# Replacing old mdp entry 'verlet-buffer-drift' by 'verlet-buffer-tolerance' Replacing old mdp entry '

**URL:** <https://gromacs.bioexcel.eu/t/replacing-old-mdp-entry-verlet-buffer-drift-by-verlet-buffer-tolerance-replacing-old-mdp-entry/11824>\
**Category:** User discussions\
**Tags:** mdp-parameters\
**Created:** [March 27, 2025, 7:12pm UTC](https://gromacs.bioexcel.eu/t/replacing-old-mdp-entry-verlet-buffer-drift-by-verlet-buffer-tolerance-replacing-old-mdp-entry/11824 "2025-03-27T19:12:35Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![salah](https://avatars.discourse-cdn.com/v4/letter/s/977dab/32.png) [@salah](https://gromacs.bioexcel.eu/u/salah)\
**Post date:** [March 27, 2025, 7:12pm UTC](https://gromacs.bioexcel.eu/t/replacing-old-mdp-entry-verlet-buffer-drift-by-verlet-buffer-tolerance-replacing-old-mdp-entry/11824/1 "2025-03-27T19:12:35Z")

</div>

GROMACS version:  
GROMACS modification: Yes/No  
Here post your question  
Replacing old mdp entry ‘verlet-buffer-drift’ by ‘verlet-buffer-tolerance’  
Replacing old mdp entry ‘nstxtcout’ by ‘nstxout-compressed’  
Replacing old mdp entry ‘xtc\_precision’ by ‘compressed-x-precision’

NOTE 1 [file grompp.mdp]:  
With Verlet lists the optimal nstlist is \>= 10, with GPUs \>= 20. Note  
that with the Verlet scheme, nstlist has no effect on the accuracy of  
your simulation.

WARNING 1 [file grompp.mdp]:  
The Berendsen thermostat does not generate the correct kinetic energy  
distribution, and should not be used for new production simulations (in  
our opinion). We would recommend the V-rescale thermostat.

---

<div class="post-metadata">

**Author:** ![obZehn](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/obzehn/32/2629_2.png) [@obZehn](https://gromacs.bioexcel.eu/u/obZehn)\
**Post date:** [March 28, 2025, 9:24am UTC](https://gromacs.bioexcel.eu/t/replacing-old-mdp-entry-verlet-buffer-drift-by-verlet-buffer-tolerance-replacing-old-mdp-entry/11824/2 "2025-03-28T09:24:51Z")

</div>

What is not clear specifically?

The replacing notes are just minor details regarding the change of wording of some keywords in the mdp file, e.g., back in the day it was `xtc_precision` and now it is `compressed-x-precision`. Probably you are using an old GROMACS mdp file, per se it’s not a problem as long as GROMACS will handle these internally.

The first note is telling you that `nstlist` can be larger as you are running on GPUs. Usually this value is overwritten to something \> 100 by GROMACS with a corresponding rescaling of the `rlist`, you can see this in the output on the screen and on the `.log` file of the simulation.

The warning is importan: never use `berendsen` neither as thermostat nor as barostat. Back in the day it was okay to switch between `berendsen` and `nose-hoover` for thermostats and `berendsen` and `parrinello-rahman` for barostats when going from equilibration to production, but nowadays it is generally better to use the newer `v-rescale` and `c-rescale` which are very robust for equilibration and produce the correct ensembles for production.