[gmx-users] RE: logfile size

[Javier Cerezo]

Not the main problem, but your dt=0.0002 (0.2fs), is it what you want?

Javier

El 06/12/11 15:59, Mark Abraham escribió:
> On 7/12/2011 12:12 AM, Mark Abraham wrote:
>> On 6/12/2011 8:48 PM, lloyd riggs wrote:
>>> Dear All,
>>>
>>> Hope this is not tooo stupid a question.  I recenttly reduced my 
>>> file sizes by increasing the time between writing.  However, if I do 
>>> an Umbrella run I get the following to my logfile output, which 
>>> makes my log file in the end larger than my trajectory;
>>>
>>> ....
>>>    Dispersion correction          V -5.49759e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49756e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49753e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49750e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49747e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49744e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49741e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49738e+03  dVdl  0.00000e+00
>>>    Dispersion correction          V -5.49735e+03  dVdl  0.00000e+00...
>>>
>>> Does anyone know how to turn this off for the logfile, I get one of 
>>> these for each step, instead of for each write step?  My write 
>>> portion of the .mdp file is;
>>>
>>> nsteps      = 20000000     ; 2 * 50000 = 100 ps
>>> dt          = 0.0002     ; 2 fs
>>> tinit       = 0
>>> ; Output control
>>> nstxout     = 100000      ; save coordinates every 0.2 ps
>>> nstvout     = 100000       ; save velocities every 0.2 ps
>>> nstenergy   = 100000       ; save energies every 0.2 ps
>>> nstcalcenergy  = 100
>>> nstfout     = 100000
>>> nstlog      = 100000       ; update log file every 0.2 ps
>>> nstxtcout   = 100000
>>>
>>
>> Check out the list of options in mdrun -h. There's one that will 
>> completely stop this printing.
>>
>> I think it's supposed to print this stuff only according to nstlog, 
>> but the code is slightly broken here. I'll suggest a fix.
>
> Nah, the code looks fine - it should only ever print at nstlog steps.
>
> Mark

-- 
Javier CEREZO BASTIDA
Ph.D. Student
Physical Chemistry
Universidad de Murcia
30100, Murcia (SPAIN)
T: (0034)868887434