StarForge par CodeHatch
Publié : August 8. 2012
Canadian studio CodeHatch sounds off on its ambitious open-world title, StarForge
Essentially a first-person action game with infinite, deformable terrain, StarForge lets players build structures a la Minecraft—but that’s only the half of it. The game allows you to build and destroy pretty much anything: Cut down trees, dig yourself a cave in the side of a mountain, or simply deform the landscape at your will. Did we mention that it does all of this with proper physics and no loading screens?
StarForge comes from a pair of Canadian coders, William Sworin and Steve Smolley, who formed a studio around the game that they called CodeHatch. “During the summer of 2010, Will was finishing his first game Silas, while I was programming experimental physics tech,” says Smolley. Sworin was playing StarCraft 2 at the time, and heard about a little game called Minecraft. Upon first playing it, he realized how much potential there was in starting a new project that joined the core gameplay of an RTS like StarCraft with the creative freedom of an open-world like Minecraft. “Will knew Minecraft would become the instant game-changer we know it to be today, inspiring and spawning many new subgenres much like Doom did in its time,” says Smolley. Indeed, Minecraft was a game that—finally—allowed players to do whatever they damn pleased.
“Gather resources, build bases, craft your gear, and survive on an alien planet,” is how Smolley describes StarForge’s core experience in its most minimal form. The premise: In a distant future, our homeworld is in peril and as a last resort, specially chosen individuals leave on a one way mission to populate another planet, taking with them as much technology and resources they could carry. “As you play, the StarForge universe will procedurally unfold itself around you in a unique way each time; more ships containing refugees like yourself land in the distance, economies form, rival clans fight for ownership of territories,” says Smolley. A cross-section of the population will rise to the top, while another wil eventually becomes lunch for some ravenous alien creature. “It's really a game made for PC gamers, and it’s a sandboxer's dream.”
From the get-go, CodeHatch has been comprised of just the two developers; only recently have they begun training an additional programmer. “This would be our first employee, and this was only made possible by everyone's continued financial support, via buying Hatch Points in our online store,” says Smolley. The duo initially chose Unity just a month after the idea of StarForge came into view. “We were curious about engines that were available, and Unity stood out for not demanding any royalty from our project,” says Smolley. “This was the first time either of us worked with Unity, but it is very easy to learn, and its workflow is phenomenal. It also has such an amazing community online, and it's hard to explain how valuable this really is. I can't imagine a better fit for our game then Unity.”
Indeed, he says there are “too many things to mention” when we ask Smolley about his favorite Unity features. “Being able to program the editor just as much as we can program the game itself was a huge help, in that it let us design our workflow exactly how we wanted,” he says. “Being able to test programming edits in real-time while the game ran was a significant development boost as well. We like the import pipeline for art, and how big projects can be easily managed—plus the way prefabs and components work is very helpful for quick development.” The Asset Store has been a boon to the developers, with such tools as NGUI becoming keys to efficiency. “The list goes on and on really,” says Smolley. “The best thing for anyone who is curious is to just try it out yourself. We hope that Unity continues to pursue their accessible, intuitive, customizable and scalable design strategy, as without it StarForge wouldn't have become what it is today.”
> The best thing for anyone who is curious is to just try it out yourself.
The project has seen its share of memorable moments. Early on, the game’s rigidbody trees were inspired by the team’s 9-5 jobs: chain-sawing and then processing hundreds of trees a day. The aim, then, has been to instill a sense of dread and suspense when a tree falls in the game, possibly crippling players if they wander into its path. Likewise, when Smolley was fiddling around with some spinning physics rectangles to test his movement code, he noticed that it was surprisingly fun to get hit by them; the team then immediately put together a gnarly slab of rusty metal in its place, forming what has come to be known as the "Slammer Turret" today. “We really just like to make stuff that’s fun: if we think of something cool, we put it in the game,” says Smolley. “It's a really iterative process. Some might call it feature creep...but well, yeah, our game style thrives on feature creep. Why would we want to stop putting stuff in?”
Early on, Smolley says the team had some graphical issues it was working through. “Image effects had a lot of issues, especially when dealing with multiple cameras with different rendering pipelines,” he says. “We wanted to have background objects have certain image effects applied, while foreground objects did not.” To do this, they used a deferred camera for the background with the image effects, and a forward camera for the foreground marked "don't clear". They expected the objects to be rendered into the scene overtop with correct depth occlusion from the background objects, after the image effects were applied to the background separately. “However what we wound up with made absolutely no sense: The foreground camera seemed to render only fragments of the foreground objects, while the image effects for the background camera would be turned on and off at random.”
Still deep into development, the team says its first, rather ambitious game has been both physically and mentally demanding. “Like most who are self-taught and in our early twenties, we had to make StarForge with very little money,” says Smolley. Because they never worked at any established game studios and have only a high school education, both have worked as general laborers throughout the development cycle. “After a long day, we'd go down into our parents' basements to work on StarForge for the rest of the night,” says Smolley. “It made us work very hard on the game when we had the chance, And since we had nothing to lose, we just put in whatever we felt like without any feeling of risk or remorse. We believe this is how most indie developers survive, month by month: working hard [on] their dream at night and during weekends. Their main piece of advice: Be ready to suffer for your art. “Be ready for a decade of making no money from it. Be fine with that, accept it, and don't let anyone tell you otherwise. You are doing this because this is who you are and what you enjoy. Make games because you love doing it. Work on them every hour of your spare time. Collaborate with other developers and be innovative. Imagine it, and make it real.”
Big idea, small budget
“Any small indie team knows that trying to out-do the big boys at their own game, playing by their rules, is a recipe for disaster,” says CodeHatch co-founder Steve Smolley. Instead of focusing its efforts on a development pipeline of hand-perfecting every pixel and polygon, the team therefore focused its energy on making new tech that enabled them to produce more in less time and provide a much more interactive, unscripted, user-directed experience. “We feel this could ‘wow’ people just as much as the next big AAA sequel from X publisher for the cinematic, Hollywood-style experience they are known for,” says Smolley. “This is why Minecraft did so well: it was pixel art, but the videos had this insane ‘wow’ factor because you had never seen anything like it before.” Their advice to other indies? To take risks where others can't. “We would also say that it is very important to create games that let users create their own fun, otherwise you'll spend months making disposable fun for them.”
CodeHatch co-founder points out that getting the project rolling initially was made easier with Unity. “Unity is very intuitive to use, and takes out almost all the trivial busy work that'd otherwise bog down your creativity,” says CodeHatch co-founder Steve Smolley. The first demo they put together used hourglass-like terrain, with each side representing where a team would be, and a forest players could chop down. “It was really fun to play and we knew at that point we had something cool,” says Smolley of the project’s early stages. “We also had the physics movement developed to an early playable state. All of this took about three months after writing our first design document.”