Oculus recently published the first installment of their ‘Developer Perspectives’ video playlists, detailed on the developer blog. Crytek’s VR movement research is presented by Julius Carter, Game Designer at the studio behind award-winning VR games 

Artificial VR locomotion—moving the player around the virtual world in ways that takes them beyond their available physical space—is an ongoing challenge for VR developers, due to the need to traverse virtual environments larger than the real play space, and its potential to cause disorientation and nausea if done incorrectly.

In order to figure out what would and wouldn’t work in the context of VR locomotion, Crytek has run a vast number of experiments, many of which informed the locomotion design of their first two VR titles, The Climb and Robinson: The Journey. Over the coming weeks, Crytek says they they will publish some 40 videos exploring those experiments and the thinking behind them. The videos will be added over time to this YouTube playlist which presently has seven already published:

Some conventional locomotion techniques, established over decades of screen-based game development, such as WASD, joystick inputs, and button sprinting can cause discomfort for some users in VR, whereas others, such as button jumping, don’t seem to be much of a problem. Alternative rotation methods like snap turning or ‘compass rotation’ have proven to be successful at reducing the chances of nausea, and variations of this are found in many VR titles as a ‘comfort mode’ option.

These methods, together with some problematic alternative tests, such as ‘scaling rotation’, where real head rotation is amplified in VR, are explained in the first set of videos. Carter also describes the importance of using an appropriate test environment for the type of experience you’re creating, and the challenges involved in testing so many types of locomotion. Much of this information will be familiar to VR developers, but it’s a useful starting point for those new to the medium.

Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.


Dominic Brennan
The trial version of Microsoft’s Monster Truck Madness probably had something to do with it. And certainly the original Super Mario Kart and Gran Turismo. A car nut from an early age, Dominic was always drawn to racing games above all other genres. Now a seasoned driving simulation enthusiast, and former editor of Sim Racer magazine, Dominic has followed virtual reality developments with keen interest, as cockpit-based simulation is a perfect match for the technology. Conditions could hardly be more ideal, a scientist once said. Writing about simulators lead him to Road to VR, whose broad coverage of the industry revealed the bigger picture and limitless potential of the medium. Passionate about technology and a lifelong PC gamer, Dominic suffers from the ‘tweak for days’ PC gaming condition, where he plays the same section over and over at every possible combination of visual settings to find the right balance between fidelity and performance. Based within The Fens of Lincolnshire (it’s very flat), Dominic can sometimes be found marvelling at the real world’s ‘draw distance’, wishing virtual technologies would catch up.
  • Lucidfeuer

    Scaling rotation is a difficult challenge, but…I’m surprised Crytek is only at this step. Obviously you have to ambiguate distance to scene object, FOV° “zonage” and linear 360° head acceleration, to get some sort of smooth scaling rotation matrice.

  • Very interesting resources for us developers!

  • When I first tried unrestricted motion in VR on my DK2 I got heavy nausea. Since then I keep following new modes of movement to avoid motion sickness but I am not keen on many of them.

    What I find interesting is that when in the real world, cycling, skateboarding, surfing, etc all have you roughly stationary while traveling and changing direction at speed, motion sickness does not happen. Is this due to being in “control” of the hardware so you can predict movement opposed to say a passenger in a car that has no control over it s and gets “Travel Sickness”? I do not know of any drivers that suffer travel sickness while driving themselves.

    So why not try methods like that in VR. A ‘Back to the Future’ hover board for instance. Maybe you need to be standing on a normal skateboard in the real world with gyro tracking on it but in VR it is a hover board that has freedom of movement. You tilt your feet left and your character turns left in the VR world. As you are making the movement yourself then you expect the motion in VR and nausea would be much reduced.

    Maybe a simple disc with a football inside it so that you can tilt at any 360 degree angle would allow fast smooth travel with full turning in VR. A kit to make such a disc would be DIY cheap too. As long as the VR world had the same disc in it then this should work and finally allow a system that most VR FPS users would crave.

    Thoughts?

    • kool

      Have you tried the aim, having the gun prop helps orient you I think.

      • No I haven’t, yeah it’s the same principle for the hands. I remember when Virtuix Omni were testing this concept out in their early days and it looked promising.

    • ender707

      • @ender707:disqus, I have seen that and think it is part way there. I imagine a more complete system where you are standing on a 3 foot diameter circular board which can tilt a full 360 but my theory is that you need the same representation of the board inside VR so your body can learn that a leg movement left in the real world creates the leg movement left in the VR world and you strafe left, if you are moving forwards and rotate slightly left you start to turn left and strafing is disabled. Your body can learn to accept this much more naturally than other input systems. For example real skiing is quite hard to learn because it restricts your natural movement, it takes a long time to master. We need to find an input system that is fast to master and cheap to build.

        To stop game developers from changing the speed or rotation of your input device in every game there should be a standard, maybe feet per second. Being consistent is a key aspect for this to succeed in VR. As you master full movement you can start to override and increase the motion speed in VR. Most games only have a walk or a run and blend very quickly between the two when using analog input. This system would need to make that much more accurate to allow greater control. Natural movement has many more levels of accelerate, sustain and decelerate than 2D/3D games offer.

        The problem with thumb-stick motion is that there is very little link to what your body expects to happen and when it goes wrong its awful. A bit like the feeling you get when putting a car in first gear and expecting to move forwards when you accelerate but you have put it in reverse by mistake. The feeling is horrible and disorientating. I guess if you brute force yourself long enough you would build up a resistance to it (ye ole sea legs) but that really hampers early adoption.

    • CazCore

      you’re on the right track, that it’s all about being in control of your view. i solved the VR FPS (game) sickness problem. :)

      • What did you do?

        • CazCore

          i should be uploading a video demonstrating it in a simple demo within a week or 2.

          • Please reply here when you do, looking forwards to it.

  • kool

    The aim has it down packed. You don’t have to move that much or stand. Farpoint has the widest range of locomotion modes to try. If you try them all before playing it eliminates any discomfort you may have.

    • polysix

      down pat.

      • kool

        It’s down packed where I’m from.