From acammarata at drexel.edu Fri Jan 20 01:45:54 2012 From: acammarata at drexel.edu (Antonio Cammarata) Date: Thu, 19 Jan 2012 19:45:54 -0500 Subject: [Wannier] seedname.dat not generated Message-ID: <1327020354.23627.26.camel@mse-cyclops.dhcp.drexel.edu> Dear All, I'm performing calculations on a perovskite structure using VASP. I got the MLWF and the band structure for my system and I tried to visualize the Wannier Centers using the seedname_centres.xyz file. I saw in this website: http://inac.cea.fr/L_Sim/BigDFT/tutorials/H3C-Wannier.html at the section "Visualizing the results with V_Sim" that each center can be colored depending on its occupation and represented by a sphere with a radius proportional to its spread. This can be done by means of the seedname.dat file that is generated by Wannier90. My problem is that I do not get this file, neither there is some keyword to generate it (such file is not even mentioned in the wannier90 user manual). How can I generate this file? Thanks a lot in advance for your kind reply. All the best Antonio Cammarata From acammarata at drexel.edu Fri Jan 20 01:34:20 2012 From: acammarata at drexel.edu (Antonio Cammarata) Date: Thu, 19 Jan 2012 19:34:20 -0500 Subject: [Wannier] Problems reading UNK* files Message-ID: <1327019660.23627.17.camel@mse-cyclops.dhcp.drexel.edu> Dear All, I'm performing calculations on a perovskite structure using VASP code. Once I got the MLWF, I reproduced the band structure that matches with the one calculated with VASP. Then I tried to plot one of the calculated Wannier functions but the program exits with this error: login2$ wannier90.x wannier90.up forrtl: severe (67): input statement requires too much data, unit 11, file ./UNK00001.1 Image PC Routine Line Source wannier90.x 0000000000578FBD Unknown Unknown Unknown wannier90.x 0000000000577AC5 Unknown Unknown Unknown wannier90.x 000000000052CCC0 Unknown Unknown Unknown wannier90.x 00000000004E7F8F Unknown Unknown Unknown wannier90.x 00000000004E77C2 Unknown Unknown Unknown wannier90.x 0000000000505827 Unknown Unknown Unknown wannier90.x 00000000004981CF w90_plot_mp_plot_ 831 plot.F90 wannier90.x 0000000000497B3A w90_plot_mp_plot_ 72 plot.F90 wannier90.x 0000000000404672 MAIN__ 177 wannier_prog.F90 wannier90.x 000000000040399C Unknown Unknown Unknown libc.so.6 0000003C7561D994 Unknown Unknown Unknown wannier90.x 00000000004038A9 Unknown Unknown Unknown login2$ Looking at the line 831 of the plot.F90 file it seems that the problem is related to the kind of integer numbers it tries to read. To check this, I try to read the first 5 integers of the UNK00001.1 file with this simple fortran program: program check integer*4 ngx, ngy, ngz, bnd, zz open(unit=1,file="UNK00001.1",FORM="UNFORMATTED") read(1)ngx, ngy, ngz, nk, nbnd print *, ngx, ngy, ngz, nk, ndnd end program and it gives the correct answer. So I modified the lines 811 and 812 of the plot.F90 in this way: integer*4 :: i,j,nsp,nat,nbnd,counter,ierr integer*4 :: loop_kpt,ik,ix,iy,iz,nk,ngx,ngy,ngz,nxx,nyy,nzz that is specifying the kind of integer and it works. My question is: does this change affect the resulting plot, maybe because now such values are of a kind not compatible with other variables? Thanks a lot in advance for your kind reply. Best Antonio Cammarata From a.mostofi at imperial.ac.uk Fri Jan 27 12:38:01 2012 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Fri, 27 Jan 2012 11:38:01 +0000 Subject: [Wannier] seedname.dat not generated In-Reply-To: <1327020354.23627.26.camel@mse-cyclops.dhcp.drexel.edu> References: <1327020354.23627.26.camel@mse-cyclops.dhcp.drexel.edu> Message-ID: <4F228C99.6080006@imperial.ac.uk> Dear Antonio, This extra output file (*.dat) must be a feature of the implementation of the interface in Abinit. Wannier90 writes the spreads to the main output file (*.wout). Also, if you set "write_proj = .true." in the input file (*.win), then the projection of each WF on the original bands in the outer window are written to the main output file as well. With this information, you should be able to achieve what you want. Best wishes, Arash -- Dr Arash A Mostofi Departments of Materials and Physics Deputy Director, CDT on Theory and Simulation of Materials Imperial College London, London SW7 2AZ, UK +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi On 20/01/2012 00:45, Antonio Cammarata wrote: > Dear All, > > I'm performing calculations on a perovskite structure using VASP. > I got the MLWF and the band structure for my system and I tried to > visualize the Wannier Centers using the seedname_centres.xyz file. > > I saw in this website: > http://inac.cea.fr/L_Sim/BigDFT/tutorials/H3C-Wannier.html > at the section "Visualizing the results with V_Sim" > that each center can be colored depending on its occupation and > represented by a sphere with a radius proportional to its spread. This > can be done by means of the seedname.dat file that is generated by > Wannier90. My problem is that I do not get this file, neither there is > some keyword to generate it (such file is not even mentioned in the > wannier90 user manual). How can I generate this file? > > Thanks a lot in advance for your kind reply. > > All the best > > Antonio Cammarata > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier > From acammarata at drexel.edu Tue Jan 31 17:19:10 2012 From: acammarata at drexel.edu (Antonio Cammarata) Date: Tue, 31 Jan 2012 11:19:10 -0500 Subject: [Wannier] seedname.dat not generated In-Reply-To: <4F228C99.6080006@imperial.ac.uk> References: <1327020354.23627.26.camel@mse-cyclops.dhcp.drexel.edu> <4F228C99.6080006@imperial.ac.uk> Message-ID: <1328026750.57901.13.camel@mse-cyclops.dhcp.drexel.edu> Dear Arash, thanks for your reply. I tried to set "write_proj = .true." but I dind't get any further information, the output file was the same I get without setting that keyword. I saw also that such keyword is not mentioned in the user guide. Thanks again for your reply. All the best Antonio On Fri, 2012-01-27 at 11:38 +0000, Arash Mostofi wrote: > Dear Antonio, > > This extra output file (*.dat) must be a feature of the implementation > of the interface in Abinit. > > Wannier90 writes the spreads to the main output file (*.wout). Also, if > you set "write_proj = .true." in the input file (*.win), then the > projection of each WF on the original bands in the outer window are > written to the main output file as well. With this information, you > should be able to achieve what you want. > > Best wishes, > > Arash > > -- > Dr Arash A Mostofi > Departments of Materials and Physics > Deputy Director, CDT on Theory and Simulation of Materials > Imperial College London, London SW7 2AZ, UK > +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi > > On 20/01/2012 00:45, Antonio Cammarata wrote: > > Dear All, > > > > I'm performing calculations on a perovskite structure using VASP. > > I got the MLWF and the band structure for my system and I tried to > > visualize the Wannier Centers using the seedname_centres.xyz file. > > > > I saw in this website: > > http://inac.cea.fr/L_Sim/BigDFT/tutorials/H3C-Wannier.html > > at the section "Visualizing the results with V_Sim" > > that each center can be colored depending on its occupation and > > represented by a sphere with a radius proportional to its spread. This > > can be done by means of the seedname.dat file that is generated by > > Wannier90. My problem is that I do not get this file, neither there is > > some keyword to generate it (such file is not even mentioned in the > > wannier90 user manual). How can I generate this file? > > > > Thanks a lot in advance for your kind reply. > > > > All the best > > > > Antonio Cammarata > > > > _______________________________________________ > > Wannier mailing list > > Wannier at quantum-espresso.org > > http://www.democritos.it/mailman/listinfo/wannier > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier From a.mostofi at imperial.ac.uk Tue Jan 31 18:48:46 2012 From: a.mostofi at imperial.ac.uk (Arash Mostofi) Date: Tue, 31 Jan 2012 17:48:46 +0000 Subject: [Wannier] seedname.dat not generated In-Reply-To: <1328026750.57901.13.camel@mse-cyclops.dhcp.drexel.edu> References: <1327020354.23627.26.camel@mse-cyclops.dhcp.drexel.edu> <4F228C99.6080006@imperial.ac.uk> <1328026750.57901.13.camel@mse-cyclops.dhcp.drexel.edu> Message-ID: <4F28297E.1070900@imperial.ac.uk> Dear Antonio, If you are using version 1.2 of Wannier90, then the keyword write_proj is there - though I think you are correct that it is not mentioned in the user guide. Because the projection of each WF onto the original set of Bloch states is trivially unity in the case of no disentanglement (ie, all WFs fully occupied), you will only see this information written to the *.wout file if you perform a calculation with disentanglement. Best wishes, Arash -- Dr Arash A Mostofi Departments of Materials and Physics Deputy Director, CDT on Theory and Simulation of Materials Imperial College London, London SW7 2AZ, UK +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi On 31/01/2012 16:19, Antonio Cammarata wrote: > Dear Arash, > > thanks for your reply. I tried to set "write_proj = .true." but I dind't > get any further information, the output file was the same I get without > setting that keyword. I saw also that such keyword is not mentioned in > the user guide. > > Thanks again for your reply. > > All the best > > Antonio > > On Fri, 2012-01-27 at 11:38 +0000, Arash Mostofi wrote: >> Dear Antonio, >> >> This extra output file (*.dat) must be a feature of the implementation >> of the interface in Abinit. >> >> Wannier90 writes the spreads to the main output file (*.wout). Also, if >> you set "write_proj = .true." in the input file (*.win), then the >> projection of each WF on the original bands in the outer window are >> written to the main output file as well. With this information, you >> should be able to achieve what you want. >> >> Best wishes, >> >> Arash >> >> -- >> Dr Arash A Mostofi >> Departments of Materials and Physics >> Deputy Director, CDT on Theory and Simulation of Materials >> Imperial College London, London SW7 2AZ, UK >> +44 (0)207 594 8154 | www.cmth.ph.ic.ac.uk/people/a.mostofi >> >> On 20/01/2012 00:45, Antonio Cammarata wrote: >>> Dear All, >>> >>> I'm performing calculations on a perovskite structure using VASP. >>> I got the MLWF and the band structure for my system and I tried to >>> visualize the Wannier Centers using the seedname_centres.xyz file. >>> >>> I saw in this website: >>> http://inac.cea.fr/L_Sim/BigDFT/tutorials/H3C-Wannier.html >>> at the section "Visualizing the results with V_Sim" >>> that each center can be colored depending on its occupation and >>> represented by a sphere with a radius proportional to its spread. This >>> can be done by means of the seedname.dat file that is generated by >>> Wannier90. My problem is that I do not get this file, neither there is >>> some keyword to generate it (such file is not even mentioned in the >>> wannier90 user manual). How can I generate this file? >>> >>> Thanks a lot in advance for your kind reply. >>> >>> All the best >>> >>> Antonio Cammarata >>> >>> _______________________________________________ >>> Wannier mailing list >>> Wannier at quantum-espresso.org >>> http://www.democritos.it/mailman/listinfo/wannier >>> >> _______________________________________________ >> Wannier mailing list >> Wannier at quantum-espresso.org >> http://www.democritos.it/mailman/listinfo/wannier > > > _______________________________________________ > Wannier mailing list > Wannier at quantum-espresso.org > http://www.democritos.it/mailman/listinfo/wannier