statistics

Statistics running TM61

Turbomole

Excited States (everything is one-core calculation)

1) Si2N2C50H54 (total 108), Ci, b3-lyp/def2-SVPD for all atoms, $scfconv 8; $denconv 1d-7; $scfinstab rpas; $soes all 7;:
a. grid m5: #PBS -l pmem=4500mb, $rpacor 4000, total wall-time : 4 days 13 hours 15 minutes and 43 seconds
b. grid m4: #PBS -l pmem=16500mb, $rpacor 15000, total wall-time : 4 days 1 hours 12 minutes and 14 seconds

# Electronic excitation spectrum of title, IRREP au singlet excitations
# excitation energy / eV  oscillator strength (length rep.)
#                    m5                      vs                      m4          (dft grid)
0.21648826016345E+01   0.12190104564554E+01     0.21649214653568E+01   0.12191490074850E+01
0.27824024075284E+01   0.26225590207116E-01     0.27825831544570E+01   0.26248868240073E-01
0.31791271449156E+01   0.12657944246559E+00     0.31794387817247E+01   0.12663991259845E+00
0.35393525583827E+01   0.12027302740696E-02     0.35395929317423E+01   0.12374855235510E-02
0.35460552106241E+01   0.19503312705987E+00     0.35461932670362E+01   0.19495707230768E+00
0.36603917077789E+01   0.48894905511338E-03     0.36603450698369E+01   0.48878861479965E-03
0.39046424759811E+01   0.93862670672506E-04     0.39047564557052E+01   0.94933369578423E-04

Almost no difference in excitation energies!

Frequencies with aoforce (aka analytical) (everything is one-core calculation)

1) Si2N2C50H54 (total 108), Ci, b3-lyp/m5/def2-SVP for all atoms,
#PBS -l pmem=5000mb
$scfconv 8; $denconv 1d-7; $scfconv 8; $denconv 1d-7; $maxcor 4000:
total wall-time : 6 days 15 hours 10 minutes and 20 seconds (without prior dscf, of course)