anything.
In retrospect I would say the limitations of the 6502 forced us, against our wills, to be clever and learn the workings of things at a deeper level. For example, we had to write our own efficient subroutines to multiply and divide 16-bit numbers, using only 8-bit addition and subtraction and bit shifting. As another example, it is possible (in fact trivial) in computer graphics to compute the pixels along a line segment from (A, B) to (C, D) without using division or computing the slope. Again it requires cleverness, though. So I think programming the 6502, especially in the days of limited computer memory, was very useful in terms of learning to think creatively and efficiently.
Moving now 15 years into the future to the year 1999, I was working on an early episode of ”Futurama.” Bender was being X-rayed (actually, ”F-rayed”), and we needed to see what was powering his mighty robot brain. Naturally, the 6502.
While I can claim responsibility for the appearance of the 6502 in ”Futurama,” I was not the most highly trained computer scientist or engineer on the ”Futurama” writing staff. I have a master’s degree in computer science from the University of California, Berkeley. However, writer Ken Keeler has a Ph.D. from Harvard in applied math, as well as a master’s from Stanford in electrical engineering (and yes, in all seriousness, Ken confirms that he does read every issue of IEEE Spectrum and occasionally looks at Transactions on Information Theory ).
No doubt Woz’s head still survives in a jar in the year 3000, and somehow it is probably wearing sandals. So it is quite possible that he provided Bender’s design to Mom’s Friendly Robot Co. in return for some extra fish food in his jar.
For more articles, go to Special Report: 25 Microchips That Shook the World.