G92. Add to summary about 2e-integrals.
- From: aiba $#at#$ volta.vmsmail.ethz.ch (Aiaz Bakassov, Phys. Chem., ETH
- Subject: G92. Add to summary about 2e-integrals.
- Date: Thu, 7 Jul 1994 15:53:44 +0200
After I posted a summary and described
MY failure with extraction of 2e-integrals
from G92, one additional mail arrived.
After some correspondence, it was decided
that the SCF=DIRECT option must be removed.
I have removed it and
I have got 2e-integrals in a quite readable format!
I thank my adviser who wished
to keep anonymous.
I must stress that the failure is ENTIRELY MINE,
and in fact EXACTLY the same advise
came originally from Alberto Gobbi
who of course couldn't know beforehand
that I will be using his advice
together with SCF=DIRECT which caused
The net result is this --
to get 2e-integrals, input the
following line (no SCF=DIRECT!!!):
#RHF 6-31G EXTRALINKS=L316 NORAFF Symm=NOINT
The only thing makes me wonder.
I thought before of removing SCF=DIRECT
(it seemed natural)
however in G92 Programmer's Guide (page 41)
it is written in front of the option "InCore":
"Insists that the SCF be done storing
the full integral list in memory.
THIS IS DONE AUTOMATICALLY IN A DIRECT
SCF CALCULATION IF SUFFICIENT MEMORY IS
See what it means ?
It means that if SCF=DIRECT
then integrals AUTOMATICALLY
are stored in memory.
If they are stored in memory,
then they are available for further
manipulations, right ?
For printing out, for instance.
It has really misled me.
And I am upset -- it seems
that each line of the manual
should checked by trial and error
for the precise meaning
of words. Who has time for that ?
Thanks once more,