1. First off, a question for those who might not be aware of what we’re discussing. What, in your words, exactly is Starshatter?

Starshatter is a military space combat simulation set in the distant future. Unlike most space sims, Starshatter allows you to directly command a wide variety of ships, from agile atmospheric and space-based fighters to giant cruisers and fleet carriers. Starshatter will take you through the complete space combat experience – from planet surface to interstellar space – with several dynamic campaigns set in a persistent simulated universe.

Starshatter is a game that I am developing in my spare time, working by myself, and almost totally from scratch. I do all the programming, game design, artwork, documentation, web site, etc. I’m on such a shoestring budget that I even wrote a 3D modeling program from scratch to build the ship models. As a result, I’ve been working on the game for almost four years now.

2. When did the idea of Starshatter first come to you, and what brought it on?

The original idea was back in 1996, or thereabouts. Back then, the game was going to be a FPS set aboard space ships and space stations. Between the first person levels, there would have been space flight sequences that allowed you to pick the next level you would play. But there wasn’t going to be any space combat, it was just a gimmick to add flexibility to the very rigid format of an FPS game. I abandoned that game after just a few months of development, but that’s where the game universe and back story were first conceived.

Several months later, I started fooling around with a first person version of Asteroids. It was going to be a bit like Han Solo’s chase through the asteroid field in The Empire Strikes Back. That was where the game engine for what is now Starshatter was originally born.

Then I got a job with Interplay, and I worked on Starfleet Academy for a while, so I had to put the Asteroids clone aside. While I was working on Starfleet, I noticed some of the game design problems that it had – how do you simulate large ship combat without boring an audience that grew up on Wing Commander and X-Wing?

After I left Interplay, I put everything together: the universe and back story for Starshatter, the engine I had started for the Asteroids clone, and a game design that included both large and small ships in combat scenarios.

So, I guess I would have to say that it has really evolved over time. There was no single moment where the game just sprang to life.

3. Which games would you say had a direct influence on the idea of Starshatter, if any?

Several! And each in different ways. There were many non-space-combat games that I liked, such as DOOM and Descent, and I wanted to capture elements of them in a space combat setting. And there were things about many of the space combat games, such as Wing Commander and Starfleet Academy and Battlecruiser 3000AD that I really wanted to avoid, as well as other things that I wanted to embrace. And over time, the Starshatter game design has come to be influenced by some of the excellent hardcore jet fighter simulations such as F22 Total Air War and Falcon 4.

At the same time, there have been some very important influences from outside gaming as well. The back story draws on themes that go back to E.E. “Doc” Smith’s Lensman series of space operas, as well as more recent works such as Vernor Vinge’s A Fire Upon the Deep. Some of the combat scenarios are inspired by David Weber and even more by Vinge’s Deepness in the Sky and Dan Simmons first two Hyperion novels.

Graphically and artistically, I’ve been influenced by things like Babylon 5 and even Star Wars Episode I.

4. I noticed in the demo that the flight model felt very “Elite-ish”, with left and right movements rolling the ship rather than turning it. Why was this flight model chosen?

The control model is driven by the physics model. Starshatter uses a fairly hardcore Newtonian flight model, especially for the atmospheric sequences. When you fly the space fighters in atmosphere, you need to perform a coordinated turn, a combination of roll and pitch to use the wings to generate lift in the direction you want to go. After a while I got tired of having to press “J” to switch between roll and yaw on the joystick, so I just learned to use airplane style controls all the time, even in space.

5. I also noticed that, again while playing the demo, that the AI was pretty cunning, to the point that it was rather hard to beat. Could you go into a bit more detail about the AI present in Starshatter?

The philosophy behind the AI design is probably unusual for a space combat simulation; the AI pilots don’t cheat. The AI controls ships that use the same physics model as the player, and the AI is limited to the same kind of control inputs that the player has. On each frame, the AI pilots get to move their joystick, throttle, and press triggers, just like you do.

The goal in designing the AI this way was to create AI pilots that were just as much fun and just as challenging to fight as real people. To help me tune the AI, the game has a mode where the AI controls all of the ships, even the player’s. That way, I can watch through the AI pilot’s eyes as he tries to fight through a mission. As I’m watching, I’m constantly evaluating “Does this seem right? Is that how I would handle the situation? Why did he turn that way? Why doesn’t he just line up the shot and finish it?”

The AI is built using a layered design, with low levels being responsible for primitive behavior, and higher levels handling more abstract reasoning. At the lowest level, the AI uses a hybrid flocking / steering model to turn the ship in the desired direction. The next layer up is the tactical layer, which is responsible for choosing targets and evaluating threats. There will actually be at least one more layer above that which is responsible for coordinating group behavior.

And if you think the AI is tough now, remember that 1) it isn’t done yet and 2) the AI fighters have about half the armor, and half the gun power, and one quarter the number of missiles that you have…

6. You’ve been working on this project for a long time. Where, would you say, are you in Starshatter’s development cycle, and how long do you think it will take to complete?

Well, I’m getting closer all the time. It’s hard to project how much longer it will take, because it is governed by how much time I get to work on it. I do plan to keep releasing demos and information right up until the game is finished. At the very least, there will be another interim release around Christmas 2001.

7. In the demo, we’re treated to a few missions involving space fighter battles, capital ship battles, and some ground attacks. What’s the feature set we can expect to see in either future installments of the demo, or the final version?

That would be telling!

Combat-wise, the demo is pretty complete. There are a few features still to add like jammers and stealth. And I need to finish take-offs and landings on planets, as well as launching and docking fighters in space. And of course, there will be ship-to-ship communication so that you can order your wingmen about.

I’m just getting started working on the dynamic campaign system. The next demo won’t have any scripted missions in it, they will all be generated automatically by the campaign engine. And the missions will be more sophisticated than they are now, with pre-set nav routes and more related force packages working together to achieve a goal.

There are many other cool features that I would like to include if time permits. But I don’t want to promise people anything that isn’t core to the design of the game.

8. Will the final release of Starshatter have any editing tools, such as a mission/campaign editor, or perhaps a way to edit ships as well?

Probably. I imagine that there might be some interest in people doing mods to use the game engine for other science fiction settings such as Babylon 5 or Space: Above and Beyond, which have more realistic flight and combat models.

9. Finally, what’s your philosophy and final goal in creating Starshatter?

My philosophy is that the Wing Commander “interactive movie” concept is an evolutionary dead end. Games that are seduced by the ability to wrap up the player’s emotions with actors and a character-driven plot become trapped by that ability. An interactive movie can’t show the player anything that hasn’t been scripted and recorded before the game even shipped.

My philosophy is that games should be about giving the player the tools and setting in which he can create a new story every time the game is played. That’s why Starshatter is based on a dynamic campaign. The game is all about experiencing new combat scenarios. There won’t even be a way to replay “failed” missions. The story will just keep unfolding with you in the middle of it all.