GearVR Instanced stereo rendering

I was looking through the forum post and there was some discussion about using multi-view to get a similar effect on the GearVR to the instanced stereo rendering that was introduced on PC at 4.11. I’ve been wondering when we can expect to see this functionality or if it would be possible to get a patch to get this functioning on our end. We are currently upgrading to 4.12 and should hopefully have that working soon. And to be clear, I’m assuming this functionality will work on all GearVR devices. Is that the case? Also, what kind of gains are you seeing with this on the render thread? Thanks!

Hey Brant,

Thanks for the question. We are working on a solution for instanced rendering for mobile platforms, but it’s still currently in development, so I can’t provide you with a changelist to integrate just yet unfortunately. According to the GearVR website, the device supports just six different devices, all Samsung Galaxy, so hopefully it won’t be a problem to make a solution that works for all of them provided the hardware supports the features required, but we’ll have to see once we decide which approach we’re going with for that change.

As far as what kind of gains we are seeing, I don’t know how directly comparable these numbers will be to mobile hardware and to the mobile forward rendering system, but in Bullet Train, instanced stereo rendering saved 1.75ms of CPU time and 0.75ms of GPU time for a 90fps+ Occulus game.

Thanks, please let us know if you have any other questions.