Hi
I'm using amber7 to simulate a protein dna complex( 6593 atoms, 372
residues) with explicit water (11955 TIP3P) and counter ions (30 K+, 13
Cl-, 2Mg).
I have simulated successfully this system on a PC cluster as well as an
equivalent system (same protocol, same number of atoms and residue, one
more covalent bond) on a single machine. However this time when I try
the simulation on my PC, I received an error message about the size of
the Non bond list. I can increase the parameter MAXPR in size.h and
recompile but what I don't understand is the difference between the
total size of non bond list : 11882628 at the beginning, 11999932 at the
error point, and the error message itself ( for me 11999932 doesn't
exceeds 12000000).... any explanations will be appreciated
Thanks,
Manu
PS: these are the last lines of my output file
---------------------------------------------------
| Local SIZE OF NONBOND LIST = 11882628
| TOTAL SIZE OF NONBOND LIST = 11882628
NSTEP = 0 TIME(PS) = 50.000 TEMP(K) = 0.00 PRESS = -1126.2
Etot = -138323.2656 EKtot = 0.0000 EPtot = -138323.2656
BOND = 1286.3115 ANGLE = 3218.1226 DIHED = 2494.3721
1-4 NB = 1516.2947 1-4 EEL = 9645.0116 VDWAALS = 13131.3806
EELEC = -169614.7587 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 0.0000 VIRIAL = 10243.0535 VOLUME = 421262.1660
Density = 1.0460
Ewald error estimate: 0.1260E-03
------------------------------------------------------------------------------
* NB pairs 294 11999932 exceeds capacity ( 12000000) 0
SIZE OF NONBOND LIST = 12000000
EWALD BOMB in subroutine ewald_list
Non bond list overflow!
check MAXPR in sizes.h
--
--------------------------------------------------------------------------------
Emmanuel GIUDICE
--------------------------------------------------------------------------------
MOUNT SINAI SCHOOL OF MEDICINE
DEPARTMENT OF PHYSIOLOGY AND BIOPHYSICS Emmanuel.Giudice.mssm.edu
One Gustave L. Levy place, Annenberg Building Tel-Labo : (212) 241-1613
21st Floor, Room 78 New York, NY 10029 Fax : (212) 860-3369
--------------------------------------------------------------------------------
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Thu Feb 26 2004 - 21:53:00 PST
