I have an OpenGL ES 3 Vertex Shader and Fragment Shader that renders particles to the screen.
I can draw a fixed number of particle sprites at an acceptable frame rate if the sprite edge sizes are fairly consistent, but if I increase the edge size considerably (e.g increase the number of pixels rendered per frame), the frame rate will dramatically drop if I attempt to render the same number of particles per frame. The only way to maintain the same frame rate with larger edge sized sprites is to draw less of them per frame.
I thought that if I added up every sprite edge dimension for a given batch, that I could use that value to determine if no more particles should be drawn for a given frame. But it appears that even with a smaller edge size batch total the frame rate can be much lower when using large edge sized particles.
Is there anyway to determine optimal pixel bandwidth to be rendered per frame?
