GPUs support UNORM formats that represent a number inside [0,1] as an 8-bit unsigned integer. In exact arithmetic, the conversion to a floating-point number is straightforward: take the integer and divide it by 255. 8-bit integers are for sure machine numbers (exactly represented) in float32 and so is 255, so if you're willing to do…

FGiesen's platform is a resource for graphics programmers and computer graphics enthusiasts, offering insights into rendering techniques, GPU programming, and real-time graphics technologies. Through articles, tutorials, and technical discussions, FGiesen offers insights into rendering algorithms, shading languages, and graphics hardware architectures. Readers can learn about graphics rendering pipelines, optimization strategies, and advanced rendering effects to create visually stunning and immersive graphics experiences.

Ryg blog

Converting an 8-bit UNORM format to a floating-point number in GPUs can be done by dividing the integer by 255, which ensures an exact result in float32. While D3D11.3 specifications consider precise conversion too expensive, practical experiments show that GPUs can handle it efficiently. For those seeking exact conversion without using doubles or divides, an alternative method involving geometric series and multiplications is proposed, achieving high precision with minimal computational cost.

Exact UNORM8 to float

Option 2: but I don’t want to use doubles either!