I’ve a peformce issue with my city-scene and beside some other improvements I’d like to make a LOD. And becouse I’m a simple man, it’ll be an verry easy one
By accident I knew, I can run to cameras at the same time (ending a dizzling double vision) My idea was to use 2 cameras with same settings, only they differ in the far clip. Then sort all my primitives into two layers ‘Big’ and ‘small’. The small once should be captured by the camera with the short horizon and the big by the other.
what I’ve done by accident, I cant on purpose … so:
- How do I run 2 cameras at the same time?
- Do I make things even worse, with this approach?
as allways, thank You verry much
So, that’s quite a different way of doing this, but I’d say even though rendering two cameras is doable, it will have a performance penalty.
For LOD you are much better keeping a list of your model entities and checking the distance from the active camera. If the distance is bigger than a threshold swap to LOD1 otherwise back to LOD0 etc.
If you have a very large number of objects then partition your entities to grids and check the distance of the objects belonging to the active grid cell in which your camera is located.
this helps me a lot. Thought of a grid myself, but detecting enter and leave, that didn’t work, but grid plus distance will work.