Info-IBMPC@WSMR-SIMTEL20.ARMY.MIL ("Info-IBMPC Digest") (04/09/91)
Info-IBMPC Digest Fri, 5 Apr 91 Volume 91 : Issue 85 Today's Editor: Gregory Hicks - Rota Spain <GHICKS@WSMR-Simtel20.Army.Mil> Today's Topics: comments on postscript printing cr/lf problems DD/HD/holes Disabling Control Break Memory Board Settings (V91 #73) Listing of interrupts Re Listing of interrupts Wuarchive Mirrors Resurrected Re: Missing 1k on PC Today's Queries: Achieve Multi I/O Plus Mistery DIP Switches EMS in VGA ram? hardware compatibility Request for info on VAX/PC/MAC Linkups native vs translation modes Search for information on LAN's Symbolic Integration WANTED: Simple Zmodem Term Prog. Toggeling To/From Turbo Mode New Upload: TCU_32A.ZIP - Turbo C Flexible menus, windows, & Forms Entry Send Replies or notes for publication to: <INFO-IBMPC@WSMR-SIMTEL20.ARMY.MIL> Send requests of an administrative nature (addition to, deletion from the distribution list, et al) to: <INFO-IBMPC-REQUEST@WSMR-SIMTEL20.ARMY.MIL> Archives of past issues of the Info-IBMPC Digest are available by FTP only from WSMR-SIMTEL20.ARMY.MIL in directory PD2:<ARCHIVES.IBMPC>. ---------------------------------------------------------------------- Date: Wed, 3 Apr 91 01:49:37 GMT From: freelsj%oak.span@Sdsc.Edu (James D. Freels) Subject: comments on postscript printing A note on tradeoffs of buying a PS printer versus using QMS-Ultrascript to translate PS output to an HP-LJ II. I have an application that generates large postscript files of lots of line draws (sort of like Tektronix plot files). I was interested in seeing whether I could save any turn-around time by using QMS-USPC to an HP-LJ II instead of the QMS-810 printer I have been using. The example postscript file is about 1.8 Mb. The computer is a 386-20 Mhz with 10 Mb memory and coprocesser. I print in serial mode to a Baytech data exchanger at 38.4 Kbaud and the Baytech in turn prints in parallel to the printers (whichever I choose). The bottle neck is at the printer not the serial communications so don't give me that argument. The straight print to the QMS-810 PS printer takes about 31 minutes for this file (picture) from the time I issue the print command until I get the page of output. The PCL file created by QMS-USPC is about 578Kb reduced from the original 1.8Mb PS file. Then I print this PCL file to the HP-LJII printer. The total turn around time for this is about 19 minutes! Since QMS-USPC is considerably cheaper than a PS printer, I argue that for this type of printing, why get a PS printer? The drawback is that QMS-USPC works best with 4Mb of your extended memory. You can tradeoff this memory for disk swap space but it will not run as fast. I am convinced that this is the way to go. I am now seriously thinking about getting an HP-DeskJet for home to do what little graphics printing I will do. Yeah, I know that a 300 dpi image on a desk-jet will take a very long time. But they are relatively inexpensive. Another comment is that I cannot detect any differences in the output quality. Furthermore, since a Deskjet is ink based, the quality there might actually be better than an HPLJII. Any comments? ------------------------------ Date: Tue, 2 Apr 91 15:23:24 CST From: "Norman R. Frech CPLS" <frech@mwraaa.army.mil> Subject: cr/lf problems >>Then I took an hour to discover that unix files use ^J as an line >>separator, while ms-dos insists on a ^M. The thing is, I knew this, >>but didn't realize the very misterious compiler errors were due to this >>effect. >>So, beware of the ^J when copying files from an unix system folks! >In addition to the CRLF programs available to correct this, I have >found what is a really good solution for me. I load the file with the >^Js into Qedit, make a small change (like deleting a single character) >and save it again. Voila! All newlines are now ^Ms. >The only problem is that I can't get Qedit to edit files bigger than >the available amount of memory. Alas. I typically use perl to fix this problem with something like the following: while (<>) { chop; print "$_\n"; } Your files can be as large as you want. Any time I have to convert a file from one format to another perl usually will do the best job. Norm Frech <frech@mwraaa.army.mil> ------------------------------ Date: Tue, 02 Apr 91 13:10:49 EST From: "Bruce H. McIntosh" <DHBHM@nervm.nerdc.ufl.edu> Subject: DD/HD/holes The reason that non-PS/2 machines cannot read a DD floppy that was formatted HD on a PS/2 is that they don't have the HD hole. Here's how it goes: The PS/2 lets you get away with formatting to HD because it hasn't got the hole sensor; thus, it can't tell that you're using a DD floppy. When you take a floppy formatted in this manner and attempt to use it on a machine that HAS the hole sensor (which is just about everybody else), that machine cannot tell that it's dealing with a HD-formatted disk. Its hole sensor is telling it, "This is a DD diskette." So, it will attempt to read what it thinks is a DD-formatted diskette, encounter sectors that are (it thinks) all messed up, and issue the dreaded "General error reading drive A" message. So, if you want to get away with using those HD-formatted DD diskettes on a non-PS/2 machine, you're most likely going to have to punch holes in them. ------------------------------ Date: Tue, 02 Apr 91 09:20:00 CST From: Mark Parr <JPARR1@UA1VM.ua.edu> Subject: Disabling Control Break Someone asked a question about disabling the Contrl-C/Break: I have a couple device drivers that will do that for you -- both work fine. One is called NOBRK.SYS and the other is BRK.DRV With the BRK.DRV, you are able to turn Control-C/Break checking on or off as needed (default is off). Send me your address and I will send you UUENCODE .ZIP files to you if you are interested. |-- JPARR1@UA1VM.BITNET -- JPARR@MIBSRV.MIB.ENG.UA.EDU Mark Parr --|-- University of Alabama in Tuscaloosa [How about sending Keith Petersen a note and asking to upload for the rest of us as well? gph] ------------------------------ Date: Tue, 2 Apr 91 16:44:00 CST From: david@wubios.wustl.edu (David J. Camp) Subject: Memory Board Settings (V91 #73) In Reply to this Note From: <Info-IBMPC Digest> [text deleted] >Date: Thu, 21 Mar 91 18:27 EST >From: FCUBBAGE%TMPLCIS.BITNET@uga.cc.uga.edu >I have a 16 bit IBM PC memory board called a RAM Parity Card with no [text deleted] Greg, Notice that there are two Subject: lines in the header lines above. This confuses my 'digest' program (included below), so that appropriate identifier. I am trying to keep the digest program simple, especially because it is possible to make it an incredible monstrosity to handle all of the weird formats. I was just hoping that it was possible for you to modify the generation of the digest so that this message took another form. For example, you could contrive an RFC822 header for that purpose, such as "Subject-Group:". This should be ignored by mail systems, so it should not cause any problems. I would appreciate hearing your response. -David- # david@wubios.wustl.edu ^ Mr. David J. Camp # # david%wubios@wugate.wustl.edu < * > +1 314 382 0584 # # ...!uunet!wugate!wubios!david v "God loves material things." # # abs (investment#1 - investment#2) << abs (anyinvestment - anydebt) # ----- cut here ----- #! /usr/bin/perl # # This is 'digest' a program to run elm on a digest as a folder. # Copyright (C) 1990 David J. Camp # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 1, or (at your option) # any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. # # david@wubios.wustl.edu ^ Mr. David J. Camp # david%wubios@wugate.wustl.edu < * > +1 314 382 0584 # ...!uunet!wugate!wubios!david v "God loves material things." # # from the elm menu, type |digest<return> # $| = 1; open (FOLDER, ">/tmp/dig$$"); while (<>) { if (/(^\([0-9]*\) |^--)----(\n|-----------------------(\n|-(\n|-------------------------(\n|---------------\n))))/) { print FOLDER "From dummy Wed Feb 29 12:12:12 1990\n"; do { $_ = <>; } until (eof() || /[\041-\177]/); } print FOLDER $_; } close (FOLDER); exec ("elm -f /tmp/dig$$ <&2 ; /bin/rm -f /tmp/dig$$"); ----- cut here ----- ------------------------------ Date: 2 Apr 91 8:39 From: Martin.Timann%cnve.rrze.uni-erlangen.dbp.de@RELAY.CS.NET Subject: Listing of interrupts Dr.Guenther W.Himmelmann asks >In 1989 there was an excellent listing of interrupts - who knows, where >to find it ( or an update ) ? We had it printed, but the printout was >lost when we had to change to a new location... Thank You! >gwh I think you mean Ralf Brown's Interrupt List. You can find it on SIMTEL20 in <MSDOS.INFO>INTER590.ZIP BTW, - excuse me I'm a novice reader of the digest and curious - who's that wizard that is hiding behind the abbreviation "gph" ? His answers are enclosed in square brackets Martin Timann Timann@cnve.rrze.uni-erlangen.dbp.de Timann@cnve.rrze.uni-erlangen.de ------------------------------ Date: Wed, 3 Apr 91 08:34:21 MST From: Gregory Hicks <GHICKS@WSMR-SIMTEL20.ARMY.MIL> Subject: Re Listing of interrupts Martin: >...Who's that wizard that is hiding behind the abbreviation "gph"? > His answers are enclosed in square brackets [Not sure what you mean by 'wizard' since I don't consider myself in that category... I'm the editor for the Digest. (Take a look at the "masthead".) By putting the answers in the individual message, I can save 'message space' for other topics. The digestify program has a limit (built in by me) of 20 messages or 30K bytes of text (whichever comes first) so that most mailers will accept the mail... The initials 'gph' stand for Gregory Paul Hicks (me). Here's a short blurb on me... I'm a Lieutenant Commander on Active Duty with the US Navy. I'm currently stationed at the Naval Base in Rota, Spain where my real job is to manage the port of Rota for US and Spanish Navy (Port Captain if you will). During the years from 85-89, I was stationed at the Joint US Military Advisory Group-Korea (Seoul and Chinhae) as a R&D Liaison Officer. Following my stint in Seoul, I was stationed at the Fleet Activity, Chinhae Korea where I took care of US Navy (and US Naval, there IS a difference) ships visiting Korea. Professionally, my college degree is in Computer Science from the University of Utah. However, I started programming in '64 just out of High School on an IBM 1610. I worked for a while with a computer service bureau in Los Angeles, at the Math Department at UCLA, moved to the Jet Propulsion Laboratory working for Bendix Field Engineering Corporation in Pasadena in '66 at the Space Flight Operations Facility (or SFOF) ans was there during the Lunar Orbiter and Surveyor space shots. I left JPL in October '67 to join the Navy in '68 (I almost was drafted). I was trained by the Navy as a Data Systems Technician. While in school at Vallejo California, I applied for a commission (as an officer), and, during my tour of duty aboard the USS MIDWAY (then CVA-41, now CV-41) was accepted into the program and sent to the University of Utah to get a degree in Computer Science. I have programmed in such languages as FORTRAN, FORTRAN II, FORTRAN IV, JOVIAL, LISP, RPG, RPG-II, various assembler languages (mostly IBM, but also some Univac machines), ... There are more, but I'd have to get out a computer language 'history' book to look up the rest of them... Personally, I'm the oldest child of 14 (8 boys, 6 girls). My father was in the Navy as a Supply Corps Officer. He retired as a Commander in 1061 while stationed in Los Angeles California. They currently reside in Ventura California. 12 of the 14 children live within two hours travel time of Ventura. Hope that helps. gph] ------------------------------ Date: Wed, 3 Apr 91 0:25:05 CST From: david@wubios.wustl.edu (David J. Camp) Subject: Wuarchive Mirrors Resurrected The wuarchive mirrors disk is now again operational. Most of the mirrors have already been restored, but alas, the simtel20 portion is the slowest. We have tweaked the timeout parameters to hasten the restoration of files. Meanwhile you may find your favorite programs unavailable. Please bear with us. Thanks, -David- # david@wubios.wustl.edu ^ Mr. David J. Camp # # david%wubios@wugate.wustl.edu < * > +1 314 382 0584 # # ...!uunet!wugate!wubios!david v "God loves material things." # # abs (investment#1 - investment#2) << abs (anyinvestment - anydebt) # ------------------------------ Date: Tue, 2 Apr 91 08:19:36 EST From: moy@xp.psych.nyu.edu Subject: Re: Missing 1k on PC I had replied earlier that the Ping-Pong virus might cause your system to "lose" 2K. The Stoned virus reduces the total memory by 1k. However, I've also found that some peripherals may reduce the *total* RAM size. The Omnibridge Sysgen floppy controller reduces the overall size of memory by 1k. (This is a disk controller that can co-exist with a second floppy controller.) Even though you install OMNIBRG.SYS like a device driver, the controller has a built-in BIOS and apparently uses the main RAM in an unorthodox way. I suppose that anything that tampers with the BIOS report of available RAM, which is at address 0000:0413h, *before* DOS comes up would change the apparent total size of your system RAM. I have not seen a system where the hard disk controller's built-in partitioning program was used to make more than one DOS volume, but I suspect these schemes steal RAM in the same way. Because DOS versions 3.30 and later provide for partitioning multiple volumes (and therefore, do not require the controller's own partitioning), this may be why different DOS versions "see" different amounts of RAM. The version of DOS used does not and should not affect the total size reported for RAM, since this is assessed by the BIOS on power-up. Moy Wong Dept. of Psychology New York University ------------------------------ Date: Wed, 3 Apr 91 01:00:54 AST From: Filipe Santos <USERVLSI%LNCC.BITNET@CUNYVM.CUNY.EDU> Subject: Achieve Multi I/O Plus Mistery DIP Switches Hi netfolks, Once more I need this kind of help from you all, as somebody again handed us a IBM PC interface card, without documentation, as usual. It's an Achieve Multi I/O Plus card, with one parallel, two serial and a game port, standard PClone fodder. There are two DIP switch blocks (4 and 8 switches), colored light blue that blends handsomely with the green PC board bellow. It's all very well, but I haven't been able to figure out which switch combinations configure what ! It's true I haven't done a thorough and painstaking research, but I was counting on a kind soul among you to avoid that. Do anyone has the manual/docs for this thing ? If so, I'd like to ask just for a mail msg describing the DIP switch settings. Thanks in advance for any help,! Filipe "JunkBin" Santos USERVLSI@LNCC.BITNET ------------------------------ Date: Mon, 01 Apr 91 11:38:11 EST From: Paco Rosich <ROSICH%EVALUN11.BITNET@CUNYVM.CUNY.EDU> Subject: EMS in VGA ram? Does anybody know of a utility to simulate expanded memory or ram-disk using the free ram in a VGA card? I have an XT with a monochrome VGA monitor and a Paradise VGA 256K card. I'm trying to put Windows to work in it. With a 640K XT, you can play 'sol' or 'reversi' but it's unreasonable to use any "hard" windows application. As a first attempt to speed execution, I choose displaying in 640x480 monochrome mode instead of grey scale. This restriction suggest another improvement: in this mode there is about 200K free in the VGA, so .. why not use it as expanded memory? By the way, this hypothetical expanded memory manager have to be specifically designed to manage memory banks in the VGA, not just use a linearly addressable region in ram. Any idea welcome. If finally no such thing exists, I'll build my own. Paco Rosich Internet: rosich@vm.ci.uv.es Bitnet: rosich@evalun11.bitnet ------------------------------ Date: Tue, 2 Apr 91 11:22:43 -0500 (EST) From: "H. Scott Matthews" <hm0i+@andrew.cmu.edu> Subject: hardware compatibility A friend of mine has a PC and an apple IIGS. Some questions: Can he use his apple imagewriter 2 with the IBM, and if so, how? Can he use his IIgs monitor with his PC? He has the PC at work, and the GS at home. He wants to be able to just get a PC CPU at home and use the old monitor and printer... scott |H. Scott Matthews | Carnegie Mellon University | |Junior |Comp.E/EPP |(412) 268-4696 ------------------------------ Date: 2 Apr 91 09:01:00 EST From: "V70NL::CODE2111" <code2111%v70nl.decnet@nusc.navy.mil> Subject: Request for info on VAX/PC/MAC Linkups Ours is a medium-size firm, with about 1000 engineers & scientists. Our group has about 30 people, consisting of about 15 analyst/engineers, 7 programmers (our particular team), plus managers and secretaries. Our team's main job is creating/modifying FORTRAN programs to do scientific calculation & plotting. The programming is done on a VAX, using 286 PC's as text-terminals. This arrangement has been in force for maybe 8 years - ever since our group first used PC's to connect to the company's UNIVAC. We run our applications on request by one of our group's analysts. This requires a large number of parameters to be properly set to yield the correct results; right now that is accomplished via a long, ill-defined (for the occaisonal user) set of NAMELIST inputs to each program. If an analyst has an unusual request (and often they do), it requires a fair amount of run/ review/rerun cycles before the output is just what they want. Also, if an app is being revised, the only person who knows how to run it is the coder, since the revision has no doubt new or changed inputs parameters that only they know about. The main reasons we stay with the VAX, even tho it costs us in connect and storage charges are: large disk space -we maintain about a dozen large applications and several large (>10 MB) data files; virtual memory - these apps use many large data arrays that can take up to 10 MB of memory; multi-user makes the source code available to all users in common; graphics library; spooled printing to a local laser; lack of portability between VAX and PC FORTRAN implementations; lack of experience with other languages; and last but not least, inertia - "if it ain't broke, don't fix it". We have been looking at changing this relationship, and have some workstations on order, with an to using them (somehow) to make our job easier. We are also considering rewriting our old apps to give them a better interface - perhaps a GUI to make the input setup easier than the current NAMELIST method. I attended a demo last week showing a DEC w/s running X-WINDOWS, with a nicely-done GUI that acted as a front end to one of our typical text-based FORTRAN apps. Very nice, easy to use. So what are our NEEDS for any new system lashup ? * applications development using FORTRAN * transfer of our current apps to new system, intact * good execution speed (run times now in minutes) * plenty of disk space * hi-res graphics to printer * make the job of system management easy (eg disk backups,etc) And our GOALS ? * reduce net costs: VAX useage vs new systems + new system mgr * make our apps easier to run- easy enough for an analyst * re-engineer apps for better coding, better documentation * source code control - version control - config. management * make apps easier to create QUESTIONS : 1. suggested alternatives to present dumb terminal/multi-user mainframe configuration ? 2. what forms of PC/VAX linkup are there (eg- X-WIN; CLIENT/SERVER, etc.) ? 3. is it feasible create a front-end to run on a PC or MAC, which would drive the app on the VAX ? (this is called a CLIENT-BASED app ?) 4. or is it preferable to run the entire app on a local machine (eg multiuser w/s) ? 5. current available products to create these front ends (Oracle, Blyth/OMNIS, who ?) 6. how best use the workstations that we'll get (eventually) ? 7. pitfalls in converting/downsizing ? 8. recommendations on using outside consultants ? ANY SUGGESTIONS OR POINTERS WOULD BE GREATLY APPRECIATED. Scott Daniels ------------------------------ Date: Tue, 02 Apr 91 11:15:09 GMT From: Jacek Holeczek <HOLECZEK%PLKTUS11@SEARN.SUNET.SE> Subject: native vs translation modes I re-read my query that was in a recent Info-IBMPC Digest and decided I needed to describe the problem more accurately. First some theory! Many (or even all) of Seagate IDE drives have two modes of "work". Those modes define how the drive is recognized by the host computer. There are two modes - one called "Native" and one called "Translation". In native mode the host computer sees the "physical" geometry of the drive, and in translation mode it sees "translated" geometry of the drive. Here is an example : ST157A translation native sectors/track 17 26 read/write heads 7 6 cylinders 733 560 guaranteed sectors 87,227 87,360 And now its time for the story. Once upon a time (not very far from here and not very long ago) there was a pretty small AT computer operating at 16MHz with IDE (AT-bus) interface on board (with 1:1 interleave). There was also a ST157AT-1 HDD in it. In the beginning this drive was preformatted in its native mode, and the average seek time was 28 miliseconds, and track-track seek time was (about) 10 miliseconds. But the owner decided to prefor- mat this HDD once more, this time in its translation mode. And the results were: average seek time 28 miliseconds (not changed), and track-track seek time 1.1 miliseconds (10 times better!) In both cases preformating was done by DISK MANAGER V4.3 with 1:1 interleave (I am sure in the first case, but not in the second). Question - why in the native mode was track-track seek time so bad? (Translation should take time?) Half of the kingdom for someone who solves this problem ! Bye, Jacek Holeczek HOLECZEK@PLKTUS11.BITNET ------------------------------ Date: Tue, 02 Apr 91 22:08:05 MET From: Ben Salemans <U216013%HNYKUN11.BITNET@CUNYVM.CUNY.EDU> Subject: Search for information on LAN's I would like to buy a LAN)network which must offer the following possibilities: 1. both DOS-pc's and Apple Mac's can be connected to the LAN 2. the LAN can be connected to the campus wide Ethernet network (with IBM-mainframes and DEC-VAX-computers) 3. the LAN can be connected to the Internet network Do you know what kind of LAN offers these three possibilities? Yours truly, Ben Salemans ------------------------------ Date: Tue, 2 Apr 1991 16:45 EST From: "James D. Burnell, WVU Student" <UN034251@WVNVAXA.WVNET.EDU> Subject: Symbolic Integration Hello, My honors calculus class recently discussed integration techniques (i.e. integration by parts, trigonometric substitution, etc.) and my teacher mentioned that he would be very interested in the source code to Macsyma, a marketed math package that can apply such techniques to symbolically solve indefinite integrals. We are fairly certain that the source code for Macsyma is under wraps (understandably). Our question: does anyone know where we could get source code in any language (preferably even pseudocode) that would supply an algorithm to do this? Code for specific integration techniques is also welcome. I checked SIMIBM.IDX and couldn't find anything. Please send responses to me and I will summarize for the net. Thank you. James D. Burnell UN034251@WVNVMS.WVNET.EDU UN034251@WVNVMS ------------------------------ Date: 2 Apr 91 17:41:54 GMT From: skwood@acsu.buffalo.edu (Scott K Wood) Subject: WANTED : Simple Zmodem Term Prog. I am currently looking for a simple, no-frills, terminal program for the IBM that supports ZModem and baud rates up to 9600. If someone could send me the file or tell me where to find it, I would appreciate it. Thanks in advance.. Scott BITNET : v092mgp5@ubvms.bitnet INTERNET : v092mgp5@ubvms.cc.buffalo.edu skwood@acsu.buffalo.edu ------------------------------ Date: Tue, 2 Apr 91 18:51:44 PST From: Ya'akov_Miles@mtsg.ubc.ca Subject: Toggeling To/From Turbo Mode I have a turbo Phoenix bios which switches speeds by toggling the control alt + keys. Internally, this must be translated to some kind of OUT instructions - one to speed the machine up, and one to slow the machine down. Can someone please post the value of the OUT instructions, and their arguments. Y.N. Miles, <MULTI@TRIUMFcl.bitnet> ------------------------------ Date: Tue, 02 Apr 91 16:19:05 CET From: Karl Keyte <KKEYTE%ESOC.BITNET@CUNYVM.CUNY.EDU> Subject: TCU_32A.ZIP - Turbo C Flexible menus, windows, & Forms Entry New upload: <MSDOS.TURBO-C> TCU_32A.ZIP TCU 3.2, form entry, menus, windows, etc. for TC, TC++, BC++ Karl ------------------------------ End of Info-IBMPC Digest V91 #85 ******************************** -------