[molpro-user] Molpro Help

Brandon Buncher brandon.buncher.12 at gmail.com
Mon Jul 15 16:35:08 BST 2013


Hi!  I've been trying to perform a variety of energy calculations on
several different bond lengths of a molecule, beginning with HF and moving
through CCSD(T), CASSCF, and MRCI.  However, whenever I try to do this, the
calculation becomes stuck on "recomputing integrals since basis changed"
right after setting all variables.  Is there a way to get around this?  I
have tried changing the number of bond length options and removing
calculation types, but they all become stuck here.  The input and output
files are below.  Thanks so much!

INPUT:
****,I3
memory,20,M
set,seward_mem=3000000
R1=[1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0
3.1 3.2 3.3 3.4 3.5] ang, R2=3.2446 ang,Theta=180.0
i=1
j=1
k=1
l=1
geometry={
I1;
I2,I1,R1(i,j,k,l);
I3,I2,R2,I1,Theta
}
basis=vtz
do i=1,#R1
hf
e(i)=energy
method(i)=program
ENDDO
table,method,e
title,Results for I3 3_2446 $basis,basis=$basis

do j=1,#R1
hf
ccsd(t)
e(j)=energy
method(j)=program
ENDDO
table,method,e
title,Results for I3 3_2446 $basis,basis=$basis

do k=1,#R1
hf
casscf
e(k)=energy
method(k)=program
ENDDO
table,method,e
title,Results for I3 3_2446 $basis,basis=$basis

do l=1,#R1
hf
mrci
e(l)=energy
method(l)=program
ENDDO
table,method,e
title,Results for I3 3_2446 $basis,basis=$basis
---



OUTPUT:
 Variables initialized (766), CPU time= 0.00 sec
 Commands  initialized (545), CPU time= 0.01 sec, 513 directives.
 Default parameters read. Elapsed time= 0.08 sec

 Checking input...
 Passed
1
                                         ***  PROGRAM SYSTEM MOLPRO  ***
                         Copyright, University College Cardiff Consultants
Limited, 2008

                                    Version 2012.1 linked 5 Sep 2012
15:26:37


 **********************************************************************************************************************************
 LABEL *
I3

 Linux-2.6.18-308.16.1.el5/molecular07.cluster(x86_64) 64 bit mpp
version                DATE: 10-Jul-13          TIME: 16:41:43
 **********************************************************************************************************************************

 SHA1:             2c68d29c09da70e1723824271fadde4bcd5f07a0
 **********************************************************************************************************************************
 SETTING SEWARD_MEM     =
0.30000000D+07
 SETTING R1(1)          =         1.30000000
ANG
 SETTING R1(2)          =         1.40000000
ANG
 SETTING R1(3)          =         1.50000000
ANG
 SETTING R1(4)          =         1.60000000
ANG
 SETTING R1(5)          =         1.70000000
ANG
 SETTING R1(6)          =         1.80000000
ANG
 SETTING R1(7)          =         1.90000000
ANG
 SETTING R1(8)          =         2.00000000
ANG
 SETTING R1(9)          =         2.10000000
ANG
 SETTING R1(10)         =         2.20000000
ANG
 SETTING R1(11)         =         2.30000000
ANG
 SETTING R1(12)         =         2.40000000
ANG
 SETTING R1(13)         =         2.50000000
ANG
 SETTING R1(14)         =         2.60000000
ANG
 SETTING R1(15)         =         2.70000000
ANG
 SETTING R1(16)         =         2.80000000
ANG
 SETTING R1(17)         =         2.90000000
ANG
 SETTING R1(18)         =         3.00000000
ANG
 SETTING R1(19)         =         3.10000000
ANG
 SETTING R1(20)         =         3.20000000
ANG
 SETTING R1(21)         =         3.30000000
ANG
 SETTING R1(22)         =         3.40000000
ANG
 SETTING R1(23)         =         3.50000000
ANG
 SETTING R2             =         3.24460000
ANG
 SETTING THETA          =
180.00000000
 SETTING I              =
1.00000000
 SETTING J              =
1.00000000
 SETTING K              =
1.00000000
 SETTING L              =
1.00000000

 Variable memory set to   20000000 words,  buffer space   230000 words

 SETTING BASIS          =    VTZ

 DO I                   =         1.00000000


 Recomputing integrals since basis changed


-- 
Brandon Buncher
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20130715/8be8e670/attachment.html>


More information about the Molpro-user mailing list