Re: [QE-users] (no subject)

Thanks, I'll make sure to try those options out! Dan On Mon, Aug 27, 2018 at 2:53 PM Paolo Giannozzi wrote: > Reducing mixing_ndim should actually reduce a bit (but not much) the > needed memory. in the very likely case of slow scf convergence, use > mixing_mode='local-TF' and increase the

Re: [QE-users] (no subject)

Thank you Paolo, I wasn't initially sure that it was really a RAM problem, because I tried reducing mixing_ndim but this didn't have any impact on the "Estimated total dynamical RAM." I am sorry that this seemed thoughtless. I tried again with K_POINTS gamma - it is running now without any

Re: [QE-users] (no subject)

Reducing mixing_ndim should actually reduce a bit (but not much) the needed memory. in the very likely case of slow scf convergence, use mixing_mode='local-TF' and increase the initial scf threshold to 10^-6 or so. Paolo On Tue, Aug 28, 2018 at 12:34 AM, Dan Gil wrote: > Thank you Paolo, > >

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear Rebeza and Hüseyin, My first e-mail was mistake that there were no lines like following. noncolin = .true lspinorb = .true The code advices me to add these two lines. Best Regards, - Orijinal Mesaj - Kimden: "Arles V. Gil Rebaza" Kime: "Quantum Espresso" Gönderilenler: 27

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

On Mon, Aug 27, 2018 at 3:13 PM, Hüsnü Kara wrote: > .. . *.rel-pbe-spn-rrkjus_psl.1.0.0.UPF > .. . *.rel-pbe-spnl-rrkjus_psl.1.0.0.UPF > .. . *.rel-pbe-spn-rrkjus_psl.1.0.0.UPF > .. . *.rel-pbe-dnl-rrkjus_psl.1.0.0.UPF > all these are fully relativistic

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

According to the error: Add lspinorb=.true. to the input Regards Fariba IASBS On Mon, 27/08/2018 09:13 AM, Hüsnü Kara wrote: > Dear Users, > > My *.in file is following: > > &&& > > calculation = 'scf' >

[QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear All, I want to make the scf calculation which is spin-polarized without spin-orbit contribution. When I try it like following: calculation = 'scf' pseudo_dir =

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear if you are performing non-colinear calculation (noncolin = .true.), you must use SO coupling and the tag nspin=4. Best Arles V. Gil Rebaza Instituto de Física La Plata Argentina 2018-08-27 9:22 GMT-03:00 Hüsnü Kara : > Dear All, > > I want to make the scf calculation which is

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear Hüsnü, If you add "noncolin" and "lspinorb" values as true, you have to use relativistic pseodopotentials and include spin orbit intetaction. I advice you to remove these two ( noncolin and lspinorb) and resubmit again. 27 Ağu 2018 Pzt 15:23 tarihinde Hüsnü Kara şunu yazdı: > Dear All, >

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear Users, My *.in file is following: &&& calculation = 'scf' pseudo_dir = '/truba/home/../qe-6.0/pseudo/' / ibrav = 0 nat = 128 ntyp = 4 starting_magnetization(1) = 1

Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution

Dear Paolo and Others, Thank you very much, Best Regards, - Orijinal Mesaj - Kimden: "Paolo Giannozzi" Kime: "Quantum Espresso" Gönderilenler: 27 Ağustos Pazartesi 2018 16:23:16 Konu: Re: [QE-users] Spin-polarized scf calculation without spin-orbit contribution On Mon, Aug 27, 2018

Re: [QE-users] (no subject)

On Sat, Aug 25, 2018 at 2:00 AM, Dan Gil wrote: [...] With the input format shown below, I get the following error from the > HPC. > > - - - - - > /tmp/slurm/job8327489/slurm_script: line 11: 111713 > Killed pw.x < sapphire.pw.in > sapphire.pw.out > slurmstepd: error: Exceeded

Re: [QE-users] How to carry out a series of simulations at the same time and output the stress and energy to one file?

Hi Hao, If you take a look at one of the PWscf examples, specifically one of the “run_example” shell scripts, you can find an example for how to do what you are asking. The book Stefano suggested will teach you how to write such scripts. There are also free alternatives online that are just a

Re: [QE-users] How to carry out a series of simulations at the same time and output the stress and energy to one file?

Dear Dr. Stephen Weitzner Thanks very much. Would you like to let me know which example I should look at? Thanks again. Cheers! Hao Weitzner, Stephen Eric 于2018年8月27日周一 下午8:22写道： > Hi Hao, > > > > If you take a look at one of the PWscf examples, specifically one of the > “run_example” shell

[QE-users] How to carry out a series of simulations at the same time and output the stress and energy to one file?

Dear All I am a beginner of quantum espresso. I tried to run single simulations and it seem to work now. What I want to investigate is the phase transformation process of silicon. I need carry out a bunch of simulations with different lattice constants. I want to output the energy and the

Re: [QE-users] How to carry out a series of simulations at the same time and output the stress and energy to one file?

https://www.dummies.com/DummiesTitle/UNIX-For-Dummies-5th-Edition.productCd-0764541471.html ___ Stefano Baroni, Trieste -- http://stefano.baroni.me > On 28 Aug 2018, at 02:01, Hao Chen wrote: > > Dear All > > I am a beginner of quantum espresso. I tried to run single simulations > and

Re: [QE-users] How to carry out a series of simulations at the same time and output the stress and energy to one file?

Thanks. I mean is there some command like in LAMMPS that can carry out simulations multitimes with just changing the parameters and output to the same file? Stefano Baroni 于2018年8月27日周一 下午7:08写道： > > https://www.dummies.com/DummiesTitle/UNIX-For-Dummies-5th-Edition.productCd-0764541471.html > >