From hinsen@ibs.ibs.fr Fri Sep 6 03:53:28 1996 Received: from ibs.ibs.fr for hinsen@ibs.ibs.fr by www.ccl.net (8.7.5/950822.1) id DAA05545; Fri, 6 Sep 1996 03:45:15 -0400 (EDT) Received: from lmshp2.ibs.fr (lmshp2.ibs.fr [192.134.36.228]) by ibs.ibs.fr (8.6.12/8.6.12) with SMTP id JAA12783; Fri, 6 Sep 1996 09:47:07 +0200 Message-Id: <199609060747.JAA12783@ibs.ibs.fr> Received: by lmshp2.ibs.fr (1.37.109.4/16.2) id AA03509; Fri, 6 Sep 96 09:44:31 +0100 Date: Fri, 6 Sep 96 09:44:31 +0100 From: Konrad Hinsen To: Tanaka_Toshimasa@takeda.co.jp Cc: chemistry@www.ccl.net In-Reply-To: (Tanaka_Toshimasa@takeda.co.jp) Subject: Re: CCL:Software from SGI PS to Mac PS? > Does anyone know of a free or shareware that can convert a SGI > postscript file to a macintosh postscript one? There is only one PostScript (well, not quite, but let's ignore the difference between level 1 and 2 for the moment). PostScript files are ordinary text files that can be transferred between different machines without too many problems. Just make sure you transfer the file as text. If you use binary transfer, the file might look messed up on the Mac (the end-of-line markers are different), but it still ought to work when sent to a printer. Another problem is making Mac programs recognize the file as a PostScript file. This might require specifying the right file type. There are utilities for that, but I can't give any details - personally I try to stay away from Macs for exactly this reason. -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsen@ibs.ibs.fr Laboratoire de Dynamique Moleculaire | Tel.: +33-76.88.99.28 Institut de Biologie Structurale | Fax: +33-76.88.54.94 41, Ave. des Martyrs | Deutsch/Esperanto/English/ 38027 Grenoble Cedex 1, France | Nederlands/Francais ------------------------------------------------------------------------------- From teuler@idris.fr Fri Sep 6 06:53:32 1996 Received: from lumiere.idris.fr for teuler@idris.fr by www.ccl.net (8.7.5/950822.1) id GAA06022; Fri, 6 Sep 1996 06:08:36 -0400 (EDT) Received: from sailor.idris.fr (teuler@sailor.idris.fr [130.84.12.91]) by lumiere.idris.fr (8.7.5/8.7.3) with ESMTP id MAA08321; Fri, 6 Sep 1996 12:08:35 +0200 Received: (from teuler@localhost) by sailor.idris.fr (8.7.3/8.7.3) id MAA13230; Fri, 6 Sep 1996 12:08:24 +0200 From: Jean-marie Teuler Message-Id: <199609061008.MAA13230@sailor.idris.fr> X-Mailer: exmh version 1.6.2 7/18/95 To: segalemb@usp.br cc: Computational Chemistry List Subject: Re: CCL:M:M93: problems in graphics subroutines In-reply-to: (Your message of Thu, 05 Sep 96 15:42:08 B.) <322F03C0.D79@usp.br> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 06 Sep 96 12:08:24 +0100 Dear Sergio, > When I tried to install the graphics subroutines of MOPAC93R2 (bz, > density, etc.) in a IBM RS6000 I had problems in linking object files.We > compile all the routines with sucess. In linking with xlf the external > reference xwin was not found, despiste the object (xdrvr.c) that > contains this routine was ok. What is happening? I had the same problem and I solved it by editing this xdrvr.c file and by renaming the "xwin_" function to (simply) "xwin". Truly, Jean-Marie Teuler /--------------------------------------------------------\ | Jean-Marie Teuler | | | CNRS-IDRIS | Messagerie : teuler@idris.fr | | Batiment 506 | | | B.P. 167 | | | 91403 Orsay Cedex | | | France | | \--------------------------------------------------------/ From yuan@nka1.med.uc.edu Fri Sep 6 07:53:31 1996 Received: from nka1.med.uc.edu for yuan@nka1.med.uc.edu by www.ccl.net (8.7.5/950822.1) id HAA06391; Fri, 6 Sep 1996 07:47:36 -0400 (EDT) Received: from localhost by nka1.med.uc.edu via SMTP (951211.SGI.8.6.12.PATCH1042/951211.SGI.AUTO) id HAA06377; Fri, 6 Sep 1996 07:46:57 -0400 Date: Fri, 6 Sep 1996 07:46:49 -0400 (EDT) From: Jie Yuan To: Tanaka_Toshimasa@takeda.co.jp cc: Computational Chemistry List Subject: Re: CCL:Software from SGI PS to Mac PS? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Sep 1996 Tanaka_Toshimasa@takeda.co.jp wrote: > Dear All, > Does anyone know of a free or shareware that can convert a SGI > postscript file to a macintosh postscript one? > When I used "Mac GS Viewer" it created a PICT file from a SGI PS, > but I could't know how to make a postscript file for mac. The simpliest way is to print from the application (in your case, GS Viewer, but it can be any application) using LaserWriter, to a file, not the printer. So, don't click the "Print" botton too soon, if you know what I mean ... :-) I heard that the core of the LW software is provided by Adobe ... :-) Cheers! Jie -- Jie Yuan, PhD - U. Cincinnati - Pharmacology & C.B. -- == POBox 670575, Cin., OH 45267-0575 = 513-558-2352 == == Jie.Yuan@UC.edu = www.uc.edu/~yuanj = using Pine == == finger -l yuanj@ucunix.san.uc.edu for my PGP pub. key= From yuan@nka1.med.uc.edu Fri Sep 6 08:22:15 1996 Received: from nka1.med.uc.edu for yuan@nka1.med.uc.edu by www.ccl.net (8.7.5/950822.1) id HAA06366; Fri, 6 Sep 1996 07:38:11 -0400 (EDT) Received: from localhost by nka1.med.uc.edu via SMTP (951211.SGI.8.6.12.PATCH1042/951211.SGI.AUTO) for id HAA06222; Fri, 6 Sep 1996 07:38:11 -0400 Date: Fri, 6 Sep 1996 07:38:04 -0400 (EDT) From: Jie Yuan To: Computational Chemistry List Subject: Re: Software from SGI PS to Mac PS? In-Reply-To: <199609060747.JAA12783@ibs.ibs.fr> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII > Does anyone know of a free or shareware that can convert a SGI > postscript file to a macintosh postscript one? This depends on the application. Although Adobe defined PostScript in the first place, its flagship application Illustrator does not use the whole set of PS language. Therefore, to use Illustrator to edit/modify a PS file, one has to "convert" it into one (with reduced command set) that Illustrator can handle. There are shareware/freeware applications that do such convertions. For example, "epsConverter". If you use GhostView, you do not need to convert. If you simply want to print a PS file, you do not have to convert. The LW Utility, DropPS and a few other utility applications can do it very easily. As you can see, sometimes you have to convert and sometimes you don't. If you want more details, you should provide us with a more specific question. You can find these applications at any info-mac mirror sites. My favorite site is: http://hyperarchive.lcs.mit.edu/HyperArchive.html I like the convenience of reading the abstracts before downloading the applications. Cheers! Jie -- Jie Yuan, PhD - U. Cincinnati - Pharmacology & C.B. -- == POBox 670575, Cin., OH 45267-0575 = 513-558-2352 == == Jie.Yuan@UC.edu = www.uc.edu/~yuanj = using Pine == == finger -l yuanj@ucunix.san.uc.edu for my PGP pub. key= From nash@chem.wisc.edu Fri Sep 6 08:53:35 1996 Received: from audumla.students.wisc.edu for nash@chem.wisc.edu by www.ccl.net (8.7.5/950822.1) id IAA06628; Fri, 6 Sep 1996 08:35:00 -0400 (EDT) Received: from [144.92.91.209] by audumla.students.wisc.edu; id HAA55390; 8.6.9W/42; Fri, 6 Sep 1996 07:34:58 -0500 X-Sender: jrnash@students.wisc.edu Message-Id: In-Reply-To: <199609060747.JAA12783@ibs.ibs.fr> References: (Tanaka_Toshimasa@takeda.co.jp) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 6 Sep 1996 07:34:54 -0500 To: CHEMISTRY@www.ccl.net From: "John R. Nash" Subject: Re: Software from SGI PS to Mac PS? Cc: Tanaka_Toshimasa@takeda.co.JP Konrad Hinsen wrote, responding to Tanaka_Toshimasa@takeda.co.JP: >> Does anyone know of a free or shareware that can convert a SGI >> postscript file to a macintosh postscript one? > >There is only one PostScript (well, not quite, but let's ignore the >difference between level 1 and 2 for the moment). PostScript files are >ordinary text files that can be transferred between different machines >without too many problems. This is correct. A relevant point is that PostScript is a printer graphic description language, not a screen graphic language. Any viewer, such as GhostScript, is merely taking the place of a printer and sending the output to the screen. Since screens are almost always of lower resolution than printers, the printed PostScript output usually looks better than what's on the screen, provided a PostScript printer is used. CambridgeSoft's ChemDraw provides a classic example of this -- the printout always looks a lot crisper than what's on the screen. Most Mac laser printers are PostScript printers, with a few of the more inexpensive ones being "QuickDraw" printers (which would need an interpreter to correctly print PS). >Another problem is making Mac programs recognize the file as a >PostScript file. This might require specifying the right file type. >There are utilities for that, but I can't give any details - >personally I try to stay away from Macs for exactly this reason. This really isn't a problem at all, as long as the PS file is correctly transferred as text. Any PostScript viewer will be automatically looking for a text file, so this is a non-issue when opening from within the application. If one uses the freeware Internet Config, a file with extension .ps will be automatically given a PostScript icon and can be configured to launch your favorite viewer when double-clicked on. Printing a PostScript file is as simple as dropping it onto an application such as Drop*PS. (Internet Config and Drop*PS are available at http://hyperarchive.lcs.mit.edu/HyperArchive.html and many other archives.) It all works rather smoothly -- personally I try to use Macs as much as possible because of this ease-of-use. The problem that the original poster may well be having is the need for an _encapsulated_ PostScript file. Whereas PostScript is a defined cross-platform language made commercially by Adobe, and as such is "one" format, EPS is an implementation that differs significantly among platforms and even applications on the same platform. The utility pstoeps (available for unix and ported to many platforms, including PC and Mac, I think) will generate some of the "standard" EPS files from PS files. Specific programs may need to have the EPS file "tweaked" in some way. I know that Adobe Illustrator comes with some utilities to help that program, and I've seen many other free/shareware utilities on the net to manipulate these files. The MIT site is also a good place to find these. The mess of standards that is EPS is unfortunate, but the damage has been done. Generally, the best course of action is to get output in raw PostScript and convert to the correct EPS format yourself, rather than mess with changing other EPS files. Good luck! -john nash p.s. I'll try to get my summary of the results of my "molecular volumes" query posted today... the news is good! -==-John R. Nash-==-Dept of Chemistry, UW-Madison-==-nash@chem.wisc.edu-==- From HORNE@CMOD.PFC.MIT.EDU Fri Sep 6 10:53:42 1996 Received: from CMODA.PFC.MIT.EDU for HORNE@CMOD.PFC.MIT.EDU by www.ccl.net (8.7.5/950822.1) id JAA06990; Fri, 6 Sep 1996 09:57:52 -0400 (EDT) Date: Fri, 6 Sep 1996 9:57:55 -0400 (EDT) From: HORNE@CMOD.PFC.MIT.EDU To: CHEMISTRY@www.ccl.net Message-Id: <960906095755.28800082@CMOD.PFC.MIT.EDU> Subject: Calculating diatomic vibrational spectrum I'm a newcomer to this field and find myself with the following problem. We have a combined experimental/theoretical program aimed at understanding processes occuring in a 75 kW microwave chemical vapor deposition system. I want to make a better connection between the 2-D (radially symmetric) simulation we have developed, and the experimental optical emission diagnostics. Therefore I'm looking for software to generate simulated spectra given local plasma and reactant species densities and temperature. We're particularly interested in the C2, CN, and CH ro-vibrational spectra. The archives of this group contain some old references to this problem but so far I've had no luck actually finding codes. If anyone can help I'd appreciate it. (In print, of course -- if this work gets that far.) Thanks Steve Horne From JERICKSON@elvax2.dowelanco.com Fri Sep 6 11:22:07 1996 Received: from elinet1.dowelanco.com for JERICKSON@elvax2.dowelanco.com by www.ccl.net (8.7.5/950822.1) id KAA07136; Fri, 6 Sep 1996 10:22:01 -0400 (EDT) From: Received: by elinet1.dowelanco.com id AA14213 (InterLock SMTP Gateway 3.0 for chemistry@www.ccl.net); Fri, 6 Sep 1996 09:22:00 -0500 Message-Id: <199609061422.AA14213@elinet1.dowelanco.com> Date: Fri, 6 Sep 1996 9:20:44 -0500 (EST) To: chemistry@www.ccl.net Subject: Re: Mopac-conjugated system > Date: Thu, 5 Sep 1996 16:32:41 -0700 > From: bear@ellington.pharm.arizona.edu (Soaring Bear) > To: chemistry@www.ccl.net > Subject: CCL:M:Mopac-conjugated system > > Hello: > My question concerns using MOPAC to calculate charge > distribution in a conjugated system, like this: > > O + > || NR2 > // \ /\ /\ /H > // \/ \ / \/ > | || N > | || > \\ // > \\ / > > I'm finding that putting various substituents around > the ring at various positions does NOT affect the charge on the > benzylic carbonyl. The carbon & oxygen keep coming up within > a couple hundredths of a charge even though there ought to be > some conjugation/resonance with the ring and substituents. > The MOPAC parameters I'm using (left over from previous > work) are: AM1 Pulay ESP XYZ Charge=1 GEO-OK > > My question is: Is there some other set of commands > that would work better with this system? > > thankyou, > Soaring Bear > > _____ ____ bear@ellington.pharm.arizona.edu > * ___) * _ \ http://ellington.pharm.arizona.edu/~bear > : (___ : |_) : Cyber-Chemist: cancer drug design topo-O O-topo > \___ \ | _ < Molecular & Nutritional Biochemist 5'. : : .*** > ___) : | |_) : Herbs, Nutrition, Natural Dentistry |'*. .*'| | > (_____/oaring |____/ear, UA New Pharmacy 404, Tucson 85721 | | *.,* | | | Soaring Bear, I repeated your MOPAC calculations (with MOPAC93) for your system using your keywords (I had to use MMOK because of the amide system) with para-substituted H, NH2 and NO2 and looked at the charge change on the carbonyl group. Here's my results: ESP charge carbonyl C carbonyl O ---------- ---------- H 0.1889 -0.2639 NH2 0.3415 -0.3073 NO2 0.2497 -0.2676 In each case the amide remained in conjugation with the phenyl ring (which is where I started it). Maybe you started the amide out of conjugation, and therefore did not see the large resonance effect. Hope this helps, Jon Erickson DowElanco CAMD jerickson@dowelanco.com From borkent@caos.kun.nl Fri Sep 6 11:53:33 1996 Received: from cammsg1.caos.kun.nl for borkent@caos.kun.nl by www.ccl.net (8.7.5/950822.1) id LAA07630; Fri, 6 Sep 1996 11:49:11 -0400 (EDT) Received: from cammsg5 (cammsg5.caos.kun.nl [131.174.88.5]) by cammsg1.caos.kun.nl (8.7.1/8.7.1) with SMTP id RAA24666 for ; Fri, 6 Sep 1996 17:49:10 +0200 (MDT) From: Hens Borkent Message-Id: <199609061549.RAA24666@cammsg1.caos.kun.nl> Subject: MMCC Results on-line To: chemistry@www.ccl.net Date: Fri, 6 Sep 1996 17:49:09 +0200 (MDT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Dear CCL readers, Talking of electronic journals: at the CAOS/CAMM Center we maintain a copy of MMCC Results in web format: see http://www.caos.kun.nl/mmcc/mmcc.html Subscription with the publisher, Bruce R. Gelin, is necessary: see http://world.std.com/~mmcc One advantage over the printed version: a text search in all issues from February 1994 till present. If there is sufficient demand, we could expand this service by incorporating more hyperlinks in the abstracts. Sincerely, -- ***** J.H. (Hens) Borkent, CAOS/CAMM Center, *CAOS * P.O. Box 9010, 6500 GL Nijmegen, The Netherlands * / * Tel 0031 24 36 52137 Fax 0031 24 36 52977 * CAMM* e-mail: borkent@caos.kun.nl ***** http://www.caos.kun.nl/staff/borkent.html From elewars@alchemy.chem.utoronto.ca Fri Sep 6 12:53:42 1996 Received: from alchemy.chem.utoronto.ca for elewars@alchemy.chem.utoronto.ca by www.ccl.net (8.7.5/950822.1) id MAA07687; Fri, 6 Sep 1996 12:03:21 -0400 (EDT) Received: (from elewars@localhost) by alchemy.chem.utoronto.ca (8.7.4/8.7.3) id MAA29042 for chemistry@www.ccl.net; Fri, 6 Sep 1996 12:03:12 -0400 (EDT) Date: Fri, 6 Sep 1996 12:03:12 -0400 (EDT) From: "E. Lewars" Message-Id: <199609061603.MAA29042@alchemy.chem.utoronto.ca> To: chemistry@www.ccl.net Subject: CHARGES AND SUBSTITUENTS 1996 Sept 6 Hello, Soaring Bear asked about MOPAC calc's on compounds like [bear@ellington.pharm.arizona.edu] O || + C NHRR / \ / \ / \\ / / \ / N | | I | | \ / \ / benzene ring He wondered why the charge on the C=O carbon was little affected by substituents on the benzene ring. I did a series of AM1 calcs, using Spartan, on O H \\ / C | / \ / \ | | | | benzene ring \ / \ / | X The electrostatic-potential charges (generally preferred to the Mulliken or the natural charges) on the carbonyl C are H OH NO2 F 0.53 0.55 0.48 0.53 So there is no obvious correlation of charge with the conventional view of the electronic proclivity of X. I have three comments: 1) The effect of X on reactivity may not be a ground-state phenomenon, but rather may be manifested in the transition state. I suspect there is something on this in the literature. 2) I think there is no good correlation between the conventional view of the electronic character of X and the C-13 chemical shift (I'd have to check this), which supports (1). 3) There is a paper on compounds similar to I above in G. B. McGaughey, E. L. Stewart and J. P. Bowen, J. Comp. Chem., 17, 1395-1405 (1996), which may be of a little use although it doesn't treat ring substitution. Errol Lewars From shenderm@U.Arizona.EDU Fri Sep 6 13:53:34 1996 Received: from bonaire.ccit.arizona.edu for shenderm@U.Arizona.EDU by www.ccl.net (8.7.5/950822.1) id NAA08167; Fri, 6 Sep 1996 13:11:17 -0400 (EDT) Received: from localhost (shenderm@localhost) by bonaire.ccit.arizona.edu (8.7.5/8.7.3) with SMTP id KAA41038 for ; Fri, 6 Sep 1996 10:11:58 -0700 Date: Fri, 6 Sep 1996 10:11:57 -0700 (MST) From: Mark D Shenderovich X-Sender: shenderm@bonaire.ccit.arizona.edu To: chemistry@www.ccl.net Subject: Re: CCL:Software from SGI PS to Mac PS? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Sep 1996 Tanaka_Toshimasa@takeda.co.jp wrote: > Dear All, > Does anyone know of a free or shareware that can convert a SGI > postscript file to a macintosh postscript one? > When I used "Mac GS Viewer" it created a PICT file from a SGI PS, > but I could't know how to make a postscript file for mac. > I appreciate your any advices. > best wishes Dear Netters: I have a closely related question. Does anyone know a program that can EDIT PS files. The minimum requirement is the ability to blow up / reduce images along each axis. Preferable platforms are SGI or IBM RS/6000. Yhank you in advance. Mark Shenderovich. From tj@eecs.uic.edu Fri Sep 6 14:53:34 1996 Received: from mail.eecs.uic.edu for tj@eecs.uic.edu by www.ccl.net (8.7.5/950822.1) id OAA08551; Fri, 6 Sep 1996 14:47:36 -0400 (EDT) Received: from bert.eecs.uic.edu (tj@bert.eecs.uic.edu [128.248.166.23]) by mail.eecs.uic.edu (8.7.5/8.7.5) with ESMTP id NAA04584; Fri, 6 Sep 1996 13:48:12 -0500 (CDT) From: tj ODonnell Received: (tj@localhost) by bert.eecs.uic.edu (8.6.10/8.6.10) id NAA22984; Fri, 6 Sep 1996 13:48:12 -0500 Message-Id: <199609061848.NAA22984@bert.eecs.uic.edu> Subject: Re: CCL:Software from SGI PS to Mac PS? To: shenderm@U.Arizona.EDU (Mark D Shenderovich) Date: Fri, 6 Sep 1996 13:48:12 -0500 (CDT) Cc: chemistry@www.ccl.net (CCL) In-Reply-To: from "Mark D Shenderovich" at Sep 6, 96 10:11:57 am X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > > Dear Netters: > > I have a closely related question. Does anyone know a program that can > EDIT PS files. The minimum requirement is the ability to blow up / reduce > images along each axis. Preferable platforms are SGI or IBM RS/6000. > Yhank you in advance. > > Mark Shenderovich. > > > Mark- On the SGI there is a program 'psrender' which I use to make SG image files (.rgb files). You can specify the width and height, which may get you what you want. Otherwise, you can edit the rgb file, say in showcase, or Photoshop (if you have it), or convert it to myriad other formats for further editing. Showcase will even convert it back to a PS file. I believe psrender is available from ftp://sgi.sgi.com/graphics/bin/ It looks like it might be called fromps or fromeps these days. Hope this helps, TJ -- *---------------------------------------------------------------------* * If you need a quote: http://www.eecs.uic.edu/~tj/quotes.html * *---------------------------------------------------------------------* | Dr. TJ O'Donnell > tj@eecs.uic.edu < http://www.eecs.uic.edu/~tj/ | *---------------------------------------------------------------------* From nick@chem.ucla.edu Fri Sep 6 15:11:46 1996 Received: from xenon.chem.ucla.edu for nick@chem.ucla.edu by www.ccl.net (8.7.5/950822.1) id OAA08484; Fri, 6 Sep 1996 14:23:55 -0400 (EDT) Received: from 128.97.34.195 (corvus.chem.ucla.edu [128.97.34.195]) by xenon.chem.ucla.edu (8.6.9/8.6.9) with SMTP id LAA00678 for ; Fri, 6 Sep 1996 11:28:34 -0700 Message-Id: <199609061828.LAA00678@xenon.chem.ucla.edu> Date: 6 Sep 1996 11:30:32 -0800 From: " nick.c " Subject: Re: CCL:Software from SGI PS to Mac PS? To: "chemistry@www.ccl.net" X-Mailer: Mail*Link PT/Internet 1.5.1 > Does anyone know of a free or shareware that can convert a SGI > postscript file to a macintosh postscript one? > When I used "Mac GS Viewer" it created a PICT file from a SGI PS, > but I could't know how to make a postscript file for mac. > I appreciate your any advices. Assuming you want to embed PS files created on an SGI into Macintosh documents such as ChemDraw or Miscrosoft Word, the first step is to convert the PS file (as folks have already mentioned there is only one kind of PS) to an EPS (encapsulated postcript) file. These EPS files can be easily imported into drawing and word processing programs. There are UNIX solutions for converting PS to EPS files, but since you specifically mentioned the Mac platform, let me suggest a great Mac utility called PS2EPS+ by Peter Lerup. Basically, all you have to do is DL the asci PS file to your Mac, then drop in on PS2EPS+ and save the rendered file as either PICT, TIFF, (or what you want) EPS. Then drop the EPS file on ChemDraw or Word and it will open an untitled document with the EPS object as the only thing in the file. Copy that object, and paste it wherever you like. Note, the image you see on your screen is a bitmapped aproximation of the object--when it prints it will look a lot nicer (smoother, sharper). PS2EPS+ requires a "bounding box" comment in the PS header, not all PS output has this comment. If yours doesn't open the PS file with a text editor (like the essential BBEdit :-), then take a look at the coordinates in the PS file and add a bounding box comment that exceeds the coords in the PS file in X, -X, Y, -Y directions. For an example of bounding box comment, look at the included example files in PS2EPS+. BTW, you can DL PS2EPS+ from your local info-mac mirror, I think I found it in the "Print" directory. Luck, -------------------= Nicholas C. DeMello, Ph.D. =-------------------- Internet: nick@chem.ucla.edu _/ _/ _/ _/_/_/ _/ _/ AOL: codeweaver _/_/ _/ _/ _/ _/ _/_/_/ CIS: 71232,766 _/ _/_/ _/ _/ _/ _/ http://www.chem.ucla.edu/~nick/ _/ _/ _/_/_/ _/ _/ From john@ibg.uu.se Fri Sep 6 15:53:42 1996 Received: from nucleus.ibg.uu.se for john@ibg.uu.se by www.ccl.net (8.7.5/950822.1) id PAA08768; Fri, 6 Sep 1996 15:35:48 -0400 (EDT) Received: from localhost by nucleus.ibg.uu.se via SMTP (940816.SGI.8.6.9/940406.SGI.AUTO) for id VAA18996; Fri, 6 Sep 1996 21:41:33 +0200 Date: Fri, 6 Sep 1996 21:41:30 +0200 (MDT) From: John Marelius To: chemistry@www.ccl.net Subject: Summary: new square root improved molecular dynamics Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII This posting summarises the results of the discussion on the usenet newsgroup comp.sys.intel on square-root calculations in molecular dynamics (MD) applications. Using a novel root algorithm and general optimisations of the MD code, performance has increased by up to 55%, depending on the simulated system. Background: The computational cost in molecular dynamics simulations (with water represented as individual molecules) is dominated by calculations of non-bonded (long-range) interaction energies and forces between pairs of atoms. In particular, the cost scales with the square of the number of atoms. The energy function has the following form: A(i,j) B(i,j) q(i)*q(j) E(i,j) = ------ - ------ + --------- r^12 r^6 r where r is the distance between atoms i and j, A and B are constants depending on the kinds of atoms and the q:s are the (partial) electric charges of the atoms. If it wasn't for Coulomb, who introduced an odd power of r (r^-1) into his law, this could be evaluatued without a square root calculation. (In cartesian space r^2 is easily obtained from the co-ordinates.) The core of the problem is that square root calculations and also divisions take up to two orders of magnitude longer than multiplications and additions to execute on most computers. One could of course consider working in another co-ordinate system but, unfortunately, other aspects of the potential energy fuctions (such as bond angles) then become more demanding. Attacking the problem: Knowing the above, plus the fact that the MD program I use spends nearly 50% of its time with calculations of sqrt(1/r^2), I decided to try to improve the overall speed by finding a faster way to calculate square roots. The initial discussion on comp.sys.intel and comp.lang.asm.x86 taught me that some types of CPUs have floating-point unit (FPU) instructions for square root, while others have not (a function library call is necessary). The speed difference between the two ways is not as big as I first expected, and most modern processors can do the square root in hardware. Approaches by me and others involving integer-math, look-up tables, Taylor polynominals or Newton-Raphson iterations ended up being either slower or less accurate than the standard function. One person who became interested in the problem was Terje Mathisen, he started thinking and coding and has overwhelmed me with a stream of ever-improving codes for fast and accurate functions to calculate sqrt(1/x). Please refer to his posting (coming soon), describing the details of the algorithm! Most modern processors have more than one execution unit for floating-point calculations (parallellism) and each of these can start working on new instructions before previous instructions are finished (pipelining). This opens for another type of optimisation, namely rewriting the progam so that it can make maximum benefit from this. Results: I incorporated the new sqrt(1/x) algorithm in the MD program. I also modified the critical sub-routines to do two or three independent calculations in parallell to avoid 'pipeline stalls'. To benchmark the new program it was tested in very short simulations of two typical systems: 1) A system dominated by protein (ca 2000 atoms + ca 1000 water atoms (=ca 333 water molecules)) 100 dynamics steps were done. 2) A system dominated by water (one 12-atom solute molecule + 1200 water atoms), where the non-bonded interactions are even more dominant in terms of CPU time. 250 dynamics steps were done. The performace tests of the new program are indeed very encouraging! The tests were run on a 400 Mhz DEC AlphaStation 500. system 1 system 2 time rel. speed time relative speed (s) (%) (s) (%) original program 37.0 100 37.4 100 new program 30.2 123 24.1 155 Initiation overheads have been subtracted. Conclusion: My week-long simulation jobs will now complete in 4.5 to 5.7 days instead! I am very grateful to Terje Mathisen for the new square root algorithm. I also acknowledge the many people all over the internet who contributed valuable information. John Marelius graduate student in molecular biotechnology +-----------------------------------------------------------------------------+ | John Marelius | | Dept. of Molecular Biology, Uppsala University | | E-mail: john@ibg.uu.se | | www: http://www.ibg.uu.se/~john | | fax: +46 18 536971 | | address: Box 590, S-751 24 Uppsala, Sweden | +-----------------------------------------------------------------------------+ From robert@pauli.utmb.edu Fri Sep 6 16:53:36 1996 Received: from nmr.utmb.edu for robert@pauli.utmb.edu by www.ccl.net (8.7.5/950822.1) id QAA08944; Fri, 6 Sep 1996 16:15:45 -0400 (EDT) Received: from pauli.utmb.edu by nmr.utmb.edu via SMTP (931110.SGI/930416.SGI.AUTO) for chemistry@www.ccl.net id AA25055; Fri, 6 Sep 96 15:18:40 -0600 Received: (from robert@localhost) by pauli.utmb.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) id PAA03401; Fri, 6 Sep 1996 15:15:12 -0500 From: "Robert Fraczkiewicz" Message-Id: <9609061515.ZM3399@pauli.utmb.edu> Date: Fri, 6 Sep 1996 15:15:12 -0500 In-Reply-To: Mark D Shenderovich "CCL:Software from SGI PS to Mac PS?" (Sep 6, 10:11am) References: X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Mark D Shenderovich , chemistry@www.ccl.net Subject: Re: CCL:Software from SGI PS to Mac PS? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Sep 6, 10:11am, Mark D Shenderovich wrote: > Subject: CCL:Software from SGI PS to Mac PS? > On Fri, 6 Sep 1996 Tanaka_Toshimasa@takeda.co.jp wrote: > Dear Netters: > > I have a closely related question. Does anyone know a program that can > EDIT PS files. The minimum requirement is the ability to blow up / reduce > images along each axis. Preferable platforms are SGI or IBM RS/6000. > Yhank you in advance. > > Mark Shenderovich. Yes. It's called vi, or emacs, or any other screen editor on any platform ;-) PS file is in ASCII format that you can directly edit. For shrinking/expanding a picture add the following command a f t e r prolog and setup blocks, but at the beginning of actual PostScript code: scale where and are numerical scaling factors in x and y direction, respectively. To move entire picture use " translate" in analogous fashion remembering that and compose translation vector in units called "dots". In 1:1 scale there are 72 dots per inch. Using these two commands and few other (rotate, moveto, etc.) I was able to produce quite complicated picture arrangements using only text editor and Ghostview to look at the final result. Hope that helps, Robert Fraczkiewicz University of Texas Medical Branch, Galveston From ngo@interval.com Fri Sep 6 17:53:35 1996 Received: from fred.interval.com for ngo@interval.com by www.ccl.net (8.7.5/950822.1) id RAA09260; Fri, 6 Sep 1996 17:47:16 -0400 (EDT) Received: from interval.interval.com (interval.interval.com [199.170.107.10]) by fred.interval.com with ESMTP id OAA28886 for ; Fri, 6 Sep 1996 14:47:11 -0700 Received: by interval.interval.com id OAA23815; Fri, 6 Sep 1996 14:47:11 -0700 Message-Id: <3.0b15.32.19960906144644.0070aed0@interval.com> X-URL: http://www.interval.com/~ngo/ X-Telephone: +1 415 842 6037 X-Sender: ngo@interval.com X-Mailer: Windows Eudora Pro Version 3.0b15 (32) Date: Fri, 06 Sep 1996 14:47:01 -0700 To: chemistry@www.ccl.net From: Tom Ngo Subject: Editing PS files Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" >On Sep 6, 10:11am, Mark D Shenderovich wrote: >> I have a closely related question. Does anyone know a program that can >> EDIT PS files. The minimum requirement is the ability to blow up / reduce >> images along each axis. Preferable platforms are SGI or IBM RS/6000. >> Thank you in advance. >> >> Mark Shenderovich. Robert Fraczkiewicz replied: >Yes. It's called vi, or emacs, or any other screen editor on any platform ;-) Here's another option. As I was surprised to find out recently, there is a third-party XTRA for Macromind FreeHand that supposedly allows one to import a PS file and edit it. Because PS is a complete language and not just a graphics format, the PS files written out by this XTRA undoubtedly are watered down compared to the imported files (if you've got code in your input file that, say, computes prime numbers, it probably won't get written to the exported file). But for the graphical elements---lines, curves, text---it does stand a chance of doing the right thing. I would imagine that a lot of the internal structure in the original file is lost; for example, if the input file contains multiple copies of an image created by transforming a single source copy, the output file will probably contain each copy explicitly. You'd have to buy Macromind FreeHand and the TechPool XTRA; check out the URL below. http://www.macromedia.com/software/xtras/freehand/techpool.html Hope this helps, --Tom Ngo =========================================================================== Tom Ngo Voice: 415/842-6037 Interval Research Corporation Fax: 415/354-0872 1801 Page Mill Rd, Bldg C Lat. 37.409, Long. -122.153 Palo Alto, CA 94304-1216 http://www.interval.com/~ngo "Never confuse motion with action." --Ben Franklin From fgonzale@lauca.usach.cl Fri Sep 6 18:53:34 1996 Received: from ralun.usach.cl for fgonzale@lauca.usach.cl by www.ccl.net (8.7.5/950822.1) id SAA09346; Fri, 6 Sep 1996 18:14:37 -0400 (EDT) Received: from lauca.usach.cl (lauca.usach.cl [158.170.64.28]) by ralun.usach.cl (8.6.11/8.6.9) with ESMTP id SAA04712 for ; Fri, 6 Sep 1996 18:18:12 -0500 Received: (from fgonzale@localhost) by lauca.usach.cl (8.6.12/8.6.12) id SAA19029; Fri, 6 Sep 1996 18:19:44 -0400 From: Fdo Danilo Gonzalez Nilo Message-Id: <199609062219.SAA19029@lauca.usach.cl> Subject: H and S with Molecular Dynamics To: chemistry@www.ccl.net Date: Fri, 6 Sep 96 18:19:43 AST Cc: fgonzale@lauca.usach.cl Dear Friends: I'm looking for any information about Entropy, Enthalpie and free Energie of solvation in caculations with Molecular Dynamics in a solvent box (PBC). I will be very grateful for any help!! thanks a lot..!!! Fernando Danilo Gonzalez N. Universidad de Santiago de Chile Facultad de Quimica y Biologia Casilla 307, Santiago-2, Chile fono: 681 2575 E-mail : fgonzale@lauca.usach.cl fax : (562) 681 2108 danilo@quantum.usach.cl URL : http://quantum.usach.cl/~danilo *************************************************************************x