30.September.02, Interview
JASON COLEMAN
The remaining founder of Big Huge Games to be interviewed by us, talks about his past and current work
Interview Comments?
Click here to let your voice be heard
| | Apolyton: |
Since we have without shame neglected doing an interview with you all these years, please introduce yourself and describe us what you do at BHG and Rise of Nations.
| | Jason Coleman: |
At least have some shame about it! Anyway, I'm the VP of Software for Big Huge Games. Besides staying out of the way of our incredible programming staff, my main programming responsibilities are currently multi-player, interface, sound, and localization. Incidentally, these are the very tasks that I thought I would avoid when we started a company J. We all jump around quite a bit on various projects and our hierarchy is very flat, but here's some idea of what the various programmer's are currently working on (okay, I don't really do anything anymore):
Jason Bestimt (JB, Besty): 3D programming, responsible for most of the graphics that you see besides terrain and interface, including unit animations and F/X.
Pranas Pauliukonis (Pancake, Cake): AI, including path-finding.
Ike Ellis (ICE): Conquer the World. It's getting very cool.
Bryant Freitag (pronounced Free-tag in his world, but we call him The Freak. We claim that he brought this nickname with him. He denies it. Fair enough.): Terrain, terrain, terrain, terrain...
Scott Lewis (We're workin' on it.): Everything scenarios--scenarios, tutorials, and scenario editor.
Ilya Kreymer (ILLLLLYAAA!!!!): Ilya was our uber-intern who has left us for something called an education. He did a great deal of work crafting our scripting language and forcing us to use splines for virtually everything, and he now works for us as a contractor.
Brian Reynolds (Hot Sauce Boy)- we actually consider Brian part of the Design department, but if you're curious about his programming role, he implements and then tweaks the actual rules of the game.
| | Apolyton: |
One of the first projects you worked on was in QA on Colonization. From chasing Brian Reynolds' bugs you've come to found a company with him. How does that feel? :)
| | Jason Coleman: |
It feels great. Brian and I have always worked effectively together. We sometimes tug in different directions, but that tension is almost always good and leads to a better game. Plus, we've become good friends.
| | Apolyton: |
How much have the things a programmer could do changed since those days? People often complain that changes in newer versions of a game are not that big.
| | Jason Coleman: |
Many of those complaints are probably directed at design more than programming. Game design has almost always been an evolutionary rather than a revolutionary process. Sure, every now and then, someone invents an entirely new genre, but that's naturally rare. Music is a valid analogy, I think. Even supposedly new genres of music are usually some mix of the old. The equivalents of rock, jazz, and country aren't invented very often. In games, what are perceived as new genres are often old genres done better or several old genres combined into one. Also, in these still early years of game development, new technology sometimes makes new genres possible.
As to what you see from the programming side, I haven't heard the complaint you describe very often, but I assume this is some sort of "Why aren't we getting more for our computer upgrades?" Part of it is that most of the computationally "easy" problems have been solved. The "next" steps tend to be subtle and/or computationally expensive, inverse kinematics for example. Also, the transition of traditionally 2D games to 3D typically requires an apparent step back. Even with all the 3D accelerators available, a lot is dependent on the available CPU and memory. To even approach the quality of pre-rendered graphics with 3D requires a great deal more horsepower.
| | Apolyton: |
A certain jackal.txt exists in my civ3 folder. Not being at Firaxis anymore, do things like that feel...weird?
| | Jason Coleman: |
It doesn't feel weird at all. Firaxis was an awesome experience, as was MicroProse before that.
| | Apolyton: |
RTS or TBS: what do you enjoy working on more?
| | Jason Coleman: |
I definitely enjoy working on RTS more. Otherwise, I wouldn't be doing it! It helps that I now prefer to play RTS. As a programmer, I also find it more challenging and interesting.
| | Apolyton: |
Do RTS games bring more difficulties to programmers due to the limitation in the available time (with the user interacting with the game continuously)?
| | Jason Coleman: |
Certainly, but that's where the fun is, right? In RTS, you essentially have 1/20 or less the time to do more work than in TBS. It's an exciting challenge.
| | Apolyton: |
How powerful are the scripting abilities in Rise of Nations? Are there many hard-coded variables (example: number of different units) that would limit the possibilities of an external editing utility?
| | Jason Coleman: |
The scripting is powerful. If anything, it's probably a little too powerful in that it closely resembles C-programming, and that may throw some people off.
As for its abilities, we pretty much follow the same sort of iterative design process that we follow in many the other aspects of the game. If a scenario designer needs some functionality, we add it. This will continue throughout the project and probably even beyond as non-BHG scenario designers request new features.
| | Apolyton: |
Common issues in strategy games are the quality of two programming tasks: the map generator and path finding. How confident do you feel about these parts in Rise of Nations?
| | Jason Coleman: |
Our "iterative design" process hugely influences both of these aspects. We play the games and, if we don't like the way maps are generated, we work on them to correct the problem. The map generation works well. Our primary focus continues to be on balance and in creating interesting decision points.
Pancake's first job is to see to it that all the units can get where they're asked to go and to do so efficiently. That sounds tough, but it's actually the easiest part of the job.
The next part involves essentially reading the players' minds and doing what they wanted or expected to happen. Needless to say, this is the hard part, and it turns out there are a LOT of these situations: "I don't care if that route was shorter, the other way was obviously safer!" or "When attacking a fort, I want my horses set to ‘auto-attack' to protect the cannons, only leaving them for short distances to attack ranged-units, but they should participate in the attack of other builds, unless..." Anyway, you get the idea. This is where our design process shines. We play the game all the time. If we don't like the way something works, other people probably won't either. Sometimes, when there's no clear solution/consensus, we might add input functionality or game options.
The less people talk about pathfinding, probably the better. We have a phrase commonly heard around the office when something really isn't working in the game: "There's blood shooting from my eye." I haven't heard that regarding pathfinding in a very long time, so I feel confident.
| | Apolyton: |
Your people in Rise of Nations will be able to find themselves a task alone. Are there other automations that you're working on to reduce micromanagement?
| | Jason Coleman: |
The general game design lends itself to less micromanagement. For instance, the default victory options (capital capture, wonder victory, and territory victory) make it so you never have to chase down every single unit or building in the game just to finish. Owning territory means you have to worry less about continually scouting seemingly insignificant areas of the map just to be sure someone doesn't grow an offensive city in your rear (no giggling). Capturing cities means you can gain infrastructure later in the game without necessarily constructing the same buildings for the Nth time.
It should be noted that "auto-working" is an option with varying multiple configurations, including "off". In addition to citizens' auto-gathering/building, caravans will automatically find a friendly route, and any unit can be set to auto-explore.
| | Apolyton: |
You've played tons of Civ2. How is it to run the same historical course (in terms of units, buildings, techs, etc) in 2 hours instead of 20?
| | Jason Coleman: |
I've never had a problem with this. Both timings are arbitrary, and it's really just a question of how much fun the game ends up being. That said, playing the Conquer the World campaigns takes many hours to complete though each scenario within Conquer the World will usually be well under an hour. This feels right for a game that encompasses strategic, operational, and tactical elements. For your standard slugfest, the game is much more operational/tactical, and shorter games seem right.
| | Apolyton: |
Is it possible that BHG will do a TBS in the future or have you been totally converted?
| | Jason Coleman: |
Never say "never" but it's not in our immediate plans. It's more likely that we'll continue incorporating TBS elements into various aspects of RTS games, perhaps blurring the lines more and more.
| | Apolyton: |
And the toughest question possible: what is your favorite civ game? :)
| | Jason Coleman: |
Fortunately, this one's easy for me, though I haven't had much time to play CivIII. CivII is definitely my favorite. Working on it was a joy, and I enjoyed playing it all along.
|
Interview Comments?
Click here to let your voice be heard
Many thanks to Jason Coleman for taking the time to answer our questions and to Tim Train and Graham Somers for their help in arraging it. Questions by Markos Giannopoulos.
INTERVIEWS INDEX PAGE
|
