These functions multiply their first argument x by FLT_RADIX (on
    binary systems, it has a value of 2) to the power of n, which is:

    x * FLT_RADIX ** n

    The definition of FLT_RADIX can be obtained by including <float.h>.