From owner-chemistry@ccl.net Sun Mar 27 13:45:01 2016 From: "Igors Mihailovs igors.mihailovs0-*-gmail.com" To: CCL Subject: CCL:G: Problems with a TD RO-CAM-B3LYP single point calculation Message-Id: <-52124-160327125424-7001-JnHS2dVHOQdm96apZsnjqQ,server.ccl.net> X-Original-From: Igors Mihailovs Content-Type: multipart/alternative; boundary=001a1140aef296ad4f052f0aa508 Date: Sun, 27 Mar 2016 19:53:58 +0300 MIME-Version: 1.0 Sent to CCL by: Igors Mihailovs [igors.mihailovs0 a gmail.com] --001a1140aef296ad4f052f0aa508 Content-Type: text/plain; charset=UTF-8 Dear Federico, We have run into OrtVc1 failed when we were working with Polarizability calculations. Only information I have found on the Internet is available here: http://www.somewhereville.com/?p=2175. Disabling fast multipole method (by "NoFMM" keyword which is denoted as "obsolete" keyword in G09 Reference) helped us. We, however, wrote to Gaussian Tech support and received further advice which I am posting here. Hello Igors, > > This is an issue we found in the part of the code that performs electric > field CPHF (as in "Polar", "Freq=Raman", etc.). The issue is triggered by > the combination of the use of molecular symmetry and the use of the FMM > techniques. For a case with no symmetry, there is no problem using FMM in > this part of the code. We will have this issue fixed in the upcoming G09 > release. > > If you are running a job that: > > - performs electric field CPHF (as in "Polar", "Freq=Raman", etc.) AND > > - on a system has symmetry, AND > > - the system is large enough that FMM techniques would be used, > > then what I recommend is to run the job turning off the use of the FMM > techniques only in Link 1002 (where the CPHF procedure is carried out) and > avoid running into the issue that we are discussing here. Please, try > adding "IOp(10/63=1)" as a workaround for the specific jobs for which the 3 > points above apply. Turning off FMM only in Link 1002 allows you to still > take advantage of these acceleration techniques in other parts of the code, > instead of turning off FMM in general as you did with "NoFMM". > Sincerely, Fernando R. Clemente, Ph.D. Technical Support Gaussian, Inc. help++gaussian.com I have no idea about why You have problems with TD ROKS calculation, but maybe You should write to Gaussian, too, for more explanations. No threats detected. www.avast.com <#m_2975292413568489481_DDB4FAA8-2DD7-40BB-A1B8-4E2AA1F9FDF2> With best wishes, Igors Mihailovs (engineer / PhD student) Institute of Solid State Physics University of Latvia 2016-03-27 2:02 GMT+02:00 Federico Javier Hernandez federico2099||gmail.com : > > Sent to CCL by: "Federico Javier Hernandez" [federico2099..gmail.com] > Does anyone know about the "OrtVc1 failed #1. Error termination via Lnk1e > in > /usr/local/g09/l914.exe" from a TD-DFT ROCAM-B3LYP calculation? > > I am trying to optimise the wavefunction in the third excited state using > the > rocam-b3lyp/aug-cc-pvdz level of theory and I get this output error: > > Convergence on wavefunction: 0.001000000000000 > *** WARNING: Number of orthogonal guesses is 2 > Iteration 1 Dimension 2 NMult 2 > Vector 1 SDiff=1.00D+00 TDiff=1.00D+00 assumed to be singlet. > Vector 2 SDiff=1.00D+00 TDiff=1.00D+00 assumed to be triplet. > CISAX will form 2 AO SS matrices at one time. > FoFJK: IHMeth= 1 ICntrl= 0 DoSepK=T KAlg= 1 I1Cent= 0 FoldK=F > FoFJK: IHMeth= 1 ICntrl= 0 DoSepK=T KAlg= 1 I1Cent= 0 FoldK=F > OrtVc1: Ph=1 IOff= 0 IPass=20 DotMx1= 1.68D-03 > OrtVc1: Ph=1 M= 2110 NPass=20 Test1= 1.00D+00 Small= 2.11D-09 VSmall= > 1.00D-12 > OrtVc1 failed #1. > Error termination via Lnk1e in /usr/local/g09/l914.exe at Fri Mar 25 > 18:32:06 > 2016. > > Thank you in advance. > > Federico> > > --001a1140aef296ad4f052f0aa508 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Dear Federico,

We have run int= o OrtVc1 failed when we were working with Polarizability calculations. Only= information I have found on the Internet is available here:=C2=A0http://www.s= omewhereville.com/?p=3D2175.

Disabling fast multipole= method (by "NoFMM" keyword which is denoted as "obsolete&qu= ot; keyword in G09 Reference) helped us. We, however, wrote to Gaussian Tec= h support and received further advice which I am posting here.

Hello Ig= ors,

This is an issue we found in the part of the code th= at=20 performs electric field CPHF (as in "Polar", "Freq=3DRaman&q= uot;, etc.). The=20 issue is triggered by the combination of the use of molecular symmetry=20 and the use of the FMM techniques. For a case with no symmetry, there is no problem using FMM in this part of the code. We will have this issue=20 fixed in the upcoming G09 release.

If you are runn= ing a job that:

- performs electric field CPHF (as= in "Polar", "Freq=3DRaman", etc.) AND

- on a system has symmetry, AND

- the sys= tem is large enough that FMM techniques would be used,

=
then what I recommend is to run the job turning off the use of the FMM=20 techniques only in Link 1002 (where the CPHF procedure is carried out)=20 and avoid running into the issue that we are discussing here. Please,=20 try adding "IOp(10/63=3D1)" as a workaround for the specific jobs= for=20 which the 3 points above apply. Turning off FMM only in Link 1002 allows you to still take advantage of these acceleration techniques in other=20 parts of the code, instead of turning off FMM in general as you did with "NoFMM".

Sincerely,
Fernando R. Clemente, Ph.D.
Technical Support
Gaussian, Inc.

I have no idea about why You have problems with = TD ROKS calculation, but maybe You should write to Gaussian, too, for more = explanations.


No threats det= ected. w= ww.avast.com
=
With best wishes,
Igors Mih= ailovs (engineer / PhD student)
Institute of Solid State Phys= ics
University of Latvia

=

2016-03-27 2:02 GMT+02:00 Federico Javier He= rnandez federico2099||gmail.com <o= wner-chemistry++ccl.net>:
Sent to CCL by: "Federico Javier Hernandez" [federico2099..gmail.com]<= br> Does anyone know about the "OrtVc1 failed #1. Error termination via Ln= k1e in
/usr/local/g09/l914.exe" from a TD-DFT ROCAM-B3LYP calculation?

I am trying to optimise the wavefunction in the third excited state using t= he
rocam-b3lyp/aug-cc-pvdz level of theory and I get this output error:

Convergence on wavefunction: 0.001000000000000
*** WARNING: Number of orthogonal guesses is 2
Iteration 1 Dimension 2 NMult 2
Vector 1 SDiff=3D1.00D+00 TDiff=3D1.00D+00 assumed to be singlet.
Vector 2 SDiff=3D1.00D+00 TDiff=3D1.00D+00 assumed to be triplet.
CISAX will form 2 AO SS matrices at one time.
FoFJK: IHMeth=3D 1 ICntrl=3D 0 DoSepK=3DT KAlg=3D 1 I1Cent=3D 0 FoldK=3DF FoFJK: IHMeth=3D 1 ICntrl=3D 0 DoSepK=3DT KAlg=3D 1 I1Cent=3D 0 FoldK=3DF OrtVc1: Ph=3D1 IOff=3D 0 IPass=3D20 DotMx1=3D 1.68D-03
OrtVc1: Ph=3D1 M=3D 2110 NPass=3D20 Test1=3D 1.00D+00 Small=3D 2.11D-09 VSm= all=3D 1.00D-12
OrtVc1 failed #1.
Error termination via Lnk1e in /usr/local/g09/l914.exe at Fri Mar 25 18:32:= 06
2016.

Thank you in advance.

Federico



-=3D This is automatically added to each message by the mailing script =3D-=
E-mail to subscribers: CHEMISTRY++ccl.n= et or use:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/s= end_ccl_message

E-mail to administrators: CHEM= ISTRY-REQUEST++ccl.net or use
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/s= end_ccl_message

Subscribe/Unsubscribe:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/chemistry/sub_un= sub.shtml

Before posting, check wait time at: http://www.ccl.net

Job: http://www.ccl.net/jobs
Conferences: http://server.ccl.net/chemist= ry/announcements/conferences/

Search Messages: http://www.ccl.net/chemistry/sear= chccl/index.shtml
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/spammers.txt

RTFI: http://www.ccl.net/chemistry/aboutccl/ins= tructions/



--001a1140aef296ad4f052f0aa508--