Isn't that also the usual argument against subnormal numbers? Anyway, you're right. According to that paper, section 4.3: https://people.eecs.berkeley.edu/~dem

Isn't that also the usual argument against subnormal numbers? Anyway, you're right. According to that paper, section 4.3: https://people.eecs.berkeley.edu/~demmel/ma221_Fall20/Dinechin_etal_2019.pdf https://people.eecs.berkeley.edu/~demmel/ma221_Fall20/Dinech... the relative error is bounded by 2^{-24} only in the interval [1.0e-6,1.0e6] for the Posit32 format, whereas it is [1.17e-38,3.4e38] for the IEEE binary32 format.