ERI_MME

Parameters for the calculation of periodic electron repulsion integrals (ERI) using the Minimax-Ewald (MME) method. Note: N_MINIMAX is the only parameter to be tuned for accuracy, all other parameters can be left to default. MME method is faster than numerical GPW. [Edit on GitHub]

Subsections

• CUTOFF_CALIB
• ERI_MME_INFO

Keywords

• CUTOFF

• DEBUG

• DEBUG_NSUM_MAX

• DEBUG_TOLERANCE

• DO_CALIBRATE_CUTOFF

• DO_ERROR_ESTIMATE

• N_MINIMAX

• PRINT_CALIB

• SUM_PRECISION

Keyword descriptions

CUTOFF: real = 3.00000000E+002

User-defined energy cutoff to be used only if DO_CALIBRATE_CUTOFF is set to .FALSE. [Edit on GitHub]

DEBUG: logical = F

Lone keyword: T

debug mode (consistency of summation methods is checked). [Edit on GitHub]

DEBUG_NSUM_MAX: integer = 1000000

restrict debug mode for non-ortho cells to this number of summands. Sums with more terms are not checked. [Edit on GitHub]

DEBUG_TOLERANCE: real = 1.00000000E-006

tolerance for rel. numerical error in debug mode. [Edit on GitHub]

DO_CALIBRATE_CUTOFF: logical = T

Lone keyword: T

Whether the energy cutoff shall be calibrated to minimize upper bound error estimate. [Edit on GitHub]

DO_ERROR_ESTIMATE: logical = T

Lone keyword: T

Whether the error due to minimax approx. and cutoff shall be estimated [Edit on GitHub]

N_MINIMAX: integer = 53

Number of terms in minimax approximation of reciprocal space potential. [Edit on GitHub]

PRINT_CALIB: logical = F

Lone keyword: T

Print detailed info on calibration. [Edit on GitHub]

SUM_PRECISION: real = 1.00000000E-016

Terms in lattice sums are ignored if absolute value smaller than this value. [Edit on GitHub]