These features are obtainable for all of our floating-point formats.

These predicates: is_signed, is_unsigned, is_finite, is_extended.

These characterizing counts:

bitwidth, precision,
nValues, nNumericValues, nFiniteValues, nNormalValues, nSubnormalValues,
nMagnitudes, nFiniteMagnitudes nNormalMagnitudes, nSubnormalMagnitudes,

nBits, nSignBits, nExpBits, nSigBits, nFracBits

These extremal values:

floatmin, floatmax, exponent_min, exponent_max
subnormal_min, subnormal_max, normal_min, normal_max