The ABINIT Users Mailing List ( CLOSED )

[Masayoshi Mikami <mmikami@yk.rim.or.jp>]

Dear all,

I would like to draw your attention about "netiquette"...

Could you please try to make the size of your post as small as possible ?
I had a problem to receive the mail of 3.6MB (very long log file) at home !! 
I think that not a few subscribers (our MLs are ***global***, about 530 
e-mail accouts) also want to receive small-sized e-mails due to their network 
environment. (That is why we prepare the splitted src_tests_*.tar.gz at the 
ABINIT web page.) We like to reduce the burden of the ML server, too.

I could suggest, following the spirit mentioned in the FAQ :
1) try to make a good summary about your results: Specific & Concise
   Just a small portion (e.g. ending part) of Log/Output files could be 
cited. No need to cite all parts of Log/Output for most cases, if whole Input 
files are attached.
2) if you still want to attach an output/log (as a whole), please compress !
   gzip, zip etc. (NB: this is no effective for binary files, of course) 
3) if you might have attachment of MByte, please stop attachment !
   (We might want to set an upper limit (say, around 500KB-1MB) which could 
automatically stop the distribution of such big-sized e-mails, like the virus 
scan ?)   

   Instead, you might put your files to an anonymous ftp server etc.
   Otherwise, it might also be a good idea to have your WEB page.
   Then you can put your outputs(and images) on your home page. 
   It could reduce the traffic of the ML effectively. 
   You could suggest by saying,
   "Please visit my homepage http://www.xxx.xxx/ for the details."
   Having your own WEB page may be more effective than posting your 
   questions daily; many posts may be overlooked in our ML inbox.
   (Heavy traffic will make the ML less effective !)
   I think that some of the subscribers might want to use a sort of 
   **WEBLOG** instead of **daily posting**, to get our attentions 
   more effectively.
(*) WEBLOG FAQ: http://webdesign.about.com/cs/weblogs/a/aa060903a.htm etc.

This is also a good occastion to think of our netiquette rule. Thanks.

Regards,
Masayoshi

On Fri, 23 Jul 2004 19:02:12 +0200
"Chun Li " <chun.li@mf.mpg.de> wrote:

> Hi, Matthieu,
> 
> Many thanks for your reply.
> I have tried to calculate the problem using modified input text with
> lower tolerances (toldfe1=1.0d-10, tolwfr2=1.0d-20) and correct Gamma
> point in another cell (0 0 1.0). 
> This time, the same eigenvalues are found for the two Gamma points in
> two different cells. But the same error message lines appeared in the
> end of the log file. And a WARNING message appeared above them (I think
> this WARNING message can be overcome by increasing the nstep):
> 
> scprqt:  WARNING -
>   nstep=   60 was not enough non-SCF iterations to converge;
>   maximum residual=  1.980E-08 exceeds tolwfr=  1.000E-20
> 
> Frankly, I don't have my band structure; I am just studying ABINIT to
> calculate it for comparison with former literatures. Then I can do other
> things with this code and the chosen PPs.
> 
> I used the Troullier-Martins PP for Zn and O, respectively. The running
> system information is:
> Processor       : Intel(R) XEON(TM) CPU 2.20GHz
> KDE version:    3.0.5
> System:         Linux
> Release:        2.4.18-64GB-SWP
> Machine:        i686
> 
> The whole log file and output file are both attached. Hope they can
> offer you enough information. Thanks again.
> 
> Chun Li
> 
> -----Original Message-----
> From: verstraete@pcpm.ucl.ac.be [mailto:verstraete@pcpm.ucl.ac.be] 
> Sent: 2004年7月23日 10:08
> To: forum@abinit.org
> Subject: Re: [abinit-forum] about the band structure calculation
> 
> 
> Hello Chun,
> 
> Your input looks ok, apart from the tolerances which are too high
> (toldfe1 
> should be 1.e-10 and especially tolwfr2 should be 1.e-20 or less), but 
> this shouldn't affect results that much.
> 
> Your two sets of eigenvalues are not the same: the second point is not 
> gamma, but equivalent to 1/2 0 0, so the difference is normal. Kpoints
> are 
> specified in reduced reciprocal coordinates.
> 
> If I understand correctly, you already have your band structure, so the 
> problem is pretty light.
> 
> Given the error, you need to send us more details on the platform you're
> 
> using and the stderr of your job. It may be some memory overflow, or a 
> real (system dependent) bug, but I doubt it. Send your full log and 
> output, and your machine details. Try running on a machine with more 
> memory, is my bet. Make sure you use the latest stable version of
> abinit, 
> to avoid rediscovering bugs.
> 
> Matthieu
> 
> On Thu, 22 Jul 2004, Chun Li  wrote:
> 
> > Dear ABINIT users,
> >
> > When I tried to calculate the band structure of ZnO, I got the
> following
> > questions. I am a beginner, and really need your help. Thanks in
> > advance.
> >
> > 1) After calculation, there are several information lines at the end
> of
> > the log file:
> >
> >
> ========================================================================
> > ========
> > ----iterations are completed or convergence reached----
> > outwf  : write wavefunction to file my3xo_DS2_WFK
> > gstate : exiting
> > == END DATASET(S)
> > ==============================================================
> >
> ========================================================================
> > ========
> > ** Address Error **
> > End of diagnostics
> >
> > What is the "address error"? I really have no idea.
> >
> > 2) In the output file, I found that the eigenvalues for nkpt are all
> > negatives, between -6 and -18. And the values for gamma point I copied
> > below:
> >
> > These are the values for origin Gamma point:
> > kpt#   1, nband= 12, wtk=  1.00000, kpt=  0.0000  0.0000  0.0000
> > (reduced coord)
> > -17.44482 -16.36490  -7.53498  -7.53498  -7.46884  -7.41847  -7.41847
> > -6.89014
> >  -6.89014  -6.81808  -6.66375  -6.66375
> >
> > These are the values for the Gamma point in another cell:
> > kpt#  49, nband= 12, wtk=  1.00000, kpt= -0.5000  1.0000  0.0000
> > (reduced coord)
> > -16.43801 -16.10491  -7.61727  -7.36561  -7.32466  -7.06549  -6.92935
> > -6.79830
> >  -6.75443  -6.71596  -6.64487  -6.49350
> >
> > They are not identical, but I don't know where the error is. Maybe I
> > have given the wrong kptbounds??
> >
> > The following is my input file. Hope someone can help me.
> >
> > # ZnO
> > # Computation of the band structure.
> > # First, a SCF density computation, and then a non-SCF band structure
> > calculation.
> > ndtset 2
> > #Dataset 1 : usual self-consistent calculation
> > ngkpt1  3 3 2
> > prtden1  1
> > toldfe1  1.0d-6
> > #Dataset 2 : the band structure
> > iscf2    -2
> > getden2  -1
> > kptopt2  -4
> > nband2   12
> > ndivk2   8 16 8 16      # 8, 16, 8 and 16 divisions of the 4 segments,
> > delimited by 5 points.
> > kptbounds2  0.0    0.0     0.0  # Gamma point
> >                   0.0    0.0     1/2  # A point
> >                   0.0    1/2     1/2  # H point
> >                   0.0    1/2     0.0  # K point
> >                  -1/2    1.0     0.0  # Gamma point in another cell.
> > prtvol2  10
> > tolwfr2  1.0d-12
> > enunit2  1
> > #Definition of the unit cell
> > acell   6.15956186  6.15956186  9.83980322
> > angdeg  90 90 120
> > #Definition of the atom types
> > ntypat 2
> > znucl 30 8
> > #Definition of the atoms
> > natom 4
> > typat 1 2 1 2
> > xred           0.0  0.0  0.0
> >                 0.0  0.0  0.3820
> >                 1/3  2/3  1/2
> >                 1/3  2/3  0.8820
> > #Definition of the planewave basis set
> > ecut   48.0
> > #Definition of the SCF procedure
> > nstep  60
> > diemac 9.0
> >
> > Any advice will be important for me.
> >
> > Best regards.
> >
> > Chun Li
> >
> >
> 
> -- 
> ===================================================================
> Matthieu Verstraete               mailto:verstraete@pcpm.ucl.ac.be 
> PCPM, Boltzmann, pl. Croix du Sud, 1            tel: 010/ 47 33 59
> B-1348 Louvain-la-Neuve  Belgium                fax: 010/ 47 34 52