Namely Height, U and V gradients and index . Maybe in some of 32 bits by some kind of an order mask for those 32 binary bits numbers the pixel value should be build from theoretically or maybe by how many numbers after comma a pixel value has. I want to do an FX map in 32bit mode the way I could split the resulting floating point image into those height, UV and index? And do this the way I still could do more or less MAx (lighter) blending?
Is there a way at all or it's totally silly idea?
AdChoices