Sir,
But if something went wrong, then "over" is rather irrelevant, isn't it?
Yes. I am planning to rerun ifI dont get any solution.
> So your simulation started from some previous point and re-ran? I don't
> see how this would be possible given what you have been posting.
The overlap was
On 8/1/13 11:44 AM, Kavyashree M wrote:
Sir,
Thank you for the suggestion.. I will make use of it in
future runs. but now the simulation is over.
But if something went wrong, then "over" is rather irrelevant, isn't it?
There was some overlap of few ns in 8core machine.
so while checking t
Sir,
Thank you for the suggestion.. I will make use of it in
future runs. but now the simulation is over.
There was some overlap of few ns in 8core machine.
so while checking the rmsd plot there was a significant
difference in rmsd between 8 core and 64 core overlapped
region of simulation.
1 mo
On 8/1/13 3:42 AM, Kavyashree M wrote:
Dear Sir,
First trajectory - traj1.xtc (0 to 5... ns)
Second trajectory - traj2.xtc (5... to 25ns)
But there is no gap in between.
for concatenating -
$ trjcat -f traj1.xtc traj2.xtc -o trajf.xtc
For nujump -
$ trjconv -f trajf.xtc -s md.tpr -pbc nojum
Dear Sir,
First trajectory - traj1.xtc (0 to 5... ns)
Second trajectory - traj2.xtc (5... to 25ns)
But there is no gap in between.
for concatenating -
$ trjcat -f traj1.xtc traj2.xtc -o trajf.xtc
For nujump -
$ trjconv -f trajf.xtc -s md.tpr -pbc nojump -o trajf.xtc
I also have a case where I
Hi Kavya,
That shouldn be a problem. Please post your sequence of commands for
concatenating and further processing of the trajectory.
Cheers,
Tsjerk
On Thu, Aug 1, 2013 at 8:12 AM, Kavyashree M wrote:
> Dear users,
>
> I ran a simulation for 25ns. First 5ns in 8 core machine
> and late part
Dear users,
I ran a simulation for 25ns. First 5ns in 8 core machine
and late part in 64 cores. It ran without any problem. The
trajectories were concatenated, jumps are removed and
rmsd was calculated. But there was sudden jump in the
rmsd curve. Is it wrong to run simulations in different
cores
7 matches
Mail list logo