ABINIT Discussion Forums

1) The tutorial does not activate the transport part. Check test files v5-85-90 or so, and the _tr input variables.

2) eph gets you the electronic thermal conductivity. For TiC this will be 0 if it's an insulator, and a large part of the full kappa comes from the phonon heat conductivity.

Update many years later: the lattice vectors in the output WFK file are always reset to the initial ones, even after relaxation. This is ugly, but needed because the lattice vectors are set by the initial R and G vectors, and are not reallocated on the fly. The number of G and their distribution in ...

This post is in the wrong forum (response functions here) In general you need to give us more info, see the nettiquette page. Also: 1) you need to make sure everything is converged to the digits you show, wrt kpts and ecut 2) your pseudopotential may not be good enough, try oncvpsp potentials and th...

Please read the netiquette and get back to us with many more details. Which version of abinit which system are you in reduced dimensions? Basically it can not find a neighborhood of inequivalent kpoint neighbors to perform the berry phase strings. This used to happen if your cell is very anisotropic...

1) if the dos still goes to negative frequencies with a finite value (not just the tails of the gaussian) you have unstable q points elsewhere in the bz which f di not show up in the bs 2) in the tetrahedron case there is no smearing at all that's the point. As to why the dos does not change with th...

Hi all,

can we do alchemical mixing with PAW potentials as well?

Hi Again No precise suggestions short of running with one thread under gdb to catch the real error. I will not have time to look into this for some days. If you want to try: 1) set the code running in parallel in the background. 2) use ps aux | grep abinit to find the process ID (pid) of the main ex...

Not being printed may simply be because of a caching problem on the nodes you use, if the writing to the log file is not flushed properly as the code dies you do not have the last chunk of text and the error.

First thing: turn MSG_ERROR it into a MSG_WARNING and try in seq and parallel.

Hi again, I have tried your input with ecut 6 and ngkpt 1 1 1 and 2 2 2, and it does crash (even in sequential), perhaps for the same reason as you: what I get is an internal consistency check which fails, and the job stops. scfcv (electric field calculation) : WARNING - The difference between pel (...

A post-doctoral research fellowship is available starting immediately (Sept 2017) in the Nanomat group at the University of Liège, on the topic of micro and mesoscopic modelling of magneto-thermal properties. The aim is to use first principles methods to understand the coupling of vibrational and ma...

The D code has not been extensively used so there may well be authentic bugs in compatibility with certain features like paw or non collinear spin. In principle if your input is wrong the code should complain rather than proceed and die. How fast a calculation can you make by reducing ecut and nkpt ...

Hi Juhl, you are mixing lots of stuff: parallel, PAW, DFT+U calculations of the constant D. Could you try (with small ecut and nkpt) to see if things crash in sequential, without U, and for norm conserving, or with combinations of the three? If a small calculation with all options on crashes, then w...

Hello Chao, if you turn off elphflag, of course, it ignores the epc and goes through normal anaddb phonons. The code is complaining it does not find one of the elements in the GKK database you have provided. Could you check all of the GKK files are correctly included? Your input files seem alright t...

Things happen automatically, you don't need to add the input variables. In particular kpt and bands are distributed. You can check explicitly with an example: run the same quick input with 2,4,8 processors and you should see a speedup (unless it is really too small a job and the i/o and initializati...

PS: we have tried looking through the source code for inconsistencies. The calls to crystal_ncwrite looks ok, but some of the integer arrays are being miswritten from within anaddb (they are fine written out in the fortran on the fly) to netcdf. The type looks ok (int32 checked with the fortran kind...

@boates: the brav flags determine integration weights in q space, which are symmetrized properly for certain Bravais Lattices _and_ certain orientations of the lattice vectors (not just any unit cell will work - this is important as results will be worse if you choose brav/=1 and have the wrong orie...

rfmagn is under development - it does not work yet.

Yes - what I mean is: 1 kpt Gamma only is enough.

Hi Seba,

for an isolated system there is never any need for k-points.

yes, this is fine. Same for q, only 0 0 0 is needed. The default shiftk 0.5 0.5 0.5 is a pain, and you should indeed compensate it, but it's not a big deal even if you leave the default: all the bands will be flat if your vacuum is well converged.

You might write Valerie Veniard (Polytechnique Palaiseau) and Nicolas Dejean (now Max Planck Hamburg): in his phd they used "2light" for the SHG (also starting from abinit), and they will certainly have insight. etsf.polytechnique.fr/system/files/Thesis_Nicolas_Tancogne-Dejean_0.pdf If the...

Is the parallelization strategy changing? Grep for npfft npkpt etc... How many processors? Can you tell if the last unoccupied bands are well converged in the non scf calculation? In passing, your positions in xred are not very symmetric (not 1/3 2/3) - could be due to strain but needs to be checked...

Interesting, thanks for the bug post. It may be simply that you have to set kptopt 3 in the finite field case and a test clause is missing. I have asked more expert people and they should get back to you. Seems it's just the time reversal symmetry which is blowing things up, but even in your kptopt ...

Hi again, strange - it does look like a "quantum of polarization" effect in Berry phases, but again I don't think Berry is used, and anyway you are not looking at P. Could you check your outputs to see if the ngfft jumps at the corresponding chi2 jumps? Or the total energy or something els...

Not enough details about your calculation (read the nettiquette!!!), but on the face of it it looks ok: the nstep 1 calculation should only provide the matrix elements. The energy derivatives are certainly not converged with nstep 1 and the DDB should be obtained from a full scf DFPT calculation.