From owner-chemistry@ccl.net Sun Jul 30 15:19:01 2006 From: "Jeff Nauss jnauss * accelrys.com" To: CCL Subject: CCL: Accelrys Customer Training for September Message-Id: <-32300-060729130323-15785-NDzRbiU7D4lrrX9PUFvgUA-x-server.ccl.net> X-Original-From: Jeff Nauss Content-Type: multipart/alternative; boundary="=_alternative 005C7851882571BA_=" Date: Sat, 29 Jul 2006 09:57:16 -0700 MIME-Version: 1.0 Sent to CCL by: Jeff Nauss [jnauss#accelrys.com] This is a multipart message in MIME format. --=_alternative 005C7851882571BA_= Content-Type: text/plain; charset="US-ASCII" Accelrys Inc. are holding the following training workshops during September 2006. These events are designed to help you get more value from your Accelrys software, helping you to better accomplish your research goals. Costs for the scheduled workshops are $500 per day for all customers. For the on-line presentations, the price is $175 per three hour session for all customers. SAN DIEGO, CA Accelrys GCG/SeqLab 30-31 August SeqWeb 1st September BTcl Scripting with Discover 12-13 Sep Introduction to Nanomaterials Modeling 14-15 Sep Library Design and Analysis with Cerius2 19-20 Sep QSAR with Cerius2 21-22 Sep Structure Based Drug Design with Cerius2 25-26 Sep Pharmacophore Generation with Catalyst 27-28 Sep BURLINGTON, MA Introduction to Quantum Mechanics Modeling 18-19 Sep Introduction to Polymer Modeling 20-21 Sep CAMBRIDGE, UK Structure-Based Design in Discovery Studio 5 Sep Protein Modeling in Discovery Studio 6 Sep Simulation in Discovery Studio 7 Sep BTcl Scripting with Discover 12-13 Sep Accord Chemistry SDK 25-26 Sep Accord Visual Objects 27 September Accord Structure Checker 28 September Accord Chemistry Cartridge 29 September INDIA Please contact Accelrys for the latest information regarding training in India for the months of August and September. ONLINE TRAINING Introduction to Discovery Studio 1 Sep (9am BST) Introduction to MS Modeling 13 Sep (4pm BST) MacVector 20 Sep (3pm PDT) MacVector 21 Sep (7am PDT) Protein-Protein Docking 21 Sep (7am PDT) Introduction to QSAR 27 Sep (4pm BST) Further details can be found on our website at http://www.accelrys.com/training/general/calendar.html and http://www.accelrys.com/services/training/general/calendar.html Questions regarding scheduling and content should be directed to workshops{}accelrys.com -- Jeffrey L. Nauss, Ph.D. Lead Training Scientist Accelrys 10188 Telesis Court, Suite 100 San Diego, CA 92121 Phone: +1-858-799-5555 Fax: +1-858-799-5100 http://www.accelrys.com/training --=_alternative 005C7851882571BA_= Content-Type: text/html; charset="US-ASCII"
Accelrys Inc. are holding the following training workshops during September 2006.  These events are designed to help you get more value from your Accelrys software,  helping you to better accomplish your research goals.

Costs for the scheduled workshops are $500 per day for all customers.  For the  on-line presentations, the price is $175 per three hour session for all customers.


SAN DIEGO, CA

Accelrys GCG/SeqLab                          30-31 August
SeqWeb                                       1st September
BTcl Scripting with Discover                 12-13 Sep
Introduction to Nanomaterials Modeling       14-15 Sep
Library Design and Analysis with Cerius2     19-20 Sep
QSAR with Cerius2                            21-22 Sep
Structure Based Drug Design with Cerius2     25-26 Sep
Pharmacophore Generation with Catalyst       27-28 Sep

BURLINGTON, MA

Introduction to Quantum Mechanics Modeling     18-19 Sep
Introduction to Polymer Modeling               20-21 Sep

CAMBRIDGE, UK

Structure-Based Design in Discovery Studio     5 Sep
Protein Modeling in Discovery Studio           6 Sep
Simulation in Discovery Studio                 7 Sep
BTcl Scripting with Discover                   12-13 Sep
Accord Chemistry SDK                           25-26 Sep
Accord Visual Objects                          27 September
Accord Structure Checker                       28 September
Accord Chemistry Cartridge                     29 September

INDIA

Please contact Accelrys for the latest information regarding training in India for the months of August and September.
 
ONLINE TRAINING

Introduction to Discovery Studio      1 Sep (9am BST)
Introduction to MS Modeling                13 Sep (4pm BST)
MacVector                            20 Sep (3pm PDT)
MacVector                            21 Sep (7am PDT)
Protein-Protein Docking                21 Sep (7am PDT)
Introduction to QSAR                      27 Sep (4pm BST)

Further details can be found on our website at http://www.accelrys.com/training/general/calendar.html and http://www.accelrys.com/services/training/general/calendar.html

Questions regarding scheduling and content should be directed to workshops{}accelrys.com

--
Jeffrey L. Nauss, Ph.D.
Lead Training Scientist
Accelrys
10188 Telesis Court, Suite 100
San Diego, CA 92121

Phone: +1-858-799-5555
Fax: +1-858-799-5100
http://www.accelrys.com/training
 --=_alternative 005C7851882571BA_=-- From owner-chemistry@ccl.net Sun Jul 30 17:38:01 2006 From: "K.Radacki K.Radacki===mail.uni-wuerzburg.de" To: CCL Subject: CCL:G: multiprocessor version of Gaussian 03 with Pentium D Message-Id: <-32301-060729043141-3871-LcGpFIc45uGYSkfShCiqtw]~[server.ccl.net> X-Original-From: "K.Radacki" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii; format=flowed Date: Sat, 29 Jul 2006 09:50:38 +0200 MIME-Version: 1.0 Sent to CCL by: "K.Radacki" [K.Radacki|,|mail.uni-wuerzburg.de] Zeki Buyukmumcu zbmumcu-.-yahoo.com wrote: >Sent to CCL by: "Zeki Buyukmumcu" [zbmumcu(a)yahoo.com] >Dear members, > >I worry if multiprocessor version of Gaussian 03 is compatible with Pentium D or Athlon X2 processor. I would be very glad if people having experience about performance of multiprocessor version of Gaussion 03 with dual core processor could inform me. best regards... > >Dr. Zeki Buyukmumcu > > > I use it for last two months. Top shows that two links 502 are working, so I belive it's compatible. :-) I had no time to make test what improvment in speed one can gain with nproc=2 with kind regards Kris -- Dr. Krzysztof Radacki Institute of Inorganic Chemistry University of Wuerzburg, Germany From owner-chemistry@ccl.net Sun Jul 30 22:10:00 2006 From: "jhammond:_:uchicago.edu" To: CCL Subject: CCL:G: transition state optimization for ethylene Message-Id: <-32302-060730025914-13172-SU+6lBNVjf3kV3LVhn4Wkw#server.ccl.net> X-Original-From: Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Date: Sun, 30 Jul 2006 01:58:44 -0500 (CDT) MIME-Version: 1.0 Sent to CCL by: [jhammond[-]uchicago.edu] I agree with previously authors who point out the spin-contamination issues with this system. Here are some additional comments: -If one wishes to study the rotation about the double bond, then the transition state would be a singlet diradical, not a triplet diradical. Real molecules do not magically change spin state without energy conversion. While the singlet and triplet diradicals are very similar in quantum chemistry, how they appear in an experiment are going to be different processes. One might be able to find the open-shell singlet wavefunction by starting from the triplet at the same geometry but if one is constructing the potential energy hypersurface cross section, the singlet must be used throughout. For the triplet, the 90 rotation from planarity is the ground-state (see "Breaking the Curse of the Non-Dynamical Correlation Problem: the Spin-Flip Method" by Krylov et al. for a qualitative picture of this). -B3LYP is not a sufficient method for this problem. In one Krylov paper (J. Chem. Phys., Vol. 118, No. 11, 15 March 2003) the potential energy surface plots with respect to rotation show a discontinuity for all SCF-based methods (HF, B3LYP, BLYP). If there is a discontinuity, the system could be susceptible to non-physical symmetry breaking. In addition, this discontinuity is non-analytic and a normal modes analysis, which is based upon a polynomial expansion about f(0), makes no sense. It's the same reason that BCS theory could not ever have been derived from perturbation theory. The two imaginary frequencies found should indicate a failure of the method. Try and fit a quadratic polynomial to the SCF curves in the Krylov paper if this isn't obvious. -The rotation barrier of ethylene is used extensively by Krylov because it is a ridiculously hard problem from a quantum chemistry perspective. It is also fundamentally multi-reference and no single configuration method will properly describe this problem. Using DFT with Pople basis sets within Gaussian is not the answer to all of life's quantum chemistry problems. I'm sorry for sounding harsh, but I feel this fact is under-appreciated among chemists. -Methods that should be sufficient for this problem include: --Krylov's spin-flip coupled-cluster --multireference perturbation theory: ---Roos's CASPT2 has been extensively applied to organic problems ---the "perturb-then-diagonalize" of Freed, among others, is sufficient but basically inaccessible to non-experts but GAMESS has MCQDP in it --multireference configuration interaction, which is accessible both with GAMESS and COLUMBUS. MRCISD should not be difficult for a minimal two-configuration reference approach. --Piecuch and Kowalski's CR-CC methods probably do it but I haven't seen a benchmark that proves it. --Any MR-CC method can do this, but those methods are impossible for non-experts. See Collect. Czech. Chem. Commun. 63, 1213 (1998) by Masik, et al. (minerva.uochb.cas.cz/Pdf/63/No08/19981213.pdf) for a BW-CC study of ethylene rotation. --EOM-CC works under certain circumstances for multireference problems. -All the above methods are very computationally expensive for a big basis set. I hope you have an account at NCSA... -If just want numbers with minimal effort, Krylov, at least, would likely have almost anything you would need, and would presumably give you the data since the results have been published. Locally, the Martinez group at UIUC could probably also calculate the results using a high-level method. Thom Dunning, also local to you, is a master of high-precision calculations and of essentially unlimited computing power. I hope this is useful and you're not personally offended by my critique of the DFT/Pople methodology. If you want triple-zeta coupled-cluster numbers with all the bells and whistles (EOM, CR, non-iterative triples) I can send you the NWChem input files to use or run them for you. The surface would take about a day on decent workstation. Jeff Hammond The University of Chicago and Pacific Northwest National Lab