Tim Sweeney

Hercules:
Welcome, Tim. What gave you the idea to create an ASCII/ANSI based game creating system?

Town of ZZT - the very first ZZT game ever, made by Tim Sweeney Tim Sweeney:
This will sound too weird to be true, but ZZT started out as a text editor. About a year before writing ZZT, I had spent over a year writing a Pascal-style programming language and editing environment on the Apple ][, which was a very cool project but I never released it. When I started on ZZT, I wanted to recreate that, starting with the text editor. So I wrote the cursor controls, for moving up, down, left, and right, and displaying text on the screen.

But then my playful side took over, and I decided to spend a few hours having fun, causing some ASCII characters block the cursor's movement (the cursor became the player), and other characters move around based on their ASCII value -- these were bullets, of course. So, the first ZZT levels started out as text files I'd write in the little text editor, then I'd hit the "run" button and the game would start going from there.

And that was WAY more fun than writing a text editor. So I started building levels and extending the editor as I thought of new features. After a few months, I had a lot of pretty big levels, and an editing tool that was really fun to use, so I decided to turn it into a shareware game. It took 9 months of part-time effort -- very part time effort; I was going to college and mowing lawns to earn money at the time. I think the first version of ZZT went online online in October 1991.

Hercules:
Were there any other people directly involved in the ZZT project? If so, what happened to them after ZZT was released?

Tim Sweeney:
The initial release of ZZT was all my work, but then the really interesting things started to happen as users started building their own levels. Best of ZZT was the first collaborative project released, and then Super ZZT. After that, the user-created levels, and entire ZZT enhanced rewrites started taking off. And they're still going.

Hercules:
Did you ever think that an online community would form because of ZZT and that it would still be very alive in the year 2000?

Tim Sweeney:
Well, it was clear that the editor fun to use, so I was hoping and expecting a few people get into it and build some levels. Maybe 10 or 20. I never expected it would be thousands! And I would have never guessed people would still be using it near its tenth aniversary.

ZZT taught me one of the most important lesson of my life: people really want freedom to create stuff. If you give users tools to expand a game, they will do tons of cool stuff you never imagined possible. They will build communities around your work, and those communities will grow to be much bigger and more interesting than the game itself. So, when we're working on a new game, we always ask ourselves: what will the community want to do with this?

With ZZT, you could build levels. With Unreal, you could also build mods, run servers, or license our engine and build an entire new game. So, what's next? What new kinds of community involvement will become possible over the next few years?

Hercules:
Did you ever create any other ZZT games besides the ones that Epic officially released?

Tim Sweeney:
It was all released.

Hercules:
Why did you create Super ZZT after ZZT? I mean, did you have any specific reason for this?

SuperZZT, the sequel to ZZT, which never was as succesful as its predecessor Tim Sweeney:
During ZZT development, I felt limited by only being able to work with one screen at a time. With Super ZZT, I wanted to create the feeling of more expansive scrolling environments, with more of an exploration aspect.

I thought the game would be a lot better that way, but it didn't really work out like that. ZZT was always way more popular. It had more of that "magic" set of limitations and tradeoffs that made it really fun to play and work with. Super ZZT had better technology and more features but was missing that special thing.

I think other game developers have experienced that kind of letdown. For example, in my mind, DOOM remains the best id Software game, and the most fun overall game, I've ever played. Though the team went on to create lots of other great games, the Quake series and so on, they haven't been able to recreate that magic feeling DOOM gave me. Maybe nobody will. It's strange.

Hercules:
Gregory Janson, who also created the GCS MegaZeux (sort of like an advanced ZZT), made Super Toolkit, which allows ZZT game creators to use 16 different colors in the editor. What do you think of this?

Tim Sweeney:
More colors are good, but I'm not so sure the game needs technology enhancements. I mean, even when it was released, in 1991, it was probably the least technologically advanced game that gained an online following at the time.

Hercules:
Do you still play recent ZZT games from time to time? If so, do you like what you see?

Tim Sweeney:
No, I make it a point in life to avoid being sentimental. Once you look back on the past as "the good old days", that's when you've reached and passed the peak of your career, and perhaps, life. I always want to be creating new stuff, software that people get enjoyment out of not just by playing our games, but by extending them, building their own levels, and growing communities. I want to do that until I drop. I'm scared of looking back into the past, out of fear that I'll get stuck in it.

Hercules:
You moved onto to other, bigger projects long ago. It must be good to know that the first thing you ever created is still used/played a lot. Does ZZT still cross your mind, sometimes?

Unreal - the game that made Tim Sweeney a famous man in the gaming world Tim Sweeney:
Yes, one of the interesting things to do is contrast ZZT and Unreal, and look at how incredibly far we've come in graphics quality in that time. But also to see how little the industry has progressed -- or maybe even gone backwards in some respects.

For example, the ZZT editor is ten times easier to learn and use than the Unreal editor. So, I always think about that. Does 3D level design need to be so much more complex? Or might there be better ways of building realistic 3D levels that is as easy as ZZT? I can't see how yet, without seriously sacrificing power and realism, but this is frequently on my mind.

The other thing that haunts me is how little programming technology has improved. While computers have become over a hundred times faster in the past ten years, the ZZT source code compiled in about 20 seconds in Turbo Pascal, and the language was so straightforward that the ZZT code almost never had bugs. Unreal takes 10 minutes to compile, and it's a lot harder now to eyeball a large C++ codebase and be confident that a function is bug-free. So are we really better off now? No, we've only done what has been necessary to take advantage of a 100X increase in computing power; we haven't really made the core aspects of our tools and work environments any better.

So, how will game development be 10 years from now? If levels take six months to build, and compiles take 5 hours each, and it costs $20 million to develop a game, then developing games won't be fun or even possible anymore. It's very clear that some fundamental assumptions in game development are going to break in the next 10 years, probably less. But it's not yet clear how. Will the game industry be taken over by huge Hollywood-style companies? Will games shift to simpler Palm Pilot style devices? Will the communities surrounding popular games grow so much that they dwarf the commercial side of the game business? I don't know, but these problems are what makes this such an incredibly exciting job.

Hercules:
I'd like to thank you for your time. Good luck on whatever you and Epic are up to!

Tim Sweeney:
You're welcome!