On Fri, Mar 18, 2016 at 3:59 AM, Sidhu, Khushwant (Dr.) <
k.sidhu.leicester.ac.uk> wrote:
> Hi,
>
> when I try to restart a simulation from a restart file (I’ve rerun &
> written multiple ones), I get the following in most of them
>
> cat tstar-25N_mdrun3.rst_39000000|grep "*"
> 109.8620092-180.0417570 558.6467279************ -94.5936925 39.7954818
> ************ -94.6608645 40.2439769************ -94.6238634 40.4988053
> ************ -94.6040732 39.9183133-103.9120614 501.3300167 -51.3830672
>
> My input file is as follows:
>
>
> production run protocol
> &cntrl
> imin=0, irest=1, ntx=5, ig=-1, ioutfm=1,
> nstlim=40000000, dt=0.002,
> ntc=2, ntf=2,
> ntb=2, cut=8.0,
> ntp=1, taup=2.0,
> ntpr =10000, ntwx=50000, ntwr=-50000, ntwv = -1
> ntt=3, gamma_ln=2.0,
> temp0=300.0,
> /
>
> cheers
>
If your simulation did not blow up, then this happens over long time
periods when particles diffuse far away. There are two solutions:
1) Use ntxo=2 to write NetCDF restart files (that way you can't overflow
the space allotted for coordinate files)
2) Set iwrap=1 to write wrapped coordinates (so the coordinates can't get
large enough to overflow the allotted space).
This has been discussed on the list many times before, so there is plenty
of information in the archives if you want to look at that more.
HTH,
Jason
--
Jason M. Swails
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Mar 18 2016 - 14:30:04 PDT
