From chemistry-request@server.ccl.net Wed Dec 12 03:30:13 2001 Received: from mbx.unige.ch ([129.194.8.72]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC8UCi30512 for ; Wed, 12 Dec 2001 03:30:13 -0500 Received: from CONVERSION-DAEMON.mbx.unige.ch by mbx.unige.ch (PMDF V6.0-025 #38753) id <0GO800I0129WU8@mbx.unige.ch> for chemistry@ccl.net; Wed, 12 Dec 2001 09:29:57 +0100 (MET) Received: from chiphy.unige.ch ([129.194.52.184]) by mbx.unige.ch (PMDF V6.0-025 #38753) with ESMTP id <0GO800HOA29W6G@mbx.unige.ch>; Wed, 12 Dec 2001 09:29:56 +0100 (MET) Date: Wed, 12 Dec 2001 09:25:16 +0100 From: Thomas Heine Subject: Re: CCL:LAPACK! Sender: thomas@mbx.unige.ch To: PRADIPTA Bandyopadhyay Cc: chemistry@ccl.net Reply-to: thomas.heine@chiphy.unige.ch Message-id: <3C17146C.6B583946@chiphy.unige.ch> Organization: University of Geneva MIME-version: 1.0 X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.0-4GB i686) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT X-Accept-Language: en References: X-Comment: This message was scanned against viruses by mbx.unige.ch. Hi Pradipta, which LAPACK version are you using? Download LAPACK 3.0, use your compiler without optimisations to create the dp library (by 1.) editing make.inc 2.) make install 3.) cd SRC make double After you've given your BLAS library path in make.inc you can proceed to testing and make double again. If everything is ok link this "certified" library and test your code. I had similar problems on different platforms, mainly caused by buggy installations by default. I.e.liblapack.a delivered by Intel MKL, scsl on SGI, dxml on Alpha and sunperf on Sun systems were not well implemented. If you run the LAPACK 3.0 test suite you'll get plenty of errors with these libraries. Good luck Thomas From chemistry-request@server.ccl.net Wed Dec 12 03:58:20 2001 Received: from bast.u.arizona.edu (IDENT:mcafiero@[128.196.133.225]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC8wKi31266 for ; Wed, 12 Dec 2001 03:58:20 -0500 Received: from localhost (mcafiero@localhost) by bast.u.arizona.edu (8.10.2+Sun/8.10.2) with ESMTP id fBC8w2i10824 for ; Wed, 12 Dec 2001 01:58:02 -0700 (MST) X-Authentication-Warning: bast.u.arizona.edu: mcafiero owned process doing -bs Date: Wed, 12 Dec 2001 01:58:02 -0700 (MST) From: Mauricio L Cafiero To: Subject: Non-canonical MO's in G98 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII hello: I would like to run an MP2 calculation in G98 with a self-defined set of non-canonical MO's. Is there a way to perform MP2 without doing HF first and by inputting my own MO's? I thank you in advance for your help. Mauricio Cafiero Doctoral Candidate: Theoretical and Computational Quantum Chemistry Dept. of Chemistry University of Arizona Tucson, AZ From chemistry-request@server.ccl.net Wed Dec 12 01:33:35 2001 Received: from onsager.cop.uop.edu ([138.9.240.7]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC6XVi28296 for ; Wed, 12 Dec 2001 01:33:35 -0500 Received: (from mccallum@localhost) by onsager.cop.uop.edu (8.11.6/8.9.3) id fBC6Xna23491 for chemistry@ccl.net; Tue, 11 Dec 2001 22:33:49 -0800 Date: Tue, 11 Dec 2001 22:33:49 -0800 From: "C. Michael McCallum" To: CCL List Subject: Re: CCL:Gaussian Flame-wars Message-ID: <20011211223349.B23466@onsager.cop.uop.edu> Reply-To: "C. Michael McCallum" References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from hultin@cc.UManitoba.CA on Tue, Dec 11, 2001 at 04:52:36PM -0600 * Phil Hultin (hultin@cc.UManitoba.CA) [011211 17:31]: > Max Valdez wrote: > > >I think Gaussian is a great tool, but it has a LOT of little secrets and > >"bugs", and the support is not so good, I don't think the spend that much > > money > >on customer support, given the price of the product. > > I absolutely agree with this statement. I think I represent a new kind of > scientist working with tools like Gaussian. I am an experimentalist, not a > computer whiz, but now that I can get serious computer power for a > reasonable dollar, I want to use serious computational tools to answer > nagging questions about organic chemistry. > > Gaussian can do WONDERFUL things, BUT its creators evidently expect that > users are just as computer-savvy and UNIX-proficient as they are. I have > wasted hours debugging failed jobs, where apparently everything was exactly > as the "User's Reference" said it should be! For instance, nowhere is it > stated that individual lines in the Route section must be 80 characters or > less - but there it is. Now, if I had been a little bit older and > remembered that back in the stone age Fortran punch cards had only 80 > columns, perhaps I would have twigged to this one sooner. I could go on and > on... > Just another viewpoint.... I would never hope or expect to walk into your organic lab and be able to safely and competently run a multi-step synthesis experiment and get (any) product, much less leave the place alive. That doesn't mean that your own lab is run poorly, or that your postdocs and students don't know their bacon, or even that you are inconsiderate because you don't leave full directions, or a list of your synthesis tricks and tips lying about in three-ring binders. Computers *are* cheap nowadays. So is good software. These facts invite the use of sophisticated scientific software by just about anyone (which is a good thing, I think). But, like any other scientific endeavor, it takes some training, experience, and yes, hidden details or "tricks" (which may just be artifacts). 100% documentation in anything is impossible, especially in s/w. Getting down on Gaussian for their "makefile policy" is kind of silly, and misses the point. I should say I don't use Gaussian, and never have. I also don't believe in the idea of proprietary knowledge. I'm not into keeping secrets and abilities so that people need me or what have you. Collaborations between experimentalists and theorists are valuable and fun, and should cross lines. But if you get bogged down in obscure details, don't be suprised, and get help from an expert, and share ideas and suggestions. That is how real progress is made. Cheers, Mike McCallum -- C. Michael McCallum | "That may be one tough nut to crack, Associate Professor | but I am one determined Chemistry, UOP | little squirrel" mmccallum@uop.edu | (209) 946-2636 | fax (209) 946-2607 From chemistry-request@server.ccl.net Wed Dec 12 03:05:29 2001 Received: from jaguar.it.wsu.edu (root@[134.121.0.73]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC85Ti29927 for ; Wed, 12 Dec 2001 03:05:29 -0500 Received: from enigma (ucis.chem.wsu.edu [134.121.41.180]) by jaguar.it.wsu.edu (8.11.6/8.11.6) with SMTP id fBC85DH02232; Wed, 12 Dec 2001 00:05:13 -0800 (PST) From: "Phillip Matz" To: "Demetrio Antonio da Silva Filho" , Subject: RE: CI-Singles - Wrong Excited State Energies? Date: Wed, 12 Dec 2001 00:05:13 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 In-Reply-To: <5.1.0.14.0.20011211160210.01b48400@deimos.email.arizona.edu> When you refer to the excited state geometry as being higher in energy are you referring to the transition energy (GS -> ES Delta Energy) or the absolute energy of the same excited state computed with different nuclear geometries? What should happen is that the absolute energy of the excited state for which you are optimizing the geometry decreases as the forces are minimized relative to the excited state wave function. The nuclear relaxation accompanying the excited state optimization will only raise the ground-state wave function's energy resulting in an apparent red-shift in the gas-phase emission spectra (by definition, otherwise the ground-state nuclear geometry was never optimized). Thus, comparing the energy of the lowest excited state (ES1) determined with a geometry optimized for the electronic ground state (GS), call it the ES1//GS energy, to the ES1 energy when the ES1 geometry is optimized, call it the ES1//ES1 energy, should allows yield energy of ES1//GS > energy of ES1//ES1. At least this should be true insofar as the original assertion that CIS is a variational method is correct. Having wrote all that I should have asked the obvious question first, did you verify the geometry converged on a local minimum and not an excited-state transition-state (any imaginary frequencies)? Regards, Phil Matz -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request@ccl.net]On Behalf Of Demetrio Antonio da Silva Filho Sent: Tuesday, December 11, 2001 3:19 PM To: CHEMISTRY@ccl.net Subject: CCL:CI-Singles - Wrong Excited State Energies? Dear CCL'ers, I am using CI-Singles (in gaussian'98) to optimize the geometry of the first singlet and triplet excited state. After getting the energies of those states, I subtracted those numbers from the ground state energy and then compared with the excited state energies obtained by running a CI-S calculation using the ground state geometry. To my surprise, I found that the first excited state energy, obtained using the ground state geometry, is SMALLER than the energy difference between the first excited state and the ground state energies. It means that when I optimize the geometry of the excited state using CI-Singles, I arrive at a geometry which energy is HIGHER than the one obtained using the ground state geometry -- the one that I used to begin the CI-S calculation. Does anybody have any idea of why I get these strange results? Those results mean that CI-Singles is not a reliable methodology to optimize excited state geometries? Thanks in advance, Demetrio Filho -= This is automatically added to each message by mailing script =- CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net From chemistry-request@server.ccl.net Tue Dec 11 21:03:40 2001 Received: from cpimssmtpu01.email.msn.com ([207.46.181.77]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC23ei23867 for ; Tue, 11 Dec 2001 21:03:40 -0500 Received: from oemcomputer ([216.161.155.247]) by cpimssmtpu01.email.msn.com with Microsoft SMTPSVC(5.0.2195.3779); Tue, 11 Dec 2001 18:02:24 -0800 Message-ID: <001901c182b1$548af660$f79ba1d8@oemcomputer> From: "catchcaf" To: , "Demetrio Antonio da Silva Filho" References: <5.1.0.14.0.20011211160210.01b48400@deimos.email.arizona.edu> Subject: Re: CCL:CI-Singles - Wrong Excited State Energies? Date: Tue, 11 Dec 2001 19:04:24 -0700 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 X-OriginalArrivalTime: 12 Dec 2001 02:02:24.0844 (UTC) FILETIME=[0AFFC0C0:01C182B1] hello: My first question is how large are the differences bewteen the CIS energies at equilibrium geometry and excited geometry? We must keep in mind that HF energies are only good to 0.001 hartree at best usually. Also, what method are you using for the ground state energy? If you are using HF for the ground state and CIS for excited states, then this is not a good comparison to begin with. Assuming you are using CIS for the ground state energy, excited states usually have more complicated wave functions are and harder to describe in a finite basis. A ground state energy computed at CIS with a certain basis will be more accurate than an excited state energy with the same basis. The best idea for accurate energy differences is to find the *most* accurate wave function you can possibly find for each state and then take differences. mauricio cafiero U of A Dept. of Chem. ----- Original Message ----- From: Demetrio Antonio da Silva Filho To: Sent: Tuesday, December 11, 2001 4:19 PM Subject: CCL:CI-Singles - Wrong Excited State Energies? > Dear CCL'ers, > > I am using CI-Singles (in gaussian'98) to optimize the geometry of the first > singlet and triplet excited state. > > After getting the energies of those states, I subtracted those numbers > from the > ground state energy and then compared with the excited state energies obtained > by running a CI-S calculation using the ground state geometry. > > To my surprise, I found that the first excited state energy, obtained using > the ground > state geometry, is SMALLER than the energy difference between the first > excited state > and the ground state energies. > > It means that when I optimize the geometry of the excited state using > CI-Singles, I arrive > at a geometry which energy is HIGHER than the one obtained using the ground > state geometry > -- the one that I used to begin the CI-S calculation. > > Does anybody have any idea of why I get these strange results? Those > results mean that CI-Singles is not a reliable methodology to optimize > excited state geometries? > > Thanks in advance, > > Demetrio Filho > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Tue Dec 11 23:06:43 2001 Received: from hotmail.com ([64.4.19.118]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBC46gi25936 for ; Tue, 11 Dec 2001 23:06:42 -0500 Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Tue, 11 Dec 2001 20:06:22 -0800 Received: from 64.230.24.138 by lw12fd.law12.hotmail.msn.com with HTTP; Wed, 12 Dec 2001 04:06:21 GMT X-Originating-IP: [64.230.24.138] From: "kaci Tizi-Ouzou" To: chemistry@ccl.net Subject: References: Pattern recognition molecules Date: Wed, 12 Dec 2001 04:06:21 +0000 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: X-OriginalArrivalTime: 12 Dec 2001 04:06:22.0427 (UTC) FILETIME=[5C24DAB0:01C182C2] Dear colleagues, Is there anybody that could suggest few references about pattern recognition in chemistry. Thanks a lot. Kaci _________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx From chemistry-request@server.ccl.net Wed Dec 12 06:13:41 2001 Received: from mserv.itpa.lt ([193.219.53.20]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCBDci02398 for ; Wed, 12 Dec 2001 06:13:39 -0500 Received: from localhost (tamulis@localhost) by mserv.itpa.lt (8.11.6/8.11.0) with ESMTP id fBCBDAU86929; Wed, 12 Dec 2001 13:13:10 +0200 (EET) (envelope-from tamulis@itpa.lt) Date: Wed, 12 Dec 2001 13:13:10 +0200 (EET) From: Arvydas Tamulis To: Demetrio Antonio da Silva Filho cc: Subject: Re: CCL:CI-Singles - Wrong Excited State Energies? In-Reply-To: <5.1.0.14.0.20011211160210.01b48400@deimos.email.arizona.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Demetrio Antonio da Silva Filho, CIS method is not correct for estimation of excitation energy because of very simplyfied theory. During your mentioned optimization of geometry in excited states using G98 HF-CIS I received your mentioned problems two years ago and additionally sometimes SCF did not reached at all in this optimization process. Seems exist some bugs in G98 CIS not allowing properly optimize the geometry. Try to ask G98 responsible persons. Presently we successfuly optimizing geometry in excited states using GAMESS-US package. Best wishes, Arvydas Tamulis ******************************************************************* Arvydas Tamulis Doctor of Natural Sciences, senior research fellow Institute of Theoretical Physics and Astronomy, Vilnius University, Theoretical Molecular Electronics Research Group, A. Gostauto 12, Vilnius 2600, Lithuania e-mail: TAMULIS@ITPA.lt; WEBsite: http://www.itpa.lt/~tamulis/ fax: +(370-2)-225361 or +(370-2)-224694 Phone: +(370-2)-620861 Home address: Didlaukio 27-40, Vilnius 2057, Lithuania Phone: +370-9919397 ******************************************************************* On Tue, 11 Dec 2001, Demetrio Antonio da Silva Filho wrote: > Dear CCL'ers, > > I am using CI-Singles (in gaussian'98) to optimize the geometry of the first > singlet and triplet excited state. > > After getting the energies of those states, I subtracted those numbers from the > ground state energy and then compared with the excited state energies obtained > by running a CI-S calculation using the ground state geometry. > > To my surprise, I found that the first excited state energy, obtained using > the ground > state geometry, is SMALLER than the energy difference between the first > excited state > and the ground state energies. > > It means that when I optimize the geometry of the excited state using > CI-Singles, I arrive > at a geometry which energy is HIGHER than the one obtained using the ground > state geometry > -- the one that I used to begin the CI-S calculation. > > Does anybody have any idea of why I get these strange results? Those > results mean that CI-Singles is not a reliable methodology to optimize > excited state geometries? > > Thanks in advance, > > Demetrio Filho > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > > From chemistry-request@server.ccl.net Wed Dec 12 09:45:09 2001 Received: from fyserv1.fy.chalmers.se ([129.16.110.66]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCEj9i06145 for ; Wed, 12 Dec 2001 09:45:09 -0500 Received: from fy.chalmers.se (ftf-pc32.fy.chalmers.se [129.16.112.162]) by fyserv1.fy.chalmers.se (8.8.8/8.8.8) with ESMTP id PAA11827 for ; Wed, 12 Dec 2001 15:43:39 +0100 (MET) Message-ID: <3C176DEA.EAC6D266@fy.chalmers.se> Date: Wed, 12 Dec 2001 15:47:06 +0100 From: Patrik Johansson Organization: CTH X-Mailer: Mozilla 4.51 [sv] (Win98; I) X-Accept-Language: sv MIME-Version: 1.0 To: chemistry@ccl.net Subject: HOMO and SCRF methods References: <20011206011124.A19953@OXYGEN.chem.nthu.edu.tw> <20011207230502.A23543@OXYGEN.chem.nthu.edu.tw> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear all, two simple (?) questions: 1. Is anyone aware of a study on different methods (HF, MP2, DFT) and basis sets applicability to using the HOMO level for an anionic specie as a model for the experimental oxidation potential? Any reason to as to why this should work/fail? 2. If the sensitivity against hydrolysis is sought for - can one use SCRF methods and the resulting shifted HOMOs as a first approximation? /Patrik From chemistry-request@server.ccl.net Wed Dec 12 09:57:34 2001 Received: from gandalf.cber.nih.gov ([128.231.52.5]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCEvYi06375 for ; Wed, 12 Dec 2001 09:57:34 -0500 Received: from localhost (rvenable@localhost) by gandalf.cber.nih.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id JAA79461; Wed, 12 Dec 2001 09:45:15 -0500 (EST) Date: Wed, 12 Dec 2001 09:45:14 -0500 From: Rick Venable To: Richard Walsh cc: chemistry@ccl.net, frisch@gaussian.com Subject: Re: CCL:Gaussian 98 makefile policy In-Reply-To: <200112102206.QAA46439@us.msp.networkcs.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > Michael Frish writes: > >To pursue your analogy, we're not saying you can't use the car, we're saying > >that we'll show you the blueprints to help you fix your car, but you can't > >start building your own inferior copies and selling or giving them to people. On Mon, 10 Dec 2001, Richard Walsh wrote: > Your points about the issues of support are fair, but the above does not > extend to the issue here. This is a NEW blue print--granted designed with > the help of yours--to "fix" (extend/improve) the performance of the same > model. > > You can't be suggesting that new blueprints are only improvements when > in the position of the original authors, but become inferior as soon > as they are given others? This brings up another point-- most of Gaussian, Inc.'s work is also "enhancements", to intellectual property (blueprints?) that is theirs only by purchase (from original author John Pople; I wonder how he feels about this). And one never seems to hear from Gaussian, Inc. (on CCL, anyway) about how to solve many of the problems that are posed on the mailing list. Based on the number of such questions, and a couple recent posts, it would also seem that Gaussian, Inc. doesn't offer much support to its paying customers in private, either. A classic case of "cash cow" syndrome. It certainly makes programs like GAMESS and Jaguar look more attractive. > What say you for this case--if no one asks for your support on someone > elses blueprints? There would seem to be a benefit to Gaussian unless > there are variables that must remain hidden from the user community. > > We have just purchased another site license for the product and continue > to like using it. =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= Rick Venable FDA/CBER/OVRR Biophysics Lab 1401 Rockville Pike HFM-419 Rockville, MD 20852-1448 U.S.A. (301) 496-1905 Rick_Venable@nih.gov ALT email: rvenable@speakeasy.org =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= From chemistry-request@server.ccl.net Wed Dec 12 07:03:47 2001 Received: from dalton.quimica.unlp.edu.ar ([163.10.18.1]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCC3ii03308 for ; Wed, 12 Dec 2001 07:03:46 -0500 Received: from bilbo.quimica.unlp.edu.ar (jubert4.quimica.unlp.edu.ar [163.10.18.18]) by dalton.quimica.unlp.edu.ar (8.10.2/8.11.2) with ESMTP id fBCC3P202915 for ; Wed, 12 Dec 2001 09:03:26 -0300 Message-Id: <5.1.0.14.0.20011212085341.009f3c10@dalton.quimica.unlp.edu.ar> X-Sender: pis_diez@dalton.quimica.unlp.edu.ar X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 12 Dec 2001 09:03:33 -0300 To: chemistry@ccl.net From: Reinaldo Pis Diez Subject: Parallel programming: mpich vs pvm? Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Hi netters, AFAIK there are two major, freely available libraries/tools to do parallel calculations, mpich and pvm (let me know if I miss something). My questions are: Which is the best in terms of performance? Which is the best to learn parallel programming for a guy who knows fortran and C? Thanks in advance. Regards, Reinaldo From chemistry-request@server.ccl.net Wed Dec 12 11:24:54 2001 Received: from zeta.qmw.ac.uk ([138.37.6.6]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGOsi08455 for ; Wed, 12 Dec 2001 11:24:54 -0500 Received: from alpha.qmw.ac.uk ([138.37.6.1]) by zeta.qmw.ac.uk with esmtp (Exim 3.32 #1) id 16ECBS-00014K-00 for chemistry@ccl.net; Wed, 12 Dec 2001 16:24:38 +0000 Received: from newton.chem.qmw.ac.uk ([138.37.52.163] helo=qmul.ac.uk) by alpha.qmw.ac.uk with esmtp (Exim 3.12 #1) id 16ECBS-0000Ij-00 for chemistry@ccl.net; Wed, 12 Dec 2001 16:24:38 +0000 Sender: pascal@ccl.net Message-ID: <3C1784C5.FB8BF74C@qmul.ac.uk> Date: Wed, 12 Dec 2001 16:24:37 +0000 From: Pascal Boulet Organization: Queen Mary, University of London X-Mailer: Mozilla 4.78C-SGI [en] (X11; I; IRIX 6.5 IP22) X-Accept-Language: en MIME-Version: 1.0 To: Computational Chemistry List Subject: Re: CCL:Gaussian Flame-wars References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi all, Phil Hultin wrote: > Gaussian can do WONDERFUL things, BUT its creators evidently expect that > users are just as computer-savvy and UNIX-proficient as they are. I have > wasted hours debugging failed jobs, where apparently everything was exactly > as the "User's Reference" said it should be! For instance, nowhere is it > stated that individual lines in the Route section must be 80 characters or > less - but there it is. Now, if I had been a little bit older and > remembered that back in the stone age Fortran punch cards had only 80 > columns, perhaps I would have twigged to this one sooner. I could go on and > on... > > So what's the point of all this? Well, maybe people would be less prone to > jump on Gaussian (as they have over this makefile thing) if they didn't feel > that the philosophy of Gaussian Inc. was similar to that of the "high > priest" - only the initiates will have the secrets revealed and then only > after years of study. A nice start would be to make the program work the > way the manual implies that it should (or to make the manual describe how > the program actually works). Even better: bring the program into the 21st > century by providing a user interface that is actually user-friendly. What > the priests and initiates regard as "a philosophy and aesthetic of > simplicity" looks quite otherwise to many of us on the outside! I am sorry but I strongly disagree with the statement that Gaussian should develop a "user-friendly" interface. This kind of development has the drawback of creating black boxes which, perhaps, are more accessible to a wider range of the scientific community but at the same time prevent those people from understanding what they are doing with those black-boxes. It's of course easier to click on a button rather than looking for command lines in the manual! In addition, the development of these "user-friendly" tools implies the use of default parameters and the complexity of these tools (e.g. Cerius2) implies that the user have to check for every parameters (with the risk that the user misses changing one of them). Actually, Gaussian is a very powerful program package provided that people spend a little part of their time exploring the manual. I don't think that an experimentalist would be able to reasonably perform, say, an XRD powder spectra, without at least knowing how the spectrometer works! Black-boxes are of course very user friendly, but in some cases it can also lead to non-reproducible results. Pascal > > Dr. Philip G. Hultin > Associate Department Head and > Associate Professor of Chemistry > University of Manitoba > Winnipeg, MB, Canada R3T 2N2 > (vox) 204-474-9814 > (fax) 204-474-7608 > mailto:hultin@cc.umanitoba.ca > http://www.umanitoba.ca/chemistry/ > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net -- Dr. Pascal Boulet Queen Mary, University of London Department of Chemistry, CCS Mile End Road London, E1 4NS UK Phone: (+44) (0)20 7882 7759 fax: (+44) (0)20 7882 7794 From chemistry-request@server.ccl.net Wed Dec 12 12:58:01 2001 Received: from socrates.cgl.ucsf.edu ([128.218.27.3]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCHw0i10643 for ; Wed, 12 Dec 2001 12:58:00 -0500 Received: from socrates.cgl.ucsf.edu (localhost [127.0.0.1]) by socrates.cgl.ucsf.edu (8.9.3/8.9.3/GSC1.7) with ESMTP id JAA349457 for ; Wed, 12 Dec 2001 09:57:45 -0800 (PST) Message-Id: <200112121757.JAA349457@socrates.cgl.ucsf.edu> To: chemistry@ccl.net Subject: somewhat obscure chemistry quote Date: Wed, 12 Dec 2001 09:57:45 -0800 From: David Konerding I'm looking for the source of a somewhat obscure chemistry quote I once came across but cannot find definitevly. I think it was written by a 19th century chemist who thought that math/computation in chemistry was wholly inappropriate. Does this ring a bell for anybody? Thanks. ----------------------------------------------------------------------------- David Konerding Email: dek@cgl.ucsf.edu WWW: http://www.cgl.ucsf.edu/home/dek ----------------------------------------------------------------------------- From chemistry-request@server.ccl.net Wed Dec 12 11:24:28 2001 Received: from cpimssmtpu10.email.msn.com ([207.46.181.85]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGOSi08430 for ; Wed, 12 Dec 2001 11:24:28 -0500 Received: from oemcomputer ([65.100.102.103]) by cpimssmtpu10.email.msn.com with Microsoft SMTPSVC(5.0.2195.4617); Wed, 12 Dec 2001 08:23:12 -0800 Message-ID: <000901c18329$95128040$67666441@oemcomputer> From: "catchcaf" To: "Christoph van Wüllen" Cc: References: <5.1.0.14.0.20011211160210.01b48400@deimos.email.arizona.edu><001901c182b1$548af660$f79ba1d8@oemcomputer> Subject: Re: CCL:CI-Singles - Wrong Excited State Energies? Date: Wed, 12 Dec 2001 09:25:12 -0700 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 X-OriginalArrivalTime: 12 Dec 2001 16:23:12.0783 (UTC) FILETIME=[4B91D1F0:01C18329] Hello: You are right, for CI with only singles, CIS ground state is equilvalent to HF, as the single excitations do not mix with the ground state. Excuse me for this error. I was thinking CI in general, and not specifically CIS. By the way, in answer to the original question, no, I do not believe CIS is a great way to model excited states, but it is a decent cheap way. ----- Original Message ----- From: Christoph van Wüllen To: catchcaf Sent: Wednesday, December 12, 2001 5:23 AM Subject: Re: CCL:CI-Singles - Wrong Excited State Energies? > Assuming you are using CIS for the ground state energy, excited states >usually Is there any difference between Hartree-Fock and CIS for ground state energies? +---------------------------------+----------------------------------+ | Prof. Christoph van Wüllen | Tele-Phone (+49) (0)30 314 27870 | | Technische Universität Sekr. C3 | Tele-Fax (+49) (0)30 314 23727 | | Straße des 17. Juni 135 | eMail | | D-10623 Berlin, Germany | Christoph.vanWullen@TU-Berlin.De | +---------------------------------+----------------------------------+ -- From chemistry-request@server.ccl.net Wed Dec 12 11:42:38 2001 Received: from dep.chem.rug.nl ([129.125.7.81]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGgbi08839 for ; Wed, 12 Dec 2001 11:42:37 -0500 Received: from [129.125.7.13] (theochem1.chem.rug.nl [129.125.7.13]) by dep.chem.rug.nl (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id RAA23402; Wed, 12 Dec 2001 17:42:02 +0100 X-Authentication-Warning: dep.chem.rug.nl: Host theochem1.chem.rug.nl [129.125.7.13] claimed to be [129.125.7.13] Mime-Version: 1.0 X-Sender: swart@hydra.chem.rug.nl Message-Id: In-Reply-To: <5.1.0.14.0.20011212085341.009f3c10@dalton.quimica.unlp.edu.ar> References: <5.1.0.14.0.20011212085341.009f3c10@dalton.quimica.unlp.edu.ar> Date: Wed, 12 Dec 2001 17:41:59 +0100 To: Reinaldo Pis Diez From: Marcel Swart Subject: Re: CCL:Parallel programming: mpich vs pvm? Cc: CHEMISTRY@ccl.net Content-Type: text/plain; charset="us-ascii" ; format="flowed" > Hi netters, > > AFAIK there are two major, freely available >libraries/tools to do parallel calculations, mpich and pvm (let me >know if I miss something). My questions are: Which is the best in >terms of performance? Which is the best to learn parallel >programming for a guy who knows fortran and C? In fact, you are wrong, there is even a third: LAM-MPI, and it is preferred over the other two. First of all, because MPI is faster and much easier to learn and use than PVM; second, because LAM-MPI is more stable (and faster) than MPICH. It can be found at: http://www.mpi.nd.edu/lam/ Marcel. From chemistry-request@server.ccl.net Wed Dec 12 11:59:12 2001 Received: from chinon.cnrs-orleans.fr ([163.9.6.107]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGxAi09219 for ; Wed, 12 Dec 2001 11:59:10 -0500 Received: (from hinsen@localhost) by chinon.cnrs-orleans.fr (8.11.2/8.11.2) id fBCGuVs27150; Wed, 12 Dec 2001 17:56:31 +0100 X-Authentication-Warning: chinon.cnrs-orleans.fr: hinsen set sender to hinsen@cnrs-orleans.fr using -f Sender: hinsen@chinon.cnrs-orleans.fr To: chemistry@ccl.net CC: Reinaldo Pis Diez Subject: Re: CCL:Parallel programming: mpich vs pvm? References: <5.1.0.14.0.20011212085341.009f3c10@dalton.quimica.unlp.edu.ar> From: Konrad Hinsen Date: 12 Dec 2001 17:56:31 +0100 In-Reply-To: <5.1.0.14.0.20011212085341.009f3c10@dalton.quimica.unlp.edu.ar> Message-ID: Lines: 33 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Reinaldo Pis Diez writes: > AFAIK there are two major, freely available > libraries/tools to do parallel calculations, mpich and pvm (let me > know if I miss something). My questions are: Which is the best in > terms of performance? Which is the best to learn parallel programming > for a guy who knows fortran and C? Another free MPI implementation is LAM. MPICH and LAM both implement MPI interface, either one will for for learning, and in most circumstances there shouldn't be significant performance differences. As for the choice between PVM and either MPI, I'd go for MPI. Both PVM and MPI follow the same basic approach (message passing), and MPI is an industry standard, whereas PVM is not. However, there is a real alternative worth looking at, the BSP model, for which there are two implementations, BSPlib from Oxford and PUB > from Paderborn. BSP (which stands for Bulk Synchronous Parallel) is a higher-level abstraction than the message-passing model which is the basis of MPI and PVM. It is much simpler to use and especially to debug (deadlocks are impossible, for example), and its long-time users also claim that it permits better communications optimization within the BSP library. See www.bsp-worldwide.org for more information about BSP. -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsen@cnrs-orleans.fr Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24 Rue Charles Sadron | Fax: +33-2.38.63.15.17 45071 Orleans Cedex 2 | Deutsch/Esperanto/English/ France | Nederlands/Francais ------------------------------------------------------------------------------- From chemistry-request@server.ccl.net Wed Dec 12 13:17:37 2001 Received: from garaged.fis.unam.mx ([132.248.33.226]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCIHbi11380 for ; Wed, 12 Dec 2001 13:17:37 -0500 Received: from yahoo.com (localhost [127.0.0.1]) by garaged.fis.unam.mx (8.11.6/8.11.6) with ESMTP id fBCIJuP09091 for ; Wed, 12 Dec 2001 12:19:56 -0600 Sender: max@garaged.fis.unam.mx Message-ID: <3C179FCC.FF2C05FD@yahoo.com> Date: Wed, 12 Dec 2001 12:19:56 -0600 From: Max Valdez X-Mailer: Mozilla 4.78 [es] (X11; U; Linux 2.4.9-ac18 i686) X-Accept-Language: en MIME-Version: 1.0 To: Computational Chemistry List Subject: Re: CCL:Gaussian Flame-wars References: <3C1784C5.FB8BF74C@qmul.ac.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear CCL'ers: Pascal wrote: > I am sorry but I strongly disagree with the statement that Gaussian should > develop a "user-friendly" interface. This kind of development has the drawback > of creating black boxes which, perhaps, are more accessible to a wider range of > the scientific community but at the same time prevent those people from > understanding what they are doing with those black-boxes. It's of course easier > to click on a button rather than looking for command lines in the manual! > > In addition, the development of these "user-friendly" tools implies the use of > default parameters and the complexity of these tools (e.g. Cerius2) implies > that the user have to check for every parameters (with the risk that the user > misses changing one of them). > > Actually, Gaussian is a very powerful program package provided that people > spend a little part of their time exploring the manual. I don't think that an > experimentalist would be able to reasonably perform, say, an XRD powder > spectra, without at least knowing how the spectrometer works! > > Black-boxes are of course very user friendly, but in some cases it can also > lead to non-reproducible results. I completly agree with you about the problem with black boxes and new users, but you cannot expect that every single user will learn all the details about any program, system, language, or anything that can be learned (computer-related or not). In the other side, it's totally correct (and intelligent) to develope tools to automatize tasks and improve performance, IMHO. Take for instance Jaguar, is a good program, very user-friendly, and supposedly faster than the competitors (that i dont know), or say NWChem (my new best reference) is as user friendly as gaussian, but is free ! even easily parallelizable without need of PVM or MPI or any parallelization tool, and the still have given me GOOD support to any simple or complicated question I have asked. How can I understand that if Gaussian gets my university's money and NWChem not ? I not trying to be rude here, but I know a couple of technicians that do good work with spectrometers (RMN) with less knowledge than me about the functionality, and I DON'T use (never have) RMN spectrometers. I don't think a program can be a black box if the source code is released, if any given user doesn't not about the system routines is because he hasn't have the time or will to learn it, that would happen with gaussian if they develope a reliable GUI (source code for de GUI not even needed to learn about the algorithms of gaussian itself). Best Regards Max From chemistry-request@server.ccl.net Wed Dec 12 11:35:25 2001 Received: from phm.utoronto.ca ([128.100.70.1]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGZPi08703 for ; Wed, 12 Dec 2001 11:35:25 -0500 Received: from phm.utoronto.ca (lpk04.phm.utoronto.ca [128.100.70.180]) by phm.utoronto.ca (8.9.3+Sun/8.9.3) with ESMTP id LAA15398; Wed, 12 Dec 2001 11:35:08 -0500 (EST) Sender: william@phm.utoronto.ca Message-ID: <3C178747.E618725D@phm.utoronto.ca> Date: Wed, 12 Dec 2001 11:35:19 -0500 From: William Wei Reply-To: william@phm.utoronto.ca Organization: Faculty of Pharmacy, U of T X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX64 6.5 IP30) X-Accept-Language: en MIME-Version: 1.0 To: Amber User , CCLer Subject: Question about CMC/MD output Content-Type: multipart/alternative; boundary="------------DCCD571562A0B0CF89089DE1" --------------DCCD571562A0B0CF89089DE1 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, dear CCLers and amber users, I have some questions while I try to understand the output of examples of amber CMC/MD. Any reply and hint is appreciated. 1. Are the Current MC energies (EMCRES) in the MCFIL the potential energies(eptot) in mdout(-o) file? If yes, the eptots in the file mdout should be found in the file MCFIL, but it does not match. If not, could you give me hint to find it out? 2. In the MSRCT file, I can not find the illustration for the format after the accumulated boltzmann probability of the species. Could you please help me on that? Thanks and have a nice day. William. -- William Wei Tel: 1-416-946-7551 (Lab.) Faculty of Pharmacy Fax: 1-416-978-8511 19 Russell Street Email: william@phm.utoronto.ca Toronto, ON. M5S 2S2, Canada william.wei@utoronto.ca --------------DCCD571562A0B0CF89089DE1 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, dear CCLers and amber users,

I have some questions while I try to understand the output of examples of amber CMC/MD. Any reply and hint is appreciated.

1. Are the Current MC energies (EMCRES) in the MCFIL the potential energies(eptot) in mdout(-o) file? If yes, the eptots in the file mdout should be found in the file MCFIL, but it does not match. If not, could you give me hint to find it out?
2. In the MSRCT file, I can not find the illustration for the format after the accumulated boltzmann probability of the species. Could you please help me on that?

Thanks and have a nice day.

William. 

-- 
William Wei                    Tel: 1-416-946-7551 (Lab.)
Faculty of Pharmacy            Fax: 1-416-978-8511
19 Russell Street              Email: william@phm.utoronto.ca
Toronto, ON. M5S 2S2, Canada          william.wei@utoronto.ca
  --------------DCCD571562A0B0CF89089DE1-- From chemistry-request@server.ccl.net Wed Dec 12 13:30:15 2001 Received: from sd.accelrys.com ([146.202.0.254]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCIUEi11637 for ; Wed, 12 Dec 2001 13:30:14 -0500 Received: (from osman@localhost) by sd.accelrys.com (8.11.0/8.11.0) id fBCIVvK22805 for chemistry@ccl.net; Wed, 12 Dec 2001 10:31:57 -0800 Received: from OGUNER-LAP.accelrys.com (dhcp388.sd.accelrys.com [172.30.201.134]) by sd.accelrys.com (8.11.0/8.11.0) with ESMTP id fBCIVuK22766; Wed, 12 Dec 2001 10:31:56 -0800 Message-Id: <5.1.0.14.2.20011212102031.027d3790@sparky2.sd.accelrys.com> X-Sender: osman@sparky2.sd.accelrys.com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 12 Dec 2001 10:29:45 -0800 To: David Konerding , chemistry@ccl.net From: "Osman F. Guner" Subject: Re: CCL:somewhat obscure chemistry quote Cc: qsar_society@accelrys.com In-Reply-To: <200112121757.JAA349457@socrates.cgl.ucsf.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id fBCIUFi11638 "Every attempt to employ mathematical methods in the study of chemical questions must be considered profoundly irrational and contrary to the spirit of chemistry. If mathematical analysis should ever hold a prominent place in chemistry --an aberration which is happily almost impossible-- it would occasion a rapid and widespread degeneration of that science." A. Compte, 1798-1857 There are some positive quotes from the same time period as well: "The more progress physical sciences make, the more they tend to enter the domain of mathematics, which is a kind of centre to which they all converge. We may even judge the degree of perfection to which a science has arrived by the facility with which it may be submitted to calculation." A. Quetelet, 1796-1874 Unfortunately, I have misplaced the actual references. Hope this helps...Osman At 09:57 AM 12/12/2001, David Konerding wrote: >I'm looking for the source of a somewhat obscure chemistry quote I once >came across but cannot find definitevly. I think it was written by a 19th >century chemist who thought that math/computation in chemistry was wholly >inappropriate. Does this ring a bell for anybody? > >Thanks. >----------------------------------------------------------------------------- >David Konerding >Email: dek@cgl.ucsf.edu >WWW: http://www.cgl.ucsf.edu/home/dek >----------------------------------------------------------------------------- > > >-= This is automatically added to each message by mailing script =- >CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins >MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH >CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 >Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net --- Osman F. Güner, Ph.D. Director, Lead Identification & Optimization Accelrys Inc. (858) 799-5341 osman@accelrys.com http://www.accelrys.com From chemistry-request@server.ccl.net Wed Dec 12 12:05:49 2001 Received: from jaguar.it.wsu.edu (root@[134.121.0.73]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCH5ni09392 for ; Wed, 12 Dec 2001 12:05:49 -0500 Received: from enigma (ucis.chem.wsu.edu [134.121.41.180]) by jaguar.it.wsu.edu (8.11.6/8.11.6) with SMTP id fBCH5WH07967; Wed, 12 Dec 2001 09:05:32 -0800 (PST) From: "Phillip Matz" To: "catchcaf" , , "Demetrio Antonio da Silva Filho" Subject: RE: CI-Singles - Wrong Excited State Energies? Date: Wed, 12 Dec 2001 09:05:33 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 In-Reply-To: <001901c182b1$548af660$f79ba1d8@oemcomputer> mauricio cafiero, Please clarify, to the best of my knowledge CIS does not lower the ground state energy. To reduce the ground state energy you must expand the CI treatment to include at least doubles (CISD). Regards, Phil Matz -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request@ccl.net]On Behalf Of catchcaf Sent: Tuesday, December 11, 2001 6:04 PM To: CHEMISTRY@ccl.net; Demetrio Antonio da Silva Filho Subject: CCL:CI-Singles - Wrong Excited State Energies? hello: My first question is how large are the differences bewteen the CIS energies at equilibrium geometry and excited geometry? We must keep in mind that HF energies are only good to 0.001 hartree at best usually. Also, what method are you using for the ground state energy? If you are using HF for the ground state and CIS for excited states, then this is not a good comparison to begin with. Assuming you are using CIS for the ground state energy, excited states usually have more complicated wave functions are and harder to describe in a finite basis. A ground state energy computed at CIS with a certain basis will be more accurate than an excited state energy with the same basis. The best idea for accurate energy differences is to find the *most* accurate wave function you can possibly find for each state and then take differences. mauricio cafiero U of A Dept. of Chem. ----- Original Message ----- From: Demetrio Antonio da Silva Filho To: Sent: Tuesday, December 11, 2001 4:19 PM Subject: CCL:CI-Singles - Wrong Excited State Energies? > Dear CCL'ers, > > I am using CI-Singles (in gaussian'98) to optimize the geometry of the first > singlet and triplet excited state. > > After getting the energies of those states, I subtracted those numbers > from the > ground state energy and then compared with the excited state energies obtained > by running a CI-S calculation using the ground state geometry. > > To my surprise, I found that the first excited state energy, obtained using > the ground > state geometry, is SMALLER than the energy difference between the first > excited state > and the ground state energies. > > It means that when I optimize the geometry of the excited state using > CI-Singles, I arrive > at a geometry which energy is HIGHER than the one obtained using the ground > state geometry > -- the one that I used to begin the CI-S calculation. > > Does anybody have any idea of why I get these strange results? Those > results mean that CI-Singles is not a reliable methodology to optimize > excited state geometries? > > Thanks in advance, > > Demetrio Filho > > > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > -= This is automatically added to each message by mailing script =- CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net From chemistry-request@server.ccl.net Wed Dec 12 11:59:54 2001 Received: from mail-02.med.umich.edu ([141.214.93.150]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCGxsi09255 for ; Wed, 12 Dec 2001 11:59:54 -0500 Received: from gwia-02-MTA by mail-02.med.umich.edu with Novell_GroupWise; Wed, 12 Dec 2001 11:59:37 -0500 Message-Id: X-Mailer: Novell GroupWise Internet Agent 6.0.1 Date: Wed, 12 Dec 2001 11:59:28 -0500 From: "Renxiao Wang" To: Subject: Summary: availabe packages for neural networks Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id fBCGxsi09256 Dear CCLers, Last week I pasted a post looking for available packages for building neural networks. And here is the summary of the answers: ---------------------------------------------------------------------- > From Ryszard Czerminski One well known package is SNNS (just search for it on the web). I would recommend also that you take a look at Support Vector Machines for this type of work. You may find a lot of freely available packages starting from www.support-vector.net ---------------------------------------------------------------------- > From Dr. Thomas Exner I would recommend the Stuttgart Neural Network Simulator (SNNS). You can receive it free from http://www-ra.informatik.uni-tuebingen.de/SNNS/ for a number of different operating systems. It can simulate a whole bunch of different neuro nets not only backpropagation and the trained net can be written out as a C subroutine to use in other programs. ---------------------------------------------------------------------- > From Didier Mathieu So far I never used neuron networks for but some colleagues who did used (and praised) SNNS: http://www-ra.informatik.uni-tuebingen.de/SNNS/ ---------------------------------------------------------------------- > From Donald B. Boyd, Ph.D. I highly recommend the following tutorial chapter. It includes information about software. K. Peterson, in Reviews in Computational Chemistry, K. B. Lipkowitz and D. B. Boyd, Eds., Wiley-VCH, New York, 2000, Vol. 16, pp. 53-140. Artificial Neural Networks and Their Use in Chemistry. ---------------------------------------------------------------------- > From Yvonne Martin The one recommended to me by Jens Sadowski and Hugo Kubinyi is that from Stuttgart University. I looked at their web site and the program looks very complete. ---------------------------------------------------------------------- > From "J.Aires de Sousa" following are some features of JATOON (website: http://www.dq.fct.unl.pt/qoa/jas/jatoon/) ---------------------------------------------------------------------- > From "Borosy, Andras {Basi~Basel}" Good idea. I used 3DNet for this purpose. I do not know if it available commercially (A.P. Borosy, P. Matyus, K. Keseru, Chemometrics & Int. Lab. Syst. 54, 107-122, 2000.) Theoretically any ANN with supervised learning is capable for that. See http://www.emsl.pnl.gov:2080/proj/neuron/neural/systems/software.html#131 Stuttgart Neural Network System is a very good choice, because it is a complex, powerfull application, run in many op. syst and free. http://www-ra.informatik.uni-tuebingen.de/SNNS/ ----------------------------------------------------------------------- As you can see, most people have mentioned "SNNS". I have downloaded that program, which seems to be a professional package right for my purpose.. Regards, Renxiao Wang, Ph.D. University of Michigan Medical School From chemistry-request@server.ccl.net Wed Dec 12 15:00:53 2001 Received: from electra.cc.umanitoba.ca (root@[130.179.16.23]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCK0qi14583 for ; Wed, 12 Dec 2001 15:00:53 -0500 Received: from hultin (hultin.chem.umanitoba.ca [130.179.48.60]) by electra.cc.umanitoba.ca (8.12.0/8.12.0) with SMTP id fBCK0aZ2024266 for ; Wed, 12 Dec 2001 14:00:36 -0600 (CST) From: "Phil Hultin" To: "Computational Chemistry List" Subject: User-Friendly vs. Computer-Machismo Date: Wed, 12 Dec 2001 14:00:36 -0600 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 My comments about a user-friendly interface for Gaussian drew some predictable results. Why is it that some computer people think that if you dislike wondering whether you should leave a blank line or not, or trying to understand why the syntax in the manual does not work, you therefore must favor a "black box" or think that you should just be able to poke any old junk in and get research-quality results? This is the clearest illustration that I could have asked for of the "computer-machismo" attitude I was mentioning in my initial post. As a synthetic chemist, I do not ask my students to use dual-pan balances - they get modern digital balances. I do not ask them to measure liquids in glass pipettes - they get gas-tight syringes. We do not boil solvents away on steam-baths, we use rotary evaporators. In other words, I do not insist that modern synthetic chemistry be accomplished with outmoded tools, just because those are the tools I may have learned on. I certainly never give them experimental procedures that I know to be erroneous and wait for them to fail before telling them the RIGHT way to do the reaction. Likewise, I would like to see computational chemistry programs (not simply Gaussian, although that is the program I have chosen to use and will continue to use) permit chemists to think about chemistry rather than the idiosyncrasies of the program. My complaint is that I am spending a disproportionate amount of time figuring out hold-overs from computer limitations of 25 years ago. Why do I have to struggle with "syntax" in what the G98 manual (page 28) says should be a "free-format" input mode? How difficult can it be to keep your manual consistent with your product, or to create a front-end that presents the valid options for each type of computation in a clear and consistent way? I don't want the program to do things for me. I too dislike a "black box" approach. I just point out that I don't enjoy struggling with undocumented bugs (or whatever you want to call them). My scientific competence is not enhanced by proving I can overcome awkward tools, not even the most powerful tools (such as G98). Users definitely should be conversant with the principles of physical and theoretical chemistry - but why are people still insisting that they have to be familiar with Fortran 77 (or whatever) and its limitations? Dr. Philip G. Hultin Associate Department Head and Associate Professor of Chemistry University of Manitoba Winnipeg, MB, Canada R3T 2N2 (vox) 204-474-9814 (fax) 204-474-7608 mailto:hultin@cc.umanitoba.ca http://www.umanitoba.ca/chemistry/ From chemistry-request@server.ccl.net Wed Dec 12 13:39:36 2001 Received: from sd.accelrys.com ([146.202.0.254]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCIdai12137 for ; Wed, 12 Dec 2001 13:39:36 -0500 Received: (from osman@localhost) by sd.accelrys.com (8.11.0/8.11.0) id fBCIfJi04357 for chemistry@ccl.net; Wed, 12 Dec 2001 10:41:19 -0800 Received: from OGUNER-LAP.accelrys.com (dhcp388.sd.accelrys.com [172.30.201.134]) by sd.accelrys.com (8.11.0/8.11.0) with ESMTP id fBCIfIK04315; Wed, 12 Dec 2001 10:41:18 -0800 Message-Id: <5.1.0.14.2.20011212103735.02820120@sparky2.sd.accelrys.com> X-Sender: osman@sparky2.sd.accelrys.com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 12 Dec 2001 10:39:07 -0800 To: qsar_society@accelrys.com, David Konerding , chemistry@ccl.net From: "Osman F. Guner" Subject: Re: QSAR - Re: CCL:somewhat obscure chemistry quote In-Reply-To: <5.1.0.14.2.20011212102031.027d3790@sparky2.sd.accelrys.com > References: <200112121757.JAA349457@socrates.cgl.ucsf.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id fBCIdai12138 My all time favorite quote is actually from Niels Bohr, I think: "Prediction is very difficult... especially about the future." At 10:29 AM 12/12/2001, Osman F. Guner wrote: >NB: Unless you reset the To: line, your reply goes to the entire list >--- > > >"Every attempt to employ mathematical methods in the study of chemical >questions must be considered profoundly irrational and contrary to the >spirit of chemistry. If mathematical analysis should ever hold a >prominent place in chemistry --an aberration which is happily almost >impossible-- it would occasion a rapid and widespread degeneration of that >science." >A. Compte, 1798-1857 > >There are some positive quotes from the same time period as well: > >"The more progress physical sciences make, the more they tend to enter the >domain of mathematics, which is a kind of centre to which they all >converge. We may even judge the degree of perfection to which a science >has arrived by the facility with which it may be submitted to calculation." >A. Quetelet, 1796-1874 > >Unfortunately, I have misplaced the actual references. Hope this >helps...Osman > >At 09:57 AM 12/12/2001, David Konerding wrote: > >>I'm looking for the source of a somewhat obscure chemistry quote I once >>came across but cannot find definitevly. I think it was written by a 19th >>century chemist who thought that math/computation in chemistry was wholly >>inappropriate. Does this ring a bell for anybody? >> >>Thanks. >>----------------------------------------------------------------------------- >>David Konerding >>Email: dek@cgl.ucsf.edu >>WWW: http://www.cgl.ucsf.edu/home/dek >>----------------------------------------------------------------------------- >> >> >>-= This is automatically added to each message by mailing script =- >>CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins >>MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH >>CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 >>Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > >--- >Osman F. Güner, Ph.D. >Director, Lead Identification & Optimization >Accelrys Inc. (858) 799-5341 >osman@accelrys.com http://www.accelrys.com > >_______________________________________________ >qsar_society mailing list >qsar_society@accelrys.com >http://nntp-mailman.sd.accelrys.com/mailman/listinfo/qsar_society --- Osman F. Güner, Ph.D. Director, Lead Identification & Optimization Accelrys Inc. (858) 799-5341 osman@accelrys.com http://www.accelrys.com From chemistry-request@server.ccl.net Wed Dec 12 16:44:17 2001 Received: from pollux.chem.umn.edu ([128.101.156.56]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCLiGi17292 for ; Wed, 12 Dec 2001 16:44:16 -0500 Received: (from cramer@localhost) by pollux.chem.umn.edu (8.9.3/8.9.3) id PAA02458; Wed, 12 Dec 2001 15:43:45 -0600 (CST) From: Christopher Cramer Message-Id: <200112122143.PAA02458@pollux.chem.umn.edu> Subject: Re: CCL:somewhat obscure chemistry quote To: osman@accelrys.com (Osman F. Guner) Date: Wed, 12 Dec 2001 15:43:45 -0600 (CST) Cc: chemistry@ccl.net In-Reply-To: <5.1.0.14.2.20011212102031.027d3790@sparky2.sd.accelrys.com> from "Osman F. Guner" at Dec 12, 2001 10:29:45 AM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > > "Every attempt to employ mathematical methods in the study of chemical > questions must be considered profoundly irrational and contrary to the > spirit of chemistry. If mathematical analysis should ever hold a prominent > place in chemistry --an aberration which is happily almost impossible-- it > would occasion a rapid and widespread degeneration of that science." > A. Compte, 1798-1857 > > There are some positive quotes from the same time period as well: > > "The more progress physical sciences make, the more they tend to enter the > domain of mathematics, which is a kind of centre to which they all > converge. We may even judge the degree of perfection to which a science > has arrived by the facility with which it may be submitted to calculation." > A. Quetelet, 1796-1874 > How nice to see something light in between the venting about Gaussian that is keeping my delete button busy... My personal favorite to add to the above hopper: "The fact is, there has been a split of chemistry into two schools since the intrusion of the Arrhenic faith, rather it should be said, the addition of a new class of worker into our profession--people without knowledge of the laboratory arts and with sufficient mathematics at their command to be led astray by curvilinear agreements; without the ability to criticise, still less of giving any chemical interpretation. The fact is, the physical chemists never use their eyes and are most lamentably lacking in chemical culture. It is essential to cast out from our midst, root and branch, this physical element and return to our laboratories. H. Armstrong, 1936 Armstrong is credited with defining the term "scientific method" and had a massive impact on chemical pedagogy in the late 1800s and early 1900s. He can perhaps be forgiven for the above crotchetiness, which was uttered at the age of 88. Chris Cramer -- Christopher J. Cramer University of Minnesota Department of Chemistry 207 Pleasant St. SE Minneapolis, MN 55455-0431 -------------------------- Phone: (612) 624-0859 || FAX: (612) 626-2006 cramer@pollux.chem.umn.edu http://pollux.chem.umn.edu/~cramer (website includes information about new textbook "Essentials of Computational Chemistry: Theories and Models") From chemistry-request@server.ccl.net Wed Dec 12 17:23:53 2001 Received: from ccl.net ([192.148.249.4]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCMNri18106 for ; Wed, 12 Dec 2001 17:23:53 -0500 Received: from arlen.ccl.net (arlen.ccl.net [192.148.249.10]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id RAA09990; Wed, 12 Dec 2001 17:23:52 -0500 (EST) Date: Wed, 12 Dec 2001 17:23:51 -0500 (EST) From: Jan Labanowski To: chemistry@ccl.net, Microelectronics@AtomicScaleDesign.Net cc: Jan Labanowski Subject: 02.01.23 4th Ann Int Symp Combinatorial Appr for New Materials Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=X-UNKNOWN Content-Transfer-Encoding: 8BIT Content-ID: COMBI BULLETIN: The COMBI 2002 Conference Newsletter Issue #2 Program Organized and Sponsored by: The Knowledge Foundation, Inc. Endorsing Publications: * Elsevier Science- Materials Science and Engineering A * High Throughput Experimentation.com * IOP - High Performance Polymers * IOP-Nanotechnology * Materials Technology * Wiley-VCH -Macromolecular Materials and Engineering COMBI 2002 - The 4th Annual International Symposium on Combinatorial Approaches for New Materials Discovery January 23 - 25, 2002 • Four Points Sheraton Hotel • San Diego, CA USA Register Online at: http://www.knowledgefoundation.com/events/8161326.htm View Complete Conference Agenda Here: http://www.knowledgefoundation.com/events/8161326_p.pdf CONVENIENTLY TIMED: The National Institute of Standards and Technology PRESENTS THE ESTABLISHMENT OF THE COMBINATORIAL METHODS CENTER AT THE NATIONAL INSTITUTE OF STANDARDS AND TECHNOLOGY Wednesday, January 23, 2002 • Four Points Sheraton Hotel • San Diego, CA 7:30am-2:30pm Website: www.nist.gov/combi. Dear Colleague, In it's fourth outstanding year the COMBI 2002 conference is scheduled for January 23-25 at the Four Points Sheraton Hotel in San Diego, CA. We're very excited to be able to provide you with a jam-packed agenda and a great line-up of speakers to follow on the success of our previous annual events. Recent Updates: * NEW! Pre-conference workshop addressing COMBI & High Throughput Synthesis Technology Transfer, Market Challenges and Investment Prospects * Dedicated exhibit space and extended viewing time to showcase products and present technical data via posters. Please contact Alan Abend soon to secure your booth space. Alan can be reached at 617-232-7400, ext. 202; aabend@knowledgefoundation.com * Delegates interested in submitting a poster should register for the conference and submit a poster abstract to me by January 2, 2002 the program brochure for details. * Discount Accommodations and Travel: A block of rooms has been allocated at a special reduced rate. Please make your reservations by January 2, 2002. When making reservations, please refer to The Knowledge Foundation. For Hotel Reservations Contact: Andersen Travel at: Phone: (508) 429-6494 or 1-800-229-6494, Fax: (508) 429-7380, Email: suek@andersentvl.com The Knowledge Foundation's official travel agent, Andersen Travel will assist you in making all or a portion of your travel arrangements. If you have not yet received a brochure for the program by mail please let me know so we can make sure you receive one. Here is the current line-up of organizations that will be discussing their work at the conference this year. There have been no program changes to date. We'll certainly keep you posted if any occur. * Accelrys Ltd. * AstraZeneca * Avantium Technologies B.V. * Bayer AG * DuPont Central R&D * ETH Zürich * General Electric Company * hte North America * National Institute of Standards and Technology * National Renewable Energy Laboratory * NovoDynamics, Inc. * Purdue University * Striatus Incorporated * Symyx Technologies, Inc. * Tokyo Institute of Technology * Universität des Saarlandes Im Stadtwald * University of Arizona * University of California at Los Angeles * University of Liverpool * University of Maryland * University of Notre Dame * University of Southampton * UOP LLC I look forward to meeting you all January 23-25 at the Four Points Sheraton Hotel in San Diego, CA. ... Serge Pan, Ph.D., Program Manager Questions/Registration Information: The Knowledge Foundation, Inc. 18 Webster Street Brookline, MA 02446 P: 617-232-7400 F: 617-232-9171 Email: custserv@knowledgefoundation.com Knowledgefoundation.com sends this information to website visitors who have chosen to receive it. If you have received this email in error please click below to be removed from the list. --- * --- * --- * --- * --- * --- * --- * --- * http://www.knowledgefoundation.com/mail.cfm From chemistry-request@server.ccl.net Wed Dec 12 13:43:58 2001 Received: from socrates.cgl.ucsf.edu ([128.218.27.3]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCIhwi12223 for ; Wed, 12 Dec 2001 13:43:58 -0500 Received: from socrates.cgl.ucsf.edu (localhost [127.0.0.1]) by socrates.cgl.ucsf.edu (8.9.3/8.9.3/GSC1.7) with ESMTP id KAA353819; Wed, 12 Dec 2001 10:43:43 -0800 (PST) Message-Id: <200112121843.KAA353819@socrates.cgl.ucsf.edu> To: David Konerding cc: chemistry@ccl.net, dek@cgl.ucsf.EDU Subject: Re: CCL:somewhat obscure chemistry quote In-Reply-To: Your message of "Wed, 12 Dec 2001 09:57:45 PST." <200112121757.JAA349457@socrates.cgl.ucsf.edu> Date: Wed, 12 Dec 2001 10:43:42 -0800 From: David Konerding David Konerding writes: > >I'm looking for the source of a somewhat obscure chemistry quote I once >came across but cannot find definitevly. I think it was written by a 19th >century chemist who thought that math/computation in chemistry was wholly >inappropriate. Does this ring a bell for anybody? Excellent, I have already received three messages that gave the quote. This will make a great quote for my thesis, which of course, is about using mathematics in chemistry :-) Now, here's a harder one. I also once came across a pair of quotes, about the use of math in chemistry, where the first said somethign like "We can compute anything" and the response was more or less "True, but it will take a very long time". Given that my simulations each took over 3 months to run (10ns of duplex RNA in a box of water using 4 cpus and AMBER6) I thought this would also be appropriate. Dave From chemistry-request@server.ccl.net Wed Dec 12 16:53:02 2001 Received: from ccmsi.jsums.edu ([143.132.103.54]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCLr2i17412 for ; Wed, 12 Dec 2001 16:53:02 -0500 Received: from Falcon [10.30.4.199] by ccmsi.jsums.edu (SMTPD32-7.04) id A1AF12710220; Wed, 12 Dec 2001 15:52:47 -0600 Message-ID: <005501c18357$5561a2e0$c7041e0a@ccmsi.jsums.edu> From: "Yevgeniy Podolyan" To: Subject: 02.03.22 Second Southern School on Computational Chemistry, Orange Beach, AL - USA Date: Wed, 12 Dec 2001 15:52:45 -0600 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Dear Colleague: We invite you to participate in the Second Southern School on Computational Chemistry organized together by the members of the NSF-EPSCOR Computational Chemistry Cluster, NSF-CREST Computational Center for Molecular Structure and Interactions, and the Army High Performance Computing Research Center. The meeting will be held at the Hampton Inn on beautiful Orange Beach, Alabama (next to Pensacola, FL) on March 22 and 23, 2002 (Friday 2 p.m. - Saturday 10 p.m.). There will be two formal lectures by William Miller from University of California at Berkeley and Vincent Ortiz from Kansas State University. In addition, we are planning 20-30-minute oral presentations of senior investigators and posters from the students. These presentations will cover all areas of computational and quantum chemistry. A special session on High Performance Computing on Large Molecular Systems is scheduled for the Friday afternoon. We will publish a volume of proceeding with abstracts and a special issue of "Structural Chemistry" with selected papers from the meeting. Please email your abstracts by February 1, 2002 to Mr. Y. Podolyan podolyan@ccmsi.jsums.edu (up to 4 pages in Microsoft Word format) with indication if you prefer oral or poster presentation and a file with your picture. These pictures will be included in the conference materials along with the list of participants. You can also register online through this web site at http://ccmsi.jsums.edu/sscc The conference fee, which includes meals (Friday dinner through Sunday breakfast), is $100. I hope to see you at our School. Jerzy Leszczynski From chemistry-request@server.ccl.net Wed Dec 12 13:50:43 2001 Received: from mail.rpi.edu (root@[128.113.22.40]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCIohi12513 for ; Wed, 12 Dec 2001 13:50:43 -0500 Received: from webmail.rpi.edu (webmail.rpi.edu [128.113.26.21]) by mail.rpi.edu (8.11.3/8.11.3) with ESMTP id fBCIoLM87502; Wed, 12 Dec 2001 13:50:21 -0500 Message-Id: <200112121850.fBCIoLM87502@mail.rpi.edu> Content-Type: text/plain Content-Disposition: inline To: P.Boulet@qmul.ac.uk From: "Dr. N. SUKUMAR" Organization: Rensselaer Polytechnic Institute Cc: chemistry@ccl.net X-Originating-Ip: 128.113.21.67 Mime-Version: 1.0 Reply-To: "Dr. N. SUKUMAR" Date: Wed, 12 Dec 2001 13:50:20 EST X-Mailer: EMUmail 4.00 Subject: Re: CCL:Gaussian Flame-wars On Wed, 12 Dec 2001 16:24:37 +0000 Pascal Boulet wrote: > It's of course > easier > to click on a button rather than looking for command lines in the manual! > I may be part of a very rapidly dwindling paleozoic minority, but I actually find it easier and more intuitive to learn simple keywords (like HF/6-31G* or CISD) than to memorize arcane heiroglyphs (icons) for every single action in this world, as seems to be the fashion nowadays. > > Black-boxes are of course very user friendly, but in some cases it can > also lead to non-reproducible results. > ESPECIALLY when the source code is not made available for these "user-friendly" programs. Dr. N. Sukumar Rensselaer Department of Chemistry From chemistry-request@server.ccl.net Wed Dec 12 15:12:18 2001 Received: from mhub16.lvs.dupont.com ([52.128.30.9]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCKCIi14869 for ; Wed, 12 Dec 2001 15:12:18 -0500 Received: from mhub14.lvs.dupont.com by mhub16.lvs.dupont.com with ESMTP for CHEMISTRY@ccl.net; Wed, 12 Dec 2001 15:11:51 -0500 Received: from CDCLN05.LVS.DUPONT.COM by mhub14.lvs.dupont.com with ESMTP for CHEMISTRY@ccl.net; Wed, 12 Dec 2001 15:11:37 -0500 Subject: Gaussian support To: CHEMISTRY@ccl.net From: "Kerwin D Dobbs" Date: Wed, 12 Dec 2001 15:11:53 -0500 Message-Id: X-MIMETrack: Serialize by Router on CDCLNMH1/DuPont_MHUB(Release 5.0.8 |June 18, 2001) at 12/12/2001 03:11:37 PM MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Gaussian 98 is a fine and very useful electronic structure program. If this software did not perform well, it would no longer reside on our computers. However, what I have been reading has not been about its performance but whether or not the support for it is good. From my own experience with Gaussian, Inc., I have only high praises for the support they have provided over the last decade. Of course, this support is the direct consequence of paying a yearly maintenance fee. So, I'm wondering ... are people/organizations purchasing a license for the Gaussian program and expecting support without paying an additional maintenance fee? Or, are they paying this maintenance fee and getting little to no support? If the latter, then they have the right to complain and take it up directly with Mike Frisch himself. Otherwise, I suggest paying close attention to what you read daily on the CCL or, even, digging into the CCL archives for help with certain aspects of using the Gaussian program. Kerwin =============================================== Kerwin Dobbs (Kerwin.D.Dobbs@usa.dupont.com) DuPont Research & Development Experimental Station P.O. Box 80328 Wilmington DE 19880-0328 phone: (302)695-8084 "The miracles of computational chemistry" =============================================== From chemistry-request@server.ccl.net Wed Dec 12 16:27:20 2001 Received: from uucp0.ash.ops.us.uu.net ([198.5.241.21]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCLRKi16818 for ; Wed, 12 Dec 2001 16:27:20 -0500 Received: from uucp0.ash.ops.us.uu.net by uucp0.ash.ops.us.uu.net with SMTP (peer crosschecked as: uucp0.ash.ops.us.uu.net [198.5.241.21]) id QQltdd16954 for ; Wed, 12 Dec 2001 21:27:03 GMT Received: from m10.UUCP by uucp0.ash.ops.us.uu.net with UUCP/RMAIL ; Wed, 12 Dec 2001 21:27:03 +0000 Received: from svega.gaussian.com by m10.gaussian.com; Wed, 12 Dec 2001 16:22:42 -0500 Received: (from frisch@localhost) by svega.gaussian.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id QAA05160 for chemistry@ccl.net; Wed, 12 Dec 2001 16:23:08 -0500 (EST) From: frisch@gaussian.com (Mike Frisch) Message-Id: <200112122123.QAA05160@svega.gaussian.com> Subject: Re: CCL:Gaussian 98 makefile policy To: chemistry@ccl.net Date: Wed, 12 Dec 2001 16:23:08 -0500 (EST) In-Reply-To: from "Rick Venable" at Dec 12, 2001 09:45:14 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > > Rick Venable writes: > > This brings up another point-- most of Gaussian, Inc.'s work is also > "enhancements", to intellectual property (blueprints?) that is theirs > only by purchase (from original author John Pople; I wonder how he feels > about this). And one never seems to hear from Gaussian, Inc. (on CCL, > anyway) about how to solve many of the problems that are posed on the > mailing list. Based on the number of such questions, and a couple > recent posts, it would also seem that Gaussian, Inc. doesn't offer much > support to its paying customers in private, either. A classic case of > "cash cow" syndrome. This is an unfair characterization. Berny Schlegel, Krishnan Raghavachari, and I have contributed substantially to the program for over twenty years, and many others have contributed nearly as long. To describe any version of Gaussian in the last two decades as "enhancements" of the original author John Pople is grossly inaccurate and does not give proper scientific credit to the long list of authors. John Pople always acknowledged the work of others. In fact, Gaussian 82 had important contributions from outside Pople's group, the majority of improvements in Gaussian 86, 88, 90, and 92 were from outside of his group, and it did not contribute at all to Gaussian 94 or 98. A glance at the references for methods added to Gaussian in the last twenty years will point to dozens of papers by the other people involved describing new science done by the participants in the form of new models and algorithms developed and implemented in the software. Some of this was done entirely by employees of Gaussian, Inc., some was done by academic groups who we help to support, and most was done via active scientific collaborations in which both the intellectual content and the code include contributions from employees and from academic collaborators. Here are a few examples, in no particular order with respect to date or importance: a. TD-DFT (Scuseria group and Gaussian). b. Solvation models (Numerous people at Gaussian, and in the Tomasi and Barone groups). c. ONIOM (Morokuma's group with contributions from people at Gaussian). d. Magnetic properties (all done at Gaussian). e. Semi-direct analytic MP2 frequencies and direct CIS frequencies (all done at Gaussian). f. Direct MCSCF energies, gradients, and frequencies (a collaboration between people at Gaussian and Robb's group). g. Conical intersection location (Schlegel and Robb). h. Linear scaling methods (Scuseria's group and Gaussian). i. Second order IRC optimization (Schlegel group). j. Redundant internal coordinate optimization (Schlegel group and Gaussian). k. CBS methods (Petersson group and Gaussian). l. Direct stability and QCSCF (Gaussian). This is just a partial list of new features and models, and does not include substantial improvements to efficiency or reliability of the algorithms, almost all of which were done entirely at Gaussian. Leaving aside the numerous contributions made by Berny Schlegel, Krishnan Raghavachari, Gary Trucks, myself, and many others to versions through Gaussian 92, I think I can safely claim that few users of Gaussian 98 old enough to have used Gaussian 92 would agree with you that there is no difference in functionality between them. In fact, I think most of our users appreciate the substantial (and successful) efforts we've made to extend the program to address new areas of chemistry which our customers had indicated were important to them. You may not like what we've done, and everyone is certainly welcome to choose whatever software they find most useful to solve their problems, but to assert that all of our contributions over the last two decades are trivial enhancements to existing code ignores a great deal of published literature. Likewise, to assert that Gaussian, Inc. has not added value to our software ignores both the new predictive capabilities added in each version and the improvements in the performance and reliability of the software. Gaussian is operated by Ph.D. scientists who take our own research and the development of our field very seriously. To facilitate the development of the field (and in contrast with normal business practice), we publish our methods and algorithms, have no secret parameters, make source code available, offer low cost site-wide licenses to academics and provide academics with free technical support. While low license fees limit our resources both with respect to new research and to technical support capacity, I think that overall our record with respect to technical innovation is excellent and that the timeliness and quality of our responses to user questions is far better than most software companies provide even when charging for support, let alone at no cost. Mike Frisch Gaussian, Inc. (In the above, I have simplified the descriptions of research by listing only the research group(s) involved in the example efforts, in order to illustrate the types of contributions. This is not meant to downplay the very important work of all the individuals at Gaussian and within the academic groups who have contributed to recent versions of our software, including James Cheeseman, John Montgomery, Slava Zakrzewski, Eric Stratmann, John Burant, Stephan Dapprich, Odon Farkas, Maurizio Cossi, Roberto Cammi, Bennedetta Mennucci, Chris Pomelli, Carlo Adamo, Simon Clifford, Joe Octerski, Philippe Ayala, Nadia Rega, Pedro Salvador, Joe Dannenberg, David Malick, Angela Rabuck, James Foresman, Istvan Kamaromi, Roberto Gomperts, Mohammad Al-Laham, Todd Keith, Chun Yang Peng, and Richard Wong.) From chemistry-request@server.ccl.net Wed Dec 12 17:30:42 2001 Received: from phm.utoronto.ca ([128.100.70.1]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCMUgi18426 for ; Wed, 12 Dec 2001 17:30:42 -0500 Received: from phm.utoronto.ca (lpk04.phm.utoronto.ca [128.100.70.180]) by phm.utoronto.ca (8.9.3+Sun/8.9.3) with ESMTP id RAA15895 for ; Wed, 12 Dec 2001 17:30:25 -0500 (EST) Sender: william@phm.utoronto.ca Message-ID: <3C17DA8C.E8E15CD0@phm.utoronto.ca> Date: Wed, 12 Dec 2001 17:30:37 -0500 From: William Wei Reply-To: william@phm.utoronto.ca Organization: Faculty of Pharmacy, U of T X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX64 6.5 IP30) X-Accept-Language: en MIME-Version: 1.0 To: CCLer Subject: Gaussian run error--util.so Content-Type: multipart/alternative; boundary="------------99A6FFAAFF637F2B3D5FF308" --------------99A6FFAAFF637F2B3D5FF308 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, dear CCLers: I just got gaussian and want to get it run. After I compiled and ran it on the supercomputer, it gave me the following message: 6554108:./g98: rld: Fatal Error: Cannot Successfully map soname 'util.so' under any of the filenames /usr/lib64/util.so:/usr/lib64/internal/util.so:/lib64/util.so:/opt/lib64/util.so: For I have no right to write in those directories. Could anyone point me out how to solve the problem. Can I give gaussian the directory that the file in and how to do that? The second question is how can I constrain the peptide bond in a plane and what is the keywords. Thank you very much. Best regards. William. -- William Wei Tel: 1-416-946-7551 (Lab.) Faculty of Pharmacy Fax: 1-416-978-8511 19 Russell Street Email: william@phm.utoronto.ca Toronto, ON. M5S 2S2, Canada william.wei@utoronto.ca --------------99A6FFAAFF637F2B3D5FF308 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, dear CCLers:

I just got gaussian and want to get it run. After I compiled and ran it on the supercomputer, it gave me the following message:
6554108:./g98: rld: Fatal Error: Cannot Successfully map soname 'util.so' under any of the filenames /usr/lib64/util.so:/usr/lib64/internal/util.so:/lib64/util.so:/opt/lib64/util.so:
For I have no right to write in those directories. Could anyone point me out how to solve the problem. Can I give gaussian the directory that the file in and how to do that?

The second question is how can I constrain the peptide bond in a plane and what is the keywords.

Thank you very much.

Best regards.

William. 

-- 
William Wei                    Tel: 1-416-946-7551 (Lab.)
Faculty of Pharmacy            Fax: 1-416-978-8511
19 Russell Street              Email: william@phm.utoronto.ca
Toronto, ON. M5S 2S2, Canada          william.wei@utoronto.ca
  --------------99A6FFAAFF637F2B3D5FF308-- From chemistry-request@server.ccl.net Wed Dec 12 18:17:27 2001 Received: from sd.accelrys.com ([146.202.0.254]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCNHRi21179 for ; Wed, 12 Dec 2001 18:17:27 -0500 Received: (from osman@localhost) by sd.accelrys.com (8.11.0/8.11.0) id fBCNJ9m31377 for chemistry@ccl.net; Wed, 12 Dec 2001 15:19:09 -0800 Received: from OGUNER-LAP.accelrys.com (dhcp388.sd.accelrys.com [172.30.201.134]) by sd.accelrys.com (8.11.0/8.11.0) with ESMTP id fBCNJ8K31352; Wed, 12 Dec 2001 15:19:08 -0800 Message-Id: <5.1.0.14.2.20011212151338.060dd7d8@sparky2.sd.accelrys.com> X-Sender: osman@sparky2.sd.accelrys.com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 12 Dec 2001 15:16:55 -0800 To: qsar_society@accelrys.com From: "Osman F. Guner" Subject: Re: CCL:somewhat obscure chemistry quote Cc: chemistry@ccl.net, dek@cgl.ucsf.edu In-Reply-To: <200112121843.KAA353819@socrates.cgl.ucsf.edu> References: Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id fBCNHRi21182 Perhaps this one? "The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the difficulty is only that the exact application of these laws leads to equations much too complicated to be soluble." P.A.M. Dirac At 10:43 AM 12/12/2001, David Konerding wrote: >Now, here's a harder one. I also once came across a pair of quotes, about >the use of math in chemistry, where the first said somethign like >"We can compute anything" and the response was more or less "True, but >it will take a very long time". Given that my simulations each took over >3 months to run (10ns of duplex RNA in a box of water using 4 cpus and AMBER6) >I thought this would also be appropriate. --- Osman F. Güner, Ph.D. Director, Lead Identification & Optimization Accelrys Inc. (858) 799-5341 osman@accelrys.com http://www.accelrys.com From chemistry-request@server.ccl.net Wed Dec 12 18:19:33 2001 Received: from ccl.net ([192.148.249.4]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBCNJXi21349 for ; Wed, 12 Dec 2001 18:19:33 -0500 Received: from TheWorld.com (pcls1.std.com [199.172.62.103]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id SAA10987 for ; Wed, 12 Dec 2001 18:19:31 -0500 (EST) Received: from vaio.world.std.com (ppp0c151.std.com [208.192.102.151]) by TheWorld.com (8.9.3/8.9.3) with ESMTP id SAA27041 for ; Wed, 12 Dec 2001 18:19:15 -0500 Message-Id: <5.1.0.14.1.20011212181654.00aa7660@pop.theworld.com> X-Sender: jle@pop.theworld.com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 12 Dec 2001 18:22:24 -0500 To: chemistry@ccl.net From: Joe Leonard Subject: re: Gaussian revised GUI Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Folks, Who's willing to pay Gaussian, Inc. to rewrite their code so as to have a better/more modern GUI? Why not just get one of the QM packages which offers graphics and use that as a GUI for Gaussian? There's GaussView, and Spartan at least used to work with Gaussian (been too long since I've used it)... I've been in this business a long time, and I've not seen terribly many cases where people wanted useability vs. feature improvements (either for their support money, or whatever). Rewriting and validating code's a SIGNIFICANT undertaking, and not one to be taken on whim. I've only seen a little GAMESS and AMBER recently, and their input (Fortran NAMELIST, I think) is 1960's technology. So it's not just Gaussian to which this argument can be applied. FWIW, I've not seen anybody redoing MOPAC's GUI, or any other dusty-deck codes. There are probably new things which I am unaware of, as I've not worked in the QM business since 1995. It would not surprise me if doing the GUI (or API) is harder work than getting the algebra of the QM code "right", too. My 2 cents, Joe jle@world.std.com From chemistry-request@server.ccl.net Wed Dec 12 20:42:26 2001 Received: from mail.rnd.lgchem.co.kr ([150.150.103.10]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD1gNi27557 for ; Wed, 12 Dec 2001 20:42:24 -0500 Subject: YAeHMOP installation at Linux? MIME-Version: 1.0 From: "=?euc-kr?B?wMy787/t?=" To: chemistry@ccl.net Message-ID: Date: Thu, 13 Dec 2001 10:42:51 +0900 X-MIMETrack: Serialize by Router on LGCHEMDJ01/LGChem/LGChemical(Release 5.0.7 |March 21, 2001) at 2001/12/13 10:43:04 AM MIME-Version: 1.0 Content-Type: text/html; charset=EUC-KR Content-Transfer-Encoding: quoted-printable Untitled

Dear CCLer.

Is there anyone who successed to installed YAeHMOP software in Linux pla= tform (redhat 7.1)?

l'd like obtain some information about it.

In my case, i got the following error massage.

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

/usr/lib/gcc-lib/i386-redhat-linux/2.96/../../../lib= f2c.so: undefined reference to 'MAIN=5F=5F'

collect2: ld returned 1 exit status

make" *** [bind] error 1

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

And my system environment i= s...

 

redhat linux 7.1

kernel-2.4.3-12

gcc-2.96-85

gcc-g77-2.96-85

glibc-2.2.4-19

f2c-20000510-5

fort77-1.18-7

 

Please let me know~.

Thank you

 

~~~~~~~~~~~~~~~~~~~~~~~~~~~
Lee Sang Uck

Research Scientist  Analytical R&D Center

LG Chem, Ltd.

Research Park
TEL : 042-866-2793
FAX : 042-863-7466
HOME : http://my.netian.com/~deugi
~~~~~~~~~~~~~~~~~~~~~~~~~~~ 

 

= From chemistry-request@server.ccl.net Wed Dec 12 19:09:54 2001 Received: from postal.sdsc.edu (IDENT:LM74bMtN1sHorMAENfGnFfoX3r3d8be6@[132.249.20.114]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD09si23615 for ; Wed, 12 Dec 2001 19:09:54 -0500 Received: from koa.sdsc.edu (IDENT:0YPfted4yxog2m2GgHJEd+kRzPhH5Lal@koa.sdsc.edu [132.249.23.116]) by postal.sdsc.edu (8.11.6/8.11.6/SDSCserver-20) with ESMTP id fBD09aL29305; Wed, 12 Dec 2001 16:09:36 -0800 (PST) Received: from localhost by koa.sdsc.edu (SGI-8.9.3/1.11-IRIXclient) with ESMTP id QAA64547; Wed, 12 Dec 2001 16:09:36 -0800 (PST) Date: Wed, 12 Dec 2001 16:09:36 -0800 From: Wibke Sudholt To: David Konerding cc: Subject: Re: CCL:somewhat obscure chemistry quote In-Reply-To: <200112121843.KAA353819@socrates.cgl.ucsf.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I do not know if you mean the Dirac quote (it is more about quantum chemistry). If yes, please take a look at the summary of my question in spring this year: http://server.ccl.net/cgi-bin/ccl/message.cgi?2001+03+15+013 Hope that helps! Could you please summarize everything to the list? Best regards, Wibke Sudholt San Diego Supercomputer Center wibke@sdsc.edu On Wed, 12 Dec 2001, David Konerding wrote: > David Konerding writes: > > > >I'm looking for the source of a somewhat obscure chemistry quote I once > >came across but cannot find definitevly. I think it was written by a 19th > >century chemist who thought that math/computation in chemistry was wholly > >inappropriate. Does this ring a bell for anybody? > > Excellent, I have already received three messages that gave the quote. > This will make a great quote for my thesis, which of course, is about > using mathematics in chemistry :-) > > Now, here's a harder one. I also once came across a pair of quotes, about > the use of math in chemistry, where the first said somethign like > "We can compute anything" and the response was more or less "True, but > it will take a very long time". Given that my simulations each took over > 3 months to run (10ns of duplex RNA in a box of water using 4 cpus and AMBER6) > I thought this would also be appropriate. > > Dave > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Wed Dec 12 20:20:36 2001 Received: from ccl.net ([192.148.249.4]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD1Kai26582 for ; Wed, 12 Dec 2001 20:20:36 -0500 Received: from socrates.cgl.ucsf.edu (socrates.cgl.ucsf.EDU [128.218.27.3]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id UAA12342 for ; Wed, 12 Dec 2001 20:20:34 -0500 (EST) Received: (from ross@localhost) by socrates.cgl.ucsf.edu (8.9.3/8.9.3/GSC1.7) id RAA457783 for chemistry@ccl.net; Wed, 12 Dec 2001 17:20:19 -0800 (PST) Date: Wed, 12 Dec 2001 17:20:19 -0800 (PST) From: Bill Ross Message-Id: <200112130120.RAA457783@socrates.cgl.ucsf.edu> To: chemistry@ccl.net Subject: CCL:Gaussian revised GUI I've been in this business a long time, and I've not seen terribly many cases where people wanted useability vs. feature improvements This has been my perception as an AMBER developer. Rewriting and validating code's a SIGNIFICANT undertaking, and not one to be taken on whim. I've only seen a little GAMESS and AMBER recently, and their input (Fortran NAMELIST, I think) is 1960's technology. So it's not just Gaussian to which this argument can be applied. One misconception is that code needs to be rewritten to support a GUI. Actually, a GUI can be made to drive the old code via its old interface, by generating the input files and optionally launching the programs. ... There are probably new things which I am unaware of, as I've not worked in the QM business since 1995. It would not surprise me if doing the GUI (or API) is harder work than getting the algebra of the QM code "right", too. Yes, the friendly user interface that is also correct is a significant undertaking in its own right. It requires enough user base to justify the expense, particularly users with money and who are not handy with the sorts of abstractions that may be involved. Given the high percentage in science of folks who are used to working at the margins with hacked-together tools, one can imagine that some part of the industry will be able to survive by providing the best/latest science, whatever the interface. Bill Ross From chemistry-request@server.ccl.net Wed Dec 12 19:58:53 2001 Received: from ucidoor.unitedcatalysts.com ([208.23.162.2]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD0wqi25867 for ; Wed, 12 Dec 2001 19:58:52 -0500 Received: by ucidoor.unitedcatalysts.com; (8.8.8/1.3/10May95) id TAA17818; Wed, 12 Dec 2001 19:58:35 -0500 (EST) Received: from 10.1.0.50 by ucismtp02.unitedcatalysts.com (InterScan E-Mail VirusWall NT); Wed, 12 Dec 2001 19:58:10 -0500 (Eastern Standard Time) Received: from lvlxch01.unitedcatalysts.com ([10.16.100.88]) by lvlmail.unitedcatalysts.com (PMDF V6.0-24 #41777) with ESMTP id <0GO900002BRZI5@lvlmail.unitedcatalysts.com> for chemistry@ccl.net; Wed, 12 Dec 2001 19:52:47 -0500 (EST) Received: by lvlxch01.unitedcatalysts.com with Internet Mail Service (5.5.2650.21) id ; Wed, 12 Dec 2001 17:45:40 -0500 Content-return: allowed Date: Wed, 12 Dec 2001 17:45:34 -0500 From: "Shobe, Dave" Subject: Summary: 3 body problems To: "'CCL'" Message-id: <157A51F55AAAD3119CD70008C7B1629D01C15794@lvlxch01.unitedcatalysts.com> MIME-version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id fBD0wri25868 Thanks to the many people who answered my question about quantum 3-bodied problems. > Maybe someone here knows this. > > It is well known that the Schrödinger equation has only been analytically > solved for two-body problems: H, He+, Li++, ... > > My question (which may sound like hair splitting to some, but is actually a > profound question) is: whether the 3-body problems He and (H2)+ are unsolved > because people have not been clever enough to find the solution, or if they > are unsolved because they CAN'T be solved using the usual functions > (logarithms, sines, cosines, etc.) Just as certain integrals can't be > expressed, and one has to make up new functions like the Bessel functions to > "solve" the integration problem. Apparently, even classical-physics analogues like Sun-Earth-Moon or Sun-Earth-Jupiter are insoluble. A web site that was provided discusses this in about the level of detail I wanted :-) http://www.physics.cornell.edu/sethna/teaching/sss/jupiter/Web/Rest3Bdy.htm. And as one of you points out, the correspondence principle implies that if the general classical case is insoluble, the general quantum case is insoluble also. As for (H2)+, i.e. a hydrogen-molecule radical-cation, whether it is solvable depends on how one defines the problem. It turns out that *for a given internuclear distance*, the problem can be solved using confocal elliptical coordinates, and this solution is discussed in some quantum chemistry textbooks. --David Shobe Süd-Chemie Inc. phone (502) 634-7409 fax (502) 634-7724 email dshobe@sud-chemieinc.com Don't bother flaming me: I'm behind a firewall. From chemistry-request@server.ccl.net Wed Dec 12 19:35:18 2001 Received: from gandalf.cber.nih.gov ([128.231.52.5]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD0ZIi25136 for ; Wed, 12 Dec 2001 19:35:18 -0500 Received: from localhost (rvenable@localhost) by gandalf.cber.nih.gov (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id TAA81211; Wed, 12 Dec 2001 19:23:31 -0500 (EST) Date: Wed, 12 Dec 2001 19:23:31 -0500 From: Rick Venable To: Mike Frisch cc: chemistry@ccl.net Subject: Re: CCL:Gaussian 98 makefile policy In-Reply-To: <200112122123.QAA05160@svega.gaussian.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > > This brings up another point-- most of Gaussian, Inc.'s work is also > > "enhancements", to intellectual property (blueprints?) that is theirs > > only by purchase (from original author John Pople; I wonder how he feels > > about this). And one never seems to hear from Gaussian, Inc. (on CCL, > > anyway) about how to solve many of the problems that are posed on the > > mailing list. Based on the number of such questions, and a couple > > recent posts, it would also seem that Gaussian, Inc. doesn't offer much > > support to its paying customers in private, either. A classic case of > > "cash cow" syndrome. On Wed, 12 Dec 2001, Mike Frisch wrote: > This is an unfair characterization. Berny Schlegel, Krishnan > Raghavachari, and I have contributed substantially to the program for > over twenty years, and many others have contributed nearly as long. To > describe any version of Gaussian in the last two decades as > "enhancements" of the original author John Pople is grossly inaccurate > and does not give proper scientific credit to the long list of > authors. John Pople always acknowledged the work of others. In fact, > Gaussian 82 had important contributions from outside Pople's group, > the majority of improvements in Gaussian 86, 88, 90, and 92 were from > outside of his group, and it did not contribute at all to Gaussian 94 > or 98. Based on the length of the response, I guess I hit a nerve there. Nonetheless, I have heard privately from several individuals who suggest that many of those "outside" enhancements were more or less absorbed by Gaussian w/o paying the scientists who created them. Further, those in academia who continue to try to develop their own code enhancements to Gaussian have to go to great lengths to prevent Gaussian, Inc. from absorbing those enhancements as well. It seems inconsistent to invoke "intellectual property" arguments over a modified Makefile, given the apparent co-opting of other people's intellectual property. So why hasn't the Pople group continued to contribute code? If Gaussian support is so great, why all the questions on CCL? =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= Rick Venable FDA/CBER/OVRR Biophysics Lab 1401 Rockville Pike HFM-419 Rockville, MD 20852-1448 U.S.A. (301) 496-1905 Rick_Venable@nih.gov ALT email: rvenable@speakeasy.org =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= From chemistry-request@server.ccl.net Wed Dec 12 22:00:32 2001 Received: from ccl.net ([192.148.249.4]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD30Wi31409 for ; Wed, 12 Dec 2001 22:00:32 -0500 Received: from rwcrmhc52.attbi.com (rwcrmhc52.attbi.com [216.148.227.88]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id WAA13355 for ; Wed, 12 Dec 2001 22:00:30 -0500 (EST) Received: from C1353359A ([12.228.114.158]) by rwcrmhc52.attbi.com (InterMail vM.4.01.03.27 201-229-121-127-20010626) with SMTP id <20011213025944.STWF403.rwcrmhc52.attbi.com@C1353359A>; Thu, 13 Dec 2001 02:59:44 +0000 Reply-To: From: "Mark Thompson" To: "Joe Leonard" , Subject: RE: Gaussian revised GUI Date: Wed, 12 Dec 2001 18:54:22 -0800 Message-ID: <000601c18381$77961970$9e72e40c@attbi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: <5.1.0.14.1.20011212181654.00aa7660@pop.theworld.com> X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Importance: Normal Joe, You hit the nail on the head. My experience is that getting a GUI to do a good job of making the setting up of complicated tasks tractable is easily as time consuming as writing the underlying modeling codes. Mark ================================= Mark Thompson Planaria Software Seattle, WA. http://www.planaria-software.com ================================= -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request@ccl.net]On Behalf Of Joe Leonard Sent: Wednesday, December 12, 2001 3:22 PM To: chemistry@ccl.net Subject: CCL:Gaussian revised GUI FWIW, I've not seen anybody redoing MOPAC's GUI, or any other dusty-deck codes. There are probably new things which I am unaware of, as I've not worked in the QM business since 1995. It would not surprise me if doing the GUI (or API) is harder work than getting the algebra of the QM code "right", too. My 2 cents, Joe jle@world.std.com From chemistry-request@server.ccl.net Wed Dec 12 22:15:42 2001 Received: from mercury.chem.nwu.edu (gwuxi@[129.105.116.2]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD3Fgi32273 for ; Wed, 12 Dec 2001 22:15:42 -0500 Received: from localhost (gwuxi@localhost) by mercury.chem.nwu.edu (8.11.3/8.11.3) with ESMTP id fBD3FGP02626; Wed, 12 Dec 2001 21:15:16 -0600 (CST) Date: Wed, 12 Dec 2001 21:15:16 -0600 (CST) From: Guosheng Wu X-X-Sender: To: "Osman F. Guner" cc: , ccl , Subject: CCL:somewhat obscure chemistry quote In-Reply-To: <5.1.0.14.2.20011212151338.060dd7d8@sparky2.sd.accelrys.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > At 10:43 AM 12/12/2001, David Konerding wrote: > >Now, here's a harder one. I also once came across a pair of quotes, about > >the use of math in chemistry, where the first said somethign like > >"We can compute anything" and the response was more or less "True, but > >it will take a very long time". Given that my simulations each took over > >3 months to run (10ns of duplex RNA in a box of water using 4 cpus and AMBER6) > >I thought this would also be appropriate. Nothing is wrong with "we can compute anything"... the following may be a better response to that: The defination of "time" really depends on what kind of info you want to get, then what kind of approach you need to use. For example, if you use full-atomic simulation on DNA(10bp) melting(in ms) by AMBER or other packages, I could not imagine how many years it will take, and what kind of results you might get; however, use simple thermodynamics or statistical mechanics, high quality results can be "Got in 60 seconds"... Guosheng Wu, From chemistry-request@server.ccl.net Wed Dec 12 22:34:22 2001 Received: from mail.rpi.edu (root@[128.113.22.40]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD3YMi00479 for ; Wed, 12 Dec 2001 22:34:22 -0500 Received: from rpi.edu (rts10p18.xyp.rpi.edu [128.113.30.27]) by mail.rpi.edu (8.11.3/8.11.3) with ESMTP id fBD3Y4M126676 for ; Wed, 12 Dec 2001 22:34:04 -0500 Message-ID: <3C1821C7.21589614@rpi.edu> Date: Wed, 12 Dec 2001 22:34:32 -0500 From: "Curt M. Breneman" X-Mailer: Mozilla 4.77 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: Gaussian Flame War - a balanced view Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Folks, I feel that I should "weigh in" on this issue - I've known Mike Frisch since the late 80's and I've always felt that he and the rest of the Gaussian, Inc. team are good people and have a great deal of integrity. In addition, their efforts have for decades provided a high quality, continuously-evolving (and stable) QM environment for general use. They don't deserve the negative language that has appeared in certain recent postings. While their makefile distribution policy (and non-competition policy) appears unusual at first, the fact that they supply source code is also unique in the world of high-functionality software products. In order to understand their reluctance to allow free distribution of software components (such as modified makefiles), one has only to look back at their experiences with trying to support an early release of Gaussian when several "unauthorized" modifications were floating around. This led to a loss of quality control over the code that reflected negatively on their ability to fix problems, but did not seem to absolve them of public blame when specific functions didn't work correctly. I believe that this exasperating experience was one of the factors that prompted the formation of a company to distribute and maintain the software and to provide version control. Mike and the rest of the Gaussian group exist in a competitive market - If their policies are perceived by the consumers as problematic, then the market will adjust accordingly. I believe that they are astute enough to adapt as needed. In my view, they add quite a bit of value to the computational chemistry community. These are my $0.02; added to make sure that the argument is balanced. Curt Breneman RPI Chemistry (p.s. I've always had prompt response from Mike or Doug when we've had problems with the installation or use of new versions of Gaussian.) Mike Frisch wrote: > > > > Rick Venable writes: > > > > This brings up another point-- most of Gaussian, Inc.'s work is also > > "enhancements", to intellectual property (blueprints?) that is theirs > > only by purchase (from original author John Pople; I wonder how he feels > > about this). And one never seems to hear from Gaussian, Inc. (on CCL, > > anyway) about how to solve many of the problems that are posed on the > > mailing list. Based on the number of such questions, and a couple > > recent posts, it would also seem that Gaussian, Inc. doesn't offer much > > support to its paying customers in private, either. A classic case of > > "cash cow" syndrome. > > This is an unfair characterization. Berny Schlegel, Krishnan > Raghavachari, and I have contributed substantially to the program for > over twenty years, and many others have contributed nearly as long. To > describe any version of Gaussian in the last two decades as > "enhancements" of the original author John Pople is grossly inaccurate > and does not give proper scientific credit to the long list of > authors. John Pople always acknowledged the work of others. In fact, > Gaussian 82 had important contributions from outside Pople's group, > the majority of improvements in Gaussian 86, 88, 90, and 92 were from > outside of his group, and it did not contribute at all to Gaussian 94 > or 98. > > A glance at the references for methods added to Gaussian in the last > twenty years will point to dozens of papers by the other people > involved describing new science done by the participants in the form > of new models and algorithms developed and implemented in the > software. Some of this was done entirely by employees of Gaussian, > Inc., some was done by academic groups who we help to support, and > most was done via active scientific collaborations in which both the > intellectual content and the code include contributions from employees > and from academic collaborators. Here are a few examples, in no > particular order with respect to date or importance: > > a. TD-DFT (Scuseria group and Gaussian). > b. Solvation models (Numerous people at Gaussian, and in the > Tomasi and Barone groups). > c. ONIOM (Morokuma's group with contributions from people at Gaussian). > d. Magnetic properties (all done at Gaussian). > e. Semi-direct analytic MP2 frequencies and direct CIS frequencies > (all done at Gaussian). > f. Direct MCSCF energies, gradients, and frequencies (a collaboration > between people at Gaussian and Robb's group). > g. Conical intersection location (Schlegel and Robb). > h. Linear scaling methods (Scuseria's group and Gaussian). > i. Second order IRC optimization (Schlegel group). > j. Redundant internal coordinate optimization (Schlegel group and > Gaussian). > k. CBS methods (Petersson group and Gaussian). > l. Direct stability and QCSCF (Gaussian). > > This is just a partial list of new features and models, and does not > include substantial improvements to efficiency or reliability of the > algorithms, almost all of which were done entirely at Gaussian. > > Leaving aside the numerous contributions made by Berny Schlegel, > Krishnan Raghavachari, Gary Trucks, myself, and many others to > versions through Gaussian 92, I think I can safely claim that few > users of Gaussian 98 old enough to have used Gaussian 92 would agree > with you that there is no difference in functionality between them. > In fact, I think most of our users appreciate the substantial (and > successful) efforts we've made to extend the program to address new > areas of chemistry which our customers had indicated were important to > them. > > You may not like what we've done, and everyone is certainly welcome to > choose whatever software they find most useful to solve their > problems, but to assert that all of our contributions over the last > two decades are trivial enhancements to existing code ignores a great > deal of published literature. Likewise, to assert that Gaussian, > Inc. has not added value to our software ignores both the new > predictive capabilities added in each version and the improvements in > the performance and reliability of the software. > > Gaussian is operated by Ph.D. scientists who take our own research and > the development of our field very seriously. To facilitate the > development of the field (and in contrast with normal business > practice), we publish our methods and algorithms, have no secret > parameters, make source code available, offer low cost site-wide > licenses to academics and provide academics with free technical > support. While low license fees limit our resources both with respect > to new research and to technical support capacity, I think that > overall our record with respect to technical innovation is excellent > and that the timeliness and quality of our responses to user questions > is far better than most software companies provide even when charging > for support, let alone at no cost. > > Mike Frisch > Gaussian, Inc. > > (In the above, I have simplified the descriptions of research by > listing only the research group(s) involved in the example efforts, in > order to illustrate the types of contributions. This is not meant to > downplay the very important work of all the individuals at Gaussian > and within the academic groups who have contributed to recent versions > of our software, including James Cheeseman, John Montgomery, Slava > Zakrzewski, Eric Stratmann, John Burant, Stephan Dapprich, Odon > Farkas, Maurizio Cossi, Roberto Cammi, Bennedetta Mennucci, Chris > Pomelli, Carlo Adamo, Simon Clifford, Joe Octerski, Philippe Ayala, > Nadia Rega, Pedro Salvador, Joe Dannenberg, David Malick, Angela > Rabuck, James Foresman, Istvan Kamaromi, Roberto Gomperts, Mohammad > Al-Laham, Todd Keith, Chun Yang Peng, and Richard Wong.) > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net From chemistry-request@server.ccl.net Wed Dec 12 23:26:11 2001 Received: from axe.humboldt.edu (root@[137.150.148.10]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD4QAi02863 for ; Wed, 12 Dec 2001 23:26:11 -0500 Received: from zoellner (hsu4938.humboldt.edu [137.150.19.74]) by axe.humboldt.edu (8.9.3/8.9.3) with SMTP id UAA09422 for ; Wed, 12 Dec 2001 20:25:54 -0800 (PST) Message-Id: <3.0.1.32.20011212202528.00be3690@axe.humboldt.edu> X-Sender: rwz7001@axe.humboldt.edu X-Mailer: Windows Eudora Light Version 3.0.1 (32) Date: Wed, 12 Dec 2001 20:25:28 -0800 To: CHEMISTRY@ccl.net From: "Robert W. Zoellner" Subject: Gaussian Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Greetings All! After reading all of the posts on Gaussian and the failings (or lack thereof) of Gaussian Inc., I decided to do a search of Chemical Abstracts ... a very simple one. I searched: Gaussian 29922 citations ADF 1111 citations MOPAC 596 citations Jaguar 378 citations Spartan 215 citations GAMESS 117 citations AMPAC 73 citations AIMPRO 17 citations I chose these as each of these is a computational chemistry (or physics) program that I have used or am currently using (or hope to use). Of course, I was unsurprised that Gaussian got so many citations, but I was surprised at the overwhelming majority toward Gaussian ... until I realized that even when I used other programs, I either checked my results with similar Gaussian calculations, or used Gaussian calculations to carry out additional calculations using basis sets that were not available in the programs that I was using. While it might be argued that some of these citations are doubtless duplicates wherein two or more of the programs are cited, a search to check that indicated that less than 100 are duplicative. So why do we use Gaussian? It IS difficult to learn unless you have old-style programming knowledge (mine was in, of all things, FORTRAN), and sometimes the manuals and books are not a clear as one might like, and it is not always easy to get help ... but it is quite obviously the "industry standard" with which to compare anything else. And, I might add, IMHO, a pretty good program if you are not prone to expecting that everything must work perfectly the first time (and as a former bench chemist, NOTHING ever worked the first time I tried an experiment). I don't know the history of how Gaussian started, or how it has related to Professor Pople or others throughout its history and after Professor Pople ceased to be an integral part of Gaussian, but I really have had nothing but good results out of the program, overall ... not always instantaneously, but ultimately. But ... if you don't like the way Gaussian sells its product, or don't like how the product works, or what Gaussian allows you to do with its product, the free market should tell you to buy something in place of Gaussian. If the company sees a drop off in sales, perhaps it will change marketing strategy. If not, perhaps your criticisms are not universal enough ... there are options to everything. Just my thoughts ... probably more than two cents worth, but you can always hit "delete"! Bob Z. Robert W. Zoellner, Ph.D. Associate Professor Department of Chemistry Humboldt State University One Harpst Street Arcata, California 95521-8299 telephone: (707)826-3244 fax: (707)826-3279 From chemistry-request@server.ccl.net Wed Dec 12 23:45:27 2001 Received: from 263.net ([202.96.44.43]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id fBD4jQi03607 for ; Wed, 12 Dec 2001 23:45:27 -0500 Received: by 263.net (Postfix, from userid 60001) id 737331CBC4864; Thu, 13 Dec 2001 12:41:50 +0800 (CST) MIME-Version: 1.0 Message-Id: <3C18318E.01355@mta1> Content-Type: Multipart/Alternative; boundary="Boundary-=_kyeEKEUpgUtKrZIFOHMRFqvillXl" Date: Thu, 13 Dec 2001 12:41:50 +0800 (CST) From: "³Â»Ô" To: chemistry@ccl.net Subject: =?gb2312?B?dGhlIHVzZSBvZiBwaGFzZSBpbiBJUkM=?= X-Priority: 3 X-Originating-IP: [218.2.137.253] X-Mailer: Coremail2.0 Copyright Tebie Ltd., 2001 --Boundary-=_kyeEKEUpgUtKrZIFOHMRFqvillXl Content-Type: Text/Html; charset="gb2312" Content-Transfer-Encoding: 8bit Dear Sir:

When I do IRC calculation using Gaussian98,I can not make the option "phase'
running,can you tell me how to use this option.My route section is
# IRC(RCFC forward phase=(44,45) maxpoints=10)

thank you
chenhui
dept chemistry Nanjing univ
P.R.China





--Boundary-=_kyeEKEUpgUtKrZIFOHMRFqvillXl Content-Type: Text/Plain; charset="gb2312" Content-Transfer-Encoding: 8bit Dear Sir: When I do IRC calculation using Gaussian98,I can not make the option "phase' running,can you tell me how to use this option.My route section is # IRC(RCFC forward phase=(44,45) maxpoints=10) thank you chenhui dept chemistry Nanjing univ P.R.China




--Boundary-=_kyeEKEUpgUtKrZIFOHMRFqvillXl--