From: Todd Richmond [trichmon@jsd.claremont.edu] Sent: Tuesday, July 01, 1997 12:09 PM To: hyperchem@hyper.com Subject: pcAnywhere Greetings, Does anyone have experience running HC5 remotely via modem using something like pcAnywhere? Since I believe it actually runs the program on the remote machine, I'm hoping that the hardware key would not be an issue. BTW, the problem with other users not being able to run HC5 on the same machine was cured by changing the prefs file to access Runtime and not Program files. Thanks to Christine Pastorek for her help... Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://enzyme.claremont.edu From: Stavrev, Krassimir Sent: Tuesday, July 01, 1997 4:31 PM To: 'hyperchem@hyper.com' Subject: Web site Dear HyperChemists, We have re-designed our web site to better serve the needs of our customers. Please take a moment and visit our site at http://www.hyper.com. Feedback is encouraged--we look forward to suggestions for new content. We are also supplementing our e-mail list server with a NNTP news server. Please send us any recommendations you may have for appropriate forums to include on the Hypercube News Server to support@hyper.com. Regards, Krassimir --- Krassimir Stavrev, PhD Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 From: Udo Haberl [hudo@darwin.helios.nd.edu] Sent: Tuesday, July 01, 1997 4:19 PM To: hyperchem@hyper.com Subject: Calculations as background jobs? Greetings, does anyone know how to run HC5 calculations (ab initio) as background jobs under Windows NT Workstation 4.0? Our problem is: we run a windows NT network with several users and everyone has his own login account. When I log out and another user logs in, the calculation has to be terminated. This is no problem in UNIX programs (Gaussian). Should it be a problem in NT? Some background programs (for example a FTP server) keep on running, when I log out. Is this also possible for HC5? If not, it should be implemented in HC6. Thanks for any help, Udo P.S.: I also miss DFT methods in Hyperchem. Also a good idea for HC6. ---------------------------------------- Udo Haberl, Dipl.-Chem. ---------------------------------------- http://www.nd.edu/~hudo/ ---------------------------------------- e-mail: hudo@darwin.helios.nd.edu ---------------------------------------- current address: Department of Chemistry and Biochemistry University of Notre Dame Notre Dame, IN 46556 USA ---------------------------------------- Kekule-Institut fuer Organische Chemie und Biochemie Universitaet Bonn Gerhard-Domagk-Str. 1, D-53121 Bonn Germany ---------------------------------------- From: victoria@acdlabs.com Sent: Wednesday, July 02, 1997 9:23 AM To: Stavrev, Krassimir Subject: Re: Web site NICE!!! >Dear HyperChemists, > >We have re-designed our web site to better serve the needs of our >customers. Please take a moment and visit our site at >http://www.hyper.com. Feedback is encouraged--we look forward >to suggestions for new content. > >We are also supplementing our e-mail list server with a NNTP news >server. Please send us any recommendations you may have for appropriate >forums to include on the Hypercube News Server to support@hyper.com. > >Regards, >Krassimir >--- >Krassimir Stavrev, PhD >Hypercube, Inc. Florida Science and Technology Park >1115 N.W. 4th Street Gainesville, Florida 32601 >Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 > > From: Stavrev, Krassimir Sent: Wednesday, July 02, 1997 10:07 AM To: 'Udo Haberl' Subject: RE: Calculations as background jobs? Hi Udo, Unfortunately HyperChem 5 cannot be run as a background job under NT. Servers typically run as NT services - HyperChem does not run as a service but as a user application. On DFT, one thing that can be done is to try to interface HyperChem and DFT codes, such as Gaussian, for example, using the open architecture and the scripting facilities of HyperChem, such as the tcl-language support. This is a promising alternative that HyperChem users can explore. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Udo Haberl [SMTP:hudo@darwin.helios.nd.edu] Sent: Tuesday, July 01, 1997 4:19 PM To: hyperchem@hyper.com Subject: Calculations as background jobs? Greetings, does anyone know how to run HC5 calculations (ab initio) as background jobs under Windows NT Workstation 4.0? Our problem is: we run a windows NT network with several users and everyone has his own login account. When I log out and another user logs in, the calculation has to be terminated. This is no problem in UNIX programs (Gaussian). Should it be a problem in NT? Some background programs (for example a FTP server) keep on running, when I log out. Is this also possible for HC5? If not, it should be implemented in HC6. Thanks for any help, Udo P.S.: I also miss DFT methods in Hyperchem. Also a good idea for HC6. ---------------------------------------- Udo Haberl, Dipl.-Chem. ---------------------------------------- http://www.nd.edu/~hudo/ ---------------------------------------- e-mail: hudo@darwin.helios.nd.edu ---------------------------------------- current address: Department of Chemistry and Biochemistry University of Notre Dame Notre Dame, IN 46556 USA ---------------------------------------- Kekule-Institut fuer Organische Chemie und Biochemie Universitaet Bonn Gerhard-Domagk-Str. 1, D-53121 Bonn Germany ---------------------------------------- From: Palocsay, Frank A [palocsfa@jmu.edu] Sent: Wednesday, July 02, 1997 9:16 AM To: hyperchem@hyper.com Subject: ChemPlus 1.6 Is it possible to download ChemPlus 1.6 from HyperCube's web page? If so, how does one do this? Frank Palocsay -- Frank A. Palocsay (540) 568 6634 palocsfa@jmu.edu From: Stavrev, Krassimir Sent: Wednesday, July 02, 1997 11:38 AM To: 'palocsfa@jmu.edu' Subject: RE: ChemPlus 1.6 Frank: ChemPlus 1.6 is a free upgrade, from version 1.5, for those who already purchased 1.5 (retail price for academics $495). Therefore it is not available as a free download. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Palocsay, Frank A [SMTP:palocsfa@jmu.edu] Sent: Wednesday, July 02, 1997 9:16 AM To: hyperchem@hyper.com Subject: ChemPlus 1.6 Is it possible to download ChemPlus 1.6 from HyperCube's web page? If so, how does one do this? Frank Palocsay -- Frank A. Palocsay (540) 568 6634 palocsfa@jmu.edu From: Stavrev, Krassimir [stavrev@hyper.com] Sent: Wednesday, July 02, 1997 12:34 PM To: 'Todd Richmond' Cc: 'hyperchem@hyper.com' Subject: RE: pcAnywhere Hi Todd, In principle, it should be possible to access a server PC and run HyperChem remotely from a client site, if both the server and the client are running pcAnywhere simultaneously. One may expect such a connection to be slow though. Unfortunately we don't have much feedback on this from HyperChem users. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Todd Richmond [SMTP:trichmon@jsd.claremont.edu] Sent: Tuesday, July 01, 1997 12:09 PM To: hyperchem@hyper.com Subject: pcAnywhere Greetings, Does anyone have experience running HC5 remotely via modem using something like pcAnywhere? Since I believe it actually runs the program on the remote machine, I'm hoping that the hardware key would not be an issue. BTW, the problem with other users not being able to run HC5 on the same machine was cured by changing the prefs file to access Runtime and not Program files. Thanks to Christine Pastorek for her help... Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://enzyme.claremont.edu From: Todd Richmond [trichmon@jsd.claremont.edu] Sent: Wednesday, July 02, 1997 2:52 PM To: hyperchem@hyper.com Subject: Re: pcAnywhere At 09:09 AM 7/1/97 -0700, you wrote: >Does anyone have experience running HC5 remotely via modem using something >like pcAnywhere? Since I believe it actually runs the program on the remote >machine, I'm hoping that the hardware key would not be an issue. Here's an answer to my own question. I installed LapLink on both machines, and was able to remotely run HC5! With the hardware lock residing on the homst machine, I was able to use the "remote control" capabilities of Laplink and run the software...since the program was executing on the machine with the lock, it ran no problem. The graphics response was fairly slow, but usable (this was on an internal LAN via ethernet). I will try it next via modem (28.8) and report, but I suspect that it will be unacceptably slow. But FYI, LapLink (and assumedly pcAnywhere) DOES work (and the file synchronization capabilities are impressive too). Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://enzyme.claremont.edu From: BestChoice@massey.ac.nz Sent: Thursday, July 03, 1997 9:46 AM To: hyperchem@hyper.com Subject: How to unsubscribe from this newsgroup!! Hallo hyperchemists I'm becoming tired of the continuing messages to this newsgroup from people who wish to get off the newsgroupt. Following are the instructions i was sent about how to unsubscribe. >Reminder to HyperChem newsgroup subscribers: > >To unsubscribe: > Send e-mail to hyperchem-request@hyper.com > ^^^^^^^ > and in the body of the message, type 'unsubscribe hyperchem'. Please send the message to the right place. Judy ###################################### Judy Edwards Department of Chemistry Massey University Palmerston North New Zealand Phone (06) 350 4817 Email BestChoice@massey.ac.nz http://www.massey.ac.nz/~wwchem/Edwards.htm ###################################### From: Nigel Phillip Ewing [ewingnp@miamiu.acs.muohio.edu] Sent: Thursday, July 03, 1997 12:23 PM To: hyperchem@hyper.com Subject: Bio+ force field To whom it may concern, Our research group has HyperChem version 4.5 for SGI and we are attempting to install the force field parameters of the latest version of CHARMM (1996) into the Bio+ force field in HyperChem by modifying the "Bio85*.txt" parameter files. However, we noticed that the torsion parameters for Bio+ (Bio85tor.txt) are in a slightly different format from that reported in the original CHARMM version (Journal of Computational Chemistry, vol. 4, no.2, 187-217 (1983)). The torsion parameters in Bio+ have two additional terms: Gamma (phase offset in degrees) and N (periodicity of the fourier terms, from 1 to 6). The latest version of CHARMM only has the force constant (half-VN), the multiplicity and the phase offset. How should we handle this in creating a new HyperChem "*tor.txt" file? How important are the fourier terms to the successful compilation of the modified force field and in its use in calculations? If they are important, how did you determine the periodicity of the fourier term (N) for a given torsion? In addition, we also noticed that the newer version of CHARMM (1996) includes the H-bonding parameters in its van der Waal's non-bonding and bonding parameters. However, the formats are completely different from the Bio+ bonding and non-bonding van der Waal's "*npr.txt, *nbd.txt" files. How do we modify the format from CHARMM (1996) to comply with what is needed in Bio+ (bio85)? Could we leave these H-bonding parameters out and still obtain useful information from calculations using this modified force field? Thanks in advance for the assistance. **************************************************** Nigel Ewing Graduate Assistant Department of Chemistry & Biochemistry Oxford, Ohio 45056 513-529-2801 513-529-1667 fax From: Ed Bertaccini, MD [bert@vapa.icon.palo-alto.med.va.gov] Sent: Thursday, July 03, 1997 6:11 PM To: hyperchem@hyper.com Subject: Pentium Pro Bug Dear Group Members, Recently Intel reported a bug in the Pentium Pro processor floating point calculations. Does anyone know if this has had an effect on any of the highly iterative floating point calculations in Hyperchem? -------signature file attached to all documents------- Ed Bertaccini, MD Assistant Professor of Anesthesia, Stanford University School of Medicine Staff Anesthesiologist and Intensivist, Palo Alto VA Health Care System Research Associate, Molecular Research Institute of Palo Alto VA office phone: 415-493-5000 ext. 65180 VA office fax: 415-852-3423 From: Stavrev, Krassimir Sent: Monday, July 07, 1997 9:49 AM To: 'Ed Bertaccini, MD' Cc: 'hyperchem@hyper.com' Subject: RE: Pentium Pro Bug Hi Ed, To this date we have determined no adverse effects which the Pentium bug will have on HyperChem results, but we are continuing to investigate the situation. If any are discovered, users will be notified promptly. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Ed Bertaccini, MD [SMTP:bert@vapa.icon.palo-alto.med.va.gov] Sent: Thursday, July 03, 1997 6:11 PM To: hyperchem@hyper.com Subject: Pentium Pro Bug Dear Group Members, Recently Intel reported a bug in the Pentium Pro processor floating point calculations. Does anyone know if this has had an effect on any of the highly iterative floating point calculations in Hyperchem? -------signature file attached to all documents------- Ed Bertaccini, MD Assistant Professor of Anesthesia, Stanford University School of Medicine Staff Anesthesiologist and Intensivist, Palo Alto VA Health Care System Research Associate, Molecular Research Institute of Palo Alto VA office phone: 415-493-5000 ext. 65180 VA office fax: 415-852-3423 From: Stavrev, Krassimir Sent: Monday, July 07, 1997 10:21 AM To: 'Nigel Phillip Ewing' Subject: RE: Bio+ force field Hi Nigel, Fourier terms are used in estimating the dihedral angles; their periodicity (N=1 to 6) depends on the number terms available for a given angle. Such a summation over N allows the program to estimate various dihedral angle configurations between the cis- and trans- conformations. Also, hydrogen bonding effects are not included explicitly in the current version of Bio+. They could be important in systems where H-bonding is present. I'll try to obtain more detailed information related to your questions and will contact you subsequently. We interested in the implementation of the new Charmm set of parameters in HyperChem, as well as parameters for other force fields, such as AMBER4, for example. Feedback from HyperChem users working in these areas will be greatly appreciated. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Nigel Phillip Ewing [SMTP:ewingnp@miamiu.acs.muohio.edu] Sent: Thursday, July 03, 1997 12:23 PM To: hyperchem@hyper.com Subject: Bio+ force field To whom it may concern, Our research group has HyperChem version 4.5 for SGI and we are attempting to install the force field parameters of the latest version of CHARMM (1996) into the Bio+ force field in HyperChem by modifying the "Bio85*.txt" parameter files. However, we noticed that the torsion parameters for Bio+ (Bio85tor.txt) are in a slightly different format from that reported in the original CHARMM version (Journal of Computational Chemistry, vol. 4, no.2, 187-217 (1983)). The torsion parameters in Bio+ have two additional terms: Gamma (phase offset in degrees) and N (periodicity of the fourier terms, from 1 to 6). The latest version of CHARMM only has the force constant (half-VN), the multiplicity and the phase offset. How should we handle this in creating a new HyperChem "*tor.txt" file? How important are the fourier terms to the successful compilation of the modified force field and in its use in calculations? If they are important, how did you determine the periodicity of the fourier term (N) for a given torsion? In addition, we also noticed that the newer version of CHARMM (1996) includes the H-bonding parameters in its van der Waal's non-bonding and bonding parameters. However, the formats are completely different from the Bio+ bonding and non-bonding van der Waal's "*npr.txt, *nbd.txt" files. How do we modify the format from CHARMM (1996) to comply with what is needed in Bio+ (bio85)? Could we leave these H-bonding parameters out and still obtain useful information from calculations using this modified force field? Thanks in advance for the assistance. **************************************************** Nigel Ewing Graduate Assistant Department of Chemistry & Biochemistry Oxford, Ohio 45056 513-529-2801 513-529-1667 fax From: Dr. P F Heelis [heelisp@newi.ac.uk] Sent: Monday, July 07, 1997 12:02 PM To: hyperchem@hyper.com Subject: chemist's database Dear fellow chemist, I am compiling a database of useful information for European chemist's. It can be found at www.infochem.co.uk I would be grateful for any contributions of info under the headings, software, on-line journals, jobs, on-line courses, books, grant sources, conferences, european societies and institutes. Regards Paul Heelis p.s. apologies for duplicate mailings Dr Paul Heelis Reader in chemistry North East Wales Institute Wrexham, Ll11 2AW UK tel +44 1978 293372 fax +44 1978 290008 heelisp@newi.ac.uk editor@netsci.u-net.com Internet Journal of Science - Biological Chemistry http://www.netsci-journal.com/ From: Dr. P F Heelis [heelisp@newi.ac.uk] Sent: Monday, July 07, 1997 12:31 PM To: hyperchem@hyper.com Subject: First Internet Conference on Photochemistry Virtual conference notice Nov 17 - Dec 12 www.netsci-journal.com/ First internet conference on photochemistry and photobiology now open for submissions Paul Heelis" Dr Paul Heelis Reader in chemistry North East Wales Institute Wrexham, Ll11 2AW UK tel +44 1978 293372 fax +44 1978 290008 heelisp@newi.ac.uk editor@netsci.u-net.com Internet Journal of Science - Biological Chemistry http://www.netsci-journal.com/ From: Stavrev, Krassimir Sent: Tuesday, July 08, 1997 8:57 AM To: 'R.A.Carney@chemistry.leeds.ac.uk' Subject: RE: Macro problrms Hi Richy, Could you please e-mail me your macro to have a look at it? Please provide also your serial number and the operating system you are using. Thanks. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Richy C [SMTP:richardc@chemistry.leeds.ac.uk] Sent: Tuesday, July 08, 1997 5:00 AM To: hyperchem@hyper.com Subject: Macro problrms I have been running an excel macro (excel 7.0) to facilitate Hyperchem 5.01a for some time now which moves a molecule to set orientations around a stationary second molecule, then at each point the energy is minimised. Data is sent back to an excel spreadsheet. However for no apparent reason the macro has ceased to work, while the macro remains unchanged. I am absolutely baffled and would greatly appriciate any comments or aid in this matter. If anyone would like to know more about the macro being used dont hesitate to ask Look forward to hearing from someone soon, Richy C (School of Chemistry, University of Leeds, England) From: Todd Richmond [richmond@jsd.claremont.edu] Sent: Friday, July 11, 1997 11:01 PM To: hyperchem@hyper.com Subject: network install problems I have a single license network install I'm trying to get to work. I have the proper "network" dongle, and have installed HC5 on the "sever" machine (PentiumPro180 with 96Meg RAM), along with the Sentinel License server. The client machine (Cyrix166 with 64 Meg RAM, connected via ethernet-10B-T) has HC5 full install. Both have the Sentinel drivers installed. I can run HC on the server machine...it says that the hardware lock is not found, looks for the license server, and finds it. However it finds it via a local connection. I know this by disabling different network options in the pref. file. The client machine never can find the license. Anyone have any ideas? I played with this for an entire afternoon and got nowhere. The manual and its addendum indicate that the client machine install should have nothing but the drivers, but this seems stupid...I don't want to run the app on the server machine. It seems that the program should be smart enough to find the license server over the network and allow the local copy to run (and tech support indicated that this should work also). TIA, Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://bebop.claremont.edu Claremont, CA 91711 http://enzyme.claremont.edu From: Morten Langgaard [ml@virgil.ruc.dk] Sent: Saturday, July 12, 1997 5:46 AM To: Todd Richmond; Hyperchem@hyper.com Subject: Re: network install problems Dear Todd, Maybe I have a solution: > I have a single license network install I'm trying to get to work. I have > the proper "network" dongle, and have installed HC5 on the "sever" machine > (PentiumPro180 with 96Meg RAM), along with the Sentinel License server. The > client machine (Cyrix166 with 64 Meg RAM, connected via ethernet-10B-T) has > HC5 full install. Both have the Sentinel drivers installed. I can run HC on > the server machine...it says that the hardware lock is not found, looks for > the license server, and finds it. However it finds it via a local > connection. I know this by disabling different network options in the pref. > file. The client machine never can find the license. Which OS are you using? I assume that it is NT, else you wouldn't have the troubles. What you probably should do is to make a complete dongle driver installation on the client PC, even though it makes no sense. (It is actually enough just to run setupx386.exe file in the drivers directory). As soon as the dongle driver is installed, the client should be able to find the lock on the server. > > Anyone have any ideas? I played with this for an entire afternoon and got > nowhere. The manual and its addendum indicate that the client machine > install should have nothing but the drivers, but this seems stupid...I > don't want to run the app on the server machine. It seems that the program > should be smart enough to find the license server over the network and > allow the local copy to run (and tech support indicated that this should > work also). > I spend about a week on this mystery in February and the tech support didn't knew about this problem then. Thanks to Ross Dickson we finally located the failure. I don't know why HyperCube haven't reported this stupid failure on the FAQ list - I think they should. Both the Manual and the paper describing the network installation are wrong. Best regards Morten Langgaard Roskilde University, Denmark From: Stavrev, Krassimir Sent: Monday, July 14, 1997 10:32 AM To: 'richmond@jsd.claremont.edu' Cc: 'hyperchem@hyper.com' Subject: RE: network install problems Todd, Morten Langgaard is right in that, if you have a Windows NT 4.0, you need to have the hardware lock device driver running on the client site as well, even though the network hardware lock is installed on the NT 4 server. This NT specific feature was discovered lately and was published subsequently in the Frequently Asked Questions section, FAQ# 2b, in April 1997; thanks are due to Victoria Barclay and Ross Dickson. See also FAQ#17 on NT/printing problem issues. Perhaps the biggest problem with network installation and the associated documentation is fingered in these postings: Why does one not need executables on the client machine, if that's where HyperChem is going to run? The source of confusion here is that there are two distinct and independent modes of operation represented by the term "networking": (1) Network licensing. In this mode, there is a network license dongle attached to a network license server machine, and the client machines obtain licenses (i.e., permission to run HyperChem) from the license server. (2) File networking. In this mode, there is a file server machine which stores most of the HyperChem files and serves them to the client machines, which may themselves even be diskless. A "network installation" of HyperChem can correspond to either of these two situations, or to a combination of both of them. Situation #2 without situation #1 is going to be quite rare, though! Furthermore, the license server does not need to be the same machine as the file server, although this too will probably be an unusual situation. I agree that the network installation manual is outdated and needs a revision. It was supplemented by an addendum sheet which explains some new features and additions that were absent in the previous instructions. A revised version of the network installation guide is under way. In the meantime, please direct requests for network licensing support to support@hyper.com. Regards, Krassimir --- Krassimir Stavrev, PhD Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Todd Richmond [SMTP:richmond@jsd.claremont.edu] Sent: Friday, July 11, 1997 11:01 PM To: hyperchem@hyper.com Subject: network install problems I have a single license network install I'm trying to get to work. I have the proper "network" dongle, and have installed HC5 on the "sever" machine (PentiumPro180 with 96Meg RAM), along with the Sentinel License server. The client machine (Cyrix166 with 64 Meg RAM, connected via ethernet-10B-T) has HC5 full install. Both have the Sentinel drivers installed. I can run HC on the server machine...it says that the hardware lock is not found, looks for the license server, and finds it. However it finds it via a local connection. I know this by disabling different network options in the pref. file. The client machine never can find the license. Anyone have any ideas? I played with this for an entire afternoon and got nowhere. The manual and its addendum indicate that the client machine install should have nothing but the drivers, but this seems stupid...I don't want to run the app on the server machine. It seems that the program should be smart enough to find the license server over the network and allow the local copy to run (and tech support indicated that this should work also). TIA, Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://bebop.claremont.edu Claremont, CA 91711 http://enzyme.claremont.edu -----Original Message----- From: Morten Langgaard [SMTP:ml@virgil.ruc.dk] Sent: Saturday, July 12, 1997 5:46 AM To: Todd Richmond; Hyperchem@hyper.com Subject: Re: network install problems Dear Todd, Maybe I have a solution: > I have a single license network install I'm trying to get to work. I have > the proper "network" dongle, and have installed HC5 on the "sever" machine > (PentiumPro180 with 96Meg RAM), along with the Sentinel License server. The > client machine (Cyrix166 with 64 Meg RAM, connected via ethernet-10B-T) has > HC5 full install. Both have the Sentinel drivers installed. I can run HC on > the server machine...it says that the hardware lock is not found, looks for > the license server, and finds it. However it finds it via a local > connection. I know this by disabling different network options in the pref. > file. The client machine never can find the license. Which OS are you using? I assume that it is NT, else you wouldn't have the troubles. What you probably should do is to make a complete dongle driver installation on the client PC, even though it makes no sense. (It is actually enough just to run setupx386.exe file in the drivers directory). As soon as the dongle driver is installed, the client should be able to find the lock on the server. > > Anyone have any ideas? I played with this for an entire afternoon and got > nowhere. The manual and its addendum indicate that the client machine > install should have nothing but the drivers, but this seems stupid...I > don't want to run the app on the server machine. It seems that the program > should be smart enough to find the license server over the network and > allow the local copy to run (and tech support indicated that this should > work also). > I spend about a week on this mystery in February and the tech support didn't knew about this problem then. Thanks to Ross Dickson we finally located the failure. I don't know why HyperCube haven't reported this stupid failure on the FAQ list - I think they should. Both the Manual and the paper describing the network installation are wrong. Best regards Morten Langgaard Roskilde University, Denmark From: Steven Desjardins [desjardi.s@fs.sciences.WLU.EDU] Sent: Monday, July 14, 1997 6:08 AM To: hyperchem@hyper.com Subject: Zindo and color Hi I would like to hear from anyone who has used zindo/s to predict the color, i.e., visible spectrum of organic dyes. How accurate have the results been? Steve Steven G. Desjardins Department of Chemistry Washington and Lee University Lexington VA 24450 (540) 463-8873 desjardi.s@wlu.edu From: Stavrev, Krassimir Sent: Monday, July 14, 1997 12:43 PM To: 'Steven Desjardins' Cc: 'hyperchem@hyper.com' Subject: RE: Zindo and color Hi Steve, One can estimate the color of the molecule from the calculated absorption envelope. Zindo/S is in my opinion the best computational tool for this kind of problems. It has been applied successfully to predict the colors of organic dyes in solution - I believe the last example was on mero-dyes; you can try to contact Dr. Zerner at zerner@qtp.ufl.edu for full references. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Steven Desjardins [SMTP:desjardi.s@fs.sciences.WLU.EDU] Sent: Monday, July 14, 1997 6:08 AM To: hyperchem@hyper.com Subject: Zindo and color Hi I would like to hear from anyone who has used zindo/s to predict the color, i.e., visible spectrum of organic dyes. How accurate have the results been? Steve Steven G. Desjardins Department of Chemistry Washington and Lee University Lexington VA 24450 (540) 463-8873 desjardi.s@wlu.edu From: Laurence Lavelle [lavelle@mbi.ucla.edu] Sent: Monday, July 14, 1997 7:07 PM To: chemistry@infomeister.osc.edu; hyperchem@hyper.com Subject: OS and Computational Speed, NT is very close to UNIX You may also be interested to know that a paper on NT SPECmarks has been posted to http://www.digital.com/fortran/ntspeed.html The results discussed are the first SPEC95 submissions using Microsoft Windows NT. Both Intel and Alpha systems were measured under NT and UNIX. Performance under NT is excellent. Summary from above www site. You do not need heroic tuning. On average, SPEC peak results are only 7% better than base results. You can use NT. SPECint_base95 and SPECfp_base95 benchmarks compiled under Windows NT achieved 88% to 102% of their performance under Unix. As compilers continue to mature under Windows NT there is every reason to expect the performance to become 1:1. You can use mainstream compilers. The results mentioned just above were achieved with Microsoft Visual C++ and Digital Visual Fortran (on both Intel and Alpha). Regards Laurence """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" Laurence Lavelle, Ph.D. University of California Los Angeles Chemistry Department and Molecular Biology Institute Laboratory of Structural Biology & Molecular Medicine Los Angeles, CA90095-1570, USA Email:LAVELLE@MBI.UCLA.EDU Phone:(310)206-8270 Fax:(310)825-0982 http://www.mbi.ucla.edu/people/lavelle/lavelle.html """"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" From: Hiroto Tachikawa [hiroto@pc.cfa.hokudai.ac.jp] Sent: Wednesday, July 16, 1997 3:45 AM To: license@hyper.com Subject: License Code Re: License Code Dir Sir: We have recently bought HyperChem for SGI work Station. Recently IP address was exchanged. Please let me know new license code. User Name : Akira Shimizu Serial No. 600-10001313 System Level Installation out-put Code: 9068 9293 4611 9151 2 Please send me it as soon as possible. Dr. Akira Shimizu Graduate School of Engineering Hokkaido University Sapporo 060 JAPAN E-mail: hiroto@pc.cfa.hokudai.ac.jp From: Steven Desjardins [desjardi.s@fs.sciences.WLU.EDU] Sent: Wednesday, July 16, 1997 5:19 AM To: Hyperchem@hyper.com Subject: visualization for G94 Hi. I know that this has been discuused before, but I wasn't paying much attention at that point. I know that HC can be used to view structures generated by Gaussian 94, but can it display anything else, i.e., orbitals, epot surfaces, energy levels, etc? Thanks. Steve Steven G. Desjardins Department of Chemistry Washington and Lee University Lexington VA 24450 (540) 463-8873 desjardi.s@wlu.edu From: Peter Roose [Peter.Roose@rug.ac.be] Sent: Wednesday, July 16, 1997 2:05 PM To: hyperchem@hyper.com Subject: Unsubscribe unsubscribe Thanks Peter Roose peter.roose@rug.ac.be Universiteit Gent Krijgslaan 281 (S4bis) 9000 Gent (Belgium) From: catmin@math.math.unibuc.ro Sent: Friday, July 18, 1997 5:19 AM To: hyperchem@hyper.com Subject: "Mailbombing to a terrorist site" Please send this message to postmaster@igc.apc.org with subject: "Mailbombing to a terrorist site" and with text: The reason of the email bombs to the address ehj@apc.igc.org is because your server is hosting a terrorist web site.The basque-spanish terrorist band have recently kidnapped and killed in 48 hours a man. His name was Miguel Angel Blanco, he is dead now because a bunch of murderers thought that his live was not worth a penny. For that and many other past aberrations the spanish peoples are very angry. The electronical community expresses his angryness sending email bombs. So Please, we beg of you to cancel the terrorist's account in order the email bombings to stop. Thanks. regards, (Your name). Please distribute this message to everybody you know, distribution lists, and USENET news groups. From: Joel Polowin [JoelP@agiss.com] Sent: Friday, July 18, 1997 12:00 PM To: 'hyperchem@hyper.com' Subject: RE: "Mailbombing to a terrorist site" > Please send this message to postmaster@igc.apc.org with subject: > "Mailbombing to a terrorist site" and with text: > > The reason of the email bombs to the address ehj@apc.igc.org is [...] > Please distribute this message to everybody you know, distribution > lists, and USENET news groups. As usual, please do NOT distribute such messages; they are almost always fraudulent, and even when they are valid they tend to propagate indefinitely, causing trouble to the innocent. For example, even if the postmaster at "igc.apc.org" -- or is that "apc.igc.org"? The error argues against the validity of the message -- DID remove the supposed offending user, s/he would *still* keep getting mailbombs. If you have sent such messages, please follow up to your recipients and to the person who sent the message to you, explaining why these chain-propagating messages are a bad idea. The E-mail system here strips off almost all message headers; I can't see who was guilty of sending the chain message to the mailing list. But, for next time -- please don't. If you have web access, perhaps the best way of investigating such chain messages is to go to DejaNews, http://www.dejanews.com . This will let you search for discussion about Usenet messages and about E-mail abuse, by specifying keywords. Joel joelp@agiss.com From: Douglas John Miller [MILLERDJ@cedarnet.cedarville.edu] Sent: Friday, July 18, 1997 4:19 PM To: hyperchem@hyper.com Subject: ZINDO/S & Electronic Spectra I'm a relative new comer to the use of HyperChem (I just obtained version 5), and I would appreciate advice on the use of ZINDO/S with CI for yielding the trends in the electronic spectra of some conjugated organic oligomers (up to about 100 atoms and a MW of about 1,000). One of my colleagues, who views HyperChem as a black box, suggested the use of ZINDO/S. We are the most interested in the gap between the ground state and the first excited state. First: Is ZINDO/S a good (or the best) method, among either the semi-empirical or ab initio methods, in HyperChem for use in showing trends in spectra of organic compounds (as you go from one oligomer to another)? I know that ZINDO/S was optimized to yield electronic spectra for transition metal compounds, but I have seen it used for purely organic compounds. 2nd: How does one establish the best CI parameters in ZINDO/S? In particular, should the Pi-Pi Overlap Weighing Factor be 0.64 or 0.585 for organic oligomer? I have concerns about the value of the SCF Convergence Limit and the use of the "Accelerate Convergence" option, since both can slightly affect the spectra. I have been using 15 occupied and 15 unoccupied orbitals in the Orbital Criterion for the CI. Thanks for any help you can offer or any references of any type that you might mention. I have already looked through the manuals and the E-mail archives at HyperCube. Dr. Douglas J. Miller Professor of Chemistry Cedarville College Cedarville, OH millerdj@cedarville.edu PS: I'm running on a P100 with 128 MG of RAM and about 2 GB hard drive space. From: Stavrev, Krassimir Sent: Monday, July 21, 1997 11:25 AM To: 'Douglas John Miller' Cc: 'hyperchem@hyper.com' Subject: RE: ZINDO/S & Electronic Spectra Dr. Miller, INDO/S is specifically parameterized for spectroscopy. This fact makes Zindo unique and perhaps the best among other semiempirical approaches which usually do not include spectroscopic data in the parameter derivations. Zindo has been applied mostly to transition-metal systems, but also to a large number of organic molecules, see http://www.qtp.ufl.edu/zindo.html, and related sites. INDO/S explores the second quantized form of the Hamiltonian from which it drops all the 3- and 4-center two-electron integrals. All 1-center two-electron integrals are considered explicitly. 2-center two-electron integrals are kept only if they are of the form , and, to maintain rotational invariance all psi(a) and psi(b) in these integrals are considered to be of s-symmetry. The only modeling of Zindo/S parameters that was done originally is at the singles-only CI level of theory to reproduce spectroscopy. 1-center one-electron integrals are obtained from the experimental ionization potentials, electron affinities and Slater-Condon factors. A minimum basis set of Slater-type functions is used to evaluate the calculated properties; a double-zeta function is used to represent transition-metal d orbitals. The interaction factors, which weigh the overlap integrals in order to evaluate the off-diagonal elements for the resonance integrals in the Wolfsberg-Helmholtz approximation =(beta(a)+beta(b))*IntFactor*S(ab)/2, were first used in the Hoffmann's work on Extended Huckel theory, in which the value of IntFactor=1.75 was suggested for sigma-sigma resonance. In Zindo/S, two distinct factors were proposed for sigma-sigma (1.267) and for pi-pi interactions - 0.585 and 0.64. The sigma-pi overlap is small and is usually neglected. The value of 0.64 has been successfully used for triplets spectroscopy, in which case it was shown to yield better results than the 0.585 factor, see the original work of Zerner and co-authors quoted in the Computational Chemistry manual, p.295. The other value, 0.585, is most commonly used in the studies of transition-metal and organic systems. To my knowledge, the 0.64 value has not been very well tested and was applied only to triplet-triplet excitations. The SCF converges usually fast for organic molecules, the acceleration (DIIS) provided in HyperChem can be useful in cases where the SCF convergence slows down, a situation most common for transition-metal systems. For spectroscopic calculations, 10^-4 kcal/mol SCF convergence limit should provide a good accuracy of the orbitals for the subsequent CI. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Douglas John Miller [SMTP:MILLERDJ@cedarnet.cedarville.edu] Sent: Friday, July 18, 1997 4:19 PM To: hyperchem@hyper.com Subject: ZINDO/S & Electronic Spectra I'm a relative new comer to the use of HyperChem (I just obtained version 5), and I would appreciate advice on the use of ZINDO/S with CI for yielding the trends in the electronic spectra of some conjugated organic oligomers (up to about 100 atoms and a MW of about 1,000). One of my colleagues, who views HyperChem as a black box, suggested the use of ZINDO/S. We are the most interested in the gap between the ground state and the first excited state. First: Is ZINDO/S a good (or the best) method, among either the semi-empirical or ab initio methods, in HyperChem for use in showing trends in spectra of organic compounds (as you go from one oligomer to another)? I know that ZINDO/S was optimized to yield electronic spectra for transition metal compounds, but I have seen it used for purely organic compounds. 2nd: How does one establish the best CI parameters in ZINDO/S? In particular, should the Pi-Pi Overlap Weighing Factor be 0.64 or 0.585 for organic oligomer? I have concerns about the value of the SCF Convergence Limit and the use of the "Accelerate Convergence" option, since both can slightly affect the spectra. I have been using 15 occupied and 15 unoccupied orbitals in the Orbital Criterion for the CI. Thanks for any help you can offer or any references of any type that you might mention. I have already looked through the manuals and the E-mail archives at HyperCube. Dr. Douglas J. Miller Professor of Chemistry Cedarville College Cedarville, OH millerdj@cedarville.edu PS: I'm running on a P100 with 128 MG of RAM and about 2 GB hard drive space. From: Jens Spanget-Larsen [jsl@virgil.ruc.dk] Sent: Tuesday, July 22, 1997 4:57 AM To: Stavrev, Krassimir; 'hyperchem@hyper.com' Subject: RE: ZINDO/S & Electronic Spectra Stavrev, Krassimir" Mon, 21 Jul 1997: > ... The other value, 0.585, is most commonly used in the studies of > transition-metal and organic systems. To my knowledge, the 0.64 value > has not been very well tested and was applied only to triplet-triplet > excitations. ... Why, then, is 0.64 adopted as the default value in HyperChem, and not 0.585 which evidently is the standard value? Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Department of Chemistry Phone: +45 46757781 + 2710 Roskilde University (RUC) Fax: +45 46757721 P.O.Box 260 E-Mail: JSL@virgil.ruc.dk DK-4000 Roskilde, Denmark http://frederik.ruc.dk/dis/chem/psos =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From: Stavrev, Krassimir Sent: Tuesday, July 22, 1997 10:54 AM To: 'Jens Spanget-Larsen' Cc: 'hyperchem@hyper.com' Subject: RE: ZINDO/S & Electronic Spectra Jens, The HyperChem manuals are pointing to the fact that various numbers have been used as interaction factors, p.295 in the CC manual, HyperChem v.5, and provide some citations on the Zindo parameterization; there is also later work on this, see for example, Inorg. Chem. 25 (1986) 2728, Theor. Chim. Acta 71 (1987) 21. In HyperChem, and in other Zindo codes, one can change the interaction factors directly prior to the calculation. Again, to my knowledge, the 0.64 value is not well tested and is used mostly for triplet excitations. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Jens Spanget-Larsen [SMTP:jsl@virgil.ruc.dk] Sent: Tuesday, July 22, 1997 4:57 AM To: Stavrev, Krassimir; 'hyperchem@hyper.com' Subject: RE: ZINDO/S & Electronic Spectra Stavrev, Krassimir" Mon, 21 Jul 1997: > ... The other value, 0.585, is most commonly used in the studies of > transition-metal and organic systems. To my knowledge, the 0.64 value > has not been very well tested and was applied only to triplet-triplet > excitations. ... Why, then, is 0.64 adopted as the default value in HyperChem, and not 0.585 which evidently is the standard value? Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Department of Chemistry Phone: +45 46757781 + 2710 Roskilde University (RUC) Fax: +45 46757721 P.O.Box 260 E-Mail: JSL@virgil.ruc.dk DK-4000 Roskilde, Denmark http://frederik.ruc.dk/dis/chem/psos =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From: Sandy Yates [sandy@ideafix.cps.unizar.es] Sent: Thursday, July 24, 1997 9:19 AM To: hyperchem@hyper.com Subject: Babel frontend for Windows Hi, I discovered a Windows NT/95 frontend (32/16-bit) for Babel (1.6) called BABELwin from the author of Re-View 3.0 (http://www.brunel.ac.uk/depts/chem/ch241s/re_view/re_view.htm ). You need to have Babel installed already. (http://mercury.aichem.arizona.edu/babel.html) or a mirror at ftp://ccl.osc.edu/pub/chemistry/software/MAC/babel/ ) You can get it from http://www.brunel.ac.uk/depts/chem/ch241s/re_view/tagon.htm You are obliged to tell Jeffrey J. Gosper that you have downloaded it. Sandy ------------------------------------------------------------------------ Sandy Yates / Area de Mecanica de Fluidos / Centro Politecnico Superior Maria de Luna, 3 / 50015 ZARAGOZA. SPAIN Phone: Spain +34 76 761000 ext 5045. Fax: Spain +34 76 761882 Email: sandy@ideafix.cps.unizar.es From: KALPNA MISTRY [K.Mistry@mmu.ac.uk] Sent: Thursday, July 24, 1997 10:02 AM To: hyperchem@hyper.com Subject: Saving snapshot files Dear Sir, I am using hyperchem 4.5 to study a peptide. I am However, having problems. I have enclosed the peptide in a periodic box 18A square and carried out geomerty optimisation. Then I start a molecular dynamics run using a snapshots file to save the snapshot, however, at some point during the run it freezes up giving an error message "error writing to c:\hyper\ent3.snp". At first I thought it was due to lack of room on my hard drive and I have deleted a subtsantial number of unwanted files. This still did not solve my problem. This is the third e-mail message I am sending you requesting your help - could you please reply. Thank you Kalpna Mistry From: Todd Richmond [trichmon@jsd.claremont.edu] Sent: Friday, July 25, 1997 12:17 PM To: hyperchem@hyper.com Subject: surfaces on proteins I am hoping someone has some insight here... We have made a series of virtual mutants of a particular protein using HC5. AMBER was used to geometry optimize, and we generally make the mutation, optimize that residue, then select all residues within about 5 angstroms, and optimize those. We would now like to generate surfaces for one or two amino acids near the mutation, and get electrostatic potentials. I understand (I think) that we need to use semi-empirical methods (AM1?), but we have problems with the calculations finishing and/or the protein "blowing up". Anybody doing this kind of thing? Have any words of wisdom? Thanks, Todd Todd Richmond, Ph.D. (909) 607-1273 (voice) Assistant Professor of Chemistry (909) 621-8588 (fax) Joint Science Department richmond@jsd.claremont.edu The Claremont Colleges http://bebop.claremont.edu Claremont, CA 91711 http://enzyme.claremont.edu From: Felipe Medrano Valenzuela [fmedrano@guaymas.uson.mx] Sent: Monday, July 28, 1997 10:42 AM To: KALPNA MISTRY Cc: hyperchem@hyper.com Subject: Re: Saving snapshot files In the past I hadv e the same problem. In my case..a hard disk surface problem occurs. I recommend you to start a scandisk program to check any hardware failture in the hard disk (scan disk is included in DOS version 6.0 or higher) and in windows 95. Sorry about my bad english..if you have any problem please contact me. > > This is the third e-mail message I am sending you requesting your > help - could you please reply. > > Thank you > Kalpna Mistry > > > From: Richy C [richardc@chemistry.leeds.ac.uk] Sent: Monday, July 28, 1997 5:25 AM To: Hyperchem Subject: Macro problems I have been running an excel macro (excel 7.0) to facilitate Hyperchem 5.01a for some time now which moves a molecule to set orientations around a stationary second molecule, then at each point the energy is minimised. Data is sent back to an excel spreadsheet. However for no apparent reason the macro has ceased to work, while the macro remains unchanged. I am absolutely baffled and would greatly appriciate any comments or aid in this matter. If anyone would like to know more about the macro being used dont hesitate to ask Look forward to hearing from someone soon, Richy C (School of Chemistry, University of Leeds, England) From: Olgun Guvench [oguvench@rocketmail.com] Sent: Monday, July 28, 1997 5:26 PM To: hyperchem@hyper.com Subject: fake water Hello HyperChem users, I have recently started using HyperChem to study simulated annealing of peptides. In vacuo, it is no problem. However, once I try to use explicit water with periodic boundary conditions large enough to envelop my molecule (using switching radii 9A and 13A), the calculation becomes prohibitively long, to put it mildly. My question is: what might be some good ways to fake the effects of an aqueous environment (if any exist)? So far I have been using a distance dependent dielectric, but I am rather nervous about my results. Any opinions and/or literature references would be extremely appreciated. Thank you. Yours, Olgun Guvench _____________________________________________________________________ Sent by RocketMail. Get your free e-mail at http://www.rocketmail.com From: Fabri Marsicano [fabri@mpc.mintek.ac.za] Sent: Tuesday, July 29, 1997 7:20 AM To: hyperchem@hyper.com Subject: Excel VBA Macro Problems Hello Hyperchem Users, I am setting up an Excel macro (Excel 7) using Excel Visual Basic to drive Hyperchem V4.5 on a Pentium 200MHz PC. Having opened up a DDE channel (named "Channel") using the DDEInitiate command, I am having unexpected difficulty with a series of statements meant to send commands to Hyperchem through DDE. The problem statements are : DDEExecute Channel, "open-file " & Directory & Molecule & ".hin" DDEExecute Channel, "do-molecular-dynamics" DDEExecute Channel, "do-optimization" When the macro executes it opens the required molecule file in the Hyperchem window, and then promptly freezes. The cursor on the Excel screen shows an hourglass and nothing whatever happens from that point on. When I run the macro in a step-wise fashion using the debug facility in Excel, I find that macro execution gets to, and Hyperchem executes, the 1st of the above statements, but the macro execution does not emerge from statement 1. If I press the ESC key on the keyboard then execution does pass to the 2nd statement and Hyperchem performs the molecular dynamics. Having done that, macro execution freezes again. I need, once again, to press the ESC key on the keyboard to pass the macro onto the 3rd statement, which Hyperchem then dutifully executes. In summary, it appears that the DDE statement in the macro passes the relevant script instruction to Hyperchem. Hyperchem carries out the instruction, but Excel never gets to find out that Hyperchem has finished the job (unless the ESC key is pressed). Is there anyone out there that has had a similar problem and can help me interpret what is happening, and suggest a way out ? Many thanks in anticipation, FABRI MARSICANO From: Stavrev, Krassimir Sent: Tuesday, July 29, 1997 9:13 AM To: 'Olgun Guvench' Subject: RE: fake water Hi Olgun, The solvent model available presently in HyperChem includes explicit waters. Reaction field or polarizable continuum models, with gradients, are on our wish-list, but are presently unavailable. The only way one can "fake" the presence of the solvent is to use Langevin Dynamics simulations, in which the forces due to the solvent are included in the calculations through adding the atomic collision and friction forces at each MD step, see the CC manual, Chapter 7. You can also try to decrease the number real waters around the solute, or avoid the use of periodic box. For large molecular systems, less expensive methods for geometry minimization, such as molecular mechanics, are strongly recommended. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Olgun Guvench [SMTP:oguvench@rocketmail.com] Sent: Monday, July 28, 1997 5:26 PM To: hyperchem@hyper.com Subject: fake water Hello HyperChem users, I have recently started using HyperChem to study simulated annealing of peptides. In vacuo, it is no problem. However, once I try to use explicit water with periodic boundary conditions large enough to envelop my molecule (using switching radii 9A and 13A), the calculation becomes prohibitively long, to put it mildly. My question is: what might be some good ways to fake the effects of an aqueous environment (if any exist)? So far I have been using a distance dependent dielectric, but I am rather nervous about my results. Any opinions and/or literature references would be extremely appreciated. Thank you. Yours, Olgun Guvench _____________________________________________________________________ Sent by RocketMail. Get your free e-mail at http://www.rocketmail.com From: Stavrev, Krassimir Sent: Tuesday, July 29, 1997 11:56 AM To: 'Fabri Marsicano' Subject: RE: Excel VBA Macro Problems Fabri: DDE (Windows Dynamic Data Exchange, I assume you are running Win-95 or NT), is affected by some regional settings in the control section. The Language in the Control Panel/Regional Settings must be English(US), please check; also the Decimal Symbol need to be set to '.', not to ','. See if this helps. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Fabri Marsicano [SMTP:fabri@mpc.mintek.ac.za] Sent: Tuesday, July 29, 1997 7:20 AM To: hyperchem@hyper.com Subject: Excel VBA Macro Problems Hello Hyperchem Users, I am setting up an Excel macro (Excel 7) using Excel Visual Basic to drive Hyperchem V4.5 on a Pentium 200MHz PC. Having opened up a DDE channel (named "Channel") using the DDEInitiate command, I am having unexpected difficulty with a series of statements meant to send commands to Hyperchem through DDE. The problem statements are : DDEExecute Channel, "open-file " & Directory & Molecule & ".hin" DDEExecute Channel, "do-molecular-dynamics" DDEExecute Channel, "do-optimization" When the macro executes it opens the required molecule file in the Hyperchem window, and then promptly freezes. The cursor on the Excel screen shows an hourglass and nothing whatever happens from that point on. When I run the macro in a step-wise fashion using the debug facility in Excel, I find that macro execution gets to, and Hyperchem executes, the 1st of the above statements, but the macro execution does not emerge from statement 1. If I press the ESC key on the keyboard then execution does pass to the 2nd statement and Hyperchem performs the molecular dynamics. Having done that, macro execution freezes again. I need, once again, to press the ESC key on the keyboard to pass the macro onto the 3rd statement, which Hyperchem then dutifully executes. In summary, it appears that the DDE statement in the macro passes the relevant script instruction to Hyperchem. Hyperchem carries out the instruction, but Excel never gets to find out that Hyperchem has finished the job (unless the ESC key is pressed). Is there anyone out there that has had a similar problem and can help me interpret what is happening, and suggest a way out ? Many thanks in anticipation, FABRI MARSICANO From: Ed Bertaccini, MD [bert@vapa.icon.palo-alto.med.va.gov] Sent: Tuesday, July 29, 1997 1:11 PM To: Fabri Marsicano; hyperchem@hyper.com Subject: DDE problems Dear Hyperchem Users, I, too, have been having difficulties with the dynamic data exchange feature of Hyperchem 5.0. Specifically, I have attempted to perform conformational searches with the Chemplus search module using molecular mechanics optimization. My calculations also halt after a single optimization and never proceed to the next search angle. When using AM1 as the method for energy calculation with optimization, the calculation proceeds just fine. I have spoken with Stavrev Krassimir on several occasions and he has been very responsive to the issue through correspondence as well as sending me a new version of the conformational search executable. None of this has worked however. I have even tried reloading my operating system from scratch and loading and running only Hyperchem, again to no avail. My system is a 200MHz pentium pro with 64 MB of RAM running Windows NT 4.0 with service pack 3 loaded plus several of the post service pack 3 hotfixes. Based on what myself and others are finding, the problem seems to be related to the dynamic data exchange. It is interesting that I do not have this problem on my 486 Win95 system at home. Either way, it sounds like more in depth investigation needs to be performed by Hyperchem or, even possibly Microsoft, to resolve this rather pressing issue. I am confident that Hyperchem will follow through with more details in the near future. -------signature file attached to all documents------- Ed Bertaccini, MD Assistant Professor of Anesthesia, Stanford University School of Medicine Staff Anesthesiologist and Intensivist, Palo Alto VA Health Care System Research Associate, Molecular Research Institute of Palo Alto VA office phone: 415-493-5000 ext. 65180 VA office fax: 415-852-3423 From: Jerzy Malecki [malecki@rose.man.poznan.pl] Sent: Wednesday, July 30, 1997 3:10 AM To: stavrev@hyper.com; hyperchem@hyper.com Subject: DDE problems Hello! to Krassimir and all HyperChem Users, I met a problem with DDE communications to HyperChem v 5.03. I am using PENTIUM 133/32MB with Windows 95 and Excel v. 7.0. I would like to perform molecular mechanics calculations on large system - say 10 to 20 solvent molecules + one solute molecule, and than switch to semi-empirical method to obtain more accurate dipole moment and formation heat for the studied molecule, which may undergoes internal rotation. These calculations are performed using an appropriate Excel macro loop procedure with variable torsion angle. However, after 5 to 20 cycles (this number changes randomly) I unexpectedly get the messages: HyperNDO: "Cannot initialize DDE connection! Your system may be unstable". HyperChem: "Computation aborted. Unable to initiate conversation with backend" This happens when DDE tries to execute semi-empirical method. The appropriate part of Excel macro looks like this: 1 =WHILE(NOT(ISBLANK(SELECTION()))) 2 =EXECUTE(Channel,"[Set-bond-torsion("&SELECTION()&")]") 3 =EXECUTE(Channel,"[name-selection(torangle)]") 4 =EXECUTE(Channel,"[restraint(torangle,"&SELECTION()&",60)]") 5 =EXECUTE(Channel,"[select-none]") 6 =EXECUTE(Channel,"[do-optimization]") 7 =EXECUTE(Channel,"[use-no-restraints]") 8 =EXECUTE(Channel,"[do-single-point]") 9 =EXECUTE(Channel,"[select-name(torangle)]") 10 =RORMULA.ARRAY(REQUEST(Channel,"total-energy"),"rc[1]") 11 =EXECUTE(Channel,"[select-none]") 12 =EXECUTE(Channel,"[select-name(molecule1)]") 13 =EXECUTE(Channel,"[calculation-method(semi-empirical)]") 14 =EXECUTE(Channel,"[do-single-point]") 15 =FORMULA.ARRAY(REQUEST(Channel,"heat-of-formation"),"rc[2]") 16 =FORMULA.ARRAY(REQUEST(Channel,"dipole-moment"),"rc[3]") 17 =FORMULA.ARRAY(REQUEST(Channel,"dipole-moment-components"),"rc[4]") 18 =EXECUTE(Channel,"[select-none]") 19 =EXECUTE(Channel,"[select-name(torangle)]") 20 =SELECT("r[1]c") 21 =NEXT() After a number of cycles with no errors unexpectedly error occurs at step 14: =EXECUTE(Channel,"[do-single-point]"). However, if one passes this step, DDE works OK until calculation reach again step 14 !!! Do you have any suggestions? Thank you very much for your help. Jerzy Malecki Institute of Molecular Physics Polish Academy of Sciences Smoluchowskiego 17, 60179 PL Poznan, Poland Phone: (4861) 612360 Fax: (4861) 684524 From: Stavrev, Krassimir Sent: Wednesday, July 30, 1997 5:02 PM To: 'hyperchem@hyper.com' Subject: Excel macros Several HyperChem users have experienced difficulties running Excel'97 macros with HyperChem 4.5 and 5. Dr. Laszlo Jicsinszky has modified his previous NewCage macro, which generates a solvent box of any molecule, to run under the Excel versions 5 through 7, and with Excel'97. The macros are downloadable from http://www.hyper.com/support/software/Macros/macros_index.html. Here are some comment lines from the NewCage.xlm script that supplement the readme-file provided earlier: ' 'This macro runs under Excel'97 only! Please, note that some commands are working differently from older Excel versions! 'major differences relate to this macro: ' '1. Of course the macro requires more space (almost twice!!!) '2. ChDir command does not change the current directory of the application started from a macro '3. Application.DDEInitiate("Application","Topic") is not suitable to check whether a progam runs or not. ' If you are using the Application.DDEInitiate("Application","Topic") and your application has not been started, this command starts it!!! ' As a direct consequence this new (and "useful") feature, your application workspace will be full with HyperChem windows after several runs. '4. Now it is not possible to convert implicitly a number and add it to a string ' We will appreciate feedback from other users and will keep the newsgroup posted on this subject. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 From: Dr. Adrian Stefan Eckert [ase@IRIS.pharmchem.uni-muenchen.de] Sent: Thursday, July 31, 1997 5:55 AM To: guenter=szeimies@rz.hu-berlin.de; rainer=hansel@uv.hu-berlin.de; rainer=witzel@rz.hu-berlin.de; rkr@koe05.chemie.hu-berlin.de; license@hyper.com Subject: New E-mail address Sehr geehrte Damen und Herren, Dear Ladies and Gentlemen, meine E-mail - Adresse hat sich geaendert. my E-mail address has changed. Meine neue E-mail - Adresse lautet: My new E-mail address is: ase@pharmchem.uni-muenchen.de (Meine bisherige E-mail - Adresse (My former E-mail address eckert@IRIS.pharm-chem.pharmazie.uni-muenchen.de wird daneben nur noch kurze Zeit gueltig sein.) will stay valuable besides the new address only for a schort period.) Regards, A. Eckert LMU________________________________________________________________________ ______________ Institute of Pharmacy and Food Chemistry Ludwig________ Dr Adrian Stefan Eckert (system administrator) Maximilians___ Sophienstr.10, D-80333 Munich, Germany University____ Tel. +49-89-5902-547, Fax -447, Munich________ E-mail: ase@pharmchem.uni-muenchen.de From: Stavrev, Krassimir Sent: Thursday, July 31, 1997 8:32 AM To: 'ase@pharmchem.uni-muenchen.de' Subject: RE: New E-mail address Dr. Eckerd: if your notification is about the newsgroup we maintain, please unsubscribe with your old e-mail address and then subscribe with the new one as described on our Web site at http://www.hyper.com/support/join_mail.html : To join the HyperChem mailing list, click here. In the body of your message, enter: subscribe hyperchem to subscribe, OR enter: unsubscribe hyperchem [your name] [your e-mail address] to unsubscribe. This will delete/include your e-mail on the list-server automatically. Regards, Krassimir --- Krassimir Stavrev, PhD, support@hyper.com Hypercube, Inc. Florida Science and Technology Park 1115 N.W. 4th Street Gainesville, Florida 32601 Voice: 352-371-7744/1-800-960-1871 Fax: 352-371-3662 -----Original Message----- From: Dr. Adrian Stefan Eckert [SMTP:ase@IRIS.pharmchem.uni-muenchen.de] Sent: Thursday, July 31, 1997 5:55 AM To: guenter=szeimies@rz.hu-berlin.de; rainer=hansel@uv.hu-berlin.de; rainer=witzel@rz.hu-berlin.de; rkr@koe05.chemie.hu-berlin.de; license@hyper.com Subject: New E-mail address Sehr geehrte Damen und Herren, Dear Ladies and Gentlemen, meine E-mail - Adresse hat sich geaendert. my E-mail address has changed. Meine neue E-mail - Adresse lautet: My new E-mail address is: ase@pharmchem.uni-muenchen.de (Meine bisherige E-mail - Adresse (My former E-mail address eckert@IRIS.pharm-chem.pharmazie.uni-muenchen.de wird daneben nur noch kurze Zeit gueltig sein.) will stay valuable besides the new address only for a schort period.) Regards, A. Eckert LMU________________________________________________________________________ ______________ Institute of Pharmacy and Food Chemistry Ludwig________ Dr Adrian Stefan Eckert (system administrator) Maximilians___ Sophienstr.10, D-80333 Munich, Germany University____ Tel. +49-89-5902-547, Fax -447, Munich________ E-mail: ase@pharmchem.uni-muenchen.de From: Joel Polowin [JoelP@agiss.com] Sent: Wednesday, July 30, 1997 4:42 PM To: 'hyperchem@hyper.com' Subject: RE: DDE problems On Wedn esday, July 30, 1997 9:10 AM, Jerzy Malecki wrote: > I met a problem with DDE communications to HyperChem v 5.03. I am > using PENTIUM 133/32MB with Windows 95 and Excel v. 7.0. I would like to > perform molecular mechanics calculations on large system - say 10 to 20 > solvent molecules + one solute molecule, and than switch to semi-empirical > method to obtain more accurate dipole moment and formation heat for the > studied molecule, which may undergoes internal rotation. These calculations > are performed using an appropriate Excel macro loop procedure with variable > torsion angle. However, after 5 to 20 cycles (this number changes randomly) > I unexpectedly get the messages: > HyperNDO: "Cannot initialize DDE connection! Your system may be > unstable". > HyperChem: "Computation aborted. Unable to initiate conversation > with backend" > This happens when DDE tries to execute semi-empirical method. I have observed the same effect, using Visual Basic. It seems to involve HyperChem not allowing enough time for the new calculation module (ndo.exe) to be activated after the last calculation module was stopped, before it tries to communicate with it. I tried to get around this manually by starting the calculation module from Visual Basic *before* I tried to change to the new calculation type, in the hope that HyperChem would then connect with the active module. It didn't help much. Joel joelp@agiss.com