[Pw_forum] BFGS ionic relax: maximum number of steps?
Hi, Whenever I run a "restart" relaxation of ions, it always terminated after one bfgs iteration, and tells me, The maximum number of steps has been reached. End of BFGS Geometry Optimization Begin final coordinates ... Even I set nstep = 1000 and electron_maxstep = 1000 So, I have to copy all coordinates into a new input to run it "from_scratch", then it will run normally, terminate after a lot of iterations. Anything wrong with my input file? A compilation error? Or it just means that the convergence has been reached? Regards, Xijun Wang The input file: * calculation = 'relax', restart_mode = 'restart', prefix= 'trip', outdir= './' pseudo_dir= '/RQusagers/wangxiju/espresso/pseudo' etot_conv_thr = 1.0E-6 forc_conv_thr = 1.0D-5 tstress = .true. , tprnfor = .true. , nstep = 1000 , dt = 150 / ibrav = 0, celldm(1) = 19.964957 nat = 64, ntyp = 2, ecutwfc = 25, ecutrho = 300 occupations = 'smearing', smearing = 'cold', degauss = 0.03D0 nspin = 2, tot_magnetization = 2 nbnd = 340 / diagonalization = 'cg', electron_maxstep = 1000, conv_thr = 1.D-6 mixing_mode = 'plain', mixing_beta = 0.1D0, mixing_ndim = 16 / ion_dynamics = 'bfgs', upscale = 100.D0 / CELL_PARAMETERS hexagonal 0.992781869 0.0 -0.001946740 0.0 0.588596312 0.0 -0.254396859 0.0 1.117512248 ATOMIC_SPECIES Zr 91.22400 Zr.pbe-nsp-van.UPF N 14.00674 N.pbe-van_bm.UPF ATOMIC_POSITIONS angstrom N 6.239044778 1.554381787 1.967061325 N 1.558801609 1.554381786 9.815800854 .. -- Dept. of Chem and Biochem, Concordia University 7141 Sherbrooke Ouest, Montreal, QC, Canada (H4B 1R6) Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23 -- next part -- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20111018/41b31e43/attachment.htm
[Pw_forum] problems with vc-relax
Hi,
You are right. Tests should be done. I was suggesting "start" from
gamma k-points to make a quick relaxation. After this, relaxation with
larger k-points could be followed if necessary.
Regards,
Xijun
On Thu, Sep 8, 2011 at 7:35 PM, WF wrote:
> Is it good enough to relax an arbitrary system with only gamma k-points? I
> thought it is necessary to do some test with k-points as this effect change
> from system to system. In my experience, more k-points is necessary for a
> metal-like system, i.e. graphene.
>
> Best regards,
>
> Wu, F
>
>> Date: Thu, 8 Sep 2011 11:12:24 -0400
>> From: xijunw at gmail.com
>> To: pw_forum at pwscf.org
>> Subject: Re: [Pw_forum] problems with vc-relax
>>
>> On Thu, Sep 8, 2011 at 10:26 AM, Elie Moujaes
>> wrote:
>> > Hello Xijun,
>>
>> Hello Elie,
>>
>> > I did not send the e-mail twice intentionally. There was a problem with
> my
>> > e-mail and it kept giving me an error about problems with Windows hotmai
>> > hence why it was sent twice without me knowing except now when I opened
> my
>> > e-mail so I apologize for that. Also I wont include blank lines in the
> input
>> > file either. ?Back to my calculations, I have restarted the
> calculations, it
>> > went fine at first but then I got the error:
>> > from scale_h: not enough memory allocated for radial FFT; try restarting
>> > with a larger cell_factor.
>>
>> Check the manual for definition of cell_factor. The default value is
>> 1.2, you could try a larger value.
>>
>> What's your new K_POINTS setting? Since you are relax the system, I
>> suggest you start from:
>>
>> ? ? ?K_POINTS {gamma}
>>
>> This will far faster. and need far less memory.
>>
>> > Another simple question: I realized when I restarted everything that the
>> > positions of the atoms were the initial ones even before starting the
>> > relaxation process where as the k-points used were the NEW ones. Is that
>> > normal? or should I have substituted the the C atoms positions by the
> most
>> > recent ones then restart again?
>>
>> Yes, restart again with NEW ATOMIC_POSITIONS and "restart_mode=
> 'from_scratch' "
>>
>> Regards,
>> ?- Xijun
>>
>> > Regards
>> > Elie
>> >
>> >> Date: Wed, 7 Sep 2011 16:20:49 -0400
>> >> From: xijunw at gmail.com
>> >> To: pw_forum at pwscf.org
>> >> Subject: Re: [Pw_forum] problems with vc-relax
>> >>
>> >> Hi, Elie,
>> >>
>> >> Please do not repeat sending the same message to the mail list. Also,
>> >> people will be more happy to help you if you keep your mail neat by
>> >> removing all the unnecessary blank lines in the input file.
>> >>
>> >> Regards,
>> >> Xijun
>> >>
>> >> On Wed, Sep 7, 2011 at 3:28 PM, Elie Moujaes
>
>> >> wrote:
>> >> > Dear all,
>> >> >
>> >> >
>> >> >
>> >> > I am relaxing a 60 atom supercell (graphene grain boundary) to make
> the
>> >> > total force on atoms mainly 0.001 Ry/au. However the vc-relax stopped
>> >> > after
>> >> > 4 days of execution and without reaching the desired force threshold.
>> >> > The
>> >> > following output was obtained :
>> >> >
>> >> >
>> >> >
>> >> > 1st relaxation process:
>> >> >
>> >> >
>> >> >
>> >> > Total force 0.091 ? Pressure =- 114.1 kbar
>> >> >
>> >> >
>> >> >
>> >> > 2nd relaxation process:
>> >> >
>> >> >
>> >> >
>> >> > total force 0.1 ? Pressure = -55 kbar
>> >> >
>> >> >
>> >> >
>> >> > (new enthalpy < old enthalpy)
>> >> >
>> >> >
>> >> >
>> >> > 3rd process:
>> >> >
>> >> >
>> >> >
>> >> > total force = 0.058 ? pressure = -11 kbar
>> >> >
>> >> >
>> >> >
>> >> > Then in the 4th process, calculations stopped. Here are the last few
>> >> > lines
>> >> > of the output:
>> >> >
>> >> >
>> >> >
>> >> > iteration # ?2 ? ? ecut= ? ?36.75 Ry ? ? beta=0.30
>> >> > ? ? ?Davidson diagonalization with overlap
>> >> > ? ? ?c_bands: ?5 eigenvalues not converged
>> >> > ? ? ?c_bands: ?1 eigenvalues not converged
>> >> > ? ? ?c_bands: ?3 eigenvalues not converged
>> >> > ? ? ?c_bands: ?1 eigenvalues not converged
>> >> > ? ? ?c_bands: ?4 eigenvalues not converged
>> >> > ? ? ?c_bands: ?4 eigenvalues not converged
>> >> > ? ? ?c_bands: ?4 eigenvalues not converged
>> >> > ? ? ?c_bands: ?4 eigenvalues not converged
>> >> > ? ? ?c_bands: ?5 eigenvalues not converged
>> >> > ? ? ?c_bands: ?5 eigenvalues not converged
>> >> > ? ? ?c_bands: ?2 eigenvalues not converged
>> >> > ? ? ?c_bands: ?2 eigenvalues not converged
>> >> > ? ? ?ethr = ?2.90E-04, ?avg # of iterations = 17.6
>> >> >
>> >> > ? ? ?negative rho (up, down): ?0.132E-01 0.000E+00
>> >> >
>> >> > ? ? ?total cpu time spent up to now is 382283.13 secs
>> >> >
>> >> > ? ? ?total energy ? ? ? ? ? ? ?= ? ?-676.77947904 Ry
>> >> > ? ? ?Harris-Foulkes estimate ? = ? ?-676.82846513 Ry
>> >> > ? ? ?estimated scf accuracy ? ?< ? ? ? 0.17363554 Ry
>> >> >
>> >> > ? ? ?iteration # ?3 ? ? ecut= ? ?36.75 Ry ? ? beta=0.30
>> >> > ? ? ?Davidson diagonalization with overlap
>> >> > ? ? ?c_bands: ?1 eigenvalues not converged
>> >> > ? ? ?c_bands: ?1 eigenvalues not converged
>> >> > ? ? ?c_bands: ?2 eigenvalues not
[Pw_forum] problems with vc-relax
On Thu, Sep 8, 2011 at 10:26 AM, Elie Moujaes
wrote:
> Hello Xijun,
Hello Elie,
> I did not send the e-mail twice intentionally. There was a problem with my
> e-mail and it kept giving me an error about problems with Windows hotmai
> hence why it was sent twice without me knowing except now when I opened my
> e-mail so I apologize for that. Also I wont include blank lines in the input
> file either. ?Back to my calculations, I have restarted the calculations, it
> went fine at first but then I got the error:
> from scale_h: not enough memory allocated for radial FFT; try restarting
> with a larger cell_factor.
Check the manual for definition of cell_factor. The default value is
1.2, you could try a larger value.
What's your new K_POINTS setting? Since you are relax the system, I
suggest you start from:
K_POINTS {gamma}
This will far faster. and need far less memory.
> Another simple question: I realized when I restarted everything that the
> positions of the atoms were the initial ones even before starting the
> relaxation process where as the k-points used were the NEW ones. Is that
> normal? or should I have substituted the the C atoms positions by the most
> recent ones then restart again?
Yes, restart again with NEW ATOMIC_POSITIONS and "restart_mode= 'from_scratch' "
Regards,
- Xijun
> Regards
> Elie
>
>> Date: Wed, 7 Sep 2011 16:20:49 -0400
>> From: xijunw at gmail.com
>> To: pw_forum at pwscf.org
>> Subject: Re: [Pw_forum] problems with vc-relax
>>
>> Hi, Elie,
>>
>> Please do not repeat sending the same message to the mail list. Also,
>> people will be more happy to help you if you keep your mail neat by
>> removing all the unnecessary blank lines in the input file.
>>
>> Regards,
>> Xijun
>>
>> On Wed, Sep 7, 2011 at 3:28 PM, Elie Moujaes
>> wrote:
>> > Dear all,
>> >
>> >
>> >
>> > I am relaxing a 60 atom supercell (graphene grain boundary) to make the
>> > total force on atoms mainly 0.001 Ry/au. However the vc-relax stopped
>> > after
>> > 4 days of execution and without reaching the desired force threshold.
>> > The
>> > following output was obtained :
>> >
>> >
>> >
>> > 1st relaxation process:
>> >
>> >
>> >
>> > Total force 0.091 ? Pressure =- 114.1 kbar
>> >
>> >
>> >
>> > 2nd relaxation process:
>> >
>> >
>> >
>> > total force 0.1 ? Pressure = -55 kbar
>> >
>> >
>> >
>> > (new enthalpy < old enthalpy)
>> >
>> >
>> >
>> > 3rd process:
>> >
>> >
>> >
>> > total force = 0.058 ? pressure = -11 kbar
>> >
>> >
>> >
>> > Then in the 4th process, calculations stopped. Here are the last few
>> > lines
>> > of the output:
>> >
>> >
>> >
>> > iteration #??2?ecut=36.75 Ry?beta=0.30
>> > ?Davidson diagonalization with overlap
>> > ?c_bands:??5 eigenvalues not converged
>> > ?c_bands:??1 eigenvalues not converged
>> > ?c_bands:??3 eigenvalues not converged
>> > ?c_bands:??1 eigenvalues not converged
>> > ?c_bands:??4 eigenvalues not converged
>> > ?c_bands:??4 eigenvalues not converged
>> > ?c_bands:??4 eigenvalues not converged
>> > ?c_bands:??4 eigenvalues not converged
>> > ?c_bands:??5 eigenvalues not converged
>> > ?c_bands:??5 eigenvalues not converged
>> > ?c_bands:??2 eigenvalues not converged
>> > ?c_bands:??2 eigenvalues not converged
>> > ?ethr =??2.90E-04,??avg # of iterations = 17.6
>> >
>> > ?negative rho (up, down):??0.132E-01 0.000E+00
>> >
>> > ?total cpu time spent up to now is 382283.13 secs
>> >
>> > ?total energy??=-676.77947904 Ry
>> > ?Harris-Foulkes estimate???=-676.82846513 Ry
>> > ?estimated scf accuracy> >
>> > ?iteration #??3?ecut=36.75 Ry?beta=0.30
>> > ?Davidson diagonalization with overlap
>> > ?c_bands:??1 eigenvalues not converged
>> > ?c_bands:??1 eigenvalues not converged
>> > ?c_bands:??2 eigenvalues not converged
>> > ?c_bands:??3 eigenvalues not converged
>> >
>> >
>> >
>> > The input of the vc-relax is:
>> >
>> >
>> >
>> >
>> >
>> > prefix='GBphonon',
>> >
>> > calculation='vc-relax',
>> >
>> > restart_mode='from_scratch',
>> >
>> > tstress=.true.,
>> >
>> > tprnfor=.true,
>> >
>> > pseudo_dir =
>> > '/exp/home/caiapo/emoujaes/espresso/espresso-4.3/Pseudo/',
>> >
>> > outdir='/exp/home/caiapo/emoujaes/espresso/Moujaes-Results/',
>> >
>> >
>> >
>> > ?/
>> >
>> > ?
>> >
>> > ibrav= 0, celldm(1) =1.889725989, nat=60, ntyp= 1, ecutwfc
>> > =36.749309,
>> > occupations='smearing', smearing='mp',degauss=0.01
>> >
>> > /
>> >
>> > ?
>> >
>> > conv_thr=1.D-6,
>> >
>> > mixing_beta=0.3D0,
>> >
>> > diago_david_ndim=2,
>> >
>> >
>> >
>> > ?/
>> >
>> >
>> >
>> >
>> >
>> > ?ion_dynamics='bfgs'
>> >
>> >
>> >
>> > /
>> >
>> > $cell
>> >
>> > cell_dynamics='bfgs',
>> >
>> >
>> >
>> > /
>> >
>> >
>> >
>> > CELL_PARAMETERS (alat)
>> >
>> > ??24.064488464???0.000772242???0.0
>> >
>> > ???0.0???6.503051170???0.0
>> >
>> >
[Pw_forum] problems with vc-relax
Hi, Elie, Please do not repeat sending the same message to the mail list. Also, people will be more happy to help you if you keep your mail neat by removing all the unnecessary blank lines in the input file. Regards, Xijun On Wed, Sep 7, 2011 at 3:28 PM, Elie Moujaes wrote: > Dear all, > > > > I am relaxing a 60 atom supercell (graphene grain boundary) to make the > total force on atoms mainly 0.001 Ry/au. However the vc-relax stopped after > 4 days of execution and without reaching the desired force threshold. The > following output was obtained : > > > > 1st relaxation process: > > > > Total force 0.091 ? Pressure =- 114.1 kbar > > > > 2nd relaxation process: > > > > total force 0.1 ? Pressure = -55 kbar > > > > (new enthalpy < old enthalpy) > > > > 3rd process: > > > > total force = 0.058 ? pressure = -11 kbar > > > > Then in the 4th process, calculations stopped. Here are the last few lines > of the output: > > > > iteration #??2?ecut=36.75 Ry?beta=0.30 > ?Davidson diagonalization with overlap > ?c_bands:??5 eigenvalues not converged > ?c_bands:??1 eigenvalues not converged > ?c_bands:??3 eigenvalues not converged > ?c_bands:??1 eigenvalues not converged > ?c_bands:??4 eigenvalues not converged > ?c_bands:??4 eigenvalues not converged > ?c_bands:??4 eigenvalues not converged > ?c_bands:??4 eigenvalues not converged > ?c_bands:??5 eigenvalues not converged > ?c_bands:??5 eigenvalues not converged > ?c_bands:??2 eigenvalues not converged > ?c_bands:??2 eigenvalues not converged > ?ethr =??2.90E-04,??avg # of iterations = 17.6 > > ?negative rho (up, down):??0.132E-01 0.000E+00 > > ?total cpu time spent up to now is 382283.13 secs > > ?total energy??=-676.77947904 Ry > ?Harris-Foulkes estimate???=-676.82846513 Ry > ?estimated scf accuracy > ?iteration #??3?ecut=36.75 Ry?beta=0.30 > ?Davidson diagonalization with overlap > ?c_bands:??1 eigenvalues not converged > ?c_bands:??1 eigenvalues not converged > ?c_bands:??2 eigenvalues not converged > ?c_bands:??3 eigenvalues not converged > > > > The input of the vc-relax is: > > > > > > prefix='GBphonon', > > calculation='vc-relax', > > restart_mode='from_scratch', > > tstress=.true., > > tprnfor=.true, > > pseudo_dir = '/exp/home/caiapo/emoujaes/espresso/espresso-4.3/Pseudo/', > > outdir='/exp/home/caiapo/emoujaes/espresso/Moujaes-Results/', > > > > ?/ > > ? > > ibrav= 0, celldm(1) =1.889725989, nat=60, ntyp= 1, ecutwfc =36.749309, > occupations='smearing', smearing='mp',degauss=0.01 > > / > > ? > > conv_thr=1.D-6, > > mixing_beta=0.3D0, > > diago_david_ndim=2, > > > > ?/ > > > > > > ?ion_dynamics='bfgs' > > > > / > > $cell > > cell_dynamics='bfgs', > > > > / > > > > CELL_PARAMETERS (alat) > > ??24.064488464???0.000772242???0.0 > > ???0.0???6.503051170???0.0 > > ???0.0???0.0???8.470514812 > > > > > > ATOMIC_SPECIES > > ?C??12.0107??C.blyp-mt.UPF > > > > ATOMIC_POSITIONS (angstrom) > > C??-11.330758616??-3.527803203???0.0 > > C??-10.659793092??-1.160339161???0.0 > > C??-12.039843315??-7.112619698???0.0 > > C??-12.041158182??-5.732701936???0.0 > > C??-10.941089654??-4.881442842???0.0 > > .. > > > > Please can anyone advice me on this? Shall I start with the new > configuration and start relaxing again? > > > > Regards > > > > Elie Moujaes > > University of Nott > > University Park > > NGT 3RD > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Dept. of Chem and Biochem, Concordia University 7141 Sherbrooke Ouest, Montreal, QC, Canada (H4B 1R6) Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23
[Pw_forum] problems with vc-relax
On Wed, Sep 7, 2011 at 3:28 PM, Elie Moujaes wrote: > Dear all, > Dear Elie, > > I am relaxing a 60 atom supercell (graphene grain boundary) to make the > total force on atoms mainly 0.001 Ry/au. However the vc-relax stopped after > 4 days of execution and without reaching the desired force threshold. The > following output was obtained : > > Please can anyone advice me on this? Shall I start with the new > configuration and start relaxing again? > I'm also doing some cell relaxation. As I understand, cell optimization is not very easy, and it is common that it does not converge after a very long run. Surely you should restart the calculation to achieve your convergence criteria. You can just change the "restart_mod = 'from_scratch' " to "restart_mode = 'restart' ", and submit the job again to continue from where you stopped. Please be advised that pw.x will read cell parameters / atomic positions, initial density, pseudopotentials from the folder of GBphonon.save, so make sure you copy the folder to computing-nodes. However, I highly suggest you do something to accelerate the convergence. For example, if you do not expect the cell angle to change (i.e. keep angles=90), you may add the following keywords to the section. cell_dofree = xyz This will greatly reduce the uncertainty of the geometry optimization. and reduce energy/force oscillation. Another great idea is to manually choose x, y and z, then do a relaxation of atomic positions only. and plot the total energy changes with x, y and z. However, be noted that different volume means different numbers of PW basis function, and energy comparison in this way could be a little confusing. Anyway, if volume change is not large, still this would help a lot. Since I'm also a learner, any more comments or correction to my comments are welcome. Thanks Xijun > > Regards > > > > Elie Moujaes > > University of Nott > > University Park > > NGT 3RD > > > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Department of Chemistry and Biochemistry, Concordia University 7141 Sherbrooke Ouest, Montreal, QC H4B 1R6 Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23
[Pw_forum] ATOMIC_POSITIONS after 'final coordinates'
Thanks Paolo. If this is true, then the 'final coordinates' is not actually final. -Xijun On Wed, Sep 7, 2011 at 11:45 AM, Paolo Giannozzi wrote: > Most likely the damped dynamics algorithm updates the > atomic positions one last time before printing them > > P. > --- > Paolo Giannozzi, Dept of Chemistry, > 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://www.democritos.it/mailman/listinfo/pw_forum > -- Department of Chemistry and Biochemistry, Concordia University 7141 Sherbrooke Ouest, Montreal, QC H4B 1R6 Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23
[Pw_forum] ATOMIC_POSITIONS after 'final coordinates'
On Wed, Sep 7, 2011 at 4:54 AM, Paolo Giannozzi wrote: > > On Sep 7, 2011, at 1:35 , Xijun Wang wrote: > >> There is an extra "ATOMIC_POSITIONS" section after the "final >> coordinates". It is neither the final coordinates > > why are you saying this? P. Because they are different. -XJ > --- > Paolo Giannozzi, Dept of Chemistry, > 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://www.democritos.it/mailman/listinfo/pw_forum > -- Department of Chemistry and Biochemistry, Concordia University 7141 Sherbrooke Ouest, Montreal, QC H4B 1R6 Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23
[Pw_forum] ATOMIC_POSITIONS after 'final coordinates'
Hi, dear espressors, I'm new to espresso (switching from CPMD) and I have a simple question regarding to the output file of a relax calculation. There is an extra "ATOMIC_POSITIONS" section after the "final coordinates". It is neither the final coordinates, nor the starting coordinates, nor cystal coordinates. What's that? See below the output file. I'd appreciate it if someone can tell me where to find the explanation to the output file in the manual. I have searched the maillist and manual online but find no answer. Xijun Wang -- Department of Chemistry and Biochemistry, Concordia University 7141 Sherbrooke Ouest, Montreal, QC, Canada (H4B 1R6) Tel: 514-848-2424-#5835 (Lab) TA office: SP175.23 *** output file of the relaxation calculation *** - ... ? ? ?Damped Dynamics: convergence achieved in ?11 steps ? ? ?End of damped dynamics calculation ? ? ?Final energy = ? -3499.4924725209 Ry Begin final coordinates ? ? ?new unit-cell volume = ? 5281.07728 a.u.^3 ( ? 782.57490 Ang^3 ) CELL_PARAMETERS (alat= 19.96079920) ? ?1.0 ? 0.0 ? 0.0 ? ?0.0 ? 0.591661302 ? 0.0 ? -0.254163181 ? 0.0 ? 1.122317102 ATOMIC_POSITIONS (angstrom) Zr ? ? ?-0.000121009 ? 0.0 ? 0.48982 Zr ? ? ? 0.050917464 ? 0.0 ? 6.659231281 ... End final coordinates ? ? ?Entering Dynamics: ? ?iteration = ? ?11 ? ? ?<vel(dt)|acc(dt)> = ? 0.78384260 ATOMIC_POSITIONS (angstrom) Zr ? ? ? 0.33474 ? 0.0 ? 0.000130486 Zr ? ? ? 0.050452867 ? 0.0 ? 6.659368418 ... Writing output data file rx.damp.save ...
