I had this problem but this has now been resolved with the latest dev version.
May 10 2016
This issue has not occurred to me in a while.
I'm guessing you've fixed it by accident.
If it reoccurs some time later, I will attempt to film it.
VR is actually used to treat PTSD.
There's too many links to list so here:
Should be mentioned that Dslyecxi did a first impressions video on the rift.
His main complaint was the resolution, which is a technology improving constantly on its own.
I wouldn't be surprised if the consumer version was over fullhd resolution.
The advantage argument isn't valid due to the low resolution of the HMD device.
It's only 640x800 per eye, only maybe 1/3 of which is primary player vision area.
Most is peripheral vision and the edges are completely unseen.
Resolution will obviously improve with future versions but Arma games have always been games that require high res view of the battlefield if you want to stay alive, something which HMDs will not provide in many years.
Clearly keyboard and mouse is a non-optimal device for any VR gaming, and controls should be bound to other handheld controller device.
Razer Hydra is a clear leader on this field at the moment.
This would need engine code to implement the tracking, but I would be happy just playing with KB&mouse for now, or hydra in mouse emulation mode.
But the discussion is seriously getting off track.
There is very little that needs to be done to make this work on the engine side.
The biggest issue is stereo rendering.
Does A3 work with any existing 3D set-up? Nvidia stereo driver or other methods?
We already have head tracking in place, and view distortion can be inserted externally.
The open-source Vireio driver already renders many games usable for Oculus Rift through a .dll hijack.
Apparently they are working on Arma engine support but currently it just crashes A2 and A3 does not react.
Naturally native support would be preferable.