Polypheme: A VR Mech Game
Premise
Polypheme is a VR game where the player takes control of a walking warmachine. The player must pilot this ‘pheme’ from its cockpit, in order to defend the last bastion of their civilisation from an unstoppable enemy. The cockpit is full of levers, buttons and other VR interactables that the player uses to march their warmachine around on the battlefield.
My Role
I pitched Polypheme and put together the team to build it. Together with seven other students, Polypheme was created in under 16 weeks as the final assignment for my Advanced Diploma. My role on the project was twofold. I worked as a lead designer and coordinated the project as the project manager. When I wasn’t doing that I worked as a narrative and mechanic designer, iterating on the interactables in the cockpit and the tutorial that tries to acclimatise players to the Pheme.
Environmental Storytelling
One of our focuses with Polypheme was to create a rich and vibrant world. The player is immediately immersed when using VR so it was a simple decision to create things in the world to tell a story. Polypheme makes extensive use of various Audio mechanics to help communicate a story to the player. These included a local radio station with a news presenter from the City Pheme. This radio station would often be interrupted by propaganda posted by the enemy faction. The tutorial is also delivered through a separate in-game radio (allowing the ‘stereo’ to be turned off). This tutorial was delivered by a voiced character called ‘Adjutant Hunter’ and would often site basic references to the ‘degens’, the enemy faction. These small touches helped to create a much more lived in space within the cockpit. Overall, the environmental storytelling helps to tie the experience together in a way that other approaches like narration wouldn’t have, and the experience will be invaluable to me in future developments.
The Incomprehensible Machine
Polypheme uses the ‘incomprehensible machine’ approach to designing puzzles to try and create an experience that is both difficult and satisfying for players. The unique perspective of VR originally created a loose, difficult to manage play space where the player was unable to competently engage the enemy. Players of Polypheme lack the situational awareness of their surroundings that they have in a first-person shooter as they have to pilot the Pheme from within, but they still have to navigate a space as if it were a FPS.
The Incomprehensible machine approach allowed us to focus on the Cockpit and all the interactions the player has with it in a different way, prioritising experimentation and exploration over ease of use. By creating a puzzle like space for the player to manage, they often found failure within the game to be less frustrating than when the controls were presented as simple. Each time they died, they gained a better understanding of how the mech worked, so they always felt like they were making progress. The initial tutorial section only focuses on getting the player accustomed to moving around the mech. It doesn’t touch on the more baffling mechanics, leaving them to the player to discover through experimentation. Overall, it helps to make an overwhelming play space a little more approachable to the player.
RTS Mechanics
One of the elements that we wanted to experiment with more when we first begun development was getting the player to manage more than just their own mech. We wanted to create a situation where the player felt like they were in charge of a small group protecting their citypheme. This led to the implementation of a Command Hub that allowed the player to issue orders within the battlespace to fellow ‘soldiers’. Trying to accommodate the basic tropes of RTS games within the VR space was an interesting design puzzle, allowing us to experiment with many different approaches to issuing commands. Overall, we settled on basic hand gestures to balance out the lack of keyboard shortcuts, with varying success. More time and iteration would have undoubtedly benefited this system, but the experience was still worth the time we put in.
Diagetic UI and VR Design
One of the key features of Polypheme was its focus on Diegetic UI. None of the user interfaces within the game are presented in a way that break the fiction. When the player is commanding their allies, they do it in fiction utilising a drop-down interface from the roof of the cockpit. Collecting reinforcements, checking objectives, using cheat codes. Everything uses the in-game UI, making it relatively intuitive to interact with.
It also meant that we could tie these facets of the UI into the mechanics of the mech. The UI uses power, and only turns on when power is available in the Pheme. Various controls would flash in and out of view when power dropped. In a longer development time, all these facets could also have begun to take effect within fire fights and other moments of intense play, tying back into the Incomprehensible machine approach to designing the cockpit.
Project Conclusion
Polypheme was developed in less than 16 weeks with a team of 8. The project was ambitious and holds amazing potential. There were a lot of time saving processes we had to utilise in order to achieve as much as we could in as little time as possible. The consequences for gameplay from some of those processes was drastic. I’m incredibly happy with how the interactions within the cockpit went, particularly since me and my team hadn’t worked on a 6 degrees of freedom VR project before, but the actual process of fighting and traversing the map feels particularly clunky.
For example, when developing Polypheme we unified the character controllers for the player, ally and enemy mech units into one single script. This allowed us to use polymorphism to cast the scripts in lots of different places and unified the different functions. It also meant that we used a single type of animator controller for all the mechs as well as a single rig and that allowed us to reuse animations across many different models. Unfortunately, this made refining the actual mech combat play experience very difficult. With all the other work that needed to happen, trying to also refine animations for the mech and clean up the odd interactions between animation driven movement and ai pathing sucked up large chunks of time. Overall, if I was to attempt this project again, I would try to move away from total mech based combat. If we had focused on a different way of implementing the player character mech and had more vehicle based enemies, a lot of our issues for refining the experience would have evaporated, giving us much more time to iterate on all those aspects of the experience that mattered.
That being said, Polypheme is an amazing piece of work. A VR project put together by a team of 8 students in less than 16 weeks, with more than half a dozen types of VR interaction, dozens of controls and underlying systems to give them consistent behaviour. All in a massive level where the player can charge around in a mech killing multiple types of enemies that fall out of the sky in drop pods. The player can call in reinforcements, issue orders to them and hear their responses in real time. Inside the cockpit there is a radar that is devoted to revealing the terrain around the player and will also show where enemies and allies are. This is just a small summary of the systems and mechanics available in Polypheme.
This project was flawed, but it was ambitious and interesting. It looks amazing and it feels great to take control of your own mech. It was one of the most rewarding things I’ve worked on in my life, and I hope that other projects can equal it in future.