[Pw_forum] discrepancy in tests

Paolo Giannozzi paolo.giannozzi at uniud.it
Mon Mar 2 12:01:13 CET 2015


On Sat, 2015-02-28 at 15:53 +0900, WANG YUANQING wrote:

>   I have checked the difference (please see below). In fact, the
> difference is  obvious. Now, I do not know where this problem comes
> from and how should I do to fix it? Can you give  me some advice?

please provide your complete output. It is very likely that this problem
arises from a compilation or library bug, but it might be related to a
specific QE version or to a specific execution mode (serial, parallel, 
with and without openmp). I do not remember any recently fixed bug in 
DFT+U forces, but you should always try the latest version (there is an 
almost final 5.1.2 version on qe-forge)

Paolo

> Thanks in advance.
> 
> Best,
> 
> Yuanqing Wang
> 
> 
> \\\\\\\\\\\\\\\\\\\\\\\\
> <      atom    1 type  1   force =     0.11319036    0.11319036    
> 0.11319036
> <      atom    2 type  1   force =     0.11316353    0.11316353    
> 0.11316353
> <      atom    3 type  2   force =    -0.25890471   -0.25890471  
> -0.25890471
> <      atom    4 type  3   force =     0.03255082    0.03255082    
> 0.03255082
> ---
> >      atom    1 type  1   force =     0.00000040    0.00000040    
> 0.00000040
> >      atom    2 type  1   force =    -0.00000101   -0.00000101   
> -0.00000101
> >      atom    3 type  2   force =    -0.14189797   -0.14189797   
> -0.14189797
> >      atom    4 type  3   force =     0.14189858    0.14189858    
> 0.14189858
> 852c849
> <      Total force =     0.530215     Total SCF correction =     0.000021
> ---
> >      Total force =     0.347578     Total SCF correction =     0.000021
> \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
> 
> 	
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum

-- 
 Paolo Giannozzi, Dept. Chemistry&Physics&Environment, 
 Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
 Phone +39-0432-558216, fax +39-0432-558222 





More information about the users mailing list