[CPMD-list] Platform-dependent output energies
Eung-Gun Kim
eunggun.kim at chemistry.gatech.edu
Fri Mar 26 21:09:08 CET 2004
Dear CPMD Mailing List Subscribers,
I have been running tests on different platforms and am getting different energy
values on different machines. The platform-dependent results dont show when I
run on the sample input files that came with the code. With my own input file,
however, I get values of the total energy as different as by 0.1 Ha upon
switching the machine. The discrepancies exist before and after geometry
optimization. Some relevant info on my input and installation is as follows:
CP Version: 3.7.2
Tested platforms:
1) IBM SP2 (AIX 5.2): xlf; MPICH 1.2.5.2; LAPACK 3.0
2) IBM Xeon cluster (RedHat 9.0): PGI Fortran compiler 5.1 (with BLAS and
LAPACK); MPICH 1.2.5.2
System:
Aluminum crystal (2x2x2 cells for a total of 32 atoms)
As you note in my input file below, I didnt use either k-point sampling or FEMD
for this metallic system. Portions from each output file are also attached at
the end of the input file.
Best regards,
E Kim
&CPMD
OPTIMIZE GEOMETRY
PCG MINIMIZE
BFGS
LSD
EMASS
900
TIMESTEP
5
MAXSTEP
10000
TRAJECTORY SAMPLE
10
&END
&DFT
FUNCTIONAL BLYP
&END
&SYSTEM
SYMMETRY
1
ANGSTROM
CELL ABSOLUTE
8.1 8.1 8.1 0.0 0.0 0.0
CUTOFF
70
CHARGE
0
MULTIPLICITY
1
&END
&ATOMS
*Al_MT_GIA_BLYP.psp KLEINMAN-BYLANDER RAGGIO=1.0
LMAX=D
32
0.000000000 0.000000000 0.000000000
2.025000000 2.025000000 0.000000000
0.000000000 2.025000000 2.025000000
2.025000000 0.000000000 2.025000000
0.000000000 0.000000000 4.050000000
2.025000000 2.025000000 4.050000000
0.000000000 2.025000000 6.075000000
2.025000000 0.000000000 6.075000000
0.000000000 4.050000000 0.000000000
2.025000000 6.075000000 0.000000000
0.000000000 6.075000000 2.025000000
2.025000000 4.050000000 2.025000000
0.000000000 4.050000000 4.050000000
2.025000000 6.075000000 4.050000000
0.000000000 6.075000000 6.075000000
2.025000000 4.050000000 6.075000000
4.050000000 0.000000000 0.000000000
6.075000000 2.025000000 0.000000000
4.050000000 2.025000000 2.025000000
6.075000000 0.000000000 2.025000000
4.050000000 0.000000000 4.050000000
6.075000000 2.025000000 4.050000000
4.050000000 2.025000000 6.075000000
6.075000000 0.000000000 6.075000000
4.050000000 4.050000000 0.000000000
6.075000000 6.075000000 0.000000000
4.050000000 6.075000000 2.025000000
6.075000000 4.050000000 2.025000000
4.050000000 4.050000000 4.050000000
6.075000000 6.075000000 4.050000000
4.050000000 6.075000000 6.075000000
6.075000000 4.050000000 6.075000000
&END
Output from the SP2 run---------------------------------------------
(K+E1+L+N+X) TOTAL ENERGY = -64.39023857 A.U.
(K) KINETIC ENERGY = 28.11394543 A.U.
(E1=A-S+R) ELECTROSTATIC ENERGY = -75.61888788 A.U.
(S) ESELF = 114.89537676 A.U.
(R) ESR = 0.00001993 A.U.
(L) LOCAL PSEUDOPOTENTIAL ENERGY = -5.89039837 A.U.
(N) N-L PSEUDOPOTENTIAL ENERGY = 13.16534369 A.U.
(X) EXCHANGE-CORRELATION ENERGY = -24.16024144 A.U.
GRADIENT CORRECTION ENERGY = -0.22825193 A.U.
Output from the Xeon run--------------------------------------------
(K+E1+L+N+X) TOTAL ENERGY = -64.27276166 A.U.
(K) KINETIC ENERGY = 28.36257064 A.U.
(E1=A-S+R) ELECTROSTATIC ENERGY = -76.82637865 A.U.
(S) ESELF = 114.89537676 A.U.
(R) ESR = 0.00001993 A.U.
(L) LOCAL PSEUDOPOTENTIAL ENERGY = -6.46519951 A.U.
(N) N-L PSEUDOPOTENTIAL ENERGY = 14.94091935 A.U.
(X) EXCHANGE-CORRELATION ENERGY = -24.28467349 A.U.
GRADIENT CORRECTION ENERGY = -0.29067887 A.U.
More information about the CPMD-list
mailing list