Im having a problem where im procedurally generating a asteroid field and some asteroids are rendering whilst others are not.
The only difference between each asteroid is the size and rotation, the shader settings are left unset. Im using monobehaviour to instantiate from a base entity, it uses a master sphere collider to define where the asteroids need to be inside, and then various box and sphere colliders to identify where no asteroids should be placed. The system works fairly slickly to generate a random/repeatable asteroid field, however at a point designated approximately -35,000 to -40,000 in X axis anything past this point is not rendered. The entity exists as per the other ones but nothing is shown. I’ve noticed if i move my asteroid belt instantiation point anywhere else where it does not go inside the 40000 it renders fine.
The above is just to signify that my procedural code for the belt has worked and created a entity at the location highlighted, unfortunately it is just not visible, I’ve used this script for the past 6 months at least in entity format without issue until i noticed this cropping issue.
The negative x rendering cut off seems to occur at approximately ~(-)33500, which is very close to 2^25 (not sure if this is significant?). I moved it to a similar position but in -33500 in z and y but that appears to be fine and it renders with no issue.
Its Hybrid Renderer V2 and the asteroids are rendered using a GameObjectConversionUtility gameobject including LOD system which for most cases works all the time! Judging by how the border is very much a square bounds it looks very much like something hardcoded as suggested in that post. If thats the case it would be good if they could define the upper/lower bounds of the rendering.
edit: In fact im sure thats case now, i just gave the asteroids a 100m/s motion in positive x and as the asteroids move beyond -335xx they are progressively rendered. I might experiment to find where the culling limits are in mean time and adjust my game accordingly.
Thanks both, I updated to preview 43 but as listed in the link the value “p.Distances[i & 3] = 32786.0f” still exists in the latest version (it appears that part of FrustrumPlanes.cs is unchanged or the change didnt update), i adjusted the value to something larger and that fixes the issue.
