From jojo@chem.ruu.nl Fri Dec 2 09:07:48 1994 From: jojo@chem.ruu.nl (Jo van Bilsen) Subject: Problem with network license server To: hyperchem@hyper.com Date: Fri, 2 Dec 94 14:23:28 MET Organization: Academic Computer Centre Utrecht, University of Utrecht, Budapestlaan 6, NL-3584 CD Utrecht, The Netherlands. Phone: +31 30 531436 Fax: +31 30 531633 We have some problems running HyperChem in combination with a novell server and the license key manager "NSRVNI.NLM". We have a server with 2 ethernet interface cards. One to a local net with PC's with only a floppy drive. The other to a WAN with internet connection. If we run HyperChem on a PC using the windows software from te server with the license key attached to the back of the PC everything works fine, on both networks. If we run the "NSRVNI.NLM" manager on the server with the license key attached we get the message "No license could be obtained .." The server gives the message curent users 0 peak users 0 We also like to know if we can restrict access to one of the two networks. Greetings Jo van Bilsen jojo@chem.ruu.nl -- --- Jo van Bilsen My tel. no. 030-531731/531436 (Holland) Mailadress : jojo@cc.ruu.nl Debugging involves rules, logic, hunches, intuition, source listings, experience tedium, and many caffeine-enriched beverages. (How to write Macintosh Software, Scott Knaster) ________________________________________________________________________ From BERGER@wuchem.wustl.edu Fri Dec 2 15:24:35 1994 From: BERGER@wuchem.wustl.edu Date: Fri, 02 Dec 1994 12:20:47 -0600 (CST) Subject: 3 different HyperChems? To: hyper@hyper.com The latest issue of "Today's Chemist at Work" listed three different "versions" (??) of HyperChem: HYPERCHEM National Biosciences, Inc. HyperChem Autodesk, Inc. HyperChem Europa Scientific Software Corp. I suppose that these are all different vendors? They give somewhat different descriptive blurbs for each address; the technical info (e.g. price and platform) are all the same. Curiously yours, Dan Berger berger@wuchem.wustl.edu ________________________________________________________________________ From GF@chemie.uni-paderborn.de Mon Dec 5 07:04:47 1994 From: "Gregor Fels" To: hyperchem@hyper.com Date: Mon, 5 Dec 1994 12:07:31 GST Subject: Macro Dear netters somebody had offered a macro for HyperChem that generates energy surface plots from Excel. Unfortiunately I missed the authors e-mail number. So, if this macro is different from the one HyperChem offers with ChemPlus I am very interested to try it. Could somebody point me to where I can get it? Gregor Dr. Gregor Fels Tel.: +49-(0)5251-602181 Universitaet-GH-Paderborn Fax: +49-(0)5251-603245 FB 13: Chemie und Chemietechnik email: gf@chemie.uni-paderborn.de Organische Chemie Warburgerstr. 100 D-33098 Paderborn Germany ________________________________________________________________________ From GF@chemie.uni-paderborn.de Mon Dec 5 08:05:41 1994 From: "Gregor Fels" To: hyperchem@hyper.com Date: Mon, 5 Dec 1994 13:06:29 GST Subject: HyperChem scripts Dear netters, I am interested in setting up a collection of organic reactions with the help of HyperChem that will assist understanding "organic textbook chemistry" and which will be re-run as script-files during lectures. Before I start doing this (over the next few months) I like to check if somebody already has put some work into this sort of thing and is willing to share the wealth. It would of course be of tremendous help for me. Gregor Dr. Gregor Fels Tel.: +49-(0)5251-602181 Universitaet-GH-Paderborn Fax: +49-(0)5251-603245 FB 13: Chemie und Chemietechnik email: gf@chemie.uni-paderborn.de Organische Chemie Warburgerstr. 100 D-33098 Paderborn Germany ________________________________________________________________________ From rbobrovs@email.tuwien.ac.at Wed Dec 7 15:05:31 1994 Date: Wed, 07 Dec 1994 18:51:35 -0100 To: hyperchem@hyper.com From: rbobrovs@email.tuwien.ac.at (Roman Bobrovsky) Subject: Dongle Probs using NT Dear netters, in order to run HyperChem under NT 3.5 I tried to install the Rainbow Driver. The dongle is connected to LPT1 at 0x3bc. Unfortunately I cannot figure out the right settings for "Actual Port" (I tried Port 1, Settings are: Bus Address=0, Bus Number=Bus 0, Bus Type = internal, Address Space = I/O) (my computer is a 486DX4, local bus, adapter card in an ISA slot). After booting, NT states that "at least one service failed to start up", and I am getting two messages in the event log: Serial: User configuration for parameter Serial1 must have PortAddress. Service Control Manager: The following boot-start or system-start driver(s) failed to load: RainPort Moreover, in the driver-setup, no ports are active (i.e. grayed). Any suggestions? Roman --- Bobrovsky Roman Institute of Organic Chemistry Vienna University of Technology Getreidemarkt 9 A-1060 Vienna AUSTRIA Phone: (++43-1) 588 01-4998 Internet: rbobrovs@email.tuwien.ac.at FidoNet: 2:310/30.33 ________________________________________________________________________ From 100046.123@compuserve.com Thu Dec 8 08:14:05 1994 Date: 08 Dec 94 07:34:08 EST From: Mike Taylor <100046.123@compuserve.com> To: "Hyperchem User's Group" Subject: Re: Dongle Probs using NT Hello, RE: Dongle Probs using NT, reported by Roman Bobrovsky at rbobrovs@email.tuwien.ac.at Roman, you should try changing the Bus Address to 3bc, if you are sure that is the address used by your LPT1 port (our computers mostly have LPT1 at 0x378). My machine is running Windows NT 3.1 with LPT1 at 0x378 and the Rainbow driver loads OK with the settings for "Actual Port 1" as Port 1, Bus Address=378, Bus Number=Bus 0, Bus Type = internal (or ISA), Address Space = I/O). However although the driver loads I get inordinate delays every time it tries to validate the licence by accessing the dongle (the program starts, but selecting File | New causes a 2-3 minute freeze). So if you can get the driver to work, I would be very interested to know if you get any similar delays. Best of luck -=Mike Taylor=- Paint Research Association, TEDDINGTON, UK via CompuServe at: 100046.123@compuserve.com ________________________________________________________________________ From hurst Thu Dec 8 13:18:49 1994 To: rbobrovs@email.tuwien.ac.at (Roman Bobrovsky) Subject: Re: Dongle Probs using NT (long) Cc: hyperchem Reply-To: hurst@hyper.com Roman Bobrovsky writes: > in order to run HyperChem under NT 3.5 I tried to install the Rainbow > Driver. The dongle is connected to LPT1 at 0x3bc. Unfortunately I cannot > figure out the right settings for "Actual Port" (I tried Port 1, Settings > are: Bus Address=0, Bus Number=Bus 0, Bus Type = internal, Address Space = > I/O) (my computer is a 486DX4, local bus, adapter card in an ISA slot). > After booting, NT states that "at least one service failed to start up", and > I am getting two messages in the event log: > > Serial: > User configuration for parameter Serial1 must have PortAddress. > > Service Control Manager: > The following boot-start or system-start driver(s) failed to load: RainPort > > Moreover, in the driver-setup, no ports are active (i.e. grayed). > > Any suggestions? The crucial point, as I think Mike Taylor mentioned, is setting the Bus Address to its correct value. Here is a draft of improved instructions that we'll be sending out with the driver in the future (steps 1 and 9-12 should help you): 1. You will need to determine how many hardware parallel ports the computer has and what their bus addresses are. One way to do this is to reboot the computer to DOS (using a bootable DOS floppy if your computer isn't configured for dual boot) and run a diagnostic program such as MSD.EXE that comes with DOS and Windows to determine the parallel port information. In MSD (Microsoft Diagnostics) type "L" to see LPT Ports information, and ignore the leading zeroes and "H" suffix (indicating hexadecimal) for the Port Address. IMPORTANT: Do not do this in a DOS box under Windows or Windows NT, as the information may not be correct! It must be done from DOS alone. 2. Under the Microsoft Windows NT Main group, select Control Panel. 3. In the Control Panel group, select Drivers. 4. When the driver list appears, select the ADD button. 5. Choose "Unlisted or Updated" from the Driver list and select the OK button. 6. Place the Rainbow port driver diskette in the disk drive, specify its location and select OK. 7. The Rainbow port driver and associated files are copied to the hard disk. The files RAINPORT.DLL, RAINPORT.HLP and RAINVDD.DLL will be copied into \%SYSTEMROOT%\SYSTEM32. The file RAINPORT.SYS will be copied into \%SYSTEMROOT%\SYSTEM32\DRIVERS. %SYSTEMROOT% is the directory where Microsoft Windows NT has been installed. 8. The Rainbow port driver setup program is automatically started. The Help button can be selected for configuration information. 9. In the Rainbow NTVDM Port mapping dialog box, leave the VDM Port addresses as the defaults of 3bc, 378, 278 and 0. For each actual hardware port that that is present (as indicated by the DOS diagnostic program that you ran) for the corresponding row in the Active column, choose the same port from the Actual Port list in that row. Thus for one parallel port you would have: Active VDM Port Actual Port ------------------------------------------------ Port 1 3bc Port 1 Port 2 378 None Port 3 278 None Port 4 0 None For two parallel ports you would have: Active VDM Port Actual Port ------------------------------------------------ Port 1 3bc Port 1 Port 2 378 Port 2 Port 3 278 None Port 4 0 None When you have finished, use the Setup Actual command button. 10. In the Configure Actual Port dialog box, choose the port that the HyperChem hardware lock is attached to, and click on Edit. 11. In the Physical Port dialog box, type in the Bus Address (determined from MSD or other hardware diagnostic program) for the port that the lock is attached to. The number should be 3bc, 378 or 278, with 378 being the most common value for LPT1. In general you can accept the defaults for Bus Number, Bus Type and Address Space. Click Ok when you are finished. 12. Back in the Rainbow NTVDM Port mapping dialog box, click on the checkbox in the Active colum beside the port that the lock is on. Thus for one parallel port you would have: Active VDM Port Actual Port ------------------------------------------------ x Port 1 3bc Port 1 Port 2 378 None Port 3 278 None Port 4 0 None For two parallel ports with the lock on the first you would have: Active VDM Port Actual Port ------------------------------------------------ x Port 1 3bc Port 1 Port 2 378 Port 2 Port 3 278 None Port 4 0 None For two parallel ports with the lock on the second you would have: Active VDM Port Actual Port ------------------------------------------------ Port 1 3bc Port 1 x Port 2 378 Port 2 Port 3 278 None Port 4 0 None Click OK when you are finished. You must restart Windows NT for the change to take effect. Rainbow has informed us that the driver will work with Windows NT 3.1 or 3.5, but will not work with some Beta copies of Windows NT 3.5 (which was code named Daytona). Hope this helps, Graham ------------ Graham Hurst Until 16 Dec 1994 at: Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com Personal email to: hurst@hyper.com ________________________________________________________________________ From rescgw@mtvms2.mtech.edu Sat Dec 10 03:20:37 1994 From: rescgw@mtvms2.mtech.edu Date: Fri, 9 Dec 1994 22:47:37 -0700 To: hyperchem@hyper.com Subject: Pentium bug. I have a Pentium with the bug. I have read Intel's posting about the one in 9 billion probably of the bug occuring. The computer support group here wants to know if I want them to pursue getting a replacement chip. My feeling is that if the bug produces that low an error rate, and affects results only beyond the 5th or 6th signficant figure, it is both a waste of time and needless pummeling of Intel to insist on a replacement. So my question is, has anyone encountered computational problems with the "defective" Pentium chips. The recent posting, which I believe was from Graham, seemed to indicate that the problem was very minor. As Intel pointed out, microprocessors generally have bugs, and the 486 actually had worse bugs than this one. ________________________________________________________________________ From rbobrovs@email.tuwien.ac.at Mon Dec 12 06:04:45 1994 Date: Mon, 12 Dec 1994 09:19:52 -0100 To: hyperchem@hyper.com From: rbobrovs@email.tuwien.ac.at (Roman Bobrovsky) Subject: Re: Dongle Probs using NT >Roman, you should try changing the Bus Address to 3bc, if you are sure that= is >the address used by your LPT1 port (our computers mostly have LPT1 at= 0x378). =20 ^^^^^ That fixed it, thnx. >However although the driver loads I get inordinate delays every time it tries to >validate the licence by accessing the dongle (the program starts, but= selecting >File | New causes a 2-3 minute freeze). So if you can get the driver to work, I >would be very interested to know if you get any similar delays. I=B4m using NT 3.5 build 807, but no delays at all :-) cheers, Roman --- Bobrovsky Roman Institute of Organic Chemistry Vienna University of Technology Getreidemarkt 9 A-1060 Vienna AUSTRIA Phone: (++43-1) 588 01-4998 Internet: rbobrovs@email.tuwien.ac.at FidoNet: 2:310/30.33 ________________________________________________________________________ From rbobrovs@email.tuwien.ac.at Mon Dec 12 06:04:52 1994 Date: Mon, 12 Dec 1994 09:22:58 -0100 To: hyperchem@hyper.com From: rbobrovs@email.tuwien.ac.at (Roman Bobrovsky) Subject: Re: Dongle Probs using NT (long) >1. You will need to determine how many hardware parallel ports the >computer has and what their bus addresses are. One way to do this is >to reboot the computer to DOS (using a bootable DOS floppy if your >computer isn't configured for dual boot) and run a diagnostic program such >as MSD.EXE that comes with DOS and Windows to determine the parallel >port information. In MSD (Microsoft Diagnostics) type "L" to see >LPT Ports information, and ignore the leading zeroes and "H" suffix >(indicating hexadecimal) for the Port Address. > >IMPORTANT: Do not do this in a DOS box under Windows or Windows NT, >as the information may not be correct! It must be done from DOS alone. That=B4s the crucial point, I ran MSD from NT and actually got the= misleading bus address. Thnx, Roman --- Bobrovsky Roman Institute of Organic Chemistry Vienna University of Technology Getreidemarkt 9 A-1060 Vienna AUSTRIA Phone: (++43-1) 588 01-4998 Internet: rbobrovs@email.tuwien.ac.at FidoNet: 2:310/30.33 ________________________________________________________________________ From hurst Mon Dec 12 11:56:20 1994 To: rescgw@mtvms2.mtech.edu Subject: Re: Pentium bug. Cc: hyperchem@hyper.com Reply-To: hurst@hyper.com rescgw@mtvms2.mtech.edu writes: > I have a Pentium with the bug. I have read Intel's posting about the > one in 9 billion probably of the bug occuring. The computer support > group here wants to know if I want them to pursue getting a replacement > chip. > > My feeling is that if the bug produces that low an error rate, and affects > results only beyond the 5th or 6th signficant figure, it is both a waste > of time and needless pummeling of Intel to insist on a replacement. > > So my question is, has anyone encountered computational problems with the > "defective" Pentium chips. The recent posting, which I believe was from > Graham, seemed to indicate that the problem was very minor. As Intel > pointed out, microprocessors generally have bugs, and the 486 actually had > worse bugs than this one. Our regression testing did show very minor differences (in 12th or 13th significant figure for four calculations out of 3,000) between a 486 and a buggy Pentium 90. Although the Pentium does have improved transcendental functions compared to a 486 which can account for differences, for the four semi-empirical geometry optimizations where we saw differences no transcendental functions were used. Our working assumption is that the differences were due to at least 4 occurences of the FDIV bug. The bug is akin to performing one in 9 billion (i.e. 1 in 9 x 10^9) divisions in single precision instead of double precision. There are lots of calculations where the momentary switch to single precision could be diastrous, but I haven't seen any reports, other than ours, on any noticeable effects of the bug in computational chemistry results. You'll have to decide whether you'll want to bother with a replacement. Intel now seems quite willing to replace chips for free, though of course there are logistical delays in getting the new chips out. The main drawback in replacing is the effort of physically replacing the chip on the motherboard and sending the old one back. The Pentium has almost 300 pins but fortunately all Pentium motherboards have ZIF (zero insertion force) sockets to make the job easier. My advice is still to replace the chip but keep using the buggy one until the replacement arrives. Cheers, Graham ------------ Graham Hurst, PhD Email to: hurst@hyper.com (until 16 Dec 1994) Until 16 Dec 1994 at: Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com ________________________________________________________________________ From oberhols@mcis.messiah.edu Mon Dec 12 12:17:53 1994 From: oberhols@mcis.messiah.edu Date: Mon, 12 Dec 94 15:27:00 +0000 Subject: Network version question To: hyperchem@hyper.com It is my understanding that there is a network version of HyperChem but not of ChemPlus. If one runs the network version of HyperChem with the hardware lock on the server, can ChemPlus be loaded on and run from the local machine? ....................................................................... Item Subject: Signature Karl M. Oberholser Internet: oberhols@mcis.messiah.edu Natural Science Dept. Voice: 717-766-2511 Messiah College Fax: 717-691-6002 Grantham, PA 17027 ________________________________________________________________________ From rbobrovs@email.tuwien.ac.at Wed Dec 14 06:03:51 1994 Date: Wed, 14 Dec 1994 11:10:54 -0100 To: hyperchem@hyper.com From: rbobrovs@email.tuwien.ac.at (Roman Bobrovsky) Subject: Error using ChemPlus under NT Dear netters, If I launch ChemPlus (vers. 1.0) from HyperChem 4.0 under NT, I get an error-message from HyperChem: Query-value: Unknown variable 'atomic-number'. As this did not appear under Windows 3.11, I wonder if anyone else was faced with that or could give me a hint how to fix it. Thnx in advance, Roman --- Bobrovsky Roman Institute of Organic Chemistry Vienna University of Technology Getreidemarkt 9 A-1060 Vienna AUSTRIA Phone: (++43-1) 588 01-4998 Internet: rbobrovs@email.tuwien.ac.at FidoNet: 2:310/30.33 ________________________________________________________________________ From hberman@acsu.buffalo.edu Sat Dec 17 02:05:11 1994 Date: Sat, 17 Dec 1994 01:57:31 -0500 From: Harvey A Berman To: hyperchem@hyper.com Subject: Information please I am interested in obtaining information on Hyperchem for modelling ligand and protein interactions. Types of information I need: 1) system requirements? 2) terminals it will run on? 3) availability? 4) price? 5) Any other things you have available Harvey Alan Berman, PhD Department of Biochemical Pharmacology SUNY at Buffalo TEL: (716) 645-2859 FAX: (716) 645-3850 ________________________________________________________________________ From Matthew.Harbowy@tjlus.sprint.com Fri Dec 23 15:33:02 1994 Date: Fri, 23 Dec 1994 15:04:00 -0500 From: Matthew.Harbowy@tjlus.sprint.com To: hyperchem@hyper.com Subject: mass spectra fragment patterns In J. Mol Struc 311 (1994) 331-341, I. Mayer and A. Gomory describe a semiempirical method for partitioning the energy to determine the two center contibution to the total energy. We are very interested in this technique and ar interested in having this method available to us, however, the only molecular modelling software we have is hyperchem. If it is possible for us to implement this technique as a programming extension, or if it is available via some other route, we would like to know. It is not clear to us how one would go about adding extensions to the hyperchem program. thank you for your assistance matt harbowy chemist thomas j. lipton, inc. matthew.harbowy@tjlus.sprint.com o-cha@cnj.digex.net "I'm the bear that went over the mountain" ________________________________________________________________________ From MPANKASK@UNMCVM.UNMC.EDU Fri Dec 30 13:04:24 1994 Date: Fri, 30 Dec 94 10:04:03 CST From: "Marvin C. Pankaskie --UNMCVM(MPANKASK)" Subject: OLE Compliance To: "HYPERCHEM USER'S GROUP" FROM: Marvin C. Pankaskie --UNMCVM(MPANKASK) I am sure this question has been asked before, but is Hyperchem 3.0 or 4.0 considered to be OLE 2.0 compliant? There is an entry in both my regedit file and in the embedding section of my win.ini file that leads me to ask this question. It is possible that I added these two entries in error, and I would like to clean up this issue. Thanks for your help. Marvin C. Pankaskie (mpankask@unmcvm.unmc.edu) Department of Pharmaceutical Sciences University of Nebraska Medical Center Omaha NE 68198-6025