From gostowskir@apsu01.apsu.edu Sun Mar 22 11:19:29 1998 Received: from apsu02.apsu.edu for gostowskir@apsu01.apsu.edu by www.ccl.net (8.8.3/950822.1) id KAA28204; Sun, 22 Mar 1998 10:43:27 -0500 (EST) Received: from gostowskir.apsu.edu by APSU02.APSU.EDU (PMDF V5.1-4 #16785) with ESMTP id <01IUYNP2THO0000QTN@APSU02.APSU.EDU> for chemistry@www.ccl.net; Sun, 22 Mar 1998 09:39:31 CST Date: Sun, 22 Mar 1998 09:44:59 -0600 From: Rudy Gostowski Subject: running GAMESS on a Pentium166MMX To: CCL group Message-id: <01IUYNP2TOWI000QTN@APSU02.APSU.EDU> MIME-version: 1.0 X-Mailer: Microsoft Internet Mail 4.70.1161 Content-type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal This may be a redundant question but I have a Gateway 166MMX with 16Mb RAM and a 2.1Gb harddrive. Is this system adequate to do ab initio calculations on molecules with 22 carbons and 17 hydrogens ? If not what upgrades would you suggest ? I have determined that the memory can go to 128 with two 64Mb DIMMs. The processor can be moved up to a 200Mhz only. In addition does anyone have experience with the Evergreen 200Mhz chip (has 64k on board rather than the Intel 32k) Finally is the current harddrive adequate or should it be replaced with a SCSI unit ? Your replys will be greatly appreciated. Rudy Gostowski Department of Chemistry Austin Peay State University Box 4547 Clarksville, TN 37044 931-648-7624 FAX 931-648-5996 gostowskir@apsu01.apsu.edu From aranyos@MIT.EDU Sun Mar 22 12:19:28 1998 Received: from MIT.EDU for aranyos@MIT.EDU by www.ccl.net (8.8.3/950822.1) id LAA28370; Sun, 22 Mar 1998 11:59:15 -0500 (EST) Received: from MIT.MIT.EDU by MIT.EDU with SMTP id AA04927; Sun, 22 Mar 98 11:59:13 EST Received: from curly.lms.mit.EDU by MIT.MIT.EDU (5.61/4.7) id AA04903; Sun, 22 Mar 98 11:59:07 EST Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sun, 22 Mar 1998 11:59:50 -0500 To: chemistry@www.ccl.net From: Attila Aranyos Subject: Top PC vs. Unix Hi, Does anyone know how fast (or slow :)) is a top PC with all you can get now (P-II 333 MhZ (may be dual) and tip-top HD contorller ) compared with a Unix workstation? thanks: attila From milan@par16.mgsl.dcrt.nih.gov Sun Mar 22 14:19:28 1998 Received: from par16.mgsl.dcrt.nih.gov for milan@par16.mgsl.dcrt.nih.gov by www.ccl.net (8.8.3/950822.1) id NAA28595; Sun, 22 Mar 1998 13:27:58 -0500 (EST) Message-Id: <199803221827.NAA28595@www.ccl.net> Received: by par16.mgsl.dcrt.nih.gov (1.38.193.4/16.2) id AA26649; Sun, 22 Mar 1998 13:17:53 -0500 Date: Sun, 22 Mar 1998 13:17:53 -0500 From: Milan Hodoscek Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Attila Aranyos Cc: chemistry@www.ccl.net Subject: CCL:Top PC vs. Unix In-Reply-To: References: X-Mailer: VM 6.34 under Emacs 20.2.1 Reply-To: milan@par10.mgsl.dcrt.nih.gov >>>>> "Attila" == Attila Aranyos writes: Attila> Hi, Does anyone know how fast (or slow :)) is a top PC Attila> with all you can get now (P-II 333 MhZ (may be dual) and Attila> tip-top HD contorller ) compared with a Unix workstation? I put some new results which answer your question on my standard benchmark table for a molecular dynamics program CHARMM. Please, see http://kihp6.cmm.ki.si/parallel/summary.html Milan Hodoscek (http://kihp6.ki.si/) From elewars@alchemy.chem.utoronto.ca Sun Mar 22 15:19:32 1998 Received: from alchemy.chem.utoronto.ca for elewars@alchemy.chem.utoronto.ca by www.ccl.net (8.8.3/950822.1) id OAA29051; Sun, 22 Mar 1998 14:30:46 -0500 (EST) Received: (from elewars@localhost) by alchemy.chem.utoronto.ca (8.7.4/8.7.3) id OAA05752 for chemistry@www.ccl.net; Sun, 22 Mar 1998 14:30:48 -0500 (EST) Date: Sun, 22 Mar 1998 14:30:48 -0500 (EST) From: "E. Lewars" Message-Id: <199803221930.OAA05752@alchemy.chem.utoronto.ca> To: chemistry@www.ccl.net Subject: RESPONSE TO EHM REF REQUEST 1998 March 22 Regarding the request for a review of the extended Hueckel method: This is a brief summary of its features; ref 2 is a review. SUMMARY OF FEATURES OF EHM ca. Jan 1998 --------------- Strengths and weaknesses of the EHM STRENGTHS One big advantage of the EHM over more elaborate semiempirical methods and ab initio and DFT is that the EHM can be applied to large systems containing almost any element. The EHM can treat esentially all elements, since the only element-specific parameter needed is the (valence-state) ionization potential [1], which is generally available. In contrast, more elaborate semiempirical methods have not been parameterized for many elements (altho' recent parameteri- zations of PM3 and MNDO for transition metals make these much more generally useful than hitherto). For ab initio and DFT methods, basis sets may not be available for elements of interest, and besides ab initio and even DFT are hundreds of times slower than the EHM and are limited to much smaller systems. The applicability of the EHM to large systems and to a variety of elements is one reason why it has been extensively applied to polymeric and solid-state structures [2]. the EHM is faster than more elaborate semiempirical methods, partly because the Fock matrix needs to be diagonalized only once to yield the eigenvalues and eigenvectors (the eigenvectors must be transformed back to those of the original nonorthogonalized matrix). In contrast, methods like AM1 and PM3 (as well as ab initio calculations) require repeated matrix diagonalization because the Fock matrix must be iteratively refined in the SCF procedure. The spartan reliance of the EHM on a paucity of empirical parameters makes it relatively easy (in the right hands) to interpret its results, which depend, in the last analysis, only on geometry (which affects overlap integrals) and ionization potentials. With a strong dose of chemical intuition this has enabled the method to yield powerful insights [3]. The applicability to large systems, including polymers and solids, containing almost any kind of atom, and the relative transparency of the physical basis of the results, are the main advantages of the EHM. Surprisingly for such a conceptually simple approach, the EHM has a theoretically-based advantage over otherwise more elaborate semiempirical methods, in that it treats orbital overlap properly: AM1 and PM3, for example, use the _neglect of differential overlap_ (NDO) approximation, meaning that they take S_ij = delta_ij, as in the simple Hueckel method. The EHM can thus give superior results [4]. The EHM is a very valuable teaching tool because it follows straightforwardly from the simple Hueckel method yet uses overlap integrals and matrix orhtogonalization in the same fashion as the mathematically more elaborate ab initio procedure. Finally, the EHM, albeit more elaborately parameterized than in its original incarnation, has recently been shown to be a serious competitor to the very useful and popular semiempirical AM1 method for calculating molecular geometries [5]. WEAKNESSES The weaknesses of the standard EHM probably arise at least in part from the fact that it does not (contrast the ab initio method) take into account electron spin or electron-electron repulsion, ignores the fact that molecular geometry is partly determined by internuclear repulsions, and makes no attempt to overcome these defects by parameterization. The standard EHM gives, by and large, poor geometries and energies. Although it predicts a CH bond length of ca. 1.0 A, it yields CC bond lengths of 1.92, 1.47 and 0.85 for ethane, ethene and ethyne, respectively, cf the actual values of 1.53, 1.33 and 1.21 A, and although the favored conformation of an alkane is usually correctly identified, the energy barriers and differences are only in modest agreement with experiment [6]. Because of this inability to reliably calculate geometries, EHM calculations are usually not used for geometry optimizations, but rather utilize experimental bond lengths and angles. [1] J. Hinze, H H. Jaffe, J Am Chem Soc 84 (1962) 540; A Stockis, R. Hoffmann 102 (1980) 2952 [2] R. Hoffmann, "Solids and Surfaces: A Chemist's View of Bonding in Extended Structures", VCH publishers, 1988. [3] (a) The Woodward-Hoffman rules: R. B. Woodward, R. Hoffmann, "The Conservation of Orbital Symmetry", Academic Press, 1970. (b) Counterintuitive orbital mixing: J. H. Ammeter, H.-B. Buergi, J. C. Thibeault, R Hoffmann, J Am Chem Soc 100 (1978) 3686. [4] Rationalizing nonplanar CC double bonds: J. Spanget-Larsen, R Gleiter, Tetrahedron, 39 (1983) 3345. [5] S. L. Dixon, P C Jurs, J Comp Chem 15 (1994) 733. [6] R. Hoffmann, J Chem Phys 39 (1963) 1397. ===================== From lmoskal@emory.edu Thu Mar 19 21:18:57 1998 Received: from gabriel.cc.emory.edu for lmoskal@emory.edu by www.ccl.net (8.8.3/950822.1) id UAA15819; Thu, 19 Mar 1998 20:27:32 -0500 (EST) Received: from curly.cc.emory.edu (lmoskal@curly.cc.emory.edu [170.140.1.66]) by gabriel.cc.emory.edu (8.8.8/8.8.8) with SMTP id UAA19490 for ; Thu, 19 Mar 1998 20:27:33 -0500 (EST) Date: Thu, 19 Mar 1998 20:27:32 -0500 (EST) From: Lyudmila V Moskaleva X-Sender: lmoskal@curly.cc.emory.edu Reply-To: Lyudmila V Moskaleva To: chemistry@www.ccl.net Subject: Summary: B3LYP Raman Activities Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear All: A few days ago I posted a question concerning calculation of Raman Activities in Gaussian94. I wish to thank everyone who responded! The answer on my question is that Raman scattering Activities and Raman depolarization Ratios require calculation of third derivatives which is only available at HF level in Gaussian. Cheol Ho Choi mentioned that it is possible to calculate them numerically by introducing an external electric field, but very painful. Regards, Lyudmila >Dear Subscribers: >Can anyone, >please, advise me on the following >subject. I am doing frequencies calculations at B3LYP level in >Gaussian94. >For some reason, in the output I get all zeros in the place of Raman >scattering Activities and Raman depolarization Ratios: > > 1 2 3 > ?A ?A ?A > Frequencies -- -549.0164 33.2578 119.9049 > Red. masses -- 7.8867 2.2972 3.9523 > Frc consts -- 1.4006 0.0015 0.0335 > IR Inten -- 0.0177 0.0940 0.0677 > Raman Activ -- 0.0000 0.0000 0.0000 > Depolar -- 0.0000 0.0000 0.0000 > >Is there any keyword or "IOP" that would force the program to calculate >these parameters or does this feature only function for HF calculations? > >Thank you in advance for any response. > >Sincerely, >Lyudmila Moskaleva > >e-mail: lmoskal@emory.edu From mao@csb0.IPC.PKU.EDU.CN Fri Mar 20 03:19:01 1998 Received: from csb0.IPC.PKU.EDU.CN for mao@csb0.IPC.PKU.EDU.CN by www.ccl.net (8.8.3/950822.1) id CAA17156; Fri, 20 Mar 1998 02:54:57 -0500 (EST) Received: by csb0.IPC.PKU.EDU.CN (920330.SGI/940406.SGI.AUTO) for CHEMISTRY@www.ccl.net id AA19586; Fri, 20 Mar 98 15:53:57 -0800 Date: Fri, 20 Mar 1998 15:53:45 -0800 (PST) From: Fenglou Mao To: "CHEMISTRY@www.ccl.net" Subject: summary:Molecular Modeling C++ class library Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear cclers, Several days ago, I sent a message for Molecular Modeling C++ class library. Many people e-mailed me to summary I recieved. I am very sorry to tell all of you that it was no Molecular Modeling C++ class library available now. About three months ago, I had begun to write a Molecular Modeling C++ class library, but one month ago I had an urgent affair to do, I will resume it about three month later. Any suggestion about the class library structure will be welcome. Best wishs! Sincerely Yours, FengLou Mao ******************************* ADD:Mr. FengLou Mao Peking University BeiJing P.R.China Tel:86-10-62751490 Fax:86-10-62751725 From d.w.price@reading.ac.uk Fri Mar 20 12:19:04 1998 Received: from sumh1.rdg.ac.uk for d.w.price@reading.ac.uk by www.ccl.net (8.8.3/950822.1) id LAA19921; Fri, 20 Mar 1998 11:41:15 -0500 (EST) Received: from indyp (actually host scsgind2) by sumh1.rdg.ac.uk; Fri, 20 Mar 1998 16:40:56 +0000 Received: from reading.ac.uk by indyp via ESMTP (940816.SGI.8.6.9/930416.SGI) for id QAA27038; Fri, 20 Mar 1998 16:40:54 GMT Sender: dave@reading.ac.uk Message-ID: <35129C14.1234E677@reading.ac.uk> Date: Fri, 20 Mar 1998 16:40:53 +0000 From: Dave Price Organization: none-what-so-ever X-Mailer: Mozilla 4.04 [en] (X11; I; IRIX 5.3 IP22) MIME-Version: 1.0 To: Comp Chem Mailing List Subject: Pair Potentials from ab initio/DFT calcs? Content-Type: multipart/alternative; boundary="------------50672FD6FE5014644E576A50" --------------50672FD6FE5014644E576A50 Content-Type: text/plain; charset="us-ascii" Dear All, I have been trying to derive a set of pair potentails for some small diatomic molecules and atoms in a metal oxide lattice using DFT calculations. The lattice ion in each case was surrounded by point charges to represent counter ions and the atom or molecule placed at distances between 1 and 10 Ang from the ion and the energy of the systems were calculated. This has been of questionable success. (I know that DFT doesn't describe vdW forces very well, if at all). Question 1: Has anyone else had success with these sort of calculations? Question 2: How does one overcome the convergence problems of intrinsically unstable geometries? Question 3: Can the problems of the lowering of symmetry by placing say an atom 10 Ang away from a ion with cubically arranged point charges cause convergence problems and be overcome? Any help would be appreciated. Cheers, Dave ------------------------------------------------------------------------ Dr. David W. Price, Tel: +44 (0)118 9875123 extn 7415 Department of Chemistry, Fax: +44 (0)118 9316331 University of Reading, mailto:d.w.price@reading.ac.uk Whiteknights, READING http://www.chem.rdg.ac.uk/g50/mmrg/dave/dave.html RG6 6AD U.K. ------------------------------------------------------------------------ --------------50672FD6FE5014644E576A50 Content-Type: text/html; charset="us-ascii" Dear All,
                I have been trying to derive a set of pair potentails for some small diatomic molecules
and atoms in a metal oxide lattice using DFT calculations.  The lattice ion in each case was surrounded
by point charges to represent counter ions and the atom or molecule placed at distances between
1 and 10 Ang from the ion and the energy of the systems were calculated.  This has been of questionable
success.  (I know that DFT doesn't describe vdW forces very well, if at all).

Question 1:  Has anyone else had success with these sort of calculations?

Question 2: How does one overcome the convergence problems of intrinsically unstable geometries?

Question 3: Can the problems of the lowering of symmetry by placing say an atom 10 Ang away from
a ion with cubically arranged point charges cause convergence problems and be overcome?

Any help would be appreciated.
                    Cheers,
                                    Dave 

 
------------------------------------------------------------------------

Dr. David W. Price,       Tel: +44 (0)118 9875123  extn 7415
Department of Chemistry,  Fax: +44 (0)118 9316331
University of Reading,    mailto:d.w.price@reading.ac.uk
Whiteknights, 
READING                http://www.chem.rdg.ac.uk/g50/mmrg/dave/dave.html
RG6 6AD 
U.K.

------------------------------------------------------------------------
  --------------50672FD6FE5014644E576A50-- From ccl@www.ccl.net Fri Mar 20 15:19:05 1998 Received: from ccl.net for ccl@www.ccl.net by www.ccl.net (8.8.3/950822.1) id OAA21159; Fri, 20 Mar 1998 14:48:16 -0500 (EST) Received: from origin.gig.usda.gov (origin.gig.usda.gov [199.75.75.100]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with SMTP id OAA08415 for ; Fri, 20 Mar 1998 14:48:14 -0500 (EST) Received: by origin.gig.usda.gov (SMI-8.6/SMI-SVR4) id OAA05823; Fri, 20 Mar 1998 14:48:15 -0500 Date: Fri, 20 Mar 1998 14:48:14 -0500 (EST) From: "Stephen R. Heller" X-Sender: srheller@origin To: jcicshelp , chemweb@ic.ac.uk, chminf-l@listserv.indiana.edu, orgchem@extreme.chem.rpi.edu Subject: Software to review Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII March 20, 1998 Subject: Computer Software for Review As the Software Review Editor for the ACS Journal of Chemical Information and Computer Science (JCICS) I often get software for review in the journal. I have one (1) new software product for review. I am looking for people who are willing to review this software product. In return for the review which is published in JCICS you get to keep the software or database. The review should be completed in 1-3 months. The length of the review is 4-10 double spaced typed pages. Sample reviews can be found in most of the recent issues of JCICS. Please try to give me some (short) reason to choose you over another person. DO NOT SAY YOU WILL REVIEW ANYTHING I HAVE AVAILABLE. Messages with such replies are trashed! I have tried this approach for about the past six years and it is working reasonably well. (REMINDER: For those who haven't finished your reviews of software sent months and months ago, this last sentence does not apply to you!) As a result, I am continuing this new method to find reviewers using this e- mail/user group system. I reserve the right to abandon this if it is a problem, or inappropriate. I will not notify people if I have found a reviewer. If you don't hear from me within a few days I have chosen someone else to review the particular package. As I get many, many, (too many) replies to this message, please do not respond after 23 March 1998 (Monday), as I am sure the software will be gone by then. I can be reached on Internet (SRHELLER@GIG.US.GOV). PLEASE BE SURE TO INCLUDE AN STREET ADDRESS, PHONE, and FAX NUMBER!!! (I often send the software by Federal Express.) Without this information I WILL NOT consider your request. Steve Heller The package I now have is: gNMR from Cherwell Scientific. gNMR is a program for simulating 1-D NMR spectra for any NMR active nucleus - in a single molecule or in a mixture. Steve Heller, NIST 820 Diamond Avenue, Room 101 Gaithersburg, MD 20899 USA Phone: 301-975-3338 FAX: 301-926-0416 E-mail: srheller@gig.usda.gov WWW: www.hellers.com/~steve From root@liposome.genebee.msu.su Sat Mar 21 11:19:17 1998 Received: from liposome.genebee.msu.su for root@liposome.genebee.msu.su by www.ccl.net (8.8.3/950822.1) id LAA24732; Sat, 21 Mar 1998 11:03:56 -0500 (EST) Received: from localhost (root@localhost) by liposome.genebee.msu.su (8.8.5/8.8.5) with SMTP id TAA18615 for ; Sat, 21 Mar 1998 19:13:26 +0300 Date: Sat, 21 Mar 1998 19:13:26 +0300 (MSK) From: root To: chemistry@www.ccl.net Subject: SUMMARY: nonmales in chemistry Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I thank to all who replied. Sorry for waiting so long: but at least I'm sure to have catched all the replies. Regards, Eugene Leitl, MSU, Polymer Sci. Dept. P.S. Replies edited. ------------------------------------------------------------------------- "Susan R. Atlas" : Count me in on your 'woman computational chemist survey'. I would identify most strongly as a theoretical chemical physicist, but I do my share of first-principles calculations. ------------------------------------------------------------------------- Anonymous (male): we recently solicited resumes for a computational chemistry position. 5 of the 80 resumes received were from persons highly likely to be female. Two were ones where no reasonable conclusions could be drawn, and the remainder were ones highly likely to be male. One can rarely, however, be certain from a resume as to a person's gender (and since that's an illegal basis for selection, it is simply something we make a guess at in order to satisfy some Department of Labor reporting requirements). ------------------------------------------------------------------------- "Cicariello-Cook, Janet" : On the CCL you suggested an informal poll to find out an approximate percentage of non-males in the comp chem field. I can think of 20 women (including myself) off the top of my head. Given some time to think about it, and go through old conference lists, etc, I'm sure I can increase this number. And this would just be people I've actually met. ------------------------------------------------------------------------- "Abby L. Parrill" : I am one of the comp chem women on the list. I am not offended at all by addresses such as 'Dear Sirs', however. ------------------------------------------------------------------------- Don Gregory : I just did a rudimentary check down through our corp. phone list, and there are some people I'm not familiar with, so this would be an underestimate. The ratio of male to female 'scientists' at MSI is roughly 16% non-male. "MSI" today is approx. 300 people, approx.40-50% of which are scientific in nature, so you can come up with whatever raw numbers seem reasonable. ------------------------------------------------------------------------- "Lisa M. Balbes, Ph.D." : At the 1997 Gordon Conference on QSAR, there were 10-20% women. At the charleston conference I just left, there were closer to 30% women. ------------------------------------------------------------------------- Terri Soar <9610538n@Lv.Levels.Unisa.Edu.Au>: I'm just replying to add my name to the "female CCL participant" tally as (and am proud to admit) a female! My supervisor is also female and is a member of the CCL community, (although I know that she won't bother replying to your email for the poll as she tends to just look at the message subject headings and ditch the ones that appear irrelevant to her work!). I tend not to get offended by the odd email that address it to Sirs as these authors are most likely from countries/cultures that just haven't caught up with equal opportunity and equal talents amongst the genders. ------------------------------------------------------------------------- In this context the following recent bit of information may also be of interest: (from) SCIENCE-WEEK - Part 1/3 A Free Weekly Digest of the News of Science March 20, 1998 4. WOMEN NOW SUBSTANTIAL PORTION OF ALL NEW US CHEMISTS The latest survey of the American Chemical Society, covering chemists and chemical engineers who graduated between July 1996 and June 1997, shows the following statistics for new women graduates (percentage of total graduates who are women): Chemistry Bachelor's Degree: 48.2% Chemistry Master's Degree: 46.2% Chemistry PhD Degree: 31.6 Chemical Engineering Bachelor's Degree: 35.4% Chemical Engineering Master's Degree: 29.3% Chemical Engineering PhD Degree: 22.9% QY: Michael Heylin mailto://edit.cen@acs.org (Chem. & Eng. News 9 Mar 98) From bausch@chem.vill.edu Sat Mar 21 13:19:17 1998 Received: from rs6chem.chem.vill.edu for bausch@chem.vill.edu by www.ccl.net (8.8.3/950822.1) id MAA25020; Sat, 21 Mar 1998 12:25:58 -0500 (EST) Received: from [153.104.73.18] (bausch.chem.vill.edu [153.104.73.18]) by rs6chem.chem.vill.edu (8.8.5/8.8.5) with ESMTP id MAA12006 for ; Sat, 21 Mar 1998 12:25:57 -0500 Date: Sat, 21 Mar 1998 12:25:57 -0500 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: CHEMISTRY@www.ccl.net From: "Joseph W. Bausch" Subject: transition state problem (?) To the CCL community: I am interested in locating a transition state and determining the energetics of a degenerate process which I'll represent as A --> TS --> A'. Using G94, I employed a QST2 calc at HF/STO-3G using A and A' HF/STO-3G geometries (with atoms ordered properly) as input. The method appeared to work fine as the transition state it found looked like what I "expected", and the relative energy is about 7 kcal/mol higher than A. I then took this geometry and did an OPT=TS calc at HF/3-21G and this job ran successfully, with the relative energy being about 5 kcal higher than A. To confirm that this transition state connected A and A', I ran an IRC calc at 321g, and this worked. So, desiring to have higher level of theory data, I then did an OPT=TS calc at HF/6-31G*, which ran successfully, this time the TS being ~4 kcal/mol higher in energy than A. The results of the calculations I'll describe next have me a bit confused. I have run OPT=TS calculations at B3LYP/6-311G* and also at MP2/6-31G*, and both of these calculations ran to completion. However, the energy of TS at B3LYP/6-311G* is ~2 kcal/mol MORE stable than A, and at MP2/6-31G* TS is MORE stable than A by ~11 kcal/mol! Has anybody out there had a similar situation to this? Can I assume that DFT and MP2 levels just aren't very good for the energetics of transition states, or are these results perhaps peculiar to the specific molecule I am investigating? If my DFT and MP2 levels of theory aren't good for this type of problem, what level of theory is considered reliable for this type of calculation? Cheers and TIA, Joe --- From lavelle@mbi.ucla.edu Sat Mar 21 23:19:21 1998 Received: from theta1.ben2.ucla.edu for lavelle@mbi.ucla.edu by www.ccl.net (8.8.3/950822.1) id WAA26660; Sat, 21 Mar 1998 22:38:45 -0500 (EST) Received: from ewald.mbi.ucla.edu (ewald.mbi.ucla.edu [128.97.39.21]) by theta1.ben2.ucla.edu (8.8.8/8.8.8) with SMTP id TAA54176; Sat, 21 Mar 1998 19:38:44 -0800 Received: from red5.mbi.ucla.edu by ewald.mbi.ucla.edu (5.65v3.2/1.1.10.5/09Oct97-1217PM) id AA09387; Sat, 21 Mar 1998 19:38:43 -0800 Message-Id: <3.0.32.19980321200021.00997250@mbi.ucla.edu> X-Sender: lavelle@mbi.ucla.edu X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 21 Mar 1998 20:00:22 -0800 To: chemistry@www.ccl.net, pc-help@gaussian.com From: Laurence Lavelle Subject: G94W and 2e- integral storage Mime-Version: 1.0 Content-Type: text/enriched; charset="us-ascii" Hi, I have some questions on integral storage, and performance benchmarks in general. I know it is often said that answers to "benchmark" questions are "system dependent" and the only way to make comparisons are real tests. This however means that one has to buy the hardware and then see "how things turn out". Quantitative answers or "real comparisons" will be much appreciated. 1) Does the latest version of G94W allow for more than one path for 2 e- integral storage. That is, can they be stored on more than one hard drive? 2) Is there a quantitative way of estimating direct SCF speed, versus integral storage on hard drive(s)? That is, knowing CPU specs like CINT95, CFP95 (e.g., SPECint95 = 7.12, SPECfp95 = 5.21); and hard drive specs (e.g., two HD each with their own SCSI controller, that have sustained data transfer of 10 MB/s) can one estimate the relative speeds of direct SCF vs conventional? Bottom line: The above two HD can store 20 MB of 2e- integrals per sec (up to a max of 2 x 9000 MB = 18000 MB). Which CPU's can recalculate more than 20 MB of 2 e- integrals per sec to make direct SCF faster than conventional? About benchmarks in general: 1) How does one know if a software package or modules within a package are more dependent on integer performance or floating point performance? 2) How does one compare performance numbers stated as CINT95(intensive integer performance), CFP95 (intensive floating point performance) or MIPS (10^6 instructions per sec)? Much appreciated. Laurence
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" Laurence Lavelle, Ph.D. University of California Los Angeles Molecular Biology Institute, and Department of Chemistry & Biochemistry Laboratory of Structural Biology & Molecular Medicine Los Angeles, CA 90095-1570, USA Email:LAVELLE@MBI.UCLA.EDU Phone (Lab): (310) 206-8270 Phone (Office): (310) 825-2083 Fax: (310) 267-1957 http://www.doe-mbi.ucla.edu/people/lavelle/lavelle.html """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" It will be a great day when schools have all the money they need, and the military has bake day sales. In nature's infinite book of secrecy A little I can read.
From jkl@ccl.net Sun Mar 22 22:19:33 1998 Received: from ccl.net for jkl@ccl.net by www.ccl.net (8.8.3/950822.1) id WAA01020; Sun, 22 Mar 1998 22:09:49 -0500 (EST) Received: from krakow.ccl.net (krakow.ccl.net [192.148.249.195]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id WAA12380; Sun, 22 Mar 1998 22:09:45 -0500 (EST) From: Jan Labanowski Received: for jkl@ccl.net by krakow.ccl.net (8.8.3/920428.1525) id WAA25932; Sun, 22 Mar 1998 22:09:44 -0500 (EST) Date: Sun, 22 Mar 1998 22:09:44 -0500 (EST) Message-Id: <199803230309.WAA25932@krakow.ccl.net> To: chemistry@www.ccl.net Subject: coupling vibrational relaxation to fluid flow Cc: jkl@ccl.net Dear Gurus, I need help... One of our customers needs a piece of software. The problem involves coupling vibrational relaxation effects to fluid flow of a gas (say oxygen). Is there a piece of software which would perform vibrational kinetics using SSH theory (or whatever which works well enough), so it is possible to solve conservation equation for oxygen along with the equation of motion. The conservation equation for oxygen will have source term made up of v-t and v-v energy transfers. They would like to model these energy transfers using the latest modeling approaches in this field. If you have any pointers, I will die (hopefully not before I can relay this information) indebted. Your, Jan Labanowski jkl@ccl.net