Page 3 of 3
Re: RIM and cutoff
Posted: Mon Sep 12, 2011 10:39 am
by Daniele Varsano
Dear Shudong Wang,
I ran your input files, and I was not able to reproduce you error.
Could you please post the report files of the setup and the rim_cut calculations?
Daniele
Re: RIM and cutoff
Posted: Mon Sep 12, 2011 2:47 pm
by sdwang
Daniele Varsano wrote:Dear Shudong Wang,
I ran your input files, and I was not able to reproduce you error.
Could you please post the report files of the setup and the rim_cut calculations?
Daniele
I upload the r_setup and r_rim_cut in the attachment.This is used v-3.2.1.and the 3.2.4 also have the same problem.
Thanks！
Re: RIM and cutoff
Posted: Mon Sep 12, 2011 5:19 pm
by Daniele Varsano
Dear Shudong Wang,
the report you post it has not the NAN problem, and it is different from the input file looks different
from the the one you posted in the previous message.
Anyway, as I can see from the reported input at the end of the report file:
Code: Select all
.-Input file : yambo.in
| xxvxc # [R XX] Hartree-Fock Self-energy and Vxc
| rim_cut # [R RIM CUT] Coulomb interaction
| FFTGvecs= 40003 RL # [FFT] Plane-waves
| RandQpts= 1000000 # [RIM] Number of random q-points in the BZ
| RandGvec= 11 RL # [RIM] Coulomb interaction RS components
| CUTGeo= "cylinder z" # [CUT] Coulomb Cutoff geometry: box/cylinder/sphere
| % CUTBox
| 0.000 | 0.000 | 0.000 | # [CUT] [au] Box sides
| %
| CUTRadius= 0.000000 # [CUT] [au] Sphere/Cylinder radius
| CUTCylLen= 0.000000 # [CUT] [au] Cylinder length
| EXXRLvcs= 85003 RL # [XX] Exchange RL components
you are setting a 0 radius, and the calculation of the cutoff potential is skipped.
I suggest you to check your input files (for instance the EXXRLvcs and FFTGvecs should consistent) and eventually run first the cutoff construction alone, look if everything is correct (for instance activating the CUTCol_test flag) and next a Sigma_x calculation.
Cheers,
Daniele
check the
Re: RIM and cutoff
Posted: Tue Sep 13, 2011 1:58 am
by sdwang
I am very sorry I post the wrong report in my previous post.And in the cut_test report , V_Rlat is also NaN.
I reupload them in attachment.
Thanks!
Re: RIM and cutoff
Posted: Tue Sep 13, 2011 3:36 pm
by Daniele Varsano
Dear Shudong Wang,
I could not reproduce your error! I ran your input files and everything went smooth.
I suggest you to repeat carefully your calculations using the new (svn) version of the code,
including also the new p2y interface.
In this way you will also benefit from the inclusion of the non local part of the pseudopotentials
in the calculations of the absorption spectra. Moreover let me suggest you to try to lint the
netcdf library in yambo as they will improve the IO, and you can also look inside of databases, which could
be useful for individuating problems.
Cheers,
Daniele