Does anyone knows why the VR camera doesn’t follow properly my Character ? As you can see on the screenshot, every components (we can see the capsule and the skeletal mesh) of the Character are placed properly in the world, but the camera just doesn’t care about it.
On the left, a listen server. In the middle, a spectator only client. On the right, my VR Mirrored window, where you can see that my VR camera is actually at the spectator pawn spot O_o.
It seems that switching collisions to “OverlapAll” on the CapsuleComponent help a bit, but the location is sometimes good and sometimes not. Here is my thought:
Since we have to set the PlayerStart point in the ground when using Vive, the engine probably doesn’t allow some components to be moved in the ground, and moves them in the next PlayerStart point it will find in the level.
If I keep only one PlayerStart it will works as expected as you can see on the screenshot.
Why does the engine starts to ■■■■■■■■■ when he has multiple PlayerStart in the level ?
I came across similar issues with my set-up using the HTC Vive. In my case, for my custom Pawn Class, I had to set:
CapsuleComponent Collision to “NoCollision”.
"Spawn Collision Handling Method" (in the class defaults) to “Always Spawn, Ignore Collisions”.
"Base eye Height" was set to 0.0.
After this, I was able to spawn 2 Pawns on the exact same spot (i.e. in the floor) in a network game.
Do you need your pawn (and it’s capsule collision component) to move around the scene? If so, you might have to create a new capsule component separate from the root one.
