Re: [Maxima-discuss] significant difference?

[Raymond T. <toy...@gm...>]

>>>>> "Schorsch" == Schorsch  <Dr.> writes:

    Schorsch> well, the difference is 6 ULPs:
    Schorsch> 0.2954883855586914
    Schorsch> 3F D2 E9 48 1E 14 A2 80

    Schorsch> 0.2954883855586907
    Schorsch> 3F D2 E9 48 1E 14 A2 74

    Schorsch> and that is not much for a numerical result from a combination of 
    Schorsch> transcendental functions.

    Schorsch> I do not know which algos are used by Maxima, but I do know that 
    Schorsch> Carlssons algorithms perform very well; usually, they are off by less 
    Schorsch> than 2 ULPs : https://arxiv.org/abs/math/9409227v1

Maxima uses Carlson's Rf function to compute the elliptic_k value.