I’ve given a lot of thought on how to write this post as I’m going to touch on very delicate subject. It’s one on which my view (and that of the team) is not completely set yet. From my point of view as a developer, I’ve witnessed how the perception of Virtual Reality is evolving. Part of this is due to a better understanding of the forces and weakness of this new medium, but it’s also due to the fear of VR, a fear that is mostly attributed to health concerns and simulation sickness acting as the flag bearer. There’s a stigma around this topic and I thought it was time to get some well-deserved new perspective on what it entails and maybe bring about a new point of view on the subject. Bear with me a little as I lay out my thoughts on this, I swear there is a point!
Evolution of VR Mentality
When we started tinkering with the DK1 back in the beginning of 2014, the VR scene was pretty much two things: first-person view horror games and rollercoasters. A lot of people saw the future of VR entertainment as that kind of experiences. It was about “thrills” and seeing the world from another person’s perspective: that was most easily associated with first-person view games.
Motion sickness was a thing, but it was mostly seen as a technological limitation, something that would go away naturally as the tech evolved. It was not uncommon to see one person feel some simulation sickness only for another to comment that “It’s going to be solved or get way better with the DK2”.
One of the first demo developed by Oculus to showcase the Rift was a first-person experience; the Tuscany Demo.This demo is what introduced a lot of developer I know to VR, and it contained the dreaded right joystick yaw control (more on that later).
The concept of “VR Legs”, the fact that you could acclimate to the discomfort caused by some VR experiences, had a pretty strong following. On the team, I remember we were telling ourselves that we would do the most hardcore experiences (like Cyber Space) in order to get better in VR. In some respects, this did work. We got better at handling some VR related discomfort!
The reason why I mention all this and what primarily led me to write this piece is that since then, there has been such a paradigm shift in what a VR game should be, that it made us reconsider many aspects of our game. Basically, as of now, creating a VR FPS is pretty much considered heresy. The demos that once were the greatest things in the world, that made me a believer in virtual reality in the first place, are now scorned by most.
While first person VR games are still coming, this evolution led to the recent insurgence of third person perspective VR content. This can be seen in numerous articles around the web but also in recent announcements at E3 like Insomniac’s “Edge of Nowhere”.
Don’t get me wrong, I’m not saying this is a bad thing! There are good reasons behind this; locomotion is way less intrusive in third person, and locomotion in VR is a problem.
Locomotion: The “Unsolvable” Problem of VR
While things like low frame-rate and high latency can create simulation sickness, the real issue lies in movements and how the vestibular system interprets what the brain is telling it. For some, moving around in translation or rotation or both can get extremely uncomfortable, extremely quickly.
In parallel to the evolution of VR content creation, developers were hard at work finding solutions to this problem. As we entered pre-production on FATED, we were already aware of some of the issues. We knew about the major concern of standard FPS control: player yaw rotation using the right stick. Some months after we started, John Carmack, one of the leading authorities in the domain, said this about it:
So, naturally, we wanted to figure it out: solve the issue and make our game the most comfortable VR experience out there. Thus came the time of the hunt for the Great VR Holy Grail! *Cue the Indiana Jones theme song*
While we found some interesting (and some horrible) ways to control rotation, most of the popular ones were already out there.
Solution #1: The “Comfort Mode” Control
There is a pretty popular movement mechanic known as the “Comfort Mode” that strives to resolve this issue. You can see it explained here.
While this solved the problem at hand for some, it did also create new ones. First, for a lot of people this was breaking immersion. There is no way that “skipping” the actual rotation movement can cut it as a “natural” way to move around. Secondly, disorientation: while using the feature sparingly could work, trying to use it in a more action based setting was quickly making people unsure as to their whereabouts in the game world. And then there is the matter of precision. If you get to a point in your game where you need to be at a certain angle, then this method is bound to fail.
While these are all new problems that may also have their own solutions, there is one single thing that still stays true: this does not work for everyone. In fact, some people reported feeling sicker using this new control mode.
Solution #2: Cockpits, Cockpits Everywhere!
The second solution came about quite naturally in various demos that actually needed them, like in games where the player is sitting in a spaceship. I use the term “cockpit” here to designated any locomotion mean that as a reference frame that the player can relate to. Racing games and mech-warrior style games are two other examples.
Quite interestingly, this had a strong positive effect on the way players experimented simulation sickness, even going as far as completely removing the unpleasant feeling for some! Great, but now that developers found out about this, the internet exploded with ways to had that frame of reference to everything. Here are some of my favorites, not always for good reasons.
There is this demo that is awesome for its novel ideas to handle movement in VR. It’s simply called “Movement Experiments”. All are interesting movement mode but the one I want to point to is the “Canvas Mode”. What this mode does is create a virtual “cockpit”, called “canvas” in this case, whenever the player rotates around. While the author says it works (and it probably does for some people), I found it quite intrusive, and immersion-breaking. Still, the demo is worth a try, so check it out!
Floating HUD (Helmet Mode)
There’s the floating HUD solution that is basically making the UI the frame of reference for the player. That, coupled with something like the visor (or helmet) of the player character can give the desired frame of reference. A good example of that can be seen in the upcoming game Adr1ft. Unfortunately, not all game settings permit this kind of implementation. A Viking with an astronaut helmet…maybe for FATED 2!
By far my favorite is the virtual nose, which is basically the natural frame of reference of humans. This research that pretty much flooded the VR space when it was first released is basically the extreme representation of the “cockpit” concept. We tested the idea on FATED, and promptly removed it. For some reason (I’m blaming low FOV), what the player ends up seeing is two partial noses, one on each side of the screen. It’s very troubling for the player and just feels out of place, I really don’t see this becoming the miracle pill to solve motion sickness.
Solution #3: Swivel Chair
The idea here is that since rotation in VR is such a problem, why not make players rotate on themselves in the real world? This does work of course, but with the headset horribly limited by a metric ton of wires, this does not always end well. Some folks out there like the people at Roto made a chair especially for VR that aims to solve that. While the idea is sound, we believe that having to invest in yet another piece of equipment to enjoy VR is not the way to go.
All in all, these solutions can work, but there is no single “magic trick” to solving the locomotion issue for everyone. This single fact is fuel for VR skeptics.
The Fear of VR
I’ve seen it numerous times: when someone mentions trying a VR demo, the automatic question is “but will I be sick?” It’s part fear of the unknown, part past experiences that were not exactly “perfect”. No one wants to say “maybe you’ll be sick testing my game”. It became imperative, at all costs, to make sure every VR experience was free of simulation sickness.
However, fear of VR is not limited only to the consumer fearing the ill effects of the technology on their health; it’s also the fear of the industry giants bringing this to market that bad content could kill the appeal. As VR development grew stronger over the last year and with the consumer version finally nearing completion, it became abundantly clear that keeping motion sickness at bay was primordial. Palmer Luckey, Oculus’ Founder, as well as John Carmack both stated it very eloquently:
“Virtual reality’s biggest enemy is bad virtual reality”. –Palmer Luckey
“The fear is if a really bad V.R. product comes out, it could send the industry back to the ’90s,” –John Carmack
But this leaves us with a question that begs to be answered…
What Is BAD Virtual Reality?
All that work leaves us with a pretty bleak picture of what we can do in virtual reality with the premise that if we want a “good” VR product, we most absolutely have zero simulation sickness, for EVERYONE. All of the above solutions can work for one person and not the other. Or it can help but not completely eliminate the effect.
Here is a small list of stuff you will not be able to experience in VR if you are very very affected by motion sickness in general.
- All experiences with movement in them.
Yikes. No rollercoasters for you!
There is a great read by Lee Perry on Gamasutra that you should really check out that go deep into the kind of thinking I believe VR content creators should adhere to. If there is one thing I would like to emphasize in this article, it’s this: “People have extremely individualized issues with VR”.
So the definition of “bad” virtual reality is going to differ from person to person, as tolerance to some movements can be handled or not. The point of this entire post is to make it clear that maybe some VR experiences will not be for everyone. While Oculus’ reluctance to push out demos that might alienate a segment of players is understandable, what would be worse is us limiting the type of experience people can enjoy. For example, while it’s true that yaw rotation makes a lot of people want to barf, for those that it does not affect (or to a lesser degree), it’s a great and intuitive way to move around.
We don’t want to limit the kind of VR content we create to encompass every demographic; it’s an impossible task that would have us throw out things we hold dear, experiences we really believe are powerful even if a smaller percentage of people can experience it comfortably. What we can do, however, is have the maximum number of options and tweaks to permit players to mold their experience to their taste and sensibility. We believe VR needs to come with its own “rating” system that will have “comfort” as its unit of measure, the same way you know if you can handle rollercoasters in an Amusement Park or if you should rather stick with the Big Wheel. We just need to be honest with our players and expose our experience as it is. Oculus already started doing this on Oculus Share with its comfort level scale, even if they seem to have shied away from the concept since then.
Well, I knew this post was going to be long but I did not quite think it would get this long! While it’s pretty much about exposing the problem, there is a second part to this that focuses way more on how we did or plan to do to resolve the simulation sickness problems in FATED. After all, we still plan to have this next to our game when it ships:
In the meantime, if you have any questions or comment on the subject, please do share! I’ll be happy to read and discuss them with you!
Mick / Lead Programmer / @Mickd777