[QE-users] QE 7.1 and gipaw
Dear QE developers, I downloaded and correctly compiled QE 7.1, but when I try to compile gipaw by "make gipaw", the following error appear: ./update_gipaw_version make -C /home/nervi/src/QE/QE71/q-e/LR_Modules make[4]: Entering directory '/home/nervi/src/QE/QE71/q-e/LR_Modules' make[4]: Nothing to be done for 'all'. make[4]: Leaving directory '/home/nervi/src/QE/QE71/q-e/LR_Modules' mpif90 -Ofast -march=native -mavx2 -fomit-frame-pointer -ftree-vectorize -funsafe-math-optimizations -ffpe-summary=invalid,zero,overflow -cpp -D__FFTW3 -D__MPI -D__SCALAPACK -I/home/nervi/src/QE/QE71/q-e//external/devxlib/src -I. -I/home/nervi/src/QE/QE71/q-e//include -I/home/nervi/src/QE/QE71/q-e/upflib -I/home/nervi/src/QE/QE71/q-e/Modules -I/home/nervi/src/QE/QE71/q-e/MBD -I/home/nervi/src/QE/QE71/q-e/FFTXlib/src -I/home/nervi/src/QE/QE71/q-e/LAXlib -I/home/nervi/src/QE/QE71/q-e/UtilXlib -I/home/nervi/src/QE/QE71/q-e/XClib -I/home/nervi/src/QE/QE71/q-e/FoX/finclude -I/home/nervi/src/QE/QE71/q-e/PW/src -I/home/nervi/src/QE/QE71/q-e/dft-d3 -I/home/nervi/src/QE/QE71/q-e/LR_Modules -I. -c gipaw_version.f90 f951: Warning: Nonexistent include directory ‘/home/nervi/src/QE/QE71/q-e/FoX/finclude’ [-Wmissing-include-dirs] make[3]: *** No rule to make target '/home/nervi/src/QE/QE71/q-e/FoX/finclude/fox_dom.mod', needed by 'xml_routines.o'. Stop. make[3]: Leaving directory '/home/nervi/src/QE/QE71/q-e/qe-gipaw-7.1/src' make[2]: *** [Makefile:7: build] Error 2 make[2]: Leaving directory '/home/nervi/src/QE/QE71/q-e/qe-gipaw-7.1' make[1]: *** [plugins_makefile:55: gipaw] Error 1 make[1]: Leaving directory '/home/nervi/src/QE/QE71/q-e/install' make: *** [Makefile:111: gipaw] Error 1 FoX directory is not present in the q-e root, and in the external directory the "fox" directory is empty. I would greatly appreciate any suggestions on how to compile gipaw on QE 7.1 Thank you, Carlo -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ The Quantum ESPRESSO community stands by the Ukrainian people and expresses its concerns about the devastating effects that the Russian military offensive has on their country and on the free and peaceful scientific, cultural, and economic cooperation amongst peoples ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] TS Dispersion
Yes, we are testing and comparing planewave calculations with dftb in the same conditions (TS and MBD). Usually I'm using the non-local vdE-df2, which, in my experience, is working well. But for this particular project we need TS (and MBD). Thanks to Paolo and to you for the reply. Carlo Il giorno mer 19 gen 2022 alle ore 22:20 Johannes Meusburger - STFC UKRI via users ha scritto: > Dear Carlo, > > > > Is there any particular reason why you want to use the TS scheme? > > > If not there are several benchmarking studies (e.g. > https://doi.org/10.1063/1.4981528 ; https://doi.org/10.1063/1.5030493 ; > https://doi.org/10.1016/j.icarus.2021.114611) demonstrating that the TS > and the D2 scheme (which works with US and PAW PP in QE) perform almost > equally well (the D2 is often even slightly better) in modelling the > geometry and material properties of a wide range of materials. Since TS is > not available yet with your PP, I suggest you could just use the D2 scheme > instead if you want to stick to the SSSP pseudopotentials. > > > > HTH Johannes > > > > Johannes Meusburger > > ISIS Neutron and Muon Source, Oxfordshire, UK > > -- > *From:* users on behalf of > Paolo Giannozzi > *Sent:* Wednesday, January 19, 2022 6:44 PM > *To:* Quantum ESPRESSO users Forum > *Subject:* Re: [QE-users] TS Dispersion > > On Wed, Jan 19, 2022 at 6:08 PM Carlo Nervi wrote: > > Anybody know whether it is a my compiler mistake or the new version of QE > (7.0?) overrides this limitation? > > > 2*no, but it's coming soon in the development version. > > Paolo > -- > Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, > Univ. Udine, via delle Scienze 206, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > This email and any attachments are intended solely for the use of the > named recipients. If you are not the intended recipient you must not use, > disclose, copy or distribute this email or any of its attachments and > should notify the sender immediately and delete this email from your > system. UK Research and Innovation (UKRI) has taken every reasonable > precaution to minimise risk of this email or any attachments containing > viruses or malware but the recipient should carry out its own virus and > malware checks before opening the attachments. UKRI does not accept any > liability for any losses or damages which the recipient may sustain due to > presence of any viruses. > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ *ICCC2020 has been postponed at 2022* ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2022) ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] TS Dispersion
Dear QE users and developers, I'm trying to use the TS dispersion in *QE 6.8*, but I get the following error: Error in routine setup (1): Tkatchenko-Scheffler not implemented with USPP/PAW Clearly I can't use the SSSP_eff_PBE Anybody know whether it is a my compiler mistake or the new version of QE (7.0?) overrides this limitation? Thank you, Carlo -- ---- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ *ICCC2020 has been postponed at 2022* ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2022) ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] NEB unexpected crash
Dear Mauro, sometimes the way in which remote shell behave could stop the child processes. If you are running programs over SSH and if the connection drops, the session will be terminated, and all the executed processes will stop. To avoid that you may use the *nohup *(No Hang Up) command. In my system, for example, I can avoid the stopping of the processes by running normally my commands in background (&) and then exiting the shell. This guarantee (according to my system settings) that the child processes will be stopped. HTH, Carlo Il giorno mar 21 set 2021 alle ore 09:47 Mauro Sgroi < maurofrancesco.sg...@gmail.com> ha scritto: > Dear Paolo, > all the 3 conditions you are reporting should not be the case. > I'm going to investigate with our HPC manager. > Best regards, > Mauro. > > Il giorno mar 21 set 2021 alle ore 08:56 Paolo Giannozzi < > p.gianno...@gmail.com> ha scritto: > >> On Mon, Sep 20, 2021 at 5:52 PM Mauro Sgroi < >> maurofrancesco.sg...@gmail.com> wrote: >> >>> >>> What does " Program stopped by user request" mean? >>> >> >> it means one of the following cases: >> - the time limit specified in input has been reached >> - a file $prefix.EXIT or $prefix.save/EXIT or something like that has >> been detected >> - signal USR1 has been sent to the process (only if compiled with a >> specific preprocessing option) >> >> Is there a way to fix this behaviour? >>> >> >> there is nothing to be fixed: one should figure out which of the three >> cases mentioned above has occurred >> >> Paolo >> >> >>> Thanks a lot and best regards, >>> Mauro Sgroi. >>> ___ >>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu) >>> users mailing list users@lists.quantum-espresso.org >>> https://lists.quantum-espresso.org/mailman/listinfo/users >> >> >> >> -- >> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, >> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy >> Phone +39-0432-558216, fax +39-0432-558222 >> >> ___ >> Quantum ESPRESSO is supported by MaX (www.max-centre.eu) >> users mailing list users@lists.quantum-espresso.org >> https://lists.quantum-espresso.org/mailman/listinfo/users > > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ *ICCC2020 has been postponed at 2022* ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2022) ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Sub optimal performance on 32 core AMD machine
Hello Pamela, I don't know whether it is clear or not, so I apologize if I repeat obvious concepts. I just bought a Threadripper 3990X with 64 core 128 threads. As far as I remember the 3960X should have 24 core - 48 threads. It is very very important to don't use more than 24 cores on 3960X . Simply forget about hyperthreading. No need to disable it in the BIOS, but simply count the real number of cores. I use gcc 9.3.0 and the new gcc 10 should be even better for AMD cpus. With openblas 0.3.12 I found that my 8-cores home Ryzen 3800X is fast as a Xeon 12 cores E5-2680 using quantum espresso 6.4.1 Carlo Il giorno mar 17 nov 2020 alle ore 19:24 Pamela Whitfield < whitfield...@gmail.com> ha scritto: > Michal > > I have a very similar use-case and looked into many of the same issues > when I got my Threadripper 3960X system at the beginning of the year to > supplement my old dual-Xeon setup. In the past few days I've been > revisiting compilation as I got hold of a Quadro GV100 for GPU acceleration > of my optimizations. > > Basically it seems as though code compiled for Zen2 either can't handle > code compiled for both MPI and OpenMP at all, or does so poorly even when > it runs. > Best performance for pw.x on v6.5 (I've been playing with GIPAW and > there's no 6.6 compatible version yet) has been with a simple gcc OpenMPI > compilation without openmp threading and with about 20 MPI cores on my 24 > core CPU. Compiling with GCC or PGI compiler made little difference, > although only the more recent PGI compilers will have zen2 optimization. > I get little benefit from Intel MKL over openblas/lapack/fftw3 even with > the debug tweaks, etc. > Puget Systems numbers with other programs suggest that OpenMP only > performs better than OpenMPI with Threadripper but I find the opposite with > QE. > I did try disabling hyperthreading in the BIOS but that made no difference > to the performance. > > GPU compilation really shows the issue with MPI/OpenMP clashing. With the > Xeons I could compile code with MKL that would run well on a Quadro K6000 > while offloading to the CPU with MPI when needed. It could still be a > compiler issue (have to use PGI with the GPU version) but it just doesn't > work with the 3960X, and some things don't thread well with pure OpenMP > (e..g dftd3 versus dftd2) so I'll still need to use separately compiled > versions of 6.5 for different problems. > > BTW with a dual CPU system you may benefit from pinning threads to > particular CPUs - it works on the dual Xeon in any case. My Threadripper > balances the load across the cores in a pretty dynamic manner and that's on > a single socket. > > Best regards > Pam Whitfield > > Independent Consultant > > > > > > Message: 1 > Date: Mon, 16 Nov 2020 15:19:04 +0100 > From: Michal Husak > To: Quantum ESPRESSO users Forum > Subject: [QE-users] Sub optimal performance on 32 core AMD machine > Message-ID: > Content-Type: text/plain; charset="UTF-8"; format=flowed > > I had purchased a new PC with 2x 16 core AMD EPYC processors . 64 > cores with hyper threading ... > I was hoping my QM programs (Quantum Espresso, CASTEP) will run on the new > system faster, than on my old 4 core i7 Intel machine (8 year old) > > To my great surprise, the opposite is almost true :-(. > My main task is scf and geometry optimization of middle sized organic > molecular crystals (abut 100 C,H,N per unit cell) ... > > I was playing with OpenMPI/OpenMP setup changes ... > I was playing with the secret MKL_DEBUG_CPU_TYPE=5 parameter > (responsible for slow run of Intel MKL compiled code on AMD) ... > > Nothing helps, the best speed is obteined when I use only 4 cores > (OpenMPI or OpenMP - results similar) ... > Using 16 or 32 cores gives almost no benefit ... > The CPU load for run on 1/4/816/32 coresponds to the nubmer of CPU > set = they try to do something ... > > Any idea what I should check, try optimize ? > > Maybe the bottleneck is memory access, not CPU power (I have 128 > GB almost not used RAM) ? > > Michal Husak > > UCT Prague > > > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ *ICCC2020 has been postponed at 2022* ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2022) ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] Sub optimal performance on 32 core AMD machine
While compiling with gcc/gfortran use -march=native option. With Intel i guess it is -xHost This will enable the CPU native instruction. With Epyc I guess a lot depends on the libraries. With gcc I used openblas and fftw3. On Ryzen they work great. HTH, Carlo Il giorno lun 16 nov 2020 alle ore 18:24 Husak Michal ha scritto: > No. I do no use hyperthreading. > I had checked both OpenMPI and OpenMP - in outup file ... > It shows what I set by the launch and enviromental variable correctly. > If I use more than 4 cores in any way, I get no speed up ... > > From: users on behalf of > Pietro Delugas > Sent: Monday, November 16, 2020 3:42:59 PM > To: Quantum ESPRESSO users Forum > Subject: Re: [QE-users] Sub optimal performance on 32 core AMD machine > > Hi > Are you using hyperthreading ? > If yes, you should try to use only 32 cores. Check the total number of > processors at the beginning of the run. > The program makes an intense usage of the cpus so that hyperthreading is > usually inefficient. > Pietro > > Sent from Mail<https://go.microsoft.com/fwlink/?LinkId=550986> for > Windows 10 > > From: Michal Husak<mailto:michal.hu...@vscht.cz> > Sent: Monday, November 16, 2020 3:19 PM > To: Quantum ESPRESSO users Forum<mailto:users@lists.quantum-espresso.org> > Subject: [QE-users] Sub optimal performance on 32 core AMD machine > > I > > ___ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ *ICCC2020 has been postponed at 2022* ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2022) ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] About the rules for skipping the comment in the script cif2qe.sh.
Dear Prof. Zhao,
the utility is provided *as it is*, shared among users for users'
convenience.
You are free to modify it or adopt other conversion utilities (like
cif2cell).
Depending also on the awk (or gawk) version small adjustment to the code is
perhaps required.
The advantage of cif2qe.sh is its simplicity in an open awk script.
Best regards,
Carlo Nervi
Il giorno mer 21 ott 2020 alle ore 15:59 Hongyi Zhao
ha scritto:
> Hi,
>
> In the script cif2qe.sh, it uses the following rules to skip the
> comment of the CIF file:
>
> $1 ~ /^\#/||/^\;/ { next }
>
> See <https://gitlab.com/QEF/q-e/-/blob/develop/PW/tools/cif2qe.sh#L250>
> for more info.
>
> But I can't figure out why should it use this rule for the lines
> beginning with ';'. According to the description on the iucr's website
> <https://www.iucr.org/resources/cif/spec/version1.1/cifsyntax>, the
> following note is given:
>
> ; at beginning of line of text delimits non-simple data values
>
> It doesn't identify comment, so really should we skip the corresponding
> line?
>
> Regards,
> --
> Associate Professor, Theory and Simulation of Materials/Open Science &
> Software
> Department of Automotive Engineering, Machine Learning and Renewable
> Energy Materials, Xingtai Polytechnic College
> NO. 552 North Gangtie Road, Xindu District, Xingtai, Hebei, China
> E-mail: hongyi.zhao at gmail.com
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
--
Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8
Fax: +39 0116707855 - Dipartimento di Chimica, via
P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
*ICCC2020 has been postponed at 2022*
ICCC 2022 28 August - 2 September 2022, Rimini, Italy: http://www.iccc2020.com
International Conference on Coordination Chemistry (ICCC 2022)
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [QE-users] ibrav -13
Forgot to mention that I used Quantum Espresso 6.6. Il giorno sab 26 set 2020 alle ore 00:23 Carlo Nervi ha scritto: > > Dear QE developers, > I have a problem with ibrav -13. > The volume compute at first is too small from what it should be from a cif > file and differs if I run with ibrav=0 > > Attached please find the cif file, the input with ibrav=-13 and with > ibrav=0 > With ibrav=-13 the first volume is computed as > unit-cell volume =1198.4953 (a.u.)^3 > > With ibrav=0 the first volume is computed as > unit-cell volume =2396.9906 (a.u.)^3 > > These are the step 0 of the scf calculations. The right volume seems to be > that one with ibrav=0. > I will appreciate any suggestions. Am I doing something wrong? > Thank you, > Carlo > > -- > > ---- > Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 > Fax: +39 0116707855 - Dipartimento di Chimica, via > P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ > > ICCC 2020 5-10 July 2020, Rimini, Italy: http://www.iccc2020.com > International Conference on Coordination Chemistry (ICCC 2020) > > <http://www.iccc2020.com/> > > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ICCC 2020 5-10 July 2020, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2020) <http://www.iccc2020.com/> ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] ibrav -13
Dear QE developers, I have a problem with ibrav -13. The volume compute at first is too small from what it should be from a cif file and differs if I run with ibrav=0 Attached please find the cif file, the input with ibrav=-13 and with ibrav=0 With ibrav=-13 the first volume is computed as unit-cell volume =1198.4953 (a.u.)^3 With ibrav=0 the first volume is computed as unit-cell volume =2396.9906 (a.u.)^3 These are the step 0 of the scf calculations. The right volume seems to be that one with ibrav=0. I will appreciate any suggestions. Am I doing something wrong? Thank you, Carlo -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ICCC 2020 5-10 July 2020, Rimini, Italy: http://www.iccc2020.com International Conference on Coordination Chemistry (ICCC 2020) <http://www.iccc2020.com/> test_-13.in Description: Binary data 1008736.cif Description: CIF chemical test test_0.in Description: Binary data ___ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users
[QE-users] cell_dofree = 'volume'
Dear QE community,
I apologize if this topic has been already discussed in the past.
I'm using QE 6.4.1.
I'd like to perform vc-relax calculations, but keeping strictly fixed the
symmetry of the crystal.
I supposed that the keyword
cell_dofree='volume'
should be the right option to use, but pw.x exit complaining that
cell_dofree='volume' can be used only if ibrav=1.
Is there any technical reasons for this?
I was supposing that simply changing only celldm(1) should work.
I tried, as alternative, cell_dynamics='damp-w', which apparently keep
almost the symmetry (but not strictly), but it shows an extremely slow
convergence (in my molecular crystal vc-relax converge in 17 bfgs steps,
while the damp-w vc-relax calculation is currently running the 127th step).
Thank you,
Carlo
P.S.: I suppose that simple deleting the below bold rows in cell_base.f90
(subroutine init_dofree) would not work..
CASE ( 'volume' )
!CALL errore(' init_dofree ', &
! ' cell_dofree = '//TRIM(cell_dofree)//' not yet
implemented ', 1 )
* IF ( ibrav /= 1 ) THEN*
*CALL errore('cell_dofree', 'Isotropic expansion is only
allowed for ibrav=1; i.e. for simple cubic', 1)*
* END IF*
iforceh = 0
iforceh(1,1) = 1
iforceh(2,2) = 1
iforceh(3,3) = 1
isotropic= .TRUE.
--
Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8
Fax: +39 0116707855 - Dipartimento di Chimica, via
P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
Re: [Pw_forum] how to fix cellside
Do you mean fixed cell? In this case: calculation = 'relax' HTH, Carlo 2018-01-16 8:13 GMT+01:00 Maxim Arsentev : > Dear Quantum-Espresso users and developers, > > I need to fix cell side during geometry optimization in espresso. How it > can be done? > > Best wishes, > Maxim Arsent'ev, Ph.D. (Chemistry) > Laboratory of research of nanostructures > Institute of Silicate Chemistry of RAS > > > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- -------- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] projwfc partially occupied levels above Fermi level
Dear Chris, what kind of system are you studying? [Fe(H2O)6]3+ is a high-spin, whereas [Fe(CN)6]3+ is a low-spin system. Depending on the strength of the field you should have low spin or high spin. I guess you have to point out the magnetic properties explicitly in the input file. Carlo 2017-11-23 8:00 GMT+01:00 Christoph Wolf : > Dear all, > > I have a probably silly question I cannot wrap my head around: > > a sixfold coordinated FeO6 gives me quite nice, texbtook like > crystal-field split t2g (dxz, yz, xy) and eg (z^2 x^2-y^2) orbitals after > running it through projwfc. I am having a hard time interpreting the > charges, however: > > total charge = 14.9828, p = 5.9918, pz= 1.9973, px= 1.9973, py= > 1.9973, > total charge = 14.9828, d = 6.7326, dz2= 0.4228, dxz= 1.9623, dyz= > 1.9623, dx2-y2= 0.4228, dxy= 1.9623, > > i.e. the t2g are doubly occupied but the eg have half an electron each > (~1e in total) which lies above the Ef by ~0.8 eV (i.e. much higher than > the broadening which was 0.005 Ry); Is this an "residual" electron from the > ligand field and why is it picket up by the eg orbitals? > > Thanks in advance for your help and time! > > Best regards, > Chris > > -- > Postdoctoral Researcher > Center for Quantum Nanoscience, Institute for Basic Science > Ewha Womans University, Seoul, South Korea > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] wrong pdos with pbesol
Dear Maxim, I personally never had great feeling with pbesol with Li and light atoms. I don't know if this could help, but in my cases it gave, after vc-relax, a cell far from the experimental data. I preferred to use pw86pbe with xdm correction and ad-hoc selected parameters (a1 = 1.2153 and a2 = 2.3704). See http://pubs.acs.org/doi/abs/10.1021/acs.jpcc.7b00821 and http://www.sciencedirect.com/science/article/pii/S0360319916305870 All the best, Carlo 2017-05-21 21:20 GMT+02:00 Максим Арсентьев : > Dear all, > > I want to calculate PDOS of Li2MnSiO4 ang dot this (Mn 3d is not at the > conduction band as in other papers): > I used this input and ultrasoft-PBESol pseudos form here > http://theossrv1.epfl.ch/Main/Pseudopotentials > > > > Many thanks, > Maxim Arsent'ev, Ph.D. (Chemistry) > Laboratory of research of nanostructures > Institute of Silicate Chemistry of RAS > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- ---- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] problem during pw4gww.x calculations
Hi Paolo! in pola_lanczos.f90 If I move the #if defined(__SCALAPACK) just after the type declaration INTEGER, EXTERNAL :: indxg2l,indxg2p REAL(kind=DP), EXTERNAL :: ddot REAL(kind=DP) :: sca INTEGER :: desc_a(9),desc_b(9) the warning disappear. It seems that the subroutine is never called in my case (no scalapack) On the other hand in the code o_1psi.f90, line 1184 call o_1psi_gamma( numv, v_states, psi1, psi2) should not be: call o_1psi_gamma_real( numv, v_states, psi1, psi2) All the best, Carlo 2017-05-16 19:55 GMT+02:00 Paolo Giannozzi : > Hi Carlo > > On Tue, May 16, 2017 at 6:08 PM, Carlo Nervi wrote: > > > in Modules: Warning: Nonexistent include directory '../ELPA/src' > > Nonrelevant problem > > > Warning: Type mismatch in argument 'state1' at (1); passed COMPLEX(8) to > > REAL(4) > > this may or may not be a problem. I find more worrying that a code > that is supposed to be double precision everywhere seems to contain > single-precision variables > > > and this: > > > > o_1psi.f90:1184:52: > > > >call o_1psi_gamma( numv, v_states, psi1, psi2) > > 1 > > Warning: Missing actual argument for argument 'l_freq' at (1) > > > > are anomalous. > > definitely so. > > > Could be these the reasons? > > for the error reported at the beginning of this thread? no, that one > is due to a missing deallocation, or to a repeated allocation (routine > called more than once, variable allocated every time and not > deallocated meanwhile). Sometimes this kind of errors may go unnoticed > and produce memory leaks. > > > I checked the code and apparently all is okay (I even move the subroutine > > orthonormalize_two_manifolds_scalapack() before its use, but does not > help. > > gfortran bug? > > or subroutine never called? > > > The subroutine is called also from self_lanczos.f90, but this is not > giving > > any errors like the following? > > Note: The following floating-point exceptions are signalling: > > IEEE_DIVIDE_BY_ZERO > > Note: The following floating-point exceptions are signalling: > > IEEE_DIVIDE_BY_ZERO IEEE_DENORMAL > > Note: The following floating-point exceptions are signalling: > > IEEE_OVERFLOW_FLAG IEEE_UNDERFLOW_FLAG IEEE_DENORMAL > > IEEE_DIVIDE_BY_ZERO and IEEE_OVERFLOW_FLAG are serious; > IEEE_UNDERFLOW_FLAG and IEEE_DENORMAL are not. > > The expert of GWL is another Paolo, not me, so there isn't much else I can > say. > > Paolo > > > > > Carlo > > > > > > > > > > 2017-05-16 16:02 GMT+02:00 Mortaza Aghtar : > >> > >> > >> Dear All, > >> > >> I want to run GW calculations for boron-nitride sheet. First I ran a scf > >> calculation > >> for only the gamma point. Then I ran pw4gww.x to get the wannier's > >> matrices. But > >> after some steps I get this error in the log file: > >> > >> > >> forrtl: severe (151): allocatable array is already allocated > >> Image PCRoutineLine > Source > >> pw4gww.x 00A998FB Unknown Unknown > >> Unknown > >> pw4gww.x 0048011D matrix_wannier_ga 274 > >> matrix_wannier_gamma.f90 > >> pw4gww.x 0042B848 go_wannier_ 104 > >> wannier.f90 > >> pw4gww.x 00412A02 produce_wannier_g 216 > >> produce_wannier_gamma.f90 > >> pw4gww.x 0040D707 MAIN__494 > >> pw4gww.f90 > >> pw4gww.x 0040C7DE Unknown Unknown > >> Unknown > >> libc.so.6 7F0EB0E1DB35 Unknown Unknown > >> Unknown > >> pw4gww.x 00000040C6E9 Unknown Unknown > >> Unknown > >> - > >> > >> All the calculations are in parallel for the wave functions. > >> > >> Thank you in advance. > >> > >> Regards, > >> Mortaza > >> > >> > >> Dr Mortaza Aghtar > >> Postdoctoral research fellow > >> Ulm University, > >> Germany > >> > >> > >> > >> > >> > >> ___ > >> Pw_forum mailing list > >> Pw_forum@pwscf.org > >> http://pwscf.org/mailman/listinfo/pw_forum > > > > > > > > > > -- > > > >
Re: [Pw_forum] problem during pw4gww.x calculations
Paolo, I don't know if this could help, but while I'm compiling qe-6.1 I got several Warnings: in Modules: Warning: Nonexistent include directory '../ELPA/src' [-Wmissing-include-dirs] in PW/src: Warning: Nonexistent include directory '../include' [-Wmissing-include-dirs] I think these are harmless. However, this: in GWW/pw4gww: pola_lanczos.f90:1433:52: call orthonormalize_two_manifolds_scalapack(old_basis, nglobal_old,v_basis, nstates_eff, threshold, new_basis, nglobal) 1 Warning: Type mismatch in argument 'state1' at (1); passed COMPLEX(8) to REAL(4) and this: o_1psi.f90:1184:52: call o_1psi_gamma( numv, v_states, psi1, psi2) 1 Warning: Missing actual argument for argument 'l_freq' at (1) are anomalous. Could be these the reasons? I checked the code and apparently all is okay (I even move the subroutine orthonormalize_two_manifolds_scalapack() before its use, but does not help. gfortran bug? The subroutine is called also from self_lanczos.f90, but this is not giving any errors like the following? Note: The following floating-point exceptions are signalling: IEEE_DIVIDE_BY_ZERO Note: The following floating-point exceptions are signalling: IEEE_DIVIDE_BY_ZERO IEEE_DENORMAL Note: The following floating-point exceptions are signalling: IEEE_OVERFLOW_FLAG IEEE_UNDERFLOW_FLAG IEEE_DENORMAL Carlo 2017-05-16 16:02 GMT+02:00 Mortaza Aghtar : > > Dear All, > > I want to run GW calculations for boron-nitride sheet. First I ran a scf > calculation > for only the gamma point. Then I ran pw4gww.x to get the wannier's > matrices. But > after some steps I get this error in the log file: > > > forrtl: severe (151): allocatable array is already allocated > Image PCRoutineLineSource > pw4gww.x 00A998FB Unknown Unknown Unknown > pw4gww.x 0048011D matrix_wannier_ga 274 > matrix_wannier_gamma.f90 > pw4gww.x 0042B848 go_wannier_ 104 > wannier.f90 > pw4gww.x 00412A02 produce_wannier_g 216 > produce_wannier_gamma.f90 > pw4gww.x 0040D707 MAIN__494 > pw4gww.f90 > pw4gww.x 0040C7DE Unknown Unknown Unknown > libc.so.6 7F0EB0E1DB35 Unknown Unknown Unknown > pw4gww.x 0040C6E9 Unknown Unknown Unknown > - > > All the calculations are in parallel for the wave functions. > > Thank you in advance. > > Regards, > Mortaza > > > Dr Mortaza Aghtar > Postdoctoral research fellow > Ulm University, > Germany > > > > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] Compiling QE 6.1 with Intel
Dear QE community, a colleague of mine is using Intel compiler and libraries to compile QE 6.1. We used compilers_and_libraries_2017.1.132 and parallel_studio_xe_2017.1.043 I got apparently normal results using "mpirun -np 21", but very different results using "mpirun -np 21 -nk 7", and relevant longer time of calculations. I suspect there is something wrong with libraries or mpi. Before to dig more deep in the debugging, do anybody is aware of any inconsistencies/anomalies using the mentioned compiler/libraries? We are using also docker... Thank you, Carlo -- -------- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] cif2qe Error
This is my output: ! Generated by using cif2qe Version 1.2 - Date: Sat Feb 11 10:37:37 CET 2017 ! _symmetry_space_group_name_H-M = 'P ! _symmetry_Int_Tables_number = 115 ! _symmetry_cell_setting = tetragonal ! a=6.926 b=6.926 c=6.41 alpha=90 beta=90 gamma=90 ! Found by cif2qe: lattice = tetragonalSpace group = P-4m2 ibrav = 6 ! ! ! Symmetry found: ! 1 +x,+y,+z[+x] [+y] [+z] ! 2 -y,+x,-z[-y] [+x] [-z] ! 3 -x,-y,+z[-x] [-y] [+z] ! 4 +y,-x,-z[+y] [-x] [-z] ! 5 -x,+y,+z[-x] [+y] [+z] ! 6 +y,+x,-z[+y] [+x] [-z] ! 7 +x,-y,+z[+x] [-y] [+z] ! 8 -y,-x,-z[-y] [-x] [-z] &CONTROL title = 'tt' calculation = 'relax' restart_mode = 'from_scratch' outdir = './1' pseudo_dir = '../PP/atompaw' prefix = 'tt' disk_io = 'none' verbosity = 'default' etot_conv_thr = 0.0001 forc_conv_thr = 0.001 nstep = 400 tstress = .true. tprnfor = .true. / &SYSTEM ibrav = 0 nat = 15 ntyp = 2 ecutwfc = 60 ecutrho = 480 vdw_corr = 'xdm' xdm_a1 = 1.2153 xdm_a2 = 2.3704 / &ELECTRONS electron_maxstep = 200 conv_thr = 1.0D-7 diago_thr_init = 1e-4 startingpot = 'atomic' startingwfc = 'atomic' mixing_mode = 'plain' mixing_beta = 0.5 mixing_ndim = 8 diagonalization = 'david' / &IONS ion_dynamics = 'bfgs' / ATOMIC_SPECIES O 15.999400 O.pw86pbe-n-kjpaw_psl.1.0.0.UPF Si 28.085500 Si.pw86pbe-n-kjpaw_psl.1.0.0.UPF ATOMIC_POSITIONS crystal O 0.189800 0.00 0.355000 O -0.00 0.189800 0.645000 O 0.810200 -0.00 0.355000 O 0.00 0.810200 0.645000 O 0.189700 0.189700 0.00 O 0.810300 0.189700 -0.00 O 0.810300 0.810300 0.00 O 0.189700 0.810300 -0.00 O 0.50 0.00 0.118300 O -0.00 0.50 0.881700 Si 0.267900 0.00 0.118400 Si -0.00 0.267900 0.881600 Si 0.732100 -0.00 0.118400 Si 0.00 0.732100 0.881600 Si 0.00 0.00 0.50 K_POINTS automatic 2 2 2 0 0 0 CELL_PARAMETERS bohr 13.08824313873762 0.00 0.00 0.00 13.08824313873762 0.00 0.00 0.00 12.11314445846205 2017-02-11 9:31 GMT+01:00 Carlo Nervi : > This is a problem of awk. Make sure you are using gawk version 4 or newer.. > > HTH, > > Carlo > > 2017-02-11 2:41 GMT+01:00 William Joseph Ivan DeBenedetti < > wj...@cornell.edu>: > >> Dear QE Users, >> >> I am attempting to convert a crystallographic information file using the >> cif2qe.sh shell, I am getting some errors which I could not find in the >> previous PW_forum, which are also not in the cif nor shell file. >> >> Commands I have used: ./cif2qe.sh [filename] and ./cif2qe.sh [filename] > >> [filename].in >> >> cif2qe.sh Version 1.2 >> Bash Version : 3.2.57(1)-release >> QE version 6.0 >> Macbook Pro i7 using iFort (2017) and MPICH 3.2 >> >> This is the error I am receiving: >> >> Williams-MacBook-Pro-2:tools WJIDeBenedetti$ ./cif2qe.sh EDI >> awk: syntax error at source line 27 >> context is >> >>> KeyW[0][ <<< 1] ="_cell_length_a"; KeyW[1][1]="" >> awk: illegal statement at source line 28 >> awk: illegal statement at source line 28 >> >> >> The following is the crystallographic information file: >> >> data_EDI >> #**
Re: [Pw_forum] cif2qe Error
This is a problem of awk. Make sure you are using gawk version 4 or newer.. HTH, Carlo 2017-02-11 2:41 GMT+01:00 William Joseph Ivan DeBenedetti : > Dear QE Users, > > I am attempting to convert a crystallographic information file using the > cif2qe.sh shell, I am getting some errors which I could not find in the > previous PW_forum, which are also not in the cif nor shell file. > > Commands I have used: ./cif2qe.sh [filename] and ./cif2qe.sh [filename] > > [filename].in > > cif2qe.sh Version 1.2 > Bash Version : 3.2.57(1)-release > QE version 6.0 > Macbook Pro i7 using iFort (2017) and MPICH 3.2 > > This is the error I am receiving: > > Williams-MacBook-Pro-2:tools WJIDeBenedetti$ ./cif2qe.sh EDI > awk: syntax error at source line 27 > context is > >>> KeyW[0][ <<< 1] ="_cell_length_a"; KeyW[1][1]="" > awk: illegal statement at source line 28 > awk: illegal statement at source line 28 > > > The following is the crystallographic information file: > > data_EDI > #*** > *** > # > # CIF taken from the IZA-SC Database of Zeolite Structures > # Ch. Baerlocher and L.B. McCusker > # Database of Zeolite Structures: http://www.iza-structure.org/databases/ > # > # All data on this site have been placed in the public domain. > # If you use this work in a scientific publication, you are obligated to > # cite its origin. > # > # The atom coordinates and the cell parameters were optimized with DLS76 > # assuming a pure SiO2 composition. > # > #*** > *** > > _cell_length_a 6.9260(0) > _cell_length_b 6.9260(0) > _cell_length_c 6.4100(0) > _cell_angle_alpha 90.(0) > _cell_angle_beta90.(0) > _cell_angle_gamma 90.(0) > > _symmetry_space_group_name_H-M 'P -4 m 2' > _symmetry_Int_Tables_number 115 > _symmetry_cell_setting tetragonal > > loop_ > _symmetry_equiv_pos_as_xyz > '+x,+y,+z' > '-y,+x,-z' > '-x,-y,+z' > '+y,-x,-z' > '-x,+y,+z' > '+y,+x,-z' > '+x,-y,+z' > '-y,-x,-z' > > loop_ > _atom_site_type_symbol > _atom_site_type_symbol > _atom_site_fract_x > _atom_site_fract_y > _atom_site_fract_z > O1O 0.18980.0.3550 > O2O 0.18970.18970. > O3O 0.50000.0.1183 > T1Si0.26790.0.1184 > T2Si0.0.0.5000 > > I have tried a few other CIFs from other databases with no luck. Has > anyone else experienced this? > > Many thanks, > > Will DeBenedetti > Cornell University > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] ecutrho <= ecutwfc?!?
ETERS {alat}
> 1.d0 0.d0 0.d0
> 0.d0 1.d0 0.d0
> 0.d0 0.d0 1.007697236034d0
>
>
> It clearly shows that ecutrho = 4*ecutwfc. The job crashes when run on 128
> processors; however when run on 16 processors (just for testing), the
> output does not crash (but terminates due to maximum allocated time being
> reached). In the heading of the 16-processor job output, I observe that the
> pseudopotential for Arsenic (As) is missing. I guess the problem might be
> with this particular pseudopotential, but I cannot spot it within the UPF
> file. The PAW pseudopotentials I use all come from pslibrary v. 1.0.0, and
> are all generated using the same QE on Stampede. The headings of the
> pseudopotentials are attached here (the whole files are too big). Please
> let me know what you think the problem might be. Thank you very much!
>
> --
> Jingyang 'Mark' Wang
> School of Applied and Engineering Physics
> Cornell University
>
> ___
> Pw_forum mailing list
> Pw_forum@pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
--
Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8
Fax: +39 0116707855 - Dipartimento di Chimica, via
P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] cif2qe conversion failure
Hello, the problem is related to your cif input file. Instead of "_atom_site_label" you should have "_atom_site_type_symbol" because the label could be any strings... HTH, Carlo 2016-04-13 18:14 GMT+02:00 Tsung-Lung Li : > Dear QE Users: > > I try to use cif2qe to convert the structure of gypsum with command > > cif2qe -i filename > filename.in > > but the resultant file contains an empty section of ATOMIC_POSITIONS. I > wonder this problem can be fixed. Thank you in advance. > > Listed below is the cif downloaded from American Mineralogist Crystal > Structure Database. > > > data_global > _chemical_name_mineral 'Gypsum' > loop_ > _publ_author_name > 'Comodi P' > 'Nazzareni S' > 'Zanazzi P F' > 'Speziale S' > _journal_name_full 'American Mineralogist' > _journal_volume 93 > _journal_year 2008 > _journal_page_first 1530 > _journal_page_last 1537 > _publ_section_title > ; > High-pressure behavior of gypsum: A single-crystal X-ray study > Note: P = 0.0001 GPa > ; > _database_code_amcsd 0004651 > _chemical_compound_source 'Valle di Caramanico, Abruzzo, Italy' > _chemical_formula_sum 'Ca S O6 H4' > _cell_length_a 6.277 > _cell_length_b 15.181 > _cell_length_c 5.672 > _cell_angle_alpha 90 > _cell_angle_beta 114.11 > _cell_angle_gamma 90 > _cell_volume 493.340 > _exptl_crystal_density_diffrn 2.318 > _symmetry_space_group_name_H-M 'C 1 2/c 1' > loop_ > _space_group_symop_operation_xyz >'x,y,z' >'1/2+x,1/2+y,z' >'x,-y,1/2+z' >'1/2+x,1/2-y,1/2+z' >'-x,y,1/2-z' >'1/2-x,1/2+y,1/2-z' >'-x,-y,-z' >'1/2-x,1/2-y,-z' > loop_ > _atom_site_label > _atom_site_fract_x > _atom_site_fract_y > _atom_site_fract_z > _atom_site_U_iso_or_equiv > Ca 0.0 0.17050 0.25000 0.01170 > S 0.0 0.32727 0.75000 0.00990 > O 0.08319 0.27218 0.59103 0.01690 > O 0.19997 0.38195 0.91298 0.01690 > O -0.20823 0.06826 -0.07831 0.02410 > H -0.25800 0.08700 -0.23400 0.03300 > H -0.24400 0.02000 -0.07700 0.04400 > > > > > > Listed below is the resultant QE file > > ! Generated by using cif2qe Version 1.2 - Date: Wed Apr 13 21:35:09 CST > 2016 > ! _symmetry_space_group_name_H-M = > ! _symmetry_Int_Tables_number = > ! _symmetry_cell_setting = > ! a=6.277 b=15.181 c=5.672 alpha=90 beta=114.11 gamma=90 > ! Found by cif2qe: lattice = monoclinicSpace group =ibrav = 0 > ! > ! > ! Symmetry found: > ! 1 x,y,z[x] [y] [z] > ! 2 1/2+x,1/2+y,z[1/2+x] [1/2+y] [z] > ! 3 x,-y,1/2+z[x] [-y] [1/2+z] > ! 4 1/2+x,1/2-y,1/2+z[1/2+x] [1/2-y] [1/2+z] > ! 5 -x,y,1/2-z[-x] [y] [1/2-z] > ! 6 1/2-x,1/2+y,1/2-z[1/2-x] [1/2+y] [1/2-z] > ! 7 -x,-y,-z[-x] [-y] [-z] > ! 8 1/2-x,1/2-y,-z[1/2-x] [1/2-y] [-z] > &CONTROL > title = 'AMS_DATA-0.0001GPa' > calculation = 'relax' > restart_mode = 'from_scratch' >outdir = './1' >pseudo_dir = '../PP/atompaw' >prefix = 'AMS_DATA-0.0001GPa' > disk_io = 'none' > verbosity = 'default' > etot_conv_thr = 0.0001 > forc_conv_thr = 0.001 > nstep = 400 > tstress = .true. > tprnfor = .true. > / > &SYSTEM > ibrav = 0 > nat = 0 > ntyp = 0 > ecutwfc = 60 > ecutrho = 480 > vdw_corr = 'xdm' >xdm_a1 = 1.2153 >xdm_a2 = 2.3704 > / > &ELECTRONS > electron_maxstep = 200 > conv_thr = 1.0D-7 >diago_thr_init = 1e-4 > startingpot = 'atomic' > startingwfc = 'atomic' > mixing_mode = 'plain' > mixing_beta = 0.5 > mixing_ndim = 8 > diagonalization = 'david' > / > &IONS > ion_dynamics =
Re: [Pw_forum] nmr chemical shift
The best approach is to compute the NMR parameters of a standard substance thet Solid State NMR people are using. That depend on the nuclei you are investigating. For example hexametilbenzene. Once you have the absolute shielding of the reference it is easy to report all the other chemical shifts. HTH, Carlo 2016-03-08 12:07 GMT+01:00 Tiana Davide : > Dear all, > > > I am calculating some nmr spectra using Gipaw. > > Since I've never done for solid state (namely periodic bulk) I was > wondering how can I calculate the chemical shift? > > Is the reference still TMS? > > if yes do I need to somehow find the TMS crystal structure or can I simply > run a single molecule TMS? > > if not how can I do it? > > > Cheers > > Davide > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] cif to pwi
000 0.00 Al 0.00 0.50 0.00 Al 0.00 0.00 0.50 Al 0.25 0.00 0.75 Al 0.25 0.50 0.25 Al 0.75 0.00 0.25 Al 0.75 0.50 0.75 Al 0.00 0.25 0.75 Al 0.00 0.75 0.25 Al 0.50 0.25 0.25 Al 0.50 0.75 0.75 Al 0.25 0.75 0.00 Al 0.25 0.25 0.50 Al 0.75 0.75 0.50 Al 0.75 0.25 0.00 O 0.264400 0.264400 0.264400 O 0.264400 0.764400 0.764400 O 0.764400 0.264400 0.764400 O 0.764400 0.764400 0.264400 O 0.014400 0.235600 0.514400 O 0.014400 0.735600 0.014400 O 0.514400 0.235600 0.014400 O 0.514400 0.735600 0.514400 O 0.485600 0.764400 0.985600 O 0.485600 0.264400 0.485600 O 0.985600 0.764400 0.485600 O 0.985600 0.264400 0.985600 O 0.764400 0.485600 0.985600 O 0.764400 0.985600 0.485600 O 0.264400 0.485600 0.485600 O 0.264400 0.985600 0.985600 O 0.235600 0.014400 0.514400 O 0.235600 0.514400 0.014400 O 0.735600 0.014400 0.014400 O 0.735600 0.514400 0.514400 O 0.735600 0.235600 0.235600 O 0.735600 0.735600 0.735600 O 0.235600 0.235600 0.735600 O 0.235600 0.735600 0.235600 O 0.014400 0.514400 0.235600 O 0.014400 0.014400 0.735600 O 0.514400 0.514400 0.735600 O 0.514400 0.014400 0.235600 O 0.485600 0.985600 0.764400 O 0.485600 0.485600 0.264400 O 0.985600 0.985600 0.264400 O 0.985600 0.485600 0.764400 K_POINTS automatic 2 2 2 0 0 0 CELL_PARAMETERS 15.27919160755835 0.00 0.00 0.00 15.27919160755835 0.00 0.00 0.00 15.27919160755835 2015-04-16 9:50 GMT+02:00 Éric Germaneau : > Here are the structures in fractional coordinates using P1 symmetry. > That may helps. > > > On 04/16/2015 03:10 PM, Winfred Mulwa wrote: > > Dear all, > Please somebody help, how do i convert the attached cif file > to pwi. > Regards >Mulwa Winfred. > D Phil Student, Computational Material Science Group, > University of the Free State - QwaQwa, > South Africa. > > > ___ > Pw_forum mailing > listPw_forum@pwscf.orghttp://pwscf.org/mailman/listinfo/pw_forum > > > -- > Éric Germaneau (艾海克), Specialist > Center for High Performance Computing > Shanghai Jiao Tong University > Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China > M:german...@sjtu.edu.cn P:+86-136-4161-6480 W:http://hpc.sjtu.edu.cn > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] Problems in gipaw EFG calculations
Thanks Filippo, however my linux settings are already "unlimited". I guess the problem is more subtle. By now I'm using the acml fftw library (version 6), which seems to be efficient (its speed seems to be comaprable to the fftw3 previously mentioned, at least with Opteron). Up to (cross fingers) no problems. :) Carlo 2015-03-13 10:04 GMT+01:00 Filippo Spiga : > On Mar 9, 2015, at 6:05 PM, Paolo Giannozzi > wrote: > > Just in case: try to replace the automatic array > > complex(dp) :: psic(dffts%nnr) > > with an allocatable array. Sometimes large automatic arrays fill the > > stack and cause strange crash. > > Or try: > > ulimit -s unlimited > > But make psic allocatable is indeed a more elegant solution. > > F > > -- > Mr. Filippo SPIGA, M.Sc. > http://filippospiga.info ~ skype: filippo.spiga > > «Nobody will drive us out of Cantor's paradise.» ~ David Hilbert > > * > Disclaimer: "Please note this message and any attachments are CONFIDENTIAL > and may be privileged or otherwise protected from disclosure. The contents > are not to be disclosed to anyone other than the addressee. Unauthorized > recipients are requested to preserve this confidentiality and to advise the > sender immediately of any error in transmission." > > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] Problems in gipaw EFG calculations
Dear QE users, for some strange reasons I have problems in runnin gipaw.x on the quartz test/example. NMR calculations runs ok, but not efg. efg calculation always exit with "Segmentation fault" error. I'm using the library fftw-3.3.4 locally compiled. Anyone know whether the -D__FFTW3 flag is incompatible with fftw-3.3.4? pw.x seems to run ok. if I compile with the -D__FFTW flag (i.e. without FFTW3), efg runs okay. Here is the output of the gipaw.x version with FFTW3 : Program received signal SIGSEGV: Segmentation fault - invalid memory reference. Backtrace for this error: #0 0x2B60E935E78F #1 0x2B60E935ED84 #2 0x2B60E9FFDE6F #3 0x969808 in n1bv_9 at n1bv_9.c:0 #4 0x9CEA57 in apply at direct.c:0 #5 0x9CB476 in apply_dit at ct.c:0 #6 0x895ECD in apply at vrank-geq1.c:0 #7 0x9CAE17 in apply at buffered.c:0 #8 0x895A16 in apply at rank-geq2.c:0 #9 0x895ECD in apply at vrank-geq1.c:0 #10 0x6B7DE2 in __fft_scalar_MOD_cft_2xy #11 0x6B3864 in __fft_parallel_MOD_tg_cft3s #12 0x6B14A3 in invfft_x_ #13 0x49899B in get_smooth_density_ #14 0x499292 in efg_ #15 0x44EBD4 in MAIN__ at gipaw_main.f90:0 [1]+ Segmentation fault gipaw.x -in quartz-efg.in > quartz-efg.out Thankx, Carlo -- ---- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] phonon parallel calculations
I did some other test with the "GRID" approach above outlined, and I cannot understand where I'm wrong. I tried different libraries during the compilation process, but while the "normal" phonon calculations runs smoothly (but slower, about 1 hour), the "GRID" approach always gave me bad (negative) frequencies (but much faster, about 10 minutes). I'm using ecut=60 and erho=480 The phonon results from without using the "-ni 8" approach is: freq (1) = -0.219517 [THz] = -7.322283 [cm-1] freq (2) = 0.237288 [THz] = 7.915064 [cm-1] freq (3) = 0.412515 [THz] = 13.760017 [cm-1] freq (4) = 0.888921 [THz] = 29.651216 [cm-1] freq (5) = 1.818322 [THz] = 60.652688 [cm-1] freq (6) = 3.267016 [THz] = 108.975924 [cm-1] freq (7) = 3.94 [THz] = 131.572497 [cm-1] freq (8) = 13.702502 [THz] = 457.066270 [cm-1] freq (9) = 15.98 [THz] = 533.257782 [cm-1] freq ( 10) = 21.306349 [THz] = 710.703307 [cm-1] freq ( 11) = 29.251729 [THz] = 975.732641 [cm-1] freq ( 12) = 31.330252 [THz] =1045.064724 [cm-1] freq ( 13) = 40.094614 [THz] =1337.412380 [cm-1] freq ( 14) = 91.247143 [THz] =3043.677078 [cm-1] freq ( 15) = 94.196285 [THz] =3142.049865 [cm-1] If I use the "-ni 8" (GRID) I got: freq (1) = -14.549865 [THz] =-485.331266 [cm-1] freq (2) = -10.853880 [THz] =-362.046453 [cm-1] freq (3) = -9.921965 [THz] =-330.961126 [cm-1] freq (4) = -9.680609 [THz] =-322.910351 [cm-1] freq (5) = -9.593462 [THz] =-320.003431 [cm-1] freq (6) = -9.461406 [THz] =-315.598531 [cm-1] freq (7) = -8.724432 [THz] =-291.015732 [cm-1] freq (8) = -4.051543 [THz] =-135.144927 [cm-1] freq (9) = 8.067478 [THz] = 269.102094 [cm-1] freq ( 10) = 19.972800 [THz] = 666.220885 [cm-1] freq ( 11) = 29.137064 [THz] = 971.907843 [cm-1] freq ( 12) = 30.784477 [THz] =1026.859612 [cm-1] freq ( 13) = 39.868296 [THz] =1329.863222 [cm-1] freq ( 14) = 91.702737 [THz] =3058.874040 [cm-1] freq ( 15) = 94.287387 [THz] =3145.088688 [cm-1] A lot of negative frequencies appear and sometimes (with a bigger system) I got a in the matdyn file (in this case an error is printed). Any idea? Anybody else had this kind of problem? Thank you, Carlo > > -- ---- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] Initial geometry optimization cannot converge
I did not read the structure, but it is not a trivial problem. You have to choose among several different approaches, and may test all of them: 1) try to draw a molecule with "chemical sense" (may using "molecular" calculations") and only later try to fit in the solid state 2) substitute an atom with that of your interest in an already optimized structure and hope... 3) try to find a suitable "space" for an extra interstitial atom and hope... None is expected to be straightforward, and all require a certain degree of "chemical experience"... HTH, Carlo 2015-02-25 3:18 GMT+01:00 Chenyang Lee : > Dear experts in PWforum, > > I am experiencing difficulties relaxing some of the structures of B-HAp. > Because the Boron group substitutions are proposed (unknown), thus the > initial structure may deviate a lot from the final optimized one. Quantum > Espresso seems not to be very capable to do these jobs (no offense) and the > SCF cycles usually exceed 500 steps and it still cannot converge. > I am attaching the input file, and I’d appreciate it if someone who is > able to identify the problem can help me solve it. > > Thank you very much, > Chenyang > __ > Chenyang Li > Undergraduate Student, Materials Science and Engineering > The Pennsylvania State University, University Park > Phone: (814)441-4312 > > ___ > > &control > calculation = 'vc-relax' > restart_mode = 'from_scratch' > prefix = ‘BHAp' > pseudo_dir = '/gpfs/home/col5422/work/HAp/pseudo/' > outdir = './scratch/' > tprnfor = .true. > forc_conv_thr = 1e-4 > etot_conv_thr = 1e-6 > nstep = 500 > / > &system > ibrav = 14 > A = 9.4166 > B = 9.4166 > C = 6.8745 > cosAC = 0.0 > cosAB = -0.5 > cosBC = 0.0 > nat = 52 > ntyp = 6 > ecutwfc = 40 > ecutrho = 400 > nosym = .true. > occupations = 'smearing' > smearing = 'gaussian' > degauss = 0.02 > / > &electrons > electron_maxstep = 500 > diagonalization = 'david' > mixing_mode = 'plain' > mixing_beta = 0.3 > conv_thr = 1e-6 > / > &ions > ion_dynamics = ‘damp' > / > &cell > cell_dynamics = ‘damp-pr' > / > > ATOMIC_SPECIES > H 1 H.pbe-kjpaw_psl.0.1.upf > B 1 B.pbe-n-kjpaw_psl.0.1.upf > C 1 C.pbe-n-kjpaw_psl.0.1.upf > O 1 O.pbe-n-kjpaw_psl.0.1.upf > P1 P.pbe-n-kjpaw_psl.0.1.upf > Ca 1 Ca.pbe-spn-kjpaw_psl.0.2.3.upf > > ATOMIC_POSITIONS crystal > H 0.00 0.00 0.40 > H 0.00 0.00 0.90 > H 0.567369 0.551274 0.972591 > H 0.629383 0.851436 0.797858 > H 0.357078 0.434312 0.660517 > H 0.809237 0.663097 0.589841 > B 0.581602 0.618142 0.689068 > C 0.310984 0.129561 0.635342 > O 0.348000 0.259000 0.073000 > O 0.741000 0.089000 0.073000 > O 0.911000 0.652000 0.073000 > O 0.259000 0.911000 0.573000 > O 0.089000 0.348000 0.573000 > O 0.259000 0.911000 0.927000 > O 0.089000 0.348000 0.927000 > O 0.348000 0.259000 0.427000 > O 0.741000 0.089000 0.427000 > O 0.911000 0.652000 0.427000 > O 0.329000 0.484000 0.25 > O 0.589000 0.466000 0.25 > O 0.516000 -0.155000 0.25 > O 0.534000 0.123000 0.25 > O 1.155000 0.671000 0.25 > O 0.877000 0.411000 0.25 > O 0.484000 1.155000 0.75 > O 0.466000 0.877000 0.75 > O-0.155000 0.329000 0.75 > O 0.123000 0.589000 0.75 > O 0.00 0.00 0.25 > O 0.00 0.00 0.75 > O 0.603664 0.524999 0.843110 > O 0.718661 0.681136 0.557507 > O 0.546871 0.739153 0.774413 > O 0.441199 0.509454 0.566674 > O 0.464549 0.231819 0.634250 > O 0.235429 0.067896 0.472588 > O 0.237756 0.091652 0.804032 > P 0.40 0.369000 0.25 > P 0.631000 0.031000 0.25 > P 0.969000 0.60 0.25 > P 0.369000 0.969000 0.75 > P 0.031000 0.40 0.75 > Ca0.33 0.67 0.001000 > Ca0.67 0.33 0.501000 > Ca0.67 0.33 0.999000 > Ca0.33 0.67 0.499000 > Ca0.246000 0.993000 0.25 > Ca0.007000 0.253000 0.25 > Ca0.747000 0.754000 0.25 > Ca0.754000 0.007000 0.75 > Ca0.993000 0.747000 0.75
[Pw_forum] amdlibm.h
Dear all, I'd like to test the library amdlibm by including the file amdlibm.h for testing and for (maybe) code speedup. This replace all the libm math library with the opteron specific instructions. For example: #undef cos #define cos amd_cos Where is the most appropriate place where to insert this include file that should be read in principle by all the source code? The library to link will be the amdlibm 3.1 I'm using also the acml 6.1.0.31 Thank you, Carlo Nervi -- ---- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] phonon parallel calculations
Thank you. Meanwhile I'm continue to get strange errors. RAM seems ok. Apparently I'm getting error only if I use the "-ni 8" option in ph.x. But it's random. Here is the last error: Program received signal SIGSEGV: Segmentation fault - invalid memory reference. Backtrace for this error: Program received signal SIGSEGV: Segmentation fault - invalid memory reference. Backtrace for this error: #0 0x2AF4EE34C78F #1 0x2AF4EE34CD84 #2 0x2AF4EEFEBE6F #3 0x2AF4EF032058 #4 0x51C509 in cgsolve_all_ #5 0x4E0486 in solve_linter_ #6 0x4BD7F6 in phqscf_ #7 0x45B666 in do_phonon_ #8 0x45758F in MAIN__ at phonon.f90:0 2015-02-20 10:06 GMT+01:00 Andrea Dal Corso : > On Fri, 2015-02-20 at 09:19 +0100, Carlo Nervi wrote: > > Dear QE developers, > > I would like to have your opinion on the procedure I followed in doing > > phonon calculations. > > I found that the use of nimages on my machine (8 numa zones) greatly > > improves the speed of some calculations, but the first (initialization?) > > part of the calculation (epsil=.true., i.e. the Electric Fields > > Calculation) is much faster when performed without the use of "-ni 8" > > switch in ph.x > > > > So, I tested the following procedure and apparently worked for a small > test > > crystal: > > > > 1) pw.x (vc-relax optimization) without using "-ni" > > 2) scf on the optimized structure containing "wf_collect=.true." and > using > > "-ni 8" > > pw.x has no image parallelization, so step 2 should be done without > images. The rest should be ok. > > HTH, > > Andrea > > > > 3) ph.x containing "only_init = .true." and without using "-ni" > > This calculate the electric fields > > 4) duplicate the directory _ph0 and copied to _ph1 ... _ph7 (in output > > directory) > > 5) ph.x containing "recover = .true." and using "-ni 8" > > This is the part where I get most of the speedup > > 6) ph.x of the input used in 5), but now without using "-ni" > > This is to collect the partial data and produce the final .out > > > > I would greatly appreciate any comments in understanding if this > procedure > > is correct. > > On a small crystal was working, but on a bigger crystal gave error. > > Either the procedure is not completely correct or I have to start to look > > for a problem in the RAM... > > > > Thank you, > > > > Carlo > > > > ___ > > Pw_forum mailing list > > Pw_forum@pwscf.org > > http://pwscf.org/mailman/listinfo/pw_forum > > -- > Andrea Dal Corso Tel. 0039-040-3787428 > SISSA, Via Bonomea 265 Fax. 0039-040-3787249 > I-34136 Trieste (Italy) e-mail: dalco...@sissa.it > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] phonon parallel calculations
Dear QE developers, I would like to have your opinion on the procedure I followed in doing phonon calculations. I found that the use of nimages on my machine (8 numa zones) greatly improves the speed of some calculations, but the first (initialization?) part of the calculation (epsil=.true., i.e. the Electric Fields Calculation) is much faster when performed without the use of "-ni 8" switch in ph.x So, I tested the following procedure and apparently worked for a small test crystal: 1) pw.x (vc-relax optimization) without using "-ni" 2) scf on the optimized structure containing "wf_collect=.true." and using "-ni 8" 3) ph.x containing "only_init = .true." and without using "-ni" This calculate the electric fields 4) duplicate the directory _ph0 and copied to _ph1 ... _ph7 (in output directory) 5) ph.x containing "recover = .true." and using "-ni 8" This is the part where I get most of the speedup 6) ph.x of the input used in 5), but now without using "-ni" This is to collect the partial data and produce the final .out I would greatly appreciate any comments in understanding if this procedure is correct. On a small crystal was working, but on a bigger crystal gave error. Either the procedure is not completely correct or I have to start to look for a problem in the RAM... Thank you, Carlo -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] EFG with GIPAW
Dear all, I compiled the svn version of gipaw (by typing make gipaw in espresso 5.1.1). While apparently the nmr gipaw job is working ok, I got the ierror message while I'm tryin to test the efg: Program received signal SIGSEGV: Segmentation fault - invalid memory reference. Backtrace for this error: #0 0x2AAB085FB78F #1 0x2AAB085FBD84 #2 0x2AAB0929AE6F #3 0x9654E8 in n1bv_9 at n1bv_9.c:0 #4 0x9CA737 in apply at direct.c:0 #5 0x9C7156 in apply_dit at ct.c:0 #6 0x891BAD in apply at vrank-geq1.c:0 #7 0x9C6AF7 in apply at buffered.c:0 #8 0x8916F6 in apply at rank-geq2.c:0 #9 0x891BAD in apply at vrank-geq1.c:0 #10 0x6B4C22 in __fft_scalar_MOD_cft_2xy #11 0x6B06A4 in __fft_parallel_MOD_tg_cft3s #12 0x6AFA63 in invfft_x_ #13 0x49871B in get_smooth_density_ #14 0x499012 in efg_ #15 0x44E974 in MAIN__ at gipaw_main.f90:0 Segmentation fault Any suggestions? thanks, Carlo -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] cif2qe problem of conversion
As already pointed out, try first to use the new and latest version of cif2qe.sh As general rule, pay attention to any residual CR/LF from Windows text file, they can scramble the logic of cif2qe.sh Carlo 2015-02-10 17:10 GMT+01:00 : > Dear users, > > > > I would like to convert a cif file to a QE file using cif2qe.sh shell, > nevertheless I have a problem in the conversion, I have no ATOMIC_POSITIONS > in my output file (see below). > > > > I am using espresso-5.1.1 with GNU Awk 4.0.1, and bash version 4.3.11 > > > > I run the code with ./cif2qe.sh (-i or –s or nothing) myfilecif > > > > *Here is my cif:* > > > > _symmetry_cell_setting 'Cubic' > > _symmetry_Int_Tables_nummber 218 > > _symmetry_space_group_name_H-M 'P -43n' > > _symmetry_space_group_name_Hall 'P -4n 2 3' > > _cell_length_a 15.2916(3) > > _cell_length_b 15.2916(3) > > _cell_length_c 15.2916(3) > > _cell_angle_alpha90.0 > > _cell_angle_beta 90.0 > > _cell_angle_gamma90.0 > > > > loop_ > > _symmetry_equiv_pos_as_xyz > > x,y,z > > x,-y,-z > > -x,y,-z > > -x,-y,z > > y,z,x > > -y,-z,x > > y,-z,-x > > -y,z,-x > > z,x,y > > -z,x,-y > > -z,-x,y > > z,-x,-y > > y+1/2,x+1/2,z+1/2 > > -y+1/2,x+1/2,-z+1/2 > > y+1/2,-x+1/2,-z+1/2 > > -y+1/2,-x+1/2,z+1/2 > > z+1/2,y+1/2,x+1/2 > > -z+1/2,-y+1/2,x+1/2 > > -z+1/2,y+1/2,-x+1/2 > > z+1/2,-y+1/2,-x+1/2 > > x+1/2,z+1/2,y+1/2 > > x+1/2,-z+1/2,-y+1/2 > > -x+1/2,-z+1/2,y+1/2 > > -x+1/2,z+1/2,-y+1/2 > > > > loop_ > > _atom_site_type_symbol > > _atom_site_fract_x > > _atom_site_fract_y > > _atom_site_fract_z > > Cs 0.3416 0.1369 0.0732 > > Cs 0.83263 0.16737 0.16737 > > Al 0.25000 0.5 0.0 > > H 0.25000 0.5 0.6 > > H 0.36116 0.5 0.0 > > > > #END > > > > *I obtain the following output:* > > > > ! Generated by using cif2qe Version - Date: mar. févr. 10 16:00:47 CET > 2015 > > ! _symmetry_space_group_name_H-M = > > ! _symmetry_Int_Tables_number = > > ! _symmetry_cell_setting = Cubic > > ! a=15.2916 b=15.2916 c=15.2916 alpha=90 beta=90 gamma=90 > > ! Found by cif2qe: lattice = cubicSpace group =ibrav = 0 > > ! > > ! > > ! Symmetry found: > > &CONTROL > > title = 'Test2' > > calculation = 'relax' > > restart_mode = 'from_scratch' > > outdir = './1' > > pseudo_dir = '../PP/atompaw' > > prefix = 'Test2' > > disk_io = 'none' > > verbosity = 'default' > > etot_conv_thr = 0.0001 > > forc_conv_thr = 0.001 > > nstep = 400 > > tstress = .true. > > tprnfor = .true. > > / > > &SYSTEM > > ibrav = 0 > > nat = 0 > > ntyp = 0 > > ecutwfc = 50 > > ecutrho = 400 > > london = .true. > > london_s6 = 0.75 > > / > > &ELECTRONS > > electron_maxstep = 200 > > conv_thr = 1.0D-7 > > diago_thr_init = 1e-4 > > startingpot = 'atomic' > > startingwfc = 'atomic' > > mixing_mode = 'plain' > > mixing_beta = 0.5 > > mixing_ndim = 8 > > diagonalization = 'david' > > / > > &IONS > > ion_dynamics = 'bfgs' > > / > > ATOMIC_SPECIES > > > > ATOMIC_POSITIONS crystal > > > > K_POINTS automatic > > 1 1 1 0 0 0 > > > > CELL_PARAMETERS > > 28.896936006399102 0.000 0.000 > > 0.00228.896936006399102 0.000 > > 0.002 0.00028.896936006399102 > > > > Any idea ??? > > > > I have exactly the same problem with other files. > > > > Thank you very much for your help. > > > > Sincerely, > > > > Romain > > > > > -- > > Dr. Romain Moury > > Research Group of Prof. Dr. Ferdi Schüth > > Department of Heterogeneous Catalysis > > Max-Planck-Institut für Kohlenforschung > > Kaiser-Wilhelm-Platz 1 > > D-45470 Mülheim an der Ruhr > > phone: +49(0)208/306-2369 > > eMail: mo...@mpi-muelheim.mpg.de > > > > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] pw 5.1.1 odd results in vc-relax
Dear Giuseppe, thanks for the reply. Yes, I kept the xdm and removed lspinorb. I do have small problems of visualization because the web-based J-ICE visualizer is not working any more with 5.1.1, however, Yes, the 50/400 Ry converged to the final volume cell of 1686 (a.u.)^3 (experimental 1835). The xdm + pbesol seems to overestimate the attractive forces, while pbesol underestimate them. Cheers, Carlo 2015-02-09 15:18 GMT+01:00 Giuseppe Mattioli : > > > > nervi@chpc111 $ awk '/unit-cell/' ULOKUB_ecut50 > > Is it 50/400 Ry? Is it a converged value? > > > unit-cell volume =1685.3884 (a.u.)^3 > ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] pw 5.1.1 odd results in vc-relax
Dear Giuseppe, thank you very much for your suggestion. My main concerns were about different results obtained on the same system, which puzzles me. Apparently my pw.x code is okay, since also the old one (4.3.2) gave similar strange results. I used pbesol, therefore I guess (if I understood correctly) I should use this input: vdw_corr = 'xdm' xdm_a1 = 0. xdm_a2 = 4.1503 (from here: http://gatsby.ucmerced.edu/wiki/XDM) I tried to add the spinorbit correction: noncolin = .true. lspinorb = .true. but the code exit with Error in routine energy_xdm (1): nspin > 2 not implemented So I run without noncolin and lspinorb. I got a much smaller cell: nervi@chpc111 $ awk '/unit-cell/' ULOKUB_ecut50_opt.out unit-cell volume =1834.6689 (a.u.)^3 new unit-cell volume = 1910.72951 a.u.^3 ( 283.14090 Ang^3 ) new unit-cell volume = 1867.80516 a.u.^3 ( 276.78016 Ang^3 ) new unit-cell volume = 1912.87224 a.u.^3 ( 283.45842 Ang^3 ) new unit-cell volume = 1905.31656 a.u.^3 ( 282.33878 Ang^3 ) new unit-cell volume = 1886.36736 a.u.^3 ( 279.53080 Ang^3 ) new unit-cell volume = 1880.73733 a.u.^3 ( 278.69651 Ang^3 ) new unit-cell volume = 1870.99986 a.u.^3 ( 277.25357 Ang^3 ) new unit-cell volume = 1854.95166 a.u.^3 ( 274.87547 Ang^3 ) new unit-cell volume = 1829.04915 a.u.^3 ( 271.03712 Ang^3 ) new unit-cell volume = 1787.37430 a.u.^3 ( 264.86154 Ang^3 ) new unit-cell volume = 1774.46108 a.u.^3 ( 262.94800 Ang^3 ) new unit-cell volume = 1757.48022 a.u.^3 ( 260.43170 Ang^3 ) new unit-cell volume = 1753.91380 a.u.^3 ( 259.90321 Ang^3 ) new unit-cell volume = 1749.78707 a.u.^3 ( 259.29169 Ang^3 ) new unit-cell volume = 1743.65835 a.u.^3 ( 258.38351 Ang^3 ) new unit-cell volume = 1737.62588 a.u.^3 ( 257.48959 Ang^3 ) new unit-cell volume = 1731.23334 a.u.^3 ( 256.54231 Ang^3 ) new unit-cell volume = 1721.50143 a.u.^3 ( 255.10019 Ang^3 ) new unit-cell volume = 1707.28097 a.u.^3 ( 252.99293 Ang^3 ) new unit-cell volume = 1687.33224 a.u.^3 ( 250.03684 Ang^3 ) new unit-cell volume = 1686.78734 a.u.^3 ( 249.95609 Ang^3 ) new unit-cell volume = 1689.29156 a.u.^3 ( 250.32718 Ang^3 ) new unit-cell volume = 1687.08639 a.u.^3 ( 250.00041 Ang^3 ) new unit-cell volume = 1685.38842 a.u.^3 ( 249.74879 Ang^3 ) new unit-cell volume = 1685.38842 a.u.^3 ( 249.74879 Ang^3 ) unit-cell volume =1685.3884 (a.u.)^3 I'm still playing trying to understand the system and its realibility. Thanks, Carlo 2015-02-09 10:59 GMT+01:00 Giuseppe Mattioli : > > Dear Carlo > Your *very exotic* crystal structure is going to be challenging for every > code for several reasons. I suppose that, for example, you will never obtain > decent lattice parameters without using a (good) contribution to > dispersion forces. As you are using PAW PPs, I suggest to try the new XDM > method. I > can report excellent results for (metal-)organic crystals. You must supply > to the code the appropriate parameters, which in the case of PBE are > > vdw_corr='xdm', > xdm_a1=0.3275, xdm_a2=2.7673, > > It should probably not affect the structural properties very much, but you > may try to check if the Iodine atoms benefit of some spin-orbit coupling > treatment. > > The cell is small, and even if your bands should be not far from flat, you > may experience some strong interaction between localized states at gamma. > You can try to use a shifted k-points mesh (e.g., 2, 2, 2, 1, 1, 1) in scf > calculations, and to reintroduce gamma in nscf ones if you need a DOS plot. > > HTH > Giuseppe > > On Sunday, February 08, 2015 12:10:25 AM Carlo Nervi wrote: > > Good point. I tried also with Castep 7.0 (same pbesol but onthe fly > pseudo) > > and it cannot reach convergence. Just increase and eat RAM without > reaching > > any convergence. > > Anyway, I run also the same two inputs also on the 4.3.2 version and the > > results are almost the same. Below I report them for comparison. > > I'm confused > > > > pw.x 5.1.1: ecut60_kp222_opt.out > > pw.x 4.3.2: ecut60_kp222_opt_pw432.out > > [ 1] unit-cell volume =1834.6689 (a.u.)^3 > >[ 1]=1834.6689 (a.u.)^3 > > [ 2] new unit-cell volume = 1916.92975 a.u.^3 ( 284.05968 Ang^3 ) [ > > 2] 1916.89038 a.u.^3 ( 284.05384 Ang^3 ) > > [ 3] new unit-cell volume = 1870.67155 a.u.^3 ( 277.20492 Ang^3 ) [ > > 3] 1870.65414 a.u.^3 ( 277.20234 Ang^3 ) > > [ 4] new unit-cell volume = 1921.05067 a.u.^3 ( 284.67033 Ang^3 ) [ > > 4] 1921.02565 a.u.^3 ( 284.3 Ang^3 )
Re: [Pw_forum] pw 5.1.1 odd results in vc-relax
) [19] new unit-cell volume = 1922.08253 a.u.^3 ( 284.82324 Ang^3 ) [19] 1920.03850 a.u.^3 ( 284.52035 Ang^3 ) [20] new unit-cell volume = 1918.60656 a.u.^3 ( 284.30816 Ang^3 ) [20] 1918.67621 a.u.^3 ( 284.31848 Ang^3 ) [21] new unit-cell volume = 1917.76107 a.u.^3 ( 284.18287 Ang^3 ) [21] 1917.87993 a.u.^3 ( 284.20048 Ang^3 ) [22] new unit-cell volume = 1917.99827 a.u.^3 ( 284.21802 Ang^3 ) [22] 1918.22644 a.u.^3 ( 284.25183 Ang^3 ) [23] new unit-cell volume = 1917.99827 a.u.^3 ( 284.21802 Ang^3 ) [23] 1918.22644 a.u.^3 ( 284.25183 Ang^3 ) [24] unit-cell volume =1917.9983 (a.u.)^3 [24] =1918.2264 (a.u.)^3 Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] pw 5.1.1 odd results in vc-relax
4.006700 N.pbesol-n-kjpaw_psl.0.1.UPF H1.007900 H.pbesol-kjpaw_psl.0.1.UPF ATOMIC_POSITIONS crystal I0.802555612 0.789940044 0.181369587 I0.197444388 0.210059956 0.818630413 C0.011037115 0.891477143 0.013032230 C0.988962885 0.108522857 0.986967770 F0.224842023 0.894645558 0.102744047 F0.775157977 0.105354442 0.897255953 F0.955951513 0.746622429 0.833467347 F0.044048487 0.253377571 0.166532653 N0.579184422 0.607816958 0.386818025 N0.420815578 0.392183042 0.613181975 C0.437800052 0.679699080 0.482871193 C0.562199948 0.320300920 0.517128807 H0.384497576 0.825751937 0.467937213 H0.615502424 0.174248063 0.532062787 C0.641400814 0.428208476 0.403043018 C0.358599186 0.571791524 0.596956982 H0.756695663 0.368817126 0.323193248 H0.243304337 0.631182874 0.676806752 K_POINTS automatic 2 2 2 0 0 0 CELL_PARAMETERS alat 1.017893850 -0.023662741 0.029149749 -0.096476858 1.067006240 0.017441278 -0.327880881 -0.358791924 1.079507559 I got the following results: unit-cell volume =1960.1695 (a.u.)^3 new unit-cell volume = 1959.31994 a.u.^3 ( 290.34126 Ang^3 ) new unit-cell volume = 1958.53703 a.u.^3 ( 290.22524 Ang^3 ) new unit-cell volume = 1957.26194 a.u.^3 ( 290.03629 Ang^3 ) new unit-cell volume = 1955.57715 a.u.^3 ( 289.78663 Ang^3 ) new unit-cell volume = 1953.25606 a.u.^3 ( 289.44268 Ang^3 ) new unit-cell volume = 1949.69460 a.u.^3 ( 288.91493 Ang^3 ) new unit-cell volume = 1947.54165 a.u.^3 ( 288.59589 Ang^3 ) new unit-cell volume = 1944.65689 a.u.^3 ( 288.16842 Ang^3 ) new unit-cell volume = 1940.96635 a.u.^3 ( 287.62153 Ang^3 ) new unit-cell volume = 1939.90715 a.u.^3 ( 287.46458 Ang^3 ) new unit-cell volume = 1939.26844 a.u.^3 ( 287.36993 Ang^3 ) new unit-cell volume = 1937.86399 a.u.^3 ( 287.16181 Ang^3 ) new unit-cell volume = 1936.35258 a.u.^3 ( 286.93784 Ang^3 ) new unit-cell volume = 1934.47381 a.u.^3 ( 286.65944 Ang^3 ) new unit-cell volume = 1932.63674 a.u.^3 ( 286.38721 Ang^3 ) new unit-cell volume = 1930.23334 a.u.^3 ( 286.03107 Ang^3 ) new unit-cell volume = 1926.68480 a.u.^3 ( 285.50523 Ang^3 ) new unit-cell volume = 1922.08253 a.u.^3 ( 284.82324 Ang^3 ) new unit-cell volume = 1918.60656 a.u.^3 ( 284.30816 Ang^3 ) new unit-cell volume = 1917.76107 a.u.^3 ( 284.18287 Ang^3 ) new unit-cell volume = 1917.99827 a.u.^3 ( 284.21802 Ang^3 ) new unit-cell volume = 1917.99827 a.u.^3 ( 284.21802 Ang^3 ) unit-cell volume =1917.9983 (a.u.)^3 -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] Input for Quantum Espresso
onstructing the input file starting from a pdb file would probably not >> be too tedious (one might lose some symmetries, however, if present as the >> pdb uses few digits in the atomic positions, in Angstroms if I am not >> wrong). I do not fully understand your question 2), do you mean that the >> structure that you built from the .cif file is a slab structure? Or that >> the .cif contains a bulk structure and you want to create a slab starting >> from that input? Well, you have first to determine the direction along the >> surface of the slab and its normal (orthogonal to the surface), and then >> add 10 Angstroms or more in that direction (and naturally adjust the >> coordinates if they are in crystal coordiates). >> >> Hopefully this helps you at least a little bit forward. >> >> Greetings from Rome, >> >>apsi >> >> -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*= >> -=*=-=*=-=*=-=*=- >> Ari Paavo Seitsonen / ari.p.seitso...@iki.fi / http://www.iki.fi/~apsi/ >> Ecole Normale Supérieure (ENS), Département de Chimie, Paris >> Mobile (F) : +33 789 37 24 25(CH) : +41 79 71 90 935 >> >> >> >> On Sun, 18 Jan 2015, Mohan maruthi sena wrote: >> >> Hi all, >>> I recently started using espresso. I want to optimize a >>> crystal structure. By looking at the previous posts , I tried to >>> use cif2qe.sh script to convert from .cif to .inp file. I tried with >>> various examples and it was working. I could not prepare inp >>> file for the attached cif file. I have visualized the cif file in vesta, >>> gaussian and then saved those .cif file. It prepared input >>> files but the atoms are too close. >>> >>> 1) Can some please let me how to give prepare input file for espresso >>> 5.1.1, using cif or pdb file? >>> >>> 2) If want to create a slab structure and maintain 10 angstorms distance >>> between the slabs? How can i do this in espresso? >>> >>> 3) I have used the following server (http://j-ice.sourceforge.net/ >>> ondemand/index.html) but could not get the input file. >>> >>> Please find the attached cif and espresso input files. >>> >>> >> ___ >> Pw_forum mailing list >> Pw_forum@pwscf.org >> http://pwscf.org/mailman/listinfo/pw_forum >> > > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] about C2/c point group symmetry of a monoclinic lattice.
Dear All,
I tried to modify the source code of cif2qe.sh, also cleaning the code a
little bit...
At my first attempt apparently the code now is working better and print the
correct number of atoms.
I would greatly appreciate if anyone will test it with some cif files.
Merry Chrstmas and Happy New Year!
Carlo
#!/bin/bash
#
# CIF to Quantum Espresso format converter
# Version 1.1 Date: 23-Dec-2014 Bug Fix
# Version 1.0 Date: 15-Mar-2014 First Full conversion
# Version 0.5 Date: 02-Oct-2013
# Version 0.4 Date: 12 Jun 2013
# Version 0.3 Date: 15 Nov 2012
#
# Copyright (C) 2012 Carlo Nervi
# This file is distributed under the terms of the
# GNU General Public License. See the file `License'
# in the root directory of the present distribution,
# or http://www.gnu.org/copyleft/gpl.txt .
#
# Use dos2unix to strip carriage returns at the end of the .cif files
#
# symmetry x,-y+1/2,z+y without ''
#
version="1.0"
USAGE="cif2qe.sh Version ${version}\nUsage: cif2qe.sh [-i] File\n( -i
uses the ibrav of QE. Do not add .cif extension!) - Requires File.cif\n"
if [ $# == 0 -o $# -gt 2 ]; then
printf "$USAGE"
exit
fi
do_ibrav=0
if [ $# == 2 ]; then
if [ $1 == "-i" ]; then
do_ibrav=1
shift
else
printf "$USAGE"
exit
fi
fi
if [ ! -f $1.cif ]; then
echo "Error. Cannot find file $1.cif"
exit
fi
awk -v FILE="$1" -v VERSION="$version" -v do_IBRAV=$do_ibrav '
BEGIN {
bohr = 0.52917721092
nfield=split("H He Li Be B C N O F Ne Na Mg Al Si P S Cl Ar K Ca Sc Ti V
Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr " \
"Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe Cs Ba La
Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb " \
"Lu Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn Fr Ra Ac Th
Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr Rf " \
"Db Sg Bh Hs Mt", AtomSymb, " ")
split("1.0079 4.0026 6.941 9.0122 10.811 12.0107 14.0067 15.9994 18.9984
20.1797 22.9897 24.305 26.9815 28.0855 30.9738 32.065 35.453 " \
"39.948 39.0983 40.078 44.9559 47.867 50.9415 51.9961 54.938 55.845
58.9332 58.6934 63.546 65.39 69.723 72.64 74.9216 78.96 79.904 " \
"83.8 85.4678 87.62 88.9059 91.224 92.9064 95.94 98 101.07 102.906
106.42 107.868 112.411 114.818 118.71 121.76 127.6 126.904 131.293 " \
"132.905 137.327 138.905 140.116 140.908 144.24 145 150.36 151.964
157.25 158.925 162.5 164.93 167.259 168.934 173.04 174.967 178.49 " \
"180.948 183.84 186.207 190.23 192.217 195.078 196.966 200.59
204.383 207.2 208.98 209 210 222 223 226 227 232.038 231.036 238.029 " \
"237 244 243 247 247 251 252 257 258 259 262 261 262 266 264 277
268", AtomMass, " ")
for (i=1; i<=nfield; i++) Atoms[AtomSymb[i]] = AtomMass[i]
#
# nKey = number of recognized Keywords
# KeyW[0][1..nKey] = recognized Keywords
# KeyW[1][1..nKey] = first synonym
# ...
#
nsynon=2
nKey=13
KeyW[0][1] ="_cell_length_a"; KeyW[1][1]=""
KeyW[0][2] ="_cell_length_b"; KeyW[1][2]=""
KeyW[0][3] ="_cell_length_c"; KeyW[1][3]=""
KeyW[0][4] ="_cell_angle_alpha";KeyW[1][4]=""
KeyW[0][5] ="_cell_angle_beta"; KeyW[1][5]=""
KeyW[0][6] ="_cell_angle_gamma";KeyW[1][6]=""
KeyW[0][7] ="_atom_site_type_symbol"; KeyW[1][7]=""
KeyW[0][8] ="_atom_site_fract_x"; KeyW[1][8]=""
KeyW[0][9] ="_atom_site_fract_y"; KeyW[1][9]=""
KeyW[0][10]="_atom_site_fract_z"; KeyW[1][10]=""
KeyW[0][11]="_symmetry_equiv_pos_as_xyz";
KeyW[1][11]="_space_group_symop_operation_xyz"
KeyW[0][12]="_symmetry_cell_setting";
KeyW[1][12]="_space_group_crystal_system"
KeyW[0][13]="_symmetry_Int_Tables_number";
KeyW[1][13]="_space_group_IT_number"
# for (i=1; i<=nKey; i++) {
# for (j=0; j= 1.) val-=1.
return val
}
function abs(numero) {
if (numero < 0) numero=-numero;
return numero
}
function Find_Lattice(a,b,c,alpha,beta,gamma) {
#
# find the bravais lattice name from lattice parameters
#
thr = 1.e-4
reticolo=""
if ( (abs(alpha-90.0) 0) Var[ivar++]=KeyW[0][tmp]; else Var[ivar++]=$1
Num_Var=ivar
} else {
ivar=0
loop_switch=0
if (tmp > 0) Var2[KeyW[0][tmp]]=$2
}
}
$1 ~ /^[^_]/ {
ivar=0
if (loop_switch==1 || loop_switch==2) {
loop_switch=2
for (i=0; i 0 ) tmpspacegroup=Int_Tables[tmptablenumber]
ibrav=Find_ibrav(tmpspacegroup, reticolo)
KP_x = int(1./(a*separation)+0.5)
KP_y = int(1./(b*separation)+0.5)
KP_z = int(1./(c*separation)+0.5)
ntyp=0
for (i=0; i___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum
Re: [Pw_forum] about C2/c point group symmetry of a monoclinic lattice.
Very probably cif2qe.sh is not working properly. It generates all the atoms by applying the symmetry found and later try to remove the equivalent atoms. Probably some atoms are not properly removed. The software is give "as it is". Carlo 2014-12-17 16:11 GMT+01:00 Paolo Giannozzi : > > On Wed, 2014-12-17 at 15:07 +0100, Carlo Nervi wrote: > > Hi, > > the cif2qe.sh -i command gave to me: > > there is something I don't understand here. The cif file contains > a line "_cell_formula_units_Z 16" which means I think that there > are 16*(GaSe2Tl) = 64 atoms in the unit cell. This is consistent > with the listing of atoms and multiplicities: > > > Tl1 Tl1+ 8 f 0.4647(6) 0.3109(5) 0.1140(9) 1. 0 > > Tl2 Tl1+ 8 f 0.2844(4) 0.0623(5) 0.3864(8) 1. 0 > > Ga1 Ga3+ 8 f 0.100(1) 0.191(2) 0.162(2) 1. 0 > > Ga2 Ga3+ 8 f 0.145(1) 0.438(1) 0.339(2) 1. 0 > > Se1 Se2- 4 e 0. 0.054(2) 0.250 1. 0 > > Se2 Se2- 4 e 0. 0.574(1) 0.250 1. 0 > > Se3 Se2- 8 f 0.207(1) 0.062(1) 0.071(2) 1. 0 > > Se4 Se2- 8 f 0.262(1) 0.310(1) 0.252(2) 1. 0 > > Se5 Se2- 8 f 0.048(2) 0.312(1) 0.438(3) 1. 0 > > The converted data contains instead 72 atoms, that is, > 18 formula units > > Paolo > > > > > ! Generated by using cif2qe Version 1.0 - Date: Wed Dec 17 16:05:24 > > CET 2014 > > ! _symmetry_space_group_name_H-M = > > ! _symmetry_Int_Tables_number = 15 > > ! _symmetry_cell_setting = > > ! a=10.779 b=10.776 c=15.663 alpha=90 beta=99.993 gamma=90 > > ! Found by cif2qe: lattice = monoclinicSpace group = C2/c ibrav > > = 13 > > ! > > ! Symmetry found: > > ! 1 x, -y, z+1/2[x] [ -y] [ z+1/2] > > ! 2 -x, -y, -z[-x] [ -y] [ -z] > > ! 3 -x, y, -z+1/2[-x] [ y] [ -z+1/2] > > ! 4x, y, z[x] [ y] [ z] > > ! 5 x+1/2, -y+1/2, z+1/2[x+1/2] [ -y+1/2] [ z+1/2] > > ! 6 -x+1/2, -y+1/2, -z[-x+1/2] [ -y+1/2] [ -z] > > ! 7 -x+1/2, y+1/2, -z+1/2[-x+1/2] [ y+1/2] [ -z+1/2] > > ! 8x+1/2, y+1/2, z[x+1/2] [ y+1/2] [ z] > > &CONTROL > >title = 'xx' > > calculation = 'relax' > > restart_mode = 'from_scratch' > > outdir = './1' > > pseudo_dir = '../PP/atompaw' > > prefix = 'xx' > > disk_io = 'none' > >verbosity = 'default' > >etot_conv_thr = 0.0001 > >forc_conv_thr = 0.001 > >nstep = 400 > > tstress = .true. > > tprnfor = .true. > > / > > &SYSTEM > >ibrav = 13 > >celldm(1) = 20.369357896686800 > >celldm(2) = 0.999721681046479, celldm(3) = > > 1.453103256331756 > >celldm(4) = 0.000 > > nat = 72 > > ntyp = 3 > > ecutwfc = 50 > > ecutrho = 400 > > london = .true. > >london_s6 = 0.75 > > / > > &ELECTRONS > > electron_maxstep = 200 > > conv_thr = 1.0D-7 > > diago_thr_init = 1e-4 > > startingpot = 'atomic' > > startingwfc = 'atomic' > > mixing_mode = 'plain' > > mixing_beta = 0.5 > > mixing_ndim = 8 > > diagonalization = 'david' > > / > > &IONS > > ion_dynamics = 'bfgs' > > / > > > > > > ATOMIC_SPECIES > > Tl1+0.00 Tl1+.pbe-n-rrkjus_psl.0.1.UPF > > Ga3+0.00 Ga3+.pbe-n-rrkjus_psl.0.1.UPF > > Se2-0.00 Se2-.pbe-n-rrkjus_psl.0.1.UPF > > > > > > ATOMIC_POSITIONS crystal > > Tl1+ 0.4647000 0.6891000 0.614 > > Tl1+ 0.2844000 0.9377000 0.8864000 > > Ga3+ 0.100 0.809 0.662 > > Ga3+ 0.145 0.562 0.839 > > Se2- 0.000 0.946 0.750 > > Se2- 0.000
Re: [Pw_forum] about C2/c point group symmetry of a monoclinic lattice.
0. -0.01(2) 0. > Se3 Se2- 0.04(1) 0.05(1) 0.01(4) 0.01(1) -0.01(2) -0.02(1) > Se4 Se2- 0.012(6) 0.021(7) 0.06(4) 0.01(1) -0.018(9) -0.018(5) > Se5 Se2- 0.05(1) 0.06(1) 0.00(4) -0.053(9) -0.01(2) -0.02(1) > loop_ > _atom_site_label > _atom_site_type_symbol > _atom_site_symmetry_multiplicity > _atom_site_Wyckoff_symbol > _atom_site_fract_x > _atom_site_fract_y > _atom_site_fract_z > _atom_site_occupancy > _atom_site_attached_hydrogens > Tl1 Tl1+ 8 f 0.4647(6) 0.3109(5) 0.1140(9) 1. 0 > Tl2 Tl1+ 8 f 0.2844(4) 0.0623(5) 0.3864(8) 1. 0 > Ga1 Ga3+ 8 f 0.100(1) 0.191(2) 0.162(2) 1. 0 > Ga2 Ga3+ 8 f 0.145(1) 0.438(1) 0.339(2) 1. 0 > Se1 Se2- 4 e 0. 0.054(2) 0.250 1. 0 > Se2 Se2- 4 e 0. 0.574(1) 0.250 1. 0 > Se3 Se2- 8 f 0.207(1) 0.062(1) 0.071(2) 1. 0 > Se4 Se2- 8 f 0.262(1) 0.310(1) 0.252(2) 1. 0 > Se5 Se2- 8 f 0.048(2) 0.312(1) 0.438(3) 1. 0 > loop_ > _atom_type_symbol > _atom_type_oxidation_number > Ga3+ 3 > Se2- -2 > Tl1+ 1 > loop_ > _citation_id > _citation_year > _citation_page_first > _citation_page_last > primary 2007 663 666 > > *Are there something wrong in my input file or primitive cell?* > > With my best wishes, > > Mutlu. > -- > Dr. Mutlu COLAKOGULLARI > Trakya Universitesi > Fen Fakultesi > Fizik Bolumu > 22030 Merkez-EDİRNE > > ___ > Pw_forum mailing list > Pw_forum@pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.ne...@unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ ___ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
[Pw_forum] SATA vs SAS
I completely agree with Axel, expecially speaking about a middle-range machine (as mine, for example, a linux home-assembled server with a careful look at the costs). Carlo 2014-08-07 14:18 GMT+02:00 Axel Kohlmeyer : > On Thu, Aug 7, 2014 at 6:56 AM, siddheshwar chopra > wrote: > > Dear Sir, > > Oh I am sorry... I think I didn't mention correctly.. I said "write > times".. > > It is actually "number of write times". Ofcourse these are very fast..But > > their life is shorter than the normal HDDs... > > Yes budget is what is the final verdict.. :-) > > not only that. but also, how much time you save, i.e. how much value > you attribute to your own time. > if an SSD will wear out and become unusable, you can easily replace it > at the fraction of the cost of the entire machine, and then - provided > an SSD scratch does significantly improve your calculation run times - > you have a faster machine all the time. > > please also note that SSD performance varies a lot and also the amount > of guaranteed writes. > > axel. > > > > > > Regards, > > > > > > On Thu, Aug 7, 2014 at 4:18 PM, Carlo Nervi > wrote: > >> > >> I'm not an expert, and maybe my infos are old, but I think you are > wrong. > >> Certain SSD have degraded performances, however the best SSD have much > >> faster access I/O time than than the best HDD. at least for medium > >> workstation. Cost for Gb is of course much higher for SSD, but in this > case > >> I would go only for a working directory to be placed on a SSD partition > (128 > >> Gb could be enough). I do not know how is the situation for high-end > >> enterprise solutions or supercomputers, but in this case i suspect that > the > >> special HDDs are not connected via simple SATA or SAS, and in these > cases > >> very probably they are much faster (again at much hgher prices) due > several > >> optimized performances (independent memory cache, for example). > >> > >> In other words, it depend how much is your budget... > >> Cheers, > >> Carlo > >> > >> > >> > >> 2014-08-07 11:42 GMT+02:00 siddheshwar chopra : > >>> > >>> Dear Sir, > >>> I just want to participate in this good discussion. I read somewhere > that > >>> Solid State Hard Disk has LESSER write times than a normal HDD. Yes > >>> undoubtedly it is the fastest we have in the market with low power > >>> consumption and zero mechanical failure risk, but at the cost of > shorter > >>> lives and more money. > >>> Any comments welcome.. > >>> > >>> Regards, > >>> > >>> > >>> On Thu, Aug 7, 2014 at 2:16 PM, Carlo Nervi > wrote: > >>>> > >>>> My two cents: even though there are differences between SAS and SATA, > >>>> they are not so huge. > >>>> From this point of view I guess that a fast Solid State Hard Disk > would > >>>> make the difference on the I/O. > >>>> > >>>> > >>>> > >>>> 2014-08-07 10:06 GMT+02:00 Paolo Giannozzi >: > >>>> > >>>>> On Wed, 2014-08-06 at 18:15 -0400, Eduardo Menendez wrote: > >>>>> > >>>>> > Is it critical to use SAS disks (compared to SATA) for good > >>>>> > performance of Quantum ESPRESSO? > >>>>> > >>>>> I guess it depends upon the amount of I/O you plan to do > >>>>> > >>>>> Paolo > >>>>> > >>>>> -- > >>>>> Paolo Giannozzi, Dept. Chemistry&Physics&Environment, > >>>>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy > >>>>> Phone +39-0432-558216, fax +39-0432-558222 > >>>>> > >>>>> ___ > >>>>> Pw_forum mailing list > >>>>> Pw_forum at pwscf.org > >>>>> http://pwscf.org/mailman/listinfo/pw_forum > >>>> > >>>> > >>>> > >>>> > >>>> -- > >>>> > >>>> > >>>> Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 > >>>> Fax: +39 0116707855 - Dipartimento di Chimica, via > >>>> P. Giuria 7, 10125 Torino, Italy.http://
[Pw_forum] SATA vs SAS
I'm not an expert, and maybe my infos are old, but I think you are wrong. Certain SSD have degraded performances, however the best SSD have much faster access I/O time than than the best HDD. at least for medium workstation. Cost for Gb is of course much higher for SSD, but in this case I would go only for a working directory to be placed on a SSD partition (128 Gb could be enough). I do not know how is the situation for high-end enterprise solutions or supercomputers, but in this case i suspect that the special HDDs are not connected via simple SATA or SAS, and in these cases very probably they are much faster (again at much hgher prices) due several optimized performances (independent memory cache, for example). In other words, it depend how much is your budget... Cheers, Carlo 2014-08-07 11:42 GMT+02:00 siddheshwar chopra : > Dear Sir, > I just want to participate in this good discussion. I read somewhere that > Solid State Hard Disk has LESSER write times than a normal HDD. Yes > undoubtedly it is the fastest we have in the market with low power > consumption and zero mechanical failure risk, but at the cost of shorter > lives and more money. > Any comments welcome.. > > Regards, > > > On Thu, Aug 7, 2014 at 2:16 PM, Carlo Nervi wrote: > >> My two cents: even though there are differences between SAS and SATA, >> they are not so huge. >> From this point of view I guess that a fast Solid State Hard Disk would >> make the difference on the I/O. >> >> >> >> 2014-08-07 10:06 GMT+02:00 Paolo Giannozzi : >> >> On Wed, 2014-08-06 at 18:15 -0400, Eduardo Menendez wrote: >>> >>> > Is it critical to use SAS disks (compared to SATA) for good >>> > performance of Quantum ESPRESSO? >>> >>> I guess it depends upon the amount of I/O you plan to do >>> >>> Paolo >>> >>> -- >>> Paolo Giannozzi, Dept. Chemistry&Physics&Environment, >>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy >>> Phone +39-0432-558216, fax +39-0432-558222 >>> >>> _______ >>> Pw_forum mailing list >>> Pw_forum at pwscf.org >>> http://pwscf.org/mailman/listinfo/pw_forum >>> >> >> >> >> -- >> >> >> Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 >> Fax: +39 0116707855 - Dipartimento di Chimica, via >> P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ >> >> >> ___ >> Pw_forum mailing list >> Pw_forum at pwscf.org >> http://pwscf.org/mailman/listinfo/pw_forum >> > > > > -- > *Dr. Siddheshwar chopra,* > > > *M.Sc., Ph.D (Physics)Assistant Professor (Physics),* > *Amity University, Noida, India.* > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ -- next part -- An HTML attachment was scrubbed... URL: http://pwscf.org/pipermail/pw_forum/attachments/20140807/0fc2b1fd/attachment.html
[Pw_forum] SATA vs SAS
My two cents: even though there are differences between SAS and SATA, they are not so huge. >From this point of view I guess that a fast Solid State Hard Disk would make the difference on the I/O. 2014-08-07 10:06 GMT+02:00 Paolo Giannozzi : > On Wed, 2014-08-06 at 18:15 -0400, Eduardo Menendez wrote: > > > Is it critical to use SAS disks (compared to SATA) for good > > performance of Quantum ESPRESSO? > > I guess it depends upon the amount of I/O you plan to do > > Paolo > > -- > Paolo Giannozzi, Dept. Chemistry&Physics&Environment, > Univ. Udine, via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/ -- next part -- An HTML attachment was scrubbed... URL: http://pwscf.org/pipermail/pw_forum/attachments/20140807/dcbf807a/attachment.html
[Pw_forum] xcrysden visualization
Regarding this question, is there any differences in calculating structures having the ibrav=-12 or ibrav=0 (and setting the corresponding CELL_PARAMETERS)? In other words the calculations should be equivalent. Is it? Also, do the calculations in both cases make use of symmetry? In this case one can use the cif2qe.sh utility to generate the ibrav=0 input file starting from the corresponding cif file. As far as I know xcrysgen can handle ibrav=0 Correct? thanks, Carlo Il 17/03/2014 13.16, Tone Kokalj ha scritto: > On Mon, 2014-03-17 at 17:23 +0530, mayank gupta wrote: >> Dear users >> >> I am not able to visualize the crystal structure with ibrav=-12 flag. > > This is correct; ibrav=-12 is not yet supported by xcrysden. > >> Please suggest the possible alternatives. > > Here is the trick how you can still see the pertinent structure with > xcrysden: > > 1. generate your input file > > 2. run pw.x on this input file for, say, only a few seconds, so that the > initial coordinates are written to output file, i.e.: > > pw.x < file.in > file.out > (press Control-C after a few seconds) > > 3. visualize the "Initial Coordinates" from this input file, i.e.: > > xcrysden --pwo file.out > > Regards, > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] phonons
Dear QE users, first of all I apologize if my is a dumb question, and I suspect it's impossible... Anyway, here it is: as chemist I'm looking for a selected vibrational frequency associated to a stretching of a particular bond in a crystal. Since among all vibrations in the crystal I'm interested only in one (the bond stretching, and eventually anything else associated with the motion of a selected atom), is there any way in QE to compute only this mode without the need to compute everything? Second question: what is the most immediate way (program or tool) to assign and find from the output the correct frequency of my vibration? Do I have to perform all the complete DOS phonons? Thank you, Carlo -- ---- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] Error in converting CIF to QE by "cif2qe.sh"
'atomic' startingwfc = 'atomic' mixing_mode = 'plain' mixing_beta = 0.5 mixing_ndim = 8 diagonalization = 'david' / &IONS ion_dynamics = 'bfgs' / ATOMIC_SPECIES Ti 47.867000 Ti.pbe-van_ak.UPF O 15.999400 O.pbe-van_ak.UPF ATOMIC_POSITIONS crystal Ti 0.000 0.000 0.000 O 0.3053000 0.3053000 0.000 Ti 0.500 0.500 0.500 O 0.8053000 0.1947000 0.500 K_POINTS automatic 5 5 8 0 0 0 CELL_PARAMETERS 8.680891628420765 0.000 0.000 0.001 8.680891628420765 0.000 0.000 0.000 5.590036668211679 HTH, Carlo -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] Error in converting CIF to QE by "cif2qe.sh"
gt; 0.1947000 > > 0.500 > > > > K_POINTS automatic > > 5 5 8 0 0 0 > > > > > > CELL_PARAMETERS > > 8.680891628420765 > > 0.000 > > 0.000 > > 0.001 > > 8.680891628420765 > > 0.000 > > 0.000 > > 0.000 > > 5.590036668211679 > > > > > > > > > > -- Dr. Lorenzo Paulatto > > IdR @ IMPMC -- CNRS & Universit? Paris 6 > > +33 (0)1 44 275 084 / skype: paulatz > > http://www-int.impmc.upmc.fr/~paulatto/ > > 23-24/4?16 Bo?te courrier 115, 4 place Jussieu > 75252 Paris > > C?dex 05 > > > > > > -Inline Attachment Follows- > > > > ___ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://pwscf.org/mailman/listinfo/pw_forum > > > > ___ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://pwscf.org/mailman/listinfo/pw_forum > > -- > Paolo Giannozzi, Dept. > Chemistry&Physics&Environment, > Univ. Udine, via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] CIF to PWSCF
Another way is to use Moldraw, written by a collegue of mine. It can be downloaded for free from: http://www.moldraw.unito.it/ Moldraw can read cif file and convert, for example, in CRYSTAL format. Remember to "Make a cell" before to export. From this format you should be able with a text editor and a minimum of efforts to write a Quantum Espresso compatible input file. HTH, Carlo Il 25/10/2012 15.02, Yue-Wen Fang ha scritto: > one way is to write a script to convert them! > i usually use castep in Materials Studio to convert the CIF files. > > Best! > > > 2012/10/24 Ben Palmer <mailto:benpalmer1983 at gmail.com>> > > Hi, > > I'm new to pwscf and I'm trying to find the relaxed co-ordinates for an > Iron crystal. I've been trying to set up the initial crystal from a cif > file. Is there a way to convert from cif to the pwscf format? > > Thanks > > Ben > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org> > http://pwscf.org/mailman/listinfo/pw_forum > > > > > -- > > Yue-Wen Fang > East China Normal Univ, China > > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] about ecutwfc convergence for total energy
Hi, as far as i know it seems a problem related with bands, so low number of k-points (for insulators). I would suggest in any cases to try to use a couple of k-points. Another important point: are you sure your system is not metallic? In this case you could try smearing. HTH, Carlo Il 08/10/2012 17.05, James Mao ha scritto: > Hi all, > > I am a little bit confused about ecutwfc convergence. I am working on > calculating > NMR of a crystal structure. Before that I need test SCF energy convergence. > However I noticed the total energy is not convergent with respect to > ecutwfc. > > It reached minimum at ecutwfc=90 then began oscillating up and down. > > Ecutwfc Total Energy > > 70 -3249.733804 > > 80 -3250.674250 > > 90 -3250.827326 > > 100 -3250.661301 > > 120 -3250.346091 > > 160 -3250.582989 > > 200 -3250.395401 > > I think higher ecutwfc means more basis functions, therefore should > always result > in a lower total energy. If I am wrong, should I keep increasing ecutwfc? > > The PP I used is (pbe-tm-new-gipaw-dc.UPF), the atoms include C, O, N, > F, S, H, P. > > Since PP is NC, I left ecutrho default. And since the cell is pretty big > (a~10, b~16, C~15), > I only calculated at gamma point. > > Any hint is helpful! Thanks. > > Best, > > James Mao > > -- > > Postdoc, Department of Chemistry, University of Pittsburgh > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum > -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] how to calculate and plot HOMO or LUMO isosurface
Il 25/07/2012 12.41, Franklin Zhan ha scritto: > ...However, O2 molecule is not magnetic material, thus I do not > know why we need to take the spin polarization into consideration. You have to review some chemistry... Try to have a look here: http://www.youtube.com/watch?v=Isd9IEnR4bw Carlo Nervi -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] GIPAW and nosymm
Dear PW community, using GIPAW, up to now, I always added 'nosym= .true.' to preliminar scf calculations with Quantum espresso 4.3.2 Is the limitation still necessary for GIPAW calculations? I'm doing calculations on a rather large system and symmetry could help a lot. BTW, the new version QE 5.0 includes some improvements in this respect? Thanks, Carlo -- ---- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] [Q-e-developers] Change in the number of qe-forge users
Hello, I do not have great experience in the field, however a lot of very good free public code is developed on http://sourceforge.net/ I have no idea is this is acceptable for the development, but I would give to it a try. Cheers, Carlo Il 18/05/2012 17.33, Filippo Spiga ha scritto: > Dear Layla, > > why moving from fusionforge to gforge? What about Trac > (http://trac.edgewall.org/)? I mean, the point is? what exactly gforge > has that we do need? And what if we will not use some of those advanced > and fancy features that gforge will provide? > > You mention a "pro-year price" that depends by the number of users. Just > for curiosity, can you tell us with the current number of users how much > is this license? > > Filippo > > -- > Mr. Filippo SPIGA, HPC and GPU Technologist > website: http://filippospiga.me ~ skype: filippo.spiga > > ?Nobody will drive us out of Cantor's paradise.? ~ David Hilbert > > > > On May 17, 2012, at 11:58 PM, Layla Martin-Samos wrote: >> Dear all, we are thinking in moving qe-forge from fusionforge to >> gforge. Unfortunately it seems that gforge ask for a commercial >> License which pro-year prince will depends on the number of users. As >> all of you know, nor Quantum ESPRESSO nor qe-forge have any direct >> financial support, so we can not afford big expenses. In order to >> reduce potential costs of the gforge commercial License, our system >> administrator has eliminated all the qe-forge users that are not bound >> to any qe-forge project. Depending on the License price then or we >> move to gforge or we stay with fusionforge. In any case we will let >> you know in order to rationalize the qe-forge users. >> >> Sorry for the inconvenience, best regards > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] Problems in running check-pw.x.j
Hi Paolo, hello community... your experience could be of great help: I also tried to use check-pw.x.j on QE 4.3.2. First I did not use mpirun: it passed with 1-2 not significant numerical discrepancy. By using "mpirun -n 2" more discrepancies appears (sometimes very different number of iterations, i.e. 86 instead of 57 in lda+U_gamma test). By using "mpirun -n 48" it crash at dipole calculation. Too fast for 48 cores? Up to now I used pw.x "apparently" without problems, but...should I be worried? QE was compiled with gfortran 4.6.1 and acml 5.0.0 (libacml.a and libacml_mv.a) and with amdlibm 3.0.1 Thanks, Carlo Il 26/03/2012 18.42, Paolo Giannozzi ha scritto: > On Mon, 2012-03-26 at 18:18 +0530, Abhinav 122 wrote: > >> I received the following error during check-pw.x.j. >> Please suggest the required steps to remove the following errors. > > most likely, your compiler is buggy. > > P. -- -------- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] HOW TO CONFIGURE ESPRESSO IN HPC CLOSTER
Dear Onyeaju, first of all I think you better have a look on the installation manual of Quantum Espresso (if you need help on Abinit, better ask to another community..) and perhaps after some trial-and-errors report a more detailed and specific question. For example machine on which you are going to install, OS, fortran compiler, mpi version, problems and oputput error after attempt in configuration, etc... I guess nobody can help you starting from scratch... I have another suggestion (sorry, I can't refrain myself, it's quite hilarious and I hope you accept the joke): look on google translator for the word "closter"... than we can ask the help of monks.. :-) Cheers, Carlo Il 28/02/2012 11.15, michael onyeaju ha scritto: > Dear QE users, > I have been having troubles in trying to configure Espresso and Abinit > in HPC closters, can any one assist me > Onyeaju, M.C > Department of Physics, > Faculty of Science, > University of Port Harcourt > Phone:+2348068118468 > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- ---- Prof. Carlo Nervi carlo.nervi at unito.it Tel:+39 0116707507/8 Fax: +39 0116707855 - Dipartimento di Chimica, via P. Giuria 7, 10125 Torino, Italy.http://lem.ch.unito.it/
[Pw_forum] GIPAW: error in output
Thanks Davide, this is really great! Please note that a similar error (IMHO) is also present in the epr_routines.f90 Many many thanks, Carlo > > On 01/-10/-28163 08:59 PM, Carlo Nervi wrote: >> The gipaw calculations proceed, apparently ends >> normally, >> but the output contains several NotaNumber (NaN). Anuone >> have an idea qhat's wrong or could please give some >> hints >> how to solve the problem? >> Thank you, >>Carlo > Problem found! Change line 356 of > qe-gipaw/src/nmr_routines.f90 from: > call greenfunction(ik,aux1, aux2 , 0.d0) > to: > call greenfunction(ik,aux1, aux2 , > (/0.d0,0.d0,0.d0/)) > > I've found that the Intel compiler is anyway filling the q > array with > zeros, but other compilers don't. As a result q(2) and > q(3) contained > random values from the RAM. > > I've tested with gfortran and I hope that now it will work > with > every compiler. > > Best, > Davide > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > >
[Pw_forum] Swapping
> Dear Quantum Espresso Users > > I think the jobs run slow when they are swapping memory to > disk. > Is there anything I can do to distribute the memory across > more nodes so > they don???t have to swap? > I want to reduce RAM usage in practice. How it is > possible? It is generally a very bad idea to swap RAM on the HD. Try to reduce the amount of RAM decreasing the size of the problem (i.e. reduce cell dimensions, use lower ecut, etc...) There is also another parameter to be considered, especially in parallel calculations: in namelist &CONTROL disk_io='minimal', or even 'none' see here: http://www.quantum-espresso.org/user_guide/node18.html HTH, Carlo
[Pw_forum] GIPAW: error in output
Okay, thanks again. addr2line -e /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x 0x44e576 /home/nervi/src/QE432/espresso-4.3.2/qe-gipaw-4.3.2/src/cgsolve_all.f90:225 It should be the line with dclamda (see below) - no idea how to furter help the developers ;-) Carlo #ifdef __PARA call mp_sum( a(1:lbnd), intra_pool_comm ) call mp_sum( c(1:lbnd), intra_pool_comm ) #endif lbnd=0 do ibnd = 1, nbnd if (conv (ibnd) .eq.0) then lbnd=lbnd+1 dclambda = CMPLX( - a(lbnd) / c(lbnd), 0.d0,kind=DP) ! !move to new position ! call zaxpy (ndmx*npol, dclambda, h(1,ibnd), 1, dpsi(1,ibnd), 1) ! !update to get the gradient ! !g=g+lam call zaxpy (ndmx*npol, dclambda, t(1,lbnd), 1, g(1,ibnd), 1) ! !save current (now old) h and rho for later use ! call zcopy (ndmx*npol, h(1,ibnd), 1, hold(1,ibnd), 1) rhoold (ibnd) = rho (ibnd) endif enddo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] GIPAW: error in output
Il 25/10/2011 18.22, Lorenzo Paulatto ha scritto: > You could try to compile using this option (all together, without spaces): > -ffpe-trap=invalid,zero,overflow,underflow,denormal > to force the code to crash at the first appearance of NaN, this could help > track down the source of the problem. Thank you Lorenzo. I did what you suggested and I got many errors like this: [chpc111:47070] Signal: Floating point exception (8) [chpc111:47070] Signal code: Floating point divide-by-zero (3) [chpc111:47070] Failing at address: 0x44e576 [chpc111:47092] [ 0] /lib64/libpthread.so.0(+0x10eb0) [0x2b972490deb0] [chpc111:47092] [ 1] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(cgsolve_all_+0xff6) [0x44e576] [chpc111:47092] [ 2] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(greenfunction_+0x1280) [0x44d1a0] [chpc111:47092] [ 3] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(paramagnetic_correction_aug_+0x1e30) [0x4718f0] [chpc111:47092] [ 4] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(suscept_crystal_+0x520f) [0x45c02f] [chpc111:47092] [ 5] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(main+0x1a7) [0x43d227] [chpc111:47092] [ 6] /lib64/libc.so.6(__libc_start_main+0xed) [0x2b9724b3db0d] [chpc111:47092] [ 7] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x() [0x43d2fd] [chpc111:47092] *** End of error message *** [chpc111:47105] *** Process received signal *** and [chpc111:47076] *** End of error message *** [chpc111:47102] [ 3] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(paramagnetic_correction_aug_+0x1e30) [0x4718f0] [chpc111:47102] [ 4] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(suscept_crystal_+0x520f) [0x45c02f] [chpc111:47102] [ 5] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x(main+0x1a7) [0x43d227] [chpc111:47102] [ 6] /lib64/libc.so.6(__libc_start_main+0xed) [0x2ac89299db0d] [chpc111:47102] [ 7] /home/nervi/src/QE432/espresso-4.3.2/GIPAW/src/gipaw.x() [0x43d2fd] [chpc111:47102] *** End of error message *** But if I run mpirun -n 1 I got the following: Fortran runtime warning: IEEE 'denormal number' exception not supported. At line 739 of file suscept_crystal.f90 (unit = 99, file = '/tmp/ceresoli/benzene.gipaw_recover') Fortran runtime error: I/O past end of record on unformatted file -- mpirun has exited due to process rank 0 with PID 47850 on node chpc111 exiting without calling "finalize". This may have caused other processes in the application to be terminated by signals sent by mpirun (as reported here). -- Any hints? Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] GIPAW: error in output
Hello, I tried to compile using the GOTO2 BLAS libs, but I got the same error in gipaw. So I tried to compile all with gfortran 4.4.5, but again I got the same NaN message, with again the following: Computing the magnetic susceptibility isolve=0ethr=0.1000E-13 k-point #1 of 1 pool # 1 ik 1 ibnd 16 linter: root not converged 0.651E+09 ik 1 ibnd 16 linter: root not converged 0.261E+30 ik 1 ibnd 16 linter: root not converged 0.362E+30 End of magnetic susceptibility calculation I'm getting frustrated... What is strange is that pw.x compiled with gfortran 4.6.1 is apparently doing very well. This is not the case of gipaw Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] GIPAW: error in output
Hello, I succesfully compiled QE 4.3.2 and the corresponding GIPAW module on a Opteron 6168 linux using gfortran 4.6, acml 5.0.0 and amdlibm 3.0.1. I did the benzene-scf.in job in the GIPAW/examples/benzene-USPP/ directory. The scf energy is the same as the reference output, which is in the same directory (apparently all is okay). I've compiled the gipaw module by the common way 'make gipaw' from the source QE dir. - I hope this was okay... The gipaw calculations proceed, apparently ends normally, but the output contains several NotaNumber (NaN). Anuone have an idea qhat's wrong or could please give some hints how to solve the problem? Thank you, Carlo ... negative rho (up, down): 0.103E-03 0.000E+00 init_paw_1: ntyp= 1 rc=1.4000 rs=0.9333 init_paw_1: ntyp= 1 rc=1.4000 rs=0.9333 init_paw_1: ntyp= 1 rc=1.4000 rs=0.9333 init_paw_1: ntyp= 1 rc=1.4000 rs=0.9333 init_gipaw_1: projectors nearly linearly dependent: ntyp = 1, l/n1/n2 = 0 2 1 0.99622328 init_gipaw_1: projectors nearly linearly dependent: ntyp = 1, l/n1/n2 = 1 2 1 0.99789339 init_paw_1: ntyp= 2 rc=0.8000 rs=0.5333 init_paw_1: ntyp= 2 rc=0.8000 rs=0.5333 init_gipaw_1: projectors nearly linearly dependent: ntyp = 2, l/n1/n2 = 0 2 1 0.99987400 GIPAW: 4.81s CPU10.71s WALL Computing the magnetic susceptibility isolve=0 ethr=0.1000E-13 k-point #1 of 1 pool # 1 ik 1 ibnd 16 linter: root not converged 0.305E+07 ik 1 ibnd 16 linter: root not converged 0.139E+31 ik 1 ibnd 16 linter: root not converged 0.122E+27 End of magnetic susceptibility calculation f-sum rule (1st term): ** 0. 0. 0.-29.8546 0. 0. 0.-29.9120 f-sum rule (2nd term): -0.3248 0. 0. 0. -0.3248 0. ... Contributions to the NMR chemical shifts: --- Core contribution in ppm: Atom 1 C pos: ( 0.00 0.107679 0.00) sigma: 200.51 Atom 2 C pos: ( 0.093253 0.053840 0.00) sigma: 200.51 Atom 3 C pos: ( 0.093253 -0.053840 0.00) sigma: 200.51 Atom 4 C pos: ( 0.00 -0.107679 0.00) sigma: 200.51 Atom 5 C pos: ( -0.093253 -0.053840 0.00) sigma: 200.51 Atom 6 C pos: ( -0.093253 0.053840 0.00) sigma: 200.51 Atom 7 H pos: ( 0.00 0.191523 0.00) sigma: 0.00 Atom 8 H pos: ( 0.165864 0.095761 0.00) sigma: 0.00 Atom 9 H pos: ( 0.165864 -0.095761 0.00) sigma: 0.00 Atom 10 H pos: ( 0.00 -0.191523 0.00) sigma: 0.00 Atom 11 H pos: ( -0.165864 -0.095761 0.00) sigma: 0.00 Atom 12 H pos: ( -0.165864 0.095761 0.00) sigma: 0.00 Bare contribution in ppm: Atom 1 C pos: ( 0.00 0.107679 0.00) sigma:NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN Atom 2 C pos: ( 0.093253 0.053840 0.00) sigma:NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN ... Total NMR chemical shifts in ppm: --- Atom 1 C pos: ( 0.00 0.107679 0.00) sigma:NaN C1anisotropy: NaNeta:0. C1sigma_xx= NaNaxis=( NaN NaN 0.00) C1sigma_yy= NaNaxis=( NaN NaN 0.00) C1sigma_zz= NaNaxis=( 0.00 0.00 1.00) Atom 2 C pos: ( 0.093253 0.053840 0.00) sigma:NaN C2anisotropy: NaNeta:0. C2sigma_xx= NaNaxis=( NaN NaN 0.00) C2sigma_yy= NaNaxis=( NaN NaN 0.00) C2sigma_zz= NaNaxis=( 0.00 0.00 1.00) -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] ibrav=12
Dear Paul, commenting your note, and as a chemist, I've been confused by the convenction used in quantum espresso. I'm using a, b, c, alpha, beta and gamma, but not always I've been able to directly translate the content of cif files into QE input... Here is an example about monoclinic input: http://www.democritos.it/pipermail/pw_forum/2010-February/015993.html I had to switch (b, c) in the cell and (y, z) in the data Finally i decided to use only triclinic for all input files, regardless of the real ibrav... Just my 2c... Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ > > > -- Messaggio originale --- > Oggetto: Re: [Pw_forum] ibrav=12 > Da: "W2AGZ" > Data:Ven, 18 Febbraio 2011, 8:43 pm > A: "'PWSCF Forum'" > -- > > Thanks, Eyvaz. The space group I???m interested in is > P21/c (14). 8 atoms per unit cell, 24 IRs. If you need > a beta tester, just ask. -Paul > > > > From: pw_forum-bounces at pwscf.org > [mailto:pw_forum-bounces at pwscf.org] On Behalf Of Eyvaz > Isaev > Sent: Friday, February 18, 2011 9:05 AM > To: PWSCF Forum > Subject: Re: [Pw_forum] ibrav=12 > > > > Dear Paul, > > In fact, I am working on this problem in order to support > in PlotPhon almost all lattice types. I will let you know > soon, hopefully. > > > Best regards, > Eyvaz. > > --- > Prof. Eyvaz Isaev, > Department of Physics, Chemistry, and Biology (IFM), > Linkoping University, Sweden > Theoretical Physics Department, Moscow State Institute of > Steel & Alloys, Russia, > isaev at ifm.liu.se, eyvaz_isaev at yahoo.com > > > > > > _ > > From: W2AGZ > To: pw_forum at pwscf.org > Sent: Fri, February 18, 2011 10:06:08 AM > Subject: [Pw_forum] ibrav=12 > > To All (especially Eyvaz): > > > > I would like to create a file for PlotPhon/Include to > support general monoclinic structures. Below is the > Include file for orthorhombic as an example ???kick > off:??� > > > > # Simple Orthorhombic lattice (X=b/a, Y=c/a) > > 1.0 0.0 0 > > 0.0 XX0 > > 0.0 0.0 YY > > # With respect to basis vectors > > 1 0.0 0.0 0.5Z > > 50 0.0 0.0 0.0G > > 50 0.50 0.0 0.0X > > 50 0.50 0.5 0.0S > > 50 0.00 0.5 0.0Y > > 50 0.00 0.0 0.0G > > 50 0.50 0.5 0.5R > > > > Now, the ???popular??� cell convention for monoclinic > structures is either ???b-axis-unique,??� or > ???c-axis-unique.??� PWgui and PWscf prefer the latter, > but that???s OK, since it???s a simple transformation, as > least in direct space, between the two. But, what is the > appropiate ???include??� for ibrav=12 ???c-axis unique???� > I???ve been to ???BILBOA,??� and the > Bouchaert-Smoluchowski-Wigner path I want is > gamma-a-u-e-z-lambda-gamma for both electrons and phonons. > Incidentally, when I was a graduate student, the legend > was that the BSW notation was invented by physicists to > purposely confuse chemists. So, please tell me how to > construct the appropriate ???scripts/lines??� and > ???include??� files accordingly. > > > > Grazi, Paul Grant > > IBM Research Staff Member Emeritus > > www.w2agz.com > > > > Addenda...I include the following at the risk of angering > Santo Paolo who may excommunicate me from the forum on the > basis of ???topic out of bounds.??� ???Mea culpa...dominus > noster...??� Therefore I apologize for this posting and > to those for whom English is not their native language, > especially Americans, Australians, Kiwis, Irish, Scots, > Welsh, Cornish, Indians and particularly Quebecois > (British Columbians excepted). > > > > 2011 is the 100th anniversary of the discovery of > superconductivity and the 25th of above liquid nitrogen in > copper oxide perovskites. I will be giving one of the > plenary talks at the upcoming APS March Meeting in > Dallas...go here > <http://meetings.aps.org/Meeting/MAR11/APS_epitome> for > the epitome. And later a reprise at the MRS April Meeting > in San Francisco. But, in addition, I will also present a > contributed talk at each meeting on ???proxy??� CuO > structures based entirely on Quantum Espresso > resources...go here > &
[Pw_forum] a brief and (maybe) stupid question
Dear all, when doing calculations (I'm interested in PW, CP and GIPAW) using PP build with PBE functional (for example), QE compiled with -DEXX flag produces the same results of the version compiled without this flag? I mean comparing the results using functionals that are available in both (with and without -DEXX) of the QE compiled versions. I apologize if this will sound as a dumb question... Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] HSE
Anna, as far as I know (perhaps others should confirm what I'm writing) the HSE functional can be employed, provided that you compile QE with -DEXX. A different problem is the generation of PP by ld1.x with the desired functional. Carlo Anna Maria Ferrari ha scritto: > Dear All > I would be interested in the use of HSE functional. > I know that for the moment the suggestion is to use usual PBE, PW91 > pseudopotentail > since HSE pseudo are not available. Am I right? > Anyway I would like to know if there is an example in the PWSCF list of > examples > or a documentation that i should know in order > to prepare a correct input for this long-range scheme? > > thanks > anna > > *** > Anna Maria Ferrari > Dipartimento Chimica IFM > Universit? di Torino > via P. Giuria 5 > 10125 Torino ITALY > > phone ++39 +11 6707563 > fax ++39 +11 670 7855 > *** > > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] B3LYP: two conflicting matching values
Dal Corso Andrea ha scritto: > On Fri, 2010-03-26 at 13:42 +0100, Carlo Nervi wrote: >> I apologize to everybody... >> it seems that it was a my mistake. I used again ld1.x and the errors >> does not pop up any more Very probably i was mixing different ld1.x >> compiled versions... >> Sorry, >> Carlo > > Presently ld1.x does not support hybrid functionals. As far as I know, > so far nobody has implemented it. > > Andrea > ??? This is confusing to me... I managed to generate a NC b3lyp PP, and apparently it works in PWSCF (compiled with -DEXX). Apparently a simple scf test worked out... Of course, I had still no time to check whether this b3lyp PP is reliable or not... but this should be (I think) another question... Am I wrong? Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] B3LYP: two conflicting matching values
I apologize to everybody... it seems that it was a my mistake. I used again ld1.x and the errors does not pop up any more Very probably i was mixing different ld1.x compiled versions... Sorry, Carlo Lorenzo Paulatto ha scritto: > On Fri, 26 Mar 2010 12:07:15 +0100, Carlo Nervi > wrote: >> P.S.: the same happens using the atomic_doc/paw_library/input/C.in by >> Lorenzo Paulatto if you modift dft='PBE' into dft='b3lyp'. > > PAW does not support BLYP nor exact exchange (yet?) so don't waste your > time on that. > > Unluckily, I have no clue about your primary question. > > regards > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] B3LYP: two conflicting matching values
Thanks Lorenzo, indeed I'm trying to generate b3lyp using the norm conserving PP. I was not very clear in my email... Unfortunately the error message appears in all ceses, including the NC PP generation.. In any cases thanks, Carlo Lorenzo Paulatto ha scritto: > On Fri, 26 Mar 2010 12:07:15 +0100, Carlo Nervi > wrote: >> P.S.: the same happens using the atomic_doc/paw_library/input/C.in by >> Lorenzo Paulatto if you modift dft='PBE' into dft='b3lyp'. > > PAW does not support BLYP nor exact exchange (yet?) so don't waste your > time on that. > > Unluckily, I have no clue about your primary question. > > regards > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] (no subject)
Dear Sergey, it is difficult to understand your question, which is too general, and it is impossible to help you if you don't provide any further informations, at least about your input files... since the error message is about the input file Please, also, provide your affiliation. Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ > > > -- Messaggio originale --- > Oggetto: [Pw_forum] (no subject) > Da: ?? > Data:Mer, 17 Marzo 2010, 7:43 pm > A: Pw_forum at pwscf.org > -- > > Hello. > I am trying to perform ph calculation after optimization > of crystal structure. First I make SCF-calculation, then > vc-relax. But when I run ph.x, I get error#20:error while > reading input file. > Explaine me please where I am going wrong. > > Thank you. > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > >
[Pw_forum] GIPAW
Yep, thanks Layla and Jonathan! Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/ > > > -- Messaggio originale --- > Oggetto: [Pw_forum] GIPAW > Da: marsamos at democritos.it > Data:Mar, 16 Marzo 2010, 7:29 pm > A: pw_forum at pwscf.org > -- > > For those who are interested in GIPAW, the past tutorial > (held in > zurich at the CECAM, 2009), is available at > http://www.gipaw.net/work.html. > > Thanks Jonathan Yates! > > bests > > Layla > > -- > Dr. L. Martin-Samos > tel. +39 040 3787 429 > CNR-DEMOCRITOS and > International School for Advanced Studies (ISAS-SISSA) > via Beirut 2-4 > 34151 Trieste Italy > > > This message was sent using IMP, the Internet Messaging > Program. > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > >
[Pw_forum] AMD Open64 Compiler Suite
Thanks Paolo, On a old dual socket opteron 248 I tried to compile PW using the default gfortran. Apparently it works. You mentioned that gfortran is buggy. Which version is buggy? I tried with 4.1.2 There are also newer versions (on another machine I have the 4.3.4), but I did no try yet... Concerning the AMD Open64 I was wondering if it is worth to try. Perhaps it would make sense only in the case of new AMD CPUs of some "power", not my "old dual socket Opteron 248" (I'd like to try the new 12 core Magny-Cours...) However, first or later I hope I'll be able to find some spare time to try to install open64 and try.... Cheers, Carlo > Carlo Nervi wrote: > >> anyone tried to compile PWSCF using the AMD Open64 >> Compiler Suite > > just tried: it works, sort of (it fails a few tests). If > you find the best options, we can add it to "configure" > > P. > -- > 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 >
[Pw_forum] AMD Open64 Compiler Suite
Dear PWSCF users, anyone tried to compile PWSCF using the AMD Open64 Compiler Suite (the last versione -afaik- is v4.2.3), possibly with the AMD Core Math Library (ACML)? I know that ACML should work, but how about Open64? After reading that Intel Compilers deliberately does not take in account of all math speedup present in non-intel CPUs, I'm considering which optimized compiler I should use on my old non-Intel based linux... Thanks, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] pw.x diagonalization (DSYGV*) failed
Dear Lorenzo, thank you for the suggestions. Finally I decided to use your PP (I'm using C, H, N, O and S). The generation of paw PP now apparently goes smoothly. However, a simple scf calculation on my system (52 atoms in the cell) cannot reach self-consistency: the energy continue to oscillate even after 200 scf cycles... (I tried with a couple of ecutoff, i.e. 40). Sorry to ask again... do you have any hints? Should I roll back to non-paw PP (I was using rrkjus)? Should I update to the latest CVS? Thanks, Carlo Lorenzo Paulatto ha scritto: > Replying to message "[Pw_forum] pw.x diagonalization (DSYGV*) failed" from > Carlo Nervi (11/03/10): >> Anyone could give me a hint where I'm wrong? >> Wrong Pseudopotentials? > > Probably, I would bet you have ghost states. > >> I tried also to use the example given in >> espresso-4.1.2/atomic_doc/paw_library/input/S.in >> however, ld1.x exit with error >> from new_paw_hamiltonian : error # 1 >> negative rho > > Remove the input parameter rmatch_augfun and it will work (already fixed > in CVS). > > cheers > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] pw.x diagonalization (DSYGV*) failed
Dear all, I generated paw PP-US using ld1.x of QE412, following some examples given by Lorenzo Paulatto in the paw directory. A simple scf calculation run with the mpirun command exits with: from pdpotrf : error #82 problems computing cholesky decomposition After reading the forum I followed the suggestion given by Paolo Giannozzi and used "pw.x -ndiag 1" and the error is now: from rdiaghg : error # 503 diagonalization (DSYGV*) failed Anyone could give me a hint where I'm wrong? Wrong Pseudopotentials? I tried also to use the example given in espresso-4.1.2/atomic_doc/paw_library/input/S.in however, ld1.x exit with error from new_paw_hamiltonian : error # 1 negative rho Thank you, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] Parallel on QuadCore
Nay Lin ha scritto:
> Dear users,
>
> i compiled mpi version pwscf 4.1.2 on quad core. with ubuntu linux
> ./pw.x shows "Parallel Version (MPI)".
> In make.sys, DFLAGS contains -D__MPI and -D_PARA.
> I have mpirun installed.
>
> may i know what is the complete mpirun command to initiate the
> calculation.
> when i run
> mpirun -np 2 ../../bin/pw.xhttp://si.scf.in>>si.scf.out
> i get error such as
> It seems that [at least] one of the processes that was started with
> mpirun did not invoke MPI_INIT before quitting (it is possible that
> more than one process did not invoke MPI_INIT -- mpirun was only
> notified of the first one, which was on node n0).
>
This looks like a problem in running/compiling with mpi (open mpi?). Try
first to compile a simple mpi program and run it with the mpirun
command. Does it works?
Carlo
--
--
Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8
Fax: +39 011 6707855 - Dipartimento di Chimica IFM
via P. Giuria 7, 10125 Torino, Italy
http://lem.ch.unito.it/
[Pw_forum] bug in data input?
Dear Miguel, thank you for the reply. Indeed also I come to the conclusion to perform a permutaion (instead of 3,1,2 I did 1,3,2, which seems to me equivalent after looking at the XCrysden screen). What is confusing, for a end-user like me, is apparently the possibility in the classical approach "A, B, C, cosAB, cosAC, cosBC" (even though I would like more the a,b,c, alfa,beta,gamma way of input) to define all these quantities without any restrictions. But this is not the case as I discovered later. The input of the "A, B, C, cosAB, cosAC, cosBC" mode works exactly like the celldm() way. This was not clear to me reading the INPUT_PW.html#id3667896 and INPUT_PW.html#id3668028 documents. Thank you, Carlo Miguel Mart?nez ha scritto: > Dear Carlo, > > You are using the standard setting, taking the B axis as the monoclinic > one. Looking at Doc/INPUT_PW.html you will see that quantum espresso > takes *C* as the monoclinic axis. For simple monoclinic cells, you just > have to perform the permutation 1,2,3 to 3,1,2. > > I've lost all hope of understanding base-centered monoclinic systems, > and I resort to ibrav=0 in such cases. > > Regards, > > Miguel > > > On Wed, 24 Feb 2010 18:52:43 +0100 Carlo Nervi > wrote: >> Dear all, >> I did some geometry optimizations with pw.x starting from a >> monoclininc cell, using Angstrom and the "A, B, C, cosAB, cosAC, >> cosBC" input methods, for example: >> >> &SYSTEM >> ibrav = 12, >> A = 8.0828 , >> B = 12.5828 , >> C = 9.7644 , >> cosAB = 0.0d0 , >> cosAC = >> -0.10713175431387366149628407325294d0 , cosBC = 0.0d0 , >> nat = 104, >> ntyp = 4, >> ecutwfc = 120 , >> / >> >> The input cif file contains the usual a, b, c, and alfa, beta=96.150, >> gamma. Therefore, I think, it should be correct to set cosAC as shown >> above. However, after open the input file with XCrysden, I realized >> that the cell was 90-90-90 (a parallelepiped). >> Where I am wrong? >> >> Playing a little bit with cosAB, cosAC and cosBC I understood that >> only cosAB makes the correct angle (96 degrees) to appears in >> XCrysden. Therefore I tried another approach: switch b and c, and to >> use the celldm() approach. Of course I had to use crystallographic >> fractional coordinates and to exchange y with z in the Atomic >> coordinates. I'm not sure this is the best approach, but apparently >> it works. I would appreciate any suggestions from the community. >> >> &SYSTEM >> ibrav = 12, >> celldm(1) = 15.274655166 , >> celldm(2) = 1.20797 , >> celldm(3) = 1.5567 , >> celldm(4) = -0.107131784 , >> nat = 104, >> ntyp = 4, >> ecutwfc = 120 , >> / >> >> Is it normal that pw.x behaves in this way? In other words, are there >> any restrictions using the "A, B, C, cosAB, cosAC, cosBC" input >> method? >> >> Thank you in advance, >> Carlo >> > > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] bug in data input?
Dear all, I did some geometry optimizations with pw.x starting from a monoclininc cell, using Angstrom and the "A, B, C, cosAB, cosAC, cosBC" input methods, for example: &SYSTEM ibrav = 12, A = 8.0828 , B = 12.5828 , C = 9.7644 , cosAB = 0.0d0 , cosAC = -0.10713175431387366149628407325294d0 , cosBC = 0.0d0 , nat = 104, ntyp = 4, ecutwfc = 120 , / The input cif file contains the usual a, b, c, and alfa, beta=96.150, gamma. Therefore, I think, it should be correct to set cosAC as shown above. However, after open the input file with XCrysden, I realized that the cell was 90-90-90 (a parallelepiped). Where I am wrong? Playing a little bit with cosAB, cosAC and cosBC I understood that only cosAB makes the correct angle (96 degrees) to appears in XCrysden. Therefore I tried another approach: switch b and c, and to use the celldm() approach. Of course I had to use crystallographic fractional coordinates and to exchange y with z in the Atomic coordinates. I'm not sure this is the best approach, but apparently it works. I would appreciate any suggestions from the community. &SYSTEM ibrav = 12, celldm(1) = 15.274655166 , celldm(2) = 1.20797 , celldm(3) = 1.5567 , celldm(4) = -0.107131784 , nat = 104, ntyp = 4, ecutwfc = 120 , / Is it normal that pw.x behaves in this way? In other words, are there any restrictions using the "A, B, C, cosAB, cosAC, cosBC" input method? Thank you in advance, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] huge memory usage in ibrav=1
Dear QE community, I recently tried to run a simple calculation using a large supercell with ibrav=1. The cell contains 13 atoms. pw.x runs, but consumes a huge amount of memory (in the example below more than 400 Mb of ram for each pw.x process). If I set A = 50.0 it consumes more than 1Gb of RAM for each process, and of course having 8 cPU and 8 Gb of RAM the calculation continously swap with the disk... Perhaps I'm doing something wrong, but I cannot find the reason. The data are derived from an analogue calculation where 104 atoms were present in the cell for a real periodic calculation. This optimization runs without any problem. I'm using QE 4.1.2. I would appreciate any suggestion. Thanks, Carlo - &CONTROL title = 'Barb_gas2' , calculation = 'relax' , restart_mode = 'from_scratch' , outdir = './' , pseudo_dir = '../PP/' , prefix = 'Barb_gas2' , disk_io = 'default' , verbosity = 'default' , etot_conv_thr = 1e-4 , forc_conv_thr = 1e-3 , nstep = 256 , iprint = 1, tstress = .true. , tprnfor = .true. , / &SYSTEM ibrav = 1, A = 10.0 , B = 10.0 , C = 10.0 , cosAB = 0.0d0 , cosAC = 0.0 , cosBC = 0.0d0 , nat = 13, ntyp = 4, ecutwfc = 120 , nosym = .true. , / &ELECTRONS electron_maxstep = 256, conv_thr = 1.0D-6 , startingpot = 'atomic' , startingwfc = 'atomic' , mixing_mode = 'plain' , mixing_beta = 0.7D0 , mixing_ndim = 8, diagonalization = 'david' , / &IONS ion_dynamics = 'bfgs' , / ATOMIC_SPECIES H1.00794 H.pbe-tm-gipaw.UPF O 15.99940 O.pbe-tm-gipaw.UPF C 12.01070 C.pbe-tm-gipaw.UPF N 14.00674 N.pbe-tm-gipaw.UPF ATOMIC_POSITIONS (angstrom) O 4.549969634.160122333.24841463 O 4.270404518.611162003.94060959 O 1.700106075.595945646.44124884 N 3.132033644.888918114.86017069 N 4.416145366.392436883.61722930 C 4.064555095.081560783.87216675 C 3.990191197.484120614.31476378 C 3.261086137.199623505.59032474 C 2.606567055.860939415.67109699 H 2.862903094.117466644.98744525 H 4.995064696.533364242.97784697 H 3.918983137.207802326.28834462 H 2.637625007.853299965.80293441 K_POINTS automatic 1 1 1 1 1 1 -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] Parallel Installation Error
Dear Masoud, did you do any recent linux update? Times ago I had similar problems using Linux Gentoo. Strange problems appear if: 1) you set up bash and Intel environment variables also for root 2) try to upgrade using root user. The solution (in my case) was to rebuild almost all the Gentoo executables, especially libraries and the gcc compiler, (Gentoo is a meta distribution: you have to compile every single package) and use the environment settings of ifort Intel compiler ONLY for the user (i.e. local .bashrc) and NOT for the root. So, use gcc for linux and intel ONLY for QE, not for system files. Also, look at the scripts included with Intel compilers: often the mkl libraries and ifort scripts contains redundant material (i.e. PATH) HTH, Carlo Masoud Nahali ha scritto: > Dear Quantum Espresso Users > > I had installed Quantum Espresso previously on my AMD (3 core) parallel. > but Now I get an error !! > I have installed Intel Compilers (version 11) included fort, icc, icpc, > and Math Kernel; and parallelized them by lam successfully. > my operating system is SUSE 11 (64 bit,Enterprise). Also mpif77 and > mpicc work well on my 3 core AMD. > > but error: Fortran 77 compiler cannot create executables !! > > linux-4koa:~/Install/QSPRESSO4.1.2/espresso-4.1.2 # ./configure > MPIF90=mpif77 F77=mpif77 CC=mpicc --enable-parallel > checking build system type... x86_64-unknown-linux-gnu > checking architecture... x86_64 > checking for ifort... ifort > checking for Fortran 77 compiler default output file name... configure: > error: Fortran 77 compiler cannot create executables > See `config.log' for more details. > > > with and without --enable-parallel I saw the same error. anyhow it is > amazing that I had installed parallel Qu! antum Espresso on this AMD and > using > CentOS5.4 (64 bit) as the Operating system. > > I appreciate so much if you help me. Many thanks > > Best Wishes > Masoud Nahali > Physical Chemistry > SUT > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] MKL 10.2 not recognized by configure script / Xeon=amd64?
I confirm that I was able to compile QE with intel ifort 11.1.056 and mkl 10.2.2.025. It added manually the libraries. Of course, you cannot use -lmkl_em64t, but: BLAS_LIBS = -lmkl_blas95_lp64 -lmkl_intel_lp64 -lmkl_intel_sequential -lmkl_core LAPACK_LIBS= -lfftw3xf_intel -lmkl_lapack95_lp64 -lmkl_intel_lp64 -lmkl_intel_sequential -lmkl_core hope this helps, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] worth recompiling ?
Hello, when i was able to compile everything and all was working, I recompiled with intel 11.0 and especially with MKL 10 libraries. There is a switch in intel 11 compiler (-xHost) that choose the best code for the CPU you have. This switch improved (at least for me) the speed. From my point of view I would give to it a try. However, there others in this forum that could properly answer (as you have probably read, now I'm in still trouble and cannot compile anymore QE...) Carlo Boris Skegin ha scritto: > Hallo. > > I managed to compile espresso-4.1 on our Linux PC cluster using > intel/9.1 > mkl/9.1 > vltmpi/mvapich_0.9.4 > > as it is said in > http://www.pwscf.org/user_guide/node13.html#SECTION00036600 : > > "The latest (July 2006) 32-bit version of ifort 9.1 works flawlessy." > > And espresso-4.1 does work in the parallel version on our cluster, at > least, if checked with simple examples. > > > Besides intel/9.1 and mkl/9.1 we have also > > intel/11.0 > intel/10.0 > intel/10.1 > > mkl/10.0 > mkl/10.1 > mkl/9.1 > > As I have some free time now, I'd like to ask you if it is worth to > recompile, say, with intel/11.0 and mkl/10.0 in order to achieve even > better performance for complicated jobs? > > Regards, > Boris Skegin > > > > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] again on OpenMPI 1.3.3
Hi again,
I am sorry to bother all the community with my compiling problems, that
a little OT, but they are quite unusual. Certainly something is wrong in
my machine (Linux Gentoo on dual Xeon 5345), but I cannot guess what.
After many tests and compiling I found that pw.x run perfectly using
"mpirun -np 2", but fail with "mpirun -np 8". The error is
"MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD with
errorcode 0".
This really I cannot understand.
Is there anyone that could give me a hint?
I succesfully compiled the "pingpong" code below with mpif90 (wrapper to
ifort). Also in this case mpirun -np 2 works, but if I put -np 4, 6 or 8
the program crash with the following message:
*** Process received signal ***
Signal: Segmentation fault (11)
Signal code: Address not mapped (1)
Failing at address: (nil)
[ 0] /lib/libpthread.so.0 [0x2b48da46b400]
[ 1] /lib/libc.so.6(fputs+0x1e) [0x2b48da6d9d0e]
[ 2] ./pingpong(main+0x206) [0x402536]
[ 3] /lib/libc.so.6(__libc_start_main+0xf4) [0x2b48da6965e4]
[ 4] ./pingpong [0x4022b9]
*** End of error message ***
I was thinking that ssh does not propagate the environment variables (so
the libraries cannot be found), but it runs on 2 cpus!
Any helps would be greatly appreciated.
Carlo
/* pingpong - measure effective bandwidth and latency */
#include "mpi.h"
#include
#include
#include
#include
#include
#define MAXSIZE (1024*1024)
#define MINSIZE (0)
#define REPEAT 50
#define INCSIZE (2)
#define INCOP *=
#define CALIBRATION_LOOPS 100
#define TAG_PING 1
#define TAG_PONG 2
/* define DETAIL if you want to create histogramms by measuring
the latency of each single ping-pong transfer */
#if 0
#define DETAIL
#include "getus.h"
#endif
#ifdef linux
#define longlong_t long long
#endif
char *buffer;
char *exename;
int min_size, max_size, inc_size, repeats;
int myrank, mysize;
static FILE* fpGlobal = NULL;
static FILE* fpDetail = NULL;
void ping (int to, int from);
void pong (int to);
int main(int argc, char **argv) {
MPI_Status status;
int first;
char fname[128];
MPI_Init(&argc, &argv);
MPI_Comm_rank(MPI_COMM_WORLD, &myrank);
MPI_Comm_size(MPI_COMM_WORLD, &mysize);
if (myrank == 0) {
strcpy(fname, argv[0]);
strcat(fname,".dat");
fpGlobal = fopen(fname,"w");
}
exename = argv[0];
if(mysize % 2 != 0) {
printf ("pingpong must be used with an even number of
processes.\n");
MPI_Finalize();
exit (1);
}
/* set run parameters */
if (argc != 4) {
printf ("usage: pingpong min_size max_size repeats\n");
printf ("using default values for this run\n");
min_size = MINSIZE;
max_size = MAXSIZE;
inc_size = INCSIZE;
repeats = REPEAT;
} else {
min_size = atoi( argv[1] );
max_size = atoi( argv[2] );
inc_size = INCSIZE;
repeats = atoi( argv[3] );
}
buffer = (char *)malloc (max_size);
/* find ping and pong processes */
if ( myrank < mysize/2 ) {
if (myrank % 2 == 0)
ping( myrank + mysize/2, myrank);
else
pong( myrank + mysize/2 );
} else {
first = (mysize/2) % 2;
if (myrank % 2 == first)
pong( myrank - mysize/2 );
else
ping( myrank - mysize/2, myrank );
}
if (myrank == 0) {
fclose(fpGlobal);
}
free (buffer);
MPI_Finalize();
}
void ping( int to, int from ) {
MPI_Status status;
double starttime, totaltime;
double getticks_overhead;
#ifdef DETAIL
longlong_t hr_start, hr_end;
longlong_t calibration = 0;
longlong_t *timings;
#endif
char fname[128];
char bytes[128];
int i, j;
int firstrun = 1;
#ifdef DETAIL
fprintf (stderr, "Calibrating...");
for (i = 0; i < CALIBRATION_LOOPS; i++) {
GETTICKS(&hr_start);
GETTICKS(&hr_end);
calibration += hr_end - hr_start;
}
getticks_overhead = ((double)calibration)/(CALIBRATION_LOOPS);
fprintf (stderr, "gethrtime() overhead is %6.3f\n", getticks_overhead);
timings = (longlong_t *)malloc (repeats*sizeof(longlong_t));
#endif
printf("pingpong from %d to %d\n\n", from, to);
fprintf(fpGlobal, "# msgsize[byte] repeats bandwidth[MB/s]
latency[us]\n");
fflush(stdout);
for( i = min_size; i <= max_size; i INCOP inc_size) {
if ((!firstrun) && (i == 0)) {
i++;
if (i > max_size)
break;
}
--
--
Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8
Fax: +39 011 6707855 - Dipartimento di Chimica IFM
via P. Giuria 7, 10125 Torino, Italy
http://lem.ch.unito.it/
[Pw_forum] openmpi 1.3.3
Hi, sorry to bother again. I recompiled OpenMPI with the suggested options, but I still have problems. I think I have something wrong in my make.sys file. First I was trying to use fftw3xf_intel libs, but ... are they incompatible with MPI? In a previous forum post I read that user were unable to use Intel FFTW3 + OpenMPI, so i switched to the internal FFTW. Paolo, thank you for the message. I know that OpenMPI and OpenMP are different, and after your suggestion I tried to recompile all (after a make clean) using only OpenMPI and do not mix them: ./configure CC=icc F77=ifort F90=ifort FC=ifort CPP=icpc FFLAGS=-xHost FCFLAGS=-xHost CFLAGS=-xHost CPPFLAGS=-xHost (the -xHost tell to Intel compiler to optimize the code for the current CPU), and I got the following part of make.sys: DFLAGS = -D__INTEL -D__FFTW -D__MPI -D__PARA IFLAGS = -I../include MOD_FLAG = -I MPIF90 = mpif90 #F90 = ifort CC = icc F77= ifort CPP= cpp CPPFLAGS = -P -traditional $(DFLAGS) $(IFLAGS) CFLAGS = -O3 -xHost $(DFLAGS) $(IFLAGS) F90FLAGS = $(FFLAGS) -nomodule -fpp $(FDFLAGS) $(IFLAGS) $(MODFLAGS) FFLAGS = -O2 -xHost -assume byterecl FFLAGS_NOOPT = -O0 -assume byterecl -xHost LD = mpif90 LDFLAGS= -static-intel -openmp LD_LIBS= BLAS_LIBS = -lmkl_em64t FLIB_TARGETS = all LAPACK_LIBS= -lmkl_em64t SCALAPACK_LIBS = The -openmp (and -parallel ?) direction to ifort should be used or not? As far as i understood only the -openmp should be passed to the linker. And only in the case I'd like to use OpenMP I should use the -openmp directive for ifort compiler. Is it? I tryed also to use "-lmkl_intel_lp64 -lmkl_sequential -lmkl_core" instead of "-lmkl_em64t": no success... Thank you, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] openmpi 1.3.3
Dear forum users, I tried to compile QE (snapshot 25-09-09 version) under linux with ifort 11.0 and MKL libraries 10.1.1.019 using OpenMPI 1.3.3 As far as I read OpenMPI 1.3 has improved performances over the previous versions... I (apparently) successfully compiled OpenMPI with Intel ifort (the hello.World program run on all 8 core of my linux pc). If I run pw.x in serial mode (OMP_NUM_THREADS=1) I got the correct results, but if I try to run pw.x using OMP_NUM_THREADS=8 either it runs forever (no convergence) reporting erroneous results, or it crash. If I run it by "mpirun -np 8 pw.x -in scf.it > scf.out &" (using OMP_NUM_THREAD=1), i get always the following message: "MPI_ABORT was invoked on rank 5 in communicator MPI_COMM_WORLD" I found the following url: http://software.intel.com/en-us/forums/intel-math-kernel-library/topic/68207/ where they explain that "...Since Open MPI considers MPI_COMM_WORLD to be a pointer it turns out to be 64-bit long. Whereas Cluster FFT was designed in times where sizeof(MPI_Comm) used to be 32-bit. In order to work correctly with Open MPI you just need to wrap the communicator as follows: DftiCreateDescriptorDM(MPI_Comm_c2f(MPI_COMM_WORLD),&desc,DFTI_DOUBLE,DFTI_COMPLEX,1,len);" I don't have enough experience to understand if this could be really relevant to the present topic, could be useful to the whole QE community that would like to compile QE using ifort and MKL, or it is totally irrelevant and unuseful... Somebody have tried to successfully use OpenMPI v 1.3.3? Maybe with ifort and MKL? I would greatly appreciate any comments. Thanks! Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] About example 22
Hello, I found also that example21 in the CVS version of 25-09-09 returns "from read_namelists : error # 1 fpmd calculation no more supported, use cp instead" Hope this helps, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] GIPAW: impossible value for nrc
Sorry, in the previous email I copy&paste another very similar error
from another job. The error message regarding the nmr-example is the
following:
If I use this generated PP the scf.out contain
Program GIPAW v.4.1CVS starts ...
Today is 23Apr2009 at 15:16:35
Parallel version (MPI)
Number of processors in use: 1
WARNING: Pseudopotential # 1 file : Si.pbe-rrkj-gipaw.UPF
WARNING: WFC # 1(3S) IS NOT CORRECTLY NORMALIZED: norm= 1.439008
WARNING: WFC HAS BEEN NOW RENORMALIZED
WARNING: Pseudopotential # 1 file : Si.pbe-rrkj-gipaw.UPF
WARNING: WFC # 2(3P) IS NOT CORRECTLY NORMALIZED: norm= 1.940790
WARNING: WFC HAS BEEN NOW RENORMALIZED
...
chi_bare vGv (VV) in 10^{-6} cm^3/mol:
-47.9269 0. 0.
0.-47.9269 0.
0. 0.-47.9269
%%
from grid_gather : error # 63936
f_out too small
%%
stopping ...
First: I cannot understand why I got the warnings after scf calculations
after simply adding the gipaw_recontruction=.true.
Sorry for bothering a lot...
Carlo
--
------
Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8
Fax: +39 011 6707855 - Dipartimento di Chimica IFM
via P. Giuria 7, 10125 Torino, Italy
http://lem.ch.unito.it/
[Pw_forum] GIPAW: impossible value for nrc
Dear all, I'm getting crazy with GIPAW... :-) Finally I did a step back and did the nmr-example provided by Seitsonen on his web pages, using the Si_pbe-20070930.UPF PP. All works! I got exactly the same results. Well, than I decided to use the Si.pbe-rrkj.in included in the pseudo_library/PBE/SR dir. Here it is, just to make sure all is okay: &input title='Si', zed=14.0, rel=1, config="[Ne] 3s2.0 3p2.0 3d-2.0" iswitch=3, dft='PBE', / &inputp lloc=2, pseudotype=1, file_pseudopw='Si.pbe-rrkj.UPF', author='ADC', / 3 3S 1 0 2.00 0.00 2.20 2.20 3P 2 1 2.00 0.00 2.40 2.40 3D 3 2 -2.00 0.20 2.40 2.40 If I use this generated PP the scf.out looks okay. Than I modified it in order to create the PP for gipaw: &input title='Si', zed=14.0, rel=1, config="[Ne] 3s2.0 3p2.0 3d-2.0" iswitch=3, dft='PBE', / &inputp lloc=2, pseudotype=1, file_pseudopw='Si.pbe-rrkj.UPF', file_recon='Si.pbe-rrkj.recon', lgipaw_reconstruction = .true. tm = .true. author='ADC', / 3 3S 1 0 2.00 0.00 2.20 2.20 3P 2 1 2.00 0.00 2.40 2.40 3D 3 2 -2.00 0.20 2.40 2.40 &test / 3 3S 1 0 2.00 0.00 2.20 2.20 3P 2 1 2.00 0.00 2.40 2.40 3D 3 2 -2.00 0.20 2.40 2.40 If I use this generated PP the scf.out contain WARNING: Pseudopotential # 1 file : H.pbe-tm-gipaw.UPF WARNING: WFC # 1(1S) IS NOT CORRECTLY NORMALIZED: norm= 1.421108 WARNING: WFC HAS BEEN NOW RENORMALIZED WARNING: Pseudopotential # 1 file : H.pbe-tm-gipaw.UPF WARNING: WFC # 2(2P) IS NOT CORRECTLY NORMALIZED: norm= 26.282579 WARNING: WFC HAS BEEN NOW RENORMALIZED Is this output from pw.x normal? Furthermore, the gipaw crash every time. One error message is: %% from cdiaghg : error #87 diagonalization (ZHEGV*) failed %% The parallel version of gipaw.x crash in a different way, mumbling something about glibc and malloc() (out of memory I suppose...) I'm using the 4.1CVS version, snapshot of January 2009. Any help would be greately appreciated! Thank you, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] GIPAW: impossible value for nrc
Dear all, I'm trying to use GIPAW module for NMR calculations. With the very helpful suggestions of Lorenzo Paulatto I generated the PAW Pseudopotentials of C and H, including gipaw. With these PP I performed geometry optimization whithin the cell (calculation='relax') and subsequently run the "calculation='scf'". When I try to run gipaw.x module, I always get from init_gipaw_1 : error #1 impossible value for nrc The values for rc, rs, nrc,nrs,nt, msh(nt) are: 0.000E+000 0.000E+000 0 0 1 745 so both nrc and nrs are zero, whereas they are required to be >=1. I tried also to copy the original PP files into the directory .save created by pw.x, but with no success. The gipaw input I used is: &inputgipaw job = 'nmr' tmp_dir = '/home/nervi/calc/Barbiturico/Michele_periodico/GIPAW_Hex/' file_reconstruction ( 1 ) = 'H.pbe-paw-gipaw.recon' file_reconstruction ( 2 ) = 'C.pbe-paw-gipaw.recon' isolve = 0 iverbosity = 1 q_gipaw = 0.01 spline_ps = .true. use_nmr_macroscopic_shape = .true. prefix = 'Hex1' / Is there anything that I missed? Could somebody give yto me further hints on what I have to do? Thank you in advance, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] again on "charge is wrong" error
Dear all, I've read some of the past messages about the "charge is wrong" error, since I faced it too. Well, I tried to increase nbnd but the result was only a longer time before crash. I read http://qe-forge.org/tracker/index.php?func=detail&aid=6&group_id=10&atid=133 that suggest a compiler problem. I used MKL 10.1.1.019 and ifort 11.0.081 compiling the version CVS4.1. Inthe input file I switched from diagonalization = 'david' to diagonalization = 'cg' and in my case magically it works. Previously was reported that the diag method does not influence the purported bug. In my case it does. Below is the input file I used. I generated PP with ld1 which includes gipaw and paw (if necssary I can post also this input file). So I cannot understand if it is a bad input file (or unlikely the PP), a compiler/library bug, or there are some other reasons. Cheers, Carlo &CONTROL title = opt_Hexametylbenzene calculation = 'scf' verbosity = 'high' restart_mode = 'from_scratch' outdir = '/home/nervi/calc/Barbiturico/Michele_periodico/Hex_opt2' wfcdir = '/home/nervi/calc/Barbiturico/Michele_periodico/Hex_opt2' pseudo_dir = '/opt/QE/pseudo/' tstress = .false. tprnfor = .true. nstep = 200 prefix = 'Hex_scf2' / &SYSTEM ibrav = 14 A = 5.2602 B = 6.1990 C = 8.0040 cosAB = -0.23883853663434945639315144328238d0 cosAC = -0.15157135729098039184458626927334d0 cosBC = -0.1769473189631088683332049444808d0 nat = 30 ntyp = 2 !nelec = 66 !nbnd = 43 nosym = .true. ecutwfc = 60 ecutrho = 400 / &ELECTRONS diagonalization = 'cg' conv_thr = 1.0d-12 mixing_mode = 'plain' mixing_beta = 0.7 electron_maxstep = 800 / ATOMIC_SPECIES H1.00794 H.pbe-tm-gipaw_CN.UPF C 12.01070 C.pbe-tm-gipaw_CN.UPF ATOMIC_POSITIONS angstrom C1.480397773 1.612313649 2.41309 C0.567917143 2.576536963 2.912448759 C2.398248842 1.016955503 3.333293664 C1.458567352 1.197000496 0.992534797 C -0.390137179 3.238330523 1.951641575 C3.424801493 0.044743239 2.809820675 H1.584075172 2.045815584 0.307303297 H0.500805783 0.728127864 0.720359738 H2.239781168 0.471645652 0.762057598 H -0.457043850 4.319305135 2.119989739 H3.849201391 2.846663344 2.063331085 H -0.094884507 3.092803886 0.908990630 H4.212652583 -0.148282042 3.538235815 H1.511394708 5.096586242 2.526556736 H3.920433573 0.445290897 1.916401289 C0.552060959 2.923991539 4.277926921 C1.470203101 2.328901527 5.167310621 C2.382255680 1.363811124 4.698964273 C3.339744478 0.703097622 5.658479146 H3.407382859 -0.378164308 5.491314032 H -0.898773017 1.094722561 5.547308946 H3.045747097 0.848022744 6.702191454 C1.491410825 2.742208586 6.617237692 H1.366101010 1.894427124 7.303910857 H2.448208749 3.213147289 6.891196900 H0.708636653 3.467720853 6.848772588 C -0.473956442 3.896108675 4.801178466 H -1.263190322 4.088213149 4.072554504 H1.439086797 -1.155263370 5.082385329 H -0.969233483 3.496601812 5.694367685 K_POINTS automatic 3 3 3 0 0 0 -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] cp.x and UPF
> Try to remove the ">" from the CP code and recompile, then > it shall work. In theory it should have worked before as > well, but it seems like the change was lost somewhere in > time. I recompiled the CP code removing the ">" and it works! Thank you Lorenzo, Carlo
[Pw_forum] cp.x and UPF
Thanks Paolo.
Below there is the output of the command
"head -55 /opt/QE/pseudo/H.pbe-rrkjus.UPF"
The pseudopotential was generated using ld1.x as included in 4.1CVS
version. There is a line with
". Also, the line
"generated=..." should not be within the PP_INFO section?
I tried to generate the pseudopotentials again (in the dir
atomic_doc/pseudo_library_PBE/SR, but I got the same output.
Is this the problem?
This means also that in my previous geometry optimization the pw.x
misunderstood ?
I tried to seek into the source code (write_upf.f90), but I cannot
figure out if something is wrong or not.
For example, st line 105 there is
write (ounps, '(//a11)', err = 100, iostat = ios) ""
but I cannot see the corresponding output. Furthermore, I found no line
code that write "" to the file, so I do not know if
all the code used to write the .UPF file is in write_upf.f90
Thanks,
Carlo
Generated using "atomic" code by A. Dal Corso (espresso distribution)
Author: ADC
Generation date: 18Mar2009
Pseudopotential type: USPP
Element: H
Functional: PBE
Suggested minimum cutoff for wavefunctions: 33. Ry
Suggested minimum cutoff for charge density: 409. Ry
The Pseudo was generated with a Scalar-Relativistic Calculation
L component and cutoff radius for Local Potential: 1 1.1000
Valence configuration:
nl pn l occ RcutRcut US E pseu
1S 1 0 1.00 1.100 1.200-0.477209
Generation configuration:
1S 1 0 1.00 1.100 1.200-0.477206
1S 1 0 0.00 1.100 1.200-0.10
2P 2 1 0.00 1.100 1.100 0.10
Pseudization used: rrkj
Paolo Giannozzi ha scritto:
> On Apr 1, 2009, at 21:16 , Carlo Nervi wrote:
>
>> However, when I try to use cp.x for MD the following message appears:
>>
>> Reading pseudopotential for specie # 1 from file :
>>/opt/QE/pseudo/H.pbe-rrkjus.UPF
>>file type is 0: Old CPV NC PP
>
> $ grep 'file type is' */*f90
> (found in CPV/read_pseudo.f90)
>
> $ vi CPV/read_pseudo.f90
> ...
>info = check_file_type( is )
>SELECT CASE (info)
>CASE (0)
> WRITE( stdout,"(3X,'file type is ',I2,': Old CPV NC
> PP')") info
> ...
>
> let us look into check_file_type:
>
>! 0 file is unknown (guess: old CPV norm-conserving format)
> ...
>! 20 file is UPF
>
> So we get 0 instead of the correct code 20, set here:
>
>OPEN( UNIT = pseudounit, FILE = TRIM(filename), STATUS = 'OLD' )
>header_loop: do while (ios == 0)
> read ( pseudounit, *, iostat = ios, err = 200) dummy
> if (matches ("", dummy) ) then
>info = 20
>exit header_loop
>
> So is not found in the upf file. Is it there or not?
>
> P.
> ---
> 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
>
>
--
--
Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8
Fax: +39 011 6707855 - Dipartimento di Chimica IFM
via P. Giuria 7, 10125 Torino, Italy
http://lem.ch.unito.it/
[Pw_forum] cp.x and UPF
I forgot to mention that I'm using the 4.1CVS version of QE. Does anybody could suggest me what I'm doing wrong? Thank you, Carlo Carlo Nervi ha scritto: > Dear all, > I suppose I've correctly generated some pseudopotentials > using the atomic tool ld1.x included in QE: > H.pbe-rrkjus.UPF > Ru.pbe-rrkjus.UPF > C.pbe-rrkjus.UPF > N.pbe-rrkjus.UPF > > The pseudoptentials apparently works fine in geometry > optimizations performed with pw.x > However, when I try to use cp.x for MD the following > message appears: > > Reading pseudopotential for specie # 1 from file : >/opt/QE/pseudo/H.pbe-rrkjus.UPF >file type is 0: Old CPV NC PP > > > from readpp : error # 2 > file format no longer supported > > Perhaps I missed some readings/informations from the > manual, but I would appreciate any suggestions. > Thank you, > Carlo >
[Pw_forum] cp.x and UPF
Dear all, I suppose I've correctly generated some pseudopotentials using the atomic tool ld1.x included in QE: H.pbe-rrkjus.UPF Ru.pbe-rrkjus.UPF C.pbe-rrkjus.UPF N.pbe-rrkjus.UPF The pseudoptentials apparently works fine in geometry optimizations performed with pw.x However, when I try to use cp.x for MD the following message appears: Reading pseudopotential for specie # 1 from file : /opt/QE/pseudo/H.pbe-rrkjus.UPF file type is 0: Old CPV NC PP from readpp : error # 2 file format no longer supported Perhaps I missed some readings/informations from the manual, but I would appreciate any suggestions. Thank you, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] Installation of PWgui
Thank you Tone! It works! I supposed (wrongly) that the pwgui-logo.gif should be created... Carlo Tone Kokalj ha scritto: > On Mon, 2009-03-16 at 14:29 +0100, Carlo Nervi wrote: >> Dear all, >> perhaps it's a dumb question, but I cannot run PWgui. >> The following packages are installed in my linux Gentoo: >> >> tcl-8.4.18 >> tk-8.4.18 >> itcl-3.3-r1 >> itk-3.3-r1 >> iwidgets-4.0.1 >> >> If I understood correctly sources should work. > > Correct! > > > >> However, after typing ./pwgui I got the response below. >> Any suggestions? >> Thanks, >>Carlo >> >> == >>This is PWgui version: 4.1CVS >> -- >> >> >> PWgui: using the system default "tclsh" interpreter >> >> PWGUI : >> /home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS >> GUIB engine : >> /home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/lib/Guib-0.5 >> >> couldn't open >> "/home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/images/pwgui-logo.gif": >> >> no such file or directory > > > Which actually means it works, but one image file is missing in tgz. > > You may fix it as: > > cd /home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/images/ > ln -s pwgui-logo-small.gif pwgui-logo.gif > > Thanks for reporting this bug. > > Regards, Tone > -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] Installation of PWgui
Dear all, perhaps it's a dumb question, but I cannot run PWgui. The following packages are installed in my linux Gentoo: tcl-8.4.18 tk-8.4.18 itcl-3.3-r1 itk-3.3-r1 iwidgets-4.0.1 If I understood correctly sources should work. However, after typing ./pwgui I got the response below. Any suggestions? Thanks, Carlo == This is PWgui version: 4.1CVS -- PWgui: using the system default "tclsh" interpreter PWGUI : /home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS GUIB engine : /home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/lib/Guib-0.5 couldn't open "/home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/images/pwgui-logo.gif": no such file or directory while executing "image create photo logo -format gif \ -file [file join $env(PWGUI) images pwgui-logo.gif]" (file "/home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/src/pwscf.itcl" line 44) invoked from within "source [file join $env(PWGUI) src pwscf.itcl]" (file "/home/nervi/src/Quantum-espresso/espresso-4.1CVS/PWgui-4.1CVS/pwgui.tcl" line 69) -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] compiling QE4.1
>> I managed to compile QE4.1 > > you mean: the CVS version ? Yes, the 4.1CVS version. I know that it is under test. But even though I'm a newbye I would like to contribute to perform some testing because I'm interested in GIPAW for simulation of solid state NMR... As far as I understood only QE and CASTEP, at the moment (ABINIT and other uses PAW), are able to use GIPAW. But maybe I'm wrong because I just entered in this world... >> Now, my question is: how can I be sure that MKL_FFTW3 is >> used? > > -D__FFTW3, link -lfftw3xf_intel or whatever you called it Okay. Thanks. Than I did correctly >> I cannot understand if the clib/fftw.h should be (or >> not?) correctly >> replaced by >> /opt/intel-10.0/Compiler/11.0/081/mkl/include/fftw/fftw.h > > clib/fftw.h is used only by FFTW v.2 > Thank you Paolo, this was one (of many) of my doubts. Therefore my binary is hopefully compiled correctly. But I'm still not convinced that MPI is working 100% as it should be (1500 s for the tests using a 8-core Xeon 5345). By the way, in one of my tests, using ./configure the make.sys wrote FDFLAGS instead of FFLAGS. I was using several switches, like LIBDIR, F90, F77, CC... Cheers, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] compiling QE4.1
Dear all, I managed to compile QE4.1 with MPI, MKL and (perhaps) with MKL_FFTW3, with better performances, but I'm still getting strange timing results. For examples, the "tests" are finished in twice the time of the reference (1500 s instead of about 700), but the example21 with 32 water molecules is done in 1'25", whereas the 64 water molecules is done in 6'50" Since mpiexec apparently run normally, I suppose QE is correctly compiled with MPI. Now, my question is: how can I be sure that MKL_FFTW3 is used? I cannot understand if the clib/fftw.h should be (or not?) correctly replaced by /opt/intel-10.0/Compiler/11.0/081/mkl/include/fftw/fftw.h I included the dir in make.sys, but the LIBOBJS (containing flib.a and clib.a) are (I suppose) linked before the LIBS, therefore I suppose that the fftw.h used is the clib/fftw.h Thank you, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] newbie update
> -lfftw3x_intel must be loaded before -lmkl (in your case, -lmkl_em64t). > Put it whenever you want, but note that LIBS = $(LAPACK_LIBS) $(BLAS_LIBS) $(FFT_LIBS) $ (MPI_LIBS) ... > P. > --- > Paolo Giannozzi, Democritos and University of Udine, Italy Thanks Paolo. Indeed I figure out myself by tryal & error method...only a little later.. sorry.. In the "mpi problem" post I describe what I did: new ifort. By the way, the new ifort 11.0.081 / mkl 10.1.1.019 works a little different. The file /opt/intel/mkl/10.1.1.019/lib/em64t/libmkl_lapack.a contains: GROUP (libmkl_intel_lp64.a libmkl_intel_thread.a libmkl_core.a) The file /opt/intel/mkl/10.1.1.019/lib/em64t/libmkl_em64t.a contains: GROUP (libmkl_intel_lp64.a libmkl_intel_thread.a libmkl_core.a) Therefore I used the same order, but substituing libmkl_intel_thread.a with libmkl_sequential In this way I can comple all. Anf the librabries order does not seems to have effects.. But the results are strange. If I use mpiexec the code crash, but if I run without mpiexec now it run at 100% CPU usage, on a single cpu...It looks like a MPI problem (mpi v. 1.0.8) (see the "mpi problem" at http://www.democritos.it/pipermail/pw_forum/2009-March/011906.html) Thanks, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] newbie update
Axel Kohlmeyer ha scritto: > how much memory does your machine have? what else is running on that machine? Yes, I'm sure it is a 8 core machine, I have 8 Gb RAM, and nothing else was running. I'm the owner and root user of this PC: a dual socket quad-core Xeon E5345 @2.33GHz on which I installed Gentoo Linux. I'm using the stable (Gentoo) mpich2-v1.0.3 However, I discovered that if mpich2 has been compiled with threads option (USE="mpe threads" emerge mpich2) I got link error messages while compiling QE with ifort. I guess there are few problems with mpich2. The latest version of mpich2 (1.0.8) is marked unstable under Gentoo. I downloaded the most recent ifort/icc/mkl compilers and libraries from the Intel site but I'm still in doubt whether to install all or to keep my gcc-4.1.2 p1.3, ifort 10.0 and mkl 9.1.023. I think the main problem lies in my poor understandings of makefile flags, resulting in linking problems with FFTW3 libraries. Right now I'm playing with make.sys and I got: ../Modules/fft_scalar.o: In function 'fft_scalar_mp_cft_b_': fft_scalar.f90(.text+0xc47): undefined reference to 'dfftw_execute_dft_' ... and many other similar link errors... For example, I'm still not sure if this is correct: DFLAGS = -D__INTEL -D__FFTW3 -D__MPI -D__PARA FFT_LIBS = -L/opt/intel/mkl/9.1.023/lib/em64t -lfftw3xf_intel MPI_LIBS = -L/usr/lib64/ -lmpich Sorry for bothering a lot, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] mpi problems
scf iterations detected Reference: 4, You got: Checking lattice-ibrav5...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.236472, You got: discrepancy in number of scf iterations detected Reference: 5, You got: Checking lattice-ibrav6-kauto...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.231540, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav6...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.232909, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav7-kauto...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.231175, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav7...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.233682, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav8-kauto...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.231428, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav8...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.231982, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav9-kauto...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.231201, You got: discrepancy in number of scf iterations detected Reference: 4, You got: Checking lattice-ibrav9...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference:-2.237083, You got: discrepancy in number of scf iterations detected Reference: 5, You got: Checking lda+U-noU...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference: -174.824658, You got: discrepancy in number of scf iterations detected Reference: 9, You got: Checking lda+U-user_ns...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference: -174.537417, You got: discrepancy in number of scf iterations detected Reference: 10, You got: Checking lda+U...application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0[unset]: aborting job: application called MPI_Abort(MPI_COMM_WORLD, 0) - process 0 discrepancy in total energy detected Reference: -174.453376, You got: discrepancy in number of scf iterations detected Reference: 15, You got: Checking lsda-cg...passed Checking lsda-mixing_TF...passed Checking lsda-mixing_localTF...passed Checking lsda-mixing_ndim...passed Checking lsda-nelup+neldw...passed Checking lsda-tot_magnetization...passed Checking lsda...passed Checking lsda, step 2 ...passed Checking md-pot_extrap1...1 FAILED with error condition! Input: md-pot_extrap1.in, Output: md-pot_extrap1.out, Reference: md-pot_extrap1.ref Aborting -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] newbie update
> Carlo Nervi wrote: > >> Total wall time (s) spent in this run: 5274.73 >> Reference: 720.04 >> >> There are few discrepancies, but they seems >> reasonables.. >> or not? > > they do. What doesn't seem reasonable to me is 2 1/2 h for > a series of tests that take 12 min not on the "ExaFlop > Machine", but on a rather standard 3GHz IntelCore2 E6850 > (the one I have in front). There is something really > strange in your setup. This is exactly what I was wondering. During the tests 4 CPUs were running at 10%, while the other four were running at 1-5%. It must be something wrong in my settings, but I cannot figure it out. Anyone could please provide a sort of "semi-standard" make.sys for ifort and mkl with a working mpi? Cheers, Carlo -- -- Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] newbie update
.discrepancy in total energy detected Reference: -55.690284, You got: -55.690283 discrepancy in number of scf iterations detected Reference: 12, You got: 14 Checking noncolin-constrain_total...discrepancy in total energy detected Reference: -55.544783, You got: -55.544784 discrepancy in number of scf iterations detected Reference: 32, You got: 30 Checking noncolin...discrepancy in pressure detected Reference: 193.22, You got: 193.53 Checking noncolin, step 2 ...passed Checking paw-atom...passed Checking paw-atom_l=2...passed Checking paw-atom_lda...passed Checking paw-atom_spin...discrepancy in total energy detected Reference: -41.264991, You got: -41.265001 discrepancy in number of scf iterations detected Reference: 6, You got: 5 Checking paw-atom_spin_lda...discrepancy in total energy detected Reference: -40.244090, You got: -40.244091 Checking paw-bfgs...discrepancy in number of scf iterations detected Reference: 7, You got: 8 Checking paw-vcbfgs...passed Checking relax-damped...passed Checking relax-el...passed Checking relax...passed Checking relax2-bfgs_ndim3...passed Checking relax2...passed Checking scf-cg...passed Checking scf-disk_io...passed Checking scf-gamma...passed Checking scf-k0...passed Checking scf-kauto...passed Checking scf-mixing_TF...passed Checking scf-mixing_beta...passed Checking scf-mixing_localTF...passed Checking scf-mixing_ndim...passed Checking scf-ncpp...discrepancy in total energy detected Reference: -15.839765, You got: -15.839767 Checking scf-wf_collect...passed Checking scf...passed Checking scf, step 2 ...passed Checking spinorbit...passed Checking spinorbit, step 2 ...passed Checking uspp-cg...passed Checking uspp-mixing_TF...passed Checking uspp-mixing_localTF...passed Checking uspp-mixing_ndim...passed Checking uspp-singlegrid...passed Checking uspp...passed Checking uspp, step 2 ...passed Checking uspp1-coulomb...passed Checking uspp1...passed Checking uspp2...discrepancy in pressure detected Reference: -30.68, You got: -30.69 Checking vc-relax1...passed Checking vc-relax2...passed Total wall time (s) spent in this run: 5274.73 Reference: 720.04 There are few discrepancies, but they seems reasonables.. or not? It's only very long... Carlo P.S.: I hope I did not post a message too much long... ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
[Pw_forum] newbie update
Dear all, finally I succesfully compiled the code using ifort, mkl and mpich2. However, I'm getting odd results, very probably due to wrong settings. I tried to run the example01 with the following settings: OMP_NUM_THREADS=1 mpiexec -n 8 The example is executed, but with a very long time. All 8 CPUs are running between 5 and 20 % of the cpu time. For some reason we are not using all the power.. I tried to run also "mpiexec -envall -n 8" (maybe the OMP_NUM_THREADS is not passed) but i got identical results. Here is the results of the first example in the example01 directory: results/si.scf.david.out Writing output data file silicon.save PWSCF: 0.50s CPU time,7.27s wall time On the other hand, if I set OMP_NUM_THREADS=8 and do not use mpiexec I got: Writing output data file silicon.save PWSCF: 7.10s CPU time,1.12s wall time Therefore results/si.scf.david.out takes 1.12s instead of 7.27s, a much more reasonable timing. Anyone could give me an advice? Thank you, Carlo -- ------ Carlo Nervi carlo.nervi at unito.it Tel:+39 011 6707507/8 Fax: +39 011 6707855 - Dipartimento di Chimica IFM via P. Giuria 7, 10125 Torino, Italy http://lem.ch.unito.it/
