home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Simtel MSDOS 1992 September
/
Simtel20_Sept92.cdr
/
msdos
/
info
/
si_173.arc
/
SI.TX3
< prev
Wrap
Text File
|
1989-07-08
|
10KB
|
204 lines
Documentation for System Information version 1.73.0
System Information (Hence referred to as SI) was written by
myself in an attempt to draw together the many and varied system
info programs that are around. I liked PC-TOOLS's CPU
identification and Peter Norton's SysInfo ROM identification
routines. I have seen some 20 or so of these utilities, written
in both BASIC, C, Assembler and Cobol! None however had ALL the
features that my program has. ---
SI has been tested on the original IBM PC, an XT clone and an
Amstrad 1640 w/EGA. It performed faultlesly on each occasion. SI
is auto-configuring, and it will correctly detect and use either
a MDA, Hercules, CGA, EGA, or VGA card. (NB: SI does not
currently fix snowing on early CGA screens, bit i'm working on
it. If you don't know what I mean, it does not effect you.)
You run SI simply by typing SI on the commandline. No switches
or input is required. The screen will clear, and a title screen
will pop up. If you run a DOS version lower than 3.0, SI will
abort with the appropriate error message. DOS 3.0 is needed for
the print spooler, CAS Comms and mouse test functions. Most users
will recognise most of the information provided, but CAS may need
to be explained a little further. My reference is Australian
Personal Computer Magazine, May 1989 pp 112. It is basicly an
extended communications protocol/standard for FAX and Modem
signals. Expect to hear more about it in the furure.
If you wish to view the Information screen, simply enter SI ?
and a screen containing my address, shareware and contribution
details are outlined.
The results can be written to the Screen, the Printer, a Disk
file or you can Quit. By using the write to disk option, you can
build up a library of results from various computers. I use this
as a list of comatible computers. It may be usefull if you go to
computer expo. You can run SI on each system, and then view the
results at heme and make up you mind based on a common referance.
PLANNED ENHANCEMENTS.
Whilst I have achieved what I wanted, I have kept on adding
more on as I discovered new things. New enhacemoents I either
have not discovered how to do, or have not had time to do
include:
* Detect presence of a
Real Time Clock --> XT and AT have different systems
for RTC's. Feedback would be
appreciated on this one.
* Determination of
Math Coprocessor type
8087, 80287, 80387. --> Now partialy fixed.
* Non destructive test
for use on a LAN system
(Present method hangs
system if LAN not present.)
* Output more information on
EGA settings, if EGA detected.
? Would anyone like more disk
statistics? This will only be
added if there is a huge demand
Like hundreds of requests, as I
have no access to a 3.5" drive,
AT 720k drives or a hard drive.
* At the suggestion of Kevin Lanton of
Dr Blaze BBS, I am now also looking
for a routine to find and report how
much EMS/LIM/Extended/Expanded mem
you have & It's version number.
Apparently there's several versions that
differ greatly??
PISSED OFF AT THE MAGAZINES SECTION.
This has nothing to do with SI, but rates a mention. The
correct abbreviation to use when commenting on floppy diskettes
is `DISK', NOT `DISC'. Disk is an abbreviation of Diskette, as
copyrighted by IBM some many years ago. Disc stands for nothing
at all. Some magazines insist on all articles being spelt this
way and it's WRONG!, SO DON'T DO IT!
Hmm. With that off my chest, back to SI.
TECHNICAL DEPARTMENT.
For those of you dying to know how, when, where, what and why
SI was written, it was written by Craig Hart of Merlinsoft Ltd as
an ongoing project over two years in starting mid 1987. SI is
coded in assembly language, and consists of eight source files of
65k total. SI is not yet finished and will continue to be updated
as the need arises. SI is assembled using A86 and debuged using
D86, both are excellent shareware programs available everywhere.
Yes, Eric I haven't registered yet, but I intend to as soon as I
get the money together and figure out how to get it to you!
I may make fragments of the code available to anyone who has a
good reason, but I will reserve the right not to publish the full
source code. There are too many pirates out there already without
encouraging more to hack in. I'd like to share the code, but with
software and viri around like they are today, I could not stand
the hassles.
FEEDBACK, COMMENTS SECTION.
As it stands, the relative CPU speed calculation is innacurate
to about +- 5%. This is due to interrupts and the like affecting
the critical timing. I had a lot of problems in this area, and I
would appreciate any suggestions on doing a more suitable
routine. I run an XT on 9.55MHz and it rates about 220%. On an
original PC (Yes, I found one to test on!), it rates exactly 100%
in 6 out of 10 runs. On an Amstrad 1640 w/EGA, it rates some
186%. SI now detects the 80386 chip. There was a bug in SI
(Version 1.72.5 or earlier) that caused the system to hang on
CPUID on an 80286. This serious error has now been fixed (To many
user's relief!). Thanks to Dr. Blaze BBS & Sysop Ron Lyth for the
help in this area.
The Machine maker/ROM BIOS maker section detects 1 version of
the Phoenix BIOS, the Amstrad 1512/1640, the Multitech 500+, the
original IBM, two types from Tandy and a 6 or 7 others. Here I
need user support. You can help by firing up Debug and typing:
-D F000:0 L FFFF (The `-' is Debug's prompt)
Most computers have a message at F000:E000 - look here first!
Look at the text on the right hand side. If you see a sentance
with your computer manufacturers name on it, pause the display
and write down (Even better: Print out) the address and the
message. (Don't worry about the Hex bytes in the middle) I
need the full number xxxx:yyyy on the left hand side, and the
exact text displayed. If there are carrage returns and line
feeds, note them as well. A submission shoud be like this:
F000:E85C "IBM ROM BIOS Version 2.03 (C) 1982 by IBM ",cr,lf,0
Any suggestions as to additions, bugs, or information on how
to determine what the machine is, please write me and I'l send
you the updated version for free! (360k technology media.)
Please feel free to append your own notes on the end of this
file. After responding, number the extention on the file one
higher than before. I.E. if you got SI.TX9, rename it to SI.T10.
Sorry, decimal numbers only Thank You.
USER RESPONSE
Name: Date: Response:
(Or Tag) DD-MM-YYYY
-----------------------------------------------------------------
Craig Hart 26-05-1989 This is the first release of the
Ver 1.6x.x doccumentation and represents the
Ver 1.71.x first marketable version of SI.
It is numbered 1.6x.x or 1.71.x
(Both are the same.)
-----------------------------------------------------------------
Craig Hart 01-06-1989 Created version 1.72.0 as the latest
Ver 1.72.0 version of SI. This version includes
the option to write ther output to
disk. A better error handler was also
introduced.
-----------------------------------------------------------------
Craig Hart 06-06-1989 Created version 1.72.5 as the latest
Ver 1.72.5 version of SI. This version corrected
a serious bug that locked the system
on an 80286 machine.
-----------------------------------------------------------------
Craig Hart 12-06-1989 Added detection of the 80386 chip
Ver 1.72.8 and removed about 1/2 a K of useless
tables from the source code.
I've deleted the now useless screen
width option, and added bus type
detection. It will detect the PS/2
Micro channel bus, or the PC/AT bus
(NB I don't guarantee this to run
on PS/2's anyway!)
Now on to Version 1.72.8.
-----------------------------------------------------------------
Craig Hart 24-06-1989 Added detection of NDP type.
Ver 1.73.0 Now SI detects 80387 or 8087/80287
(Can't figure out how to tell b/w
87 & 287!) Added pagination to screen
output routine.
Now on to version 1.73.0
-----------------------------------------------------------------
Kevin Lentin 01-07-1989 Here's the detection for an Olivetti
VER 1.73.0 M28 AT Clone. In memory at F000:6009
search for:
"COPYRIGHT (C) OLIVETTI 0/01/86",0,0,0
"Version 2.07",13,10,$FF,0
** My many thanks to K.L. for this one **
<Merlin>
-----------------------------------------------------------------
The next user should write in here.