Copyright (C) 2010-2014 ABINIT group (XG)
This file is distributed under the terms of the GNU General Public License, see
~abinit/COPYING or
http://www.gnu.org/copyleft/gpl.txt .
For the initials of contributors, see ~abinit/doc/developers/contributors.txt .
Many thanks to the contributors to the ABINIT project between
June 2013 and January 2014. These release notes
are relative to improvements of ABINITv7.6 with respect to v7.4.
The list of contributors includes :
B. Amadon, G. Antonius, T. Applencourt, A. Berger, J.-M. Beuken, J. Bieder, F. Bottin, J. Bouchet, F. Bruneval, D. Caliste,
M. Delaveau, B. Dorado, C. Espejo, A. Garcia, L. Genovese, M. Giantomassi, Y. Gillet, X. Gonze, N. Holzwarth,
F. Jollet, G. Jomard, J. Junquera, A. Martin, C. Martins, Renato Miceli,
Takeshi Nishimatsu, S. Ponce, Y. Pouillon, T. Rangel, M. Stengel, M. Torrent, M. Verstraete, J. Wiktor, B. Xu
It is worth to read carefully
all the modifications that are mentioned in the present file,
and examine the links to help files or test cases ...
This might take some time ...
Xavier
__________________________________________________________________
Version 7.6, released on January 13, 2014.
Changes with respect to version 7.4 :
A. WARNINGS.
B. Most noticeable achievements (for users)
C. Changes in the package, for developers
D. Other changes (or on-going developments, not finalized)
__________________________________________________________________
A. WARNINGS AND IMPORTANT REMARKS
(nothing for this version)
__________________________________________________________________
B. Most noticeable achievements
B.1 A large set of PAW atomic data files, covering 71 elements (=all except lanthanides or Z>85, with nevertheless Rn and Lu),
with the GGA PBE functional, has been generated and tested.
See http://www.abinit.org/downloads/PAW2/JTH-TABLE/index.html .
These atomic data files have been validated in ABINIT against all-electron calculations using the delta package, http://molmod.ugent.be/deltacodesdft.
A paper has been written, and is available on arXiv : http://arxiv.org/abs/1309.7274 .
These files are written in the XML format (that can also be read by GPAW).
In the above-mentioned paper, this new JTH table is also compared with two other sets of PAW atomic pseudopotentials,
from GPAW, and also from K.F. Garrity, J.W. Bennet, K.M. Rabe, D. Vanderbilt, arXiv:1305.5973v1
See the Test v7#64 . Other tests using the PAW XML format : tests v7#60, 61, 63.
Work by F. Jollet, M. Torrent and N. Holzwarth.
B.2 The Dynamical Mean Field Theory (DMFT) with Continuous Time Quantum Monte Carlo solver is available and fully parallelised.
New input variables : dmft_t2g dmftctqmc_check dmftctqmc_correl dmftctqmc_gmove dmftctqmc_grnns dmftctqmc_meas dmftctqmc_mrka
dmftctqmc_mov dmftctqmc_order dmftqmc_l dmftqmc_n dmft_seed dmftqmc_therm
See also input variables usedmft and dmft_solv
New tests v7#27, 28, 29 and paral #92.
Work by J. Bieder and B. Amadon
B.3 Introduction of the new input variable usepotzero, allowing to change the definition of the average of the
local potential. The convention of PW-SCF can now be used (as well as the usual convention of ABINIT).
The cross validation of the results between the two codes can be made much stronger than previously, especially
in the charged supercell case.
A paper has been accepted in Phys. Rev. B :
"Consistent treatment of charged systems within periodic boundary conditions:
The Projector Augmented-Wave and pseudopotential methods revisited"
See the test v7#26 .
F. Bruneval, JP Crocombette, X. Gonze, M. Torrent and F. Jollet
B.4 The parallelism on the atomic site, in case of PAW, has been finalised, both in the case of ground-state calculations
and DFPT (dielectric tensor and phonons - not Z*).
See test mpiio#62
Work by M. Delaveau and M. Torrent
__________________________________________________________________
C. Changes for the developers
C.1 A new nightly bot has been added : milou_g95_snofbfpe
This bot is activated with a whole set of debugging options, -ffpe-trap=invalid,zero,overflow -finit-real=snan
allowing to trap invalid operations, divide-by-zero, overflows. Activated nightly.
The accompanying cleaning of the code has been performed, so that all tests(non-parallel) execute now without detecting any error of that kind.
Other bots (no nightly) : tikal_intel14 ( with openmpi 1.6.5 )
Work by J.-M. Beuken
C.2 PLASMA and ELPA are now detected automatically by the build system.
In parallel, some tests are going on for multi-core architecture, and some adaptation on MIC (e.g. nonlop.F90).
Work by M. Torrent
C.3 The doc subsystem has been enabled.
Work by Y. Pouillon
C.4 There has been a deep cleaning of exception treatment.
E.g. Now, the "PERS" mode is used for WARNINGs, COMMENTs and ERRORs, while leave_test is not used anymore,
and abinit_comm_leave is not used anymore (now we **always** call MPI_ABORT is something goes wrong).
Work by M. Giantomassi
C.5 The creation of sections of output written in YAML (see e.g. http://en.wikipedia.org/wiki/YAML ) has started
The build system supports YAML, and The YAML library has been integrated in the build system (TRIO connector for YAML)
Work by M. Giantomassi and Y. Pouillon
__________________________________________________________________
D. Other changes
(or on-going developments, not yet finalized).
D.1 Introduction of the new input variable "accuracy", currently under tuning.
See test v7#64
Work by F. Jollet
D.2 Work on the autoparal of the GW code
The new input variable max_ncpus has been created, allowing to diagnose the most efficient proc distribution.
(same meaning as -paral_kgb but more general and safer)
See test tutoparal#gspw_01
Work by M. Giantomassi
D.3 Work on the zero-point motion and temperature dependence of the gap (and electronic structure) is going on :
- introduction of the new input variable use_nonscf_gkk , to allow correct and fast calculation of the contribution
of the k points outside of the irreducible part of the Brillouin zone ;
- parallelisation, see test paral#60
Work by S. Poncé
D.4 Work on the VdW-DF and VdW-WF function is going on :
introduction of several new input variable (vdw_df_damin, vdw_df_ngpts, vdw_df_phisoft, vdw_df_rsoft, vdw_df_tolerance, vdw_df_tweaks),
see test vdwxc#03 .
Corrections for the VdW-WF .
Work by Y. Pouillon + C. Espejo
D.5 PAW + Bigdft project
The ABINIT PAW library (libpaw) is in progress, see src/42_libpaw.
It contains:
- all the PAW object definitions and associated methods and functions,
- all the tool to read the PAW datasets (pseudo potential files)
It is now possible to perform wavelet-based PAW calculations ; the sequential
and parallel versions are available. However this feature is in beta stage.
It is also possible to follow the ABINIT path for the potential/density mixing, noticeably
with the DIIS algorithm, both with norm-conserving pseudopotentials and PAW atomic data.
See tests bigdft#33 and 34, and bigdft_paral#01 and 02
Work by T. Rangel + M Torrent, with the help of Luigi Genovese, Damien Caliste, and J.-M. Beuken
D.6 The partial DOS can now be computed in case of spinorial calculations.
Test v7#06 exercices this capability, and also address the case of spin DOS.
Work done by M. Verstraete
D.7 On-going efforts for the computation of elastic tensor within PAW.
Work done by A. Martin and M. Torrent
D.8 On-going efforts on the development of Hartree-Fock.
Hartree-Fock calculations are now coded within Norm-Conserving PSP formalism,
with the use of symetries and parallelism (not included in v7.6, though - developments to be merged in v7.7).
Still in the testing phase.
PAW + Hartree-Fock under development
From C. Martins (helped by M. Torrent and B. Amadon)
D.9 On-going efforts for the computation of U within RPA. The current implementation is restricted to static U, for non entangled correlated bands.
Work by B. Amadon, and T. Applencourt.
D.10 On-going work on PAW+DFPT : improving the stability, the parallelism has been fully tested and improved.
See test mpiio#62
Work by M. Torrent and B. Dorado.
D.11 Work on the use of the PLASMA library.
by M. Torrent.
D.12 Continued efforts on the development of the abipy library
Soon to be alpha tested (watch the anouncement on the forum).
From M. Giantomassi
D.13 Allow for the use of different ngfft in DFPT and GS.
From M. Torrent
D.14 On-going work on the computation of electron-phonon based transport properties.
By B. Xu and M. Verstraete
D.15 Work towards k-centred G-spheres
By A. Berger (+ M. Giantomassi)
D.16 Reading of NC pseudopotential common with SIESTA, using a XML format (with FOC as library for reading).
See tests fox#01 and 02
A paper is currently under writing.
By A. Garcia, J. Junquera, M. Verstraete, X. Gonze
D.17 Doppler broadening of positron annihilation spectroscopy
Sequential version is available with full PAW. The parallel version in progress
Gives information on the electron momentum distribution in the sample = defect signature
Work by M. Torrent, Julia Wiktor, Gerald Jomard
D.18 Post-process : Response function from Molecular Dynamics
To be included in APPA
Phonon spectrum obtained from trajectories (with a minimal set of coeffs)
Work by Francois Bottin, Johann Bouchet, and Alexandre Martin
D.19 Output of phonon isosurface is working
Should provide a test ?
By Bin Xu
D.20 Numerous miscellaneous additional bug fixes (to the sources, as well as to the build system,
including patches for the fallbacks), and improvements of documentation by :
B. Amadon, G. Antonius, J.-M. Beuken, J. Bieder, F. Bruneval, C. Espejo,
M. Giantomassi, Y. Gillet, X. Gonze, F. Jollet, Renato Miceli,
Takeshi Nishimatsu, Y. Pouillon, T. Rangel, M. Stengel, M. Torrent, M. Verstraete, B. Xu