Ive currently wrote an infinite terrain viewer, with cpu based editing. As im sure you will know, i am finding dynamic editing a touch slow with cpu based editing.
I’d like to abuse a seperate camera/render that is basically just the texture fed into a post process shader to gpu edit it.
All this is working to a fashion.
However, I would like to set the gpu hacked scene/camera to render the image in a higher float point precision without effecting my normal scenes details.
This is where im struggling as no matter what texture settings i set it to it still seems to render as the default lower precision.
Im guessing this part cant be done using Texture::Frgb32/16 etc of the render target texture.
Anyone know where i should be setting this up for it to work correct?