Sponsored By

Featured Blog | This community-written post highlights the best of what the game industry has to offer. Read more like it on the Game Developer Blogs.

The History of a Forgotten Computer - PART 1

This is the true story of how one if the first home computers came to be. And the small group of people who created the software for the system. This is part of our history and a story I hope everyone will enjoy. Part 1 of 2

Matt Powers

April 25, 2014

28 Min Read

The first article I ever wrote for Gamasutra was entitled, "A Brief History of Videogame Development."  That article had a brief look into a story about a group of early game developers.  That short story of mine piqued interest in a number of people who asked me to pursue this fascinating tale further.  And so I did, and here you have it.  I hope you enjoy this look into some of our video game history that was pretty much unknown, until now.  - Matt Powers.

Startup screen from "Goldilocks and the Three Bears" on the CyberVision


A number of computers were created in the 1970's.  They included the IBM PC and Apple.  Microsoft was just a startup company.  The hobbyists were using soldering irons and circuit boards and working out of garages.  Computers intrigued a number of young people who went on to be the founders of both the computer and video game industry.  This is a story about one of these early computers and the group of people that made it.  The computer they created was all but lost to memory though the people went on to make their mark in the computer industry. 

PART 1 - It was a dark and stormy night...

This story is true.
The names have not been changed.
And no one is innocent.

It had been raining all day and the stormy weather continued into the evening.  The phone rang.  Insistently the ringing continued throughout the house.  Janey answered the phone and after a brief exchange with her husband she hung up.  "Strange conversation," she thought; "What are those guys up to?"  Her husband John and his friend Joe had gone out earlier on an errand.  John had just called to tell her that he probably wouldn't be coming home tonight.  He and Joe were going to pull an all-nighter.  As he put it, "We have an operating system we need to write."

John and Joe were driving John's recently purchased used car, a 1968 VW Westfalia Camper.  It was a nice car, didn't have many miles on it, and had enough room for the family.  While used, some new paint had been added to the van - Janey had recently painted a peace sign on the spare wheel cover and balanced off the peace sign with bright yellow smiley faces on all four hubcaps. 

John and Joe were driving in downtown Columbus, Ohio, and their errand was to purchase a CompuColor 8001, a microcomputer with color display and built-in BASIC language.  They had recently started their own business and had a contract with Warner Cable in Columbus to develop a real-time network monitoring display system.  They had called various shops around Columbus and found a dealer for the CompuColor 8001. They were on their way to pick it up.

The VW camper parked in front of the small electronics shop.  This was not the best area of town; there were vacant lots all round - the shop looked quiet, but appeared to be open for business.  It had been raining all night and the pair of men were glad to finally find their destination.

They approached the store and went in the front door.  The shop looked empty, but the jingling from the bell on the door must have alerted someone.  They told the clerk what they were looking for.  After inquiring about what they were going to use the CompuColor 8001 for, the clerk asked them if they were programmers.  Joe and John said they were. Without saying anything, just a gesture to wait there, the clerk walked to the back of the shop.  He came back after a few minutes and asked John and Joe to follow him.  They went to the back of the store and up a flight of narrow stairs to the 2nd floor.  They entered a small, shadowy room containing a table strewn with electrical components.  At a bench near the door was Jim, the owner of the store, hunched over a circuit board with a soldering gun in his hand.  Three other men were hovering around a table in the middle of the room, obviously anxious and distracted.  There was a aura of urgency and tension in the room.

Joe and John cautiously moved forward until they were able to make out what everyone's attention was focused on.  There were blue prints containing circuit diagrams, circuit boards, wires, and various other electrical components strewn about the table.  One of the men looked up, eyes unfocused, clearly just becoming aware there were strangers in the room.  After adjusting his glasses, the man straightened, looked at John and Joe and asked, "Do you know anything about computer programming?"  

John and Joe looked at each other - what have they stumbled into?  The men around the table explained that they were working on a prototype of a home computer to demonstrate to a conference of Montgomery Ward buyers.  If the buyers liked what they saw, Montgomery Ward would sell the home computer in their stores and put it in their store catalog.  It was a multi-million-dollar opportunity.  The prototype was working, but the designer/engineer had only written a crude demonstration program.  They needed a full-blown software suite if they were going to win the contract.  Joe and John asked when they needed to do the demonstration.  After furtive glances to each other, one responded, "Monday, two days from now."  After some brief negotiation on what they would charge, John asked for a phone.  He needed to call his wife - he and Joe would not be coming home that night.

This occurred in 1977.  In 1977, microwave ovens were just becoming available - the biggest wattage was 625 watts, all the controls were analog.  There were no cell phones; telephone answering machines used cassette tapes.  The largest TV available was 25  inches (diagonally).  People were buying 8-track tape players.  A component stereo system was a record player with two 5 inch speakers. 

In early 1977, Steve Wozniak, Steve Jobs, and Mike Markkula filed for incorporation of Apple Computer.  The corporation purchased the Apple Computer partnership for $5,308.96.  That same year, Bill Gates and Paul Allen signed a partnership agreement to officially create the Microsoft company.  Also, in 1977, A working model of the first Radio Shack computer was demonstrated to company president, Charles Tandy.  That same year Commodore showed a prototype PET computer at the Winter Consumer Electronics Show.

That evening in 1977 John and Joe created the operating system for the computer which would become the CyberVision, a computer that eventually was sold by Montgomery Ward.  This computer, while one of the first groundbreaking computers, and the first truly multi-media computer, is barely known today.

And, John and Joe did buy that CompuColor 8001 they were looking for.  And, they finished their project for QUBE (the world's first "two-way" cable system).  But, significantly,  that night also started John, Joe, and their company down the path of video game software development.

op·er·at·ing sys·tem
noun: operating system; plural noun: operating systems

The first task for John and Joe was to write a rudimentary Operating System (or "OS") for this new machine.  They needed to design and write - basically create - the fundamental functions which would operate this piece of hardware. 

Operating systems perform basic tasks, such as recognizing input from the keyboard, sending output to the display screen, keeping track of files and directories on the disk, and controlling peripheral devices such as disk drives and printers.  http://www.webopedia.com/TERM/O/operating_system.html

This needed to be done in machine language and entered all together, without error, in hexadecimal.  Hexadecimal is a numbering system used by computers which is a base 16 system(going from zero to F which is 15) versus our common numbering system which is base 10(zero though 9).  For example, in hexadecimal the number 14 would be E.  The number 23 would be 17.  To get a program into this early computer the instructions needed to be entered one hex number at a time.  It took multiple tries to get this 2K block of binary data debugged and entered flawlessly so it could then be burned into a ROM. 

Read-only memory (ROM) is a class of storage medium used in computers and other electronic devices. Data stored in ROM can only be modified slowly or with difficulty, so it is mainly used to distribute firmware (software that is very closely tied to specific hardware, and unlikely to need frequent updates).
 (From Wikipedia.)

To help ensure the code would work, John and Joe wrote out the code on paper and then John read out the numbers as Joe entered them using the keypad.  As Joe recalls:

I literally locked up while entering an endless string of hexadecimal numbers into the prototype.  This was at the end of two (or three?!) days of non-stop design, programming and hand converting 1802 assembly language into raw hex so we could bootstrap the system.  I hit a proverbial wall and couldn't continue to do something as simple as entering the hex digits as John was reading them to me.

I left my car at the office, and John drove me home to my apartment where I probably slept for 24 hours.  - Joe Miller

The first "bench" prototype of what become the CyberVision didn't have a ROM image, but it did have RAM at the address the ROM would eventually be. 

RAM (pronounced ramm) is an acronym for random access memory, a type of computer memory that can be accessed randomly; that is, any byte of memory can be accessed without touching the preceding bytes. RAM is the most common type of memory found in computers and other devices, such as printers. (From Webopedia.)

So, John and Joe had to load the control logic for CyberVision (i.e. the OS) by hand into that address range and debug it, hoping that no one kicked the power plug out and wiped out the system.  If the power was lost while entering the code they would have to start over from the beginning.  If a wrong number was entered or any mistake at all, they would have to start over.

Once the team had that system code debugged, they were able to burn it into ROM and the system would run that on power-up.  After that point, they didn't need to manually enter long byte strings.  They could then load data from tape.

Joe and John completed the OS and demonstration software in time to show to the Montgomery Ward buyers. The contract was secured and a home computer business was launched.

relating to or using a system of numerical notation that has 16 rather than 10 as its base.

The Business Meeting

The offices of the company John and Joe founded, Authorship Resource Inc. (ARI), were located in The Atrium on Doubletree Avenue in Columbus, Ohio.  Entering the front door of The Atrium, one was presented with a small indoor forest of plants and paths.  Each company's front office faced out into the "jungle."  Visitors to ARI strolled through this indoor forest to their office.

ARI had just moved into their offices and had the latest in office furniture that could be found at used office furniture resale shops.  Janey and Joe did the shopping.  The best furniture was reserved for the front office.  They wanted to give a good impression to the outside world.

Real wood was out of the price range, so a tasteful desk laminated in brown fake teak wood faced the window looking out at the atrium.  Janey spent many hours at that desk typing letters, reports, and announcements on a gently-used IBM Selectric (or "golfball") typewriter.  This wonder of the age came with a box of golfball type heads so you could print using italics, bold, and various fonts.  An IBM ad at the time for the Selectric reads:

The IBM Selectric is a unique typewriter.
Conventional machines have 44 type bars...and moving carriages.  Not the Selectric.  Instead, it has a golf ball sized typing element carrying a complete set of 88 characters and symbols.   This element sprints across the paper, typing out character faster than the eye can follow.
From:  http://ak.pinterest.com/pin/412712753322035077/

A wooden love seat with open arms and imitation brown leather cushions, a matching coffee table with a chrome table lamp, and a huge potted plant completed the look.  Later on, a hanging philodendron in a macramé basket was added to complement the outside atrium plants.

ARI was a no-tie office.  The employees dressed in what today we would call business casual.  This included high-waisted jeans and trousers with wide, flared legs.  As Janey recalls,

"We wore these pants with wide bottoms called 'bells.'  There were also pants with even wider bottoms called 'big bells.'  When you wore these bells you had to watch out for the furniture.  If you walked too closely to a table leg, the bell bottoms could grab the leg and then get wrapped around the furniture."

One of the perks of the location of the ARI offices was that it was located across the street from a Anheuser-Busch brewery.  And that brewery had a tasting room for visitors and tour groups.

ARI acquired the contract to create all the software for the CyberVision.  One of the first business meetings ARI had regarding the CyberVision was held at the Anheuser-Busch tasting room.  It seemed like a good idea at the time, free beer along with their meeting.  This was a time where most business was a coat and tie affair.  But John, Joe, and their associates were part of a new era of business - the software business.  No ties required. 

After the meeting adjourned, the software developers were leaving the tasting room when one of the staff from Anheuser-Busch stopped John. 

"Excuse me sir, it appears you were having a business meeting in our tour group tasting room?"

"Yes," John replied, "Thank you so much; the beer was excellent; the meeting went very well."

The employees response was short and to the point, "In the future, could you please not have your business meetings in our tasting room."

What is in a Name?

Everything was moving quickly around this time.  The hardware was being finalized and manufacturing was to begin soon.  With the Operating System completed, the team at ARI was off and running to actually develop the first packages of software to be sold with the new machine.  And it was exactly that, just a machine - it needed a name.  Montgomery Ward's needed the name finalized for the catalog and the team at ARI needed the name to finish packaging design for the cassettes, ROM startup screen, etc.

It was Joe who finally came up with the name, "CyberVision."  As Joe recalls:

It just seemed obvious to me that this was a new application in the living room for the whole family that would eventually be as ubiquitous as the television.  Cybernetics was a general term for the science of communication and became associated with the computer revolution at the time.  (CDC computers at Battelle were called Cyber 170s, for example.)  So, I just did the obvious and created the combination of cybernetics and television, CyberVision.   - Joe Miller

The year was 1978

In November of 1978, television audiences were able to watch the Star Wars Holiday Special which aired on CBS.  1978 was quite a year for computers and technology.  Some other notable achievements included:

  • The Xerox Corporation created a TCP enhancement: Internet Protocol (IP), a separate program that handled the routing of individual messages.  Combined with the TCP protocol from 1979, the two became known as TCP/IP and represented the standard system used in most large networks.

  • The first BBS (Bulletin Board System) created by Ward Christen (USA), member of a Chicago user group, went "on air" that year.

  • Epson announced the MX-80 dot matrix printer which established a new standard in high performance printing for a low price.

  • INTEL introduced the 16-bit-8086 chip and the coprocessor 8087.

  • PC's grew into complete systems: keyboard, display, floppy drives, memory, BIOS etc.  Memory was expanded up to 16 Kb.

  • Apple Computer demonstrated its first working prototype Apple II disk drive at the Consumer Electronics Show in Las Vegas.

The cover of TIME magazine in February of 1978 featured a collage of images featuring a digital calculator, a digital watch, C3-PO, and a PDP-11/34.  The editor of Time wrote in that issue about the new addition to the Time staff, the PDP-11/34:

PDP etc. could hardly have arrived at a more propitious moment, for in this issue Time presents a special 15-page section entitled “The Computer Society.”  The report explains just how the world of electronic sorcery works, and examines its impact on our daily lives.  To make such a complicated technical phenomenon understandable, a team of six correspondents, five writers, four reporter-researchers and three photographers spent a month interviewing scientists, visiting manufacturing plants and trying out the newest and most exciting computerized products.
- TIME Magazine, February 20th, 1978

Time Magazine, February 20th, 1978

That Christmas in 1978 the CyberVision 2001 first became available to Christmas shoppers at Montgomery Ward. 

Making One of the First Video Games

For the rollout of the CyberVision, ARI needed to develop a suite of software.  The demos to Montgomery Ward were successful, and they were ready to stock the CyberVision 2001 in their stores.  Now the rush was on to create the content, the games for the system.

There were also some games for the awesomely named CyberVision 2001, Montgomery Ward’s brief 1978 entry into the personal computer business. This $399 1802-based machine added color to the equation.  It’s also one of the most obscure and rare micros ever made; there are just not many machines or “cybersettes” of games in the wild.  It was finding a CyberVision ad in an old Ward’s catalog that sent me on my 1802 vision quest and led me to CHIP-8…

From:  http://incrediblystrangegames.com/post/68960258945/do-you-recall-the-least-popular-microprocessor-of-all

One of the first games they developed was called Sub Chase.  John developed the game but at this time everything still needed to be entered into the computer via keypad by hexadecimal, a long and tedious process that involved many, many hours hunched over "the Cosmac." 

Startup screen from "Sub Chase" on the CyberVision

New processes had to be developed for the creation of the software on the CyberVision.  ARI was blazing a path that few, if any, had gone down before.  And because of the limited nature of the system they were creating software for, they found that more preproduction made for better development.

Screenshot from the CyberVision game, Close Encounters of the Third Kind

From the beginning, ARI developed a set of tools and a process that made it easier to develop apps for the system.  They made a practice of developing storyboards of all the titles on special templates they had printed.  ARI also had templates that made it easy to convert bitmap artwork into strings for the app, and they built a library of "special effects" that was reused across apps to allow an author to request a range of graphic effects (wipes, animations, etc.)  And, because ARI used studio-recorded audio in most of their apps, they pre-scripted everything, including sound effects, so they could get in and out of the studio efficiently. 

I have samples of all of these forms somewhere, and I have the original storyboard that Brenda did for the "Close Encounters" demo we did in New York for Barry Diller and Michael Eisner, both of whom were at Paramount Pictures in 1978.  - Joe Miller

ARI had a staff of people now working to create the games needed for the CyberVision.  The backgrounds of these people were varied as  "software developer" wasn't something you could actually advertise for at the time.  Previous experiences varied from educator, writer, actress, to radio disc jockey.

One of these early employees at ARI,  Jeff Schwamberger, recalls some of the tricks they learned while developing the software:

PATTERNS. We were all getting pretty tried of creating graphics with just vast expanses of solid color.  I don’t remember the exact details, but red was something like hexadecimal B4, green was C4 and so on.  Either by accident or design (probably the former), I discovered that if you used, say, B5 instead of B4 for solid red, you’d get a small checkerboard pixel, letting us display patterns — vastly adding to richness of the images.  I’m joking of course.  But anything was an improvement over solid color.

A5s. A5s were really a pretty brilliant technology for syncing images and logic.  But they were tricky devils.  We’d expanded and occupied more space, moving into the next door offices.  After arranging all the equipment, Joe had spent hours bundling cables to keep them from hanging loose from the back of the machines.  The first time we tried to master one of the programs I’d developed, it was a complete disaster.  Nothing would sync properly.  I thought it was my code. It turned out that bundling the cables — e.g., a data cable with a power cable — was causing electronic interference that was garbling the A5 signal.  Letting the cables dangle was the solution.

MORE A5s. When we mastered the programs, you could actually hear the A5s as you laid them down — a brief, but distinctive electronic pulse.  They were (to me at least) the most nerve-wracking part of the program to get right.  After one very late night session mastering a program, I went home and fell into a fitful sleep.  Maddeningly, every dream segment was initiated by an A5.  Talk about feeling like some sort of cyborg.

- Jeff Schwamberger

The first CyberVision units contained:

  • 1802 microprocessor

  • 1K ROM

  • 4K RAM (2K program memory and 2K video memory)

  • The big breakthrough was the cassette system. The CyberVision used a standard stereo cassette with one channel for audio and one channel for data. The data channel was 2000 baud. Because it was a two-channel playback, the data could be synchronized with the audio.

The software development system for the CyberVision:

  • RCA Cosmac Development System, Model CDP 1802

  • 1802 microprocessor

  • Two 8-inch floppy disk drives

  • All programming was done in 1802 Assembly Language.

One of the other earlier games was a version of the traditional game, Hangman.  As one of the early employees of ARI, Brenda Laurel, recalls from a talk she gave in 1979:

It was this tiny screen 100-something by 50-something.  I can't do the powers of age stuff.  But it was a little screen and four colors.  And when this guy came in, all he could do was show the alphabet on the screen, but he had placed 10,000 units of it with Montgomery Wards and had to show them on Monday.

So Joe said he would give me a refrigerator full of Dr. Pepper and we'd work with it.  By Monday morning we had some dancing letters and an address book.  These were like radical ideas.
This was the first graphical display I had seen in those days.  So I gave up full-time acting work to work for Joe because he had this idea that Montgomery Ward thought we should build interactive fairy tales.

So I'm coding.  Within two weeks I'm writing code.  It's like a bee flying, you know.  And I'm doing lip synching.  Romanian hangman with 16 lip positions.  We had 2-K of RAM, loading at 200 BAUD from cassette tape. And when there was a drop-out you were hosed.  You know, one little piece of oxide fell off and it was over.

- from Brenda's 1997 talk to the San Francisco Webgrrls - http://gos.sbc.edu/l/laurel.html


Janey, Jeff, Brenda, and Ken at the ARI offices - 1978

The Coding Process

In the ARI offices the most used equipment was the development system. 

It is important to remember that at this time there were no PC's, no laptops, no desktop computers, no text editors, no debuggers, and so forth that are so common today.  The developers wrote everything in assembly language.  And there was no sophisticated tools for debugging.  Today coders use sophisticated debuggers with which you can watch the code execute in real-time, pause execution, examine code in memory, etc...  None of that existed at this time. 

The first step after storyboarding the application was to map out the code with code diagrams.  This was done by hand with pencil and paper.  There were plastic templates that you could use to draw the various logic symbols.  Next they would write the 1802 assembly language that would do what the code design said.  The code was written by typing it on a DecWriter Keyboard connected to the COSMAC.

Brenda Laurel and Ken Balthaser at ARI
The DecWriter is the keyboard at Brenda's left

As Ken Balthaser helps explain:

There was no monitor so we had to read what we had typed directly from the paper.  When you were done entering your code you would invoke the COSMAC to assemble it and produce the binary file.  You would ready the CyberVision to accept the download and then download it.  When complete, the code would begin to execute (hopefully).  The CyberVision(CV) was hooked up to a TV set sitting next to it so that you could see what your code was doing, or not doing.  During execution, those LED's on the CV would blink madly, showing what part of memory was executing.  The series of LED's represented hexadecimal numbers that represented the memory location where the code was executing.  These LED's were the only clue to what was happening.  So, if the program started to run and then suddenly got hung up, the LED's would show, in hex, where in memory the program stopped.  - Ken Balthaser

As for the non-programmer coders at ARI, they used the MICAL system, mapping out their graphics screens and sprites (even though they weren't technically sprites) using the special coding forms.  MICAL was a binary interpreter invented by John and Joe to make it easier to program the CyberVision.  It simplified certain things like erase the screen, put a graphic image (sprite) at a certain X and Y on the screen, display a full screen of a graphic, erase a graphic and so forth.  With these primitive instructions they were able to create graphics and animations.  If MICAL would not do what was needed, Joe or John to write a special subroutine that could be called.  A subroutine library system was created - there were binders of printouts of special subroutines and a master list  of their names and what they did.  Included in the binder were floppies containing these routines.  As new ones were created, they were added to the library and the MICAL coders could use this library to enhance their applications.  

When writing code for the CyberVision the code size was a very important factor.  The CyberVision had just 2K of RAM for code and 2K of RAM for the display (which was 128 x 96 pixels).  So, the team always had to be careful to keep their chunks of code under 2K.  This was all being done at a time when there were no monitors much less development environments or debuggers.  Anything that was needed to create software had to be generated from scratch - including all the tools and processes. 

Manufacturing and demos

The CyberVision used an audio cassette to store its data.  It had a stereo head for playback; the left track was for audio and the right track was for data.  It was common back then to encode data on a cassette, but the CyberVision's data rate was significantly faster than any other computer of its time - 2,000 baud (bits per second).  The applications that ARI developed for the CyberVision used both channels.  While the audio was playing from the left track, data was loaded from the right track.  Some applications used the audio to tell a story and the data to synchronize the animation.  A string of "A5" hexadecimal bytes was recorded as a marker to signal that the program could go to the next step.

The ROM image was duplicated by RCA in New Jersey.  At that time in the industry, a customer for a ROM had to physically sign off on a number of forms that represented steps in the manufacturing process.  It sped up the process by several days to weeks for someone to be there at the plant so sign off on these steps could occur immediately.  As Joe recalls:

On Monday, 8/15/77, I hand carried the ROM image to an RCA plant in New Jersey to start the manufacturing process.  So, at that point the ROM was "locked" in for the initial units, less than 30 days from that initial mysterious meeting on Broad St.  To that date, John and I had expended over 615 man-hours on the project pretty much consuming all of our waking hours, at the expense of other ARI projects and customers we had at that time.  - Joe Miller

The manufacturing of the audio cassettes was done in St. Louis.  Initially the duplication of the cassettes was not straightforward due to the unique nature of the 2000 baud circuit in the CyberVision and how it read the tape.  Again, Joe Miller recalls:

The story about duplicating the tapes is a long one so I won't go into it in detail here, but the short version was we finally discovered that due to the unique nature of the 2000 baud circuit we needed to "regenerate" the digital data track as the tape was being duplicated.  I wrote a small program that ran on a specially modified CyberVision unit, called REGEN, that took care of this problem and we were finally off and running with tape duplication for the initial units.  (We solved this problem another way later and didn't need to use REGEN for high speed duplication after this first run.)  - Joe Miller

During those early presentations things didn't always go as planned.  The technology was new and not perfect.  When there would be a software glitch, Joe's usual response would be to remove the cassette tape, bend over and blow on the recorder heads.   For some unknown reason, this almost always solved the problem.


The History of a Forgotten Computer

If you liked this article or have any questions about it please leave a comment.  Keep a look out for Part 2 which should be available within the next week.  For more articles written by Matt Powers you can visit:


If you would like to contact Matt:  [email protected]



Read more about:

Featured Blogs

About the Author(s)

Daily news, dev blogs, and stories from Game Developer straight to your inbox

You May Also Like