ARES: A ROOM SCALE VR NARRATIVE
Amit Garg, Josh Fisher, Wesley Wang. Advised by Prof. Janet Murray
In the near future, visitors embark on a dangerous space adventure. Within moments, the journey goes haywire. Lost and alone, the player finds themselves stranded. This interactive narrative makes use of Impossible Spaces, a VR natural-locomotion concept for room-scale experiences that enables the user to enjoy seemingly limitless natural movement throughout a digital world without breaking their sense of presence. This distinctive, immersive experience tests users’ survival skills and offers a fresh and exciting experience.
ANTECEDENT RESEARCH and insights
We started the project with first making a list of all well known VR experiences available for the HTC Vive and then we tried them all out one by one to see what worked and what didn't.
The learnings we gathered from this experience are as follows:
- Most VR experiences made players use either joystick movement or teleportation movement to traverse their surroundings.
- Joystick movement made us nauseous and VR sick.
- Teleportation movement while widely used, still disoriented us.
- Most interactions used haptic rumble to reinforce interactions.
- Trigger button on the HTC Vive controllers was universally used to grab objects instead of the grip as it felt more natural to the players.
- Some games like Budget Cuts try to solve the disorientation caused by teleportation by showing a preview of the destination before players made the movement action final.
- None of the games used any ways to redirect or move the players in a manner that they don't reach the edges. They simply placed barriers.
After our VR antecedents research, we decided to come up with an exciting scenario that would suit a room scale setting and also be expansive in its intended scope. We actively tried to employ successful interaction principles we learned from our antecedents review as much as we could in the narrative.
After watching a talk on the use of pre-visualizations in Dreamworks' VR Process and a video on Stanford's VR prototyping method, we decided to try role playing in real life to figure out how an interactor will interact and move within the spaces we had conceptualized.
In this role playing phase, we let our imagination take over and only followed the limitation that once we reach the end of the physical play space i.e. we are about to walk into a real wall, we would need to turn the interactors around without making it obvious to them that they are out of the physical play area.
To better capture player movement, I made a rough top down layout diagram in Sketch showing each space the player would explore and how the player would move in a high level manner in that space. This served as a great reference throughout the building process.
1st ROUGH prototype in unity
During the coding, I realized that if we tricked the player into thinking they entered a narrow shimmy space where their field of view is limited by the space, they wouldn't be able to detect if we were changing the architecture in the room they were just in.
So, I coded a simple demo in Unity to test this hypothesis out and decided to test this during GVU Demo Showcase held in Fall 2016.
After demo'ing this simple test at GVU Demo Day Showcase we quickly came to three conclusions:
1. People loved the interaction of breaking rocks.
2. People were squeezing their bodies to fit into narrow virtual corridors.
3. Most importantly, changing the level drastically immediately made them realize that something was off.
IMpossible spaces and locomotion
The Impossible Spaces concept was first published by researchers from ICT at USC in 2012. Impossible Spaces are self-overlapping architectures that are manipulated when the visitor isn’t paying attention to them. The are termed as "Impossible" because such a space cannot exist in real life architecture.
Impossible Spaces use self-overlapping architecture to maximize the natural walking space in an immersive environment.
Designing impossible spaces with narrative intent
After discovering the concept of Impossible Spaces and how similar it was compared to our original idea of turning people around, we decided to use the concept of Impossible Spaces to affect how we build our environments and how much we let two sections of our level overlap.
Building it in UNIty
We used Unity and the opensource VRTK library to build the experience. We also use invisible Trigger Planes that when hit by the player load the next part of the scene without them realizing. We use a framework of grouping objects under parent objects called Appears and Disappears.
Appears and Disappear groups had some common elements which are not supposed to change with the point of view of the player. The sections of the level which are not common are the ones that are being loaded in and out without the player realizing it.
Lastly, we also heavily used the free assets from the Unity Asset Store to our advantage as they saved us a lot of time during development.
Play testing and impressions by players
We did user tests with 13 people and found that when asked to draw the path they took in Ares, most people perceived the space as a downward spiral path. This was a significant result as they were actually going roughly along a square-ish path along the perimeter of the physical play area.
People also reported high sense of presence when they were administered the Sense of Presence Questionnaire devised by Witmer and Singer et al.
No one reported any motion sickness due to our experience which was tested by the Simulator Sickness Questionnaire devised by Kennedy, Lane, Berbaum, & Lilienthal.
ARES at CHI Student game Competition 2017
Ares was published and chosen to demo as a FINALIST in CHI Student Game Design Competition 2017 held at Denver in the innovative game design category by a panel consisting of eminent game researchers, namely, Josh Tanenbaum and Florian Floyd Mueller.
I setup and demonstrated Ares for a live audience at CHI with a lot of help from this great talk by Owlchemy Labs and got lots of feedback on our narrative as well as our interactions.
People were excited when they were breaking rocks, timid when crossing the canyon with a fall and were squeezing their bodies into a narrower volume when going through the shimmy and crawl spaces.