Path: utzoo!utgpu!watmath!watdragon!trillium!bamcpherson
From: bamcpherson@trillium.waterloo.edu (Brent McPherson)
Newsgroups: comp.sys.ibm.pc
Subject: Cursor question, Int 0x10 service 0x03
Keywords: insert cursor, problem, certain machines
Message-ID: <9821@watdragon.waterloo.edu>
Date: 19 Nov 88 21:10:51 GMT
Sender: daemon@watdragon.waterloo.edu
Distribution: comp
Lines: 40


Hi, 
	I calculate a half-height insert cursor from the information
	received through interrupt 0x10, service 0x03. 
	What I do is take the old cursors bottom scan line, divide this
	by two. This now becomes the new top scan line of the insert cursor.

	eg.

	regs.h.ah = 0x03;                /* read cursor position */
	intr( 0x10, &regs );             /* perform video interrupt */
	old_cursor = regs.x.cx;          /* save normal cursor */
	ins_cursor = (((old_cursor & 0xff) >> 1) << 8) + (old_cursor & 0xff);


	The cursor is set by using interrupt 0x10, service 0x01 and
	the cursor (old or ins) is stored into CX.

	There seems to a problem on certain machines. (the plain IBM PC
	with a monochrome monitor I think) Basically instead of getting a
	half-height insert cursor you get a cursor in the middle of the
	character. Also a mode change using "mode" seems to correct the 
	problem.

                normal       desired      actual
		cursor     ins cursor    ins cursor
		+----+       +----+       +----+
		|    |       |    |       |    |
		|    |       |####|       |####|
		|####|       |####|       |    |
		+----+       +----+       +----+

	I do not have access to a machine where the problem occurs and
	cannot investigate this problem. Does anyone know if
	this a known bug? Am I doing something wrong? Has anyone else
	had this problem? This code is called from within a TSR.

	Thanks in advance,
--
Brent