Warren Robinett Interview:
A. Merrill's Talks to the Programmer of "Adventure" for the Atari 2600
The following interview with programmer Warren Robinett is a huge departure from the usual Arthur's Hall content. Warren Robinett was a programmer for the Atari Corporation, and was responsible for the design of the classic Atari 2600 game “Adventure”. The game was a visualization of earlier text adventure games, and a huge influence on later games like The Legend of Zelda. “Adventure” is also famous for containing the world's first “Easter egg”, since the programmer put his signature in a secret room within the game. The interview was conduced in 1998 by A. Merrill (a friend of Arthur's Hall) and posted on a college web site that no longer exists. I'm a video game geek at heart, so I hope some of you will enjoy this as much as I do.
Q: What is your education?
BA Rice University 1974 in "Computer Applications to Language and Art" MS Berkeley 1976 Computer Science
Q: When did you start working for Atari, and how old were you?
I worked there November 1977 to June 1979. I was 26 when I started.
Q: Did you do any game programming before Adventure?
I did the Atari 2600 cartridge Slot Racers as my first game. Then I started Adventure, got sort of stuck, did the Atari 2600 Basic Programming cartridge, and finished Adventure and Basic Programming at the same time. This all happened in about 18 months, which sounds very short now, but remember that the game cartridges had 4K of memory back then. They filled up fast.
Q: Did you carry around the idea for Adventure in your head for awhile before you programmed it, or was it an off-the-cuff sort of thing?
It was directly inspired by the original text Adventure, created by Don Woods and Willie Crowther, which was sweeping the world of the old ArpaNet in 1978. I played it with my friend Julius Smith at the Stanford Artificial Intelligence Lab, where he worked.
I thought that the idea of a network of rooms, with objects you could pick up and move around, plus creatures that moved around on their own was really cool. Of course, it was all done with text descriptions in the text Adventure, but I thought it would be possible to make a video game version of that idea.
However, my boss at Atari knew that the text Adventure game required hundreds of K on a mainframe, whereas the Atari 2600 had 4K of ROM, 128 bytes of RAM, and a 1 Mhz 6502 processor. So he thought it was impossible, and told me not to do it. So I did it anyway.
My idea was to show one room at a time on the TV screen, and use the video game joystick to control the room-to-room motion by running of the edge of the screen (in the text game this was done by typing "Go North" or "Go East." For the functions of picking up and dropping objects, I planned to use the joystick button (analogous to typing "Take Bird" or "Drop Bird"). And for all the other things you could type in the text game to use the objects to do things (like "Kill Snake" or "Read Magazine") I planned to do by letting the player make two objects touch each other, because the Atari 2600 hardware provided a collision detection mechanism. This overall scheme turned out to work pretty well, although I didn't actually have it all figured out from the start, and it took me a while to figure out certain parts of it.
Q: How do you find the secret room?
The secret room is inside the black castle in Games 2 and 3, in a part of the Catacombs (the maze where you can only see a little ways). There is an isolated section in the catacombs that you can only get to if you use the bridge to cross the maze walls, and inside that little isolated pocket is an object (not mentioned in the game's documentation, because I didn't tell anyone about it): the Gray Dot. The dot was just one pixel in size, to make it seem as insignificant as possible. It was also the same color as the background color in all the rooms, so it was invisible in most places.
You had to take the dot and use it to get through a side wall below and to the right of the Yellow Castle, and then you got into the secret room, which had my signature in it: "Created by Warren Robinett."
Q: I've heard of game programmers putting their initials somewhere in the game, but people putting their whole name in the game was just about unheard of. Were you afraid of being reprimanded?
Well, in the tight world of 4K memory, the graphics for my signature did use up 5% of the ROM, but I actually added it at the very end, from memory that was left over.
Yes, I figured the powers that be would not be happy about that little trick, and would remove it from the game as soon as they found out that the signature was in there. But if a few hundred thousand cartridges got manufactured first, at least those would be out in the world.
What actually happened was that by the time some kid or kids had discovered the secret room and Atari found out, I didn't work there any more. So they couldn't really punish me, and the manager of game software at that time decided little hidden surprises in games, which he called "Easter Eggs" were kind of cool. Also, it would have cost $10,000 to make a new mask for the ROM in the cartridge. So what happened was that Atari left the secret room in the game.
Q: In the games produced by ActiVision, the programmers name was right on the cart and there was a picture of them in the instruction booklet. In games produced by Atari, the programmer was not mentioned at all. Did this lack of recognition bother you and your fellow programmers?
Yes, this was part of the motivation to put my signature in the game.
Q: Did this have any impact on your decision to leave Atari?
No, not really. I had worked really hard for a couple of years, and was ready for a change. My salary of $22,000 per year (with no royalties or any kind of incentive) was just an average engineer's salary, so I could get another job later, when I was done going off and traveling.
Three of us who designed games at Atari--the other two were Tom Reuterdahl and Jim Huether--were out drinking one night. This was after some Atari designers had left to start ActiVision and Imagic. Those guys, who used to be down the hall from us, were now all multi-millionaires. How did we miss the boat? Were we stupid, or what? So we formed the Dumb-Shits Club. The requirement for membership were that you had to have designed games for Atari and never made any money from it.
Q: What other 2600 games did you design?
Slot Racers and Basic Programming for the Atari 2600, and later Rocky's Boots for the Apple II.
Q: Were there any games that you designed that weren't finished or released?
No, all of mine were released.
Q: What is your favorite 2600 game?
I don't know. I never played them much. I was a designer, not a player.
Q: Do you currently own a 2600?
Yes. I have a 2600 and my old cartridges in a box in the attic. I fire it up every once in a while. My two sons are 3 and 9 months, so someday I'll let them play Adventure.
Q: How do you respond to Adventure detractors that claim that the dragons look more like psycho-ducks?
Well, what can I say--the dragons do look like ducks. It was just a different era of game design. One person, the game programmer, had the idea, wrote the code, created the graphics, and created the sound effects. I guess I was a better programmer than artist, but hey, I kind of like those duck-dragons.
Q: How long did it take you to program Adventure?
I started it in April 1978 and worked on it for several months. But there were some issues that weren't quite resolved, so I paused for a while while I worked on another cartridge, and then finished in Early 1979. So a total of about 1 year elapsed time.
The unresolved issues were, I think, I had room to room motion, and movable objects, and a dragon that chased you from screen to screen, and the rooms that had walls, I hadn't thought of the sword for killing dragons yet or the bat, and hadn't figured out how to do good mazes on the 2600.
Q: How many Adventure carts sold?
I was told about one million.
Q: Was the instruction book written solely by you, or did you work with someone to write and format it?
I had very little to do with the manual for Adventure. There was a group at Atari that produced the game manuals. Steve Harding wrote the manual for Adventure, I believe. He played the game and then wrote it up as he saw fit. I may have reviewed it to check for mistakes. I talked to him some as the game was under development -- we weren't miles apart. But the manual was not my product, just the game.
Steve started out by saying "An evil magician has stolen the Enchanted Chalice and has hidden it somewhere in the Kingdom." There is no evil magician in the game, but Steve apparently thought a setting was needed to explain why the player had to go get the Chalice. Also, I called it the Holy Grail when I was working on the game, but I guess that was too touchy from a religious point of view for him.
I named the Dragons (Grundle, Rhindle, and Yorgle) which he used. I also had a name for the bat (Knubberrub) which was suggested by a friend, but I guess he thought that name was too weird.
Needless to say, I neglected to tell Steve about the secret room.
Q: What are you up to nowadays? Are you still involved in the computer gaming industry?
Then I got into virtual reality, first as NASA in California and then I moved to work at the University of North Carolina. I'm thinking about getting into educational software again.
Right after Atari I got involved with some educators and we started a company, The Learning Company, to make computer games to teach things to kids.