Minimization error with CG
Luca
bellu... at unisi.it
Fri Oct 5 14:21:14 UTC 2007
Hi all,
regular MD or annealing procedure are right but
when a I perform minimization with CG (I can't use BFGS) of my
protein
(4647 atoms and no water) I have this error:
.......
......
*******************************************************************************
*** BRENT - NUMBER OF ENERGY EVALUATIONS :
4 ***
*******************************************************************************
-------- Informations at step = 4 ------------
Optimization Method = CG
Total Energy = -9.6915636844
Real energy change = -0.3249158998
Decrease in energy = YES
Convergence check :
Max. step size = 0.3175224352
Conv. limit for step size = 0.0030000000
Convergence in step size = NO
RMS step size = 0.0237387989
Conv. limit for RMS step = 0.0015000000
Convergence in RMS step = NO
Max. gradient = 0.1033238799
Conv. limit for gradients = 0.0004500000
Conv. for gradients = NO
RMS gradient = 0.0077247606
Conv. limit for RMS grad. = 0.0003000000
Conv. for gradients = NO
---------------------------------------------------
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-9.691563684351616
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-10.144738185773022
WARNING| Particles: 1365 840 at distance [au]: 0.96785100
less than: 1.31885811; increase EMAX_SPLINE.
*************************************
*** ERROR in build_neighbor_lists ***
*************************************
*** GEOMETRY wrong or EMAX_SPLINE too small! ***
===== Routine Calling Stack =====
12 build_neighbor_lists
11 build_fist_neighbor_lists
10 list_control
9 fist_force_control
8 cp_eval_at
7 cg_eval1d
6 cg_mnbrak
5 LinMin
4 cp_cg_main
3 geoopt_cg
2 qs_geo_opt
1 CP2K
I tray some combination but I don't perform my minimizzation...(SPLINE
generation are
very slow ) with SAVE_MEM T option i had this error:
*******************************************************************************
*** STARTING GEOMETRY
OPTIMIZATION ***
*** CONJUGATE
GRADIENTS ***
*******************************************************************************
-------- Informations at step = 0 ------------
Optimization Method = SD
Total Energy = -8.6496756962
---------------------------------------------------
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-8.649675696171034
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-8.769097957696978
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-8.919447375783333
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-8.996111041205189
ENERGY| Total FORCE_EVAL ( FIST ) energy (a.u.):
-8.484884286061162
*******************************************************************************
*** MNBRACK - NUMBER OF ENERGY EVALUATIONS :
5 ***
*******************************************************************************
*******************************************************************************
*** BRENT - NUMBER OF ENERGY EVALUATIONS :
3 ***
*******************************************************************************
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Stack trace terminated abnormally.
Follow input section:
&FORCE_EVAL
METHOD FIST
&MM
&FORCEFIELD
PARM_FILE_NAME ./par_amber2charmm.inp
PARMTYPE CHM
EI_SCALE14 0.8333
VDW_SCALE14 1.0
&SPLINE
UNIQUE_SPLINE T
EMAX_SPLINE 0.5
&END SPLINE
&END FORCEFIELD
&POISSON
&EWALD
EWALD_TYPE NONE
ALPHA 0.44
GMAX 64
O_SPLINE 6
&END EWALD
PERIODIC NONE
&END POISSON
&PRINT
# &FF_INFO
# &END FF_INFO
&END PRINT
&END MM
&SUBSYS
&CELL
ABC 57.000 60.0 53.000
UNIT ANGSTROM
PERIODIC NONE
&END CELL
&TOPOLOGY
COORD_FILE_NAME 1ys1_as.pdb
COORDINATE PDB
CONN_FILE_NAME 1ys1_a.psf
CONNECTIVITY psf
&END TOPOLOGY
&END SUBSYS
&END FORCE_EVAL
&GLOBAL
PRINT_LEVEL LOW
PROJECT 1ys1_asFIST
RUN_TYPE GEO_OPT
SAVE_MEM T
&PERFORMANCE
CPUTIME T
MEMORY T
&END
&END GLOBAL
&MOTION
&GEOOPT
OPTIMIZER CG
&END GEOOPT
&MD
ENSEMBLE NVE
STEPS 150000
TIMESTEP 1.0
TEMPERATURE 300.
# TEMP_TOL 25.0
ANNEALING 0.9
&END MD
&PRINT
&RESTART ON
&END
&RESTART_HISTORY OFF
&END
&TRAJECTORY
EACH 100
FORMAT XMOL
&END
&VELOCITIES OFF
&END
&END PRINT
&CONSTRAINT
&HBONDS
&END HBONDS
&END CONSTRAINT
&END MOTION
More information about the CP2K-user
mailing list