Hi,<br><br>Thank you. I do have access to the 2010.1 sources so, if you send me the necessary files I can make the modifications locally.<br><br>Best regards,<br>Hugo<br><br><div class="gmail_quote">On Fri, Aug 5, 2011 at 1:10 PM, Tatiana Korona <span dir="ltr"><<a href="mailto:tania@tiger.chem.uw.edu.pl">tania@tiger.chem.uw.edu.pl</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Dear Hugo,<br>
<br>
Just one word of warning. Transition density matrices were so far used with SYMMETRIC one-electron operator matrices, like DM, QM etc. They cannot be used with ANTIsymmetric operators, like VELO (D/DX etc.). The reason for this are interchanged virt-occ and occ-virt parts or the density (what of course has no effect for traces tr D Op, if Op is symmetric, so all electric dipole transition moments in the usual length gauge etc. are OK). I have found this only recently. Sorry for that. If you have access to the molpro source 2010.1 and can recompile your own copy, I can send you the fortran files which should be exchanged, then you will have all your problems solved (problems with saving densities and reading them by matrop, too). Before I make an official patch, I have to contact other developers, because I made changes that can affect their code, and some of them are on vacation now.<br>
<br>
Best wishes,<br>
<br>
Tatiana<br>
<br>
On Tue, 12 Jul 2011, hugo wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dear Tatiana,<br>
<br>
Thank you for your answer. I'll try the 2006.1 release until this is fixed in newer versions.<br>
<br>
Best regards,<br>
Hugo<div><div></div><div class="h5"><br>
<br>
On 07/11/2011 01:23 PM, Tatiana Korona wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dear Hugo,<br>
<br>
The problem results from changing of functionality of one subroutine (not mine) which serves in molpro to save density<br>
matrices. Prof. Werner promised to look at this when he will be less busy. It is unfortunately difficult to propose a<br>
workaround for the time being apart from using a 2006.1 version instead of 2010.1 one. I will see him on WATOC next week<br>
and hope to find a solution during this conference. I will keep you informed if something positive changes in 2010.1.<br>
<br>
Best wishes,<br>
<br>
Tatiana<br>
<br>
On Tue, 5 Jul 2011, Hugo Martiniano wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Thank you for your attention.<br>
<br>
Regards,<br>
Hugo<br>
<br>
On Mon, Jul 4, 2011 at 3:12 AM, Tatiana Korona<br>
<<a href="mailto:tania@tiger.chem.uw.edu.pl" target="_blank">tania@tiger.chem.uw.edu.pl</a>><u></u>wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dear Hugo,<br>
<br>
Yes, it seems to be an error introduced after the version 2006.4.<br>
I will look how to solve this problem.<br>
<br>
Best wishes,<br>
<br>
Tatiana<br>
<br>
<br>
On Sat, 2 Jul 2011, Hugo Martiniano wrote:<br>
<br>
Hello,<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
I am trying to save a transition density matrix from an EOM-CCSD<br>
calculation.<br>
If I am not mistaken, according to the manual, this can be done by setting<br>
trans=2<br>
and choosing a record, either with the dm directive or the densave<br>
keyword.<br>
<br>
The problem is that I can't get any one of these to work.<br>
<br>
With densave I get the following error:<br>
<br>
******************************<u></u>******************************<u></u>****<br>
********************<br>
<br>
<br>
EOM-CCSD properties program. Author: T. Korona (2003)<br>
<br>
******************************<u></u>******************************<u></u>****<br>
********************<br>
<br>
<br>
<br>
Records on file 2<br>
<br>
IREC NAME TYPE OFFSET LENGTH IMPLEMENTATION EXT PREV<br>
PARENT MPP_STATE<br>
1 500 VAR <a href="tel:4096.%20102002" value="+14096102002" target="_blank">4096. 102002</a>. df 0 0<br>
0 1<br>
2 610 BASINP 106098. 8192. df 0 0<br>
0 1<br>
3 700 GEOM 114290. 27110. df 0 0<br>
0 1<br>
4 1000 BASIS 141400. 933. df 0 0<br>
0 1<br>
5 2100 RHF 142333. 7552. df 0 0<br>
0 1<br>
6 5555 149885. 2527727. df 0 0<br>
0 1<br>
7 5040 2677612. 300. df 0 0<br>
0 1<br>
8 5700 2677912. 10624. df 0 0<br>
0 1<br>
<br>
READM: RECORD= 5700.2 EXTENSION= 0 OFFSET= 0. ADDRESS=<br>
2677912. LEN= 20 IMPLEMENTATION=df<br>
CURRENT WRITTEN FILE LENGTH: 2677912.TRYING TO READ UP TO ADDRESS:<br>
2677932.<br>
? Error<br>
? Trying to read beyond end of file<br>
? The problem occurs in readm<br>
<br>
ERROR EXIT<br>
CURRENT STACK: MAIN<br>
<br>
<br>
If I use dm I get a different one:<br>
<br>
Natural orbitals saved on record 5600.2 (orbital set 2)<br>
<br>
Bra < 1.1|, Ket | 2.1><br>
<br>
EOM-CCSD trans. dipole moment 0.00000000 0.00000000<br>
0.36439832 < 1.1| 2.1><br>
<br>
<br>
?ERROR IN WRITE_DUMP: OVERWRRITING RECORD 5600.2 ENTRY<br>
DENSITY/TRANSITION SET= 2 WITH ILLEGAL LENGTH. NEW= 577 OLD= 300<br>
<br>
ERROR EXIT<br>
CURRENT STACK: MAIN<br>
<br>
<br>
A sample input file is below:<br>
<br>
memory,100,m<br>
file,2,w_tdm.wfu<br>
symmetry,nosym<br>
geometry<br>
3<br>
<br>
O -0.32532 0.22659 0.00000<br>
H 0.62464 0.21785 0.00000<br>
H -0.65067 -0.66595 0.00000<br>
endg<br>
basis=vdz;hf;{ccsd;eom,2.1,**<u></u>trans=2,densave=5600.2;}<br>
<br>
<br>
Am I missing something in my input or is there a bug in the program?<br>
<br>
The molpro version used is 2010.1 patch level 20 (latest).<br>
<br>
Thank you,<br>
Hugo<br>
<br>
<br>
</blockquote>
Dr. Tatiana Korona<br>
<a href="http://tiger.chem.uw.edu.pl/**staff/tania/index.html" target="_blank">http://tiger.chem.uw.edu.pl/**<u></u>staff/tania/index.html</a><<a href="http://tiger.chem.uw.edu.pl/staff/tania/index.html" target="_blank">http://<u></u>tiger.chem.uw.edu.pl/staff/<u></u>tania/index.html</a>><br>
Quantum Chemistry Laboratory<br>
University of Warsaw<br>
Pasteura 1, PL-02-093 Warsaw, POLAND<br>
<br>
<br>
`The man who makes no mistakes does not usually make anything.'<br>
Edward John Phelps (1822-1900)<br>
<br>
</blockquote>
<br>
</blockquote>
<br>
Dr. Tatiana Korona <a href="http://tiger.chem.uw.edu.pl/staff/tania/index.html" target="_blank">http://tiger.chem.uw.edu.pl/<u></u>staff/tania/index.html</a><br>
Quantum Chemistry Laboratory<br>
University of Warsaw<br>
Pasteura 1, PL-02-093 Warsaw, POLAND<br>
<br>
<br>
`The man who makes no mistakes does not usually make anything.'<br>
Edward John Phelps (1822-1900)<br>
<br>
</blockquote>
<br>
</div></div></blockquote><div><div></div><div class="h5">
<br>
Dr. Tatiana Korona <a href="http://tiger.chem.uw.edu.pl/staff/tania/index.html" target="_blank">http://tiger.chem.uw.edu.pl/<u></u>staff/tania/index.html</a><br>
Quantum Chemistry Laboratory<br>
University of Warsaw<br>
Pasteura 1, PL-02-093 Warsaw, POLAND<br>
<br>
<br>
`The man who makes no mistakes does not usually make anything.'<br>
Edward John Phelps (1822-1900)<br>
</div></div></blockquote></div><br>