Are MD parameters from xtb passed on to QCxMS?

[tobigithub]

Hi,
I was wondering if the MD parameters for XTB are passed on to QCxMS?
Do they have any influence when we add them to the qcxms.in file? That includes:

mdtemp 298.15 # MD thermostat/initial siman/GBSA temperature
mdtime 50.0 # MD run time in ps
mddump 1000. # dump (=optimize) structure in siman every mddump fs
mddumpxyz 50. # MD dump step in fs for xyz output file, MUST BE .eq. mdstep for power IR spectra
velodump 0 # set to 1 if dumps (trj file) should contain velocities
nvt 1 # use thermostat (=1, =0 for NVE)
mdskip 500 # skip interval in -mdav, -mdopt
mdstep 4.0 # MD time step in fs (automatically determined if < 0), could be 4-5 fs with shake =2, md_hmass=4
md_hmass 4 # increase hydrogen mass to this value in amu (at const. tot. mass) allowing large time steps (=0 off)
shake 2 # shake on (=0: off which is default) for XH bonds only (=1), or all bonds (=2)
nsiman 3 # number of siman annealing blocks
ewin_conf 20.0 # energy window (kcal) for considering conformers
tend 1000 # highest siman annealing temperature (very system specific)
enan 0 # include enantiomers in siman (=1)
````
Source: http://bbs.keinsci.com/thread-10056-1-1.html

Thank you!
T.

[JayTheDog]

Hi,
no, the QCxMS Input is handled completely different from the tblite library. GFN-xTB is only used to do single points with the given settings. You can, however, influence the MD by different parametes which can be found in the qcxms docs. I've just added some more keywords.
However, some of the settings you request do not make sense in the aspect of QCxMS (nvt, shake, ewin_conf), so it might not be of many use to implement this as a feature.

[tobigithub]

@JayTheDog
Thanks!
Tobias