Arrowhead Game Studios Plays to Win

While still students at Luleå University of Technology in 2008, Anton and some developer friends entered the Swedish Game Awards at the Royal Institute of Technology. They developed and eneterd a game prototype that later became ‘Magicka’, and won the Game of the Year award. After taking advice from the staff at their university, they seized the opportunity and left their studies behind to start their own development studio.

Original Vision

As they began researching the essentials of running a company, the team entered an incubator program where they shared a small office together and continued to develop ‘Magicka’. “However, after two months, we realized that we had made a lot of mistakes and needed to make some fundamental changes,” said Anton.

“For one thing, we switched to working with a 3D game engine that we built in-house, and rewrote the entire game from scratch, aware that we had to make it commercially viable. Initially, it had a steep learning curve and was designed for a players seeking an intensely challenging gaming experience. With this in mind, we remade the game visually and changed the mechanics, but were careful to keep what was fun.

“To make your vision a commercially viable product, it's important to distill it down to its core. Once you have the very essence of your vision, try to wrap it in context that consumers will be familiar with, but never hide it. Make sure that the core is always visible. Many developers start with a great idea that everyone gets excited about, but then forget it throughout the course of development and lose that original vision.”

Game Plan

Throughout the process, they learned the importance of planning everything, from development to finances - especially when they ran out of money a third of the way into development. Anton said, “Planning is one of the things that we’re still working hard to improve, but the best way to make sure people can always contribute is to keep everyone informed. If artists don't know what to do, it's not that they haven’t been assigned a task, it's that they don’t know what suits the game.”

Communication relates to team morale as well, which was vital during the peiod of rewriting ‘Magicka’. “Being brutally open with each other has been a keystone of Arrowhead's development process,” said Anton. “If an asset or feature that someone created isn't good enough, they need to know it. If you tiptoe around the issue or worse, change their asset or feature without telling them the game will suffer, people will lose trust in each other and interest in the game.”

GAME ENGINES

Since that early decision to rebuild ‘Magicka’ using a game engine, Arrowhead has learned a lot about choosing and working with game engines. They are among the early adopters of the Stingray game engine from Autodesk. “While we had built a custom game engine in-house for ‘Magicka’,  on ‘Helldivers’ we realized that we couldn’t do everything ourselves. We wanted to focus on making games and to build our game the way that we had envisioned it. The Stingray Engine allowed us to do that,” said Anton.

He advised that when shopping around for engines, it’s best to try to keep an open mind and select the best one for each project, but in fact, a major time investment is involved in switching engines because artists, designers and programmers need to learn new workflows. Therefore, switching engines may not be worthwhile unless a team is working on a project that is radically different from their previous games.

“For ‘Helldivers’, we weren’t interested in engines that were essentially level editors. Modding tools and level editors are great for getting big results quickly, or if you're creating something similar to the game the tools were originally built for,” Anton said. “But such tools tend to be more restrictive. A pure game engine like Stingray has a steeper learning curve, but will allow you to create just about anything, even games that the engine’s developers would have never dreamed it would be used for.”

Avoiding Dependencies

They were looking for a flexible, lightweight engine that they could expand on, modify and use any way they liked. They didn’t and still don’t want any dependencies that are hard coded into the engine. “If, for example, the physics engine is developed by a third party that you as a game developer have no relation to and you find a bug, how do you solve that?” Anton said.

“Some of the larger engines may take several months to release a patch to fix the issue. If the engine is modular enough, you have the option to revert to an older version of the library or swap it for a different one entirely. Stingray still has some dependencies, but there are degrees to ‘hard coded’. My opinion is that the interfaces to those libraries have been sufficiently abstracted to allow us to make changes as we see fit.

“For most of us, Stingray has been the first commercial engine we have used. We've only recently started looking into Unreal Engine 4 and opinions are mixed. On one hand Unreal is a smoother experience than Stingray, but it can also be more restrictive. Unreal lets both novices and professionals get going straight away, but places limitations on what you can do. Stingray may require setup and research, but its more open, modular nature will allow you to do just about anything.”

Creative Control

Early on, the team had wanted to avoid publishers to maintain creative control over their games, but since the first project, ‘Magicka’, was bigger than anticipated, they rethought their strategy. “We made a trailer to gain exposure and spent some time marketing it to the best of our abilities, pushing it out to all the big sites like IGN and Gamespot. It worked! We were signed soon after, and Paradox Interactive became the publisher of ‘Magicka’.

We've had the good fortune to work with great publishers that have given us almost complete creative freedom. One surprising benefit is the on-going demand to see results. If you're just doing something on your own volition, it's too easy to over complicate something or get stuck treading water. If someone else expects results from you by the end of the month, you'll usually rise to challenge and want to impress them.

Helldivers

For ‘Helldivers’, the Arrowhead team wanted to create a game that held the essence of what they look for as gamers themselves. “We developed it as a top-down shooter, and because I was obsessed at that time with global politics, we also planned to make a commentary on geo-political warfare. Furthermore, we’re huge geeks, so naturally we love our Star Wars, Starship Troopers and Aliens. We mixed all of that together to make ‘Helldivers’ - a geo-political, warfare, top-down shooter, packed with pop culture references.

Sony Computer Entertainment, the publisher, asked Arrowhead to develop ‘Helldivers’ cross-platform for the PS4 and PS Vita. They agreed right away, not realizing how difficult it would be. “We figured it’s impossible, so let’s do it anyways. Miraculously, it worked out in the end, but there were many hurdles. Sony was very impressed, in fact, and asked how we managed to do it with a team of only three programmers.”

Cross-platform Development

Anton suggested a few factors a developer should consider before starting work on a cross-platform game, for example, the differences in the platforms’ controls, and what platform-specific features the game involves. He also recommended identifying the weakest of the target platforms, noting that one platform may have a weaker GPU than the others or less RAM and so on, and keeping the game to those specifications.
 
He explained, “We began developing ‘Helldivers’ for the PS3. When adapting the game for the PS4, we had relatively few issues in bringing the game up and increasing the fidelity of it all. However, the PS Vita for mobile came into the development schedule fairly late, and we discovered that it is much easier to scale up on a more powerful platform than it is to scale back to a weaker one.