AI War Interview


posted 12/10/2009 by Charles Husemann
other articles by Charles Husemann
One Page Platforms: PC
We like indie titles here at Gaming Nexus, especially good indie titles.  Which is why we followed up our review of AI Wars with an interview request.  Here's what we got back:

Could you introduce yourself and talk a little bit about your role on the project?

Sure. My name is Chris Park, and I’m the lead designer and sole programmer for AI War: Fleet Command. I’ve been doing hobbyist level design, game design, mods, campaigns, and so forth since I was a kid, but I’ve been a business software developer for the last eight years. Since 2003 I’ve been getting more and more into actually programming my own games, as opposed to developing hobbyist content for other people’s titles. I founded Arcen Games in hopes of eventually making a living off of my work in games, rather than business software. As of December, that dream is finally becoming a reality, so it’s a very exciting time.

Why don’t you tell us a little about your team, your design philosophy, and what made you decide to jump into the 4X world.
Arcen started out as just me, and actually I was working on AI War for a good half year before the company even existed. In early 2009 I contracted a great up-and-coming composer from UNC, Pablo Vega, to do the music on AI War, and in December he’s going to be joining our team fulltime as well in order to handle all of our music and sound work. For the initial release of AI War, all the way back in May of this year, that was the entire team of staff – though for AI War I had my three awesome alpha testers of my dad, my uncle, and my uncle’s colleague.

I’ve had a longstanding weekly co-op RTS session with them since 1998 or so (although Civilization IV, Neverwinter Nights, and a few other non-RTS titles did sneak into our sessions), and it was my frustration over the current strategy offerings that led me to create AI War. Other titles just didn’t emphasize co-op enough or in the right way, or didn’t have random maps or multiplayer savegames, or enough depth for long-term play, or – and this is the clincher – strong enough AI to hold our interest for more than a few months. AI War was born out of a desire for a specific sort of collaborative strategy-intensive experience that I wasn’t finding in the AAA titles.

After the initial release of AI War, we started having some moderate financial success and were able to grow the team. Philippe Chabot joined the team fulltime in August as our new artist, and the game is amazingly better thanks to his work. Lars Bull had been the most prolific contributor out of all my various beta testers, and joined the team on a part-time basis both to help out with AI War, and so that we could create an upcoming casual puzzle game together. Calvin Southwood was one of our most active early fans, and he also joined the team on a part-time basis to help us with techwriting, forum moderation, and otherwise. And then, of course, there are hundreds of fans from our forums who have contributed to the ongoing design discussion about AI War and its expansions since our release, and have helped make our 2.0 version a reality.

When it comes to my design philosophy, I’m all about efficiency and keeping an eye toward the big picture as development proceeds. In the business software world, I was accustomed to billing clients by the hour and needing to keep time spent to a minimum for their sake and mine, and that mindset has definitely carried over to Arcen. Gameplay and fun-factor are the key foci for our games, and we focus on polishing those through iterative design processes. Another thing I’ve learned from business programming is the value of a Subject Matter Expert, or SME, on a project. For this reason, I only make games that I actually want to play myself, so that I can act as my own SME when evaluating if it is any good or not. Even so, we seek fan feedback early and often with prerelease beta and even alpha versions of our software. Thus we wind up with a whole team of motivated SMEs contributing insights, advice, and ideas, and I think that really shows in the end products.

Could you talk about what separates AI War from other 4X games like Galactic Civilization II?
Well, AI War is realtime, unlike most other 4X games, which tend to be turn-based. But more significantly than that, I made a very RTS-style control scheme for the game. I basically wanted an RTS game that played over a 4X sort of time span, with the sort of strategic choices that you normally only see in a turn-based game. Many RTS games wind up with cop-out strategies such as just rushing, or just turtling for a while and then using a huge mass of guys to overwhelm your opponent, but you don’t see that sort of thing much at all in the TBS/4X world. Using the galactic scale and all of the hard choices that our AI Progress metric creates for players, but in full realtime, you wind up with much more of a true RTS/4X hybrid than I think has been seen before.

Most RTS games have very streamlined interfaces, and very few if any subscreens, and this is the sort of approach I took with AI War. As much as I love some of the great 4X games, I’m not in love with the volume and depth of all the subscreens that tend to show up there. One key thing in a turn-based game is that you have basically infinite time to think about any given move, and so in that context subscreens are not so much an issue. RTS games are much more about prioritization and keeping all the different parts of your military empire in good standing as you proceed; in a complex RTS, there is not enough time to manage every little detail, so you have to trust to the unit AI to a certain extent, and manage only those aspects of the game that are most important at the time. That sort of mentality creates a fundamental difference in feel between AI War and other 4X titles. AI War has been called the first “grand strategic 4X tower defense RTS,” and I think that’s as apt a subgenre title as anything – the more you play the game, the less it feels like any other title out there at the moment.
Page 1 of 4