From h332262@sirius.cab.u-szeged.hu Mon Sep 29 05:27:21 1997 Received: from sirius.cab.u-szeged.hu for h332262@sirius.cab.u-szeged.hu by www.ccl.net (8.8.3/950822.1) id FAA01196; Mon, 29 Sep 1997 05:10:38 -0400 (EDT) Received: (qmail 21961 invoked by uid 32262); 29 Sep 1997 09:08:43 -0000 Date: Mon, 29 Sep 1997 11:08:43 +0200 (MET DST) From: Kiss Gergo X-Sender: h332262@sirius To: CHEMISTRY@www.ccl.net Subject: Re: RasMol problem under RH Linux 4.2 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sat, 27 Sep 1997, Kiss, Gergo wrote: > Hello everybody, > > I have just compiled the RasMol package on my PC under linux. It sent no > error messages, but when I want to start it, it says: > > RasMol Molecular Renderer > Roger Sayle, August 1995 > Version 2.6 > [8bit version] > > No suitable display detected! > RasMol> > > I have the Xfree86 3.3.1 server for X11, and it works fine(except rasmol). > Please help me, I need RasMol! > > Hi again, The problem is solved. I used the X11 in 16bpp and compiled RasMol with -DEGHTBIT in the Makefile. When I changed it to -DSIXTEENBIT and recompiled, the problem disappeared. Thanks for everyone who replied. Bye, Kiss, Gergo h332262@sirius.cab.u-szeged.hu From wibke@theochem.uni-duesseldorf.de Mon Sep 29 08:27:23 1997 Received: from lithium.theochem.uni-duesseldorf.de for wibke@theochem.uni-duesseldorf.de by www.ccl.net (8.8.3/950822.1) id IAA01757; Mon, 29 Sep 1997 08:10:53 -0400 (EDT) Received: from caesium.theochem.uni-duesseldorf.de (caesium.theochem.uni-duesseldorf.de [134.99.82.7]) by lithium.theochem.uni-duesseldorf.de (8.8.3/8.8.3) with ESMTP id OAA09307; Mon, 29 Sep 1997 14:10:54 +0200 (MET DST) Received: (from wibke@localhost) by caesium.theochem.uni-duesseldorf.de (8.8.3/8.8.3) id OAA03487; Mon, 29 Sep 1997 14:10:53 +0200 (MET DST) Date: Mon, 29 Sep 1997 14:10:53 +0200 (MET DST) From: Wibke Sudholt Message-Id: <199709291210.OAA03487@caesium.theochem.uni-duesseldorf.de> Subject: Summary: CASSCF/CASPT2 question Cc: wibke@lithium.theochem.uni-duesseldorf.de To: chemistry@www.ccl.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-MD5: GZS2MVzHcZ0uWeNBcKHjUQ== Dear CCLers, some time ago (sorry for my late reply, I was on a conference in the meantime) I posted a question regarding a systematic criterion for the inspection of CASSCF and CASPT2 calculations which I will try to summarize now. I do not get many answers (nevertheless, thanks very much to all who responded) and most of them were dealing with the choose of the right active space - but so called "chemical intuition" seem to be no very "systematic" criterion for me. Since I do not absolutely want to change the active space, I would still be happy to get more answers for my original question printed below. Our intermediate solution is to use state average calculations (the states lie very close together in CASSCF but not in CASPT2) although we first do not want to choose this "way out". As someone was interested in, the calculations were done with MOLCAS 3. Dipl.-Chem. Wibke Sudholt Institute of Theoretical Chemistry Heinrich-Heine-University Duesseldorf Germany wibke@theochem.uni-duesseldorf.de The original question was: ???????????????????????????????????????????????????????????????????????????????? Dear CCLers, perhaps someone of you can help me with this problem: I am doing excited state CASPT2 calculations with a CASSCF reference on a medium size organic molecule. Sometimes different starting orbitals lead to different CASSCF results probably because of several wavefunction minima. I suppose that this can be caused by an unsufficient active space, but I would run out of time and computational resources if I enlarge it. I also do not want to use state average calculations. So I need a systematic criterion to choose safely the "right" results. I am thinking of the lowest CASSCF energy (variation principle) or the highest weight of the CASSCF reference in the CASPT2 calculation (lowest perturbation). What would you prefer? Or are there other criterions? Can I compare calculations with different active spaces in the same way? I hope this question is not too stupid, but the subject is still relatively new to me. Any suggestions or pointers to the literature would be welcome. I will try to summerize. Thank you very much in advance Dipl.-Chem. Wibke Sudholt Institute of Theoretical Chemistry Heinrich-Heine-University Duesseldorf Germany wibke@theochem.uni-duesseldorf.de ???????????????????????????????????????????????????????????????????????????????? The answers were (except of calls for a summary - I hope I have not forgotten somebody): !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Dear Wibke, Firstly you can virtually guarantee that using a different set of orbitals as input for the CAS will lead to different wavefunctions. (although CAS is a FCI-type procedure the wavefunction will only be invarient w.r.t. orbital rotations WITHIN your CAS space- i.e. rotations w.r.t. core/virtual space which would change the spaces are important and should be "eliminated" through the choice of orbitals to put in your active space) On how to chose the correct orbitals/active space size this is still one of the black arts in modern computational chemistry, there are many texts which cover this question. The simplest solution is to use uhf natural orbitals and include those orbitals which have populations 1.98>n>0.01 but this is totally arbitrary and not that useful for reactivity where you may be interested in the formation of particular a bond(s). In this case the best way is to go empirical using your favourite molecular visualiser to look at your guess orbitals, and include the "right things" (e.g. sigma_g/sigma_u for a single bond dissociation). Other that that symmetry can be of use in making these choices - and i assume that i savailable if you intend to be running caspt2 afterwards. the main point i'd make is be very very sure you have the right active space [size + orbitals] before starting to run any pt2 calcn's - as they will only be reliable as the reference. noj -- Dr. Noj Malcolm Theory Group, Dept. of Chemistry, University of Manchester, Oxford Road, Manchester. M13 9PL e-mail noj.malcolm@man.ac.uk inter-bat http://mchhpi.ch.man.ac.uk/~mbdtsnm/ !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Dear Mr. Sudholt! I have some experience in the CASSCF calculations and probably would be able to be helpful. First of all, which program do you use? I hope it's MOLPRO, which is really cute in CASSCF. You can also control many things (especially when your system has some symmetry) with MOLPRO. The problem with the active space is common. Really, one should know the problem (molecule, # of orbitals/electrons) to judge. If you send me this information, it would be very helpful. In general, the solution is to keep your eye on the chemical meaning of the orbitals. Fro example, when studying a bond dissociation, in the region of the PES with small bond distances the mentioned bond could go down in energy and would be switched with higher lying orbitals. Usually the reason for choosing CASSCF is a clear idea on which orbitals would be mixed/correlated in the system/processes under study, but definitely not the variational principle -- using CASSCF, one looks for "static" correlation, without which the process just cannot be described properly. So if you will to discuss the problem in details, please, send a brief description of the problem to me. Sincerely, Dmitry Khoroshun. dima@euch4e.chem.emory.edu !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! From gryko@jsucc.jsu.edu Mon Sep 29 09:27:23 1997 Received: from jsucc.jsu.edu for gryko@jsucc.jsu.edu by www.ccl.net (8.8.3/950822.1) id JAA02010; Mon, 29 Sep 1997 09:10:55 -0400 (EDT) Received: from localhost (gryko@localhost) by jsucc.jsu.edu (8.8.7/8.8.5) with SMTP id NAA08698 for ; Mon, 29 Sep 1997 13:09:31 GMT Date: Mon, 29 Sep 1997 08:09:31 -0500 (CDT) From: Jan Gryko To: chemistry@www.ccl.net Subject: Alpha vs Pentium Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I am interested comparing performance of Alpha and Pentium PC on a typical DFT or MD code under Linux or NT. If you have any comments about Alpha vs Pentium please respond and I will post a summary to the list. Thanks J. Gryko Jacksonville State University From cyrillo@lua.ifi.unicamp.br Mon Sep 29 10:27:24 1997 Received: from sol.ifi.unicamp.br for cyrillo@lua.ifi.unicamp.br by www.ccl.net (8.8.3/950822.1) id JAA02153; Mon, 29 Sep 1997 09:42:30 -0400 (EDT) Received: (from daemon@localhost) by sol.ifi.unicamp.br (8.8.5/8.8.5) id KAA06079 for ; Mon, 29 Sep 1997 10:42:50 -0300 (EST) Received: from polimero2.ifi.unicamp.br(143.106.6.149) via SMTP by ifi-gw.unicamp.br, id smtpd006077; Mon Sep 29 10:42:50 1997 Received: (from cyrillo@localhost) by polimero2.ifi.unicamp.br (8.6.9/8.6.9) id KAA12925 for CHEMISTRY@www.ccl.net; Mon, 29 Sep 1997 10:40:15 -0300 From: Marcio Cyrillo - pos Message-Id: <199709291340.KAA12925@polimero2.ifi.unicamp.br> Subject: LOCALIZE keyword. To: CHEMISTRY@www.ccl.net (Lista de Quimica) Date: Mon, 29 Sep 1997 10:40:15 -0300 (EST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Dear Netters, I wonder if someone could help me explaning how the mopac's keyword LOCALIZE works. Any help will be appreciated. Thanks in advance, M.C. =-------------------------------------------------= Marcio Cyrillo - http://www.ifi.unicamp.br/~cyrillo Graduate Student State University of Campinas - Unicamp Campinas - SP - Brazil Applied Physics Department room 82 - phone +55 19 788 2383 home: +55 19 234 3494 =-------------------------------------------------= From qibvigap@lg.ehu.es Mon Sep 29 10:34:23 1997 Received: from lgsx01.lg.ehu.es for qibvigap@lg.ehu.es by www.ccl.net (8.8.3/950822.1) id JAA02175; Mon, 29 Sep 1997 09:54:40 -0400 (EDT) Received: from lgdx04.lg.ehu.es by lgsx01.lg.ehu.es (4.1/4.7 ) id AA22497; Mon, 29 Sep 97 15:53:03 +0100 Received: from localhost by lgdx04.lg.ehu.es; (5.65v3.2/1.1.8.2/08Oct96-0444PM) id AA28124; Mon, 29 Sep 1997 15:57:04 +0100 Date: Mon, 29 Sep 1997 15:57:04 +0100 (GMT+0100) From: Pablo Vitoria Garcia To: ccl Subject: Broken-simmetry states in G94 Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi all, I recently became interested in the calculation of the interactions in transition metal dimers. I am aware of the broken-simmetry state, spin projection technique of Noodleman (JCP 74,5737(1981) and other papers). All the calculations using this method are done (as far as I am aware) with the ADF program system. I must say that I don't fully understand the theory behind the method. Besides, I only have access to G94 for density functional methods. Could somebody explain to me how to do these type of calculations using Gaussian94? I mean, is there some 'recipe' to do it? I will also appreciate any help with the theory itself: references, suggestions,... Thank you very much for your help and time Pablo -------------------------------------------------------------------------------- Pablo Vitoria Garcia Departamento de Quimica Inorganica, Facultad de Ciencias Universidad del Pais Vasco (UPV/EHU) Apartado 644, E-48080 Bilbao SPAIN e-mail: qibvigap@lg.ehu.es Phone: +34 4 4647700 Ext. 2450 -------------------------------------------------------------------------------- From roth@organik.uni-erlangen.de Mon Sep 29 11:27:31 1997 Received: from faui45.informatik.uni-erlangen.de for roth@organik.uni-erlangen.de by www.ccl.net (8.8.3/950822.1) id KAA02339; Mon, 29 Sep 1997 10:39:50 -0400 (EDT) Received: from derioc1.organik.uni-erlangen.de (derioc1.organik.uni-erlangen.de [131.188.128.1]) by uni-erlangen.de with ESMTP id QAA19434 (8.7.6/7.5c-FAU); for ; Mon, 29 Sep 1997 16:39:46 +0200 (MET DST) Received: (from roth@localhost) by organik.uni-erlangen.de id QAA13958 (8.8.7/7.4f-FAU); for CHEMISTRY@www.ccl.net; Mon, 29 Sep 1997 16:39:40 +0200 (GMT-2:00) From: Henryette Roth Message-Id: <199709291439.QAA13958@derioc1.organik.uni-erlangen.de> Subject: Dallas ACS Meeting : "Computational Reaction Mechanisms" To: CHEMISTRY@www.ccl.net Date: Mon, 29 Sep 1997 16:39:38 +0200 (MET) Return-Receipt-To: roth@organik.uni-erlangen.de Receipt-Requested-To: roth@organik.uni-erlangen.de X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Dallas ACS Meeting : "Computational Reaction Mechanisms" Symposium ___________________________________________________________________ The above symposium will take place from Sunday to Tuesday (March 29 - 31, 1998) at the Dallas ACS meeting. Submitted lectures and posters are invited. Deadline for submissions is Oct. 10. All aspects of computational reaction mechanisms and techniques to treat them will be eligible for the symposium. .. Tim Clark Tim Clark Computer-Chemie-Centrum Naegelsbachstrasse 25 D-91052 Erlangen Germany clark@organik.uni-erlangen.de Tel: +49-9131-852948 FAX: +49-9131-856565 From berriz@chasma.harvard.edu Mon Sep 29 14:27:25 1997 Received: from chasma.harvard.edu for berriz@chasma.harvard.edu by www.ccl.net (8.8.3/950822.1) id NAA03655; Mon, 29 Sep 1997 13:32:11 -0400 (EDT) Received: by chasma.harvard.edu (AIX 3.2/UCB 5.64/4.03) id AA17925; Mon, 29 Sep 1997 13:29:41 -0400 Date: Mon, 29 Sep 1997 13:29:41 -0400 From: berriz@chasma.harvard.edu (Gabriel Berriz) Message-Id: <9709291729.AA17925@chasma.harvard.edu> To: chemistry@www.ccl.net Subject: Dihedral angle restoring force Dear netters: I'm looking for computationally efficient code (or algorithm) to compute the restoring forces (magnitudes and directions) corresponding to a potential dependent on the dihedral angle formed by three consecutive bonds along a chain. The desired forces act on the four atoms forming the three bonds. Any pointers would be much appreciated. Thank you. Gabriel Berriz berriz@chasma.harvard.edu From brian@ernie.chem.wsu.edu Mon Sep 29 15:27:27 1997 Received: from cheetah.it.wsu.edu for brian@ernie.chem.wsu.edu by www.ccl.net (8.8.3/950822.1) id OAA04040; Mon, 29 Sep 1997 14:48:10 -0400 (EDT) Received: from ernie.chem.wsu.edu (ernie.chem.wsu.edu [134.121.43.23]) by cheetah.it.wsu.edu (8.8.5/8.8.5) with SMTP id LAA09497 for ; Mon, 29 Sep 1997 11:47:21 -0700 (PDT) Received: by ernie.chem.wsu.edu (AIX 4.1/UCB 5.64/4.03) id AA51838; Mon, 29 Sep 1997 11:45:17 -0700 Received: from gateway.photon.com by bert.chem.wsu.edu (AIX 3.2/UCB 5.64/4.03) id AA24395; Sat, 27 Sep 1997 13:41:10 -0700 Received: from worf.moldyn.com ([172.16.2.22]) by gateway.photon.com with SMTP id <5993-2>; Sat, 27 Sep 1997 13:26:15 -0700 Received: by worf.moldyn.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) for brian@bert.chem.wsu.edu id QAA29118; Sat, 27 Sep 1997 16:41:01 -0400 Date: Sat, 27 Sep 1997 13:41:01 -0700 From: dnc@worf.moldyn.com (Donovan Chin) Message-Id: <199709272041.QAA29118@worf.moldyn.com> To: brian@bert.chem.wsu.edu Subject: IMPORTANT MESSAGE FROM CHARMM-BBS Sender: brian@ernie.chem.wsu.edu __________________T_H_E__C_H_A_R_M_M__B_B_S_______________________________ Dear all, The charmm-bbs has been dead for some time. The machine that ran this bbs expired a few months ago. The bbs probably will not continue from my old group at Harvard (Whitesides) for obvious reasons; its not feasible to run charmm-bbs from here at Moldyn. Charmm-bbs may reappear if there are any brave soles out there. The charmm-bbs requires knowledge of Unix, unix mail, and Perl. Charmm-bbs is very robust: the bbs requires very little maintenance once implemented. I have placed what log files I have of posted messages and responses to the bbs over the last 5 years on ftp.photon.com/incoming (anonymous login). The log files are: chmbbs_jun92_to_dec92.log.gz chmbbs_jan93_to_dec93.log.gz chmbbs_jan94_to_dec94.log.gz chmbbs_jan95_to_oct95.log.gz chmbbs_jan96_to_current.log.gz You will not be able to see these files, so you must type each name explicitly. I hope at least some of you have benefited from charmm-bbs over the years. Thanks to BB for providing some legitimacy during the initial inception of charmm-bbs; thanks to MK for patience and no legal actions; thanks to JL for providing the bbs software. Regards, Donovan -------------------------------------------------------------------- Donovan N. Chin (v) 617-354-3124 x21 Moldyn Inc. (f) 617-491-4522 955 Massachusetts Ave., 5th Floor (e) dnc@moldyn.com Cambridge, MA 02139-3180 --------------------------------------------------------------------