[gmx-developers] Bug with continuation from checkpoint with gromacs 4.0.7

[Mark Abraham]

On 17/03/2010 7:59 PM, Alexey Shvetsov wrote:
> On Среда 17 марта 2010 10:47:50 Mark Abraham wrote:
>> On 17/03/2010 6:39 PM, Alexey Shvetsov wrote:
>>> On Вторник 16 марта 2010 05:45:22 Mark Abraham wrote:
>>>> On 16/03/2010 10:21 AM, Alexey Shvetsov wrote:
>>>>> On Вторник 16 марта 2010 01:45:42 Roland Schulz wrote:
>>>>>> Hi,
>>>>>>
>>>>>> what is nsteps in the mdp file?
>>>>>>
>>>>>> Did the simulation up to this point really did 3377000 steps?
>>>>>>
>>>>>> Roland
>>>>>
>>>>> nstep is 50000000 (10ns with 2fs timestep)
>>>>
>>>> No. Like I suggested, the run you asked for with this .tpr is already
>>>> finished.
>>>>
>>>>>>>>> starting mdrun 'Protein in water'
>>>>>>>>> 500000 steps,   1000.0 ps (continuing from step 3377000,   6754.0
>>>>
>>>> Mark
>>>
>>> That looks strange. Since after restart from same tpr its already reached
>>> ~7ns. How can i check number of steps in tpr?
>>
>> Do the same as I suggested for checking your .cpt... use gmxcheck!
>>
>> Mark
>
> gmxcheck output for previos crashed and uresumable run
>
> # Atoms  390476
> Last frame         -1 time 6741.000
>
>
> Item        #frames Timestep (ps)
> Step             1
> Time             1
> Lambda           1
> Coords           1
> Velocities       1
> Forces           0
> Box              1
>
>
> gmxcheck output for new run with same tpr for last checkpoint
>
> # Atoms  390476
> Last frame         -1 time 7030.000
>
>
> Item        #frames Timestep (ps)
> Step             1
> Time             1
> Lambda           1
> Coords           1
> Velocities       1
> Forces           0
> Box              1

Great. I've no idea what these output mean. :-)

Mark