Games are inevitable. People play games. Even animals play games. Pick up a rock and you can make a game with it. Add a stick and you have more game possibilities. It’s just natural for human beings to invent ways to play.
Pretty much as soon as there were computers, there were computer games. Games like Colossal Cave: Adventure and Star Trek, the whole PLATO system, and many others were created on huge mainframe computers, which only large corporations and universities owned. There was no game industry. There were no game companies as we know them today. But there were games.
The first home computers were game machines. Pong was essentially a single-purpose computer, as was the Odyssey. Over time, game machines became more sophisticated, and more general home computers began to appear—the TRS-80, Commodore PET, Altair kits and others. Early home computers weren’t great game platforms, but they always had games.
The first home computers to achieve widespread popularity, the Apple II, Commodore’s VIC-20 and C-64, and the Atari 400 and 800 systems boasted full color displays. This crop of early home computers allowed users to do real work as well as play games, but games proliferated on them, while productivity was hampered by the limitations of the systems themselves and the slow evolution of technology. For instance, at the beginning, the Apple II and Commodore systems relied upon slow, linear tape drives to load and store data. Floppy drives eventually appeared, but during the early years there was no such thing as a hard drive, and memory was counted in bytes, not kilobytes, megabytes or gigabytes. They also printed text all in caps on lines only 40 characters wide. (I wrote my first book using an Apple II with all caps, 40 character lines, using a black and white TV as a monitor. Talk about headaches.)
One game-changing productivity innovation appeared first on the Apple II—VisiCalc, the first real spreadsheet. It might seem hard to imagine that something as utilitarian as a spreadsheet program could revolutionize the world of computers, but VisiCalc was a wonder in its time. It was a very sophisticated toy in some ways, and a major empowerment to businesses and individuals at the same time.
Then came the giant—Big Blue. IBM was a huge, respected, and very serious company. They didn’t make games. They made giant computers, and they were the dominant brand in the mainframe and business world. They were International Business Machines, with emphasis on business.
So when IBM introduced their first home computer, it was a big hairy deal. And when Lotus introduced Lotus 1-2-3, a spreadsheet even more powerful than VisiCalc, the combination rocked the world. And that’s where Microsoft first appears in the story, because they provided the operating system for the IBM PC. The Disk Operating System. DOS. The story of how Bill Gates managed to secure the operating system deal with IBM has been told many times and I won’t go into detail about it here. But this is where the Microsoft story begins.
The effect of the IBM PC combined with Lotus 1-2-3 can’t be overstated. They took the revolution in personal computers and made it legit. They made it business. And they made Microsoft rich, even though they were still small and nowhere near as important as Big Blue.
The First Game
Chronologically, Microsoft’s first game was an unlicensed, but mostly faithful to the original version of Will Crowther and Don Woods’ Colossal Caves: Adventure text-based game for the TRS-80, published in 1979 and later ported to Apple II and the brand-new IBM PC in 1981. The game was part of a new Consumer Products Division that Microsoft formed in 1979 under Vern Rayburn, one of the earliest Microsoft hires. But the story is even more convoluted, and it starts with an argument between Gordon Letwin and Bill Gates.
Letwin had written a disk operating system, H-DOS, for the Heathkit H8 and also his own version of BASIC. But then a very young Bill Gates showed up at Heath Company and started working to convince them to adopt Microsoft’s product. Letwin challenged Gates, pointing out some differences between their products. One of the key features he brought up was that his version would immediately check syntax if you entered a bad line of code, where Microsoft’s product didn’t. Stuff like that. Gates is later quoted as saying, “Anyway, he was being very sarcastic about that, telling me how dumb that was.”
Three versions of Microsoft Adventure—TRS-80, Apple II, IBM PC.
Heath decided to implement Microsoft BASIC instead of Letwin’s version, but in typical Bill Gates fashion, he later hired Letwin, apparently valuing the man’s talent more than his tact. And it was Letwin who produced Microsoft’s first game. Technically proficient, he found a way to extract the content of the original Colossal Caves: Adventure code from the original FORTRAN, and recreate it with only a few intentional changes, mostly in jokes for programers and some simplication of puzzles. He also solved several technical problems with memory and code access by using a memory expander and random access, at the time very expensive, floppy disk to load and access the games, instead of cassette, which was the most common way of loading games like Scott Adams’ series of very popular text adventures.
Microsoft Adventure was never licenced or authorized, but Crowther and Woods never sought any credit or royalties to the game, and so Microsoft was never challenged in their use of the game.
Microsoft Flight Simulator
The game people remember as Microsoft’s first official “game” was Flight Simulator. Originally developed by Bruce Artwick and his partner Stu Moment, and released in late in 1979*, it first appeared on the Apple II and was published by their company, SubLOGIC. An improved version was licensed to Microsoft and was released late in 1982. Although Flight Simulator was not technically a game but a simulation of flight, it did include both a crop-dusting mode and a legitimate game called World War I Ace where you could dogfight while flying a Sopwith Camel.
According to at least one source, Microsoft and IBM both tried to get the Flight Simulator license, but Artwick chose Microsoft. The same (unverified) source also states that Microsoft’s 18th employee, Vern Raburn, was Artwick’s main contact. Although PC version was released as Microsoft Flight Simulator in 1982, it was still developed by SubLOGIC until 1988, which is when Artwick left and founded a new company, BAO (Bruce Artwick Organization), taking Flight Simultator with him.
Flight Simulator was very successful before the Microsoft deal, having sold well on a number of platforms, and the Microsoft version remained consistently profitable for several deccades. By the end of 1999 it had sold more than 21 million copies. Fight Simulator also set a Guinness world record as the longest running video game series (more than 32 years) as of September 2012.
Back in 1982, Microsoft was growing quickly, largely because its lucrative deal with IBM, but they were still in some ways figuring out what they wanted to become, other than the dominant force in the microcomputer industry. So it’s not clear how Flight Simulator figured into their plans, which had, in addition to their operating systems, featured forays into hardware peripherals like the Z-80 SoftCard, an upcoming computer of their own called MSX, and their first mouse; applications such as an early version of Word; and even a book publishing division.
Why, then, did Microsoft acquire the Flight Simulator license? I haven’t found anyone who can answer that question. Perhaps it was an experiment. Possibly someone just saw it as a successful product on other systems, and decided to acquire it as yet another direction for the fast-expanding company to enter into. What is is clear in retrospect is that publishing Microsoft Flight Simulator was not the start of a grand plan to become a major player in the computer game field.
The Added Benefit of Flight Sim
Whatever the reason for having Flight Simulator, it ended up serving a purpose that most people would not have guessed. According to Russ Glaeser, a veteran of both BAO and Microsoft’s ACES studio formed around Flight Simulator, “one of the interesting things about early flightsim was that it basically just went around DOS. It was written in one hundred percent assembly language. It just hit the hardware directly. And it was such a good test of Intel compatibility that that’s what they used it for. That was one of their tests to see if new chips coming out were Intel compatible.” In fact, Flight Simulator was such a good predictor of compatibility that it was used in testing all the way until Windows 7.
Jon Solon had moved from Minneapolis to Seattle largely because of his love of hang gliding, but as an early hire at Microsoft, he began in the applications test division. After doing some testing on Project and Word, he had the opportunity to test Flight Simulator 3 by virtue not only of his hang gliding experience, but because he was also a licensed pilot.
Solon and Artwick were partially responsible for turning Flight Simulator into a Windows testing application. He says, “It was rather dramatic in the sense that Bruce Artwick and I had benchmarked the performance of the early Windows, and we basically submitted a memo to Bill Gates and Brad Allchin, Silverberg—all the main Windows folks—and said that Flight Simulator version 5 and Space Simulator version 1 would NOT be Windows products, and here’s why.” In the memo, they demonstrated the constraits involved in writing to the screen quickly, especially in aerial maneuvers such as banking the aircraft or even moving quickly through the galaxy in Space Simulator. Instead of Windows, he says, “we had to do it in DOS using expanded memory - not extended memory, but expanded.”
Solon says with pride that he and his team were responsible for nudging the engineering team to address Window’s deficiencies in graphics. They would not be the last ones to point this out, but they may possibly have been the first.
*Part II of this volume focuses heavily on Windows’ ability to support games and the story behind the technology that allowed fast-paced games to excel on PCs, despite Windows.)
The Purchase
On December 12, 1995, Microsoft announced the purchase of BAO, stating “Under the terms of the agreement, Artwick will consult with Microsoft in the design and development of new titles while the majority of BAO’s development team will relocate to Microsoft’s Redmond campus.”
Glaeser was one of the BAO developers who relocated. “We were in the process of working on Flightsim 95 when they announced to us that we were all going to be moving to Seattle. At that time we were located in Champaign, which is 120 miles outside of Chicago. They did the transition in a very strange way. They didn’t want to be disruptive, and by not being disruptive, they made it way worse. They basically didn’t want to bring the whole team down for 2 weeks while everybody moved, so they just took one or two people and moved them, one at a time, which meant that some part of the team was unavailable for 2 weeks. And that went on for 6 months.” This was only the first of several awkward transitions of companies acquired by Microsoft over the years. More examples later…
Eventually, things got sorted out and the ACES division of Microsoft continued to develop and consistently improve Flight Simulator until the studio was closed down in 2012.
Microsoft released one other DOS game in 1982 called Olympic Decathlon (also known as Microsoft Decathlon), which had originally appeared on the TRS-80 in 1980 and the Apple II in 1981. Although there were well over 2000 games available for DOS systems—many of them classics—only Flight Simulator, Olympic Decathlon, and Space Simulator (published in 1994) were from Microsoft.
Time Pilots
In addition to becoming a product manager on Flight Simulator, Jon Solon also spent many years working on Space Simulator. Because space is so vast, there were specific challenges to developing a real simulator for space travel in DOS. One of the compromises they made was to use time in a unique way - essentially as a gas pedal. If you wanted to go long distances, you essntially sped up time to get there in a reasonable period of time, instead of months, years, or light years. Solon got a lot of joy out it, though. Coming to work, he would “boot it up and, using the joystick, fly out of Cape Kennedy and do a quick shot to the Moon, one orbit around the Moon, then come back to Earth and glide back down for a landing, using time as a gas pedal.”
During his time on Flight Simulator and Space Simulator, Solon had the opportunity to experience several commercial and military flight simulators and to meet several astronauts - most notably Donald Williams who had flown the Space Shuttle twice and was an F4 pilot during the Vietnam war with 745 carrier landings. Williams was very helpful in many ways, including facilitating several visits to NASA’s Johnson Space Center.
Solon also gave presentations about their work, including one at Cornell University where he was hosted by a professor named Carl Maas, a colleague of Carl Sagan. Sagan keynoted the Cornell Flight Simulation Conference, and after the event, over a banquet, some cocktails, and conversation, “Carl and I slipped out the back door and secretly drove to the professor’s house where we—my boss Tony Garcia, Leigh Cole and I—we showed Space Simulator to Carl Sagan.” They were hoping to get his thoughts on the product.
Sagan was impressed enough to request a repeat showing the next day, where he also invited his wife and son, his father-in-laws, and writing partner, Ann Duryan and her father. “A real thrill for me was rapidly approaching Mars at high speed and then turning the spacecraft 180 degrees to fire the thrusters to slow down so that I could be captured in Mars orbit, and I looked to my right and there sitting on Carl’s knee was his young son, Sammy, 3 or 4 years old, with his eyes as big as saucers, and I thought to myself, ‘What the hell? I’m showing Carl’s boy how to orbit Mars.’”
*Solon kept many momentoes of his time on Microsoft’s simulators, including some prized autographs. See last page of the Online Appendix.
Tomorrow Makers
Former Associated Press science writer Grant Fjermedal started working at Microsoft in the 1980s after writing The Tomorrow Makers: A Brave New World of Living-Brain Machines. Originally published by Macmillan in 1986, it was later republished by Microsoft Press in 1988. “I went to MIT, Carnegie Mellon, Stanford AI Labs, think tanks, DEC in the Midwest and in Japan. I wrote this book about computers, artificial intelligence, robotics, and creating our evolutionary successors,” says Fjermedal.
One of Fjermedal’s readers was young Bill Gates. “I was at a conference at their campus and Gates came up to me and said he’d read the book and liked it. That’s how small the company was. I had a nametag on and Gates just wandered up to me.”
Through his connection with Gates, Fjermedal started working as a contractor at Microsoft. One of his interests was astronomy, and he had just published a book on astronomy for Putnam. He had another book in mind that would combine actual science along with science fiction, and although Microsoft’s publishers liked the idea, they actually had something else in mind for Fjermedal. They said that they had a product in the works called Space Simulator and suggested that he meet with the team and consider writing a book about the new product. He and the team hit it off from the beginning, and while working with them he wrote the book: Adventures in Space Simulator. Next, they asked him if he’d be interested in writing help files for the game. He had never written help files before, but he thought it would be fun.
And so Fjermedal began to carve out a unique role within the organization. He wasn’t a designer, an artist, or even a manager. Sometimes he was a writer, but he was also a researcher who was more than willing to experiment in a new medium.
“So I end up with an office at Microsoft in Building 10 and started working crazy hours—12-14 hours a day just researching this. And in the meantime, whenever there was a delay in Space Simulator, they said would you be interested in doing something else? And I think one of the first things I did for them was the Windows Entertainment Pack. I’d just get the games, start playing with them, and rather than just doing a help file, I would write these wacky stories, like for Blackjack I wrote about this Dr. Blackjack who traveled, played in all the major card places around the world… across the Atlantic, Monte Carlo. And the people really loved it. It was fun. And when the work got hot and heavy on Space Simulator, I had a custom-made futon that was sleeping bag width rolled up underneath my desk, and I kept a sleeping bag there. I kept cereal there. I was getting quite well known for sleeping in my office.”
Wacko UI
At one point, Fjermedal was asked to write help files for Fight Simulator, which was still a DOS game at the time. He set up three computers, one playing the Mac version of Flight Sim, and the two others playing the current and upcoming versions in DOS.
What struck Fjermedal almost immediately was how confusing the user interface was. “Even the Mac commands were pretty wacko.” And as for the DOS versions, they were just illogical and cumbersome. So he wrote a 4-page memo that proposed a complete restructuring of the UI that went back to BAO and was adopted.
Gates Sees the Future and It’s GUI
Despite these early forays into game publishing, Microsoft didn’t see itself as a game publisher, and with DOS looking more and more outdated after the successful introduction of the Apple Macintosh, one of Microsoft’s most serious initiatives was to create a new operating system that could move them into the modern age of graphical user interfaces. Bill Gates had seen the writing on the wall as early as the late ‘70s when he, like Jobs and many others, saw the Xerox PARC GUI, and perhaps again in 1982 at COMDEX, where he saw the early PC graphical interface of Visi On from VisiCorp.
Even as late as 1984, Gates was working closely with Apple, sending top programmers to create versions of Word, MultiPlan, Chart, and File for the Mac debut. DOS was very successful for many years, but the future was GUI, and Gates knew it. Apple’s Macintosh without doubt signaled the oncoming obsolescence of DOS, and though they were fierce competitors, Gates and Jobs were sometimes collaborators, sometimes parties to lawsuits over the ensuing years.
But there was a time when they were friendly, as this quote from http://www.mac-history.net/apple/2011-01-30/microsofts-relationship-with-apple reveals.
“With the rise of the Apple II in the late seventies Microsoft became more and more successful – even before the IBM PC was invented. When Apple developed the Macintosh Bill Gates and his team were the most important software partner – despited the fact that Microsoft was also the driving force behind the IBM PC and the PC clones. And Steve Jobs even invited Bill Gates for the preview of the Mac: The high point of the October 1983 Apple sales conference in Hawaii was a skit based on a TV show called The Dating Game. Jobs played emcee, and his three contestants, whom he had convinced to fly to Hawaii, were Bill Gates and two other software executives, Mitch Kapor and Fred Gibbons. As the show’s jingly theme song played, the three took their stools.”
To make a long story short, when Jobs refused Gates’ offer to help Apple license the Mac OS, amid falling Mac sales in 1985, Microsoft responded by launching Windows 1.0, which soured the relationship and resulted in a lawsuit by Apple against Microsoft and Hewlett-Packard, and an ultimate compromise between the companies. Yet Windows, weak as it was at the time, was destined to take over and shake up not only the computer industry, but the game industry, which unlike Microsoft, had been very busy publishing DOS games and wasn’t eager to change to a new operating system.
Almost Zorked
Zork was one of the most popular text adventure games in the early 1980s and the product that launched the quintessential text adventure game company, Infocom. In 1980, Joel Berez remembers sending a query about Zork, which was first launched on a PDP-11, to several publishers, including Microsoft. “We got a letter back from Vern Rayburn, who I think was one of the top people at Microsoft then, telling us that they already had the Adventure game, and didn’t really think the market would support another text adventure.” Berez says that they made a deal with Personal Software to publish Zork on TRS-80 and Apple II, but after a year, the company decided to focus entirely on productivity software, specifically the groundbreaking spreadsheet, VisiCalc. They got back the rights to Zork and Berez, on the verge of graduating from MIT’s Sloan business school, conferred with his team and they decided to become publishers of their own products. “In retrospect Rayburn did is a great favor by rejecting Zork.”
Years later, at a launch event for a new PC clone, Berez met Bill Gates for the first time. Somehow the conversation turned to the issue of Rayburn’s rejection of the game to which Gates responded that he had known about Zork and would have overriden Rayburn’s decision, adding “If I had gotten my hands on it, you never would have gotten it back.”
In an aside, Berez mentioned that Mitch Kapor, the founder of Lotus, the developers of the breakout spreadsheet product Lotus 1-2-3, was a classmate of his. He adds, “I believe that Mitch spent some time in marketing at Personal Software, which perhaps is where he got the idea to create his own version of the spreadsheet.”
*According to one source, on Dec. 31, 1979