Info-IBMPC@WSMR-SIMTEL20.ARMY.MIL ("Info-IBMPC Digest") (12/24/90)
Info-IBMPC Digest Sun, 23 Dec 90 Volume 90 : Issue 213
Today's Editor:
Gregory Hicks - Rota Spain <GHICKS@WSMR-Simtel20.Army.Mil>
Today's Topics:
bug in PKLITE?
Downloading Binary Files with Kermit under Telix
Looking for an "idle" algorithm
Modems
Paradox list
Re: Virus scanners
Today's Queries:
Disk Manager and Miniscribe drive problem
Disk Utility to Read or Modify Boot Sectors
divide overflow error when running WP 5.1
Non-DOS Restore from DOS Backup?
Deleting the File NUL
PC Fax products that works with Windows
Two modems/one works/one doesn't
Virus Scanners (V90 #191)
Wanted: Wildcat BBS demo
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: Thu, 20 Dec 90 15:06:10 PST
From: Ian_Gay@cc.sfu.ca
Subject: bug in PKLITE?
The compressed file produced by PKLITE compression of ED.EXE (from
version 2.71 of PC-Write) crashes my system when run. This is the only
example so far. Other programs have compressed and run successfully.
The version I have is PKLTE10.EXE from wuarchive.
Test results: I find PKLITE 20%-50% slower than LZEXE, the compressed
files are 2% to 5% smaller.
------------------------------
Date: Thu, 20 Dec 90 15:59:39 cst
From: Michael H. Riddle <riddle@hoss.unl.edu>
Subject: Downloading Binary Files with Kermit under Telix
In comp.sys.ibm.pc.digest you write:
>I am having trouble downloading binary files useing Kermit under Telix
>3.12 (registered!).. When I try to download a .ZIP file that I got
>through FTP, I set Telix (using ALT-O) to receive a binary file. But
>when I begin the transfer, it just sits there waiting for the file
>header. If I set Telix to receive the file as a text type file, it
>transfers alright, but doesn't run when I get it on my system. Any
>suggestions?
>David Hart (HART53@SNYPOTVA.bitnet)
I've been told you need to uuencode the file first, transfer it, and
then use a PC-based uudecode utility (there should be some at simtel20,
I would think.
I had the same problem and before I found an answer, found a compiled
version of sz (ZMODEM send) residing on the Unix box I use.
This probably isn't what you wanted to hear, but....
riddle@hoss.unl.edu | University of Nebraska
postmaster%inns@iugate.unomaha.edu | College of Law
mike.riddle@f27.n285.z1.fidonet.org | Lincoln, Nebraska, USA
------------------------------
Date: Thu, 20 Dec 90 09:57:21 +0100
From: S89406316%HSEPM1.HSE.NL@uga.cc.uga.edu
Subject: Looking for an "idle" algorithm
[John Larson is looking for an "idle" algorithm]
John,
You're right in saying that hooking up to the keyboard interrupt (int
9h) or checking for disk activity (int 13h) is probably not a good
idea, because the CPU could be doing something else.
But the CPU is definitely not doing anything else (well, hardly anyway)
when it's waiting for keyboard input (int 16h, service 0 - It slipped
my mind if there was a "wait for a key" function in Int 16h for
extended keyboards. I haven't got any doc at hand, sorry)
So, you hook up to Int 16h and check if a "key wait" function is
requested. If no, do the original Int 16h via a far JMP.
If the function requested WAS "key wait", do the following: record the
values of the time-of-day counters and add 10 minutes (providing for
hour timer rollover of course). Check if there's a key available (call
the original int 16h with AH=1 or copy the code from the BIOS listing)
If there is a key available, call the original int 16h, "wait for key",
then return.
If there was no key available, check if the current time-of-day
counters are above or equal to the recorded "start time +10 minutes".
If they do, you can do whatever you want but don't call any DOS Int 21h
functions. Remember to re-record the "start time+10" when your routine
returns.
Note: You must do an STI to enable (timer tick and keyboard)
interrupts, while your Int 16h handler is active.
Another thing you should know is, that DOS has an undocumented Int 28h,
which it calls frequently when it waits for the keyboard. The PRINT
program uses this interrupt for determining if it can read data from
disk: You can safely call any Int 21h function that has nothing to do
with keyboard, i.e. all functions with AH>0E (??? look this up!). That
includes disk services. Remember to always call the original Int 28h
when your handler is finished.
You could also hook up to the Int 21h itself, and with the Int 8 (timer
interrupt) check if the Int 21h takes 10 minutes or more to return
(which will probably be caused by DOS waiting for the keyboard.)
Conclusion: A good definition for "idle" is: "*Waiting* for the user to
press a key" (NOT: doing ANYTHING while the user is not pressing a
key). A combination of the techniques described above is probably the
most elegant solution for your problem.
Jac Goudsmit
Hogeschool Eindhoven, The Netherlands
S89406316@HSEPM1.HSE.NL
Happy holidays to everyone...
CU in 1990one
------------------------------
Date: Thu, 20 Dec 90 20:33:00 CST
From: 55SRWLGS <55srwlgs%SACEMNET.AF.MIL@uga.cc.uga.edu>
Subject: Modems
I have only used the U.S. Robotecs modem. These are Hayes compatible,
Hayes being the standard for modems for PC type micros. I'd rate the
external better than the internal. We use externals at work. At home I
have an internal, which I expect to replace in time. It's behavior is
squirrely.
In other words, I can upload using any protocol, but after the first 8
months, I could only get it to download using KERMIT and even parity. I
have not been able to find a solution, either.
The externals at work all work ok, however. Zenith computer stores
used to sell them. You would probabaly get the best prices through
Computer Shopper.
Frank Starr
55srwlgs@sacemnet.af.mil
------------------------------
Date: Thu, 20 Dec 90 13:55 EST
From: Michael Gordon <GORDONM@MAX.CC.DENISON.EDU>
Subject: Paradox list
LISTSERV@BRUFPB host the list PARADOX@BRUFPB
------------------------------
Date: Thu, 20 Dec 90 16:03:13 cst
From: riddle@hoss.unl.edu (Michael H. Riddle)
Subject: Re: Virus scanners
In digest V90 #191, Robert_Slade@cc.sfu.ca writes:
>I suspect that the call for a "safe" virus detector, protector,
>disinfector, et al was prompted by the reports that SCAN, one of the
>best known, is *not* safe to use. The fact is the files purporting to
>be versions 65, 68 and 70 of SCAN have been found to contain viri or
>trojan programs. The latest version that I know of is 67C (67 and 67B
>were functional but had bugs).
>An alternate is F-PROT whose author, Fridrik Skulason, can be reached
>at frisk@rhi.hi.is. This is the program that I most recommand. It is
>a very complete suite of virus detection and security utilities.
>Of the commerical software, I note that VirexPC is written by Ross
>Greenburg, well known in anti-viral circles for Flu-Shot.
Unfortunately, any utility is subject to the kind of tampering and
forgery that SCAN suffered. Starting with version 72, McAfee is
distributing it using the -AV autoverification function of PKZIP. I
suspect an adroit villian could find a way to fake the -AV signature,
but it would be a lot harder.
Still, you need to know your source. For me, that means McAfee's
board, or someplace like simtel20.army.mil which is known to get the
product straight from McAfee. (Or Greenburg or whomever. The point is
to know your source.)
riddle@hoss.unl.edu | University of Nebraska
postmaster%inns@iugate.unomaha.edu | College of Law
mike.riddle@f27.n285.z1.fidonet.org | Lincoln, Nebraska, USA
------------------------------
Subject: Today's Queries:
Date: Thu, 20 Dec 90 08:20:43 EST
From: R. Allen Wilkinson x3383 <urt@magi.ncsl.nist.gov>
Subject: Disk Manager and Miniscribe drive problem
Hello everyone,
I am looking for a way to fix a driver problem.
First the driver that I am having problems with is "dmdrvr.bin". I
would like to stop using it if possible. I have a hard drive which
needs it. The driver causes conflicts with other drivers and obviously
takes up memory. The only reason I see needing the driver is to allow
my system to recognize my hard drives special configuration. Are their
ROM chips that can fix this? Or some other kind of fix to free my
memory and resolve conflicts?
Drive: Miniscribe Model 3180E
Number of Heads: 7
Number of Cylinders: 1250
ROM bios: DTK 386 Bios Version 4.25 06/12/89
Driver: dmdrvr.bin from Ontrack Computer Systems Inc.
Disk Manager Version 4.02
Any help or insight would be greatly appreciated,
R. Allen Wilkinson urt@magi.ncsl.nist.gov
------------------------------
Date: Thu, 20 Dec 1990 21:18:19 EST
From: THATERW@VAX.CS.SUNYHSCSYR.EDU (Bill Thater)
Subject: Disk Utility to Read or Modify Boot Sectors
Can anybody recommend a PD/Shareware Disk Utility package (read "not
too expensive 'cuase I gotta buy it myself") that will allow me to
read/modify the boot sectors, FATs, et all on floppies and hard disks?
I will need it to work on all types of disks. Please reply direct to me
at the address below (note: Please try to use the Bitnet address, our
mailer doesn't always let me get Internet mail :) ) and I'll sumerize
to the list if I get any answers.
Thank you all for your time and effort.
Bill
Bill Thater
Centro Parking Inc.
Voice: (315) 464-4539
E-mail: THATERW@SNYSRYV1.BITNET
THATERW@VAX.CS.SUNYHSCSYR.EDU
THATERW@139.127.2.1
------------------------------
Date: Thu, 20 Dec 90 16:27 CST
From: <UABSSG02%ASNUAB.BITNET@CUNYVM.CUNY.EDU>
Subject: divide overflow error when running WP 5.1
I am unable to run Word Perfect 5.1 on my PC-XT (Zenith Data
Systems, (Model ZF-148-41). After loading WP, when I try to run the
program, I get a 'Divide Overflow' error. The message keeps scrolling
and I have to reboot the computer.
I get the same error when trying to change the system date or time
from Norton Control Center. All other packages that I've tried are
working just fine. (Lotus 123, Dbase III +, WordStar 3.3, and many
more). I'm wondering if the problem has something to do with changing
the system date. I do not have a math co-processor.
The zenith manual says, 'This message typically results when some
process has created an illegal processor state. Normally this message
is enountered in math processes. If encountered during testing, it
could indicate possible problems in the CPU, the Monitor ROM, system
memory, or the co-processor.' This does not make much sense to me.
I would appreciate it if knowledgeable persons on the net can help
solve this problem. Please reply to me directly, if you prefer. Thank
you very much.
(System config. : 8088 processor, 2 external disk drives, 20meg
hard drive, DOS 3.3. )
Suresh Grandhi
uabssg02@asnuab.bitnet
------------------------------
Date: Thu, 20 Dec 90 15:44 EST
From: <JFL4%PSUVM.BITNET@uga.cc.uga.edu>
Subject: Non-DOS Restore from DOS Backup?
I apologize if this question has been asked ad nauseum - is there a
utility, PD or commercial, which allows a selective restore from a DOS
backup? I have a 17 disk backup which will not restore with any
version of 3.X or 4.0 DOS. It stops after the second disk and says
"Files not contained on Source disks" or something to that effect.
I only need two or three files off of this backup so there's a pretty
good chance they're there - it's just that DOS won't let me get past
the second disk.
Any help would be appreciated. Please respond directly and if there's
sufficient interest I'll summarize to the board.
Thanks,
Jeff Luck
Penn State
------------------------------
Date: Thu, 20 Dec 90 8:47:25 CST
From: Thom Tellez <tellez@STL-06SIMA.ARMY.MIL>
Subject: NUL, the file
How may I delete a file called NUL?
I've tried the easy way. I've looked at FAQ files. I've RTFM. Still
I have the file.
I do not know how I created it, but I would like help or suggestions
for getting rid of it.
The file shows in the root directory of my c: drive as 59 bytes long.
I cannot, however discover any contents to it (with VB's LIST for
instance). My machine is a 386 running 4DOS 3.02a, DOS 3.3., QEMM 5.1.
Others similarly configured have not had this problem. The file size
and data stamp have not changed since I first noticed it.
Thanks for whatever help you can give. /\
/* \
/ * \
Season's Greetings from Thom '''||'''
------------------------------
Date: Thu, 20 Dec 90 12:09:55 CDT
From: Tom Cervenka <CTCT100%UICVMC.BITNET@UICVM.uic.edu>
Subject: PC Fax products that works with Windows
Is anybody using or know of someone who uses a fax board or somesort
of interface and also runs Windows 3? I've tried The Complete PC and
Gamma Fax and neither of them will work when Windows 3 is running. If
you send replies directly to me I'll summarize them for the rest of the
list. Thanks.
Tom Cervenka - Univ of Il at Chicago, Info Cntr, Phone 6-7739
Internet: ctct100@uicvmc.aiss.uiuc.edu Prodigy: CMGB18A
------------------------------
Date: Thu, 20 Dec 90 10:56:00 EST
From: VFM702%SCRANTON.BITNET@uga.cc.uga.edu
Subject: Two modems/one works/one doesn't
I have a tale of two modems and three computers. I will do my best to
simplify a very confusing situation. My apologies in advance if I
dont't use the right terms since I don't know the phone terminology
that well.
The two modems are:
An Everex 2400 external
A Practical Peripherals 2400 external
The three computers are:
A PS/2 Model 50
A Compaq 286 laptop
A Compaq 286 desktop
The Everex now works on all three computers; the Practical Peripherals
works on none of the computers. Software: Kermit (2.32 and 3.0).
Up until its use on these computers, the Practical Peripherals modem
has worked fine, mostly on PS/2 Model 70s and 30s.
The one obvious difference is that the three computers listed above
(Model 50 and two Compaqs) are in a building with a PBX system
supporting voice and data lines. When the Practical Peripherals
worked, it was being run through a standard (?) Centrex system in other
buildings.
NONE of the modems worked when connected through this PBX system. A
phone line was run through spare jacks in the wall which are not on the
voice/data system, but still part of this building's phone system. That
is when the Everex started to work but the Practical Peripherals still
did not.
It is very important to us that we get these Practical Peripherals
modems to work in this building so any idea or suggestions would be
most welcome.
Thanks in advance.
Vince Merkel
Computing Systems
University of Scranton
800 Linden St. Scranton, PA 18510-4668
717-941-7637
BITNET: VFM702@SCRANTON.BITNET
INTERNET: VFM702@JAGUAR.UCS.UOFS.EDU
------------------------------
Date: Thu, 20 Dec 90 15:02:43 CDT
From: Tom Cervenka <CTCT100%UICVMC.BITNET@UICVM.uic.edu>
Subject: Virus Scanners (V90 #191)
The author of the note suggests that the VIRUSCAN is unsafe because of
trojan versions and suggests another product instead. Actually, the
problem is not with the program but with the distribution. VIRUSCAN is
shareware so you are encoraged to pass it along. It is this practice
that make leaves us vulnerable to the trojans. You can be sure you have
an authentic version however if you simply run the validation program
and call the bulletin board to verify the checksums.
Since new viruses are created each month any virus defense is going to
require frequent updates and direct mail is too expensive what with the
price of disk media and postage. I think that the CVIA system is one of
the best, safest and least expensive.
Tom Cervenka - Univ of Il at Chicago, Info Cntr, Phone 6-7739
Internet: ctct100@uicvmc.aiss.uiuc.edu Prodigy: CMGB18A
------------------------------
Date: Thu, 20 Dec 90 16:51:33 -0500
From: dsrekrg@prism.gatech.edu (Rob Gibson)
Subject: Wanted: Wildcat BBS demo
Can someone tell me where I can find an up to date Wildcat demo BBS at
an ftp site? Simtel only has a version from last year.
Thank you,
Rob Gibson, WREK 91.1 MHz, 40000 watts
Georgia Tech, Atlanta Georgia, 30332 404/894-2468
uucp: ...!{allegra,amd,hplabs,seismo,ut-ngp}!gatech!prism!dsrekrg
ARPA: dsrekrg@prism.gatech.edu
------------------------------
End of Info-IBMPC Digest V90 #213
*********************************
-------