Question on subnormal descriptions in "Section 3.3 Subnormals" of Version 3.0: 2025-07-29 #12
Description
IEEE Std 754 value sets include subnormals. A value with trailing significand field T and exponent field E is interpreted as 1.T × 2E−bias except when all bits of the exponent bitfield E are zero, in which case the value is 0.T × 2−bias.
Shouldn't the subnormal value actually need to be 0.T × 2(1−bias) to ensure continuity between normals and subnormals ?