home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Cream of the Crop 1
/
Cream of the Crop 1.iso
/
PROGRAM
/
TP061492.ARJ
/
06-14-92.TPC
Wrap
Text File
|
1992-06-14
|
41KB
|
1,285 lines
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 01-01-00 00:00:00
From
To
Subject
--- WM v2.01/92-0100
* Origin: A.C.E. of Spades (615)383-4381 The B.A.N. board (1:116/33)
* Tossed by SFToss v1.00b on 92/04/09 12:01:00
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 01:04:18
From Trevor Carlsen
To Dj Murdoch
Subject Re: Tp 6 Exe Slow?
DM> This takes 25 seconds to execute, on a 486-33 under Desqview as the
DM> only active window.
This was what I considered a good opportunity to (very roughly) compare the
overhead of DV. On my similar machine the above took 6.919 seconds.
It would seem that DV imposes a significant performance penalty.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/07 13:19:56
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 00:21:27
From Trevor Carlsen
To Tony Nugent
Subject Type-ing it Right!!
TN> I have since got a reply (from my local Australian Pascal echo)
TN> that nails the problem, and it was a subtle error in using SET
TN> typing.
TN> What should be done in this case is to the following:
TN> type
TN> CharDigit = '0'..'9';
TN> Digits = SET OF CharDigit;
TN> Postcode : array[1..4] of Digits;
TN> { Postcode is a SET OF CharDigits!!! }
TN> after which the appropriate error code is generated.
TN> Very subtle difference, eh? I thought I had sets understood, but
TN> there are finer points to everything we learn!
TN> Of course, loading any data into this variable will, in "real
TN> life", need to be tested as belong to type Digits before it is
TN> put there.
I do not recall seeing that reply locally and do not agree with it. Doing
it that way will prevent Postcode being displayed and will double its size
to 8 bytes. Hardly what you would want.
I would forget about the typing and do your own range-check on the characters.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/07 13:19:56
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 00:56:25
From Trevor Carlsen
To BRIAN PAPE
Subject PC CLOCK BYTE
BP> On all of the previous posts about the PC clock byte at ($40:$6C)?,
BP> people have said that you can directly map that memory into a Longint
BP> variable. But, it seems to me that since the turbo Longint type is
BP> mapped like this:
BP> xxxx xxxx xxxx xxxx
BP> 1 2 3 4
BP> but in DOS, the dword is stored like this:
BP> xxxx xxxx xxxx xxxx
BP> 4 3 2 1
BP> you have to swap the bytes around to get the right value!
BP> I tried it by mapping a Longint, but it didn't work, so I swapped the
BP> bytes:
BP> clock := (swap(clock shr 16) + (swap(clock) shl 16)
BP> and this works... I would appreciate it if someone told me how they
BP> did it without swapping the bytes around...
There is no need to do any "swapping around". The 4 bytes at $40:$6c are a
standard longint type.
I do not understand what you mean by having to "swap the bytes around to get
the right value". A longint is constructed in the standard Intel architecture
style of -
LSB .. MSB.
So the following -
type
long = array[0..3] of byte;
var
demo : longint;
demoA: long absolute demo;
x : 0..3;
begin
demo := 1;
for x := 0 to 3 do
write(demoA[x]:4);
end.
will result in
1 0 0 0
To roughly calculate the hour of the day -
var
hour : word absolute $40:$6e;
and the word at $40:$6c will return the minutes if divided by 1092.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/07 13:19:56
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 09:09:18
From Dj Murdoch
To Geoffrey Anderson
Subject Re: inverse of a polynomial
GA> y = 2x^0 + 3x^1 ...
GA> I want to find g(y), that is, solve for x in terms of y,
GA> for any such polynomial.
There isn't necessarily a unique solution - a polynomial of degree D (i.e.
terms up to and including x^D) could have as many as D solutions. There
are exact formulas for the solutions up to D=4, but they're really only practica
up to D=2, so you need an approximation technique.
One is bisection: just guess at a couple of x values which bracket y, then
replace one end by the midpoint until you find something pretty close to g(y).
For odd D, the guess is easy, but it may not be so easy for even D.
Take a look in Numerical Recipes [in Pascal], by Press et al, for a good discuss
on and some code.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/08 12:11:05
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 09:16:08
From Dj Murdoch
To Frank Mccormick
Subject Re: Rewriting Output
FM> Hope to get some help here...
FM> When you subvert direct video in TP 4/5/6 by assigning
FM> Output to a null file and rewriting it, how do you get
FM> back to a normal situation ?
FM> I have written an ANSI based door which does the
FM> above...but then when the door returns to the BBS, the
FM> windows no longer work because Output has been redirected.
Use AssignCRT.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/08 12:11:05
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-08-92 01:59:24
From Trevor Carlsen
To David Vohwinkel
Subject Memory Management
DV> Does anyone know of a way to make a Pascal program remove
DV> itself from memory so that you can call an external program
DV> that is bigger than 60k wich is the limit I think... I want
DV> to write a Pascal program that 'shells out' to a seperate
DV> program that is say 250k...
There is no 60K limit on the EXECing of a child process. Check out your manual
or help system on EXEC.
To swap ITSELF out of memory before an exec is another (very complex) matter.
To demonstrate the process would be beyond the scope of a message. Suffice
it to say that it is not strictly possible. Some part of the original program
MUST remain present in RAM at all times.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/08 19:58:22
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-08-92 02:05:04
From Trevor Carlsen
To Daniel Walton
Subject File Pointer size
DW> I am writing a program that uses frequent access to the disk
DW> and for some reason Turbo Pascal 6 is assigning an integer
DW> value to the file pointer. Therefore when I have the
DW> Blockread size set to 1 byte I cannot read in more than 32K
DW> worth of data. So how do I set the file pointer to a long
DW> integer?
Without you demonstrating what you are doing it is impossible to determine
what is wrong. The help system demo shows how to use BlockRead adequately.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/08 19:58:22
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-08-92 08:03:23
From Dj Murdoch
To Trevor Carlsen
Subject Re: Tp 6 Exe Slow?
TC> This was what I considered a good opportunity to (very
TC> roughly) compare the overhead of DV. On my similar
TC> machine the above took 6.919 seconds.
TC> It would seem that DV imposes a significant performance penalty.
It usually seems to be less than 70% overhead. Perhaps I was wrong about
that window being the only active one. Hmmm....
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/09 17:09:58
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-08-92 08:16:26
From Dj Murdoch
To Trevor Carlsen
Subject Re: Tp 6 Exe Slow?
DM> This takes 25 seconds to execute, on a 486-33 under Desqview as the
DM> only active window.
TC> This was what I considered a good opportunity to (very
TC> roughly) compare the overhead of DV. On my similar
TC> machine the above took 6.919 seconds.
TC> It would seem that DV imposes a significant performance penalty.
Just redid the tests. On a barebones machine, or one with just QEMM loaded,
I get about 16 or 17 seconds. In a single Desqview window I get the same.
Running two Desqview windows, with one just sitting at the prompt, I get
25 seconds. (That must have been what I had the first time.) Running the
same program simultaneously in the two windows, I get 33 seconds.
So, it looks as though the Desqview overhead is pretty minimal - better than
I thought - provided you don't leave windows sitting idle.
I don't know why you got 7 seconds though - what sort of machine have you
got? Mine is a DTK, with supposedly a 256K cache. Norton's old SYSINFO gives
it a 51.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/09 17:09:58
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-08-92 08:21:14
From Dj Murdoch
To Daniel Walton
Subject Re: File Pointer size
DW> I am writing a program that uses frequent access to
DW> the disk and for
DW> some reason Turbo Pascal 6 is assigning an integer value
DW> to the file pointer.
DW> Therefore when I have the Blockread size set to 1 byte I
DW> cannot read in more
DW> than 32K worth of data. So how do I set the file pointer
DW> to a long integer?
You're probably not setting the block size when you use BlockRead. That's
one reason not to use it - it's a very easy mistake to make. (You need to
say Reset(filevar,1).) Use a TBufStream or a TDOSStream instead.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/09 17:09:58
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 21:32:16
From Mark Ouellet
To Jud Mccranie
Subject Re: Impact Of .Tpu Size O
On 03 Jun 92, you, Jud Mccranie, of 1:3645/10.0 wrote...
MO>>>> Remember using CRT installs video routines which
MO>>>> handle screen writes, those same routines you showed people
MO>>>> how to bypass to write ANSI screens.
MO>>
JM>>> That wasn't me.
JM>>>
MO>>>> So naturely, anything used must be kept by the
MO>>>> linker but try using other routines such as WRITE and
MO>>>> WRITELN and GOTOXY etc... you'll see they were discarded
MO>>>> when you first compiled without using them.
MO>>
MO>> Jud,
MO>> Allow me to refresh your memory ;-)
JM> No, you're still wrong. I NEVER wrote the paragraph above. Note that
JM> it is a quote of a quote. Someone wrote to someone else, I quoted, and
JM> it was left in there. You are assuming that the original was from me,
JM> someone replied. But actually it was person A to person B, then I
JM> replied to that message, and quoted some of it, and that was in there.
JM> I did make a comment that "uses crt" adds to the exe file, even if you
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
JM> don't use anything from it, but I *never* said what you said I said.
^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Jud,
That's funny because what I said you said was
exactly that: "That you said the CRT adds to the size
eventhough you don't use anything".
And I was trying to explain to you, since from a
previous message in which you said you didn't know WHY the
crt unit added to the .exe size even if you used nothing.
If you had just read the message and taken it for
what it was, an explanation, instead of jumping on your
paranoia wagon again, we wouldn't have to clutter this echo
with these messages.
If you have anything more to say on the subject, I
will only reply to Netmail from now on.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:01
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 21:44:13
From Mark Ouellet
To Sean Ocker
Subject Re: SBlaster...
On 02 Jun 92, you, Sean Ocker, of 1:117/369.0 wrote...
SO> Call 1:117/349. It is a two-line BBS with 3.8+ gigs. It has 4 CD ROMS
SO> and one is the Magnum Sights and Sounds. FREQ's are not allowed off it
SO> (CD's are too slow) but they have a 14.4 on one of the lines.
SO>
SO> My address is 1:117/369. Sorry about that.
Sean,
Thanks, I might take a look but I prefer Freq'ing as
logging on as a human caller often makes the difference
between spending a few bucks for a file and a large LD bill.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:01
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 21:46:34
From Mark Ouellet
To David Masaki
Subject Re: Compressing Binary Sorted Trees
On 02 Jun 92, you, David Masaki, of 1:345/27.0 wrote...
DM> You're right in that what I described earlier would produce a binary
DM> sorted tree with all it's nodes on one side and produce the exact
DM> opposite effect of what I wanted. What I want is for the tree after a
DM> bunch of insertions which would make it less efficient to be compressed
DM> (made into it's most efficient form). How would that be done?
Well David,
as I stated in the original message, AVL or Balanced
Binary Trees do balance the tree on each insertion so as to
insure a minimum search path for any node.
I think many B+Tree libraries are available, Borland
used to make one ie: Turbo Database Toolbox, which is not
sold anymore I think. Last version was for TP 4.0 I think
but might be compatible with newer versions of TP.
Turbo Power also produces a database library which
includes B+Tree indexes I think.
Other than that you would have to find a good book
on AVL or B+Trees, and if you do let me know as I have been
searching for a good one too.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:01
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 21:51:17
From Mark Ouellet
To Kevin Higgins
Subject Re: Looking for SHAZAM (TV App. code gen
On 03 Jun 92, you, Kevin Higgins, of 1:128/74.0 wrote...
KH> Ah, another person who spent time uploading there, to build a decent
KH> file ratio, then got cut off when Turbo City decided to only grant
KH> access to those who would cough up money, eh?
Kevin,
Wrong assertion, I simply found the file there after
some one recomended their library but since I couldn't freq
form it I was simply trying to find another source.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:02
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 21:54:04
From Mark Ouellet
To David Rye
Subject Re: Pascal source
On 26 May 92, you, David Rye, of 1:240/1.4 wrote...
DR> If it weren't for Turbo City I never would have been able to keep up
DR> to date of various techniques and tips that have kept me employed as a
DR> programmer.
Yes they do have some very nice stuff. The zone 2
Turbo City is not bad either. Lots of SB/Adlib stuff along
with TP stuff.
DR> I think Pam requires a password for freqs, you might want to drop
DR> her some netmail and ask about that though.
Well it seems from the what I received with FILES
that you must be a registered member to even Freq.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:02
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-07-92 22:12:26
From Mark Ouellet
To Ryan Brown
Subject Re: e-mail
On 01 Jun 92, you, Ryan Brown, of 1:203/674.0 wrote...
RB> What is pegasus?
It's an e-mail system for Novell Networks. It also
allows file attaches between stations.
And it's FREEware.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/09 17:10:02
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 02:34:54
From Trevor Carlsen
To Peter Beeftink
Subject Writing directly to disk sectors.
PB>... How do I truncate to "the old length" though?
Look up truncate in the manual.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:44
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 03:50:20
From Trevor Carlsen
To Peter Beeftink
Subject getftime()
PB> As I found out the hard way, before you can successfully use
PB> getftime, you have to open (reset();) the file in question.
PB> Now, is it possible to read the time_of_creation of a
PB> directory(-file)????
Just use FindFirst/FindNext with the appropriate mask. The creation time
can be determined from the timestamp.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 03:00:48
From Trevor Carlsen
To Frank Mccormick
Subject Rewriting Output
FM> When you subvert direct video in TP 4/5/6 by assigning
FM> Output to a null file and rewriting it, how do you get back
FM> to a normal situation
This is covered in the manual... but not obviously! AssignCRT is the procedure
you are looking for. So to return stdout to the normal crt setup -
assignCRT(output); rewrite(output);
(Chapter 15 - The CRT unit. p 199)
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 03:53:08
From Trevor Carlsen
To Tony Nugent
Subject MCGA Tutorial #5
TN> Hey, great stuff. Should be more of this type of thing.
TN> However, I did not get all the code as it got cut out either
TN> in transit through the systems (I'm in Brisbane Aust), or by
TN> my offline reader...
TN> ... While appreciated by people such as myself (don't know
TN> about those that zap this stuff log distance around the
TN> world :-), your longer messages need to be cut shorter and
TN> posted as several smaller parts. It does look like
TN> interesting code, but it's not complete for me :-(
The message will be complete on the system. (It was complete here) Perhaps
you could get it by a direct message read rather than by your reader.
I have had several requests from echo users to enforce this echo's 150 line
message limit. I don't intend taking a strong stance on that, as the full
text of the messages are almost always available on the main board. It is
(or certainly should be) only the crippled readers that are doing the truncating
In fact, in the majority of cases the complete text is probably available
in your own packet.
Even so, it would be considered "good form" if all those posting long messages
would do everybody the courtesy of abiding by the echo rule of no messgaes
over 150 lines. For those whose readers truncate at 100 lines - tough. Get
a decent reader! :-)
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 03:57:52
From Trevor Carlsen
To Patrick Craig
Subject Lots of "thanks"
PC> Though the lingo is a bit over the top of my head, I do appreciate
PC> the information. Thanks!
Patrick, I don't normally worry too much about "thank-you" type messages but
this echo does have a rule on this (see excerpt below). The reason I point
this out to you is that there were ten such messages from you today! Your
sysop should have the complete text of the echo rules available. They are
posted monthly.
8. No "thank-you" or "no content" or "rubbish" messages. Sysops spend a
great deal of time and money to enable the distribution of echoes
such as this. Please respect this and avoid messages such as "Thank
you. Just what I needed" or "I agree" etc. By observing this
etiquette you will be helping to ensure greater participation in the
future.
Thanks. (No need to reply to this.)
Trevor Carlsen
Moderator.
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 07:59:50
From Trevor Carlsen
To Greg Williams
Subject Memory Management
TC> To swap ITSELF out of memory before an exec is another
TC> (very complex) matter. To demonstrate the process
TC> would be beyond the scope of a message. Suffice it to
TC> say that it is not strictly possible. Some part of
TC> the original program MUST remain present in RAM at all
TC> times.
GW> Hmm.... the only reason we tend to leave only the "reload" fragment of
GW> the program in memory (while swapping the rest out) is to do just that
GW> (reload, when the child process is finished). However, would it not
GW> be possible to do something similar to what the UNIX fork()and exec()
GW> command combination does, and totally overwrite the parent process
GW> with the new process?
If the process was being managed by the OS then there should not be great
difficulty. When the process being swapped out is doing the swapping, I cannot
see how there can be an alternative to retaining some of that process in RAM.
I have Turbo Power's "PopDos" loaded as my first TSR. I can "pop" to DOS
with 620K of free memory from any program if the TSR determines that it is
safe to do so. This means that an application that is running is swapped
out in its entirety but the swapper is the TSR and NOT the application.
As a matter of interest, PopDos has been a Godsend for me during my program
development work. If I have a lockup, I just pop to dos, save the contents
of my Ramdrives and then reboot. Without it my ramdrive stuff may be lost.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 09:04:32
From Trevor Carlsen
To Jud Mccranie
Subject Re: Tp 6 Exe Slow?
JM> P.S. I'm uploading it here, ANAGRTST.ZIP, 3K, simplified source code,
JM> sample data file. TP 6.0 (with G+ even) takes about 8% longer to
JM> execute than TP 5.5.
Mmmmm... Looks like your sysop expects a password. After the Yoohoo/2U2 handsha
e I was summarily dismissed.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 09:07:57
From Trevor Carlsen
To Justin Henry
Subject Removing 32 Bytes from File
JH> How would I go about removing the first 32 bytes from an untyped
JH> file? Read 1 byte 32 times, then read and write the rest? Would
JH> someone suggest an easier way? By the way, I am trying to remove
JH> the 32-Byte Header from a Sound Blaster .VOC file.
Declare the file as untyped with a record size of 1. Seek to offset 32 and
read however many bytes you want into an appropriately sized and declared
buffer.
var
f : file;
buffer : array[1..32768] of byte;
result : word;
begin
assign(f,filename);
reset(f,1);
seek(f,32);
BlockRead(f,buffer,sizeof(buffer),result);
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 09:50:48
From Trevor Carlsen
To Rich Veraa
Subject Your problem
I cannot locate your exact problem but it appears to be an out-of-range error
on the variable index. Check that out carefully.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/09 19:32:45
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 20:53:00
From Norbert Igl
To Peter Beeftink
Subject getftime()
> Hello All!
> As I found out the hard way, before you can successfully use
> getftime, you have to open (reset();) the file in question. Now, is
> it possible to read the time_of_creation of a directory(-file)????
Shure i do....
-------------------------8<---------------------------------
uses dos;
[...]
function GetDirTime(Name:PathStr):longint;
Var Sr : SearchRec;
begin
findfirst(Name, DIRECTORY, Sr);
while ( DosError = 0 ) and ((Sr.Attr and DIRECTORY) <> 0)
do FindNext(Sr);
If DosError = 0
then GetDirTime := Sr.Time
else GetDirTime := -1;
end;
[...]
Var Dt : DateTime;
L : LongInt;
begin
L:= GetDirTime('C:\WINDOWS');
if L <> -1 then
begin
UnpackTime(L, Dt);
with Dt do
writeln('C:\WINDOWS was created at ',Day,'.',month,'.',Year,' - '
,Hour:2,':',Min:2,':',Sec:2 );
end else
writeln('C:\WINDOWS not found....');
end.
------------------------->8---------------------------------
sorry for the typos, if any...
Bye from Germany, Norbert
--- FD 2.02/FastEcho
* Origin: STOP READING! You're leaving the MSG-sector (2:241/5300.3)
* Tossed by SFToss v1.00b on 92/06/10 07:40:08
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 07:50:20
From Dj Murdoch
To Justin Henry
Subject Re: Removing 32 Bytes from File
JH> How would I go about removing the first 32 bytes from an untyped
JH> file? Read 1 byte 32 times, then read and write the rest? Would
JH> someone suggest an easier way? By the way, I am trying to remove
JH> the 32-Byte Header from a Sound Blaster .VOC file.
If you're using TP 6.0, the easiest way is using streams:
uses
objects; var
infile, outfile : TDOSStream; begin
infile.init('inputfilename',stOpenRead);
infile.seek(32); { remember byte 32 is the 33rd byte }
outfile.init('outputfilename',stCreate);
outfile.copyfrom(infile, infile.Getsize-32);
infile.done;
outfile.done; end.
The same thing can be done as you suggest, but it'll be much slower. It can
be done with about equal speed using BlockRead/BlockWrite, but won't be as
simple: you'll have to set up a buffer, and loop until you've copied the
whole file.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/10 10:56:01
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 07:55:03
From Dj Murdoch
To Jud Mccranie
Subject Re: Topics..
JM> Sorry, I didn't realize that. I thought that was the reason for the
JM> "lessons", to lessen (no pun intended) the traffic in elementry
JM> questions here. As you may have noticed, however, I'm often quick to
JM> ---
JM> * Origin: Business Connection 1.2Gigs (912)247-6977 (1:3645/10)
A lot of your messages have had the last line (or more?) chopped off like
this. I've noticed that some mail editors mess up if the message you enter
doesn't have a blank line at the end - perhaps you should check yours out.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/10 10:56:02
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-11-92 09:10:27
From Trevor Carlsen
To Max Maischein
Subject File Pointer size
> I am writing a program that uses frequent access to the
> disk and for
> some reason Turbo Pascal 6 is assigning an integer value to the
> file pointer.
> Therefore when I have the Blockread size set to 1 byte I cannot
> read in more
> than 32K worth of data. So how do I set the file pointer to a
> long integer?
MM> This is the limit of DOS ;-)
No it is not... the file pointer is a longint and bears no relationship to
the number of bytes that can be read at a time. The most that can be read
is 64K not 32K. Without seeing what is being done, we can only surmise.
Looks to me like an integer type is being used instead of a word type.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/11 11:49:41
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-11-92 09:13:34
From Trevor Carlsen
To John Black
Subject Getting the current directory string
JB> ... I can't find a way to get what
JB> the current directory is...
Look up GetDir in the manual.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/11 11:49:41
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-11-92 09:15:16
From Trevor Carlsen
To Timothy Lang
Subject Making TSR Programs
TL> I know how to make normal programs in TP, but how can I make
TL> TSR programs? I have heard that I need to make memory
TL> directives and I would need some type of HOT KEY mechanism
TL> and so on... What I want to do is take a flight manager
TL> program I am working on and make it so that it loads and
TL> then is activated at any time by hot key. It does not have
TL> to do anything when it is not active.
TSRs are both complex and tricky. Professionals usually prefer to rely on
rock-solid, fully debugged commercial toolkits to do this. Best I have seen
is "TSRs Made Easy" by Turbo Power. This is a subset of their major toolbox,
Object Professional.
Ross Wentworth's shareware TSR library (forgotten its name and I have not
got it) is another that seems to get good reviews.
If you particularly want to do this yourself then you have months of reading
books and experimentation ahead of you if it is to be done properly.
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/11 11:49:41
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-11-92 09:55:49
From Trevor Carlsen
To Scott Munger
Subject XXX
SM> The program needs a password to continue its execution. I
SM> want it to (instead of displaying the password) replace the
SM> password with a x everytime someone types it in until they
SM> press enter. Just like your personal password when you log
SM> on to a bbs. Any help would be great.
Untested.
function Password(prompt: string): string;
const
BSpace= #8;
CR = #13;
space = #32;
var
st : string;
len : byte absolute st;
done : boolean;
col,
row : byte;
begin
done := false; len := 0;
write(prompt);
col := WhereX; row := WhereY;
repeat
st[succ(len)] := ReadKey;
if st[succ(len)] = #0 then
st[succ(len)] := ReadKey { discard extended keys }
else case UpCase(st[succ(len)]) of
BSpace : if len > 0 then begin
dec(len);
gotoXY(col+len,row);
write(space);
gotoXY(col+len,row);
end;
CR :done := true;
'A'..'Z': begin
write('*'); { display * instead of character entered }
inc(len);
end;
end; { case }
until done;
Password := st;
writeln;
end; { Password }
I hope that there are no serious logic flaws or obvious errors there! :-)
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/11 11:49:41
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-11-92 12:11:23
From Trevor Carlsen
To Dj Murdoch
Subject Re: Tp 6 Exe Slow?
TC> This was what I considered a good opportunity to (very
TC> roughly) compare the overhead of DV. On my similar
TC> machine the above took 6.919 seconds.
TC> It would seem that DV imposes a significant performance penalty.
DM> Just redid the tests. On a barebones machine, or one with
DM> just QEMM loaded, I get about 16 or 17 seconds. In a single
DM> Desqview window I get the same. Running two Desqview
DM> windows, with one just sitting at the prompt, I get 25
DM> seconds. (That must have been what I had the first time.)
DM> Running the same program simultaneously in the two windows,
DM> I get 33 seconds.
DM> So, it looks as though the Desqview overhead is pretty
DM> minimal - better than I thought - provided you don't leave
DM> windows sitting idle.
DM> I don't know why you got 7 seconds though - what sort of
DM> machine have you got? Mine is a DTK, with supposedly a 256K
DM> cache. Norton's old SYSINFO gives it a 51.
Mine is an Australian built 486/33 with 128K cache running under DrDOS 6.0
and 4DOS.
However :-) I cannot let you go on believing a part-truth! I cheated!
I *did* use your program code with the only alteration being the adding of
a timer unit. But I used very different compiler directives to you! With
your compiler directives it took 15.213 seconds! (You may recall that you
had at least R+ I+ S+) It makes a huge difference when they are changed.
The other possible difference is I run Eagle Performance Software's system unit.
Whilst the figure I quoted was truthful, the comparison was invalid. It was
a "teaser" message. (I am not a vindictive person - I just get even! :-)
We're 15 all now!)
TeeCee
--- TC-ED v2.01
* Origin: The Pilbara's Pascal Centre (+61 91 732930) (3:690/644)
* Tossed by SFToss v1.00b on 92/06/11 11:49:41
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-10-92 08:18:33
From Dj Murdoch
To Daniel Torrey
Subject Re: C++ and Pascal
DT> I'm trying to reuse some code in Pascal units. I know
DT> that I have to declare the routines extern pascal, but I
DT> don't know how to link code that's stored in a Turbo
DT> Pascal 6.0 .TPU file into a C++ program. Does anyone know
DT> how to do this? I can't find any mention of it in either
DT> the Borland C++ or Turbo Pascal manuals. Help!
I don't think it's possible. TLINK doesn't understand .TPU files at all.
You can, with a lot of work, link C++ .OBJ files into TP programs - take
a look in the manual for all the restrictions and details.
--- Msg V3.2
* Origin: Murdoch's_Point - - (1:221/177.40)
* Tossed by SFToss v1.00b on 92/06/11 15:56:00
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-09-92 21:12:02
From Mark Ouellet
To Rich Veraa
Subject Re: your modem
On 04 Jun 92, you, Rich Veraa, of 1:135/907.0 wrote...
RV> A message from Trevor Carlsen to Rich Veraa was released into the
RV> bitstream 30 May 92 09:48.
RV>
TC>> I'd say there is a bug, rather than a
TC>> problem with DV interaction problems.
TC>> Most likely with pointers. If you let
TC>> me know where I can FREQ it from (and
TC>> the filename), I'll pick it up and check it out for you.
TC>> (I note that your number is unpublished)
RV> Trevor, a friend of mine tried to send that with a 9600 v32, but could
RV> only connect with you at 2400.. We're not familiar with PEP -- is there
RV> some sort of secret to connect with it at 9600?
Rich,
I'm afraid it isin't Trevor who is at fault, nor is
it his PEP modem. I ran up a 150 $Can LD bill trying to
connect with many ZONE 3 modems over a week-end. Tried my
trusty Courrier HST at first but when that didn't work I
tried borrowing my friend's HST DS which does V32, V32bis.
I couldn't get a good connection then either. I
suspect the phone link I used was just too bad for the
connection. Maybe the trans-continent connection was the
problem. Satelite transmissions are sometimes hard because
of the delays involved in the reply.
It is puzzleing though as I have had no problems
with Zone 2 or almost none, mostly a case of my HST finding
itself face-to-face with a V32/bis.
I was about to try sending my mail to my Zone 3 gate
(which would be 1:3/0 since I'm in zone 1) but never got
around to calling him up to make sure it was ok.
Sorry TC for the long OFF-TOPIC post, it has nothing
to do with Pascal but it is of interrest to others I think.
Best regards,
Mark Ouellet.
--- ME2
* Origin: Governements, Proof of Peter's principle (Fidonet 1:240/1.4)
* Tossed by SFToss v1.00b on 92/06/11 15:56:01
▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
Conference 4
Date 06-12-92 15:09:00
From Norbert Igl
To BRIAN PAPE
Subject CLOCK PROBLEM
Hallo BRIAN,
am Dienstag, 09 Juni 1992 schrieb BRIAN PAPE an ALL:
BP> program testclk;
BP> uses crt;
BP> var
BP> a,b,c,d:longint;
BP> timer:longint absolute $0:$469;
--------------------------------
That's the problem...
Bios:Timertick is Timer : Longint absolute $40:$6C; (* = 0:$46C *)
you got a wrong val in your low-word, so .....
Gruss aus Bonn, Norbert
--- GoldED 2.40
* Origin: May the source be with you... (2:241/5300.3)
* Tossed by SFToss v1.00b on 92/06/13 20:57:32