RE: Attempt to reassemble colorforth: the Mark Slicker bug

Hoi Albert,

I downloaded your color-bg.com image - unfortunately I no longer have a PC
that runs Chuck's ( or your ) binary. Something to do with an upgrade a year
or so back.
But I can confirm that your image produces the same red/green/blue colour
bars at the top of the screen as Chuck's binary.
Since the colour bars are probably an attempt at displaying the colorForth
logo, this means that most of the code is working - maybe just the address
of the video RAM buffer is wrong - so you're on the right track! Perhaps you
could merge your current code with the one of the other cf's which use other
methods of determining the video RAM address...

BTW if you click on color-bg.com using Windows 98 SE ( and presumably
earlier versions ) and answer YES to the "Run in MSDOS mode" option, Windows
will add a line to the c:\autoexec.bat file something like :
"CALL ........\COLOR-BG-COM" .
This will execute the colorForth binary every time Windows is started.
For most Windows users, this is a one-way operation, as you cannot get past
colorForth to boot Windows in order to disable it. IIRC there was an irate
post about this on clf when color.com was first released.

To return to normal Windows ( 98 SE )operation :
Press F8 as Windows is starting, and select "4. step-by-step confirmation".
Press "Y" repeatedly (about 15 to 20 times) until the "CALL
........\COLOR-BG-COM" line when you press "N".
Press "Y" again a few more times... Windows will then tidy up its mess, and
all will return to normal.

Hopefully anyone who has made their Win98 system comatose in this way will
be able to fix it, in order to read this email, and so find out how to fix
it.... ;)

Regards

Howerd  8^)







-----Original Message-----
From: Albert van der Horst [mailto:[email protected]]
Sent: 20 May 2005 15:16
To: [email protected]
Cc: Albert van der Horst
Subject: [colorforth] Attempt to reassemble colorforth: the Mark Slicker
bug


Hoi Folks,

I believe I am now in a position to reassemble colorforth.
Surely I can reassemble to exact the same COLOR.COM, but that is
not useful in itself.

As a first attempt to do something useful, I have tried to
remove the bug that was recently pointed out by Mark Slicker.
(And that took me so long to understand.)

Small as this change is, it involves a change in quite some
code addresses, which affects large parts of the executable,
but a change of data addresses is still largely circumvented.

Unfortunately, I can't run the original COLOR.COM that this is
based on. So I would ask for someone who can run the original
COLOR.COM to try out the changed version COLOR-BG.COM and report
back what happens. (I got kind of stuck with Howards colorforth,
then decided to start with the version from www.colorforth.com.)

It is available through anonymous ftp from
        ftp://80.100.243.19/pub/colorforth/ms-bug.zip

(Log in as anonymous or ftp. Most browser do this automatically.)

Apart from the colorforth this archive contains the source,
a difference-file, and the assembler/disassembler with a
script to assemble (on a Linux machine).

Note that the colorforth screens are still represented as
reassemblable hex dumps. But the icons look nice, I think.
The undiffed source is straight from the disassembler/crawler
colorcm-1-37.dis. So be prepared to find some strange things.

Groetjes Albert


--
Albert van der Horst,Oranjestr 8,3511 RA UTRECHT,THE NETHERLANDS
Economic growth -- like all pyramid schemes -- ultimately falters.
[email protected] http://home.hccnet.nl/a.w.m.van.der.horst

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
Main web page - http://www.colorforth.com