cancel
Showing results for 
Search instead for 
Did you mean: 

Designing around motion sickness

SiggiG
Protege
Greetings fellow Rifters!

I'd be very interested to hear your experience with motion sickness with the Oculus Rift, and theories about what causes it and how we can reduce it when designing our games. Testing how different things work out can often be uncomfortable so I'd love to share our findings and hear from the rest of you.

We've mostly been doing a few simple demos and trying out the included Tuscany demo, as well as a game prototype we're developing where you sit in a cockpit.

From what I've seen the lower brightness level of a space environment as well as having a static reference point when you sit inside a vehicle vastly reduces the motion sickness many people have felt.
The cockpit seems to give the brain a static reference similar to how looking at the horizon can help people feeling less car- or seasick.
The lower brightness helping might simply be due to the brain sensing less movement around you since a lot of it is black, or reducing the ghosting on the LCD panel?

Of the scenes we've tested it's been the worst in the Tuscany demo, and based on my gut feeling (pun intended) I think that has something to do with the jarring movement of suddenly being propelled forwards without sensing any motion. This seems to be reduced if the player moves around slowly.

I have also tested some scenes which taxed the GPU and reduces the FPS down to 30-40 fps, and this caused even more motion sickness.

So from what I've seen:
* High FPS is a must (we knew that already)
* Having a static reference like a cockpit around you helps a lot.
* Slower movement speeds when walking around in 1st person, this favors exploration over "run and gun" games.
* "Easing in" the movement speed instead of going from 0 to 50 in 1 second might help.
* Lower brightness scenes seem to reduce motion sickness.
CCP Games, EVE: Valkyrie developer | @SiggiGG
59 REPLIES 59

Alkapwn
Protege
I just posted this in the thread mentioned above, but it seems as this thread might apply a bit better.

After using the Rift for a while and getting the hang of things in VR, I thought I would step the game up and see if I could cause myself to get motion sickness.Two things that I quickly noticed. Strafing and strafe walking were the biggest culprits.

Strafing left and right repeatedly had an odd feel to it, I think due to the fact that we don't do it fairly often in real life, especially at that speed.

Strafe walking, as best as I can describe it is when you're moving forward/backward while also strafing. This has a very un-natural feel to it and is accentuated when you start turning/rotating while doing this. Most of my co-workers seemed to be fine until they started doing weird combinations of strafing and rotating together which seemed to bring on motion sickness quite rapidly. I think brantlew has it completely correct with the rotations being a huge factor.

I don't know if there's an easy way to solve for this in software aside from possibly lowering the speed of lateral movement or something like that. The issue seems to be you're using something other than your body to facilitate turning, which may just be a limitation of seated gameplay. I think that something that would help this from a hardware standpoint would be an additional tracker. This tracker could then be placed on either the chest, back, or waist. Then the player movement would simply be forward, back, left and right. When the user wants to turn their character, they would rotate heir body to turn the character in game. This would be in a natural way that the player's body and mind would be used to. For some reason I found it easier to get nauseous while sitting down rather than standing up.

I hope this helps somehow.

SiggiG
Protege
Pretty cool and relevant video that shows the difference positional tracking can make:
http://www.youtube.com/watch?v=R0-dsbeasgA
CCP Games, EVE: Valkyrie developer | @SiggiGG

craigotron
Honored Guest
It occurred to me that amusement parks have had to develop very good strategies for just this very problem. When you consider that a big part of their business model is stuffing people with food that upsets stomachs, they must have gotten very good at it over the years. Does anyone know anyone who works in this industry? Ideal candidate of course would be a Disney Imagineer. I'd love to hear their thoughts on the matter.

It's also encouraging to think that people do go to amusement parks, make themselves nauseated, and then happily and excitedly do it again while paying good money for it all the while. Bodes well for VRs long-term prospects.

I had another thought, which is rooted in solid physics, but is still probably crackpot when applied to human biology. Would simulating a subtle red/blue shift when players move away from or toward objects make our brains acknowledge that, yes, that is how the laws of the universe behave and therefore act as a counterweight to the confusion coming from the inner ear?

So, those are my two thoughts. I'd like to hear yours.

SiggiG
Protege
"craigotron" wrote:

I had another thought, which is rooted in solid physics, but is still probably crackpot when applied to human biology. Would simulating a subtle red/blue shift when players move away from or toward objects make our brains acknowledge that, yes, that is how the laws of the universe behave and therefore act as a counterweight to the confusion coming from the inner ear?


Afaik humans don't move fast enough to see redshift so our brains probably don't care? 🙂
CCP Games, EVE: Valkyrie developer | @SiggiGG

craigotron
Honored Guest
Hence my crackpot disclaimer 😄

Since some developers are building things where FTL travel is possible, this might be a natural way to explore if this thought is pseudo-scientific hokum or possibly useful.

@Notch - my Rift headset is pointed at your avatar

kingtut
Honored Guest
I've just created https://developer.oculus.com/wiki/ind ... n_Sickness to summarise the discussions. Obviously feel free to edit.

Note that discussions etc should definitely take place on the forums, but I think the wiki is a better way to collect and collate the different threads which can and will pop up on this subject.

SSJKamui
Honored Guest
I also read about a suggestion of forced pauses and game session duration limits, which could prevent overusing VR tech. The problem is, Nintendo exactly tried that with the Virtual Boy and that was one reason for the commercial demise of the device. So, this might not be an appropriate solution.

t0pquark
Honored Guest
I was reading this thread before work this morning, and then right before lunch something hit me that might be able to explain some of the issues with strafing.

The whole concept of strafing in a video game is that it’s a shortcut to expressing something you otherwise couldn’t easily do WITHOUT a HMD, which is to look a different direction than you are moving. When you use this shortcut WITH a HMD, it confuses your body.

Consider what happens when you press the “A” or “D” key. You start running (not side stepping) to the left or right, even though your point of view hasn't changed. This was needed because “W” always moves you towards your view (which is also locked to your rectical). If you watch an in game character do this, they almost instantly snap their body sideways and begin to run, while their head hasn't moved at all. In the game, the head is the base point and body swivels.

But what is happening with your HMD in the real world?

You could very well be sitting with your head still completely square to your shoulders, which is absolutely not what body thinks SHOULD be happening in the real world. In order to “circle strafe” in real life you would be moving parallel to your waist/shoulders and would have to be tracking your target over your shoulder with your head. The body is the base point and the head swivels.

I don’t have my dev kit yet, but something I would propose testing is to remove the strafe keys altogether and change “A” and “D” to rotate your hips/torso. When you are walking or running, your hips point the direction you are moving. You’ll need to relearn the technique a little bit, but it’s still fully possible (many "giant robot" games work this way already) and much closer to how the body actually works.

A follow up part of this would be that if your hips achieve a certain delta of rotation from your head (and/or shoulders, which is controlled by the mouse/right-stick), they would need to start dragging your view with them, again, reflecting what happens in the real world.

brantlew
Adventurer
@t0pquark: You are right about strafing being unrealistic in the real world. Your solution seems reasonable, but how does it fundamentally change the effect? If I understand you correctly you are changing the control mechanism of strafing from the act of moving your finger from the W to the D key to the act of pressing the W key + pressing the D key for a short time and releasing it. But your head and torso are still in the same relative positions in the real world and you are still receiving the same dissonant inertial and visual signals. It does have the advantage that the transition is continuous instead of instantaneous, but not sure if that will improve the effect or not. I guess it's worth a shot.

LaurentMT
Honored Guest
I can remember my first experiences with 3D games in the 90's (Wolfenstein 3D) and it was just awfull :lol:
After just 10 or 20mn, you were sure to be sick (queasiness, headaches, ...).
But the industry has progressed very quickly in this domain.

I guess it should be a bit more complex with VR.
The funny thing is that motion sickness seems related to the detection of hallucinations and that our main goal with VR is just about that : trying to provide a hallucination as credible as possible.
Avoiding motion sickness seems a interesting challenge !

But I'm quite sure that at least, we'll be able to define some how-to adapted to the current hardware, which combined with regular practice, will help to avoid motion sickness... as much as possible.