Info-IBMPC@USC-ISIB.ARPA (Info-IBMPC Digest) (01/25/86)
Info-IBMPC Digest Saturday, 25 January 1986 Volume 5 : Issue 13
This Week's Editor: Eliot Moore <Elmo@USC-ISIB>
Today's Topics:
Administravia - Errata
The Truth About Lotus
Lotus and GM
GM Buys Lotus
DU.BAS and DUMP.ASM added to library
GREP modifications by Joe Newcomer
NASH-MORE, aka PC-MORE, added...
MARK3.ASM added to library
Tandy 1000 & SCO Xenix
Ejecting Paper
Default cluster size and LS.COM
Some Speedup info from Compuserve IBM PRO SIG
Cranium 10 Query
DeskView Query
----------------------------------------------------------------------
Date: 24 Jan 1986 00:01:35 PST
Subject: Administravia - Errata
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
Reply-To: INFO-IBMPC@USC-ISIB.ARPA
Digests V5 #10 and #11 were incorrectly titled "Wednesday January 23."
We sincerely apologize for any inconvenience.
Elmo
-------
------------------------------
Date: Fri, 24-Jan-86 10:34:37 EDT
From: David Farber <farber%pcpond.pc.udel.edu@Louie.UDEL.EDU>
Subject: The Truth About Lotus
Sorry to disappoint, folks. GM bought a majority interest in Group Lotus plc
of Hethel, England. This is the famous racing company founded by Colin Chapman
in 1958 (I think).
GM wanted to buy the whole thing, but Toyota owns 20% or there abouts & won't
sell.
GM is trying to boost it's image in the racing/specialist-engineering world,
I would guess.
Steve.
[19 Million was a bit cheap.... -wab]
------------------------------
Date: 24 Jan 1986 15:32-EST
Sender: NGALL@G.BBN.COM
Subject: Re: Lotus and GM
From: NGALL@G.BBN.COM
To: info-ibmpc@USC-ISIB.ARPA
GM did not acquire any part of Lotus Development Corp.
They acquired majority interest in Lotus, Ltd. the British automotive
manufacturer. (I was fooled by the headline in the NYT for a minute
myself!)
-- Nick Gall
------------------------------
Date: Fri 24 Jan 86 17:16:06-PST
From: Brad A. Silverberg <SILVERBERG@SRI-CSL.ARPA>
Subject: GM Buys Lotus
To: info-ibmpc@USC-ISIB.ARPA
Sorry, but the Lotus that GM bought a majority share of is the
automobile manufacturer in England, makers of such illustrious
cars as the Elan, Elite, and Europa. They also make Formula 1
racers.
Mitch Kapor would have a heart attack if GM had bought 50 % of
Lotus Development for only $19 million!
Brad.
------------------------------
Date: 22 Jan 1986 18:54:12 PST
Subject: DU.BAS and DUMP.ASM added to library
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The following utilities have been added to our public domain software library:
DU.BAS This program lists the number of bytes occupied by each
directory on the default disk. For each directory, it
indicates:
Actual number of bytes in files
Number of bytes in sectors to contain those files
Number of files
Storage efficiency.
Documentation is included in the program.
<STEHLE@SRI-TSC> 12/31/85
DUMP.ASM This is a filter that transforms the standard input stream
into a HEX and ASCII dump and sends it to the standard output.
Both input and output can be redirected, e.g. DUMP in-name >
out-name. If just a filename is given on the command line,
e.g. DUMP filename, the file will be read and dumped to the
console. By Richard G. Markley with enhancements by
Ted Shapin.
<Ted Shapin, BEC.SHAPIN@USC-ECL> 12/18/85
-------
------------------------------
Date: 22 Jan 1986 19:25:23 PST
Subject: GREP modifications by Joe Newcomer
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
GREP.C, modified by Joe Newcomer, has replaced the older version in the
library and is available for snarphing by interested individuals. Joe's
comments follow:
From: Joe.Newcomer@a.sei.cmu.edu
To: info-ibmpc@usc-isib.arpa
Subject: PC-DOS GREP
I took the copy of GREP which had been in <INFO-IBMPC> and made a
few changes. Some changes were gratuitous: I like case folding,
filenames and line numbers to come out with no effort on my part.
I set the defaults to behave this way. I removed the comments around
various tolower() operations and replaced the folding with a runtime
test, defaulted to fold but the -x (for eXact) switch disables this.
-n and -f now have the opposite meaning, turning off line numbers and
filenames. :0 and :9 indicate numeric patterns.
More significant changes however were required to run under PC-DOS.
Since the shell does no wildcard expansion, I had to put this in
the source. I also added a -s switch which will scan any subdirectory
which matches the basic file pattern (using *.* for the filename).
To simplify my life, I used a standard library package, CTOOLS-2 from
Blaise Computing, to provide the first and next filescan operations;
careful reading of the documentation is required, plus reading the
source, to discover how to make this work properly when recursively
scanning subdirectories; the 'drbuffer' operations save and restore
the state, apparently successfully. If someone is inclined to write
the low-level code to make this independent of the libraries, feel free.
Also, Lattice C did not properly handle the ': ' pattern because the
space separates the : and what follows; I changed it to accept :. for
space.
I then successfully used it to find a file I had lost (of COURSE it is
in that sub-sub-directory, an obvious place to put it...how could I
have forgotten!?) So it seems to work.
joe
-------
------------------------------
Date: 23 Jan 1986 15:51:50 PST
Subject: NASH-MORE, aka PC-MORE, added...
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
Don Nash's latest version of his MORE program has been added to our
public domain software library. As with all our other snarfable files,
it's on host USC-ISIB, directory <INFO-IBMPC>. Guest account is ANONYMOUS,
and password is GUEST.
NASH-MORE.C Originally called PC-MORE but renamed with the author's
nash-more.doc permission so that it won't be confused with other
PC-MORE's, this is a possible replacement for the PC-DOS
utility MORE.COM. NASH-MORE accepts filenames on the
command line as well as from pipes. It allows forward
paging by the screenful, half screenful, or line. It will
also rewind the file, call up an editor on the file, push
to an inferior COMMAND.COM, and do other niceties. Written
in CI C86, version 2.20G.
<Donald L. Nash, DLNASH@NGP.UTEXAS.EDU> 12/13/85
-------
------------------------------
Date: 23 Jan 1986 18:18:14 PST
Subject: MARK3.ASM added to library
From: Koji Okazaki <swg.Koji@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The following utility has been added to our library:
MARK3.ASM A utility to set attributes of files, flicking the mode
mark3.doc bits. After assembling and linking, run it through
EXE2BIN to create a .COM file. Usage is as follows:
MARK path\filespec /xxxx, where x may be:
R for read-only,
H for hidden,
S for system,
A for archive,
N for normal (all attribute bits off),
Q for quite (no messages)
All other attributes are ignored if N is present.
The filespec may include wildcards.
<Ted Shapin, BEC.SHAPIN@USC-ECL> 1/8/86
-------
------------------------------
Date: Fri, 24 Jan 86 9:39:13 EST
From: Brint Cooper <abc@BRL.ARPA>
To: kimery@ford-wdl1.arpa
cc: info-ibmpc@usc-isib.arpa
Subject: Tandy 1000 & SCO Xenix
Actually, no one to my knowledge has ever claimed '100%' PC
compatibility. Other Tandy machines are 'more' compatible than is the
1000.
However, there is a growing third party market for memory expansion
boards for the T1000. These boards include the DMA chip, sockets for
512K memory, RS232, and real time clock/calendar with accompanying
software. Some also include a ramdisk program. So, while you cannot
use the 'cheap' PC memory boards, you can get most of the commonly
needed functions in one expansion slot.
And you don't need any accessory for the parallel printer port. It is
standard on the T1000.
We have a T1000 in the family and are very happy with it. Of course, we
didn't try to get UNIX(tm) on the cheap. We're running MS-DOS, and
every program written for the IBM that we've tried has worked just fine.
Disclaimer: I don't bother with this, but I have no affiliation with
Tandy; opinions and views expressed are entirely my own.
Brint Cooper
ARPA: abc@brl.arpa
UUCP: ...{seismo,decvax,cbosgd}!brl-bmd!abc
------------------------------
Date: 24 Jan 1986 11:39:35 PST
Subject: Ejecting Paper
From: Craig Milo Rogers <ROGERS@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
After printing a file on my Epson I need to eject a blank page
so I can tear off the page(s) I just printed. Well, I've gotten too
lazy to stand up and hit the appropriate buttons on the FX-80. Instead
I run the program EJECT, which outputs a single form feed (^L) to PRN:.
An alternative is to use the DOS PRINT command: PRINT \DEV\NUL.
This can be put in a batch command file.
Here is the source for EJECT.ASM. Assemble, link, and convert
it to a .COM file in the usual manner.
Craig Milo Rogers
-----
PAGE 60,132
TITLE Eject -- Force a Form Feed On the Printer
; eject.asm 23 Jan 86 Craig Milo Rogers at USC/ISI
; Created this program to force a form feed on a line printer.
CODE_SEG SEGMENT
ORG 100H
ASSUME CS:CODE_SEG
ASSUME DS:CODE_SEG
BEGIN:
MOV AH,05H ; DOS function "Printer Output".
MOV DL,0CH ; An ASCII Form Feed.
INT 21H ; Print on on the display.
INT 20H ; All done, exit.
CODE_SEG ENDS
END BEGIN
-------
-------
------------------------------
Date: Fri, 24 Jan 86 12:27:40 PST
From: prandt!reynolds@AMES-NAS.ARPA (Don Reynolds)
To: amelia!info-ibmpc@usc-isib.ARPA
Subject: Default cluster size and LS.COM
Cc: reynolds@AMES-NAS.ARPA
Some comments on the following message from Vol. 5, Issue 11, of this
Digest:
------------------------------
> Date: Thu, 23 Jan 86 08:09 PST
> From: DGilbert.ES@Xerox.COM
> Subject: Cluster Size for 20 MB Hard Disk
>
> My thanks to Dan Hickey. Under DOS 3.1 I deleted the active partition
> on my 20 mb drive ( IBM PC/XT ) and then recreated it. This using
> FDISK. After using FORMAT/s/v, the cluster size dropped from 8k to
> 2k. Why do it? Well, my remaining free space went from approx. 3 mb
> up to 8 mb. Well worth the change. Apparantly, most hard disk you
> buy, such as the 20 mb unit I purchased from Priority One, already are
> partitioned with dos 2.1 format. FDISK tells you only that there is
> already a dos partiton. USE FDISK ANYWAY.
> -----------------
I have a PC/XT (circa 1984) with the standard 10 MB hard disk. The disk
has not been re-formatted since I upgraded from DOS 2.0 to 3.1, but my
impression is that it would set the cluster size to 4k, (not 8k) for the
10 MB XT disk, and 2k for the faster 20 MB AT disk. I know the current
cluster size is 4k, and do not understand the default 8k cluster size on
the 20 MB disk from Priority One.
Large cluster sizes may make sense if all you have is large files, but
small files dominate on my disk too, Doug, and I would save about 1 MByte
of space by going to 2k clusters on the 10 MB XT disk.
For anyone concerned about disk utilization, I highly recommend a public
domain utility LS.COM by R. Edward Nather (modified for Lattice C by
Jeffry Spidle). I use it in a simple command:
ls -saR | sort /r > dirsort.prn
This gives a listing of directories in order of decreasing space
occupied. A bit of text editing on DIRSORT.PRN to add quote marks and a
^L at the end permits plots using Lotus 1-2-3. It's easy to use this
information to migrate the lower priority directories with BAC.COM (PC
Magazine, Aug. 20, 1985, pp. 197) or with your favorite utility.
I found LS.LBR on the NASA BBS. If the LS.C and QSORT.C sources are not
in the Info-IBMPC Digest library, I will be happy to make them
available. An excerpt of LS.DOC follows at the end of this message.
Why did IBM use a 2k cluster size on the AT disk and 4k on the XT?
Smaller cluster sizes require more seeks and therefore slower
performance. The AT disk is so much faster than the older XT model that
the more efficient packing of the smaller clusters can be afforded.
Methods for setting the cluster size have been discussed earlier in this
digest (See Vol. 4, Issue 103, dated 7 Sept. 1984).
Best,
Don
-----------------ls - a Unix-like directory listing program for MS-DOS
(c) 1984, R. Edward Nather
Syntax: ls [-aclrstuR] [(path)name ... ]
Options may appear in any order, grouped or separated; if separate,
each must be preceded by a dash. The name(s) may refer to files or
directories. If no name is given the current directory is listed.
Options:
(none) Show filenames (only) sorted alphabetically
-a all: include system files, hidden files, "." and ".."
-c columnar: list information in 1-column format
-l long listing: include file's size, date, time, attributes
-r reverse the sorting direction
-s report size(s) only
-t sort by time of last file modification
-u include actual disk use, with totals & available space
-R recursively list all subdirectories
--------------
------------------------------
Subject: Some Speedup info from Compuserve IBM PRO SIG
Date: Thu, 23 Jan 86 22:47:15 -0500
From: Dave Farber <farber@huey.udel.EDU>
January 22, 1986
This file was created in response to a request to make a reference guide for
speeding up the IBM PC-AT available in the CompuServe IBM SIG data libraries.
It contains information gathered from a number of IBM PC-AT users, companies,
and my own experiences.
When the PC-AT was released, the fact that IBM had chosen to socket the clock
crystal was greeted with much interest and speculation, and it's this ability
to remove the crystal that is the key factor in speeding up the AT. The crystal
is contained in a 1/2x3/8" silver housing that plugs into its socket located
near the 80286 chip, behind the center opening for the hard disk. The crystal
can be removed easily by inserting a small nonmagnetic screwdriver or table
knife between the crystal and socket, and pushing the crystal toward the AT's
red power switch, then lifting it up and out of the retaining clip. A faster
crystal can then be insterted to increase the operating speed of the 80286 CPU
chip which is driven at 1/2 the crystal speed. WARNING: Changing the crystal
may void the warranty so save the original crystal to reinstall if your system
needs servicing.
CRYSTAL SPEED-UP PRODUCTS
Crystals:
The AT comes with a 12 mhz thick lead crystal, giving a 6 mhz CPU
speed, but research done by ARIEL shows that (providing the new
speed limiting ROM isn't installed) 99% of ATs will run with 16
mhz, 85% with 18 mhz, and 68% with 20 mhz crystals. Two types of
replacement crystals are available:
1) Crystals with thin wire leads:
These crystals are very cheap and readily available, but
the lead wires are very thin, so may have to be bent double
to make good contact in the AT's crystal socket. These
crystals are better used for making speed switches, than
for use on the AT's motherboard. [ Herm Fischer discovered
that the gold pins from an RS232 connector can be soldered
to thin lead crystals providing a better fit. -wab]
2) Crystals with thick leads:
These crystals are more expensive, but properly fit the AT's
crystal socket. ARIEL emphasizes the importance of using
high quality, military grade thick lead crystals (which they
sell) in the AT, and their claims may be valid as Peter
Norton reported having problems with some cheap thin lead
crystals which resolved when he switched to quality thick
lead crystals, but I have used cheap thin lead crystals from
B.G. Micro without problems, and found them to work well in
homebrew speed switches.
Speed Switches:
Some cases have been reported of software that won't install (probably
due to the copy-protection) or run (due to certain types of timing
loops) with the faster crystals installed, and speeded-up ATs will not
work with some local area networks. To work around these problems, one
can use a speed switch to switch back and forth between several speeds,
using a slower speed for speed limited software ar LAN use, and faster
speeds for number crunching, etc.. A homebrew speed switch can easily
be made for < $10, using 2 crystals and a small DPDT switch connected
to 2 wires that plug into the crystal socket (I have used one for
months without problems), or a variety of commercial products are
available, most of which consist of a switch and 2 to 5 crystals
mounted on an expansion slot cover bracket (this prevents the slot from
being used by any expansion boards) and connected to 2 wires plugged
into the crystal socket.
VARIABLE OSCILLATORS
Another approach to increasing the AT's clock speed is to use a variable
oscillator instead of replacements crystals. BGI is currently offering such a
product.
FAST MEMORY CHIPS
When operating the AT at faster speeds, faster RAM chips may be required. In
some cases, faster RAMs are required on memory expansion boards, than on the AT
motherboard itself (This seems to be especially true of paged memory boards).
On the AT motherboard, 150ns chips generally work well at CPU speeds of 8 mhz,
and some users have found them adequate at CPU speeds as high as 10 mhz, though
one supplier of fast RAMs claimed that 150ns chips used in a 10 mhz AT would
get flaky from heat degeneration after 8 - 9 months of use. Basic guidelines
for motherboard RAMs would be for CPU speeds < 9mhz, use 150ns; for 9 - 10 mhz,
use 120ns; > 10 mhz, use 100ns. These same guidelines will also hold true for
some of the extended memory boards (such as the AST Advantage), but others may
require 120ns or 100ns only, or may not work at all at high speeds. The ability
of paged memory boards (such as the Rampage AT) to function at > 6mhz has been
erratic to date, and they may not work at all at high speeds, though the makers
of these boards are trying to solve these problems. In contrast, there have
been reports of AST Advantage boards working at CPU speeds of 12 mhz or
greater.
[I spoke with Chris Symes of Tall Tree. They use 150 ns chips exclusively.
Their memory boards run reliably at 18.7 Mhz. Tall Tree boards provide a memory
paging scheme which can be used to expand PC memory beyond 640K or the boards
can be used as straight AT memory bypassing the paging hardware. The paging
hardware does not slow access time. They reccommend not using crystals faster
than 18Mhz and insist you should have no problems at this rate.
Crystals faster than 18 Mhz won't help on a standard AT anyway. The 80286 has a
few bytes of on board cache. Usually memory fetches will overlap with
computation. At speeds greater than 18 Mhz calculations proceed faster than the
AT can fetch memory. The processor will get memory bound and insert processor
wait states in order to wait for memory. Peter Norton's speed test will still
falsely show improvement as his loop of multiply divide pairs has a higher
ratio of compute-load per byte of instruction than most programs. My
conclusion is that unless one designs an AT clone from scratch to run with a 20
Mhz crystal, don't try speeding up your AT past 18 Mhz -wab]
802886s
Many, if not most ATs are shipped with 80286s rated at 6 mhz, and though they
usually work fine at 8 mhz, for those wanting to run the AT at 10 mhz or more,
the 80286-6 may have to be pulled from its socket and replaced with a faster
chip. There are 10, 12, 14, and 16 mhz 80286s available, though the supporting
chips may not be able to handle speeds > 12mhz. Reportedly, old ATs with the
orange motherboard may need to have some of the chips unsoldered and replaced
with faster chips before they can function at high speeds. AT users who want to
run at very high speeds can send their motherboard to BGI for any necessary
modifications and testing. It's believed that IBM will release a more expensive
8 mhz AT this year.
80287s
In the AT, the 80287 math coprocessor is normally driven at 1/3 the crystal
speed, which is normally 4 mhz. The standard 4 mhz 80287s may not work with
crystals faster than 16 mhz. There are 8, 10, and 12 mhz 80287s available for
use in speeded-up ATs, though even with a 20 mhz crystal, the 80287 would be
driven at < 7 mhz, so for maximum speed, small boards are available that plug
into the 80287 socket and drive the 80287 at 8 mhz.
80386s
Intel has been demonstrating an AT with a small board that holds an 80386 and
plugs into the 80286 socket, turning the AT into a limited 80386 machine.
Another company has a similar product but has chosen mot to market it. Several
companies have 80386 products for the AT in Beta test, so it's likely that
80386 products will be released this year for the AT, similar to the 80286
products already being sold for the PC and XT. BGI will be offering an 80386
motherboard for the AT, probably in March.
NEC CPU CHIPS
It has been rumored for several months that NEC will be releasing another one
of their V series chips called the V-60, which will be a faster 80286 clone.
Dvorak reports that it will have a built-in math coprocessor. If they achieve
true 80286 compatibility, putting one in an AT with a fast crystal could be
interesting.
SOFTWARE PATCHES
Floppy Drive:
Some AT users have experienced "Disk Read Error" type messages when
accessing the floppy drive at CPU speeds > 8 mhz due to timeout errors.
Patches are available to prevent this. A public domain version is
available in the CompuServe IBM SIG data libraries, and several
companies offer commercial versions that also speed-up the keyboard.
NEW AT ROM:
The new AT ROM being installed in ATs having IBM's new 30 MB hard disk, has a
timing routine that prevents the system from operating at speeds > 6 mhz.
Currently, I know of only 4 ways users wanting to speed up these ATs can deal
with this problem:
PROM Blasters: Use one to copy and/or patch a ROM.
ROM Patch: Send your motherboard to BGI and get the ROM patched as part
of their motherboard upgrade package.
PHOENIX ROM: Buy the Phoenix ROM for the AT and turn your AT into an
AT clone with the compatibility of a Compaq 286. (You
must use the Compaq GW BASIC).
Circuit Boards: The AT-2 board from AMS defeats the ROM timing routine.
The best solution is to refuse to accept an AT with the new ROM from your
dealer, thus casting a vote with your buying power against what appears to be a
marketing move by IBM prior to the release of a faster, more expensive version
of the AT.
PRODUCTS AND SUPPLIERS
The product discriptions here consist of information supplied by the
manufacturer or distributor which is subject to change, so check with the
company for the latest info.
AMS (818-810-8443):
High quality thick lead crystals
AT-1 Board: Lets you switch manually between 3 speeds. Has a system
reboot button. Occupies an expansion slot.
AT-2 Board: Same as AT-1, but lets you change speed from the keyboard
and defeats the timing routine in the new AT ROM.
ARIEL (201-788-2788):
High quality thick lead crystals
Fast 80286s
Floppy drive timeout and keyboard speedup patches
Beta testing 80386 product, and board to defeat the timing routine in
the new AT ROM
BGI (215-538-3900):
100ns RAMs
80286s: 10, 12, 14, and 16 mhz
80287s: 8, 10, and 12 mhz
Variable Oscillator Board: Lets you select CPU speeds up to 12 mhz
using DIP switches. Are working on a software driver to allow
speed switching from the keyboard. Will hold an 80287 and drive
it at up to 12 mhz.
Fast RAM Board: Memory expansion up to 16 MBs with ports
Fast AT Motherboard Upgrade: Send in your AT motherboard and they will
fix it to enable operation at maximum speed.
80386 Motherboard For AT: Uses a variable oscillator to drive current
80386s at 14.5 mhz and future faster 80386s at 20 mhz CPU
speed.
Supposed to be available in March.
B.G. Micro (214-271-5546):
Low cost crystals with thin wire leads for making your own speed
switch.
Megahertz (801-355-8857):
Speed Switches
Fast Crystals
8 mhz 80287 Board
8 mhz 80287s
PHOENIX ROM: For AT, to replace the new AT ROM with the timing routine.
Floppy drive timeout and keyboard speedup patches
Microprocessors Unlimited (918-267-4961):
100ns 256K RAMs
MicroWay (617-746-7341):
8 mhz 80287 Board
8 mhz 80287s
Ronsco Systems (312-534-1303):
5-Speed Speed Switch
-------------------------------------------------------------------------------
Stephen Sisler
72366,3724
------------------------------
From: <@ge-crd.arpa:davidsen@kbsvax>
Subject: Cranium 10
To: <info-ibmpc@usc-isib>
Date: Wed, 22 Jan 86 13:06:14 est
From: Davidsen <davidsen@kbsvax.steinmetz>
Posted-Date: Wed, 22 Jan 86 13:06:14 est
To: info-ibmpc@usc-isib.arpa
Has anyone tried the "Crainum 10" compatible from 1 Computer Store? It
is supposedly running at 10MHz, runs DOS, pc/ix, and xenix, and costs
$3050 quantity one.
------------------------------
Date: Wed 22 Jan 86 10:03:47-PST
From: Steve Dennett <DENNETT@SRI-NIC.ARPA>
Subject: DeskView
To: info-ibmpc@USC-ISI.ARPA
Two postings have lauded Quarterdeck's "Deskview" over MS Windows. Would
either (or both) of these folks please post a review of "Deskview" expanding
on it's capabilities (and perhaps contrasting them with Windows)? Thanks.
Steve Dennett
dennett@sri-nic.arpa
-------
------------------------------
End of Info-IBMPC Digest
************************
-------