Philipp Spaniol (spanni)

1 reply · posted

Unity Eyetracking and colliders

hello community :)I'm working as a student on a unity VR project with eyetracking. Right now I'm setting up colliders for buildings and I'm wondering, what collider to use. For more accurate data i'd like to use mesh collider, since they are way closer to the shape of the building then some box colliders. Since the buildings are static and i haven't found any good article about that, I'd like to ask if someone can explain me, if there would be a hard performance crash when i create mesh colliders for static objects. If there is just a ray hitting the surface, i hope that the issue wouldn't be that hard if I use mesh colliders, but i really don't know. any opinions on that?

  • I guess it would depend on how many mesh colliders you are needing to compute. If it is just a few objects I would say it wouldn't be a huge performance hit but having every object as a mesh collider is going to make the physics work overtime since the geometry is quite complex and it is after-all trying to calculate what would happen in a collision from different angles.

    Does the VR struggle more with the mesh object calcs?

    I'd recommend maybe cutting your objects up into more basic shapes too if you are able and see what that does on performance. If they are static and never going to move (just passive collision objects) then no one will ever know the difference if your house is made of 4 objects versus one. You'd probably get a little more control on your colliders that way, right?

    I guess I wouldn't go crazy with that because then you just sacrificed complex collision calcs with more small collision calcs.