The “Zorfon” Mystery: Answers from a Golden-Age Video Game Developer
Programmer Noelie Alito reveals the story behind her one-of-a-kind Atari prototype cartridge
“Hello, are you the … Noelie Alito who worked for GCC in the early 1980s and programmed the game Moon Patrol (among others) for the Atari 2600?”
I was pretty sure I’d sent my letter to the right mailing address. But after waiting for weeks with no reply, I started to wonder if I’d ever learn anything more about the super unusual cartridge that I’d recently came to possess.
As my hope really started to fade, on the morning of my birthday I woke up to find a response in my email inbox.
Noelie Alito not only answered my questions about the odd cartridge in my possession, ZORFON PATROL, she kindly answered a number of interview questions, allowing me a glimpse into what it was like working for General Computer Corporation doing programming for Atari games in the early 1980s.
Here’s how it all happened.
THE ATARI 2600 SCENE TODAY
It doesn’t happen often that something new and unexpected shows up on the now over-40-years-old Atari 2600 scene. Occasionally something previously unknown to the collecting community surfaces, like the Sears picture label cartridge for Superman (first discussed in 2003), the Atari Inc. picture label for Flag Capture (2006), and the box and manual for the still-mysterious Air Raid (2010). Even better, every now and then an unreleased game will be brought to light thanks to a generous and thoughtful programmer, like Paul Walters’ Arkyology was in August of 2018.
Internet forums like AtariAge, and database sites like Atarimania and AtariBoxed, have become indispensable tools for sharing, documenting, and cataloguing a part of video gaming history that might otherwise be largely lost to time. It’s best not to think for too long about how much Atari history may already be lost due to bit-rot, negligence, or other reasons.
When it comes to Zorfon Patrol, the data stored on the EPROMs could have become corrupted or erased. The cartridge’s original owner could have simply thrown it away. The thrift shop to where it had been brought could easily have trashed it. The guy who got it from his friend who worked at the thrift shop might have discarded it because it didn’t have a “normal” Atari label like all the other cartridges. Or it could have been lost in transit to me.
Thankfully, none of those things happened. I made sure of the last one.
The cartridge that I drove to personally pick up from a seller last year was certainly something new to the Atari scene, too, but in its own unique way. It’s one-of-a-kind, as confirmed by its creator, and to my knowledge nothing exactly like it in the Atari 2600 world has been found before.
GENERAL COMPUTER CORPORATION
The story of how Atari ended up working with the upstart General Computer Corporation is fairly well known. Unauthorized arcade enhancement kits for Atari’s Missile Command led to litigation between GCC and Atari in 1981, which led to a settlement that resulted in Atari contracting with GCC to handle programming of some of the biggest name home video game titles of the early 80s. The burgeoning tech company founded by MIT dropouts ended up creating well-received home console versions of arcade hits like Ms. Pac-Man, Jungle Hunt, Pole Position, Galaxian, and Dig Dug. According to Doug Macrae, co-founder and former Chairman of GCC, 77 titles in total were shipped for Atari over the span of a few years.
One of the first 2600 titles that Atari tasked GCC with programming was the arcade game Moon Patrol (originally released in arcades by Irem in Japan, licensed and imported by Williams Electronics to North America). Moon Patrol was successful enough in arcades in the second-half of 1982 to be an obvious choice to be ported to numerous home video game systems and computers, and the consensus seems to be that GCC did an admirable job with the 2600 port.
ATARI PROTOTYPES
At least six prototype versions of 2600 Moon Patrol have surfaced over the years, and they allow a glimpse into the refinement process of the creation of a video game. Over time, one sees features added and removed from the game as the prototypes progress. Some changes are minor, others are more significant. The oldest known prototype had no title screen included, among numerous other differences; the latest known prototype is identical to the final retail release.
For GCC-programmed Atari games, the large majority of prototype/review cartridges that have been found sport the more commonly seen black Atari “Loaner Cartridge” labels. On rare occasions, however, a prototype cartridge turns up with a blue loaner label. These may have only been for internal GCC use, i.e. for games in developmental stages and not meant to be distributed to Atari or anywhere else. At least one prototype of 5200 Choplifter exists with a black Atari lab loaner label over top of the internal GCC blue label, perhaps indicating that it was first reviewed internally before being sent to Atari for review.
THE AUCTION
I spotted an interesting-looking blue-labeled GCC prototype cartridge in an eBay game lot auction in October of 2018. One wouldn’t have known that a prototype was included in the lot from the title of the auction listing, or even from the first seven of the twelve posted photos. Only two of the twelve showed the prototype’s label in any detail. I consider myself pretty fastidious when it comes to scouring eBay for unusual Atari stuff, but I could have easily missed it myself if not for some good fortune.
I have a few prototypes in my collection, but I immediately felt that this one could be something special, and I decided that I’d go all-in on trying to win it. Why?
One: that title! I was beyond intrigued. What could it be? What on Earth is a “Zorfon”?! I was pretty sure the cartridge had something to do with Moon Patrol, based on the fact that…
…Two: It was hand-signed by Noelie Alito, a GCC programmer on the team that made Moon Patrol for the 2600. If nothing else, a prototype signed by its programmer is certainly something that doesn’t show up every day. Everything pointed to this being something extraordinary.
The day before the end of the auction, I’d settled on a maximum amount that I’d bid in the closing seconds. There were enough people watching the listing (some of whom had already placed bids) that I felt pretty strongly that someone with deeper pockets would top my maximum. But I knew I had to try. Not wanting to trust an auto-bidding tool, I set a 2:30 AM alarm in order to place my bid at the very end of the auction. Waking up, I bid the most that I’ve ever bid on *any* auction, and held my breath as the final seconds ticked away.
Somehow… my bid ended up being just enough. I won. I was nearly shaking with excitement after the confirmation notification popped up. With my mind racing, wondering what exactly it was that I’d just won, it took a while for me to fall back asleep.
The following day, I contacted the seller to ask if I could pick up the lot in person. I figured that they would agree considering that I live not too far away, and that I would be saving them the time and effort to package up a sizable lot of cartridges for shipping. Left unsaid at the time was the fact that I wanted to avoid risking a potentially one-of-a-kind item getting lost in the mail. After a couple of anxious days trading messages, we agreed to meet up the following weekend.
I met the seller in a parking lot in his neighborhood, and I quickly sifted through the box of cartridges, looking for the only one I really cared about. Once I saw that it was in there, I paid the seller, thanked him, and asked him two questions that I’d had on my mind since first seeing the auction:
“Do you know a Noelie Alito?” (I also mentioned her married name in case the seller only knew her by that name.) He did not.
“Do you know who owned these cartridges?” The auction listing stated that the carts had been stored in a drawer for years, so I’d hoped the seller knew something about their provenance. The seller somewhat reluctantly admitted to me that he’d got them from a friend at a secondhand shop in nearby Potomac, Maryland, so no, he didn’t know that either.
LOOKING FOR ANSWERS
I drove home, still not knowing exactly what it was that I had bought! Holding it in my hand, I could easily feel that it was heavier than a standard cartridge due to the larger PCB and EPROMs inside. Arriving home, I immediately fired up my Atari 2600. The expected attract-mode screen of Moon Patrol popped up. Then, seconds later, the title screen appeared.
My jaw dropped. I ran to grab my laptop to fire up Moon Patrol on the Atari 2600 Stella emulator so I could compare the two games side by side. Obviously, there was something noticeably different about the title screen!
The handwritten note on the label about controlling the music led me to believe that it was an almost-final release, and the screen graphics and game sounds brought me to the same conclusion.
So… what was this? Was there a chance that there was some kind of licensing issue with the Moon Patrol name? Had Atari considered giving the home version a unique name that they thought was more interesting? Was it something else entirely? I was too excited to think straight.
Googling for “Zorfon” didn’t bring up anything that seemed even tangentially relevant. I knew then that I was going to have to try to track down the original programmers to get answers. Doing some research, I learned that Noelie Alito did, at one time, have a connection to Potomac, Maryland. A little sleuthing led to figuring out what I believed to be a current address, and I composed a snail-mail letter. I dropped a small photo print of the cartridge into the envelope as well, in case it might help refresh her memory.
When I didn’t receive a response for quite some time, I decided to send an email to Dr. Mark Ackerman, a professor at the University of Michigan, and another programmer from the GCC team who worked on 2600 Moon Patrol (among various other titles). He quickly replied with the opinion that I’d come to own what he called an “Xmas cartridge”. He elaborated:
“I did one of Ms. Pac [Man] that had “Jeremy” in place of the title screen for my nephew (unfortunately, I’m pretty sure his family tossed it at some point) and then the owners/management of GCC allowed others to also burn a few personalized cartridges too.”
This certainly felt more like the “right” answer to me — certainly more reasonable than my theory about a potential licensing issue, anyway. I thanked him for his help and continued to wait to hear from Noelie Alito, especially since he said he had no idea what “Zorfon” meant.
It made my month when Noelie finally emailed me a few weeks later.
THE INTERVIEW
Over the course of multiple emails back-and-forth, Noelie answered pretty much all of my questions. She said that she wasn’t looking for any publicity or fame, but that if I wanted to “plumb [her] questionable memory” about her time at GCC, she would oblige.
What follows is a reconstructed interview based on my original letter and our subsequent emails. It has been edited for readability.
Rob Wanenchak: What’s your educational/programming background, and what was your path to the programming industry? Had programming always been an interest of yours?
NA: Near the end of my Freshman year at MIT, I went to see my Freshman Advisor in his electronic lab in a basement. He wasn’t there, but his serf grad student talked to me about potential majors. I was interested in physics, math and the introductory software. I remember the grad student, seemingly at home in his Pit of Despair, advising me against math and physics (“There’s no money in that.”) Software could be applied to any discipline, so he suggested I go with the EECS major (“Course VI”), and I could change it later. At the time, the first year of EECS (which would be my sophomore year) was joint software and hardware, and the classes forked into Electrical Engineer or Computer Science depending on your preference after that.
In 1980, when Computer Science was a young discipline, my MIT software class was about 1/3 female, which might have been Peak Female for that department. After that, women tended to infiltrate the traditional white-shirt-and-pocket-protector ranks of the long-established engineering disciplines… and men went into software.
[One of GCC’s founding employees] came to MIT to do the interviews, and I remember that when he asked me the questions (“What is the difference between a compiler and an interpreter?”), he kept checking my answers against what was written down. It was clear he wasn’t a tech guy. I didn’t know 6502 at the time, but they and I expected it to be trivial to learn it as I went (many MIT software problem sets involved implementing solutions in languages learned on the fly).
RW: Did you go straight from your coursework at MIT to work at GCC?
NA: I spent my 9th semester doing my MIT “undergrad thesis” (wrote a program for a professor), which was the only graduation requirement I had left. I joined GCC in January of 1981.
{Author’s note: Based on other evidence, this was likely misremembered or typed incorrectly. January of 1983 was when she likely began at GCC.}
RW: How long had GCC existed when you were hired on?
NA: I don’t know how long they had been in existence. There were already upwards of 30 people there.
If I remember correctly, General Computer Company was a “contractual entity” between Atari and General Computer Corporation, and Atari has “right of first refusal” for all GCCo. products, though technically, Atari couldn’t take ownership of any games licensed to other companies (like Namco and the Pac-Man franchise). Atari arranged for joint license agreements to make 2600 versions of a variety of (non-Atari) arcade games. GCC had already made a Ms. Pac-Man cart and the arcade games Food Fight (modest sales?) and Quantum (which didn’t get traction with arcade owners because, IIRC, their experience with high failure rates of the vector-graphics hardware used in Tempest).
RW: What was it like being hired by GCC? Were you excited, or was it just a job on the way to other things, or was it something else entirely?
NA: I’ve never been an ambitious person, but I keep getting into situations that some people would give their right arms for (my student job was a tape ape at the lab that processed astronomy satellite data, for instance).
My previous summer internship involved prototyping 3D projections on an 8-bit x 3-color x 512 x 512 graphics array. When I was toured through the GCC lab my guide pointed out that the 2600 work used 128 bytes of RAM. I heard “128 bytes” but I figured he had misspoken and had meant 128k. Boy was I wrong.
I worried about not being a “fun” enough person for GCC, but fortunately the work required enough detail and focus that I knew I would be a contributor. I worked with conversion projects (i.e., from arcade to home versions), and was not bothered by the fact that all the “sexy” work was in creating entirely new arcade games (including hardware). On top of that, besides the profit-sharing all employees got, the conversion workers got a slice of the royalties on retail sales of the carts. I went to two Consumer Electronics Shows (Las Vegas and Chicago), and the whole company went to Florida for Disney World, plus a shuttle launch.
I was there a total of 20 months, but it was a very intense 20 months.
RW: You may know about this, but in case you don’t… there are a number of photos from your GCC days here, on Steve Szymanski’s page.
NA: Wow. I didn’t know because I have had a varied and distracting work history since then. (Although I do remember going back to visit shortly after they switched to after-market installation of Mac hard drives, and found it terribly depressing after the go-go days of video games.) I remember Steve S. was best friends with [Allen] “Alien” Welles, my one-time group leader and an all-around nice guy. In fact, the last time I remember spending any amount of time thinking about GCC was when I learned of Alien’s death many years ago, so terribly sad.
RW: So… what’s the story behind the existence of Zorfon Patrol?
Noelie Alito: Zorfon Patrol was essentially a bootleg copy for a young relative in Potomac, but I didn’t think it was a risk of general piracy due to the relative difficulty of making the carts (as opposed to tape-based games). I’m not sure if that version of MP used an anti-piracy trick, which involved the code attempting to write over itself, effectively a NOP on the PROM version. The cartridges were relatively hard to pirate compared to other contemporary formats of software, so the risk of personal bootlegs was minimal. The cartridges were, in effect, their own dongles.
RW: Is “Zorfon” just a nonsense, alien-sounding word that you made up? Or is there another explanation for it?
NA: If I remember correctly, it was a type of family epithet, like “a Farkel” (from Laugh-In) or “a Gumby” (from Monty Python). I remember coding the bitmap on graph paper while commuting by rail (Boston’s “Purple line”).
RW: Did you make any other personalized carts?
NA: I made no other personalized carts, ZP or otherwise. There is a very remote possibility that there was an earlier attempt at ZP left at GCC, but we were pretty aggressive at re-using EEPROMs and the little socketed cartridge boards.
RW: What were the circumstances of giving the game to your younger relative? Was it for a special occasion?
NA: It was a chance to do something “cool” for him.
Do you recall if it was given to them before Moon Patrol was sold at retail to the general public?
NA: It definitely went out to him before the cart launched. It was a trivial effort: The most time I spent on it was filling in and tweaking the title page bitmap on graph paper while heading back home on the commuter rail.
RW: I haven’t found any significant differences from the eventual retail release apart from the alternate title screen. Do you happen to remember if you made any other changes?
NA: I don’t even remember if it had the full rock/hole/rocketship map coded in, or whether it had the standard anti-piracy trick of writing over itself (harmless on ROMs). Throughout my software career I’ve come to realize that the developer goes through so many variations of design or implementation that they tend to remember less about the final application (or interface) than the users of their software, just as I was more familiar with the final hardware/firmware interface than its original designer might have been.
RW: The copy protection you mentioned — was it a ROM checksum test that would fail if someone tried copying the code?
NA: I wasn’t describing a ROM checksum, though we might have had that for QC/verification rather than anti-piracy. The anti-piracy would just have been real ROM game code that overwrote other essential code. If that real code executed on a cart, it would have no effect (be essentially a NOP), but if it ran downloaded into RAM, it would trash the program.
RW: Was this anti-piracy code only used during development? I don’t think any retail versions of Atari-published 2600 games are known to have code that does this, but I could be wrong.
NA: By the time I started at GCC others had created a basic skeleton for all 2600 games, including the sound driver and the dual-ROM switching handler, so that later developments could concentrate on gameplay implementation. It may be that this anti-piracy trick was only used on 7800 carts.
The original 2600 carts had an address range limited by the cartridge connector. The expanded carts had a simple element which, when strobed, would toggle which ROM was being read, almost doubling the amount of ROM we could work with. Typically the gameplay would be on one ROM and the display code (juggling the background, ball, player, etc., bitmaps) was on the other. Since we didn’t know which ROM would come up when the cart was powered up, one ROM booted the game-start code and the other started up strobing the toggle, sending it to the game-start ROM. Again, this is part of the basic software infrastructure that later game converters inherited.
RW: Is there anything on this AtariProtos page, or the linked pages at the bottom that discuss various work-in-progress stages of the Moon Patrol prototype, that you can further explain or expand upon?
NA: The sound was created by a dedicated sound/music person [Patty Goodson] working on all the carts, and many of the bitmaps (like the buggy and flying ships) were created by graphic designers. For all I can remember, I might have done the bitmaps for the rocks and holes, but it would have been a trivial issue in any case, maybe a few hours’ work.
When I was assigned Moon Patrol as my first project, they told me it would be a fairly straightforward conversion, a simple project for a newbie, but a large part of the effort was the transition at ground level, where all the bitmaps would have to be reset in the time of one scan line. I remember asking one of the developers who had predicted it would be an easy job whether he knew of a special trick at that point and his response was on the lines of “oh, yeah, that does sound hard”. The missile lengths (implemented using BALL register, I think) were a bitch, and ended up being one of those frustrating design trade-offs. Flickering was never an option at GCC, thank the gods.
RW: I think it would be interesting to Atari fans to know the reason why some things were removed from the earliest prototype versions of Moon Patrol (like the checkpoint letters), if you remember.
NA: I don’t specifically remember about the checkpoint letters, but if they were in the arcade version I might have put them in early on, but later sacrificed them to the Goddess of Limited ROM and Super-duper-extremely Limited RAM. That happened a lot: Gameplay Uber Alles.
RW: What other games did you work on during your time at GCC? Some sources I’ve seen say that you also worked on 7800 Pole Position II and 7800 Dig Dug.
NA: I teamed up with Jacques [Hugon] to port Pole Position II. It worked well, because I came in before 9am and Jacques typically wandered in well after noon, meaning we each had dedicated time on our shared in-circuit emulator. We were close to finished when the word came down that it would be the default power-up game for the 7800. Because it would take extra work to make room for the boot code, the first 7800s were shipped with separate Pole Position II cartridges. The annoying part was that we told them not to use those ROMs because we were working on a bug that happened at the advanced level, but they shipped them anyway. We finally identified the bug (hard to reproduce because the display had to be unusually busy), and the fixed version went in the 7800 boot ROM.
Coding was also inevitably tight for the 7800 because we wanted to put in so many more game features. At one point, Jacques and I discussed how neat it would be if we could animate the wheels coming off like the arcade version. I remember spending an intense morning trying to strip down enough code and animations to make room for the boot section. The next day I came in and Jacques was celebrating that he “found” enough room to make the flying wheels animation. He’d done a good job, so that, rather than back out his changes, we were going to have to find somewhere else to make room for the boot space. Then we had a major snow storm, and many of us stayed late rather than risk the streets. A group of about five of us spent the time poring over Pole Position II printouts to find a few bytes here and there, some of us sprawled on the lab floor, scowling away. We found the needed bytes, but I laugh at how different those tasks were from the software engineering training I’d worked so hard at.
After Pole Position II was shipped, I was brought in late on 7800 Xevious to code in the pattern of attack (which classes of fighters started at which point in the flyover), a straightforward, detailed but non-creative task, like coding in the pattern of rocks and holes in Moon Patrol, with a minimal learning curve for me. The primary design team, who had pretty much burnt out on Xevious, was being pulled onto other projects, so I was like a relief pitcher.
My office-mate Ava-Robin [Cohen] did the 2600 Dig Dug conversion, and she might have also done the 7800 version.
RW: Do you remember any games being worked on independently that never saw the light of day? Or was it pretty much all-arcade-conversion-business, all-the-time?
NA: I don’t know if the Road Warrior arcade game even made it to Beta Test in an arcade (I think they used “1001 Plays” in Central Square on Massachusetts Avenue as their testbed).
We were one of the three companies in competition to design the next generation base unit (ultimately the 7800). The other two competitors were Amiga (then just a graphics chip producer) and some obscure Japanese company called Nintendo.
RW: What happened at the end of your time working at GCC?
NA: Chainsaw Jack Tramiel took over Atari [in July of 1984] and the GCCo. conversion work was being phased out.
We switched to working on games for Apple Lisa computers and the hardware people started developing a hard disk drive for them. Kevin [Curran, one of GCC’s founders] said that, at one point, we had more engineers working on Apple stuff than Apple itself.
Eventually, GCC laid off ~2/3 of the software group (me, included) and switched to being an aftermarket hard disk upgrade for Macintoshes. I visited a few months later and it was so depressing. It looked to me like a sweatshop of assemblers.
RW: After leaving GCC, where did you end up working?
NA: I was an OS software engineer at a datacom switch company called Bytex. When I came in, the software people were put into a “corral” in the cinderblock building while a real building was being built for the rapidly-growing startup. Some of the engineers were rather timid, and were surprised when I would argue with one of the co-founders (from MIT) about design. My co-workers didn’t realize that he actually enjoyed arguing with other engineers as a break from the stuffy company managers.
It was so different from working on cramped software for mass-produced 1970s technology. (I tested my automatic-backup software by pulling processor boards out while the device was running.) The end users of the switch would be routing millions of dollars worth of financial transactions per second, so they paid good money for a device with minimal downtime.
RW: Did you keep anything from your GCC programming days? Source code, promotional materials, etc.?
NA: I have the framed 2600 Moon Patrol box that GCC “awarded” me, and my husband made me a similar one from a 7800 Xevious box.
RW: What are/were some of your favorite video games?
NA: In our arcade closet I became a master of Robotron. I liked Xevious, too, and Battlezone. But mostly Robotron.
RW: When was the last time you played Moon Patrol?
NA: When I was at GCC (though my husband and I used the do the tune as commentary on some household chores).
VIDEO GAME HISTORY PRESERVATION AND YOU
Preservation of the stories surrounding the creation of video games is often just as important as the preservation of the games themselves. In this case, the discovery of an unusual video game artifact led to its programmer sharing her perspective on her time working in the early video game industry. One doesn’t need to stumble across a unique prototype to add to the annals of video game history, though — this is something in which anyone with a genuine interest and passion can take part. You can support any number of preservation institutions and museums, most of which rely on donations and contributions to operate. Or, choose some game history that you think needs to be further examined, and do what you can to help fill gaps in the existing record.
Regardless, an unlimited number of interesting stories remain to be told, and if anything can be inferred from this article and interview, I hope that it’s exactly that. While the number of stories may be limitless, the time to tell them is not. As the pioneers of the early video game industry grow older and the technology that they created and used ages with them, research opportunities and methodologies diminish.
Please, if you have an opportunity to contribute to the documentation and preservation of video game history — and that goes for any of it; even contemporary video game development eventually becomes video game history! — don’t pass it up. If you have source code, development materials, marketing paraphernalia, prototype hardware or games— it all counts! — consider getting it into the hands of someone who will make sure that generations to come can make the most of it.
Follow the author on Twitter at @AtariSpot
