by Steven Kent
One of the first programmers Atari hired was Alan Miller, a graduate of Cal-Berkeley who had become addicted to such early coin-op games as Space Race and Tank during his final year of college. He responded to an Atari employment ad in 1977.
My first interview was with Larry Wagner, the guy who headed up the software group, and one of their hardware engineers, Joe Decuir.
The boss of the Micro Electronics Group, Bob Brown, was away at the time I was interviewing, so I didn’t have a chance to meet him. Bob was one of the technology leaders who I really respect. He understood technology and how to motivate technical people. I was very happy to work for him.
—Alan Miller
Unlike the engineers in the coin-op division who enjoyed socializing as a group, the people programming games for the Video Computer System formed cliques and never integrated. Owen Rubin, a coin-op engineer who helped design computer development systems for the consumer division, described the environment as “cutthroat.” Coin-op engineers generally stayed at Atari for years, but VCS people seldom lasted more than a year or two.
I was just there for a year and a half.
There really were kind of two rival cliques when I was there. There was one group: Al Miller, Bob Whitehead, Dave Crane, and Larry Kaplan, who, for some reason, kind of formed their own little clique.
Me and my two friends, Tom Reuterdahl and Jim Huether, were sort of another clique.
—Warren Robinett, early VCS programmer
Life in the consumer division was not as wild as it was in coin-op. The programmers did not engage in petty battles with the maintenance department or go to movies. They did not pull pranks on their managers. Coin-op engineers worked in teams, consumer programmers worked alone. Most important, through its first year and a half, the consumer division was a financial black hole. Coin-op earned millions of dollars, much of which was used to cover losses accrued by the consumer division.
Unlike coin-op engineers, who wrote their code and then gave it to data entry people to input, consumer programmers entered their own code. Wagner generally assigned them their first game, but they were usually allowed to come up with their own concepts after that.
It typically took about 3 or 4 months to make a game, so the process was very fast paced.
Most of the first games were based on existing coin-op games. I came on board and they assigned me to do a game called Surround that was similar to a number of arcade games. At least I didn’t have to come up with a concept. That helped speed things along.
—Alan Miller
In the beginning, programmers were responsible for creating every element of their games. The same person who created the concept was also responsible for the programming, art, and even the sound effects. The look of VCS games improved, however, when an artist named Marilyn Churchill was brought in from the marketing department to help. A lot of the first games for the VCS were based on popular arcade and board games of the time.
As they started their second generation of games, Atari’s VCS programmers came up with their own concepts, developed new skills, and accelerated up the learning curve. To make their deadlines, programmers worked long hours, for which they felt inadequately compensated.
After my first game, I was given free rein there to come up with all my own concepts and implement them myself. We did all the implementation in those days, including the music and the graphics.
We did the music, what little there was. I can’t say I’m a great musician, but I like music a lot.
I thought making games was a pretty creative, unique act that warranted compensation. I think I was making $27,000 or $30,000 a year at that time. It was not aggressive engineering compensation, frankly, even for that era.
—Alan Miller
The first games for the VCS were fairly plain. After designing Surround, Alan Miller went on to create cartridges based on Hangman and Concentration. Warren Robinett’s first game was Slot Racers. Other early games included Pong and Breakout.
Within a year, however, the games became more intricate. David Crane created a football game and Miller created Basketball, a home video game shown from a 3D perspective.
Of the games I did at Atari, the one I like the best was the Basketball game. It was one-on-one or one against the computer. VCS hardware was designed explicitly to do the Tank kind of game, and the basketball game took sports to a new level of realism on the VCS. It had a really good playability.
I was on the basketball team in high school and loved the sport. Wish I was better at it.
—Alan Miller
Change Comes to Atari
The team had four games ready by the time Atari unveiled the VCS at the Consumer Electronics Show in June 1977. By the time the system was released in October, there were nine game cartridges: Combat, Street Racer, Air-Sea Battle, Surround, Blackjack, Basic Math, Indy 500, Video Olympics (variations of Pong), and Starship.
The Video Computer System retailed for $199 and came with a cartridge called Combat. Designed by Larry Kaplan and Larry Wagner, Combat supposedly contained twenty-seven unique games. Most of the games, however, were variations of the arcade game Tank. In fact, the menu of games on the outside of the Combat cartridge contained the following list:
1–5 TANK®
6–9 TANK-PONG™
10–14 INVISIBLE-TANK™
15–20 BIPLANE
21–27 JET-FIGHTER™
Most of the variations in the games involved trading out bullets for missiles and empty battlefields for mazes.
Despite shipping problems and slow sales through Christmas, the VCS outsold the Fairchild Channel F. This was not enough for Warner chairman Steve Ross. He was disappointed with Atari’s sales and wondered if purchasing the company had been a mistake. Bushnell’s answer, abandoning the VCS and developing a more powerful home console, infuriated him.
Bushnell’s struggles were well known throughout Atari. When he was forced out of the company in 1978, many people believed the company had lost its soul. They expected Warner to impose a stricter culture. When Ray Kassar introduced himself as Bushnell’s replacement at a company-wide meeting, nobody trusted him.
Ray came on to run the company about a year after I was there. I don’t have positive feelings about him at all. He had no understanding or appreciation of the industry, no understanding or appreciation of fundamental technology. He was destined to run that company into the ground.
—Alan Miller
Naturally, when I came, they were all very suspicious. People get nervous when any new guy comes in. They were afraid, I don’t deny that, but I really had great respect for the programmers because I knew that’s where the products came from. I did everything to encourage them.
—Ray Kassar
Kassar’s East Coast “high society” mannerisms offended many Atari employees. The programmers and engineers at Atari did not care about Kassar’s Harvard credentials or his tailored suits. They viewed these traits as peculiarities and made fun of them in a rash of Kassar-jokes.
In Kassar’s mind, his job was to raise corporate revenues. As he surveyed Atari’s situation, he recognized that the company needed a focused marketing plan for its consumer products. He needed to build a quality-assurance program. Sears was complaining about defective VCS units, and no one knew how to respond. In fact, according to Kassar, the company’s relationship with its retail partners was beginning to disintegrate.
Kassar was ready to fix Atari’s problems, but he demanded rewards for his work. He filled his office with expensive furniture and converted the executive dining room into a place of fine dining. Chefs from many of San Francisco’s most expensive restaurants were brought in to prepare daily meals.
Ray came on board to run the company and, frankly, nobody liked him. He was just a totally different kind of guy. The tech guys used to wear shorts to work. Ray was a really high society type and people just couldn’t identify with him. Didn’t like him. I don’t think it was because of anything that he did ov
ertly, it’s just that he was different.
He always wore very heavy cologne and you could literally smell if he had been through the area. People used to joke about that all the time.
—Alan Miller
Like Kassar or hate him, one fact that no one could dispute was that Atari grew exponentially under his supervision. In 1977, the year before Kassar became CEO, Atari had $75 million in sales. Under Kassar, Atari became the fastest-growing company in the history of the United States,* as the company’s sales exceeded $2 billion within three years.
We went from $75 million to $2.2 billion and made a lot of money. They don’t talk about all the money we made for the company. One year we made $400 million after taxes. It was the most profitable company in the world.
—Ray Kassar
The First Easter Egg
They decided that security was really important, so they installed one of those magnetic keypad systems. It was universally hated by all the tech types, who were sort of anarchists anyway. You had to have your little electronic key to get through the doors, and there was no way to get around it.
One night Warren Robinett went down to the cafeteria to get some food. It was late, and he had forgotten his wallet upstairs. His key, the little mag card he needed to get back in the office, was in his wallet [so he was stuck]. He started looking around the building for some way to get back in and he found that the tool room for the coin-op people was open, so he took some tools out and literally broke down the door to consumer engineering. The alarms did not go off.
It turned out that the security system recorded the comings and goings of legitimate employees. You could break down the door and it would not register a thing.
—Alan Miller
Atari’s coin-op engineers felt that Ray Kassar did not appreciate their accomplishments and accused him of paying attention only to the game designers in the consumer division. What they did not know was that the consumer division designers also disliked him.
By all appearances, Kassar did not trust his employees. Shortly after taking over, he had an extensive security system installed. Employees had to carry magnetic identification cards to enter buildings and secured areas. Though electronic-security systems were fairly standard in the computer industry, many Atari employees counted Kassar’s increasing security as one more step toward destroying the company’s relaxed culture.
Kassar’s policy about programmers not receiving publicity infuriated all of the company’s designers, but one consumer programmer, Warren Robinett, found a way around it. He had just finished his first game, Slot Racers, and decided to make his next project a graphic version of Adventure, the pioneering all-text computer game created by Will Crowther and Don Woods.
Like the computer game, Robinett’s Adventure would take place in a medieval world with dragons and caverns. The original game, however, took place in an enormous universe. In order to beat the game, players had to create maps. Because Robinett’s game was for the VCS, it was restricted in size to 4K of code. The VCS, with its memory limitations and joystick controller, was not suited for text-based games. Robinett had to draw his dungeon and dragons.
I played Adventure at the Stanford Artificial Intelligence lab. One of my roommates was a grad student. He took me over there and we played it and I thought that it was a really cool, great, amazing thing. I had just finished Slot Racers and I was trying to figure out what game to do next. I decided to do a video-game version of Adventure.
It presented several problems because it was all text. You’d get a text description of the room you were in and what was around you, and you’d make commands like, “pick up wand,” “take bird,” “go north,” “go south,” “wave wand,” and things like that. It was all noun-verb descriptions for movements or actions, and it took up quite a bit of memory to give these text descriptions.
I decided I’d do the “go north/southeast” thing with the joystick and I’d show one room at a time graphically on the screen. The rooms were all interconnected. If you drove your cursor off the edge of the screen, you popped into the next room.
—Warren Robinett
In the original Adventure, players found weapons and other inventory, much of which they carried with them throughout the game. Because Robinett had to show graphic representations of each item, he limited players to carrying one inventory item at a time. Making the proper selection for each situation was crucial, since a certain sword might defeat one enemy but be useless against another. Throughout the game, pesky bats tried to fly away with whatever objects players carried.
I made the decision to allow you to carry just one object at a time, and that turned out to be a good thing because it meant you had to make strategic choices. If you had a treasure and a weapon and you wanted to go somewhere, you had to pick which one you were going to take.
It was also a good choice because the graphics on the 2600 were so limited and it kept things from getting too cluttered on the screen.
—Warren Robinett
Robinett began Adventure in the days when programmers were expected to create their own artwork. He describes his dragons as looking like ducks and admits that the entire game looked a bit primitive. When he was about halfway through, he got bogged down and started another project. He did not return to finish Adventure for nearly six months.
When he returned, Robinett decided to create a hidden room. The room would have a special surprise for anyone who found it, and the keys to open the room would be readily available, but Robinett made the keys and location of the room so obscure that he doubted that anyone would ever discover them.*
To access Robinett’s secret room, you had to find “the dot,” a single gray pixel in the center of a wall of the exact same color. If your cursor touched the single interactive dot on that noninteractive wall, it would indicate that you could pick it up.
I called it “the dot” and it was just one pixel. It was the smallest, most insignificant little object you could possibly have, and it was gray. It was the same color as the background. That made it even more insignificant because [even if you found it], you could lose it and maybe not find it again.
It was hidden in part of one of the mazes in which you couldn’t see very far. The area was even inaccessible—you had to use the bridge to cross the wall to get into it. You had to make a map of the whole maze and then you would discover that there was one little tiny chamber that you couldn’t get to unless you used the bridge to cross the wall. And then if you went in there, you’d run into the dot and you could pick it up.
If you picked up this little dot, the one pixel dot that was hidden inside the inaccessible part of a large maze, and you brought it back and you messed around with it long enough, you found that it could get you through this wall and into the secret room in which I filled the screen with the words, “Created by Robinett.” It [the message] was in every color in the rainbow because I made the graphics go through the entire color palette. I wanted my name in colored lights.
—Warren Robinett
No one knew about Robinett’s secret room. He did not tell his friends at work about his little prank. If word got out, he would have been fired. It cost approximately $10,000 to manufacture games at that time. Robinett’s secret room took up 5 percent of the storage on the Adventure cartridge, and he was afraid that if Atari executives discovered it, they would insist on deleting it and remastering the game.
I was the only person creating the game, and nobody went through our programs with a fine-toothed comb to see what might be in there. The hard part was keeping it a secret for a year until the game came out. I didn’t even tell my two buddies, Jim Huether and Tom Reuterdahl. I felt that if I couldn’t keep the secret myself, how could I expect them to keep a juicy secret like that?
—Warren Robinett
Atari manufactured nearly 300,000 copies of Adventure. In 1980, after Robinett had left Atari, a 12-year-old boy from Salt Lake City sent a letter to Atari to inform the company about a strange thing he
had discovered in the game Adventure. He had found the dot and opened Robinett’s secret room.
Robinett’s prank created a sensation. Arnie Katz, Joyce Worley, and Bill Kunkel, the publishers of a magazine called Electronic Games, reported the story. They referred to the room as an “Easter egg.” The popularity of Robinett’s hidden room was also noticed at Atari. In the future, entire games would be built around hidden surprises.
The Great Migration
The 1978 introduction of Space Invaders ignited interest in consumer video games, as well as arcade games. Atari’s Video Computer System did not sell particularly well through the 1977 Christmas season, but its sales were better than expected throughout the rest of the year. As Christmas 1978 approached, however, a new competitor emerged.
Magnavox returned to the video-game industry with Odyssey 2: a game console that the electronics manufacturer hoped to pass off as something more by adding a built-in keyboard. The keyboard did not fool consumers into believing that Odyssey 2 was a computer. Consumers did not see Odyssey 2 as somehow being on a par with Apple. It was a video game system, and despite having launched the first home game system, Magnavox could not hope to compete with Atari. By the end of 1978, Atari had sold its entire inventory of over 400,000 warehoused VCSs and had to step up the production of new units.
