like realistically if you store the environment as a sparse voxel octree and use some Fancy Math to have dynamic step size in the raycasting it should be fairly fast
@chjara running your raycaster on the custom gpu architecture attached to your cpu architecture running your own os compiled with your own compiler in your own language
Gregory's Server
@chjara running your raycaster on the custom gpu architecture attached to your cpu architecture running your own os compiled with your own compiler in your own language