Provide reliable, accurate down to several ULP's and fast numerical framework to compute generalized Fermi-Dirac integrals aiming at full floating-point parameter coverage, generality, precision control and speed.
MIT License
0
stars
0
forks
source link
Large errors of Ffermi_derivatives_m_n_quad for k>>16 #6
Precision for large eta=16, e.g. eta=64, and eta<0 or tiny eta around 0.0 reduced or completely wrong result. Error is from dbl_exp algorithm, once switched to Sommerfeld series problem vanish.
Precision for large eta=16, e.g. eta=64, and eta<0 or tiny eta around 0.0 reduced or completely wrong result. Error is from dbl_exp algorithm, once switched to Sommerfeld series problem vanish.