Dear Jibiao,
One explanation could be that the thresholds you chose are not tight enough.
You could check whether the problem is alleviated if you tighten these.
For the scf calculation, set conv_thr in electrons to something much
tighter than the default, I would suggest 1.0D-10.
Also your threshold for tr2_ph in inputph can be tighter. If the
discrepancy decreases, than the frequencies are just imprecise due to
bad convergence.
Best,
Leonid Kahle
On 10/9/19 3:48 PM, Jibiao Li wrote:
Dear Prof. Paolo Giannozzi,
The two computers have the exact same OS (ubuntu), compilers (ifort),
libraries, and parallel environments installed. It really strange why
different results produced by the same inputs. Can anyone help me?
Best
------------------
/*Dr. Jibiao Li, */
/*Department of Material Science and Engineering*/
/*Yangtze Normal University*/
/*Juxian Dadao 16#, Fuling, Chongqing, China*/
/*Email: [email protected], [email protected],
[email protected]*/
/*Homepage: https://www.researchgate.net/profile/Jibiao_Li*/
------------------ Original ------------------
*From:* "p.giannozzi"<[email protected]>;
*Date:* Wed, Oct 9, 2019 09:42 PM
*To:* "Quantum ESPRESSO users Forum"<[email protected]>;
*Subject:* Re: [QE-users] Computer dependent phonon frequencies
Hard to say with no knowledge (or even with the knowledge) of which
computers, compilers, libraries, parallel execution, ... produce
strange results.
Paolo
On Wed, Oct 9, 2019 at 2:19 PM Jibiao Li <[email protected]
<mailto:[email protected]>> wrote:
Dear QE experts,
I am studying mixed H2O/OH phase on Ag(111) using QE 6.4.1. To my
suprise, however, to different computers gave totally different
vibrational frequencies, with the exact same input files. Is there
anybody to help me?
Computer A gave
freq ( 46 - 46) = 24.2 [cm-1] --> A I+R
freq ( 47 - 47) = 72.8 [cm-1] --> A I+R
freq ( 48 - 48) = 137.6 [cm-1] --> A I+R
freq ( 49 - 49) = 199.5 [cm-1] --> A I+R
freq ( 50 - 50) = 308.7 [cm-1] --> A I+R
freq ( 51 - 51) = 379.5 [cm-1] --> A I+R
freq ( 52 - 52) = 585.2 [cm-1] --> A I+R
freq ( 53 - 53) = 800.0 [cm-1] --> A I+R
freq ( 54 - 54) = 872.2 [cm-1] --> A I+R
freq ( 55 - 55) = 901.9 [cm-1] --> A I+R
freq ( 56 - 56) = 998.4 [cm-1] --> A I+R
freq ( 57 - 57) = 1529.7 [cm-1] --> A I+R
freq ( 58 - 58) = 2914.9 [cm-1] --> A I+R
freq ( 59 - 59) = 3064.3 [cm-1] --> A I+R
freq ( 60 - 60) = 3610.2 [cm-1] --> A I+R
Computer B gave
freq ( 46 - 46) = 33.0 [cm-1] --> A I+R
freq ( 47 - 47) = 45.1 [cm-1] --> A I+R
freq ( 48 - 48) = 54.1 [cm-1] --> A I+R
freq ( 49 - 49) = 78.2 [cm-1] --> A I+R
freq ( 50 - 50) = 82.4 [cm-1] --> A I+R
freq ( 51 - 51) = 99.1 [cm-1] --> A I+R
freq ( 52 - 52) = 108.9 [cm-1] --> A I+R
freq ( 53 - 53) = 114.4 [cm-1] --> A I+R
freq ( 54 - 54) = 188.3 [cm-1] --> A I+R
freq ( 55 - 55) = 590.6 [cm-1] --> A I+R
freq ( 56 - 56) = 1101.8 [cm-1] --> A I+R
freq ( 57 - 57) = 1533.4 [cm-1] --> A I+R
freq ( 58 - 58) = 2812.1 [cm-1] --> A I+R
freq ( 59 - 59) = 3644.5 [cm-1] --> A I+R
freq ( 60 - 60) = 3786.8 [cm-1] --> A I+R
*Why the two computers give totally different results? *
*
*
&CONTROL
calculation = 'scf' ,
restart_mode = 'from_scratch' ,
outdir = './' ,
pseudo_dir = '/home/jibiaoli/pseudo/' ,
prefix = 'sym' ,
tstress = .true. ,
tprnfor = .true. ,
/
&SYSTEM
ibrav = 0,
celldm(1) = 9.456004324,
nat = 20,
ntyp = 3,
ecutwfc = 49 ,
ecutrho = 411 ,
occupations = 'smearing' ,
degauss = 0.02 ,
smearing = 'methfessel-paxton' ,
vdw_corr = 'DFT-D',
/
&ELECTRONS
mixing_beta = 0.3 ,
diagonalization = 'david' ,
/
CELL_PARAMETERS alat
0.866025404 0.500000000 0.000000000
0.866025404 -0.500000000 0.000000000
0.000000000 0.000000000 5.400000000
ATOMIC_SPECIES
Ag 107.8682 Ag.pbe-n-kjpaw_psl.1.0.0.UPF
O 15.99940 O.pbe-n-kjpaw_psl.1.0.0.UPF
H 1.00800 H.pbe-kjpaw_psl.1.0.0.UPF
ATOMIC_POSITIONS angstrom
O 5.986597700 0.051902808 11.800640138
H 6.530086903 0.896034963 11.935550794
H 6.519207423 -0.788895855 11.940829013
O 2.800029893 0.013830142 11.748050047
H 3.756125095 0.022425916 11.956803063
Ag -0.454154628 0.136836718 9.568098196
Ag 5.307442766 0.123896111 9.577760992
Ag 2.445911493 0.129358587 9.587754292
Ag 2.920157088 1.499919274 7.139192063
Ag 1.485120252 -0.981971793 7.160426239
Ag 4.380244685 -0.987689704 7.134282475
Ag 2.889028971 -1.667975200 4.717759990 0 0 0
Ag 4.333537867 0.833997281 4.717759990 0 0 0
Ag 5.778055146 -1.667970358 4.717759990 0 0 0
Ag 0.000000000 0.000000000 2.358879995 0 0 0
Ag 5.778052352 0.000009682 2.358879995 0 0 0
Ag 2.889026176 0.000004841 2.358879995 0 0 0
Ag 2.889023381 1.667984881 0.000000000 0 0 0
Ag 1.444514485 -0.833987600 0.000000000 0 0 0
Ag 4.333540661 -0.833982759 0.000000000 0 0 0
K_POINTS automatic
6 6 1 0 0 0
phonons at Gamma
&inputph
tr2_ph=1.0d-14,
prefix='sym',
alpha_mix=0.1,
fildyn='phG.dyn',
amass(1)=107.8682,
amass(2)=15.999,
amass(3)=1.0079,
outdir='./'
nat_todo=5,
/
0.0 0.0 0.0
1 2 3 4 5
------------------
/*Dr. Jibiao Li, */
/*Department of Material Science and Engineering*/
/*Yangtze Normal University*/
/*Juxian Dadao 16#, Fuling, Chongqing, China*/
/*Email: [email protected] <mailto:[email protected]>,
[email protected] <mailto:[email protected]>,
[email protected] <mailto:[email protected]>*/
/*Homepage: https://www.researchgate.net/profile/Jibiao_Li*/
_______________________________________________
Quantum ESPRESSO is supported by MaX
(www.max-centre.eu/quantum-espresso
<http://www.max-centre.eu/quantum-espresso>)
users mailing list [email protected]
<mailto:[email protected]>
https://lists.quantum-espresso.org/mailman/listinfo/users
--
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list [email protected]
https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list [email protected]
https://lists.quantum-espresso.org/mailman/listinfo/users
