Hi Jose,
I think the issue here is that the code should NOT be segfaulting. Quitting
with a linmin error is fine but a segfault it should not do. Hence why you
need to make sure all the test cases pass and why I inquired if this was in
serial / parallel, which version of AMBER etc.
All the best
Ross
From: owner-amber.scripps.edu [mailto:owner-amber.scripps.edu] On Behalf Of
Jose Borreguero
Sent: Friday, September 05, 2008 8:44 AM
To: amber.scripps.edu
Subject: Re: AMBER: help...sander "segmentation fault" after 2hours of
minimization
Total energy decreased from -45404.6511 to -57070.4772.
Minimal RMS=0.000721 at time_step=20100 (still higher than limit_RMS=0.0005)
Is there an automatic way to flag a high-energy bond ? This is a solvated
system with 14000 atoms :)
On Fri, Sep 5, 2008 at 11:20 AM, David A. Case <case.biomaps.rutgers.edu>
wrote:
On Fri, Sep 05, 2008, Jose Borreguero wrote:
> I am trying to understand what went wrong with my minimization. The last
> lines of the output file do not indicate anything weird going on, or maybe
> the BOND energy term is indicating somethin (at the beginning of the
> minimization, BOND=115.7477 and at the end, BOND*=4411.3524 *).
> Where else should I look for pointers to the problem?
It is the *total* energy that is minimized -- did that go downhill from
the starting point?
It looks like your minimization may have converged, reaching an rms of
less than 0.0005, at some step after the last printed step. If so,
there is no reason, per se, to think that something is wrong. You can
examine the structure to see if there are individual bonds that are
leading to the large-ish bond energy, and so on.
...dac
> Here are the last lines of the output file:
> *NSTEP ENERGY RMS GMAX NAME NUMBER
> 20150 -5.7069E+04 1.1228E-03 4.5015E-02 H1 9884
>
> BOND = 4411.3524 ANGLE = 93.5743 DIHED =
> 212.3167
> VDWAALS = 10160.1516 EEL = -75579.7172 HBOND =
> 0.0000
> 1-4 VDW = 156.0090 1-4 EEL = 3475.8361 RESTRAINT =
> 1.9324
> EAMBER = -57070.4772*
> And here's the input file:
> *&cntrl
> imin=1, maxcyc=100000, ntmin=0, ntr=1,
> cut=10.0, nsnb=10, ntc=1, ntf=1, drms=0.0005
> &end
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
to majordomo.scripps.edu
--
Jose M. Borreguero
Postdoctoral Associate
Oak Ridge National Laboratory
P.O. Box 2008, M.S. 6164
Oak Ridge, TN 37831
phone: 865-241-3071 fax: 865-576-5491
Email: borreguerojm.ornl.gov
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
to majordomo.scripps.edu
Received on Sun Sep 07 2008 - 06:07:51 PDT
