[gmx-users] problems with output frequency and timestamps

[Grazia Daminelli]

Hi !

I am working with Gromacs 3.2.1 and I am having a problem with the 
output frequency during energy minimization.

This is my em.mdp file:

    title               =  energy minimization
    cpp                 =  /lib/cpp
    define              =  -DFLEXIBLE
    constraints         =  none
    integrator          =  steep
    nsteps              =  5000
    nstlist             =  10
    ns_type             =  grid
    coulombtype         =  Switch
    vdw-type            =  Switch
    rlist               =  1.1
    rcoulomb            =  1.0
    rvdw                =  1.0
    ;
    ;       Energy minimizing stuff
    ;
    emtol               =  100.0
    emstep              =  0.01
    ;
    ;        Output
    ;
    nstxout             =  50
    nstvout             =  50
    nstfout             =  50
    nstlog              =  100
    nstenergy           =  50


The calculation converged to the required emtol in 1939 steps.
When I check the trajectory with gmxcheck -f I get:

    Checking file traj.trr
    trn version: GMX_trn_file (single precision)
    Reading frame       0 time   64.000
    # Atoms  8997
    Reading frame      30 time 1939.000
    Timesteps at t=1897 don't match (64, 42)
    Last frame         30 time 1939.000


    Item        #frames
    Step            31
    Time            31
    Lambda          31
    Coords          31
    Velocities       0
    Forces          31
    Box             31


while gmxcheck -e generates a long list of :

    Timesteps at t=3 don't match (3, 1)
        ...................
    Timesteps at t=1933 don't match (2, 1)
    Reading frame   1530 time 1935.000
    Timesteps at t=1936 don't match (1, 2)
    Last frame read 1532

    Found 1533 frames.

now my questions:
1)with nst?out=50 and 1939 steps I would expect 39 frames in the 
trajectory (why do I get 31??)

2) what does it mean: Timesteps ... don't match??

3)why do I get no velocities in the output trajectory?

3)when I run g_rms I get 31 values (for 31 frames) with completly 
strange timestamps. Why??
     64
    127
    190
    253
    317
    379
    443
    506
    569
    632
    695
    759
    822
    885
    948
    1011
    1075
    1137
    1201
    1264
    1327
    1391
    1454
    1517
    1580
    1643
    1706
    1769
    1833
    1897
    1939

4) why do I get 1533 energy values when I run g_energy ??? Also in this 
case the timestamps are not to understand?


thanks for your help

Grazia