We recently partnered with Victory Motorcycles (Polaris Industries) to create a first-of-its-kind virtual reality ride that transported International Motorcycle Show attendees in Chicago to Sturgis—through the Badlands and Needles of South Dakota—in the middle of winter.
Victory VR Experience, Chicago IMS
We experiment a lot at space150. spaceLab at space150 exists as a research and development incubator for emerging technology and media, the learnings of which we bring forward in our client work. When projects are completed, we often just see the shiny end result which hides a lot of the thinking and complexity involved. What follows here is an in-depth look at our early learnings and experimentation with VR and an examination of how that research informed our approach to this project. Because VR is so new, we also wanted to share our learnings from the creation process of #VictoryVR and the results of the final reveal to attendees at IMS Chicago.
Virtual Reality Experimentation
Our experimentation with VR began with the original Oculus Rift headset in the spring of 2013. We signed up for the Kickstarter and soon received the original Oculus headset. This was released shortly after Google Glass and while there was a lot of buzz around Glass (we even experimented with Glass and Computer Vision), when we tried Oculus for the first time we knew this was something much closer to reality. You can read about VR and watch videos, but you need to try it yourself to really understand it.
When the Oculus headset was first released there were a small handful of demos to try. The “Tuscany” demo, from Oculus, was a cliffside house that you could walk in and around.
Oculus Tuscany demo. Photo credit: Oculus
The Tuscany demo made it clear that Oculus was on a path to commercializing this technology enabling an amazing experience at a reasonable consumer price point (the original Oculus was 00). We were finally at a point where all of the hard work done in labs and research facilities was getting ready for the mass market. VR did not happen overnight, but it very quickly became much more feasible.
Another early and influential Oculus demo was the “Riftcoaster.” This virtual rollercoaster quickly became the default “go-to” demo for us and for many other people. It was simple to understand and it was a fully guided experience. All you had to do was sit down and wear the headset to experience it.
Riftcoaster demo. Photo credit: Wemo.io
This demo was built using the Unreal Engine; the VR rollercoaster was simply added to an environment that already existed. This showed how VR could be added to existing games and experiences.
The first time people put on the Oculus headset, we often see what we call “VR Face,” an open face smile and a “wow.” When you look around, you quickly get it, and it’s powerful. I’ve been a gamer and technologist my whole life, and there was nothing like seeing this for the first time. You are fully immersed in the experience through sight and sound, and you are disconnected from the real world. Most people often don’t even realize they are making this face.
A real, actual, VR Face
We have a fun office. It’s high energy and we have a lot of the latest tech around. When we get new things there’s usually a rush to see them. When the first Oculus Rift arrived at space150, the demand to try it was unprecedented. It turns out this high interest in Oculus gave us some great insights.
After a few days, we learned one thing quickly: many people experienced VR illness from the Riftcoaster rollercoaster demo. The entire demo is only about one minute long, and reactions were strong even during that short time. After the ride, many people had the same symptoms of motion sickness (nausea and dizziness). Many people needed to sit down after, and a few people had a strong enough reaction where they needed to stop the experience early. It didn’t seem to have anything to do with age or gender; it was a random occurrence across the range of people that experienced this. We tried a number of demos, and with less intense experiences, reactions were less intense. We knew at this point that the design of the experience would have a big impact on how people would react to it, and more importantly, be able to handle it.
We also noticed a curious thing: when people were in a VR experience, on the virtual rollercoaster for instance, they would move and lean as they would on the real thing. They would lean naturally and look ahead into corners. With our understanding of motion sickness, we knew we could use this to our advantage in the future.
When Victory approached us with the potential to use VR for a project, we wanted to be sure it was a good fit. While we experiment early with emerging technologies, we aren’t fans of using something just because it’s new. We always want to be sure we’re using technology to help our partners tell stories and create experiences that solve business problems.
With this in mind, we dug into the Victory business with their team, and quickly learned a powerful thing: when people sit on a Victory motorcycle, they are very likely to purchase a Victory bike in the future. The Victory team had two goals for the Chicago IMS show:
- Position Victory as an innovator in its category
- Drive traffic to its IMS booth and get “butts on seats”
How could we get people to sit on their bikes over all of the other competing bikes at the show? Often, the insight comes quickly, and the hard part is figuring out how to solve for it. In this case, the connection was an easy one: let’s put our riders on a Victory Motorcycle, and use VR to transport them to a sunny place far away from the Chicago winter. By doing this, we’ve done two key things:
- The rider has spent five minutes on a Victory Motorcycle (and we know this is a step towards purchase)
- They’ve had a new and thrilling experience, one that they’ve never had before, and we know this can create a strong and positive brand association
In the end, this was a win-win scenario. The attendees at Chicago IMS were able to see something genuinely new, and Victory has strong exposure to potential new customers and potential future fans.
Virtual Reality Challenges
VR is new, and while new things are always exciting, new tech presents unique challenges. There is a lot of experimentation in the VR space because it isn’t all figured out yet. There’s a discussion going on right now around how to tell stories in VR, and of course there are people on both sides of this issue. Some are very skeptical and resistant this new medium, while other people look at this as a genuinely new way to create experiences and tell stories. We couldn’t be more excited about the future of VR. That said, we wanted to outline some of the key challenges we found as we made this experience because shared knowledge helps us all do this better in the future.
Challenge: VR Illness
VR illness is a real thing. We touched on this earlier and you can read a lot about it elsewhere. We don’t have the exact details around why it happens (Oculus has its POV along with ways to avoid it), or why it only happens to certain people. The situation with VR is simple: a certain percentage of people will have a negative physical reaction to virtual experiences. A very mellow and relaxed experience will be something that most people can handle. A fast and thrilling experience will be too much for some people, but will be more thrilling for the people that can handle it.
A lot of very smart people are on this problem, and it’s a complicated one to solve. The CEO of Oculus, Brendan Iribe, is very prone to VR sickness and in many ways he’s their benchmark. They won’t release the final consumer product until he’s able to have an extended experience in the rift (interview here). Other things may sound silly, but aren’t: Purdue University is researching how a “fake nose” in the 3D experience can lessen the effects of VR illness. This may be a real and important thing. In the end, it’s likely many things combined that contribute to VR illness, and it will be a journey to solve them.
Motion and VR sickness can come from a number of things. One source is a disconnect between what you see in the virtual space and what you feel in the physical space. This is at the core of why we see people lean in a way that mirrors what they are seeing in a virtual environment. We wanted to give people the ability to do this, and to encourage it as a core part of the experience. To accomplish this, we envisioned a stand that would support the bike to allow the riders to lean and steer in a natural way.
Victory is a brand that’s all about American muscle and power. Given this, we couldn’t create an experience that was a mellow “lean back” type of experience. Our partners at Victory wanted us to push the experience, and they wanted to make it as exciting as possible. We needed to find the right balance between an experience that’s too mellow and an experience that’s too much.
Challenge: Guided Experience vs. Active (Game) Experience
This sounds like a simple decision to make, but it’s actually one of the more complicated things we had to face. A lot of the VR experiences are either fully guided “on rails” (like the rollercoaster example) or wide open (games that give you complete control). For this project, we needed to be sure what we created was a good positive brand experience while giving the rider some control over how mellow or fast their experience is. It turns out the grey area in between these two is a challenging place to be.
A fully guided experience gives the rider no control; everyone has the same experience. This was an option for us, but we really wanted to let people have some control over the speed of the experience. A fully active and open experience would allow people to do things that we didn’t want them to do, for example driving off the road. Allowing for this much control creates many problems, e.g., Do you end the experience early if they drive off the road? Do you reset them to the middle of the road? Do you put invisible barriers in place to prevent them from doing this? All of these are very unrealistic, and we wanted to create a thrilling yet realistic experience.
We’ll explore this in more detail in the following sections. As you decide to give some control, it’s a challenge to determine how to give some agency to the person that is experiencing this, while keeping enough control to ensure it’s a good experience that’s positive and on-brand. The more control you give the rider, the more dynamic the experience will be and this means more work and planning to create the full experience.
The Victory VR Setup at Chicago IMS
Challenge: Control and the Virtual Self
We have some passionate riders at space150, and we immediately started thinking about how much control to give the rider. Throttle, braking, shifting, steering and leaning are all core to a riding experience – why not include them all? We knew that we would have a diverse audience: non-riders and experienced riders, non-gamers and gamers. We also knew with high certainty that this would be their first experience with VR. We decided that the best way to create an immersive, but all-inclusive experience would be to guide them through the experience while giving them throttle control to speed up or slow down at will.
One problem this creates is that the rider can go too fast. If a rider had the throttle maxed and was heading into a corner we would need to govern the speed down to an appropriate level to safely to make the corner. Given all the control options we had, we felt the benefit of giving some control was worth the small disconnect that could happen in these scenarios.
The acceleration also allowed us to have some fun with our “virtual self” in the experience. As your real hand moved the throttle on the real bike, the virtual hand moved accordingly. It’s a unique experience to see this in VR, and a cool thing to be able to bring to people for the first time. It’s another way to connect the real and virtual worlds. VR has the ability to manipulate our mind-body connection and it is a powerful thing to experience.
Virtual hand mirroring real hand movement
Speaking of your virtual self, a critical piece of any VR experience is the virtual “you.” In the real world, you see your hands, feet, and torso. In many virtual experiences, this is left out. It’s a real challenge to do this well, and to make it look believable. In experiences where your body and self are not represented, you exist as a floating camera in the air. As we started to play with early demos, the experience of seeing your body is a powerful one, so we knew that we needed to include it. When people look down and see a virtual version of themselves, it’s a much higher level of immersion and it grounds you in the experience.
Riftcoaster HD, body (self) is not represented
Titans of Space, body is represented
We knew we’d have male and female riders that varied in height and weight. With six weeks to build the entire experience, we solved for this by creating a gender neutral model in full leather and gloves. We targeted an average height and weight to accommodate a wide range of riders. In the VR space, your mind will make a strong connection to a virtual self, even if it’s not exactly you. Our experience was designed around having people grip the handlebars during this whole ride – so your real hands would match the position of your virtual hands. We used this natural alignment to our advantage.
We designed leaning as a core part of the experience because we wanted people to be able to mirror what was happening in the virtual space. At the show, the vast majority of people did lean in the real world as they would in the virtual world. One fascinating thing we observed is that non-riders (people that had never driven a motorcycle before) leaned naturally along with that was happening in the virtual experience. For them, the leaning felt natural, even though it was a brand new experience, and they’ve never done the real thing.
To complement people’s natural inclination to lean, a custom motorcycle stand was built that allowed the bike to lean from side to side so riders could mirror curves as they rode through the experience.
Takeaway: Creating immersive VR experiences goes far beyond just having good graphics and visuals. The ability to control and customize the intensity of the experience is key, and allowing people to move naturally and see their virtual self helps ground the experience while creating a strong connection between the real and virtual worlds.
Challenge: The Virtual Environment
Choosing the location of the virtual ride was a big part of our initial planning. After discussing a few of the most iconic rides in the country with the Victory team, we landed on a “best of Sturgis, South Dakota” ride through the Badlands and the Needles Highway. Our goal was to create an experience that was about two minutes long and allow riders to escape to this sunny summer destination.
We created a “best of” condensed ride that brings together the best moments of both areas. The details are important: the Badlands and Needles areas have unique natural geological features that we wanted to incorporate. We also wanted to include the iconic Wall Drug signs (you can’t miss them, right?). If people have been there before, we want them to see these iconic things to bring them back. If they haven’t been to this location in real life, we wanted to bring forward the truly unique elements of the Sturgis ride.
Iconic YOLO sign in the Badlands
Entering Needles Hwy.
As we started to create the environments and the road through it, we realized that we had a lot to do in only two minutes, and that the balance of straightaways, long sweeping curves and tight corners had to be balanced. Fortunately, we have a lot of riders at space, and feedback from Billy, our CEO, and others helped us get the corners down so they were smooth and felt natural with how the bike moved and leaned. Shifting into and out of corners was another important detail that took a lot of refinement. If we had more time, we’d still probably be tweaking these two things. As we started to refine the path through both environments, we learned that while similar, they each had enough variation that we needed to treat them as two very distinct environments, each with unique challenges. While there was some overlap, it wasn’t as much as we had initially hoped for.
Takeaway: Focus is critical. Going into the design of this, we felt that the Badlands and Needles area should both be included, and would be similar enough to have a lot of overlap. If we had to do this again, we would focus on just one area, likely the Badlands. When it all came together, the openness, the straightaways, and the long gentle corners really felt amazing. We could have created enough variation in this one environment and could have put more work into one place, rather than dividing it among two. One of our goals at the end the experience was to leave people wanting more and we could have achieved this ending near the needles, or by seeing that area off in the distance. Focus is always a challenge, but it’s always important to balance quality and quantity.
Challenge: Immersion through Sight, Sound and Feel
When you look into the Oculus headset, you see two separate images. Each eye sees a slightly different view of the world, which is how you get the convincing 3D effect with VR. When you see videos or screenshots of VR experiences, you’ll often see this view of what each eye is seeing.
Stereo Oculus view (note each eye view is slightly different)
If you look closely at the image above, you’ll notice that the images are different. If you overlay them on top of each other, you can see it more clearly:
Overlapped stereo Oculus view to show differences
The reason that VR is setup like this is to mimic how we see in the real world. Our left and right eyes are separated by a few inches, and each sees a slightly different view of the world, and this allows us to see in 3D. For VR, this means that every time we draw our virtual world, we need to do it twice, once for each eye. We’re basically required to do twice the work as a normal 3D experience, and this impacts how detailed and rich we can make the virtual world. Additionally, for the VR experience to appear smooth and fluid, we need to draw the scene at 75 times per second (75 hz) or higher. If you dip below this, the whole experience appears choppy, and this can quickly lead to VR illness and a subpar experience.
Modern videogames can look truly amazing, and many of the VR experiences we’ve seen don’t have the same level of visual fidelity. We knew that performance would be an issue, so we planned ahead and purchased the fastest gaming computers we could buy. This included the very latest NVIDIA graphics card that has some VR specific features. Our thinking was that we knew we’d hit a wall with performance, so we wanted to push that as far as we could with fast hardware.
The computer, liquid cooling, and the NVIDIA GTX 980
Once we had the virtual world roughed in, we quickly found our performance limits. Creating these environments is always a balance between the size of the world, number of things in the world, and the visual quality of the objects in the world. All things equal, a smaller world can have more detail, while a larger environment will have less detail.
We knew we wanted to push the visuals beyond what we’d seen in other VR experiences, so we had to find the correct balance of these things. When we were planning the environment, we all decided on a “pre-dusk” ride that would have the sun low in the sky, and allows us to have very warm, end-of-day lighting. We also wanted to use the low sun to our advantage, and have that be a big part of the experience. As you’re winding along the roads, you see the sun, it creates dramatic shadows, and we used some other lens effects to simulate riding in a helmet. All of these things add layers of complexity and visual depth that help mask the fact you’re in a virtual world.
We wanted the initial moments of the ride to be the most impactful, giving a great first impression and to garner the “VR face” we mentioned earlier. To do this, we focused our energy on creating the most dramatic view right when you start. This meant an amazing sky, driving towards the sunset, and a large view of the open world. We also wanted to slowly introduce riders to curves, so we started mellow then ramped them up to become more intense.
One performance issue common with games has to do with the amount of detail meshes we can use (things like foliage, rocks, shrubs, trees, and leaves). The more of these you have, the more organic and natural things will look, and the more you can use to hide transitions between two different objects (like the ground and the trunk of a tree). We kept adding more until we hit our performance wall, then backed off a bit.
The model of the bike was a key feature, and needed to be true to form
Takeaway: In the end, we like where we ended up, but we really wanted to push things further. If you want to create impressive visuals you’ll need powerful hardware to do so. The process of game development and creating these virtual worlds always involves a series of compromises and tradeoffs, and you are required to make challenging decisions at all points along the way.
While much of the focus on VR is all about the visuals, audio is a critical part of the VR experience, and is often not looked at with equal importance. Once we made the decision to allow users to have some control over the throttle and speed of the ride, our audio needs changed: we needed to create a dynamic sound engine that would allow us to represent the correct sound for all of the different speeds that our riders would be experiencing.
To create a whole dynamic sound engine, we grabbed a bike and rented a dyno at an amazing local performance shop. We also found one of the leading vehicle sound recording engineers in the game industry to help us record this. We needed to record all of the individual parts of riding (engine start, idle, acceleration, various speeds, etc), so we could assemble them together in real time during the experience.
Ultimately, this was worth it and made a huge difference in quality and sense of immersion. When you hit the throttle and hear the engine rev, it’s a strong one-to-one connection and you know you are in control. When you’ve heard the real bike, and you ride in the simulation, you immediately recognize the sound of the bike as the real thing.
Dyno and Sound Recording
Takeaway: Audio is often under utilized in experiences like this. As we gave control to the user, the complexity of the simulation increased, and we went from a static fully controlled audio track to one that needed to react to users input in real time.
VR is a very immersive experience that takes over vision and hearing. Many experiences add additional elements such as fans for flying or skydiving, mist to simulate water, and even smells. For this project we really wanted to make it feel like you were on a motorcycle that was running, so we asked the question, is there any way to have the actual engine running? We quickly learned the answer to this was no, so we had to look elsewhere.
We have a lot of collective experience in the audio world, and we knew about the existence of tactile transducers. Tactile transducers give you the amazing low-end bass of a huge sound system in a very compact form. These are used in a number of scenarios: in pro audio, in military and training simulations, in theme park rides, and obviously in gaming.
Tactile transducer motion & sound testing
While we were in the planning phase of this project, I was at CES, and was able to look at a few of these in person and talk directly to the engineers and product designers at Earthquake Sound. We quickly settled on a unit that would be perfect for this project. When they arrived, we connected the transducers to the bike, and ran some test sounds through the system – and it was amazing. It took some work and experimenting to tune the levels and frequency of the sound, but once we did this we could very accurately simulate the feel of a real motorcycle with its engine running. In connection with the audio of the actual motorcycle, the effect was incredibly realistic. When we have people run through this experience, it’s very fun to see their reaction when the engine “starts.”
Takeaway: It’s a given that VR experiences will have video and audio elements. Depending on the experience that you’re creating, adding additional elements that further enhance the immersion is very powerful. For us, the feeling of a motorcycle is important, and it wouldn’t have been the same experience without the tactile transducer.
Challenge: Integrated Controls
Many VR experiences at trade shows start by just placing the headset and headphones on you, and handing you a game controller. This can work well if you’re a gamer and have experience with game controls, but we wanted to be more integrated and natural with our input.
A typical Oculus experience, CES 2015
We made the decision to use the throttle input on the bike. We wanted to wire in direct to the engine if possible, but we didn’t know exactly how to do this. We knew that in the automotive world there are standard interfaces that allow companies like Automatic to get at all of the details of your car (throttle, etc.), enabling any car to be a “smart” car. We didn’t know if this existed in the motorcycle world, so we started to investigate. It turns out there is no equivalent for motorcycles, so we needed to find a way to work with this specific bike. James Squires, our SVP of engineering, had some experience with engines and thought there would be a way to tie directly into the throttle body, which links the engine to the throttle. A quick teardown of the bike confirmed this indeed worked, and our experience in gaming helped us to understand that we could directly tap into this control. We ordered a custom control board from a game company in the UK, and were on our way. In the end, we were able to turn our real Victory Motorcycle into a giant USB joystick.
Integrated USB throttle control testing
Takeaway: Experience outside of the pure computing and digital space is important so you can understand what’s possible. While it’s easy to use game controllers or other “out of the box” controls, we wanted the input and interface to be native to the bike. This meant we needed to have a better understanding of how the engine controls work to be able to integrate natively, and have a simpler, more natural interface.
Challenge: Trade Show Environments
Trade Show environments can bring unique challenges, and the Chicago IMS was no exception. They are loud events, and this particular show had a booth nearby where they occasionally fired up and revved a real motorcycle (we’re looking at you, Triumph).
Much of our onboarding process was presented in the actual VR experience, but it was still very important for us to share some of the details of the ride directly with the rider. The loud environment presented a challenge, and we may move some more of the onboarding information more into the experience in the future.
The crowded IMS showfloor
Another environmental challenge was lighting: the Oculus headset determines the orientation (rotation) of your head, as well as the position of your head as you move your head back and forth or front and back, to position you properly in the virtual world. To accomplish this, the Oculus headset has a camera that detects infrared light, and the Oculus headset has a number of infrared emitting lights. This light is not visible to us, but is used as an invisible way to detect the position of the Oculus headset. During all of the testing we did, we never encountered any issues with the camera. When we transitioned to the trade show environment, we saw some jitter and quickly found that the lights in the trade show were emitting infrared light, causing this interference. We were able to adjust the camera to minimize the issue, but when we do this again, we’ll want to have more control over how much light reaches the headset, and construct a darkened area or canopy to isolate the setup from outside interference.
The Oculus infrared camera
Takeaway: Trade Shows are busy, loud, and noisy environments. It’s always important to scout ahead as much as possible to understand environmental constraints. Nothing compares to real testing, in the final environment, especially when you are using an emerging technology like VR.
We knew that most people experiencing this ride would be in VR for the first time. We also know that when you try VR and you first look around the virtual environment it’s a real “wow” moment. Our goal was to make this moment as big as possible. It has been our experience with several VR demos that there is little onboarding and the user is immediately placed directly into the virtual environment. Most people need time to adjust their headset to see clearly, and to adjust headphones and any other components of the experience. We wanted separate these two things, and provide the best reveal possible. We had to onboard people in the right way.
Our approach was to be very specific about what happens when people start the experience:
- Get people seated comfortably on the bike. At this point, we explain what they’re in for, how the experience works, how long it will be, and to let us know if they experience any VR illness or any other issues.
- We fit the Oculus headset on the rider, and explain that they’ll need to do some adjusting to see properly. For people that wear glasses especially, the headset needs to be fitted well, and this can take a few tries.
- At this point, the experience begins, and the rider sees welcome screens that show them how to interact with the experience and also demonstrates how to use the main throttle control input.
At this point, the rider has been seated on the bike, has a general idea of what they’re in for, and the headset has been clearly adjusted. They’ve had an intentionally minimal VR experience that shows them just enough to know that’s going on, without giving away what is next. When the rider hits the throttle to start the experience, they hear the bike start, feel the rumble of the bike (the tactile transducer), and the virtual world fades in to reveal the virtual bike, and their virtual self – all at once. They are in an enclosed motorcycle trailer, and after a few seconds, the door opens and they see the much larger world they’re in.
World / Environment reveal
Takeaway: It’s important to understand your audience, and their experiences, especially with a new medium like this. We were able to assume that this would be the first VR experience for our riders, and we onboarded people with that assumption. With this approach, we were able to maximize the wow factor, and leave a strong and lasting impression.
The Chicago IMS show was over Valentine’s Day weekend in February. Our significant others were prepped for this ahead of time, and we were also ready to spend the weekend sequestered at the convention center and the Hilton Rosemont. We haven’t seen much published information about these types of experiences, so we wanted to share what we learned.
The welcome sign for the experience
We had 24 hours of open booth time over three days, and we ran 297 people through the experience, which works out to just under five minutes per person. The line started at the beginning of the first day, and never let up. At times, the wait was over an hour – people were patient and excited – and we talked to them to find out what they knew about Victory and VR in general. After the experience, we ran everyone through a quick survey to gather feedback that would help us improve the experience in the future. Talking to people immediately after the ride was a thrilling experience and helped us get the best feedback possible.
Out of 297 riders, we had just five people stop the experience early due to VR illness. This number was much lower than we expected – we were guessing, worst case scenario, this could be as high as five or ten percent of all riders. To see the actual number this low was a good sign that we designed the experience appropriately.
The IMS show confirmed our thoughts that awareness of VR and Oculus would be low. This is clearly a new technology, and very few people have actually tried it. Additionally, we were able to introduce the Victory brand to many new people, which was one of our primary goals. Over 90 percent of the virtual riders were new to Victory, and this was their first connection with the brand. The riders also had a lot of feedback on what else they’d like to see in the future, and where they’d like to go next.
In the end, we set out to bring people a new experience, and this was certainly the case with more than 90 percent of the riders being non-Victory owners. We’re proud to have introduced a lot of potential customers to Victory Motorcycles in a very hands-on and innovative way.
The physical and digital worlds are increasing blending together, and virtual reality is a new medium unlike anything before it. Our partners at Victory were willing to experiment with this new technology, and we knew that we could pull this off.
You’ll see more of this experience in the future. We learned a lot from what we did, and have some amazing takeaways and learnings to bring forward in this and other VR work we’re doing. As more of these experiences are created, everyone is gaining a better understanding of what it’s best at.
In the end, you can look at this VR experience, and many others, and see something that looks very simple. In the end, that’s our goal: we want to deal with the complexity and bring a simple, compelling experience to a mass audience.
With Virtual Reality, the devil is truly in the details.