Dear ambers,
My umberlla sampling simulation with Amber 16 in GTX1080
ofter quit anormalously.
Example in the below, shows overflow of Fortran format
Between 130000 and 135000.
Oddly enough, the simulation continued by the end, 500000 steps in this case.
No proper restart file was generated, of course.
NMR restraints: Bond = 1.781 Angle = 0.000 Torsion = 0.000
===============================================================================
NSTEP = 130000 TIME(PS) = 300.000 TEMP(K) = 364.42 PRESS = 0.0
Etot = 7127.4152 EKtot = 40129.9688 EPtot = -33002.5535
BOND = 8368.8312 ANGLE = 16807.3350 DIHED = 12576.8921
1-4 NB = 12125.2584 1-4 EEL = -58738.1525 VDWAALS = -21559.1837
EELEC = -2583.6596 EHBOND = 0.0000 RESTRAINT = 0.1257
EAMBER (non-restraint) = -33002.6792
------------------------------------------------------------------------------
NMR restraints: Bond = 0.126 Angle = 0.000 Torsion = 0.000
===============================================================================
wrapping first mol.: -108509.81400 2405238.82200 1007158.64500
NSTEP = 135000 TIME(PS) = 310.000 TEMP(K) =********* PRESS = 0.0
Etot = ************** EKtot = ************** EPtot = **************
BOND = -0.0000 ANGLE = 4242447.3325 DIHED = 261422.4660
1-4 NB = 0.0000 1-4 EEL = 0.0024 VDWAALS = **************
EELEC = -10707.1762 EHBOND = 0.0000 RESTRAINT = **************
EAMBER (non-restraint) = **************
------------------------------------------------------------------------------
The following simulation gives an error message
"*s in the inpcrd file often indicate an overflow of the Fortran format used
to store coordinates in the inpcrd/restart files. This often happens when
coordinates are not wrapped into the center cell (when iwrap = 0) and some
particles diffuse too far away. Try restarting from your last good restart
file and setting iwrap=1 or using a NetCDF restart file format. See the
Amber manual for details"
However, "IWRAP" is set to 1, so that I guess that
overflow of Fortran format occurs not due to natural particle diffusion.
If anyone know such situation and overcome it.
Could you give me advices.
Best regards,
IK
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu May 18 2017 - 16:00:03 PDT
