From bth20 at cam.ac.uk Tue Feb 1 01:32:07 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Tue, 1 Feb 2005 00:32:07 -0000 Subject: [Pw_forum] tetrahedra error In-Reply-To: <41FE7353.4030305@physics.rutgers.edu> Message-ID: Thanks for the recent advice on the Constrained VC relaxation On a different subject: I am trying out the procedure for calculating density of states (using dos.x). I am trying to do it for Cobalt. I have run the scf calculation successfully. When I do the next stage (as in example08), I get the following error message: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from tetrahedra : error # 13 cannot remap grid on k-point list %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I'm not sure what the problem is. Here is my input: &CONTROL calculation = 'nscf' , restart_mode = 'from_scratch' , pseudo_dir = '/scratch/bth/Programs/Espresso/pseudo/' , prefix = 'Co', verbosity = 'high', / &SYSTEM ibrav = 4, celldm(1) = 4.738, celldm(3) = 1.62, nat = 2, ntyp = 1, ecutwfc = 30 , ecutrho = 80 , nspin = 2, starting_magnetization(1) = 1.0, occupations = 'tetrahedra' / &ELECTRONS mixing_beta = 0.7, conv_thr = 1.0d-7 / &IONS upscale = 10, / &CELL / ATOMIC_SPECIES Co 58.93 Co.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {crystal} Co 0.353333333 0.666666667 0.250000000 1 1 1 Co 0.666666667 0.333333333 0.750000000 1 1 1 K_POINTS {automatic} 12 12 12 1 1 1 Any thoughts appreciated Ben From degironc at sissa.it Tue Feb 1 09:13:46 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 01 Feb 2005 09:13:46 +0100 Subject: [Pw_forum] tetrahedra error References: Message-ID: <41FF3A3A.6010502@sissa.it> from the source code (PW/kpoint_grid.f90): .... do n=1,nks do nk=1,nkr if (equiv(nk).eq.n) go to 20 end do ! this failure of the algorithm may indicate that the displaced grid ! (with k1,k2,k3.ne.0) does not have the full symmetry of the lattice call errore('tetrahedra','cannot remap grid on k-point list',n) 20 continue end do .... which is indeed your case since grid 12 12 12 1 1 1 does not have the full hcp symmetry try 12 12 12 0 0 1 stefano Ben Hope wrote: >Thanks for the recent advice on the Constrained VC relaxation > >On a different subject: > >I am trying out the procedure for calculating density of states (using >dos.x). I am trying to do it for Cobalt. > >I have run the scf calculation successfully. When I do the next stage (as in >example08), I get the following error message: > >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from tetrahedra : error # 13 > cannot remap grid on k-point list >%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > >I'm not sure what the problem is. > >Here is my input: > >&CONTROL > > calculation = 'nscf' , > > restart_mode = 'from_scratch' , > > pseudo_dir = '/scratch/bth/Programs/Espresso/pseudo/' , > > prefix = 'Co', > > verbosity = 'high', > >/ > >&SYSTEM > > ibrav = 4, > > celldm(1) = 4.738, > > celldm(3) = 1.62, > > nat = 2, > > ntyp = 1, > > ecutwfc = 30 , > > ecutrho = 80 , > > nspin = 2, > > starting_magnetization(1) = 1.0, > > occupations = 'tetrahedra' > >/ > >&ELECTRONS > > mixing_beta = 0.7, > > conv_thr = 1.0d-7 > >/ > >&IONS > > upscale = 10, > >/ > >&CELL > >/ > >ATOMIC_SPECIES > > Co 58.93 Co.pbe-nd-rrkjus.UPF > >ATOMIC_POSITIONS {crystal} > > Co 0.353333333 0.666666667 0.250000000 1 1 1 > > Co 0.666666667 0.333333333 0.750000000 1 1 1 > >K_POINTS {automatic} > > 12 12 12 1 1 1 > > > > >Any thoughts appreciated > >Ben > > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From cvasse at fy.chalmers.se Tue Feb 1 11:33:01 2005 From: cvasse at fy.chalmers.se (Vasile Chis) Date: Tue, 01 Feb 2005 11:33:01 +0100 Subject: [Pw_forum] q2r.x for bigger systems Message-ID: <41FF5ADD.4070706@fy.chalmers.se> Dear users. I have a problem with q2r.x for bigger systems. The system is a 19 layer slab. The dynamical matrices are calculated over a 4x4 MP-mesh I have already edited the q2r.f90 in order to handle larger number of layers. My problem is that it doesn't happen anything when running it. Input file: cat > q2r.in < Hi everyone, There was a post about this error on January 18th but I don't see a solution posted. I am now getting an identical error. When I run the attached input (on an IBM SP3) I get : / NEW-OLD atomic charge density approx. for the potential/ /*** error in Message Passing (mp) module ***/ /*** error code: 8912 /This happens between ionic steps of a BFGS structural relaxation. / /I looked at the code, and I think what it tells me is that a parameter mp_msgsiz_max (=100000000) is being exceeded. Maybe the amount of information that needs to be passed between processors is too big ? Should I just increase this parameter and recompile ? (or will this blow a fuse ? ) Thanks for any help! Paul P.S. Grazie Gerardo & Yingwen for your help with my last question. I got sidetracked and I still haven't compiled on my Opteron but I will post my solution when I do. __________________________________________ &CONTROL title = 'water' , calculation = 'relax' , restart_mode = 'from_scratch' , outdir = '/home1/u9/tangney/Water/PWSCF/Configurations/Monomer/Test', pseudo_dir = '/home1/u9/tangney/Water/PWSCF', disk_io = 'high' , verbosity = 'high' , iprint = 1 , nstep = 200 , tstress = .true. , tprnfor = .true. , forc_conv_thr=1.0d-5, / &SYSTEM ibrav = 1, celldm(1) = 40.0, nat = 3, ntyp = 2, ecutwfc = 85.0d0, nosym = .true. , nbnd = 10, nelec = 8, / &ELECTRONS electron_maxstep = 200, startingwfc = 'random' , conv_thr = 1.0d-12, / &IONS ion_dynamics = 'bfgs' , / ATOMIC_SPECIES O 16.00000 O.blyp.upf H 1.00000 h.ps.upf ATOMIC_POSITIONS bohr O -14.627030000 -2.682920000 -3.082660000 H -13.707290000 -1.368870000 -2.037010000 H -13.472600000 -3.918840000 -3.916950000 K_POINTS automatic 1 1 1 0 0 0 / / -- ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Dr. Paul Tangney Theory of Nanostructured Materials Facility The Molecular Foundry Lawrence Berkeley National Lab. E-mail: PTTangney at lbl.gov 1 Cyclotron Road, Bldg 66 Phone: (510) 642-2635 Berkeley, CA 94720 Fax : (510) 643-9345 ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050201/894fa36b/attachment.htm From eyvaz_isaev at yahoo.com Wed Feb 2 10:57:06 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Wed, 2 Feb 2005 01:57:06 -0800 (PST) Subject: [Pw_forum] q2r.x for bigger systems In-Reply-To: <41FF5ADD.4070706@fy.chalmers.se> Message-ID: <20050202095706.9960.qmail@web60304.mail.yahoo.com> Hi, > > My problem is that it doesn't happen anything when > running it. It is not so clear what actually happens. If absolutly nothing happens with command q2r.x out it means that it is waiting for input from command line which is very unlikely. In my expertize there must be a message from q2r.x if something goes wrong with input file. May be you just entered q2r.x without " > Input file: > > cat > q2r.in < &input > nr1=4,nr2=4,nr3=1,zasr=.false.,fild='19slab.fc' > &end > 6 > dyn1.dyn > dyn2.dyn > dyn3.dyn > dyn4.dyn > dyn5.dyn > dyn6.dyn > EOF > > Hopefully it's only some kind of input problem. > Anyone encountered this kind of problem? > Best regards > /Vasse > > -- > __________________________________ > Vasile Chis > Interface Modelling Group > Department of Physics > G?teborgs Universitet > email :cvasse at fy.chalmers.se > __________________________________ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail From giannozz at nest.sns.it Wed Feb 2 15:30:57 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 2 Feb 2005 15:30:57 +0100 Subject: [Pw_forum] error in mp module In-Reply-To: <4200219E.5060601@civet.berkeley.edu> References: <4200219E.5060601@civet.berkeley.edu> Message-ID: <200502021530.57668.giannozz@nest.sns.it> On Wednesday 02 February 2005 01:41, Paul Tangney wrote: > There was a post about this error on January 18th but > I don't see a solution posted. I replied directly to the person who submitted the error, hoping that nobody would run such a big calculation any time soon... > I am now getting an identical error [..] I looked at the code, > and I think what it tells me is that a parameter mp_msgsiz_max > (=100000000) is being exceeded. correct. There are a few checks in the code that a number is not out of range, such as for instance larger than some arbitrarily defined "unreasonably large" number. For instance, 5000 is considered "unreasonably large" for a FFT grid dimension. I think that your charge density size is exceeding the "unreasonably large" value. Just increase mp_msgsiz_max and recompile ("make pw" or whatever you need): there shouldn't be any side effect. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From subhra at iitg.ernet.in Thu Feb 3 05:22:35 2005 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Thu, 3 Feb 2005 09:52:35 +0530 (IST) Subject: [Pw_forum] Warnings Message-ID: Hi, I am trying to test the convergence of Gamma and X phonons for a simple metal in fcc structure. I got the convergence of Gamma phonons using a high plane-wave cut off (60 Ryd), a dense k-space mesh(22x22x22) using M-P smearing and without any offset in the k-mesh. Using the conjigate-greadient diagonalization for the scf part I always got warnings '1 eigenvector did not converge' at the last iteration. Therefore, I used Davidson diagonalization which took care of the warning. When I try to test the convergence of X phonons, I used the same set of parameters for the nscf calculation. Using the Davidson diagonalization for the nscf part I get hundreds of warning '1 eigenvector did not converge'. I then used Conjugate-gradient diagonalization for the nscf part to see if it takes care of the problem. To my astonishment, the warning is still there, albeit less in number. To my understanding, under such circumstances, one should not trust the X phonon results. Can somebody suggest a way out? Regards, Subhradip -- ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology, Guwahati North Guwahati, Assam-781039 India E-mail: subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) ************************************************************************ From cvasse at fy.chalmers.se Wed Feb 2 14:22:39 2005 From: cvasse at fy.chalmers.se (Vasile Chis) Date: Wed, 02 Feb 2005 14:22:39 +0100 Subject: [Pw_forum] q2r.x for bigger systems Message-ID: <4200D41F.6070506@fy.chalmers.se> Hi again, The gamma dynamical matrix is calculated, which I call "dyn1.dyn". The error message I get is: - 1525-006 The STATUS= specifier in the OPEN statement for unit 1 cannot be set to OLD because - the file aldyn1.dyn does not exist. The program will stop. But, the file does so exist. I have edited the q2r.f as I said last time and also changed, for q(3,48) to q(3,57) and so on for all that which needed to be changed. Any clues? Best regards, Vasse >Hi, >/ />/ My problem is that it doesn't happen anything when />/ running it. / >It is not so clear what actually happens. If >absolutly nothing happens with command q2r.x /out it means that it is waiting for input from />command line which is very unlikely. In my expertize >there must be a message from q2r.x if something goes >wrong with input file. >May be you just entered q2r.x without "One comment not related to the problem. Set up >"zasr=.true." if you like to get correct frequencies >for the Gamma point. By the way, did you calculate the >dynamical matrix for the Gamma point? From your input >it is not clear. >Bests, >Eyvaz. >/ />/ Input file: />/ />/ cat > q2r.in </ &input />/ nr1=4,nr2=4,nr3=1,zasr=.false.,fild='19slab.fc' />/ &end />/ 6 />/ dyn1.dyn />/ dyn2.dyn />/ dyn3.dyn />/ dyn4.dyn />/ dyn5.dyn />/ dyn6.dyn />/ EOF />/ />/ Hopefully it's only some kind of input problem. />/ Anyone encountered this kind of problem? />/ Best regards />/ /Vasse />/ / -- __________________________________ Vasile Chis Interface Modelling Group Department of Physics G?teborgs Universitet email :cvasse at fy.chalmers.se __________________________________ From Myeong.Lee at asu.edu Thu Feb 3 06:06:03 2005 From: Myeong.Lee at asu.edu (Myeong.Lee at asu.edu) Date: Wed, 02 Feb 2005 22:06:03 -0700 (MST) Subject: [Pw_forum] pwscf question Message-ID: <1107407163.4201b13b7fec3@webmail.asu.edu> Dear Sir, I used PWSCF for several molecules, and for some molecules I got both positive and negative values of imaginary k. So, I wonder if imaginary k can have negative value depending on the property of the molecule or if I did something wrong. (My guess is that if the band edge appears at k=pi/a, it has negative imaginary k.) I really appreciate your time. Sincerely, Myeong Lee. From giannozz at nest.sns.it Thu Feb 3 09:29:09 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 3 Feb 2005 09:29:09 +0100 Subject: [Pw_forum] Warnings In-Reply-To: References: Message-ID: <200502030929.09387.giannozz@nest.sns.it> On Thursday 03 February 2005 05:22, Subhradip Ghosh wrote: > I always got warnings '1 eigenvector did not converge' at the > last iteration [...]. To my understanding, under such circumstances, > one should not trust the X phonon results. not necessarily: default convergence thresholds are quite strict. Moreover in a metal you should always have some empty bands. The highest bands may be unreliable, but they do not contribute to the energy or to vibrational frequencies Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Thu Feb 3 09:37:28 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 3 Feb 2005 09:37:28 +0100 Subject: [Pw_forum] q2r.x for bigger systems In-Reply-To: <4200D41F.6070506@fy.chalmers.se> References: <4200D41F.6070506@fy.chalmers.se> Message-ID: <200502030937.28174.giannozz@nest.sns.it> On Wednesday 02 February 2005 14:22, Vasile Chis wrote: > The error message I get is: > > - 1525-006 The STATUS= specifier in the OPEN statement for unit 1 cannot be > set to OLD because - the file aldyn1.dyn does not exist. The program will > stop. > > But, the file does so exist. add a "call system('ls -l '//TRIM(filin))" before the open statement. If the file is there, replace OPEN(unit=1,file=filin,status='old',form='formatted') with OPEN(unit=1,file=TRIM(filin),status='old',form='formatted') (it shouldn't be needed, but who knows). If it still doesn't work, recompile q2r.x with boundary check enabled and see what happens. P. -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From degironc at sissa.it Thu Feb 3 09:43:09 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 03 Feb 2005 09:43:09 +0100 Subject: [Pw_forum] q2r.x for bigger systems References: <4200D41F.6070506@fy.chalmers.se> Message-ID: <4201E41D.9010005@sissa.it> Vasile Chis wrote: > Hi again, > > The gamma dynamical matrix is calculated, which I call "dyn1.dyn". > > The error message I get is: > > - 1525-006 The STATUS= specifier in the OPEN statement for unit 1 > cannot be set to OLD because > - the file aldyn1.dyn does not exist. The program will stop. > > But, the file does so exist. check your input file. or write it again from scratch. If the code want to open aldyn1.dyn is becouse you tell it to do so.... You may have inserted strange control character in your input. > > I have edited the q2r.f as I said last time and also changed, > for q(3,48) to q(3,57) and so on for all that which needed to be changed. Why did you so ? I doubt you should need that . Stefano de Gironcoli From bth20 at cam.ac.uk Thu Feb 3 11:23:11 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Thu, 3 Feb 2005 10:23:11 -0000 Subject: [Pw_forum] magnetic moments In-Reply-To: Message-ID: Hi guys How can I calculate the magnetic moment of each atom? (I've set nspin = 2, starting_magnetization(1) = 1.0 etc but can't see it in the output) Regards Ben From swblelia at sw.ehu.es Thu Feb 3 15:35:26 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Thu, 03 Feb 2005 15:35:26 +0100 Subject: [Pw_forum] about example 03 (Al relaxation) Message-ID: <420236AE.7030007@sw.ehu.es> Hello: I have a short question. I am trying to relax Magnesium surfaces in the same way as example03 for aluminium. When I run the example what I don't understand is that only Z-direction is relaxed. How come that atoms do not relax moving in X and Y direction? As far as I understand from the input, there are not fixed directions of atom positions so in principle they should all move in 3 directions. regards From degironc at sissa.it Thu Feb 3 17:35:43 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 03 Feb 2005 17:35:43 +0100 Subject: [Pw_forum] about example 03 (Al relaxation) References: <420236AE.7030007@sw.ehu.es> Message-ID: <420252DF.9090801@sissa.it> Aritz Leonardo wrote: > Hello: > > I have a short question. I am trying to relax Magnesium surfaces in > the same way as example03 for aluminium. > > When I run the example what I don't understand is that only > Z-direction is relaxed. How come that atoms do not relax moving in X > and Y direction? As far as I understand from the input, there are not > fixed directions of atom positions so in principle they should all > move in 3 directions. symmetry (which is found by the code and preserved during relaxation) is the reason stefano deGironcoli From cbarreteau at cea.fr Thu Feb 3 19:04:11 2005 From: cbarreteau at cea.fr (Cyrille Barreteau) Date: Thu, 3 Feb 2005 19:04:11 +0100 (CET) Subject: [Pw_forum] about example 03 (Al relaxation) In-Reply-To: <420252DF.9090801@sissa.it> References: <420236AE.7030007@sw.ehu.es> <420252DF.9090801@sissa.it> Message-ID: <1262.81.56.70.96.1107453851.squirrel@dsm-mail> If you take a less symmetric surface such as a vicinal surface or a reconstructed (missing row reconstruction for example) then you will find x,y,z relaxation... cyrille barreteau > Aritz Leonardo wrote: > >> Hello: >> >> I have a short question. I am trying to relax Magnesium surfaces in >> the same way as example03 for aluminium. >> >> When I run the example what I don't understand is that only >> Z-direction is relaxed. How come that atoms do not relax moving in X >> and Y direction? As far as I understand from the input, there are not >> fixed directions of atom positions so in principle they should all >> move in 3 directions. > > symmetry (which is found by the code and preserved during relaxation) > is the reason > > stefano deGironcoli > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From subhra at iitg.ernet.in Fri Feb 4 03:52:52 2005 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Fri, 4 Feb 2005 08:22:52 +0530 (IST) Subject: [Pw_forum] Warnings In-Reply-To: <200502030929.09387.giannozz@nest.sns.it> References: <200502030929.09387.giannozz@nest.sns.it> Message-ID: On Thu, 3 Feb 2005, Paolo Giannozzi wrote: > On Thursday 03 February 2005 05:22, Subhradip Ghosh wrote: > > > I always got warnings '1 eigenvector did not converge' at the > > last iteration [...]. To my understanding, under such circumstances, > > one should not trust the X phonon results. > > not necessarily: default convergence thresholds are quite strict. > Moreover in a metal you should always have some empty bands. > The highest bands may be unreliable, but they do not contribute > to the energy or to vibrational frequencies > > Paolo > > Thanks. While running the nscf calculations I get the same warning a couple of times. Presumably, two eigenvectors have problem with convergence. Should I ignore it as well? Subhradip -- ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology, Guwahati North Guwahati, Assam-781039 India E-mail: subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) ************************************************************************ From subhra at iitg.ernet.in Fri Feb 4 07:56:18 2005 From: subhra at iitg.ernet.in (Subhradip Ghosh) Date: Fri, 4 Feb 2005 12:26:18 +0530 (IST) Subject: [Pw_forum] Convergence of Gamma phonon Message-ID: Dear All, I have a question regarding convergence of Gamma phonon for metallic cases. I understand that to check the convergence of Gamma phonon frequencies, one should test convergence with respect to PW cut-off, k-point mesh, smearing parameter sequentially. My question is: what is the criteria for acceptable convergence? Is it only if the difference in gamma phonon frequencies are ~1% witrh respect to all of the above parameters? Is a higher tolerance(say 3%) for gamma phonons permitted? I am asking this for a specific reason. I am doing fcc Copper. For the Gamma phonon, increasing the PW cut-off upto 60 Ryd hasn't given me convergence ~1%. I am using a k-mesh of 12x12x12 without offset. It seems strange to me that such a high cut-off is required. Am I going the wrong way? Should I first look for the convergence with respect to k-mesh? Subhradip -- ************************************************************************ Dr. Subhradip Ghosh Assistant Professor Department of Physics Indian Institute of Technology, Guwahati North Guwahati, Assam-781039 India E-mail: subhra at iitg.ernet.in Phone: +91 361 2582717(O) +91 361 2584717(R) ************************************************************************ From gj224 at cam.ac.uk Thu Feb 3 14:42:41 2005 From: gj224 at cam.ac.uk (G. Jones) Date: 03 Feb 2005 13:42:41 +0000 Subject: [Pw_forum] compilation In-Reply-To: References: Message-ID: Dear Forum I attempting to compile PW_scf on a parallel machine but am getting the following error: do you know the reason? test -d bin || mkdir bin ( cd Modules; make all ) ( cd clib ; make all ) ( cd flib ; make all ) if test -d PW ; then ( cd PW ; make all ) ; fi fpp -P -D__SUN -D__MPI -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I/opt/SUNWhpc/include -I. -I../include -I../Modules -I../PW -I../PH cft_3.f90 -o cft_3.F90 "command line", warning: bad option: -o , ignored. mpf90 -O3 -dalign -xchip=ultra3 -xarch=v9b -xlic_lib=sunperf -I/opt/SUNWhpc/include -I. -I../include -I../Modules -I../PW -I../PH -M. -M../Modules -M../PW -M../PH -D__SUN -D__MPI -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c cft_3.F90 -o cft_3.o subroutine cft_3 (f, n1, n2, n3, nx1, nx2, nx3, igrid, sign) ^ "cft_3.F90", Line = 178, Column = 12: ERROR: Global name "CFT_3" is defined as a SUBROUTINE at line 131 (cft_3.F90) and as a SUBROUTINE in this program unit. f90comp: 221 SOURCE LINES f90comp: 1 ERRORS, 0 WARNINGS, 0 OTHER MESSAGES, 0 ANSI *** Error code 1 make: Fatal error: Command failed for target `cft_3.o' Current working directory /auto/users/gj224/espresso/PW *** Error code 1 make: Fatal error: Command failed for target `pw' franklin-1$pwd /users/gj224/espresso ______________________ Glenn Jones ___________________________ Department of Chemistry Lensfield Road University of Cambridge UK CB2 1EW 01223 336508 ___________________________ From baroni at sissa.it Fri Feb 4 08:47:12 2005 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 4 Feb 2005 08:47:12 +0100 Subject: [Pw_forum] Convergence of Gamma phonon In-Reply-To: References: Message-ID: <85ae3dca59e3e0e9bb0b0d747ea88846@sissa.it> Dear Subhadip: On Feb 4, 2005, at 7:56 AM, Subhradip Ghosh wrote: > Dear All, > > I have a question regarding convergence of Gamma phonon for metallic > cases. > > I understand that to check the convergence of Gamma phonon frequencies, > one should test convergence with respect to PW cut-off, k-point mesh, > smearing parameter sequentially. > > My question is: what is the criteria for acceptable convergence? > > Is it only if the difference in gamma phonon frequencies are ~1% witrh > respect > to all of the above parameters? > > Is a higher tolerance(say 3%) for gamma phonons permitted? only you can say. if you want to compare with an experimental datum which is know to, say, 10%, I would say that 3% is quite acceptable. if instead, you want to calculate a derived quantity which depends sensitively on the phonon frequency, then a higher accuracy may be required. you have also to be aware of the overall inaccuracy coming from systematic approximations (mainly, in your case, the lda/gga, and the use of pseudopotentials). it may be silly to pursue very high numerical accuracy when you know from the beginning that the best calculations you can possibly do will always be affected by such systematic errors. the estimate of such systematic errors, in turn, is often more the matter of an art than of a science ... only experience can guide you in this unexplored land ... > > I am asking this for a specific reason. > > I am doing fcc Copper. For the Gamma phonon, increasing the PW cut-off > upto 60 Ryd hasn't given me convergence ~1%. I am using a k-mesh of > 12x12x12 without offset. > > It seems strange to me that such a high cut-off is required. first-row elements and first-row transition metals are known to have the "hardest" pseudopotentials. first-row transition metals more than first-row elements. the heavier, the worst. overall, copper is one of the worst cases. I do not find anything strange in the lack of convergence at 60 Ry. of course, it depends on the specific pseudopotential (choice of the core radii, etc.). this is THE case where the use of ultra-soft pseudopotentials would be kind of mandatory (that should allow to decrese the cutoff to below 30, Ry, I would say) > Am I going the wrong way? Should I first look for the convergence with > respect to k-mesh? the convergence with respect to the cutoff depends on the properties of the atomic core, that with respect to kpoints on the properties of the crystal. they are independent and they should be checked independently. hope this clarifies a bit Stefano B. > > Subhradip > > -- > *********************************************************************** > * > Dr. Subhradip Ghosh > Assistant Professor > Department of Physics > Indian Institute of Technology, Guwahati > North Guwahati, Assam-781039 > India > E-mail: subhra at iitg.ernet.in > Phone: +91 361 2582717(O) > +91 361 2584717(R) > *********************************************************************** > * > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Stefano Baroni --- SISSA & DEMOCRITOS National Simulation Center via Beirut 2-4 34014 Trieste Grignano / [+39] 040 3787 406 (tel) -528 (fax) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 3527 bytes Desc: not available Url : /pipermail/attachments/20050204/f64e4379/attachment.bin From baroni at sissa.it Fri Feb 4 09:07:35 2005 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 4 Feb 2005 09:07:35 +0100 Subject: [Pw_forum] about example 03 (Al relaxation) In-Reply-To: <1262.81.56.70.96.1107453851.squirrel@dsm-mail> References: <420236AE.7030007@sw.ehu.es> <420252DF.9090801@sissa.it> <1262.81.56.70.96.1107453851.squirrel@dsm-mail> Message-ID: <91ffd757dd584c0ef380e6421e2d1872@sissa.it> Dear Leonardo: all what has been said is correct. Let me add a general theoretical consideration which may possibly help clarify the matter. In any finite system, vanishing forces on individual atoms are a NECESSARY and SUFFICIENT condition for mechanical equilibrium (whether the equilibrium is stable or not depends on the second derivatives of the energy: squared phonon frequencies in our case). In an infinite system (such as a surface), instead, fanisihing forces are a NECESSARY, but NOT SUFFICIENT, condition for mechanical equilibrium. As strange as this fact might seem, it is indeed rather intuitive. Think of a textbook linear harmonic chain with first-neighbor interactions. If all the nearest-neighbor distances are equal, then the forces acting on individual "atoms" will vanish. Nevertheless, the energy per atom of the system will obviously depend on such distances, and in general will be minimal for a well determined value of them, only. If you cut the linear chain (you create a "surface"), then the degrees of freedom perpendicular to the surface (the "z" axis) will relax. In a 3D system, the degrees of freedom parallel to the surface may or may not relax, depending on structure of the surface. If the surface unit cell has only one atom (or else if its symmetry is sufficiently high), then the parallel degrees of freedom will not relax, even though the surface is not in mechanical equilibrium. What is missing, to define the SUFFICIENT conditions for mechanical equilibrium in an infinite system is the STRESS (whose concept is a generalization of that of PRESSURE). This, however, wil be the subject of a different thread ... Take care Stefano B. On Feb 3, 2005, at 7:04 PM, Cyrille Barreteau wrote: > If you take a less symmetric surface such as a vicinal > surface or a reconstructed (missing row reconstruction for > example) then you will find x,y,z relaxation... > cyrille barreteau > >> Aritz Leonardo wrote: >> >>> Hello: >>> >>> I have a short question. I am trying to relax Magnesium surfaces in >>> the same way as example03 for aluminium. >>> >>> When I run the example what I don't understand is that only >>> Z-direction is relaxed. How come that atoms do not relax moving in X >>> and Y direction? As far as I understand from the input, there are not >>> fixed directions of atom positions so in principle they should all >>> move in 3 directions. >> >> symmetry (which is found by the code and preserved during relaxation) >> is the reason >> >> stefano deGironcoli >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Stefano Baroni --- SISSA & DEMOCRITOS National Simulation Center via Beirut 2-4 34014 Trieste Grignano / [+39] 040 3787 406 (tel) -528 (fax) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 3139 bytes Desc: not available Url : /pipermail/attachments/20050204/b1f2b6c1/attachment.bin From giannozz at nest.sns.it Fri Feb 4 09:25:49 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 4 Feb 2005 09:25:49 +0100 Subject: [Pw_forum] compilation In-Reply-To: References: Message-ID: <200502040925.49305.giannozz@nest.sns.it> On Thursday 03 February 2005 14:42, G. Jones wrote: > -D__SUN [...] -D__FFTW -D__USE_INTERNAL_FFTW I think that -D__SUN implies that you use sunperf libraries for fft. Do not define -D__FFTW -D__USE_INTERNAL_FFTW P, -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From massimiliano.bonomi at mi.infn.it Fri Feb 4 10:15:46 2005 From: massimiliano.bonomi at mi.infn.it (Massimiliano Bonomi) Date: Fri, 04 Feb 2005 10:15:46 +0100 Subject: [Pw_forum] wf_collect problem Message-ID: <42033D42.5030407@mi.infn.it> Dear users, I'm testing the new option wf_collect in espresso package version 2.1.2 on two different parallel machines. The first one has a Infiniband network, Intel compiler v. 8.1 and mpich-vmi 2.0 the second one a Myrinet network with pgi compiler. In the first case the wf_collect option does work correctly, in the second the scf calculation stops at this point: > entering subroutine stress ... > > total stress (ryd/bohr**3) (kbar) P= -0.12 > -0.00007282 0.00000000 0.00000000 -10.71 0.00 0.00 > 0.00000000 -0.00007282 0.00000000 0.00 -10.71 0.00 > 0.00000000 0.00000000 0.00014323 0.00 0.00 21.07 > > > Writing file out.save for program phonon Any idea of the problem?? Thanks in advance. Massimiliano Bonomi. From bth20 at cam.ac.uk Fri Feb 4 11:41:06 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Fri, 4 Feb 2005 10:41:06 -0000 Subject: [Pw_forum] relax problem In-Reply-To: Message-ID: Hi I am performing a relax calculation on a very thin Cobalt cylinder. The relaxation process converges if it is non spin-polarized (nspin=1) but not if it is spin-polarized (nspin=2). I have tried reducing the mixing beta but it didn't help. Any ideas? Here is the input file: &CONTROL calculation = 'relax' , restart_mode = 'from_scratch' , pseudo_dir = '/scratch/bth/Programs/Espresso/pseudo/' , prefix = 'CoCyl01', verbosity = 'high', / &SYSTEM ibrav = 8, celldm(1) = 4.0695, celldm(2) = 6.0, celldm(3) = 6.0, nat = 10, ntyp = 1, ecutwfc = 30 , ecutrho = 80 , occupations = 'smearing', degauss = 0.03, nspin = 2, starting_magnetization(1) = 1.0, / &ELECTRONS mixing_beta = 0.5 conv_thr = 1.0d-7 / &IONS upscale = 10, / &CELL / ATOMIC_SPECIES Co 58.93 Co.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {crystal} Co 0.0 0.0 0.0 1 1 1 Co 0.0 0.1026784617 0.0 1 1 1 Co 0.0 -0.1026784617 0.0 1 1 1 Co 0.0 0.0513392309 0.0889221563 1 1 1 Co 0.0 -0.0513392309 0.0889221563 1 1 1 Co 0.0 0.0513392309 -0.0889221563 1 1 1 Co 0.0 -0.0513392309 -0.0889221563 1 1 1 Co 0.5 0.0513392309 0.0296407188 1 1 1 Co 0.5 -0.0513392309 0.0296407188 1 1 1 Co 0.5 0.0 -0.0592814375 1 1 1 K_POINTS {automatic} 7 1 1 1 1 1 ==================================== Ben Hope C.U.H.&H. Club Captain St. John's College CAMBRIDGE CB2 1TP Tel: - Home: (01223) 578417 - Office: (01223) 760312 - Mobile: 07742 517432 www.cam.ac.uk/societies/cuhh ==================================== From eyvaz_isaev at yahoo.com Fri Feb 4 11:53:25 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 4 Feb 2005 02:53:25 -0800 (PST) Subject: [Pw_forum] relax problem In-Reply-To: Message-ID: <20050204105325.53119.qmail@web60308.mail.yahoo.com> Hi, At least your ecutrho seems to be very small. It should be at least 4 times larger than encut. In fact, this parameter is very imoprtant for US pseudopotentials. Besides I do not see a smearing method you used, though you use "occupation=smearing". So, you should add "smearing=m-v", say. You can try another method, if you like. Hope it helps. Bests, Eyvaz. --- Ben Hope wrote: > Hi > > I am performing a relax calculation on a very thin > Cobalt cylinder. The > relaxation process converges if it is non > spin-polarized (nspin=1) but not > if it is spin-polarized (nspin=2). I have tried > reducing the mixing beta but > it didn't help. > > Any ideas? > > Here is the input file: > > &CONTROL > > calculation = 'relax' , > > restart_mode = 'from_scratch' , > > pseudo_dir = > '/scratch/bth/Programs/Espresso/pseudo/' , > > prefix = 'CoCyl01', > > verbosity = 'high', > > / > > &SYSTEM > > ibrav = 8, > > celldm(1) = 4.0695, > > celldm(2) = 6.0, > > celldm(3) = 6.0, > > nat = 10, > > ntyp = 1, > > ecutwfc = 30 , > > ecutrho = 80 , > > occupations = 'smearing', > > degauss = 0.03, > > nspin = 2, > > starting_magnetization(1) = 1.0, > > / > > &ELECTRONS > > mixing_beta = 0.5 > > conv_thr = 1.0d-7 > > / > > &IONS > > upscale = 10, > > / > > &CELL > > / > > ATOMIC_SPECIES > > Co 58.93 Co.pbe-nd-rrkjus.UPF > > ATOMIC_POSITIONS {crystal} > > Co 0.0 0.0 0.0 1 1 1 > > Co 0.0 0.1026784617 0.0 1 1 1 > > Co 0.0 -0.1026784617 0.0 1 1 1 > > Co 0.0 0.0513392309 0.0889221563 1 1 1 > > Co 0.0 -0.0513392309 0.0889221563 1 1 1 > > Co 0.0 0.0513392309 -0.0889221563 1 1 1 > > Co 0.0 -0.0513392309 -0.0889221563 1 1 > 1 > > Co 0.5 0.0513392309 0.0296407188 1 1 1 > > Co 0.5 -0.0513392309 0.0296407188 1 1 1 > > Co 0.5 0.0 -0.0592814375 1 1 1 > > K_POINTS {automatic} > > 7 1 1 1 1 1 > > > > ==================================== > Ben Hope > C.U.H.&H. Club Captain > St. John's College > CAMBRIDGE > CB2 1TP > > Tel: - Home: (01223) 578417 > - Office: (01223) 760312 > - Mobile: 07742 517432 > > www.cam.ac.uk/societies/cuhh > ==================================== > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail From Alain.Allouche at up.univ-mrs.fr Fri Feb 4 13:19:46 2005 From: Alain.Allouche at up.univ-mrs.fr (Alain Allouche) Date: Fri, 04 Feb 2005 13:19:46 +0100 Subject: [Pw_forum] cp.x Message-ID: <42036862.5060104@up.univ-mrs.fr> Dear cp.x_users, Does somebody uses cp.x on IBM SP4 under AIX system ? Apparently some subroutines do not pass correctly their argument (eigs for example). Thanks -------------- next part -------------- A non-text attachment was scrubbed... Name: Alain.Allouche.vcf Type: text/x-vcard Size: 469 bytes Desc: not available Url : /pipermail/attachments/20050204/9938b5f1/attachment.vcf From fsoyalp at yyu.edu.tr Fri Feb 4 22:15:25 2005 From: fsoyalp at yyu.edu.tr (Fethi SOYALP) Date: Fri, 4 Feb 2005 23:15:25 +0200 (EET) Subject: [Pw_forum] LO-TO splitting In-Reply-To: <20050104205001.28481.70587.Mailman@democritos.sissa.it> References: <20050104205001.28481.70587.Mailman@democritos.sissa.it> Message-ID: <1137.194.27.35.27.1107551725.squirrel@194.27.35.27> > You guess? The acoustic mode is 3-fold degenerate. The other three =20 > modes are split in a singlet (LO) plus a doublet (TO). Why LT and TO =20 > modes are split would require some more explanation, and I would like =20= > > to urge you to consult any text-book in solid-state physics to =20 > understant why. If after studying your favorite text you still do not =20= > > understand, please revert to us and we will give further help. Dear Stefano, Thank you very much for your help about LO-TO splitting in example06. Now I calculate vibration modes for a CaF2 structure.I am confused because there isn?t degenerate some direcion, for egzample (110). How can I determine which modes are LO-TO and which modes are LA-TA when degeneracy disapear. according to the output file, first two and last five eigenvectors are below. for first three eigenvectors, because of the small modes, I can say first three modes are acoustic, last six modes are optic .Near the X point (last four eigenvectors) acoustic and optic branches are crossing. I can?t determine acouistic and optic modes from their greatnes. please gave me some comments. I have need your help. q = 0.0250 0.0250 0.0000 ************************************************************************** omega( 1) = 0.512971 [THz] = 17.110993 [cm-1] ( -0.406694 0.025294 0.406694 -0.025294 0.000000 0.000000 ) ( -0.404596 0.057287 0.404596 -0.057287 0.000000 0.000000 ) ( -0.408577 -0.006711 0.408577 0.006711 0.000000 0.000000 ) omega( 2) = 0.542446 [THz] = 18.094187 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.574443 0.045210 ) ( -0.002253 -0.014223 -0.002253 -0.014223 -0.570445 0.090350 ) ( 0.000000 0.014401 0.000000 0.014401 -0.577556 0.000000 ) omega( 3) = 0.671957 [THz] = 22.414210 [cm-1] ( 0.031989 0.406461 0.031989 0.406461 0.000000 0.000000 ) ( 0.063866 0.403232 0.063866 0.403232 -0.020143 0.003190 ) ( 0.000000 0.408258 0.000000 0.408258 0.020394 0.000000 ) omega( 4) = 9.419107 [THz] = 314.189680 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.129018 0.010154 ) ( -0.075525 -0.476846 -0.075525 -0.476846 0.157578 -0.024958 ) ( 0.000000 0.482790 0.000000 0.482790 0.159542 0.000000 ) omega( 5) = 9.420889 [THz] = 314.249133 [cm-1] ( -0.007168 -0.091078 -0.007168 -0.091078 0.000000 0.000000 ) ( 0.017643 0.111392 0.017643 0.111392 0.674393 -0.106813 ) ( 0.000000 0.112780 0.000000 0.112780 -0.682799 0.000000 ) omega( 6) = 9.448169 [THz] = 315.159105 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.097607 0.490380 -0.097607 -0.490380 0.000000 0.000000 ) ( -0.019693 -0.499612 0.019693 0.499612 0.000000 0.000000 ) omega( 7) = 9.751590 [THz] = 325.280207 [cm-1] ( 0.246460 0.309902 -0.246460 -0.309902 0.000000 0.000000 ) ( -0.282494 -0.302995 0.282494 0.302995 0.000000 0.000000 ) ( -0.231618 -0.343456 0.231618 0.343456 0.000000 0.000000 ) omega( 8) = 9.773977 [THz] = 326.026956 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.531826 -0.041856 ) ( -0.025955 -0.163873 -0.025955 -0.163873 -0.543362 0.086060 ) ( 0.000000 0.165916 0.000000 0.165916 -0.550135 0.000000 ) omega( 9) = 9.775480 [THz] = 326.077074 [cm-1] ( -0.029580 -0.375850 -0.029580 -0.375850 0.000000 0.000000 ) ( 0.060880 0.384382 0.060880 0.384382 -0.231528 0.036670 ) ( 0.000000 0.389173 0.000000 0.389173 0.234414 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 17.111000 TA (because all atoms vibrate same direction) 18.094200 LA (L because the vibrational eigenvector has non-vanishing z component; the wavevector is directed along xy, hence the mode is transverse) 22.414200 ?A 314.189697 O (O because small atoms vibrations are bigger than small atom) 314.249115 ?O 315.159088 TO 325.280212 TO 326.027008 ?O 326.077087 ?O q = 0.0500 0.0500 0.0000 ************************************************************************** omega( 1) = 0.622850 [THz] = 20.776190 [cm-1] ( 0.405477 -0.048461 -0.405477 0.048461 0.000000 0.000000 ) ( 0.392739 -0.111248 -0.392739 0.111248 0.000000 0.000000 ) ( 0.407894 0.015560 -0.407894 -0.015560 0.000000 0.000000 ) omega( 2) = 0.715136 [THz] = 23.854512 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.570236 0.090317 ) ( -0.008890 -0.027362 -0.008890 -0.027362 -0.547730 0.177968 ) ( 0.000000 0.028770 0.000000 0.028770 -0.575918 0.000000 ) omega( 3) = 1.066602 [THz] = 35.578241 [cm-1] ( -0.064032 -0.404285 -0.064032 -0.404285 0.000000 0.000000 ) ( -0.125672 -0.386777 -0.125672 -0.386777 0.038912 -0.012643 ) ( 0.000000 -0.406682 0.000000 -0.406682 -0.040915 0.000000 ) omega( 4) = 9.355051 [THz] = 312.052975 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.204219 -0.032345 ) ( 0.139849 0.430410 0.139849 0.430410 -0.249827 0.081174 ) ( 0.000000 -0.452560 0.000000 -0.452560 -0.262684 0.000000 ) omega( 5) = 9.362934 [THz] = 312.315930 [cm-1] ( 0.022737 0.143558 0.022737 0.143558 0.000000 0.000000 ) ( -0.057382 -0.176604 -0.057382 -0.176604 -0.608907 0.197846 ) ( 0.000000 -0.185692 0.000000 -0.185692 0.640243 0.000000 ) omega( 6) = 9.454900 [THz] = 315.383604 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.068326 0.495310 -0.068326 -0.495310 0.000000 0.000000 ) ( 0.088078 -0.492181 -0.088078 0.492181 0.000000 0.000000 ) omega( 7) = 9.758643 [THz] = 325.515459 [cm-1] ( 0.101694 0.381763 -0.101694 -0.381763 0.000000 0.000000 ) ( -0.168110 -0.379074 0.168110 0.379074 0.000000 0.000000 ) ( -0.042742 -0.412469 0.042742 0.412469 0.000000 0.000000 ) omega( 8) = 9.831911 [THz] = 327.959444 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.486764 -0.077096 ) ( -0.080649 -0.248212 -0.080649 -0.248212 -0.468179 0.152121 ) ( 0.000000 0.260985 0.000000 0.260985 -0.492272 0.000000 ) omega( 9) = 9.837634 [THz] = 328.150317 [cm-1] ( -0.054427 -0.343638 -0.054427 -0.343638 0.000000 0.000000 ) ( 0.107811 0.331810 0.107811 0.331810 -0.350098 0.113754 ) ( 0.000000 0.348885 0.000000 0.348885 0.368115 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 20.776199 23.854500 35.578201 312.053009 312.315887 315.383606 325.515503 327.959412 328.150299 last five eigenvector. q = 0.9000 0.9000 0.0000 ************************************************************************** omega( 1) = 5.112475 [THz] = 170.534951 [cm-1] ( 0.453049 -0.311934 -0.453049 0.311934 0.000000 0.000000 ) ( -0.301186 0.089491 0.301186 -0.089491 0.000000 0.000000 ) ( -0.191063 0.249433 0.191063 -0.249433 0.000000 0.000000 ) omega( 2) = 5.481749 [THz] = 182.852665 [cm-1] ( -0.172532 0.530999 -0.172532 0.530999 0.000000 0.000000 ) ( -0.175797 0.241963 -0.175797 0.241963 0.078318 0.056902 ) ( 0.000000 0.299083 0.000000 0.299083 -0.096807 0.000000 ) omega( 3) = 7.581087 [THz] = 252.879509 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.078334 0.025452 ) ( 0.093007 -0.128014 0.093007 -0.128014 0.540610 0.392776 ) ( 0.000000 0.158234 0.000000 0.158234 0.668231 0.000000 ) omega( 4) = 7.745696 [THz] = 258.370336 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.493523 -0.160355 ) ( -0.228501 0.314505 -0.228501 0.314505 0.203237 0.147660 ) ( 0.000000 -0.388749 0.000000 -0.388749 0.251214 0.000000 ) omega( 5) = 7.974611 [THz] = 266.006149 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.360316 -0.346659 0.360316 0.346659 0.000000 0.000000 ) ( 0.495263 0.068665 -0.495263 -0.068665 0.000000 0.000000 ) omega( 6) = 8.411441 [THz] = 280.577334 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.576853 -0.187431 ) ( 0.229436 -0.315791 0.229436 -0.315791 -0.086098 -0.062554 ) ( 0.000000 0.390339 0.000000 0.390339 -0.106424 0.000000 ) omega( 7) = 10.869387 [THz] = 362.566124 [cm-1] ( -0.022070 0.067924 -0.022070 0.067924 0.000000 0.000000 ) ( 0.140863 -0.193881 0.140863 -0.193881 0.498735 0.362352 ) ( 0.000000 -0.239650 0.000000 -0.239650 -0.616470 0.000000 ) omega( 8) = 11.043721 [THz] = 368.381318 [cm-1] ( 0.123427 -0.222049 -0.123427 0.222049 0.000000 0.000000 ) ( 0.341637 -0.317828 -0.341637 0.317828 0.000000 0.000000 ) ( 0.089575 -0.457937 -0.089575 0.457937 0.000000 0.000000 ) omega( 9) = 11.085726 [THz] = 369.782482 [cm-1] ( -0.071758 0.220847 -0.071758 0.220847 0.000000 0.000000 ) ( 0.232682 -0.320260 0.232682 -0.320260 -0.294668 -0.214089 ) ( 0.000000 -0.395863 0.000000 -0.395863 0.364229 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 170.535004 182.852707 252.879501 258.370300 266.006104 280.577301 362.566101 368.381287 369.782501 q = 0.9250 0.9250 0.0000 ************************************************************************** omega( 1) = 5.167330 [THz] = 172.364732 [cm-1] ( 0.442339 -0.329097 -0.442339 0.329097 0.000000 0.000000 ) ( -0.287867 0.123077 0.287867 -0.123077 0.000000 0.000000 ) ( -0.200616 0.240351 0.200616 -0.240351 0.000000 0.000000 ) omega( 2) = 5.378558 [THz] = 179.410579 [cm-1] ( -0.129809 0.540692 -0.129809 0.540692 0.000000 0.000000 ) ( -0.138284 0.271397 -0.138284 0.271397 0.064520 0.032874 ) ( 0.000000 0.304596 0.000000 0.304596 -0.072412 0.000000 ) omega( 3) = 7.584358 [THz] = 252.988617 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.042364 -0.010171 ) ( -0.054594 0.107147 -0.054594 0.107147 -0.610927 -0.311283 ) ( 0.000000 -0.120254 0.000000 -0.120254 -0.685660 0.000000 ) omega( 4) = 7.808341 [THz] = 260.459944 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.440308 -0.105709 ) ( -0.193961 0.380669 -0.193961 0.380669 0.160424 0.081740 ) ( 0.000000 -0.427235 0.000000 -0.427235 0.180048 0.000000 ) omega( 5) = 7.946584 [THz] = 265.071270 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.405802 0.292104 0.405802 -0.292104 0.000000 0.000000 ) ( 0.228960 -0.444497 -0.228960 0.444497 0.000000 0.000000 ) omega( 6) = 8.319704 [THz] = 277.517305 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.662283 0.159000 ) ( -0.164136 0.322134 -0.164136 0.322134 0.072537 0.036959 ) ( 0.000000 -0.361540 0.000000 -0.361540 0.081410 0.000000 ) omega( 7) = 10.871653 [THz] = 362.641727 [cm-1] ( -0.013202 0.054992 -0.013202 0.054992 0.000000 0.000000 ) ( 0.084463 -0.165768 0.084463 -0.165768 0.582623 0.296861 ) ( 0.000000 -0.186045 0.000000 -0.186045 -0.653893 0.000000 ) omega( 8) = 11.050990 [THz] = 368.623808 [cm-1] ( 0.097862 -0.233025 -0.097862 0.233025 0.000000 0.000000 ) ( 0.276325 -0.376438 -0.276325 0.376438 0.000000 0.000000 ) ( 0.075308 -0.460858 -0.075308 0.460858 0.000000 0.000000 ) omega( 9) = 11.075187 [THz] = 369.430936 [cm-1] ( -0.055930 0.232966 -0.055930 0.232966 0.000000 0.000000 ) ( 0.192932 -0.378651 0.192932 -0.378651 -0.254209 -0.129526 ) ( 0.000000 -0.424970 0.000000 -0.424970 0.285306 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 172.364700 179.410599 252.988602 260.459900 265.071289 277.517303 362.641693 368.623810 369.430908 q = 0.9500 0.9500 0.0000 ************************************************************************** omega( 1) = 5.207227 [THz] = 173.695562 [cm-1] ( 0.393952 -0.387044 -0.393952 0.387044 0.000000 0.000000 ) ( -0.254228 0.181294 0.254228 -0.181294 0.000000 0.000000 ) ( -0.185762 0.250982 0.185762 -0.250982 0.000000 0.000000 ) omega( 2) = 5.302235 [THz] = 176.864690 [cm-1] ( 0.086726 -0.547564 0.086726 -0.547564 0.000000 0.000000 ) ( 0.095328 -0.293390 0.095328 -0.293390 -0.045832 -0.014892 ) ( 0.000000 -0.308488 0.000000 -0.308488 0.048190 0.000000 ) omega( 3) = 7.587883 [THz] = 253.106224 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.019256 -0.003050 ) ( -0.025701 0.079099 -0.025701 0.079099 -0.663000 -0.215422 ) ( 0.000000 -0.083170 0.000000 -0.083170 -0.697119 0.000000 ) omega( 4) = 7.859171 [THz] = 262.155459 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.345835 -0.054775 ) ( -0.142325 0.438032 -0.142325 0.438032 0.114290 0.037135 ) ( 0.000000 -0.460575 0.000000 -0.460575 0.120172 0.000000 ) omega( 5) = 7.926161 [THz] = 264.390041 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.264399 -0.424374 -0.264399 0.424374 0.000000 0.000000 ) ( -0.120319 0.485307 0.120319 -0.485307 0.000000 0.000000 ) omega( 6) = 8.244308 [THz] = 275.002344 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.776467 0.122980 ) ( -0.094861 0.291952 -0.094861 0.291952 0.047725 0.015507 ) ( 0.000000 -0.306977 0.000000 -0.306977 0.050181 0.000000 ) omega( 7) = 10.872635 [THz] = 362.674492 [cm-1] ( -0.006135 0.038735 -0.006135 0.038735 0.000000 0.000000 ) ( 0.039328 -0.121039 0.039328 -0.121039 0.649278 0.210963 ) ( 0.000000 -0.127268 0.000000 -0.127268 -0.682692 0.000000 ) omega( 8) = 11.055931 [THz] = 368.788605 [cm-1] ( 0.067935 -0.242448 -0.067935 0.242448 0.000000 0.000000 ) ( 0.194892 -0.424640 -0.194892 0.424640 0.000000 0.000000 ) ( 0.054132 -0.464082 -0.054132 0.464082 0.000000 0.000000 ) omega( 9) = 11.066863 [THz] = 369.153260 [cm-1] ( -0.038426 0.242610 -0.038426 0.242610 0.000000 0.000000 ) ( 0.138374 -0.425871 0.138374 -0.425871 -0.186942 -0.060741 ) ( 0.000000 -0.447787 0.000000 -0.447787 0.196562 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 173.695602 176.864700 253.106201 262.155487 264.390015 275.002289 362.674500 368.788605 369.153290 q = 0.9750 0.9750 0.0000 ************************************************************************** omega( 1) = 5.231455 [THz] = 174.503706 [cm-1] ( 0.243465 -0.496342 -0.243465 0.496342 0.000000 0.000000 ) ( -0.158826 0.268252 0.158826 -0.268252 0.000000 0.000000 ) ( -0.114907 0.289795 0.114907 -0.289795 0.000000 0.000000 ) omega( 2) = 5.255377 [THz] = 175.301680 [cm-1] ( 0.043417 -0.551666 0.043417 -0.551666 0.000000 0.000000 ) ( 0.048621 -0.306979 0.048621 -0.306979 -0.023775 -0.003766 ) ( 0.000000 -0.310806 0.000000 -0.310806 0.024071 0.000000 ) omega( 3) = 7.590739 [THz] = 253.201497 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.004963 0.000391 ) ( 0.006713 -0.042387 0.006713 -0.042387 0.695815 0.110206 ) ( 0.000000 0.042915 0.000000 0.042915 0.704489 0.000000 ) omega( 4) = 7.895472 [THz] = 263.366337 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.198742 -0.015641 ) ( -0.076350 0.482056 -0.076350 0.482056 0.060190 0.009533 ) ( 0.000000 -0.488065 0.000000 -0.488065 0.060941 0.000000 ) omega( 5) = 7.913739 [THz] = 263.975668 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.131005 -0.482533 -0.131005 0.482533 0.000000 0.000000 ) ( -0.053908 0.497085 0.053908 -0.497085 0.000000 0.000000 ) omega( 6) = 8.191441 [THz] = 273.238867 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 -0.916819 -0.072155 ) ( 0.030660 -0.193581 0.030660 -0.193581 -0.016851 -0.002669 ) ( 0.000000 0.195994 0.000000 0.195994 -0.017061 0.000000 ) omega( 7) = 10.872943 [THz] = 362.684763 [cm-1] ( -0.001575 0.020019 -0.001575 0.020019 0.000000 0.000000 ) ( 0.010114 -0.063855 0.010114 -0.063855 0.692254 0.109642 ) ( 0.000000 -0.064651 0.000000 -0.064651 -0.700883 0.000000 ) omega( 8) = 11.058790 [THz] = 368.883976 [cm-1] ( -0.036765 0.248500 0.036765 -0.248500 0.000000 0.000000 ) ( -0.104468 0.455560 0.104468 -0.455560 0.000000 0.000000 ) ( -0.031917 0.466293 0.031917 -0.466293 0.000000 0.000000 ) omega( 9) = 11.061549 [THz] = 368.976011 [cm-1] ( -0.019585 0.248848 -0.019585 0.248848 0.000000 0.000000 ) ( 0.072335 -0.456708 0.072335 -0.456708 -0.099078 -0.015692 ) ( 0.000000 -0.462400 0.000000 -0.462400 0.100313 0.000000 ) ************************************************************************** diagonalizing the dynamical matrix ... 174.503693 175.301697 253.201508 263.366302 263.975708 273.238892 362.684814 368.884003 368.976013 q = 1.0000 1.0000 0.0000 ************************************************************************** omega( 1) = 5.239579 [THz] = 174.774706 [cm-1] ( 0.104210 -0.002667 -0.774879 0.019464 0.000000 0.000000 ) ( -0.436565 0.010966 0.058711 -0.001502 0.000000 0.000000 ) ( -0.436565 0.010966 0.058711 -0.001502 0.000000 0.000000 ) omega( 2) = 5.239579 [THz] = 174.774706 [cm-1] ( -0.774540 -0.030066 -0.104163 -0.004093 0.000000 0.000000 ) ( -0.058685 -0.002306 -0.436374 -0.016939 0.000000 0.000000 ) ( -0.058685 -0.002306 -0.436374 -0.016939 0.000000 0.000000 ) omega( 3) = 7.591846 [THz] = 253.238404 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.707107 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.707107 0.000000 ) omega( 4) = 7.909569 [THz] = 263.836578 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.340127 -0.006494 0.619793 0.011327 0.000000 0.000000 ) ( 0.340127 0.006494 -0.619793 -0.011327 0.000000 0.000000 ) omega( 5) = 7.909569 [THz] = 263.836578 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.616368 -0.066041 0.338223 -0.036518 0.000000 0.000000 ) ( -0.616368 0.066041 -0.338223 0.036518 0.000000 0.000000 ) omega( 6) = 8.171481 [THz] = 272.573075 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 1.000000 0.000296 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) omega( 7) = 10.872996 [THz] = 362.686515 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.707107 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.707107 0.000000 ) omega( 8) = 11.059725 [THz] = 368.915167 [cm-1] ( -0.150757 -0.065210 0.289515 0.123338 0.000000 0.000000 ) ( -0.539143 -0.229683 0.280745 0.121436 0.000000 0.000000 ) ( -0.539143 -0.229683 0.280745 0.121436 0.000000 0.000000 ) omega( 9) = 11.059725 [THz] = 368.915167 [cm-1] ( 0.306189 -0.072661 0.159606 -0.038807 0.000000 0.000000 ) ( -0.297223 0.072268 -0.570194 0.135311 0.000000 0.000000 ) ( -0.297223 0.072268 -0.570194 0.135311 0.000000 0.000000 ) ************************************************************************** 174.774704 174.774704 253.238403 263.836609 263.836609 272.573090 362.686493 368.915192 368.915192 Best Regarts Fethi Soyalp From aouahab at yahoo.fr Sun Feb 6 10:46:29 2005 From: aouahab at yahoo.fr (Abdelouahab O.) Date: Sun, 6 Feb 2005 10:46:29 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster Message-ID: <20050206094629.10964.qmail@web26403.mail.ukl.yahoo.com> Hello ALL I have a pc linux cluster with ifort8.0, mkl7.2, and LAM7.0.6: the parallel compilation cant be finilized usin g the -D__LAM -D__PARA, the error apppers when compiling broadcast.f90, when I remove -D__PARA, no erros but the run is not performed in parallel, and the lam says that it seems that the programm runnig under lam does no call mpi_init or (some thing like that). What is the solution to this problem? thanks A. OUAHAB Universit? de OUARGLA Algeria D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ From giannozz at nest.sns.it Sun Feb 6 15:54:54 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sun, 6 Feb 2005 15:54:54 +0100 Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: <20050206094629.10964.qmail@web26403.mail.ukl.yahoo.com> References: <20050206094629.10964.qmail@web26403.mail.ukl.yahoo.com> Message-ID: <200502061554.55068.giannozz@nest.sns.it> On Sunday 06 February 2005 10:46, Abdelouahab O. wrote: > I have a pc linux cluster with ifort8.0, mkl7.2, and > LAM7.0.6: the parallel compilation cant be finilized > using the -D__LAM -D__PARA, the error apppears when > compiling broadcast.f90 which error? From degironc at sissa.it Mon Feb 7 09:39:20 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 07 Feb 2005 09:39:20 +0100 Subject: [Pw_forum] LO-TO splitting References: <20050104205001.28481.70587.Mailman@democritos.sissa.it> <1137.194.27.35.27.1107551725.squirrel@194.27.35.27> Message-ID: <42072938.6080100@sissa.it> Fethi SOYALP wrote: > >Thank you very much for your help about LO-TO splitting in example06. >Now I calculate vibration modes for a CaF2 structure.I am confused >because there isn?t degenerate some direcion, for egzample (110). How can >I determine which modes are LO-TO and which modes are LA-TA when >degeneracy disapear. according to the output file, first two and last five >eigenvectors are below. for first three eigenvectors, because of the >small modes, I can say first three modes are acoustic, last six modes are >optic .Near the X point (last four eigenvectors) acoustic and optic >branches are crossing. I can?t determine acouistic and optic modes from >their greatnes. please gave me some comments. I have need your help. > away from high symmetry lines there may not be purely longitudinal and purely transverse modes, but all modes have mixed character. In any case the DEFINITION of a longitudinal modes is that its eigenvector is parallel to the phonon vewavector, the DEFINITION of a transverse mode is that the eigenvector is orthogonal to it. Apply the definitions. best regards, Stefano de Gironcoli From Myeong.Lee at asu.edu Mon Feb 7 22:49:45 2005 From: Myeong.Lee at asu.edu (Myeong.Lee at asu.edu) Date: Mon, 07 Feb 2005 14:49:45 -0700 (MST) Subject: [Pw_forum] imagianry k-values Message-ID: <1107812985.4207e2794a20a@webmail.asu.edu> Dear users, For the complex band structure, I got only negative imaginary k-values for alkane, but for alkene I also got positive values as well as negative k-values. (file name: bands.alkane.im, bands.alkene.im) So, I wonder if imaginary k-values can be positive in case that the band edge appears at k=pi/a like alkene or if I've got the wrong result. Thank you. From aouahab at yahoo.fr Tue Feb 8 14:52:05 2005 From: aouahab at yahoo.fr (Abdelouahab O.) Date: Tue, 8 Feb 2005 14:52:05 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: <200502061554.55068.giannozz@nest.sns.it> Message-ID: <20050208135205.84691.qmail@web26403.mail.ukl.yahoo.com> Hello Thanks for your relpay; This is make.sys ======================================================= # make.sys. Generated from make.sys.in by configure. CC = gcc CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) $(IFLAGS) # See include/machine.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW IFLAGS = -I. -I../include -I../Modules -I../PW -I../PH CPP = gcc -E CPPFLAGS = $(DFLAGS) $(IFLAGS) F77 = ifort F90 = ifort FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) F77FLAGS = $(FFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(DFLAGS) F77FLAGS_NOOPT = -O0 LD = ifort LDFLAGS = -Vaxlib -O3 -tpp7 $(LIBOBJS) $(LIBS) LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a # LIBS must contain the location of all needed external libraries LIBS = -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam # MYLIB can be one of the following (depending on LIBS): # blas : compile the local copy of blas routines # lapack : compile the local copy of lapack routines # blas_and_lapack : all of the above - use this for a quick test # or if you don't have an optimized blas/lapack library # lapack_ibm : compile only lapack routines not present in IBM ESSL # use this together with IBM ESSL # lapack_t3e : compile only lapack routines not present in T3E scilib # use this together with T3E scilib # lapack_mkl : compile only lapack routines not present in Intel MKL # use this together with Intel MKL MYLIB = lapack_mkl AR = ar ARFLAGS = ruv ======================================================= And make all gives: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% make[1]: Entering directory `/home1/wahab/pwscf2.0.4/pwscf/Modules' ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c io_global.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c kind.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c parallel_include.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c shmem_include.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c mp_global.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c mp.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c recvec.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c stick_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c berry_phase.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c basic_algebra_routines.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c constants.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c parameters.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c io_files.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c bfgs_module.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c cell_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c clocks.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c control_flags.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c parallel_types.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c descriptors.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c electrons_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c energies.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c fft_types.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c mp_buffers.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c fft_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c fft_scalar.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c formats.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c griddim.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c input_parameters.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c mp_wave.f90 fortcom: Warning: mp_wave.f90, line 130: A dummy argument with an expl icit INTENT(OUT) declaration is not given an explicit value. [PW] SUBROUTINE splitwf ( pw, pwt, ngwl, ig_l2g, mpime, nproc, root, comm ) ...........................^ fortcom: Warning: mp_wave.f90, line 23: A dummy argument with an expli cit INTENT(OUT) declaration is not given an explicit value. [PWT] SUBROUTINE mergewf ( pw, pwt, ngwl, ig_l2g, mpime, nproc, root, comm ) ...............................^ fortcom: Warning: mp_wave.f90, line 314: A dummy argument with an expl icit INTENT(OUT) declaration is not given an explicit value. [RHO] SUBROUTINE splitrho(rho, rhot, ngl, ig_l2g, mpime, nproc, root) ..........................^ fortcom: Warning: mp_wave.f90, line 226: A dummy argument with an expl icit INTENT(OUT) declaration is not given an explicit value. [RHOT] SUBROUTINE mergerho(rho, rhot, ngl, ig_l2g, mpime, nproc, root) ...............................^ fortcom: Warning: mp_wave.f90, line 490: A dummy argument with an expl icit INTENT(OUT) declaration is not given an explicit value. [IGL] SUBROUTINE splitig(igl, igtot, ngl, mpime, nproc, root, comm) .........................^ fortcom: Warning: mp_wave.f90, line 396: A dummy argument with an expl icit INTENT(OUT) declaration is not given an explicit value. [IGTOT] SUBROUTINE mergeig(igl, igtot, ngl, mpime, nproc, root, comm) ..............................^ ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c io_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c ions_base.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c parser.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c pseudodata.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c pseudo_types.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c ptoolkit.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c read_cards.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c read_namelists.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c readpseudo.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c smallbox.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c splinelib.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c version.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c wavefunctions.f90 make[1]: Leaving directory `/home1/wahab/pwscf2.0.4/pwscf/Modules' ( cd clib; make all ) make[1]: Entering directory `/home1/wahab/pwscf2.0.4/pwscf/clib' gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c c_mkdir.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c c_today.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c cpflush.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c cptimer.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c factor.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c fft_stick.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c indici.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c memstat.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c qsort.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c readocc.c gcc -O3 -fomit-frame-pointer -D__LINUX -D__INTEL -D__LAM -D__PARA -DHA S_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -I. -I../include -I../Modules -I../PW -I../PH -c round2.c ar ruv clib.a c_mkdir.o c_today.o cpflush.o cptimer.o factor.o fft_sti ck.o indici.o memstat.o qsort.o readocc.o round2.o ar: creating clib.a a - c_mkdir.o a - c_today.o a - cpflush.o a - cptimer.o a - factor.o a - fft_stick.o a - indici.o a - memstat.o a - qsort.o a - readocc.o a - round2.o make[1]: Leaving directory `/home1/wahab/pwscf2.0.4/pwscf/clib' ( cd flib; make all ) make[1]: Entering directory `/home1/wahab/pwscf2.0.4/pwscf/flib' ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c avrec.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c erf.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c iceil.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c iglocal.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c inv3.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c latgen.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c ngnr_set.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c recips.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c scnds.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c simpsn.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c sort.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c sph_bes.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c transto.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c volume.f90 ar ruv flib.a avrec.o erf.o iceil.o iglocal.o inv3.o latgen.o ngnr_set .o recips.o scnds.o simpsn.o sort.o sph_bes.o transto.o volume.o ar: creating flib.a a - avrec.o a - erf.o a - iceil.o a - iglocal.o a - inv3.o a - latgen.o a - ngnr_set.o a - recips.o a - scnds.o a - simpsn.o a - sort.o a - sph_bes.o a - transto.o a - volume.o ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c blockset.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c gridsetup.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c hangup.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c ilcm.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c indxg2l.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c indxg2p.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c indxl2g.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c infog1l.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c infog2l.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c localdim.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c localindex.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c matmul.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c npreroc.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c numroc.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c ownerof.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c reduce.f90 ar ruv ptools.a blockset.o gridsetup.o hangup.o ilcm.o indxg2l.o indxg 2p.o indxl2g.o infog1l.o infog2l.o localdim.o localindex.o matmul.o np reroc.o numroc.o ownerof.o reduce.o ar: creating ptools.a a - blockset.o a - gridsetup.o a - hangup.o a - ilcm.o a - indxg2l.o a - indxg2p.o a - indxl2g.o a - infog1l.o a - infog2l.o a - localdim.o a - localindex.o a - matmul.o a - npreroc.o a - numroc.o a - ownerof.o a - reduce.o ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c eispack.f90 touch eispack ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c lapack_mkl.f ar ruv flib.a lapack_mkl.o a - lapack_mkl.o touch lapack_mkl make[1]: Leaving directory `/home1/wahab/pwscf2.0.4/pwscf/flib' if test -d PW ; then ( cd PW ; make all ) ; fi make[1]: Entering directory `/home1/wahab/pwscf2.0.4/pwscf/PW' ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c aainit.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c para.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c pwcom.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c add_efield.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c becmod.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c rbecmod.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c add_vuspsi.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c addusdens.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c addusforce.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c addusstress.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c allocate_fft.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c allocate_locpot.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c allocate_nlpot.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c allocate_wfc.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c atomic_rho.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c atomic_wfc.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c bachel.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__ FFTW -D__USE_INTERNAL_FFTW -c bfgs.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_bess.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c bp_c_phase.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c bp_calc_btq.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_dbess.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c bp_qvan3.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_radin.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c bp_strings.f90 ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_ylm_q.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_zgedi.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -c bp_zgefa.f ifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW -c broadcast.f90 fortcom: Error: broadcast.f90, line 146: This name has already been assigned a data type. [MPI_DOUBLE_PRECISION] INTEGER MPI_REAL4, MPI_DOUBLE_PRECISION, MPI_REAL16 .........................^ compilation aborted for broadcast.f90 (code 1) make[1]: *** [broadcast.o] Error 1 make[1]: Leaving directory `/home1/wahab/pwscf2.0.4/pwscf/PW' make: *** [pw] Error 2 --- Paolo Giannozzi a ?crit : > On Sunday 06 February 2005 10:46, Abdelouahab O. > wrote: > > > I have a pc linux cluster with ifort8.0, mkl7.2, > and > > LAM7.0.6: the parallel compilation cant be > finilized > > using the -D__LAM -D__PARA, the error apppears > when > > compiling broadcast.f90 > > which error? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Tue Feb 8 17:24:33 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Tue, 8 Feb 2005 17:24:33 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: <20050208135205.84691.qmail@web26403.mail.ukl.yahoo.com> Message-ID: On Tue, 8 Feb 2005, Abdelouahab O. wrote: AO> Hello AO> Thanks for your relpay; AO> This is make.sys hi, the current version of pwscf (or Quantum-ESPRESSO) is 2.1.2. with many new features, bugfixes and improvements. if you have installed you LAM/MPI properly for the intel compilers you should be able to just use CC=mpicc F77=mpif77 F90=mpif90 alternatively you can try to use CC=env LAMHCC=icc mpicc F77=env LAMHF77=ifort mpif77 F90=env LAMHF77=ifort mpif77 using the mpiXXX wrappers will make sure, that you include the proper headers that are matching your LAM installation. AO> ======================================================= AO> # make.sys. Generated from make.sys.in by configure. AO> CC = gcc AO> CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) AO> $(IFLAGS) AO> # See include/machine.h.README for a list of AO> precompilation options AO> # (possible arguments to -D or -U) and their meaning AO> DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA AO> -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW AO> IFLAGS = -I. -I../include -I../Modules -I../PW AO> -I../PH AO> CPP = gcc -E AO> CPPFLAGS = $(DFLAGS) $(IFLAGS) AO> F77 = ifort AO> F90 = ifort AO> FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) AO> F77FLAGS = $(FFLAGS) AO> F90FLAGS = $(FFLAGS) -nomodule -fpp $(DFLAGS) AO> F77FLAGS_NOOPT = -O0 AO> LD = ifort AO> LDFLAGS = -Vaxlib -O3 -tpp7 $(LIBOBJS) $(LIBS) AO> LIBOBJS = ../flib/ptools.a ../flib/flib.a AO> ../clib/clib.a AO> # LIBS must contain the location of all needed AO> external libraries AO> LIBS = -L/opt/intel/mkl72/lib/32 AO> -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam AO> # MYLIB can be one of the following (depending on AO> LIBS): AO> # blas : compile the local copy of blas routines AO> # lapack : compile the local copy of lapack AO> routines AO> # blas_and_lapack : all of the above - use this for a AO> quick test AO> # or if you don't have an optimized AO> blas/lapack library AO> # lapack_ibm : compile only lapack routines not AO> present in IBM ESSL AO> # use this together with IBM ESSL AO> # lapack_t3e : compile only lapack routines not AO> present in T3E scilib AO> # use this together with T3E scilib AO> # lapack_mkl : compile only lapack routines not AO> present in Intel MKL AO> # use this together with Intel MKL AO> MYLIB = lapack_mkl AO> AR = ar AO> ARFLAGS = ruv AO> ======================================================= AO> AO> And make all gives: AO> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% AO> make[1]: Entering directory AO> `/home1/wahab/pwscf2.0.4/pwscf/Modules' [...] AO> fortcom: Warning: mp_wave.f90, line 130: A dummy AO> argument with an expl AO> icit INTENT(OUT) declaration is not given an explicit AO> value. [PW] AO> SUBROUTINE splitwf ( pw, pwt, ngwl, ig_l2g, AO> mpime, nproc, root, AO> comm ) AO> ...........................^ AO> fortcom: Warning: mp_wave.f90, line 23: A dummy AO> argument with an expli this looks like a 'feature', that only the recent intel fortran (and the DEC/Compaq) compilers choke on. this may have been fixed in the newer releases. i remember to have seen several of those with version 2.1 which are now fixed. AO> -D__USE_INTERNAL_FFTW -c broadcast.f90 AO> fortcom: Error: broadcast.f90, line 146: This name has AO> already been assigned a data type. AO> [MPI_DOUBLE_PRECISION] AO> INTEGER MPI_REAL4, MPI_DOUBLE_PRECISION, AO> MPI_REAL16 AO> .........................^ AO> compilation aborted for broadcast.f90 (code 1) AO> make[1]: *** [broadcast.o] Error 1 AO> make[1]: Leaving directory AO> `/home1/wahab/pwscf2.0.4/pwscf/PW' AO> make: *** [pw] Error 2 i cannot reproduce this on my desktop with pwscf v2.0.4, my installation of LAM/MPI version 7.0.6 and the intel compiler version 8.1. in my distribution broadcast.f90 only has 41 lines, so your error on line 146 looks very suspicious. same is for the content, 'MPI_REAL4' is only present in MPICH installations and not in LAM/MPI. if you have administrator priviledge on your machine, you can download my LAM/MPI package as rpm from: http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-linux.html#mpi these rpm are compiled special to be compatible to the intel compilers as well as g77 or the pgi compilers. you simply need to use mpiifort/mpiicc instead of mpif77/mpicc. regards, axel. AO> AO> --- Paolo Giannozzi a ?crit : AO> > On Sunday 06 February 2005 10:46, Abdelouahab O. AO> > wrote: AO> > AO> > > I have a pc linux cluster with ifort8.0, mkl7.2, AO> > and AO> > > LAM7.0.6: the parallel compilation cant be AO> > finilized AO> > > using the -D__LAM -D__PARA, the error apppears AO> > when AO> > > compiling broadcast.f90 AO> > AO> > which error? AO> > _______________________________________________ AO> > Pw_forum mailing list AO> > Pw_forum at pwscf.org AO> > http://www.democritos.it/mailman/listinfo/pw_forum AO> > AO> AO> AO> AO> AO> AO> AO> AO> D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! AO> Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ AO> _______________________________________________ AO> Pw_forum mailing list AO> Pw_forum at pwscf.org AO> http://www.democritos.it/mailman/listinfo/pw_forum AO> AO> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From bth20 at cam.ac.uk Wed Feb 9 16:19:25 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Wed, 9 Feb 2005 15:19:25 -0000 Subject: [Pw_forum] efermit error Message-ID: Hi there I am performing a dos calculation on Cobalt. (As previously mentioned) I have done the initial scf calculation. Now when I do the nscf part I get: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% from efermit : error # 1 unexpected error %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% I'm not sure why. My input file is &CONTROL calculation = 'nscf' , restart_mode = 'from_scratch' , pseudo_dir = '/scratch/bth/Programs/Espresso/pseudo/' , prefix = 'Co', verbosity = 'high', / &SYSTEM ibrav = 4, celldm(1) = 4.738, celldm(3) = 1.62, nat = 2, ntyp = 1, ecutwfc = 25 , ecutrho = 100 , nspin = 2, starting_magnetization(1) = 1.0, occupations = 'tetrahedra', / &ELECTRONS mixing_beta = 0.7, conv_thr = 1.0d-7, / &IONS upscale = 10, / &CELL / ATOMIC_SPECIES Co 58.93 Co.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {crystal} Co 0.353333333 0.666666667 0.250000000 1 1 1 Co 0.666666667 0.333333333 0.750000000 1 1 1 K_POINTS {automatic} 12 12 12 0 0 1 Can anyone see the problem? Thanks Ben ==================================== Ben Hope C.U.H.&H. Club Captain St. John's College CAMBRIDGE CB2 1TP Tel: - Home: (01223) 578417 - Office: (01223) 760312 - Mobile: 07742 517432 www.cam.ac.uk/societies/cuhh ==================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050209/5827a344/attachment.htm From alcantar at phys.ksu.edu Thu Feb 10 23:08:08 2005 From: alcantar at phys.ksu.edu (Alcantara Ortigoza, Marisol) Date: Thu, 10 Feb 2005 16:08:08 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Message-ID: Hi, I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using PBE pseudopotential provided by pwscf. First I calculated the lattice constant which turned to be ~ 3.67 A, in good agreement with experiment (3.61, an error less than 2%). Therefore, I proceed to relax the surface and I found these results: > Relaxation of Cu100 with a = 3.67 A > > ATOMIC_POSITIONS (alat) > Cu 0.500000000 0.500000000 -2.022823825 > Cu 0.000000000 0.000000000 -2.022823918 > Cu 0.000000000 0.500000000 -1.489744169 > Cu 0.500000000 0.000000000 -1.489744169 > Cu 0.500000000 0.500000000 -1.000543208 > Cu 0.000000000 0.000000000 -1.000543383 > Cu 0.000000000 0.500000000 -0.511527906 > Cu 0.500000000 0.000000000 -0.511527906 > Cu 0.000000000 0.000000000 0.000000000 > Cu 0.500000000 0.500000000 0.000000000 > Cu 0.500000000 0.000000000 0.511527906 > Cu 0.000000000 0.500000000 0.511527906 > Cu 0.000000000 0.000000000 1.000543383 > Cu 0.500000000 0.500000000 1.000543208 > Cu 0.500000000 0.000000000 1.489744169 > Cu 0.000000000 0.500000000 1.489744169 > Cu 0.000000000 0.000000000 2.022823918 > Cu 0.500000000 0.500000000 2.022823825 > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00000000 0.00000000 > -0.00067738 > atom 2 type 1 force = 0.00000000 0.00000000 > -0.00067707 > atom 3 type 1 force = 0.00000000 0.00000000 > -0.00006941 > atom 4 type 1 force = 0.00000000 0.00000000 > -0.00006941 > atom 5 type 1 force = 0.00000000 0.00000000 > -0.00067574 > atom 6 type 1 force = 0.00000000 0.00000000 > -0.00067565 > atom 7 type 1 force = 0.00000000 0.00000000 > -0.00063682 > atom 8 type 1 force = 0.00000000 0.00000000 > -0.00063682 > atom 9 type 1 force = 0.00000000 0.00000000 > 0.00000000 > atom 10 type 1 force = 0.00000000 0.00000000 > 0.00000000 > atom 11 type 1 force = 0.00000000 0.00000000 > 0.00063682 > atom 12 type 1 force = 0.00000000 0.00000000 > 0.00063682 > atom 13 type 1 force = 0.00000000 0.00000000 > 0.00067565 > atom 14 type 1 force = 0.00000000 0.00000000 > 0.00067574 > atom 15 type 1 force = 0.00000000 0.00000000 > 0.00006941 > atom 16 type 1 force = 0.00000000 0.00000000 > 0.00006941 > atom 17 type 1 force = 0.00000000 0.00000000 > 0.00067707 > atom 18 type 1 force = 0.00000000 0.00000000 > 0.00067738 > > That means that, > > 1st and 2nd layers separated by ~ + 6.62% > completely wrong result!!! > 2nd and 3rd layers approached by ~ - 2.16% > completely wrong result!!! > Which is clearly wrong according to the experiment, which predicts: 1st and 2nd layers approach by ~ -2.4 % 2nd and 3rd layers separate by ~ +1.0 % Therefore, I was advised to calculate carefully the lattice constant using the Murnaghan equation of state and repeat the relaxation. But, this time I had to be more careful and calculate the stress too while relaxing the system. I found a lattice constant just 0.1% larger than the one I had been using before: a = 3.6737 A And these were the results I found after relaxing the system: > Relaxation of Cu100 with a = 3.6737 A (Murnaghan fitting) > > ATOMIC_POSITIONS (alat) > Cu 0.500000000 0.500000000 -1.996326571 > Cu 0.000000000 0.000000000 -1.996326572 > Cu 0.000000000 0.500000000 -1.511684867 > Cu 0.500000000 0.000000000 -1.511684867 > Cu 0.500000000 0.500000000 -1.009845482 > Cu 0.000000000 0.000000000 -1.009845452 > Cu 0.000000000 0.500000000 -0.505544119 > Cu 0.500000000 0.000000000 -0.505544119 > Cu 0.000000000 0.000000000 0.000000000 > Cu 0.500000000 0.500000000 0.000000000 > Cu 0.500000000 0.000000000 0.505544119 > Cu 0.000000000 0.500000000 0.505544119 > Cu 0.000000000 0.000000000 1.009845452 > Cu 0.500000000 0.500000000 1.009845482 > Cu 0.500000000 0.000000000 1.511684867 > Cu 0.000000000 0.500000000 1.511684867 > Cu 0.000000000 0.000000000 1.996326572 > Cu 0.500000000 0.500000000 1.996326571 > > Forces acting on atoms (Ry/au): > > atom 1 type 1 force = 0.00000000 0.00000000 > 0.00023390 > atom 2 type 1 force = 0.00000000 0.00000000 > 0.00023392 > atom 3 type 1 force = 0.00000000 0.00000000 > -0.00011805 > atom 4 type 1 force = 0.00000000 0.00000000 > -0.00011805 > atom 5 type 1 force = 0.00000000 0.00000000 > 0.00087029 > atom 6 type 1 force = 0.00000000 0.00000000 > 0.00087025 > atom 7 type 1 force = 0.00000000 0.00000000 > -0.00009007 > atom 8 type 1 force = 0.00000000 0.00000000 > -0.00009007 > atom 9 type 1 force = 0.00000000 0.00000000 > 0.00000000 > atom 10 type 1 force = 0.00000000 0.00000000 > 0.00000000 > atom 11 type 1 force = 0.00000000 0.00000000 > 0.00009007 > atom 12 type 1 force = 0.00000000 0.00000000 > 0.00009007 > atom 13 type 1 force = 0.00000000 0.00000000 > -0.00087025 > atom 14 type 1 force = 0.00000000 0.00000000 > -0.00087029 > atom 15 type 1 force = 0.00000000 0.00000000 > 0.00011805 > atom 16 type 1 force = 0.00000000 0.00000000 > 0.00011805 > atom 17 type 1 force = 0.00000000 0.00000000 > -0.00023392 > atom 18 type 1 force = 0.00000000 0.00000000 > -0.00023390 > > total stress (ryd/bohr**3) (kbar) P= > -8.85 > -0.00008010 0.00000000 0.00000000 -11.78 0.00 > 0.00 > 0.00000000 -0.00008010 0.00000000 0.00 -11.78 > 0.00 > 0.00000000 0.00000000 -0.00002029 0.00 0.00 > -2.99 > > 1st and 2nd layers approached by ~ -3.07 % > good agreement with exp.!!! > > 2nd and 3rd layers separated by ~ +0.37 % > not too bad ... > We were intrigued by these opposite and inconsistent results under a negligible change in the lattice parameter so, I was advised to repeat the calculation with the old lattice constant value but, now also calculating the stress, since maybe I had been doing something wrong... Well, fortunately, I got the same relaxed positions, same small final forces, but what was really amazing was the stress I found: total stress (ryd/bohr**3) (kbar) P= 231.63 0.00160923 0.00000000 0.00000000 236.73 0.00 0.00 0.00000000 0.00160923 0.00000000 0.00 236.73 0.00 0.00000000 0.00000000 0.00150533 0.00 0.00 221.44 That's it, everything is consistent now, the systems wants to expand because we imposed a little bit smaller lattice constant. Our question is, how is it possible that the stress increases so much under such a small change of the lattice constant?, how can we explain that we find completely opposite results under such a small change of the lattice constant? We definitely conclude that either I'm missing something or Cu PBE pseudopotentail is dangerously oversensitive so that we cannot trust. Please let me know your comments. Thank you very much. Best regards. Marisol Alcantara Ortigoza. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050210/b3207c3d/attachment.htm From marzari at MIT.EDU Fri Feb 11 02:02:45 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Thu, 10 Feb 2005 20:02:45 -0500 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In-Reply-To: References: Message-ID: <420C0435.4010301@mit.edu> In these calculations it is essential to have an accurate k-point sampling and some finite temperature broadening. What were you using in both cases ? How many bands did you have above the Fermi energy ? nicola Alcantara Ortigoza, Marisol wrote: > Hi, > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using > PBE pseudopotential provided by pwscf. > --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu From giannozz at nest.sns.it Fri Feb 11 16:28:00 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 11 Feb 2005 16:28:00 +0100 Subject: [Pw_forum] efermit error In-Reply-To: References: Message-ID: <200502111628.00980.giannozz@nest.sns.it> On Wednesday 09 February 2005 16:19, Ben Hope wrote: > from efermit : error # 1 > unexpected error and also something like "Only input k-points are used (inequivalent points not generated)" at the beginning of the run? it is a bug. For the time being, change in routine PW/setup.f90 the line ltest = ( nks /= input_nks ) .AND. & ( .NOT. lscf ) .AND. ( .NOT. ( lphonon .OR. lraman ) ) into ltest = ( nks /= input_nks ) .AND. ( .NOT. ( ltetra ) .AND. & ( .NOT. lscf ) .AND. ( .NOT. ( lphonon .OR. lraman ) ) Gaussian smearing with an automatic k-point grid may have the same problem, by the way Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From aouahab at yahoo.fr Fri Feb 11 16:59:57 2005 From: aouahab at yahoo.fr (Abdelouahab O.) Date: Fri, 11 Feb 2005 16:59:57 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: Message-ID: <20050211155957.2631.qmail@web26402.mail.ukl.yahoo.com> Thanks for replay. I have update my intel fortran compiler to 8.1, and downlaod espresso2.1.2; the lam precompiled for fedora core (lam-7.0.6-2ak6.i386.rpm) and install it on my system The make.sys is =================== # make.sys. Generated from make.sys.in by configure. CC = mpicc CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) $(IFLAGS) # See include/defs.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW FDFLAGS = $(DFLAGS) IFLAGS = -I. -I../include -I../Modules -I../PW -I../PH CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) F77 = mpif77 F90 = mpiifort FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) F77FLAGS = $(FFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) F77FLAGS_NOOPT = -O0 LD = mpiifort LDFLAGS = $(LIBOBJS) $(LIBS) LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a # LIBS must contain the location of all needed external libraries LIBS = -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam # MYLIB can be one of the following (depending on LIBS): # blas : compile the local copy of blas routines # lapack : compile the local copy of lapack routines # blas_and_lapack : all of the above - use this for a quick test # or if you don't have an optimized blas/lapack library # lapack_ibm : compile only lapack routines not present in IBM ESSL # use this together with IBM ESSL # lapack_t3e : compile only lapack routines not present in T3E scilib # use this together with T3E scilib # lapack_mkl : compile only lapack routines not present in Intel MKL # use this together with Intel MKL MYLIB = lapack_mkl AR = ar ARFLAGS = ruv RANLIB = echo ========================================================================= the result after ./makedeps.sh ;make all is: /////////////////////////////////////////// make[1]: Entering directory `/home/wahab/pw2.1.2/Modules' mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c io_global.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c kind.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c parameters.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c atom.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c parallel_include.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c shmem_include.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c mp_global.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c mp.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c recvec.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c stick_base.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c berry_phase.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c basic_algebra_routines.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c constants.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c io_files.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c parser.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c bfgs_module.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c cell_base.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c check_stop.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c clocks.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c control_flags.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c cp_emass.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c fft_types.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c mp_buffers.f90 mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c fft_base.f90 fortcom: Error: fft_base.f90, line 691: This name does not have a type, and must have an explicit type. [MPI_DOUBLE_COMPLEX] call mpi_alltoallv (f_aux(1), sendcount, sdispls, MPI_DOUBLE_COMPLEX, f_in(1), & -------------------------------------------------------^ compilation aborted for fft_base.f90 (code 1) /usr/bin/mpif77: No such file or directory make[1]: *** [fft_base.o] Error 1 make[1]: Leaving directory `/home/wahab/pw2.1.2/Modules' make: *** [mods] Error 2 [root at cn1 pw2.1.2]# /////////////////////////////////////////// I dont understand why the last line "/usr/bin/mpif77: No such file or directory" appears, when I look for mpif77 this is whats the system gives: [root at cn1 pw2.1.2]# which mpif77 /usr/bin/mpif77 and when trying to execute the command mpif77: [root at cn1 pw2.1.2]# mpif77 /usr/lib/gcc-lib/i386-redhat-linux/3.3.2/libfrtbegin.a(frtbegin.o)(.text+0x32): In function `main': : undefined reference to `MAIN__' collect2: ld returned 1 exit status mpif77: No such file or directory What can I do to make it work? Thanks Axel Kohlmeyer wrote:On Tue, 8 Feb 2005, Abdelouahab O. wrote: AO> Hello AO> Thanks for your relpay; AO> This is make.sys hi, the current version of pwscf (or Quantum-ESPRESSO) is 2.1.2. with many new features, bugfixes and improvements. if you have installed you LAM/MPI properly for the intel compilers you should be able to just use CC=mpicc F77=mpif77 F90=mpif90 alternatively you can try to use CC=env LAMHCC=icc mpicc F77=env LAMHF77=ifort mpif77 F90=env LAMHF77=ifort mpif77 using the mpiXXX wrappers will make sure, that you include the proper headers that are matching your LAM installation. AO> ======================================================= AO> # make.sys. Generated from make.sys.in by configure. AO> CC = gcc AO> CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) AO> $(IFLAGS) AO> # See include/machine.h.README for a list of AO> precompilation options AO> # (possible arguments to -D or -U) and their meaning AO> DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA AO> -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW AO> IFLAGS = -I. -I../include -I../Modules -I../PW AO> -I../PH AO> CPP = gcc -E AO> CPPFLAGS = $(DFLAGS) $(IFLAGS) AO> F77 = ifort AO> F90 = ifort AO> FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) AO> F77FLAGS = $(FFLAGS) AO> F90FLAGS = $(FFLAGS) -nomodule -fpp $(DFLAGS) AO> F77FLAGS_NOOPT = -O0 AO> LD = ifort AO> LDFLAGS = -Vaxlib -O3 -tpp7 $(LIBOBJS) $(LIBS) AO> LIBOBJS = ../flib/ptools.a ../flib/flib.a AO> ../clib/clib.a AO> # LIBS must contain the location of all needed AO> external libraries AO> LIBS = -L/opt/intel/mkl72/lib/32 AO> -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam AO> # MYLIB can be one of the following (depending on AO> LIBS): AO> # blas : compile the local copy of blas routines AO> # lapack : compile the local copy of lapack AO> routines AO> # blas_and_lapack : all of the above - use this for a AO> quick test AO> # or if you don't have an optimized AO> blas/lapack library AO> # lapack_ibm : compile only lapack routines not AO> present in IBM ESSL AO> # use this together with IBM ESSL AO> # lapack_t3e : compile only lapack routines not AO> present in T3E scilib AO> # use this together with T3E scilib AO> # lapack_mkl : compile only lapack routines not AO> present in Intel MKL AO> # use this together with Intel MKL AO> MYLIB = lapack_mkl AO> AR = ar AO> ARFLAGS = ruv AO> ======================================================= AO> AO> And make all gives: AO> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% AO> make[1]: Entering directory AO> `/home1/wahab/pwscf2.0.4/pwscf/Modules' [...] AO> fortcom: Warning: mp_wave.f90, line 130: A dummy AO> argument with an expl AO> icit INTENT(OUT) declaration is not given an explicit AO> value. [PW] AO> SUBROUTINE splitwf ( pw, pwt, ngwl, ig_l2g, AO> mpime, nproc, root, AO> comm ) AO> ...........................^ AO> fortcom: Warning: mp_wave.f90, line 23: A dummy AO> argument with an expli this looks like a 'feature', that only the recent intel fortran (and the DEC/Compaq) compilers choke on. this may have been fixed in the newer releases. i remember to have seen several of those with version 2.1 which are now fixed. AO> -D__USE_INTERNAL_FFTW -c broadcast.f90 AO> fortcom: Error: broadcast.f90, line 146: This name has AO> already been assigned a data type. AO> [MPI_DOUBLE_PRECISION] AO> INTEGER MPI_REAL4, MPI_DOUBLE_PRECISION, AO> MPI_REAL16 AO> .........................^ AO> compilation aborted for broadcast.f90 (code 1) AO> make[1]: *** [broadcast.o] Error 1 AO> make[1]: Leaving directory AO> `/home1/wahab/pwscf2.0.4/pwscf/PW' AO> make: *** [pw] Error 2 i cannot reproduce this on my desktop with pwscf v2.0.4, my installation of LAM/MPI version 7.0.6 and the intel compiler version 8.1. in my distribution broadcast.f90 only has 41 lines, so your error on line 146 looks very suspicious. same is for the content, 'MPI_REAL4' is only present in MPICH installations and not in LAM/MPI. if you have administrator priviledge on your machine, you can download my LAM/MPI package as rpm from: http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-linux.html#mpi these rpm are compiled special to be compatible to the intel compilers as well as g77 or the pgi compilers. you simply need to use mpiifort/mpiicc instead of mpif77/mpicc. regards, axel. AO> AO> --- Paolo Giannozzi a ?crit : AO> > On Sunday 06 February 2005 10:46, Abdelouahab O. AO> > wrote: AO> > AO> > > I have a pc linux cluster with ifort8.0, mkl7.2, AO> > and AO> > > LAM7.0.6: the parallel compilation cant be AO> > finilized AO> > > using the -D__LAM -D__PARA, the error apppears AO> > when AO> > > compiling broadcast.f90 AO> > AO> > which error? AO> > _______________________________________________ AO> > Pw_forum mailing list AO> > Pw_forum at pwscf.org AO> > http://www.democritos.it/mailman/listinfo/pw_forum AO> > AO> AO> AO> AO> AO> AO> AO> AO> D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! AO> Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ AO> _______________________________________________ AO> Pw_forum mailing list AO> Pw_forum at pwscf.org AO> http://www.democritos.it/mailman/listinfo/pw_forum AO> AO> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum --------------------------------- D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! Cr?ez votre Yahoo! Mail -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050211/405bbbd8/attachment.htm From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Fri Feb 11 17:41:22 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Fri, 11 Feb 2005 17:41:22 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: <20050211155957.2631.qmail@web26402.mail.ukl.yahoo.com> Message-ID: On Fri, 11 Feb 2005, Abdelouahab O. wrote: AO> Thanks for replay. AO> AO> I have update my intel fortran compiler to 8.1, and downlaod AO> espresso2.1.2; the lam precompiled for fedora core AO> (lam-7.0.6-2ak6.i386.rpm) and install it on my system ok. a lot of work, but i hope it will be worth the effort... AO> The make.sys is AO> =================== AO> # make.sys. Generated from make.sys.in by configure. AO> CC = mpicc AO> CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) $(IFLAGS) you may want to add -funroll-loops here which should make the FFTs a little faster, but i suppose you're more interested in getting it to work at, right? AO> # See include/defs.h.README for a list of precompilation options AO> # (possible arguments to -D or -U) and their meaning AO> DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW there should be a -D__MPI here. AO> FDFLAGS = $(DFLAGS) AO> IFLAGS = -I. -I../include -I../Modules -I../PW -I../PH AO> CPP = cpp AO> CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) AO> F77 = mpif77 this should be 'F77 = mpiifort'. using mpif77 will use the g77. while there is no problem with that per se, you may get into trouble mixing objects from g77 with objects from the intel compiler. AO> F90 = mpiifort AO> FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) again. using '-O2 -unroll' should be faster than '-O3'. AO> F77FLAGS = $(FFLAGS) AO> F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) AO> F77FLAGS_NOOPT = -O0 AO> LD = mpiifort AO> LDFLAGS = $(LIBOBJS) $(LIBS) here you should use either 'LDFLAGS = -static-libcxa $(LIBOBJS) $(LIBS)' if you have the download version of the intel-8.1 compiler, or 'LDFLAGS = -i-static $(LIBOBJS) $(LIBS)' if you have the latest update from http://premier.intel.com/ ( $ > rpm -q intel-ifort8 intel-ifort8-8.1-023 ) [...] AO> mpiifort -Vaxlib -O3 -tpp7 -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__LAM -D__PARA -D__FFTW -D__USE_INTERNAL_FFTW -c fft_base.f90 AO> fortcom: Error: fft_base.f90, line 691: This name does not have a type, and must have an explicit type. [MPI_DOUBLE_COMPLEX] AO> call mpi_alltoallv (f_aux(1), sendcount, sdispls, MPI_DOUBLE_COMPLEX, f_in(1), & AO> -------------------------------------------------------^ AO> compilation aborted for fft_base.f90 (code 1) AO> /usr/bin/mpif77: No such file or directory AO> make[1]: *** [fft_base.o] Error 1 AO> make[1]: Leaving directory `/home/wahab/pw2.1.2/Modules' AO> make: *** [mods] Error 2 AO> [root at cn1 pw2.1.2]# AO> /////////////////////////////////////////// AO> I dont understand why the last line "/usr/bin/mpif77: No such file or directory" appears, when I look for mpif77 this is whats the system gives: this is from the fact, that /usr/bin/mpiifort is a small shell script, that actually calls mpif77 with the LAMHF77 environment variable set to ifort. AO> [root at cn1 pw2.1.2]# which mpif77 AO> /usr/bin/mpif77 AO> and when trying to execute the command mpif77: AO> [root at cn1 pw2.1.2]# mpif77 AO> /usr/lib/gcc-lib/i386-redhat-linux/3.3.2/libfrtbegin.a(frtbegin.o)(.text+0x32): In function `main': AO> : undefined reference to `MAIN__' AO> collect2: ld returned 1 exit status AO> mpif77: No such file or directory AO> What can I do to make it work? AO> Thanks i have run configure like this: LIBDIRS=$HOME/lib F77=mpiifort F90=mpiifort CC=mpicc ./configure and then edited the make.sys file a little bit (the final one is attached for reference). please remember that you need to do a make clean, whenever you change any compiler flags to get a consistent compilation. regards, axel. [...] -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. -------------- next part -------------- # make.sys. Generated from make.sys.in by configure. CC = mpicc CCFLAGS = -O3 -fomit-frame-pointer -funroll-loops $(DFLAGS) $(IFLAGS) # See include/defs.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__LINUX -D__INTEL -D__MPI -D__PARA -D__FFTW -D__LAM -D__USE_INTERNAL_FFTW FDFLAGS = $(DFLAGS) IFLAGS = -I. -I../include -I../Modules -I../PW -I../PH CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) F77 = mpiifort F90 = mpiifort FFLAGS = -O2 -unroll -tpp7 -assume byterecl $(IFLAGS) F77FLAGS = $(FFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) F77FLAGS_NOOPT = -O0 LD = mpiifort -i-static LDFLAGS = $(LIBOBJS) $(LIBS) LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a # LIBS must contain the location of all needed external libraries LIBS = -L/home/akohlmey/lib -latlas_p4 # MYLIB can be one of the following (depending on LIBS): # blas : compile the local copy of blas routines # lapack : compile the local copy of lapack routines # blas_and_lapack : all of the above - use this for a quick test # or if you don't have an optimized blas/lapack library # lapack_ibm : compile only lapack routines not present in IBM ESSL # use this together with IBM ESSL # lapack_t3e : compile only lapack routines not present in T3E scilib # use this together with T3E scilib # lapack_mkl : compile only lapack routines not present in Intel MKL # use this together with Intel MKL MYLIB = AR = ar ARFLAGS = ruv RANLIB = echo From alcantar at phys.ksu.edu Fri Feb 11 17:50:33 2005 From: alcantar at phys.ksu.edu (Alcantara Ortigoza, Marisol) Date: Fri, 11 Feb 2005 10:50:33 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Message-ID: Hi, Thank you very much for replying. I did not change anything in the input file for each case, but the lattice constant. So, in both cases the k-point sampling was 9X9X1. This mesh gave us 15 special kpoints, while the temperature broadening was 0.05 ryd. 1) For a = 3.67 A Fermi energy = 5.3079 ev Highest band energy = 10.9354 ev # bands above the Fermi energy : 17 2) For a = 3.6737 A Fermi energy = 5.5509 ev Highest band energy = 11.2380 ev # bands above the Fermi energy : 17 Thank you very much. Marisol Alcantara Ortigoza. -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Nicola Marzari Sent: Thursday, February 10, 2005 7:03 PM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In these calculations it is essential to have an accurate k-point sampling and some finite temperature broadening. What were you using in both cases ? How many bands did you have above the Fermi energy ? nicola Alcantara Ortigoza, Marisol wrote: > Hi, > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using > PBE pseudopotential provided by pwscf. > --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From ejl7240 at chemail.tamu.edu Fri Feb 11 17:49:27 2005 From: ejl7240 at chemail.tamu.edu (Eduardo J. Lamas) Date: Fri, 11 Feb 2005 10:49:27 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In-Reply-To: Message-ID: <000001c51059$a607ccf0$0ee35ba5@ch0316> Hi what about having a very well converged wavefunction, that is very important for having accurate forces (and not that important for energies). Regards, Eduardo -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Alcantara Ortigoza, Marisol Sent: Friday, February 11, 2005 10:51 AM To: pw_forum at pwscf.org Subject: RE: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Hi, Thank you very much for replying. I did not change anything in the input file for each case, but the lattice constant. So, in both cases the k-point sampling was 9X9X1. This mesh gave us 15 special kpoints, while the temperature broadening was 0.05 ryd. 1) For a = 3.67 A Fermi energy = 5.3079 ev Highest band energy = 10.9354 ev # bands above the Fermi energy : 17 2) For a = 3.6737 A Fermi energy = 5.5509 ev Highest band energy = 11.2380 ev # bands above the Fermi energy : 17 Thank you very much. Marisol Alcantara Ortigoza. -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Nicola Marzari Sent: Thursday, February 10, 2005 7:03 PM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In these calculations it is essential to have an accurate k-point sampling and some finite temperature broadening. What were you using in both cases ? How many bands did you have above the Fermi energy ? nicola Alcantara Ortigoza, Marisol wrote: > Hi, > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using > PBE pseudopotential provided by pwscf. > --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From ferretti.andrea at unimore.it Fri Feb 11 18:35:38 2005 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Fri, 11 Feb 2005 18:35:38 +0100 (CET) Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In-Reply-To: Message-ID: dear Marisol, addressing the point of nicola, the convergence the convergence of kpt mesh should be carefully checked when dealing with metals.... here a broadening of 0.05 Ry = 0.68 eV (!!) seems to me to be too large, probably missing the detailed structure of energy bands and the position of the Fermi level for instance.... which results in unconverged physical properties..... the broadening value should be chosen in accordance with the choice of the kpt mesh, and, to my (limited) experience, some "reasonable" values are between 0.004 and 0.01 Ry... I don't know if this may explain the disagreement and oscillation in results you observed but the broadening you set looks quite 'unusual'.... hope it helps Andrea -- Andrea Ferretti INFM National Research Center on nanoStructures and bioSystems at Surfaces (S3) Dipartimento di Fisica, Universita' di Modena e Reggio Emilia Via Campi 213/A I-41100 Modena, Italy Tel: +39 059 2055283 Fax: +39 059 367488 E-mail: ferretti.andrea at unimore.it URL: http://www.nanoscience.unimo.it On Fri, 11 Feb 2005, Alcantara Ortigoza, Marisol wrote: > Hi, > > Thank you very much for replying. > > I did not change anything in the input file for each case, but the > lattice constant. So, in both cases the k-point sampling was 9X9X1. This > mesh gave us 15 special kpoints, while the temperature broadening was > 0.05 ryd. > > 1) For a = 3.67 A > > Fermi energy = 5.3079 ev > > Highest band energy = 10.9354 ev > > # bands above the Fermi energy : 17 > > 2) For a = 3.6737 A > > Fermi energy = 5.5509 ev > > Highest band energy = 11.2380 ev > > # bands above the Fermi energy : 17 > > Thank you very much. > > Marisol Alcantara Ortigoza. > > > > -----Original Message----- > From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On > Behalf Of Nicola Marzari > Sent: Thursday, February 10, 2005 7:03 PM > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential > > > > > In these calculations it is essential to have an > accurate k-point sampling and some finite temperature broadening. > > What were you using in both cases ? How many bands did you have above > the Fermi energy ? > > nicola > > > Alcantara Ortigoza, Marisol wrote: > > > Hi, > > > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using > > PBE pseudopotential provided by pwscf. > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From alcantar at phys.ksu.edu Fri Feb 11 19:14:15 2005 From: alcantar at phys.ksu.edu (Alcantara Ortigoza, Marisol) Date: Fri, 11 Feb 2005 12:14:15 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Message-ID: Hi, I used the default value for convergence threshold: 1e-6. And, for this threshold the wavefunction converged well. Thank you. Marisol. -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Eduardo J. Lamas Sent: Friday, February 11, 2005 10:49 AM To: pw_forum at pwscf.org Subject: RE: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Hi what about having a very well converged wavefunction, that is very important for having accurate forces (and not that important for energies). Regards, Eduardo -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Alcantara Ortigoza, Marisol Sent: Friday, February 11, 2005 10:51 AM To: pw_forum at pwscf.org Subject: RE: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Hi, Thank you very much for replying. I did not change anything in the input file for each case, but the lattice constant. So, in both cases the k-point sampling was 9X9X1. This mesh gave us 15 special kpoints, while the temperature broadening was 0.05 ryd. 1) For a = 3.67 A Fermi energy = 5.3079 ev Highest band energy = 10.9354 ev # bands above the Fermi energy : 17 2) For a = 3.6737 A Fermi energy = 5.5509 ev Highest band energy = 11.2380 ev # bands above the Fermi energy : 17 Thank you very much. Marisol Alcantara Ortigoza. -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Nicola Marzari Sent: Thursday, February 10, 2005 7:03 PM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential In these calculations it is essential to have an accurate k-point sampling and some finite temperature broadening. What were you using in both cases ? How many bands did you have above the Fermi energy ? nicola Alcantara Ortigoza, Marisol wrote: > Hi, > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by using > PBE pseudopotential provided by pwscf. > --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From alcantar at phys.ksu.edu Fri Feb 11 19:26:02 2005 From: alcantar at phys.ksu.edu (Alcantara Ortigoza, Marisol) Date: Fri, 11 Feb 2005 12:26:02 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Message-ID: Dear Andrea, Thanks for replying. It's worth to try those values , however the same value for broadening was used in both calculations and the volume of the two systems was almost the same (differing by just 0.9%), then the Brilloruin zone differed by 0.9% too. The k-point sampling density, therefore, should be same for both systems. Thus, if the broadening were a problem for the first calculation, it should also be a problem for the second one, but for the second calculation I got the correct result according to the experiment. Best regard. Marisol. -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Andrea Ferretti Sent: Friday, February 11, 2005 11:36 AM To: PWSCF forum list Subject: RE: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential dear Marisol, addressing the point of nicola, the convergence the convergence of kpt mesh should be carefully checked when dealing with metals.... here a broadening of 0.05 Ry = 0.68 eV (!!) seems to me to be too large, probably missing the detailed structure of energy bands and the position of the Fermi level for instance.... which results in unconverged physical properties..... the broadening value should be chosen in accordance with the choice of the kpt mesh, and, to my (limited) experience, some "reasonable" values are between 0.004 and 0.01 Ry... I don't know if this may explain the disagreement and oscillation in results you observed but the broadening you set looks quite 'unusual'.... hope it helps Andrea -- Andrea Ferretti INFM National Research Center on nanoStructures and bioSystems at Surfaces (S3) Dipartimento di Fisica, Universita' di Modena e Reggio Emilia Via Campi 213/A I-41100 Modena, Italy Tel: +39 059 2055283 Fax: +39 059 367488 E-mail: ferretti.andrea at unimore.it URL: http://www.nanoscience.unimo.it On Fri, 11 Feb 2005, Alcantara Ortigoza, Marisol wrote: > Hi, > > Thank you very much for replying. > > I did not change anything in the input file for each case, but the > lattice constant. So, in both cases the k-point sampling was 9X9X1. > This mesh gave us 15 special kpoints, while the temperature broadening > was 0.05 ryd. > > 1) For a = 3.67 A > > Fermi energy = 5.3079 ev > > Highest band energy = 10.9354 ev > > # bands above the Fermi energy : 17 > > 2) For a = 3.6737 A > > Fermi energy = 5.5509 ev > > Highest band energy = 11.2380 ev > > # bands above the Fermi energy : 17 > > Thank you very much. > > Marisol Alcantara Ortigoza. > > > > -----Original Message----- > From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On > Behalf Of Nicola Marzari > Sent: Thursday, February 10, 2005 7:03 PM > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential > > > > > In these calculations it is essential to have an > accurate k-point sampling and some finite temperature broadening. > > What were you using in both cases ? How many bands did you have above > the Fermi energy ? > > nicola > > > Alcantara Ortigoza, Marisol wrote: > > > Hi, > > > > I was trying to relax Cu100 using 9 p(2x2) layers + 9 vacuum by > > using PBE pseudopotential provided by pwscf. > > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum > _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From wlyim at puccini.che.pitt.edu Fri Feb 11 17:00:49 2005 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Fri, 11 Feb 2005 11:00:49 -0500 (EST) Subject: [Pw_forum] parallel ph.x in Opteron Message-ID: Dear PWSCF community, Recently I updated pwscf to version 2.1.2 into Opteron cluster using PGI 5.1 compiler. The serial version of pw.x and ph.x seemed ok. The job can be finished without error. I have also the parallel verion. In parallel run, pw.x is working, however, I got some error by parallel ph.x. I have compiled the parallel program with lam.7.1.1 and mpich.1.2.6, both of them gave the same error message. ======================================================================== Program PHONON v.2.1.2 starts ... Today is 11Feb2005 at 16:58:21 Parallel version (MPI) Number of processors in use: 2 R & G space division: nprocp = 2 Ultrasoft (Vanderbilt) Pseudopotentials Reading file xe.save ... read complete Reading file xe.save ... read complete Planes per process (thick) : nr3 = 45 npp = 23 ncplane = 2025 Proc/ planes cols G planes cols G columns G Pool (dense grid) (smooth grid) (wavefct grid) 1 23 672 16322 23 672 16322 189 2513 2 22 673 16325 22 673 16325 190 2516 0 45 1345 32647 45 1345 32647 379 5029 nbndx = 4 nbnd = 4 natomwfc = 4 npwx = 2045 nelec = 8.00 nkb = 1 ngl = 332 0: DEALLOCATE: memory at 439d1440 not allocated ----------------------------------------------------------------------------- One of the processes started by mpirun has exited with a nonzero exit code. This typically indicates that the process finished in error. If your process did not finish in error, be sure to include a "return 0" or "exit(0)" in your C code before exiting the application. PID 19789 failed on node n0 (192.168.100.27) with exit status 1. ----------------------------------------------------------------------------- ============================================================================= Can anyone have similar problem? I will be very happy if anyone can help me to solve this problem. Many thanks in advance. Best regards, William Yim From giannozz at nest.sns.it Sat Feb 12 15:06:58 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Sat, 12 Feb 2005 15:06:58 +0100 Subject: [Pw_forum] parallel ph.x in Opteron In-Reply-To: References: Message-ID: <200502121506.58730.giannozz@nest.sns.it> On Friday 11 February 2005 17:00, wlyim at puccini.che.pitt.edu wrote: > Recently I updated pwscf to version 2.1.2 into Opteron cluster using > PGI 5.1 compiler. The serial version of pw.x and ph.x seemed ok. [...] > In parallel run, pw.x is working, however, I got some error by parallel > ph.x. it looks like a compiler problem. Get the latest available patches for the PGI compiler; if nothing changes, try g95. There is a version for 64-bit AMD. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From proffess at yandex.ru Sat Feb 12 16:57:38 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Sat, 12 Feb 2005 18:57:38 +0300 (MSK) Subject: [Pw_forum] nscf calculation Message-ID: <420E2772.000004.20563@pantene.yandex.ru> Dear PWscf users and authors, I performed a scf calculation of a system (36 atoms). After that I tried to perform a nscf calculation (for DOS calculation). I used 'tetrahedra' method for occupations and automatic generation of k-points (400 1 1 0 0 0). I did not use the 'restart' option, but I got: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% task # 34 from davcio : error # 0 i/o error in davcio %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% What is wrong in my case? Sergey From aouahab at yahoo.fr Sat Feb 12 17:25:37 2005 From: aouahab at yahoo.fr (Abdelouahab O.) Date: Sat, 12 Feb 2005 17:25:37 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: Message-ID: <20050212162537.564.qmail@web26410.mail.ukl.yahoo.com> Thanks; I've changed the make.sys like you's # make.sys. Generated from make.sys.in by configure. CC = mpicc CCFLAGS = -O3 -fomit-frame-pointer -funroll-loops $(DFLAGS) $(IFLAGS) # See include/defs.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__LINUX -D__INTEL -D__MPI -D__PARA -D__FFTW -D__LAM -D__USE_INTERNAL_FFTW FDFLAGS = $(DFLAGS) IFLAGS = -I. -I../include -I../Modules -I../PW -I../PH CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) F77 = mpiifort F90 = mpiifort FFLAGS = -O2 -unroll -tpp7 -assume byterecl $(IFLAGS) F77FLAGS = $(FFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) F77FLAGS_NOOPT = -O0 LD = mpiifort -i-static LDFLAGS = -static-libcxa $(LIBOBJS) LIBOBJS = ../flib/ptools.a ../flib/flib.a ../clib/clib.a # LIBS must contain the location of all needed external libraries LIBS = -L/home/wahab/atals -latlas_p4 -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam # MYLIB can be one of the following (depending on LIBS): # blas : compile the local copy of blas routines # lapack : compile the local copy of lapack routines # blas_and_lapack : all of the above - use this for a quick test # or if you don't have an optimized blas/lapack library # lapack_ibm : compile only lapack routines not present in IBM ESSL # use this together with IBM ESSL # lapack_t3e : compile only lapack routines not present in T3E scilib # use this together with T3E scilib # lapack_mkl : compile only lapack routines not present in Intel MKL # use this together with Intel MKL MYLIB = AR = ar ARFLAGS = ruv RANLIB = echo ////////////////////////////////////// but the line: LIBS = -L/home/wahab/atals -latlas_p4 -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 because i think the library in your "/home/akohlmey/lib" is atlas libs; so I added the atlas library with the mkl since I have the 8.1.20 version of intel complier i used "-static-libcxa $(LIBOBJS)" make clean;./makedeps.sh make all I got: make[1]: Entering directory `/home/wahab/pw2.1.2/Modules' mpiifort -O2 -unroll -tpp7 -assume byterecl -I. -I../include -I../Modules -I../PW -I../PH -nomodule -fpp -D__LINUX -D__INTEL -D__MPI -D__PARA -D__FFTW -D__LAM -D__USE_INTERNAL_FFTW -c io_global.f90 ........ ........ (making clib, flib is ok) ........ but all programs in PW/ were compiled but linking them fails: mpiifort -i-static -o memory.x memory.o add_efield.o add_vuspsi.o addusdens.o addusforce.o addusstress.o allocate_fft.o allocate_locpot.o allocate_nlpot.o allocate_wfc.o atomic_rho.o atomic_wfc.o bachel.o becmod.o bfgs.o bp_bess.o bp_c_phase.o bp_calc_btq.o bp_dbess.o bp_qvan3.o bp_radin.o bp_strings.o bp_ylm_q.o c_bands.o c_gemm.o ccalbec.o ccgdiagg.o cdiagh.o cdiaghg.o cegterg.o cft3.o cft3s.o cft_3.o cft_sgi.o cft_sp.o cft_sun.o cft_t3e.o cfts_3.o cgramg1.o checkallsym.o checksym.o cinitcgg.o clean_pw.o close_files.o compute_dip.o compute_scf.o constraints_module.o coset.o cryst_to_car.o cubicsym.o d_matrix.o data_structure.o davcio.o delta_e.o deriv_drhoc.o diis_base.o real_diis_module.o complex_diis_module.o diropn.o divide.o divide_et_impera.o dndepsilon.o dndtau.o dprojdepsilon.o dprojdtau.o dqvan2.o drhoc.o dvloc_of_g.o dynamics.o efermig.o efermit.o electrons.o eqvect.o error_handler.o estimate.o ewald.o ewald_dipole.o fftw.o force_cc.o force_corr.o force_ew.o force_hub.o force_lc.o force_us.o forces.o functionals.o g_psi.o g_psi_mod.o gen_at_dj.o gen_at_dy.o gen_us_dj.o gen_us_dy.o ggen.o gk_sort.o gradcorr.o gweights.o h_1psi.o h_psi.o hexsym.o hinit0.o hinit1.o init_ns.o init_paw_1.o init_paw_2.o init_pool.o init_run.o init_us_1.o init_us_2.o init_at_1.o init_vloc.o input.o interpolate.o io_pot.o ions.o irrek.o iweights.o kpoint_grid.o lchk_tauxk.o linmin.o lsda_functionals.o make_pointlists.o mix_pot.o mix_rho.o mode_group.o move_ions.o multable.o n_plane_waves.o new_ns.o ns_adj.o newd.o noncol.o openfil.o ortho.o orthoatwfc.o output_tau.o para.o paw.o potinit.o print_clock_pw.o psymrho.o punch.o pw_gemm.o pwcom.o qvan2.o rdiaghg.o read_conf_from_file.o read_file.o read_ncpp.o read_pseudo.o readin.o readnewvan.o readvan.o regterg.o remove_atomic_rho.o reset_k_points.o restart.o restart_from_file.o restart_in_electrons.o restart_in_ions.o rgen.o rho2zeta.o rotate_wfc.o rotate_wfc_gamma.o ruotaijk.o s_1psi.o s_axis_to_ca.o s_gemm.o s_psi.o save_in_cbands.o save_in_electrons.o save_in_ions.o scala_cdiag.o scala_cdiaghg.o scala_utils.o scale_h.o scopy_t3e.o seqopn.o set_fft_dim.o set_hubbard_l.o set_kplusb.o set_kplusq.o set_kup_and_kdw.o set_rhoc.o set_vrs.o setlocal.o setqf.o setup.o setupkpt.o sgam_at.o sgam_at_mag.o sgam_ph.o sgama.o show_memory.o smallg_q.o spinor.o sph_ind.o startup.o stop_pw.o stres_cc.o stres_ewa.o stres_gradcorr.o stres_har.o stres_hub.o stres_knl.o stres_loc.o stres_us.o stress.o struct_fact.o sum_band.o sumkg.o sumkt.o summary.o swap.o symrho.o symtns.o symvect.o symz.o tabd.o trntns.o trnvecc.o trnvect.o tweights.o update_pot.o updathes.o upf_to_internal.o usnldiag.o v_of_rho.o vcsmd.o vcsubs.o vhpsi.o vloc_of_g.o vloc_psi.o vpack.o w0gauss.o w1gauss.o wfcinit.o wgauss.o write_config_to_file.o write_ns.o wsweight.o ../Modules/atom.o ../Modules/basic_algebra_routines.o ../Modules/berry_phase.o ../Modules/bfgs_module.o ../Modules/cell_base.o ../Modules/check_stop.o ../Modules/clocks.o ../Modules/constants.o ../Modules/control_flags.o ../Modules/fft_base.o ../Modules/fft_scalar.o ../Modules/fft_types.o ../Modules/functionals.o ../Modules/input_parameters.o ../Modules/io_base.o ../Modules/io_files.o ../Modules/io_global.o ../Modules/ions_base.o ../Modules/kind.o ../Modules/mp_buffers.o ../Modules/mp_global.o ../Modules/mp_wave.o ../Modules/mp.o ../Modules/path_base.o ../Modules/path_formats.o ../Modules/path_variables.o ../Modules/path_opt_routines.o ../Modules/path_io_routines.o ../Modules/parallel_include.o ../Modules/parameters.o ../Modules/parser.o ../Modules/pseudo_types.o ../Modules/read_cards.o ../Modules/read_namelists.o ../Modules/readpseudo.o ../Modules/recvec.o ../Modules/shmem_include.o ../Modules/stick_base.o ../Modules/supercell.o ../Modules/uspp.o ../Modules/version.o ../Modules/wavefunctions.o -static-libcxa ../flib/ptools.a ../flib/flib.a ../clib/clib.a ld: cannot find -lirc_s /usr/bin/mpif77: No such file or directory make[1]: *** [memory.x] Error 1 make[1]: Leaving directory `/home/wahab/pw2.1.2/PW' make: *** [pw] Error 2 What is tjhe meaning of ld: cannot find -lirc_s ? thanks Axel Kohlmeyer wrote:On Tue, 8 Feb 2005, Abdelouahab O. wrote: AO> Hello AO> Thanks for your relpay; AO> This is make.sys hi, the current version of pwscf (or Quantum-ESPRESSO) is 2.1.2. with many new features, bugfixes and improvements. if you have installed you LAM/MPI properly for the intel compilers you should be able to just use CC=mpicc F77=mpif77 F90=mpif90 alternatively you can try to use CC=env LAMHCC=icc mpicc F77=env LAMHF77=ifort mpif77 F90=env LAMHF77=ifort mpif77 using the mpiXXX wrappers will make sure, that you include the proper headers that are matching your LAM installation. AO> ======================================================= AO> # make.sys. Generated from make.sys.in by configure. AO> CC = gcc AO> CCFLAGS = -O3 -fomit-frame-pointer $(DFLAGS) AO> $(IFLAGS) AO> # See include/machine.h.README for a list of AO> precompilation options AO> # (possible arguments to -D or -U) and their meaning AO> DFLAGS = -D__LINUX -D__INTEL -D__LAM -D__PARA AO> -DHAS_ZHEGVX -D__FFTW -D__USE_INTERNAL_FFTW AO> IFLAGS = -I. -I../include -I../Modules -I../PW AO> -I../PH AO> CPP = gcc -E AO> CPPFLAGS = $(DFLAGS) $(IFLAGS) AO> F77 = ifort AO> F90 = ifort AO> FFLAGS = -Vaxlib -O3 -tpp7 $(IFLAGS) AO> F77FLAGS = $(FFLAGS) AO> F90FLAGS = $(FFLAGS) -nomodule -fpp $(DFLAGS) AO> F77FLAGS_NOOPT = -O0 AO> LD = ifort AO> LDFLAGS = -Vaxlib -O3 -tpp7 $(LIBOBJS) $(LIBS) AO> LIBOBJS = ../flib/ptools.a ../flib/flib.a AO> ../clib/clib.a AO> # LIBS must contain the location of all needed AO> external libraries AO> LIBS = -L/opt/intel/mkl72/lib/32 AO> -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam AO> # MYLIB can be one of the following (depending on AO> LIBS): AO> # blas : compile the local copy of blas routines AO> # lapack : compile the local copy of lapack AO> routines AO> # blas_and_lapack : all of the above - use this for a AO> quick test AO> # or if you don't have an optimized AO> blas/lapack library AO> # lapack_ibm : compile only lapack routines not AO> present in IBM ESSL AO> # use this together with IBM ESSL AO> # lapack_t3e : compile only lapack routines not AO> present in T3E scilib AO> # use this together with T3E scilib AO> # lapack_mkl : compile only lapack routines not AO> present in Intel MKL AO> # use this together with Intel MKL AO> MYLIB = lapack_mkl AO> AR = ar AO> ARFLAGS = ruv AO> ======================================================= AO> AO> And make all gives: AO> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% AO> make[1]: Entering directory AO> `/home1/wahab/pwscf2.0.4/pwscf/Modules' [...] AO> fortcom: Warning: mp_wave.f90, line 130: A dummy AO> argument with an expl AO> icit INTENT(OUT) declaration is not given an explicit AO> value. [PW] AO> SUBROUTINE splitwf ( pw, pwt, ngwl, ig_l2g, AO> mpime, nproc, root, AO> comm ) AO> ...........................^ AO> fortcom: Warning: mp_wave.f90, line 23: A dummy AO> argument with an expli this looks like a 'feature', that only the recent intel fortran (and the DEC/Compaq) compilers choke on. this may have been fixed in the newer releases. i remember to have seen several of those with version 2.1 which are now fixed. AO> -D__USE_INTERNAL_FFTW -c broadcast.f90 AO> fortcom: Error: broadcast.f90, line 146: This name has AO> already been assigned a data type. AO> [MPI_DOUBLE_PRECISION] AO> INTEGER MPI_REAL4, MPI_DOUBLE_PRECISION, AO> MPI_REAL16 AO> .........................^ AO> compilation aborted for broadcast.f90 (code 1) AO> make[1]: *** [broadcast.o] Error 1 AO> make[1]: Leaving directory AO> `/home1/wahab/pwscf2.0.4/pwscf/PW' AO> make: *** [pw] Error 2 i cannot reproduce this on my desktop with pwscf v2.0.4, my installation of LAM/MPI version 7.0.6 and the intel compiler version 8.1. in my distribution broadcast.f90 only has 41 lines, so your error on line 146 looks very suspicious. same is for the content, 'MPI_REAL4' is only present in MPICH installations and not in LAM/MPI. if you have administrator priviledge on your machine, you can download my LAM/MPI package as rpm from: http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-linux.html#mpi these rpm are compiled special to be compatible to the intel compilers as well as g77 or the pgi compilers. you simply need to use mpiifort/mpiicc instead of mpif77/mpicc. regards, axel. AO> AO> --- Paolo Giannozzi a ?crit : AO> > On Sunday 06 February 2005 10:46, Abdelouahab O. AO> > wrote: AO> > AO> > > I have a pc linux cluster with ifort8.0, mkl7.2, AO> > and AO> > > LAM7.0.6: the parallel compilation cant be AO> > finilized AO> > > using the -D__LAM -D__PARA, the error apppears AO> > when AO> > > compiling broadcast.f90 AO> > AO> > which error? AO> > _______________________________________________ AO> > Pw_forum mailing list AO> > Pw_forum at pwscf.org AO> > http://www.democritos.it/mailman/listinfo/pw_forum AO> > AO> AO> AO> AO> AO> AO> AO> AO> D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! AO> Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ AO> _______________________________________________ AO> Pw_forum mailing list AO> Pw_forum at pwscf.org AO> http://www.democritos.it/mailman/listinfo/pw_forum AO> AO> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum --------------------------------- D?couvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! Cr?ez votre Yahoo! Mail -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050212/ce762628/attachment.htm From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Sat Feb 12 21:19:36 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Sat, 12 Feb 2005 21:19:36 +0100 (CET) Subject: [Pw_forum] Can I get pw.x and memory.x for my linux pc cluster In-Reply-To: <20050212162537.564.qmail@web26410.mail.ukl.yahoo.com> Message-ID: On Sat, 12 Feb 2005, Abdelouahab O. wrote: AO> AO> Thanks; AO> AO> I've changed the make.sys like you's [...] AO> AO> LD = mpiifort -i-static AO> AO> LDFLAGS = -static-libcxa $(LIBOBJS) you should use either of '-i-static' or '-static-libcxa' but not both. which one depends on your compiler version. if you have the update from the support site, you should use '-i-static', for the (older) download version, you should use '-static-libcxa'. [...] AO> AO> LIBS = -L/home/wahab/atals -latlas_p4 -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -L/usr/lib/lam AO> [...] AO> AO> but the line: LIBS = -L/home/wahab/atals -latlas_p4 AO> -L/opt/intel/mkl72/lib/32 -lmkl_lapack -lmkl_ia32 AO> AO> because i think the library in your "/home/akohlmey/lib" is atlas AO> libs; so I added the atlas library with the mkl as far as PWscf is concerned, ATLAS and MKL provide the same functionality, so you need only one of them. AO> ../Modules/uspp.o ../Modules/version.o ../Modules/wavefunctions.o -static-libcxa ../flib/ptools.a ../flib/flib.a ../clib/clib.a AO> AO> ld: cannot find -lirc_s libirc_s.a is distributed with both icc and ifort, so there may be an installation problem. i have no problem at all. in fact, today i found some time to compile 'clean' serial and lam-7.0.6 versions for pentium 4, athlon (xp), and opteron. you can find the binaries in a few hours at: http://ftp.theochem.ruhr-uni-bochum.de/outgoing/axel_kohlmeyer/pwscf-bin/ regards, axel kohlmeyer. [...] -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From wlyim at puccini.che.pitt.edu Sun Feb 13 23:38:44 2005 From: wlyim at puccini.che.pitt.edu (wlyim at puccini.che.pitt.edu) Date: Sun, 13 Feb 2005 17:38:44 -0500 (EST) Subject: [Pw_forum] Re: Re: parallel ph.x in Opteron (Paolo Giannozzi) In-Reply-To: <20050212162601.26189.32324.Mailman@democritos.sissa.it> Message-ID: Thanks for your suggestion. I am waiting for the ifc compiler. By the way, I have checked the optimization flag and the subroutines. I found that all the object files can be compiled with the default configure settings (FFLAGS = -fast -r8 $(IFLAGS)). However, for a specific object file dvqpsi_us_only.o, I needed to lower the optimization level and changed it to (FFLAGS = -O1 -Mnoframe -Mlre -r8 $(IFLAGS) $(FDFLAGS)). So, now my parallel pw.x and ph.x works, where all object files were compiled by default setting and dvqpsi_us_only.o by specific setting. Any bug in dvqpsi_us_only.f90 or PGI 5.1 compiler? Many thanks in advance. Best regards, William Yim > Message: 2 > From: Paolo Giannozzi > Organization: Scuola Normale Superiore di Pisa > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] parallel ph.x in Opteron > Date: Sat, 12 Feb 2005 15:06:58 +0100 > Reply-To: pw_forum at pwscf.org > > On Friday 11 February 2005 17:00, wlyim at puccini.che.pitt.edu wrote: > > > Recently I updated pwscf to version 2.1.2 into Opteron cluster using > > PGI 5.1 compiler. The serial version of pw.x and ph.x seemed ok. [...] > > In parallel run, pw.x is working, however, I got some error by parallel > > ph.x. > > it looks like a compiler problem. Get the latest available patches > for the PGI compiler; if nothing changes, try g95. There is a version > for 64-bit AMD. > > Paolo > > From degironc at sissa.it Mon Feb 14 09:47:19 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 14 Feb 2005 09:47:19 +0100 Subject: [Pw_forum] nscf calculation References: <420E2772.000004.20563@pantene.yandex.ru> Message-ID: <42106597.6010805@sissa.it> Your grid (400 1 1 0 0 0) looks strange to me. why do you need so many points in the b_1 direction ? however... this should not the source of the problem stefano Sergey Lisenkov wrote: >Dear PWscf users and authors, > >I performed a scf calculation of a system (36 atoms). After that I tried to perform a nscf calculation (for DOS calculation). I used 'tetrahedra' method for occupations and automatic generation of k-points (400 1 1 0 0 0). I did not use the 'restart' option, but I got: > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > task # 34 > from davcio : error # 0 > i/o error in davcio > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > >What is wrong in my case? > Sergey >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From proffess at yandex.ru Mon Feb 14 10:21:37 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Mon, 14 Feb 2005 12:21:37 +0300 (MSK) Subject: [Pw_forum] nscf calculation In-Reply-To: <42106597.6010805@sissa.it> References: <420E2772.000004.20563@pantene.yandex.ru> <42106597.6010805@sissa.it> Message-ID: <42106DA1.000008.01467@colgate.yandex.ru> Dear Stefano, Thank you for your message. I calculate 1D-system, so I need to use k-points only in one direction. Probably, I should use less k-points, because I use tetrahedron method. But I found what is a problem. It is a very simple - no enough disk space. I looked into the files and found that the size of 1 wfc file is around 3 Gb, number of wfc files is a number of processors, and size of .save file is around 15 Gb. They are very very huge. This is a source of my problem. So, because, I have no enough disk space for such calculations, is it possible to reduce the size of these files? Thanks, Sergey From proffess at yandex.ru Mon Feb 14 10:41:42 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Mon, 14 Feb 2005 12:41:42 +0300 (MSK) Subject: [Pw_forum] eigenvalues not converged Message-ID: <42107256.000002.22271@soapbox.yandex.ru> Dear PWscf authors, I have such kind warnings and before I looked both in manual and forum. As people explained, this warning can be ignored, if it is not present in the last steps of self-consistency. But I got these warnings during NSCF calculations. Should I ignore them? k( 61) = ( 0.4958678 0.0000000 0.0000000), wk = 0.0330579 G cutoff = 2523.7443 (2055649 G-vectors) FFT grid: (108,200,200) G cutoff = 1261.8722 ( 726943 G-vectors) smooth grid: ( 72,144,144) nbndx = 460 nbnd = 115 natomwfc = 192 npwx = 1825 nelec = 192.00 nkb = 384 ngl = 18515 The initial potential is read from file 44-t2.pot Starting wfc are atomic total cpu time spent up to now is 533.89 secs Band Structure Calculation Davidson diagonalization (with overlap) WARNING: 4 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 1 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 6 eigenvalues not converged WARNING: 6 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 4 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 6 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 5 eigenvalues not converged WARNING: 6 eigenvalues not converged WARNING: 18 eigenvalues not converged WARNING: 1 eigenvalues not converged .... Sergey From degironc at sissa.it Mon Feb 14 10:44:57 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 14 Feb 2005 10:44:57 +0100 Subject: [Pw_forum] nscf calculation References: <420E2772.000004.20563@pantene.yandex.ru> <42106597.6010805@sissa.it> <42106DA1.000008.01467@colgate.yandex.ru> Message-ID: <42107319.9080108@sissa.it> Sergey Lisenkov wrote: >Dear Stefano, > >Thank you for your message. I calculate 1D-system, so I need to use k-points only in one direction. Probably, I should use less k-points, because I use tetrahedron method. But I found what is a problem. It is a very simple - no enough disk space. I looked into the files and found that the size of 1 wfc file is around 3 Gb, number of wfc files is a number of processors, and size of .save file is around 15 Gb. They are very very huge. This is a source of my problem. So, because, I have no enough disk space for such calculations, is it possible to reduce the size of these files? > > If you reduce the number of points the dimension of those files will be correspondingly reduced. I really cannot immagine what system needs 400 points in a given direction. stefano >Thanks, > Sergey >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From proffess at yandex.ru Mon Feb 14 10:53:04 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Mon, 14 Feb 2005 12:53:04 +0300 (MSK) Subject: [Pw_forum] nscf calculation In-Reply-To: <42107319.9080108@sissa.it> References: <420E2772.000004.20563@pantene.yandex.ru> <42106597.6010805@sissa.it> <42106DA1.000008.01467@colgate.yandex.ru> <42107319.9080108@sissa.it> Message-ID: <42107500.000014.01418@pantene.yandex.ru> Dear Stefano, I reduced the number of k-points up to (100 1 1 0 0 0), only 51 k-points in total. But the size of the files is reduces a little. I am not sure that for DOS calculations 51 k-points is enough. Sergey From degironc at sissa.it Mon Feb 14 11:13:29 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 14 Feb 2005 11:13:29 +0100 Subject: [Pw_forum] nscf calculation References: <420E2772.000004.20563@pantene.yandex.ru> <42106597.6010805@sissa.it> <42106DA1.000008.01467@colgate.yandex.ru> <42107319.9080108@sissa.it> <42107500.000014.01418@pantene.yandex.ru> Message-ID: <421079C9.9060100@sissa.it> The disk memory used by pwscf.wfc file is LINEAR in the numer of k-points used. stefano Sergey Lisenkov wrote: >Dear Stefano, > >I reduced the number of k-points up to (100 1 1 0 0 0), only 51 k-points in total. But the size of the files is reduces a little. I am not sure that for DOS calculations 51 k-points is enough. > > Sergey >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > > From verissim at df.ufscar.br Mon Feb 14 15:08:37 2005 From: verissim at df.ufscar.br (Marcos Verissimo Alves) Date: Mon, 14 Feb 2005 12:08:37 -0200 Subject: [Pw_forum] Compilation of PWSCF 2.1.2 In-Reply-To: <30243ba2.549ce163.823ae00@expms6.cites.uiuc.edu> References: <30243ba2.549ce163.823ae00@expms6.cites.uiuc.edu> Message-ID: <1108390117.19792.18.camel@rocknroll.df.ufscar.br> Hi all, I have downloaded the espresso package and I am trying to compile it on a 64bit AMD running Mandrake Linux, using ifort 8.1 and the free mkl libraries. I ran configure and then make pw. Apparently, the clib and flib libraries are built successfully, with no problems reported. Everything goes fine until the compiler tries to link all objects: ld: warning: i386:x86-64 architecture of input file `../clib/clib.a(c_mkdir.o)' is incompatible with i386 output ld: warning: i386:x86-64 architecture of input file `../clib/clib.a(cptimer.o)' is incompatible with i386 output ld: warning: i386:x86-64 architecture of input file `../clib/clib.a(fft_stick.o)' is incompatible with i386 output ld: warning: i386:x86-64 architecture of input file `../clib/clib.a(indici.o)' is incompatible with i386 output ld: warning: i386:x86-64 architecture of input file `../clib/clib.a(qsort.o)' is incompatible with i386 output ifort: error: Fatal error in ld, terminated by segmentation violation make[1]: *** [memory.x] Error 1 make[1]: Leaving directory `/usr/local/software/instala/espresso/PW' make: *** [pw] Error 2 Can anyone help me fix this? Thanks in advance, Marcos -- Dr. Marcos Ver?ssimo Alves Departamento de F?sica Universidade Federal de S?o Carlos S?o Carlos, SP - Brasil http://joselitosn.df.ufscar.br/~dm/equipe/verissim.htm *********************************** Dr. Marcos Verissimo Alves Physics Department Universidade Federal de Sao Carlos Sao Carlos, SP - Brazil http://joselitosn.df.ufscar.br/~dm/equipe/verissim-en.htm From massimiliano.bonomi at mi.infn.it Mon Feb 14 15:13:47 2005 From: massimiliano.bonomi at mi.infn.it (Massimiliano Bonomi) Date: Mon, 14 Feb 2005 15:13:47 +0100 Subject: [Pw_forum] post-processing problems Message-ID: <4210B21B.4040504@mi.infn.it> Dear users, I'm trying to make a 3d charge plot of a gold cluster made of 8 atoms using espresso 2.1.2 on a linux cluster with pgi compiler. The program pp.x stops with this error: Program POST-PROC v.2.1.2 starts ... Today is 14Feb2005 at 14:54:10 Parallel version (MPI) Number of processors in use: 30 R & G space division: nprocp = 30 Reading file out.save ... read complete Reading file out.save ... read complete Planes per process (thick) : nr3 =288 npp = 10 ncplane =82944 Planes per process (smooth): nr3s=216 npps= 8 ncplanes=46656 Proc/ planes cols G planes cols G columns G Pool (dense grid) (smooth grid) (wavefct grid) 1 10 2123 402807 8 1188 168800 307 22139 ...................... 29 9 2123 402805 7 1188 168796 307 22135 30 9 2123 402805 7 1188 168788 306 22134 0 288 63697******* 216 356535063625 9209 664101 nbndx = 53 nbnd = 53 natomwfc = 72 npwx = 21112 nelec = 88.00 nkb = 144 ngl = 16610 *** error in Message Passing (mp) module *** *** error code: 8912 Any idea of the problem?? Can it be due to the big wavefunctions (15 Giga!)?? Thanks for your help... -- Massimiliano Bonomi Dipartimento di Fisica, Universit? degli Studi di Milano via Celoria, 16 - 20133 Milano, Italy Ph. +39 02 50317 717 From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Mon Feb 14 15:28:32 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Mon, 14 Feb 2005 15:28:32 +0100 (CET) Subject: [Pw_forum] Compilation of PWSCF 2.1.2 In-Reply-To: <1108390117.19792.18.camel@rocknroll.df.ufscar.br> Message-ID: On Mon, 14 Feb 2005, Marcos Verissimo Alves wrote: MVA> Hi all, hi! MVA> MVA> I have downloaded the espresso package and I am trying to compile it on MVA> a 64bit AMD running Mandrake Linux, using ifort 8.1 and the free mkl MVA> libraries. I ran configure and then make pw. Apparently, the clib and MVA> flib libraries are built successfully, with no problems reported. MVA> Everything goes fine until the compiler tries to link all objects: MVA> MVA> ld: warning: i386:x86-64 architecture of input file MVA> `../clib/clib.a(c_mkdir.o)' is incompatible with i386 MVA> output you seem to be using the 32bit version of the intel compiler. in that case you also have to compile the c modules in 32bit mode. e.g by using the corresponding icc or 'gcc -m32' instead of gcc as c compiler. to take advantage of the 64bit addressing, however, you need to download the EM64T version of the intel compilers from intel's premier support site (l_fce_pc_8.1.024.tar.gz). note the 'fce' instead of the 'fc' for the 32bit/itanium compiler archive. regards, axel. MVA> ld: warning: i386:x86-64 architecture of input file MVA> `../clib/clib.a(cptimer.o)' is incompatible with i386 MVA> output MVA> ld: warning: i386:x86-64 architecture of input file MVA> `../clib/clib.a(fft_stick.o)' is incompatible with i386 MVA> output MVA> ld: warning: i386:x86-64 architecture of input file MVA> `../clib/clib.a(indici.o)' is incompatible with i386 MVA> output MVA> ld: warning: i386:x86-64 architecture of input file MVA> `../clib/clib.a(qsort.o)' is incompatible with i386 MVA> output MVA> ifort: error: Fatal error in ld, terminated by segmentation MVA> violation make[1]: *** [memory.x] Error 1 MVA> make[1]: Leaving directory MVA> `/usr/local/software/instala/espresso/PW' make: *** [pw] Error 2 MVA> MVA> Can anyone help me fix this? MVA> MVA> Thanks in advance, MVA> MVA> Marcos MVA> MVA> -- ======================================================================= Dr. Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/ ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Mon Feb 14 15:32:47 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 14 Feb 2005 15:32:47 +0100 Subject: [Pw_forum] post-processing problems In-Reply-To: <4210B21B.4040504@mi.infn.it> References: <4210B21B.4040504@mi.infn.it> Message-ID: <200502141532.47701.giannozz@nest.sns.it> On Monday 14 February 2005 15:13, Massimiliano Bonomi wrote: > *** error in Message Passing (mp) module *** > *** error code: 8912 see http://www.democritos.it/pipermail/pw_forum/2005-February/001990.html http://www.democritos.it/pipermail/pw_forum/2005-February/001992.html -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Feb 14 15:41:49 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 14 Feb 2005 15:41:49 +0100 Subject: [Pw_forum] Re: Re: parallel ph.x in Opteron (Paolo Giannozzi) In-Reply-To: References: Message-ID: <200502141541.49081.giannozz@nest.sns.it> On Sunday 13 February 2005 23:38, wlyim at puccini.che.pitt.edu wrote: > for a specific object file dvqpsi_us_only.o, I needed to lower the > optimization level [...]. So, now my parallel pw.x and ph.x works, > where all object files were compiled by default setting and > dvqpsi_us_only.o by specific setting. now I remember that I had a similar problem with the PGI compiler a few weeks ago. It also turned out to be a problem with optimization in that routine > Any bug in dvqpsi_us_only.f90 or PGI 5.1 compiler? to the best of my knowledge there is nothing wrong in dvqpsi_us_only.f90 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From konstantin_kudin at yahoo.com Mon Feb 14 21:09:51 2005 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Mon, 14 Feb 2005 12:09:51 -0800 (PST) Subject: [Pw_forum] parallel ph.x in Opteron In-Reply-To: <200502121506.58730.giannozz@nest.sns.it> Message-ID: <20050214200952.31099.qmail@web52009.mail.yahoo.com> --- Paolo Giannozzi wrote: > it looks like a compiler problem. Get the latest available patches > for the PGI compiler; if nothing changes, try g95. There is a version > > for 64-bit AMD. > And of course IFC 8.1 is always an option. It is downloadable from Intel's Premier support site. 64-bit IFC 8.1 works on Opteron just fine. The 64-bit version is very similar to the 32-bit one, with pretty much the same compiler options. Kostya __________________________________ Do you Yahoo!? Take Yahoo! Mail with you! Get it on your mobile phone. http://mobile.yahoo.com/maildemo From proffess at yandex.ru Tue Feb 15 09:10:39 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Tue, 15 Feb 2005 11:10:39 +0300 (MSK) Subject: [Pw_forum] DOS calculation Message-ID: <4211AE7F.000050.00991@tide.yandex.ru> Dear PWscf authors and users, I calculate the DOS using the next region: Emin=-8.0 Emax=8.0 DeltaE=0.05 But in output file I see the next region: # E (eV) dos(E) Int dos(E) -8.000 0.1333E+02 0.6664E+00 ......... 3.000 0.1065E+02 0.1192E+03 3.050 0.1748E+01 0.1192E+03 The question is: why the region is different that I put? Sergey From degironc at sissa.it Tue Feb 15 09:16:40 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 15 Feb 2005 09:16:40 +0100 Subject: [Pw_forum] DOS calculation References: <4211AE7F.000050.00991@tide.yandex.ru> Message-ID: <4211AFE8.9040704@sissa.it> Sergey Lisenkov wrote: >Dear PWscf authors and users, > >I calculate the DOS using the next region: > >Emin=-8.0 Emax=8.0 DeltaE=0.05 > >But in output file I see the next region: > ># E (eV) dos(E) Int dos(E) > -8.000 0.1333E+02 0.6664E+00 >......... > 3.000 0.1065E+02 0.1192E+03 > 3.050 0.1748E+01 0.1192E+03 > >The question is: why the region is different that I put? > > do you have enough bands to reach +8 eV ? stefano From proffess at yandex.ru Tue Feb 15 09:23:25 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Tue, 15 Feb 2005 11:23:25 +0300 (MSK) Subject: [Pw_forum] DOS calculation In-Reply-To: <4211AFE8.9040704@sissa.it> References: <4211AE7F.000050.00991@tide.yandex.ru> <4211AFE8.9040704@sissa.it> Message-ID: <4211B17D.000003.04755@soapbox.yandex.ru> Dear Stefano, >do you have enough bands to reach +8 eV ? The number of bands is 115. Does it mean that the region I put is fictitious? I can change only Emin, bu Emax is always the same. Sergey From giannozz at nest.sns.it Tue Feb 15 09:56:30 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 15 Feb 2005 09:56:30 +0100 Subject: [Pw_forum] DOS calculation In-Reply-To: <4211AE7F.000050.00991@tide.yandex.ru> References: <4211AE7F.000050.00991@tide.yandex.ru> Message-ID: <200502150956.30688.giannozz@nest.sns.it> On Tuesday 15 February 2005 09:10, Sergey Lisenkov wrote: > I calculate the DOS using the next region: > > Emin=-8.0 Emax=8.0 DeltaE=0.05 > > But in output file I see the next region: > > # E (eV) dos(E) Int dos(E) > -8.000 0.1333E+02 0.6664E+00 > ......... > 3.000 0.1065E+02 0.1192E+03 > 3.050 0.1748E+01 0.1192E+03 > > The question is: why the region is different that I put? Emax is set to min(Emax, Eup), where Eup is the upper band energy, plus 3 times the gaussian broadening (if any). There shouldn't be any eigenvalue beyond 3.05 eV. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From degironc at sissa.it Tue Feb 15 17:29:54 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Tue, 15 Feb 2005 17:29:54 +0100 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential References: Message-ID: <42122382.6090008@sissa.it> What about convergence with respect to ecutwfc and ecutrho ? stefano Alcantara Ortigoza, Marisol wrote: >Hi, > >I used the default value for convergence threshold: 1e-6. And, for this >threshold the wavefunction converged well. > >Thank you. > >Marisol. > > > From Myeong.Lee at asu.edu Tue Feb 15 22:40:00 2005 From: Myeong.Lee at asu.edu (Myeong.Lee at asu.edu) Date: Tue, 15 Feb 2005 14:40:00 -0700 (MST) Subject: [Pw_forum] complex band structure Message-ID: Dear users, Has any of you ever calculated complex band strucuture? Then, has any of you got bands for positive imaginary k? Thanks. From eyvaz_isaev at yahoo.com Tue Feb 15 23:55:01 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 15 Feb 2005 14:55:01 -0800 (PST) Subject: [Pw_forum] Typos In-Reply-To: Message-ID: <20050215225503.7373.qmail@web60309.mail.yahoo.com> Hi, I would like to note that there is just a typos in README file for example03. Namely, for the second example where Al(001) surface relaxation is considered. According to the text ibrav=4, which is wrong, because ibrav=4 is for a hexagonal lattice. So, it should be really tetragonal with ibrav=6. Bests, Eyvaz. __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail From virginie.quequet at polytechnique.fr Thu Feb 17 16:22:10 2005 From: virginie.quequet at polytechnique.fr (Virginie Quequet) Date: 17 Feb 2005 16:22:10 +0100 Subject: [Pw_forum] cut-off radius for radial integrations Message-ID: <1108653730.22428.25.camel@gurosai.polytechnique.fr> Hello I try to change to 60 a.u. the cut-off radius for radial integration, which is put to 10 a.u. in the code (rcut variable). It seems to me that it should change very little the results, but it is not the case. The pressure change a lot and when I try to find the equilibrium lattice parameter, the pressure is like a sinusoide and the minimum of the energy is so far that I didn't continue searching it. My pseudopotential is definied up to 60 a.u, so I don't understand why the calcul is so false. Do you have a clue? Thanks -- Virginie Quequet From alcantar at phys.ksu.edu Thu Feb 17 17:23:51 2005 From: alcantar at phys.ksu.edu (Alcantara Ortigoza, Marisol) Date: Thu, 17 Feb 2005 10:23:51 -0600 Subject: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential Message-ID: Dear Stefano, Yes, you're right. That is the answer. For the "good" calculation (new lattice constant) I was using ecutwfc = 30 and ecutrho = 240 ryd, while for the "wrong" calculation (old lattice constant) I used the default values (ecutwfc =25 ryd, ecutrho =100 ryd). So, after you mentioned the ecutwfc I repeat the calculation with the old lattice constant but with ecutwfc = 30 and ecutrho = 240 ryd. I found quite similar results to the ones obtained with the new lattice constant; particularly the relaxation of first and second layer are almost the same (3.09% and 0.37% respectively), although the bulk-like layers behaved a little bit different. Stress is quite small too. I guess now I need to check the convergence of the energy cutoff just to make sure I can relay on these values. Thank you very much all of you. Marisol -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Stefano de Gironcoli Sent: Tuesday, February 15, 2005 10:30 AM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Relaxation of Cu100 with PBE pseudopotential What about convergence with respect to ecutwfc and ecutrho ? stefano Alcantara Ortigoza, Marisol wrote: >Hi, > >I used the default value for convergence threshold: 1e-6. And, for this >threshold the wavefunction converged well. > >Thank you. > >Marisol. > > > _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From matteoc at MIT.EDU Fri Feb 18 03:21:31 2005 From: matteoc at MIT.EDU (Matteo Cococcioni) Date: Thu, 17 Feb 2005 21:21:31 -0500 Subject: [Pw_forum] structural relaxation with espresso 2.1.2 in parallel Message-ID: <1108693291.4215512bd17a7@webmail.mit.edu> Dear pwscf users in our lab we have found some problems in doing structural relaxation with the latest version of the espresso package (2.1.2). We find that, using bfgs (or old-bfgs) algorithm, the structural optimization sistematically stops after the second ionic step. The process doesn't crash but freezes without any error message, after writing: NEW-OLD atomic charge density approx. for the potential We are compiling the code with the latest intel compiler (ifc_8.1.023) and using the lam mpi parallel environment (lam 7.0.6 compiled with icc_8.1.026 besides the above fortran compiler). The problem only appears in parallel jobs and seems to be somehow related to the architecture of the machine. In fact it only shows up if we run the job on more than one node. On the contrary, if we run a parallel job on the two proc of a dual xeon for instance, the problem doesn't appear. We have tried to increase the parameter mp_msgsiz_max in the Modules/mp.f90 routine as suggested few days ago by Paul Tangney and Paolo Giannozzi, but it didn't solve the problem. Actually the problem seems to be unrelated to the size of the system: we tried with very small ones (as the one whose input is attached below) and we had the same situation. We thought it could be a problem related to the use of several local disks on different machines, but using a shared directory as outdir didn't help. Does anyone have any idea to solve this problem? Many thanks in advance for any help. Best regards, Matteo %%%%%%%%%%%%%%%%%%%%%% Input file %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% &control calculation = 'relax' restart_mode='from_scratch' nstep=20 iprint=10 outdir='./' pseudo_dir = './' / &system ibrav= 8 celldm(1) = 4.0d0 celldm(2) = 1.0d0 celldm(3) = 2.0d0 nat= 2 ntyp= 2 nelec=10 nbnd=10 nspin = 2 ecutwfc =25.0 ecutrho =200.0 occupations = 'smearing' degauss=0.0024 xc_type = 'PBE' starting_magnetization(1)=0.0 / &electrons electron_maxstep = 200 mixing_beta = 0.3 conv_thr = 1.0d-8 / &ions / ATOMIC_SPECIES C 12.0107d0 CUSPBE.RRKJ3 O 15.9994d0 O.PBE.US.RRKJ3 ATOMIC_POSITIONS {bohr} C 2.0 2.0 2.0 1 1 1 O 2.0 2.0 5.0 1 1 1 K_POINTS {automatic} 1 1 1 0 0 0 From konstantin_kudin at yahoo.com Fri Feb 18 04:04:30 2005 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Thu, 17 Feb 2005 19:04:30 -0800 (PST) Subject: [Pw_forum] structural relaxation with espresso 2.1.2 in parallel In-Reply-To: <1108693291.4215512bd17a7@webmail.mit.edu> Message-ID: <20050218030430.93334.qmail@web52007.mail.yahoo.com> Dear Matteo, This is a known issue. See this link for details: http://www.democritos.it/pipermail/pw_forum/2005-January/001923.html Kostya --- Matteo Cococcioni wrote: > > Dear pwscf users > > in our lab we have found some problems in doing structural relaxation > with the > latest version of the espresso package (2.1.2). We find that, using > bfgs (or > old-bfgs) algorithm, the structural optimization sistematically stops > after the > second ionic step. The process doesn't crash but freezes without any > error > message, after writing: > > NEW-OLD atomic charge density approx. for the potential > > We are compiling the code with the latest intel compiler > (ifc_8.1.023) and using > the lam mpi parallel environment (lam 7.0.6 compiled with icc_8.1.026 > besides > the above fortran compiler). > > The problem only appears in parallel jobs and seems to be somehow > related to the > architecture of the machine. In fact it only shows up if we run the > job on more > than one node. On the contrary, if we run a parallel job on the two > proc of a > dual xeon for instance, the problem doesn't appear. > > We have tried to increase the parameter mp_msgsiz_max in the > Modules/mp.f90 > routine as suggested few days ago by Paul Tangney and Paolo > Giannozzi, but it > didn't solve the problem. Actually the problem seems to be unrelated > to the > size of the system: we tried with very small ones (as the one whose > input is > attached below) and we had the same situation. > > We thought it could be a problem related to the use of several local > disks on > different machines, but using a shared directory as outdir didn't > help. > > Does anyone have any idea to solve this problem? > > Many thanks in advance for any help. > > Best regards, > > Matteo > > > > %%%%%%%%%%%%%%%%%%%%%% Input file %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > &control > calculation = 'relax' > restart_mode='from_scratch' > nstep=20 > iprint=10 > outdir='./' > pseudo_dir = './' > / > &system > ibrav= 8 > celldm(1) = 4.0d0 > celldm(2) = 1.0d0 > celldm(3) = 2.0d0 > nat= 2 > ntyp= 2 > nelec=10 > nbnd=10 > nspin = 2 > ecutwfc =25.0 > ecutrho =200.0 > occupations = 'smearing' > degauss=0.0024 > xc_type = 'PBE' > starting_magnetization(1)=0.0 > / > &electrons > electron_maxstep = 200 > mixing_beta = 0.3 > conv_thr = 1.0d-8 > / > &ions > / > ATOMIC_SPECIES > C 12.0107d0 CUSPBE.RRKJ3 > O 15.9994d0 O.PBE.US.RRKJ3 > ATOMIC_POSITIONS {bohr} > C 2.0 2.0 2.0 1 1 1 > O 2.0 2.0 5.0 1 1 1 > K_POINTS {automatic} > 1 1 1 0 0 0 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail From proffess at yandex.ru Fri Feb 18 11:14:21 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 18 Feb 2005 13:14:21 +0300 (MSK) Subject: [Pw_forum] DOS calculation In-Reply-To: <200502150956.30688.giannozz@nest.sns.it> References: <4211AE7F.000050.00991@tide.yandex.ru> <200502150956.30688.giannozz@nest.sns.it> Message-ID: <4215BFFD.000030.10952@soapbox.yandex.ru> Dear Stefano and Paolo, Thank you very much for your replies. I plotted the DOS and I wand to know where is Fermi level? Sergey From giannozz at nest.sns.it Fri Feb 18 11:37:29 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 18 Feb 2005 11:37:29 +0100 Subject: [Pw_forum] DOS calculation In-Reply-To: <4215BFFD.000030.10952@soapbox.yandex.ru> References: <4211AE7F.000050.00991@tide.yandex.ru> <200502150956.30688.giannozz@nest.sns.it> <4215BFFD.000030.10952@soapbox.yandex.ru> Message-ID: <200502181137.29682.giannozz@nest.sns.it> On Friday 18 February 2005 11:14, Sergey Lisenkov wrote: > Thank you very much for your replies. I plotted the DOS > and I wand to know where is Fermi level? in the output of the scf calculation. Or otherwise: plot the integrated DOS (it is also calculated and printed in the same file) and see where it crosses the number of electrons P. -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From proffess at yandex.ru Fri Feb 18 12:20:14 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 18 Feb 2005 14:20:14 +0300 (MSK) Subject: [Pw_forum] DOS calculation In-Reply-To: <200502181137.29682.giannozz@nest.sns.it> References: <4211AE7F.000050.00991@tide.yandex.ru> <200502150956.30688.giannozz@nest.sns.it> <4215BFFD.000030.10952@soapbox.yandex.ru> <200502181137.29682.giannozz@nest.sns.it> Message-ID: <4215CF6E.00000B.30703@colgate.yandex.ru> Dear Paolo, >in the output of the scf calculation. Or otherwise: plot the >integrated DOS (it is also calculated and printed in the same >file) and see where it crosses the number of electrons I did not use the 'tetrahedra' option for scf calculations. I will restart them with this option and also will look in DOS output. Thanks. Sergey From giannozz at nest.sns.it Fri Feb 18 14:56:32 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 18 Feb 2005 14:56:32 +0100 Subject: [Pw_forum] cut-off radius for radial integrations In-Reply-To: <1108653730.22428.25.camel@gurosai.polytechnique.fr> References: <1108653730.22428.25.camel@gurosai.polytechnique.fr> Message-ID: <200502181456.32538.giannozz@nest.sns.it> On Thursday 17 February 2005 16:22, Virginie Quequet wrote: > I try to change to 60 a.u. the cut-off radius for radial integration, you shouldn't: the radial integration is performed on short-range functions only. Beyond 10 a.u. all you have is numerical noise. > which is put to 10 a.u. in the code (rcut variable). It seems to me > that it should change very little the results, but it is not the case. please provide a test case (input and output) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From matteoc at MIT.EDU Fri Feb 18 15:19:21 2005 From: matteoc at MIT.EDU (Matteo Cococcioni) Date: Fri, 18 Feb 2005 09:19:21 -0500 Subject: [Pw_forum] structural relaxation with espresso 2.1.2 in parallel In-Reply-To: <20050218030430.93334.qmail@web52007.mail.yahoo.com> References: <20050218030430.93334.qmail@web52007.mail.yahoo.com> Message-ID: <1108736361.4215f96918c3c@webmail.mit.edu> Hi Konstantin and thank you very much for your suggestion. I apology with the whole pwscf community for bothering with known and solved problems. Matteo Quoting Konstantin Kudin : > Dear Matteo, > > This is a known issue. See this link for details: > > http://www.democritos.it/pipermail/pw_forum/2005-January/001923.html > > Kostya > > --- Matteo Cococcioni wrote: > > > > > Dear pwscf users > > > > in our lab we have found some problems in doing structural relaxation > > with the > > latest version of the espresso package (2.1.2). We find that, using > > bfgs (or > > old-bfgs) algorithm, the structural optimization sistematically stops > > after the > > second ionic step. The process doesn't crash but freezes without any > > error > > message, after writing: > > > > NEW-OLD atomic charge density approx. for the potential > > > > We are compiling the code with the latest intel compiler > > (ifc_8.1.023) and using > > the lam mpi parallel environment (lam 7.0.6 compiled with icc_8.1.026 > > besides > > the above fortran compiler). > > > > The problem only appears in parallel jobs and seems to be somehow > > related to the > > architecture of the machine. In fact it only shows up if we run the > > job on more > > than one node. On the contrary, if we run a parallel job on the two > > proc of a > > dual xeon for instance, the problem doesn't appear. > > > > We have tried to increase the parameter mp_msgsiz_max in the > > Modules/mp.f90 > > routine as suggested few days ago by Paul Tangney and Paolo > > Giannozzi, but it > > didn't solve the problem. Actually the problem seems to be unrelated > > to the > > size of the system: we tried with very small ones (as the one whose > > input is > > attached below) and we had the same situation. > > > > We thought it could be a problem related to the use of several local > > disks on > > different machines, but using a shared directory as outdir didn't > > help. > > > > Does anyone have any idea to solve this problem? > > > > Many thanks in advance for any help. > > > > Best regards, > > > > Matteo > > > > > > > > %%%%%%%%%%%%%%%%%%%%%% Input file %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > &control > > calculation = 'relax' > > restart_mode='from_scratch' > > nstep=20 > > iprint=10 > > outdir='./' > > pseudo_dir = './' > > / > > &system > > ibrav= 8 > > celldm(1) = 4.0d0 > > celldm(2) = 1.0d0 > > celldm(3) = 2.0d0 > > nat= 2 > > ntyp= 2 > > nelec=10 > > nbnd=10 > > nspin = 2 > > ecutwfc =25.0 > > ecutrho =200.0 > > occupations = 'smearing' > > degauss=0.0024 > > xc_type = 'PBE' > > starting_magnetization(1)=0.0 > > / > > &electrons > > electron_maxstep = 200 > > mixing_beta = 0.3 > > conv_thr = 1.0d-8 > > / > > &ions > > / > > ATOMIC_SPECIES > > C 12.0107d0 CUSPBE.RRKJ3 > > O 15.9994d0 O.PBE.US.RRKJ3 > > ATOMIC_POSITIONS {bohr} > > C 2.0 2.0 2.0 1 1 1 > > O 2.0 2.0 5.0 1 1 1 > > K_POINTS {automatic} > > 1 1 1 0 0 0 > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > > __________________________________ > Do you Yahoo!? > Yahoo! Mail - Helps protect you from nasty viruses. > http://promotions.yahoo.com/new_mail > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From proffess at yandex.ru Mon Feb 21 08:51:28 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Mon, 21 Feb 2005 10:51:28 +0300 (MSK) Subject: [Pw_forum] Strange last k-point Message-ID: <42199300.000024.21568@colgate.yandex.ru> Dear PWscf authors and users, I performed nscf calculation with tetrahedra option and automatic generated k-points (200 1 1 0 0 0). For me it seems to be very strange the coordinate of the last k-point: number of k points= 101 (tetrahedron method) cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0100000 k( 2) = ( 0.0050000 0.0000000 0.0000000), wk = 0.0200000 k( 3) = ( 0.0100000 0.0000000 0.0000000), wk = 0.0200000 k( 4) = ( 0.0150000 0.0000000 0.0000000), wk = 0.0200000 k( 5) = ( 0.0200000 0.0000000 0.0000000), wk = 0.0200000 k( 6) = ( 0.0250000 0.0000000 0.0000000), wk = 0.0200000 k( 7) = ( 0.0300000 0.0000000 0.0000000), wk = 0.0200000 k( 8) = ( 0.0350000 0.0000000 0.0000000), wk = 0.0200000 k( 9) = ( 0.0400000 0.0000000 0.0000000), wk = 0.0200000 k( 10) = ( 0.0450000 0.0000000 0.0000000), wk = 0.0200000 k( 11) = ( 0.0500000 0.0000000 0.0000000), wk = 0.0200000 k( 12) = ( 0.0550000 0.0000000 0.0000000), wk = 0.0200000 k( 13) = ( 0.0600000 0.0000000 0.0000000), wk = 0.0200000 k( 14) = ( 0.0650000 0.0000000 0.0000000), wk = 0.0200000 k( 15) = ( 0.0700000 0.0000000 0.0000000), wk = 0.0200000 .... k( 93) = ( 0.4600000 0.0000000 0.0000000), wk = 0.0200000 k( 94) = ( 0.4650000 0.0000000 0.0000000), wk = 0.0200000 k( 95) = ( 0.4700000 0.0000000 0.0000000), wk = 0.0200000 k( 96) = ( 0.4750000 0.0000000 0.0000000), wk = 0.0200000 k( 97) = ( 0.4800000 0.0000000 0.0000000), wk = 0.0200000 k( 98) = ( 0.4850000 0.0000000 0.0000000), wk = 0.0200000 k( 99) = ( 0.4900000 0.0000000 0.0000000), wk = 0.0200000 k( 100) = ( 0.4950000 0.0000000 0.0000000), wk = 0.0200000 k( 101) = ( -0.5000000 0.0000000 0.0000000), wk = 0.0100000 Why is '-0.5 0.0 0.0' ? I gess it should be 0.5 0.0 0.0 Thanks, Sergey From baroni at sissa.it Mon Feb 21 08:58:49 2005 From: baroni at sissa.it (Stefano Baroni) Date: Mon, 21 Feb 2005 08:58:49 +0100 Subject: [Pw_forum] Strange last k-point In-Reply-To: <42199300.000024.21568@colgate.yandex.ru> References: <42199300.000024.21568@colgate.yandex.ru> Message-ID: <3a2c142a55e849b64cde2f8e940d6a6d@sissa.it> I do not really have a clue, but (x,0,0) is equivalent to (-x,0,0) because of time-reversal invariance. Whatever the reason might be for the k-point generator to stick a minus sign, it should make no difference ... Cheers, Stefano B. On Feb 21, 2005, at 8:51 AM, Sergey Lisenkov wrote: > Dear PWscf authors and users, > > I performed nscf calculation with tetrahedra option and automatic > generated k-points (200 1 1 0 0 0). For me it seems to be very strange > the coordinate of the last k-point: > > number of k points= 101 (tetrahedron method) > cart. coord. in units 2pi/a_0 > k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = > 0.0100000 > k( 2) = ( 0.0050000 0.0000000 0.0000000), wk = > 0.0200000 > k( 3) = ( 0.0100000 0.0000000 0.0000000), wk = > 0.0200000 > k( 4) = ( 0.0150000 0.0000000 0.0000000), wk = > 0.0200000 > k( 5) = ( 0.0200000 0.0000000 0.0000000), wk = > 0.0200000 > k( 6) = ( 0.0250000 0.0000000 0.0000000), wk = > 0.0200000 > k( 7) = ( 0.0300000 0.0000000 0.0000000), wk = > 0.0200000 > k( 8) = ( 0.0350000 0.0000000 0.0000000), wk = > 0.0200000 > k( 9) = ( 0.0400000 0.0000000 0.0000000), wk = > 0.0200000 > k( 10) = ( 0.0450000 0.0000000 0.0000000), wk = > 0.0200000 > k( 11) = ( 0.0500000 0.0000000 0.0000000), wk = > 0.0200000 > k( 12) = ( 0.0550000 0.0000000 0.0000000), wk = > 0.0200000 > k( 13) = ( 0.0600000 0.0000000 0.0000000), wk = > 0.0200000 > k( 14) = ( 0.0650000 0.0000000 0.0000000), wk = > 0.0200000 > k( 15) = ( 0.0700000 0.0000000 0.0000000), wk = > 0.0200000 > .... > k( 93) = ( 0.4600000 0.0000000 0.0000000), wk = > 0.0200000 > k( 94) = ( 0.4650000 0.0000000 0.0000000), wk = > 0.0200000 > k( 95) = ( 0.4700000 0.0000000 0.0000000), wk = > 0.0200000 > k( 96) = ( 0.4750000 0.0000000 0.0000000), wk = > 0.0200000 > k( 97) = ( 0.4800000 0.0000000 0.0000000), wk = > 0.0200000 > k( 98) = ( 0.4850000 0.0000000 0.0000000), wk = > 0.0200000 > k( 99) = ( 0.4900000 0.0000000 0.0000000), wk = > 0.0200000 > k( 100) = ( 0.4950000 0.0000000 0.0000000), wk = > 0.0200000 > k( 101) = ( -0.5000000 0.0000000 0.0000000), wk = > 0.0100000 > > Why is '-0.5 0.0 0.0' ? I gess it should be 0.5 0.0 0.0 > > Thanks, > Sergey > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --- Stefano Baroni --- SISSA & DEMOCRITOS National Simulation Center via Beirut 2-4 34014 Trieste Grignano / [+39] 040 3787 406 (tel) -528 (fax) Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 2987 bytes Desc: not available Url : /pipermail/attachments/20050221/82efc0a6/attachment.bin From aaron at chips.ncsu.edu Sun Feb 20 21:15:01 2005 From: aaron at chips.ncsu.edu (aaron at chips.ncsu.edu) Date: Sun, 20 Feb 2005 15:15:01 -0500 (EST) Subject: [Pw_forum] rdiaghg in pw.x -- NEB In-Reply-To: <1108653730.22428.25.camel@gurosai.polytechnique.fr> Message-ID: Hi, I am getting the following error in the january '05 release of ESPRESSO. The calculation is an neb. The platform is ibm-sp4. The offended routine is a library call in lapack (essl I believe). In the flow of neb this happens after the first set of energies are calculated. A few images are converged and then one fails. Any suggestions? DPOTRF : 2538-2148 The matrix (ARG NO. 3) is not positive definite. The leading minor of order (205) has a nonpositive determinant. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from rdiaghg : error # 410 info =/= 0 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... Thanks Aaron From bth20 at cam.ac.uk Mon Feb 21 14:35:45 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Mon, 21 Feb 2005 13:35:45 -0000 Subject: [Pw_forum] relax calculation including spin Message-ID: Hi I am running a relax calculation on a cobalt structure. When I turn spin off (nspin = 1) it converges. When I have spin on (nspin = 2) it does not converge - it fails on one of the self-consistent calculations. That is it reaches iteration 100 without falling within the convergence threshold which I've set as easy as 10d-5. I'm not sure how to proceed other than setting the threshold even higher (mixing beta already at 0.1). Is it possible to allow it further iterations? Can I get the program to continue with the relaxation calculation even though it's not converged to within the set limit? Thanks Ben. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050221/eeea2179/attachment.htm From bth20 at cam.ac.uk Mon Feb 21 15:20:30 2005 From: bth20 at cam.ac.uk (Ben Hope) Date: Mon, 21 Feb 2005 14:20:30 -0000 Subject: [Pw_forum] relaxation via forces Message-ID: Also, Is it possible to get pwscf to perform the relaxation purely by minimizing forces rather than via a total energy calculation? Ben -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20050221/f5fc3bc6/attachment.htm From giannozz at nest.sns.it Mon Feb 21 16:29:11 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 21 Feb 2005 16:29:11 +0100 Subject: [Pw_forum] rdiaghg in pw.x -- NEB In-Reply-To: References: Message-ID: <200502211629.11648.giannozz@nest.sns.it> On Sunday 20 February 2005 21:15, aaron at chips.ncsu.edu wrote: > DPOTRF : 2538-2148 > The matrix (ARG NO. 3) is not positive definite. > The leading minor of order (205) has a nonpositive determinant. I am afraid you are out of luck. For some reason your overlap matrix (appearing in the ultrasoft pseudopotential formalism) is not positive definite as it should be. I inquired some time ago on a similar case and couldn't find any good reason (and not even a bad one) for such behavior. So the only suggestion is to use conjugate-gradient diagonalization (slow but safe) or diis (fast but very unsafe). Please post from the same e-mail you used to register: posts appearing to come from non-registered e-mails are held for moderator approval. Due to the large number of spam messages, such posts are automatically deleted after a while. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From aloysius at physics.usyd.edu.au Tue Feb 22 00:55:26 2005 From: aloysius at physics.usyd.edu.au (Aloysius Soon) Date: Tue, 22 Feb 2005 10:55:26 +1100 Subject: [Pw_forum] Band structure calculation for Cu2O Message-ID: <421A74EE.1010101@physics.usyd.edu.au> Dear all, I'm hoping to calculate the band structure for Cu2O but is having a hard time finding the high symmetry points for this structure. Can any kind soul please help? Any sort of data base or reference text will be greatly appreciated. Cheers! Regards, Aloysius -- Aloysius Soon SL Ph.D. Research Student Condensed Matter Theory Group School of Physics A28, Room 361 The University of Sydney, Australia Phone: +61 2 903 65389 Fax: +61 2 935 17726 Email: aloysius at physics.usyd.edu.au Web: http://www.physics.usyd.edu.au/~aloysius/ From eyvaz_isaev at yahoo.com Tue Feb 22 13:47:23 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 22 Feb 2005 04:47:23 -0800 (PST) Subject: [Pw_forum] Band structure calculation for Cu2O In-Reply-To: <421A74EE.1010101@physics.usyd.edu.au> Message-ID: <20050222124723.66001.qmail@web60304.mail.yahoo.com> Hi, According to webelements.com Cu2O is in a simple cubic lattice. So, you can use the data contained in a message I sent some months ago. The message contained an attached file with explanation how the band structure could be constructed. Nevertheless, high symmetry points for a simple cubic lattice are: Gamma - 0, 0, 0 X - 1/2, 0, 0 M - 1/2,1/2,0 R -1/2,1/2,1/2 Bests, Eyvaz. --- Aloysius Soon wrote: > Dear all, > I'm hoping to calculate the band structure for Cu2O > but is > having a hard time finding the high symmetry points > for this > structure. > Can any kind soul please help? Any sort of data base > or > reference text will be greatly appreciated. > Cheers! > > > > Regards, > Aloysius > > -- > Aloysius Soon SL > Ph.D. Research Student > Condensed Matter Theory Group > School of Physics A28, Room 361 > The University of Sydney, Australia > Phone: +61 2 903 65389 > Fax: +61 2 935 17726 > Email: aloysius at physics.usyd.edu.au > Web: http://www.physics.usyd.edu.au/~aloysius/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? The all-new My Yahoo! - What will yours do? http://my.yahoo.com From virginie.quequet at polytechnique.fr Tue Feb 22 14:46:09 2005 From: virginie.quequet at polytechnique.fr (Virginie Quequet) Date: 22 Feb 2005 14:46:09 +0100 Subject: [Pw_forum] Re: cut-off radius for radial integrations In-Reply-To: <20050219063855.3568.70972.Mailman@democritos.sissa.it> References: <20050219063855.3568.70972.Mailman@democritos.sissa.it> Message-ID: <1109079969.11326.90.camel@gurosai.polytechnique.fr> Tank you for your reply. It seems quite systematic, because my first example was for titanium and I tried for silicon with the input of example01 (david) with a pseudo-potential with non-local part defined up to 60 a.u. and I obtain this rcut = 10 a.u. ! total energy = -15.89013028 ryd estimated scf accuracy < 6.7E-09 ryd band energy sum = 0.62621995 ryd one-electron contribution = 4.74024554 ryd hartree contribution = 1.07806794 ryd xc contribution = -4.80868510 ryd ewald contribution = -16.89975867 ryd convergence has been achieved Forces acting on atoms (Ry/au): atom 1 type 1 force = 0.00000000 0.00000000 0.00000000 atom 2 type 1 force = 0.00000000 0.00000000 0.00000000 Total force = 0.000000 Total SCF correction = 0.000000 entering subroutine stress ... total stress (ryd/bohr**3) (kbar) P= -17.45 -0.00011860 0.00000000 0.00000000 -17.45 0.00 0.00 0.00000000 -0.00011860 0.00000000 0.00 -17.45 0.00 0.00000000 0.00000000 -0.00011860 0.00 0.00 -17.45 rcut = 60 a.u. ! total energy = -15.90406433 ryd estimated scf accuracy < 5.9E-09 ryd band energy sum = 0.61224706 ryd one-electron contribution = 4.72633379 ryd hartree contribution = 1.07803417 ryd xc contribution = -4.80867362 ryd ewald contribution = -16.89975867 ryd convergence has been achieved Forces acting on atoms (Ry/au): atom 1 type 1 force = 0.00000000 0.00000000 0.00000000 atom 2 type 1 force = 0.00000000 0.00000000 0.00000000 Total force = 0.000000 Total SCF correction = 0.000000 entering subroutine stress ... total stress (ryd/bohr**3) (kbar) P= -25.60 -0.00017399 0.00000000 0.00000000 -25.60 0.00 0.00 0.00000000 -0.00017399 0.00000000 0.00 -25.60 0.00 0.00000000 0.00000000 -0.00017399 0.00 0.00 -25.60 Virginie Quequet On Thursday 17 February 2005 16:22, Virginie Quequet wrote: > I try to change to 60 a.u. the cut-off radius for radial integration, you shouldn't: the radial integration is performed on short-range functions only. Beyond 10 a.u. all you have is numerical noise. > which is put to 10 a.u. in the code (rcut variable). It seems to me > that it should change very little the results, but it is not the case. please provide a test case (input and output) Paolo From yonasb at yahoo.com Tue Feb 22 17:16:23 2005 From: yonasb at yahoo.com (Yonas Abraham) Date: Tue, 22 Feb 2005 08:16:23 -0800 (PST) Subject: [Pw_forum] pwcond.x reference Message-ID: <20050222161624.78541.qmail@web60807.mail.yahoo.com> Hi, did any one know what reference was used to implement the complex band structure calculation (pwcond.x). I try to look at the source code or the INITI_PWCOND file for reference and didn't find any. I have PRB 70 045417 (2004) by Alexander Smognunov. did by any chance that the code is based on that paper? thankx yonas From ml_wangus at yahoo.com Tue Feb 22 18:25:12 2005 From: ml_wangus at yahoo.com (Ming Wang) Date: Tue, 22 Feb 2005 09:25:12 -0800 (PST) Subject: [Pw_forum] Is E_cut higher, the results is better? Message-ID: <20050222172512.70516.qmail@web51902.mail.yahoo.com> Hi, I used different E_cut(35 Ryd, 135Ryd) to calculate molecular polarizability. The results are a little difference, I wonder whether the results from high E_cut is more reliable than lower E_cut results. I realized the E_cut is around 25 Ryd when the USPP is generated. Does this mean I should choose E_cut around 25 Ryd? Thanks! Ming __________________________________ Do you Yahoo!? The all-new My Yahoo! - Get yours free! http://my.yahoo.com From varki at ts.infn.it Tue Feb 22 16:17:09 2005 From: varki at ts.infn.it (Varki Pagan (Peressi)) Date: Tue, 22 Feb 2005 16:17:09 +0100 (CET) Subject: [Pw_forum] choice of the pseudo, Ni(110) Message-ID: Hi all, i'm using pw in order to study a Ni(110) surface. I' m now just fixing the bulk properties, so i ran some loops to study the variation of the minimum value of the energy and of the alat in the bulk. after doing that stuff, i tried using a different pseudo, obtaining quite different results. i' m sorry for probably it' s a silly question, but how do i choose my pseudo ? first used pseudo : Perdew-Zunger (LDA) exch-corr nonlinear core-correction semicore state d in valence Rabe Rappe Kaxiras Joannopoulos (ultrasoft) obtaining a0 = 6.47 alat and Emin = -85.72192 Ry second used pseudo : Perdew-Burke-Ernzerhof (PBE) exch-corr nonlinear core-correction semicore state d in valence Rabe Rappe Kaxiras Joannopoulos (ultrasoft) obtaining a0 = 6.66 alat and Emin = -85.89509 Ry others parameters fixed in both the runs ( obtained as best choice for a desired accuracy after different evaluation runs) : &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '/home/espresso/pseudo/', outdir='/tmp/' prefix='ni' tprnfor = .true., tstress = .true. / &system ibrav=2, celldm(1) =6.80, nat=1, ntyp=1, nspin = 2, starting_magnetization(1)=0.7, ecutwfc = 24.0, ecutrho = 288.0, occupations='smearing', smearing='methfessel-paxton', degauss=0.01 / &electrons diagonalization='' conv_thr = 1.0e-8 mixing_beta = 0.7 / ATOMIC_SPECIES Ni 58.69 NiUS.RRKJ3.UPF ATOMIC_POSITIONS Ni 0.0 0.0 0.0 K_POINTS (automatic) 5 5 5 1 1 1 So, i apologize once more 'cause probably it's a silly question, and thank you all in advance for your attention. bye, varki pagan. From dalcorso at sissa.it Tue Feb 22 18:41:23 2005 From: dalcorso at sissa.it (Andrea Dal Corso) Date: Tue, 22 Feb 2005 18:41:23 +0100 Subject: [Pw_forum] pwcond.x reference In-Reply-To: <20050222161624.78541.qmail@web60807.mail.yahoo.com> References: <20050222161624.78541.qmail@web60807.mail.yahoo.com> Message-ID: <1109094084.3772.10.camel@dhpc-5-33.sissa.it> On Tue, 2005-02-22 at 08:16 -0800, Yonas Abraham wrote: > Hi, > > did any one know what reference was used to implement > the complex band structure calculation (pwcond.x). I > try to look at the source code or the INITI_PWCOND > file for reference and didn't find any. I have PRB 70 > 045417 (2004) by Alexander Smognunov. did by any > chance that the code is based on that paper? > Yes. Other papers were the pwcond.x code has been used are: A. Smogunov, A. Dal Corso, and E. Tosatti, Ballistic conductance of Ni nanowire with a magnetization reversal, Surf. Sci. 566-568, 390 (2004) F. Picaud, A. Smogunov, A. Dal Corso, and E. Tosatti, Complex band structures and decay length in polyethylene chains, J. Phys.: Cond. Mat. 15, 3731 (2003). A. Smogunov, A. Dal Corso, and E. Tosatti, Complex band structure with ultrasoft pseudopotentials: fcc Ni and Ni nanowire, Surf. Sci. 532-535, 549 (2003) Best regards, Andrea -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From giannozz at nest.sns.it Tue Feb 22 18:55:44 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 22 Feb 2005 18:55:44 +0100 Subject: [Pw_forum] choice of the pseudo, Ni(110) In-Reply-To: References: Message-ID: <200502221855.44984.giannozz@nest.sns.it> On Tuesday 22 February 2005 16:17, Varki Pagan (Peressi) wrote: > first used pseudo : [...] > Perdew-Zunger (LDA) exch-corr > [...] obtaining a0 = 6.47 alat and Emin = -85.72192 Ry > second used pseudo : [...] > Perdew-Burke-Ernzerhof (PBE) exch-corr > [...] obtaining a0 = 6.66 alat and Emin = -85.89509 Ry "different exchange-correlation" is something more than simply "different pseudos". It is well known that gradient-corrected functionals (like PBE) tend to yield larger equilibrium lattice parameters than LDA. ======================================= Please post from the same e-mail you used to register: posts appearing to come from non-registered e-mails will be automatically deleted (starting from tomorrow) ======================================= Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From yonasb at yahoo.com Tue Feb 22 19:27:44 2005 From: yonasb at yahoo.com (Yonas Abraham) Date: Tue, 22 Feb 2005 10:27:44 -0800 (PST) Subject: [Pw_forum] pwcond.x reference In-Reply-To: <1109094084.3772.10.camel@dhpc-5-33.sissa.it> Message-ID: <20050222182744.44709.qmail@web60808.mail.yahoo.com> Thank you Andrea! --yonas --- Andrea Dal Corso wrote: > > > > On Tue, 2005-02-22 at 08:16 -0800, Yonas Abraham > wrote: > > Hi, > > > > did any one know what reference was used to > implement > > the complex band structure calculation > (pwcond.x). I > > try to look at the source code or the INITI_PWCOND > > file for reference and didn't find any. I have PRB > 70 > > 045417 (2004) by Alexander Smognunov. did by any > > chance that the code is based on that paper? > > > > Yes. Other papers were the pwcond.x code has been > used are: > > > A. Smogunov, A. Dal Corso, and E. Tosatti, > Ballistic conductance of Ni nanowire with a > magnetization reversal, > Surf. Sci. 566-568, 390 (2004) > > F. Picaud, A. Smogunov, A. Dal Corso, and E. > Tosatti, > Complex band structures and decay length in > polyethylene chains, > J. Phys.: Cond. Mat. 15, 3731 (2003). > > A. Smogunov, A. Dal Corso, and E. Tosatti, > Complex band structure with ultrasoft > pseudopotentials: fcc Ni and Ni > nanowire, > Surf. Sci. 532-535, 549 (2003) > > > Best regards, > > Andrea > -- > Andrea Dal Corso Tel. > 0039-040-3787428 > SISSA, Via Beirut 2/4 Fax. > 0039-040-3787528 > 34014 Trieste (Italy) e-mail: > dalcorso at sissa.it > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ===== --yonas http://www.yonasb.com From eyvaz_isaev at yahoo.com Tue Feb 22 23:17:56 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Tue, 22 Feb 2005 14:17:56 -0800 (PST) Subject: [Pw_forum] Is E_cut higher, the results is better? In-Reply-To: <20050222172512.70516.qmail@web51902.mail.yahoo.com> Message-ID: <20050222221757.34886.qmail@web60305.mail.yahoo.com> Hi, What concerns your question outlined in Subject field, I don't think so. At least, not always results calculated using higher cutoff energy are agreed better with experimental. Definitely, for US pseudopotentials you don't need to use 135Ry, cutoff energy around 25-35Ry should be OK for USPP. But always you have to make some tests for total energy calculations, increasing ecutwfc, in order be sure about convergency of the total energy with respect to number of plane waves used for expansion of the electron wave function. Using converged ecutwfc, when difference between two last successive total energies is a desired tolerance (say, 1mRy), means that results you obtained depend weakly on your basis choice. Bests, Eyvaz. --- Ming Wang wrote: > Hi, > I used different E_cut(35 Ryd, 135Ryd) to > calculate > molecular polarizability. The results are a little > difference, I wonder whether the results from high > E_cut is more reliable than lower E_cut results. I > realized the E_cut is around 25 Ryd when the USPP is > generated. Does this mean I should choose E_cut > around > 25 Ryd? > > Thanks! > > Ming > > > > > __________________________________ > Do you Yahoo!? > The all-new My Yahoo! - Get yours free! > http://my.yahoo.com > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail From giannozz at nest.sns.it Tue Feb 22 23:16:35 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 22 Feb 2005 23:16:35 +0100 Subject: [Pw_forum] Re: cut-off radius for radial integrations In-Reply-To: <1109079969.11326.90.camel@gurosai.polytechnique.fr> References: <20050219063855.3568.70972.Mailman@democritos.sissa.it> <1109079969.11326.90.camel@gurosai.polytechnique.fr> Message-ID: <200502222316.36370.giannozz@nest.sns.it> On Tuesday 22 February 2005 14:46, Virginie Quequet wrote: > I tried for silicon with the input of example01 (david) with a > pseudo-potential with non-local part defined up to 60 a.u. I don't see any significant difference in results for 4 < rcut < 60 in PW/setup.f90. What do you mean exactly by 'a pseudo-potential with non-local part defined up to 60 a.u.' ? Paolo -- Paolo Giannozzi e-mail: p.giannozzi at sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From cbarreteau at cea.fr Wed Feb 23 08:28:06 2005 From: cbarreteau at cea.fr (Cyrille Barreteau) Date: Wed, 23 Feb 2005 08:28:06 +0100 (CET) Subject: [Pw_forum] choice of the pseudo, Ni(110) In-Reply-To: <200502221855.44984.giannozz@nest.sns.it> References: <200502221855.44984.giannozz@nest.sns.it> Message-ID: <60491.132.166.21.1.1109143686.squirrel@dsm-mail> The choice of exchange-correlation functional is not an easy task. As a rule of thumb I would say that GGA is better for 3d metals while LDA is better for 5d. And 4d is in between.. For example LDA leads to the wrong magnetic behaviour of Iron (mainly because the lattice constant found in LDA is too small). But if you look at surface energies of Pt you will find anormally low values in GGA... But maybe people looking at interaction of molecules with a Pt surface would find GGA better...... It is a kind of cookery:-) cyrille -- ================================================ Cyrille Barreteau | phone:+33 (0)1 69 08 29 51 CEA Saclay | fax : +33 (0)1 69 08 84 46 DSM/DRECAM/SPCSI | email: cbarreteau at cea.fr Batiment 462 | 91191 Gif sur Yvette Cedex FRANCE ~~~~~~~~~~~~~~~~~~~~~~~~ http://www-drecam.cea.fr/spcsi/index.php http://www-drecam.cea.fr/Images/Pisp/cbarreteau/cbarreteau_fr.html ================================================ On Mar 22 f?vrier 2005 18:55, Paolo Giannozzi a ?crit : > On Tuesday 22 February 2005 16:17, Varki Pagan (Peressi) wrote: > >> first used pseudo : [...] >> Perdew-Zunger (LDA) exch-corr >> [...] obtaining a0 = 6.47 alat and Emin = -85.72192 Ry > >> second used pseudo : [...] >> Perdew-Burke-Ernzerhof (PBE) exch-corr >> [...] obtaining a0 = 6.66 alat and Emin = -85.89509 Ry > > "different exchange-correlation" is something more than simply > "different pseudos". It is well known that gradient-corrected > functionals (like PBE) tend to yield larger equilibrium lattice > parameters than LDA. > > ======================================= > Please post from the same e-mail you used to register: posts > appearing to come from non-registered e-mails will be > automatically deleted (starting from tomorrow) > ======================================= > > Paolo > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 > Piazza dei Cavalieri 7 I-56126 Pisa, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From linda.luowei at gmail.com Tue Feb 22 19:14:22 2005 From: linda.luowei at gmail.com (wei luo) Date: Tue, 22 Feb 2005 19:14:22 +0100 Subject: [Pw_forum] Help for reading data Message-ID: Hello, I am new user. I try to get Raman frequency using pqscf software. Now I get the *.dnyG file. But I don't know how to read it and identify the Raman vibration mode. How should I analysize the omega (i)? Such as : phonons of Ti2SiC at Gamma 3 8 4 5.7800000 0.0000000 4.2300000 0.0000000 0.0000000 0.0000000 1 'Ti ' 43627.15044733520 2 'Si ' 25600.30616389483 3 'C ' 10945.52072012734 1 1 0.0000000 0.5773503 0.3870450 2 1 0.5000000 0.2886751 2.5020450 3 1 0.5000000 0.2886751 3.8429550 4 1 0.0000000 0.5773503 1.7279550 5 2 0.0000000 0.5773503 3.1725000 6 2 0.5000000 0.2886751 1.0575000 7 3 0.0000000 0.0000000 0.0000000 8 3 0.0000000 0.0000000 2.1150000 Diagonalizing the dynamical matrix q = ( 0.000000000 0.000000000 0.000000000 ) ************************************************************************** omega( 1) = -0.927703 [THz] = -30.945054 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.355144 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.355144 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.355144 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.355144 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.351826 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.351826 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.352085 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.352085 0.000000 ) omega( 2) = 1.085249 [THz] = 36.200241 [cm-1] ( -0.008551 0.000000 -0.350944 0.000000 0.000000 0.000000 ) ( -0.008551 0.000000 -0.350944 0.000000 0.000000 0.000000 ) ( -0.008551 0.000000 -0.350944 0.000000 0.000000 0.000000 ) ( -0.008551 0.000000 -0.350944 0.000000 0.000000 0.000000 ) ( -0.008817 0.000000 -0.361895 0.000000 0.000000 0.000000 ) ( -0.008817 0.000000 -0.361895 0.000000 0.000000 0.000000 ) ( -0.008525 0.000000 -0.349875 0.000000 0.000000 0.000000 ) ( -0.008525 0.000000 -0.349875 0.000000 0.000000 0.000000 ) omega( 3) = 1.085249 [THz] = 36.200241 [cm-1] ( 0.350944 0.000000 -0.008551 0.000000 0.000000 0.000000 ) ( 0.350944 0.000000 -0.008551 0.000000 0.000000 0.000000 ) ( 0.350944 0.000000 -0.008551 0.000000 0.000000 0.000000 ) ( 0.350944 0.000000 -0.008551 0.000000 0.000000 0.000000 ) ( 0.361895 0.000000 -0.008817 0.000000 0.000000 0.000000 ) ( 0.361895 0.000000 -0.008817 0.000000 0.000000 0.000000 ) ( 0.349875 0.000000 -0.008525 0.000000 0.000000 0.000000 ) ( 0.349875 0.000000 -0.008525 0.000000 0.000000 0.000000 ) omega( 4) = 3.602149 [THz] = 120.155570 [cm-1] ( 0.002007 0.000000 0.413248 0.000000 0.000000 0.000000 ) ( -0.002007 0.000000 -0.413248 0.000000 0.000000 0.000000 ) ( 0.002007 0.000000 0.413248 0.000000 0.000000 0.000000 ) ( -0.002007 0.000000 -0.413248 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.001934 0.000000 0.398047 0.000000 0.000000 0.000000 ) ( -0.001934 0.000000 -0.398047 0.000000 0.000000 0.000000 ) omega( 5) = 3.602149 [THz] = 120.155570 [cm-1] ( -0.413248 0.000000 0.002007 0.000000 0.000000 0.000000 ) ( 0.413248 0.000000 -0.002007 0.000000 0.000000 0.000000 ) ( -0.413248 0.000000 0.002007 0.000000 0.000000 0.000000 ) ( 0.413248 0.000000 -0.002007 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( -0.398047 0.000000 0.001934 0.000000 0.000000 0.000000 ) ( 0.398047 0.000000 -0.001934 0.000000 0.000000 0.000000 ) omega( 6) = 4.881638 [THz] = 162.834991 [cm-1] ( -0.000051 0.000000 0.142614 0.000000 0.000000 0.000000 ) ( 0.000051 0.000000 -0.142614 0.000000 0.000000 0.000000 ) ( 0.000051 0.000000 -0.142614 0.000000 0.000000 0.000000 ) ( -0.000051 0.000000 0.142614 0.000000 0.000000 0.000000 ) ( 0.000242 0.000000 -0.677733 0.000000 0.000000 0.000000 ) ( -0.000242 0.000000 0.677733 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) omega( 7) = 4.881638 [THz] = 162.834991 [cm-1] ( 0.142614 0.000000 0.000051 0.000000 0.000000 0.000000 ) ( -0.142614 0.000000 -0.000051 0.000000 0.000000 0.000000 ) ( -0.142614 0.000000 -0.000051 0.000000 0.000000 0.000000 ) ( 0.142614 0.000000 0.000051 0.000000 0.000000 0.000000 ) ( -0.677733 0.000000 -0.000242 0.000000 0.000000 0.000000 ) ( 0.677733 0.000000 0.000242 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) omega( 8) = 4.904280 [THz] = 163.590265 [cm-1] ( 0.000000 0.000000 0.000000 0.000000 0.419278 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.419278 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.419278 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.419278 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.385243 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 -0.385243 0.000000 ) omega( 9) = 6.074220 [THz] = 202.615506 [cm-1] ( 0.000597 0.000000 -0.176369 0.000000 0.000000 0.000000 ) ( 0.000597 0.000000 -0.176369 0.000000 0.000000 0.000000 ) ( 0.000597 0.000000 -0.176369 0.000000 0.000000 0.000000 ) ( 0.000597 0.000000 -0.176369 0.000000 0.000000 0.000000 ) ( -0.002182 0.000000 0.644609 0.000000 0.000000 0.000000 ) ( -0.002182 0.000000 0.644609 0.000000 0.000000 0.000000 ) ( 0.000505 0.000000 -0.149203 0.000000 0.000000 0.000000 ) ( 0.000505 0.000000 -0.149203 0.000000 0.000000 0.000000 ) omega(10) = 6.074220 [THz] = 202.615506 [cm-1] ( -0.176369 0.000000 -0.000597 0.000000 0.000000 0.000000 ) ( -0.176369 0.000000 -0.000597 0.000000 0.000000 0.000000 ) ( -0.176369 0.000000 -0.000597 0.000000 0.000000 0.000000 ) ( -0.176369 0.000000 -0.000597 0.000000 0.000000 0.000000 ) ( 0.644609 0.000000 0.002182 0.000000 0.000000 0.000000 ) ( 0.644609 0.000000 0.002182 0.000000 0.000000 0.000000 ) ( -0.149203 0.000000 -0.000505 0.000000 0.000000 0.000000 ) ( -0.149203 0.000000 -0.000505 0.000000 0.000000 0.000000 ) omega(11) = 7.659691 [THz] = 255.501486 [cm-1] ( -0.000071 0.000000 -0.500000 0.000000 0.000000 0.000000 ) ( -0.000071 0.000000 -0.500000 0.000000 0.000000 0.000000 ) ( 0.000071 0.000000 0.500000 0.000000 0.000000 0.000000 ) ( 0.000071 0.000000 0.500000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) omega(12) = 7.659691 [THz] = 255.501486 [cm-1] ( -0.500000 0.000000 0.000071 0.000000 0.000000 0.000000 ) ( -0.500000 0.000000 0.000071 0.000000 0.000000 0.000000 ) ( 0.500000 0.000000 -0.000071 0.000000 0.000000 0.000000 ) ( 0.500000 0.000000 -0.000071 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) omega(13) = 7.811757 [THz] = 260.573900 [cm-1] ( -0.000071 0.000000 0.445932 0.000000 0.000000 0.000000 ) ( 0.000071 0.000000 -0.445932 0.000000 0.000000 0.000000 ) ( 0.000071 0.000000 -0.445932 0.000000 0.000000 0.000000 ) ( -0.000071 0.000000 0.445932 0.000000 0.000000 0.000000 ) ( -0.000051 0.000000 0.319826 0.000000 0.000000 0.000000 ) ( 0.000051 0.000000 -0.319826 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) ( 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 ) Best linda From swblelia at sw.ehu.es Wed Feb 23 09:48:01 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Wed, 23 Feb 2005 09:48:01 +0100 Subject: [Pw_forum] Mg pseudopotential Message-ID: <421C4341.4060709@sw.ehu.es> Hi, I am trying to calculate surface phonons of Mg using "bhs" pseudopotential provided in the pwscf web. The phonon frequencies resulting from the calculations are too high. To improve this I think that it is necessary to use a PP with nonlinear correction, this way the equilibrium lattice constant is bigger and frequencies low down a bit. Does anybody know how to obtain a norm conserving pseudo with linear correction which is NOT ULTRASOFT? thanks From virginie.quequet at polytechnique.fr Wed Feb 23 11:27:28 2005 From: virginie.quequet at polytechnique.fr (Virginie Quequet) Date: 23 Feb 2005 11:27:28 +0100 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #557 - 11 msgs In-Reply-To: <20050223063745.24752.18876.Mailman@democritos.sissa.it> References: <20050223063745.24752.18876.Mailman@democritos.sissa.it> Message-ID: <1109154447.22000.127.camel@gurosai.polytechnique.fr> I mean that the two pseudo-potentials for Si given with espresso SiPBE_nc.UPF and Si.vbc.UPF have their non-local part cut at 10 a.u. (only 957 points on a mesh of 1141 for SiPBE_nc.UPF and only 359 points on a mesh of 431 for Si.vbc.UPF). So when I use one of this pseudo-potential and change rcut, nothing happens, but if I use instead a pseudo with non-local part cut at 80 a.u., I see the difference. Best regards Virginie Quequet On Tuesday 22 February 2005 14:46, Virginie Quequet wrote: > I tried for silicon with the input of example01 (david) with a > pseudo-potential with non-local part defined up to 60 a.u. I don't see any significant difference in results for 4 < rcut < 60 in PW/setup.f90. What do you mean exactly by 'a pseudo-potential with non-local part defined up to 60 a.u.' ? Paolo -- Paolo Giannozzi e-mail: p.giannozzi at sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy --__--__-- _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From varki at ts.infn.it Wed Feb 23 11:43:42 2005 From: varki at ts.infn.it (Varki Pagan (Peressi)) Date: Wed, 23 Feb 2005 11:43:42 +0100 (CET) Subject: [Pw_forum] Re: choice of the pseudo, Ni(110) In-Reply-To: References: Message-ID: On Tue, 22 Feb 2005, Varki Pagan (Peressi) wrote: Thank you all, now it' s clear to me how to proceed. you very kind, thank you, best regards, varki. > > Hi all, > > i'm using pw in order to study a Ni(110) surface. > > I' m now just fixing the bulk properties, so i ran some loops to study the > variation of the minimum value of the energy and of the alat in the bulk. > > after doing that stuff, i tried using a different pseudo, obtaining quite > different results. > > i' m sorry for probably it' s a silly question, but how do i choose my pseudo > ? > > first used pseudo : > > Perdew-Zunger (LDA) exch-corr > nonlinear core-correction > semicore state d in valence > Rabe Rappe Kaxiras Joannopoulos (ultrasoft) > > obtaining a0 = 6.47 alat and Emin = -85.72192 Ry > > > second used pseudo : > > Perdew-Burke-Ernzerhof (PBE) exch-corr > nonlinear core-correction > semicore state d in valence > Rabe Rappe Kaxiras Joannopoulos (ultrasoft) > > obtaining a0 = 6.66 alat and Emin = -85.89509 Ry > > > others parameters fixed in both the runs ( obtained as best choice for a > desired accuracy after different evaluation runs) : > > &control > calculation='scf' > restart_mode='from_scratch', > pseudo_dir = '/home/espresso/pseudo/', > outdir='/tmp/' > prefix='ni' > tprnfor = .true., > tstress = .true. > / > &system > ibrav=2, celldm(1) =6.80, nat=1, ntyp=1, > nspin = 2, starting_magnetization(1)=0.7, > ecutwfc = 24.0, ecutrho = 288.0, > occupations='smearing', smearing='methfessel-paxton', degauss=0.01 > / > &electrons > diagonalization='' > conv_thr = 1.0e-8 > mixing_beta = 0.7 > / > ATOMIC_SPECIES > Ni 58.69 NiUS.RRKJ3.UPF > > ATOMIC_POSITIONS > Ni 0.0 0.0 0.0 > K_POINTS (automatic) > 5 5 5 1 1 1 > > > > So, i apologize once more 'cause probably it's a silly question, and thank > you all in advance for your attention. > > bye, > > varki pagan. > > > From giannozz at nest.sns.it Wed Feb 23 11:48:03 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 23 Feb 2005 11:48:03 +0100 Subject: [Pw_forum] Help for reading data In-Reply-To: References: Message-ID: <200502231148.03733.giannozz@nest.sns.it> On Tuesday 22 February 2005 19:14, wei luo wrote: > Now I get the *.dnyG file. But I don't know how to read it and > identify the Raman vibration mode. ======================================= Please post from the same e-mail you used to register: posts appearing to come from non-registered e-mails will be automatically deleted (effective NOW: this post is the last one I rescue) ======================================= "make tools" compiles code "dynmat.x" in the pwtools/ directory. It reads the dynamical matrix file at q=0 and can - apply the acoustic sum rule, so that you get omega=0 for translation modes at q=0 - add the non-analytical terms that are responsible for the TO-LO splitting, calculate IR cross sections (provided that the dynamical matrix file contains the effective charges, option "epsil=.true." in ph.x) - produce a file with eigendisplacements that can be directly plotted by "molden" - Raman cross sections can also be calculated but it is much trickier It shouldn't be difficult to classify phonon modes in a 8-atom system. In any case, the bunch of numbers following each frequency are, for each atom: Re(u_x) Im(u_x) Re(u_y) Im(u_y) Re(u_z) Im(u_z), where u is the normalized eigendisplacement. The imaginary terms are obviously zero for q=0 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From massimiliano.bonomi at mi.infn.it Wed Feb 23 11:59:23 2005 From: massimiliano.bonomi at mi.infn.it (Massimiliano Bonomi) Date: Wed, 23 Feb 2005 11:59:23 +0100 Subject: [Pw_forum] multiplicity in lsda calculation Message-ID: <421C620B.6070802@mi.infn.it> Dear users, I'm doing a lsda calculation with a gold cluster of 8 atoms using both CPMD and espresso 2.1.2 package. With the first one I can specify the multiplicity of the state (singolet, triplet...), but it's very slow in reaching convergence. How can i specify the multiplicity with pwscf (which is much faster)?? Thanks in advance. -- Massimiliano Bonomi Dipartimento di Fisica, Universit? degli Studi di Milano via Celoria, 16 - 20133 Milano, Italy Ph. +39 02 50317 717 From emenendez at macul.ciencias.uchile.cl Wed Feb 23 14:55:19 2005 From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P) Date: Wed, 23 Feb 2005 10:55:19 -0300 (CLST) Subject: [Pw_forum] Re: Band structure calculation for Cu2O In-Reply-To: <20050223063745.24752.18876.Mailman@democritos.sissa.it> References: <20050223063745.24752.18876.Mailman@democritos.sissa.it> Message-ID: Dear Aloysius, Eyvaz has already given you the solution for Cu2O . For next structures you find in your career, have three advices 1. Install XCRYSDEN, following the link at http://www.democritos.net/scientific.php This program can read a PWSCF input file (Menu File/Open structure/Open PWSCF Input file). Then, with the menu Tools/k-path selection you find a figure of the Brillouin zone, where clicking at the symmetry points you retrieve the coordinates. XCRYSDEN can also be called from PWGUI to visualize the structure. It is advisable to always visualize the structure before do gross calculations. 2. It has been said that there is nothing more practical than a good theory. Read the chaper on the reciprocal lattice in any deep book on solid State Physics, eg. the chaper II of Solid State Physics, by G. Grosso and G. Pastori-Parravicini. I guess that Ashcroft & Mermin's book has a similar chapter. 3. Find this book author = {C. J. Bradley and A. P. Cracknell}, title = {The mathematical theory of symmetry in solids}, publisher = {Clarendon Press}, year = {1972}. It has tables and figures of the high symmetry points for all crystal classes, and many more issues related to the symmetry of solids. Have succes, Eduardo A. Menendez Proupin Department of Physics Faculty of Science University of Chile Las Palmeras 3425 ?u?oa, Santiago Chile Phone: 56+2+678 74 11 http://fisica.ciencias.uchile.cl/~emenendez/ > > Message: 1 > Date: Tue, 22 Feb 2005 10:55:26 +1100 > From: Aloysius Soon > To: pw_forum at pwscf.org > Subject: [Pw_forum] Band structure calculation for Cu2O > Reply-To: pw_forum at pwscf.org > > Dear all, > I'm hoping to calculate the band structure for Cu2O but is > having a hard time finding the high symmetry points for this > structure. > Can any kind soul please help? Any sort of data base or > reference text will be greatly appreciated. > Cheers! > > > > Regards, > Aloysius > From marzari at MIT.EDU Wed Feb 23 15:30:23 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 23 Feb 2005 09:30:23 -0500 Subject: [Pw_forum] multiplicity in lsda calculation In-Reply-To: <421C620B.6070802@mi.infn.it> References: <421C620B.6070802@mi.infn.it> Message-ID: <421C937F.6010901@mit.edu> Dear Massimiliano, using Gamma-sampling only (as appropriate in this case) there should be a fixed-occupations option described somewhere in the docs (it would be trivial but not terribly useful to expand the input flags to have this option imposed at every k-point). Sometimes it is difficult (or impossible) to converge some of these simple systems (atoms, molecules, or clusters in vacuum); I presume this is due to level-crossings instabilities, that affect iterative selfconsistent approaches (as is the case with PWSCF). We have experienced quite a few of these problems in recent weeks. A finite electronic temperature would help smooth the level-crossing instabilities, and thus aid convergence, but then you are not anymore guaranteed to converge to the state of definite S_z you were originally looking for. (Note: finite electronic T approaches should be used to improve k-point sampling accuracy, more than ground-state convergence; e.g. in direct-minimization approaches convergence to the ground state is reached variationally irrespectively of the electronic T, but in iterative scf approaches a finite T helps out. See http://nnn.mit.edu/phd for a detailed explanation). A simple solution to this problem that I have been mulling would be introducing two Fermi energies - i.e. combining an electronic temperature with a constraint of fixed n_up-n_down (and, given that nelec is fixed -> fixed n_up and fixed n_down). It should be very easy to implement, and we might give it a try in the next few weeks (but you'd be also welcome to try, or interact with one of my students on this - Heather Kulik). Any comments ? (By the way, it would be helpful to have in the docs a note that says that for finite temperature calculations the "total energy" printed by the code is actually nicola PS: note that in DFT you have states of definite S_z, but not S^2, so the notation singlet or triplet is only approximate. E.g. a real triplet has S = 1 and S_z = 1, 0 or -1 in hbar units. Massimiliano Bonomi wrote: > Dear users, > I'm doing a lsda calculation with a gold cluster of 8 atoms using both > CPMD and espresso 2.1.2 package. > With the first one I can specify the multiplicity of the state > (singolet, triplet...), but it's very slow in reaching convergence. > How can i specify the multiplicity with pwscf (which is much faster)?? > > Thanks in advance. > -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu From marzari at MIT.EDU Wed Feb 23 15:41:51 2005 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 23 Feb 2005 09:41:51 -0500 Subject: [Pw_forum] multiplicity in lsda calculation In-Reply-To: <421C937F.6010901@mit.edu> References: <421C620B.6070802@mi.infn.it> <421C937F.6010901@mit.edu> Message-ID: <421C962F.6010900@mit.edu> > Any comments ? (By the way, it would be helpful to have in the > docs a note that says that for finite temperature calculations > the "total energy" printed by the code is actually the total electronic free energy E-TS, and that the correction for metals is +1/2 TS. Summing the "total energy" E-TS + the "correction for metals" 1/2 TS you obtain E - 1/2 TS, or the "corrected energy" introduced by Gillan and De Vita. In the case of Fermi-Dirac or Gaussian smearing, the corrected energy depends only to orders greater than two in the temperature, and thus one can use large smearings (.01 to .05 Ry) in true metallic systems, and still have a reasonably good estimate of the zero-smearing energy (anything else - e.g. forces or stresses, will still be greatly influenced by this temperature). A more general solution to this problem is to use Methfessel-Paxton or cold-smearing approaches, where the quadratic dependence of the "total energy" E-TS on temperature is removed "ab-initio". Finally, in an atom or a cluster you might need to be careful. If I got all of this correctly (Stefano dG, please chip in), the "total energy", being E-TS, will have an S term different from zero as soon as you have degenerate orbitals with fractional fillings; since what you really need there is E, just add twice the "correction for metals" to the "total energy"), making sure that the E you get does not change significantly with T. :-) nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu From aaron at chips.ncsu.edu Wed Feb 23 19:01:16 2005 From: aaron at chips.ncsu.edu (aaron at chips.ncsu.edu) Date: Wed, 23 Feb 2005 13:01:16 -0500 (EST) Subject: [Pw_forum] rdiaghg in pw.x -- NEB In-Reply-To: <200502211629.11648.giannozz@nest.sns.it> Message-ID: Hi , I am reporting that it seems I was able to solve this problem I was having by reducing the ds parameter under the namelist IONS. From virginie.quequet at polytechnique.fr Thu Feb 24 09:01:46 2005 From: virginie.quequet at polytechnique.fr (Virginie Quequet) Date: 24 Feb 2005 09:01:46 +0100 Subject: [Pw_forum] Re: cut-off radius for radial integrations In-Reply-To: <20050223063745.24752.18876.Mailman@democritos.sissa.it> References: <20050223063745.24752.18876.Mailman@democritos.sissa.it> Message-ID: <1109232106.22007.1159.camel@gurosai.polytechnique.fr> Sorry, I make a mistake with the subject in my reply. I post it again. I mean that the two pseudo-potentials for Si given with espresso SiPBE_nc.UPF and Si.vbc.UPF have their non-local part cut at 10 a.u. (only 957 points on a mesh of 1141 for SiPBE_nc.UPF and only 359 points on a mesh of 431 for Si.vbc.UPF). So when I use one of this pseudo-potential and change rcut, nothing happens, but if I use instead a pseudo with non-local part cut at 80 a.u., I see the difference. Best regards Virginie Quequet From: Paolo Giannozzi Organization: Scuola Normale Superiore To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Re: cut-off radius for radial integrations Date: Tue, 22 Feb 2005 23:16:35 +0100 Reply-To: pw_forum at pwscf.org On Tuesday 22 February 2005 14:46, Virginie Quequet wrote: > I tried for silicon with the input of example01 (david) with a > pseudo-potential with non-local part defined up to 60 a.u. I don't see any significant difference in results for 4 < rcut < 60 in PW/setup.f90. What do you mean exactly by 'a pseudo-potential with non-local part defined up to 60 a.u.' ? Paolo From giannozz at nest.sns.it Thu Feb 24 09:26:04 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 24 Feb 2005 09:26:04 +0100 Subject: [Pw_forum] Re: cut-off radius for radial integrations In-Reply-To: <1109232106.22007.1159.camel@gurosai.polytechnique.fr> References: <20050223063745.24752.18876.Mailman@democritos.sissa.it> <1109232106.22007.1159.camel@gurosai.polytechnique.fr> Message-ID: <200502240926.04709.giannozz@nest.sns.it> On Thursday 24 February 2005 09:01, Virginie Quequet wrote: > I mean that the two pseudo-potentials for Si given with espresso > SiPBE_nc.UPF and Si.vbc.UPF have their non-local part cut at 10 a.u. > (only 957 points on a mesh of 1141 for SiPBE_nc.UPF and only 359 > points on a mesh of 431 for Si.vbc.UPF). So when I use one of this > pseudo-potential and change rcut, nothing happens, but if I use instead > a pseudo with non-local part cut at 80 a.u., I see the difference. the above means nothing to me. If you use a different pseudo, you get different results. A pseudo cannot have "a non-local part cut at 80 a.u." : the nonlocal part is, by construction, zero beyond the matching radius, i.e. a few a.u. . The local part has a known behavior (-Z/r) at large r and this is accounted for analytically. Please explain EXACTLY what you are doing (i.e. in such a way that one can reproduce it) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Thu Feb 24 09:51:22 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 24 Feb 2005 09:51:22 +0100 Subject: [Pw_forum] Mg pseudopotential In-Reply-To: <421C4341.4060709@sw.ehu.es> References: <421C4341.4060709@sw.ehu.es> Message-ID: <200502240951.22232.giannozz@nest.sns.it> On Wednesday 23 February 2005 09:48, Aritz Leonardo wrote: > Does anybody know how to obtain a norm conserving pseudo > with linear correction which is NOT ULTRASOFT? from the web page http://www.pwscf.org/pseudo.htm: ------------------------------------------------------------ If you do not find here the PP you need, you may try to [...] generate it, using the PP generation code included in directory atomic/. See directory atomic_doc/ for available documentation. Other available PP generation packages include: David Vanderbilt's code (UltraSoft PPs) http://www.physics.rutgers.edu/~dhv/uspp/index.html The Fritz Haber code (Norm-Conserving PPs) http://www.fhi-berlin.mpg.de/th/fhi98md/fhi98PP Jos?-Lu?s Martins' code (Norm-Conserving PPs) http://bohr.inesc-mn.pt/~jlm/pseudo.html The first two codes produce PPs either in UPF format or in a format that can be converted to UPF using the utilities contained in directory upftools. Remember: always test the PPs on simple test systems before proceeding to serious calculations. ------------------------------------------------------------ My old version of the 'atomic' code is still available in http://web1.sns.it/~giannozz/software.html Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From swblelia at sw.ehu.es Thu Feb 24 12:49:53 2005 From: swblelia at sw.ehu.es (Aritz Leonardo) Date: Thu, 24 Feb 2005 12:49:53 +0100 Subject: [Pw_forum] what is alphamix(niter) Message-ID: <421DBF61.6090504@sw.ehu.es> Hi, I was having some trouble with a phonon calculation because of the converging criteria could not be reached. Changing the default value of alphamix to 0.2 seems to work... but does anybody know what am I really changing? and what is the integer inside brackets? thanks a lot. From aaron at chips.ncsu.edu Thu Feb 24 14:33:21 2005 From: aaron at chips.ncsu.edu (aaron at chips.ncsu.edu) Date: Thu, 24 Feb 2005 08:33:21 -0500 (EST) Subject: [Pw_forum] what is alphamix(niter) In-Reply-To: <421DBF61.6090504@sw.ehu.es> Message-ID: From giannozz at nest.sns.it Thu Feb 24 16:27:03 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 24 Feb 2005 16:27:03 +0100 Subject: [Pw_forum] what is alphamix(niter) In-Reply-To: References: Message-ID: <200502241627.03611.giannozz@nest.sns.it> On Thursday 24 February 2005 14:33, aaron at chips.ncsu.edu wrote: > From the Doc files: > > INPUT_PH: > > alpha_mix(niter) mixing factor (for each iteration) for alpha_mix(1)=0.7 > vnew(in) = alpha_mix*vold(out) + (1-alpha_mix)*vold(in) actually it is more complex than this. The above is "simple mixing", but what the phonon code uses is a more sophisticated "Broyden mixing": D.D.Johnson, PRB 38, 12807 (1988). The meaning of alpha_mix is anyway the same: how much of the output potential is added to the input potential for the next iteration. > the integer in parentheses is the iteration number for which you wish > to set the alpha_mix. this is a possibility that was added years ago for some cases of difficult convergence. It may be ignored: just set alpha_mix(1) to the desired value for all iterations Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From slzhu at imr.ac.cn Fri Feb 25 10:44:23 2005 From: slzhu at imr.ac.cn (Shenglong Zhu) Date: Fri, 25 Feb 2005 17:44:23 +0800 Subject: [Pw_forum] Help for converting castep's pseudopotential file to pwscf' file format Message-ID: <20050225094159.AE70311276B@democritos.sissa.it> Hello, All, I try to code converting CASTEP's pseudopotential file to PWSCF's format, since CASTEP has fullly-tested potential files for nearly all elements. Are there anybody who know the format of the CASTEP's format? Best regards, Shenglong Zhu From eyvaz_isaev at yahoo.com Fri Feb 25 11:06:57 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 25 Feb 2005 02:06:57 -0800 (PST) Subject: [Pw_forum] Help for converting castep's pseudopotential file to pwscf' file format In-Reply-To: <20050225094159.AE70311276B@democritos.sissa.it> Message-ID: <20050225100657.86084.qmail@web60304.mail.yahoo.com> Hi, Please pay attention to a fact that CASTEP is a commercial product and be sure that you are not violating Copyright. Bests, Eyvaz. --- Shenglong Zhu wrote: > Hello, All, > > I try to code converting CASTEP's pseudopotential > file to PWSCF's format, since CASTEP has > fullly-tested potential files for nearly all > elements. > > Are there anybody who know the format of the > CASTEP's format? > > Best regards, > > Shenglong Zhu > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________ Do you Yahoo!? Yahoo! Mail - now with 250MB free storage. Learn more. http://info.mail.yahoo.com/mail_250 From breezejd at lsbu.ac.uk Fri Feb 25 14:58:09 2005 From: breezejd at lsbu.ac.uk (Jonathan Breeze) Date: Fri, 25 Feb 2005 13:58:09 +0000 Subject: [Pw_forum] Help for converting castep's pseudopotential file to pwscf' file format In-Reply-To: <20050225100657.86084.qmail@web60304.mail.yahoo.com> References: <20050225100657.86084.qmail@web60304.mail.yahoo.com> Message-ID: <200502251358.09820.breezejd@lsbu.ac.uk> Please note that there is also an academic distribution of CASTEP which is free (no cost!) to academic instutions under an agreement between the UK C-P consortium, Accelrys (MSI) and Daresbury laboratory UK (enquiries to ukpc at dl.ac.uk). The most recent version (3.02) comes with a pseudopotential generator written by Victor Milman of MSI (vmilman at msi.com) which uses David Vanderbilt's ultrasoft pp generation code. Since there are utilities to convert Dave Vanderbilts potentials to PWSCF UPF format also, you could generate essentially the same potentials for both CASTEP and PWSCF if you were so inclined. This way I'm sure you won't be violating any licensing rules. Regards, Jonathan On Friday 25 February 2005 10:06, Eyvaz Isaev wrote: > Hi, > > Please pay attention to a fact that CASTEP is a > commercial product and be sure that you are not > violating Copyright. > > Bests, > Eyvaz. > > --- Shenglong Zhu wrote: > > Hello, All, > > > > I try to code converting CASTEP's pseudopotential > > file to PWSCF's format, since CASTEP has > > fullly-tested potential files for nearly all > > elements. > > > > Are there anybody who know the format of the > > CASTEP's format? > > > > Best regards, > > > > Shenglong Zhu > > > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > __________________________________ > Do you Yahoo!? > Yahoo! Mail - now with 250MB free storage. Learn more. > http://info.mail.yahoo.com/mail_250 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From ceresoli at sissa.it Fri Feb 25 18:33:45 2005 From: ceresoli at sissa.it (Davide Ceresoli) Date: Fri, 25 Feb 2005 18:33:45 +0100 Subject: [Pw_forum] Na PW91 pseudo messes up FPMD Message-ID: <421F6179.7010000@sissa.it> Hi, I've got a problem in running FPMD with the Cl PW91 pseudo found on the pwscf web site (Cl.pw91-mt.UPF). Here is an excerpt of the output: At the very beginning, at the zero-th step: TOTAL ENERGY = NaN A.U. KINETIC ENERGY = 1557.7764407689 A.U. ELECTROSTATIC ENERGY = 7566210.6129575055 A.U. ESELF = 1595.7691216057 A.U. ESR = 7566073.8892419031 A.U. PSEUDOPOTENTIAL ENERGY = -527.9284639858 A.U. N-L PSEUDOPOTENTIAL ENERGY = NaN A.U. EXCHANGE-CORRELATION ENERGY = -279.6680250007 A.U. etc... and the initial forces on Cl atoms are NaN (but the forces on sodium atoms are not NaN). (I'm using espresso-2.1.2 and espresso from CVS 3 days ago). Other pseudopotentials for Cl (PZ, BLYP) work fine. PW calculations instead, they are always fine. Can I use the same pseudopotentials in PW and FPMD (norm-conserving), or there is some caveat? Thanks, Davide From eyvaz_isaev at yahoo.com Fri Feb 25 18:51:40 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 25 Feb 2005 09:51:40 -0800 (PST) Subject: [Pw_forum] Na PW91 pseudo messes up FPMD In-Reply-To: <421F6179.7010000@sissa.it> Message-ID: <20050225175140.88888.qmail@web60304.mail.yahoo.com> Hi, > Can I use the same pseudopotentials in PW and FPMD > (norm-conserving), > or there is some caveat? Yes, I think so. FPMD and PW use the same interface and PsPs. The best choice should be PsP in the UPF format. Good luck, Eyvaz. __________________________________ Do you Yahoo!? Yahoo! Mail - Find what you need with new enhanced search. http://info.mail.yahoo.com/mail_250 From eyvaz_isaev at yahoo.com Fri Feb 25 21:00:49 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Fri, 25 Feb 2005 12:00:49 -0800 (PST) Subject: [Pw_forum] Help for reading data In-Reply-To: <200502231148.03733.giannozz@nest.sns.it> Message-ID: <20050225200050.7748.qmail@web60307.mail.yahoo.com> Dear Paolo, Unexpectedly I have received the next message from PHP at ts.infn.it This address is no longer active Does it means that something is wrong with my account in the forum? Bests, Eyvaz. __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail From ejl7240 at chemail.tamu.edu Fri Feb 25 22:44:08 2005 From: ejl7240 at chemail.tamu.edu (Eduardo J. Lamas) Date: Fri, 25 Feb 2005 15:44:08 -0600 Subject: [Pw_forum] Pt PP problem with RRKJ3 - one pseudo wave is zero Message-ID: <002f01c51b83$22747390$0ee35ba5@ch0316> Dear List members: I am testing a PP I generated with the PP program that comes with the espresso package. After drawing the pseudowavefunctions I found that the column associated to the s wave is all zeros. Surprisingly the PP performs well in the transferability tests especially in that channel (that has occupancy different from zero). I am wondering why is that and if there could be any problem with the subroutine that calculate the pseudo waves in the PP package. Any comment will be appreciated. Regards, Eduardo PS.find attached the PP input files. Any comment on the parameters used will also be appreciated. I have to use latter correction, because if not used the all electron wfc will fail to converge. But I am departing from the recipe given in the manual for obtaining revPBE. Is that a serious problem? &input title='Pt', zed=78.0, rel=1, iswitch=3, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, latt=1, config='[Xe] 4f14 5d9 6s1 6p0', dft='revPBE', / &test nconf=1 pseudotype=3, configts(1)= '5d9 6s1 6p0', file_wavefunctionsps='wavesPS.wfc', file_pseudopw='Pt_revPBE_RRKJ3_v5.UPF', / &inputp lloc=0, file_pseudopw='Pt_revPBE_RRKJ3_v5', zval=10.d0, nlcc=.false., latt=1, rcore=1.2, rcloc=2.6, file_logderps='logderps.wri', / 3 5D 3 2 9.00 0.00 1.90 2.40 6P 2 1 0.00 0.00 2.50 2.50 6S 1 0 1.00 0.00 2.30 2.30 And the file for PP testing is: &input title='Pt', zed=78.0, rel=1, iswitch=2, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, config='[Xe] 4f14. 5d9 6s1 6p0', dft='revPBE', / &test nconf=3 pseudotype=3, file_pseudo='Pt_revPBE_RRKJ3_v5', configts(1)= '5d10 6s0 6p0', configts(2)= '5d9 6s0.75 6p0.25', configts(3)= '5d9 6s0 6p0', / From hushujun at mail.sdu.edu.cn Sat Feb 26 15:52:55 2005 From: hushujun at mail.sdu.edu.cn (=?gb2312?B?uvrK9778?=) Date: Sat, 26 Feb 2005 22:52:55 +0800 Subject: [Pw_forum] About exchange-correlation functional is PP Message-ID: <309429575.09244@mail.sdu.edu.cn> Dear Drs, I generated the new PPs of atoms: Zn and O, but using the different exchange-correlation functional, PBE and PW91 respectively, and then used them for the pw.x calculation. Error appeared at the beginning of the output file. But when using the some functional, it is successful. Is it necessary to use the same one? Best wishes Shujun Hu ____________________________ hushujun at mail.sdu.edu.cn From hushujun at mail.sdu.edu.cn Sat Feb 26 16:03:41 2005 From: hushujun at mail.sdu.edu.cn (Shujun Hu) Date: Sat, 26 Feb 2005 23:03:41 +0800 Subject: [Pw_forum] About energy of a single atom Message-ID: <309430221.10006@mail.sdu.edu.cn> Dear Drs, I tried to calculate the energy of a single atom in order to gain the bond energy. Although such information is given in the PP, i donnt trust it. One of the important parameters is the atomic distance. Maybe 10 times of the atomic radius between atoms is better, and such system may be considered as "alone". Yes or no? And the E-cutoff, how to set? Best wishes Shujun Hu ____________________________ hushujun at mail.sdu.edu.cn From eyvaz_isaev at yahoo.com Sat Feb 26 15:51:45 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sat, 26 Feb 2005 06:51:45 -0800 (PST) Subject: [Pw_forum] About exchange-correlation functional is PP In-Reply-To: <309429575.09244@mail.sdu.edu.cn> Message-ID: <20050226145145.90660.qmail@web60301.mail.yahoo.com> Hi, Yes, you have to use pseudopotentials with the same exchange-correlation functionals. Otherwise you get an error "inconsistent DFT read" and PW stops. Bests, Eyvaz. --- ?????? wrote: > > Dear Drs, > I generated the new PPs of atoms: Zn and O, but > using the different > exchange-correlation functional, PBE and PW91 > respectively, and then used them for > the pw.x calculation. Error appeared at the > beginning of the output file. But when > using the some functional, it is successful. Is it > necessary to use the same one? > Best wishes > > Shujun Hu > ____________________________ > hushujun at mail.sdu.edu.cn > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From proffess at yandex.ru Sun Feb 27 13:24:47 2005 From: proffess at yandex.ru (Sergey Lisenkov) Date: Sun, 27 Feb 2005 15:24:47 +0300 (MSK) Subject: [Pw_forum] error in during bands.x Message-ID: <4221BC0F.000019.17915@camay.yandex.ru> Dear PWscf authors and users, I tried to use bands.x and after a long time I got the next message: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from punch_band : error # 5 increase maxdeg %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... 2 I looked at sources and found that maxdeg=4 (max allowed degeneracy). I am interested what is the value I should use? Thanks, Sergey From eyvaz_isaev at yahoo.com Sun Feb 27 22:45:00 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 27 Feb 2005 13:45:00 -0800 (PST) Subject: [Pw_forum] error in during bands.x In-Reply-To: <4221BC0F.000019.17915@camay.yandex.ru> Message-ID: <20050227214500.36911.qmail@web60307.mail.yahoo.com> Hello Sergei, --- Sergey Lisenkov wrote: > I looked at sources and found that maxdeg=4 (max > allowed degeneracy). I am interested what is the > value I should use? Perhaps you have eigenvalues with more than 4-fold degeneration. Try maxdeg>4, say maxdeg=10 (just as an example, to be more precise, you have to look at your bands file), recompile the program and try again. Bests, Eyvaz. __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail From eyvaz_isaev at yahoo.com Sun Feb 27 23:40:32 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 27 Feb 2005 14:40:32 -0800 (PST) Subject: [Pw_forum] About energy of a single atom In-Reply-To: <309430221.10006@mail.sdu.edu.cn> Message-ID: <20050227224032.20336.qmail@web60301.mail.yahoo.com> Hi, --- Shujun Hu wrote: > I tried to calculate the energy of a single atom in > order to gain the bond energy. It is not so clear what do you mean. Did you calculate the energy using atomic program or using pw.x and supercell approximation? > Although such information is given in the PP, i > donnt trust it. Sorry, but I don't see any information about bond energy in a PsP file, say, for Al. There is just the total energy for atomic configuration used for PsP generation. So, it will be different for different atomic configurations. And methods, too. Do not trust, do not use. >One of the important parameters is the atomic > distance. Maybe 10 times of the atomic radius > between atoms is better, and such system may be > considered as "alone". Yes or no? > And the E-cutoff, how to set? The distance could be approved just on the base of tests with different size of a supercell where there is your atom. The choice of the cutoff energy (ecutwfc) depends on a pseudopotential in use. If you would like to use norm-conserving pseudopotential for 3d-metals, and the first-row elements (like C, O), you should use cutoff energy around 100Ry to get good results. For test purpose ~80Ry is OK, I believe. For ultrasoft pseudopotentials usually the energy is around 25-35Ry. Bests, Eyvaz. __________________________________ Do you Yahoo!? Yahoo! Mail - now with 250MB free storage. Learn more. http://info.mail.yahoo.com/mail_250 From hushujun at mail.sdu.edu.cn Mon Feb 28 02:53:00 2005 From: hushujun at mail.sdu.edu.cn (Shujun Hu) Date: Mon, 28 Feb 2005 09:53:00 +0800 Subject: [Pw_forum] About energy of a single atom Message-ID: <309555580.05774@mail.sdu.edu.cn> Dear Isaev, I mean I want to calculate the bond energy of O2 molecular, to test the PP, using such methord followed: E(single atom)*2-E(O2 molecular)=>E(bond) , as E(single atom) including the energy of valence electrons only. The value of E(single atom) can be obtained from the PP file as "totle energy". Right? The biggest difficulty for me is obtaining the energy of O2 molecular under the periodic boundary condition. You know the distance between O2 moleculars is much larger than the bond length. How can i set up the value of celldm(1), celldm(2)... in the pw.x calculating? Best wishes Shujun Hu ____________________________ hushujun at mail.sdu.edu.cn From sunsr at ihep.ac.cn Mon Feb 28 07:53:34 2005 From: sunsr at ihep.ac.cn (Shaorui Sun) Date: Mon, 28 Feb 2005 14:53:34 +0800 Subject: [Pw_forum] about the elastic constants computation Message-ID: <200502280633.j1S6XPt7021044@mail.ihep.ac.cn> Dear all: who could tell me how to identify the ealstic constants such as C11,C12 and C44, with pwscf? Best Regards! S. R. Sun --------------------------------------------- Beijing Synchrotron Radiation Facility Institute of High Energy Physics Chinese Academy of Sciences P. O. Box 918, 100049 Beijing P. R. China Tel: 0086+10 88236710 email:sunsr at ihep.ac.cn From eyvaz_isaev at yahoo.com Mon Feb 28 08:40:44 2005 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Sun, 27 Feb 2005 23:40:44 -0800 (PST) Subject: [Pw_forum] about the elastic constants computation In-Reply-To: <200502280633.j1S6XPt7021044@mail.ihep.ac.cn> Message-ID: <20050228074044.94508.qmail@web60305.mail.yahoo.com> Hi, --- Shaorui Sun wrote: > who could tell me how to identify the > ealstic constants such as C11,C12 and C44, with > pwscf? You can evaluate these constants by means of total energy calculations. I.e. you have to calculate the total energy for undistorted and strained (usually, tetragonal and monoclinic straines) lattices. And then fitting E(\delta)-E(0) with respect to distorsion by an appropriate function. Calculations oh these parameters via phonons seem to be less accurate due to the fact that the slope of acoustic modes in the long-wave limit (i.e. near the \Gamma point) is very sensitive to parameters used in such a kind of calculations. Bests, Eyvaz. __________________________________ Do you Yahoo!? Yahoo! Mail - 250MB free storage. Do more. Manage less. http://info.mail.yahoo.com/mail_250 From hushujun at mail.sdu.edu.cn Mon Feb 28 09:47:00 2005 From: hushujun at mail.sdu.edu.cn (Shujun Hu) Date: Mon, 28 Feb 2005 16:47:00 +0800 Subject: [Pw_forum] insufficient virtual memory for a Zn48O48 system!!!!!!!! Message-ID: <309580420.24578@mail.sdu.edu.cn> Dear Drs, I try to do the calculation for a 96-atom system, Zn48O48 (before doping, for the purpose of comparing). There are totle 864 electrons, 12 for Zn and 6 for O respectively. And the number of bands is set to be 450. After running the memory.x system gave such information: ********************************************************** Program memory v.2.1.2 starts ... Today is 28Feb2005 at 15:49:49 Estimated Max memory (k-point code): 14690.79Mb nonscalable memory = 123.65Mb scalable memory = 3654.75Mb nonscalable wspace = 148.32Mb scalable wspace =10764.07Mb (diag:10764.07Mb, mix: 287.04Mb) Estimates based on 16-byte complex, 8-byte reals, 4-byte integers ********************************************************** As my pc taking only 512M physical momory, I selected "conjugate-gradient diagonalization" mode. Then i run memory.x again: ********************************************************** Program memory v.2.1.2 starts ... Today is 28Feb2005 at 15:54:12 Estimated Max memory (k-point code): 4048.99Mb nonscalable memory = 97.94Mb scalable memory = 3654.75Mb nonscalable wspace = 9.27Mb scalable wspace = 287.04Mb (diag: 23.92Mb, mix: 287.04Mb) Estimates based on 16-byte complex, 8-byte reals, 4-byte integers ********************************************************** Then i run pw.x programme. Unfortunately such information appeared: ********************************************************** ...... threshold for starting DIIS: 0.0010 reduced basis size: 3 forrtl: severe (41): insufficient virtual memory Image PC Routine Line Source pw.x 082622F4 Unknown Unknown Unknown Stack trace terminated abnormally. ********************************************************** Noting that there is insufficient virtual memory (swap partition is working, but only 2G for Redhat linux), I set up an 5G active swapfile to add to the virtual memory. But same information appeared as above. Maybe such system is too large. I want to know why it cannt work as there is enough virtual memory. Should i add the physical memory? Best wishes Shujun Hu ____________________________ hushujun at mail.sdu.edu.cn From degironc at sissa.it Mon Feb 28 10:34:34 2005 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 28 Feb 2005 10:34:34 +0100 Subject: [Pw_forum] Mg pseudopotential References: <421C4341.4060709@sw.ehu.es> Message-ID: <4222E5AA.8030308@sissa.it> Aritz Leonardo wrote: > Hi, > > I am trying to calculate surface phonons of Mg using "bhs" > pseudopotential provided in the pwscf web. The phonon frequencies > resulting from the calculations are too high. To improve this I think > that it is necessary to use a PP with nonlinear correction, this way > the equilibrium lattice constant is bigger and frequencies low down a > bit. > > Does anybody know how to obtain a norm conserving pseudo with linear > correction which is NOT ULTRASOFT? The pseudopotential Mg.pz-n-vbc.UPF that you can find at the pwscf web page is a norm-conserving PP with nlcc and it has been used in a number of systems, including thermal expansion of Mg surfaces. look at the details item in the pseudopotential page. best regards, stefano > > thanks > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Mon Feb 28 10:55:45 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 28 Feb 2005 10:55:45 +0100 Subject: [Pw_forum] error in during bands.x In-Reply-To: <20050227214500.36911.qmail@web60307.mail.yahoo.com> References: <20050227214500.36911.qmail@web60307.mail.yahoo.com> Message-ID: <200502281055.45573.giannozz@nest.sns.it> On Sunday 27 February 2005 22:45, Eyvaz Isaev wrote: > --- Sergey Lisenkov wrote: > > I looked at sources and found that maxdeg=4 (max > > allowed degeneracy). I am interested what is the > > value I should use? > > Perhaps you have eigenvalues with more than 4-fold > degeneration. Try maxdeg>4, say maxdeg=10 (just as an > example, to be more precise, you have to look at your > bands file), recompile the program and try again. the algorithm assumes that all states whose energies are closer than eps, presently set to 0.001 Ry, are degenerate. This information is used to deal with band crossing at high symmetry points. Accidental degeneracies can occur, though. I think that maxdeg can be safely increased, even if group theory says the opposite. Alternatively, eps can be reduced to prevent accidental degeneracies. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Feb 28 12:06:48 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 28 Feb 2005 12:06:48 +0100 Subject: [Pw_forum] Pt PP problem with RRKJ3 - one pseudo wave is zero In-Reply-To: <002f01c51b83$22747390$0ee35ba5@ch0316> References: <002f01c51b83$22747390$0ee35ba5@ch0316> Message-ID: <200502281206.48669.giannozz@nest.sns.it> On Friday 25 February 2005 22:44, Eduardo J. Lamas wrote: > After drawing the pseudowavefunctions I found that the column > associated to the s wave is all zeros. it is a known problem, but it doesn't affect the generation of the PP > I have to use latter correction, because if not used the all electron > wfc will fail to converge. don't take too seriously those warnings: usually they are present only in the first scf iterations. In any case, you can use a different (ionized) reference configuration. > But I am departing from the recipe given in the > manual for obtaining revPBE. Is that a serious problem? the Latter correction forces the potential to have the correct asymptotic behavior. It is a sort of "poor-man" self-interaction correction. In principle one should perform the calculation in the solid or molecule using the Latter correction as well, to be consistent, but this is of course not possible. In most cases the asymptotic behavior of the potential is not that important in a solid, but in some cases it is. In DFT it is wrong anyway. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From giannozz at nest.sns.it Mon Feb 28 12:25:42 2005 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 28 Feb 2005 12:25:42 +0100 Subject: [Pw_forum] Na PW91 pseudo messes up FPMD In-Reply-To: <421F6179.7010000@sissa.it> References: <421F6179.7010000@sissa.it> Message-ID: <200502281225.42537.giannozz@nest.sns.it> On Friday 25 February 2005 18:33, Davide Ceresoli wrote: > and the initial forces on Cl atoms are NaN ... Sodium Nitride is a nasty compound ! > Can I use the same pseudopotentials in PW and FPMD > (norm-conserving) or there is some caveat? you can. Please post a job demostrating the problem you are observing (with enough information to reproduce it) Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From axel.kohlmeyer at theochem.ruhr-uni-bochum.de Mon Feb 28 14:56:29 2005 From: axel.kohlmeyer at theochem.ruhr-uni-bochum.de (Axel Kohlmeyer) Date: Mon, 28 Feb 2005 14:56:29 +0100 Subject: [Pw_forum] insufficient virtual memory for a Zn48O48 system!!!!!!!! In-Reply-To: <309580420.24578@mail.sdu.edu.cn> Message-ID: <200502281356.j1SDuTk01633@yello.theochem.ruhr-uni-bochum.de> On Mon, 28 Feb 2005 16:47:00 +0800 "Shujun Hu" wrote: [...] > Dear Drs, > I try to do the calculation for a 96-atom system, Zn48O48 (before doping, for the > purpose of comparing). There are totle 864 electrons, 12 for Zn and 6 for O > respectively. And the number of bands is set to be 450. After running the memory.x dear shujun hu, wow, this is quite a large system. > system gave such information: > ********************************************************** > Program memory v.2.1.2 starts ... > Today is 28Feb2005 at 15:49:49 > Estimated Max memory (k-point code): 14690.79Mb > nonscalable memory = 123.65Mb > scalable memory = 3654.75Mb > nonscalable wspace = 148.32Mb > scalable wspace =10764.07Mb (diag:10764.07Mb, mix: 287.04Mb) > Estimates based on 16-byte complex, 8-byte reals, 4-byte integers > ********************************************************** > As my pc taking only 512M physical momory, I selected "conjugate-gradient whoa. with that kind of machine, you'll have to wait _ages_ until you'll see a result. also the machine will grind your harddrive easily to death. > Noting that there is insufficient virtual memory (swap partition is working, but > only 2G for Redhat linux), I set up an 5G active swapfile to add to the virtual > memory. But same information appeared as above. > Maybe such system is too large. I want to know why it cannt work as there is > enough virtual memory. Should i add the physical memory? you may have enough virtual memory in theory, but you have to remember that you have only a 32bit cpu, and that will restrict your maximum available address space. under the best circumstances (the proper kernel and a compiler with a 'clean' runtime), there should be a limit of 3GB minus some change for your maximum address space. most of the time, you'll be limited to 2GB or less. what you need is a 64bit machine (or even better a 64bit SMP machine) with at least 8GB of memory. with anything less, your machine will come almost to a halt, as swapspace is so much slower than physical memory. for testing, i'd recommend you cut the system size in half in each direction so that you have a unitcell of 12 atoms, take a denser k-point mesh and use that to pre-optimize your structure and get a feeling for what works (and what doesn't) and how long the 8x larger system will take to compute. regards, axel kohlmeyer. > > Best wishes > > Shujun Hu > ____________________________ > hushujun at mail.sdu.edu.cn > > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- ======================================================================= Axel Kohlmeyer e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de Lehrstuhl fuer Theoretische Chemie Phone: ++49 (0)234/32-26673 Ruhr-Universitaet Bochum - NC 03/53 Fax: ++49 (0)234/32-14045 D-44780 Bochum http://www.theochem.ruhr-uni-bochum.de ======================================================================= If you make something idiot-proof, the universe creates a better idiot.