it seems to me that there isn’t any way to prevent the shader to output to the render target and the camera depth render target. Is this correct? I would like to profile on low-end machines the difference without the camera depth render target as I don’t even think we are using it.
It seems that what I wrote is not true, so I am investigating further