The World of Computer Software/The World of Computer Software.iso/v93n00y.zip/V93N004.IBM

Internet Message Format | 1993-01-17 | 35KB

From:    WF02::IN%"[e-mail]" 16-JAN-1993 09:17:35.11
To:    James Gerber <[e-mail]>
CC:
Subj:    Info-IBMPC Digest V93 #6

Return-path: $$INFOPC
 <@WS5.CIS.TEMPLE.EDU:$$[e-mail]>
Received: from JNET-DAEMON by GRAD.CIS.TEMPLE.EDU; Sat, 16 Jan 93 08:05 EST
Received: From TEMPLEVM(MAILER) by TMPLCIS with Jnet id 4214 for
 GERBER@TMPLCIS; Sat, 16 Jan 93 08:05 EDT
Received: from TEMPLEVM.BITNET (NJE origin LISTSERV@TEMPLEVM) by VM.TEMPLE.EDU
 (LMail V1.1a/1.7e) with BSMTP id 6253; Sat, 16 Jan 1993 08:04:36 -0500
Date: Fri, 15 Jan 1993 02:15:39 GMT+1
From: Info-IBMPC Digest <[e-mail]>
Subject: Info-IBMPC Digest V93 #6
Sender: Info-IBMPC redistribution list <$$[e-mail]>
To: James Gerber <[e-mail]>
Reply-to: [e-mail]

Info-IBMPC Digest           Fri, 15 Jan 93       Volume 93 : Issue   6

Today's Editor:
         Gregory Hicks - Rota Spain <[e-mail]>

Today's Topics:
                      ATM fonts for OS/2 (2 msgs)
                    Combining NETBEUI and TCP/IP (?)
                   System Security and OS/2 (2 msgs)
                        Fixed point sqrt anyone?
                         Flush keyboard buffer
                    FORTRAN compiler needed quickly
                          Getting IRQ2 to work
                     Hot-key to kill WPS? (2 msgs)
                      How to disable WPS (3 msgs)
                    Microsoft Mouse Driver - SUMMARY
                       multiple DOS windows, how?
          Network drive identification/ MOUNT command for DOS
                   Null Modem Cable Dilusion (2 msgs)

Send Replies or notes for publication to: <[e-mail]>

Send requests of an administrative nature (addition to, deletion from
the distribution list, et al) to: <[e-mail]>

Addition and Deletion requests for UK readers should be sent to:
<[e-mail]>

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: 31 Dec 92 20:35:24 GMT
From: Mike Levis <[e-mail]>
Subject: ATM fonts for OS/2

 [e-mail] (Qazi Ahmad) writes:

>  I just d/l a ATM font file from HOBBES and really have no idea how to
>use the font! I do not know how to install it in OS/2. ANy ideas?

1.  unzip the file
2.  double-click on the ``OS/2 System'' folder
3.  double-click on the ``System Setup'' folder
4.  double-click on the ``Font Palette'' program-object
5.  click on the ``Edit font...'' button
6.  click on the ``Add...'' button
7.  type the name of the directory where the fonts are
8.  follow rest of instructions, etc...

You might want to read about font installation.  Double-click on
``Master Help Index'' and look at the ``fonts (screen)'' topics.

>Ahmad
>[e-mail]
--
======= Mike Levis   [e-mail] =======
::  ftp ftp-os2.nmsu.edu (128.123.35.151) for OS/2  ::  .--.
::  software & information.   Get /pub/os2/00Index  :: (OS/2)
::::::::  & /pub/os2/all/info/faq/faq20h.txt  ::::::::  ~--~

------------------------------

Date: 5 Jan 93 20:24:37 GMT
From: David Charlap <[e-mail]>
Subject: ATM fonts for OS/2

One bug that was fixed with the SP (XR 6055), that i don't think was
mentioned after all this time: Fonts converted with PFM2AFM work with
WPS now.  They didn't under GA.

I now have some menus and windows opening in "Viking 12" and "Mystical
14" with no problem.
--
  |)  David Charlap          | .signature confiscated by FBI due to
 /|_  [e-mail] | an ongoing investigation into the
((|,)                        | source of these .signature virusses
 ~|~

------------------------------

Date: 5 Jan 93 10:07:12 GMT
From: Frank Maas <[e-mail]>
Subject: Combining NETBEUI and TCP/IP (?)
Keywords: tcp/ip, tcp, ip, netbeui, pcnfs, nfs, lanman, Lan Manager
Summary: Does a driver with both protocols exist?

First of all, I know this is probably not the correct group for this
question, but I really didn't find a group that suited the subject. And
since it is about a MS-DOS driver...


Problem: Currently I'm using a LAN which is connected to both an Unix
server and an OS/2 server. The first can be reached when I load PC-NFS
on my machine. To reach the latter, I have to load NETBEUI (Lan
Manager).  Since I have to access both servers (Unix for E-Mail,
connections to the world :-), etc; OS/2 for NOTES, Programs, etc), I
have to reboot my machine each time I want to do something on the other
server. Concurrent access is not possible.

What I want: A driver that implements both the NETBEUI and the PC-NFS
layer. So, NOTES (which has to use NETBEUI) sees NETBEUI, and telnet
sees PC-NFS (or TCP/IP to be correct).  MS Windows should see PC-NFS
when MSW is set up for PC-NFS and it should see NETBEUI when it is set
up for NETBEUI.

So the question(s): Does such a driver exist?  Did you have the same
problem and solved it differently?  Do you see other solutions?

You can follow-up to this article, but if this is the wrong place, you
could also mail me.

With Regards (and Best Whishes for '93),

Frank Maas (e-mail: [e-mail])

------------------------------

Date: 30 Dec 92 06:41:10 GMT
From: CrazyMan <[e-mail]>
Subject: System Security and OS/2

[e-mail] wrote:
:  [e-mail] (Kyler Laird) writes:
: > [e-mail] (larry.a.shurr) writes:

: > I'd sure love some simple security that would keep people from deleting
: > files like config.sys.  I thought that something like that was available
: > for OS/2, but I can't recall.
:
: Hmmm, I must be simpleminded, but setting permission to read-only
: should keep the casual user from doing to much harm.

: bob - [e-mail]

That will work.  Also, don't forget about the ALT-F1 trick when booting
your computer.  That will copy the CONFIG.SYS, and .INI files from the
/OS2/INSTALL directory back to the proper place.  So, if someone
deleted CONFIG.SYS you just ALT-F1 when you boot up, and it's back
again.

Here's the actual quote from the OS/2 README file:

Warning:  This procedure removes any desktop customization done since
installing the operating system.

1.  Press Ctrl+Alt+Del to restart the computer.

2.  Before the OS/2 Logo panel appears, hold down Alt+F1 for 20
seconds.

The current versions of the CONFIG.SYS, OS2.INI, and OS2SYS.INI files
are renamed and replaced by the default installation versions in the
\OS2\INSTALL subdirectory.  If you have a STARTUP.CMD file, it also is
renamed.  This deactivates the startup feature.

I suppose you could replace the relevant files in these directories
with later versions, so that once you have the system running
correctly, you could go back to that version of the system, instead of
the default installation.

- Kevin Lowey ([e-mail])
>>>>> Anonymous FTP to FTP.USASK.CA for DOS, OS/2, and Windows programs <<<<<

------------------------------

Date: 30 Dec 92 19:37:51 GMT
From: Curtis Roelle <roelle@uars_mag.jhuapl.edu>
Subject: System Security and OS/2

[e-mail] (CrazyMan) writes:

>[e-mail] wrote:
>:  [e-mail] (Kyler Laird) writes:
>: > [e-mail] (larry.a.shurr) writes:
>: >
>: > I'd sure love some simple security that would keep people from deleting
>: > files like config.sys.  I thought that something like that was available
>: > for OS/2, but I can't recall.
>: >
>: > --kyler
>:
>: Hmmm, I must be simpleminded, but setting permission to read-only
>: should keep the casual user from doing to much harm.

>: bob - [e-mail]

>That will work.  Also, don't forget about the ALT-F1 trick when booting your
>computer. ...

I expect that technique is how the "uninstallation" of the Service Pack
changes are implemented (to return your system to its prior status
before installing the service pack).  The following quote is from the
"OS/2 2.0 Service Pak (10/92) - README" ...

          Many customers are aware of the ALT+F1
          key sequence to restore a desktop to its
          original state. After installing this
          Service Pak, ALT+F1 will return your
          system to its state as it was
          immediately before the Service Pak was
          installed.

Curt [e-mail]

------------------------------

Date: 6 Jan 93 10:44:01 GMT
From: Eyal Lebedinsky <[e-mail]>
Subject: Fixed point sqrt anyone?

 [e-mail] (Adam Billyard) writes:

>[e-mail] (the Dodger) writes:

>> Well? Does anyone know of a quick and dirty way to get a reasonably
>>accurate square root of a fixed point integer number? My numbers are
>>stored in a 16.16

>It depends on what "reasonably accurate" means.  I used a
>Newton-Raphson approximation in a 3D pool game I did a while back, and
>I got enough accuracy and quickly enough for my needs..

>       Adam
>
>NB If you're looking to do vector normalization then you probably need
>quite a lot of accuracy.

Here is how I do it. This code is from a flight simulator. I use 16 bit
ints but need to take sqrt of 32bit numbers (the result of squaring the
ints).

The program uses a table for the starting value. In practice I write
the tables to a file which I later include into the source.

#define FASTCALL _fastcall              /* microsoft style */

typedef unsigned char   Uchar;
typedef unsigned int    Uint;
typedef unsigned short  Ushort;
typedef unsigned long   Ulong;

static Uint     far sqrtab0[256] = {0};
static Uint     far sqrtab1[256] = {0};
static Uint     far sqrtab2[256] = {0};
static Uint     far sqrtab3[256] = {0};

static int far
lsqrt (Ulong x)                         /* used for initialization only */
{
        long    e;
        Ulong   r, t;

        if (x & 0xffff0000L)
                r = 662 + x / 17916;
        else if (x & 0x0000ff00L)
                r = 3 + x / 70;
        else
                r = 2 + x / 11;

        do {
                t = x / r;
                e = (long)(r - t) / 2;
                r = (r + t) / 2;
        } while (e);
        return ((int)r);
}

static void far
set_lsqrt (void)                        /* initialise tables */
{
        int     i;

        for (i = 0; i < 256; ++i) {
                sqrtab3[i] = lsqrt (i*256L*256L*256L);
                sqrtab2[i] = sqrtab3[i] >> 4;
                sqrtab1[i] = sqrtab2[i] >> 4;
                sqrtab0[i] = sqrtab1[i] >> 4;
        }
}

static Uint far FASTCALL
my_sqrt (Ulong x)
{
        register Uint   r, t;
        register int    e, ind;

        ind = 0;
        if (x & 0xff000000L) {
                if (x & 0xc0000000L) {
                        ind = 1;
                        x /= 4;
                }
                r = sqrtab3[x>>24];
        } else if (x & 0x7fff0000L)
                r = sqrtab2[x>>16];
        else if (x & 0x007fff00L)
                r = sqrtab1[x>>8];
        else
                return (sqrtab0[x]);

        do {
                t = (Uint)(x / r);
                e = (int)(r - t) / 2;
                r -= e;
        } while (e);

        if (ind)
                return (t+r);
        else
                return ((t+r)/2);
}

main ()
{
        set_lsqrt ();

        return (0);
}

------------------------------

Date: 1 Jan 93 18:25:02 GMT
From: "Jon K. Salmon" <[e-mail]>
Subject: Flush keyboard buffer


Lee,

> I have installed my own ISR to chain the INT16h ROM BIOS keyboard driver.
> Therefore the keyboard buffer must be cleared before returning to the
> interrupted program, otherwise the keypress remains in the buffer and is
> read with the next buffer read.

I would suggest either

               Mov    Ax,40h                ;
               Mov    Ds,Ax                 ;  Ds -> BIOS data area
               Cli                          ; Disable interrupts.
               Mov    Ax,[1Ah]              ; Set head and tail pointers
               Mov    [1Ch],Ax              ;  equal.
               Sti                          ; Enable interrupts.

or

    @@_Check:
               Mov    Ah,01h                ; Function = Check keyboard
               Int    16h                   ; Call BIOS.
               Jz     @@_Done               ; Exit if nothing in buffer.
               Xor    Ah,Ah                 ; Function = Read keyboard
               Int    16h                   ; Call BIOS.
               Jmp    @@_Check              ; Jump to check for keypress.
    @@_Done:

The first method clears the buffer at a low level by setting the head
and tail pointers of the circular queue equal, effectively emptying the
queue.

The second method simply uses the BIOS routines to check for a
keypress, and, if one is found, to read (and discard) it.  The routine
just keeps reading and discarding keys until there are none left in the
buffer.  With this approach you may wish to consider using the extended
keyboard functions, too (11h and 10h, respectively).  Also, you may
have to make pseudo-calls to Int 16h to bypass your own Int 16h
handler.

               Pushf                        ; Simulate interrupt.
               Call   [OldInt16h]           ; Call previous handler.

Anyway, I would recommend the second approach.  Why?  If some other TSR
has taken over Int 16h as a means of stuffing the buffer, setting the
low-level queue to empty will NOT empty any buffer this other TSR
manages internally, whereas the Int 16h method will.

-- Jon

------------------------------

Date: 31 Dec 92 15:48:03 GMT
From: Dave Tutelman <[e-mail]>
Subject: FORTRAN compiler needed quickly
Summary: The net comes through again

My thanks to those who helped me solve my problem, and it has been
solved.  A couple of days ago, I wrote:

>I need to borrow, rent, or buy (cheaply) a FORTRAN compiler (or
>interpreter) for MSDOS.  Can be VERY simple, as it's for a project for
>my son's beginning FORTRAN course in High School.  (He accidentally
>blew away a week or two of work from his BACKUP floppy [arrgh!] in
>class, and wants to reconstruct it on his PC at home during the Xmas
>break.)  I suspect this is a one-time, couple-of-day need (as the
>course ends in January, and the next course is PASCAL).

I received quite a few helpful suggestions.  As it turned out, the
majority of the suggestions cited a freeware package (BCF77 ver 1.3b),
available at most of the major archive sites. Definitely at WU and
SIMTEL.  Check the PD MS-DOS archives, directory "fortran", for the
file BCF7713B.ZIP, if you want a copy; its about a quarter meg of
download.

Anyway, I downloaded it, and it seems to be just what the doctor ordered,
or close enough anyway.
 - Seems to run on anything Intel from my 386-40 to my son's 8086 box.
 - Compiles and links Fortran-77 pretty quickly and hassle-free.
 - Includes an inconvenient but workable debugger.
 - The docs (originally in German) have been augmented by an English
   translation.
 - The progress and error messages from the compiler, linker, and debugger
   are in German.  While this will prove a challenge to my son, his
   background in languages should allow him to handle it with some effort,
   and will probably be a good experience in the long run.

Thanks to
        Khalid Sarwar Warraich ([e-mail])
        Darin Latimer ([e-mail])
        Rick Hoffbeck ([e-mail])
        Hannes Blankenstein ([e-mail])
        Bob Hood ([e-mail])
for pointing me in the right direction, and especially to
        Andre Koestli of Stuttgart
for creating the package and making it freely available.

Thanks for other suggestions to:
 - Joi Ellis ([e-mail]) for suggesting Lahey's Personal Fortran,
   available mail order for $89.
 - Doug Sewell ([e-mail]) for pointing out that Microsoft's Fortran
   compiler is currently on sale for $149.
 - Steve Hovater ([e-mail]) for offering to lend me his Fortran while
   I needed it.

Thanks for the help, everyone.

Dave

------------------------------

Date: 30 Dec 92 22:19:36 GMT
From: ochealth <[e-mail]>
Subject: Getting IRQ2 to work

 [e-mail] writes:
>
>I have one more question though.  It was explained to me a while back that
>IRQ3 would be answered first if IRQ3 and IRQ4 both had data coming in, and
>to solve a problem where moving the mouse would abort a download because
>the mouse was on IRQ3, and the mode on IRQ4.
>
>Now, if my COM3 is on IRQ2, redirected as IRQ9, then will its interupt
>be taken first because IRQ2 comes before IRQ3, or will it be treated as IRQ9,
>coming after IRQ5 in the order of hierchy.

priority
IRQ0
IRQ1
IRQ2
        IRQ8
        IRQ9
        IRQ10
        IRQ11
        IRQ12
        IRQ13
        IRQ14
        IRQ15
IRQ3
IRQ4
IRQ5
IRQ6
IRQ7

It will be taken as IRQ9, and because of the AT setup, it will come
before IRQ3,4,5,6,7

[e-mail]

------------------------------

Date: 1 Jan 93 17:32:40 GMT
From: "Timothy F. Sipples" <[e-mail]>
Subject: Hot-key to kill WPS?

[e-mail] (Paul Prescod) writes:
>As far as I am concerned, IBM really, really, wasted Ctrl-Alt-Del.
>C-A-D is a hardware interrupt, so as long as the handler isn't
>corrupted, it should ALWAYS get the system's attnetion.  And I've never
>had a case when I couldn't reboot with ctrl-alt-del.  So why couldn't
>the C-A-D handler bring up a menu: "Kill WPS, Start cmd.exe session,
>attempt shutdown, Reboot."
>For those that REALLY wanted to fast reboot, (though it's not a good
>idea) they could hit ctrl-alt-del and del again.  It's not like it's a
>lot of extra hassle.

Actually, any keystroke is a hardware interrupt.  There's nothing
particularly special about CTRL-ALT-DEL.

You've hit upon what the result will probably be, namely that
CTRL-ALT-DEL will have to be struck twice, but its basic function will
be preserved.

From what I've been able to gather, IBM was severely annoyed that
Microsoft decided to redefine CTRL-ALT-DEL.  I suppose IBM's viewpoint
was that CTRL-ALT-DEL (on a single user system, anyway) should be
preserved as the "I've fallen and I can't get up" key, and do an
immediate, safe reboot (it does).  Other keystrokes (e.g. CTRL-ESC)
should be used for task switching and shutting down tasks.

However, for better or worse, Microsoft has redefined CTRL-ALT-DEL
(although they probably weren't the first).  You may have noticed that
the Service Pak introduces a warning message when you use CTRL-ALT-DEL
indicating that the system is rebooting.  I expect this will evolve to
something like, "Press CTRL-ALT-DEL to reboot or ESC to abort."  I
think that's the best solution -- CTRL-ALT-DEL is a bit too dangerous.

Reviewers have complained that OS/2 doesn't seem to offer the
CTRL-ALT-DEL facility in Windows, whereby it can shut down errant
applications (sometimes).  It does, in fact.  CTRL-ESC is extremely
robust -- this is a protected operating system, after all.  I suppose
IBM felt that an extra keystroke wasn't required to convey robustness
(whether it actually exists or not).

Timothy F. Sipples      | Read the OS/2 FAQ List 2.0h, available from
[e-mail] | 128.123.35.151, anonymous ftp, in /pub/os2/all/info
Dept. of Econ., Univ.   | /faq, or from [e-mail] (send "HELP")
of Chicago, 60637       | [Read the List, THEN post to ONE OS/2 newsgroup.]

------------------------------

Date: 3 Jan 93 17:46:59 GMT
From: "Timothy F. Sipples" <[e-mail]>
Subject: Hot-key to kill WPS?

[e-mail] (Tom J Parry) writes:
>> However, for better or worse, Microsoft has redefined CTRL-ALT-DEL
>> (although they probably weren't the first).
>Well, DesqView redefined CTRL-ALT-DEL quite a while ago. I found it
>EXTREMELY useful. It is redefined in such a way that you can still argue
>that its function has not changed - it reboots the machine. In the DV case
>(and WinEnh I presume) it reboots the _virtual_ machine, thus killing the
>app. DV allows CTRL-SHIFT-METABUCKEY-DEL (can't remember offhand) to reboot
>the physical machine.

Now just one question: how am I supposed to explain the concept of a
"virtual machine" to someone who hasn't used computers before, in ten
words or less?

Let's take the case of OS/2 here.  If I'm using the System Editor,
that's not a virtual machine.  If I'm using a seamless copy of
WordPerfect for Windows, that's a virtual machine (possibly with a copy
of 1-2-3 for Windows running alongside).  If I'm using DOS WordPerfect,
that's another virtual machine.  Now when exactly does this
CTRL-ALT-DEL take effect on the virtual machine?  You mean, it behaves
differently if I'm in the System Editor or DeScribe?  :-)  You mean, if
I try it in WordPerfect for Windows, 1-2-3 for Windows also comes down?
:-)

This sort of borders on a religious issue, and I did suggest the remedy
that I prefer now that we're in this mess (namely that, when you press
CTRL-ALT-DEL, you are given the opportunity to press it again to reboot
the system or hit ESC to back out).  However, we're supposed to be
trying to make computers easy (not harder or more confusing) to use,
and we've still got a long way to go.  It's real simple right now:
CTRL-ALT-DEL reboots the system, CTRL-ESC switches among running tasks
and open windows (and lets you close anything out there).  CTRL-ESC
DOES allow you to shut down a virtual machine -- the function is
already in there.  CTRL-ALT-DEL isn't going to do any better -- this is
a protected mode operating system, after all.  If CTRL-ESC doesn't
work, then IBM's programmers need to find out what the problem is and
correct it.  (ALT-F4 for tasks running in windows, DOS, Windows, or
OS/2, accomplishes the same thing, by the way.)

For ten years plus CTRL-ALT-DEL has meant "reboot the system."  Why
should that change (in standalone OS/2, anyway, which is single user)?
The only thing I think that needs to be changed is to make it idiot
proof -- you have to hit it twice to get the system to restart.

Timothy F. Sipples      | Read the OS/2 FAQ List 2.0h, available from
[e-mail] | 128.123.35.151, anonymous ftp, in /pub/os2/all/info
Dept. of Econ., Univ.   | /faq, or from [e-mail] (send "HELP")
of Chicago, 60637       | [Read the List, THEN post to ONE OS/2 newsgroup.]

------------------------------

Date: 30 Dec 92 14:45:54 GMT
From: Harry Hsiao <[e-mail]>
Subject: How to disable WPS

 Jeremy Mathers writes:
> [e-mail] (Kenneth A. Kahn) writes:
>...
>>>Is there any way to disable the OS/2 WPS when booting OS/2?
>...
>>In your CONFIG.SYS change the line that reads PROTSHELL=C:/OS2/PMSHELL.EXE
>>to PROTSHELL=C:/OS2/CMD.EXE;  be warned though that this environment will not
>>support any PM applications, only Full Screen.  Also there is only a single
>>screen group; i.e. you can't switch between different processes.
>
>I think the SHFTRUN solution is better.  How to do it is documented
>elsewhere in this group, so I won't go into it, but it allows you to
>boot quickly to a command prompt if you choose to.  I have found it very
>handy.

Kenneth's solution did work, however, I could not shutdown nor
Ctrl-Alt-Del after I disable WPS.

Does SHFTRUN provide a better solution? I missed the posting you
mentioned and I don't have access to the old postings. Would you please
repost it here?

Harry Hsiao

------------------------------

Date: 2 Jan 93 18:24:34 GMT
From: "Timothy F. Sipples" <[e-mail]>
Subject: How to disable WPS

 [e-mail] writes:
>Is there any way to disable the OS/2 WPS when booting OS/2?

Yes.  In your CONFIG.SYS, change the SET RUNWORKPLACE line in your
CONFIG.SYS to read:

SET RUNWORKPLACE=C:\OS2\CMD.EXE

changing the drive letter, if necessary.

See the OS/2 FAQ List for details.  At IBM you may obtain it internally
-- contact [e-mail] if you need instructions.

--
Timothy F. Sipples      | Read the OS/2 FAQ List 2.0h, available from
[e-mail] | 128.123.35.151, anonymous ftp, in /pub/os2/all/info
Dept. of Econ., Univ.   | /faq, or from [e-mail] (send "HELP")
of Chicago, 60637       | [Read the List, THEN post to ONE OS/2 newsgroup.]

------------------------------

Date: 2 Jan 93 23:03:27 GMT
From: "Kenneth A. Kahn" <[e-mail]>
Subject: How to disable WPS

> Timothy F. Sipples writes:
> [e-mail] writes:
>>Is there any way to disable the OS/2 WPS when booting OS/2?
>
>Yes.  In your CONFIG.SYS, change the SET RUNWORKPLACE line in your
>CONFIG.SYS to read:
>
>SET RUNWORKPLACE=C:\OS2\CMD.EXE
>
>changing the drive letter, if necessary.
>
>See the OS/2 FAQ List for details.  At IBM you may obtain it
>internally -- contact [e-mail] if you need instructions.
>

Tim,

Technically speaking, this does not disable the WPS, it is still
running the Desktop.  All the DLLs that support the WPS are still
active.  If the need is to totally disable the WPS, you need to use
PROTSHELL=C:\OS2\CMD.EXE; of course this will not support any PM
applications.

------------------------------

Date: 6 Jan 93 10:13:54 GMT
From: Brian D Milner <[e-mail]>
Subject: Microsoft Mouse Driver - SUMMARY

Microsoft Mouse driver v8.20 - command line switches - Summary

I recieved some very prompt replies to my post, all via email. I
include the text from a couple of them below.

It seems that the driver only supports the switch /Y - which is
explained below. The driver contains text stating:

                /i /z /c /b invalid

                Parameters /U /E /X no longer supported

Also the driver repeats all its instructions in 9 languages - English,
French, Dutch, German, Swedish, Finnish, Spanish, Portugese, and
Italian.

My warmest thanks to everyone who responded to my post.

Response #1 of 2
^^^^^^^^^^^^^^^^

If you find out, let me know.  Microsoft issues those mouse drivers with
just about everything, and doesn't include any documentation.  I've
figured out that
/Cx => on comx
/Sxx  => sensitivity from 10 to 100
/Hxx
/Vxx  => horizontal and vertical sensitivity, if you want different
values.

The strange thing is that the Microsoft driver sometimes (rarely) loads
and runs with my Chicony trackball, and sometimes it doesn't.  Also,
have you ever looked inside the driver?  It has instructions in about
20 different languages.  Strange.  I have no idea how to get them to
work.  Setting country= doens't do it.  Not that I really need my mouse
to send me error messages in Norewigan.

Response #2 of 2

There are no switches for the 8.20 ver of MOUSE.COM except for the "/Y"
one which means "no hardware cursor". This switch is needed on some
Video 7 and a few other graphics cards because the card uses a graphic
cursor (sprite) instead of the standard block cursor in DOS programs.
This messes up mouse support for DOS programs under Windows.

To control the mouse program make changes in the MOUSE.INI file,
normally contained in your C:\WINDOWS directory. This file controls the
mouse program.

|Brian Milner, The Computer Centre, Brunel University , West London, UK   |
+-------------------------------------------------------------------------+
|[e-mail] | To shut down Internet; Post to the newsgroups|
+--------------------------+       saying you think 'Lost in Space'       |
                           |       is better than 'Startrek - TOS'        |

------------------------------

Date: 2 Jan 93 06:05:57 GMT
From: [e-mail]
Subject: multiple DOS windows, how?

 [e-mail] (Jiansheng Zhao) writes:
>
> I have been using OS/2 for several months. There is still one
>question which I am afraid of asking: how can I open more than two DOS
>(or OS/2) windows?  Creating shadows all points to the same (ONE)
>session. Can anyone help?

> Jiansheng Zhao
> Dept. Chem. Eng, Univ. of B. C., Canada. V6T 1Z4
> [e-mail]

I'm sure you'll get a couple of response for this one.

Just go to the object settings. Go to WINDOW. Change the OBJECT OPEN
BEHAVIOR to CREATE NEW WINDOW instead of DISPLAY EXISTING WINDOW.

This question bugged me too when I first started using OS/2. Until
someone pointed this out to me.

------------------------------

Date: 31 Dec 92 17:16:01 GMT
From: Mark Buckaway <[e-mail]>
Subject: Network drive identification/ MOUNT command for DOS

UNIX sysadmins know about the MOUNT command to add file systems to
thier unix box. I would like to do something similar for DOS. That is
add a "file system" (ie. another drive) as a directory off my main
drive. DOS includes the JOIN command but this is incompatible with
network drives. I am running Netware Lite V1.1.

Some time back, I saw articles that were mentioning that it was
possible to change the "device descriptor" (please forgive me if this
is NOT the correct term) to make a network drive LOOK like a hard drive
to DOS. My idea encorporates writing a same utility to change the
network drive "into" a hard drive, spawn to the JOIN command, and
switch the drive back to a network drive. If it is possible to fool DOS
in this manner, this should work.

All reference material I have says nothing of how to determine if a
network drive is such and not a hard drive. Does anyone know who to do
this? Or have code segments either in assembly or C on how to poke
around memory for these items. If all else fails, anyone know of a good
reference book I can RTFM?

I am very knowledgible in C, know something about assembly. Any help
would be appreciated.

Mark

Club InSaNiTy BBS   |   BBS:[phone] |   [e-mail]
DataSoft Consulting | Voice:[phone] |   uunorth!cibbs!root
Mark Buckaway       | Toronto, Ont. CANADA  |   1@4653 (WWIVnet)

------------------------------

Date: 2 Jan 93 12:20:00 GMT
From: Dick Zeitlin <[e-mail]>
Subject: Null Modem Cable Dilusion

RB> . pins 2 and 3 were cross-connected.
  > . pin 4 on one was connected to (6 *and* 8) on the other, (& vice versa)
  > . pins 5 were connected.
  > . pins 1 7 and 9 were left unnconnected.
  > . Pin 7 is supposed to be signal ground, so to compensate I clipped
  > . a jumper from one metal chasis to the other.

There are many incarnations of null modem cables.  If you want a "max"
async null modem, do the following.

         25 Pin                                   9 Pin
        A       B                               A       B
       ___     ___                             ___     ___
        1       1       Chassis Ground
        2       3       SD  to  RD              2       3
        3       2       RD  to  SD              3       2
        4       5       RTS to  CTS             7       8
        5       4       CTS to  RTS             8       7
        6       8,20    DSR to  CD, DTR         6       1,4
        7       7       Signal Ground           5       5
        8,20    6       CD,DTR to DSR           1,4     6

If you have to choose between using the Signal Ground or Chassis Ground
on the 25 pin cable, use the Signal Ground.  But you should really use
both.

        D/

 . QMPro 1.0 41-3345 . Bring back A-N airways!!    [e-mail]

------------------------------

Date: 2 Jan 93 01:11:16 GMT
From: Rob Bairos <[e-mail]>
Subject: Null Modem Cable Dilusion Confusion

Hi, a couple of days ago, I posted my problem concerning file transfers
between my two PC's.  The following is the solution and my experiences
if anyone's who's interested.  Thanks for the responses!

After fiddling with mismatching connections for a couple of hours, I
salvaged two joystick cords from my old colecovision and gemini, (these
contain at least 8/9 possible wires in them).  I used the Radio Shack
null modem adapter as a guide.  The two cords were connected on a one
to one basis except for the following:


 pins 2 and 3 were cross-connected.
 pin 4 on one was connected to (6 *and* 8) on the other, (& vice versa)
 pins 5 were connected.
 pins 1 7 and 9 were left unnconnected.
 Pin 7 is supposed to be signal ground, so to compensate I clipped
 a jumper from one metal chasis to the other.

After several scotch-tape pieces later, I plugged them into the 9-pin
ports where the mice usually fit in.

I used PCplus and was about to give up cuz I almost didn't notice what
I typed on one keyboard showed up on the other screen. I set it to max
baud rate, threw in an extra stop bit (figured it would be more stable)
and disabled all carrier-detect options.  (don't know why carrier
detect wasn't detecting).  Anyways, it worked beautifully*.

*Well, until I disabled disk-caching software and *disabled* turbo
speed on the one PC, the max average cps was about 2700 using ZModem.
After I got it up to 9900 - 10,000cps with strange 10sec+ pauses
appearing spurratically.

That's my story.  There's better software out there for the actual
transfer.  I had to keep shelling-out, and creating new directories
manually.

One small strike against the RS-232 conspiracy!

Rob Bairos

------------------------------

End of Info-IBMPC Digest V93 #6
*******************************
-------