I mentioned in a previous post that doing exact UNORM or SNORM conversions to float in hardware was not particularly expensive, but didn't go into detail how. Let's rectify that! (If you haven't read that post yet, please start there if you need an explanation of what the UNORM and SNORM formats are, as well…

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

The post explains the hardware method for converting UNORM and SNORM numerical formats to floating-point (float) values, specifically targeting the IEEE 754 binary32 format. It discusses the basic principles, special cases, and general strategy for conversion, including handling of special values and rounding steps. Techniques used include bit shifts, normalization, and rounding processes to ensure accurate results.

UNORM and SNORM to float, hardware edition