Topic Options
#672 - 12/07/03 09:48 PM EOF during read
freesky Offline
Forum Member

Registered: 11/20/03
Posts: 76
Hi, all

When I use the LN algotithm to perform MD simulation, the below errors occur, could anyone tell me how to solve it. Thanks!



CHARMM> DYNA LNX REST NSTEP 2000000 TIME 0.0005 -
CHARMM> NPRINT 100 IPRFRQ 250 IHTFRQ 0 IEQFRQ 0 NTRFRQ 100 -
CHARMM> FIRSTT 300.0 IUNREA 60 IUNWRI 61 IUNCRD 30 IUNVEL -1 -
CHARMM> KUNIT 50 IUNO -1 NSAVC 100 NSAVV 100 NSNOS 10 ISVFRQ 1000
-
CHARMM> TSTRUC 300
IUNREA = 60 IUNWRI = 61 IUNOS = -1
IUNCRD = 30 IUNVEL = -1 KUNIT = 50

***** LEVEL -3 WARNING FROM *****
***** EOF during read
******************************************
BOMLEV ( -1) IS REACHED - TERMINATING. WRNLEV IS 5



_________________________
---- State Key Lab of Solid State Microstructure Nanjing University P.R.C

Top
#673 - 12/08/03 04:25 AM Re: EOF during read [Re: freesky]
jb007 Offline
Forum Member

Registered: 09/30/03
Posts: 250
Loc: Florida State
Is it the initial startup? Because then "STARt" should be in the dynmix input command line instead of "RESTa". If not, then the restart-file (unit 60) is incomplete (perhaps you've got an automated unix/linux script that submits jobs subsequently). The term 'ISVFRQ 1000' asks CHARMM to save the restart file (which contains the current atomic velocities) every 1000 step. I have learned that the use of ISVFRQ is somewhat historic, back in the ancient time, when computer resources were less than they are today. Of course - increased CPU:power in itself doesn't automatically ensure accurate predictions...

Top

Moderator:  John Legato, lennart