Hi,
Without more information we can't help,
1) compiler and library versions
2) your arch.make
3) your input files
Den tir. 17. nov. 2020 kl. 22.01 skrev D. Bennett :
> Dear siesta users,
>
> I'm running some calculations using a parallel version of siesta-psml,
> and I'm finding that some
Hi,
This is not totally unsurprising.
When you hit 128 cores you are effectively only having 30 orbitals per
core. This means when the diagonalization takes place it's blocks will be
~30x30 matrices.
You are reaching the bottleneck of the diagonalization routines which takes
up a considerable
fference that i
> am getting is very small. So, could you please tell me if I am going in the
> right direction?
>
> On Mon, Sep 21, 2020 at 6:56 PM Nick Papior wrote:
>
>> How did you do it? How were your unit conversions etc.?
>>
>> Remember that sisl automatical
at I want to make a calculation for the leads two times..one time
> with spin up and the other time with all spins down how can I make such
> thing ? In other words what are the options that I should add to the input
> file of the leads ?
>
> On Wed, Sep 23, 2020 at 11:01 PM Nick Pa
Please be more specific in your question, it isn't clear to us what
anti-parallel means?
Den tir. 22. sep. 2020 kl. 22.05 skrev rayan moukhadder <
rayanroro321...@gmail.com>:
> Dear siesta users,
> I want to perform a calculation in transiesta code with two graphene
> sheets sandwiching a cobalt
ity Silchar.
>
> On Mon, Sep 7, 2020 at 12:24 PM Nick Papior wrote:
>
>> Hi,
>>
>> Den søn. 6. sep. 2020 kl. 16.36 skrev Harkishan Dua > >:
>>
>>> Dear sir,
>>> I am trying to generate the potential plot for monolayer CdS and Bilay
Den fre. 4. sep. 2020 kl. 22.00 skrev zhyphy :
> Dear All,
> I have some questions in the process of learning Green's function.I hope
> someone can help me.
> (1) The surface Green function contains all the informations about
> interaction between electrodes and scatter region, but for
you'll find that
the voxel spacing (along Z) should correspond nicely with your mismatch.
>
> Attached with this email are the plots for CdS monolayer and Bilayer
> systems.
>
> On Wed, Sep 2, 2020 at 2:11 PM Boubacar Traore
> wrote:
>
>> Thanks Nick for this upda
The same flag applies to transiesta. :)
On Tue, 1 Sep 2020, 22:01 El-abed Haidar,
wrote:
> Good afternoon,
>
> In siesta you can use UseData to continue a siesta job, what command
> should I use to make sure that TranSIESTA continues from where it stops
> especially if the job was not
You simply do "siesta xxx.fdf"
TranSiesta is enabled through the fdf-option "SolutionMethod".
As for the I-V curve, yes, I would recommend you to compile siesta with
NetCDF4 support.
Den fre. 28. aug. 2020 kl. 22.00 skrev MANISH KR MOHANTA <
manish.ph16...@inst.ac.in>:
> Dear Experts,
>
> I am
Hi,
In this it seems you don't need the filtering that macroave is capable of.
In that case you can use sisl to create the data for you.
The command would be something like this:
sgrid siesta.VH --average 0 --average 1 --out VH_z.dat
which will write a 2 column data file with z coordinates (in
Could you try with a nonzero eta for the electrodes?
Same goes for transiesta calculations with bias.
Say a value of 1e-4 eV.
One cannot use eta==0 since the dos explodes if you hit an eigenstate.
On Sat, 25 Jul 2020, 07:12 赵鸿远, wrote:
> Dear all,
> I am using siesta4.1-b4 version. I want
you are aiming for?*
>
Yeah, sounds correct.
>
>
> El-abed Haidar | Doctor of Philosophy (Science)
> Condensed Matter Theory (CMT) Group| School of Physics
> THE UNIVERSITY OF SYDNEY | NSW | 2006
>
>
>
>
>
> *From: *Nick Papior
> *Sent: *Friday, 10 July 2020
ng as they are converged) and then increase them only in
> Tbtrans.
>
> A couple of questions though:
>
>1. @Nick Papior when you increase your k points
>in Tbtrans that wont change the kpoints in TSHS files because those are
>TranSIESTA. Is that correct? Because th
The basic principle of transiesta is as with siesta.
A rough outline would be this:
0) Create your system, decide upon the electrodes. Ensure your device
screens off the defect towards the electrodes (meaning you'll need some
additional electrode layers)
1) You converge your density in your
How do you want to read it?
In sisl there is a recent addition which allows you to iterate the values
in the file
python
import sisl as si
wfsx = si.get_sile("file.WFSX")
for es in wfsx.yield_eigenstate():
print(es, es.shape)
###
This lets you use the values in the file, if needed :)
TranSiesta does not use the same file formats.
1) Eigenvalues cannot be calculated using the NEGF formalism, so you won't
get the eig file, ever.
2) If you use siesta-4.1 or later you will only get a projected DOS output
from TBtrans in the *.TBT.nc files. Subsequently you should use sisl,
In conjunction with making Siesta easier to install (and the future PSML
incorporation) I have made some scripts which should ease this.
I here attach the scripts which I have tested (only a few times), any
feedback is very welcome.
They should more or less be self-contained. :)
@Nicolas please
Thanks for bringing this to the attention of the community!
Den ons. 17. jun. 2020 kl. 22.06 skrev I. Camps :
> Dear SIESTers,
>
> I would like to invite you to visit the Materials Modeling Stack Exchange
> forum (in public beta).
>
> The site is dedicated for those interested in building a
ed for k-point convergence.
But you should converge system parameters before doing NEGF calculations.
This includes basis set optimization + mesh-cutoff etc.
>
>
>
>
>
> On Tue, Jun 16, 2020 at 12:36 AM Nick Papior wrote:
>
>> Hi,
>>
>> 1. Great that you are usin
(CMT) Group| School of Physics
> THE UNIVERSITY OF SYDNEY | NSW | 2006
>
>
>
>
>
> *From: *Nick Papior
> *Sent: *Tuesday, 16 June 2020 11:35 PM
> *To: *El-abed Haidar
> *Cc: *siesta-l
> *Subject: *Re: [SIESTA-L] [***Posible SPAM***]Sin Asunto/Subject is blank
Hi,
1. Original poster. Could you please try read the manual. Your problem are
multiple, the electronic structure parameters are insufficient and you are
not using the electrodes as they should. I.e. they are not bulk like.
Your simulation requires the use of real-space self-energies, see
ou and looking forward to your thoughts!
>
> EL-abed
>
>
>
> El-abed Haidar | Doctor of Philosophy (Science)
> Condensed Matter Theory (CMT) Group| School of Physics
> THE UNIVERSITY OF SYDNEY | NSW | 2006
>
>
>
>
>
> *From: *El-abed Haidar
> *Sent: *Tue
Hi,
1. Great that you are using buffer atoms to get the *correct* behaviour of
your electrodes
2. However, you are using MoS2 as electrodes, which is semiconducting. This
is not a good idea since the NEGF formalism requires the electrodes are
*bulk-like* and this cannot be enforced if the
options to manually specify the scaling factor.
>
>
> And one more question! is there any graphical package that I can use to
> visualize the bond current vectors inserted xcrysden?
>
I don't get what you mean? You can use xcrysden?
>
> Thanks.
> Sincerely,
> SH.
>
> On
ind file:
> zmumps_struc.hCan't find file: zmumps_struc.hCan't find file:
> zmumps_struc.hCan't find file: zmumps_struc.hCan't find file:
> zmumps_struc.hCan't find file: zmumps_struc.hCan't find file:
> zmumps_struc.hCan't find file: zmumps_struc.hCan't find file:
> zmumps_struc.h*
>
e)
> Condensed Matter Theory (CMT) Group
>
> | School of Physics
> THE UNIVERSITY OF SYDNEY | NSW | 2006
>
>
>
> *From: *Nick Papior
> *Sent: *Monday, 18 May 2020 6:01 AM
> *To: *siesta-l
> *Subject: *Re: [SIESTA-L] Transietsa/sisl problem
>
>
>
> Hi,
>
ulk: true
> - transmission out: false [TBT.T.Out]
> - transmission out correction: false [TBT.T.Out]
> - transmission out correction (eigen): false [TBT.T.Out,
> TBT.T.Eig]
> - transmission -> elec_1: f
!!!
>
>
>
> Regards,
> SH.
>
> On Sat, May 16, 2020 at 12:30 AM Nick Papior wrote:
>
>> Hi,
>>
>> You probably haven't calculated bond currents. Please check your tbtrans
>> input options.
>>
>> / Nick
>>
>> On Thu, 14 M
../Src/alloc.F90(2343):
> error #6406: Conflicting attributes or multiple declaration of name.
> [DIE] call die('alloc_err: allocate error')---^compilation aborted for
> /home/icamps/temp/SIESTA/siesta-4.1-b4/Util/COOP/../../Src/alloc.F90 (code
> 1)make: *** [../../Obj/arch.make:128:
Hi,
You probably haven't calculated bond currents. Please check your tbtrans
input options.
/ Nick
On Thu, 14 May 2020, 22:13 shirin zandi, wrote:
> Dear Transiesta/sisl user
>
> I am trying to use sisl code to analyze the Transiesta obtained data, But
> when I use this code the below error
ry. Can i do same with 4.0.2 version?
>
> On Thu, 14 May 2020 at 01:30, Nick Papior wrote:
>
>> This can be done if you use siesta-4.1-b4 and if compiled with NetCDF
>> support.
>>
>> Then you will get a file named: *.TBT.nc
>> Then you can use sisl (htt
This can be done if you use siesta-4.1-b4 and if compiled with NetCDF
support.
Then you will get a file named: *.TBT.nc
Then you can use sisl (https://github.com/zerothi/sisl) to extract orbital
resolved DOS by doing something like this:
sdata siesta.TBT.nc --atom 10-12[3,4] --DOS --out
Could you try and do (in Util/COOP):
make clean
make parallel.o
make
and give me the output if it fails?
Den fre. 8. maj 2020 kl. 22.10 skrev I. Camps :
> Hello SIESTers,
>
> I successfully compiled SIESTA with the attached arch.make (link
>
nd or in front of
the scattering region electrode (semi-inf-dir -a3 or semi-inf-dir +a3,
respectively).
Hope that is correct and thank you again for your time.
>
> EL-abed
>
> Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for
> Windows 10
>
>
>
> *Fr
Hi,
This is already fixed in a coming release. :)
You can't get out the wavefunction coefficients in the code (without adding
code).
If you want it you can use sisl https://github.com/zerothi/sisl
to extract the information and post-calculate the eigenvalues again.
Den søn. 3. maj 2020 kl.
ics
> THE UNIVERSITY OF SYDNEY | NSW | 2006
>
> ------
> *From:* siesta-l-requ...@uam.es on behalf of
> Nick Papior
> *Sent:* Saturday, 2 May 2020 7:07 PM
> *To:* siesta-l
> *Subject:* Re: [SIESTA-L] Still confused about K points in siesta,
> transies
Hi,
Great you have tried to fully understand this issue.
Den man. 20. apr. 2020 kl. 22.01 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Following up on that: In the manual it says that if we use* TBT kgrid
> Monkhorst Pack (block):*
> This means it would act as the same as * kgrid
Please see this discussion:
https://www.mail-archive.com/siesta-l@uam.es/msg09386.html
Den tir. 10. mar. 2020 kl. 22.10 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Good afternoon,
> I was doing transiesta and tbtrans calculation with siesta 4.1-b4 version
> I wanted to find the DOS
Hi,
This has no influence, it is a message about the Pulay mixing algorithm
which tries to solve the problem using SVD since the inversion failed.
When an inversion fails in a Pulay mixing scheme it means that two
iterations are extremely close and thus the matrix becomes non-invertible.
When
This has been fixed in later revisions, please try 4.1-b4 :)
On Sat, 7 Mar 2020, 22:01 Rajesh Thakur,
wrote:
> Dear All,
> I tried to run a Molecular Dynamics (MD) calculation for a one-dimensional
> system (Periodic in Z-direction only). In the non-periodic direction, the
> distance between
Sorry for the delay.
I have now runned your big system, and I am not able to reproduce the
problem.
I would suggest you try with other linear algebra packages.
The standard shipped BLAS+LAPACK packages are also *extremely* slow.
Den ons. 4. mar. 2020 kl. 22.12 skrev Giacomo Giorgi :
> Dear all,
In Siesta 4.1 you can do projected transmissions, please see the tbtrans
manual in section 4.6.2 (Projected transmissions).
Den fre. 21. feb. 2020 kl. 22.07 skrev Jingxian Yu <
jingxian...@adelaide.edu.au>:
> Dear colleagues,
>
> I have computed transmission coefficients for the junction device,
Please see the tbtrans manual, just after TBT.DOS.A.All, there is a small
discussion.
Basically DOS.Gf = \sum_electrodes DOS.A + DOS.bound-states
So if there are no bound states you have that the sum of spectral density
of states equals the Green function density of states.
Den fre. 21. feb.
An electrode calculation should not use the transiesta solution. :)
On Wed, 12 Feb 2020, 22:04 Dwaipayan Chakraborty, wrote:
> Thank you very much for your reply.
>
> The input file is like that:
>
> SystemName 3X1X1_elec
> SystemLabel 3X1X1_elec
>
> NumberOfAtoms 18
> NumberOfSpecies 2
>
>
00. 0.
>> %endblock LatticeVectors
>> for the elec-x and elec-z, respectively.
>> So, based on this example should I change the order of lattice vectors
>> for the electrodes (y-direction) which are perpendicular to the device
>> region (z-direction)?
>>
In Siesta 4.1 you don't need to specify TS.TBT.PDOSFrom/To. They are not
used.
If you have the *.TBT.nc files you can simply use the sdata command to
extract PDOS for individual atoms.
Please go through TB_04 example in the TranSiesta tutorial here:
https://github.com/zerothi/ts-tbt-sisl-tutorial
rality5;>
> Sender
> notified by
> Mailtrack
> <https://mailtrack.io?utm_source=gmail_medium=signature_campaign=signaturevirality5;>
> 01/22/20,
> 11:15:34 AM
>
> On Wed, Jan 22, 2020 at 6:01 AM Nick Papior wrote:
>
>> Hi,
>>
>> Your 3rd electr
Hi,
Your 3rd electrode and device lattice vector are not parallel. Please make
sure they are parallel. :)
Den man. 20. jan. 2020 kl. 22.04 skrev Ziba Torkashvand <
zi.torkashv...@gmail.com>:
> Hello everybody
> I want to do a transiesta calculation with Siesta 4.1-b4.
> In my device, two bulk
Hi,
Yes, it first reads the atomic species and atomic information from the fdf
file. Then it proceeds to read the XV file and reads the coordinates and
lattice vectors.
Den tor. 16. jan. 2020 kl. 22.02 skrev D. Bennett :
> Dear siesta users,
>
> I have a very quick technical question: is it
ONLY define transport along a single lattice vector. But this lattice
vector may have components along each of the 3 Cartesian directions.
> Thank you for your time really appreciate it
> El abed
>
> Sent from my iPhone
>
> On 12 Dec 2019, at 8:16 am, Nick Papior wrote:
>
&g
Den tir. 10. dec. 2019 kl. 22.04 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Good evening
> I was just wondering if we have a unit cell with skew vectors ( c vector
> for example having y and z components) I have the following TRANSIESTA
> questions:
>
This is only possible in 4.1 and
Your monkhorst pack grid input block should contain 3 lines of:
integer integer integer float
This is probably why you see the error.
Den tir. 10. dec. 2019 kl. 22.12 skrev RAHUL SURESH :
> Hi
>
> SystemName GRAPHENE
> SystemLabel GH
>
> Systemtype = molecule
>
> NumberOfAtoms
In your electrode calculation, look for:
superc: Internal auxiliary supercell: 3 x 5 x 3 = 45
This should have a 3 along the semi-infinite direction.
Your electrode in your attached output does seem too short.
Den tir. 10. dec. 2019 kl. 22.03 skrev El-abed Haidar <
I guess you are doing a gamma point calculation. In that case avtrans and
trans files are exactly the same, hence it only writes avtrans.
On Tue, 3 Dec 2019, 22:02 El-abed Haidar,
wrote:
> Good evening,
> I am not sure why TRANS file is not printing out as one of the final
> output files.
> I
Hi,
This is a limitation in 4.0 and prior versions.
In 4.1 this restriction has been lifted and you can do calculations on
skewed cells.
Den tor. 21. nov. 2019 kl. 22.00 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Good afternoon,
> I have a question concerning the unit cell for a
If you are using 4.0 and older versions then please check the output for
details. It should clearly state why it stopped without writing the TSDE.
If you are using 4.1 and newer versions, then it is because you are using
the wrong flags. Please carefully read the manual which explains all new
We don't officially have any distribution packages for Siesta.
So you'll have to install Siesta manually. :)
/ Nick
Den tir. 24. sep. 2019 kl. 22.00 skrev Jyotirmoy Deb <
deb.jyotirmo...@gmail.com>:
> Dear Sir/Madam,
> Whether it is possible to install mpi version of SIESTA using yum install
>
Essentially you need to add so many electrode layers that the relaxation
does not induce changes in the layers closest to the electrode. You can
test the effect by adding more electrode layers and changing what you
constrain. You want the physical properties to be converged as a function
of
When you duplicate a system you fold the bandstructure. However, they refer
to the same system in a new reference supercell.
If you *only* duplicate your system you should be able to *visually*
decipher which bands belongs to the original band-structure by following
the folding.
Den søn. 8. sep.
You could use sisl (https://github.com/zerothi/sisl) which can convert
several file formats to other file formats.
It your case you could run:
sgeom input.xyz output.fdf
Which does what you need.
On Sat, 7 Sep 2019, 22:01 RAHUL SURESH, wrote:
> Hi Herbert
>
> I need the input in fdf format.
:
> Thank you for your reply.
>
> @Nick: I attached both files: input and output.
> @Berna: They are different schemes but the output is shown twice with
> different values.
>
>
> []'s,
>
> Camps
>
>
> On Mon, Sep 2, 2019 at 5:09 PM Nick Papior wrote:
>
&
Could you share your input fdf? And also which version of siesta you are
using?
Den søn. 1. sep. 2019 kl. 22.01 skrev I. Camps :
> Hello SIESTers,
>
> I am trying to analyze the charge transfer when a metal is absorbed on a
> surface. In order to do that, I calculated the charges using different
. Camps :
> Thanks Nick.
>
> Unfortunately, the only xml output that I have is related to the ion files
> :(.
>
> Should I have to re-calculate?
>
> []'s,
>
> Camps
>
>
> On Thu, Aug 8, 2019 at 5:00 PM Nick Papior wrote:
>
>> If you also store
If you also store the xml output you can get the spin polarization
information there, key=siesta:stot and key=siesta:svec (only for
non-colinear).
Den tir. 6. aug. 2019 kl. 22.00 skrev I. Camps :
> Dear SIESTers,
>
> In previous versions of SIESTA (aka 3.1), when running a spin polarized
>
No, you probably cannot do this since you need a "screening" layer which
ensures the electrode has the electronic structure of a bulk part.
Den man. 8. jul. 2019 kl. 22.04 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Good afternoon all.
> I was wondering if I calculate a system made of
>
> --
> *From:* siesta-l-requ...@uam.es on behalf of
> Nick Papior
> *Sent:* Monday, 24 June 2019 8:23 PM
> *To:* siesta-l
> *Subject:* Re: [SIESTA-L] Thoughts on PDOS siesta vs PDOS tbtrans
>
> In general,
>
> A 0 bias calculati
Hi,
Den ons. 26. jun. 2019 kl. 22.02 skrev Fazle Subhan :
> Dear Dr. Nick
>
> Previously I work on some nanoribbons for calculating the I-V curve where
> I simply use k-points along the z-direction. Therefore in the P.TRANS file
> I get only one k point (means is it due to the one dimension
; --
> *From:* siesta-l-requ...@uam.es on behalf of
> Nick Papior
> *Sent:* Monday, 24 June 2019 8:23 PM
> *To:* siesta-l
> *Subject:* Re: [SIESTA-L] Thoughts on PDOS siesta vs PDOS tbtrans
>
> In general,
>
> A 0 bias calculation should
Den fre. 21. jun. 2019 kl. 22.02 skrev El-abed Haidar <
ehai2...@uni.sydney.edu.au>:
> Good evening,
> I was wondering for the spin PDOS extracted from Tbtrans, should I
> multiply the values obtained by -1
>
No!
> because I have noticed PDOS spin down calculated via TBTRANs for many
> different
In general,
A 0 bias calculation should yield a somewhat equivalent PDOS in siesta vs.
tbtrans. If this is the case it means that you have set up your system in a
sufficient manner. This is to say that the electrodes already screen off
the junction sufficiently (please spend some time on why this
Try a make clean, otherwise you may have some problematic flags in your
arch.make.
One should not need to edit the arch.make to compile tbtrans.
Den ons. 24. apr. 2019 kl. 22.07 skrev I. Camps :
> Hello,
>
> After successfully compiled siesta-4.1-b4, I moved to
> *siesta-4.1-b4/Util/TS/TBtrans*
> I "solved" removing from arch.make.
>
> Thank you very much.
>
> Camps
>
> On Mon, Apr 22, 2019 at 5:01 PM Nick Papior wrote:
>
>> You have a mixed usage of FPPFLAGS_* and LDFLAGS. So in the end you don't
>> have the include directories for NetCDF (or any of
You have a mixed usage of FPPFLAGS_* and LDFLAGS. So in the end you don't
have the include directories for NetCDF (or any of your other libraries) in
your FPPFLAGS variable (since you do FPPFLAGS = ... which overwrites what
you did).
Simply carefully go through your variables/flags and either
-np ... siesta RUN.fdf > RUN.out
or equivalent, then send the output again.
Den man. 15. apr. 2019 kl. 22.02 skrev I. Camps :
> Hi Nick,
>
> Here it is.
>
> []'s,
>
> Camps
>
>
> On Sun, Apr 14, 2019 at 5:02 PM Nick Papior wrote:
>
>> Could you please do a tar
Could you please do a tar.gz or zip file? rar is horrible ;)
Den lør. 13. apr. 2019 kl. 22.01 skrev I. Camps :
> Hello Nick,
>
> Here are the files.
>
> []'s,
>
> Camps
>
>
> On Fri, Apr 12, 2019 at 5:02 PM Nick Papior wrote:
>
>> Could you attach the o
Could you attach the output, fdf and psf files?
Den tor. 11. apr. 2019 kl. 22.01 skrev I. Camps :
> Hello,
>
> My system is a nanotube with 123 atoms (B, N, Ni). I am running a geometry
> optimization and the calculations are stopped (after 10 CG steps) with the
> following message:
>
> "*Sparse
Probably the easiest way is to use sisl (www.github.com/zerothi/sisl) which
is a Python module.
Something like this:
import sisl
H = sisl.get_sile('RUN.fdf').read_hamiltonian()
print(H[0, 1]) # the Hamiltonian element between orbital 1 and 2
Since H is a sparse matrix you can also loop the
Hi,
Den søn. 31. mar. 2019 kl. 22.02 skrev Rodrigo Garcia Amorim <
rgamo...@id.uff.br>:
> Dear developers,
>
> I saw in the manual that is possible calculate the phonon
> transmission. Regarding this topic, my questions are:
>
> 1) Could I use similar setup of transport calculation?
>
Yes
There are many issues in relaxing a geometry with TranSiesta, besides the
system you have chosen which has a band-gap.
1) For gapped electrodes one cannot be sure the electrostatics in your
electrode calculation and in the device calculation coincide *perfectly*
and hence you will have a mismatch
> On Sat, Mar 9, 2019 at 4:00 PM Nick Papior wrote:
>
>> Could you please delete all TBTGF* files, and rerun the calculation?
>>
>> Alternatively, see if increasing the eta value helps.
>>
>> Den fre. 8. mar. 2019 kl. 22.02 skrev Day, Zacharie :
>>
>>
ow can I calculate Mulliken charge distribution on the system?
>
Search for Mulliken in the siesta manual, it explicitly specifies how this
is done. :)
>
> Best regards
> Mina
>
> Get Outlook for Android <https://aka.ms/ghei36>
>
>
>
>
> On Fri, Mar 22
Qtot is the valence charge for your system. So everything seems fine!
NetCharge is an additional charge added to your system and thus would
change Qtot correspondingly.
Den tor. 21. mar. 2019 kl. 22.11 skrev Mina Sedighi :
> Dear all,
>
>
> I solved SCF convergence problem by playing with
Yes, there is a difference between the siesta and tbtrans.
Siesta is a fully periodic calculation.
TBtrans is a semi-infinite calculation where one or more directions are
replaced by bulk electrodes. Additionally, TBtrans allows non-equilibrium
calculations.
These two systems are not equivalent
Could you please delete all TBTGF* files, and rerun the calculation?
Alternatively, see if increasing the eta value helps.
Den fre. 8. mar. 2019 kl. 22.02 skrev Day, Zacharie :
> Hello,
>
> A user on my cluster is trying to process something with Siesta 4.0.2
> TBTrans_rep. The process works up
You can't, per see. :)
TranSiesta allows one to remove periodic boundary conditions by adding a
semi-infinite direction where a truly bulk semi-infinite electrode is
placed.
Siesta, however, intrinsically solves the potential in a periodic box.
Den tir. 26. feb. 2019 kl. 22.05 skrev Mina
As the error message states:
One of your variables has unit eV but expects Ry/Bohr
This is your error: (a force is not eV, rather Ry/Bohr or its other
variants, eV/Ang)
MD.MaxForceTol 0.04 eV
Den man. 25. feb. 2019 kl. 22.01 skrev Mina Sedighi :
> Dear Siesta users and developers,
>
>
> I am
Many of them *should* be very small. There are floating point round-off
errors which makes it very hard to reproduce exactly the same numbers.
So you should look at which quantities are un-equal, and if they are, are
they within a respectable difference.
Den man. 25. feb. 2019 kl. 22.03 skrev
Besides encountering the error that I mentioned in
> my last email, I receive the following warning for some atoms:
>
>
> siesta: WARNING: Atoms "A" "B" too close: rij =0.00 Ang
>
>
> Thank you for your advice.
>
>
> Best regards,
>
> Mi
This may happen due to two atoms getting too close.
Could you check your atomic coordinates at the given MD step?
Den man. 18. feb. 2019 kl. 22.12 skrev Mina Sedighi :
> Dear Siesta users,
>
>
> I perform a parallel relaxation simulation with siesta version 4.1-b3 on
> cluster. I face the
Probably the electrode TSHS files are not present in the directory.
Please ensure you follow the correct steps of a transiesta calculation:
1) Calculate each of the electrodes with transiesta
2) Specify the electrode TSHS files in the device fdf file
3) Run the device system.
Den lør. 16. feb.
Dear Aleksandar,
This is not surprising, the referenced article does use the minimal
unit-cell for the fully periodic systems (3a and 3c). See the figure
text "Minimal unit-cells in the transverse transmission direction have been
used due to the periodicity.", it is also stated in the body text.
#WriteWaveFunctions .true.
is commented out. Remove the # and it should work. :)
Den man. 11. feb. 2019 kl. 22.19 skrev Xiaoning Zang <
xiaoning.z...@kaust.edu.sa>:
> Dear Siesta users,
>
> I need to get the real-space wave functions of molecules using denchar
> which requires the output file
Since Siesta is a code using periodic boundary conditions any atomic
coordinate translated by a lattice vector is an equivalent system.
You can easily try this yourself by calculating two different structures
with one of them having a single atom displaced by a lattice vector.
Den ons. 6. feb.
Could you please use Siesta 4.0 as that is the latest stable release, see
if that also seg-faults.
If you still see seg-faults you should try and compile siesta/tbtrans with
modified flags for debugging purposes, see here:
https://answers.launchpad.net/siesta/+faq/2779
Den ons. 6. feb. 2019 kl.
ick:
> Thank you very much for your reply. I can not understand your
> explanation well. Would you like to give me an example using openmp or mpi?
> Thank you again for your kind help.
> Sincerely
> Ren-hui Zheng
>
>
>
>
>
>
> From: Nick Papior
> Date: 2019-02
These messages are written to std-out and std-err.
Simply do:
siesta ... 2>/dev/null &
If you want to suppress those messages from stderr while running in the
background.
Den man. 4. feb. 2019 kl. 22.04 skrev I. Camps :
> Hi SIESTers,
>
> I would like to know how to avoid the output from
Dear Ren-hui Zheng,
Siesta can benefit from MPI, OpenMP or both.
The benefit may be briefly summarized as follows:
MPI has distribution across orbitals, hence a basic advice is that you get
performance scaling with # of cores == # of atoms (rule of thumb)
OpenMP also distributes across orbitals,
I apply this command for all involved molecules?
>
>
> Best,
>
> Mina
> --
> *From:* siesta-l-requ...@uam.es on behalf of
> Nick Papior
> *Sent:* January 23, 2019 5:13:06 PM
> *To:* siesta-l
> *Subject:* Re: [SIESTA-L] rigid | molecule comma
Note this is only implemented in the 4.1-beta releases.
The rigid specification goes in the Geometry.Constraint block:
%block Geometry.Constraints
rigid [1 -- 3]
rigid [4 -- 10]
%endblock
"average" the force on atoms 1-3 and move 1-3 equivalently.
"average" the force on atoms 4-10 and move
101 - 200 de 893 matches
Mail list logo