Elk Activity
Brought to you by:
jkdewhurst,
sangeetasharma
Activity for Elk
-
J. K. Dewhurst created a blog post
Elk version 10.6.11 released
-
Elk released /elk-10.6.11.tgz
-
Pankaj Verma posted a comment on discussion Elk Users
Dear Michael, You have used mommtfix where fixed moment to each atom (if required ) in a specific direction is assigned. So my question is that why we further need to give moment to each atoms in block where species is defined in last section of above code. I will be thankful for your suggestion. Regards Pankaj Verma
-
Trishu Verma posted a comment on discussion Elk Developers
Hello Peter, Thank you for replying. I also checked the same on different machine and there it is generating same KPOINTS.OUT files, as expected. Therefore, I also think think that you are right, it may be because of some compilation issue. The version of elk is "10.4.5" compiled with intel 2021.4.0's mpiifort. I am attaching the make.inc file for your reference. Once again thank you for your important suggestion. Regards.
-
Peter Elliott posted a comment on discussion Elk Developers
Hi Trishu, I tried your 2 input files and get the same kpoint grid for both. As you say, the kpoint grid shouldn't change if you just double your magnetic field. So that behavior shouldn't happen. I suspect the issue is with the compiler, but I don't have much knowledge about them. Could you give some more details on the version of ELK and the compilers and libraries you used. -Peter
-
It would be really helpful, if anyone have any clue or suggestion regarding the same. Regards.
-
Dear Sir/Ma'am, As per my understanding/knowledge, co-ordinates of k-points sampled should only depend upon crystal structure and symmetry, density of k-points(given by user) and if there is any shift or not. Please correct me if I am wrong. But I am getting sometime random total no. of k-points. I did just changed the external magnetic field and temperature. I am attaching, the elk.in and KPOINTS.OUT files for your reference. Is there any way to make elk use given KPOINTS.OUT instead of generating...
-
Hi All, I don't think there was a bug -- the DOS code hasn't changed in a long time. The IDOS is defined to be the TDOS minus the sum of the PDOS components. Regards, Kay.
-
Long Yu posted a comment on discussion Elk Users
Dear Jerry, Has this bug been fixed in the new version? Regards, Long
-
Markus posted a comment on discussion Elk Users
What are you comparing to? Is it FLAPW calculations or maybe some sort of pseudopotential calculation? The basis sets play a big role in the description of localized states. You can tweak your convergence parameters by just setting high or vhighq and see what happens. Easier than testing all parameters individually. Once you get what you expect, you can try to find the responsible parameter.
-
The x-axis is Hartree, so multiply by 27.211 to get eV. Note that in the spin polarized case, the DOS is in states / Hr / unit cell / spin! In the spin unpolarized case, the DOS is spin degenerate, so you get twice the number of states.
-
TOP GUN posted a comment on discussion Elk Users
DOS UNITS https://sourceforge.net/p/elk/discussion/897820/thread/f832f6b7/?limit=25#eb4a
-
Long Zhang modified a comment on discussion Elk Users
It's noted in the code " DOS units are states/Hartree/unit cell". But the energy (horizontal) axis in PDOS_Sss_Aaaaa.OUT plotted out to be like in eV. Is it really eV or Hartree?
-
Tried to adjust U with different values, those 4f states below Fermi level are always at around -6 eV or lower.
-
Tried to adjust U with different values, those 4f states below Fermi level are always very low at around -6 eV or lower.
-
Hi All, I am trying to calculate the ground state properties of ferromagnetic Gd, using dft+u. It has been studied before in the literatures, and it's expected to have occupied 4f states at around 4 eV below Fermi level. I tried to use ELK to reproduce the result but so far I am not successful. Can anyone tell me where I did wrong? The elk.in that I used is copied below. Thanks, Long ------------------------------ elk.in -------------------------------------- tasks 0 10 spinorb .true. spinpol .true....
-
Tried to adjust U with different values, those 4f states below Fermi level are always very low at around -6 Har or lower.
-
Hi All, I am trying to calculate the ground state properties of ferromagnetic Gd, using dft+u. It has been studied before in the literatures, and it's expected to have occupied 4f states at around 4 eV below Fermi level. I tried to use ELK to reproduce the result but so far I am not successful. Can anyone tell me where I did wrong? The elk.in that I used is copied below, and the attached picture is the PDOS I got. Thanks, Long ------------------------------ elk.in --------------------------------------...
-
Hi All, I am trying to calculate the ground state properties of ferromagnetic Gd, using dft+u. It has been studied before in the literatures, and it's expected to have occupied 4f states at around 4 eV below Fermi level. I can't reproduce the result using ELK. Can anyone tell me where I did wrong? The elk.in that I used is copied below, and the attached picture is the PDOS I got. Thanks, Long ------------------------------ elk.in -------------------------------------- tasks 0 10 spinorb .true. spinpol...
-
Zhiwei Li posted a comment on discussion Elk Users
Dear Kay, Thank you for your test. I made some further calculations using your settings and your results is reproduceable. However, when I turn on dft+u with the following set: dft+u 1 4 1 2 5 The magnetic moments will start from the c-axis direction and relax to the ab-plane direction in the end with about 10 T external field. Best regards, Zhiwei
-
It's noted in the code " DOS units are states/Hartree/unit cell". Just want to confirm so that means the horizontal axis in PDOS_Sss_Aaaaa.OUT is in Hartree? Thanks. Edit: it's confirmed.
-
Hi All, I am trying to calculate ground state properties of ferromagnetic Gd, using dft+u. The PDOS is expected to have occupied 4f states at around 4 eV below Fermi level. I have not been successful in reproducing this feature. Appreciate any help/suggestion to help to tune this calculation. The elk.in that I used is copied below, and the attached picture is the PDOS I got. Thanks, Long ------------------------------ elk.in -------------------------------------- tasks 0 10 spinorb .true. spinpol...
-
It's noted in the code " DOS units are states/Hartree/unit cell". Just want to confirm so that means the horizontal axis in PDOS_Sss_Aaaaa.OUT is in Hartree? Thanks.
-
(deleted)
-
Hi All, Is the energy unit (in horizontal axis) in DOS and PDOS eV or Hartree? I would like to confirm this but can't find it explicitly said anywhere. Thanks, Long
-
Hi All, Is the energy unit (for horizontal axis) in DOS and PDOS eV or Hartree? It looks like eV instead of Hartree. I would like to confirm this but can't find it explicitly said anywhere. Thanks, Long
-
CY ZHUNG modified a comment on discussion Elk Users
Using ifx on Fedora 42 with 4 cores laptop and optimized with -O2 level, all tests can passed. Because it only suport -O2 level optimization. The problem may be the compitable between elk and ifx, i think.
-
Using ifx on Fedora 42 with 4 core laptop and optimized with -O2 level, all tests also could passed. Because it only suport -O2 level optimization. The problem is compitable between elk and ifx, i think.
-
Using ifx on Fedora 42 with 4 core laptop and optimized with -O2 level, all tests also could passed. Because it only suport -O2 level optimization. The problem is compitable between elk and ifx i think.
-
Using ifx on Fedora 42 with 4 core laptop and optimized with -O2 level. all tests passed. because it only suport -O2 level optimization. I think it is compitable problem between elk and ifx.
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any suggestion. Assume I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo basis for example. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb...
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any discussion. Say I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb can be expanded in...
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any discussion. Say I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb can be expanded in...
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any discussion. Say I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb can be expanded in...
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any discussion. Say I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb can be expanded in...
-
Dear ELK users and developers, I am looking for instruction/idea to do a post-processing thing for a project of mine. Thanks in advance for any discussion. Say I have done a proper ground state calculation for some crystal structure ABO3, using APW+lo. I want to do an inner product of the eigen state, psi(k,n), with a free-atomic orbital Orb (e.g. s, px, py, pz .... ) at site A or B or O. Something very very close to free atomic orbital probably will do too. Anyway, if the Orb can be expanded in...
-
Hybrid OpenMP+MPI is usually faster than MPI alone for large systems. For small systems, I think thread creation and cache coherency can cause too much overhead for nested OpenMP to be worth it. Hopefully this will improve as compilers, kernels and architectures advance. Elk uses its own thread management scheme which limits the number threads to the total number of cores. This seems to be more efficient than using OMP_DYNAMIC=true. If your calculations are for large systems, then the compute time...
-
Olle Heinonen posted a comment on discussion Elk Users
Thanks, Kay. I am running now with OMP_NUM_THREADS=96 and so far the code runs (I recompiled with the the -fcheck compiler flags). I am not sure I understand why OMP_PROC_BIND=false is needed - on some nodes we have, the intersocket communications are slow and I usually want to confine batches of threads or tasks to one socket. I am also a bit mystified by why OMP_STACKSIZE=512M works. I had tried to set the OMP stack size to a much larger number (as my nodes have boatloads of RAM) but that did not...
-
Are you using -fcheck=all ? If so, then remove this option because it won't allow passing of unallocated arrays. You may be reaching the limit of 4-byte integers for addressing arrays. The maximum address is 2,147,483,647. For a real(8) array this would correspond to 16GB. You may also reach the limit of the size of arrays that MPI routines can transfer. We had this problem with BSE matrices and had to transfer them in smaller chunks. Some things you can try: a different compiler set lowq=.true.,...
-
Dear LongYu. You are correct: there is indeed a bug in the way the wavefunction plotting routine handles symmetry. I'll have this fixed for the next release. In the meantime, you can use an earlier version of the code to plot wavefunctions. Thanks and regards, Kay.
-
Dear Kay, The attachment is my input file. Thank you for your prompt reply! best LongYu
-
Hi Olle, Segmentation faults in Elk occur (almost always) when the code runs out of stack space. You'll need to set: export OMP_NUM_THREADS=32 export OMP_PROC_BIND=false export OMP_STACKSIZE=512M ulimit -Ss 524288 before running the code. Alternatively use the elk.sh script provided. The arrays taumt and tauir are not allocated except for meta-GGA calculations. We abuse the Fortran Standard by passing them unallocated to potxc where they are not referenced by the subroutine. This works for all the...
-
Hi Kay. Thanks for response. Sorry it took a while to find your response. I get the seg fault with ramdisk .false. and mixsdb=2. The error, as I just noted in a post a few seconds ago, seems to stem from taumt not being allocated: At line 39 of file potks.f90 Fortran runtime error: Allocatable actual argument 'taumt' is not allocated. I am not sure why this shows up - I am palying with LSDA or various GGA flavors for a spin-polarized system, no meta-GGA so I am not sure why this error shows up. I...
-
Hi. I am running elk on some simple systems that are spin-polarized and have spin-orbit coupling turned on. I get a segmentation fault sometimes (typically for larger systems, > 10-ish atoms) irrespective of using the gnu or Intel coompiler suites. Compiling with gnu flags -fcheck=all shows an error in output: At line 39 of file potks.f90 Fortran runtime error: Allocatable actual argument 'taumt' is not allocated This seems to be an array for kinetic energy density. I get the error irrespective of...
-
Hi Malte, Do you mean integrated along the other two dimensions? If so, then you can use tasks 172 and 173 to produce the once-integrated 2D and full 3D EMD plots. Regards Kay.
-
Hi LongYu, Could you post your input file? Thanks, Kay.
-
Hi Zhiwei, I ran tasks 0 highq .true. bfieldc 0.0 0.0 0.1 spinpol .true. spinorb .true. nosource .true. avec 13.35162285 0.4665733494E-12 0.000000000 -6.675811423 11.56284457 0.000000000 0.000000000 0.000000000 26.80242178 atoms 3 : nspecies 'Mn.in' : spfname 6 : natoms; atpos, bfcmt below 0.33333334 -0.33333331 0.00222096 0.00000400 -0.00000100 -0.00000040 -0.33333337 0.33333331 -0.00222099 0.00000400 -0.00000100 -0.00000040 0.33333331 -0.33333337 0.49777904 0.00000400 -0.00000100 -0.00000040 -0.33333331...
-
Dear all, Using the same input file, elk-9 ends normally, while elk-10.6.2 reports the following error Error(roteuler): matrix improper or not unitary Determinant : 0.000000000 If the developer can fix this bug, it would be greatly appreciated. best LongYu
-
-
Dear all, Calculated the example of GaAs in the program example (elk-10.6.2), but abnormal fluctuations occurred near 0 frequency. The calculation of other structures has the same problem. Is this a bug in the program? Any help will be appreciated. Best LongYu
-
Dear all, Calculated the example of GaAs in the program example, but abnormal fluctuations occurred near 0 frequency. The calculation of other structures has the same problem. Is this a bug in the program? Any help will be appreciated. Best LongYu
-
Dear all, Calculated the example of GaAs in the program example, but abnormal fluctuations occurred near 0 frequency. The calculation of other structures has the same problem. Is this a bug in the program? Any help will be appreciated. Best YuLong
-
Jerry Li modified a comment on discussion Elk Users
I’m trying to calculate the density of states (DoS), band structure, and optical properties of CsPbBr₃ using ELK 10.6.2 with GW+BSE, but the DoS output looks unusual—very noisy and spiky. Do you have any idea what might be causing this? I've attached my elk.in file for reference. Thanks! ~~~~~~~~~elk.in~~~~~~~~ tasks 0 ! ground-state calculation 10 20 22 620 120 ! compute momentum matrix elements 121 ! compute RPA dielectric function with no local field contributions 180 ! generate RPA dielectric...
-
I’m trying to calculate the density of states (DoS), band structure, and optical properties of CsPbBr₃ using ELK 10.6.2 with GW+BSE, but the DoS output looks unusual—very noisy and spiky. Do you have any idea what might be causing this? I've attached my elk.in file for reference. Thanks! ~~~~~~~~~elk.in~~~~~~~~ tasks 0 ! ground-state calculation 10 20 22 620 120 ! compute momentum matrix elements 121 ! compute RPA dielectric function with no local field contributions 180 ! generate RPA dielectric...
-
I’m trying to calculate the density of states (DoS), band structure, and optical properties of CsPbBr₃ using ELK 10.6.2 with GW+BSE, but the DoS output looks unusual—very noisy and spiky. Do you have any idea what might be causing this? I've attached my elk.in file for reference. Thanks!
-
I am trying to calculate DoS, band structure and optical properties of CsPbBr3 with ELK GW+BSE, but the DoS looks weird, very sparky.
-
I am trying to calculate DoS, band structure and optical properties of CsPbBr3 with ELK GW+BSE, but the DoS looks weird, very sparky.
-
Zixuan Li posted a comment on discussion Elk Users
Hello I'm a new user to elk. Recently I'm studying the TDDFT module. I notice that the peak power density is calculated through the formula t1*(w^2)/(8*pi*solsc), where t1 equals to the sum of the A_0^2 in three directions. I'm curious that should the peak power density be calculated through solsc*t1*(w^2)/(8*pi) ? Or if there some mistake in my derivation ? And I wonder where did the value of solsc is stored, I can't find it in the source file. Besides, could anyone tell me how to convert the elk's...
-
Test_017 passed all tests on my Ubuntu 22.04 with intel Oneapi 23.2 li@p720:~/DFT/Elk/tests/test_017$ elk.sh ┌─────────────────────────────────┐ │ Elk code version 10.6.2 started │ └─────────────────────────────────┘ OMP: Info #277: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead. Number of MPI processes : 1 Number of OpenMP threads per MPI process : 8 Total number of threads (MPI x OpenMP) : 8 Maximum OpenMP nesting level : 4 Number of OpenMP threads at first nesting...
-
Test_017 passed all tests on my Ubuntu 22.04 with intel Oneapi 23.2
-
Andrew Shyichuk modified a comment on discussion Elk Users
(deleted)
-
I've had this bug today with Elk 10.6.2 and libXC 6.2.2. The fact that Yuri had this with libxc 7.0.0 is peculiar. Anyway, I applied a dirty hack of removing the abusing subroutines completely, and at least it compiled. Consider the modified file with the parts commented out.
-
I've had this bug today with Elk 10.6.2 and libXC 6.2.2. The fact that Yuri had this with libxc 7.0.0 is peculiar. Anyway, I applied a dirty hack of removing the abusing subroutines completely, and at least it compiled. Consider the modified file with the parts commented out.
-
Dear Youzhao, Your make.ink simply disables LibXC, as the "SRC_LIBXC = libxcifc_stub.f90" line is not commented out (is active).
-
Good news. Bad news is Test_017 could not pass the test. I tried it on CachyOS, Fedora 42 and Ubuntu 25.2 with intel Oneapi 2025.2 set and -O3 optimzation level. one of the test.log is attached here.
-
Glad to hear it!
-
Andreas Fischer posted a comment on discussion Elk Users
This seems to be fixed for me with v10.6.2
-
Elk version 10.6.2 released
-
-
Hi ELK users and developers, I am trying to calculate the Mn3Si2Te6 compound under external magnetic field. To do this, I made the setting in my elk.in file bfieldc: 0 0 B with reducebf=1.0 B is the desired external magnetic field. My intention is to see the evolution of the noncollinear magnetic structure with B along the c-axis. Therefore, I used the source-free version of the code with PBE settings: nosource=.true. and sxcscf=1.14 Also, a small random field (much smaller than B) within the muffin-tin...
-
Hi Ellaine, Thank you for your suggestion. After I set rgkmax to 6.0 and gmaxvr to 20.0, the problem seemed to go away, but a new one appeared. "Error(eveqnfvr): diagonalisation failed DSYGVX returned INFO = 32600 The leading minor of the overlap matrix of order 8679 is not positive definite Order of overlap matrix : 23921" Do you have any idea? Many thanks Jerry
-
Dear Maam, Newer features are added in newer ELK versions. How does one get the idea about the associated "TASK Number" with each newer feature, "newer input parameters with newer features" and "ideal example/input" of that newer feature etc. It would be of great help to ELK users, if the ELK developers have a word of advice about the above. surinder On Fri, Jul 4, 2025 at 12:12 PM Sangeeta Sharma sangeetasharma@users.sourceforge.net wrote: Dear Cy and TopGun, Cy: Usually, if the computers are too...
-
Dear Lars and Sangeeta. Thanks a lot for your suggestions. Several weeks ago. I tried to optimize my study objects, several atoms with highly symmetry. Geo optimization of the structures calculated by Quantum Espresso and Castep were finished in one hour while that done by elk had not finished in a whole night. Follow your advices, i will retry vkloff with shift this weekend. Yours Sincerely!
-
Yes, it is reproduced on my workstation with compiling "F90_OPTS = -O3 -march=core-avx512" but on my dell notebook. all tests passed with compiling "F90_OPTS = -O2 ……" the speed will slow down when the code compiling with Flags with "-O2" ?
-
Dear Lars and Sangeeta. Thanks a lot for your suggestions. Several weeks ago. I tried to optimize my study objects, several atoms with highly symmetry. Geo optimization of the structures calculated by Quantum Espresso and Castep were finished in on hour while that done by elk had not finished in a whole night. Follow your advices, i will retry vkloff with shift this weekend. Yours Sincerely!
-
Sangeeta Sharma posted a comment on discussion Elk Users
Dear Cy and TopGun, Cy: Usually, if the computers are too small or the system size is too big, one can use special k-points in the BZ (Chadi-Cohen points or Monkhorst-Pack grids etc.) to perform the integral over the BZ. These k-points are special points which are representative of the zone. You can see some more details about them in this paper for example: https://journals.aps.org/prb/abstract/10.1103/PhysRevB.8.5747 If you do not want to use such special set of k-points, just set vkloff (k-point...
-
Dear Lars, Is the ELK manual fully updated ? surinder On Tue, Jul 1, 2025 at 11:27 PM Lars Nordström larsnordstrom@users.sourceforge.net wrote: Dear Cy, All "freshers" should read the manual, do you agree? I doubt you have read the manual distributed with the code and downloadable from homepage. In the manual you will find how the k-point grid is set up. If you are too lazy to do that as a courtesy I will give you a summary of theinformation you would find: there is a variable called VKLOFF that...
-
Lars Nordström posted a comment on discussion Elk Users
Dear Cy, Congratulations you have passed first level, to read the manual. Next level is to understand the manual. (I will give you a free hint: read the beginning of the manual) If you pass next level, you are allowed to be active in this forum. Until then, take care! /Lars
-
Dear Lars. Based on the explaination, the shift chould be turned off by setting Vkloff to 0, 0, 0 ? Is this right? This is not suggested in the manul. Yours Sincerly
-
Dear Lars. Based on the explaination, the shift chould be turned off by set Vkloff to 0, 0, 0 ? Is this right? This is not suggested in the manul. Yours Sincerly
-
Dear Cy, All "freshers" should read the manual, do you agree? I doubt you have read the manual distributed with the code and downloadable from homepage. In the manual you will find how the k-point grid is set up. If you are too lazy to do that as a courtesy I will give you a summary of theinformation you would find: there is a variable called VKLOFF that shifts the k-point mesh in a way that is common in special k-point approaches. "Vlkoff settings confuse and stop fresher to use elk in their future...
-
Dears. ELK use LAPW method with vlkoff settings. As a fresh elk user, I just just understand Kmesh density and do not know how to confirm the vkloff blocks. So, i create the topic. If set vlkoff: Vx, Vy, Vz, KPONTS: Kx, Ky, Kz Could Kmesh densty be determined through Vx/Kx, Vy/Ky, Vz/Kz ? And so, Can you develop one tool or calculator to make it easier to be understood and confirmed ? Vlkoff settings confuse and stop fresher to use elk in their future simulation. Could it been explain more in de...
-
Dears. ELK use LAPW method with vlkoff settings. As a fresh elk user, I just just understand Kmesh density and do not know how to confirm the vkloff blocks. So, i create the topic. If set vlkoff: Vx, Vy, Vz, KPONTS: Kx, Ky, Kz Could Kmesh densty be determined through Vx/Kx, Vy/Ky, Vz/Kz ? And so, Can you develop one tool or calculator to make it be easier to be understand and confirmed ? Vlkoff settings confuse and stop fresher to use elk in their future simulation. Could it been explain more in...
-
Yes, it is reproduced on my machine with compiling "F90_OPTS = -O3 -march=core-avx512"
-
Yes, it is reproduce on my machine with compiling "F90_OPTS = -O3 -march=core-avx512"
-
Dears. ELK use LAPW method with vlkoff settings. As a fresh elk user, I just just understand Kmesh density and do not know how to confirm the vkloff blocks. So, i create the topic. If sets vlkoff: Vx, Vy, Vz, KPONTS: Kx, Ky, Kz Could Kmesh densty be determined through Vx/Kx, Vy/Ky, Vz/Kz ? And so, Can you develop one tool or calculator to make it be easier to be understand and confirmed ? Vlkoff settings confuse and stop fresher to use elk in their future simulation. Could it been explain more in...
-
Dears. ELK use LAPW method with vlkoff settings. As a fresh elk user, I just just understand Kmesh density and do not know how to confirm the vkloff blocks. So, i create the topic. If sets vlkoff: Vx, Vy, Vz, KPONTS: Kx, Ky, Kz Could Kmesh densty be determined Vx/Kx, Vy/Ky, Vz/Kz ? And so, Can you develop one tool or calculator to make it be easier to be understand and confirmed ? Vlkoff settings confuse and stop fresher to use elk in their future simulation. Could it been explain more in detail...
-
Hi Kay, same setup but and libs, but with version 10.5.16 the following test-failures occur: Test 015 (MPI & non-MPI) ┌────────────────────┐ │ Current task : 500 │ └────────────────────┘ Info(elk): MPI process 1 idle for task 500 Info(elk): MPI process 3 idle for task 500 Info(elk): MPI process 2 idle for task 500 Error(testcheck): variable 18 outside tolerance for quantity 'Coupled tensor moments' TEST_820.OUT_ (correct value) : 0.144345801193E-02 TEST_820.OUT : 0.146266498838E-02 absolute difference...
-
Just discovered a minor bug in the current version of Elk (10.5.16): the sum of all the component PDOS and IDOS does not equal the TDOS. This issue wasn't present in earlier versions.
-
Youzhao Lan posted a comment on discussion Elk Users
Dear Yuri, I can link them successfully. My make.inc attached for you. CentOS 7.8, gcc 9.2, libxc 7.00 Best regards. Youzhao Lan
-
Hi Yuri, I'm not sure why that's happening. Could you try nm libxc.a | grep xc_func_set_fhc_enforcement nm libxcf03.a | grep xc_func_set_fhc_enforcement in the elk/src directory. Regards, Kay.
-
Yuri posted a comment on discussion Elk Users
Build breaks: /usr/local/bin/ld: libxcf03.o: in function __xc_f03_lib_m_MOD_xc_f03_func_set_fhc_enforcement': libxcf03.f90:(.text+0x50f): undefined reference toxc_func_set_fhc_enforcement' /usr/local/bin/ld: libxcf03.f90:(.text+0x51b): undefined reference to `xc_func_set_fhc_enforcement' collect2: error: ld returned 1 exit status Version: 10.5.16 libxc-7.0.0 FreeBSD 14.2
-
Hi, It seems that you are not converged w.r.t. the number of empty states. Elk sets up the second-variational Hamiltonian using the first-variational basis. The second-variational Hamiltonian includes the terms associated with magnetism. Thus you should increase nempty until the quantity you're calculating is properly converged. I used nempty 32 which produced a smooth band structure. Regards, Kay.
-
Thank you, Kay! Your team has done great work!
-
Dear All, Potential-only meta-GGA functionals (like Tran-Blaha '09) have been restored to Elk version 10.5.16. Libxc version 7 is required for this release. Regards, Kay.
-
Dear Jerry, Markus and Subrata, Potential-only meta-GGA functionals have now be restored in version 10.5.16. You can use Tran-Blaha '09 for gaps, see elk/examples/meta-GGA/Si-Tran-Blaha. If your system size is not too large, then the GW approximation is usually more reliable for accurate fundamental gaps. Regards, Kay.
-
Dear Olle, You can try disabling the RAM disk with ramdisk .false. This will free up some RAM. Reducing the size of the subspace dimension for Broyden mixing with mixsdb 2 may also help. Regards, Kay.
-
Elk version 10.5.16 released
-
-
Nagarjuna Patra posted a comment on discussion Elk Users
Dear Elk Developers and Users, I am using Elk's task=105 to compute the 3D nesting function for my system, and I successfully obtained the output files NEST3D.OUT and NESTING.OUT.Please verify if the input tags i have used is correct or not. I am attaching the input file for MgB2 as example. However, I couldn’t find any official documentation or examples that explain how to interpret or extract useful information from these files. Could you please help clarify the following points? 1.In NEST3D.OUT...
1 >