(29 Apr 94)
******************************
* *
* Section 3 - Input Examples *
* *
******************************
The distribution tape contains a number of short test
examples, named EXAM*.INP. You should run all of these
tests to be sure you have installed GAMESS correctly. The
correct answers are shown in the comments preceeding each
of the short input tests. The "correct" answers are
obtained on a RS/6000 computer, other machines may differ
in the last energy digit, or the last couple of gradient
digits.
The examples are listed in the final pages of this
section, and serve as a useful tutorial about the kind of
calculations GAMESS can do.
Example Description
------- -----------
1 CH2 RHF geometry optimization
2 CH2 UHF + gradient
3 CH2 ROHF + gradient
4 CH2 GVB + gradient
5 CH2 CI
6 CH2 MCSCF geometry optimization
7 HPO RHF + gradient
8 H2O RHF
9 H2O MCSCF
10 H2O RHF + hessian
11 HCN RHF IRC
12 HCCH RHF geometry optimization
13 H2O RHF properties
14 H2O CI transition moment
15 C2- GVB/ROHF on 2-pi-u state
16 Si GVB/ROHF on 3-P state
17 CH2 GVB/ROHF + hessian
18 P2 RHF with effective core potential
19 CH2 spin-orbit coupling
20 NH3 MP2 trudge optimization
21 CH3 OS-TCSCF hessian
22 H3CN UHF + UMP2 energy
23 SiH3- PM3 geometry optimization
24 H2O SCRF test case
25 internal coordinate example
The distribution tape contains a number of larger test
jobs, named BENCH*.INP. It is a good idea to run a few of
these tests along with all the short ones described above.
BENCH tests 1, 2, 3, 4, 6, and 10 are of moderate length.
The others are considerably longer, and you might not want
to run them if your machine is a "well supported" model.
GAMESS benchmarks
The distribution tape for GAMESS contains some larger
BENCH*.INP tests. These jobs are a more strenuous test of
GAMESS, as more memory, CPU, and disk space is required to
run them. They also serve as a good comparison of various
machines performance. These jobs are
number of disk
Test Molecule Method Basis set AOs CSFs Mbytes
1 SiC2H6 RHF direct SCF 6-31G* 61 1 14
2 SiH2 MCSCF 6-31G** 29 51 5
3 Si2H4 second order CI 6-31G* 46 4600 64
4 SiC3H8 RHF 6-31G* 80 1 38
5 C3H4 MCSCF + gradient DH(d) 53 20 84
6 O2+ CI transition Duij(2d) 60 504 43
7 OHBr MCSCF 3-21G(d,p) 49 110 55
8 SnC5H6 GVB-PP(3) 3-21G* 96 6 105
9 C6H8 UHF DH(d,p) 130 1 175
10 P2H4+ ROHF + gradient DH(d) 56 1 14
11 HPPH second order CI 6-31G* 42 55017 490
12 SbC4H4NO2 RHF + gradient 3-21G* 110 1 111
13 FOOF RHF + hessian 6-31G* 60 1 98
Performance for a representative subset of these jobs
is reported on the next page. All times were obtained
with the GAMESS version of Dec 6, 1989 or later. Any
older version of GAMESS will run significantly slower than
the times reported here.
The machines tested, their location, operating system,
and FORTRAN version are summarized. Machines with vector
capability are marked "v". All runs are for single CPUs,
although many of the machines are multiprocessor models.
Times are measured in CPU seconds, and are normalized to the
performance of IBM's desktop 42 MHz RS/6000 system.
It is important to note that the machines were in
general running other jobs at the time the benchmarks were
run, and that background load can have a big influence on
the reproducibility of times. Of course, the size and
characteristics of the disks used, and total system RAM
also influence performance. Most of the jobs were run
using 750,000 words in the GAMESS dynamic memory pool, but
the larger systems frequently use more, again influencing
the times. Usually, only one trial run was made. All of
these variables mean the results should be regarded only
as an informal benchmark.
The operating systems tested are
microVAX II - NDSU Chemistry, VMS 5.0, FORTRAN 5.3
VAXstation 3200 - NDSU Chemistry, VMS 5.2, FORTRAN 5.3
SUN SPARCstn 1 - U. Utah, SunOS 4.0.3c, f77 1.2
SUN SPARCstn 370 - U. Alberta, SunOS, f77 (&Iowa State)
DECstation 3100 - NDSU Chemistry, Ultrix 3.1, MIPS f77 1.0
486DX2/66 MHz - United Tech., Linux 0.99.13, f2c+gcc
DECstation 5000-120 - NDSU Chemistry, Ultrix 4.2, f77 3.1
(this is a 20 Mhz R3000A, there are many other models)
SUN SPARCstn 2 - U.Texas-Dallas, SunOS 4.1.1, f77 1.3.1
v Stardent 3000 - NEC corporation
Sumitomo SG330 - Mie University, Japan
SGI 4D/380S - Iowa State, Irix 3.3.1, MIPS FORTRAN 2.00
v Convex C220 - Univ. Twente, UNIX v8.0, fc v6.0.0.0
SGI Indy - ISU Chemistry, Irix 5.1.1.3, f77 4.0 -mips2
v Cray Y-MP EL/232 - Mitsui Petrochem, Unicos 6.16, cft77 5.0.3
SGI 4D/440VGX - U.Penn, Irix 3.3.2
IBM RS/6000 m520 - NDSU chemistry, AIX 3.1.2, xlf 1.01
DECstation 5000-240 - Ames Lab, Ultrix 4.2a, f77 3.2
SGI 4D/35 - NDSU Chemistry, Irix 4.0.1, f77 3.4.1
IBM RS/6000 m530 - NDSU chemistry, AIX 3.1.5, xlf 2.2
HP 9000-720 - Tohoku University, Japan
VAX 7610 - 3M Corp, VMS 5.5-2, FORTRAN 6.0-31
Sun SPARCstn 10 - Mie U., SunOS 4.1.3, f77 2.0.1
HP 9000-715 - Mat. Res. Lab, Victoria - HPUX, f77 +O2
SGI Indigo2 (r4000, 100 MHz) - Iowa State, Irix 4.0.5H, f77 -mips2
SGI Indigo Elan (R4000, 100 MHz) - SGI, Irix 4.0.5F, f77 -mips2
SGI Challenge L (R4400, 100 MHz) - SGI, Irix 5.0beta, f77 -mips2
IBM RS/6000 m350 - ISU chemistry, AIX 3.2.2, xlf 2.2.1
v VAX 9000-410 - Aristotle U. Thessaloniki, VMS 5.5 (no DXML)
v IBM 3090-600J - U.Minnesota, AIX/370 1.2, VS FORTRAN 2.5.0
IBM RS/6000 m560 - U.Alberta, AIX 3.2.3, xlf 2.3
DEC AXP 3000-300 - Virginia Tech, OSF/1
DEC AXP 3000-400S - Iowa State, openVMS 1.0, FORTRAN 6.0 (133 MHz)
DEC AXP 3000-400S - Iowa State, OSF/1 1.2, DEC FORTRAN 3.3 (133 MHz)
SGI Indigo2 (R4400, 150 MHz) - Odense U., Irix 5.1.1, f77 -mips2
Fujitsu M1800/20 - RIKEN Japan, UXP/M
v NEC SX-3 - NEC Systems Lab, SX-3/22, SUPER-UX 3.1H, f77sx
IBM RS/6000 m370 - ISU chemistry, AIX 3.2.3E, xlf 2.3.0
IBM RS/6000 m580 - United Technologies Corp.
HP 9000-735 -Australian Nat. U. - HPUX, f77 +O2
v Cray Y-MP/864 - SDSC (San Diego), Unicos 6.1.6, cft77 5.0.3.14
HITAC M-880 - U.Tokyo, VOS3/AS, FORT77/HAP V25-OA
v IBM ES/9000 m900 - CNSF (Cornell), AIX 1.2.1, VS FORTRAN 2.5.0
v Fuj. VP2200 - Australian Nat. U., UXP/M V10L10, F77 EX/VP V12L10
DEC AXP 3000-500X - U.Minnesota, OSF/1 (200 MHz)
DEC AXP 3000-600S - Texas Tech., OSF/1 1.3
v Cray C90 - NERSC, Unicos 7.C, cft77 5.0.3.11
IBM RS/6000 m590 - Norsk Hydro MolSim-Lab., AIX 3.2.5 xlf 2.3
-----------------------------------------------------------------
ROHF grad RHF E RHF hess MCSCF E
BENCH10 BENCH4 BENCH13 BENCH7
------- ------ ------- ------
microVAX II 7644 ( ) 8716 (67.0) 47620 (86.6) 47472 (74.4)
VAXstn 3200 2395 ( ) 2525 (19.4) 13075 (23.8) 14396 (22.6)
Sun SPARCstn 1 1242 ( ) 1430 (11.0) 6632 (12.1) 7547 (11.8)
Sun SPARCstn 370 902 ( ) 859 (6.61) 3945 (7.17) 4314 (6.76)
DECstn 3100 693 (6.36) 689 (5.30) 3070 (5.58) 4056 (6.36)
486DX2/66 578 (5.30) 696 (5.35) 2943 (5.35) 3752 (5.88)
DECstn 5000-120 485 (4.45) 583 (4.48) 2302 (4.19) 3089 (4.84)
Sun SPARCstn 2 540 ( ) 537 (4.13) 2366 (4.30) 2634 (4.12)
v Stardent 3000 462 ( ) 486 (3.73) 1638 (2.98) 1560 (2.44)
Sumitomo SG330 280 (2.11) 337 (2.59) 1791 (3.26) 1982 (3.11)
SGI 4D/380S 346 ( ) 325 (2.50) 1608 (2.92) 1841 (2.89)
v Convex C220 309 ( ) 329 (2.53) 1163 (2.11) 1459 (2.29)
SGI Indy 290 (2.66) 325 (2.50) 1292 (2.35) 1147 ( )
v Cray Y-MP EL/232 296 ( ) 321 (2.47) 975 (1.77) 945 (1.48)
SGI 4D/440VGX 296 ( ) 288 (2.21) 1414 (2.57) 1688 (2.65)
IBM RS/6000 m520 263 ( ) 284 (2.18) 1137 (2.07) 1591 (2.49)
DECstn 5000-240 232 (2.13) 294 (2.26) 1142 (2.08) 1449 (2.27)
SGI 4D/35 236 (2.17) 255 (1.96) 1451 (2.64) 1392 (2.18)
IBM RS/6000 m530 170 (1.56) 204 (1.57) 782 (1.42) 1012 (1.59)
VAX 7610 161 (1.48) 172 (1.32) 830 (1.51) 888 (1.39)
HP 9000-720 138 (1.27) 144 (1.11) 785 (1.43) 1023 (1.60)
HP 9000-715 141 (1.29) 158 (1.21) 715 (1.30) 829 (1.30)
Sun SPARCstn 10 125 (1.15) 155 (1.19) 691 (1.26) 710 (1.11)
SGI Indigo2 133 (1.22) 137 (1.05) 711 (1.29) 701 (1.10)
SGI Indigo Elan 118 (1.08) 125 (0.96) 645 (1.17) 597 (0.94)
IBM RS/6000 m350 109 ( 1 ) 130 ( 1 ) 550 ( 1 ) 638 ( 1 )
SGI Challenge L 105 (0.96) 114 (0.88) 619 (1.13) 547 (0.86)
v VAX 9000-410 127 (1.17) 120 (0.92) 441 (0.80) 633 (0.99)
v IBM 3090-600J/VF 120 ( ) 119 (0.92) 420 (0.76) 467 (0.73)
DEC AXP 3000-300 95 (0.87) 110 (0.85) 475 (0.86) 532 (0.83)
IBM RS/6000 m560 83 (0.76) 94 (0.72) 394 (0.72) 487 (0.76)
VMS DEC AXP 3000-400S 84 (0.77) 93 (0.72) 320 (0.58) 437 (0.68)
SGI Indigo2 (150) 70 (0.64) 78 (0.60) 434 (0.78) 392 (0.61)
Fujitsu M1800/20 61 (0.56) 66 (0.51) 310 (0.56) 360 (0.56)
v NEC SX-3 66 (0.60) 87 (0.67) 197 (0.36) 157 (0.25)
IBM RS/6000 m370 64 (0.59) 74 (0.57) 366 (0.67) 391 (0.61)
OSF DEC AXP 3000-400S 72 (0.66) 73 (0.56) 324 (0.59) 388 (0.61)
IBM RS/6000 m580 59 (0.54) 69 (0.53) 322 (0.59) 381 (0.60)
HP 9000-735 50 (0.46) 59 (0.45) 291 (0.53) 365 (0.57)
v Cray Y-MP/864 60 (0.55) 68 (0.52) 188 (0.34) 197 (0.31)
HITAC M-880 54 (0.50) 59 (0.45) 297 (0.54) 362 (0.57)
v IBM ES/9000-900 53 (0.49) 59 (0.45) 201 (0.37) 187 (0.29)
DEC AXP 3000-600S 52 (0.48) 53 (0.41) 230 (0.42) 279 ( )
v Fujitsu VP2200 50 (0.46) 57 (0.44) 148 (0.27) 127 (0.20)
DEC AXP 3000-500X 46 (0.42) 50 (0.38) 214 (0.39) 261 (0.41)
v Cray C90 42 (0.39) 49 (0.38) 129 (0.23) 137 (0.21)
IBM RS/6000 m590 33 (0.30) 33 (0.25) 147 (0.28) 202 (0.32)
-----------------------------------------------------------------
Note that HF energy and gradient runs are pretty much
scalar, whereas hessian runs and MCSCF/CI use a vector
processor effectively.
Many of the timings preceed the Oct 18, 1991 version
of GAMESS, which converges in fewer iterations for B10.
! EXAM01.
! 1-A-1 CH2 RHF geometry optimization using GAMESS.
!
! Although internal coordinates are used (COORD=ZMAT),
! the optimization is done in Cartesian space (NZVAR=0).
! This run uses a criterion (OPTTOL) on the gradient
! which is much tighter than is normally used.
!
! This job tests the sp integral module, the RHF module,
! and the geometry optimization module.
!
! Using the default search METHOD=BAKER,
! FINAL E= -37.2322678015, 8 iters, RMS grad= 0.0264308
! FINAL E= -37.2308175316, 8 iters, RMS grad= 0.0320881
! FINAL E= -37.2375723414, 7 iters, RMS grad= 0.0056557
! FINAL E= -37.2379944432, 7 iters, RMS grad= 0.0017904
! FINAL E= -37.2380387851, 8 iters, RMS grad= 0.0003389
! FINAL E= -37.2380397692, 6 iters, RMS grad= 0.0000030
!
$CONTRL SCFTYP=RHF RUNTYP=OPTIMIZE COORD=ZMT NZVAR=0 $END
$SYSTEM TIMLIM=2 MEMORY=100000 $END
$STATPT OPTTOL=1.0E-5 $END
$BASIS GBASIS=STO NGAUSS=2 $END
$GUESS GUESS=HUCKEL $END
$DATA
Methylene...1-A-1 state...RHF/STO-2G
Cnv 2
C
H 1 rCH
H 1 rCH 2 aHCH
rCH=1.09
aHCH=110.0
$END
! EXAM02.
! 3-B-1 CH2 UHF calculation on methylene ground state.
!
! This test uses the default choice, COORD=UNIQUE, to
! enter the molecule. Only the symmetry unique atoms
! are given, and they must be given in the orientation
! which GAMESS expects.
!
! This job tests the UHF energy and the UHF gradient.
! In addition, the orbitals are localized.
!
! The initial energy is -37.228465066.
! The FINAL energy is -37.2810867258 after 11 iterations.
! The unrestricted wavefunction has <S**2> = 2.013.
! The RMS gradient is 0.027589766.
! The dipole moment is 0.016188.
! The spin density at Hydrogen is -0.0167104.
! Mulliken, Lowdin charges on C are -0.020584, 0.018720.
! FINAL localization sums are 30.57 and 25.14 Debye**2.
!
$CONTRL SCFTYP=UHF MULT=3 RUNTYP=GRADIENT LOCAL=BOYS $END
$SYSTEM TIMLIM=1 MEMORY=100000 $END
$BASIS GBASIS=STO NGAUSS=2 $END
$GUESS GUESS=HUCKEL $END
$DATA
Methylene...3-B-1 state...UHF/STO-2G
Cnv 2
Carbon 6.0
Hydrogen 1.0 0.0 0.82884 0.7079
$END
! EXAM03.
! 3-B-1 CH2 ROHF calculation on methylene ground state.
! The wavefunction is a pure triplet state (<S**2> = 2),
! and so has a higher energy than the second example.
!
! For COORD=CART, all atoms must be given, and as in the
! present case, these may be in an unoriented geometry.
! GAMESS deduces which atoms are unique, and orients
! the molecule appropriately. The geometry here is thus
! identical to the second example.
!
! This job tests the ROHF wavefunction and gradient code.
! It also tests the direct SCF procedure.
!
! The initial energy is -37.228465066.
! The FINAL energy is -37.2778767089 after 7 iterations.
! The RMS gradient is 0.027505548.
! The dipole moment is 0.025099.
! The Hydrogen atom spin density is 0.0129735.
! Mulliken, Lowdin charges on C are -0.020346, 0.019470.
!
$CONTRL SCFTYP=ROHF MULT=3 RUNTYP=GRADIENT COORD=CART $END
$SYSTEM TIMLIM=1 MEMORY=100000 $END
$SCF DIRSCF=.TRUE. $END
$BASIS GBASIS=STO NGAUSS=2 $END
$GUESS GUESS=HUCKEL $END
$DATA
Methylene...3-B-1 state...ROHF/STO-2G
Cnv 2
Hydrogen 1.0 0.82884 0.7079 0.0
Carbon 6.0
Hydrogen 1.0 -0.82884 0.7079 0.0
$END
! EXAM04.
! 1-A-1 CH2 TCSCF calculation on methylene.
! The wavefunction has two configurations, exciting
! the carbon sigma lone pair into the out of plane p.
!
! Note that the Z-matrix used to input the molecule
! can include identifying integers after the element
! symbol, and that the connectivity can then be given
! using these labels rather than integers.
!
! This job tests the GVB wavefunction and gradient.
!
! The initial GVB-PP(1) energy is -37.187342653.
! The FINAL energy is -37.2562020559 after 10 iters.
! The GVB CI coefs are 0.977505 and -0.210911, giving
! a pair overlap of 0.64506.
! Mulliken, Lowdin charges for C are 0.020810, 0.055203.
! The RMS gradient = 0.019618475.
! The dipole moment is 1.249835.
!
$CONTRL SCFTYP=GVB RUNTYP=GRADIENT COORD=ZMT $END
$SYSTEM TIMLIM=1 MEMORY=100000 $END
$BASIS GBASIS=STO NGAUSS=2 $END
$SCF NCO=3 NSETO=0 NPAIR=1 $END
$DATA
Methylene...1-A-1 state...GVB...one geminal pair...STO-2G
Cnv 2
C1
H1 C1 rCH
H2 C1 rCH H1 aHCH
rCH=1.09
aHCH=99.0
$END
! normally a GVB-PP calculation will use GUESS=MOREAD
$GUESS GUESS=HUCKEL $END
! EXAM05
! 1-A-1 CH2 CI calculation.
! The wavefunction is a full CI within the minimal
! basis, except the 1s carbon orbital is constrained
! to double occupancy. 56 configurations are generated
! by the $DRT group, and 2 CI roots are found.
!
! This job tests the GUGA-CI code, note that gradients
! are not available for CI.
!
! State 1 EIGENvalue = -37.2798768747, c(1) = 0.932805.
! State 2 EIGENvalue = -37.0372982302, c(3) = 0.933057.
! The ground state dipole moment is 1.226756 Debye.
!
$CONTRL SCFTYP=CI $END
$SYSTEM TIMLIM=1 MEMORY=100000 $END
$BASIS GBASIS=STO NGAUSS=2 $END
$DATA
Methylene...1-A-1 state...CI...STO-2G basis
Cnv 2
Carbon 6.0
Hydrogen 1.0 0.0 0.82884 0.7079
$END
!
! Usually GUESS=MOREAD is used to get converged MOs
! from a prior run using any sort of SCF (except UHF).
!
$GUESS GUESS=HUCKEL $END
$DRT GROUP=C2V IEXCIT=6 NFZC=1 NDOC=3 NVAL=3 $END
! look at 1-1A1 and 2-1A1 states
$GUGDIA NSTATE=2 $END
$GUGDM NFLGDM(1)=1,1 $END
! EXAM06.
! 1-A-1 CH2 MCSCF methylene geometry optimization.
! The two configuration ansatz is the same as used in
! the fourth example.
!
! The optimization is done in internal coordinates,
! as NZVAR is non-zero. Since a explicit $ZMAT is
! given, these are used for the internal coordinates,
! rather than those used to enter the molecule in
! the $DATA. (Careful examination of this trivial
! triatomic's input shows that $ZMAT is equivalent
! to $DATA in this case. You would normally give
! $ZMAT only if it is somehow different.)
!
! This job tests the MCSCF wavefunction and gradient.
!
! At the initial geometry:
! The initial energy is -37.187342653,
! the FINAL E= -37.2562020530 after 8 iterations,
! the RMS gradient is 0.0256392.
!
! After 4 steps,
! FINAL E= -37.2581791665, RMS gradient=0.0000177,
! r(CH)=1.1243569, ang(HCH)=98.8161138, dipole=1.212792
!
$CONTRL SCFTYP=MCSCF RUNTYP=optimize NZVAR=3 COORD=ZMT $END
$SYSTEM TIMLIM=6 MEMORY=100000 $END
$BASIS GBASIS=STO NGAUSS=2 $END
$DATA
Methylene...1-A-1 state...MCSCF/STO-2G
Cnv 2
C
H 1 rCH
H 1 rCH 2 aHOH
rCH=1.09
aHOH=99.0
$END
$ZMAT IZMAT(1)=1,1,2, 1,1,3, 2,2,1,3 $END
!
! Normally one starts a MCSCF run with converged SCF
! orbitals, as Huckel orbitals normally do not converge.
! Even if they do converge, the extra iterations are
! very expensive, so use MOREAD for your runs!
!
$GUESS GUESS=HUCKEL $END
$MCSCF MAXIT=10 $END
!
! two active electrons in two active orbitals.
!
$DRT GROUP=C2V FORS=.TRUE. NMCC=3 NDOC=1 NVAL=1 $END
! EXAM 14.
! CI transition moments. Water, using RHF/STO-3G MOs.
! All orbitals are occupied, transition is 1-1A1 to 2-1A1.
!
! E(STATE 1)= -75.0101113548, E(STATE 2)= -74.3945819375
! Dipole LENGTH is <Q>=0.392614
! Dipole VELOCITY is <d/dQ>=0.368205
!
$CONTRL SCFTYP=CI RUNTYP=TRANSITN UNITS=BOHR $END
$SYSTEM TIMLIM=1 MEMORY=100000 $END
$BASIS GBASIS=STO NGAUSS=3 $END
$DATA
WATER MOLECULE...STO-3G...TRANSITION MOMENT
CNV 2
OXYGEN 8.0 0.0 0.0 0.0
HYDROGEN 1.0 0.0 1.428 -1.096
$END
! standard SD-CI calculation
$DRT1 GROUP=C2V IEXCIT=2 NFZC=1 NDOC=4 NVAL=2 $END
$TRANST $END
--- RHF ORBITALS --- GENERATED AT 09:24:04 18-FEB-88
WATER MOLECULE...STO-3G...TRANSITION MOMENT
E(RHF)= -74.9620539825, E(NUC)= 9.2384802989, 8 ITERS
$VEC1
1 1 9.94117078E-01...
$END
! EXAM 15.
! C2- diatom, in the electronic state doublet-pi-u.
! This illustrates a open shell SCF calculation, using
! fed in coupling coefficients, and the GVB/ROHF code.
!
! The FINAL energy is -75.5579181071 after 11 iterations.
!
$CONTRL SCFTYP=GVB MULT=2 ICHARG=-1 UNITS=BOHR $END
$SYSTEM TIMLIM=15 MEMORY=170000 $END
$BASIS GBASIS=DH NDFUNC=1 POLAR=DUNNING $END
$DATA
C2-...DOUBLET-PI-UNGERADE...OPEN SHELL SCF
DNH 4
CARBON 6.0 0.0 0.0 -1.233
$END
$GUESS GUESS=MOREAD NORB=30
NORDER=1 IORDER(5)=7,5,6 $END
$SCF NCO=5 NSETO=1 NO=2 COUPLE=.TRUE.
F(1)=1.0, 0.75
ALPHA(1)=2.0, 1.5, 1.00
BETA(1)=-1., -.75, -0.5 $END
--- RHF ORBITALS --- GENERATED AT 14:05:16THU MAR 24/88
CC R(C-C) = 2 * 1.233 BOHR BAS=831+1D
E(RHF)= -75.3856001855, E(NUC)= 14.5985401460, 18 ITERS
$VEC
1 1-7.06500288E-01...
$END
! EXAM 16.
! ROHF/GVB on Si 3-P state, using Gordon's 6-31G basis.
!
! The purpose of this example is two-fold, namely to
! show off the open shell capabilities of the GVB code,
! and to emphasize that the 6-31G basis for Si in GAMESS
! is Mark Gordon's version. The basis stored in GAMESS is
! completely optimized, whereas Pople's uses the core from
! from a 6-21G set, reoptimizing only the -31G part.
! The energy from Pople's basis would be only -288.828405.
!
! Jacobi diagonalization is intrinsically slow, but in this
! case results in pure subspecies in degenerate p irreps.
! In fact, these may be labeled in the highest Abelian
! subgroup of the atomic point group Kh.
!
! The FINAL energy is -288.8285729745 after 8 iterations.
!
$CONTRL SCFTYP=GVB MULT=3 $END
$SYSTEM TIMLIM=2 MEMORY=100000 KDIAG=3 $END
$BASIS GBASIS=N31 NGAUSS=6 $END
$DATA
Si...3-P term...ROHF in full Kh symmetry
Dnh 2
Silicon 14.
$END
$GUESS GUESS=HUCKEL $END
$SCF NCO=6 NSETO=1 NO=3 COUPLE=.TRUE.
F(1)=1.0, 0.333333333333333
ALPHA(1)=2.0, 0.66666666666667, 0.16666666666667
BETA(1)=-1.0, -0.33333333333333, -0.16666666666667
$END
! EXAM 17.
! Analytic hessian for an open shell SCF function.
! Methylene's 1-B-1 excited state.
! FINAL energy= -38.3334724780 after 7 iterations.
! The FREQuencies are 1224.19, 3563.44, 3896.23
!
$CONTRL SCFTYP=GVB MULT=1 RUNTYP=HESSIAN UNITS=BOHR $END
$SYSTEM TIMLIM=4 MEMORY=100000 $END
$SCF NCO=3 NSETO=2 NO(1)=1,1 NPAIR=0 $END
$ZMAT IZMAT(1)=1,1,2, 1,1,3, 2,2,1,3 $END
$GUESS GUESS=HUCKEL $END
$BASIS GBASIS=STO NGAUSS=3 $END
$DATA
METHYLENE...1-B-1 STATE...ROHF...STO-3G BASIS
CNV 2
CARBON 6.0 0.0 0.0 0.0041647278
HYDROGEN 1.0 0.0 1.8913952563 0.7563907037
$END
! EXAM 18.
! effective core potential...diatomic P2...RHF/CEP-31G*
! See Stevens,Basch,Krauss, J.Chem.Phys. 81,6026-33(1984).
! GAMESS FINAL E= -12.6956517413, RMS gradient=0.000354618
! A separate run gives E(P)= -6.32635, so De= 26.95 kcal/mol
!
$CONTRL SCFTYP=RHF RUNTYP=GRADIENT ECP=SBK NZVAR=1 $END
$SYSTEM TIMLIM=15 MEMORY=200000 $END
$GUESS GUESS=HUCKEL $END
$ZMAT IZMAT(1)=1,1,2 $END
$DATA
diatomic phosphorous
Dnh 4
PHOSPHORUS 15.0 0.0 0.0 0.9395
SBK
D 1
1 0.45 1.0
$END
! EXAM 19.
! Spin-orbit coupling example.
! This run duplicates the one electron result shown in Table
! 3 of T.R.Furlani,H.F.King, J.Chem.Phys. 82, 5577-83(1985).
! Note that the lower multiplicty DRT1 generates two CSFs,
! with one of the singlet-delta states formed by xx-yy, and
! the desired singlet-sigma-plus as the 2nd root: xx+yy.
! E(3s-) = -54.9382257056
! E(1s+) = -54.7988368491
! SOC(z) = -114.3851
!
$CONTRL SCFTYP=CI MULT=3 RUNTYP=SPINORBT UNITS=BOHR $END
$SYSTEM TIMLIM=2 MEMORY=100000 $END
! triplet-sigma-minus to singlet-sigma-plus SOC
$TRANST NFZC=3 NOCC=5 NUMVEC=1 NUMCI=2 IROOTS(1)=2,1 $END
$DRT1 GROUP=C4V IEXCIT=2 NFZC=3 NDOC=1 NVAL=1 $END
$DRT2 GROUP=C4V IEXCIT=2 NFZC=3 NALP=2 $END
!
! Since the 1e- spin orbit integrals cannot use L shells,
! we must input the 6-31G set for nitrogen by hand.
!
$DATA
Imidogen radical
Cnv 4
NITROGEN 7.0
S 6
1 4173.511460 0.001834772
2 627.457911 0.01399463
3 142.902093 0.06858655
4 40.234329 0.2322409
5 12.820213 0.4690699
6 4.390437 0.3604552
S 3
1 11.626362 -0.1149612
2 2.716280 -0.1691175
3 0.772218 1.145852
P 3
1 11.626362 0.06757974
2 2.716280 0.3239073
3 0.772218 0.7408951
S 1
1 0.212031 1.0
P 1
1 0.212031 1.0
HYDROGEN 1.0 0.0 0.0 1.9748
N31 6
$END
--- ROHF ORBITALS --- GENERATED AT 12:04:18 29 MAR 90 ( 88)
IMIDOGEN RADICAL
E(ROHF)= -54.9382257007, E(NUC)= 3.5446627507, 8 ITERS
$VEC1
1 1 9.97073281E-01...
$END
! EXAM 20.
! MP2 (frozen core) geometry optimization of NH3.
! Since GAMESS cannot compute MP2 gradients, this uses the
! slow nongradient TRUDGE optimization, in HINT coords.
!
! The optimized geometry from Table 6.2 in Hehre et al.,
! "Ab initio Molecular Orbital Theory" is
! r(N-H)=1.017 Angs, <(H-N-H)=106.3 deg.
! This run gives r(N-H)=1.0176 Angs, <(H-N-H)=106.21 deg.
! with a final E(MP2)=-56.3542104882 Eh
!
$CONTRL MPLEVL=2 COORD=HINT SCFTYP=RHF RUNTYP=TRUDGE $END
$SYSTEM TIMLIM=30 MEMORY=100000 $END
$TRUDGE OPTMIZ=GEOMETRY NPAR=2 IEX(1)=21,22 P(1)=1.02 $END
$GUESS GUESS=HUCKEL $END
$BASIS GBASIS=N31 NGAUSS=6 NDFUNC=1 $END
$DATA
NH3...6-31G*...MP2...TRUDGE OPTIMIZATION OF GEOMETRY
CNV 3
NITROGEN 7. LC 0.0 0.0 0.0 - O K
HYDROGEN 1. PCC 1.0 68.34 0.0 + O K I
$END
! EXAM 21.
! Open shell two configuration SCF analytic hessian.
! M.Duran, Y.Yamaguchi, H.F.Schaefer III
! J.Phys.Chem. 1988, 92, 3070-3075.
! Least motion insertion of CH into H2, which leads to
! a 3rd order hypersaddle point on the 2-B-1 surface.
!
! Literature values are
! FINAL E=-39.25104, C1=0.801, C2=-0.598
! FREQ= 4805i, 1793i, 1317i, 989, 2914, 3216
! GAMESS obtains
! FINAL E=-39.2510351248, C1=0.801141, C2=-0.598476
! FREQ= 4805.55i, 1793.09i, 1317.50i,
! FREQ= 988.80, 2913.51, 3216.43
!
$CONTRL SCFTYP=GVB MULT=2 RUNTYP=HESSIAN $END
$SYSTEM TIMLIM=25 MEMORY=100000 $END
$GUESS GUESS=MOREAD NORB=16 NORDER=1 IORDER(4)=6,4,5 $END
$SCF NCO=3 NSETO=1 NO=1 NPAIR=1 CICOEF(1)=0.7,-0.7 $END
$DATA
Insertion of CH into H2...OS-TCSCF ansatz...DZ basis
CNV 2
CARBON 6.0 0.0000000000 0.0000000000 -0.0001357549
S 6
1 4232.61 0.002029
2 634.882 0.015535
3 146.097 0.075411
4 42.4974 0.257121
5 14.1892 0.596555
6 1.9666 0.242517
S 1
1 5.1477 1.0
S 1
1 0.4962 1.0
S 1
1 0.1533 1.0
P 4
1 18.1557 0.018534
2 3.9864 0.115442
3 1.1429 0.386206
4 0.3594 0.640089
P 1
1 0.1146 1.0
HYDROGEN 1.0 0.0000000000 0.0000000000 1.0922959062
DH 0 1.2 1.2
HYDROGEN 1.0 0.0000000000 0.4152229538 -1.4824967459
DH 0 1.2 1.2
$END
--- these are 2-A1 ROHF vectors ---
--- ROHF ORBITALS --- GENERATED AT 08:23:42 27 JUN 90 (178)
E(ROHF)= -39.2316245004, E(NUC)= 8.0760320442, 12 ITERS
$VEC
1 1 6.01223299E-01...
! EXAM22.
!
! 3-A-2 H3CN UMP2/6-31G*//UHF/6-31G*
!
! The FINAL UHF energy= -94.0039683676 after 13 iters.
! The E(MP2) energy= -94.2315757758
!
$CONTRL SCFTYP=UHF MULT=3 RUNTYP=ENERGY MPLEVL=2
COORD=ZMT $END
$SYSTEM TIMLIM=30 MEMORY=300000 $END
$BASIS GBASIS=N31 NGAUSS=6 NDFUNC=1 NPFUNC=0 $END
$GUESS GUESS=HUCKEL $END
$DATA
Methylnitrene...UHF/6-31G* structure
Cnv 3
N
C 1 rCN
H 2 rCH 1 aHCN
H 2 rCH 1 aHCN 3 120.0
H 2 rCH 1 aHCN 3 -120.0
rCN=1.4329216
rCH=1.0876477
aHCN=110.21928
$END
! EXAM23.
! semiempirical calculation, using the MOPAC/GAMESS combo
! AM1 gets the geometry disasterously wrong!
!
! initial geometry, MNDO AM1 PM3
! FINAL HEAT OF FORMATION 105.14088 96.45997 46.89387
! RMS gradient 0.1405472 0.1008587 0.0366232
! final geometry (# steps), 8 12 6
! FINAL HEAT OF FORMATION 46.45649 -1.81730 -2.79646
! RMS gradient 0.0000425 0.0000063 0.0000172
! r(SiH) 1.42117 1.45811 1.52103
! a(HSiH) 101.960 120.000 96.276
!
$CONTRL RUNTYP=OPTIMIZE COORD=ZMT ICHARG=-1 $END
$SYSTEM TIMLIM=5 MEMORY=200000 $END
$BASIS GBASIS=PM3 $END
$DATA
Silyl anion...comparison of semiempirical models
Cnv 3
Si
H 1 rSiH
H 1 rSiH 2 aHSiH
H 1 rSiH 2 aHSiH 3 aHSiH -1
rSiH=1.15
aHSiH=110.0
$END
! EXAM24.
! Self-consistent reaction field test, of water in water.
! Cavity radius is calculated from the 1.00 g/cm**3 density.
! FINAL energy is -74.9666740755 after 12 iterations
! Induced dipole= -0.03663, RMS gradient= 0.033467686
!
$contrl scftyp=rhf runtyp=gradient coord=zmt $end
$system memory=300000 $end
$basis gbasis=sto ngauss=3 $end
$guess guess=huckel $end
$scrf radius=1.93 dielec=80.0 $end
$data
water in water, arbitrary geometry
Cnv 2
O
H 1 rOH
H 1 rOH 2 aHOH
rOH = 0.95
aHOH = 104.5
$end
! EXAM25.
! Illustration of coordinate systems for geometry searches.
! Arbitrary molecule, chosen to illustrate ring, methyl on
! ring, methine H10, imino in ring, methylene in ring.
!
! H8 H9
! \|
! H7-C6 O1---O5 H13
! \ / \ /
! C2 C4
! / \ / \
! H10 N3 H12
! |
! H11
!
! The initial AM1 energy is -48.6594935
! initial RMS final E final RMS #steps
! Cartesians 0.0200113 -48.7022520 0.0000283 50
! dangling Z-mat 0.0600637 ... OO bond crashes on 1st step
! good Z-matrix 0.0232915 -48.7022508 0.0000299 20
! nat. internals 0.0209855 -48.7022565 0.0000192 15
!
$contrl scftyp=rhf runtyp=optimize coord=zmt $end
$system memory=300000 $end
$statpt hess=guess nstep=100 nprt=-1 npun=-2 $end
$basis gbasis=am1 $end
$guess guess=huckel $end
$data
Illustration of coordinate systems
C1
O
C 1 rCOa
N 2 rCNa 1 aNCO
C 3 rCNb 2 aCNC 1 wCNCO
O 4 rCOb 3 aOCN 2 wOCNC
C 2 rCC 1 aCCO 5 wCCOO
H 6 rCH1 2 aHCC1 1 wHCCO1
H 6 rCH2 2 aHCC2 1 wHCCO2
H 6 rCH3 2 aHCC3 1 wHCCO3
H 2 rCHa 1 aHCOa 5 wHCOOa
H 3 rNH 2 aHNC 1 wHNCO
H 4 rCHb 5 aHCOb 1 wHCOOb
H 4 rCHc 5 aHCOc 1 wHCOOc
rCOa=1.43
rCNa=1.47
rCNb=1.47
rCOb=1.43
aNCO=106.0
aCNC=104.0
aOCN=106.0
wCNCO=30.0
wOCNC=-30.0
rCC=1.54
aCCO=110.0
wCCOO=-150.0
rCH1=1.09
rCH2=1.09
rCH3=1.09
aHCC1=109.0
aHCC2=109.0
aHCC3=109.0
wHCCO1=60.0
wHCCO2=-60.0
wHCCO3=180.0
rCHa=1.09
aHCOa=110.0
wHCOOa=100.0
rNH=1.01
aHNC=110.0
wHNCO=170.0
rCHb=1.09
rCHc=1.09
aHCOb=110.0
aHCOc=110.0
wHCOOb=150.0
wHCOOc=-100.0
$end
To use Cartesian coordinates:
--- $contrl nzvar=0 $end
To use conventional Z-matrix, with dangling O-O bond:
--- $contrl nzvar=33 $end
To use well chosen internals, with all ring bonds closed:
--- $contrl nzvar=33 $end
--- $zmat izmat(1)=1,1,2, 1,2,3, 1,3,4, 1,4,5, 1,5,1,
2,1,2,3, 2,5,4,3, 3,5,1,2,3, 3,1,5,4,3,
1,6,2, 2,6,2,1, 3,6,2,1,5,
1,6,7, 1,6,8, 1,6,9, 2,7,6,2, 2,8,6,2, 2,9,6,2,
3,7,6,2,1, 3,8,6,2,1, 3,9,6,2,1,
1,10,2, 2,10,2,1, 3,10,2,1,5,
1,11,3, 2,11,3,2, 3,11,3,2,1,
1,12,4, 2,12,4,5, 3,12,4,5,1,
1,13,4, 2,13,4,5, 3,13,4,5,1 $end
To use natural internal coordinates:
$contrl nzvar=44 $end
$zmat izmat(1)=1,1,2, 1,2,3, 1,3,4, 1,4,5, 1,5,1, ! ring !
2,5,1,2, 2,1,2,3, 2,2,3,4, 2,3,4,5, 2,4,5,1,
3,5,1,2,3, 3,1,2,3,4, 3,2,3,4,5, 3,3,4,5,1, 3,4,5,1,2,
1,2,6, 2,6,2,1, 2,6,2,3, 4,6,2,1,3, ! methyl C !
1,6,7, 1,6,8, 1,6,9, ! methyl Hs !
2,7,6,8, 2,8,6,9, 2,9,6,7, 2,9,6,2, 2,7,6,2, 2,8,6,2,
3,7,6,2,1,
1,10,2, 2,10,2,1, 2,10,2,3, 2,10,2,6, ! methine !
1,11,3, 2,11,3,2, 2,11,3,4, 4,11,3,2,4, ! imino !
1,12,4, 1,13,4, ! methylene !
2,12,4,13, 2,12,4,3, 2,13,4,3, 2,12,4,5, 2,13,4,5
ijS(1)=1,1, 2,2, 3,3, 4,4, 5,5, ! ring !
6,6, 7,6, 8,6, 9,6,10,6,
7,7, 8,7, 9,7,10,7,
11,8,12,8,13,8,14,8,15,8,
11,7,12,9, 14,9,15,9,
16,10, 17,11,18,11, 19,12, ! methyl C !
20,13, 21,14, 22,15, ! methyl Hs !
23,16, 24,16, 25,16, 26,16, 27,16, 28,16,
23,17, 24,17, 25,17,
24,18, 25,18,
26,19, 27,19, 28,19,
27,20, 28,20,
29,21,
30,22, 31,23,32,23,33,23, 32,24,33,24, ! methine !
34,25, 35,26,36,26, 37,27, ! imino !
38,28, 39,29, ! methylene !
40,30, 41,30, 42,30, 43,30, 44,30,
41,31, 42,31, 43,31, 44,31,
41,32, 42,32, 43,32, 44,32,
41,33, 42,33, 43,33, 44,33
Sij(1)=1.0, 1.0, 1.0, 1.0, 1.0, ! ring !
1.0, -0.8090, 0.3090, 0.3090, -0.8090,
-1.1180, 1.8090, -1.8090, 1.1180,
0.3090, -0.8090, 1.0, -0.8090, 0.3090,
-1.8090, 1.1180, -1.1180, 1.8090,
1.0, 1.0,-1.0, 1.0, ! methyl C !
1.0, 1.0, 1.0, ! methyl Hs !
1.0, 1.0, 1.0,-1.0,-1.0,-1.0,
2.0,-1.0,-1.0,
1.0,-1.0,
2.0,-1.0,-1.0,
1.0,-1.0,
1.0,
1.0, 2.0,-1.0,-1.0, 1.0,-1.0, ! methine !
1.0, 1.0,-1.0, 1.0, ! imino !
1.0, 1.0, ! methylene !
4.0, 1.0, 1.0, 1.0, 1.0,
1.0,-1.0, 1.0,-1.0,
1.0, 1.0,-1.0,-1.0,
1.0,-1.0,-1.0, 1.0 $end
