Skip to content

Releases: MRChemSoft/mrchem

MRChem-v1.1.4: Bugfix release

17 Jan 13:03
8196932
Compare
Choose a tag to compare

Fixed

  • CRITICAL: sign error in para contribution to magnetic properties

MRChem-v1.1.3: Bugfix release

23 Aug 10:25
Compare
Choose a tag to compare

Fixed

  • Compatibility issue with Python-3.12

MRChem-v1.1.2: Bugfix release

20 Jan 08:51
b6e3564
Compare
Choose a tag to compare

Fixed

  • Updated MRCPP to v1.4.2 (fixing compilation issues)

MRChem-v1.1.1: Bugfix release

02 Dec 17:59
744b3e5
Compare
Choose a tag to compare

Fixed

  • Use dynamic threshold in 0th iteration (initial guess) SCRF
  • Bundling issue with pyparsing in input parser
  • Bug in CUBEfile parser

MRChem-v1.1.0: Minor release

26 Sep 13:08
87ec995
Compare
Choose a tag to compare

Added

  • Solvent effects through cavity-free PCM
  • Geometric derivative
  • Geometry optimization through QCEngine (beware of numerical instabilities)
  • cube initial guess: read generic CUBE files as starting orbitals
  • sad_gto initial guess: using 3-21G AO basis instead of Hydrogen functions
  • SAD input files for 5th row elements (Rb-Xe)
  • Tunable physical constants through user input

Changed

  • Updated MRCPP to v1.4.1
  • Improved Exchange efficiency and scaling
  • Improved MPI parallelization

MRChem-v1.0.2: Bugfix release

14 Oct 07:59
b5fbff5
Compare
Choose a tag to compare

Changed

  • Updated MRCPP to v1.4.0
  • Updated Eigen to v3.4.0

Fixed

  • Compile error with Eigen-3.4 due to conversion std::array<double> -> Eigen::VectorXd

MRChem-v1.0.1: Bugfix release

04 Dec 15:17
9309e3e
Compare
Choose a tag to compare

Changed

  • Moved CI from Travis to GitHub Actions
  • Abort on failure when reading MW guess

Fixed

  • Minor issues with the printed output
  • Error in NMR tensor output (dia term printed transposed)
  • Precision issue with CHK guess (now full world_prec is used)
  • MPI dead-lock when mixing shared-mem potentials with orbital bank
  • SAD guess for large systems (Eigen threading error in matrix multiply)
  • CORE/SAD guess for heavy atoms (force refinement for sharp s-functions)

MRChem-v1.0.0: Major release

28 Oct 10:22
9f5fff1
Compare
Choose a tag to compare

We are proud to announce the first major release of MRChem!
We added many features, fixed bugs, and improved overall usability of the software.

Features

MRChem can be used for:

  • Hartree-Fock
  • Kohn-Sham DFT (LDA/GGA/hybrid)
  • Restricted (closed-shell) and unrestricted
  • External electric field
  • Ground state energy
  • Dipole moment
  • Quadrupole moment
  • Polarizability (static/dynamic)
  • Magnetizability
  • NMR shielding
  • Density/orbital plots

Usability

The build system and launcher scripts have been rewritten from scratch to ensure
a smooth and productive user experience. The code can be compiled from source
or installed through popular package managers, such as Spack, Conda, EasyBuild.
MRChem can also be used with MolSSI QCEngine and as a Singularity container.

Contributors

Radovan Bast (@bast), Magnar Bjørgve (@bjorgve), Roberto Di Remigio (@robertodr),
Luca Frediani (@ilfreddy), Gabriel Gerez (@Gabrielgerez), Stig Rune Jensen (@stigrj),
Jonas Juselius (@juselius), Rune Sørland Monstad (@runesorland), Peter Wind (@gitpeterwind).

MRChem-v1.0.0-alpha3: Pre-release

10 Oct 15:36
Compare
Choose a tag to compare
Pre-release

Added

  • Dagger terms in exchange hessian
  • Screening of exchange contributions

Changed

  • Updated MRCPP to v1.3.5
  • Updated XCFun to v2.1.0
  • Updated parselglossy to v0.7.0
  • Removed runtest dependency
  • Removed parselglossy as runtime dependency
  • Encapsulated MRChem and MRCPP parallelizations

Fixed

  • Installation path for SAD files
  • Misc CMake fixes to enable packaging

MRChem-v1.0.0-alpha2: Pre-release

29 Jun 22:39
Compare
Choose a tag to compare
Pre-release

Added since alpha1

  • New JSON output

Changed since alpha1

  • Improved mrchem launcher script
  • Improved documentation
  • Input keywords for v1 defined and fixed
  • Updated MRCPP to v1.2.0
  • Updated XCFun to v2.0.1
  • Updated nlohmann/json to v3.6.1