From owner-chemistry@ccl.net Sat Oct 20 05:54:01 2007 From: "=?ISO-8859-1?Q?Sina_T=FCreli?= sinatureli[*]gmail.com" To: CCL Subject: CCL: Large Computation Time Leap in Mopac and A Question About pm6 Message-Id: <-35428-071020054956-4371-debSR5paiocSWWZKWnc0PQ.@.server.ccl.net> X-Original-From: "=?ISO-8859-1?Q?Sina_T=FCreli?=" Content-Type: multipart/alternative; boundary="----=_Part_4387_6377840.1192873783791" Date: Sat, 20 Oct 2007 12:49:43 +0300 MIME-Version: 1.0 Sent to CCL by: "=?ISO-8859-1?Q?Sina_T=FCreli?=" [sinatureli__gmail.com] ------=_Part_4387_6377840.1192873783791 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Yes the issue was something similar to that (I needed adjust the molecule to reduce its charge and also use the keyword CHARGE), James J. P. Stewart has kindly answered my question on that matter. Thanks though. On 10/19/07, David Gallagher gallagher.da++gmail.com < owner-chemistry::ccl.net> wrote: > > > Sent to CCL by: David Gallagher [gallagher.da[A]gmail.com] > One possible reason for running very slowly in previous versions of > MOPAC2007, is using the default RHF for a metal containing compound > with an odd number of electrons. If it is supposed to be a radical, > then you should try UHF for these situations. The latest version of > MOPAC2007 can automatically adjust the settings for metal containing > compounds with an odd number of electrons. > > If this is not the issue, then if you send me the molecule file, I > would be happy to take a closer look. > > Regards, > David Gallagher > CAChe Research > > > At 12:15 AM 10/15/2007, Sina T reli sinatureli:_:gmail.com wrote: > > >Sent to CCL by: "Sina T reli" [sinatureli!^!gmail.com] > > > >- I am using mopac2007 to do some geometry optimizations with mndod > >and pm6. My first subject was a molecule of size 141 which was > >handled quite nicely and fast and ended about in 30 minutes. However > >when I added a ligand that was about of 20 atoms in size, the > >calculation time leaped to like 2 days. The cycles werent even > >initiated before half a day passed... I used .mopac as the first > >output file and when I wanted to optimized the ligated one I used > >.xyz format with some atoms fixed to their coordinates. Is there > >something I am doing wrong or is this leap expected? And also I read > >that mopac 2007 isnt very practical for molecules of size above 600. > >Say if I use it for a molecule of size 400 atoms would it take too > >long on a 3.8 ghz 1gb ram machine? (I might plan to continue the > >work in a advanced computational center then...) > > > >Thanks> > > > ------=_Part_4387_6377840.1192873783791 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Yes the issue was something similar to that (I needed adjust the molecule to reduce its charge and also use the keyword CHARGE), James J. P. Stewart has kindly answered my question on that matter. Thanks though.

On 10/19/07, David Gallagher gallagher.da++gmail.com <owner-chemistry::ccl.net> wrote:

Sent to CCL by: David Gallagher [gallagher.da[A]gmail.com]
One possible reason for running very slowly in previous versions of
MOPAC2007, is using the default RHF for a metal containing compound
with an odd number of electrons. If it is supposed to be a radical,
then you should try UHF for these situations. The latest version of
MOPAC2007 can automatically adjust the settings for metal containing
compounds with an odd number of electrons.

If this is not the issue, then if you send me the molecule file, I
would be happy to take a closer look.

Regards,
David Gallagher
CAChe Research


At 12:15 AM 10/15/2007, Sina T reli sinatureli:_:gmail.com wrote:

>Sent to CCL by: "Sina  T  reli" [sinatureli!^!gmail.com]
>
>- I am using mopac2007 to do some geometry optimizations with mndod
>and pm6. My first subject was a molecule of size 141 which was
>handled quite nicely and fast and ended about in 30 minutes. However
>when I added a ligand that was about of 20 atoms in size, the
>calculation time leaped to like 2 days. The cycles werent even
>initiated before half a day passed... I used .mopac as the first
>output file and when I wanted to optimized the ligated one I used
>.xyz format with some atoms fixed to their coordinates. Is there
>something I am doing wrong or is this leap expected? And also I read
>that mopac 2007 isnt very practical for molecules of size above 600.
>Say if I use it for a molecule of size 400 atoms would it take too
>long on a 3.8 ghz 1gb ram machine? (I might plan to continue the
>work in a advanced computational center then...)
>
>Thanks



E-mail to subscribers: CHEMISTRY::ccl.net or use:
      http://www.ccl.net/cgi-bin/ccl/send_ccl_message

E-mail to administrators: CHEMISTRY-REQUEST::ccl.net or use
      http://www.ccl.net/cgi-bin/ccl/send_ccl_message

Subscribe/Unsubscribe:
      http://www.ccl.net/chemistry/sub_unsub.shtml

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

Job: http://www.ccl.net/jobs
Conferences: http://server.ccl.net/chemistry/announcements/conferences/

Search Messages: http://www.ccl.net/htdig  (login: ccl, Password: search)
      

RTFI: http://www.ccl.net/chemistry/aboutccl/instructions/




------=_Part_4387_6377840.1192873783791-- From owner-chemistry@ccl.net Sat Oct 20 14:55:01 2007 From: "Anatoli Korkin a_korkin{:}yahoo.com" To: CCL Subject: CCL: NGC2009 meeting will be held at McMaster Uinversity, Canada Message-Id: <-35429-071019193401-3954-v6lGafrV6RatJ+nivftnTA|*|server.ccl.net> X-Original-From: "Anatoli Korkin" Date: Fri, 19 Oct 2007 19:33:57 -0400 Sent to CCL by: "Anatoli Korkin" [a_korkin.:.yahoo.com] We are pleased to announce that our next NGC2009 meeting will be held in August 2009 at McMaster Uinversity, Hamilton, Ontario as a joint event with 14th Canadian Semiconductor Technology Conference (CSTC2009). The conference topics will include nanoelectronics, photonics and renewable energy research and technology development. Following the tradition established at NGC2007 meeting in Arizona, NGC2009 & CSTC2009 conference will also include technology-business interface sessions to assist scientists, engineers, and industry leaders in development of entrepreneurial ventures based on disruptive nanotechnology innovations. The press release from Nano & Giga Forum network, which also includes information about the special issues of Nanotechnology and Solid State Electronics is avalable at the following web page: http://asdn.net/ngc2007/ngc2009atMcMaster.shtml For more information and to join NGC2009 & CSTC2009 meeting, please, contact us at organizers^-^asdn.net We look forward to see you at our next meeting in Canada! NGC Organizers From owner-chemistry@ccl.net Sat Oct 20 16:52:01 2007 From: "Uwe Huniar uwe.huniar=cosmologic.de" To: CCL Subject: CCL: Parallel runs in Turbomole Message-Id: <-35430-071020165015-5522-XWXgvD4UZrdoGA5ohIEz0g[A]server.ccl.net> X-Original-From: Uwe Huniar Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-15; format=flowed Date: Sat, 20 Oct 2007 22:50:29 +0200 MIME-Version: 1.0 Sent to CCL by: Uwe Huniar [uwe.huniar]~[cosmologic.de] Hi, your $TURBODIR setting points to a TmoleX installation and not to a non-GUI based installation. TmoleX does not contain the parallel version of Turbomole (not yet supported by the GUI), hence the shared libraries are usually not present - although it seems that the parallel version has been installed there in addition by hand, could that be? Replacing the serial binaries by the parallel ones might even work with TmoleX, but that has never been tested... Anyway, the problem stems from system settings or the installation. So please contact the Turbomole Support and/or your local Turbomole administrator. Parallel benchmarks have been done some time ago, but they are simply too old to be meaningful nowadays. We are planning to publish a white book about the parallel version of Turbomole together with HP, you will find the results on our web site (www.cosmologic.de) as soon as they are available. Regards, Uwe Gonzalo Jim nez-Os s gjimenez,,unizar.es schrieb: > Sent to CCL by: "Gonzalo Jim nez-Os s" [gjimenez[-]unizar.es] > Dear CCL'ers > > In the last days I've doing some benchmarking with Turbomole, and I am experiencing some kind of trouble when submitting parallel jobs. Whereas the serial jobs run in a really straightforward and quick fashion, every parallel run crashes at its first calculation stages. I do believe that I have set up the parallel environment correctly by following the manual guidelines: > > export PARA_ARCH=MPI > export PATH=TURBODIR/bin/sysname:PATH > export PARNODES=2 > > My sysname seems to be OK (em64t-unknown-linux-gnu_mpi), but when I launch the executable (e.g. jobex -ri) the calculation crashes after entering my ssh password. Here is the output: > > ***************************** > convgrep will be taken out of the TURBODIR directory > Password: > /opt/soft/tmoleX/TURBOMOLE/bin/em64t-unknown-linux-gnu_mpi/ridft_mpi: error while loading shared libraries: libguide.so: cannot open shared object file: No such file or directory > /opt/soft/tmoleX/TURBOMOLE/bin/em64t-unknown-linux-gnu_mpi/ridft_mpi: error while loading shared libraries: libguide.so: cannot open shared object file: No such file or directory > MPI Application rank 0 exited before MPI_Init() with status 127 > OPTIMIZATION CYCLE 1 > Password: > /opt/soft/tmoleX/TURBOMOLE/bin/em64t-unknown-linux-gnu_mpi/rdgrad_mpi: error while loading shared libraries: libguide.so: cannot open shared object file: No such file or directory > MPI Application rank 0 exited before MPI_Init() with status 127 > relax ended abnormally > program stopped. > relax ended abnormally > program stopped. > run relax for a cartesian step > relax ended abnormally > program stopped. > ERROR: Module relax failed to run properly - please check output job.1 and job.last for the reason > ***************************** > > It seems like it does not find de shared libraries that I have located in TURBODIR/libso/em64t-unknown-linux-gnu_mpi. It is a bit confusing to me... > Any idea about that? I am currently using Intel EM64T Dual Core processors. > > On the other hand, do you know about any benchmark results centered on the scaling ability of Turbomole? > > Regards, > > Gonzalo> > > > > -- ------------------------------------------------------------------------------ Dr. Uwe Huniar COSMOlogic GmbH&Co.KG Burscheider Str. 515 D-51381 Leverkusen Germany Tel. +49 - 2171 - 36 36 65 Fax +49 - 2171 - 73 16 89 EMail: uwe.huniar||cosmologic.de turbomole||cosmologic.de info||turbomole.com From owner-chemistry@ccl.net Sat Oct 20 20:21:00 2007 From: "Brian Salter-Duke brian.james.duke#gmail.com" To: CCL Subject: CCL: New book about modern valence bond theory Message-Id: <-35431-071020174445-9963-8teP1pLOtMmKbNL5aCd0IA()server.ccl.net> X-Original-From: "Brian Salter-Duke" Content-Disposition: inline Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=UTF-8 Date: Sun, 21 Oct 2007 06:46:57 +1000 MIME-Version: 1.0 Sent to CCL by: "Brian Salter-Duke" [brian.james.duke*gmail.com] This book could well be very important but it certainly is not the first book since Pauling. There are several multi-author volumes on modern VB theory. There are volumes in the Springer lecture series by Harcourt and by Epiotis. More importantly there is the single author volume by George Gallup, "Valence Bond Methods: Theory and applications", Cambridge University Press, 2002. Brian. On 10/19/07, Anastassia Alexandrova anastassia.alexandrova%x%yale.edu wrote: > > Sent to CCL by: Anastassia Alexandrova [anastassia.alexandrova.-*-.yale.edu] > Hello! > > Thank you for sharing this exciting news! However, I would like to object that > this is the first book since Pauling. There is "Valency and Bonding" by > Weinhold and Landis (Cambridge, 2005). Lovely book. > > Best > --------------------------------------- > Anastassia Alexandrova, Ph.D. > Yale University > Department of Chemistry > 225 Prospect Street > New Haven, CT 06520-8107 > Phone: 203-432-6288 > Fax: 203-432-6144 > anastassia.alexandrova^_^yale.edu > http://zarbi.chem.yale.edu/~anastassia/ > --------------------------------------- > God Rules Over Mankind, Animals, Cosmos and Such > > > > Quoting "David Danovich david.danovich^^huji.ac.il" : > > > > > Sent to CCL by: "David Danovich" [david.danovich() huji.ac.il] > > > > Hello, > > > > We would like to bring to the attention of the CCL community the > > First Book on the Application of Modern Valence Bond Theory Since > > Pauling's "The Nature of The Chemical Bond". > > > > Sason Shaik and Philippe C. Hiberty > > > > Please use link below in order to get information about the book > > > > http://yfaat.ch.huji.ac.il/Sason-site/Shaik_Hiberty_BookCover.pdf > > > > David > > ________________________________________________________________ > > Dr. David Danovich, Department of Organic Chemistry, The Hebrew University, > > Edmond J. Safra Campus - Givat Ram, 91904 Jerusalem, Israel > > http://yfaat.ch.huji.ac.il/david.html, david.danovich-*-huji.ac.il > > FAX:(+972)-2-6584033, Phone:(+972)-2-6586934(w), Mobile:(+972)-544-768669http://www.ccl.net/chemistry/sub_unsub.shtmlConferences: > > http://server.ccl.net/chemistry/announcements/conferences/> -- Brian Salter-Duke (aka Brian Duke) Brian.James.Duke-*-gmail.com