ABINIT Discussion Forums

Dear antonio, I implemented the wfoptalg 114 (new LOBPCG version) (BTW wfoptalg 4 should not be used) My remarks : 1: all outputs give the same energy +/- 0.8 meV -> so I guess the calculation is correct (at least for the wave functions and the density) 2: the density are all converge pretty nicely ...

Hi, My opinion is that there is something with you MPI installation. 0) Remove optimization and scalapack 1) Check that the MPI you use was compiled with intel 17 and not an other version of intel nor gnu. 2) Check the optimizations of the MPI with your admin. There might be some variables used to a...

Study of energetic materials by first-principles methods. Theoretical development and applications. A postdoctoral position is available in France, at the "Commissariat à l'Énergie Atomique" (CEA, Bruyères le Châtel, south of Paris, France). We are interested here in the properties of ene...

Study of energetic materials by first-principles methods. Theoretical development and applications. A postdoctoral position is available in France, at the "Commissariat à l'Énergie Atomique" (CEA, Bruyères le Châtel, south of Paris, France). We are interested here in the properties of ene...

Abinit does not use complex for the most part of the code. We usually have arrays of the shape (2,*) with (1,*) the real part and (2,*) the imaginary part. Fortran does not check the type of the argument for pure subroutine. Therefore abinit arrays can be seen as 1D-complex array or 2D-real arrays i...

Hi In the version 8.0.8 you cannot calculate a partial DOS with full parallelization. So remove autoparal. You can only parallelized over k-points. (this is to be corrected in the next release) When the calculation of PDOS is done, you will have one file per atom with its partial DOS. You should be ...

Hi

Your latex installation misses the a4wide file.
You can install it or juste don't try to make the documentation.

Cheers

It means the total number of wave function coefficients is larger than the maximal value you can store in a "int" type, Namely, number of kpt*number of bands*number of spinors * number of plane waves is larger than 2^32-1 To handle such a calculation, you can 1) Recompile to user 64bit int...

Hi,

The typical use is
[code]
./abinit < tbase1_x.files > log &
[/code/

Note that "<" means read the standard input from the following file while ">" means send the standard output to the following file.
Does this work ?
Cheers.
Jordan

Thx JM !

Yes, you are completely right. That's why I wrote we have a new implementation of that routine that is much more efficient but still not very rich in capabilities. The new version of this routine is much faster in cpu time so it should answer to your question. and BTW, thanks for the feedbacks. They...

Hi there,

Do you know how to run a single test that is not numbered ?
Let say I want to run onlly [tutorespfn][trf2_3]. How do I do that ?

For a number test like [v7][t34] I would do


Ideads ?

Cheers

Jordan

What do you mean by "too much time" .... ?
We have a new implementation of this routine that is much faster but still limited in capabilities compared to opernlb.
We have been working on that point for several month.
Stay tuned

Hi,

You encountered a --too-- usual error in abinit when running in parallel. At the moment, there is no fix except changing by hand npband/bandpp/npfft instead of using autoparal.
Very soon, abinit will have a newer version of the algorithm causing this error which should then be avoided.

Cheers

Hi

Your acell is wrong.
Abinit expects bohr but you use angstrom. So you can change you acell to bohr (*0.52918) or add the word angstrom after the three parameters.

With such a small acell (3.89 bohr) your system is under high pressure and thus becomes metallic.

Cheers

Jordan

Hi,

It seems you did not define the variable acell which is the lattice parameters a b and c
For a perovkite like PbTiO3 you should have acell = 7.3 7.3 7.3 bohr (check litterature for a better value)

Cheers

Hi,

You should indeed compile with -lxcf90 -lxc flags together and use -L/path/to/libxc/libraries
During the abinit configuration step this is set using


Cheers

Hi, Depending on you cpu, your librairies, and so one, deviations are expected. It is hard to say if they're significant or not in you case since they are quite larger than the deviation extracted from our test farm. This being said, our test farm is not exhaustive so the expected deviation may be w...

Hi, Depending on the CPU you have on your localcomputer and on you cluster, this could be due to a vectorization issue with intel16. Could you try to build abinit with the --enable-avx-safe-mode on your cluster and try again. Also could you check if the energies and forces are the same the moment th...

Hi,

Just like this, I have no clue to help you with your error.
Could you try to use the latest 8.0.8b version of abinit ? www.abinit.org

Cheers

Jordan

You can use eivec=2 to print the eigen vectors. If you want a tool that directly reads the anaddb output you can use animemd in the abiout package : m Open the file output with anaddb and switch to mode phonon with the comman :mode phonons :add QX QY QZ IMODE where QX QY QZ are one qpt and IMODE the...

Hi,
t2g are 3 times degenerated and eg only 2 times. You should observe these degeneracies on the PDOS.
Cheers

Hi, This probably means you have the usual issue with the lobpcg algorithm which is automatically activated with paral_kgb. Changing the processors distribution changes the way the algorithm works and so avoid or produce an error. There is currently no cure available for public abinit, but hopefully...

Hi, Ok sorry, Could you check that in your first run, your fft grids (ngfft) at the end of the first run and at the beginning of the second are the same ? If not, add the corresponding first ngfft in your second run. Also, I see that you converge the energy (toldfe), you should try to converge the d...

Hi, Although this should not happen... you can try for simplicity to recompile abinit adding --with-trio-flavor="fox" when you do the ./configure step. Of course, if you already have --with-trio-flavor, just add fox like this : --with-trio-flavor="whatyouhave+fox" Cheers Jordan