By Jason "loonyboi" Bergman
ark Frohnmayer deserves some major credit. Aside from the fact that he worked on the game that proved that teamplay doesn't have to be lame, he helped to make the world's first first person shooter that handles mile long outdoor environments as easily as Doom handled enormous mazes. The game is of course, Starsiege: Tribes, and the sheer size of these levels is reason alone to check it out. I took it upon myself to track down one of the guys behind the game, and Mark Frohnmayer was happy to help.
How'd you come to be in the games industry?
Right after I graduated from high school I applied to Dynamix for a summer job as a game tester. After a few weeks of that I started doing programming in their Mac group. When I got back from college (spring of 95) I started working full time and have been doing so since.
You actually graduated from college? Congrats...what was your major?
I majored in Electrical Engineering and Computer Sciences at UC Berkeley.
Have you worked at any other game companies?
What other games have you worked on?
I did most of the graphics engine for A10-II: Silent Thunder, and a little graphics work on some other titles.
Would you say that working on Starsiege: Tribes was a completely different experience, or did it prepare you somewhat?
Tribes was definitely a completely different experience. I came onto A10-II in the second half of the project to do graphics work. On Tribes I was there from the beginning and did a great deal more of the design and architecture work. Also the Tribes project was a significantly more ambitious project.
You've been with Tribes since its inception...how did the project come about?
We had a lot of people who liked the FPS genre, but, coming from a flight-sim background, didn't like the indoor-only restriction of other games of that type. We thought the union of indoor FPS play and outdoor vehicle based combat would be pretty cool.
Did the original version have more of an emphasis on vehicle combat then?
Not really - we always wanted vehicles as an enhancement to, not a replacement for first-person action. Still there was an assumption that outdoors = vehicles (tank sims, flight sims, etc.).
Why a first person shooter?
Because we enjoy the genre and thought we could add something to it.
The game's fundamental emphasis is on huge outdoor environments. Considering the fact that it hadn't been done before when you started working on it (at least not in a first person shooter, anyway) was this a major challenge?
Adding huge outdoor environments to a FPS was difficult technologically, however making it fun to play was. We definitely wanted the game to focus on person-to-person combat - Starsiege was the vehicular combat game, but making navigation of the large worlds fun meant not spending 20 minutes walking from one base to another.
How did the two games (Tribes and Starsiege) develop simultaneously? Was one given a higher priority than the other?
For the first year and a half we spent the majority of the time co-developing the engine, then each team focused on its project. The marketing effort for Starsiege started before Tribes because it was anticipated that Starsiege would ship before Tribes, but both projects were given a very high priority.
Was the game originally conceived as having a major emphasis on multiplayer, or did it ever actually have a single player game attached to it?
Tribes actually started out as a single-player game with some multiplayer capabilities, but gradually shifted focus as we completed more and more of the multiplayer functionality and left hanging the single player game. Eventually we decided we could spend a really long time making both a single player and multiplayer game or we could try to really shine at one or the other. At that point the choice was pretty clear.
What was the proposed single-player game like? Was it story-driven?
We kicked around a bunch of ideas, ending up with a mission-based design to advance the plot - futuristic Special Forces unit faces some alien race on a far-distant planet.
Was teamplay always the planned emphasis?
By the time we actually had a game up and running teamplay was a given - we played a ton of Quake CTF and QWCTF and found the team-oriented gameplay to be a lot of fun.
Were bots ever considered as AI-driven teammates?
It was an idea that we kicked around for a while but making the bots smart enough to be good teammates would have taken a long time.
Were there other vehicles that never made it into the final version of the game?
We had ground vehicles up and running, but getting them into a shippable state (balanced, fun, cool) would have taken more time than we had.
Let's talk a bit about the graphical side of the engine...the game shipped with only hardware acceleration via Glide. In retrospect, do you regret focusing all your energies entirely on Glide support?
We actually didn't focus all our efforts on Glide - we had a D3D version as well, but the performance just wasn't there. We make heavy use of dynamically generated textures and D3D would just choke on our texture set.
Credits: Illustration © 1999 Rowan Crawford. This interview is © 1999 Jason Bergman & Mark Frohnmayer. All other content is © 1999 loonyboi productions. Unauthorized reproduction is strictly prohibited, so don't try it...or we'll damn you.