Forum Discussion
Motion Sickness - Causes and Possible Solutions
I thought I'd start a thread for people who have experienced motion sickness with the Rift so we can identify the issues and come up with solutions and possibly a few guidelines to help developers create an experience that will have much less of a tendency to upset the player's senses. In my experience, it isn't one little thing that directly causes motion sickness with the Rift, but rather lots of small inconsistencies that pile up and end up with you feeling queasy.
Demos/Games with Level of Disorentation
Planet1 Oculus Demo (minimal to no disorientation - great head positioning and gradual momentum movement modeling)
First Law Oculus Demo (minimal disorientation due to very sharp and immediate movements)
Museum of the Microstar (moderate disorientation due to lower framerate/disorienting motion mechanics)
TF2 (moderate/high disorientation due to motion mechanics)
Heli Hell 2 Oculus Demo (high disorientation due to what I believe to be bad head positioning)
I'll continue to add to the list as I spend more time with the Rift. Thanks to everyone developing the great demos that are currently out!
Immediate Momentum Increase/Halt
Issue: When any movement key is pressed/released and movement immediately goes full speed or comes to a full stop. It makes me feel like my body is floating forward at a 45 degree angle and my legs are lagging behind the rest of my body when motion stops. Very apparent in the Tuscany demo as well as in TF2.
Solution: There definitely needs to be a build up and wind down when motion is initiated or halted modeled to imitate real life movement.
Body Orientation
Issue: This is one of the worst contributors in my opinion, especially if the movement isn't smooth. In the Tuscany demo I end up closing my eyes and re-orienting my body with the mouse to avoid looking at the jerky motion it creates while standing still.
Solution: Possibly some kind of timer after the player stops moving that automatically (and smoothly) re-orients the player model to that direction. I think most first person demos would benefit form adding in a simple wheelchair or cart model that the user can sit in and control like a vehicle (since most people are going to be sitting while playing). In my opinion this would be much easier to implement until the nuances of player movement are nailed down.
No Sound / Inconsistent Sound
Issue: If you're in the Tuscany demo and you're hearing the echoes consistent with your office or whatever real world space you're playing in, this is another contributor (albeit minor).
Solution: Any sound would be better than no sound I think. Stock looped audio taken from a building of comparable size would work well. If no sound engine is present, recommend the user put on headphones and play music. Just something to block out glaringly inconsistent spatial audio.
Head Orientation & Modeled In-game Player Bodies
Issue: Having your view zeroed in so that you're 30 degrees off from the orientation of your shoulders in game. Very weird.
Solution: A simple 'zero' orientation key should be standard if you're featuring an in-game player body.
Aircraft "OH SHIT" Movements / Crashing
Issue: Obviously one of the hazards of flying a helicopter or any aircraft, but when it happens in the Rift it can cause serious motion sickness (I nearly puked when I lost control of my helicopter in the Heli Hell demo).
Solution: Possible control scheme 'easy mode' to prevent players from being able to crash so easily. If you are developing an aircraft sim, an on-rails 'tourist' mode would be a cool option as most of the enjoyment is from looking out of the vehicle.
Lack of "Head Bobbing" & Movement
Issue: Head fixed in space with little or no movement when moving forward or performing any other type of movement is another factor that contributes to disorientation.
Solution: Head movements in the real world need to be studied very closely and implemented as close to the real thing as possible. Walking produces a slight up and down movement. Running produces a more exaggerated and noticeable jarring 'impact' on the vision. Side strafing is even more exaggerated. Etc...
Poor Frame rate
Issue: Anything under a certain threshold can be disorienting (I'm going to guess it's around 60 fps).
Solution: Frame rate should take priority.
Demos/Games with Level of Disorentation
Planet1 Oculus Demo (minimal to no disorientation - great head positioning and gradual momentum movement modeling)
First Law Oculus Demo (minimal disorientation due to very sharp and immediate movements)
Museum of the Microstar (moderate disorientation due to lower framerate/disorienting motion mechanics)
TF2 (moderate/high disorientation due to motion mechanics)
Heli Hell 2 Oculus Demo (high disorientation due to what I believe to be bad head positioning)
I'll continue to add to the list as I spend more time with the Rift. Thanks to everyone developing the great demos that are currently out!
11 Replies
- I would put "Lack of configurable options" as a sort of meta-issue. It seems there are large variations between people and one person's solution is another person's problem.
- Hi,
I tried to add some smoothness to the right stick control of the gamepad to prevent nausea.
It is now a little more natural and less confusing but I don't know if anything was improved regarding nausea.
How to install and try it:
- I modified the OculusWorldDemo found in the Oculus SDK 0.2.1, so you need this installed.
- I built the exe using Visual Studio 2012 which seems to require some direct3d dlls that I added to the package.
- Feel free to try it and give your impressions, you can download binaries for Windows + src here:
https://www.dropbox.com/s/u5dkfkohvd2zkz1/GamepadRotationSmoothing.zip
Keys:
[1]: smoothness on/off
[2] and [3]: adjust smoothness "lag time". - Why more threads? There are at least 2 big ones + a wiki entry here on this subject :)
"SiggiG" wrote:
Why more threads? There are at least 2 big ones + a wiki entry here on this subject :)
I did a brief search of the forum and didn't get any results when I made the thread, but I see the Wiki page now. Thank you for the heads up :)"320x200" wrote:
I would put "Lack of configurable options" as a sort of meta-issue. It seems there are large variations between people and one person's solution is another person's problem.
I second this. I think for now the #1 thing people really need to realize is that there is a large variation in people and lists of dos-and-don'ts should not be compiled from personal experience or be considered authoritative for now.
Some honest to goodness science is going to have to be done. Makes me kind of excited :)- A wiki is good for synthetizing knowledge.
But maybe a forum is better to share some experiences like mine a few posts up. - No no no, wiki is good for collating established knowledge or on the other hand just collecting a lot of anecdotes. There won't be any real knowledge until people do the actual science, until then it will just be some kind of "common knowledge" or "conventional wisdom" that is actually unsubstantiated.
Not saying it won't be really useful as a stop gap in lieu of the actual research, but I shudder at the idea that a wiki, of all things, is a synthesis of knowledge when the primary research hasn't been done yet. Wikipedia can only begin to be trust worthy because it is supposed to be a collection of already established facts.
But who am I kidding. People out in the real world seem to have very different ideas of what counts as knowledge than I have in academia ;) - To my mind, the wiki is a way to store lists and information. The forum is a place to discuss information.
I absolutely agree that the science needs to be done. When it is done, the results can be stored or referenced to on the wiki, but the wiki is just a repository of information. Currently it is a collection of hypotheses, anecdotes, and a small amount of scientific papers. As hypotheses get upgraded to theories, these can be differentiated (with reference to the evidence/experiments) on the wiki, with the ultimate aim of a collection of 'laws of VR'. - I should probably calm down a little and remember I'm among friends :)
My main concern is that we do not create a set of "must do" and "must not do" items too quickly. I am beginning to think that for now VR is going to have to be very configurable. Maybe it will always have to be very configurable. I'm afraid people might put a big stamp on their games that says "Works for me!" and dismiss anybody who gets sick as a wussy.
On the other hand, I think if we collected all the things that will make people sick, built constraints into a game to prevent anybody from getting sick, I wonder what the odds are that we'd end up with something incredibly boring :)
One interesting person is Jamie Hyneman, he claimed to never get motion sickness at all. Would be interesting to figure out how many people have this trait and what kind of experiences in VR they would enjoy since they can go nuts and feel OK. Would these people feel like motion constrained VR is too tame? (If we take the constraints off of these people in multiplayer games is that considered cheating?)
Should VR games be more forgiving and let you skip parts that are too uncomfortable? Just to make an example, lets say I have a level that has a lot of swinging rope bridges and 90% of my player testers love them but 10% puke their guts out. Should I kill what is fun for 90% or should I let 10% opt out?
Enough rambling... - Sure the wiki is a good collaboration tool.
But I prefer avoiding to insert noise in it if the information has not reached a minimal quality through explicit "peer review".
On the matter of is it ok to have only 100% puke free games, I would also agree to let people experience stronger stuff. Classic games also do that by discriminating blind people.
But my fear is more that what will happen if VR never reaches the 100% nausea free games area.
