Blowing Things Up

It’s a safe bet that many of today’s multimedia developers found their first inspiration for the possibilities of combining computers with audio-visual elements in the hit video arcade games of the late 1970s and early 1980s. Dave Theurer developed two of those game classics, Missile Command and Tempest, while a programmer at Atari. As one of the superstar game designers of that era, Theurer encountered first-hand the challenge of preparing thousands of images for use in games. That, plus the needs of his computer artist friends, led him to develop DeBabelizer, a software tool that, since its commercial release by Equilibrium Technologies in January 1993, has become indispensable for working with images in multimedia development and other graphical applications. Equilibrium recently released a low-priced version of the program called DeBabelizer Lite. Morph caught up with Dave Theurer in San Francisco to talk about the challenges of programming with graphics and other multimedia elements.

Morph Meets Video Game Pioneer and DeBabelizer Author Dave Theurer

Morph: Did you have a game idea in mind when you joined Atari ?

Dave: No, I just wanted to blow stuff up. When I was a kid, I had a big old chemistry lab in my basement and I’d make contact explosives, little bombs, rockets that would go up and blow up. So when I went to college, at Wheaton College in Chicago, I thought I’d be a chemist, but I found out you don’t blow stuff up in chemistry, you just do titration experiments. I switched to physics and that was pretty good until it started getting really theoretical, particle physics and so on, and it got so far away from real life that I just said the hell with it, I want something that will keep me in touch with humanity. So with two years to go I switched to psychology and that’s what I got my degree in. I learned all about motivation and getting people addicted and it came in real handy for designing video games. We studied how to get pigeons to do things, and if you use the same techniques on people, it works.

Morph: When did you first encounter a computer?

Dave: They used a computer for college records and such. In psychology a lot of it was statistics and analysis of results. They were calculating it all manually and I said, “That’s ridiculous, we’ve got this computer here,” so I just figured out how to use the computer and did all my stuff on that. I just got a manual on FOCAL, this cheesy little language sort of like BASIC and !earned how to use it. I worked for the college data processing center for about six months after graduating, then found a job at a company called Bunker Ramo, that built supermarket computer systems, making an incredible $9,000 per year as a junior programmer. When they ran into problems in that business I called up National Semiconductor because they were building the same kind of systems. I moved to California and joined National Semiconductor in 1976. While I was working in Connecticut for Bunker Ramo, I went out with some friends to a bar one night and I played Pong for the first time. I saw it was made by Atari, and I said, "That's what I want to do." I didn't know where Atari was so I went to the library and tried to look it up, but I couldn't find it. So I decided that it was a nice thought, but it probably won't happen. And then it was like the gods took over, plucking me down at National Semiconductor in an office with a guy who, six months later, quit and became manager of programmers of Atari. He hired me without even an interview. It couldn’t have been easier. I ended up right in the spot where I wanted to be, and I certainly didn’t guide myself there.

Morph: Were you taken by what you actually experienced in Pong or more by the potential? Pong was really a pretty basic game.

Dave: When I saw the game I didn't think anything futuristic at all. I just thought, this is fun and I wanted to do it more and more. I put more quarters in.

Morph: People sometimes complain about the less-than-robust nature of today’s authoring systems and other development tools. What did you have to work with when you programmed Missile Command and Tempest?

Dave: When I got to Atari it was really very primitive. They didn’t have any graphics design tools. Basically, when they wanted to do a game, they’d just say, “Here, do this game.” They’d pair you with a hardware designer. the programmer had to do everything: Write the whole operating system from scratch, design all the graphics, write the tools to do the graphics, create sound tools, and do the sound design - everything from scratch for every game. The first game I did was Four Player Soccer. They said, ‘Do soccer.” I said, “I hate soccer.” They said, You're new here, do soccer.” So I did it in about four months, and it was fun. They had those big trackballs back then. four or five inches in diameter. To debug this game, we'd get four guys in there just wailing away on these monster trackballs, by the end of the day you’d be soaking wet. Then they said they had a game idea they wanted me to do. The Russians are attacking the U.S. with nuclear missiles and your job is to defend the country. When I walked out of the office, my spine was tingling, I knew instantly it was going to be a hot game.

Morph: You wrote it in assembly language to run on the 6502 microprocessor. What sort of development computer did you use?

Dave: It was a little black box that ran FORTH and hooked on to the game board. We wrote our programs up by hand and someone would type them in and give us back a paper tape to feed into the black box and download to the board. The game boards were really primitive; Missile Command was only about 11K total RAM. And there was no way to get the games to run fast enough by programming in high level languages on bigger systems, with the compilers as slow as they were.

Morph: Do you think that having to program within those constraints helped you?

Dave: It really damaged me. I’m still trying to save RAM. You don’t have to do that any more. When you’ve got a hundred bucks in your pocket you don’t have to buy a burger. In the first stages of doing a game we’d test it on guys around the office. Once you had it to the point where you could field test it we’d take it to a local bar the Oasis in Palo Alto is where we put Missile Command. We’d go out and watch people playing it, all kinds of people lined up around it. It was such a rush to see people fighting over your game, because a lot of games you’d send out and nobody would play them.

Morph: Those were heady days.

Dave: Missile Command was one of the first color games. Tempest was the first color vector game. My next game was I, Robot. That was the first game to use three dimensional solid objects. It went to Siggraph in 1984 and won some sort of commendation. It was really cool because you could rotate and display 300 or 400 polygons per frame and that was a big deal at the time. I, Robot came out in 1984 and it had a Big Brother theme but nobody picked up on it. There was this big eye in the sky looking at you and it blinked, and you could do stuff while the eye was shut, but if you did stuff while the eye was open it would zap you. I, Robot also had a second mode for people who didn’t like video games, called Doodle City. It was a 3D graphics tool. You could take all these objects in the game and paint on the screen with them. You could record them and play them back. You could make things reflect, rotate objects, etc. I’ve had a lot of people who came up to me and said they had dropped a lot of acid and spent a lot of time in Doodle City.

Morph: You stayed at Atari until 1990.

Dave: When I did Missile Command I lived near Ames Research Center down in Mountain View. They were always sending out these U2 flights, they go straight up and sound like an atomic bomb exploding. I’d hear those things and it would terrify me. And I’d wake up in the middle of the night from a nightmare where I’d see these streaks coming in, and I’d be up in the Santa Cruz mountains and I’d see it hit Sunnyvale and I’d know I had about 45 seconds until the blast reached me. I had those nightmares once a month for a year after I finished Missile Command, I had internalized the game so much while working on it. Having to do all this other stuff raised hell with my personal life, too. If there’s anything I could communicate in this interview it is that people should watch out for their personal lives when they get sucked into these projects. It’s really seductive and all consuming. My entire existence was creating these games. They’re holding out this huge carrot, all the money you’re going to make. And it’s fun, so you think, OK, if I can just make it for two or three years, then I’ll work on my life. I see a lot of that going on now.

Morph: We published an article a few months back about balancing development goals with life goals.

Dave: It’s important. Nobody is looking out for these poor programmers and artists.

Morph: At the Outpost we talk a lot about the idea that programming, as we know it, is dead.

Dave: When I did Missile Command and Tempest, it was basically a one-man job, but after that I started getting help on the graphics and sound. It’s evolved to the point where it’s like making a movie. You’ve got a sound department and an art department, you give them your requirements, they put it in the schedule, then hire the people. By 1990, every arcade game was a kick, punch, and jump game, and I hate those games. During my last few years at Atari I had been dealing with artists. When I did a game, sometimes the artists wouldn’t quite do what they were supposed to do; sometimes they’d give you a ton of graphics and they were all a little off. Like two pixels shifted to the left, or instead of using the 16-color palette that you asked for they’d use a 13-color palette. And you’d be sitting there with 200 pictures to fix, so what are you going to do? We had a VAX computer by then and I’d write all these little programs to fix the problems with the artwork to save my butt. I finally got tired of what I was doing at Atari and I quit. A friend of mine, probably one of the greatest video game artists on the planet, Allan Murphy, quit Atari. He wanted to be able to design graphics on the Macintosh for the Atari, and go back and forth between the Atari ST and the Macintosh. The tools on the Atari ST were pretty pathetic, and he liked the Macintosh, so I wrote him a PICT to NEO converter. Then he wanted an IBM format, so I added that. I added more functionality as he required it. It was a nice change from Atari, working at home on my Mac with Think C, with a source level debugger, just sitting there and clicking away and figuring out the problems. Several more of my artist friends started using DeBabelizer and they would call me and request more functions so I’d stick them in the program, and it was getting bigger and bigger. After some site licensing, we finally released it commercially in January 1993. My goal was to get it priced so that my artist friends could buy it. So we came up with a $300 list price which would give us a street price of under $200.

Morph: Is maintaining and improving DeBabelizer a full time job?

Dave: I’ve got a list of about 800 things my artist friends want me to put in it. One of the things that people ask for most is the ability to read PostScript files. And we want to go into the CYMK world so that we can do better things in desktop publishing.

Morph: Ever tempted to get back into developing games or other content?

Dave: People are constantly trying to get me to do it, but not now. I did that for 11 years, and now I want to do this for awhile, build tools for other developers. This way I can empower more people.

Morph: Although they do leave a lot to be desired in terms of robustness and functionality, authoring tools allow non-technical people to create content.

Dave: Get rid of the programmer, that’s the most important thing. If you can get rid of that extra link, get straight from the creator’s mind to the content, then you don’t have to do that extra step where you always lose something. There’s always a compromise. A direct link is always the best and when you have to go through a detour, like a programmer, there’s the possibility of messing things up. The programmer’s going to have a personality, want to do things his or her own way, and get tired one day. If you just had your own design tools to work with directly, you’d have much better results. The good thing about DeBabelizer is that it runs plug-ins, which means that I don’t have to be doing everything for everybody. I can let other software developers come up with cool image processing tools that just plug right into DeBabelizer Kai’s Power Tools for example. Why reinvent the wheel? That’s what’s going to be happening over the next few years with Apple Script, you’ll be able to pick out a special function of the program and use that function. So I’m a fan of object-oriented programming, the idea of selling functionality instead of applications, functionality that an end user can wrap around his or her needs. The relationship between the end user and the programmer will change dramatically.

OCR Scanned From Morph's Outpost on the Digital Frontier - May, 1994 issue.