How The Community Is Building Unreal Tournament

Unreal Tournament [official site] represents a new and interesting way of developing games. Beyond the Early Access periods now common on Steam or the mostly-advertising open betas used for every major multiplayer game, UT is fully free and developed by its community. Thanks to Unreal Engine 4's availability, it already has an editing suite that rivals the best, despite not having left pre-Alpha. This means that what would usually be mods put out many months after release are an integral part of the development process, shaping the core game. I spoke to Lead Developer Steve Polge about the influence of the community on development.

The pure amount of stuff the UT community puts out despite its relatively small size and youth is impressive. Every week a developer livestream highlights some new piece of art, weapon model, map, gun design or idea for how movement should function. Diving onto the official site and forums I found even more, including the images peppering this article that come from a particularly lovely set of screenshots by Viktor V, who goes by the handle Polyneutron. I asked Steve whether, when Epic started the project, they thought they would get this sort of response:

We had never done anything like this before, so while we knew Unreal Tournament had a passionate community of fans, we weren't sure how much community involvement in the development of the new UT we would get. We've been very pleased with the level and quality of contributions in every area, and now that UE4 is freely available to everyone and the game is starting to take shape, we're seeing even more contributions and involvement from the community. Its actually kind of overwhelming - it's hard for us to keep up with everything the community is working on.

One particularly prolific producer, Gooba, has been hard at work creating concept art for weapons. I'm a fan of his Minigun work, particularly the third one here, a more sleek and futuristic style than we usually see for a bullet-spewer and matching with some of his other designs. However, his most well-received project has been the Rocket Launcher, which is now official and had animation and model work started on it. Once Epic's in-house art team is finished with their current project, the Link Gun, they will start putting Gooba's launcher into the game. These guns join the Flak Cannon, Shock Rifle and Enforcer as community created weapons, all three of which have first versions of their models in current builds.

Giving this much influence over to the community, including iconic elements like the design of the Flak Cannon and Shock Rifle, things that players will see thousands of times in the game, is risky. Is there a worry that mistakes will be made, or that their attempts will not match up with Epic's vision for the game?

Our goal is to make a great Unreal Tournament game that is a great competitive shooter and appeals to a wide audience. While we can't satisfy everyone's individual ideas about what is right for UT, having a wide variety of perspectives is extremely valuable. We want to stay true to what is special about Unreal Tournament, but at the same time make a game that appeals to a broader audience than just the passionate fans that are currently engaged with us.

They are getting a broad range of designs: another contributor, Aberiu, had his concept for the Sniper Rifle accepted. It's much sleeker than other guns used so far, with less cosmetic extras and flair, but clean blocks of colour and an aura of efficiency to it. Part of the strength of this open development platform and, to an extent, the Unreal Tournament universe and genre as a whole is that there is room for alternative designs. If someone feels that this Sniper Rifle doesn't aesthetically match with other guns, or wants to create their own set of armaments from their imaginary space-guns megacorp, they can. Skins such as this will be available through the already-running UE4 Marketplace either freely or at a small fee from which Epic will take a cut.

Of course, it isn't just colour choices and inevitable hats that Epic is turning to the community for:

The community has had a lot of influence on the movement system and weapon functionality and balance, both through feedback and by prototyping and testing their own ideas. They've also helped us with our direction on performance, readability of gameplay, and how we support community created content on servers. Their contributions and feedback on level design, both in terms of competitive flow and innovative designs, have been invaluable.

That movement system Steve mentioned has gone through many iterations and arguments. It was the primary focal point for the game's design given its massive influence over how weapons are balanced and maps are made, so naturally it was the first thing the community was asked about. They responded with waves of posts and suggestions, obsessing over even the tiniest minutia of jump distances and speeds. Some of the most significant work came from raxxy, who you may remember put together the free-to-download super-early builds, and another forum user, Sir_Brizz. Together they created a movement system prototyper that allowed anyone to easily test new values and possibilities without the then-paid-for Editor.

The pair's own work with the tool created a framework that influenced the movement system as it exists now, in a nearly final state. They implemented an option to slide after falling, keeping some amount of momentum from a jump or dodge. It's a brilliant addition, speeding up movement around a map and making it feel much more natural. A side effect causes there to now be a rudimentary form of wall-running, letting you stick to them for short periods. This specifically is something still being experimented with by the community, finding the sweet-spot of allowing for skill-based movement without wrecking map design.

That has always been one of the main worries. A very contentious early decision was to remove double jumping as it made maps more awkward to build, requiring needlessly complicated calculations on any gaps or vertical climbs. It also solved the issue of UT 2004's 'dodge-jump' which could send players rocketing around at undesigned-for speeds and was faster than simply running in almost all circumstances. However, many saw this as taking away options from the player. It was the first argument to spike such a major reaction, but wouldn't be the last as everything from Rocket Launcher alt-fire modes to damage values of AoE projectiles are discussed daily.

Developers do take part in these discussions, posting on the official forums quite regularly. You can read all of Steve's posts here. He is not as talkative as Pete "YemYam" Hayes, a senior modeler who lives on the weapon forums. There he gives feedback on concepts and posts the threads of art he and Epic have decided to turn into official materials. Equally, many members of the team spend time in the official IRC channel, an unprecedented level of interaction for a big studio. I asked Steve if he was worried about the effect this might have on development - could too many disparate elements fail to gel into a cohesive game?

We love Unreal Tournament, and we don't want to lose what we think makes it special. Most importantly though, we want to make a game that realizes this experience and makes it meaningful and compelling for a large number of players, both past Unreal Tournament fans and new players. Having a large community of players enjoying this game and creating for this game is a vital part of how we define success.

It will be fascinating to see how this open development morphs as the game grows. Can a chat channel be maintained if there are thousands of users there? Will developers be able to post on official forums without being hounded, as they often are elsewhere? Once a larger group of newer players move in, will their ideas for the game integrate with those of the old school fans? Will they even be attracted in the first place or will the decisions made in development, influenced by that smaller, hardcore group, ward off the wider gaming audience?

Unreal Tournament's open development raises more questions than it answers, but for now it's fascinating to watch.