Bumped nmesh to 1024 in m_filter.f90
Certain FilterCutoff values did not result in usable basis functions. This change should resolve most of the problems. This result will change all basis-functions using the FilterCutoff option and destroys any backward compatibility for such calculations. Since the filtering is only applied when FilterCutoff is specified the majority of calculations should be unaffected.
This commit is contained in:
parent
7a3e974f3a
commit
cb00aa4540
|
@ -88,7 +88,7 @@ PRIVATE ! Nothing is declared public beyond this point
|
|||
integer, parameter :: dp = selected_real_kind(14,100)
|
||||
|
||||
! Internal parameters for filter subroutine
|
||||
integer, parameter:: nmesh = 128 ! Number of radial integr. points
|
||||
integer, parameter:: nmesh = 1024 ! Number of radial integr. points
|
||||
integer, parameter:: minj = 10 ! Min. num. of Bessel functions
|
||||
real(dp),parameter:: njkr = 0.65_dp ! Num. Bess. funcs. / (kc*rc)
|
||||
real(dp),parameter:: emax = 1.e-2_dp ! Min. accepted eigval
|
||||
|
|
|
@ -1 +1 @@
|
|||
siesta-4.1--944
|
||||
siesta-4.1--945
|
||||
|
|
Loading…
Reference in New Issue