These applications run in what looks like dos text mode, but they have an
arrow shaped mouse cursor.
I think it's done by modifying the character set on the fly so that the four
cells that underlie the cursor get set to special characters which are the
same as written, except they have the appropriate mouse fragments ORed in.
Then when the mouse moves again, those four are replaced and the next four are
modified.
Anyway, I don't want to write the code, but I want my app to look like that.
The specific examples that I know of are: Dos 6.2 DEFRAG, Norton Anti Virus
(NAV), and the Norton SMARTCAN dos app.
Thanks
You're looking for the Precise Point mouse driver thingy. Have a look
around for a file with a name like PP111.ZIP or something. I've got that
here (except it's an ARJ file, but close enough)but there may be a later
version than 1.11 (which is what this one is). I can mail it to you if you
like, but it's about 80k compressed.
-Dale
This is my .sig.
Hope you like it.
I'm pretty sure that those utilities actually sqitch into graphics
mode and copy everyting across to the graphics screen. I have a
Hercules monocrome card (don't laugh) and it always uses the
standard # (<- a full block) cursor, and not the graphics one.
Also, you might notice that check boxes will turn into circles,
and some other nice things ...
I think what you might have to do is programme the cursor with your
own character, but it won't scroll smothly across the screen then.
--
Jim Richards, | Genasys II Pty Ltd
| 13th Level, 33 Berry St, North Sydney, NSW, Australia
| Phone: +61-2-954-0022 (-9930 FAX)
| Internet: ji...@genasys.com.au
No, they use text mode with modified character set. The Herc card is unable
to modify the character set, so it uses the block cursor.
If you want to see evidence of this, look at what happens to wide characters
like 'm' when the mouse is near them. Because of the kludgy way the VGA
handles character cells, the right side of the m gets widened when the mouse
is near it.
You can get C or Pascal code to do this stuff yourself, in MOU105.ZIP. It's
available on lots of ftp sites.
Duncan Murdoch
Well, although you are "pretty sure" you are completely *WRONG*. You should
not write any postings on subjects you do not know anything about. I could
explain exactly what happens in which mode, but I don't feel like it right now.
Mail me (preferably on KHEK...@hut.nl") and I will explain.
Greetings,
Kilian Hekhuis
I thought I read somewhere that the Hercules card is the only one that does not
allow the user to redefine the character set. If this is true then it would
explain why you are not seeing a "graphic" mouse cursor.
Paul Battersby
>In article <bart.11....@fast.net>, Bart Addis <ba...@fast.net> wrote:
>>
>>I think it's done by modifying the character set on the fly so that the four
>>cells that underlie the cursor get set to special characters which are the
>>same as written, except they have the appropriate mouse fragments ORed in.
>>Then when the mouse moves again, those four are replaced and the next four
>are
>>modified.
>>
>>The specific examples that I know of are: Dos 6.2 DEFRAG, Norton Anti Virus
>>(NAV), and the Norton SMARTCAN dos app.
> I'm pretty sure that those utilities actually sqitch into graphics
> mode and copy everyting across to the graphics screen. I have a
> Hercules monocrome card (don't laugh) and it always uses the
> standard # (<- a full block) cursor, and not the graphics one.
> Also, you might notice that check boxes will turn into circles,
> and some other nice things ...
> I think what you might have to do is programme the cursor with your
> own character, but it won't scroll smothly across the screen then.
It does stay in text mode - just creates a new set of text bit maps for the
characters surrounding the cursor (4 off). Do a printscreen of a Norton
display and see what characters get printed where the mouse cursor is.
For the other fancy 'graphics' characters bit maps are created too. A whole
new set of characters can be displayed by redirecting the 'video character
set' interrupt vector to your own set.
David Forsyth da...@iwr.ru.ac.za
Institute for Water Research Rhodes Unversity South Africa
----------------------------------------------------------------------
An end user, when asked to respond with a 'Y' or an 'N', invariably
presses '8'.
Karen Hooten
You are not completely right on this. The 'video character set' interrupt is used
only in graphics mode, not in text mode. In text mode the character map(s) are
stored in different vidmem bit planes.
Kilian Hekhuis
Bzzzt! Wrong! You may wish to look at these programs running on a EGA/VGA.
You will then see what we mean...Of course on VGA systems text mode
characters are *9* pixels wide (the 9th column of pixels on the rhs are
always blank, except for ASCII 192..223 where the 9th column duplicates the
8th.
>|>
>|> It does stay in text mode - just creates a new set of text bit maps for the
>|> characters surrounding the cursor (4 off). Do a printscreen of a Norton
>|> display and see what characters get printed where the mouse cursor is.
Or run the program in a window under Windoze.
>|> For the other fancy 'graphics' characters bit maps are created too. A whole
>|> new set of characters can be displayed by redirecting the 'video character
>|> set' interrupt vector to your own set.
> You are not completely right on this. The 'video character set' interrupt is used
> only in graphics mode, not in text mode. In text mode the character map(s) are
> stored in different vidmem bit planes.
Changing the font in both text and graphics modes is just a BIOS call away.
In most cases to make it stick across mode changes (eg from MODE CO80 or
graphics<->text mode) you will need to patch the int 10h function 0 handler.
[WARNING: gratuitous advertising follows] I've written a TSR what does just
that. Font files are stored separately. If you want a copy mail me and I'll
see what I can do. But this program doesn't have anything to do with
text mode arrow mouse cursors (other than software fonts) [END of gratuitous
advertising]
-------------------------------------------------------------------------------
Michael Deegan mdee...@pilot.stu.cowan.edu.au Perth, Western Australia
GCS -d+ H s+:- g+++ p0 !au a- w+(++) v@ c++ UA+ p? L !3 E? N++(+++) K- W+ M+ V-
-po+ Y t(+) !5 !j R G+(')('')('''') tv b+ D++ B? e+ u* h! f? !r>* !n !x+
-------------------------------------------------------------------------------