FLINT3.7.0-dev
/

DFLOAT_EXP_EPS_1, DFLOAT_EXP_EPS_2, DFLOAT_EXP_EPS_3, DFLOAT_EXP_EPS_4, DFLOAT_EXP_EPS

Synopsis#include <flint/dfloat.h>
DFLOAT_EXP_EPS_1source
DFLOAT_EXP_EPS_2source
DFLOAT_EXP_EPS_3source
DFLOAT_EXP_EPS_4source
DFLOAT_EXP_EPS(n)source

Relative error bounds of the exponential kernels, about 2−51.02^{-51.0}, 2−100.22^{-100.2}, 2−151.02^{-151.0} and 2−201.62^{-201.6}: for −746≤x0≤709.9-746 \le x_0 \le 709.9, dN_exp returns exp⁡(x)(1+δ)\exp(x)(1 + \delta) with ∣δ∣≤|\delta| \le DFLOAT_EXP_EPS_N, up to an additional absolute error of 2−10702^{-1070} when the result is subnormal.

Cite this page

The FLINT team, “DFLOAT_EXP_EPS_1, DFLOAT_EXP_EPS_2, DFLOAT_EXP_EPS_3, DFLOAT_EXP_EPS_4, DFLOAT_EXP_EPS (dfloat.h),” FLINT: Fast Library for Number Theory, documentation, version 3.7.0-dev, 2026. https://flintlib.org/doc/dfloat/DFLOAT_EXP_EPS_1.html (generated 2026-11-02).