Info-IBMPC@WSMR-SIMTEL20.ARMY.MIL (02/18/89)
Info-IBMPC Digest Sat, 18 Feb 89 Volume 89 : Issue 24
Today's Editor:
Gregory Hicks - Chinhae Korea <COMFLEACT@Taegu-EMH1.army.mil>
Today's Topics:
Regarding subst'ing drives...
B-tree Indexed file access in 'C' (BPLUS)
Direct disk access
DOCPRINT uploaded to Simtel20
Epsilon EEL files uploaded to Simtel20
F11 and F12
DSP Help
Neural Network Computing Systems Request
Long PATH Strings
Reverse Video for VT100 Terminal Driver
Reply to Unknown Card (370 Option)
Memory Parity error on Zenith AT
New files uploaded to Simtel20
Odd Character in V 89 N 12 (2 msgs)
overlays in Turbo-C
Painter's Apprentice uploaded to Simtel20
Trouble with MS Windows (2 msgs)
Write protecting hardware
----------------------------------------------------------------------
Date: Mon, 13 Feb 89 10:25:27 PST
From: jkimble@crash.cts.com
Subject: Regarding subst'ing drives...
SUBST is an excellent way to shorten your execution path, but make sure
you change the parameter "LASTDRIVE =" in your CONFIG.SYS file. If it's
set to something like "E" and you attempt create a "higher" logical drive,
you'll be rewarded with an error message.
------------------------------
Date: 14 Feb 89 06:35:56 PST (Tuesday)
From: "hugh_davies.WGC1RX"@Xerox.COM
Subject: B-tree Indexed file access in 'C' (BPLUS)
I have a copy of BPLUS. It has at least one serious bug, which renders it
useless, at least on the IBM PC (or at least on the 8086 machines I've
tried it on).
It manages the database in memory, until it runs out of space, or you
close the file, when it writes the database back to disk. In order to do
this, it uses a lot of the 'C' memory allocation and moving routines. The
problem is the use of memcpy for overlapping areas in the procedure
moveup. As the man page says, "Character movement is performed differently
in different implementations. Thus overlapping moves may yield surprises".
In this case it means that writing new records into an existing index does
not work correctly. The bug is exhibited by the test program provided in
the ARC, and shows itself by the last 10 records written back into a file
all having the same key, even though they should be different.
Regards,
Hugh Davies
(HUGE.WGC1RX@XEROX.COM)
------------------------------
Date: 14 Feb 89 11:53:26 gmt
From: D.Murphy%edinburgh.ac.uk@NSS.Cs.Ucl.AC.UK
Subject: Direct disk access
I've got a DOS disk here with some corrupted sectors. I've used the Ultra
utilities to look at them and am told that the CRC is corrupted, however,
there is no indication of whether this is because the CRC or the data is
dead. Because of the discrepancy between the data and the CRC DOS won't
look at the disk.
Is there a way I can address disk tracks/sectors directly and read ALL the
data, including the CRC from the disk, without DOS getting in the way and
stopping this, and if so, how ? If there is software which'll let me do
this then fine, but I don't mind programming round it (anything accessible
via Turbo C v1.5 ?).
Oh yes, the disk was given to me by a friend (they have experimental data
on it) with the message "it is too late to lecture me on backups" :-).
D.Murphy@uk.ac.edinburgh
Chemistry Department
Edinburgh Univ.
------------------------------
Date: Tuesday, 7 February 1989 12:33-MST
From: BP3%PSUVM.BITNET@CUNYVM.CUNY.EDU
Subject: DOCPRINT uploaded to Simtel20
Now available from Simtel20:
Filename Type Bytes CRC
Directory PD1:<MSDOS.PRINTER>
DOCPRINT.ARC BINARY 83456 0C51H
DocPRINT is the alternative to using your word processor for formating and
printing .DOC files you get with ShareWare programs. DocPRINT will let
you use different size paper, printer font, and even allows you to print
on BOTH sides of the paper using a two pass print method without creating
TEMP files.
So far DocPRINT works only with EPSON or compatible printers, but the next
version will have Customizable Printing.
------------------------------
Date: Friday, 10 February 1989 13:08-MST
From: Brad Keister <KEISTER@uinpla.npl.uiuc.edu>
Subject: Epsilon EEL files uploaded to Simtel20
I've uploaded two more Epsilon EEL files to SIMTEL20. Their descriptions
follow this note. Many thanks to those who responded to my request on
info -ibmpc! At this point all EEL files of which I'm aware from previous
listings are now posted on SIMTEL20.
Brad Keister
Department of Physics
University of Illinois
KEISTER@UIUCNPL
keister@uinpla.npl.uiuc.edu
CALC.E A calculator for Epsilon. Results of calculations may
be placed in the file.
Yuval Rakavy <yuval%humus.huji.ac.il@RELAY.CS.NET> 4/22/87
CXTND.E Extended C mode. This code takes where Lugaru left off
and adds lots of spiffy stuff for C hamsters.
<John Kubiatowicz, kubitron@athena.mit.edu 3/27/87>
---
[The files are now in Simtel20's pd1:<msdos.eel> directory]
------------------------------
Date: Mon, 13 Feb 89 10:06:00 MST
From: <pgaughan@NMSU.Edu>
Subject: F11 and F12
Jeff Clough writes
> I'm already aware of the BIOS call that will return an extended key
>code. That's not what I'm after. I need a way to put the keyboard into
>a mode that will cause it to return the F11 and F12 key codes even when
>INT 16h with AH=0 is used. (I'm beginning to wonder if it is really
>possible.)
This sounds like a job for a handy-dandy TSR user written interupt
routine! The're relatively easy to write and if you know what you're
doing the routine shouldn't be more that 10 lines of intallation code and
10 lines of interupt handling. PC magazine's utility section is a good
place to look for examples.
Disclaimer: I have no connections whatsoever with above mentioned magazine
(including a subscription).
Patrick Gaughan Programmer of Gor
pgaughan@nmsu.edu New Mexico State University
[Such a routine already exists in the Lending Library in file
PD1:<msdos.keybd>f11f12.arc... Source and executables included. gph]
------------------------------
Date: Sun, 12 Feb 89 13:18:35 +0200
From: O M <FNMELNIK%WEIZMANN.bitnet@cunyvm.cuny.edu>
Subject: DSP help
Has anybody heard of a card for the IBM XT,AT,386... which allows fast DSP
(digital signal Processing) of picture frames. We have some material about
a card which supposably does some specific DSP functions at lightning
speed (15 Mflops) but we can't really make-out from the prospectus if it
is powerful enough to do Image processing.
What we need is a card which will execute a 1024x1024 2-dienssional FTt at
good speeds (in seconds not hours). The problem seems to emenate from the
fact that the card we have information on can't access the picture memory
directly...the card must DMA the data to its private data block and only
then can it be processed (that is very cumbersome when we have 1024x1024
bytes to deal with in each picture).
Please write back quickly....to my address not the listser
Thanx,
-Ofer
Acknowledge-To: <FNMELNIK@WEIZMANN>
------------------------------
Date: Mon, 13 Feb 89 23:16:14 GMT (Postman Pat ver 3.1)
From: Drew <SCR596%CYBER2.CENTRAL.BRADFORD.AC.UK@CUNYVM.CUNY.EDU>
Subject: Neural Network Computing Systems Request
This is very much a speculative message, however anybodys help would be
gratefully received.
Could I ask who, if anyone, is working on or is interested in Neural
Network Computing systems. If you could email replies to me I will pass
them on to my supervisor who has asked me to put out this call. Perhaps
we could start up a discussion group?
Thanks,
Drew Radtke, Phone: (0274) 733466 x532
Science & Society,
University of Bradford, Bradford, West Yorkshire,
BD7 1DP.
------------------------------
Date: 14 Feb 89 09:34:39 PST (Tuesday)
From: "George_C._Burkitt.ElSegundo"@Xerox.COM
Subject: Long PATH Strings
> I need to overcome the 130 char limit (or so) of the PATH string in
>DOS. I'm using MS-DOS 3.21, which allows to increase the environment
>using SHELL /E argument. This, however, does not allow having long search
>PATH for executable files.
The 2-14-89 issue of PC magazine User - to user section suggests a way to
greatly shorten the Path string, by use of the Subst command and the use
of logical drives to locate files.
------------------------------
Date: Tue, 14 Feb 89 17:58:32 -0500
From: Brian C. Anderson <castillo@media-lab.media.mit.edu>
Subject: Reverse Video for VT100 Terminal Driver
I'm running Xenix on an AT and I need to add support for "highlighted
lines" to a vt100 terminal driver. Does anyone know what ESC-codes are
used to set intensity-on and intensity-off on at vt100? Or is this
something that just isn't supported on this kind of terminal? Please
reply directly, I doubt anyone else cares about this problem :-)
------------------------------
Date: Tue, 14 Feb 89 15:23:44 EST
From: Tal Wyatt (HDL) <twyatt@BRL.MIL>
Subject: Reply to Unknown Card (370 Option)
Hans Pronk asks about a double-card labeled PC-370 P-card. That is the
so-called "370 option" for XT's (there is a very similar product with 16
address lines for the AT) which was made by IBM to allow XT's to emulate
370 mainframes. We have several of these for XT's and a bunch for AT's.
My advice is to give it back to whoever gave it to you. We note the
following:
(1) the double-card contains several 68000's with microcode to
emulate the 370 instruction set, but needs a third card to emulate an IBM
3277 (they don't make that anymore) or an IBM 3278/79 if this is the AT
version of the product. The double-card also contains 1/2 Meg of ram
which is bank-switched to provide several megabytes of 370-type virtual
memory.
(2) Very specialized software is required to run it at all --
available from IBM only, of course, if they haven't stopped selling it
altogether. The software is called VM/PC and costs about $1000. The
latest version we have is Release 2.01.
(3) The software works usefully only in connection with a 370
mainframe running VM/CMS, using the 327x emulator card for communications
(VM/PC also runs the emulation). We got stuck with trying to use it with
a 370 running MVS and the only thing it does for us is allow very slow
file transfers (compared with other 327x emulation software). The
software is probably great for someone who dreams in CMS and speaks 370
processor machine language fluently (really!), but not for people less
conversant with 370 stuff.
(4) The hardware is beastly slow, compared to a real mainframe IBM.
You can imagine waiting 5-10 minutes for a compilation that takes 5
seconds on the 370 mainframe. Typical.
Hope this helps.
Tal Wyatt
------------------------------
Date: 10 Feb 89 12:22:00 GMT-15:00
From: "T. DEJANE" <motu_7a@wmms-srf-yoko.arpa>
Subject: Memory Parity error on Zenith AT
In Vol 89 no 17 Steve Zelazny asks about Memory parity interupt error on a
Zenith AT (IBM clone)
> I'm working on a project in the physics department. I'm trying to debug
>an error of the form "memory parity interrupt." The code in which the
>error occurs is a data acquistion program.
>Data is input from a photosensor which records the intensity of light.
>The data is written into memory via a D.M.A. (Direct Memory Access)
>operation. The first time the data is manipulated in the code produces
>the "memory parity interrupt." Actually, the error doesn't occur all the
>Steve Zelazny (v055l9y3@ubvms.bitnet)
Parity is checked each time a memory retrive is done. Is it posible
that you have a memory problem? Or maybe the add-in card that you are
using to capature data is the problem? Most memory problems will not be
found by the power-on-test done by your computer.
Try running the ROM based memory test. To run this, type <CTRL><ALT><INS>
followed by 'TEST' at the prompt. Select 'memory test' from the
following menu. If the ROM test does not reveal a problem, try running
the disk based Diagnostic tests in both the fast mode and the normal mode.
If any of these tests show a problem, remove the add-in card that is being
used to capture data and re-run the failing test to ensure that the card
is not introducing the error. I would be interested to hear if any of
this helps.
Hope this helps.
T. D. DeJane (motu_7a@wmms-srf-yoko.arpa)
[For information, DeJane provides technical assistance in the data
processing and computer repair arena to US Navy units operating in the
Western Pacific Area. He has been working in the computer repair field
since around 1969. (He and I were at the Navy's computer repair school at
about the same time.) gph]
------------------------------
Date: Sun, 12 Feb 1989 01:26 MST
From: Keith Petersen <w8sdz@WSMR-SIMTEL20.ARMY.MIL>
Subject: New files uploaded to Simtel20
[--forwarded message--]
From: "Leslie C. Brown" <lbrown@BRL.MIL>
I've uploaded the following files to Simtel20:
pd1:<msdos.database>
PCFDB1A.ARC and PCFDB2A.ARC updated PCFile-dB executables and utilities
The document file PCFDB3.ARC remains unchanged.
pd1:<msdos.editor>DBXLMAC.ARC dBASE and Foxbase macros for Qedit
pd1:<msdos.editor>GALXY242.ARC Word processor replaces GALAXY23.ARC
pd1:<msdos.editor>G-MERGE1.ARC Merge addresses/letters to produce form letters
pd1:<msdos.opus>OMMM_130.ARC Opus mailer
pd1:<msdos.games>CRAZY8.ARC Crazy 8 card game
pd1:<msdos.zmodem>PCZ0115.ARC a free clone of DSZ file trans program
pd1:<msdos.dskutl>QCOPY.ARC another quick disk copying program
---
Thanks, Les!
--Keith
------------------------------
Date: Sun, 12 Feb 89 16:16:10 EST
From: "Robert E. Zaret" <ZARET@MITVMA.MIT.EDU>
Subject: Odd Character in V 89 N 12
In digest Volume 89 Issue 12, the second message (a query about putting
the current directory into an environmental variable) includes characters
that put my printer in expanded mode. The characters were in the phrase
"Something like CDIR=C:g?" where "g" is the first character and ? means
the rest of the phrase was garbled.
Ah well, no real damage done; just another adventure in the world of
computers.
------------------------------
Date: Mon, 13 Feb 89 08:12:42 MST
From: Gregory Hicks <GHICKS@WSMR-SIMTEL20.ARMY.MIL>
Subject: Odd Character in V 89 N 12
Well, I have two explanations. The RAW text has a 'back-arrow' just after
the ':' (Should be <back-slash>) and the other has a ^[ in it. The TENEX
version has a $ (TENEX version is the back-routed copy) there.
If it is ^[, that's <ESC> and the following chars are commands to the
printer.
Sorry for not catching that.
Regards,
Gregory Hicks
-------
------------------------------
Date: Mon, 13 Feb 89 11:02 MET
From: "Anneke Sicherer-Roetman"
<ANNEKE%RUUCLA.SURFNET%HASARA5.BITNET@CUNYVM.CUNY.EDU>
Subject: overlays in Turbo-C
Hello networkers,
Is there anybody out there who can tell me how to create and use overlays
in Turbo-C. Currently I have version 1.5, but I have understood that 2.0
does not support overlays either. Thaks in advance for your help.
Anneke Sicherer-Roetman,
State University Utrecht, Netherlands
------------------------------
Date: Saturday, 11 February 1989 20:41-MST
From: nelson at clutx.clarkson.edu
Subject: Painter's Apprentice uploaded to Simtel20
I have just uploaded Painter's Apprentice to Simtel20.
Filename Type Bytes CRC
Directory PD1:<MSDOS.GRAPHICS>
PA.ARC BINARY 108148 8D5FH
Painter's Apprentice is a bitmap editor in the genre of MacPaint, PC
Paintbrush, Dr. Halo, etc. Painter's Apprentice requires an EGA or VGA
and a mouse driver compatible with Microsoft's.
-russ
------------------------------
Date: Wed, 8 Feb 89 13:58 EST
From: <VY7106%BINGVAXA.BITNET@CORNELLC.ccs.cornell.edu>
Subject: Trouble with MS Windows
Hi ho. I'm having trouble with MS Windows v2.03 on an IBM PC/XT with
640K, CGA, running it at about 8 MHz from a 20 meg drive. Whenever I try
to execute a .PIF file, I get the message "Not enough memory...", however
it will still execute any file made for windows, like Write, Paint, and
some shareware stuff, like QVT and Command Post. I know I'm not short on
memory, because I've even tried it with VRAM simulating 128k EMS, and I
know it's not a disagreement with a TSR program, because I've tried it
with none present. For some reason, Windows just won't run a .PIF file.
Anybody else with a similar problem or solution?
If it's not too late to add to my message, I'm using DOS 3.30 with
BUFFERS=20 and FILES=30 and STACKS=9,128.
Rick Dickens
VY7106@BINGVAXA
------------------------------
Date: Fri, 10 Feb 89 02:00:22 MST
From: Gregory Hicks <GHICKS@WSMR-SIMTEL20.ARMY.MIL>
Subject: Trouble with MS Windows
You don't say what version of DOS you're using, but the 'out of memory'
happens to me when I don't have the "FILES=XX" set high enough...
(FILES=XX should be in your CONFIG.SYS file, and probably should be about
XX= 30 or 40...)
Regards,
Gregory Hicks
------------------------------
Date: Wed, 8 Feb 89 10:57:22 PST
From: Frank_Barchard@cc.sfu.ca
Subject: Write protecting hardware
When the question of password protection for harddrives was brought up, it
reminded me of my solution to protecting my modem from unauthorized usage.
I set the AT &L1 option(leased line) and wrote the configuration to
non-volitile RAM. The only way to use the modem then, is to type AT &L0
once. The same approach could be used for hardware, I expect.
------------------------------
End of Info-IBMPC Digest
************************
-------