OpenGL Programming Guide: The Official Guide to Learning OpenGL, Version 4.3, Eighth Edition

10- and 11-bit Unsigned Floating-Point Values

For normalized color values in the range [0, 1], unsigned 10- and 11-bit floating-point formats may provide a more compact format with better dynamic range than either floating-point values, or OpenGL’s unsigned integer pixel formats. The maximum representable values are 65204 and 64512, respectively.

Routines for converting floating-point values into 10-bit unsigned floating-point values, and vice versa, are shown below.

Click here to view code image

#define UF11_EXPONENT_BIAS  15#define UF11_EXPONENT_BITS  0x1F#define UF11_EXPONENT_SHIFT 6#define UF11_MANTISSA_BITS  0x3F#define UF11_MANTISSA_SHIFT (23 - UF11_EXPONENT_SHIFT)#define UF11_MAX_EXPONENT \  (UF11_EXPONENT_BITS << UF11_EXPONENT_SHIFT) ...