Here fixed bugs in Chemcraft are enlisted; more exactly, these are only bugs which are not always revealed immediately by the user, and which can theoretically lead to publications of wrong results:

06/27/2025, b772:  -   (Important) we found a new bug with visualization of MOs from Orca output files. If the basis set description contained more than 9 contracted Gaussians for an atomic orbital, Chemcraft read only the first of them, and this could lead to drawing wrong pictures of MOs. This problem occurred with big basis sets and transition metals; it could be revealed by the user with “Check orbitals” button. Probably this problem led to drawing wrong pictures very seldom, because the first contracted gaussians are usually the most important. To avoid such cases in future, possibly the users should sometimes build at least 1-2 cubes of MOs of their molecules for checking the correctness of Chemcraft pictures (cube files are always visualized properly) before publicizing the pictures of MOs. To compare cubes produced by Chemcraft and by other programs, you can press the “Explicit coordinates” button in the MOs generator window and paste the coordinates from the .cube file.

The version b722 is available with "Newest Windows version of Chemcraft" link at Download page.

05/16/2025, b770:  -   We found a bug, or rather a flaw in previous versions of Chemcraft: they didn't compute the electronic entropy when calculating the thermodynamic functions like Gibbs energy. This didn't matter for molecules with multiplicity 1 (we checked only such molecules when comparing the results computed by Chemcraft and by other programs like Gaussian). Now we have implemented the computation of electronic entropy, and get a full match with Gaussian (and almost full match with ORCA, with some difference caused by using different algorithms for calculating the vibrational entropy);

   -   Fixed a bug with visualization of Priroda output files with relaxed PES scan: previously, if some scan steps weren't converged, they were not visualized by Chemcraft;

   -   Fixed a small bug with the option “Show MO energies diagram” for Orca files with restricted-open computations (e.g. ROB3LYP): previously the orbital occupied with a single electron was not marked as alpha-occupied with only one vertical arrow;

   -   Fixed a bug or a flaw of versions starting from b688 with visualization of some Orca output files with CASSCF computations and NEVPT2/QD-NEVPT2 additional jobs, containing multiple tables with identical names: because the spectrum tables in these files could contain different X values, Chemcraft could draw the spectrum improperly for graphs like "ABSORPTION SPECTRUM VIA TRANSITION ELECTRIC DIPOLE MOMENTS", as it reads only one X values column (i.e. Chemcraft thinks that the X values are always the same in all spectra). Current version b770 do not visualize the tables with different X values at all, it shows that they were not read properly.

04/11/2025, b763:

- Fixed a bug with the mass spectrum analyzer: previously it showed the graph of a peak of a fragment ion with errors, if this ion contained atoms like Be, F, Na, etc., which contain only one stable isotope. This bug didn't harm the main feature of this tool – enumeration of possible fragment ions;

- Not a bug, but an annoying problem: previous version (b762) showed by default the enantiomers in a wrong way (though this can be easily altered via “Tools/Preferences/View axes orientation”). Now we have fixed this.

03/13/2025, b762:

-  Fixed a small bug of the previous version, b748: when the vibrational frequencies were computed by Hessian, and when the normal modes were computed, Chemcraft used the values of displacements in mass-weighted coordinates, while in programs like Gaussian or Orca, the Cartesian coordinates are used for this; because of that, for molecules with heavy atoms, in the vibrational modes involving these atoms the Cartesian displacements (amplitudes shown by the Chemcraft animator) were very big. Probably this problem was almost unimportant for the overwhelming majority of cases;

-  Fixed a bug with visualization of TDDFT spectra from Orca5 output files containing also spin-orbit corrected absorption spectra: the latest version visualized such files improperly.

01/11/2024, b748:

- In previous versions, the option “Tools/Mass spectrum analysis tools/Calculate the mass of current molecule with custom distribution of isotopes” could work improperly for molecules containing atoms like Cl, if you select isotopes like 37Cl: Chemcraft calculated the mass of Cl as 0 for these cases, because there is also an isotope 36Cl with zero percentage. Probably this bug was easily revealed by the user if it occurred.

09/28/2024, b726:

 - Previous versions sometimes showed incorrectly the structures read from xdatcar files (for non-cubic cells);

 - With previous build 688, some .xyz files with multiple structures were not read correctly.

11/29/2023, b688:

-   Fixed a problem with visualization of Quantum Espresso files which contained lattice vectors with “non-standard” orientation (e.g. if A.y is nonzero).

03/21/2022, b622:

-  (IMPORTANT) We found and fixed a bug in previous versions with rendering MOs from Orca output files. MOs containing F or G functions were rendered improperly, because we did not know that Orca uses phase factors for some AOs (e.g. F+2 orbitals are equal in Gaussian and Orca while F+3 orbitals have opposite signs in these programs). We are sorry for this problem.

07/22/2021, b610:

-   Fixed a bug which caused problems when visualizing TD spectrum from large Orca output files (when the number of bands in the spectrum is bigger than 1000).

07/02/2015, b436:

   -   Fixed some bugs with visualization of Gaussian file with frequency computation with frozen atoms;
    -   Fixed some bugs with visualization of PDB files.