From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: choose the input parameter of 'cmass', no matter what kind of VCSMD you use e.g., 'nm','cm' or 'mm'. The values of cmass for these kinds of MD are quite different. I can get a good result for geometry minisation using 'cm' method (I have shown the input parameters in my last email). The cmass in 'cm' should be the real cell mass from my experience. While cmass in 'mm' could be very small(say 0.0001), as I discussed with Guido , but i don't have any experience for 'nm' and 'mm' MD. I guess that the reason you can not get a good minisation at a give pressure is because you choose a wrong CMASS, also it doesn't make sense that "increasing the mass of the unit cell to damp the oscillations..". It depends on what kind of MD you use. So you can try to do geometry minisation using 'cm' and, if possible, let me know your results.But sometimes I got error message "cdiaghg: error", I haven't figure out yet what is the reason for the error. Best Wishes Yanming Ma Steacie Inst. for Molecular Sciences Naional Research Council of Canada _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: PWSCF, say 1.2, came out already. I can not find this new version in the PWSCF homepage. I just wonder that how can I get this new version. Best Regards Yanming Ma PhD Steacie Institute for Molecular Sciences, National Research Councils of Canada. 100 Sussex K1A 0R6 _________________________________________________________________ ???????? MSN Explorer: http://explorer.msn.com/lccn From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: step 40 it continue increasing. It seems that it still hasn't converged. I am not clear the principle for this Gaussian broadening procedures. In my case, the Fermi energy is around 19.47 eV. I am wondering if it is reasonable if I take the positive number (in the final 40 step) as final lambda value. But at the step 40, the calculated lambda is at fermi energy 18.9 eV which is far away from 19.47eV. Any idea on the principle for Gaussian broadening procedures. Can I take the positive lambda after 40 step Gaussian broadening and by 35 step negative number? THanks Yanming Ma PhD Steacie Institute for Molecular Sciences, National Research Councils of Canada. 100 Sussex K1A 0R6 _________________________________________________________________ ???????????????????????????? MSN Messenger: http://messenger.msn.com/cn From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: the slaves receive by MPI_Recv() with in a loop and again send all data set to the master by using MPI_Recv() with in a loop.. The loop is used for how many data they receive and how many data they send just keep the track... $ time mpirun -np 5 a.out 0 - MPI_RECV : Message truncated [0] Aborting program ! [0] Aborting program! I am processor 4I am processor 3I am processor 2I am processor 1/home/software/mpich-1.2.5/bin/mpirun: line 1: 1699 Broken pipe /home/st027364/sort/a.out -p4pg /home/st027364/sort/PI1551 -p4wd /home/st027364/sort real 0m7.002s user 0m0.100s sys 0m0.110s Thanks.. Shamim --------------------------------- Do you Yahoo!? Friends. Fun. Try the all-new Yahoo! Messenger --0-219863464-1086902058=:25743 Content-Type: text/html; charset=us-ascii
Hello,
When I run a MPI program then the following error has shown. I dont know why..Some times it shows and sometimes not...sometimes work properly....
What the problem?
What i have done is here.. 
From master node i send 55 data set to 5 processors by a MPI_SEND() with in a loop...
the slaves receive by MPI_Recv() with in a loop and again send all data set to the master by using MPI_Recv() with in a loop..
The loop is used for how many data they receive and how many data they send just keep the track...
 
 
$ time mpirun -np 5 a.out
0 - MPI_RECV : Message truncated
[0]  Aborting program !
[0] Aborting program!
I am processor 4I am processor 3I am processor 2I am processor 1/home/software/mpich-1.2.5/bin/mpirun: line 1:  1699 Broken pipe             /home/st027364/sort/a.out -p4pg /home/st027364/sort/PI1551 -p4wd /home/st027364/sort
real    0m7.002s
user    0m0.100s
sys     0m0.110s
 
Thanks..
Shamim

Do you Yahoo!?
Friends. Fun. Try the all-new Yahoo! Messenger --0-219863464-1086902058=:25743-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: Hi Timothy, Just some thoughts on your job... Have you tried a more physical distance between the two Si atoms? It can be very difficult to optimize the electronic wavefunction when they are almost right on top of each other. Also from what I understand, the procedure can be to fix the atoms, optimize the electronic wavefunction (e-kinetic energy goes to approximately zero) and then begin with moving the ions (in fact this becomes more than one input file). Finally if you can show your output we can more correctly diagnose your problem(s). I hope this helps, Aaron On Tue, 16 Nov 2004, Timothy A.V. Teatro wrote: > Hi. I'm really sorry to ask a question like this. I wrote this input > file a while ago. I have been fiddling with it for a while now, and I > can't figure out why I can't get two simple little silicon atoms to > move! I am accustomed to using PWscf, but I am not trying to switch to > FPMD for more accurate dynamics. But I cannot get any dynamics out of > this thing! Even when I try putting the Si atoms very close to each > other, where the forces should be enormous I can't get them to move. > > Thanks a lot for looking at this. I'm sure I did something very silly, > which is going to embarrass me later. > > Here is a verbatim listing of the input script: > > #!/bin/sh > > PW_ROOT=$HOME/cpmd/FPMD > PSEUDO=$HOME/cpmd/pseudo > TMP_DIR=$HOME/cpmd/tmp0 > > if [ ! -d results ]; then > mkdir results > fi > cd results > > rm -rf $TMP_DIR/* > > cat > sih.cp1.in << EOF > > &CONTROL > title = ' Silicon ', > calculation = 'cp', > restart_mode = 'from_scratch', > pseudo_dir = '$PSEUDO/', > outdir='$TMP_DIR/', > ndr = 51, > ndw = 51, > nstep = 50, > iprint = 10, > isave = 100, > tstress = .TRUE., > tprnfor = .TRUE., > dt = 5.0d0, > etot_conv_thr = 1.d-9, > ekin_conv_thr = 1.d-4, > prefix = 'si' > / > > &SYSTEM > ibrav = 1, > celldm = 10.61, > nat = 2, > ntyp = 1, > nbnd = 4, > nelec = 8, > ecutwfc = 8.0, > xc_type = 'BLYP' > / > > &ELECTRONS > emass = 400.d0, > emass_cutoff = 2.5d0, > orthogonalization = 'ortho', > ortho_eps = 5.d-8, > ortho_max = 15, > electron_dynamics = 'sd', > ! electron_damping = 0.3, > electron_velocities = 'zero', > electron_temperature = 'not_controlled', > / > > &IONS > ion_dynamics = 'verlet', > ion_damping = 0.2, > ion_radius(1) = 0.8d0, > ion_radius(2) = 0.8d0, > ion_velocities = 'zero', > ion_temperature = 'nose', > tempw = 1000, > fnosep = 1.2, > tolp=100, > ion_nstepe = 10 > / > > &CELL > cell_dynamics = 'pr', > cell_velocities = 'zero', > press = 0.0d0, > / > > ATOMIC_SPECIES > Si 28.086 Si.vbc.UPF 4 > > ATOMIC_POSITIONS > Si 0.00 0.00 0.00 > Si 0.00 0.00 0.10 > > K_POINTS > 1 > 0.0 0.0 0.0 1.0 > EOF > > $PW_ROOT/par2.x < sih.cp1.in > Si.md64H.out > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > --__--__-- Message: 2 From: Paolo Giannozzi Organization: Scuola Normale Superiore di Pisa To: pw_forum at pwscf.org Subject: Re: [Pw_forum] error in phonon calculation Date: Wed, 17 Nov 2004 18:24:36 +0100 Reply-To: pw_forum at pwscf.org On Wednesday 17 November 2004 13:34, Jose C. Conesa wrote: > Concerning the answers by Stefano, Paolo and Eyvaz to my question: > > From Stefano: > > The message you mention should come out when performing a calculation > > with lsda flag on and fixed occupation . is this your situation ? > > No. The system is indeed spin-polarized (nspin = 2, > starting_magnetization(1) = 1) and uses lda, but I have > "occupations = 'tetrahedra' " are you sure? Phonon calculation in metals with the tetrahedron method is not implemented: the code stops immediately if you try. Only smearing is implemented. > From Paolo: > >actually this message shows up also in a spin-polarized calculation > > with free occupations but no smearing or tetrahedron technique. > > This case used to be forbidden in the self-consistent calculation, > > but I think it is allowed now, so the message might be obsolete. > > Well, then: should I disregard the message, and rely on the calculation > (from this point of view) after all? I am not sure, but it seems to me that in a spin-polarized system the number of occupied bands might be incorrectly computed by the phonon code in that particular case Please do not post to pw_forum-admin at pwscf.org 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 End of Pw_forum Digest ------=_NextPart_000_0005_01C4CD67.FF6DAD80 Content-Type: application/octet-stream; name="Si.md2.out" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="Si.md2.out" =3D----------------------------------------------------------------------= --------=3D=0A= AB-INITIO COSTANT PRESSURE MOLECULAR DYNAMICS=0A= A Car-Parrinello Parallel Code=0A= =3D----------------------------------------------------------------------= --------=3D=0A= =0A= Version: 2.0.4 - Thu Aug 5 11:04:37 EDT 2004 =0A= Authors: Carlo Cavazzoni, Guido Chiarotti, Sandro Scandolo,=0A= Paolo Focher, Gerardo Ballabio=0A= =0A= =3D----------------------------------------------------------------------= --------=3D=0A= =0A= THIS RUN WAS STARTED ON: Thu Nov 18 12:10:39 2004 =0A= =0A= Tasks =3D 1 This task id =3D 0=0A= Warning: card &CELL ignored=0A= Warning: card CELL_DYNAMICS =3D 'PR', ignored=0A= Warning: card CELL_VELOCITIES =3D 'ZERO', ignored=0A= Warning: card PRESS =3D 0.0D0, ignored=0A= Warning: card / ignored=0A= =0A= =0A= =0A= MD PARAMETERS READ FROM STANDARD INPUT=0A= -------------------------------------=0A= =0A= Job Title: Silicon = =0A= Restart Mode =3D from_scratch , Number of MD Steps =3D 50=0A= Print out every 10 MD Steps=0A= Reads from unit =3D 51, Writes to unit =3D 51=0A= Alat =3D 10.610000=0A= Celldm =3D 10.610000 0.000000 0.000000 0.000000 0.000000 = 0.000000=0A= MD Simulation time step =3D 5.0000=0A= Electronic fictitious MASS =3D 400.00=0A= Verlet algorithm for electron dynamics=0A= =0A= =0A= Atomic Pseudopotentials Parameters=0A= ----------------------------------=0A= =0A= ATOMIC PSEUDOPOTENTIAL for SPECIE : 1=0A= Read from file /home/tteatro/cpmd/pseudo//Si.vbc.UPF = =0A= Type is UPF and NONLOCAL. =0A= Local component is ..... : 3=0A= Non local components are : 1 2=0A= Pseudo charge : 4.000, pseudo radius : 0.800=0A= =0A= Pseudopotentials Grid : Channels =3D 3, Mesh =3D 431=0A= dx =3D 0.02500000000001=0A= point radius pseudopotential=0A= 1 0.0013082599 -9.254371 0.000000 0.000000=0A= 107 0.0185171615 -9.251778 0.000000 0.000000=0A= 215 0.2755303950 -8.700627 0.000000 0.000000=0A= 431 61.0041973233 -0.065569 0.000000 0.000000=0A= =0A= Atomic wavefunction Grid : Channels =3D 2, Mesh =3D 431=0A= dx =3D 0.02500000000001=0A= Channels occupation number : 0.0000 0.0000=0A= point radius wavefunction=0A= 1 0.0013082599 0.000184 0.000001=0A= 107 0.0185171615 0.002610 0.000112=0A= 215 0.2755303950 0.046837 0.024953=0A= 431 61.0041973233 0.000000 0.000000=0A= =0A= Real Mesh Report =0A= ---------------- =0A= Global Dimensions Local Dimensions Processor Grid=0A= .X. .Y. .Z. .X. .Y. .Z. .X. .Y. .Z.=0A= 20 20 20 20 20 20 1 1 1=0A= =0A= Stick Mesh=0A= ----------=0A= nst =3D 293, nstw =3D 69=0A= PEs n.st n.stw n.g n.gw=0A= 1 293 69 3743 461=0A= 0 293 69 3743 461=0A= =0A= =0A= Reciprocal Space Mesh=0A= ---------------------=0A= Large Mesh Number of G Small Mesh Number = of G=0A= PE Global Local Max Local Global Local = Max Local=0A= 1 3743 3743 3743 461 461 = 461=0A= =0A= =0A= Estimated Sizes of the problem=0A= ------------------------------=0A= dimension of the problem (byte/pe) : 2850840=0A= =0A= * Reciprocal space initialization ( from ggen ) :=0A= =0A= =0A= Real Space vectors (cart. coord., units: ALAT):=0A= A1 =3D 1.00 0.00 0.00=0A= A2 =3D 0.00 1.00 0.00=0A= A3 =3D 0.00 0.00 1.00=0A= =0A= Primitive vectors (units: 2*PI/ALAT):=0A= B1 =3D 1.00 0.00 0.00=0A= B2 =3D 0.00 1.00 0.00=0A= B3 =3D 0.00 0.00 1.00=0A= =0A= MAXIMUM INDICES (N1H,..., IND1,... ) :=0A= 20 20 20 **** **** 1=0A= =0A= Scaled atomic positions from standard = input=0A= Species 1 atoms =3D 2=0A= Si 0.000000 0.000000 0.000000=0A= Si 0.000000 0.000000 0.018850=0A= =0A= =0A= * TEMPERATURE CONTROL OF IONS WITH NOSE THERMOSTAT=0A= =0A= NOSE MASS: 571624.4828 TEMPERATURE (K): 1000.0000=0A= IONIC DEGREES OF FREEDOM: 3.=0A= =3D=3D 6892 TIME STEPS PER NOSE OSCILL=0A= =0A= =0A= =0A= * FORMF_BASE TIMING: 242.55224609=0A= =0A= ggen (sec) : 0.004=0A= newg (sec) : 0.003=0A= strucf (sec) : 0.000=0A= formf (sec) : 0.288=0A= Occupation number from init=0A= 2.00 2.00 2.00 2.00=0A= =0A= Wave Initialization: random initial wave-functions=0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= =0A= Integral Simulation Time (AU): 0.00=0A= =0A= =0A= TOTAL ENERGY =3D 75.7474806584 A.U.=0A= KINETIC ENERGY =3D 7.9834829918 A.U.=0A= ELECTROSTATIC ENERGY =3D 65.7506096323 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.2595082869 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 5.9034075310 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -1.6305112099 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 3521.4477026658 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.381393E-01 0.497722E-01 -0.400045E+03=0A= -0.473994E-01 0.523415E-01 0.399987E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= 0.00 0.00 0.00 0.00=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 0 0.00000 0.000 75.74748 0.0000 75.74748 75.74748 = 0.4=0A= MAIN: doions =3D F=0A= 1 0.55370 0.000 74.62782 0.0000 74.62782 75.18153 = 0.0=0A= MAIN: doions =3D F=0A= 2 0.53663 0.000 73.53987 0.0000 73.53987 74.07650 = 0.0=0A= MAIN: doions =3D F=0A= 3 0.51317 0.000 72.49591 0.0000 72.49591 73.00908 = 0.0=0A= MAIN: doions =3D F=0A= 4 0.48234 0.000 71.51018 0.0000 71.51018 71.99251 = 0.0=0A= MAIN: doions =3D F=0A= 5 0.44393 0.000 70.59826 0.0000 70.59826 71.04219 = 0.0=0A= MAIN: doions =3D F=0A= 6 0.39963 0.000 69.77325 0.0000 69.77325 70.17288 = 0.0=0A= MAIN: doions =3D F=0A= 7 0.35249 0.000 69.04254 0.0000 69.04254 69.39502 = 0.0=0A= MAIN: doions =3D F=0A= 8 0.30575 0.000 68.40674 0.0000 68.40674 68.71249 = 0.0=0A= MAIN: doions =3D F=0A= 9 0.26198 0.000 67.86085 0.0000 67.86085 68.12283 = 0.0=0A= =0A= =0A= * MD STEP <> 10 <> =0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= MAIN: doions =3D F=0A= =0A= Integral Simulation Time (AU): 50.00=0A= =0A= =0A= TOTAL ENERGY =3D 67.3962560734 A.U.=0A= KINETIC ENERGY =3D 5.3471192594 A.U.=0A= ELECTROSTATIC ENERGY =3D 60.8800984863 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.3213204309 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 5.7406864600 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -2.2503277014 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 2412.0054683125 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.166919E+00 0.151716E+00 -0.399834E+03=0A= -0.223136E+00 0.151760E+00 0.399774E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= -17.68 -8.87 -4.10 9.82=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 10 0.22273 0.000 67.39626 0.0000 67.39626 67.61899 = 0.0=0A= MAIN: doions =3D F=0A= 11 0.18864 0.000 67.00269 0.0000 67.00269 67.19132 = 0.0=0A= MAIN: doions =3D F=0A= 12 0.15970 0.000 66.66965 0.0000 66.66965 66.82935 = 0.0=0A= MAIN: doions =3D F=0A= 13 0.13554 0.000 66.38729 0.0000 66.38729 66.52283 = 0.0=0A= MAIN: doions =3D F=0A= 14 0.11557 0.000 66.14690 0.0000 66.14690 66.26247 = 0.0=0A= MAIN: doions =3D F=0A= 15 0.09916 0.000 65.94099 0.0000 65.94099 66.04015 = 0.0=0A= MAIN: doions =3D F=0A= 16 0.08570 0.000 65.76337 0.0000 65.76337 65.84908 = 0.0=0A= MAIN: doions =3D F=0A= 17 0.07466 0.000 65.60895 0.0000 65.60895 65.68361 = 0.0=0A= MAIN: doions =3D F=0A= 18 0.06554 0.000 65.47364 0.0000 65.47364 65.53918 = 0.0=0A= MAIN: doions =3D F=0A= 19 0.05798 0.000 65.35416 0.0000 65.35416 65.41214 = 0.0=0A= =0A= =0A= * MD STEP <> 20 <> =0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= MAIN: doions =3D F=0A= =0A= Integral Simulation Time (AU): 100.00=0A= =0A= =0A= TOTAL ENERGY =3D 65.2478959668 A.U.=0A= KINETIC ENERGY =3D 4.9931321626 A.U.=0A= ELECTROSTATIC ENERGY =3D 60.2354145661 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.0222376953 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 4.6087648766 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -2.5671779432 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 2431.8023919531 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.745937E-01 0.804582E-01 -0.399634E+03=0A= -0.137135E+00 0.772021E-01 0.399785E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= -15.60 -9.72 -8.64 1.55=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 20 0.05165 0.000 65.24790 0.0000 65.24790 65.29955 = 0.0=0A= MAIN: doions =3D F=0A= 21 0.04631 0.000 65.15277 0.0000 65.15277 65.19908 = 0.0=0A= MAIN: doions =3D F=0A= 22 0.04176 0.000 65.06710 0.0000 65.06710 65.10886 = 0.0=0A= MAIN: doions =3D F=0A= 23 0.03784 0.000 64.98956 0.0000 64.98956 65.02740 = 0.0=0A= MAIN: doions =3D F=0A= 24 0.03444 0.000 64.91907 0.0000 64.91907 64.95350 = 0.0=0A= MAIN: doions =3D F=0A= 25 0.03145 0.000 64.85473 0.0000 64.85473 64.88618 = 0.0=0A= MAIN: doions =3D F=0A= 26 0.02882 0.000 64.79582 0.0000 64.79582 64.82464 = 0.0=0A= MAIN: doions =3D F=0A= 27 0.02648 0.000 64.74174 0.0000 64.74174 64.76821 = 0.0=0A= MAIN: doions =3D F=0A= 28 0.02437 0.000 64.69197 0.0000 64.69197 64.71634 = 0.0=0A= MAIN: doions =3D F=0A= 29 0.02248 0.000 64.64609 0.0000 64.64609 64.66857 = 0.0=0A= =0A= =0A= * MD STEP <> 30 <> =0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= MAIN: doions =3D F=0A= =0A= Integral Simulation Time (AU): 150.00=0A= =0A= =0A= TOTAL ENERGY =3D 64.6037376489 A.U.=0A= KINETIC ENERGY =3D 4.9152859994 A.U.=0A= ELECTROSTATIC ENERGY =3D 59.9003570129 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.0388805555 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 4.5313480878 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -2.7043728957 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 2434.3792268399 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.383148E-01 0.391792E-01 -0.399573E+03=0A= -0.765900E-01 0.354219E-01 0.399790E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= -13.76 -8.35 -7.98 -2.81=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 30 0.02076 0.000 64.60374 0.0000 64.60374 64.62450 = 0.0=0A= MAIN: doions =3D F=0A= 31 0.01920 0.000 64.56459 0.0000 64.56459 64.58378 = 0.0=0A= MAIN: doions =3D F=0A= 32 0.01776 0.000 64.52836 0.0000 64.52836 64.54613 = 0.0=0A= MAIN: doions =3D F=0A= 33 0.01645 0.000 64.49482 0.0000 64.49482 64.51127 = 0.0=0A= MAIN: doions =3D F=0A= 34 0.01524 0.000 64.46376 0.0000 64.46376 64.47899 = 0.0=0A= MAIN: doions =3D F=0A= 35 0.01412 0.000 64.43497 0.0000 64.43497 64.44909 = 0.0=0A= MAIN: doions =3D F=0A= 36 0.01308 0.000 64.40830 0.0000 64.40830 64.42139 = 0.0=0A= MAIN: doions =3D F=0A= 37 0.01212 0.000 64.38360 0.0000 64.38360 64.39572 = 0.0=0A= MAIN: doions =3D F=0A= 38 0.01123 0.000 64.36071 0.0000 64.36071 64.37193 = 0.0=0A= MAIN: doions =3D F=0A= 39 0.01040 0.000 64.33951 0.0000 64.33951 64.34991 = 0.0=0A= =0A= =0A= * MD STEP <> 40 <> =0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= MAIN: doions =3D F=0A= =0A= Integral Simulation Time (AU): 200.00=0A= =0A= =0A= TOTAL ENERGY =3D 64.3198920756 A.U.=0A= KINETIC ENERGY =3D 4.8112435517 A.U.=0A= ELECTROSTATIC ENERGY =3D 59.7213613035 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.0866420992 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 4.6385683035 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -2.7646389838 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 2368.1901019246 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.230367E-01 0.191356E-01 -0.399592E+03=0A= -0.410947E-01 0.160041E-01 0.399760E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= -12.85 -7.60 -7.45 -5.01=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 40 0.00962 0.000 64.31989 0.0000 64.31989 64.32951 = 0.0=0A= MAIN: doions =3D F=0A= 41 0.00890 0.000 64.30174 0.0000 64.30174 64.31064 = 0.0=0A= MAIN: doions =3D F=0A= 42 0.00823 0.000 64.28495 0.0000 64.28495 64.29318 = 0.0=0A= MAIN: doions =3D F=0A= 43 0.00761 0.000 64.26942 0.0000 64.26942 64.27703 = 0.0=0A= MAIN: doions =3D F=0A= 44 0.00703 0.000 64.25508 0.0000 64.25508 64.26211 = 0.0=0A= MAIN: doions =3D F=0A= 45 0.00649 0.000 64.24184 0.0000 64.24184 64.24833 = 0.0=0A= MAIN: doions =3D F=0A= 46 0.00599 0.000 64.22962 0.0000 64.22962 64.23561 = 0.0=0A= MAIN: doions =3D F=0A= 47 0.00553 0.000 64.21834 0.0000 64.21834 64.22386 = 0.0=0A= MAIN: doions =3D F=0A= 48 0.00510 0.000 64.20794 0.0000 64.20794 64.21303 = 0.0=0A= MAIN: doions =3D F=0A= 49 0.00470 0.000 64.19835 0.0000 64.19835 64.20305 = 0.0=0A= =0A= =0A= * MD STEP <> 50 <> =0A= =0A= =0A= Total integrated electronic density=0A= in G-space =3D 8.000000 in R-space =3D 8.000000=0A= MAIN: doions =3D F=0A= =0A= Integral Simulation Time (AU): 250.00=0A= =0A= =0A= TOTAL ENERGY =3D 64.1895177932 A.U.=0A= KINETIC ENERGY =3D 4.7121644588 A.U.=0A= ELECTROSTATIC ENERGY =3D 59.6577018164 A.U.=0A= ESELF =3D 15.9576912161 A.U.=0A= ESR =3D 64.2069878907 A.U.=0A= PSEUDOPOTENTIAL ENERGY =3D -2.1093835189 A.U.=0A= N-L PSEUDOPOTENTIAL ENERGY =3D 4.7103053625 A.U.=0A= EXCHANGE-CORRELATION ENERGY =3D -2.7812703256 A.U.=0A= VAN DER WAALS ENERGY =3D 0.0000000000 A.U.=0A= EMASS KINETIC ENERGY =3D 2297.8043154331 A.U.=0A= =0A= =0A= =0A= Atomic Positions and Velocities (AU)=0A= =0A= Si 0.0000 0.0000 0.0000 0.0000E+00 0.0000E+00 0.0000E+00=0A= Si 0.0000 0.0000 0.2000 0.0000E+00 0.0000E+00 0.0000E+00=0A= =0A= Center of mass displacement (a.u.): 0.000000=0A= =0A= Cell Variables (AU)=0A= =0A= 10.61000000 0.00000000 0.00000000=0A= 0.00000000 10.61000000 0.00000000=0A= 0.00000000 0.00000000 10.61000000=0A= =0A= System Density [g/cm^3] : 0.5270=0A= =0A= Forces on Ions (AU)=0A= =0A= -0.128058E-01 0.909501E-02 -0.399628E+03=0A= -0.203916E-01 0.694197E-02 0.399729E+03=0A= =0A= Total Stress (KB)=0A= =0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= 0.00000000 0.00000000 0.00000000=0A= =0A= Eigenvalues (eV), kp =3D 1 , spin =3D 1=0A= =0A= -12.55 -7.39 -7.31 -6.22=0A= =0A= =0A= Partial temperatures (for each ionic specie) =0A= Species Temp (K) MSD (AU)=0A= 1 0.00 0.0000=0A= =0A= =0A= NFI EKINC TEMP ETOT EKCELL ECONS ETTT = SEC.=0A= 50 0.00433 0.000 64.18952 0.0000 64.18952 64.19385 = 0.0=0A= =0A= RESTART FILE WRITTEN ON UNIT fort.51 COMPLETED IN 0.013 SEC.=0A= =0A= =0A= Execution time statistics (SEC)=0A= MD step mean time .. 0.031=0A= Total time ......... 1.570=0A= =0A= Statistica FFT sulle wave functions=0A= Numero totale di fft . 612=0A= Tempo medio per loop . 0.00458=0A= Tempo medio per fft .. 0.00038=0A= Tempo totale fft ..... 0.23367=0A= =0A= =0A= Statistica FFT sui potenziali=0A= Numero totale di fft . 461=0A= Tempo medio per loop . 0.00963=0A= Tempo medio per fft .. 0.00107=0A= Tempo totale fft ..... 0.49134=0A= =0A= =0A= PC3FFT TIMINGS=0A= FFTXW FFTYW FFTZW TRASW FFTXP FFTYP FFTZP TRASP=0A= 0.000 0.164 0.022 0.035 0.000 0.180 0.062 0.179=0A= =0A= =0A= Averaged Physical Quantities=0A= accomulated this run=0A= EKINC 0.11249 0.11249 (AU)=0A= TOTEL ENERGY 66.00010 66.00010 (AU)=0A= TEMPERATURE 0.00000 0.00000 (K/1000)=0A= ENTHALPY 66.00010 66.00010 (AU)=0A= ECONS 66.00010 66.00010 (AU)=0A= PRESSURE 0.00000 0.00000 (Kbar)=0A= VOLUME 1194.38998 1194.38998 (AU)=0A= =0A= =0A= =0A= mp_buffers: high_watermark (bytes): 93760=0A= =0A= mp: high_watermark (bytes): 64000=0A= main: Closing unit : 28=0A= main: Closing unit : 41=0A= main: Closing unit : 42=0A= main: Closing unit : 43=0A= From cploop, etot =3D 0.000000000000000E+000=0A= =0A= THIS RUN WAS ENDED ON: Thu Nov 18 12:10:41 2004 =0A= =0A= =3D----------------------------------------------------------------------= --------=3D=0A= User time (sec.) : 1.7700=0A= System time (sec.) : 0.1900=0A= JOB DONE.=0A= =3D----------------------------------------------------------------------= --------=3D=0A= ------=_NextPart_000_0005_01C4CD67.FF6DAD80-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: tefield LOGICAL ( default = .FALSE. ) If .TRUE. a sawlike potential is added to the bare ionic potential. iedir INTEGER 1, 2 or 3. Used only if tefield is .TRUE.. The direction of the electric field is parallel to the bg(:,edir) reciprocal lattice vector ( So the potential is constant in planes defined by the mesh points ) emaxpos REAL ( default = 0.5D0 ) Position of the maximum of the sawlike potential within the unit cell ( 0 < emaxpos < 1 ) eopreg REAL( default = 0.1D0 ) Part of the unit cell where the sawlike potential decreases. ( 0 < eopreg < 1 ) eamp REAL ( default = 0.001 a.u. ) Amplitude of the electric field (in a.u.) ( 1 a.u. = 51.44 10^10 V/m ) From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: c epsil if .true. in a q=0 calculation for a non metal the .false. c macroscopic dielectric constant of the system is c computed. c Best Regards, Andrea On Mon, 2004-11-29 at 11:30, xywu wrote: > For the encode is unshown, i send it again. > > Dear all PWscf users, > > I want to investigate the electric field effect by pwscf . How to use > pwscf by density-functional perturbation theory or self-consistent finite > electric field. I did not find those from the manual. > > Thank you in advance! > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- 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 bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: If [...] the code stops at the first namelist (``control'') and you are running on a PC cluster: your communication library (MPI) might not be properly configured to allow input redirection (so that you are effectively reading an empty file). See section "Running on parallel machines'', or inquire with your local computer wizard (if any). In the section "Running on parallel machines'' you find: Some implementations of the MPI library may have problems with input redirection in parallel. If this happens, use the option -in (or -inp or -input), followed by the input file name. Example: pw.x -in input -npool 4 > 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 _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: ds REAL ( default = 1.5D0 ) optimization step length ( Hartree atomic units ) my ds is now: ds = 0.5D0, I was previously using the default and getting the error below. Thanks for your help, Aaron On Mon, 21 Feb 2005, Paolo Giannozzi wrote: > 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. > > > From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: 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) more info can be found using: grep -i alpha_mix $PW_HOME/*/*.f90 and grep -i alphamix $PW_HOME/*/*.f90 the integer in parentheses is the iteration number for which you wish to set the alpha_mix. Also once the alpha_mix is set, each consecutive iteration takes on the value of the last one if it is not specified. On Thu, 24 Feb 2005, Aritz Leonardo wrote: > 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. > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: DGETRF subroutine which calls the XERBLA error handler (in lapack.f). I ran example24 without problem. So the error might be due to my system or the pseudo-potential, but with so many parameters, for me it's hard to see what went wrong. I am ignorant to programming, so digging into the source code is not a good cnoice. I ask for your hints and suggestions. here are some further information and questions. 1) the error appears after printing out the ionic contribution. So it's somewhere in calculating efg_corr. 2) Could it be something related to metallic phase? I used band smearing. 3) Does the recon file have more restriction than being norm-conserving? I noticed that in atomic_doc/pseudo-gen directory, two files were used to generate Si pseudopotential and reconstruction file. In si_nc_paw.in, there's a line file_pseudopw='SiPBE_nc.bad'. Why does one need separate files for generating pseudopoetial and reconstruction file? 4) Is it related to symmetry? I am calculating a bulk system with no symmtry except 3 translational vectors. I used ibrav=14. I appreciate your help and suggestions! Lilong From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: ------------------------------------------------------------ Best regards Xiyong Wu(X.Y.Wu) E_mail:xywu at imr.ac.cn From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: "IBM with ESSL ? ESSL contains a version of ZHPEV that is incompatible=20 with ZHPEV provided by LAPACK. If __AIX is defined, the former is used. Load ESSL before LAPACK. Paolo " I am very confused, that means I have to install a liblapack.a separately together with essl?=20 =20 Thanks and regards, LIU, LEI -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On Behalf Of Gerardo Ballabio Sent: Friday, May 06, 2005 3:57 PM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] compiling problems of espresso-2.1.3 onIBM-rs6000??? On 05/06/05 05:08:11, Liu Lei wrote: > Yes, but essl is supposed to have all the math lib functions. > LIBS=3D"-lessl -L/usr/lib -llapack" does not work on our machine, as > there is no liblapack.a file in the folder. >=20 > Is there any way to solve this problem? Yes, if lapack library is not installed on your machine, you can set =20 "MYLIB =3D lapack" in the make.sys file after configuration (configure =20 will probably set "MYLIB =3D blas_and_lapack", but if you do have essl, = you don't need blas). Gerardo _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: DOS of the Fermi energy. I wanted do some research on the DOS of Ef with the increasing pressures. what should I do? Best regards Paolo Giannozzi ?????? On Friday 03 June 2005 02:03, Lijun Zhang wrote: > I wanted to calculate the accurate DOS at Fermi energy > of some metal, how to do it? the DOS at Fermi energy is not printed (except in the calculation of electron-phonon coefficients) but the DOS at any energy can be printed using postprocessing code "dos.x". The dos are calculated in PP/dosg.f90 and flib/dost.f90, respectively for gaussian broadening and tetrahedra 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 ------------------------------------------------- Lijun Zhang National Lab of Superhard Materials, Jilin University, ChangChun 130012, P.R.China --------------------------------- DO YOU YAHOO!? ????????G?????????????????????????????????????? --0-245608481-1117853416=:54864 Content-Type: text/html; charset=gb2312 Content-Transfer-Encoding: 8bit
Hi Paolo,
Thanks for your helps.
In fact I have calculated the DOS with tetrahedra method,
but I meet a problem.In scf step there is a Ef,and I integrate
the total DOS,I also get a "fermi energy" at which the integrated
DOS is equal to the number of total electrons in cell.
But there is discrepancies between the two energy,
and there is also  discrepancies(sometimes very notable)
between the two "DOS at Fermi energy".
Is this due to the calculation uncertainty?
From this point I don't know how to determine the accurate
DOS of the Fermi energy.
I wanted do some research on  the DOS of Ef with the increasing
pressures.
what should I do?
Best regards

Paolo Giannozzi <giannozz at nest.sns.it> ??????
On Friday 03 June 2005 02:03, Lijun Zhang wrote:

> I wanted to calculate the accurate DOS at Fermi energy
> of some metal, how to do it?

the DOS at Fermi energy is not printed (except in the calculation
of electron-phonon coefficients) but the DOS at any energy can be
printed using postprocessing code "dos.x". The dos are calculated
in PP/dosg.f90 and flib/dost.f90, respectively for gaussian broadening
and tetrahedra

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


-------------------------------------------------
Lijun Zhang
National Lab of Superhard Materials,
Jilin University,
ChangChun 130012,
P.R.China

DO YOU YAHOO!?
????????G?????????????????????????????????????? --0-245608481-1117853416=:54864-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: PP for Ti, and only two PBE and PW91 GGA PPs of Ba. So, where I can download a GGA PP of Ti, or a LDA PP of Ba? BTW, since the BaTiO3 sometimes is a tetragonal structure. What is the difference between the choice ibrav=3D6 or ibrav=3D7? Thank you very much. Jian From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: ------=_Part_29619_18213343.1132067736716 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Dear all,

I am going to do something with the surface phonons of metals, especially for a surface with an absorbate. But I have no idea how can I fulfil this.
I have read and tried the example02, but I am not sure if I should do the s= ame way as it?
For a surface simulated by a slab supercell, there would be two surfaces, actually, how can I mimic the bulk like substrate? By including thicker slab or just fixed one side as is done in a surface relaxation?

If anyone can share the script to do so, it will be highly appreciate.

Thank you.

Konzern

From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: ------=_Part_29619_18213343.1132067736716-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: (actually, if you have the entire pdf, I am interested for a personal copy...) Thank you very much Regards Jean-Yves Raty -------------------------------------------------------------------------= --- ----------------------- FNRS-Universite de Liege Tel : + 32-4-3663747 Institut de Physique B5 Fax : + 32-4-3662990 B4000 Sart-Tilman Email : jyraty at ulg.ac.be Belgique =20 -----Original Message----- From: pw_forum-admin at pwscf.org [mailto:pw_forum-admin at pwscf.org] On = Behalf Of Paolo Giannozzi Sent: Wednesday, May 10, 2006 3:11 PM To: pw_forum at pwscf.org Subject: Re: [Pw_forum] Using another basis set On Wednesday 10 May 2006 10:01, Diep Quang Vinh wrote: > In the code pw, I know that we use plane wave as the basis set, > is it possible for us to import different set and run the code?=20 "to import results obtained from different basis sets" is in principle possible, even if not trivial. To use a different basis set, you may = want to use a different code. > do you have an idea on how one calculates the KB projector within the=20 > code? see attached, eq. 41 . Integrals are performed numerically=20 on a radial grid. P. --=20 Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513=20 Piazza dei Cavalieri 7 I-56126 Pisa, Italy From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: "configure ARCH=3Dcygwin F77=3Dg95 F90=3Dg95 CC=3Dgcc" This all runs fine as does the "make all" command (although a few warnings= about REAL(8)/COMPLEX(8) inconsistencies are issued by the compiler)=2E When I try running a few examples they fail as follows ---------------------------------------- $ run=5Fexample /c/users/oeffner/tmp/espresso-3=2E1/examples/example04 : starting This example shows how to use pw=2Ex to perform molecular dynamics for 2- and 8-atom cells of Si starting with compressed bonds along (111)=2E executables directory: /c/Users/Oeffner/Tmp/espresso-3=2E1/bin pseudo directory: /c/Users/Oeffner/Tmp/espresso-3=2E1/pseudo temporary directory: C:/Users/Oeffner/tmp/espresso-3=2E1/tmp checking that needed directories and files exist=2E=2E=2E done running pw=2Ex as: /c/Users/Oeffner/Tmp/espresso-3=2E1/bin/pw=2Ex=20 cleaning C:/Users/Oeffner/tmp/espresso-3=2E1/tmp=2E=2E=2E done running the MD calculation for Si in a 2 atom cell=2E G-point=2E=2E=2EAt= line 103 of file error=2Ef90 Traceback: not available, compile with -ftrace=3Dframe or -ftrace=3Dfull STOP 2 done cleaning C:/Users/Oeffner/tmp/espresso-3=2E1/tmp=2E=2E=2E done running the MD calculation for Si in a 8 atom cell=2E G-point=2E=2E=2EAt= line 103 of file error=2Ef90 Traceback: not available, compile with -ftrace=3Dframe or -ftrace=3Dfull STOP 2 done cleaning C:/Users/Oeffner/tmp/espresso-3=2E1/tmp=2E=2E=2E done running the MD calculation for Si in a 2 atom cell=2E G3X-points=2E=2E=2E= At line 103 of file error=2Ef90 Traceback: not available, compile with -ftrace=3Dframe or -ftrace=3Dfull STOP 2 done /c/users/oeffner/tmp/espresso-3=2E1/examples/example04: done ---------------------------------------- One of the output files, si=2Emd8=2Eout, then reads as follows ------------------------------------------- Program PWSCF v=2E3=2E1 starts =2E=2E=2E Today is 19Jun2006 at 14: 3: 6=20 Ultrasoft (Vanderbilt) Pseudopotentials Current dimensions of program pwscf are: ntypx =3D 10 npk =3D 40000 lmax =3D 3 nchix =3D 6 ndmx =3D 2000 nbrx =3D 14 nqfx =3D 8 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= %% %%% from readpp : error # 3 file /c/Users/Oeffner/Tmp/espresso-3=2E1/pseudo/Si=2Evbc=2EUPF not fo= und %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= %% %%% stopping =2E=2E=2E ------------------------------------------- But on my system the Si=2Evbc=2EUPF is present as evident from the followi= ng command: ------------------------------------------- $ ls $PSEUDO=5FDIR/Si=2Evbc=2EUPF /c/Users/Oeffner/Tmp/espresso-3=2E1/pseudo/Si=2Evbc=2EUPF ------------------------------------------- Does anyone know what could be wrong=3F This is the first time ever I've worked with espresso so any advice would be welcome=2E Thanks, Rob -------------------------------------------------------------------- mail2web - Check your email from the web at http://mail2web=2Ecom/ =2E From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: pw.x determines first the symmetry operations (rotations) of the Bravais lattice; then checks which of these are symmetry operations of the system (including if needed fractional translations). This is done by rotating (and translating if needed) the atoms in the unit cell and verifying if the rotated unit cell coincides with the original one. If a symmetry operation contains a fractional translation that is incompatible with the FFT grid, it is discarded in order to prevent problems with symmetrization. Typical fractional translations are 1/2 or 1/3 of a lattice vector. If the FFT grid dimension along that direction is not divisible respectively by 2 or by 3, the symmetry operation will not transform the FFT grid into itself. Bests, Eyvaz. --- lan haiping wrote: > Thank you , Eyvaz. > I still have a confusion : why does the modified > ecutrho affect the > system's symmetry ? > > Regards > > Hai-Ping > > On 8/20/06, Eyvaz Isaev > wrote: > > > > OK, let me hint more. > > > > Anatase has a Pearson symbol tI12 or space group > > I4_1/amd. So, ibrav should be 7, but you have > chosen a > > monoclinic lattice, i.e. ibrav=14. The space group > has > > 16 symmetry elements, but your choice after > correcting > > ecutrho gives only 4 symmetry elements . > > > > Imagine FCC lattice and its conventional basis > > vectors now. According to you now we have to put > > ibrav=4 or 5. But ibrav=2 for FCC lattice. > > > > In addition, I also checked your input file via > > XCrysDen before I replied previously. Yes, there > is no > > problem with it, but a figure does not look like > > anatase though you used nonzero ibrav. XcrysDen > > usually replicates atomic positions if > ibrav.neq.0. > > For example, if you put ibrav=0 for FCC cell and > > specify conventional CELL_PARAMETERS you will find > a > > rhombohedron, but using ibrav=2 and default basis > > vectors you find a cubic cell. > > > > Bests, > > Eyvaz. > > > > > > --- lan haiping wrote: > > > > > Dear Eyvaz, > > > Thank you. > > > I checked the strcuture again using XCRYSDEN, > > > and found no problems related to the lattices. > > > > > > Regards, > > > > > > Hai-Ping > > > > > > > > > > > > On 8/20/06, Eyvaz Isaev > > > wrote: > > > > > > > > Hi, > > > > > > > > It seems your lattice choice is not correct. > > > Please > > > > have a look at > > > > > http://cst-www.nrl.navy.mil/lattice/struk/c5.html > > > > > > > > Bests, > > > > Eyvaz. > > > > > > > > --- lan haiping wrote: > > > > > > > > > Dear All, > > > > > I also have calculated gamma phonon > frequencies > > > of > > > > > anatase TiO2 . > > > > > I run ph.x caculation after relaxing the > > > structure, > > > > > but 9 of 18 frequencies > > > > > are minus. > > > > > > > > > > Would you please give me some hints and > help? > > > > > > > > > > input settings are below, thank you in > advance ! > > > > > > > > > > *phonons of TiO2 at Gamma > > > > > &inputph > > > > > tr2_ph=1.0d-14, > > > > > recover=.true., > > > > > epsil=.true., > > > > > prefix='tio2', > > > > > fildyn='TiO2-gamma.dyn', > > > > > outdir='/home/haiping/tmp/' > > > > > / > > > > > 0.0 0.0 0.0* > > > > > ** > > > > > *Relaxing calculation input :* > > > > > *&CONTROL > > > > > title = 'TiO2' , > > > > > calculation = 'relax', > > > > > verbosity = 'default', > > > > > tprnfor = .true., > > > > > outdir = '/home/haiping/tmp/', > > > > > prefix = 'tio2', > > > > > disk_io = 'default', > > > > > nstep = 200, > > > > > pseudo_dir = > > > > > '/home/haiping/espresso/pseudo/', > > > > > / > > > > > &SYSTEM > > > > > ibrav= 14, > > > > > > > > > > > celldm(1)=10.2856,celldm(2)=1.0,celldm(3)=1.0, > > > > > > > > > > celldm(4)=-0.759357,celldm(5)=-0.759357, > > > > > celldm(6)=0.518714 > > > > > nat = 6 , > > > > > ntyp = 2 , > > > > > nbnd = 30 , > > > > > ecutwfc = 30.0000000000, > > > > > ecutrho = 160.000, > > > > > occupations = 'fixed' , > > > > > / > > > > > > > > > > > > > > > &ELECTRONS > > > > > electron_maxstep = 100, > > > > > conv_thr = 1.0e-07, > > > > > mixing_mode = 'plain' , > > > > > startingpot = 'atomic' , > > > > > startingwfc = 'atomic' , > > > > > mixing_beta = 0.300000000, > > > > > mixing_ndim = 8, > > > > > diagonalization = 'david_overlap' , > > > > > diago_david_ndim = 8, > > > > > / > > > > > > > > > > &IONS > > > > > ion_dynamics = 'bfgs', > > > > > pot_extrapolation = > 'first_order', > > > > > wfc_extrapolation = > 'first_order', > > > > > > > > > > / > > > > > > > > > > ATOMIC_SPECIES > > > > > O 15.999 O.pw91-van_ak.UPF > > > > > Ti 47.867 Ti.pw91-nsp-van.UPF > > > > > ATOMIC_POSITIONS crystal > > > > > Ti 0.00000 0.00000 0.00000 > > > > > Ti 0.75000 0.25000 0.50000 > > > > > O 0.20800 0.20800 -0.00000 > > > > > O 0.95800 0.45800 0.50000 > > > > > O 0.54200 0.04200 0.50000 > > > > > O 0.79200 0.79200 -0.00000 > > > > > K_POINTS automatic > > > > > 6 6 7 0 0 0 > > > > > > > > > > * > > > > > > > > > > > > > > > On 8/19/06, Ezad Shojaee > > > > > > > > wrote: > > > > > > > > > > > > hi > > > > > > i am trying to calculate phonon > frequencies at > > > > > gamma point of > > > > > > TiO2(anatase) > > > > > > and i have relaxed the structure enough (i > > > think > > > > > )but performing PH.Xafter > > > > > > scf, 2of 18 frequencies are minus! > > > > > > i want to know that what technical problem > do > > > i > > > > > have? does anyone > > > > > > calculate > > > > > > these beore? > > > > > > any suggestion will be appreciated > > > > > > > > > > > > > > > > > > > > > > > > > --- lan haiping wrote: > > > > > Dear Eyvaz, > > > Thank you. > > > I checked the strcuture again using XCRYSDEN, > > > and found no problems related to the lattices. > > > > > > Regards, > > > > === message truncated === __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: questions...the answers they seek are clearly present in the current available resources. It just takes a little elbow grease to find them. It's much easier to simply post a question to the forum. I don't support this...researching one's own problem is far more educational than having someone answer it for you. We all learned this in the 8th grade. That said, I feel that these wiki ideas are great, as they would make it easier to find answers. I don't know if this is planned but the wikis should also contain the reference papers which disuss the thoery that goes into pwscf. As far as the developers are concerned, I am truly impressed with their patience and competence. It is intimidating enough for a graduate student and pwscf beginner to post to the forum, it would be much worse if those answering the questions were condescending. Instead, they treat every question more or less equally, and for that I applaud them. Thank you for all of your hard work and patience. Eric Abel Ph.D. Student, M.I.T. >From: "Axel Kohlmeyer" >Reply-To: pw_forum at pwscf.org >To: pw_forum at pwscf.org >Subject: Re: [Pw_forum] Answers and Questions >Date: Thu, 31 Aug 2006 10:33:14 -0400 > >On 8/31/06, Fernando A Reboredo wrote: >>Dear All, > >dear fernando, > >thanks for sharing your thoughts with us. > >>I feel compelled to write because I am always amazed by the degree of >>patience that the developers of these ESPRESO package show towards the >>questions in this mailing list. I also confess that I feel embarrassed to >>ask my own questions since I know that I will be stealing research time of >>colleagues. I thank the developers for making this package available. > >you may be 'stealing' some research time, but you also have to realize >that even trivial beginner's questions occasionally point out flaws in >either >the implementation or the documentation. furthermore, even if you may >not be the most qualified person to respond, you may still have encountered >a similar problem, so you would actually be giving back some of the >'stolen' >time by responding and reporting your own experiences. this is how >communities >do work: if everybody contributes, everybody benefits. > >>However, there are two types of questions 1) the ones that refer to the >>inner reason of the error messages (that I am not qualified to answer) 2) > >well, in my personal experience, trying to figure out what error messages >in a code mean (by looking at the sources), has helped me a lot to >understand >how program packages work and trying to figure out whether this is a >legitimate problem, a bug in the code or just a flaw in the input file, >is a good training to avoid problems for the next input you may need to >write and may make you more qualified. > >>the ones that refer to a chapter of "Introduction to Solid State Theory" >>or >>"Quantum Mechanics". Since I feel I am not better teacher that Aschoft or >>Kohen I also remain quiet there. > >see above. even when you may not be the better teacher right now, >responding to the best of your knowledge may actually _make_ you >a better teacher. a lot of it is just a matter of practice. you can always >start small and only provide answers to problems where you feel confident >and refer to the literature for the rest. > >on top of that, a good way to contribute (and that applies to everybody) >would be to collect frequently asked questions and their answers from the >mailing list archives and integrate them into the quantum-espresso wiki >pages. i've done something similar for a different project a couple of >years >back and it was _extremely_ helpful in getting a better understanding, >while at the same time the works was more that of an editor, i.e., you >didn't have to be an expert, but just take the available answers and edit >them into one more consistent text. this is somewhat time consuming, >but given the large number of people in this forum, it should not be >so much if this is shared amongst them. as i wrote before: everybody >contributes, everybody benefits. > >>I interrupt my silence to suggest to askers to think whether a question is >>type 1 or 2. > >part of the problem of a beginner in using tools like quantum espresso >is, that you frequently cannot tell, where this problem originates from. >many people answering here realize this fact (everybody has been through >that in some way at some point in time) and are willing to give people >some leeway at the beginning (there is no real gain from being rude >over e-mail regardless) and only get increasingly irritated when people >start taking advantage of that. > >ok. i guess this is enough 'preaching' for me for today. please everybody >give this some thought and (hopefully) help us to make the QE project >even better and even more fun than it is already right now. > >thanks for reading and ciao, > axel. >> >> >>Thanks again for the hard work and patience. >> >> >> >>I am Fernando A. Reboredo ORNL (and I approve this message) > > >-- >======================================================================= >Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania >Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 >tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 >======================================================================= >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 From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: which is only exhibited by a substance in the presence of an externally applied magnetic field. Similar definition is done in Kittel's textbook. Hopefully, now you can draw a conclusion. Bests, Eyvaz. --- Lijun Zhang wrote: > Dear all, > > I want to perform the phonon calculations of > ZnF2, which is diamagnetic. > Should I think of spin polarization? > > Best regards > > Lijun Zhang > National Lab of Superhard Materials > Jilin University > Changchun,130012 > P.R.China > Email: lijun_physics at yahoo.com.cn > > > --------------------------------- > Mp3??????-?????????????? __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: Gamma (0,0,0) N (0, 1/2, 0) X (0,0,1/2) P (1/4,1/4,1/4) Z (1/2,1/2,-1/2) I don't recall getting any errors or strange results when choosing paths bet= ween these points. You should search for this thesis as a good reference fo= r titania. Luke ----- Original Message -----=20 From: lan haiping=20 To: pw_forum at pwscf.org=20 Sent: Monday, October 23, 2006 12:38 AM Subject: Re: [Pw_forum] question about bands calculation Dear Luke, thanks for your reply. my structure is anatase titania. Actually, i set the k-point paths according to high symmetry lines such as= \Gamma[0,0,0] to Z [0,0,0.5] , Z to Q[0,0.5,0.5] etc. Is it right ? Regards, hai-ping =20 On 10/22/06, lukethulin at netscape.net wrote:=20 Hai-ping, I don't think you're going about a band structure calculation correctly.= The scf calculation should only use automatic k-point selection of (4,4,4)= or something like that. The following nscf calculation needs you to input=20= k-paths along high symmetry lines that vary by crystal structure. What form= of titania are you modeling, anatase, rutile?=20 Luke ----- Original Message -----=20 From: lan haiping=20 To: pw_forum=20 Sent: Saturday, October 21, 2006 2:31 PM Subject: [Pw_forum] question about bands calculation =20 Hi! i have a question about bands calculation . Just after finished a 'scf' calculation for tio2, i performed a serie= s (about 5) 'nscf' calculations with different k-grid paths. Except k-grid= path, all settings are the same for these series ' nscf ' calculation. =20 In some nscf-output, the values of "Highest occupied level and lowest= unoccuped level" are found. But these values are not found in the other=20= nscf-output . I donot know what the reasons are related to . Morever, i also= came to a calculation failed with complains "Too many eigenvalues are not c= onverged". =20 Regards, hai-ping ------=_NextPart_000_0037_01C6F67B.A6EF51B0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
hai-ping,
 
From a 2003 thesis by Giovanni Cangiani, bu= t=20 obviously there are other sources:
 
Gamma (0,0,0)
N (0, 1/2, 0)
X (0,0,1/2)
P (1/4,1/4,1/4)
Z (1/2,1/2,-1/2)
 
I don't recall getting any errors or strang= e=20 results when choosing paths between these points.  You should search fo= r=20 this thesis as a good reference for titania.
 
Luke
----- Original Message -----
lan=20 haiping
Sent: Monday, October 23, 2006 12:3= 8=20 AM
Subject: Re: [Pw_forum] question ab= out=20 bands calculation

Dear Luke,
thanks for your reply. my structure is anatase titania.
Actually, i set the k-point paths according to high symmetry lines su= ch=20 as \Gamma[0,0,0] to Z [0,0,0.5] , Z to Q[0,0.5,0.5]  etc.
Is it right ?
 
Regards,
hai-ping

 
On 10/22/06, lukethulin at netscape.net= <lukethulin at netscape.net<= /A>> wrote:=20
Hai-ping,
 
I don't think you're going about a band= =20 structure calculation correctly.  The scf calculation should only u= se=20 automatic k-point selection of (4,4,4) or something like that.  The= =20 following nscf calculation needs you to input k-paths along high symmetr= y=20 lines that vary by crystal structure.  What form of titania are you= =20 modeling, anatase, rutile?
 
Luke
----- Original Message -----
Sent: Saturday, October 21, 200= 6 2:31=20 PM
Subject: [Pw_forum] question ab= out=20 bands calculation

 
Hi!
 
  i have a question about bands calculation .
 Just after finished a 'scf' calculati= on=20 for tio2, i performed a series (about 5) =20 'nscf' calculations with  different k-grid=20 paths. Except k-grid path, all settings are the same for these series=20= '=20 nscf '  calculation.  
In some nscf-output,  the values of "Highest occupied level=20= and=20 lowest unoccuped level"  are  found. But  these values=20= are=20 not found in the other nscf-output . I donot know what the reasons are= =20 related to . Morever, i also came to a calculation failed with complai= ns=20 "Too many eigenvalues are not converged". 
 
Regards,
hai-ping
 
 
------=_NextPart_000_0037_01C6F67B.A6EF51B0-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: the selfconsistent threshold (conv_thr=1.d-6 in the &electrons namelist). This is usually fine for total energy which is variational, but already for structural relaxation it is often reduced so as to get more accurate forces. Phonon calculations are even more sensitive to the quality of self-consistency of your unperturbed system. Try to reduce conv_thr to 1.d-8 in the scf (and nscf) steps and see if this improves things. InN is a (wide band gap) semiconductor: there is no real need to use a metallic setup which, by the way, is incompatible with the calculation of dielectric constant and effective charges, that you will probably need to get correct phonons frequencies at gamma. I would expect that 2 or 10 Chadi-Cohen points should be enough for this system (automatic grids 2 2 2 1 1 1 and 4 4 4 1 1 1, respectively). hope this helps, stefano On Sun, 10 Dec 2006, e kb wrote: > Dear Stefano , QE users, > Thank you for your interest, > > Here are the scf, nscf and ph.x input files respectively, for X(0,0,1). > For X(0,1,1) the scf file is the same, in nscf input I just changed the phonon namelist to (0.00 1.00 1.00) so not to make the mail longer and unreadable, I am skipping it and just adding ph.X input for X(0,1,1). > > I should note that this was a part of test calculations for the new pseudo for Indium that I recently generated using the 'atomic' code. The cut off energy that I used for these calculations (34 Ryd) is chosen to be enough to represent correct lattice constants, but still the pseudo and the value of cut off energy needs further examination. That is why I would not be surprised to see any discrepiencies in the frequencies. But again, I could not understand the differences that occured for the same point. > With a verbose output I checked the phonon polarization vectors, > As far as I understand, the calculation is done for the same modes for the two points. > > I do not want to bother you with the whole output files, > if this is a very simple problem that an experienced/careful user can notice from the inputs, > otherwise I can attach the output files. > > Buona domenica, > Emine Kucukbenli > ------------ > &control > calculation = 'scf' > restart_mode='from_scratch', > prefix='34_vib', > tprnfor = .true. > pseudo_dir = '$PSEUDO_DIR' , > etot_conv_thr = 1.0D-5 > / > &system > ibrav= 2, > celldm(1)=9.593140000 > nat= 2, > ntyp= 2, > ecutwfc =34.0, > ecutrho =272.0, > occupations = 'smearing' > smearing = 'gaussian' > degauss = 0.004, > / > &electrons > mixing_mode = 'plain' > mixing_beta = 0.7 > / > ATOMIC_SPECIES > In 114.818 In_us_2.upf > N 14.0067 N.pbe-van_ak.UPF > ATOMIC_POSITIONS { crystal } > In 0.00000000000 0.00000000000 0.0000000000000 > N 0.25000000000 0.25000000000 0.2500000000000 > K_POINTS { automatic } > 8 8 8 0 0 0 > ------------ > &control > calculation = 'phonon' > restart_mode='from_scratch', > prefix='34_vib', > tprnfor = .true. > pseudo_dir = '$PSEUDO_DIR' , > etot_conv_thr = 1.0D-5 > / > &system > ibrav= 2, > celldm(1)=9.593140000 > nat= 2, > ntyp= 2, > ecutwfc =34.0, > ecutrho =272.0, > occupations = 'smearing' > smearing = 'gaussian' > degauss = 0.004, > / > &electrons > mixing_mode = 'plain' > mixing_beta = 0.7 > / > &phonon > xqq(1) = .00, xqq(2) = .00, xqq(3) = 1.00 > / > ATOMIC_SPECIES > In 114.818 In_us_2.upf > N 14.0067 N.pbe-van_ak.UPF > ATOMIC_POSITIONS { crystal } > In 0.00000000000 0.00000000000 0.0000000000000 > N 0.25000000000 0.25000000000 0.2500000000000 > K_POINTS { automatic } > 8 8 8 0 0 0 > ------------------------- > > &inputph > tr2_ph=1.0d-14, > amass(1)=114.818, > amass(2)=14.0067, > prefix='34_vib', > fildyn='34_vib.dyn', > / > 0.00 0.00 1.00 > -------------------------------- > > > &inputph > tr2_ph=1.0d-14, > amass(1)=114.818, > amass(2)=14.0067, > prefix='34_vib', > fildyn='34_vib.dyn', > / > 0.00 1.00 1.00 > ---------------------------------- > X(0,0,1) > omega( 1) = 3.163928 [THz] = 105.537987 [cm-1] > omega( 2) = 3.163928 [THz] = 105.537987 [cm-1] > omega( 3) = 6.030058 [THz] = 201.142433 [cm-1] > omega( 4) = 14.775784 [THz] = 492.870383 [cm-1] > omega( 5) = 14.775784 [THz] = 492.870383 [cm-1] > omega( 6) = 16.372030 [THz] = 546.115769 [cm-1] > X(0,1,1) > omega( 1) = 2.153190 [THz] = 71.823161 [cm-1] > omega( 2) = 2.153190 [THz] = 71.823161 [cm-1] > omega( 3) = 6.239891 [THz] = 208.141736 [cm-1] > omega( 4) = 14.879198 [THz] = 496.319913 [cm-1] > omega( 5) = 14.879198 [THz] = 496.319913 [cm-1] > omega( 6) = 16.354833 [THz] = 545.542147 [cm-1] > ---------------------------------- > > Stefano Baroni wrote: > Dear Emin, dear all: all my apologies! > > Nicola was of course right, and I was WRONG! (011) IS indeed equivalent to (001), because the latter is equivalent to (-1,0,0) and the difference between (011) and (-100) is (111) which IS an FCC reciprocal-lattice vector. The coordinates of the K point are (3/4,3/4,0). With reference to the figure mentioned in the link below, K is the intersection among the planes x+y+z=3/2, z=0, and x=y, hence the above coordinates. > > To come to your original question. Are the symmetries/degeneracies of the results what you expect? Can you make the input/output of your calculation accessible to us? > > > Stefano > > On Dec 10, 2006, at 2:46 PM, e kb wrote: > > Thank you for your answer Stefano, > Of course this explains the different frequencies, > and that also means that I misunderstand the kpoints given in example 5, > about band structure for an fcc, > and Dr. Marzari's explanation to another QE user, > who was confused like me. > http://www.democritos.it/pipermail/pw_forum/2006-June/004365.html > > Thank you again, > Emine Kucukbenli > > Stefano Baroni wrote: > Emine: (001) and (011) are *NOT* two "X" points in the FCC Brillouin zone. The difference between the second and the first (010) is *NOT* a reciprocal-space vector. If it was, X would also be, which you will agree it is not. As a matter of fact, (011) has its own nickname (it is called "K"). No surprise, therefore, that frequencies @K are different from frequencies @X. Are these arguments clear and convincing enough? Stefano > > On Dec 9, 2006, at 11:10 PM, e kb wrote: > > Dear QE users, > I am performing phonon calculations for a zinc blende structure, > I calculated the frequencies at high symmetry points, evaluating a single point at each time. > So I followed the procedure of scf, nscf and ph.x calculation one by one for L(0.5 0.5 0.5) - Gamma - X(0,0,1) - X(0,1,1) points. > In these calculations every parameter (else than q-vector in &phonon list in nscf calculation and phonon card for ph.x input of course) is kept fixed. > > Although I expect the frequencies at two X points - (0,0,1) and (0,1,1) - to be the same, > the difference in TA modes' frequencies are around 30 cm-1, > in the other ones, around 5 cm-1. > > I would like to attribute this to unconverged lattice parameters, cut off energies or thresholds etc. but what I do not understand is why the program treats these two points differently, although they should be equivalent. So I expect to see the same value for frequency at these points, even if it is wrong, highly negative, or sth else that it should not be. > > I will appreciate any answers. > Thank you. > > Emine Kucukbenli > > > > > > --------------------------------- > Everyone is raving about the all-new Yahoo! Mail beta. > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > > > --------------------------------- > Check out the all-new Yahoo! Mail beta - Fire up a more powerful email and get things done faster. > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > > --------------------------------- > Have a burning question? Go to Yahoo! Answers and get answers from real people who know. From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: the selfconsistent threshold (conv_thr=1.d-6 in the &electrons namelist). This is usually fine for total energy which is variational, but already for structural relaxation it is often reduced so as to get more accurate forces. Phonon calculations are even more sensitive to the quality of self-consistency of your unperturbed system. Try to reduce conv_thr to 1.d-8 in the scf (and nscf) steps and see if this improves things. InN is a (wide band gap) semiconductor: there is no real need to use a metallic setup which, by the way, is incompatible with the calculation of dielectric constant and effective charges, that you will probably need to get correct phonons frequencies at gamma. I would expect that 2 or 10 Chadi-Cohen points should be enough for this system (automatic grids 2 2 2 1 1 1 and 4 4 4 1 1 1, respectively). hope this helps, stefano On Sun, 10 Dec 2006, e kb wrote: > Dear Stefano , QE users, > Thank you for your interest, > > Here are the scf, nscf and ph.x input files respectively, for X(0,0,1). > For X(0,1,1) the scf file is the same, in nscf input I just changed the phonon namelist to (0.00 1.00 1.00) so not to make the mail longer and unreadable, I am skipping it and just adding ph.X input for X(0,1,1). > > I should note that this was a part of test calculations for the new pseudo for Indium that I recently generated using the 'atomic' code. The cut off energy that I used for these calculations (34 Ryd) is chosen to be enough to represent correct lattice constants, but still the pseudo and the value of cut off energy needs further examination. That is why I would not be surprised to see any discrepiencies in the frequencies. But again, I could not understand the differences that occured for the same point. > With a verbose output I checked the phonon polarization vectors, > As far as I understand, the calculation is done for the same modes for the two points. > > I do not want to bother you with the whole output files, > if this is a very simple problem that an experienced/careful user can notice from the inputs, > otherwise I can attach the output files. > > Buona domenica, > Emine Kucukbenli > ------------ > &control > calculation = 'scf' > restart_mode='from_scratch', > prefix='34_vib', > tprnfor = .true. > pseudo_dir = '$PSEUDO_DIR' , > etot_conv_thr = 1.0D-5 > / > &system > ibrav= 2, > celldm(1)=9.593140000 > nat= 2, > ntyp= 2, > ecutwfc =34.0, > ecutrho =272.0, > occupations = 'smearing' > smearing = 'gaussian' > degauss = 0.004, > / > &electrons > mixing_mode = 'plain' > mixing_beta = 0.7 > / > ATOMIC_SPECIES > In 114.818 In_us_2.upf > N 14.0067 N.pbe-van_ak.UPF > ATOMIC_POSITIONS { crystal } > In 0.00000000000 0.00000000000 0.0000000000000 > N 0.25000000000 0.25000000000 0.2500000000000 > K_POINTS { automatic } > 8 8 8 0 0 0 > ------------ > &control > calculation = 'phonon' > restart_mode='from_scratch', > prefix='34_vib', > tprnfor = .true. > pseudo_dir = '$PSEUDO_DIR' , > etot_conv_thr = 1.0D-5 > / > &system > ibrav= 2, > celldm(1)=9.593140000 > nat= 2, > ntyp= 2, > ecutwfc =34.0, > ecutrho =272.0, > occupations = 'smearing' > smearing = 'gaussian' > degauss = 0.004, > / > &electrons > mixing_mode = 'plain' > mixing_beta = 0.7 > / > &phonon > xqq(1) = .00, xqq(2) = .00, xqq(3) = 1.00 > / > ATOMIC_SPECIES > In 114.818 In_us_2.upf > N 14.0067 N.pbe-van_ak.UPF > ATOMIC_POSITIONS { crystal } > In 0.00000000000 0.00000000000 0.0000000000000 > N 0.25000000000 0.25000000000 0.2500000000000 > K_POINTS { automatic } > 8 8 8 0 0 0 > ------------------------- > > &inputph > tr2_ph=1.0d-14, > amass(1)=114.818, > amass(2)=14.0067, > prefix='34_vib', > fildyn='34_vib.dyn', > / > 0.00 0.00 1.00 > -------------------------------- > > > &inputph > tr2_ph=1.0d-14, > amass(1)=114.818, > amass(2)=14.0067, > prefix='34_vib', > fildyn='34_vib.dyn', > / > 0.00 1.00 1.00 > ---------------------------------- > X(0,0,1) > omega( 1) = 3.163928 [THz] = 105.537987 [cm-1] > omega( 2) = 3.163928 [THz] = 105.537987 [cm-1] > omega( 3) = 6.030058 [THz] = 201.142433 [cm-1] > omega( 4) = 14.775784 [THz] = 492.870383 [cm-1] > omega( 5) = 14.775784 [THz] = 492.870383 [cm-1] > omega( 6) = 16.372030 [THz] = 546.115769 [cm-1] > X(0,1,1) > omega( 1) = 2.153190 [THz] = 71.823161 [cm-1] > omega( 2) = 2.153190 [THz] = 71.823161 [cm-1] > omega( 3) = 6.239891 [THz] = 208.141736 [cm-1] > omega( 4) = 14.879198 [THz] = 496.319913 [cm-1] > omega( 5) = 14.879198 [THz] = 496.319913 [cm-1] > omega( 6) = 16.354833 [THz] = 545.542147 [cm-1] > ---------------------------------- > > Stefano Baroni wrote: > Dear Emin, dear all: all my apologies! > > Nicola was of course right, and I was WRONG! (011) IS indeed equivalent to (001), because the latter is equivalent to (-1,0,0) and the difference between (011) and (-100) is (111) which IS an FCC reciprocal-lattice vector. The coordinates of the K point are (3/4,3/4,0). With reference to the figure mentioned in the link below, K is the intersection among the planes x+y+z=3/2, z=0, and x=y, hence the above coordinates. > > To come to your original question. Are the symmetries/degeneracies of the results what you expect? Can you make the input/output of your calculation accessible to us? > > > Stefano > > On Dec 10, 2006, at 2:46 PM, e kb wrote: > > Thank you for your answer Stefano, > Of course this explains the different frequencies, > and that also means that I misunderstand the kpoints given in example 5, > about band structure for an fcc, > and Dr. Marzari's explanation to another QE user, > who was confused like me. > http://www.democritos.it/pipermail/pw_forum/2006-June/004365.html > > Thank you again, > Emine Kucukbenli > > Stefano Baroni wrote: > Emine: (001) and (011) are *NOT* two "X" points in the FCC Brillouin zone. The difference between the second and the first (010) is *NOT* a reciprocal-space vector. If it was, X would also be, which you will agree it is not. As a matter of fact, (011) has its own nickname (it is called "K"). No surprise, therefore, that frequencies @K are different from frequencies @X. Are these arguments clear and convincing enough? Stefano > > On Dec 9, 2006, at 11:10 PM, e kb wrote: > > Dear QE users, > I am performing phonon calculations for a zinc blende structure, > I calculated the frequencies at high symmetry points, evaluating a single point at each time. > So I followed the procedure of scf, nscf and ph.x calculation one by one for L(0.5 0.5 0.5) - Gamma - X(0,0,1) - X(0,1,1) points. > In these calculations every parameter (else than q-vector in &phonon list in nscf calculation and phonon card for ph.x input of course) is kept fixed. > > Although I expect the frequencies at two X points - (0,0,1) and (0,1,1) - to be the same, > the difference in TA modes' frequencies are around 30 cm-1, > in the other ones, around 5 cm-1. > > I would like to attribute this to unconverged lattice parameters, cut off energies or thresholds etc. but what I do not understand is why the program treats these two points differently, although they should be equivalent. So I expect to see the same value for frequency at these points, even if it is wrong, highly negative, or sth else that it should not be. > > I will appreciate any answers. > Thank you. > > Emine Kucukbenli > > > > > > --------------------------------- > Everyone is raving about the all-new Yahoo! Mail beta. > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > > > --------------------------------- > Check out the all-new Yahoo! Mail beta - Fire up a more powerful email and get things done faster. > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > > > > > --------------------------------- > Have a burning question? Go to Yahoo! Answers and get answers from real people who know. _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum --------------------------------- Cheap Talk? Check out Yahoo! Messenger's low PC-to-Phone call rates. --0-1202871954-1165791937=:33393 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit
 
Dear Stefano,
Thank you, I will study the relevant part for pert. potential., do the suggested tests and modifications,
and send the result back to close the case.
But a fast point that I wanted to note that is,
InN turns out to be a metal according to my calculations,
without several corrections used, like GW or SIC.
That was the reason of the k-mesh.
Thank you for your help.

Emine Kucukbenli

Stefano de Gironcoli <degironc at sissa.it> wrote:

If I remember correctly there used to be a term in the phonon
calculation in which the perturbing potential was written as DeltaV(r) =
exp(+iqr)*\DeltaVperiodic(r) and \DeltaVperiodic was expanded up to
the charge density cutoff ... this term was therefore slightly different
for two q wavevectors that would otherwise be equivalent. This never
gave any noticeable difference in real calculations and I would be
surprised if this would be the cause of your results but a way to check
this is to increase the charge density cutoff and see what happens as
in this case the problem should disappear when the potential description
become complete.

From your input it looks to me that you are using the default value for
the selfconsistent threshold (conv_thr=1.d-6 in the &electrons namelist).
This is usually fine for total energy which is variational, but already
for structural relaxation it is often reduced so as to get more accurate
forces.

Phonon calculations are even more sensitive to the quality of
self-consistency of your unperturbed system. Try to reduce conv_thr to
1.d-8 in the scf (and nscf) steps and see if this improves things.

InN is a (wide band gap) semiconductor: there is no real need to use a
metallic setup which, by the way, is incompatible with the calculation
of dielectric constant and effective charges, that you will probably need
to get correct phonons frequencies at gamma.
I would expect that 2 or 10 Chadi-Cohen points should be enough for this
system (automatic grids 2 2 2 1 1 1 and 4 4 4 1 1 1, respectively).

hope this helps,

stefano

On Sun, 10 Dec 2006, e kb wrote:

> Dear Stefano , QE users,
> Thank you for your interest,
>
> Here are the scf, nscf and ph.x input files respectively, for X(0,0,1).
> For X(0,1,1) the scf file is the same, in nscf input I just changed the phonon namelist to (0.00 1.00 1.00) so not to make the mail longer and unreadable, I am skipping it and just adding ph.X input for X(0,1,1).
>
> I should note that this was a part of test calculations for the new pseudo for Indium that I recently generated using the 'atomic' code. The cut off energy that I used for these calculations (34 Ryd) is chosen to be enough to represent correct lattice constants, but still the pseudo and the value of cut off energy needs further examination. That is why I would not be surprised to see any discrepiencies in the frequencies. But again, I could not understand the differences that occured for the same point.
> With a verbose output I checked the phonon polarization vectors,
> As far as I understand, the calculation is done for the same modes for the two points.
>
> I do not want to bother you with the whole output files,
> if this is a very simple problem that an experienced/careful user can notice from the inputs,
> otherwise I can attach the output files.
>
> Buona domenica,
> Emine Kucukbenli
> ------------
> &control
> calculation = 'scf'
> restart_mode='from_scratch',
> prefix='34_vib',
> tprnfor = .true.
> pseudo_dir = '$PSEUDO_DIR' ,
> etot_conv_thr = 1.0D-5
> /
> &system
> ibrav= 2,
> celldm(1)=9.593140000
> nat= 2,
> ntyp= 2,
> ecutwfc =34.0,
> ecutrho =272.0,
> occupations = 'smearing'
> smearing = 'gaussian'
> degauss = 0.004,
> /
> &electrons
> mixing_mode = 'plain'
> mixing_beta = 0.7
> /
> ATOMIC_SPECIES
> In 114.818 In_us_2.upf
> N 14.0067 N.pbe-van_ak.UPF
> ATOMIC_POSITIONS { crystal }
> In 0.00000000000 0.00000000000 0.0000000000000
> N 0.25000000000 0.25000000000 0.2500000000000
> K_POINTS { automatic }
> 8 8 8 0 0 0
> ------------
> &control
> calculation = 'phonon'
> restart_mode='from_scratch',
> prefix='34_vib',
> tprnfor = .true.
> pseudo_dir = '$PSEUDO_DIR' ,
> etot_conv_thr = 1.0D-5
> /
> &system
> ibrav= 2,
> celldm(1)=9.593140000
> nat= 2,
> ntyp= 2,
> ecutwfc =34.0,
> ecutrho =272.0,
> occupations = 'smearing'
> smearing = 'gaussian'
> degauss = 0.004,
> /
> &electrons
> mixing_mode = 'plain'
> mixing_beta = 0.7
> /
> &phonon
> xqq(1) = .00, xqq(2) = .00, xqq(3) = 1.00
> /
> ATOMIC_SPECIES
> In 114.818 In_us_2.upf
> N 14.0067 N.pbe-van_ak.UPF
> ATOMIC_POSITIONS { crystal }
> In 0.00000000000 0.00000000000 0.0000000000000
> N 0.25000000000 0.25000000000 0.2500000000000
> K_POINTS { automatic }
> 8 8 8 0 0 0
> -------------------------
>
> &inputph
> tr2_ph=1.0d-14,
> amass(1)=114.818,
> amass(2)=14.0067,
> prefix='34_vib',
> fildyn='34_vib.dyn',
> /
> 0.00 0.00 1.00
> --------------------------------
>
>
> &inputph
> tr2_ph=1.0d-14,
> amass(1)=114.818,
> amass(2)=14.0067,
> prefix='34_vib',
> fildyn='34_vib.dyn',
> /
> 0.00 1.00 1.00
> ----------------------------------
> X(0,0,1)
> omega( 1) = 3.163928 [THz] = 105.537987 [cm-1]
> omega( 2) = 3.163928 [THz] = 105.537987 [cm-1]
> omega( 3) = 6.030058 [THz] = 201.142433 [cm-1]
> omega( 4) = 14.775784 [THz] = 492.870383 [cm-1]
> omega( 5) = 14.775784 [THz] = 492.870383 [cm-1]
> omega( 6) = 16.372030 [THz] = 546.115769 [cm-1]
> X(0,1,1)
> omega( 1) = 2.153190 [THz] = 71.823161 [cm-1]
> omega( 2) = 2.153190 [THz] = 71.823161 [cm-1]
> omega( 3) = 6.239891 [THz] = 208.141736 [cm-1]
> omega( 4) = 14.879198 [THz] = 496.319913 [cm-1]
> omega( 5) = 14.879198 [THz] = 496.319913 [cm-1]
> omega( 6) = 16.354833 [THz] = 545.542147 [cm-1]
> ----------------------------------
>
> Stefano Baroni wrote:
> Dear Emin, dear all: all my apologies!
>
> Nicola was of course right, and I was WRONG! (011) IS indeed equivalent to (001), because the latter is equivalent to (-1,0,0) and the difference between (011) and (-100) is (111) which IS an FCC reciprocal-lattice vector. The coordinates of the K point are (3/4,3/4,0). With reference to the figure mentioned in the link below, K is the intersection among the planes x+y+z=3/2, z=0, and x=y, hence the above coordinates.
>
> To come to your original question. Are the symmetries/degeneracies of the results what you expect? Can you make the input/output of your calculation accessible to us?
>
>
> Stefano
>
> On Dec 10, 2006, at 2:46 PM, e kb wrote:
>
> Thank you for your answer Stefano,
> Of course this explains the different frequencies,
> and that also means that I misunderstand the kpoints given in example 5,
> about band structure for an fcc,
> and Dr. Marzari's explanation to another QE user,
> who was confused like me.
> http://www.democritos.it/pipermail/pw_forum/2006-June/004365.html
>
> Thank you again,
> Emine Kucukbenli
>
> Stefano Baroni wrote:
> Emine: (001) and (011) are *NOT* two "X" points in the FCC Brillouin zone. The difference between the second and the first (010) is *NOT* a reciprocal-space vector. If it was, X would also be, which you will agree it is not. As a matter of fact, (011) has its own nickname (it is called "K"). No surprise, therefore, that frequencies @K are different from frequencies @X. Are these arguments clear and convincing enough? Stefano
>
> On Dec 9, 2006, at 11:10 PM, e kb wrote:
>
> Dear QE users,
> I am performing phonon calculations for a zinc blende structure,
> I calculated the frequencies at high symmetry points, evaluating a single point at each time.
> So I followed the procedure of scf, nscf and ph.x calculation one by one for L(0.5 0.5 0.5) - Gamma - X(0,0,1) - X(0,1,1) points.
> In these calculations every parameter (else than q-vector in &phonon list in nscf calculation and phonon card for ph.x input of course) is kept fixed.
>
> Although I expect the frequencies at two X points - (0,0,1) and (0,1,1) - to be the same,
> the difference in TA modes' frequencies are around 30 cm-1,
> in the other ones, around 5 cm-1.
>
> I would like to attribute this to unconverged lattice parameters, cut off energies or thresholds etc. but what I do not understand is why the program treats these two points differently, although they should be equivalent. So I expect to see the same value for frequency at these points, even if it is wrong, highly negative, or sth else that it should not be.
>
> I will appreciate any answers.
> Thank you.
>
> Emine Kucukbenli
>
>
>
>
>
> ---------------------------------
> Everyone is raving about the all-new Yahoo! Mail beta.
>
> ---
> Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
>
> Please, if possible, don't send me MS Word or PowerPoint attachments
> Why? See: http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>
>
>
>
>
>
>
>
>
> ---------------------------------
> Check out the all-new Yahoo! Mail beta - Fire up a more powerful email and get things done faster.
>
> ---
> Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste
> [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)
>
>
> Please, if possible, don't send me MS Word or PowerPoint attachments
> Why? See: http://www.gnu.org/philosophy/no-word-attachments.html
>
>
>
>
>
>
>
>
>
>
> ---------------------------------
> Have a burning question? Go to Yahoo! Answers and get answers from real people who know.
_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://www.democritos.it/mailman/listinfo/pw_forum

Cheap Talk? Check out Yahoo! Messenger's low PC-to-Phone call rates. --0-1202871954-1165791937=:33393-- From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: fixed_magnetization(3) REAL (default 0.d0) value of the total magnetization to be maintained fixed when constrained_magnetization='total' Hopefully this is the one you need. Bests, Eyvaz. Regards Waheed _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum ____________________________________________________________________________________ 8:00? 8:25? 8:40? Find a flick in no time with the Yahoo! Search movie showtime shortcut. http://tools.search.yahoo.com/shortcuts/#news From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: save some data from the self-consistent calculations. Is this data-saving flag equivalent to the previous way of computing the el-ph coupling with a second very dense non self-consistent mesh? In the case of not being equivalent, how small are the differences? On another ground, are displaced Monkhorst-Pack grids compatible with electron phonon calculations? I'd say no, since the displacement is grid dependent, and this wouldn't assure that k+q falls in a point in fine k. Any comments? Thanks in advance, Miguel ---------------------------------------- Miguel Mart=EDnez Canales Dto. F=EDsica de la Materia Condensada UPV/EHU Facultad de Ciencia y Tecnolog=EDa Apdo. 644 48080 Bilbao (Spain) Fax: +34 94 601 3500 Tlf: +34 94 601 5437 ---------------------------------------- "If you're a mad scientist, you can use GPLv2'd software for your evil plans to take over the world" Linus Torvalds From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: be included before -llapack for this application. Which is a little unusual. # For IBM machines with essl: load essl BEFORE lapack ! # remember that LAPACK_LIBS precedes BLAS_LIBS in loading order They have a macro called __ESSL below which probably alters the calling convention for the calls in ESSL that are different than LAPACK, so it would probably be bad to put -llapack first in this case. Also, to help with debugging, are you willing to send us a link to where you obtained your source code or the location of the tar file on iceberg? Thank you, ************************************************************************** Any comments would be helpful. Thank you, Vivek From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: be included before -llapack for this application. Which is a little unusual. # For IBM machines with essl: load essl BEFORE lapack ! # remember that LAPACK_LIBS precedes BLAS_LIBS in loading order They have a macro called __ESSL below which probably alters the calling convention for the calls in ESSL that are different than LAPACK, so it would probably be bad to put -llapack first in this case. Also, to help with debugging, are you willing to send us a link to where you obtained your source code or the location of the tar file on iceberg? Thank you, ************************************************************************** Any comments would be helpful. Thank you, Vivek From bogus@does.not.exist.com Tue Jul 10 14:20:16 2007 From: bogus@does.not.exist.com () Date: Tue, 10 Jul 2007 12:20:16 -0000 Subject: No subject Message-ID: (swap space is on). Has anyone had a similar problem? Cheers, Vivian ------=_Part_61425_27160904.1182224617373 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi,

I can successfully run all the examples provided with espresso using mpirun, however, when I attempt to run my own calculations the program crashes with a memory allocation failure.  This occurs when real memory is exhausted.  From examining "top" output, I can see that no virtual memory gets allocated (swap space is on).

Has anyone had a similar problem?

Cheers,

Vivian
------=_Part_61425_27160904.1182224617373-- From liyanpcl at yahoo.com.cn Sun Jul 1 03:36:29 2007 From: liyanpcl at yahoo.com.cn (li yan) Date: Sun, 1 Jul 2007 09:36:29 +0800 (CST) Subject: [Pw_forum] =?gb2312?q?=BB=D8=B8=B4=A3=BA=20Re:=20[Pw=5Fforum]=20Re:about=20the=20dis?= =?gb2312?q?placive=20instabilities?= In-Reply-To: <38226797-F160-4FCE-A07B-5F1ED5B433C3@sissa.it> Message-ID: <76881.29908.qm@web15615.mail.cnb.yahoo.com> Dear Stefano, I'm sorry. I did not explain my question clearly. I found a softening transverse acaustic mode at X (TA(X)) point on the brillouin zone boundary in a NaCl structure. Since the first high-pressure phase that the experiment observed is CsCl structure, could the soft mode be considered as the driven force of the transition from NaCl to CsCl? If it could not, may i give an intermediate state between the NaCl and CsCl phase if I can get a energy favored structure in terns of the softening mode in the NaCl structure? It is a monoclinic phase, obtained by moving the atoms along the displacements (order parameters) corresponding to the softening mode. Thank you very much for any advice. Best regards. Stefano Baroni ?????? On Jun 15, 2007, at 3:20 AM, li yan wrote: Dear Stefano ?????I'm so sorry and giving my delayed thanks to you, ?????I am not sure if i can continue the discussion of my previous question. I have?read the papers?you suggested. And the authors proposed that the softening mode coupled with the lattice strain can induced a second or a slightly first order pahse transition. But I stil have doubts in the displacive instabilitis. ???The character of the ?transition from NaCl to CsCl is thougt to be?first order in experiment. May I?cosider this??transition?to be induced by the?lattice instability if I?found?the softening mode in the ?NaCl phase?at the?pressure?which is consitent with the experiment??lt;/DIV> which normal-mode displacement would bring the NaCl structure to CsCl? what would be the order parameter of the transition? S. --- Stefano Baroni - SISSA?&?DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) Please, if possible, don't?send me MS Word or PowerPoint attachments Why? See:?http://www.gnu.org/philosophy/no-word-attachments.html --------------------------------- ????????????3.5G??????20M?????? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070701/8580b96f/attachment.htm From amos at errno.com Mon Jul 2 04:54:54 2007 From: amos at errno.com (Amos Leffler) Date: Sun, 01 Jul 2007 20:54:54 -0600 Subject: [Pw_forum] Example21 Problem In-Reply-To: References: Message-ID: <468868FE.9010500@errno.com> Dear Forum, I am unable to compile the program using the commands, "-ftrace=frame" or "-ftrace=full" or I dont know how to formulate the command properly. The response is "no such command". I have approximately 28 Gb empty space on my disk and 256 Mb of RAM. I always assumed that if more than 256 Mb is needed there will be paging to the swap space. This would slow the program but not stop it. I checked "ulimit" and it replied unlimited. At this point I have no idea of what is needed. Hopefully someone can help. Amos Leffler Axel Kohlmeyer wrote: > On Fri, 29 Jun 2007, Amos Leffler wrote: > > > amos, > > how much memory do you have on the machine that you are running on? > how large are your ulimit settings? > > the 32 water example requires about 1.2GB address space and > about 400MB physical memory to run. for the larger systems > accordingly more. > > when a job stops complaining about not being able to allocate > memory, it is likely that you either do not have enough memory > or address space available, or that your account settings > prohibit that. > > cheers, > axel. > > AL> Dear Forum, > AL> Attached is the result of the following input in example 21: > AL> ./run_example 256 > AL> It calculates values for all sizes up to 256 and skips the remainder. > AL> However, I am puzzled that the calculations take very little time > AL> certainly less than a minute for each one. The "README" note says it > AL> takes a long time and should include more than what is shown. There may > AL> be more to this than I understand. > AL> > AL> Amos Leffler > AL> Axel Kohlmeyer wrote: > AL> > On 6/27/07, Amos Leffler wrote: > AL> >> Dear Forum, > AL> >> I found later after sending this note that the program ran (in > AL> >> background) but did the calculation for all of the numbers of water > AL> >> molecules. The original input did have as the last entry a number (64) > AL> >> so there still is a problem. > AL> > > AL> > i disagree. please explain first the existence of the -ftrace=full > AL> > which is obviously causing the problems you see. omit it and > AL> > the script will run fine. > AL> > > AL> > axel. > AL> > > AL> >> > AL> >> Amos Leffler > AL> >> > AL> >> Axel Kohlmeyer wrote: > AL> >> > On Tue, 26 Jun 2007, Amos Leffler wrote: > AL> >> > > AL> >> > AL> Dear Forum, > AL> >> > AL> Attached are the details of a problem I ran into > AL> >> trying to > AL> >> > AL> execute example21. I am using SuSE Linux 10.2 which has > AL> >> worked well > AL> >> > AL> up to this point so I think the problem is in the run_example > AL> >> file. > AL> >> > > AL> >> > well, i would say the problem is in your using of -ftrace=full. why? > AL> >> > example21 is special and requires as an argument the maximum number > AL> >> > of water molecules that you want to run (32-512). if you provide > AL> >> > > AL> >> > the error message stems from the numerical compares of the first > AL> >> > argument of the script, $1, with the corresponding number of waters. > AL> >> > the way it is done, allows to have a default value of '0' (i.e. > AL> >> > don't run any of those tests). your -ftrace=full argument expands > AL> >> > to a non-numerical value, hence the error messages. in all other > AL> >> > scripts it will be ignored, anyways. > AL> >> > > AL> >> > cheers, > AL> >> > axel. > AL> >> > > AL> >> > > AL> >> > AL> > AL> >> > AL> Amos Leffler > AL> >> > AL> > AL> >> > AL> Walnut Creek, CA. > AL> >> > AL> > AL> >> > AL> unaffiliated > AL> >> > AL> > AL> >> > AL> > AL> >> > > AL> >> > > AL> >> > AL> >> _______________________________________________ > AL> >> Pw_forum mailing list > AL> >> Pw_forum at pwscf.org > AL> >> http://www.democritos.it/mailman/listinfo/pw_forum > AL> >> > AL> >> > AL> > > AL> > > AL> > AL> > > From akohlmey at cmm.chem.upenn.edu Mon Jul 2 05:37:49 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 1 Jul 2007 23:37:49 -0400 (EDT) Subject: [Pw_forum] Example21 Problem In-Reply-To: <468868FE.9010500@errno.com> Message-ID: On Sun, 1 Jul 2007, Amos Leffler wrote: AL> Dear Forum, AL> I am unable to compile the program using the commands, AL> "-ftrace=frame" or "-ftrace=full" or I dont know how to formulate the AL> command properly. The response is "no such command". amos, these are not commands but flags to the g95 compiler. these kind of problems are not specific to QE. to get over this, i recommend you get some sort of linux/unix training so that these rather elementary issue don't get in the way in the future (there are plenty of books, online resources, workshops, linux user groups, knowledgable colleages/friends etc.). AL> I have approximately 28 Gb empty space on my disk and 256 Mb of AL> RAM. I always assumed that if more than 256 Mb is needed there will be AL> paging to the swap space. This would slow the program but not stop it. you need to have about 1.5GB of swap space configured. this has to be done explicitely. use the 'free' command to verify it. here is the output from my laptop (512MB + 1GB swap): [akohlmey at zero akohlmey]$ free -t total used free shared buffers cached Mem: 510888 494480 16408 0 15996 302508 -/+ buffers/cache: 175976 334912 Swap: 1020116 61892 958224 Total: 1531004 556372 974632 that being said, any computational code will perform horribly, if you have to go into swap. unlike 20 or so years ago, disk space is now relatively much much slower compared to RAM, so using swap is not much of an option. with 256MB RAM you'll have at best about 200MB left to run cp.x in (depending on what else you have running on your machine and whether you are running X or not). that will bring your machine to a grinding halt already with the 32 water example and you will run out of address space with the 64 water (i assume you are running with a 32bit OS). AL> I checked "ulimit" and it replied unlimited. AL> At this point I have no idea of what is needed. Hopefully AL> someone can help. fist of all, you need more physical memory or else you'll kill you machine (it is are desktop and not a notebook i hope?). better yet you run on a parallel machine (cluster). cheers, axel. AL> AL> Amos Leffler AL> AL> Axel Kohlmeyer wrote: AL> > On Fri, 29 Jun 2007, Amos Leffler wrote: AL> > AL> > AL> > amos, AL> > AL> > how much memory do you have on the machine that you are running on? AL> > how large are your ulimit settings? AL> > AL> > the 32 water example requires about 1.2GB address space and AL> > about 400MB physical memory to run. for the larger systems AL> > accordingly more. AL> > AL> > when a job stops complaining about not being able to allocate AL> > memory, it is likely that you either do not have enough memory AL> > or address space available, or that your account settings AL> > prohibit that. AL> > AL> > cheers, AL> > axel. AL> > AL> > AL> Dear Forum, AL> > AL> Attached is the result of the following input in example 21: AL> > AL> ./run_example 256 AL> > AL> It calculates values for all sizes up to 256 and skips the remainder. AL> > AL> However, I am puzzled that the calculations take very little time AL> > AL> certainly less than a minute for each one. The "README" note says it AL> > AL> takes a long time and should include more than what is shown. There may AL> > AL> be more to this than I understand. AL> > AL> AL> > AL> Amos Leffler AL> > AL> Axel Kohlmeyer wrote: AL> > AL> > On 6/27/07, Amos Leffler wrote: AL> > AL> >> Dear Forum, AL> > AL> >> I found later after sending this note that the program ran (in AL> > AL> >> background) but did the calculation for all of the numbers of water AL> > AL> >> molecules. The original input did have as the last entry a number (64) AL> > AL> >> so there still is a problem. AL> > AL> > AL> > AL> > i disagree. please explain first the existence of the -ftrace=full AL> > AL> > which is obviously causing the problems you see. omit it and AL> > AL> > the script will run fine. AL> > AL> > AL> > AL> > axel. AL> > AL> > AL> > AL> >> AL> > AL> >> Amos Leffler AL> > AL> >> AL> > AL> >> Axel Kohlmeyer wrote: AL> > AL> >> > On Tue, 26 Jun 2007, Amos Leffler wrote: AL> > AL> >> > AL> > AL> >> > AL> Dear Forum, AL> > AL> >> > AL> Attached are the details of a problem I ran into AL> > AL> >> trying to AL> > AL> >> > AL> execute example21. I am using SuSE Linux 10.2 which has AL> > AL> >> worked well AL> > AL> >> > AL> up to this point so I think the problem is in the run_example AL> > AL> >> file. AL> > AL> >> > AL> > AL> >> > well, i would say the problem is in your using of -ftrace=full. why? AL> > AL> >> > example21 is special and requires as an argument the maximum number AL> > AL> >> > of water molecules that you want to run (32-512). if you provide AL> > AL> >> > AL> > AL> >> > the error message stems from the numerical compares of the first AL> > AL> >> > argument of the script, $1, with the corresponding number of waters. AL> > AL> >> > the way it is done, allows to have a default value of '0' (i.e. AL> > AL> >> > don't run any of those tests). your -ftrace=full argument expands AL> > AL> >> > to a non-numerical value, hence the error messages. in all other AL> > AL> >> > scripts it will be ignored, anyways. AL> > AL> >> > AL> > AL> >> > cheers, AL> > AL> >> > axel. AL> > AL> >> > AL> > AL> >> > AL> > AL> >> > AL> AL> > AL> >> > AL> Amos Leffler AL> > AL> >> > AL> AL> > AL> >> > AL> Walnut Creek, CA. AL> > AL> >> > AL> AL> > AL> >> > AL> unaffiliated AL> > AL> >> > AL> AL> > AL> >> > AL> AL> > AL> >> > AL> > AL> >> > AL> > AL> >> AL> > AL> >> _______________________________________________ AL> > AL> >> Pw_forum mailing list AL> > AL> >> Pw_forum at pwscf.org AL> > AL> >> http://www.democritos.it/mailman/listinfo/pw_forum AL> > AL> >> AL> > AL> >> AL> > AL> > AL> > AL> > AL> > AL> AL> > AL> AL> > AL> > AL> AL> _______________________________________________ AL> Pw_forum mailing list AL> Pw_forum at pwscf.org AL> http://www.democritos.it/mailman/listinfo/pw_forum AL> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From ferretti.andrea at unimore.it Mon Jul 2 11:57:28 2007 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Mon, 2 Jul 2007 11:57:28 +0200 (CEST) Subject: [Pw_forum] Want test crash In-Reply-To: <016301c7ba5b$64e136e0$d245c0de@masj> References: <137301.42076.qm@web60324.mail.yahoo.com> <016301c7ba5b$64e136e0$d245c0de@masj> Message-ID: Dear Mashao, it seems the espresso datafile cannot be read by WanT, some reasons might be: - the file ~test01/SCRATCH/silicon.save/datafile.xml is corrupted or empty - some portability issue due to the compiler/architecture combination occurred (even if I think this is somehow unlike since espresso is working) could you please post the make.sys file (from the WanT home directory) as well as a description of your machine ?? by the way: 1) did you run ./run.sh bands ? this would not run the whole test but only a specific calculation (bands interpolation using wannier functions); instead, use ./run.sh dft (to run espresso calculations) ./run.sh want (to compute the remaining part of the test) 2) WanT-2.1-RC1 is a "release compliant" version of the code (i.e. a pre-release): consider to update it to the stable version WanT-2.1 (to be released in few days) hope it helps andrea > Dear Sir: > I am a newbie for pwscf and want. I have compiled the parallel version > of PWSCF(espresso-3.2) and WANT(want-2.1-RC1). I excute the > tests of PWSCF and everything is OK! However, when I excute > the tests of WANT, the error appeared . > when I excute test01 for run.sh bands, the bands.x crashed and > the error in bands.out as following. > > DFT-data fmt automaticaly detected: qexml > > *** from PE : 0 > *** in routine : lattice_read_ext > *** error msg. : reading lattice > *** error code : 1 > *** aborting *** > -- Andrea Ferretti National Research Center S3, CNR-INFM ( http://s3.infm.it ) Dip. Fisica, Univ. di Modena e Reggio E. (Italy) Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti URL: http://www.nanoscience.unimo.it Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From mashao_jie at 163.com Mon Jul 2 15:16:37 2007 From: mashao_jie at 163.com (mashaojie163) Date: Mon, 2 Jul 2007 21:16:37 +0800 Subject: [Pw_forum] Want test crash References: <137301.42076.qm@web60324.mail.yahoo.com> <016301c7ba5b$64e136e0$d245c0de@masj> Message-ID: <005801c7bcab$39aa7390$d245c0de@masj> Thank you for your reponse. I excute ./run.sh all . and the other calculations terminated normally until the band.x excute. Another question: I can not compile want successfully with intel compiler. So I compiled it with PGI compiler. I have compiled PWSCF with intel compiler successfully! my make.sys file as following make.sys. Generated from make.sys.in by configure. # ----------------------------- # compilation rules .SUFFIXES : .SUFFIXES : .o .c .f .f90 .f90.o: $(CPP) $(CPPFLAGS) $< -o $*.F90 $(MPIF90) $(F90FLAGS) -c $*.F90 -o $*.o .f.o: $(F77) $(FFLAGS) -c $< .c.o: $(CC) $(CFLAGS) -c $< # ----------------------------- # compiler definitions CC = pgcc CFLAGS = -fast $(DFLAGS) $(IFLAGS) CPP = cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) F90 = pgf90 MPIF90 = /share/data/software/pgif/linux86-64/6.0/bin/mpif90 F90FLAGS = $(FFLAGS) $(FDFLAGS) $(IFLAGS) $(MODFLAGS) F77 = pgf77 FFLAGS = -fast -r8 FFLAGS_NOOPT = -O0 LD = /share/data/software/pgif/linux86-64/6.0/bin/mpif90 LDFLAGS = -Bstatic AR = ar ARFLAGS = ruv RANLIB = ranlib # ----------------------------- # libraries BLAS_LIBS = -lacml LAPACK_LIBS = -lacml FFT_LIBS = MPI_LIBS = MASS_LIBS = INTERNAL_LIBS = # ----------------------------- # application-specific settings # See include/defs.h.README for a list of precompilation options # (possible arguments to -D or -U) and their meaning DFLAGS = -D__PGI -D__FFTW -D__USE_INTERNAL_FFTW -D__MPI -D__PARA FDFLAGS = $(DFLAGS) IFLAGS = -I../include -I../include MODFLAGS = -I. -I../iotk -I../libs -I../wannier -I../transport LIBOBJS = ../libs/libwant.a ../iotk/libiotk.a ../ctools/libctools.a # LIBS must contain the location of all needed external libraries LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $(MPI_LIBS) $(MASS_LIBS) ----- Original Message ----- From: "Andrea Ferretti" To: "PWSCF forum list" Sent: Monday, July 02, 2007 5:57 PM Subject: Re: [Pw_forum] Want test crash > > Dear Mashao, > > it seems the espresso datafile cannot be read by WanT, some reasons might > be: > - the file ~test01/SCRATCH/silicon.save/datafile.xml is > corrupted or empty > - some portability issue due to the compiler/architecture combination > occurred (even if I think this is somehow unlike since espresso is > working) > > could you please post the make.sys file (from the WanT home directory) as > well as a description of your machine ?? > > by the way: > > 1) did you run > ./run.sh bands > ? > this would not run the whole test but only a specific > calculation (bands interpolation using wannier functions); > instead, use > ./run.sh dft (to run espresso calculations) > ./run.sh want (to compute the remaining part of the test) > > 2) WanT-2.1-RC1 is a "release compliant" version of the code (i.e. a > pre-release): consider to update it to the stable version WanT-2.1 > (to be released in few days) > > > hope it helps > > andrea > > >> Dear Sir: >> I am a newbie for pwscf and want. I have compiled the parallel version >> of PWSCF(espresso-3.2) and WANT(want-2.1-RC1). I excute the >> tests of PWSCF and everything is OK! However, when I excute >> the tests of WANT, the error appeared . >> when I excute test01 for run.sh bands, the bands.x crashed and >> the error in bands.out as following. >> >> DFT-data fmt automaticaly detected: qexml >> >> *** from PE : 0 >> *** in routine : lattice_read_ext >> *** error msg. : reading lattice >> *** error code : 1 >> *** aborting *** >> > > > -- > Andrea Ferretti > National Research Center S3, CNR-INFM ( http://s3.infm.it ) > Dip. Fisica, Univ. di Modena e Reggio E. (Italy) > Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti > URL: http://www.nanoscience.unimo.it > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From ferretti.andrea at unimore.it Mon Jul 2 16:12:04 2007 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Mon, 2 Jul 2007 16:12:04 +0200 (CEST) Subject: [Pw_forum] Want test crash In-Reply-To: <005801c7bcab$39aa7390$d245c0de@masj> References: <137301.42076.qm@web60324.mail.yahoo.com> <016301c7ba5b$64e136e0$d245c0de@masj> <005801c7bcab$39aa7390$d245c0de@masj> Message-ID: Dear Mashao, > Thank you for your reponse. > I excute ./run.sh all . and the other calculations > terminated normally until the bands.x excute. ok... did you run the whole test serially or in parallel ?? note that only the transport part of WanT support parallelization at the moment if the test crashed when doing bands.x calculaiton, it means it was successful in e.g. wannier localization, is it true ?? > Another question: I can not compile want successfully with > intel compiler. > So I compiled it with PGI compiler. I have compiled > PWSCF with intel compiler successfully! > please, send also a log of the compilation failure with the intel compiler; moving to PGI, try first compiling serially (which should be the defualt for WanT configure) -- Andrea Ferretti National Research Center S3, CNR-INFM ( http://s3.infm.it ) Dip. Fisica, Univ. di Modena e Reggio E. (Italy) Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti URL: http://www.nanoscience.unimo.it Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From chen_shao_hua197 at yahoo.com.tw Mon Jul 2 19:10:21 2007 From: chen_shao_hua197 at yahoo.com.tw (=?big5?q?=B3=AF=20=A4=D6=B5=D8?=) Date: Tue, 3 Jul 2007 01:10:21 +0800 (CST) Subject: [Pw_forum] Does anyone have old format N.pz-vbc NCPP ? Message-ID: <310754.48290.qm@web73010.mail.tp2.yahoo.com> Dear members, Does anyone have old format N.pz-vbc NCPP ? Because I want use it to change valence charge zp. Thanks max Department of Physics, National Taiwan University,Taiwan ____________________________________________________________________________________ ????????????? - ???? Yahoo!??????http://tw.info.yahoo.com/seal/index.html From chaohao.mse at gmail.com Wed Jul 4 15:07:11 2007 From: chaohao.mse at gmail.com (Chaohao Hu) Date: Wed, 4 Jul 2007 15:07:11 +0200 Subject: [Pw_forum] about the pseudopotential of Iridium (Ir) Message-ID: <6502ee650707040607q6564c9fk54b20d622718cc5e@mail.gmail.com> Dear pwscf users, I want to consider the spin orbit coupling effect during my calculations, but I can not find the pseudopotential of Iridium (Ir) considering the full relativistic effect from the PWSCF web. Who can help me generate it? I will appreciate his great help. Best regards, C. H. Hu -- ============================ C.H. Hu Postdoctoral fellow Chimie et Physico-Chimie appliquees Institut Francais du Petrole (IFP) Rueil-Malmaison, France E_mail: chaohao.mse at gmail.com ============================ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070704/cb34472f/attachment.htm From yukihiro_okuno at fujifilm.co.jp Thu Jul 5 09:46:59 2007 From: yukihiro_okuno at fujifilm.co.jp (yukihiro_okuno at fujifilm.co.jp) Date: Thu, 5 Jul 2007 16:46:59 +0900 Subject: [Pw_forum] erro in virtual.x for VCA Message-ID: Dear PWscf Users. I'm now tring to use virtual crystal approximation (VCA), and use virtual.x for making mixing pseudo potential, But when i use virtual.x with specifing two pseudo potential file, error had occured like from virtual :error # 1 grid interpolation is not working yet I think this is the inconsistent for the grid number of pseudo potential, but how should i cure the pseudo potential? ( I use the down loaded pseudo potential from PWScf site.) Sincerely, Yukihiro Okuno, From yukihiro_okuno at fujifilm.co.jp Thu Jul 5 13:24:12 2007 From: yukihiro_okuno at fujifilm.co.jp (yukihiro_okuno at fujifilm.co.jp) Date: Thu, 5 Jul 2007 20:24:12 +0900 Subject: [Pw_forum] erro in virtual.x for VCA 2 ( can I use splinelib.f90?) Message-ID: Dear PWscf Users and Developer. I see virtual.f90 and it is not implemented for interpolation for pseudo-potential yet. But the code are written like if(interpolate) then -------- call dosplineint(....) ------ endif and can I use /Module/splinelib.f90 for this dosplineint with slight modification of code ? Or , are there problem for such a spline interpolation for virtual crystal approximation? (Why stop to use dosplineint in virtual.f90 ?) Sincerely, Yukihiro Okuno. From degironc at sissa.it Thu Jul 5 13:52:32 2007 From: degironc at sissa.it (degironc) Date: Thu, 05 Jul 2007 13:52:32 +0200 Subject: [Pw_forum] erro in virtual.x for VCA 2 ( can I use splinelib.f90?) In-Reply-To: References: Message-ID: <468CDB80.6010304@sissa.it> the code was adapted to allow for interpolation but not tested as it was not needed at that time. If you test it and find that it is working properly we could remove the error message... otherwise report about the problems. Thank you, Stefano de Gironcoli SISSA and DEMOCRITOS yukihiro_okuno at fujifilm.co.jp wrote: > Dear PWscf Users and Developer. > > > I see virtual.f90 and it is not implemented for > interpolation for pseudo-potential yet. > > But the code are written like > > if(interpolate) then > -------- > call dosplineint(....) > ------ > endif > > and can I use /Module/splinelib.f90 for this dosplineint with > slight modification of code ? > > Or , are there problem for such a spline interpolation for virtual > crystal approximation? > (Why stop to use dosplineint in virtual.f90 ?) > > Sincerely, > > > Yukihiro Okuno. > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From helen at fh.huji.ac.il Thu Jul 5 14:17:31 2007 From: helen at fh.huji.ac.il (Helen) Date: Thu, 5 Jul 2007 15:17:31 +0300 Subject: [Pw_forum] modified Hartree exchange potential Message-ID: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> Hi, I want to modifiy the Hartree exchange potential to take into account a screened coulomb potential. Which files need to be altered? I noticed that a subroutine hartree(..) is called but couldn't find it. Also I wanted to change the exchange functional for a new functional. I changed Modules/functionals.f90 to call a new functional, iexch=8, ELSEIF (iexch == 8) THEN ! 'bnl' CALL slater_bnl(rs, ex, vx) and then added the new functional in flib/functionals.f90 as a new subroutine under the slater one. !!----------------------------------------------------------------------- subroutine slater_bnl (rs, ex, vx) !----------------------------------------------------------------------- ! bnl ! USE kinds implicit none real(DP) :: rs, ex, vx, kf, q real(DP), parameter :: f= -0.687247939924714d0, alpha = 2.0d0/3.0d0, & pi34 = 0.6203504908994d0, pie=3.141592653589793d0 ,gam=0.5d0 ! pi34=(3/4pi)^(1/3) ! f = -9/8*(3/2pi)^(2/3) ! kf=3d0*pie**2*pi34/rs q=gam/kf ex=(kf/(2d0*pie))*(q**2*(((2d0*dsqrt(pie)/q)*derf(1d0/q))-1d0+ & ((q**2-2d0)*(1d0-dexp(-1d0/q**2))))-1.5d0) vx=(-gam/(6d0*pie))*((-4d0*dexp(-1d0/q**2)/q) & +2d0*q**2*(1d0-dexp(-1d0/q**2)) & -2d0*(dexp(-1d0/q**2)*(q**2-2d0)/q)-2d0*dsqrt(pie)*derf(1d0/q) & -2d0*q+2d0*q*(q**2-2d0)*(1d0-dexp(-1d0/q**2))+4d0*dsqrt(pie)*derf(1d0/q)) & +(2d0*gam/(3d0*pie*q))*(q**2*(((2d0*dsqrt(pie)/q)*derf(1d0/q))-1d0+ & ((q**2-2d0)*(1d0-dexp(-1d0/q**2))))-1.5d0) ! return end subroutine slater_bnl Is this sufficient, or are there any other routines I need to change? Thanks for your help, Helen Eisenberg -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070705/a23c4caf/attachment.htm From mverissi at ictp.it Thu Jul 5 14:36:01 2007 From: mverissi at ictp.it (Marcos Verissimo Alves) Date: Thu, 5 Jul 2007 14:36:01 +0200 (CEST) Subject: [Pw_forum] Error 61 in rdiaghg In-Reply-To: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> References: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> Message-ID: <40897.10.2.0.25.1183638961.squirrel@webmail3.ictp.trieste.it> Hi all, I am trying to run a simple calculation (scf, ibrav=2, 2 atoms in the unit cell, gamma point) using espresso 3.0 but I'm getting from rdiaghg : error # 61 info =/= 0 How can I solve this? If necessary, I can post my input and pseudo files. Thanks in advance, Marcos -- Dr. Marcos Verissimo Alves Post-Doctoral Fellow Condensed Matter and Statistical Physics Sector International Centre for Theoretical Physics Trieste, Italy -------- I have become so addicted to vi that I try to exit OpenOffice by typing :wq! From mverissi at ictp.it Thu Jul 5 14:43:33 2007 From: mverissi at ictp.it (Marcos Verissimo Alves) Date: Thu, 5 Jul 2007 14:43:33 +0200 (CEST) Subject: [Pw_forum] Please disregard my previous message... In-Reply-To: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> References: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> Message-ID: <44117.10.2.0.25.1183639413.squirrel@webmail3.ictp.trieste.it> Sorry for posting a question that already had an answer in the pwscf archives. I have just seen a possible solution to the problem. Marcos -- Dr. Marcos Verissimo Alves Post-Doctoral Fellow Condensed Matter and Statistical Physics Sector International Centre for Theoretical Physics Trieste, Italy -------- I have become so addicted to vi that I try to exit OpenOffice by typing :wq! From eyvaz_isaev at yahoo.com Thu Jul 5 14:49:32 2007 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 5 Jul 2007 05:49:32 -0700 (PDT) Subject: [Pw_forum] Error 61 in rdiaghg In-Reply-To: <40897.10.2.0.25.1183638961.squirrel@webmail3.ictp.trieste.it> Message-ID: <25978.10120.qm@web60321.mail.yahoo.com> Hi Marcos, Similar problem was discussed before in the forum many times. Briefly, the problem is due to a fact that an eigenvalue can not converge with a machine precision in this particular case. One reason is that a structure you used is not correct. So, have a look at your input file again. If it is not the case use "conjugated-gradient" method for diagonalization in &electrons. Bests, Eyvaz. --- Marcos Verissimo Alves wrote: > Hi all, > > I am trying to run a simple calculation (scf, > ibrav=2, 2 atoms in the unit > cell, gamma point) using espresso 3.0 but I'm > getting > > from rdiaghg : error # 61 > info =/= 0 > > How can I solve this? If necessary, I can post my > input and pseudo files. > > Thanks in advance, > > Marcos > > -- > Dr. Marcos Verissimo Alves > Post-Doctoral Fellow > Condensed Matter and Statistical Physics Sector > International Centre for Theoretical Physics > Trieste, Italy > > -------- > > I have become so addicted to vi that I try to exit > OpenOffice by typing :wq! > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, and Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Pinpoint customers who are looking for what you sell. http://searchmarketing.yahoo.com/ From mverissi at ictp.it Thu Jul 5 15:03:34 2007 From: mverissi at ictp.it (Marcos Verissimo Alves) Date: Thu, 5 Jul 2007 15:03:34 +0200 (CEST) Subject: [Pw_forum] Error 61 in rdiaghg In-Reply-To: <25978.10120.qm@web60321.mail.yahoo.com> References: <40897.10.2.0.25.1183638961.squirrel@webmail3.ictp.trieste.it> <25978.10120.qm@web60321.mail.yahoo.com> Message-ID: <45704.10.2.0.25.1183640614.squirrel@webmail3.ictp.trieste.it> Hi Eyvaz, I am trying to run a test for NdS, which has the NaCl structure. I generated a GGA pseudo for Nd based on an existing LDA one using Vanderbilt's code, converted it to UPF (successfully according to uspp2upf.x) and wanted to test it to see its performance in a simple calculation. I am using ibrav=2, and one atom at 0.00 0.00 0.00 and another at 0.50 0.50 0.50 - is this the correct setting for the NaCl structure? Now I have tried to change the diagonalization method to davidson, cg and diis, to no avail. The error is still there, and at the very beginning of the scf cycle (the code stops at "Starting wfc are atomic"). Marcos > Hi Marcos, > > Similar problem was discussed before in the forum > many times. > Briefly, the problem is due to a fact that an > eigenvalue can not converge with a machine precision > in this particular case. > One reason is that a structure you used is not > correct. So, have a look at your input file again. > If it is not the case use "conjugated-gradient" method > for diagonalization in &electrons. > > Bests, > Eyvaz. > > --- Marcos Verissimo Alves wrote: > >> Hi all, >> >> I am trying to run a simple calculation (scf, >> ibrav=2, 2 atoms in the unit >> cell, gamma point) using espresso 3.0 but I'm >> getting >> >> from rdiaghg : error # 61 >> info =/= 0 >> >> How can I solve this? If necessary, I can post my >> input and pseudo files. >> >> Thanks in advance, >> >> Marcos >> >> -- >> Dr. Marcos Verissimo Alves >> Post-Doctoral Fellow >> Condensed Matter and Statistical Physics Sector >> International Centre for Theoretical Physics >> Trieste, Italy >> >> -------- >> >> I have become so addicted to vi that I try to exit >> OpenOffice by typing :wq! >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> > > > ------------------------------------------------------------------- > Prof. Eyvaz Isaev, > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, > Russia, and > Condensed Matter Theory Group, Uppsala University, Sweden > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > > ____________________________________________________________________________________ > Pinpoint customers who are looking for what you sell. > http://searchmarketing.yahoo.com/ > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Dr. Marcos Verissimo Alves Post-Doctoral Fellow Condensed Matter and Statistical Physics Sector International Centre for Theoretical Physics Trieste, Italy -------- I have become so addicted to vi that I try to exit OpenOffice by typing :wq! From paulatto at sissa.it Thu Jul 5 15:30:21 2007 From: paulatto at sissa.it (Lorenzo Paulatto) Date: Thu, 5 Jul 2007 15:30:21 +0200 (CEST) Subject: [Pw_forum] modified Hartree exchange potential In-Reply-To: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> References: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> Message-ID: <3440.147.122.5.194.1183642221.squirrel@webmail.sissa.it> On Thu, July 5, 2007 14:17, Helen wrote: > Hi, > I want to modifiy the Hartree exchange potential to take into account a > screened coulomb potential. Which files need to be altered? I noticed that > a subroutine hartree(..) is called but couldn't find it. it is v_h(..) in PW/v_of_rho.f90; remember that the potential you use in the scf code must be consistent with the one you have used in the atomic (ld1) code to generate the pseudopotential. The latter is computed in atomic/hartree.f90 > Also I wanted to change the exchange functional for a new functional. > I changed Modules/functionals.f90 to call a new functional, iexch=8, > and then added the new functional in flib/functionals.f90 as a new > subroutine under the slater one. > [snip] > Is this sufficient, or are there any other routines I need to change? Same as above: the atomic code must be consistent, otherwise unpredictable things *may* happen. bye -- Lorenzo Paulatto +39 040 3787 312 http://people.sissa.it/~paulatto/ ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by SquirrelMail http://www.squirrelmail.org/ From mashao_jie at 163.com Thu Jul 5 15:42:00 2007 From: mashao_jie at 163.com (mashaojie163) Date: Thu, 5 Jul 2007 21:42:00 +0800 Subject: [Pw_forum] Want test crash References: <137301.42076.qm@web60324.mail.yahoo.com> <016301c7ba5b$64e136e0$d245c0de@masj> <005801c7bcab$39aa7390$d245c0de@masj> Message-ID: <019101c7bf0a$4a5178e0$d245c0de@masj> Dear Sir: did you run the whole test serially or in parallel ?? > note that only the transport part of WanT support parallelization at the > moment I think the error is resulted in because I compiled want and excute want in parallel. So now I compile want and excute it seriallly 1) I compiled it by ./configue F90=pgf90 ; make all everything is OK 2) and I excute DFT calculation in parallel I modify PARA_PREFIX="mpirun -np 2 -machinefile ../nodefile" ./run.sh dft 3) I excute want serially PARA_PREFIX= ./run.sh want everything is OK until band.x crash. The error in band.out is DFT-data fmt automaticaly detected: qexml *** from PE : 0 *** in routine : lattice_read_ext *** error msg. : reading lattice *** error code : 1 *** aborting *** 4) ./run.sh check running CHECK... ######################### File: disentangle.out Variable This File Reference % Difference iteration 1 1 0.000000 ok omega_i 26.785130 26.785104 0.000001 ok omega_i_avrg 6.696282 6.696276 0.000001 ok status CONVERGED CONVERGED ######################### File: wannier.out Variable This File Reference % Difference iteration 8500 8500 0.000000 ok center_sum 28.627843 28.627818 0.000001 ok omega_i 26.785130 26.785104 0.000001 ok omega_d 0.000006 0.000006 0.000000 ok omega_od 1.842708 1.842709 0.000001 ok omega_tot 28.627843 28.627818 0.000001 ok omega_avrg 7.156961 7.156955 0.000001 ok onsite_hr 5.215476 5.215477 0.000000 ok status END_OF_ITER END_OF_ITER Then I ecute test02. 1) .PARA_PREFIX="mpirun -np 2 -machinefile ../nodefile" ./run_Cu.sh dft every thing is ok 2) .PARA_PREFIX= ./run_Cu.sh want running DISENTANGLE calculation... CRASH done running WANNIER calculation... CRASH done running BANDS calculation... CRASH done running DOS calculation... CRASH done running PLOT calculation... CRASH done The same error as above occur in disentangle.out wannier.out ...etc. DFT-data fmt automaticaly detected: qexml *** from PE : 0 *** in routine : lattice_read_ext *** error msg. : reading lattice *** error code : 1 *** aborting *** Why the error occur? ----- Original Message ----- From: "Andrea Ferretti" To: Sent: Monday, July 02, 2007 10:12 PM Subject: (????????-????????????????????)Re: [Pw_forum] Want test crash > Dear Mashao, > >> Thank you for your reponse. >> I excute ./run.sh all . and the other calculations >> terminated normally until the bands.x excute. > > ok... > did you run the whole test serially or in parallel ?? > note that only the transport part of WanT support parallelization at the > moment > > if the test crashed when doing bands.x calculaiton, it means it was > successful in e.g. wannier localization, is it true ?? > > >> Another question: I can not compile want successfully with >> intel compiler. >> So I compiled it with PGI compiler. I have compiled >> PWSCF with intel compiler successfully! >> > > please, send also a log of the compilation failure with the intel > compiler; moving to PGI, try first compiling serially (which should be > the defualt for WanT configure) > > -- > Andrea Ferretti > National Research Center S3, CNR-INFM ( http://s3.infm.it ) > Dip. Fisica, Univ. di Modena e Reggio E. (Italy) > Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti > URL: http://www.nanoscience.unimo.it > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From eyvaz_isaev at yahoo.com Thu Jul 5 16:43:07 2007 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 5 Jul 2007 07:43:07 -0700 (PDT) Subject: [Pw_forum] Error 61 in rdiaghg In-Reply-To: <45704.10.2.0.25.1183640614.squirrel@webmail3.ictp.trieste.it> Message-ID: <44019.57871.qm@web60311.mail.yahoo.com> Hi Marcos, > > I am trying to run a test for NdS, which has the > NaCl structure. I generated a GGA pseudo for Nd > based on an existing LDA one using Vanderbilt's > code, converted it to UPF (successfully according to > uspp2upf.x) and wanted to test it to see its > performance in a simple calculation. >I am using ibrav=2, and one atom at 0.00 0.00 0.00 > and another at 0.50 0.50 0.0 - is this the correct > setting for the NaCl structure? Of course, this is OK. >Now I have tried to change the diagonalization >method to davidson, cg and diis, to no avail. The >error is still there, and at the very beginning of >the scf cycle (the code stops at "Starting wfc are >atomic"). So, I can suggest that this is due to the Nd pseudopotential. Did you try to calculate ground state properties of hypotetical fcc and bcc phases for Nd? Bests, Eyvaz. > > Marcos > > > Hi Marcos, > > > > Similar problem was discussed before in the forum > > many times. > > Briefly, the problem is due to a fact that an > > eigenvalue can not converge with a machine > precision > > in this particular case. > > One reason is that a structure you used is not > > correct. So, have a look at your input file again. > > If it is not the case use "conjugated-gradient" > method > > for diagonalization in &electrons. > > > > Bests, > > Eyvaz. > > > > --- Marcos Verissimo Alves > wrote: > > > >> Hi all, > >> > >> I am trying to run a simple calculation (scf, > >> ibrav=2, 2 atoms in the unit > >> cell, gamma point) using espresso 3.0 but I'm > >> getting > >> > >> from rdiaghg : error # 61 > >> info =/= 0 > >> > >> How can I solve this? If necessary, I can post my > >> input and pseudo files. > >> > >> Thanks in advance, > >> > >> Marcos > >> > >> -- > >> Dr. Marcos Verissimo Alves > >> Post-Doctoral Fellow > >> Condensed Matter and Statistical Physics Sector > >> International Centre for Theoretical Physics > >> Trieste, Italy > >> > >> -------- > >> > >> I have become so addicted to vi that I try to > exit > >> OpenOffice by typing :wq! > >> > >> _______________________________________________ > >> Pw_forum mailing list > >> Pw_forum at pwscf.org > >> > http://www.democritos.it/mailman/listinfo/pw_forum > >> > > > > > > > ------------------------------------------------------------------- > > Prof. Eyvaz Isaev, > > Theoretical Physics Department, Moscow State > Institute of Steel & Alloys, > > Russia, and > > Condensed Matter Theory Group, Uppsala University, > Sweden > > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com > > > > > > > > > ____________________________________________________________________________________ > > Pinpoint customers who are looking for what you > sell. > > http://searchmarketing.yahoo.com/ > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > -- > Dr. Marcos Verissimo Alves > Post-Doctoral Fellow > Condensed Matter and Statistical Physics Sector > International Centre for Theoretical Physics > Trieste, Italy > > -------- > > I have become so addicted to vi that I try to exit > OpenOffice by typing :wq! > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, and Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________Ready for the edge of your seat? Check out tonight's top picks on Yahoo! TV. http://tv.yahoo.com/ From ferretti.andrea at unimore.it Thu Jul 5 17:14:20 2007 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Thu, 5 Jul 2007 17:14:20 +0200 (CEST) Subject: [Pw_forum] Want test crash In-Reply-To: <019101c7bf0a$4a5178e0$d245c0de@masj> References: <137301.42076.qm@web60324.mail.yahoo.com> <016301c7ba5b$64e136e0$d245c0de@masj> <005801c7bcab$39aa7390$d245c0de@masj> <019101c7bf0a$4a5178e0$d245c0de@masj> Message-ID: > > DFT-data fmt automaticaly detected: qexml > > *** from PE : 0 > *** in routine : lattice_read_ext > *** error msg. : reading lattice > *** error code : 1 > *** aborting *** > 4) ./run.sh check > running CHECK... > try to compile with -D__IOTK_SAFEST (to be added to FFLAGS and FDFLAGS after make.sys generation) Andrea Andrea Ferretti National Research Center S3, CNR-INFM ( http://s3.infm.it ) Dip. Fisica, Univ. di Modena e Reggio E. (Italy) Tel: +39 059 2055301; Fax: +39 059 374794; Skype: andrea_ferretti URL: http://www.nanoscience.unimo.it Please, if possible, don't send me MS Word or PowerPoint attachments Why? See: http://www.gnu.org/philosophy/no-word-attachments.html From akohlmey at cmm.chem.upenn.edu Thu Jul 5 16:18:20 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 5 Jul 2007 10:18:20 -0400 (EDT) Subject: [Pw_forum] Error 61 in rdiaghg In-Reply-To: <45704.10.2.0.25.1183640614.squirrel@webmail3.ictp.trieste.it> Message-ID: On Thu, 5 Jul 2007, Marcos Verissimo Alves wrote: MVA> Hi Eyvaz, MVA> MVA> I am trying to run a test for NdS, which has the NaCl structure. I MVA> generated a GGA pseudo for Nd based on an existing LDA one using MVA> Vanderbilt's code, converted it to UPF (successfully according to MVA> uspp2upf.x) and wanted to test it to see its performance in a simple MVA> calculation. I am using ibrav=2, and one atom at 0.00 0.00 0.00 and MVA> another at 0.50 0.50 0.50 - is this the correct setting for the NaCl MVA> structure? Now I have tried to change the diagonalization method to MVA> davidson, cg and diis, to no avail. The error is still there, and at the MVA> very beginning of the scf cycle (the code stops at "Starting wfc are MVA> atomic"). hi marcos, that is frequently a sign of a bad pseudopotential (check in the psp generation log for negative density), but also some BLAS/LAPACK libraries are more sensitive than others. try switching between ATLAS, MKL, ACML, GOTO-BLAS, and plain BLAS/LAPACK. ciao, axel. MVA> Marcos MVA> MVA> > Hi Marcos, MVA> > MVA> > Similar problem was discussed before in the forum MVA> > many times. MVA> > Briefly, the problem is due to a fact that an MVA> > eigenvalue can not converge with a machine precision MVA> > in this particular case. MVA> > One reason is that a structure you used is not MVA> > correct. So, have a look at your input file again. MVA> > If it is not the case use "conjugated-gradient" method MVA> > for diagonalization in &electrons. MVA> > MVA> > Bests, MVA> > Eyvaz. MVA> > MVA> > --- Marcos Verissimo Alves wrote: MVA> > MVA> >> Hi all, MVA> >> MVA> >> I am trying to run a simple calculation (scf, MVA> >> ibrav=2, 2 atoms in the unit MVA> >> cell, gamma point) using espresso 3.0 but I'm MVA> >> getting MVA> >> MVA> >> from rdiaghg : error # 61 MVA> >> info =/= 0 MVA> >> MVA> >> How can I solve this? If necessary, I can post my MVA> >> input and pseudo files. MVA> >> MVA> >> Thanks in advance, MVA> >> MVA> >> Marcos MVA> >> MVA> >> -- MVA> >> Dr. Marcos Verissimo Alves MVA> >> Post-Doctoral Fellow MVA> >> Condensed Matter and Statistical Physics Sector MVA> >> International Centre for Theoretical Physics MVA> >> Trieste, Italy MVA> >> MVA> >> -------- MVA> >> MVA> >> I have become so addicted to vi that I try to exit MVA> >> OpenOffice by typing :wq! MVA> >> MVA> >> _______________________________________________ MVA> >> Pw_forum mailing list MVA> >> Pw_forum at pwscf.org MVA> >> http://www.democritos.it/mailman/listinfo/pw_forum MVA> >> MVA> > MVA> > MVA> > ------------------------------------------------------------------- MVA> > Prof. Eyvaz Isaev, MVA> > Theoretical Physics Department, Moscow State Institute of Steel & Alloys, MVA> > Russia, and MVA> > Condensed Matter Theory Group, Uppsala University, Sweden MVA> > Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com MVA> > MVA> > MVA> > MVA> > ____________________________________________________________________________________ MVA> > Pinpoint customers who are looking for what you sell. MVA> > http://searchmarketing.yahoo.com/ MVA> > _______________________________________________ MVA> > Pw_forum mailing list MVA> > Pw_forum at pwscf.org MVA> > http://www.democritos.it/mailman/listinfo/pw_forum MVA> > MVA> MVA> MVA> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Thu Jul 5 17:40:45 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 5 Jul 2007 17:40:45 +0200 Subject: [Pw_forum] modified Hartree exchange potential In-Reply-To: <3440.147.122.5.194.1183642221.squirrel@webmail.sissa.it> References: <00de01c7befe$75f79940$cd604084@fh.huji.ac.il> <3440.147.122.5.194.1183642221.squirrel@webmail.sissa.it> Message-ID: On Jul 5, 2007, at 15:30 , Lorenzo Paulatto wrote: >> I want to modifiy the Hartree exchange potential to take into >> account a >> screened coulomb potential. Which files need to be altered? I >> noticed that >> a subroutine hartree(..) is called but couldn't find it. > > it is v_h(..) in PW/v_of_rho.f90; remember that the potential you > use in > the scf code must be consistent with the one you have used in the > atomic > (ld1) code to generate the pseudopotential. The latter is computed in > atomic/hartree.f90 also note that the scf code implicitly assumes the 1/r dependence when calculating the G=0 term of the potential >> Also I wanted to change the exchange functional for a new functional. >> I changed Modules/functionals.f90 to call a new functional, iexch=8, >> and then added the new functional in flib/functionals.f90 as a new >> subroutine under the slater one. >> [snip] >> Is this sufficient, or are there any other routines I need to change? Routines "set_dft_from_name" and "_from_indices" should be modified accordingly > Same as above: the atomic code must be consistent, otherwise > unpredictable > things *may* happen. In principle, pseudopotentials should be generated using the same functional that is used in the scf calculation, but nothing dramatic will happen otherwise. In any event: it is wise to try new functionals on atoms first. The atomic code uses the same routines as the scf code for functionals. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From lanhaiping at gmail.com Thu Jul 5 17:46:20 2007 From: lanhaiping at gmail.com (lan haiping) Date: Thu, 5 Jul 2007 23:46:20 +0800 Subject: [Pw_forum] Strange behavior of scf calculation under finite-field Message-ID: Dear All, I am studying electric-enthalpy method implemented in pw.x with example31. There are some problems i cannot solve by myself. Firstly, when i set "ecutrho" explictly, calculations under finite-field are failed without any complain information, such as CRASH report. My server is of 8GB memory, and no such crash for other scf calculations . I had tried 'cg' and 'david' methods, and cannot solve this problems. Secondly, when i set occupations to 'mv' smearing scheme with degauss=0.03 a.u. , the scf calculation of finite-field situation is very difficult to reach convergence. I have tried to increase or decrease mixing_ndim, but failed. Besides, i have tried to increase nbnd , which seems no effect on scf convergence. Would you please give me some suggestion on parameters setting for scf calculation under finite-field ? Thanks for your time and attention . Any suggestion or comment is appreciated . Best Wishes, H.P -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070705/b3876ae9/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu Jul 5 17:03:25 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 5 Jul 2007 11:03:25 -0400 (EDT) Subject: [Pw_forum] Strange behavior of scf calculation under finite-field In-Reply-To: Message-ID: On Thu, 5 Jul 2007, lan haiping wrote: HPL> Dear All, HPL> HPL> I am studying electric-enthalpy method implemented in pw.x with HPL> example31. There are some problems i cannot solve by myself. HPL> Firstly, when i set "ecutrho" explictly, calculations under finite-field HPL> are failed without any complain information, such as CRASH report. My server please provide an input that can reproduce this. HPL> is of 8GB memory, and no such crash for other scf calculations . I had tried HPL> 'cg' and 'david' methods, and cannot solve this problems. HPL> Secondly, when i set occupations to 'mv' smearing scheme with degauss=0.03 HPL> a.u. , the scf calculation of finite-field situation HPL> is very difficult to reach convergence. I have tried to increase or calculations under external field _are_ more difficult to converge. how does your field strength compare to the value in example31? HPL> decrease mixing_ndim, but failed. Besides, i have tried to increase HPL> nbnd , which seems no effect on scf convergence. Would you please give me HPL> some suggestion on parameters setting for scf calculation under finite-field HPL> ? it would be much easier to detect problems or make suggestions, if you'd post the relevant part of your input file (or another input that has the exact same problem). cheers, axel. HPL> HPL> Thanks for your time and attention . Any suggestion or comment is HPL> appreciated . HPL> HPL> Best Wishes, HPL> H.P HPL> HPL> HPL> HPL> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From conor.hogan at roma2.infn.it Thu Jul 5 18:27:44 2007 From: conor.hogan at roma2.infn.it (Conor Hogan) Date: Thu, 5 Jul 2007 18:27:44 +0200 (CEST) Subject: [Pw_forum] (still) inconsistent pp.x and projwfc.x ... Message-ID: <41803.141.108.248.231.1183652864.squirrel@141.108.248.231> Dear forum, I experienced some problems a while ago http://www.democritos.it/pipermail/pw_forum/2007-March/006105.html on ordering of states in a metalloporphyrin molecule being different according to pp.x and projwfc.x, and the problem has not gone away :( I did try Paolo's suggestion to avoid {Gamma} http://www.democritos.it/pipermail/pw_forum/2007-March/006128.html and I also tried fixed occ vs smearing; 3.2-cvs/3.2/3.1; pgi and intel; gopenmol vs xcrysden; tuesday vs wednesday. To no avail. Then very cleverly after all these tests I decided to do the same on a bare Co atom, and the problem is there again (doh!). To summarise: Co: 3d7 4s2 (3.2-cvs of yesterday 4Jul2007, ifort 9.1, serial run) pp.x and visual analysis: band up down 6 s s 5 dx2-y2 dx2-y2 4 dxy dxy 3 dxz dxz 2 dyz dyz 1 dz2 dz2 projwfc.x, state #1 = s (l=0,m=1), states #2-#6 are d (l=1,m=1-5) 6 #1 #1 5 #6 #3+#4 4 #5 #3+#4 3 #3+#4 #6 2 #3+#4 #5 1 #2 #2 presumably, the #3 (50%) +#4 (50%) are the dxz+dyz combinations. Since my cell is not massive, the degeneracy is broken enough I think that you should have clearly distinguishable bands. The problem is of course more clear in the Co-porphryin case, although there it was the pp.x results that appeared unphysical, which is very weird (and unsatisfying) I attach the basic inputs, the Co pseudo is the PBE from the website. I really hope that someone can look into this, either to point out that I'm completely off the mark, am running one of the pp codes wrong, or to put my mind at ease and identify a bug - because if it is, its the worst kind of bug: code runs fine but gives *sometimes* wrong results! Many thanks, Conor ######## scf input &control calculation = 'scf' verbosity= 'high', etot_conv_thr=1.0e-4, prefix='Co_atom', pseudo_dir = './', disk_io='minimal', wf_collect=.true. outdir='./tmp' / &system ibrav= 8, celldm(1)=14., celldm(2) = 1.00, celldm(3) = 0.75 nat=1, ntyp=1, nspin =2, multiplicity = 4, noncolin = .false., nbnd = 10, ecutwfc = 30.0, ecutrho=240., occupations='from_input' / &electrons conv_thr = 1.0d-7 mixing_beta = 0.7, startingpot='atomic' / ATOMIC_SPECIES Co 58.00 Co.pbe-nd-rrkjus.UPF ATOMIC_POSITIONS {angstrom} Co 0.5000000000 0.5000000000 0.5000000000 0 0 0 K_POINTS {crystal} 1 0.0 0.0 0.0 1.0 OCCUPATIONS 1.0 1.0 1.0 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 0.4 1.0 0.0 0.0 0.0 0.0 ############# pp.x input (or something like it) &inputpp prefix = 'Co_atom' outdir = './' filplot = 'Co_atom.dat' plot_num= 7 spin_component = 2, kpoint=1 kband=7, / &plot nfile = 1 filepp(1) = 'Co_atom.dat' weight(1) = 1.0 iflag = 3 output_format = 5 fileout = 'Co_atom_2_7.xcrys' x0(1) =-0.00, x0(2)=-0.00, x0(3) = -0.00, e1(1) =0.5, e1(2)=0.0, e1(3) = 0.0, e2(1) =0.0, e2(2)=0.5, e2(3) = 0.0, e3(1) =0.0, e3(2)=0.0, e3(3) = 0.5, nx=30, ny=30, nz =30 / ############### projwfc input &inputpp prefix='Co_atom', outdir='./', ngauss = 0, degauss = 0.01, lsym = .false., filproj='banana' &end ---- Dr. Conor Hogan --- ----- --- --- Dipartimento di Fisica e CNR-INFM | : | : Universita' di Roma "Tor Vergata" |-- : --- |--- Tel: +39 06 72594908 | : | : Fax: +39 06 2023507 --- : --- : http://www.fisica.uniroma2.it/~cmtheo-group/ European Theoretical Spectroscopy Facility http://www.etsf.eu Q. What do you call a fish with no eyes? A. A fsh. (Tommy Cooper) The early bird gets the worm, but the second mouse gets the cheese. - S. Wright If you go through a lot of hammers each month, I don't think it necessarily means you're a hard worker. It may just mean that you have a lot to learn about proper hammer maintenance - J. Handey From srijan.india at gmail.com Thu Jul 5 22:48:04 2007 From: srijan.india at gmail.com (Srijan Kumar Saha) Date: Fri, 6 Jul 2007 02:18:04 +0530 Subject: [Pw_forum] Strange behavior of phonon calculation Message-ID: <3fcc7e670707051348l2ef8e9b4yc31eae3d3f523ad3@mail.gmail.com> Dear pwscf users, My calculated total energy for a metal (spin-less) is total energy = -22.79040567 Ry and correcponding gamma phonon frequency = 736 cm-1 When I swicth on the spin tag to specify the spin polarized calculation for the same sample, then I got total energy = -22.79040581 Ry and correcponding gamma phonon frequency = 750 cm-1 while all the other input parameters are exactly same. In latter case also total magnetization is zero i,e, equal number of up and down electrons then why this difference????? This difference is not expected at all. Is it due to any bug within the spin-polarized phonon codes or am I missing something???? Looking forward to your kind reply. With best regards, Srijan Dept. of Physics CCMT, India -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070706/30866c60/attachment.htm From lanhaiping at gmail.com Fri Jul 6 08:22:48 2007 From: lanhaiping at gmail.com (lan haiping) Date: Fri, 6 Jul 2007 14:22:48 +0800 Subject: [Pw_forum] Strange behavior of scf calculation under finite-field In-Reply-To: References: Message-ID: Thank you, Axel. My calculation is modified from example31, and the script is attached below. The field strength i set for calculation is 1.0e-3 a.u., the same as all examples implemented in Q.E. I want to do a 'relax' calculation after a 'scf' calculation under finite-field, but the calculation always cannot reach a successful convergence after one or more 'bfgs' steps. Is there any trick to accelerate or reach scf convergence ? I have tried several times with different schemes , such as increasing bands number , adding smearing , or decreasing /increasing mixing parameters. Unfortranately , there seems unsuccessful. Regards, H.P On 7/5/07, Axel Kohlmeyer wrote: > > On Thu, 5 Jul 2007, lan haiping wrote: > > HPL> Dear All, > HPL> > HPL> I am studying electric-enthalpy method implemented in pw.x with > HPL> example31. There are some problems i cannot solve by myself. > HPL> Firstly, when i set "ecutrho" explictly, calculations under > finite-field > HPL> are failed without any complain information, such as CRASH report. My > server > > please provide an input that can reproduce this. > > HPL> is of 8GB memory, and no such crash for other scf calculations . I > had tried > HPL> 'cg' and 'david' methods, and cannot solve this problems. > HPL> Secondly, when i set occupations to 'mv' smearing scheme with > degauss=0.03 > HPL> a.u. , the scf calculation of finite-field situation > HPL> is very difficult to reach convergence. I have tried to increase or > > calculations under external field _are_ more difficult to converge. > how does your field strength compare to the value in example31? > > HPL> decrease mixing_ndim, but failed. Besides, i have tried to increase > HPL> nbnd , which seems no effect on scf convergence. Would you please > give me > HPL> some suggestion on parameters setting for scf calculation under > finite-field > HPL> ? > > it would be much easier to detect problems or make suggestions, > if you'd post the relevant part of your input file (or another > input that has the exact same problem). > > cheers, > axel. > > HPL> > HPL> Thanks for your time and attention . Any suggestion or comment is > HPL> appreciated . > HPL> > HPL> Best Wishes, > HPL> H.P > HPL> > HPL> > HPL> > HPL> > > -- > ======================================================================= > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu > Center for Molecular Modeling -- University of Pennsylvania > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 > ======================================================================= > 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 > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070706/1e94dc67/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: t03_run.sh Type: application/x-sh Size: 15050 bytes Desc: not available Url : /pipermail/attachments/20070706/1e94dc67/attachment.sh From giannozz at nest.sns.it Fri Jul 6 09:57:58 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 6 Jul 2007 09:57:58 +0200 Subject: [Pw_forum] (still) inconsistent pp.x and projwfc.x ... In-Reply-To: <41803.141.108.248.231.1183652864.squirrel@141.108.248.231> References: <41803.141.108.248.231.1183652864.squirrel@141.108.248.231> Message-ID: <05EF2E3C-3C1D-48B4-A829-561CF2E7FEC6@nest.sns.it> On Jul 5, 2007, at 18:27 , Conor Hogan wrote: > [...] that I'm completely off the mark, am running one of the pp codes > wrong, or to put my mind at ease and identify a bug the second you say, but you have a good excuse: the documentation is confusing. More exactly, it is the logic of the LSDA calculation that is confusing: different spin polarizations appear as copies of the same k-point. So if you want to plot spin-down, you have to specify kpoint=2 (in your case, with one k-point), not spin_polarization=2 (used only in the noncollinear case). The documentation has been made more explicit on this point. Thank you for pointing out this Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Fri Jul 6 10:16:21 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 6 Jul 2007 10:16:21 +0200 Subject: [Pw_forum] Strange behavior of phonon calculation In-Reply-To: <3fcc7e670707051348l2ef8e9b4yc31eae3d3f523ad3@mail.gmail.com> References: <3fcc7e670707051348l2ef8e9b4yc31eae3d3f523ad3@mail.gmail.com> Message-ID: <15C6B8B7-C4B9-470E-8B09-FFC4AB1C288C@nest.sns.it> On Jul 5, 2007, at 22:48 , Srijan Kumar Saha wrote: > My calculated total energy for a metal (spin-less) is > > total energy = -22.79040567 Ry and > corresponding gamma phonon frequency = 736 cm-1 > > > When I switch on the spin tag to specify the spin polarized > calculation for > the same sample, then I got > > total energy = -22.79040581 Ry and corresponding > gamma phonon frequency = 750 cm-1 > > while all the other input parameters are exactly same. please provide a job demostrating such behaviour Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From calderin at physics.queensu.ca Fri Jul 6 14:50:26 2007 From: calderin at physics.queensu.ca (Lazaro Calderin) Date: Fri, 06 Jul 2007 14:50:26 +0200 Subject: [Pw_forum] A bug? Message-ID: <1183726226.32552.137.camel@localhost.localdomain> Hi, There is a possible hopefully not harmful bug in pw.x. If you use the same pseudo with the upf and old format there are a couple of line at the beginning that do not report the same info and in the case of upf it is wrong. For example for a pseudo for Al with the two formats you get the differences: 35c35 < Exchange-correlation = SLA PZ NOGX NOGC (1100) --- > Exchange-correlation = LDA (1100) 52,53c52,53 < PSEUDO 1 is Al zval = 3.0 lmax= 2 lloc= 0 < (in numerical form: 1135 grid points, xmin = 0.00, dx = 0.0000) --- > PSEUDO 1 is Al zval = 3.0 lmax= 2 lloc= 2 > (in numerical form: 1135 grid points, xmin = -7.00, dx = 0.0125) In the case of the upf lloc, xmin and dx are all zero while for the old format they are reported correctly. To reproduce the bug generate the pseudo potentials using, i.e., espresso/atomic_doc/pseudo-gen/al.in and run example03 for the two pseudo potentials in the two formats. The results for the two pseudopotential are pretty much the same as compared with the reference. That is why I think this buggy is not harmful to the calculations. Regards, Lazaro -- ============================================================= Dr. Lazaro Calderin, Ph.D. Dpt. of Theoretical, Atomic and Optical Physics Univ. of Valladolid, 47005 Valladolid. Spain Phone: +34 983 423891 Fax: +34 983 423013 E-mails: calderin at fta.uva.es or calderin at physics.queensu.ca From giannozz at nest.sns.it Fri Jul 6 15:20:56 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 6 Jul 2007 15:20:56 +0200 Subject: [Pw_forum] A bug? In-Reply-To: <1183726226.32552.137.camel@localhost.localdomain> References: <1183726226.32552.137.camel@localhost.localdomain> Message-ID: <977DD94C-55EA-4544-B973-348ADB4F819C@nest.sns.it> On Jul 6, 2007, at 14:50 , Lazaro Calderin wrote: > There is a possible hopefully not harmful bug in pw.x. definitely not harmful in any way. The printout has been fixed: http://www.democritos.it:8888/O-sesame/chngview?cn=4604 Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From degironc at sissa.it Fri Jul 6 15:26:50 2007 From: degironc at sissa.it (degironc) Date: Fri, 06 Jul 2007 15:26:50 +0200 Subject: [Pw_forum] A bug? In-Reply-To: <1183726226.32552.137.camel@localhost.localdomain> References: <1183726226.32552.137.camel@localhost.localdomain> Message-ID: <468E431A.8010406@sissa.it> I agree the printout is different but this is because in the UPF format the radial grid is given as a numerical array while in the original pseudo it is given in terms of xmin, dx (and zmesh which is not reported...) the two grids are (should) be identical and all results should be exactly the same as, I understand, it is the case... maybe we should change the printout when a numerical radial grid is used (xmin and dx do not make sense anymore). Stefano de Gironcoli SISSA & DEMOCRITOS, Trieste (Italy) Lazaro Calderin wrote: > Hi, > > There is a possible hopefully not harmful bug in pw.x. > If you use the same pseudo with the upf and old format there are a > couple of line at the beginning that do not report the same info and in > the case of upf it is wrong. > > For example for a pseudo for Al with the two formats you get the > differences: > > 35c35 > < Exchange-correlation = SLA PZ NOGX NOGC (1100) > --- > >> Exchange-correlation = LDA (1100) >> > 52,53c52,53 > < PSEUDO 1 is Al zval = 3.0 lmax= 2 lloc= 0 > < (in numerical form: 1135 grid points, xmin = 0.00, dx = 0.0000) > --- > >> PSEUDO 1 is Al zval = 3.0 lmax= 2 lloc= 2 >> (in numerical form: 1135 grid points, xmin = -7.00, dx = 0.0125) >> > > In the case of the upf lloc, xmin and dx are all zero while for the old > format they are reported correctly. > > To reproduce the bug generate the pseudo potentials using, i.e., > espresso/atomic_doc/pseudo-gen/al.in and run example03 for the two > pseudo potentials in the two formats. > > The results for the two pseudopotential are pretty much the same as > compared with the reference. That is why I think this buggy is not > harmful to the calculations. > > Regards, > Lazaro > > > > From lmoreira at fc.unesp.br Fri Jul 6 20:13:18 2007 From: lmoreira at fc.unesp.br (Leandro Moreira de Campos Pinto) Date: Fri, 06 Jul 2007 15:13:18 -0300 Subject: [Pw_forum] Compilation problems Message-ID: <20070706151318.yp3cad9tokckkk88@webmail.fc.unesp.br> Dears, when I tried to compile espresso-3.2 I got the following error: ----------------------------------- ar: creating libiotk.a ar: iotk_attr+CHARACTER1_0.o: No such file or directory make[2]: ** [libiotk.a] Erro 1 make[2]: Saindo do diret?rio `/home/quimica/Desktop/Programas/espresso-3.2/iotk/src' make[1]: ** [libiotk.a] Erro 2 make[1]: Saindo do diret?rio `/home/quimica/Desktop/Programas/espresso-3.2/iotk' make: ** [libiotk] Erro 2 ------------------------------------ Could anybody help me with this??? Thanks. Leandro -- ************************************************** * Leandro Moreira de Campos Pinto * * Grupo de Eletrocat?lise e Rea??es Superficiais * * Departamento de Qu?mica * * Faculdades de Ci?ncias, UNESP, C.P. 473 * * Bauru, SP, CEP 17033-360, Brasil * * lmoreira at fc.unesp.br * * http://www.fc.unesp.br/grupo_dafc * ************************************************** ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From akohlmey at cmm.chem.upenn.edu Fri Jul 6 19:38:29 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 6 Jul 2007 13:38:29 -0400 (EDT) Subject: [Pw_forum] Compilation problems In-Reply-To: <20070706151318.yp3cad9tokckkk88@webmail.fc.unesp.br> Message-ID: On Fri, 6 Jul 2007, Leandro Moreira de Campos Pinto wrote: LM> Dears, dear leandro, LM> when I tried to compile espresso-3.2 I got the following error: please update to the latest patchlevel (3.2.2). LM> LM> ----------------------------------- LM> ar: creating libiotk.a LM> ar: iotk_attr+CHARACTER1_0.o: No such file or directory LM> make[2]: ** [libiotk.a] Erro 1 LM> make[2]: Saindo do diret?rio LM> `/home/quimica/Desktop/Programas/espresso-3.2/iotk/src' LM> make[1]: ** [libiotk.a] Erro 2 LM> make[1]: Saindo do diret?rio LM> `/home/quimica/Desktop/Programas/espresso-3.2/iotk' LM> make: ** [libiotk] Erro 2 LM> ------------------------------------ LM> LM> Could anybody help me with this??? there must have been an error before that. try 'make clean ; make >& mk.out' and have a close look at the generated file mk.out. i also recommend to turn off native language support (NLS) via setting the environment variable LC_ALL to the value of C whenever running any scientific software. a lot of code are written under the assumptions of a regular ASCII locale and thus weird things can happen. also you make your error messages less- or even unreadable by people not familiar with your local language. please have a look at http://www.democritos.it/pipermail/pw_forum/2007-April/006264.html for an example of a problem cause by use of NLS. cheers, axel. LM> LM> Thanks. LM> LM> Leandro LM> LM> LM> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From ghaderyan at gmail.com Sat Jul 7 22:53:49 2007 From: ghaderyan at gmail.com (saman ghaderyan) Date: Sun, 8 Jul 2007 00:23:49 +0330 Subject: [Pw_forum] phonon Message-ID: hi can i change K_POINTS for per q_vector to calculation phonon freq (I calculated phonon freq with deferent k-point and when i wanted to use from q2r.x this massage showed 'unknow this file ' thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070708/78d199ba/attachment.htm From ezadshojaee at hotmail.com Sun Jul 8 09:20:32 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 8 Jul 2007 07:20:32 +0000 Subject: [Pw_forum] problem with parallel calculation Message-ID: Hi trying to calculate phonon for a grid in the parallel mode, i have this error message in the second point: MPI_Recv: message truncated (rank 0, comm 9)Rank (0, MPI_COMM_WORLD): Call stack within LAM:Rank (0, MPI_COMM_WORLD): - MPI_Recv()Rank (0, MPI_COMM_WORLD): - main() how can i fix this? thanx _________________________________________________________________ With Windows Live Hotmail, you can personalize your inbox with your favorite color. www.windowslive-hotmail.com/learnmore/personalize.html?locale=en-us&ocid=TXT_TAGLM_HMWL_reten_addcolor_0607 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070708/f5159db7/attachment.htm From srijan.india at gmail.com Sun Jul 8 17:14:20 2007 From: srijan.india at gmail.com (Srijan Kumar Saha) Date: Sun, 8 Jul 2007 20:44:20 +0530 Subject: [Pw_forum] Strange behavior of phonon calculation In-Reply-To: <15C6B8B7-C4B9-470E-8B09-FFC4AB1C288C@nest.sns.it> References: <3fcc7e670707051348l2ef8e9b4yc31eae3d3f523ad3@mail.gmail.com> <15C6B8B7-C4B9-470E-8B09-FFC4AB1C288C@nest.sns.it> Message-ID: <3fcc7e670707080814k76d5ae0dv6a271f1c290fa627@mail.gmail.com> Dear pwscf users, My calculated gamma phonon frequency for Si (spin-less) is = 510 cm-1 When I switch on the spin tag to specify the spin polarized calculation for the same sample, then I got corresponding gamma phonon frequency = 214 cm-1 while all the other input parameters are exactly same. Paolo wrote: please provide a job demostrating such behaviour Dear Paolo, I'm also sending the following input file what is demostrating such behaviour by attachment. ************************************************************************************** #!/bin/sh executable_dir=./ tmp_dir=./tmp pseudo_dir=./pseudo cat > si.scf.in << EOF &control calculation='scf', restart_mode='from_scratch', prefix='si' pseudo_dir = '$pseudo_dir/', outdir='$tmp_dir/' / &system ibrav = 2, celldm(1) =10.20, nat= 2, ntyp= 1, ecutwfc = 18.0, nelec = 8.00, nbnd =10 nspin=2, tot_magnetization=0, occupations='smearing', smearing='fd', degauss=0.005, / &electrons mixing_beta = 0.7 conv_thr = 1.0d-8 / ATOMIC_SPECIES Si 28.086 Si.vbc.UPF ATOMIC_POSITIONS Si 0.00 0.00 0.00 Si 0.25 0.25 0.25 K_POINTS 10 0.1250000 0.1250000 0.1250000 1.00 0.1250000 0.1250000 0.3750000 3.00 0.1250000 0.1250000 0.6250000 3.00 0.1250000 0.1250000 0.8750000 3.00 0.1250000 0.3750000 0.3750000 3.00 0.1250000 0.3750000 0.6250000 6.00 0.1250000 0.3750000 0.8750000 6.00 0.1250000 0.6250000 0.6250000 3.00 0.3750000 0.3750000 0.3750000 1.00 0.3750000 0.3750000 0.6250000 3.00 EOF $ECHO " running the scf calculation...\c" $executable_dir/pw.x < si.scf.in > si.scf.out $ECHO " done" # phonon calculation at Gamma cat > si.phG.in << EOF phonons of Si at Gamma &inputph tr2_ph=1.0d-14, prefix='si', amass(1)=28.08, outdir='$tmp_dir/', fildyn='si.dynG', / 0.0 0.0 0.0 EOF $ECHO " running the phonon calculation at Gamma...\c" $executable_dir/ph.x < si.phG.in > si.phG.out $ECHO " done" ******************************************************************************** To run the spin less calculation I just remove tot_magnetization tag and reset nspin=1 and I get correct Gamma point phonon frequency (510 cm-1) whereas above input file for spin calculation will give us a completely wrong result (214 cm-1). Looking forward to your valuable suggestions. With best regards, Srijan -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070708/5a92de5b/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: run Type: application/octet-stream Size: 1444 bytes Desc: not available Url : /pipermail/attachments/20070708/5a92de5b/attachment.obj From akohlmey at cmm.chem.upenn.edu Sun Jul 8 21:02:30 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 8 Jul 2007 15:02:30 -0400 (EDT) Subject: [Pw_forum] problem with parallel calculation In-Reply-To: Message-ID: On Sun, 8 Jul 2007, Ezad Shojaee wrote: ES> ES> Hi ES> trying to calculate phonon for a grid in the parallel mode, i have ES> this error message in the second point: ES> MPI_Recv: message truncated (rank 0, comm 9)Rank (0, ES> MPI_COMM_WORLD): Call stack within LAM:Rank (0, MPI_COMM_WORLD): - ES> MPI_Recv()Rank (0, MPI_COMM_WORLD): - main() ES> how can i fix this? hi ezad, i don't know the phonon code much, but from this little piece of information, probably nobody can tell what is wrong. this error message comes from the underlying MPI library, so please let us first know, whether you can run any MPI program successfully. if yes, please follow the usual advice for bug/problem reports with any software, i.e., provide some of the text leading up to this and afterwards, if any (so that people can locate where it crashes in the code), what type and version of machine/operating system/compiler/MPI/quantum espresso you are using and ideally also the input. since your report is about a parallel execution, please indicate, whether your input would run in serial, as sometimes errors in the input can show up as MPI errors. cheers, axel. ES> thanx ES> _________________________________________________________________ ES> With Windows Live Hotmail, you can personalize your inbox with your favorite color. ES> www.windowslive-hotmail.com/learnmore/personalize.html?locale=en-us&ocid=TXT_TAGLM_HMWL_reten_addcolor_0607 -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From mbaris at metu.edu.tr Mon Jul 9 07:16:36 2007 From: mbaris at metu.edu.tr (mbaris at metu.edu.tr) Date: Mon, 09 Jul 2007 08:16:36 +0300 Subject: [Pw_forum] problem with parallel calculation In-Reply-To: References: Message-ID: <20070709081636.hwod4z98tgbkkso4@webmail2.metu.edu.tr> Dear Ezad, > MPI_Recv: message truncated (rank 0, comm 9)Rank (0, > MPI_COMM_WORLD): Call stack within LAM:Rank (0, MPI_COMM_WORLD): - > MPI_Recv()Rank (0, MPI_COMM_WORLD): - main() > how can i fix this? > This indicates that your main node received a message with different size than expected (from node 9?), which shouldn't happen. As I can see, you are using LAM. It is hard to locate the exact source of this kind of error (especially when using LAM), but here are some quick tips: 1) Are you sure your nodes are initialized properly, are there anything in your nodes (firewall etc.) that may be appending extra output? 2) Are you sure the ports LAM use are assigned solely for LAM (ex. look for /etc/services) 3) Are all nodes executing the same code (i.e. when you are not using shared storage)? 4) Are all your nodes homogeneous? (i.e. same kernel,glibc,libraries similar hardware) 5) One of your daemons may have died prematurely, check for cpu quotas, or your particular calculation may have caused something like a segmentation fault in one of the nodes. Hope this helps, O. Baris Malcioglu METU Ankara From dalcorso at sissa.it Mon Jul 9 09:03:18 2007 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Mon, 09 Jul 2007 09:03:18 +0200 Subject: [Pw_forum] Strange behavior of phonon calculation In-Reply-To: <3fcc7e670707080814k76d5ae0dv6a271f1c290fa627@mail.gmail.com> References: <3fcc7e670707051348l2ef8e9b4yc31eae3d3f523ad3@mail.gmail.com> <15C6B8B7-C4B9-470E-8B09-FFC4AB1C288C@nest.sns.it> <3fcc7e670707080814k76d5ae0dv6a271f1c290fa627@mail.gmail.com> Message-ID: <1183964598.3283.4.camel@dhpc-5-02.sissa.it> On Sun, 2007-07-08 at 20:44 +0530, Srijan Kumar Saha wrote: > > > Dear pwscf users, > > My calculated gamma phonon frequency for Si (spin-less) is > = 510 cm-1 When I switch on the spin tag to specify the spin > polarized > calculation for the same sample, then I got corresponding > gamma phonon frequency = 214 cm-1 > while all the other input parameters are exactly same. > > > Paolo wrote: > please provide a job demostrating such behaviour > > Dear Paolo, > I'm also sending the following input file what is demostrating such > behaviour > by attachment. > > ************************************************************************************** > #!/bin/sh > executable_dir=./ > tmp_dir=./tmp > pseudo_dir=./pseudo > cat > si.scf.in << EOF > &control > calculation='scf', > restart_mode='from_scratch', > prefix='si' > pseudo_dir = '$pseudo_dir/', > outdir='$tmp_dir/' > / > &system > ibrav = 2, celldm(1) =10.20, nat= 2, ntyp= 1, > ecutwfc = 18.0, nelec = 8.00, nbnd =10 > nspin=2, tot_magnetization=0, The phonon code does not work with constrained magnetization. Nobody programmed this part so far. Actually in this example the flag tot_magnetization=0 is not needed because Si is non magnetic. If you remove the flag the phonon code should work also with nspin=2. Hope this helps, Andrea > occupations='smearing', smearing='fd', degauss=0.005, > / > &electrons > mixing_beta = 0.7 > conv_thr = 1.0d-8 > / > ATOMIC_SPECIES > Si 28.086 Si.vbc.UPF > ATOMIC_POSITIONS > Si 0.00 0.00 0.00 > Si 0.25 0.25 0.25 > K_POINTS > 10 > 0.1250000 0.1250000 0.1250000 1.00 > 0.1250000 0.1250000 0.3750000 3.00 > 0.1250000 0.1250000 0.6250000 3.00 > 0.1250000 0.1250000 0.8750000 3.00 > 0.1250000 0.3750000 0.3750000 3.00 > 0.1250000 0.3750000 0.6250000 6.00 > 0.1250000 0.3750000 0.8750000 6.00 > 0.1250000 0.6250000 0.6250000 3.00 > 0.3750000 0.3750000 0.3750000 1.00 > 0.3750000 0.3750000 0.6250000 3.00 > EOF > $ECHO " running the scf calculation...\c" > $executable_dir/pw.x < si.scf.in > si.scf.out > $ECHO " done" > > # phonon calculation at Gamma > cat > si.phG.in << EOF > phonons of Si at Gamma > &inputph > tr2_ph=1.0d-14, > prefix='si', > amass(1)=28.08, > outdir='$tmp_dir/', > fildyn='si.dynG', > / > 0.0 0.0 0.0 > EOF > $ECHO " running the phonon calculation at Gamma...\c" > $executable_dir/ph.x < si.phG.in > si.phG.out > $ECHO " done" > > > ******************************************************************************** > To run the spin less calculation I just remove tot_magnetization > tag > and reset nspin=1 and I get correct Gamma point > phonon frequency (510 cm-1) whereas above input file for spin > calculation > will give us a completely wrong result (214 cm-1). > > Looking forward to your valuable suggestions. > > With best regards, > Srijan -- Andrea Dal Corso Tel. 0039-040-3787428 SISSA, Via Beirut 2/4 Fax. 0039-040-3787528 34014 Trieste (Italy) e-mail: dalcorso at sissa.it From helen at fh.huji.ac.il Mon Jul 9 13:31:51 2007 From: helen at fh.huji.ac.il (Helen) Date: Mon, 9 Jul 2007 14:31:51 +0300 Subject: [Pw_forum] modified explicit exchange potential Message-ID: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> Thank you Paolo and Lorenzo to the replies to my previous e-mail. I think I didn't explain what I wanted to modify clearly enough. We modify the explicit exchange to include a descreened potential: K_x[n]=-0.25 Int[ P[n](r,r')^2 * u(r-r') ] d3r d3r' where P[n](r,r') is the density matrix of non-interacting particles and u(r-r') is the descreened interaction (normally the coulomb interaction u(r)=1/r, but in our case a different function). Which files do I need to change for the modified K_x and its potential? Is it exx.f90 in PW? These explicit exchanges are also used in Hartree-Fock and B3LYP calculations. Thank you for your help, Helen Eisenberg -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070709/d6a96917/attachment.htm From degironc at sissa.it Mon Jul 9 14:47:08 2007 From: degironc at sissa.it (degironc) Date: Mon, 09 Jul 2007 14:47:08 +0200 Subject: [Pw_forum] modified explicit exchange potential In-Reply-To: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> References: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> Message-ID: <46922E4C.10305@sissa.it> Helen wrote: > Thank you Paolo and Lorenzo to the replies to my previous e-mail. I > think I didn't explain what I wanted to modify clearly enough. We > modify the explicit exchange to include a descreened potential: > K_x[n]=-0.25 Int[ P[n](r,r')^2 * u(r-r') ] d3r d3r' > where P[n](r,r') is the density matrix of non-interacting particles > and u(r-r') is the descreened interaction (normally the coulomb > interaction u(r)=1/r, but in our case a different function). > > Which files do I need to change for the modified K_x and its > potential? Is it exx.f90 in PW? These explicit exchanges are also used > in Hartree-Fock and B3LYP calculations. > > Thank you for your help, > Helen Eisenberg Yes, the part of the code performing HF-like exchange is in exx.f90. You need to compile the code specifying -DEXX at compilation time... HF, B3LYP and PBE0 are implemented (but no corresponding pseudopotential are available so one need to use some PBE or LDA one and overwite the XC by the input_dft flag in the system namelist) ... Beware: it is extreeeemely slow. As you wll see the EXX part requires some work to be made more useful... if you are interested in working in that line let me know... :-) .. by the way, full affiliation would be appreciated. Stefano de Gironcoli SISSA and DEMOCRITOS, Trieste, Italy From lanhaiping at gmail.com Mon Jul 9 17:59:05 2007 From: lanhaiping at gmail.com (lan haiping) Date: Mon, 9 Jul 2007 23:59:05 +0800 Subject: [Pw_forum] bfgs crash for relax calculation under finite-electric fieldcaclulation Message-ID: Hi, All I have come to bfgs crash with error information below when i did relax calculation of graphene system under finite-homogeneous-electric field. The finite-homogeneous-field is perpendicular to the graphene plane with E=0.001 a.u.. I just want to perform calculation of static dielectric properties of this 2-D system. " trust_radius < trust_radius_min resetting bfgs history %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= %% from bfgs : error # 1 bfgs history already reset at previous step %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= %%%%% stopping ... *************************************************************************= ****** " I checked mailing-list several times , and didnot find the related solutions, would you please give me some suggestion and comments ? Best Wishes, H.P -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070709/5365682b/attachment.htm From degironc at sissa.it Mon Jul 9 18:19:57 2007 From: degironc at sissa.it (degironc) Date: Mon, 09 Jul 2007 18:19:57 +0200 Subject: [Pw_forum] bfgs crash for relax calculation under finite-electric fieldcaclulation In-Reply-To: References: Message-ID: <4692602D.1040007@sissa.it> This means that the calculation cannot proceeds further because, I guess, the required threshold is below the noise level of the energy and forces you compute. In order to reduce the noise you could increase the scf threshold and/or increase the ecutrho to improve the accuracy of XC energy calculation. Are you sure the your relaxation has not already converged ? stefano an haiping wrote: > Hi, All > > I have come to bfgs crash with error information below when i did > relax calculation of graphene system under finite-homogeneous-electric > field. The finite-homogeneous-field is perpendicular to the graphene > plane with E= 0.001 a.u.. > I just want to perform calculation of static dielectric properties of > this 2-D system. > " > trust_radius < trust_radius_min > > resetting bfgs history > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= > > %% > > from bfgs : error # 1 > > bfgs history already reset at previous step > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= > %%%%% > > stopping ... > > *************************************************************************= > > ****** > " > > I checked mailing-list several times , and didnot find the related > solutions, would you please give me some suggestion and comments ? > > Best Wishes, > > H.P > > > -- > Hai-Ping Lan > Department of Electronics , > Peking University , Bejing, 100871 > lanhaiping at gmail.com , hplan at pku.edu.cn > From eyvaz_isaev at yahoo.com Tue Jul 10 00:43:42 2007 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 9 Jul 2007 15:43:42 -0700 (PDT) Subject: [Pw_forum] phonon In-Reply-To: Message-ID: <75102.25391.qm@web60325.mail.yahoo.com> Dear Saman, First of all, please do not forget your affiliation. And then can you please explain more precisely what is your problem. Bests, Eyvaz. --- saman ghaderyan wrote: > hi > can i change K_POINTS for per q_vector to > calculation phonon freq > (I calculated phonon freq with deferent k-point and > when i wanted to use > from q2r.x this massage showed 'unknow this file ' > thanks > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, and Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC From lanhaiping at gmail.com Tue Jul 10 10:43:20 2007 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 10 Jul 2007 16:43:20 +0800 Subject: [Pw_forum] bfgs crash for relax calculation under finite-electric fieldcaclulation In-Reply-To: <4692602D.1040007@sissa.it> References: <4692602D.1040007@sissa.it> Message-ID: Thank you, Stefano. I have rechecked my system, and found there was a mistake of atomic coordinates in my script file . Sorry for my mind-absent . Regards, H.P On 7/10/07, degironc wrote: > > This means that the calculation cannot proceeds further because, I > guess, the required threshold is below the noise level of the energy and > forces you compute. > In order to reduce the noise you could increase the scf threshold and/or > increase the ecutrho to improve the accuracy of XC energy calculation. > Are you sure the your relaxation has not already converged ? > stefano > > an haiping wrote: > > Hi, All > > > > I have come to bfgs crash with error information below when i did > > relax calculation of graphene system under finite-homogeneous-electric > > field. The finite-homogeneous-field is perpendicular to the graphene > > plane with E= 0.001 a.u.. > > I just want to perform calculation of static dielectric properties of > > this 2-D system. > > " > > trust_radius < trust_radius_min > > > > resetting bfgs history > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= > > > > %% > > > > from bfgs : error # 1 > > > > bfgs history already reset at previous step > > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%= > > %%%%% > > > > stopping ... > > > > > *************************************************************************= > > > > ****** > > " > > > > I checked mailing-list several times , and didnot find the related > > solutions, would you please give me some suggestion and comments ? > > > > Best Wishes, > > > > H.P > > > > > > -- > > Hai-Ping Lan > > Department of Electronics , > > Peking University , Bejing, 100871 > > lanhaiping at gmail.com , hplan at pku.edu.cn > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070710/86d8f7dc/attachment.htm From lanhaiping at gmail.com Tue Jul 10 12:03:33 2007 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 10 Jul 2007 18:03:33 +0800 Subject: [Pw_forum] convergence problem of graphene system under homogeneous finite-field. Message-ID: Hi. Everyone. I am performing some test works on graphene system with electric-enthalpy approach which was developed by Umari & Souza, to examine its dielectric properties. My calculation came to scf convergence problem when i applied E=0.001 a.u.that is perpendicular to the graphene plane, the total-energy of this system oscillates up and down as shown below : " total energy = -22.10984675 Ry total energy = -31.05580980 Ry total energy = -21.97194327 Ry total energy = -22.49028278 Ry total energy = -22.62216901 Ry total energy = -23.00354121 Ry total energy = -24.57862186 Ry total energy = -21.98569845 Ry total energy = -24.92584359 Ry total energy = -24.96325979 Ry total energy = -22.09410112 Ry total energy = -32.84912151 Ry total energy = -22.00923042 Ry total energy = -27.67266074 Ry total energy = -29.60728014 Ry total energy = -21.49630504 Ry total energy = -23.72936622 Ry total energy = -23.50209359 Ry total energy = -22.52577866 Ry " Would anyone please give me some suggestion or comments ,et al ? My running script is given below. Best Wishes, H.P graphene.scf.efield2.in " &control calculation='scf' restart_mode='restart', prefix='graphene-SL', gdir=3, nppstr=1, lelfield=.true., nberrycyc=6 pseudo_dir='/home/haiping/espresso-3.2/pseudo', outdir='/home/haiping/tmp' / &system ibrav= 4, celldm(1)=4.648726267482302, celldm(3)= 5, nat= 2, ntyp= 1, ecutwfc = 30.0, nosym=.true. / &electrons diagonalization='cg', conv_thr = 1.0d-8, mixing_beta = 0.01, startingwfc='random', efield=0.001, / ATOMIC_SPECIES C 12.0107 C.pz-rrkjus.UPF ATOMIC_POSITIONS C 0.0000 0.000000 0.0 C 0.3333 0.666667 0.0 K_POINTS 16 0.0 0.0 0.0 1 0.0 0.25 0.0 1 0.0 0.50 0.0 1 0.0 0.75 0.0 1 0.25 0.0 0.0 1 0.25 0.25 0.0 1 0.25 0.50 0.0 1 0.25 0.75 0.0 1 0.50 0.0 0.0 1 0.50 0.25 0.0 1 0.50 0.50 0.0 1 0.50 0.75 0.0 1 0.75 0.0 0.0 1 0.75 0.25 0.0 1 0.75 0.50 0.0 1 0.75 0.75 0.0 1 " -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070710/f0e11ce0/attachment.htm From ezadshojaee at hotmail.com Wed Jul 11 12:14:11 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Wed, 11 Jul 2007 10:14:11 +0000 Subject: [Pw_forum] k_point generation in ph.x for a grid Message-ID: Hi i have two problems, the first is a little bit silly! when we want to calculate phonons for a grid, the code generates k_points for each point in the grid which half of them have no weight ( except for the Gamma point) what's the point? and the second: the code generates different # of k_points for a same point of grid in the serial mode and the parallel mode with all other of parameters the same. what is this about? _________________________________________________________________ With Windows Live Hotmail, you can personalize your inbox with your favorite color. www.windowslive-hotmail.com/learnmore/personalize.html?locale=en-us&ocid=TXT_TAGLM_HMWL_reten_addcolor_0607 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070711/007cf565/attachment.html From giannozz at nest.sns.it Wed Jul 11 12:42:36 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 11 Jul 2007 12:42:36 +0200 Subject: [Pw_forum] k_point generation in ph.x for a grid In-Reply-To: References: Message-ID: <1EB77D52-A9A3-4E6A-81B0-B3F50C19C814@nest.sns.it> On Jul 11, 2007, at 12:14 , Ezad Shojaee wrote: > when we want to calculate phonons for a grid, the code generates > k_points > for each point in the grid which half of them have no weight > ( except for the > Gamma point) what's the point? for finite q one needs both \psi_k and \psi_k+q . The latter are those with zero weight (this was useful in some cases but maybe it is obsolete) > the code generates different # of k_points for a same point of grid > in the > serial mode and the parallel mode with all other of parameters the > same no it doesn't Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From shruba at gmail.com Wed Jul 11 17:31:57 2007 From: shruba at gmail.com (shruba at gmail.com) Date: Wed, 11 Jul 2007 11:31:57 -0400 Subject: [Pw_forum] pseudopotentical from dacapo Message-ID: <822f4ec80707110831x4f0cadcer7d05181a17e10474@mail.gmail.com> Hi all, I have a question regarding ultrasoft pseudopotential ,can I use dacapo pseudopotential in pwscf. if yes how can I convert binary files to pwscf format It will be great if you can refer me some other sources for vanderbilt ultrasoft pseudopotential, I looked at DACAPO website and found they have more pseudopotentials than vanderbilt ultrasoft pseudopotential site. So I am wondering if there are more sources. Thanks in advance shruba -- shruba gangopadhyay graduate student department of chemistry university of central florida orlando, FL-32826 'friendship doubles joys and reduces sorrows by half' (Francis Bacon). -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070711/b80305ac/attachment.htm From giannozz at nest.sns.it Wed Jul 11 17:52:38 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 11 Jul 2007 17:52:38 +0200 Subject: [Pw_forum] pseudopotentical from dacapo In-Reply-To: <822f4ec80707110831x4f0cadcer7d05181a17e10474@mail.gmail.com> References: <822f4ec80707110831x4f0cadcer7d05181a17e10474@mail.gmail.com> Message-ID: On Jul 11, 2007, at 17:31 , shruba at gmail.com wrote: > can I use dacapo pseudopotential in pwscf. yes you can > if yes how can I convert binary files to pwscf format uspp2upf.x converts binary files. Those in dacapo repository are big-endian. Set compiler options for uspp2upf.x so that it reads big-endian files: http://www.democritos.it/pipermail/pw_forum/2007-June/006813.html http://www.democritos.it/pipermail/pw_forum/2007-June/006823.html Other possibility: use the "reform.x" utility http://www.democritos.it/pipermail/pw_forum/2007-June/006816.html If you do not understand what we are talking about, please read some documentation or inquire with somebody knowledgeable at your university. Asking the same question once again will waste your time and ours. Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From ghaderyan at gmail.com Wed Jul 11 23:14:53 2007 From: ghaderyan at gmail.com (saman ghaderyan) Date: Thu, 12 Jul 2007 00:44:53 +0330 Subject: [Pw_forum] phonon Message-ID: thanks and sorry for my older mail, i don't know how the k-points accounted for each q-vector (for example k-points {automatic} 10 10 10 0 0 0 for q-vector 0 0 0 47mesh and for 0.5 0.5 1, 1120 mesh ) ?? i want to use deferent K-points for any q_vector (manually )for achieve phonon frequency. MS saman ghaderyan Buali sina university, Iran From eyvaz_isaev at yahoo.com Thu Jul 12 10:46:23 2007 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Thu, 12 Jul 2007 01:46:23 -0700 (PDT) Subject: [Pw_forum] phonon In-Reply-To: Message-ID: <254747.36078.qm@web60323.mail.yahoo.com> Dear Saman, > i don't know how the k-points accounted for each > q-vector (for example k-points {automatic} 10 10 > 10 0 0 0 for q-vector 0 0 0 47mesh and for > 0.5 0.5 1, 1120 mesh ) ?? My apologies, what do you mean 47mesh and 1120 mesh? Usually q-vectors are used to calculate dynamical matrices which are then converted to the force constant matrices by means of Fast Fourier Transformation. Number of q-points at this stage has to be not too large, around 10-30 depending on a system considered. k-points are used for self-consistent calculations and there is no connection between these q-, and k- point meshes. > i want to use deferent K-points for any q_vector > (manually )for achieve phonon frequency. What is a reason to use different k-points set for different q-points? Do you like to perform a convergency test over k-points? If so, you can do 1. scf (pw.x) 2. nscf for a q-point (pw.x) 3. phonon(ph.x) Hope it helps. Bests, Eyvaz. > MS saman ghaderyan > Buali sina university, Iran > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ------------------------------------------------------------------- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, and Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, eyvaz_isaev at yahoo.com ____________________________________________________________________________________ Now that's room service! Choose from over 150,000 hotels in 45,000 destinations on Yahoo! Travel to find your fit. http://farechase.yahoo.com/promo-generic-14795097 From helen at fh.huji.ac.il Thu Jul 12 13:40:58 2007 From: helen at fh.huji.ac.il (Helen) Date: Thu, 12 Jul 2007 14:40:58 +0300 Subject: [Pw_forum] modified explicit exchange potential References: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> <46922E4C.10305@sissa.it> Message-ID: <000801c7c479$83be1d30$cd604084@fh.huji.ac.il> Hi, I have a few questions regarding exx: 1. I was reading in the examples the README file for EXX_example, that when calculating the band structure you have to use the same k-points as are contained in the scf calculation. Currently I want to calculate the band structure in Si along the 100 axis, similar to example01 (though using the different functional for exx which I need to implement). Can I just use the standard scf input k-points and tag the k-points I need at the end (appropriately weighted) in the scf input file? If so what weighting will they need? e.g K_POINTS 20 0.1250000 0.1250000 0.1250000 1.00 0.1250000 0.1250000 0.3750000 3.00 0.1250000 0.1250000 0.6250000 3.00 0.1250000 0.1250000 0.8750000 3.00 0.1250000 0.3750000 0.3750000 3.00 0.1250000 0.3750000 0.6250000 6.00 0.1250000 0.3750000 0.8750000 6.00 0.1250000 0.6250000 0.6250000 3.00 0.3750000 0.3750000 0.3750000 1.00 0.0 0.0 0.0 1.0? 0.0 0.0 0.1 1.0? 0.0 0.0 0.2 1.0? 0.0 0.0 0.3 1.0? 0.0 0.0 0.4 1.0? 0.0 0.0 0.5 1.0? 0.0 0.0 0.6 1.0? 0.0 0.0 0.7 1.0? 0.0 0.0 0.8 1.0? 0.0 0.0 0.9 1.0? 0.0 0.0 1.0 1.0? 2. In README it states that the error involved in discarding the q=0 term goes as 1/(NQS*Omega). This fact is utilized in removing the error. Will this still be valid if the e- interaction is not U(r)=1/r but U(r)=efr(gam r)/r ? [Fourier Transform=exp(-q^2/4*gam^2)*4Pi/q^2, so divergence as q->0 is same as U(r)=1/r, so I think they should be the same) 3. Is there anywhere in exx.f90 where the interaction is defined and then used throughout the code. i.e is there a specific place in the code where I can change the interaction, or do I need to look through the whole code? Thanks so much for your help, Dr.Helen Eisenberg Post-doctoral researcher, The Fritz Haber Center for Molecular Dynamics, The Hebrew University of Jerusalem, Jerusalem 91904 Israel. ----- Original Message ----- From: "degironc" To: Sent: Monday, July 09, 2007 3:47 PM Subject: Re: [Pw_forum] modified explicit exchange potential > Helen wrote: >> Thank you Paolo and Lorenzo to the replies to my previous e-mail. I think >> I didn't explain what I wanted to modify clearly enough. We modify the >> explicit exchange to include a descreened potential: >> K_x[n]=-0.25 Int[ P[n](r,r')^2 * u(r-r') ] d3r d3r' >> where P[n](r,r') is the density matrix of non-interacting particles and >> u(r-r') is the descreened interaction (normally the coulomb interaction >> u(r)=1/r, but in our case a different function). >> Which files do I need to change for the modified K_x and its potential? >> Is it exx.f90 in PW? These explicit exchanges are also used in >> Hartree-Fock and B3LYP calculations. >> Thank you for your help, >> Helen Eisenberg > Yes, the part of the code performing HF-like exchange is in exx.f90. > You need to compile the code specifying -DEXX at compilation time... HF, > B3LYP and PBE0 are implemented (but no corresponding pseudopotential are > available so one need to use some PBE or LDA one and overwite the XC by > the input_dft flag in the system namelist) ... > Beware: it is extreeeemely slow. > > As you wll see the EXX part requires some work to be made more useful... > if you are interested in working in that line let me know... :-) .. by the > way, full affiliation would be appreciated. > > Stefano de Gironcoli > SISSA and DEMOCRITOS, Trieste, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From shanavas at barc.gov.in Fri Jul 13 14:38:08 2007 From: shanavas at barc.gov.in (Shanavas K. V.) Date: Fri, 13 Jul 2007 18:08:08 +0530 Subject: [Pw_forum] Fermi surface nesting function Message-ID: <1184330288.4608.15.camel@athena.purnima> Hi all, I am trying to write a code to calculate fermi surface nesting function for an AFM system. Is there any code already available for it? I found a paper http://arxiv.org/pdf/cond-mat/0504077 using the NF, but can somebody guide me to some paper that explain the basic theory & method that can help in writing the code? Thanks in advance, regards, shanavas Scientist, Bhabha Atomic Research Centre, Mumbai, India From njuxuyuehua at gmail.com Fri Jul 13 10:27:54 2007 From: njuxuyuehua at gmail.com (xu yuehua) Date: Fri, 13 Jul 2007 16:27:54 +0800 Subject: [Pw_forum] relax question Message-ID: hello all i am doing relax. in my inputfile, i use etot_conv_thr=1.D-6 forc_conv_thr=1.D-6 and the output is atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 atom 2 type 1 force = 0.00000081 -0.00000108 0.00000033 atom 3 type 1 force = -0.00000435 -0.00000425 -0.00000160 atom 4 type 1 force = -0.00000312 -0.00000321 -0.00000080 atom 5 type 1 force = -0.00000446 -0.00000036 -0.00000273 atom 6 type 1 force = -0.00000048 0.00000230 0.00000112 atom 7 type 1 force = 0.00000164 0.00000118 0.00000060 atom 8 type 1 force = 0.00000131 -0.00000040 0.00000145 atom 9 type 2 force = -0.00000362 0.00000154 -0.00000201 atom 10 type 2 force = -0.00000029 0.00000191 0.00000302 atom 11 type 2 force = 0.00000576 -0.00000134 0.00000194 atom 12 type 2 force = 0.00000390 0.00000276 0.00000233 Total force = 0.000015 Total SCF correction = 0.000054 SCF correction compared to forces is too large, reduce conv_thr The maximum number of steps has been reached. my qustion is ,force at every atom is ok ,but why the program asks to reduce conv_thr my conv_thr is not enough small ? or my total relax step (50) is not adequate? is there any principle about these parameter? need your help thanks a lot -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070713/b1791c59/attachment.htm From lanhaiping at gmail.com Fri Jul 13 10:52:50 2007 From: lanhaiping at gmail.com (lan haiping) Date: Fri, 13 Jul 2007 16:52:50 +0800 Subject: [Pw_forum] relax question In-Reply-To: References: Message-ID: Dear yuehua i think your forc_conv_thr=1.D-6 is too tight for your calculation .. Usually, For most calculation 1.D-4 is very fine.. On 7/13/07, xu yuehua wrote: > > hello all > i am doing relax. > in my inputfile, > i use etot_conv_thr=1.D-6 > forc_conv_thr=1.D-6 > and the output is > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > atom 2 type 1 force = 0.00000081 -0.00000108 0.00000033 > atom 3 type 1 force = -0.00000435 -0.00000425 - > 0.00000160 > atom 4 type 1 force = -0.00000312 -0.00000321 -0.00000080 > atom 5 type 1 force = -0.00000446 -0.00000036 -0.00000273 > atom 6 type 1 force = -0.00000048 0.00000230 > 0.00000112 > atom 7 type 1 force = 0.00000164 0.00000118 0.00000060 > atom 8 type 1 force = 0.00000131 -0.00000040 0.00000145 > atom 9 type 2 force = -0.00000362 0.00000154 -0.00000201 > atom 10 type 2 force = -0.00000029 0.00000191 0.00000302 > atom 11 type 2 force = 0.00000576 -0.00000134 0.00000194 > atom 12 type 2 force = 0.00000390 0.00000276 > 0.00000233 > > Total force = 0.000015 Total SCF correction = 0.000054 > SCF correction compared to forces is too large, reduce conv_thr > > The maximum number of steps has been reached. > my qustion is ,force at every atom is ok ,but why the program asks to > reduce conv_thr > my conv_thr is not enough small ? > or my total relax step (50) is not adequate? > is there any principle about these parameter? > need your help > thanks a lot > -- > Xu Yuehua > physics Department of Nanjing university > China > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070713/bb46f38d/attachment.htm From dingxunlei at gmail.com Fri Jul 13 08:12:56 2007 From: dingxunlei at gmail.com (Ding Xunlei) Date: Fri, 13 Jul 2007 14:12:56 +0800 Subject: [Pw_forum] relax question Message-ID: <46976bcc.201f5e0a.6fd2.fffffa01@mx.google.com> Dear Yuehua, The default number is : etot_conv_thr=1.D-4 forc_conv_thr=1.D-3 So, 1. Your setting for forc_conv_thr=1.D-6 is too small. 2. If you do need to set force like that, you need to set etot_conv_thr even smaller, saying 1.D-7. And also, you need to set conv_thr (for scf) smaller than etot_conv_thr, saying 1.D-8. Hope it help. Yours sincerely, Ding 2007-07-13 ______________________________________________ ======= 2007-07-13 16:27:54 You have written: ======= >hello all >i am doing relax. >in my inputfile, >i use etot_conv_thr=1.D-6 > forc_conv_thr=1.D-6 >and the output is > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > atom 2 type 1 force = 0.00000081 -0.00000108 0.00000033 > atom 3 type 1 force = -0.00000435 -0.00000425 -0.00000160 > atom 4 type 1 force = -0.00000312 -0.00000321 -0.00000080 > atom 5 type 1 force = -0.00000446 -0.00000036 -0.00000273 > atom 6 type 1 force = -0.00000048 0.00000230 0.00000112 > atom 7 type 1 force = 0.00000164 0.00000118 0.00000060 > atom 8 type 1 force = 0.00000131 -0.00000040 0.00000145 > atom 9 type 2 force = -0.00000362 0.00000154 -0.00000201 > atom 10 type 2 force = -0.00000029 0.00000191 0.00000302 > atom 11 type 2 force = 0.00000576 -0.00000134 0.00000194 > atom 12 type 2 force = 0.00000390 0.00000276 0.00000233 > >Total force = 0.000015 Total SCF correction = 0.000054 > SCF correction compared to forces is too large, reduce conv_thr > > The maximum number of steps has been reached. >my qustion is ,force at every atom is ok ,but why the program asks to >reduce conv_thr >my conv_thr is not enough small ? >or my total relax step (50) is not adequate? >is there any principle about these parameter? >need your help >thanks a lot >-- >Xu Yuehua >physics Department of Nanjing university >China > >_______________________________________________ >Pw_forum mailing list >Pw_forum at pwscf.org >http://www.democritos.it/mailman/listinfo/pw_forum > = = = = = = = = = = = = = = = = = = = = From njuxuyuehua at gmail.com Fri Jul 13 15:44:59 2007 From: njuxuyuehua at gmail.com (xu yuehua) Date: Fri, 13 Jul 2007 21:44:59 +0800 Subject: [Pw_forum] relax question In-Reply-To: <46976bcc.201f5e0a.6fd2.fffffa01@mx.google.com> References: <46976bcc.201f5e0a.6fd2.fffffa01@mx.google.com> Message-ID: thanks for above Mr Ding and Mr Lan answers. and another related question is whether conv_thr is in &ELECTRONS? is it about scf ? i don not understand how a flow chart when relaxing ions and its details? hope your kind help 2007/7/13, Ding Xunlei : > > Dear Yuehua, > > The default number is : > etot_conv_thr=1.D-4 > forc_conv_thr=1.D-3 > > So, 1. Your setting for forc_conv_thr= 1.D-6 is too small. > 2. If you do need to set force like that, you need to set > etot_conv_thr even smaller, saying 1.D-7. And also, you need to set TRUE > smaller than etot_conv_thr, saying 1.D-8. > > Hope it help. > > Yours sincerely, > Ding > 2007-07-13 > ______________________________________________ > > ======= 2007-07-13 16:27:54 You have written: ======= > > >hello all > >i am doing relax. > >in my inputfile, > >i use etot_conv_thr=1.D-6 > > forc_conv_thr=1.D-6 > >and the output is > > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > > atom 2 type 1 force = 0.00000081 -0.00000108 > 0.00000033 > > atom 3 type 1 force = -0.00000435 -0.00000425 - > 0.00000160 > > atom 4 type 1 force = -0.00000312 -0.00000321 - > 0.00000080 > > atom 5 type 1 force = - 0.00000446 -0.00000036 - > 0.00000273 > > atom 6 type 1 force = -0.00000048 0.00000230 > 0.00000112 > > atom 7 type 1 force = 0.00000164 0.00000118 > 0.00000060 > > atom 8 type 1 force = 0.00000131 -0.00000040 > 0.00000145 > > atom 9 type 2 force = -0.00000362 0.00000154 - > 0.00000201 > > atom 10 type 2 force = -0.00000029 0.00000191 > 0.00000302 > > atom 11 type 2 force = 0.00000576 -0.00000134 > 0.00000194 > > atom 12 type 2 force = 0.00000390 0.00000276 > 0.00000233 > > > >Total force = 0.000015 Total SCF correction = 0.000054 > > SCF correction compared to forces is too large, reduce conv_thr > > > > The maximum number of steps has been reached. > >my qustion is ,force at every atom is ok ,but why the program asks to > >reduce conv_thr > >my conv_thr is not enough small ? > >or my total relax step (50) is not adequate? > >is there any principle about these parameter? > >need your help > >thanks a lot > >-- > >Xu Yuehua > >physics Department of Nanjing university > >China > > > >_______________________________________________ > >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 > -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070713/72a067be/attachment-0001.htm From lanhaiping at gmail.com Fri Jul 13 15:51:24 2007 From: lanhaiping at gmail.com (lan haiping) Date: Fri, 13 Jul 2007 21:51:24 +0800 Subject: [Pw_forum] relax question In-Reply-To: References: <46976bcc.201f5e0a.6fd2.fffffa01@mx.google.com> Message-ID: Dear yuehua you should do a look at $QE/Doc/INPUT_PW, you can find all that you want to know. Regards, On 7/13/07, xu yuehua wrote: > > thanks for above Mr Ding and Mr Lan answers. > and another related question is whether conv_thr is in &ELECTRONS? is > it about scf ? > i don not understand how a flow chart when relaxing ions and its details? > hope your kind help > 2007/7/13, Ding Xunlei > >: > > > > Dear Yuehua, > > > > The default number is : > > etot_conv_thr=1.D-4 > > forc_conv_thr=1.D-3 > > > > So, 1. Your setting for forc_conv_thr= 1.D-6 is too small. > > 2. If you do need to set force like that, you need to set > > etot_conv_thr even smaller, saying 1.D-7. And also, you need to set > > TRUE smaller than etot_conv_thr, saying 1.D-8. > > > > Hope it help. > > > > Yours sincerely, > > Ding > > 2007-07-13 > > ______________________________________________ > > > > ======= 2007-07-13 16:27:54 You have written: ======= > > > > >hello all > > >i am doing relax. > > >in my inputfile, > > >i use etot_conv_thr=1.D-6 > > > forc_conv_thr=1.D-6 > > >and the output is > > > atom 1 type 1 force = 0.00000290 0.00000095 - > > 0.00000364 > > > atom 2 type 1 force = 0.00000081 -0.00000108 > > 0.00000033 > > > atom 3 type 1 force = -0.00000435 -0.00000425 - > > 0.00000160 > > > atom 4 type 1 force = -0.00000312 -0.00000321 - > > 0.00000080 > > > atom 5 type 1 force = - 0.00000446 -0.00000036 - > > 0.00000273 > > > atom 6 type 1 force = -0.00000048 0.00000230 > > 0.00000112 > > > atom 7 type 1 force = 0.00000164 0.00000118 > > 0.00000060 > > > atom 8 type 1 force = 0.00000131 -0.00000040 > > 0.00000145 > > > atom 9 type 2 force = -0.00000362 0.00000154 - > > 0.00000201 > > > atom 10 type 2 force = -0.00000029 0.00000191 > > 0.00000302 > > > atom 11 type 2 force = 0.00000576 -0.00000134 > > 0.00000194 > > > atom 12 type 2 force = 0.00000390 0.00000276 > > 0.00000233 > > > > > >Total force = 0.000015 Total SCF correction = 0.000054 > > > SCF correction compared to forces is too large, reduce conv_thr > > > > > > The maximum number of steps has been reached. > > >my qustion is ,force at every atom is ok ,but why the program asks to > > >reduce conv_thr > > >my conv_thr is not enough small ? > > >or my total relax step (50) is not adequate? > > >is there any principle about these parameter? > > >need your help > > >thanks a lot > > >-- > > >Xu Yuehua > > >physics Department of Nanjing university > > >China > > > > > >_______________________________________________ > > >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 > > > > > > -- > Xu Yuehua > physics Department of Nanjing university > China > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20070713/6a9c85c5/attachment.htm From degironc at sissa.it Fri Jul 13 17:24:54 2007 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 13 Jul 2007 17:24:54 +0200 Subject: [Pw_forum] relax question In-Reply-To: References: Message-ID: <1184340294.469799461369c@webmail.sissa.it> a tipical bond stiffness, if I recall correctly, is in the range of 0.1-1.0 Ry/bohr^2 thus a threshold of 1.d-6 Ry/bohr in the force means that you are asking a numerical accuracy in your interatomic distances of the order of 1.d-5-1.d-6 bohr... are you sure you really need it ? best regards, stefano deGironcoli, SISSA & DEMOCRITOS Quoting xu yuehua : > hello all > i am doing relax. > in my inputfile, > i use etot_conv_thr=1.D-6 > forc_conv_thr=1.D-6 > and the output is > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > atom 2 type 1 force = 0.00000081 -0.00000108 0.00000033 > atom 3 type 1 force = -0.00000435 -0.00000425 -0.00000160 > atom 4 type 1 force = -0.00000312 -0.00000321 -0.00000080 > atom 5 type 1 force = -0.00000446 -0.00000036 -0.00000273 > atom 6 type 1 force = -0.00000048 0.00000230 0.00000112 > atom 7 type 1 force = 0.00000164 0.00000118 0.00000060 > atom 8 type 1 force = 0.00000131 -0.00000040 0.00000145 > atom 9 type 2 force = -0.00000362 0.00000154 -0.00000201 > atom 10 type 2 force = -0.00000029 0.00000191 0.00000302 > atom 11 type 2 force = 0.00000576 -0.00000134 0.00000194 > atom 12 type 2 force = 0.00000390 0.00000276 0.00000233 > > Total force = 0.000015 Total SCF correction = 0.000054 > SCF correction compared to forces is too large, reduce conv_thr > > The maximum number of steps has been reached. > my qustion is ,force at every atom is ok ,but why the program asks to > reduce conv_thr > my conv_thr is not enough small ? > or my total relax step (50) is not adequate? > is there any principle about these parameter? > need your help > thanks a lot > -- > Xu Yuehua > physics Department of Nanjing university > China > ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From njuxuyuehua at gmail.com Sat Jul 14 10:20:16 2007 From: njuxuyuehua at gmail.com (xu yuehua) Date: Sat, 14 Jul 2007 16:20:16 +0800 Subject: [Pw_forum] relax question In-Reply-To: <1184340294.469799461369c@webmail.sissa.it> References: <1184340294.469799461369c@webmail.sissa.it> Message-ID: about why i use etot_conv_thr as well as force_conv_thr= 1.d-6 Ry as usual,in vasp and castep software,about 1.d-4 ev is ok why it is so different between these DFT software ? i don not understand. need your help 2007/7/13, Stefano de Gironcoli : > > a tipical bond stiffness, if I recall correctly, is in the range of > 0.1-1.0 > Ry/bohr^2 thus a threshold of 1.d-6 Ry/bohr in the force means that you > are > asking a numerical accuracy in your interatomic distances of the order of > 1.d-5-1.d-6 bohr... are you sure you really need it ? > best regards, > stefano deGironcoli, SISSA & DEMOCRITOS > > Quoting xu yuehua : > > > hello all > > i am doing relax. > > in my inputfile, > > i use Bohr> and the output is > > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > > atom 2 type 1 force = 0.00000081 -0.00000108 > 0.00000033 > > atom 3 type 1 force = -0.00000435 -0.00000425 - > 0.00000160 > > atom 4 type 1 force = - 0.00000312 -0.00000321 - > 0.00000080 > > atom 5 type 1 force = -0.00000446 -0.00000036 - > 0.00000273 > > atom 6 type 1 force = -0.00000048 0.00000230 > 0.00000112 > > atom 7 type 1 force = 0.00000164 0.00000118 > 0.00000060 > > atom 8 type 1 force = 0.00000131 -0.00000040 > 0.00000145 > > atom 9 type 2 force = -0.00000362 0.00000154 - > 0.00000201 > > atom 10 type 2 force = - 0.00000029 0.00000191 > 0.00000302 > > atom 11 type 2 force = 0.00000576 -0.00000134 > 0.00000194 > > atom 12 type 2 force = 0.00000390 0.00000276 > 0.00000233 > > > > Total force = 0.000015 Total SCF correction = 0.000054 > > SCF correction compared to forces is too large, reduce conv_thr > > > > The maximum number of steps has been reached. > > my qustion is ,force at every atom is ok ,but why the program asks to > > reduce conv_thr > > my conv_thr is not enough small ? > > or my total relax step (50) is not adequate? > > is there any principle about these parameter? > > need your help > > thanks a lot > > -- > > Xu Yuehua > > physics Department of Nanjing university > > China > > > > > > > ---------------------------------------------------------------- > SISSA Webmail https://webmail.sissa.it/ > Powered by Horde http://www.horde.org/ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070714/a92acd4f/attachment.html From marzari at MIT.EDU Sat Jul 14 17:59:51 2007 From: marzari at MIT.EDU (Nicola Marzari) Date: Sat, 14 Jul 2007 11:59:51 -0400 Subject: [Pw_forum] relax question In-Reply-To: References: <1184340294.469799461369c@webmail.sissa.it> Message-ID: <4698F2F7.2020702@mit.edu> xu yuehua wrote: > > about why i use etot_conv_thr > as well as force_conv_thr= 1.d-6 Ry > as usual,in vasp and castep software,about 1.d-4 ev is ok > why it is so different between these DFT software ? > i don not understand. > need your help The definition of what a convergence threshold is is probably different. Check into the source code of pwscf where etot_conv_thr is used, and look at the documentation of the other codes. It's like when you change cars - a British car on fifth gear will go 100 miles/h, an Italian car will go 180 km/h. They all do the same thing. 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 2586534 marzari at mit.edu http://quasiamore.mit.edu From ezadshojaee at hotmail.com Sun Jul 15 07:17:57 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 15 Jul 2007 05:17:57 +0000 Subject: [Pw_forum] k_point generation in ph.x for a grid Message-ID: Hi i have talked about it before, but i am not still satisfied! when i do a serial run, for example the ph.x generates 400 k_points for one of the points in the grid, but when i do the same in the parallel mode deviding the k_points between 4, it generates 112 k_points for the same point in the grid which the eigenvalues of 12 of them are exactly zero! and ofcourse the weights are different. and the interesting part is that the conclusion is different from the serial one!! what is the problem? Thanx for any help _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/049102a1/attachment.htm From maximoff at berkeley.edu Sun Jul 15 08:11:25 2007 From: maximoff at berkeley.edu (Sergey Maximoff) Date: Sat, 14 Jul 2007 23:11:25 -0700 Subject: [Pw_forum] Pseudopotentials for TPSS Message-ID: Folks: The distribution version of QE 3.1.1 includes norm-conserving TPSS pseudopotentials for a couple of atoms. The header of the UPF file for H says: "Generated using ld1 code (maybe, or maybe not) Author: unknown Generation date: as well " Looking at the supplied with QE 3.1.1 pseudopotential code reveals statement if (dft_is_meta()) call errore('setyp','meta-GGA not implemented yet', 1) that implies that running ld1.x, at least in the given form, cannot produce TPSS pseudopotentials. I am therefore asking three questions . a) what "(maybe, or maybe not) " code in the header above is? b) Does this "(maybe, or maybe not)" belong to the public domain? c) If the answer to the latter question is 'yes', how can I acquire this "(maybe, or maybe not) "? Regards, Sergey From niuli1978 at yahoo.com.cn Sun Jul 15 08:55:13 2007 From: niuli1978 at yahoo.com.cn (li niu) Date: Sun, 15 Jul 2007 14:55:13 +0800 (CST) Subject: [Pw_forum] decomoposition of the atomic contribution to the vibrational spectrum Message-ID: <712523.52415.qm@web15007.mail.cnb.yahoo.com> Dear all, I onced asked this problem but could still not solved it. My question is about the projection of vibrational spectrum. In order to further decompose the carbon contribution to the vibrational spectrum, as in the paper "Vibrational properties of trtrahedral amorphous carbon from first principles"[APL 75,644(1999)], we should define for each bond a "stretching" vector in the space of the 3N (N is the number of atoms) displacement. The components of each vector involves the displacement of two atoms in the direction of the bond and with opposite orientations. We use these vectors as a (non-orthonormal) basis of the stretching subspace and define the bending subspace as the complement of the stretching subspace. My problems: How can I construct the stretching subspace and orthorgonalize these vectors in 3N space? I wonder if someone has written a code for calculating it,and want to share with others? need your help, any help will be appreciated! best regards! Niu Li Harbin Institue of Technology People's Republic of China --------------------------------- ????????-3.5G???20M??? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/4caab625/attachment.htm From ding at sissa.it Sun Jul 15 11:29:03 2007 From: ding at sissa.it (Xunlei Ding) Date: Sun, 15 Jul 2007 11:29:03 +0200 Subject: [Pw_forum] relax question In-Reply-To: References: <1184340294.469799461369c@webmail.sissa.it> Message-ID: <1184491743.4699e8df8d37f@webmail.sissa.it> Hi, In the vasp forum, one set the parameter like this: EDIFF: 1E-07 EDIFFG: -0.0005 And in the reply it is said, "Anyways, your ediffg is very low. A strict force criteria is -0.001, and many people would use -0.01 or higher. If you look at the change in the energy as you go from a force of 0.01 eV/Ang to 0.001 eV/Ang, it is usually negligible." You can check the difference of results when you set force_conv_thr as 1.d-3 and 1.d-4. If the difference is negligible, why use too strict threshold? Yours, Ding Xunlei Quoting xu yuehua : > about why i use etot_conv_thr > as well as force_conv_thr= 1.d-6 Ry > as usual,in vasp and castep software,about 1.d-4 ev is ok > why it is so different between these DFT software ? > i don not understand. > need your help > > > > 2007/7/13, Stefano de Gironcoli : > > > > a tipical bond stiffness, if I recall correctly, is in the range of > > 0.1-1.0 > > Ry/bohr^2 thus a threshold of 1.d-6 Ry/bohr in the force means that you > > are > > asking a numerical accuracy in your interatomic distances of the order of > > 1.d-5-1.d-6 bohr... are you sure you really need it ? > > best regards, > > stefano deGironcoli, SISSA & DEMOCRITOS > > > > Quoting xu yuehua : > > > > > hello all > > > i am doing relax. > > > in my inputfile, > > > i use Bohr> and the output is > > > atom 1 type 1 force = 0.00000290 0.00000095 -0.00000364 > > > atom 2 type 1 force = 0.00000081 -0.00000108 > > 0.00000033 > > > atom 3 type 1 force = -0.00000435 -0.00000425 - > > 0.00000160 > > > atom 4 type 1 force = - 0.00000312 -0.00000321 - > > 0.00000080 > > > atom 5 type 1 force = -0.00000446 -0.00000036 - > > 0.00000273 > > > atom 6 type 1 force = -0.00000048 0.00000230 > > 0.00000112 > > > atom 7 type 1 force = 0.00000164 0.00000118 > > 0.00000060 > > > atom 8 type 1 force = 0.00000131 -0.00000040 > > 0.00000145 > > > atom 9 type 2 force = -0.00000362 0.00000154 - > > 0.00000201 > > > atom 10 type 2 force = - 0.00000029 0.00000191 > > 0.00000302 > > > atom 11 type 2 force = 0.00000576 -0.00000134 > > 0.00000194 > > > atom 12 type 2 force = 0.00000390 0.00000276 > > 0.00000233 > > > > > > Total force = 0.000015 Total SCF correction = 0.000054 > > > SCF correction compared to forces is too large, reduce conv_thr > > > > > > The maximum number of steps has been reached. > > > my qustion is ,force at every atom is ok ,but why the program asks to > > > reduce conv_thr > > > my conv_thr is not enough small ? > > > or my total relax step (50) is not adequate? > > > is there any principle about these parameter? > > > need your help > > > thanks a lot > > > -- > > > Xu Yuehua > > > physics Department of Nanjing university > > > China > > > > > > > > > > > > > ---------------------------------------------------------------- > > SISSA Webmail https://webmail.sissa.it/ > > Powered by Horde http://www.horde.org/ > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > > > > > -- > Xu Yuehua > physics Department of Nanjing university > China > ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From mpayami at aeoi.org.ir Sun Jul 15 11:13:26 2007 From: mpayami at aeoi.org.ir (Mahmoud Payami) Date: Sun, 15 Jul 2007 13:43:26 +0430 Subject: [Pw_forum] problem in parallel execution Message-ID: <001a01c7c6c0$66cefc50$206510ac@aeoi.org.ir> Dear PW users, Some times when I prepare a parallel environment using "mpdboot" command, and run pw.x (both R & G and k point parallel), at the mid of the execution the number of pools (automatically?!) reduces to smaller values and the running process slows down. That is, when I use "mpdtrace" and I get, say, 10 node names, in the mid of execution this command renders, say 3 node numbers and the program is running in those new 3 pools. I would be grateful if you please comment on how can I control this reduction of the number of pools. Bests, Mahmoud Payami -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/eacec5ab/attachment.htm From baroni at sissa.it Sun Jul 15 12:48:02 2007 From: baroni at sissa.it (Stefano Baroni) Date: Sun, 15 Jul 2007 06:48:02 -0400 Subject: [Pw_forum] relax question In-Reply-To: <1184491743.4699e8df8d37f@webmail.sissa.it> References: <1184340294.469799461369c@webmail.sissa.it> <1184491743.4699e8df8d37f@webmail.sissa.it> Message-ID: <692B131C-B6CA-421B-9FD6-03B0E2FD6237@sissa.it> Hi All: a small further comment to explicitly stress a point that has already been raised more or less implicitly. One runs simulations not just for the sake of it, but to calculate properties, right? And properties are numbers, right? Well, the right questions to ask at the start of a simulation concern the influence of the input parameters on the *specific numbers* one wants to calculate (i.e. the *specific property* one wants to study). Of course, the energy is a very important property, and it is very correct to ask oneself what the influence of the input parameters on the accuracy of the calculated energy would be (not to mention its conservation, if one runs a MD simulation). It is not the only one, though. If you want to study the equilibrium geometry, for instance, it is the accuracy of bond lengths and bond angles that would matter, and the softer the system, the more difficult it will be to achieve a given target accuracy (this point has been discussed in this very thread, I think). All in all, I would recommend that before asking what value a given parameter should be given, one should ask oneself "why should I care about this parameter?". A proper answer to the latter question will almost invariably also answer the first. I understand that the usage of canned packages such as QE or vasp encourages a somewhat uncritical approach to the many, many, parameters needed to fine tune a well done simulation. This is one more reason, I think, to make an effort to ask sometimes general questions before specific ones. Cheers - Stefano On Jul 15, 2007, at 5:29 AM, Xunlei Ding wrote: > Hi, > > In the vasp forum, one set the parameter like this: > EDIFF: 1E-07 > EDIFFG: -0.0005 > > And in the reply it is said, "Anyways, your ediffg is very low. A > strict force > criteria is -0.001, and many people would use -0.01 or higher. If > you look at > the change in the energy as you go from a force of 0.01 eV/Ang to > 0.001 eV/Ang, > it is usually negligible." > > You can check the difference of results when you set force_conv_thr > as 1.d-3 and > 1.d-4. If the difference is negligible, why use too strict threshold? > > Yours, > Ding Xunlei > > > Quoting xu yuehua : --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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 -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/65291560/attachment-0001.htm From giannozz at democritos.it Sun Jul 15 12:51:06 2007 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sun, 15 Jul 2007 12:51:06 +0200 Subject: [Pw_forum] k_point generation in ph.x for a grid In-Reply-To: References: Message-ID: <200707151251.06039.giannozz@democritos.it> On Sunday 15 July 2007 07:17, Ezad Shojaee wrote: > when i do a serial run, for example the ph.x generates 400 k_points for one > of the points in the grid, but when i do the same in the parallel mode > deviding the k_points between 4, it generates 112 k_points for the same > point in the grid which the eigenvalues of 12 of them are exactly zero! please provide an example showing the behavior you are describing -- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at democritos.it Sun Jul 15 12:49:22 2007 From: giannozz at democritos.it (Paolo Giannozzi) Date: Sun, 15 Jul 2007 12:49:22 +0200 Subject: [Pw_forum] Pseudopotentials for TPSS In-Reply-To: References: Message-ID: <200707151249.22618.giannozz@democritos.it> On Sunday 15 July 2007 08:11, Sergey Maximoff wrote: > a) what "(maybe, or maybe not) " code in the header above is? nothing related to TPSS -- Paolo Giannozzi, Democritos and University of Udine, Italy From ezadshojaee at hotmail.com Sun Jul 15 14:45:17 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 15 Jul 2007 12:45:17 +0000 Subject: [Pw_forum] =?windows-1256?q?k=5Fpoint_generation_in_ph=2Ex_for_a_?= =?windows-1256?q?grid=FD?= Message-ID: Hi, this is the output file of the code ph.x, in the parallel mode which ran on 8 cpu's and defining npool=4, (deviding the k_points to 4 parts) as you see qualitatively, the last 12 k_points in the list have zero instead of their eigenvalues. The # of k_points for the same point in the grid is dependent to the number of pools (may be trivial) but in all cases there are some k_points with zero in their eigenvalues and also the final result differs! where may be the problem? Thanx for any suggestion! Calculation of q = ### Planes per process (thick) : nr3 = 72 npp = 36 ncplane = 5184 Planes per process (smooth): nr3s= 48 npps= 24 ncplanes= 2304 Proc/ planes cols G planes cols G columns G Pool (dense grid) (smooth grid) (wavefct grid) 1 36 1095 31079 24 553 10976 190 2259 2 36 1096 31078 24 554 10989 191 2260 0 72 2191 62157 48 1107 21965 381 4519 number of k points= 136 cart. coord. in units 2pi/a_0 k( 1) = ( 0.0000000 0.0000000 0.0000000), wk = 0.0246914 k( 2) = ( -0.2500000 0.0000000 0.0994589), wk = 0.0000000 . . . k( 135) = ( -0.3750000 -0.2500000 0.2486474), wk = 0.0246914 k( 136) = ( -0.6250000 -0.2500000 0.3481063), wk = 0.0000000 Band Structure Calculation Davidson diagonalization with overlap . . End of band structure calculation k = 0.0000 0.0000 0.0000 band energies (ev): -49.1240 -49.0635 -25.5839 -25.3454 -25.3454 -25.2735 -25.2735 -25.1898 -10.4104 -9.4033 -9.2501 -8.9222 3.2904 3.4159 3.4165 3.7791 5.1375 5.1380 5.1959 6.4344 6.4402 6.4405 6.7349 6.7353 . . k = 0.0000-0.3750 0.3481 band energies (ev): -49.1038 -49.0825 -25.4806 -25.4059 -25.3684 -25.3096 -25.2518 -25.2464 -9.9189 -9.5090 -9.2043 -9.0437 3.1063 3.2098 3.6205 3.6679 4.7623 5.0586 5.6381 5.7352 6.0354 6.3572 6.5759 6.7423 k = 0.2500 0.3750-0.2486 band energies (ev): 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 . . k =-0.6250-0.2500 0.3481 band energies (ev): 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 _________________________________________________________________ Play free games, earn tickets, get cool prizes! Join Live Search Club.? http://club.live.com/home.aspx?icid=CLUB_wlmailtextlink -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/d0ef4ba5/attachment-0001.htm From njuxuyuehua at gmail.com Sun Jul 15 14:57:34 2007 From: njuxuyuehua at gmail.com (xu yuehua) Date: Sun, 15 Jul 2007 20:57:34 +0800 Subject: [Pw_forum] relax question In-Reply-To: <692B131C-B6CA-421B-9FD6-03B0E2FD6237@sissa.it> References: <1184340294.469799461369c@webmail.sissa.it> <1184491743.4699e8df8d37f@webmail.sissa.it> <692B131C-B6CA-421B-9FD6-03B0E2FD6237@sissa.it> Message-ID: thanks for all your replies.i shall try it according to your advices. 2007/7/15, Stefano Baroni : > > Hi All: > > a small further comment to explicitly stress a point that has already been > raised more or less implicitly. > > One runs simulations not just for the sake of it, but to calculate > properties, right? And properties are numbers, right? > > > Well, the right questions to ask at the start of a simulation concern the > influence of the input parameters on the *specific numbers* one wants to > calculate (i.e. the *specific property* one wants to study). Of course, > the energy is a very important property, and it is very correct to ask > oneself what the influence of the input parameters on the accuracy of the > calculated energy would be (not to mention its conservation, if one runs a > MD simulation). It is not the only one, though. If you want to study the > equilibrium geometry, for instance, it is the accuracy of bond lengths and > bond angles that would matter, and the softer the system, the more difficult > it will be to achieve a given target accuracy (this point has been discussed > in this very thread, I think). > > > All in all, I would recommend that before asking what value a given > parameter should be given, one should ask oneself "why should I care about > this parameter?". A proper answer to the latter question will almost > invariably also answer the first. > > > I understand that the usage of canned packages such as QE or vasp > encourages a somewhat uncritical approach to the many, many, parameters > needed to fine tune a well done simulation. This is one more reason, I > think, to make an effort to ask sometimes general questions before specific > ones. > > > Cheers - Stefano > > > On Jul 15, 2007, at 5:29 AM, Xunlei Ding wrote: > > Hi, > > > In the vasp forum, one set the parameter like this: > EDIFF: 1E-07 > EDIFFG: -0.0005 > > > And in the reply it is said, "Anyways, your ediffg is very low. A strict > force > criteria is -0.001, and many people would use -0.01 or higher. If you look > at > the change in the energy as you go from a force of 0.01 eV/Ang to 0.001eV/Ang, > it is usually negligible." > > > You can check the difference of results when you set force_conv_thr as > 1.d-3 and > 1.d-4. If the difference is negligible, why use too strict threshold? > > > Yours, > Ding Xunlei > > > > > Quoting xu yuehua : > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070715/d136c3e8/attachment.htm From ghaderyan at gmail.com Sun Jul 15 23:43:20 2007 From: ghaderyan at gmail.com (saman ghaderyan) Date: Mon, 16 Jul 2007 01:13:20 +0330 Subject: [Pw_forum] pseudopotential Message-ID: hi i made pseudo potential for Zn (NC-LDA) to calculation phonon freq (ZnS), the result of testing was acceptable , and the phonon freq were nice but ZnS lattice parameter is too small (about 8.8 a.u) can i use this pseudo potential? thanks From ghaderyan at gmail.com Sun Jul 15 23:44:11 2007 From: ghaderyan at gmail.com (saman ghaderyan) Date: Mon, 16 Jul 2007 01:14:11 +0330 Subject: [Pw_forum] pseudopotential Message-ID: hi i made pseudo potential for Zn (NC-LDA) to calculation phonon freq (ZnS), the result of testing was acceptable , and the phonon freq were nice but ZnS lattice parameter is too small (about 8.8 a.u) can i use this pseudo potential? thanks saman ghaderyan buali sina university , Iran From shanavas at barc.gov.in Mon Jul 16 22:55:11 2007 From: shanavas at barc.gov.in (Shanavas K. V.) Date: Tue, 17 Jul 2007 02:25:11 +0530 Subject: [Pw_forum] Fermi surface nesting function In-Reply-To: <1184330288.4608.15.camel@athena.purnima> References: <1184330288.4608.15.camel@athena.purnima> Message-ID: <1184619311.4397.5.camel@athena.purnima> Hi all, I am replying to my own query! (I thought it might help somebody) After a lot of search I found that some form of NF is implemented in Abinit (with the name: nesting factor). Will go through it carefully to see if it is the same thing. regards, shanavas Scientist, Bhabha Atomic Research Centre, Mumbai, India On Fri, 2007-07-13 at 18:08 +0530, Shanavas K. V. wrote: > Hi all, > > I am trying to write a code to calculate fermi surface nesting function > for an AFM system. > Is there any code already available for it? I found a paper > http://arxiv.org/pdf/cond-mat/0504077 using the NF, but can somebody > guide me to some paper that explain the basic theory & method that can > help in writing the code? > > Thanks in advance, > > regards, > shanavas > > Scientist, > Bhabha Atomic Research Centre, > Mumbai, India From giannozz at nest.sns.it Mon Jul 16 09:37:44 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 16 Jul 2007 09:37:44 +0200 Subject: [Pw_forum] =?utf-8?q?k=5Fpoint_generation_in_ph=2Ex_for_a_grid?= =?utf-8?b?4oCO?= In-Reply-To: References: Message-ID: On Jul 15, 2007, at 14:45 , Ezad Shojaee wrote: > this is the output file of the code ph.x this is NOT a sample job. Please provide a sample job --- Paolo Giannozzi, Democritos and University of Udine, Italy From ding at sissa.it Mon Jul 16 09:44:12 2007 From: ding at sissa.it (Xunlei Ding) Date: Mon, 16 Jul 2007 09:44:12 +0200 Subject: [Pw_forum] pseudopotential In-Reply-To: References: Message-ID: <1184571852.469b21ccccc92@webmail.sissa.it> Dear Saman, What's the experimental value for the lattice parameter? And LDA always underestimates lattice parameters. Best regards, Ding Quoting saman ghaderyan : > hi > i made pseudo potential for Zn (NC-LDA) to calculation phonon freq > (ZnS), the result of testing was acceptable , and the phonon freq > were nice > but ZnS lattice parameter is too small (about 8.8 a.u) > can i use this pseudo potential? > thanks > saman ghaderyan > buali sina university , Iran > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From ykanai at berkeley.edu Mon Jul 16 10:20:26 2007 From: ykanai at berkeley.edu (Yosuke Kanai) Date: Mon, 16 Jul 2007 01:20:26 -0700 Subject: [Pw_forum] Pseudopotentials for TPSS In-Reply-To: References: Message-ID: <469B2A4A.30302@berkeley.edu> Hi Sergey I am not sure about this particular PP, but I have an old version of PP generation code (by Paolo G.), in which Xiaofei Wang implemented TPSS in it. He (Xiaofei) is no longer in the field, but I have this code and can send it to you if you like. Yosuke Sergey Maximoff wrote: > Folks: > > The distribution version of QE 3.1.1 includes norm-conserving TPSS > pseudopotentials for a couple of atoms. The header of the UPF file > for H says: > > "Generated using ld1 code (maybe, or maybe not) > Author: unknown Generation date: as well " > > Looking at the supplied with QE 3.1.1 pseudopotential code reveals > statement > > if (dft_is_meta()) call errore('setyp','meta-GGA not implemented > yet', 1) > > that implies that running ld1.x, at least in the given form, cannot > produce TPSS pseudopotentials. > > I am therefore asking three questions . > > a) what "(maybe, or maybe not) " code in the header above is? > > b) Does this "(maybe, or maybe not)" belong to the public domain? > > c) If the answer to the latter question is 'yes', how can I acquire > this "(maybe, or maybe not) "? > > > Regards, > > Sergey > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- -------------------------------- Yosuke Kanai, Ph.D. Berkeley Nanosciences and Nanoengineering Institute University of California, Berkeley Email : ykanai at berkeley.edu Phone : (510)-643-1157 From degironc at sissa.it Mon Jul 16 10:45:41 2007 From: degironc at sissa.it (degironc) Date: Mon, 16 Jul 2007 10:45:41 +0200 Subject: [Pw_forum] Pseudopotentials for TPSS In-Reply-To: <469B2A4A.30302@berkeley.edu> References: <469B2A4A.30302@berkeley.edu> Message-ID: <469B3035.8010404@sissa.it> PG told me that that version was lost forever ... it would be nice to have it and to integrate its content in the distribution. Stefano de Gironcoli, SISSA & DEMOCRITOS Yosuke Kanai wrote: > Hi Sergey > > I am not sure about this particular PP, but I have an old version of PP > generation code > (by Paolo G.), in which Xiaofei Wang implemented TPSS in it. > > He (Xiaofei) is no longer in the field, but I have this code and can > send it to you if you like. > > Yosuke > > > Sergey Maximoff wrote: > >> Folks: >> >> The distribution version of QE 3.1.1 includes norm-conserving TPSS >> pseudopotentials for a couple of atoms. The header of the UPF file >> for H says: >> >> "Generated using ld1 code (maybe, or maybe not) >> Author: unknown Generation date: as well " >> >> Looking at the supplied with QE 3.1.1 pseudopotential code reveals >> statement >> >> if (dft_is_meta()) call errore('setyp','meta-GGA not implemented >> yet', 1) >> >> that implies that running ld1.x, at least in the given form, cannot >> produce TPSS pseudopotentials. >> >> I am therefore asking three questions . >> >> a) what "(maybe, or maybe not) " code in the header above is? >> >> b) Does this "(maybe, or maybe not)" belong to the public domain? >> >> c) If the answer to the latter question is 'yes', how can I acquire >> this "(maybe, or maybe not) "? >> >> >> Regards, >> >> Sergey >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum >> >> > > > From helen at fh.huji.ac.il Mon Jul 16 11:44:35 2007 From: helen at fh.huji.ac.il (Helen) Date: Mon, 16 Jul 2007 12:44:35 +0300 Subject: [Pw_forum] modified explicit exchange potential References: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> <46922E4C.10305@sissa.it> Message-ID: <002101c7c78d$eb220560$cd604084@fh.huji.ac.il> Hi, I sent a previous message which wasn't answered, but which I've since been working on, so I think I have found some answers myself. I'd really appreciate Dr. de Gironcoli's input as I'm sure he is a lot more knowledgable than I am. 1. I asked a question about band-structure calculations when using exx.f90, as in the examples it states that only k-points used in the original scf calculation can be used to calculate band structure. I think a way to get round this would be to in the scf input file, first list the k-points with normal weights for scf calculation, followed by a list of k-points for band structure calculation with very small weights. This way the values at these points are calculated but they won't effect any BZ integrations. (For direct band gaps just use Monkhorst-pack k-point generator which includes k=(0, 0, 0)). 2. The error involved in discarding the q=0 term goes as 1/(NQS*Omega). This fact is utilized in removing the error. I think this will still be valid if the e- interaction is not U(r)=1/r but U(r)=efr(gam r)/r ? [Fourier Transform=exp(-q^2/4*gam^2)*4Pi/q^2, so divergence as q->0 is same as U(r)=1/r, so I think they should be the same) 3. I think that in exx.f90 all I need to do to account for a descreened potential is to change fac(ig): fac(ig)=e2*fpi*dexp(-qq/(4d0*gam))/(tpiba2*qq + yukawa ) * grid_factor I did this in the routines exxenergy2 and in vexx. I haven't touched exx_divergence, do I need to? Are there any other places which need to be changed? Thanks so much for your help, Dr.Helen Eisenberg Post-doctoral researcher, The Fritz Haber Center for Molecular Dynamics, The Hebrew University of Jerusalem, Jerusalem 91904 Israel. ----- Original Message ----- Hi, I have a few questions regarding exx: 1. I was reading in the examples the README file for EXX_example, that when calculating the band structure you have to use the same k-points as are contained in the scf calculation. Currently I want to calculate the band structure in Si along the 100 axis, similar to example01 (though using the different functional for exx which I need to implement). Can I just use the standard scf input k-points and tag the k-points I need at the end (appropriately weighted) in the scf input file? If so what weighting will they need? e.g K_POINTS 20 0.1250000 0.1250000 0.1250000 1.00 0.1250000 0.1250000 0.3750000 3.00 0.1250000 0.1250000 0.6250000 3.00 0.1250000 0.1250000 0.8750000 3.00 0.1250000 0.3750000 0.3750000 3.00 0.1250000 0.3750000 0.6250000 6.00 0.1250000 0.3750000 0.8750000 6.00 0.1250000 0.6250000 0.6250000 3.00 0.3750000 0.3750000 0.3750000 1.00 0.0 0.0 0.0 1.0? 0.0 0.0 0.1 1.0? 0.0 0.0 0.2 1.0? 0.0 0.0 0.3 1.0? 0.0 0.0 0.4 1.0? 0.0 0.0 0.5 1.0? 0.0 0.0 0.6 1.0? 0.0 0.0 0.7 1.0? 0.0 0.0 0.8 1.0? 0.0 0.0 0.9 1.0? 0.0 0.0 1.0 1.0? 2. In README it states that the error involved in discarding the q=0 term goes as 1/(NQS*Omega). This fact is utilized in removing the error. Will this still be valid if the e- interaction is not U(r)=1/r but U(r)=efr(gam r)/r ? [Fourier Transform=exp(-q^2/4*gam^2)*4Pi/q^2, so divergence as q->0 is same as U(r)=1/r, so I think they should be the same) 3. Is there anywhere in exx.f90 where the interaction is defined and then used throughout the code. i.e is there a specific place in the code where I can change the interaction, or do I need to look through the whole code? Thanks so much for your help, Dr.Helen Eisenberg Post-doctoral researcher, The Fritz Haber Center for Molecular Dynamics, The Hebrew University of Jerusalem, Jerusalem 91904 Israel. From: "degironc" To: Sent: Monday, July 09, 2007 3:47 PM Subject: Re: [Pw_forum] modified explicit exchange potential > Helen wrote: >> Thank you Paolo and Lorenzo to the replies to my previous e-mail. I think >> I didn't explain what I wanted to modify clearly enough. We modify the >> explicit exchange to include a descreened potential: >> K_x[n]=-0.25 Int[ P[n](r,r')^2 * u(r-r') ] d3r d3r' >> where P[n](r,r') is the density matrix of non-interacting particles and >> u(r-r') is the descreened interaction (normally the coulomb interaction >> u(r)=1/r, but in our case a different function). >> Which files do I need to change for the modified K_x and its potential? >> Is it exx.f90 in PW? These explicit exchanges are also used in >> Hartree-Fock and B3LYP calculations. >> Thank you for your help, >> Helen Eisenberg > Yes, the part of the code performing HF-like exchange is in exx.f90. > You need to compile the code specifying -DEXX at compilation time... HF, > B3LYP and PBE0 are implemented (but no corresponding pseudopotential are > available so one need to use some PBE or LDA one and overwite the XC by > the input_dft flag in the system namelist) ... > Beware: it is extreeeemely slow. > > As you wll see the EXX part requires some work to be made more useful... > if you are interested in working in that line let me know... :-) .. by the > way, full affiliation would be appreciated. > > Stefano de Gironcoli > SISSA and DEMOCRITOS, Trieste, Italy > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From msn77 at berkeley.edu Mon Jul 16 11:47:26 2007 From: msn77 at berkeley.edu (Sergey N. Maximoff) Date: Mon, 16 Jul 2007 02:47:26 -0700 (PDT) Subject: [Pw_forum] Pseudopotentials for TPSS Message-ID: <2153.75.18.171.134.1184579246.squirrel@calmail.berkeley.edu> Folks, Thank you very much for your comments. It would indeed be cool if pseudopotebtial generator for meta-GAAs become at a part of the distribution... if the existing code comes back from nirvana, I would be really happy to have it. The Original TPSS paper states that TPSS significantly improves the surface XC energies over the parent GGA, PBE (for the jellium). I would like to check if this functional is any good for reaction barriers over metal surfaces... There has been recent JCP indicating that TPSS sometimes is as good as hybrid functinals for reaction barriers over materials with a gap... Regards, Sergey From giannozz at nest.sns.it Mon Jul 16 14:14:36 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 16 Jul 2007 14:14:36 +0200 Subject: [Pw_forum] problem in parallel execution In-Reply-To: <001a01c7c6c0$66cefc50$206510ac@aeoi.org.ir> References: <001a01c7c6c0$66cefc50$206510ac@aeoi.org.ir> Message-ID: On Jul 15, 2007, at 11:13 , Mahmoud Payami wrote: > Some times [...] at the mid of the execution the number of pools > (automatically?!) reduces to smaller values the number of pools is fixed: the code reads from the command line how many pools and behaves accordingly. I don't think the number of pools and the number of processors in a pool can change! Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From mpayami at aeoi.org.ir Mon Jul 16 16:28:51 2007 From: mpayami at aeoi.org.ir (Mahmoud Payami) Date: Mon, 16 Jul 2007 18:58:51 +0430 Subject: [Pw_forum] problem in parallel execution References: <001a01c7c6c0$66cefc50$206510ac@aeoi.org.ir> Message-ID: <000601c7c7b5$a1b09220$206510ac@aeoi.org.ir> > >> Some times [...] at the mid of the execution the number of pools >> (automatically?!) reduces to smaller values > > the number of pools is fixed: the code reads from the command line > how many pools and behaves accordingly. I don't think the number > of pools and the number of processors in a pool can change! Dear Paolo, Thank you very much for your response. As I understaood, if the number of pools and/or the number of processors in the pools reduces (somehow) from what the program has read from the command line, the program certainly stops. So, something should be wrong in the mpich2 response to the "mpdtrace" command. Bests, mahmoud payami From gnkrumah at ictp.it Mon Jul 16 18:57:20 2007 From: gnkrumah at ictp.it (Nkrumah-Buandoh George_Kofi) Date: Mon, 16 Jul 2007 18:57:20 +0200 (CEST) Subject: [Pw_forum] Which Fermi level is right? Message-ID: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> I am calculating the band structure and density of states for YVO4. First I relaxed the atomic positions, and then used the relaxed positions of the atoms for doing scf calculation, and then used the same relaxed atomic positions for nscf calculation of the density of states. For the scf calculation I used occupation='smearing', the fermi energy is 7.5440 eV. For the density of states I used occupation='tetrahedra', and the Fermi energy is 6.4196 eV Why do the two different methods give two different results? Which of these is the correct Fermi energy level? George K Nkrumah-Buandoh International Centre for Theoretical Physics, Trieste Italy & Department of Physics, University of Ghana, Accra Ghana From nanomate at gmail.com Tue Jul 17 03:58:14 2007 From: nanomate at gmail.com (Kenneth Wong) Date: Tue, 17 Jul 2007 11:58:14 +1000 Subject: [Pw_forum] Charge difference query Message-ID: <68e1226a0707161858p69f8b18agcae099fa6ccf2c21@mail.gmail.com> Dear PWscf, I wish to get some help regarding the calculation of the charge difference. For example, if H2O was adsorbed onto TiO2 i want to see the charge difference this is how i would go about it Charge density of a. (H2O + TiO2) b. (TiO2) c. (H2O) Charge density difference = a - (b+c) So i would save the the a, b, c, intermediate files all in e.g. separate calculation &inputPP filplot='inter(1)' ...........then 'inter(2) then 'inter(3)' (i guess they will all have to be in different names? inter(1), inter(2), inter(3) as a, b, c, respectively) Now, here is where i am lost..... &plot nfile = 3 filepp(3)=inter, weight(3) =weight(1)*rho(1) - ( weight(2)*rho(2)+weight(3)*rho(3)), i don't know what to do in this section, in regards to how to use weight(i), filepp(i) Thank you very much, regards, Ken. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070717/c47bd9d9/attachment-0001.htm From akohlmey at cmm.chem.upenn.edu Tue Jul 17 03:08:41 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 16 Jul 2007 21:08:41 -0400 (EDT) Subject: [Pw_forum] Charge difference query In-Reply-To: <68e1226a0707161858p69f8b18agcae099fa6ccf2c21@mail.gmail.com> Message-ID: On Tue, 17 Jul 2007, Kenneth Wong wrote: KW> Dear PWscf, dear ken, KW> I wish to get some help regarding the calculation of the charge difference. KW> For example, if H2O was adsorbed onto TiO2 i want to see the charge KW> difference this is how i would go about it KW> KW> Charge density of KW> a. (H2O + TiO2) KW> b. (TiO2) KW> c. (H2O) KW> Charge density difference = a - (b+c) KW> KW> So i would save the the a, b, c, intermediate files all in e.g. KW> separate calculation the following is just what i gathered from reading the documentation. so it may or may not work, but at least it should put you into the right direction... KW> &inputPP KW> filplot='inter(1)' this would be the output file name: 'densdiff' ...........then 'inter(2) then 'inter(3)' nope. KW> (i guess they will all have to be in different names? inter(1), inter(2), KW> inter(3) as a, b, c, respectively) KW> KW> Now, here is where i am lost..... KW> &plot KW> nfile = 3 KW> filepp(3)=inter, here you provide 3 array element for input file and weights. filepp(1)='h2o_plus_tio2' filepp(2)='h2o' filepp(3)='tio2' KW> weight(3) =weight(1)*rho(1) - ( weight(2)*rho(2)+weight(3)*rho(3)), weight(1)=1.0 weight(2)=-1.0 weight(3)=-1.0 that should give you what you want. the postprocess job will just loop over all files, pick the desired property and sum it after multiplying with the weight into the result data and finaly write it out in the desired format. does that make sense to you? cheers, axel. KW> KW> i don't know what to do in this section, in regards to how to use weight(i), KW> filepp(i) KW> KW> KW> Thank you very much, KW> KW> regards, Ken. KW> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From niuli1978 at yahoo.com.cn Tue Jul 17 04:45:08 2007 From: niuli1978 at yahoo.com.cn (li niu) Date: Tue, 17 Jul 2007 10:45:08 +0800 (CST) Subject: [Pw_forum] decomoposition of the atomic contribution to the vibrational spectrum In-Reply-To: Message-ID: <957755.92156.qm@web15005.mail.cnb.yahoo.com> Dear all, I onced asked this problem but could still not solved it. My question is about the projection of vibrational spectrum. In order to further decompose the carbon contribution to the vibrational spectrum, as in the paper "Vibrational properties of trtrahedral amorphous carbon from first principles"[APL 75,644(1999)], we should define for each bond a "stretching" vector in the space of the 3N (N is the number of atoms) displacement. The components of each vector involves the displacement of two atoms in the direction of the bond and with opposite orientations. We use these vectors as a (non-orthonormal) basis of the stretching subspace and define the bending subspace as the complement of the stretching subspace. My problems: How can I construct the stretching subspace and orthorgonalize these vectors in 3N space? Which one code may I use? need your help, any help will be appreciated! best regards! Niu Li Harbin Institue of Technology People's Republic of China --------------------------------- Mp3???-??????? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070717/9ef0206d/attachment.htm From akohlmey at cmm.chem.upenn.edu Tue Jul 17 05:19:06 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Mon, 16 Jul 2007 23:19:06 -0400 (EDT) Subject: [Pw_forum] decomoposition of the atomic contribution to the vibrational spectrum In-Reply-To: <957755.92156.qm@web15005.mail.cnb.yahoo.com> Message-ID: On Tue, 17 Jul 2007, li niu wrote: LN> Dear all, dear niu li (and others that feel, they don't get the help they need), LN> I onced asked this problem but could still not solved it. My question LN> is about the projection of vibrational spectrum. if you do not get an answer to a question, simply repeating it will not improve your chances to get an answer. on the contrary, it may even irritate people. instead you should ask yourself, why did nobody answer? perhaps your question was confusing, so you should try to ask in a different way. perhaps nobody knows the answer. then repeating the question will not help. perhaps nobody is interested in solving your problem. then you should try to make it more attractive to solve your problem. perhaps your question is so simple, that people think you can figure it out by yourself. then you should think about the problem again. most likely, it is a combination of several of the points listed above. LN> In order to further decompose the carbon contribution to the LN> vibrational spectrum, as in the paper "Vibrational properties of LN> trtrahedral amorphous carbon from first principles"[APL LN> 75,644(1999)], we should define for each bond a "stretching" vector LN> in the space of the 3N (N is the number of atoms) displacement. The the most obvious question: did you ask the authors of this paper? perhaps they already have a solution to your problem? LN> components of each vector involves the displacement of two atoms in LN> the direction of the bond and with opposite orientations. We use LN> these vectors as a LN> (non-orthonormal) basis of the stretching subspace and define the bending LN> subspace as the complement of the stretching subspace. what you describe sounds like rather straightforward vector/matrix algebra. have you thought about programming it yourself? LN> My problems: LN> How can I construct the stretching subspace and orthorgonalize these LN> vectors in 3N space? well, you essentially wrote what you have to do yourself. LN> Which one code may I use? LN> LN> need your help, any help will be appreciated! (almost) everybody appreciates help. however, if you cannot come up with some incentive that motivates people to help you (an interesting challenge, a project worth doing, people feeling obliged to help you, because you were helping them, money, ...), you'll be stuck with your problem. this is the kind of contract that you enter when using open source software. best regards, axel. LN> best regards! LN> LN> Niu Li LN> Harbin Institue of Technology LN> People's Republic of China LN> LN> LN> --------------------------------- LN> Mp3??????-?????????????? -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From hqzhou at nju.edu.cn Tue Jul 17 08:13:56 2007 From: hqzhou at nju.edu.cn (Huiqun Zhou) Date: Tue, 17 Jul 2007 14:13:56 +0800 Subject: [Pw_forum] decomoposition of the atomic contribution to thevibrational spectrum References: <957755.92156.qm@web15005.mail.cnb.yahoo.com> Message-ID: <04e001c7c839$af3585c0$1d00a8c0@solarflare> Li Niu: Nobody but linear algebra can help you;-) Go to find a textbook, Gram Schmidt orthogonalization will help you understand the process of vector orthogonalization. For actual numerical computation, you'd better use Householder transformation in your case as your 3N vectors are numeric. Hope this give you a start point. Huiqun Zhou at Nanjing University, China ----- Original Message ----- From: li niu To: pw_forum at pwscf.org Sent: Tuesday, July 17, 2007 10:45 AM Subject: [Pw_forum] decomoposition of the atomic contribution to thevibrational spectrum Dear all, I onced asked this problem but could still not solved it. My question is about the projection of vibrational spectrum. In order to further decompose the carbon contribution to the vibrational spectrum, as in the paper "Vibrational properties of trtrahedral amorphous carbon from first principles"[APL 75,644(1999)], we should define for each bond a "stretching" vector in the space of the 3N (N is the number of atoms) displacement. The components of each vector involves the displacement of two atoms in the direction of the bond and with opposite orientations. We use these vectors as a (non-orthonormal) basis of the stretching subspace and define the bending subspace as the complement of the stretching subspace. My problems: How can I construct the stretching subspace and orthorgonalize these vectors in 3N space? Which one code may I use? need your help, any help will be appreciated! best regards! Niu Li Harbin Institue of Technology People's Republic of China ------------------------------------------------------------------------------ Mp3???-??????? ------------------------------------------------------------------------------ _______________________________________________ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070717/a219bbc7/attachment.htm From lanhaiping at gmail.com Tue Jul 17 11:11:55 2007 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 17 Jul 2007 17:11:55 +0800 Subject: [Pw_forum] Which Fermi level is right? In-Reply-To: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> References: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> Message-ID: Dear George, I am not familiar with all these smearing scheme, and just try to explain as far as i know. When we do calculation with "smearing", fractional occupations are allowed. For different smearing scheme, the occupation functions are defined differently. So, it should be nonsense to discuss the absolute values of fermi energy obtained from different smearing schemes. hope help ! H.P On 7/17/07, Nkrumah-Buandoh George_Kofi wrote: > > I am calculating the band structure and density of states for YVO4. First > I relaxed the atomic positions, and then used the relaxed positions of the > atoms for doing scf calculation, and then used the same relaxed atomic > positions for nscf calculation of the density of states. For the scf > calculation I used occupation='smearing', the fermi energy is 7.5440 eV. > For the density of states I used occupation='tetrahedra', and the Fermi > energy is 6.4196 eV > > Why do the two different methods give two different results? Which of > these is the correct Fermi energy level? > > George K Nkrumah-Buandoh > International Centre for Theoretical Physics, Trieste Italy & Department > of Physics, University of Ghana, Accra Ghana > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070717/42734cb2/attachment.htm From giannozz at nest.sns.it Tue Jul 17 12:05:15 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 17 Jul 2007 12:05:15 +0200 Subject: [Pw_forum] Which Fermi level is right? In-Reply-To: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> References: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> Message-ID: <6CEDAC00-46AF-40D7-B0A3-EBB0A46A0272@nest.sns.it> On Jul 16, 2007, at 18:57 , Nkrumah-Buandoh George_Kofi wrote: > [...] For the scf calculation I used occupation='smearing', the > fermi energy > is 7.5440 eV. For the density of states I used > occupation='tetrahedra', and > the Fermi energy is 6.4196 eV > > Why do the two different methods give two different results? they don't. If both calculations are performed with a k-point grid that allows a decent description of the Fermi surface, and if the gaussian broadening has an appropriate value, you should get similar results for both KS eigenvalues and Fermi energy. If you don't , you should carefully check your input data. Remember that atomic positions are read from input data, not from the data file, in a nscf calculation Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From marzari at MIT.EDU Tue Jul 17 15:15:03 2007 From: marzari at MIT.EDU (Nicola Marzari) Date: Tue, 17 Jul 2007 09:15:03 -0400 Subject: [Pw_forum] Which Fermi level is right? In-Reply-To: References: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> Message-ID: <469CC0D7.50807@mit.edu> Dear All, an additional comment here - all smearing schemes will converge to the same identical result in the limit of zero smearing/temperature, and accurate sampling. The tetrahedron method is not a smearing scheme, but again all correct calculations hsould converge to the same value. This has been often repeated, but it is worthwhile testing all of this on one's own, to figure out what is really happening. nicola lan haiping wrote: > Dear George, > I am not familiar with all these smearing scheme, and just try to > explain as far as i know. > When we do calculation with "smearing", fractional occupations are allowed. > For different smearing scheme, the occupation functions are defined > differently. > So, it should be nonsense to discuss the absolute values of fermi > energy obtained from different smearing schemes. -- --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu From konstantin_kudin at yahoo.com Tue Jul 17 16:56:50 2007 From: konstantin_kudin at yahoo.com (Konstantin Kudin) Date: Tue, 17 Jul 2007 07:56:50 -0700 (PDT) Subject: [Pw_forum] decomoposition of the atomic contribution to the vibrational spectrum In-Reply-To: <957755.92156.qm@web15005.mail.cnb.yahoo.com> Message-ID: <653369.10850.qm@web57706.mail.re3.yahoo.com> Dear Li, As with most non-standard requests, you got to be prepared to work for it :-) Anyway, I think this kind of analysis is known as "potential energy distribution" (PED), and was reasonably popular for molecules in the old days. There is a program that did the PED with the Gaussian vibrational output, it is here: http://www.ccl.net/cca/software/SOURCES/FORTRAN/gar2ped/index.shtml That should give you a starting point, naturally, you will still have to deal with formats, add 3 degrees of freedom which are no longer rotations for 3D periodic systems, and define a bunch of internal coordinates which would be messy in the case of the 3D interconnected carbon. Even if do not actually need the PED, the program still has most of the components you are looking for. So go ahead and please report back on your success! Good luck! Kostya --- li niu wrote: > Dear all, > I onced asked this problem but could still not solved it. My > question > is about the projection of vibrational spectrum. > In order to further decompose the carbon contribution to the > vibrational spectrum, as in the paper "Vibrational properties of > trtrahedral amorphous carbon from first principles"[APL > 75,644(1999)], we should define for each bond a "stretching" vector > in the space of the 3N (N is the number of atoms) displacement. The > components of each vector involves the displacement of two atoms in > the direction of the bond and with opposite orientations. We use > these vectors as a > (non-orthonormal) basis of the stretching subspace and define the > bending > subspace as the complement of the stretching subspace. > My problems: > How can I construct the stretching subspace and orthorgonalize > these > vectors in 3N space? > Which one code may I use? > > need your help, any help will be appreciated! > > best regards! > > Niu Li > Harbin Institue of Technology > People's Republic of China > > > --------------------------------- > Mp3??????-?????????????? > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > ____________________________________________________________________________________ Fussy? Opinionated? Impossible to please? Perfect. Join Yahoo!'s user panel and lay it on us. http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7 From lanhaiping at gmail.com Tue Jul 17 17:18:58 2007 From: lanhaiping at gmail.com (lan haiping) Date: Tue, 17 Jul 2007 23:18:58 +0800 Subject: [Pw_forum] Which Fermi level is right? In-Reply-To: <469CC0D7.50807@mit.edu> References: <35228.10.50.27.227.1184605040.squirrel@webmail2.ictp.trieste.it> <469CC0D7.50807@mit.edu> Message-ID: Thank you for clarification of my understanding. Regards, H.P On 7/17/07, Nicola Marzari wrote: > > > > Dear All, > > an additional comment here - all smearing schemes will converge > to the same identical result in the limit of zero smearing/temperature, > and accurate sampling. > > The tetrahedron method is not a smearing scheme, but again all correct > calculations hsould converge to the same value. > > This has been often repeated, but it is worthwhile testing all > of this on one's own, to figure out what is really happening. > > nicola > > > > > lan haiping wrote: > > Dear George, > > I am not familiar with all these smearing scheme, and just try to > > explain as far as i know. > > When we do calculation with "smearing", fractional occupations are > allowed. > > For different smearing scheme, the occupation functions are defined > > differently. > > So, it should be nonsense to discuss the absolute values of fermi > > energy obtained from different smearing schemes. > > -- > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 2586534 marzari at mit.edu http://quasiamore.mit.edu > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -- Hai-Ping Lan Department of Electronics , Peking University , Bejing, 100871 lanhaiping at gmail.com, hplan at pku.edu.cn -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070717/e404c518/attachment.htm From ghaderyan at gmail.com Tue Jul 17 23:14:04 2007 From: ghaderyan at gmail.com (saman ghaderyan) Date: Wed, 18 Jul 2007 00:44:04 +0330 Subject: [Pw_forum] pseudopotential Message-ID: thanks dear Ding the experimental lattice parameter is about 10.2 , i test a lot of code but i can't find better result than this , when the lattice parameter was nice the result of testing (pseudo potential)was too bad , and phonon freq too i change the orbital d to valance part but after one day just for Gamma with cut of large ,the result was too bad (phonon freq) , can you help me thanks again From degironc at sissa.it Tue Jul 17 23:34:34 2007 From: degironc at sissa.it (degironc) Date: Tue, 17 Jul 2007 23:34:34 +0200 Subject: [Pw_forum] pseudopotential In-Reply-To: References: Message-ID: <469D35EA.2020901@sissa.it> 3d orbital energy in Zn is very close to the 4s one and they should be included in valence. Ultrasoft pseudopotentials are needed to obtain an affordable cut-off. If for some reason you do not want to use USPP then you need to consider the 3d orbitals as core but you must at least generate the NCPP with the nonlinear core correction. Decent lattice parameters and phonon frequencies have been obtained in this way for a number of II-VI semiconductors (Phys.Rev.B 47, 3588 (1993)). best regards, stefano de Gironcoli, SISSA & DEMOCRITOS, Trieste (Italy) saman ghaderyan wrote: > hi > i made pseudo potential for Zn (NC-LDA) to calculation phonon freq > (ZnS), the result of testing was acceptable , and the phonon freq > were nice > but ZnS lattice parameter is too small (about 8.8 a.u) > can i use this pseudo potential? > thanks > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From degironc at sissa.it Wed Jul 18 01:08:40 2007 From: degironc at sissa.it (degironc) Date: Wed, 18 Jul 2007 01:08:40 +0200 Subject: [Pw_forum] modified explicit exchange potential In-Reply-To: <002101c7c78d$eb220560$cd604084@fh.huji.ac.il> References: <010501c7c21c$bed23af0$cd604084@fh.huji.ac.il> <46922E4C.10305@sissa.it> <002101c7c78d$eb220560$cd604084@fh.huji.ac.il> Message-ID: <469D4BF8.8060301@sissa.it> Dear Helen, Helen wrote: > Hi, > > I sent a previous message which wasn't answered, but which I've since > been working on, so I think I have found some answers myself. I'd > really appreciate Dr. de Gironcoli's input as I'm sure he is a lot > more knowledgable than I am. > > 1. I asked a question about band-structure calculations when using > exx.f90, as in the examples it states that only k-points used in the > original scf calculation can be used to calculate band structure. I > think a way to get round this would be to in the scf input file, first > list the k-points with normal weights for scf calculation, followed by > a list of k-points for band structure calculation with very small > weights. This way the values at these points are calculated but they > won't effect any BZ integrations. (For direct band gaps just use > Monkhorst-pack k-point generator which includes k=(0, 0, 0)). In the present status of the EXX part calculating a "simple" band-structure is not so simple. It is certainly possible (as you say) to perform a calculation with a regular grid that contains gamma and most high symmetry points (specifying nbnd>nelec/2 in order to have access to some empty bands). This will also contains some point along (001) and can give an idea of the dispersion. I do not see any easy way (actually I dont see any way) by which one can calculate the band structure in a point that does not belong to a regular grid without implementing something new. The problem is that Vx is defined as a sum over the BZ = C * sum_q f(k+q) sum_G /(q+G)2 where phi_k+q are the self-consistently calculated occupied bands in k+q and q is a point in a regular (unshifted) grid. This implies that in order to calculate the bands in k the occupied bands in all k+q are needed..which you tipically don't have A possibility would be to change the definition of Vx to = C * sum_k' f(k') sum_G /(k'-k+G)2 where now k' is a regular (shifted or unshifted) grid. The point here is that the grid of k'-points is always the same as k changes, as opposed to the grid of k+q that is always different.. this would allow to do a band calculation in a simpler way because one could keep the occupied phi_k' from the scf calculation. This requires some coding in exx and probably some reorganization in the management of files in order not to overwrite the files containing the phi_k' in a nscf calculation... I'm not sure if the two calculations (sum over q or sum over k') are strictly equivalent in the scf case when one tries to reduce the BZ sum in Vx to a subset of the points... I think they are not equivalent but I don't know which one is "better". Another option could be: 1) calculate the (occupied and some empty) bands in a regular grid of points, 2) use the wannier90 code to extract the maximally localized wannier-functions, 3) use them to build a tight-binding representation of the hamiltonian that allows you to 4) fourier interpolate the bandstructure throughout the BZ... It looks complicated but it could be the simplest option... and if automated and shown to be accurate it could actually be used also to speed up the EXX part in general...maybe > 2. The error involved in discarding the q=0 term > goes as 1/(NQS*Omega). This fact is utilized in removing the error. I > think this will still be valid if the e- interaction is not U(r)=1/r but > U(r)=efr(gam r)/r ? [Fourier Transform=exp(-q2/4*gam2)*4Pi/q2, so > divergence as q->0 is same as U(r)=1/r, so I think they should be the > same) the q=0 therm is problematic for two reasons: i) it contains an integrable divergence and this is taken care of by the Gygi-Baldereschi trick and ii) the remaining non-divergent term in the q->0 cannot be calculated in q=0 because it is a 0/0 limit. This is why there is the extrapolation trick. I think this still remain a problem even if U(r) is erf(gam r)/r because the problem is the leading q2 dependence of ||^2 which is not easy to extract... If instead one would like to replace the 1/r with a short range interaction (yukawa or other) this limit should be well defined and one would not need extrapolation. For this and other reasons hybrid functionals with a screened short-range non-local exchange might have some computational advantages. > 3. I think that in exx.f90 all I need to do to account for a > descreened potential is to change fac(ig): > fac(ig)=e2*fpi*dexp(-qq/(4d0*gam))/(tpiba2*qq + yukawa > ) * grid_factor > I did this in the routines exxenergy2 and in vexx. I haven't touched > exx_divergence, do I need to? Are there any other places which need to > be changed? I'm not sure... exx_divergence is basically the Ewald energy of a set of unit point charges located at the Bravais lattice vectors+uniform compensating background. If you change the interaction I expect this number to change... try to see how your exchange energy converge for an isolated system (an atom or a molecule) as a function on the box-size, w/o the extrapolation trick. If the divergence is correctly calculated it should converge as 1/omega, if not I would expect something like 1/boxsize. best regards, Stefano de Gironcoli, SISSA & DEMOCRITOS, Trieste (Italy) From lyu7 at ncsu.edu Wed Jul 18 05:09:42 2007 From: lyu7 at ncsu.edu (Liping YU) Date: Tue, 17 Jul 2007 23:09:42 -0400 Subject: [Pw_forum] can't minimize electrons by cp.x in some case Message-ID: <469D8476.50400@ncsu.edu> Dear CP users, By using cp.x (Version 3.2.2), I successfully run the electronic minimization for alpha-Al2O3 in the orthorhombic supercell with 0001 along z. However, for the same system, but with all atoms' original x-coordinates changed into z-coordinates [ correspondingly, y->x, z->y, and celldm(1)->celldm(3), celldm(2)->cellcm(1) and celldm(3)->celldm(2) ], I obtained exploded results in the same calculation of electronic minimization. The forces in the later calculation are unreasonable huge. I played with restart_mode, dt, emass, emass_cutoff, electro_dynamics and orthogonalization, but obtained the similar results for the later. Does anybody have an idea on it? Thanks a lot! Here are the results and input files: ============================================================================== -------------------------------------------------------------------------------- 'Wrong' results of the later case:: -------------------------------------------------------------------------------- * Physical Quantities at step: 1 PERFORMING CONJUGATE GRADIENT MINIMIZATION OF EL. STATES cg_sub: missed minimum, case 3, iteration 1 cg_sub: missed minimum, case 3, iteration 2 cg_sub: missed minimum, case 3, iteration 3 cg_sub: missed minimum, case 3, iteration 4 cg_sub: missed minimum, case 3, iteration 5 cg_sub: missed minimum, case 3, iteration 6 cg_sub: missed minimum, case 3, iteration 7 cg_sub: missed minimum, case 3, iteration 8 total energy = 538.09920 Hartree a.u. kinetic energy = 1245.91322 Hartree a.u. electrostatic energy = -489.43267 Hartree a.u. esr = 0.00005 Hartree a.u. eself = 1141.77281 Hartree a.u. pseudopotential energy = -46.33444 Hartree a.u. n-l pseudopotential energy = 63.60468 Hartree a.u. exchange-correlation energy = -235.65159 Hartree a.u. average potential = 0.00000 Hartree a.u. Eigenvalues (eV), kp = 1 , spin = 1 ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************************************************************** ******************************** Allocated memory (kb) = 237186 CELL_PARAMETERS 15.70448170 0.00000000 0.00000000 0.00000000 24.64098148 0.00000000 0.00000000 0.00000000 9.01629394 System Density [g/cm^3] : 3.9297 Center of mass square displacement (a.u.): 0.000000 Total stress (GPa) ****************** ****************** ****************** ****************** ****************** ****************** ****************** ****************** ****************** ATOMIC_POSITIONS Al 0.268496E+01 -0.455430E+00 0.451284E+01 Al 0.105372E+02 -0.455117E+00 0.466802E-02 Al 0.531009E+01 0.468791E+00 0.946897E-02 ...... Forces acting on atoms (au): Al -0.248373E+00 -0.291618E+01 0.131940E+01 Al 0.113470E+01 -0.410432E+01 0.202596E-01 Al 0.251123E-01 -0.219640E+01 0.789481E-01 Al -0.436733E-02 -0.301101E+01 0.144421E+01 Al -0.472089E+00 -0.242857E+01 0.312572E+00 Al 0.426014E+00 -0.204010E+01 0.184244E+01 Al 0.596109E+00 -0.699927E+00 0.192207E+01 Al 0.866630E+00 -0.126521E+01 0.553334E+00 Al 0.356806E+00 0.526145E+00 0.142158E+00 Al 0.430449E+00 0.535675E+00 0.293962E+01 Al -0.132803E+00 0.191428E+01 -0.450394E+00 Al 0.625795E-01 0.196445E+01 0.219006E+01 Al -0.353702E+00 0.197857E+01 0.125822E+01 Al -0.219280E+00 0.353727E+01 -0.367691E+00 Al 0.112175E+01 0.236952E+01 -0.728411E-02 Al -0.563573E+00 0.348400E+01 0.134256E+01 Al -0.220073E+01 0.211831E+01 0.445998E+00 Al 0.407965E+01 0.369462E+01 0.200353E+01 Al -0.870430E+00 0.220623E+01 0.137657E+01 Al 0.114890E+01 0.417221E+01 -0.251779E+01 Al 0.137358E+01 -0.187841E+01 -0.661891E+00 Al -0.459417E+01 -0.428507E+01 0.267088E+01 Al -0.163996E+01 -0.262860E+01 -0.451911E+00 Al 0.315362E+01 -0.406970E+01 0.231598E+01 O 0.346055E+46 0.167063E+46 -0.429934E+46 O 0.194413E+46 0.632090E+45 -0.182378E+46 O -0.367866E+46 -0.255433E+46 0.490183E+46 O -0.335031E+46 -0.195137E+46 -0.784473E+46 O 0.105093E+46 -0.634591E+45 0.103351E+46 O 0.318632E+45 0.376460E+45 0.246231E+45 O -0.729536E+45 0.453911E+45 -0.114576E+46 O 0.705261E+46 -0.657916E+46 0.123020E+46 O -0.848081E+45 0.978335E+45 -0.111124E+46 O 0.188616E+44 -0.965380E+45 -0.706223E+45 O -0.379249E+45 0.205877E+45 0.634545E+45 O -0.442910E+46 -0.474109E+45 0.265680E+46 O -0.115670E+46 0.581940E+45 0.664210E+45 O 0.328088E+46 -0.422694E+46 -0.542608E+45 O 0.483259E+45 0.379970E+46 -0.187528E+46 O -0.981912E+45 -0.276819E+46 -0.154096E+46 O -0.105872E+46 0.136467E+44 -0.653914E+44 O -0.703848E+45 -0.101888E+46 -0.165355E+46 O 0.917974E+45 0.253408E+46 -0.465626E+46 O -0.890023E+45 0.927226E+45 -0.150334E+46 O -0.387531E+45 0.181674E+46 -0.245774E+45 O -0.125154E+46 0.150074E+46 0.158393E+46 O 0.267528E+44 0.170237E+46 -0.254081E+46 O 0.108675E+46 0.102464E+45 -0.214929E+46 O -0.284874E+46 0.170581E+46 -0.796977E+45 O 0.563682E+46 0.403288E+46 0.121271E+46 O 0.347754E+45 0.604159E+45 -0.932441E+45 O -0.456776E+46 0.159613E+46 -0.269005E+46 O 0.134099E+46 -0.141831E+46 0.206923E+46 O -0.113675E+46 0.622588E+45 -0.226152E+46 O -0.109569E+46 0.571525E+45 0.432158E+46 O -0.836097E+46 -0.108399E+46 0.537673E+46 O 0.111064E+46 -0.301310E+44 -0.536402E+45 O -0.603030E+46 -0.201443E+46 -0.812679E+45 O 0.518096E+46 -0.238284E+46 0.905447E+44 O 0.568088E+46 -0.628589E+46 -0.695370E+46 ------------------------------------------------------------------------------------------ Input file for the above results (the later case) ------------------------------------------------------------------------------------------ &CONTROL calculation = "cp", restart_mode = "restart", nstep = 1, iprint = 1, isave = 1, dt = 1.D0, tstress = .TRUE., tprnfor = .TRUE., etot_conv_thr = 1.D-10, ekin_conv_thr = 1.D-8, prefix = "o-alo", pseudo_dir = "../../pseudo/", outdir = "./tmp/", / &SYSTEM ibrav = 8, celldm(1) = 15.7044817, celldm(2) = 1.56904137, celldm(3) = 0.57412235, nat = 60, ntyp = 2, ecutwfc = 30.D0, ecutrho = 300.d0, nr1b = 30, nr2b = 30, nr3b = 30 / &ELECTRONS emass = 400.D0, emass_cutoff = 4.D0, orthogonalization = 'Gram-Schmidt', startingwfc = 'random', ampre = 0.02, tcg = .true., passop = 0.3, maxiter = 250, / &IONS ion_dynamics = "none", ion_damping = 0.0, ion_positions = 'from_input', greasp=1.0, o-alo.cp.cg.in (22%) Al 7.921473606 15.971975291 4.512788702 Al 2.694314090 16.896107944 4.517763841 Al 10.546602633 16.896414345 0.009582439 Al 5.302831858 20.078484811 0.009606589 Al 13.155159701 20.078486621 4.517743025 Al 0.075949597 21.003161771 0.004794067 Al 7.928278522 21.003156550 4.512935734 O 7.710741155 -2.043769533 7.642312579 O -0.141621955 -2.043829875 3.134084677 O 2.902749682 -2.050015479 7.642377402 O 10.754987676 -2.049713078 3.134311843 O 5.306687008 -2.046772139 2.764307187 O 13.158897981 -2.046724568 7.272498826 O 5.092257966 2.063048454 5.896313554 O 12.944492444 2.063267791 1.388179219 O 2.688265902 2.060188463 1.757927226 O 10.540661233 2.060478908 6.266082043 O 0.284303173 2.057468884 5.896279716 O 8.136532580 2.057686863 1.388127542 O 0.071157477 6.166900746 2.764063146 O 7.923368669 6.166948993 7.272176343 O 2.475335124 6.169732809 7.642234377 O 10.327579475 6.170021630 3.134020880 O 5.519541909 6.163882972 3.134050053 O 13.371899127 6.163938481 7.642111773 O 7.711192321 10.276734419 1.388073580 O -0.141169534 10.276674514 5.896304177 O 5.307234596 10.273539335 6.266122096 O 13.159444652 10.273588312 1.757934671 O 2.903110438 10.270543671 1.388178022 O 10.755349769 10.270848468 5.896244326 O 2.690619950 14.380890578 7.272314156 O 10.543014297 14.381180661 2.764159060 O 5.094762571 14.383429111 3.134152831 O 12.946997671 14.383647526 7.642287787 O 0.286676935 14.377747036 3.134227234 O 8.138906084 14.377964980 7.642379617 O 2.476487646 18.490034104 1.388364305 O 10.328730061 18.490320236 5.896577459 O 0.072491775 18.487597696 6.266180844 O 7.924702090 18.487645669 1.758066145 O 5.520961025 18.484346553 5.896615377 O 13.373317086 18.484400398 1.388556392 ============================================================================================== ----------------------------------------------------------------------------- OUTPUT for the former: ----------------------------------------------------------------------------- ....... * Physical Quantities at step: 1 PERFORMING CONJUGATE GRADIENT MINIMIZATION OF EL. STATES total energy = -732.78366 Hartree a.u. kinetic energy = 275.84787 Hartree a.u. electrostatic energy = -873.80938 Hartree a.u. esr = 0.00005 Hartree a.u. eself = 1141.77281 Hartree a.u. pseudopotential energy = -57.54289 Hartree a.u. n-l pseudopotential energy = 175.38242 Hartree a.u. exchange-correlation energy = -252.66169 Hartree a.u. average potential = 0.00000 Hartree a.u. Eigenvalues (eV), kp = 1 , spin = 1 -14.11 -13.65 -13.65 -13.09 -13.09 -12.98 -12.72 -12.72 -12.53 -12.53 -12.30 -12.09 -11.90 -11.81 -11.77 -11.77 -11.68 -11.68 -11.64 -11.55 -11.53 -11.48 -11.48 -11.47 -11.46 -11.37 -11.37 -11.31 -11.31 -11.30 -11.30 -11.29 -11.29 -11.22 -11.20 -11.16 -1.82 -1.41 -1.41 -1.05 -0.88 -0.88 -0.84 -0.79 -0.79 -0.66 -0.47 -0.45 -0.35 -0.35 -0.28 -0.28 -0.12 -0.12 0.01 0.02 0.03 0.03 0.06 0.06 0.38 0.48 0.65 0.77 0.78 0.88 1.00 1.00 1.13 1.20 1.20 1.20 1.20 1.32 1.37 1.37 1.66 1.66 1.78 1.82 2.09 2.09 2.10 2.36 2.36 2.41 2.41 2.53 2.57 2.57 2.59 2.59 2.71 2.72 2.72 2.94 2.94 3.01 3.01 3.29 3.42 3.42 3.53 3.53 3.54 3.63 3.66 3.69 3.70 3.70 3.71 3.71 3.72 3.72 3.75 3.77 3.79 3.84 3.84 3.97 3.97 4.05 4.16 4.34 4.35 4.35 4.52 4.52 4.53 4.58 4.58 4.59 4.65 4.65 4.75 4.75 4.78 4.83 4.83 4.88 4.88 4.98 5.02 5.04 Allocated memory (kb) = 230722 CELL_PARAMETERS 9.01629400 0.00000000 0.00000000 0.00000000 15.70448171 0.00000000 0.00000000 0.00000000 24.64098151 System Density [g/cm^3] : 3.9297 Center of mass square displacement (a.u.): 0.000000 Total stress (GPa) -6.04271817 -0.00027176 -0.00002169 -0.00027176 4.11174552 -0.22254777 -0.00002169 -0.22254777 0.46893890 ATOMIC_POSITIONS Al 0.451284E+01 0.268496E+01 -0.455430E+00 Al 0.466802E-02 0.105372E+02 -0.455117E+00 Al 0.946897E-02 0.531009E+01 0.468791E+00 ..... Forces acting on atoms (au): Al 0.417762E-04 -0.677787E-03 -0.691584E-03 Al 0.495211E-04 -0.682236E-03 -0.707502E-03 Al -0.205621E-04 0.675129E-03 0.667576E-03 Al -0.234378E-04 0.669865E-03 0.671738E-03 Al -0.390579E-04 -0.639510E-03 -0.676457E-03 Al -0.368408E-04 -0.644786E-03 -0.671837E-03 Al 0.458185E-04 0.712668E-03 0.711855E-03 Al 0.381743E-04 0.708301E-03 0.696171E-03 Al 0.483879E-04 -0.660296E-03 -0.683827E-03 Al 0.397437E-04 -0.665540E-03 -0.670208E-03 Al -0.289318E-04 0.677116E-03 0.642819E-03 Al -0.368054E-04 0.671593E-03 0.656349E-03 Al -0.256374E-04 -0.685579E-03 -0.663989E-03 Al -0.334716E-04 -0.690695E-03 -0.680101E-03 Al 0.482031E-04 0.650814E-03 0.694218E-03 Al 0.508795E-04 0.645649E-03 0.698681E-03 Al 0.586784E-04 -0.706187E-03 -0.702973E-03 Al 0.560339E-04 -0.711430E-03 -0.697780E-03 Al -0.306276E-04 0.677379E-03 0.684187E-03 Al -0.226141E-04 0.672064E-03 0.668847E-03 Al -0.330825E-04 -0.633730E-03 -0.656039E-03 Al -0.244428E-04 -0.639551E-03 -0.643206E-03 Al 0.398327E-04 0.675616E-03 0.668444E-03 Al 0.471589E-04 0.669170E-03 0.682309E-03 O -0.254709E-03 0.532287E-02 -0.755145E-03 O -0.241602E-03 0.534011E-02 -0.753724E-03 O -0.278910E-03 -0.536292E-02 0.775544E-03 O -0.290868E-03 -0.536467E-02 0.756420E-03 O -0.172332E-02 0.313578E-05 -0.131023E-04 O -0.172698E-02 0.101175E-04 -0.148365E-04 O 0.260230E-03 0.537528E-02 -0.712651E-03 O 0.265478E-03 0.537157E-02 -0.726318E-03 O 0.173689E-02 0.356061E-04 0.181250E-04 O 0.173488E-02 0.198960E-04 0.680726E-05 O 0.262848E-03 -0.530875E-02 0.715801E-03 O 0.257000E-03 -0.530945E-02 0.702960E-03 O -0.173448E-02 0.996019E-05 -0.175891E-04 O -0.173254E-02 0.180220E-04 -0.178225E-04 O -0.269582E-03 0.536955E-02 -0.731083E-03 O -0.258501E-03 0.536808E-02 -0.748297E-03 O -0.262685E-03 -0.534357E-02 0.754220E-03 O -0.248664E-03 -0.536075E-02 0.755388E-03 O 0.259182E-03 0.535002E-02 -0.732640E-03 O 0.247551E-03 0.536818E-02 -0.734215E-03 O 0.173874E-02 -0.109692E-04 0.940481E-05 O 0.174329E-02 -0.365810E-05 0.782602E-05 O 0.247631E-03 -0.534783E-02 0.726465E-03 O 0.260530E-03 -0.534942E-02 0.713336E-03 O -0.172934E-02 -0.121098E-04 0.108589E-04 O -0.172990E-02 -0.262559E-04 -0.170945E-05 O -0.263294E-03 0.530948E-02 -0.735133E-03 O -0.269159E-03 0.530898E-02 -0.746165E-03 O -0.273903E-03 -0.537115E-02 0.743730E-03 O -0.268530E-03 -0.537168E-02 0.731360E-03 O 0.260322E-03 0.537313E-02 -0.705011E-03 O 0.248830E-03 0.536996E-02 -0.721939E-03 O 0.174869E-02 -0.111007E-05 -0.340138E-04 O 0.174765E-02 0.626865E-05 -0.308590E-04 O 0.245844E-03 -0.535335E-02 0.731714E-03 O 0.234987E-03 -0.537152E-02 0.729755E-03 ---------------------------------------------------------- INPUT file for the former case: ---------------------------------------------------------- &CONTROL calculation = "cp", restart_mode = "from_scratch", nstep = 1, iprint = 1, isave = 1, dt = 4.D0, tstress = .TRUE., tprnfor = .TRUE., etot_conv_thr = 1.D-10, ekin_conv_thr = 1.D-8, prefix = "o-alo", pseudo_dir = "../../pseudo/", outdir = "./tmp/", / &SYSTEM ibrav = 8, celldm(1) = 9.016294, celldm(2) = 1.741789, celldm(3) = 2.732939 nat = 60, ntyp = 2, ecutwfc = 30.D0, ecutrho = 300.d0, nr1b = 30, nr2b = 30, nr3b = 30 / &ELECTRONS emass = 800.D0, emass_cutoff = 4.D0, orthogonalization = 'Gram-Schmidt', startingwfc = 'random', ampre = 0.02, tcg = .true., passop = 0.3, maxiter = 250, / &IONS ion_dynamics = "none", ion_damping = 0., ion_positions = 'from_input', greasp=1.0, ion_radius(1) = 0.8d0, ion_temperature = 'not_controlled', / ATOMIC_SPECIES Al 26.981538 Al.pw91-n-van_ak.UPF O 15.9994 O.pw91-van_ak.UPF ATOMIC_POSITIONS (bohr) Al 4.512840552 2.684961535 -0.455429738 Al 0.004668024 10.537248030 -0.455117214 Al 0.009468974 5.310085697 0.468791447 Al 4.517600670 13.162309898 0.468803796 Al 0.009394854 0.066768164 3.651494112 Al 4.517566306 7.918991937 3.651502237 Al 4.512656525 2.692310931 4.575652737 Al 0.004546173 10.544601266 4.575961120 Al 0.004600602 5.302200765 7.757965484 Al 4.512758723 13.154527493 7.757967932 Al 0.009310437 0.076044480 8.682658005 Al 4.517463841 7.928371279 8.682659781 Al 4.517544635 2.686152973 11.865003001 Al 0.009423549 10.538443392 11.865315242 Al 0.004616936 5.312009565 12.789274351 Al 4.512779077 13.164234008 12.789286578 Al 0.004665903 0.069249842 15.971969814 Al 4.512788702 7.921473606 15.971975291 Al 4.517763841 2.694314090 16.896107944 Al 0.009582439 10.546602633 16.896414345 Al 0.009606589 5.302831858 20.078484811 Al 4.517743025 13.155159701 20.078486621 Al 0.004794067 0.075949597 21.003161771 Al 4.512935734 7.928278522 21.003156550 O 7.642312579 7.710741155 -2.043769533 O 3.134084677 -0.141621955 -2.043829875 O 7.642377402 2.902749682 -2.050015479 O 3.134311843 10.754987676 -2.049713078 O 2.764307187 5.306687008 -2.046772139 O 7.272498826 13.158897981 -2.046724568 O 5.896313554 5.092257966 2.063048454 O 1.388179219 12.944492444 2.063267791 O 1.757927226 2.688265902 2.060188463 O 6.266082043 10.540661233 2.060478908 O 5.896279716 0.284303173 2.057468884 O 1.388127542 8.136532580 2.057686863 O 2.764063146 0.071157477 6.166900746 O 7.272176343 7.923368669 6.166948993 O 7.642234377 2.475335124 6.169732809 O 3.134020880 10.327579475 6.170021630 O 3.134050053 5.519541909 6.163882972 O 7.642111773 13.371899127 6.163938481 O 1.388073580 7.711192321 10.276734419 O 5.896304177 -0.141169534 10.276674514 O 6.266122096 5.307234596 10.273539335 O 1.757934671 13.159444652 10.273588312 O 1.388178022 2.903110438 10.270543671 O 5.896244326 10.755349769 10.270848468 O 7.272314156 2.690619950 14.380890578 O 2.764159060 10.543014297 14.381180661 O 3.134152831 5.094762571 14.383429111 O 7.642287787 12.946997671 14.383647526 O 3.134227234 0.286676935 14.377747036 O 7.642379617 8.138906084 14.377964980 O 1.388364305 2.476487646 18.490034104 O 5.896577459 10.328730061 18.490320236 O 6.266180844 0.072491775 18.487597696 O 1.758066145 7.924702090 18.487645669 O 5.896615377 5.520961025 18.484346553 O 1.388556392 13.373317086 18.484400398 Thanks a lot for your help. Liping -- Liping YU, PhD Candidate Department of Physics & Center for High Performance Simulation North Carolina State University, Raleigh, NC 27695, USA tel: 1-919-5157228 From jibiaoli at gmail.com Wed Jul 18 07:01:41 2007 From: jibiaoli at gmail.com (Clark Lee) Date: Wed, 18 Jul 2007 13:01:41 +0800 Subject: [Pw_forum] On Workfunction Message-ID: Dear all, calculations of STM maps can also give a value of workfunction. But I noticed that it is substantially different from a result using the conventional method which is related to the subtraction of Fermi energy from a potential. My question in which one is reasonable? -- Yours sincerely, Clark Lee State Key Lab of Corrosion and Protection (SKLCP) Institute of Metal Research (IMR) Chinese Academy of Sciences (CAS), China Phone: 024-23971339 Email: jibiaoli at imr.ac.cn or jibiaoli at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070718/4dad5290/attachment-0001.htm From hande at newton.physics.metu.edu.tr Wed Jul 18 07:42:56 2007 From: hande at newton.physics.metu.edu.tr (Hande Ustunel) Date: Wed, 18 Jul 2007 08:42:56 +0300 (EEST) Subject: [Pw_forum] On Workfunction In-Reply-To: Message-ID: Dear Clark, They usually match pretty well. Two things you should be careful about : 1) Make sure your plane-averaged potential is converged well enough in the vacuum region 2) Units. The Fermi energy is reported in the output in eV whereas the potential values are in Ryd unless I'm mistaken. Best wishes, Hande On Wed, 18 Jul 2007, Clark Lee wrote: > Dear all, > calculations of STM maps can also give a value of workfunction. But I > noticed that it is substantially different from a result using the > conventional method which is related to the subtraction of Fermi energy from > a potential. My question in which one is reasonable? > > -- Hande Ustunel Department of Physics Office 439 Middle East Technical University Ankara 06531, Turkey Tel : +90 312 210 3264 http://www.physics.metu.edu.tr/~hande From glaweh at physik.fu-berlin.de Wed Jul 18 10:07:52 2007 From: glaweh at physik.fu-berlin.de (Henning Glawe) Date: Wed, 18 Jul 2007 10:07:52 +0200 Subject: [Pw_forum] Normalization of the atomic wave functions Message-ID: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> Moin, I am at the momemt trying to understand how the atomic states used in an LDA+U calculation are normalized. I inserted code to print out the normalization into the k-loop of the new_ns subroutine: [...] COMPLEX(DP) :: hg_norm [...] ALLOCATE(wfcatom(npwx,natomwfc)) DO ik = 1, nks IF (lsda) current_spin = isk(ik) IF (nks.GT.1) READ (iunigk) npw, igk IF (nks.GT.1) CALL davcio (evc, nwordwfc, iunwfc, ik, - 1) CALL davcio (swfcatom, nwordatwfc, iunsat, ik, - 1) CALL davcio (wfcatom, nwordatwfc, iunat, ik, -1) DO i = 1, natomwfc hg_norm=ZDOTC(npw, wfcatom(1,i), 1, swfcatom(1,i), 1) write( 43, '(i4.4,1x,i1.1,1x,i3.3,1x,2e20.10)') ik, current_spin, i, hg_norm ENDDO [...] so basically I am calculating <\phi^I_k|S|\phi^I_k> (the S operator being related to the relaxed norm conservation in the case of ultrasoft pseudopotentials). I assumed this to be always 1. However, running the FeO example in the QE distribution, I obtain a spektrum of values, most of them between 0.9 and 1.1. The norm of one atomic state of the example system and its spin-counterpart are spread quite widely, between 0.3 and 3.85 (?). So somehow I get the impression that I do not understand how the atomic states are normalized in pwscf... -- c u henning From degironc at sissa.it Wed Jul 18 11:01:20 2007 From: degironc at sissa.it (degironc) Date: Wed, 18 Jul 2007 11:01:20 +0200 Subject: [Pw_forum] Normalization of the atomic wave functions In-Reply-To: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> References: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> Message-ID: <469DD6E0.20607@sissa.it> the atomic wavefunctions in the pseudopotential are normalized, and in the LDA+U they are used as such... however for each k-point the code builds the bloch-type linear combination of atomic wavefunctions. The norm of this bloch-like wfc depends on the overlap between wfc in neighboring cells AND on the wfc spread. The 3d orbital are likely to be rather localized but the 4s wfc are probably very delocalized. this is why the calculated norm of the bloch-type wfc is significantly different from 1. The projection in LDA+U is however defined on the individual atoms (even if technically bloch states are involved) and these wfc are normalized. hope this helps, stefano de Gironcoli, SISSA and DEMOCRITOS, Trieste (Italy) Henning Glawe wrote: > Moin, > I am at the momemt trying to understand how the atomic states used in an > LDA+U calculation are normalized. I inserted code to print out the > normalization into the k-loop of the new_ns subroutine: > > [...] > COMPLEX(DP) :: hg_norm > [...] > ALLOCATE(wfcatom(npwx,natomwfc)) > > DO ik = 1, nks > IF (lsda) current_spin = isk(ik) > IF (nks.GT.1) READ (iunigk) npw, igk > IF (nks.GT.1) CALL davcio (evc, nwordwfc, iunwfc, ik, - 1) > > CALL davcio (swfcatom, nwordatwfc, iunsat, ik, - 1) > CALL davcio (wfcatom, nwordatwfc, iunat, ik, -1) > DO i = 1, natomwfc > hg_norm=ZDOTC(npw, wfcatom(1,i), 1, swfcatom(1,i), 1) > write( 43, '(i4.4,1x,i1.1,1x,i3.3,1x,2e20.10)') ik, current_spin, i, hg_norm > ENDDO > [...] > > so basically I am calculating <\phi^I_k|S|\phi^I_k> (the S operator > being related to the relaxed norm conservation in the case of ultrasoft > pseudopotentials). I assumed this to be always 1. > > However, running the FeO example in the QE distribution, I obtain a spektrum > of values, most of them between 0.9 and 1.1. The norm of one atomic state of > the example system and its spin-counterpart are spread quite widely, between > 0.3 and 3.85 (?). > > So somehow I get the impression that I do not understand how the atomic > states are normalized in pwscf... > > From marzari at MIT.EDU Wed Jul 18 16:16:25 2007 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 18 Jul 2007 10:16:25 -0400 Subject: [Pw_forum] can't minimize electrons by cp.x in some case In-Reply-To: <469D8476.50400@ncsu.edu> References: <469D8476.50400@ncsu.edu> Message-ID: <469E20B9.6070609@mit.edu> Liping YU wrote: > Dear CP users, > By using cp.x (Version 3.2.2), I successfully run the electronic > minimization for alpha-Al2O3 in the orthorhombic supercell with 0001 > along z. However, for the same system, but with all atoms' original > x-coordinates changed into z-coordinates [ correspondingly, y->x, z->y, > and celldm(1)->celldm(3), celldm(2)->cellcm(1) and celldm(3)->celldm(2) > ], I obtained exploded results in the same calculation of electronic > minimization. The forces in the later calculation are unreasonable huge. > I played with restart_mode, dt, emass, emass_cutoff, electro_dynamics > and orthogonalization, but obtained the similar results for the later. > Does anybody have an idea on it? Thanks a lot! Dear Liping, this is interesting. It looks like your cell was rotated correctly (esr and eself are the same). Note that you are using conjugate gradients, so emass and dt and electron_dynamics should not matter (those are for Car-Parrinello). Starting from scratch, and with Gram-Schmidt orthogonalization, the only thing that could go wrong is that your initial step in the first steepest descent direction is too large (or too small), so that the quadratic minimization is not working (cg takes an electronic step in the direction of the steepest descent/conj grad for the electrons, and with the initial value of the energy, initial value of the derivate of the energy with respect to the search direction, and value of the energy for a finite step, it guesses by parabolic interpolation where it should go (cg_sub.f90). Can you send me the first ~20 lines of convergence.dat, for the run you did, starting from scratch, with the Gram-Schnmidt ortho ? This is just a reminder for all of us that the CG/ensemble-DFT output should be made more clear. 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 2586534 marzari at mit.edu http://quasiamore.mit.edu From niuli1978 at yahoo.com.cn Wed Jul 18 16:36:41 2007 From: niuli1978 at yahoo.com.cn (li niu) Date: Wed, 18 Jul 2007 22:36:41 +0800 (CST) Subject: [Pw_forum] decomoposition of the atomic contribution to the vibrational spectrum In-Reply-To: Message-ID: <860446.74151.qm@web15008.mail.cnb.yahoo.com> Niu Li wrote, Thank you Axel, Huiqun Zhou and Kostya to the replies to my previous e-mail. Thanks! best regards, Niu Li Harbin Institue of Technology People's Republic of China --------------------------------- ????????-3.5G???20M??? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070718/5471674b/attachment.htm From lyu7 at ncsu.edu Wed Jul 18 16:55:57 2007 From: lyu7 at ncsu.edu (Liping YU) Date: Wed, 18 Jul 2007 10:55:57 -0400 Subject: [Pw_forum] can't minimize electrons by cp.x in some case In-Reply-To: <469E20B9.6070609@mit.edu> References: <469D8476.50400@ncsu.edu> <469E20B9.6070609@mit.edu> Message-ID: <469E29FD.1030707@ncsu.edu> Nicola Marzari wrote: > Liping YU wrote: > >> Dear CP users, >> By using cp.x (Version 3.2.2), I successfully run the electronic >> minimization for alpha-Al2O3 in the orthorhombic supercell with 0001 >> along z. However, for the same system, but with all atoms' original >> x-coordinates changed into z-coordinates [ correspondingly, y->x, z->y, >> and celldm(1)->celldm(3), celldm(2)->cellcm(1) and celldm(3)->celldm(2) >> ], I obtained exploded results in the same calculation of electronic >> minimization. The forces in the later calculation are unreasonable huge. >> I played with restart_mode, dt, emass, emass_cutoff, electro_dynamics >> and orthogonalization, but obtained the similar results for the later. >> Does anybody have an idea on it? Thanks a lot! >> > > > Dear Liping, > > > this is interesting. It looks like your cell was rotated correctly > (esr and eself are the same). > > Note that you are using conjugate gradients, so emass and dt and > electron_dynamics should not matter (those are for Car-Parrinello). > > Starting from scratch, and with Gram-Schmidt orthogonalization, the only > thing that could go wrong is that your initial step in the first > steepest descent direction is too large (or too small), so that > the quadratic minimization is not working (cg takes an electronic > step in the direction of the steepest descent/conj grad for the > electrons, and with the initial value of the energy, initial > value of the derivate of the energy with respect to the search > direction, and value of the energy for a finite step, it guesses > by parabolic interpolation where it should go (cg_sub.f90). > > Can you send me the first ~20 lines of convergence.dat, for the run > you did, starting from scratch, with the Gram-Schnmidt ortho ? > > Dear Dr. Marzari, Thanks for your help. Here are the lines from convergence.dat (starting from scratch and with Gram-Schimidt ortho). As you said, there is something wrong with the initial step guess. Liping ------------------ later case: ------------------ 1 302.959497331307261 0.000000000000000000E+00 0.000000000000000000E+00 CG1 302.9594973313************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 2 208.080403140153123 0.000000000000000000E+00 0.000000000000000000E+00 CG1 208.0804031402************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 3 114.115527389165749 0.000000000000000000E+00 0.000000000000000000E+00 CG1 114.1155273892************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 4 224.448589174938689 0.000000000000000000E+00 0.000000000000000000E+00 CG1 224.4485891749************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 5 -189.519897289624879 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -189.5198972896************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 6 -202.822713906861907 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -202.8227139069************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 7 -123.127813772902329 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -123.1278137729************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 8 285.742937450055365 0.000000000000000000E+00 0.000000000000000000E+00 CG1 285.7429374501************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 9 180.849983087926148 0.000000000000000000E+00 0.000000000000000000E+00 CG1 180.8499830879************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 10 138.998489312365251 0.000000000000000000E+00 0.000000000000000000E+00 CG1 138.9984893124************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 11 397.002136613081007 0.000000000000000000E+00 0.000000000000000000E+00 CG1 397.0021366131************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 12 573.437725667723953 0.000000000000000000E+00 0.000000000000000000E+00 CG1 573.4377256677************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 13 606.704262898569027 0.000000000000000000E+00 0.000000000000000000E+00 CG1 606.7042628986************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 14 253.128716475577534 0.000000000000000000E+00 0.000000000000000000E+00 CG1 253.1287164756************************************************************ CG2 1.0000000 0.3000000 0.0000000********** 15 247.113139577618142 0.000000000000000000E+00 0.000000000000000000E+00 CG1 247.1131395776 NaNQ NaNQ -NaNQ CG2 1.0000000 0.3000000 -NaNQ -NaNQ 16 247.113139577618142 0.000000000000000000E+00 0.000000000000000000E+00 CG1 247.1131395776 NaNQ NaNQ NaNQ CG2 1.0000000 -NaNQ -NaNQ NaNQ 17 247.113139577618142 0.000000000000000000E+00 0.000000000000000000E+00 CG1 247.1131395776 NaNQ NaNQ NaNQ CG2 1.0000000 -NaNQ -NaNQ NaNQ 18 247.113139577618142 0.000000000000000000E+00 0.000000000000000000E+00 CG1 247.1131395776 NaNQ NaNQ NaNQ CG2 1.0000000 -NaNQ -NaNQ NaNQ 19 247.113139577618142 0.000000000000000000E+00 0.000000000000000000E+00 CG1 247.1131395776 NaNQ NaNQ NaNQ CG2 1.0000000 -NaNQ -NaNQ NaNQ --------------------------------------------------- the former case: --------------------------------------------------- 1 297.040229170397595 0.000000000000000000E+00 0.000000000000000000E+00 CG1 297.0402291704 -327.1583825617 -386.8660741503 -363.1527920080 CG2 1.0000000 0.3000000 0.4258172 0.4826634 2 -363.152792007997846 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -363.1527920080 -272.6635984330 -798.6105501259 -287.3075660536 CG2 1.0000000 0.8516344 0.4057333-0.0870868 3 -416.146500451263023 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -416.1465004513 115.7092400707 -868.5552470882 -89.9981814793 CG2 1.0000000 0.3000000 0.1212124-0.3604576 4 -512.223995671078455 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -512.2239956711 -141.8478201946 -613.6722624131 -497.0361487692 CG2 1.0000000 0.3000000 0.0950392-0.0748551 5 -554.138403096254137 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -554.1384030963 -533.7970198382 -609.0294320983 -587.0819504104 CG2 1.0000000 0.3000000 0.1382032 0.3000813 6 -587.081950410440527 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -587.0819504104 -377.3688251167 -678.6723303982 -634.8275753183 CG2 1.0000000 0.2764064 0.0982342 0.2606476 7 -634.827575318320669 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -634.8275753183 -676.7993942356 -677.2199851317 -677.2082852143 CG2 1.0000000 0.1964684 0.2182027 0.4998620 8 -677.208285214284047 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -677.2082852143 -676.0642528498 -704.3739416357 -702.5816388666 CG2 1.0000000 0.4364055 0.2159526 0.4670116 9 -702.581638866591447 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -702.5816388666 -700.5851296605 -715.2442264718 -713.9268025492 CG2 1.0000000 0.4319052 0.2080517 0.4479797 10 -713.926802549238118 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -713.9268025492 -707.4353164568 -719.9881555276 -719.5109712385 CG2 1.0000000 0.4161035 0.1705982 0.4606371 11 -719.510971238478533 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -719.5109712385 -722.3990664141 -723.7343003271 -723.7523936499 CG2 1.0000000 0.3411965 0.2183968 0.5021421 12 -723.752393649912165 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -723.7523936499 -723.1457809956 -726.2415588235 -726.2158213828 CG2 1.0000000 0.4367935 0.2065009 0.4948301 13 -726.215821382789727 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -726.2158213828 -726.7889039815 -727.8122697250 -727.8224784840 CG2 1.0000000 0.4130018 0.2293637 0.5031973 14 -727.822478484035628 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -727.8224784840 -727.6146930470 -728.9700744499 -728.9856879341 CG2 1.0000000 0.4587275 0.2198269 0.5068027 15 -728.985687934085718 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -728.9856879341 -728.9412069499 -729.8090884078 -729.8214151150 CG2 1.0000000 0.4396539 0.2169357 0.5074852 16 -729.821415115000491 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -729.8214151150 -729.7308456310 -730.4596782049 -730.4646708176 CG2 1.0000000 0.4338714 0.2097418 0.5039111 17 -730.464670817577371 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -730.4646708176 -730.3411536991 -730.9289569303 -730.9318604187 CG2 1.0000000 0.4194837 0.1973871 0.5031268 18 -730.931860418665337 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -730.9318604187 -730.9724441314 -731.3171886586 -731.3230804993 CG2 1.0000000 0.3947742 0.2028776 0.5076452 19 -731.323080499311345 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -731.3230804993 -731.2069299467 -731.6544765905 -731.6641691423 CG2 1.0000000 0.4057551 0.1876667 0.5146238 20 -731.664169142343098 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -731.6641691423 -731.5335053634 -731.9343749836 -731.9396539059 CG2 1.0000000 0.3753334 0.1692201 0.5097683 21 -731.939653905895852 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -731.9396539059 -732.3534680804 -732.3539154273 -732.3538954898 CG2 1.0000000 0.3000000 0.3101934 0.4999759 22 -732.353895489767524 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.3538954898 -732.5815423493 -732.6376776645 -732.6389847191 CG2 1.0000000 0.6203867 0.4294048 0.5023029 23 -732.638984719072710 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.6389847191 -732.4570455855 -732.7456255078 -732.7450664851 CG2 1.0000000 0.8588095 0.3246891 0.4973789 24 -732.745066485118855 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7450664851 -732.7433421661 -732.7703120338 -732.7703396748 CG2 1.0000000 0.6493783 0.3193265 0.5005474 25 -732.770339674779962 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7703396748 -732.7708818332 -732.7783288661 -732.7783271028 CG2 1.0000000 0.6386531 0.3249360 0.4998896 26 -732.778327102833373 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7783271028 -732.7786415094 -732.7817093070 -732.7817095054 CG2 1.0000000 0.6498721 0.3328603 0.5000293 27 -732.781709505350022 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7817095054 -732.7822523959 -732.7830161436 -732.7830162805 CG2 1.0000000 0.6657206 0.3772782 0.5000524 28 -732.783016280478023 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7830162805 -732.7825486578 -732.7833716066 -732.7833715616 CG2 1.0000000 0.7545565 0.2992071 0.4999366 29 -732.783371561580680 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7833715616 -732.7834898682 -732.7835395086 -732.7835395165 CG2 1.0000000 0.5984142 0.3876581 0.5000236 30 -732.783539516533210 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7835395165 -732.7835156018 -732.7836122459 -732.7836122427 CG2 1.0000000 0.7753162 0.3601526 0.4999777 31 -732.783612242669847 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836122427 -732.7836121658 -732.7836405694 -732.7836405702 CG2 1.0000000 0.7203052 0.3599086 0.5000140 32 -732.783640570163243 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836405702 -732.7836367298 -732.7836515737 -732.7836515736 CG2 1.0000000 0.7198172 0.3330223 0.4999956 33 -732.783651573598377 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836515736 -732.7836527383 -732.7836560289 -732.7836560289 CG2 1.0000000 0.6660447 0.3582026 0.4999969 34 -732.783656028890164 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836560289 -732.7836555063 -732.7836575386 -732.7836575387 CG2 1.0000000 0.7164051 0.3316316 0.5000064 35 -732.783657538653642 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836575387 -732.7836574543 -732.7836580830 -732.7836580830 CG2 1.0000000 0.6632631 0.3196973 0.5000009 36 -732.783658083049204 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836580830 -732.7836581336 -732.7836582713 -732.7836582713 CG2 1.0000000 0.6393946 0.3446389 0.5000137 37 -732.783658271287095 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836582713 -732.7836582713 -732.7836583424 -732.7836583424 CG2 1.0000000 0.6892778 0.3446758 0.5000177 38 -732.783658342354329 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836583424 -732.7836583406 -732.7836583677 -732.7836583677 CG2 1.0000000 0.6893517 0.3389761 0.4999944 39 -732.783658367739577 0.000000000000000000E+00 0.000000000000000000E+00 CG1 -732.7836583677 -732.7836583685 -732.7836583779 -732.7836583779 CG2 1.0000000 0.6779523 0.3458048 0.5000547 > This is just a reminder for all of us that the CG/ensemble-DFT output > should be made more clear. > > nicola > > > -- Liping YU, PhD Candidate Department of Physics & Center for High Performance Simulation North Carolina State University, Raleigh, NC 27695, USA tel: 1-919-5157228 From marzari at MIT.EDU Wed Jul 18 17:14:50 2007 From: marzari at MIT.EDU (Nicola Marzari) Date: Wed, 18 Jul 2007 11:14:50 -0400 Subject: [Pw_forum] can't minimize electrons by cp.x in some case In-Reply-To: <469E29FD.1030707@ncsu.edu> References: <469D8476.50400@ncsu.edu> <469E20B9.6070609@mit.edu> <469E29FD.1030707@ncsu.edu> Message-ID: <469E2E6A.90400@mit.edu> Dear Liping, excellent. Let's move this discussion to personal email, but just start by using a smaller passop. Let me comment on what the code writes, below: > ------------------ > later case: > ------------------ > 1 302.959497331307261 0.000000000000000000E+00 0.000000000000000000E+00 > CG1 > 302.9594973313************************************************************ > CG2 1.0000000 0.3000000 0.0000000********** > 2 208.080403140153123 0.000000000000000000E+00 0.000000000000000000E+00 > CG1 > 208.0804031402************************************************************ > CG2 1.0000000 0.3000000 0.0000000********** > --------------------------------------------------- > the former case: > --------------------------------------------------- > 1 297.040229170397595 0.000000000000000000E+00 0.000000000000000000E+00 > CG1 297.0402291704 -327.1583825617 -386.8660741503 -363.1527920080 > CG2 1.0000000 0.3000000 0.4258172 0.4826634 > 2 -363.152792007997846 0.000000000000000000E+00 0.000000000000000000E+00 > CG1 -363.1527920080 -272.6635984330 -798.6105501259 -287.3075660536 > CG2 1.0000000 0.8516344 0.4057333-0.0870868 Let's discuss former case - the first three lines refer to the first CG step. The total energy is 297.040..., from random wavefunctions. The code takes an initial step of 0.3 in the search direction, the new energy is -327.158, and with the parabolic interpolation it decides that the perfect step should have been 0.4258... , with a predicted value of the energy of -386.866 . Instead, when the energy is recalculated for a 0.4258 step, it finds -363.152. This is still good, but in the first steps the problem is highly non-quadratic, and that is what the fourth number of the third line shows (for a perfect parabola, the ratio between the energy decrease predicted by a quadratic interpolator/extrapolator should be exactly 0.5 of the energy decrease predicted by the linear extrapolator - in the later steps, this fourth number is close to 0.5). CG step 2 - you start from the energy calculated before (-363.152...) and take an optimal step (assuming that the curvature of the parabola has not changed). This step of 0.85 in the search direction gives you a new energy at the minimum (extrapolated !) of -798, and a minimum coordinate of 0.40. Once the code goes there, it finds only -287, still good, etc... Note that the first few steps are highly nonlinear, but quickly it settles down. 1) solution for you - to use a smaller passop for the code, 0.1 or smaller, and see how things change. 2) suggestion to improve the code - the initial step could always be fixed, or only midly reduced or augmented, with caution, following the trend from previous steps. I think moving from 0.3 to 0.85 from one iteration to the next is too much. 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 2586534 marzari at mit.edu http://quasiamore.mit.edu From giannozz at nest.sns.it Wed Jul 18 19:20:09 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Wed, 18 Jul 2007 19:20:09 +0200 Subject: [Pw_forum] can't minimize electrons by cp.x in some case In-Reply-To: <469D8476.50400@ncsu.edu> References: <469D8476.50400@ncsu.edu> Message-ID: <13ECE92B-E335-44D3-B8A7-4038E68D8F7D@nest.sns.it> On Jul 18, 2007, at 5:09 , Liping YU wrote: > By using cp.x (Version 3.2.2), I successfully run the electronic > minimization for alpha-Al2O3 in the orthorhombic supercell with 0001 > along z. However, for the same system, but with all atoms' original > x-coordinates changed into z-coordinates [ correspondingly, y->x, z- > >y, > and celldm(1)->celldm(3), celldm(2)->cellcm(1) and celldm(3)->celldm > (2) > ], I obtained exploded results in the same calculation of electronic > minimization. does this happen only with 'cg', or also in other cases? P. --- Paolo Giannozzi, Democritos and University of Udine, Italy From ezadshojaee at hotmail.com Thu Jul 19 11:27:31 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Thu, 19 Jul 2007 09:27:31 +0000 Subject: [Pw_forum] problem with parallel mode Message-ID: Hi i have this error message while running ph.x in the parallel mode: MPI_Recv: message truncated (rank 0, comm 9)Rank (0, MPI_COMM_WORLD): Call stack within LAM:Rank (0, MPI_COMM_WORLD): - MPI_Recv()Rank (0, MPI_COMM_WORLD): - main() I can run MPI successfully.The specs:1- machine=AMD opteron dualcore64bit,2- OS=CentOS64bit 4.4,3- compiler=intel fortran9.1,4- MPI= lamMPI 7.1.3,5- espresso=3.2and my input runs in serial without problem. Any suggestion will be appreciated! _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070719/85dc65b9/attachment.htm From giannozz at nest.sns.it Thu Jul 19 14:35:02 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 19 Jul 2007 14:35:02 +0200 Subject: [Pw_forum] problem with parallel mode In-Reply-To: References: Message-ID: On Jul 19, 2007, at 11:27 , Ezad Shojaee wrote: > i have this error message while running ph.x in the parallel mode: and what happened to the error you reported a few days ago? P. --- Paolo Giannozzi, Democritos and University of Udine, Italy From ezadshojaee at hotmail.com Thu Jul 19 15:14:50 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Thu, 19 Jul 2007 13:14:50 +0000 Subject: [Pw_forum] problem with parallel mode Message-ID: Hi I'm glad you remind it! that is still on and it appears when the later error message seems to be gone!(the message seems to be about MPI) Thanx _________________________________________________________________ Make every IM count. Download Windows Live Messenger and join the i?m Initiative now. It?s free.? http://im.live.com/messenger/im/home/?source=TAGWL_June07 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070719/e79f6483/attachment.htm From emenendez at macul.ciencias.uchile.cl Thu Jul 19 16:00:12 2007 From: emenendez at macul.ciencias.uchile.cl (Eduardo Ariel Menendez P) Date: Thu, 19 Jul 2007 10:00:12 -0400 (CLT) Subject: [Pw_forum] electrical conductivity of metallic surfaces Message-ID: Hi, I wonder if it is possible to calculate the electrical conductivity of a metallic surface, lets say, gold. In principle it should be possible calculating the Fermi surface and using Boltzmann's kinetic equation, but considering the amount of atoms needed for a surface calculation this Fermi surface should be extremely nested. a) Is it practical to obtain such a Fermi surface with espresso and xcrysden? b) Is it sound (theoretically) to calculate the surface's conductivity (for currents parallel to the surface) using the Boltzmann equation formalism? Thanks Eduardo Eduardo A. Menendez Proupin Department of Physics Faculty of Science University of Chile Las Palmeras 3425 ?u?oa, Santiago Chile Phone: (56)(2) 978 74 11 Fax : (56)(2) 271 29 73 http://fisica.ciencias.uchile.cl/~emenendez/ From akohlmey at cmm.chem.upenn.edu Fri Jul 20 07:28:28 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Fri, 20 Jul 2007 01:28:28 -0400 Subject: [Pw_forum] problem with parallel mode In-Reply-To: References: Message-ID: <7b6913e90707192228h139ac5d0v29556a6372fd6512@mail.gmail.com> On 7/19/07, Ezad Shojaee wrote: > > Hi > I'm glad you remind it! that is still on and it appears when the later > error message seems to be gone!(the message seems to be about MPI) ezad, ...and still you are ignoring paolo's request of posting an easy to follow way to _reproduce_ what you are seeing. with a few (generic) lines of error output and some snippets of program output, one has to be a psychic to debug your problems. there can be thousands of reasons, ranging from you doing something very stupid to having found a real bug. so until you give people a chance to debug your problems, you'll be on your own. cheers, axel. > Thanx > > ________________________________ > Play free games, earn tickets, get cool prizes! Join Live Search Club. > Join Live Search Club! > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From glaweh at physik.fu-berlin.de Fri Jul 20 13:37:46 2007 From: glaweh at physik.fu-berlin.de (Henning Glawe) Date: Fri, 20 Jul 2007 13:37:46 +0200 Subject: [Pw_forum] Normalization of the atomic wave functions In-Reply-To: <469DD6E0.20607@sissa.it> References: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> <469DD6E0.20607@sissa.it> Message-ID: <20070720113746.GA4013@ruprecht.physik.fu-berlin.de> On Wed, Jul 18, 2007 at 11:01:20AM +0200, degironc wrote: > The projection in LDA+U is however defined on the individual atoms (even > if technically bloch states are involved) and these > wfc are normalized. > hope this helps, ok, in the pen-and-paper check this seems ok. The reason for my original question was: In my system (CaCuO2 in an 2D-AFM sqrt(2)*sqrt(2)*1 unit cell), when using atomic states as the projectors for the projectors, I am getting a lot of matrix elements of the occupation matrix significantly larger than 1 (between 4 and 5(!)). The Hubbard term in the total energy, as I understand from Matteo Cococcioni's thesis (and the meaning of the word 'occupation'), seems only well-defined as long as these matrix elements are between 0 and 1. Now I am trying to investigate where these occupation values are coming from and started by looking at the single scalar products used in the calculation of the n. Finding out that already these values are too big (i.e. larger than 1) in my system, I tried to investigate the normalization of the atomic wave functions and checked also the known-working example of FeO. In the FeO example, the occupation matrix elements are also slightly bigger than 1 (1.002 in feo_standard.out), but I was thinking that this was related to numerical errors. Now I am asking myself (and the forum ;) ): is there any part in the formalism and/or code which restricts the occupation to the right regime i.e. 0..1 ? -- c u henning From matteo at umn.edu Fri Jul 20 17:13:21 2007 From: matteo at umn.edu (Matteo Cococcioni) Date: Fri, 20 Jul 2007 10:13:21 -0500 Subject: [Pw_forum] Normalization of the atomic wave functions In-Reply-To: <20070720113746.GA4013@ruprecht.physik.fu-berlin.de> References: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> <469DD6E0.20607@sissa.it> <20070720113746.GA4013@ruprecht.physik.fu-berlin.de> Message-ID: <46A0D111.2070208@umn.edu> Dear Henning, the problem you report is probably due to the fact that some pseudo potentials contain non-normalized atomic wave functions. while this doesn't have any effect on "normal" (non LDA+U) because the normalization of the Kohn-Sham orbitals is always imposed, in the case of LDA+U calculation this may lead to meaningless occupations like the ones you find. The problem is disappearing with the generation of pseudo potentials with normalized atomic wfcs. But few unnormalized ones are still around. For the moment a quick fix would be to impose the normalization of atomic wfc inside orthoatwfc.f90. Attached to this message please find a modified version of this routine that is able to do this. Be ware this has been obtained modifying the routin in the cvs code so may be different from the one you are using now. To activate the normalization of at wfc in the input file you should use: U_projection=="norm-atomic". Please keep in mind this is a quick fix to your problem. This normalization probably introduces a dependence of the energy on the lattice parameters that is not accounted for in the calculation of stresses. If in "normal" (non problematic) LDA+U calculations give occupations slightly bigger than 1 (e.g. 1.002) this is due to lack of orthogonalization of the atomic wfc of one atomic sites with those belonging to neighbor sites. It's probably not terribly important untill it gets to bigger values (e.g. 1.5) as it may happen with extended orbitals as Stefano de Gironcoli wrote. Hope this helps, Matteo Henning Glawe wrote: > On Wed, Jul 18, 2007 at 11:01:20AM +0200, degironc wrote: > >> The projection in LDA+U is however defined on the individual atoms (even >> if technically bloch states are involved) and these >> wfc are normalized. >> hope this helps, >> > > ok, in the pen-and-paper check this seems ok. > The reason for my original question was: In my system (CaCuO2 in an > 2D-AFM sqrt(2)*sqrt(2)*1 unit cell), when using atomic states as the > projectors for the projectors, I am getting a lot of matrix elements of the > occupation matrix significantly larger than 1 (between 4 and 5(!)). > > The Hubbard term in the total energy, as I understand from Matteo > Cococcioni's thesis (and the meaning of the word 'occupation'), seems only > well-defined as long as these matrix elements are between 0 and 1. > > Now I am trying to investigate where these occupation values are coming > from and started by looking at the single scalar products used > in the calculation of the n. Finding out that already these values are too > big (i.e. larger than 1) in my system, I tried to investigate the > normalization of the atomic wave functions and checked also the > known-working example of FeO. > > In the FeO example, the occupation matrix elements are also slightly bigger > than 1 (1.002 in feo_standard.out), but I was thinking that this was related > to numerical errors. Now I am asking myself (and the forum ;) ): is there any > part in the formalism and/or code which restricts the occupation to the > right regime i.e. 0..1 ? > > -------------- next part -------------- A non-text attachment was scrubbed... Name: orthoatwfc.f90 Type: text/x-fortran Size: 6493 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20070720/91663b75/attachment.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: matteo.vcf Type: text/x-vcard Size: 294 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20070720/91663b75/attachment.vcf From nizar.aouni at gmail.com Mon Jul 23 18:51:56 2007 From: nizar.aouni at gmail.com (nizar aouni) Date: Mon, 23 Jul 2007 11:51:56 -0500 Subject: [Pw_forum] No symmetry! output file message Message-ID: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> Dear PWSCF Users, I run SCF calculations on quartz successefully. When looking to the output file, there is a message " No symmetry!" and I am using trigonal symmetry. Could you please tell me where is the error if it exists. The input file is: &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '/home/pseudo/' outdir= '/home/pwscftemp/' prefix='quartz' tstress = .true. tprnfor = .true. forc_conv_thr = 8.0D-2 nstep = 10 / &system ibrav = 4, celldm(1) =9.2838, celldm(3) =1.1000, nat= 6, ntyp=2, ecutwfc = 30.0, ecutrho = 150.0 / &electrons conv_thr = 1.0e-6 mixing_beta = 0.5 / ATOMIC_SPECIES Si 28.086 Si.pbe-rrkj.UPF O 15.9994 O.pbe-rrkjus.UPF ATOMIC_POSITIONS O 0.41211 0.26563 0.11865 O 0.14648 -0.26563 -0.11865 O 0.73438 0.14648 -0.21468 O 0.58789 -0.14648 0.21468 Si 0.46875 0.00000 0.00000 Si 0.53125 0.53125 0.33333 K_POINTS (Automatic) 2 2 2 1 1 1 Thank you in advanace for your help. -- --------------------------------------------- Aouni Nizar PhD Student Department of Mechanical Engineering, University of Houston ------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070723/995a1b41/attachment.htm From nanomate at gmail.com Tue Jul 24 09:24:07 2007 From: nanomate at gmail.com (Kenneth Wong) Date: Tue, 24 Jul 2007 17:24:07 +1000 Subject: [Pw_forum] davcio #10 error for projwfc.x Message-ID: <68e1226a0707240024s55aac5f8wc07cac6dcd07dba7@mail.gmail.com> Dear PWscf, I tried the projwfc.x on two of my own different pw.xcalculations but i keep getting this error message Calling projwave .... IOS = ** from davcio : error # 10 i/o error in davcio i have tried the dos.x calculation and that has worked properly, i have tried to find a soluation myself but this was all i could find from a previous query relating to this same problem - "it depends. The above error means that the code could not read some data files, presumably because their format was not what was expected. In order to find out why, one needs to know exactly under which (reproducible) circumstances this happens." - Paolo Giannozzi I can't seem to work out whats wrong, this is my projwfc input &inputpp prefix='pwscf', outdir='/data/kwong/single2', DeltaE=0.1, Emax=20.0, Emin=-20.0, ngauss=1, filpdos='pwscf.pdos', / thanks, Ken. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070724/80583fa4/attachment.htm From giannozz at nest.sns.it Tue Jul 24 09:49:01 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 24 Jul 2007 09:49:01 +0200 Subject: [Pw_forum] No symmetry! output file message In-Reply-To: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> References: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> Message-ID: <4B4E8BFC-7BD9-446D-8EA9-64F786D257A1@nest.sns.it> On Jul 23, 2007, at 18:51 , nizar aouni wrote: > When looking to the output file, there is a message " No symmetry!" > and I am using trigonal symmetry. > Could you please tell me where is the error if it exists. it doesn't exist, and everything is explained in detail here: http://www.quantum-espresso.org/wiki/index.php/Troubleshooting_% 28PWscf%29 sect.1.21 and 1.22 > forc_conv_thr = 8.0D-2 too large Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Tue Jul 24 10:16:08 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 24 Jul 2007 10:16:08 +0200 Subject: [Pw_forum] davcio #10 error for projwfc.x In-Reply-To: <68e1226a0707240024s55aac5f8wc07cac6dcd07dba7@mail.gmail.com> References: <68e1226a0707240024s55aac5f8wc07cac6dcd07dba7@mail.gmail.com> Message-ID: <8E00CDE8-8EE6-4891-B9A0-1C5C17E3EDFD@nest.sns.it> On Jul 24, 2007, at 9:24 , Kenneth Wong wrote: > i have tried to find a solution myself that's an excellent initiative > but this was all i could find from a previous query relating to > this same problem - > "it depends. The above error means that the code could not read > some data files, presumably because their format was not what > was expected. In order to find out why, one needs to know > exactly under which (reproducible) circumstances this happens." "under which (reproducible) circumstances" means: - for which code version - for which compiler - in serial or parallel execution - whether the tests and examples work or not - whether the error is reproducible - whether the error always occurs or only in some specific cases, and which cases: - for all runs/input data or only for some specific runs/input data - for more than one hardware/software (compiler/libraries) configuration, or only on a specific one In any event, check if files "outdir"/"prefix".wfc (produced by the previous scf or non-scf run) exist Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From glaweh at physik.fu-berlin.de Wed Jul 25 12:03:43 2007 From: glaweh at physik.fu-berlin.de (Henning Glawe) Date: Wed, 25 Jul 2007 12:03:43 +0200 Subject: [Pw_forum] Normalization of the atomic wave functions In-Reply-To: <46A0D111.2070208@umn.edu> References: <20070718080752.GA4036@ruprecht.physik.fu-berlin.de> <469DD6E0.20607@sissa.it> <20070720113746.GA4013@ruprecht.physik.fu-berlin.de> <46A0D111.2070208@umn.edu> Message-ID: <20070725100343.GA16837@physik.fu-berlin.de> On Fri, Jul 20, 2007 at 10:13:21AM -0500, Matteo Cococcioni wrote: > the problem you report is probably due to the fact that some pseudo > potentials contain non-normalized atomic > wave functions. while this doesn't have any effect on "normal" (non > LDA+U) because the normalization > of the Kohn-Sham orbitals is always imposed, in the case of LDA+U > calculation this may lead to meaningless > occupations like the ones you find. The problem is disappearing with the > generation of pseudo potentials with > normalized atomic wfcs. But few unnormalized ones are still around. I already suspected something like this ;) I backported your file to the 3.2.2, I'll attach it to this mail in case anyone else is interested. I did a quick check with the U=0 case, and indeed the occupation numbers are reasonable (i.e. strictly below 1). Thanks a lot! -- c u henning -------------- next part -------------- A non-text attachment was scrubbed... Name: cococcioni-ldau_norm-atomic_HG-backport-3.2.2.patch Type: text/x-diff Size: 1380 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20070725/bfa2014f/attachment.patch From nizar.aouni at gmail.com Wed Jul 25 22:10:41 2007 From: nizar.aouni at gmail.com (nizar aouni) Date: Wed, 25 Jul 2007 15:10:41 -0500 Subject: [Pw_forum] No symmetry! output file message In-Reply-To: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> References: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> Message-ID: <73f71e150707251310l507cc8f5s3d943ee13d3547@mail.gmail.com> Hi, The output file contains the following: bravais-lattice index = 4 lattice parameter (a_0) = 9.2838 a.u. unit-cell volume = 762.2556 (a.u.)^3 number of atoms/cell = 6 number of atomic types = 2 kinetic-energy cutoff = 30.0000 Ry charge density cutoff = 150.0000 Ry convergence threshold = 1.0E-06 beta = 0.5000 number of iterations used = 8 plain mixing Exchange-correlation = SLA PW PBE PBE (1434) celldm(1)= 9.283800 celldm(2)= 1.000000 celldm(3)= 1.100000 celldm(4)= 0.000000 celldm(5)= 0.000000 celldm(6)= 0.000000 crystal axes: (cart. coord. in units of a_0) a(1) = ( 1.000000 0.000000 0.000000 ) a(2) = ( -0.500000 0.866025 0.000000 ) a(3) = ( 0.000000 0.000000 1.100000 ) reciprocal axes: (cart. coord. in units 2 pi/a_0) b(1) = ( 1.000000 0.577350 0.000000 ) b(2) = ( 0.000000 1.154701 0.000000 ) b(3) = ( 0.000000 0.000000 0.909091 ) PSEUDO 1 is Si zval = 4.0 lmax= 2 lloc= 0 (in numerical form: 883 grid points, xmin = 0.00, dx = 0.0000) PSEUDO 2 is O (US) zval = 6.0 lmax= 2 lloc= 0 Version 0 0 0 of US pseudo code Using log mesh of 1269 points The pseudopotential has 4 beta functions with: l(1) = 0 l(2) = 0 l(3) = 1 l(4) = 1 Q(r) pseudized with 0 coefficients, rinner = 0.000 0.000 0.000 0.000 0.000 atomic species valence mass pseudopotential Si 4.00 28.08600 Si( 1.00) O 6.00 15.99940 O ( 1.00) *No symmetry*! Cartesian axes site n. atom positions (a_0 units) 1 O tau( 1) = ( 0.4121100 0.2656300 0.1186500 ) 2 O tau( 2) = ( 0.1464800 -0.2656300 -0.1186500 ) 3 O tau( 3) = ( 0.7343800 0.1464800 -0.2146800 ) 4 O tau( 4) = ( 0.5878900 -0.1464800 0.2146800 ) 5 Si tau( 5) = ( 0.4687500 0.0000000 0.0000000 ) 6 Si tau( 6) = ( 0.5312500 0.5312500 0.3333300 ) Is there any other explanation ? I am getting this error frequently in my output files and when going to submit the phonon calculations I got an error message " i/o davcio", that's why I want to make sure that this message doesn't have any effect on the Ph.x calculations. Thank you vey much for your time, Regards ------------------------------ >* When looking to the output file, there is a message " No symmetry!" *>* and I am using trigonal symmetry. *>* Could you please tell me where is the error if it exists. * it doesn't exist, and everything is explained in detail here: http://www.quantum-espresso.org/wiki/index.php/Troubleshooting_% 28PWscf%29 sect.1.21 and 1.22 >* forc_conv_thr = 8.0D-2 * too large Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy -- --------------------------------------------- Aouni Nizar PhD Student Department of Mechanical Engineering, University of Houston ------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070725/f13b6949/attachment.htm From yinzping at gmail.com Thu Jul 26 02:11:34 2007 From: yinzping at gmail.com (Zhiping Yin) Date: Wed, 25 Jul 2007 17:11:34 -0700 Subject: [Pw_forum] [pw_forum]Problem with pseudopotential generation: the case of Po Message-ID: <523370bf0707251711h1764b64fo692ec3f2206b50b2@mail.gmail.com> Dear PWscf users, I tried to generate a pseudopotential for Po with the ld1 code. I tried to generate both an USPP and a NCPP. The USPP and NCPP were generated with no error/warning messages in the output files. However, when I tried to test the generated pseudopotential, it ran into problems. I had no experience on pseudopotential generation before. Can someone help me to check if there is any mistake with the input files? Thanks very much. The input file for the generation of the USPP: &input title='Po', zed=84.0, rel=2, iswitch=3, rlderiv=2.50, eminld=-5.0, emaxld=5.0, deld=0.02, nld=5, config='[Xe] 4f14. 5d10. 6s2 6p4', dft='PZ' / &inputp pseudotype=3, lloc=0, file_pseudopw='Po.rel.us.5d.RRKJ3', nlcc=.true., / 9 5D 3 2 4.00 0.00 2.20 2.30 1.50 5D 3 2 0.00 -0.20 2.20 2.30 1.50 5D 3 2 6.00 0.00 2.30 2.40 2.50 5D 3 2 0.00 -0.20 2.30 2.40 2.50 6P 2 1 2.00 0.00 2.80 2.80 0.50 6P 2 1 0.00 -0.20 2.80 2.80 0.50 6P 2 1 2.00 0.00 2.90 2.90 1.50 6P 2 1 0.00 -0.20 2.90 2.90 1.50 6S 1 0 2.00 0.00 2.50 2.50 0.50 The USPP was generated with no error in the output file. The input for the testing of the USPP: &input title='Po', zed=84.0, rel=2, iswitch=2, rlderiv=2.50, eminld=-5.0, emaxld=5.0, deld=0.02, nld=5, config='[Xe] 4f14. 5d10. 6s2 6p4', dft='PZ' / &test nconf=1 file_pseudo='Po.rel.us.5d.RRKJ3.UPF', configts(1)='4f14. 5d10. 6s2 6p4' / The input file for the generation of the NCPP: &input title='Po', zed=84.0, rel=2, iswitch=3, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, config='[Xe] 4f14. 5d10. 6s2 6p4', dft='PZ' / &inputp pseudotype=2, lloc=0, file_pseudopw='Po.rel.5d.RRKJ3', nlcc=.true., / 5 5D 3 2 4.00 0.00 2.10 2.30 1.50 5D 3 2 6.00 0.00 2.10 2.30 2.50 6P 2 1 2.00 0.00 2.70 2.70 0.50 6P 2 1 2.00 0.00 2.80 2.80 1.50 6S 1 0 2.00 0.00 2.40 2.40 0.50 Again, the NCPP was generated with no error in the output file. The input for the testing of the NCPP: &input title='Po', zed=84.0, rel=2, iswitch=2, rlderiv=2.50, eminld=-4.0, emaxld=4.0, deld=0.02, nld=5, config='[Xe] 4f14. 5d10. 6s2 6p4', dft='PZ' / &test nconf=1 file_pseudo='Po.rel.5d.RRKJ3.UPF', configts(1)='4f14. 5d10. 6s2 6p4' / Sincerely, Zhiping Yin -- Zhiping Yin graduate student department of physics UC Davis -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070725/33dd392f/attachment.htm From akohlmey at cmm.chem.upenn.edu Thu Jul 26 01:58:23 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Wed, 25 Jul 2007 19:58:23 -0400 (EDT) Subject: [Pw_forum] [pw_forum]Problem with pseudopotential generation: the case of Po In-Reply-To: <523370bf0707251711h1764b64fo692ec3f2206b50b2@mail.gmail.com> Message-ID: On Wed, 25 Jul 2007, Zhiping Yin wrote: ZY> Dear PWscf users, dear zhiping yin, ZY> I tried to generate a pseudopotential for Po with the ld1 code. I tried to ZY> generate both an USPP and a NCPP. The USPP and NCPP were generated with ZY> no error/warning messages in the output files. However, when I tried to test ZY> the generated pseudopotential, it ran into problems. I had no experience on ZY> pseudopotential generation before. Can someone help me to check if there is under those circumstance, it is highly recommended to first try to reproduce some existing and easy to test/compare pseudopotentials, so that you can practice your skills in generating them. ZY> any mistake with the input files? Thanks very much. it would have been very helpful, if you had stated _what_ the error was that you get. ZY> &test ZY> nconf=1 ZY> file_pseudo='Po.rel.us.5d.RRKJ3.UPF', ZY> configts(1)='4f14. 5d10. 6s2 6p4' ZY> / my guess would be that the error lies in specifying 4f14 here. you didn't put any f electrons in the valence, why should you need to add them here? please remember, that for transferability tests, you also should test different electron configurations. see the input examples in the atomic_doc/pseudo-test subdirectory. cheers, axel. [...] -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Thu Jul 26 09:20:17 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 26 Jul 2007 09:20:17 +0200 Subject: [Pw_forum] No symmetry! output file message In-Reply-To: <73f71e150707251310l507cc8f5s3d943ee13d3547@mail.gmail.com> References: <73f71e150707230951j3da11f96of66dc1cc6ba3df2b@mail.gmail.com> <73f71e150707251310l507cc8f5s3d943ee13d3547@mail.gmail.com> Message-ID: <463307A9-FFA6-44AF-BFA6-F59661BCC1C7@nest.sns.it> On Jul 25, 2007, at 22:10 , nizar aouni wrote: > I am getting this error frequently in my output files Once again: this is not an error. If the code says that there are no symmetries, there are no symmetries (at least according to the criterion the code uses to decide if a symmetry is there or not). --- Paolo Giannozzi, Democritos and University of Udine, Italy From njuxuyuehua at gmail.com Thu Jul 26 11:00:44 2007 From: njuxuyuehua at gmail.com (xu yuehua) Date: Thu, 26 Jul 2007 17:00:44 +0800 Subject: [Pw_forum] which can be deleted in scratch ? Message-ID: hello all: i have a foolish question .in my scratch,there are a lot of files which suffixes are such bar1, bar2,com1,com2, dvkb31,dvkb32,dwf1,dwf2,ebar1,.ebar2........and file folder *.save. these cost much space .so my question is which of these are useless after finish computing , which can be canceled ? hope your help .thanks -- Xu Yuehua physics Department of Nanjing university China -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070726/b70a5bc0/attachment.htm From helen at fh.huji.ac.il Thu Jul 26 13:36:15 2007 From: helen at fh.huji.ac.il (Helen) Date: Thu, 26 Jul 2007 14:36:15 +0300 Subject: [Pw_forum] Fw: modified explicit exchange potential Message-ID: <009901c7cf79$2cb23f20$cd604084@fh.huji.ac.il> I was wondering if anyone had some answers to these questions about exx.f90? Thanks Helen (Hebrew University, Jerusalem) ----- Original Message ----- From: "Helen" To: "degironc" Sent: Monday, July 23, 2007 2:48 PM Subject: Re: [Pw_forum] modified explicit exchange potential > Hi, > Thanks so much for your reply. I've been looking through exx.f90 and there > are a few fundamental things I don't understand. > > 1. I've read Gygi's paper and I understand you convert the integral into a > term with no divergence and a diverging analytical term which can be > evaluated. > Int g/(x-x0) => Int (g-g0)/(x-x0) + Int(g0/(x-x0)) > as is shown in equation (5) of the paper (PRB 34, 4405). > However in the code it looks like you're just multiplying U by a > factor(8/7). How does this fit in with Gygi and where does 8/7 come from? > > 2. I still don't understand what exxdiv does/calculates? > > Thanks for your help, > Helen From akohlmey at cmm.chem.upenn.edu Thu Jul 26 17:45:22 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Thu, 26 Jul 2007 11:45:22 -0400 (EDT) Subject: [Pw_forum] which can be deleted in scratch ? In-Reply-To: Message-ID: On Thu, 26 Jul 2007, xu yuehua wrote: XY> hello all: XY> i have a foolish question .in my scratch,there are a lot of files which XY> suffixes are such bar1, XY> bar2,com1,com2, dvkb31,dvkb32,dwf1,dwf2,ebar1,.ebar2........and file folder XY> *.save. XY> these cost much space .so my question is which of these are useless XY> after finish computing , if there is no more job running and you don't plan to run any post processing, you can safely delete them. the postprocessing usually only needs the *.save folders. you can control the names of those files with the 'prefix' flag. cheers, axel. XY> which can be canceled ? XY> XY> hope your help .thanks XY> XY> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From naromero at gmail.com Thu Jul 26 23:49:11 2007 From: naromero at gmail.com (Nichols A. Romero) Date: Thu, 26 Jul 2007 17:49:11 -0400 Subject: [Pw_forum] spin-orbit references Message-ID: <6ac064b60707261449v3234e7d8te51985945401e264@mail.gmail.com> Hi, Does anyone know of a good reference (perhaps a student dissertation) that describs the more technical details involved in treating spin-orbit coupling within DFT? (For example, when to treat the system pertubatively, self-consistently, pseudpotential, etc.) Thanks, -- Nichols A. Romero, Ph.D. High Performance Technologies, Inc. DoD User Productivity Enhancement & Technology Transfer (PET) Group Reston, VA 443-567-8328 (C) 410-278-2692 (O) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070726/e653a3be/attachment.htm From luotengf at msu.edu Fri Jul 27 00:13:58 2007 From: luotengf at msu.edu (Tengfei Luo) Date: Thu, 26 Jul 2007 18:13:58 -0400 Subject: [Pw_forum] How to compare total energy calculation results with references? References: Message-ID: <021601c7cfd2$43bd4df0$4e8a0923@engineering.egr.msu.edu> Dear all: I'm new to PWscf. I tried the first example which includes the total energy calculation of Si. The result I got from this example was: ! total energy = -15.84452726 Ry I believe this is for the total energy of the two Si atoms system. However, from the reference data here: http://physics.nist.gov/PhysRefData/DFTdata/Tables/14Si.html The total energy is -288.198397 a.u.. The caluclated data is very different from this. Is this because of using of pesudopotential? How can I get the real totoal energy? I would appreciate any help! Thank you! Tengfei -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070726/292722fb/attachment.htm From yinzping at gmail.com Fri Jul 27 00:50:29 2007 From: yinzping at gmail.com (Zhiping Yin) Date: Thu, 26 Jul 2007 15:50:29 -0700 Subject: [Pw_forum] [pw_forum] how to set up spin-orbital coupling calculation in PWscf Message-ID: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> Dear PWscf users, I tried to set up spin-orbital coupling calculation in PWscf without spin polarization. I used a full relativistic NCPP and set lspinorb=.true.. However, the output is identical with lspinorb=.false. I wonder what is the correct way to set up a spin-orbital coupling calculation with/without spin polarization. What is the requirement with the pseudopotential. Thanks very much. Best, Zhiping Yin -- Zhiping Yin graduate student department of physics UC Davis -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070726/0c517c01/attachment.htm From baroni at sissa.it Fri Jul 27 08:22:03 2007 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 27 Jul 2007 08:22:03 +0200 Subject: [Pw_forum] [pw_forum] how to set up spin-orbital coupling calculation in PWscf In-Reply-To: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> References: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> Message-ID: <16935E3F-8524-4F73-AC54-8478B42E5EC4@sissa.it> I am not sure I understand the question, nor would I probably be able to answer if I did. However, you may want to think about the fact that in the presence of SO coupling, a system is always (locally) spin polarized. You see the point? Stefano On Jul 27, 2007, at 12:50 AM, Zhiping Yin wrote: > Dear PWscf users, > > I tried to set up spin-orbital coupling calculation in PWscf > without spin polarization. I used a full relativistic NCPP and set > lspinorb=.true.. However, the output is identical with > lspinorb=.false. I wonder what is the correct way to set up a spin- > orbital coupling calculation with/without spin polarization. What > is the requirement with the pseudopotential. Thanks very much. > > Best, > Zhiping Yin > > -- > Zhiping Yin > graduate student > department of physics > UC Davis > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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 -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/b8d6e6fa/attachment.htm From baroni at sissa.it Fri Jul 27 08:36:31 2007 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 27 Jul 2007 08:36:31 +0200 Subject: [Pw_forum] spin-orbit references In-Reply-To: <6ac064b60707261449v3234e7d8te51985945401e264@mail.gmail.com> References: <6ac064b60707261449v3234e7d8te51985945401e264@mail.gmail.com> Message-ID: You may want to try this, and references quoted therein. http://www.sissa.it/cm/thesis/2007/moscaconte.pdf Stefano On Jul 26, 2007, at 11:49 PM, Nichols A. Romero wrote: > Hi, > > Does anyone know of a good reference (perhaps a student > dissertation) that > describs the more technical details involved in treating spin-orbit > coupling within > DFT? (For example, when to treat the system pertubatively, > self-consistently, pseudpotential, etc.) > Thanks, > -- > Nichols A. Romero, Ph.D. > High Performance Technologies, Inc. > DoD User Productivity Enhancement & > Technology Transfer (PET) Group > Reston, VA > 443-567-8328 (C) > 410-278-2692 (O) > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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 -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/2f31f0f4/attachment-0001.htm From baroni at sissa.it Fri Jul 27 08:44:35 2007 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 27 Jul 2007 08:44:35 +0200 Subject: [Pw_forum] How to compare total energy calculation results with references? In-Reply-To: <021601c7cfd2$43bd4df0$4e8a0923@engineering.egr.msu.edu> References: <021601c7cfd2$43bd4df0$4e8a0923@engineering.egr.msu.edu> Message-ID: <55D0489B-A47A-4526-9806-C04B50C79BEA@sissa.it> On Jul 27, 2007, at 12:13 AM, Tengfei Luo wrote: > Dear all: > I'm new to PWscf. I tried the first example which includes the > total energy calculation of Si. The result I got from this example > was: > > ! total energy = -15.84452726 Ry > > I believe this is for the total energy of the two Si atoms system. > However, from the reference data here: > > http://physics.nist.gov/PhysRefData/DFTdata/Tables/14Si.html > > The total energy is -288.198397 a.u.. The caluclated data is very > different from this. Is this because of using of pesudopotential? YES > How can I get the real totoal energy? 1) Do an all-electron calculation 2) If you want something aproximate, you may want to add to the PWscf total energy twice the difference berween the all-electron energy of an isolated atom and its energy calculated with the _same_ pseudopotential that you use for the solid-state calculation. Best if the difference is calculated between the all-electron atomic energy obtained by a real-space code and the pseudopotential atomic energy calculated with plane waves (you can do that with PWscf, with some care, if you wish). The reason is PW calculations introduve extra numerical errors due to the limited basis set. > I would appreciate any help! > > Thank you! you are most welcome > > Tengfei please, do not forget to add your affiliation to your signature Stefano --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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 -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/c7a0812b/attachment.htm From yinzping at gmail.com Fri Jul 27 08:52:11 2007 From: yinzping at gmail.com (Zhiping Yin) Date: Thu, 26 Jul 2007 23:52:11 -0700 Subject: [Pw_forum] [pw_forum] how to set up spin-orbital coupling calculation in PWscf In-Reply-To: <16935E3F-8524-4F73-AC54-8478B42E5EC4@sissa.it> References: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> <16935E3F-8524-4F73-AC54-8478B42E5EC4@sissa.it> Message-ID: <523370bf0707262352o5970866ft45f9139632c9bce6@mail.gmail.com> Hi Stefano and other users, Stefano, thank you for your reply. One needs to include SO coupling basically because the atom is heavy, which makes the SO coupling large, no matter it is spin polarized or not. In a metallic system (not ferromagnetic nor antiferromagnetic ), I don't know how to do a spin-polarized calculation, or more exactly why we need it spin polarized. I do know how to do SO coupling calculation without spin polarization in WIEN2K code. I don't if it is possible to do so in PWscf too. Anyway, how to set up a SO coupling calculation, with spin polarization if needed, in PWscf? Thanks. Best, Zhiping Yin On 7/26/07, Stefano Baroni wrote: > > I am not sure I understand the question, nor would I probably be able to > answer if I did. However, you may want to think about the fact that in the > presence of SO coupling, a system is always (locally) spin polarized. You > see the point? Stefano > On Jul 27, 2007, at 12:50 AM, Zhiping Yin wrote: > > Dear PWscf users, > > I tried to set up spin-orbital coupling calculation in PWscf without spin > polarization. I used a full relativistic NCPP and set lspinorb=.true.. > However, the output is identical with lspinorb=.false. I wonder what is the > correct way to set up a spin-orbital coupling calculation with/without spin > polarization. What is the requirement with the pseudopotential. Thanks very > much. > > Best, > Zhiping Yin > > -- > Zhiping Yin > graduate student > department of physics > UC Davis > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > > --- > Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste > [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) > > Please, if possible, don't send me MS Word or PowerPoint attachments > Why? See: http://www.gnu.org/philosophy/no-word-attachments.html > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Zhiping Yin graduate student department of physics UC Davis -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070726/5051bbbc/attachment-0001.htm From dalcorso at sissa.it Fri Jul 27 08:58:25 2007 From: dalcorso at sissa.it (Dal Corso Andrea) Date: Fri, 27 Jul 2007 08:58:25 +0200 Subject: [Pw_forum] [pw_forum] how to set up spin-orbital coupling calculation in PWscf In-Reply-To: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> References: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> Message-ID: <1185519505.3323.16.camel@dhpc-5-02.sissa.it> On Thu, 2007-07-26 at 15:50 -0700, Zhiping Yin wrote: > Dear PWscf users, > > I tried to set up spin-orbital coupling calculation in PWscf without > spin polarization. I used a full relativistic NCPP and set > lspinorb=.true.. However, the output is identical with > lspinorb=.false. I wonder what is the correct way to set up a > spin-orbital coupling calculation with/without spin polarization. What > is the requirement with the pseudopotential. Thanks very much. > The following input variables are relevant for a spin-orbit calculation: noncolin=.true. lspinorb=.true. starting_magnetization To make a spin-orbit calculation noncolin must be true. If it is false lspinorb is set to false no matter what you give as input. If starting_magnetization is set to zero (or not given) the code makes a spin orbit calculation without spin magnetization (it assumes that time reversal is true and it does not calculate the magnetization). The states are still two component spinors but the total magnetization is zero. If starting_magnetization is different from zero it makes a non collinear spin polarized calculation with spin orbit. The final spin magnetization might be zero or different from zero depending on the system. Furthermore to make a spin-orbit calculation you must use fully relativistic pseudopotentials at least for the atoms in which you think that spin orbit is large. If all the pseudopotentials are scalar relativistic the calculation becomes equivalent to a noncolinear calculation without spin orbit. Hope this helps, Andrea > Best, > Zhiping Yin > > -- > Zhiping Yin > graduate student > department of physics > UC Davis > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- 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 baroni at sissa.it Fri Jul 27 09:17:18 2007 From: baroni at sissa.it (Stefano Baroni) Date: Fri, 27 Jul 2007 09:17:18 +0200 Subject: [Pw_forum] [pw_forum] how to set up spin-orbital coupling calculation in PWscf In-Reply-To: <1185519505.3323.16.camel@dhpc-5-02.sissa.it> References: <523370bf0707261550s2c46c00av1e61c0755ba2a9b1@mail.gmail.com> <1185519505.3323.16.camel@dhpc-5-02.sissa.it> Message-ID: Zhiping: I wish Andrea's reply fully clarifies the practical side of the problem. I would like to make sure, though, that my previous note (saying that in the presence of SO interactions electronic states are always psin polarized) is properly understood. Here is a summary of how SO coupling was explained to me long, LONG, ago. In the reference frame of any electron the electron is - by definition - at rest, but the other charges (including nuclei - or screened nuclei, i.e. frozen ions) are not. The other charges are therefore associated to currents that generate a magnetic field which, in turns, interacts with the spin at rest, determining an alignement of the electron spin along its direction. Hence, the electron is polarized. The magnitude and direction of the magnetic field depends on the position of the electron, and so the resulting spin polarization will depend on position. In order to describe a spin polarization that depends on position, you will need to describe electron states with two-component spinors. This is what the "noncolin=.true." keyword is sipposed to mean ("non colin" for "non collinear (spin) polarization). Hope this clarifies a bit more. Also, remember: computer codes (Wien2k, pwscf, you name it) are fun, but the reason why they work and the way they work is even more fun! Have fun with electronic structure calculations! Stefano On Jul 27, 2007, at 8:58 AM, Dal Corso Andrea wrote: > On Thu, 2007-07-26 at 15:50 -0700, Zhiping Yin wrote: >> Dear PWscf users, >> >> I tried to set up spin-orbital coupling calculation in PWscf without >> spin polarization. I used a full relativistic NCPP and set >> lspinorb=.true.. However, the output is identical with >> lspinorb=.false. I wonder what is the correct way to set up a >> spin-orbital coupling calculation with/without spin polarization. >> What >> is the requirement with the pseudopotential. Thanks very much. >> > > The following input variables are relevant for a spin-orbit > calculation: > > noncolin=.true. > lspinorb=.true. > > starting_magnetization > > To make a spin-orbit calculation noncolin must be true. If it is false > lspinorb is set to false no matter what you give as input. > > If starting_magnetization is set to zero (or not given) the code > makes > a spin orbit calculation without spin magnetization (it assumes that > time reversal is true and it does not calculate the magnetization). > The states are still two component spinors but the total magnetization > is zero. > > If starting_magnetization is different from zero it makes a non > collinear spin polarized calculation with spin orbit. The final spin > magnetization might be zero or different from zero depending on the > system. > > Furthermore to make a spin-orbit calculation you must use fully > relativistic pseudopotentials at least for the atoms in which you > think > that spin orbit is large. If all the pseudopotentials are scalar > relativistic the calculation becomes equivalent to a noncolinear > calculation without spin orbit. > > Hope this helps, > > Andrea > >> Best, >> Zhiping Yin >> >> -- >> Zhiping Yin >> graduate student >> department of physics >> UC Davis >> _______________________________________________ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://www.democritos.it/mailman/listinfo/pw_forum > -- > 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 --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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 -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/149f236f/attachment-0001.htm From cyrille.barreteau at cea.fr Fri Jul 27 09:00:50 2007 From: cyrille.barreteau at cea.fr (BARRETEAU Cyrille) Date: Fri, 27 Jul 2007 09:00:50 +0200 Subject: [Pw_forum] RE : spin-orbit references References: <6ac064b60707261449v3234e7d8te51985945401e264@mail.gmail.com> Message-ID: <18E4AC9C2FAC504E994A6759DE5ABE8D20F151@DIODON.extra.cea.fr> Dear Stefano, I am not the one who asked the question but I want thank you anyway for this reference that is very close to my personal interest.. best wishes 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 cyrille.barreteau 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 http://www-drecam.cea.fr/Phocea/Membres/Cours/index.php ================================================================== -------- Message d'origine-------- De: pw_forum-bounces at pwscf.org de la part de Stefano Baroni Date: ven. 27/07/2007 08:36 ?: PWSCF Forum Objet : Re: [Pw_forum] spin-orbit references You may want to try this, and references quoted therein. http://www.sissa.it/cm/thesis/2007/moscaconte.pdf Stefano On Jul 26, 2007, at 11:49 PM, Nichols A. Romero wrote: > Hi, > > Does anyone know of a good reference (perhaps a student > dissertation) that > describs the more technical details involved in treating spin-orbit > coupling within > DFT? (For example, when to treat the system pertubatively, > self-consistently, pseudpotential, etc.) > Thanks, > -- > Nichols A. Romero, Ph.D. > High Performance Technologies, Inc. > DoD User Productivity Enhancement & > Technology Transfer (PET) Group > Reston, VA > 443-567-8328 (C) > 410-278-2692 (O) > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum --- Stefano Baroni - SISSA & DEMOCRITOS National Simulation Center - Trieste [+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype) 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: application/ms-tnef Size: 3710 bytes Desc: not available Url : http://www.democritos.it/pipermail/pw_forum/attachments/20070727/07e60d41/attachment.bin From ezadshojaee at hotmail.com Fri Jul 27 10:45:18 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Fri, 27 Jul 2007 08:45:18 +0000 Subject: [Pw_forum] problem with parallel ph.x Message-ID: Hii have attached the output file of ph.x (calculation of IFC's for a cubic grid ), you can see what i was trying to describe in the nscf calculation part of the 2nd point in the grid( where some eigenvalues are zero )what's the matter?and also i have this error message later: MPI_Recv: message truncated (rank 0, comm 9)Rank (0, MPI_COMM_WORLD): Call stack within LAM:Rank (0, MPI_COMM_WORLD): - MPI_Recv()Rank (0, MPI_COMM_WORLD): - main() these are the specs of ours: 1- machine=AMD opteron dualcore64bit, 2- OS=CentOS64bit 4.4, 3- compiler=intel fortran9.1, 4- MPI= lamMPI 7.1.3, 5- espresso=3.2 and i have different datas depending on the # of npools!! i should say that ph.x goes correctly in the serial mode, and i am able to run scf calculation in the parallel mode (with the same results of the serial)hope this could be enough for buggin(the input is quite known by the info above) _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/0f1a727e/attachment-0001.htm -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ph.out Url: http://www.democritos.it/pipermail/pw_forum/attachments/20070727/0f1a727e/attachment-0001.txt From giannozz at nest.sns.it Fri Jul 27 13:05:22 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 27 Jul 2007 13:05:22 +0200 Subject: [Pw_forum] problem with parallel ph.x In-Reply-To: References: Message-ID: On Jul 27, 2007, at 10:45 , Ezad Shojaee wrote: > (the input is quite known by the info above) no it's not: there are several pieces of information, needed to produce the input, that are missing from the output. Please provide the input that produced the output you sent, for both the scf and the nscf calculations. --- Paolo Giannozzi, Democritos and University of Udine, Italy From uccaati at ucl.ac.uk Fri Jul 27 16:05:31 2007 From: uccaati at ucl.ac.uk (Antonio Tilocca) Date: Fri, 27 Jul 2007 15:05:31 +0100 (BST) Subject: [Pw_forum] compilation of bfgs_modules.f90 on Sun Message-ID: For some reason, the Sun compiler stops while compiling bfgs_modules.f90, complaining that the .dot. and .times. operators are not defined in gdiis_step. Adding "USE basic_algebra_routines" at the beginning of this routine apparently fixes the problem, but I am not sure why the problem is there in first place. Antonio Tilocca Chemistry/UCL/UK From ceresoli at sissa.it Fri Jul 27 16:19:43 2007 From: ceresoli at sissa.it (Davide Ceresoli) Date: Fri, 27 Jul 2007 16:19:43 +0200 Subject: [Pw_forum] compilation of bfgs_modules.f90 on Sun In-Reply-To: References: Message-ID: <46A9FEFF.2080506@sissa.it> Antonio Tilocca wrote: > For some reason, the Sun compiler stops while compiling bfgs_modules.f90, > complaining that the .dot. and .times. operators are not defined > in gdiis_step. Adding "USE basic_algebra_routines" at the > beginning of this routine apparently fixes the problem, but I am not sure > why the problem is there in first place. > > Antonio Tilocca > Chemistry/UCL/UK The SUN compiler needs you to repeat the line: USE basic_algebra_routines in every CONTAIN'ed subroutine. For instance, in bfgs_modules.f90, after: SUBROUTINE gdiis_step(), add: USE basic_algebra_routines It's already in the CVS. Davide From proffess at yandex.ru Fri Jul 27 16:15:03 2007 From: proffess at yandex.ru (Sergey Lisenkov) Date: Fri, 27 Jul 2007 18:15:03 +0400 Subject: [Pw_forum] compilation of bfgs_modules.f90 on Sun In-Reply-To: 1630000000149917724 References: 1630000000149917724 Message-ID: <325561185545703@webmail17.yandex.ru> We had the same problem with pathscale compiler (v3.0) on Sun Opteron cluster and this trick solved a problem. According to pathscale support team, it was a compiler bug ( as Paolo always tells us:-). Maybe the Sun compiler has the same bug... Regards, Sergey 27.07.07, 18:05, Antonio Tilocca : > For some reason, the Sun compiler stops while compiling bfgs_modules.f90, > complaining that the .dot. and .times. operators are not defined > in gdiis_step. Adding "USE basic_algebra_routines" at the > beginning of this routine apparently fixes the problem, but I am not sure > why the problem is there in first place. > Antonio Tilocca > Chemistry/UCL/UK > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Fri Jul 27 22:34:29 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Fri, 27 Jul 2007 22:34:29 +0200 Subject: [Pw_forum] compilation of bfgs_modules.f90 on Sun In-Reply-To: <325561185545703@webmail17.yandex.ru> References: 1630000000149917724 <325561185545703@webmail17.yandex.ru> Message-ID: <99B10837-B06E-4D97-8561-A4840ADDE082@nest.sns.it> On Jul 27, 2007, at 16:15 , Sergey Lisenkov wrote: > We had the same problem with pathscale compiler (v3.0) on Sun Opteron > cluster and this trick solved a problem. According to pathscale > support team, > it was a compiler bug ( as Paolo always tells us:-). not always, but often ... Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From ihsanas at yahoo.com Sat Jul 28 15:52:37 2007 From: ihsanas at yahoo.com (احسان عريقات) Date: Sat, 28 Jul 2007 06:52:37 -0700 (PDT) Subject: [Pw_forum] charge density difference Message-ID: <72872.77654.qm@web52305.mail.re2.yahoo.com> Dear pwscf users I am trying to plot charge density difference (using version 3.2).I can made IrO.charge ,Irclean .charge and O.charge but when I tried to make charge density difference ,I have error my file is as shown below &inputpp / &plot nfile = 3 filepp(1) = 'IrO.charge' filepp(2) ='Irclean.charge' filepp(3) ='O.charge' weight(1) = 1.0 weight(2) = -1.0 weight(3) = -1.0 iflag = 2 output_format = 2 fileout = '1x1charge.rho.dat' e1(1) =0.0, e1(2)=0.0, e1(3) = 1.0, e2(1) =1.0, e2(2)=1.0, e2(3) = 0.0, nx=60, ny=64 / The error forrtl: severe (174): SIGSEGV, segmentation fault occurred Image PC Routine Line Source pp.x 082634E0 Unknown Unknown Unknown pp.x 0826029A Unknown Unknown Unknown pp.x 08239EE5 Unknown Unknown Unknown pp.x 0823970C Unknown Unknown Unknown pp.x 0806F2E1 Unknown Unknown Unknown pp.x 0804B06D Unknown Unknown Unknown pp.x 08049B4B Unknown Unknown Unknown pp.x 08049AC6 Unknown Unknown Unknown libc.so.6 001267E4 Unknown Unknown Unknown pp.x 08049A01 Unknown Unknown Unknown Can someone help me regards Ihsan physics department Jordan ____________________________________________________________________________________ Moody friends. Drama queens. Your life? Nope! - their life, your story. Play Sims Stories at Yahoo! Games. http://sims.yahoo.com/ From akohlmey at cmm.chem.upenn.edu Sun Jul 29 00:09:23 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sat, 28 Jul 2007 18:09:23 -0400 (EDT) Subject: [Pw_forum] charge density difference In-Reply-To: <72872.77654.qm@web52305.mail.re2.yahoo.com> Message-ID: On Sat, 28 Jul 2007, ????? ?????? wrote: I> Dear pwscf users I> I am trying to plot charge density difference (using I> version 3.2).I can made IrO.charge ,Irclean .charge I> and O.charge but when I tried to make charge density [...] I> Can someone help me not without additional information. please quote additional text before the segfault message. please also post the inputs to generate the three .charge files. and finally, please always provide information about the platform you are running on (cpu, OS, compiler etc.). have you verified, that this is not a bug fixed with the updates to 3.2? cheers, axel. I> regards I> Ihsan I> physics department I> Jordan I> I> I> I> I> ____________________________________________________________________________________ I> Moody friends. Drama queens. Your life? Nope! - their life, your story. Play Sims Stories at Yahoo! Games. I> http://sims.yahoo.com/ I> _______________________________________________ I> Pw_forum mailing list I> Pw_forum at pwscf.org I> http://www.democritos.it/mailman/listinfo/pw_forum I> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From ezadshojaee at hotmail.com Sun Jul 29 07:43:18 2007 From: ezadshojaee at hotmail.com (Ezad Shojaee) Date: Sun, 29 Jul 2007 05:43:18 +0000 Subject: [Pw_forum] problem with parallel ph.x Message-ID: the input files for scf and phonon calculation are attached> From: giannozz at nest.sns.it> Date: Fri, 27 Jul 2007 13:05:22 +0200> To: pw_forum at pwscf.org> Subject: Re: [Pw_forum] problem with parallel ph.x> > > On Jul 27, 2007, at 10:45 , Ezad Shojaee wrote:> > > (the input is quite known by the info above)> > no it's not: there are several pieces of information, needed> to produce the input, that are missing from the output.> Please provide the input that produced the output you sent,> for both the scf and the nscf calculations.> > ---> Paolo Giannozzi, Democritos and University of Udine, Italy> > > _______________________________________________> Pw_forum mailing list> Pw_forum at pwscf.org> http://www.democritos.it/mailman/listinfo/pw_forum _________________________________________________________________ With Windows Live Hotmail, you can personalize your inbox with your favorite color. www.windowslive-hotmail.com/learnmore/personalize.html?locale=en-us&ocid=TXT_TAGLM_HMWL_reten_addcolor_0607 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20070729/b02e5e9b/attachment.htm -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ph.in Url: http://www.democritos.it/pipermail/pw_forum/attachments/20070729/b02e5e9b/attachment.txt -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: scf.in Url: http://www.democritos.it/pipermail/pw_forum/attachments/20070729/b02e5e9b/attachment-0001.txt From brsahu at physics.utexas.edu Mon Jul 30 01:25:44 2007 From: brsahu at physics.utexas.edu (brsahu at physics.utexas.edu) Date: Sun, 29 Jul 2007 18:25:44 -0500 Subject: [Pw_forum] bfgs error Message-ID: <20070729182544.patrnnmxf0o4ok4o@mail.ph.utexas.edu> Dear PWSCF users, I encountered an error in a calculation (carbon slabs, ultrasoft PP, ecutwfc = 30 Ry and ecutrho = 250 Ry) CASE: energy_new > energy_old new trust radius = 0.0000495950 bohr trust_radius < trust_radius_min resetting bfgs history %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from bfgs : error # 1 bfgs history already reset at previous step %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... The solution suggested in the mailing list (scf convergence criteria can not be greater than the force criteria) is contrary to my inputs as Total force = 0.003851 Total SCF correction = 0.000054 entering subroutine stress ... total stress (Ry/bohr**3) (kbar) P= -8.86 -0.00015635 0.00000000 0.00000000 -23.00 0.00 0.00 0.00000000 -0.00001219 -0.00000281 0.00 -1.79 -0.41 0.00000000 -0.00000281 -0.00001204 0.00 -0.41 -1.77 number of scf cycles = 61 number of bfgs steps = 47 energy old = -370.1139086454 Ry energy new = -370.1138810384 Ry and in the input file I have forc_conv_thr = 1.0D-3 conv_thr = 1.0d-9 etot_conv_thr is default any suggestions? Sahu UT-Austin, USA From akohlmey at cmm.chem.upenn.edu Mon Jul 30 01:04:09 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Sun, 29 Jul 2007 19:04:09 -0400 (EDT) Subject: [Pw_forum] bfgs error In-Reply-To: <20070729182544.patrnnmxf0o4ok4o@mail.ph.utexas.edu> Message-ID: On Sun, 29 Jul 2007 brsahu at physics.utexas.edu wrote: dear sahu, BS> Dear PWSCF users, BS> BS> I encountered an error in a calculation (carbon slabs, ultrasoft PP, BS> ecutwfc = 30 Ry and ecutrho = 250 Ry) BS> BS> CASE: energy_new > energy_old BS> BS> new trust radius = 0.0000495950 bohr BS> BS> trust_radius < trust_radius_min BS> BS> resetting bfgs history [...] BS> The solution suggested in the mailing list (scf convergence criteria BS> can not be greater than the force BS> criteria) is contrary to my inputs it is not entirely clear to me, how i should understand this, but having a tight SCF convergence parameter (conv_thr, about two or three orders of magnitude smaller than forc_conv_thr) is a pre-requiste for well-converged forces. what you are seeing is that your forces are still fluctating a lot with small displacements (the reason for the BFGS resets). so you have to consider other sources. what comes to my mind are: the density cutoff, bad pseudopotentials, k-point convergence/smearing. your total force is already very close to the desired accuracy, so it may be something that has only a smaller impact on forces, hence my suggestion would be to try with cranking up the density cutoff to 300ry. cheers, axel. BS> BS> as BS> BS> Total force = 0.003851 Total SCF correction = 0.000054 BS> BS> BS> entering subroutine stress ... BS> BS> total stress (Ry/bohr**3) (kbar) P= -8.86 BS> -0.00015635 0.00000000 0.00000000 -23.00 0.00 0.00 BS> 0.00000000 -0.00001219 -0.00000281 0.00 -1.79 -0.41 BS> 0.00000000 -0.00000281 -0.00001204 0.00 -0.41 -1.77 BS> BS> BS> number of scf cycles = 61 BS> number of bfgs steps = 47 BS> BS> energy old = -370.1139086454 Ry BS> BS> energy new = -370.1138810384 Ry BS> BS> and in the input file I have BS> BS> forc_conv_thr = 1.0D-3 BS> conv_thr = 1.0d-9 BS> BS> etot_conv_thr is default BS> BS> any suggestions? BS> BS> Sahu BS> UT-Austin, USA BS> _______________________________________________ BS> Pw_forum mailing list BS> Pw_forum at pwscf.org BS> http://www.democritos.it/mailman/listinfo/pw_forum BS> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From giannozz at nest.sns.it Mon Jul 30 10:00:09 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 30 Jul 2007 10:00:09 +0200 Subject: [Pw_forum] bfgs error In-Reply-To: <20070729182544.patrnnmxf0o4ok4o@mail.ph.utexas.edu> References: <20070729182544.patrnnmxf0o4ok4o@mail.ph.utexas.edu> Message-ID: <7FAE8BB1-70F9-4CEC-B5BE-066CFD986651@nest.sns.it> On Jul 30, 2007, at 1:25 , brsahu at physics.utexas.edu wrote: > The solution suggested in the mailing list (scf convergence criteria > can not be greater than the force criteria) is contrary to my inputs > as > > Total force = 0.003851 Total SCF correction = 0.000054 other solutions to this problem have been suggested in the mailing list. My favorite one is "do not waste time trying to over-optimize your atomic positions". The forces you have are already very small. Stop the optimization and go to the beach :-) Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Mon Jul 30 10:03:07 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 30 Jul 2007 10:03:07 +0200 Subject: [Pw_forum] charge density difference In-Reply-To: <72872.77654.qm@web52305.mail.re2.yahoo.com> References: <72872.77654.qm@web52305.mail.re2.yahoo.com> Message-ID: On Jul 28, 2007, at 15:52 , ????? ?????? wrote: > I am trying to plot charge density difference (using version 3.2) > [...] > forrtl: severe (174): SIGSEGV, segmentation fault occurred first of all, check if the error occurs if you plot just one of the files. The three files should be produced by a run with the same cell, and if I remember correctly the first one should contain the largest number of atoms Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From giannozz at nest.sns.it Tue Jul 31 16:18:44 2007 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Tue, 31 Jul 2007 16:18:44 +0200 Subject: [Pw_forum] problem with parallel ph.x In-Reply-To: References: Message-ID: On Jul 29, 2007, at 7:43 , Ezad Shojaee wrote: > the input files for scf and phonon calculation are attached this is what I get on a sp5 machine, with the CVS version of the code: http://www.fisica.uniud.it/~giannozz/public/ph.tar.gz It is only part of the calculation, but it looks correct to me, or at least I cannot see the anomalies you get Paolo --- Paolo Giannozzi, Democritos and University of Udine, Italy From akohlmey at cmm.chem.upenn.edu Tue Jul 31 15:52:29 2007 From: akohlmey at cmm.chem.upenn.edu (Axel Kohlmeyer) Date: Tue, 31 Jul 2007 09:52:29 -0400 (EDT) Subject: [Pw_forum] problem with parallel ph.x In-Reply-To: Message-ID: On Tue, 31 Jul 2007, Paolo Giannozzi wrote: PG> PG> On Jul 29, 2007, at 7:43 , Ezad Shojaee wrote: PG> PG> > the input files for scf and phonon calculation are attached PG> PG> this is what I get on a sp5 machine, with the CVS version of PG> the code: http://www.fisica.uniud.it/~giannozz/public/ph.tar.gz PG> It is only part of the calculation, but it looks correct to me, or PG> at least I cannot see the anomalies you get same here on x86_64 with intel 9.1, and openmpi. ezad, since you are using separate diretory for temporary files, are you sure that you don't accidentally run two jobs with the same 'prefix'? also i noticed that the job creates a lot of disk i/o. i would suggest to check the machine hosting the disk(s) for i/o errors in the kernel log. cheers, axel. PG> PG> Paolo PG> --- PG> Paolo Giannozzi, Democritos and University of Udine, Italy PG> PG> PG> _______________________________________________ PG> Pw_forum mailing list PG> Pw_forum at pwscf.org PG> http://www.democritos.it/mailman/listinfo/pw_forum PG> -- ======================================================================= Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot. From ihsanas at yahoo.com Tue Jul 31 18:29:46 2007 From: ihsanas at yahoo.com (احسان عريقات) Date: Tue, 31 Jul 2007 09:29:46 -0700 (PDT) Subject: [Pw_forum] charge density difference Message-ID: <922940.49632.qm@web52311.mail.re2.yahoo.com> Thanks you Paolo a lot c/a of the three cells were not the same this was the problem . regards Ihsan physics department- Jordan ____________________________________________________________________________________ Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC From michaelv at Princeton.EDU Tue Jul 31 23:14:34 2007 From: michaelv at Princeton.EDU (Michael Vargas (michaelv@Princeton.EDU)) Date: Tue, 31 Jul 2007 17:14:34 -0400 Subject: [Pw_forum] LD1 - Choosing a 3D rcutus Message-ID: Hello, I'm pretty new at using this program. I am trying to make an US-PP for Oxygen over a large energy range. I am using an unbound 3D for the local channel. I was wondering if anyone could give me some insight on how to choose my cut-off radius for this? I am currently using about 1.5 bohr for the rcut and rcutus, as this works well for agreement in the logarithmic derivatives, but I'm not sure if there should be some deeper reason. This is an unbound 3D, so I can't see the wavefunction to guess at a reasonable radius. Thank you for any help you can give. -- Mike