Non-Euclidean VR - Hidden Gem or Dead End?
When I first tried out Tea for God, I was absolutely blown away. I had never tried Non-Euclidean (also referred to as Impossible Rooms, Twisted Spaces) VR games before and I thought to myself “why is this not used in EVERY game?”. If you are unfamiliar, these games utilize overlapping rooms that otherwise could not exist in a physical space to give the illusion of walking through a constantly changing environment. I was a huge fan of Games like Onward and Pavlov and absolutely love VR shooters, but I couldn’t understand why this genre hadn’t been adapted into such a format. It was just so… immersive. That was the inspiration for me to create Bullet Style and I’m incredibly happy with how the game has turned out.
There are plenty of positives to such a format: the ability to physically walk, more natural usage of the environment such as when hiding behind cover or ducking under obstacles, and almost complete avoidance of issues related to vertigo or headaches due to thumb-stick based movement often found in VR games. With all of that said, the topic of this conversation isn’t about how great Non-Euclidean games are; It’s about the limitations and why I personally feel that there are very good reasons as to why these games are not more wide spread.
Play Space Constraints: Non-Euclidean games require a decent amount play space in order to function well and truly immerse players in the game. I have personally found that no less than a 6.5sqft x 6.5sqft area is necessary to really enjoy such games and even then, more space really does add to the immersion and reduce the claustrophobia of being in tight environments. Many players do not have access to a large space and may find themselves unable to effectively enjoy games as I have personally experienced through some of my own players. Such an issue can lead players to avoiding these types of games and restrict your market to players who have adequate space.
Dizziness and Repetitive Movement: Many Non-Euclidean games utilize some form of “looping” in their level design where players walk in circles through the play space. In many cases, these games will incorporate mechanics such as moving platforms or environments, obstacles to duck under or crawl through, or changes in direction. With that said, some players simply do not enjoy the feeling of walking in the same area repeatedly even if the game incorporates methods to mitigate the repetitive motion that comes from walking through a limited play space.
Design Considerations: Though the implementation of a Non-Euclidean world in VR will differ by game, a fundamental aspect of this type of system (in my experience) is the detection of where the player is looking/positioned in relation to the world. Not only is this a fairly large task to undertake with very little external resources to draw upon (I largely had to come up with my own system for such a game), care must also be taken for the design of the environment and what the player can see at any given time. This applies to props and entities within this world as well as the layout of the level itself. Because of the constant overlap of areas within a Non-Euclidean environment due to the traversal of the same space repeatedly, entities and props must be positioned in such a way that they cannot be observed by the player at the incorrect time.
Performance Considerations: Similar to the previous point, Non-Euclidean VR games also pose performance considerations as you are effectively splitting these worlds into traversable “chunks” or “rooms” that the player will move through in sequence depending on the layout. In my experience, these rooms typically encompasses anywhere from 1/3 to 2/3 of the play space (though this could vary heavily depending on layout, multiple pathways, etc.). The overlapping nature of such a system requires some degree of change in the environment through these various rooms to keep it immersive. This means that one must design environments that are either modified or changed every few steps by the player. Compounding this over dozens of rooms creates some considerations for chunking/loading/culling objects as needed to keep overhead performance as minimal as possible.
With all of this said, I cannot wait to pursue more of my own Non-Euclidean games as well as continually improve what I currently have available. Amid all of these constraints, I still feel that this is a format of game that needs to be utilized more. In an ideal future, omni-directional, realistic treadmills become relatively cheap and available. But in the meantime, I’ll be pursuing these types of games as they are the cheapest and most accessible option currently available for realistic walking in VR.