The ABINIT Users Mailing List ( CLOSED )

["H.L.Shih" <hlshih@gate.sinica.edu.tw>]

Dear Abinit experts,

I tried to calculate the Gamma point frequencies of metalic oxide 
material, and adopting two type of psedopotential from official webpage of 
abinit. However, two shots using different potential sets have different 
results. The one adopting FHI, TM type pseudopotentials is about 10k times 
larger than the result carried out from Troullier-Martins (pspnc) ones. 
Furthermore, the answer carried out from 2 nodes is different from the 
one from 4 nodes (4 node parallel job seems have larger phonon 
frequencies). It seems to me the result from Troullier-Martins 
pseudopotential is much reasonable than the others. 

I was wondering why might cause this error, any suggestion to profile the 
problem I've encountered. Thanks in advance.

BTW, the version I adopted is "4.3.3". Another differences for two 
calculations is, for Troullier-Martins, 32bit executable is compiled. And 
for FHI potential sets, I tried with 64bit binary of abinip running on AMD 
Opteron 2.0 GHz linux box.

FHI Gamma point frequencies:
-----------------------------
 Phonon frequencies in cm-1    :
- -6.596885E+07 -6.159965E+07 -4.911165E+07 -4.656295E+07 -4.114280E+07
- -3.651928E+07 -2.650503E+07 -2.216013E+07 -1.803439E+07 -1.581312E+07
- -1.395147E+07 -1.231242E+07 -1.039826E+07 -9.472266E+06 -6.698459E+06
- -6.262442E+06 -5.730117E+06 -4.044703E+06 -2.413126E+06  4.583825E+04
-  1.312479E+05  5.339116E+05  6.844839E+06  1.363061E+07  2.825319E+07
-  3.041465E+07  3.109165E+07  3.316058E+07  3.347192E+07  3.491152E+07
-  3.549429E+07  3.620444E+07  4.073205E+07  4.090598E+07  4.097474E+07
-  4.382139E+07  4.847548E+07  5.167990E+07  5.341588E+07  5.529741E+07
-  1.031570E+08  1.331766E+08


Troullier-Martins pseudopotentials:
-----------------------------------
 Phonon frequencies in cm-1    :
- -2.136235E+02 -2.136068E+02 -2.135971E+02 -2.109442E+02 -2.097088E+02
- -2.095180E+02 -1.840092E+02 -1.839582E+02 -1.204800E+02 -1.129393E+02
- -1.116381E+02  1.855254E+02  1.907553E+02  2.044435E+02  2.075331E+02
-  2.696133E+02  2.796680E+02  2.814017E+02  2.994630E+02  2.995635E+02
-  3.055038E+02  3.055691E+02  3.056160E+02  3.497053E+02  3.501484E+02
-  3.501864E+02  3.816764E+02  3.816988E+02  3.817299E+02  4.236426E+02
-  4.240534E+02  4.241525E+02  4.489436E+02  4.489927E+02  5.824469E+02
-  5.835664E+02  5.836782E+02  5.885160E+02  5.890513E+02  5.892054E+02
-  6.274730E+02  6.423617E+02



BR,
J 

-- 
---------------------------------
Jason Shih
Academia Sinica Computer Center
No.128, Sec. 2, Academia Rd., 
Nangang District, Taipei City 115,Taiwan (R.O.C.)
Tel: +886-2-27899960
Fax: +886-2-27899949
---------------------------------