# Real space computation, testing suite, H atom # Basis is plane-waves. #Definition of the SCF procedure nstep 100 # Maximal number of SCF cycles toldfe 1.0d-8 # Tolerence on difference of energy iscf 2 # simple mixing on potential #SCF preconditioner iprcel 0 # use diemix diemac et consorts diemac 3.0 # value for molecules in empty boxes diemix 0.4 # mixing # Force no symetries nsym 1 symrel 1 0 0 0 1 0 0 0 1 #Definition of the unit cell acell 3*5 rprim 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.0 natom 1 nband 2 #Hartree solver icoulomb 1 # Use Poisson solver ndtset 5 nscforder1 8 # Order of scaling used by kernel nscforder2 14 nscforder3 16 nscforder4 20 nscforder5 24 optforces 0 # Don't compute the forces inside the SCF loop #Definition of the k-point grid ngkpt 1 1 1 nshiftk 1 shiftk 0. 0. 0. #Definition of the atom types ntypat 1 # There is only one type of atom znucl 1 # The keyword "znucl" refers to the atomic number #Definition of the atoms typat 1 # All possible atoms are type 1. xred 0.5 0.5 0.5 # chkprim 0 #Exchange-correlation functional ixc 1 # LDA Teter Pade parametrization #Definition of the planewave basis set ecut 50 # This cut-off is too low for real calculations # but for the test, it is OK. # Avoid print densities, wavefunctions and eigenvalues prteig 0 prtden 0 prtwf 0 pp_dirpath "$ABI_PSPDIR" pseudos "PseudosGTH_pwteter/01h.pspgth" #%% #%% [setup] #%% executable = abinit #%% need_cpp_vars = #%% [files] #%% files_to_test = #%% t20.out, tolnlines = 0, tolabs = 0.000e+00, tolrel = 0.000e+00 #%% [paral_info] #%% max_nprocs = 1 #%% [extra_info] #%% authors = D. Caliste #%% keywords = #%% description = #%% Simple test for real space computation (using a wavelet based poisson solver), #%% no forces, just an H atom in an empty box. The cut-off is unrealistic. This #%% test tests the following part of the code : #%% * psp spline generation for real space ; #%% * local part of potential from pseudo ; #%% * ion-ion interaction computation (instead of Ewald) ; #%% * Hartree potential with Poisson's solver. #%%