[gmx-users] parameters for output control

[Justin Lemkul]

On 8/20/12 1:15 PM, mohammad agha wrote:
> Dear Gromacs Users,
>
> I have one problem about the way of computation in the Gromacs with frame and time. (I work with MARTINI CG)
> I have one md.mdp file that the parameters of output control in it is as follows:
> nsteps = 30000000
> dt = 0.030
> ; Output control
> nstxout = 2000
> nstvout = 2000
> nstenergy = 2000
> nstlog = 2000
> nstxtcout = 2000
> xtc_precision = 1000
> Consequently, total time of my system is 900000 ps.When I want to obtaine some information about my system for example RDF by g_rdf command, computation of Gromacs perfome as follows:
> Reading frame 200 time 12000.000
> .................300
> .................400
> .................500
> .................600
> .................700
> ...........
> ...........
> ...........
> Reading frame 2000 time 120000.000
> .................3000
> .................4000
> .................5000
> .................6000
> .................7000
> ...........
> ...........
> ...........
> Last frame 15000 time 900000.000
> It means that until frame 2000 reading is 100 to 100 in frames and from 2000 to end, it is 1000 to 1000.

I don't understand this assertion.  You have frames every 60 ps, as expected. 
Console output is buffered so that the programs are not wasting time continually 
updating the frame number and time value.

> Also, I did with this parameters:
> ; Output control
> nstxout = 200
> nstvout = 200
> nstenergy = 100
> nstlog = 200
> nstxtcout = 200
> xtc_precision = 100
> but the result of reading the frames was the same previous.
> I want to Gromacs read frames 100 to 100 not 1000 to 1000, because I find problem when I use -b in g_rdf command. for eaxample, with -b 80000, reading is only until 860040 ps, Reading frame 13000 time 860040.000 ,not until end of simulation (900000).

This sounds like the same issue above.  The console output may be incomplete. 
It's merely a convenience to have an estimate of how long the calculation is taking.

> Also, I want to have smaller systems with smaller volume by convert 2000 to 10000 for parameters for output control but I want reading from frame for computation of properties become 100 frames to 100 frames or 10 frames to 10 frames.
>

I also don't understand this statement so I can't offer anything to particularly 
useful to help.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================