Hi,
Is this the same run than the one you mentioned in the first email of
this thread? (in that one, you mentioned the problem at NSTEP = 1300000).
OK, Can you re-run the same simulation (don't forget to change ig from
-1 to the number specified in mdout) for NSTEP=400000 and NSTEP=500000
and check/visualize the restart files? If the error is reproducible, you
should be able to see a problem at NSTEP=500000 (or 520000). What is
important is to see if this problem occur already before (NSTEP=400000).
Gerald.
On 04/03/2015 10:41 PM, Quynh Vo wrote:
> Hi, this is the output from grep command
>
> NSTEP = 20000 TIME(PS) = 1020.000 TEMP(K) = 297.48 PRESS =
> 0.0
> NSTEP = 40000 TIME(PS) = 1040.000 TEMP(K) = 300.58 PRESS =
> 0.0
> NSTEP = 60000 TIME(PS) = 1060.000 TEMP(K) = 298.74 PRESS =
> 0.0
> NSTEP = 80000 TIME(PS) = 1080.000 TEMP(K) = 300.06 PRESS =
> 0.0
> NSTEP = 100000 TIME(PS) = 1100.000 TEMP(K) = 299.26 PRESS =
> 0.0
> NSTEP = 120000 TIME(PS) = 1120.000 TEMP(K) = 313.25 PRESS =
> 0.0
> NSTEP = 140000 TIME(PS) = 1140.000 TEMP(K) = 296.79 PRESS =
> 0.0
> NSTEP = 160000 TIME(PS) = 1160.000 TEMP(K) = 299.04 PRESS =
> 0.0
> NSTEP = 180000 TIME(PS) = 1180.000 TEMP(K) = 299.51 PRESS =
> 0.0
> NSTEP = 200000 TIME(PS) = 1200.000 TEMP(K) = 294.61 PRESS =
> 0.0
> NSTEP = 220000 TIME(PS) = 1220.000 TEMP(K) = 302.24 PRESS =
> 0.0
> NSTEP = 240000 TIME(PS) = 1240.000 TEMP(K) = 298.35 PRESS =
> 0.0
> NSTEP = 260000 TIME(PS) = 1260.000 TEMP(K) = 295.77 PRESS =
> 0.0
> NSTEP = 280000 TIME(PS) = 1280.000 TEMP(K) = 299.57 PRESS =
> 0.0
> NSTEP = 300000 TIME(PS) = 1300.000 TEMP(K) = 300.06 PRESS =
> 0.0
> NSTEP = 320000 TIME(PS) = 1320.000 TEMP(K) = 298.38 PRESS =
> 0.0
> NSTEP = 340000 TIME(PS) = 1340.000 TEMP(K) = 295.67 PRESS =
> 0.0
> NSTEP = 360000 TIME(PS) = 1360.000 TEMP(K) = 296.26 PRESS =
> 0.0
> NSTEP = 380000 TIME(PS) = 1380.000 TEMP(K) = 299.82 PRESS =
> 0.0
> NSTEP = 400000 TIME(PS) = 1400.000 TEMP(K) = 291.37 PRESS =
> 0.0
> NSTEP = 420000 TIME(PS) = 1420.000 TEMP(K) = 294.48 PRESS =
> 0.0
> NSTEP = 440000 TIME(PS) = 1440.000 TEMP(K) = 300.86 PRESS =
> 0.0
> NSTEP = 460000 TIME(PS) = 1460.000 TEMP(K) = 300.35 PRESS =
> 0.0
> NSTEP = 480000 TIME(PS) = 1480.000 TEMP(K) = 298.76 PRESS =
> 0.0
> NSTEP = 500000 TIME(PS) = 1500.000 TEMP(K) = 296.49 PRESS =
> 0.0
> wrapping first mol.: -1412287.72681 1327010.77339 1006973.72210
> NSTEP = 520000 TIME(PS) = 1520.000 TEMP(K) =********* PRESS =
> 0.0
> wrapping first mol.: -3188244.88551 2988967.15637 2541362.54131
> NSTEP = 540000 TIME(PS) = 1540.000 TEMP(K) =********* PRESS =
> 0.0
> wrapping first mol.: -4726609.32026 4458801.91818 3929349.31811
> NSTEP = 560000 TIME(PS) = 1560.000 TEMP(K) =********* PRESS =
> 0.0
> wrapping first mol.: -6271732.30144 5886992.10717 5413198.12471
> NSTEP = 580000 TIME(PS) = 1580.000 TEMP(K) =********* PRESS =
> 0.0
> wrapping first mol.: -7287253.59696 6945599.13615 6810626.98843
> wrapping first mol.: -7287253.59696 6945599.13615 6810626.98843
> Best regards,
> Quynh Vo
>
>
> Hi,
>
> Could you give the results of the following command on our mdout file:
> egrep 'NSTEP|wrapping first mol' mdout
> This should output the NSTEP line as well as when it does the wrapping
> to see if this happens suddenly or not.
> Thanks,
>
> Gerald.
>
> On Thursday, April 2, 2015 11:42 AM, Quynh Vo <ziczac_nq.yahoo.com> wrote:
>
>
> Hi,
>
> I would like to say yes to your question; it worked well with nmropt=0.
> I ran several MD simulations (not umbrella sampling) before and never
> experienced this problem. And also for these regular simulations, nmropt
> was set to 1 for slow heating but no error whatsoever.
> Best regards,
> Quynh Vo
>
>
>
>
>
> On Thursday, April 2, 2015 11:35 AM, Jason Swails
> <jason.swails.gmail.com> wrote:
>
>
> On Thu, 2015-04-02 at 18:33 +0000, Quynh Vo wrote:
> > Hi,
> >
> >
> > This is pmemd.cuda serial version 12.3.1 08/07/2013. When I did
> > test.cuda a few tests failed (from 4 to 7 depending on the GPUs). The
> > diff files didn't look that bad (in my opinion, i can be wrong). For
> > example one of the test failure:
>
> Does it work if you set nmropt=0? I know this doesn't help much (but
> Amber 14 has a lot more support for NMR restraints in the GPU code than
> Amber 12 does).
>
> This would at least help us narrow down where the problems could be.
>
>
> -Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
>
--
____________________________________________________________________________
Prof. Gerald MONARD
SRSMC, Université de Lorraine, CNRS
Boulevard des Aiguillettes B.P. 70239
F-54506 Vandoeuvre-les-Nancy, FRANCE
e-mail : Gerald.Monard.univ-lorraine.fr
tel. : +33 (0)383.684.381
fax : +33 (0)383.684.371
web : http://www.monard.info
____________________________________________________________________________
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 03 2015 - 15:00:03 PDT
