Phonons with vasp

Message

saikat · #1 Post by **saikat** » Fri Mar 29, 2013 7:30 pm

Hi,

I am trying to run phonon calculations with 32 nodes and 4 ppn (128 prcs). If I use the NPAR tag it complains and terminates saying remove the NPAR tag. Because, the Kpoints was not suitable and it needs to be readjusted. If I do that, then it stops after iteration 1, saying bad termination of one of your procs. This is kind of large system so I need to use as many procs as possible. I am using vasp 5.3 and IRION =8. I am using 2x2x2 supercell.
Any help would be greatly acknowledged.

Thanks a lot.

hatdau · #2 Post by **hatdau** » Sun Mar 31, 2013 12:16 am

Please upload input files.

saikat · #3 Post by **saikat** » Sun Mar 31, 2013 6:26 pm

I am using the following INCAR ...
-------
PREC = Accurate
ENCUT = 420
IBRION = 8
EDIFF = 1.0e-08
IALGO = 38
ISMEAR = 0; SIGMA = 0.1
LREAL = Auto
ADDGRID = .TRUE.
LWAVE = .FALSE.
LCHARG = .FALSE.
LEPSILON=.TRUE.
! LPLANE = .FALSE.
! NPAR = 8
! NSIM = 1

I tried both using the last three commented lines and also without them .. but no success ...I tried NSIM=4 as mentioned in other comments from the vasp users.

saikat · #4 Post by **saikat** » Tue Apr 02, 2013 4:15 pm

Hello,

Any help on the issue with phonon calculations? I am keen to learn more about the solution. Is there a bus in the new version which may cause this?
https://www.vasp.at/index.php/news/bugf ... onse-nmr-f

Help of any kind would be greatly appreciated.

Thanks.