We are happy to announce the new MidasCpp 2025.10.0 release. Just follow the instruction in the manual or directly go to our GitLab page to clone the git repository or to download the source code.
Midas 2025.10.0
Changes have been made to some PES-keywords to make input less verbose in most cases.
See manual for full set of changes.
New Features
- The
midascppexecutable can now dump the contents of /proc/self/maps to stdout using the command-line option --dump-memory-map.
The memory map is dumped at the beginning of the calculation. - Adds an implementation of the Recursive TENsor expansion (RTEN). See manual for details.
- The following keywords were added under
#1 SinglePoint:#2 IoLevel,#2 DeterminantThr,#2 IgnoreRotationand#2 AllowImproperRotation.
See the manual for details.
Changes
- Removed keyword
#3 AddFitFuncsConvervand added keyword#3 AddFitFuncsScheme,
which can now be used to toggle betweenCONSERVATIVEandLEGACY.
CONSERVATIVEis the new default. - Added
#3 BasisTypeunder#2 Basis.BSPLINE-basis is now the default. Change basis by setting the#3 BasisTypekeyword.
Most, if not all, production calculations should use a BSPLINE basis. #2 DynamicAdgaExtis now on by default. Disable by setting#2 DynamicAdgaExttoFALSE.- PES will now automatically add a single multilevel, if no multilevels are given and only a single SinglePoint is defined.
This will allow users to skip the#2 MultiLevelkeyword under PES if only doing a single level. SP_GENERICwill now automaticallychdirto the specific SPC scratch-dir. This means one no longer has to
manually change the directory in theRunScript. TheRunScriptis still passed the scratch-dir as an argument, meaning that old
RunScripts which didchdirto the scratch-dir should still work as they do now.Kabschis now the default option for the keyword#2 RotationSchemeunder#1 SinglePoint.
Bugfixes
- Fixed an issue where only supplying a single integer to
#3 AnalyzeStates(i.e. analyze the same number of states for all modes) would make the program crash.
One can now supply a single integer which is then applied to all modes. - Fixed a bug in the LU-based determinant, det(A) = det(PLU).
Let us know if you encounter any problems.