[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IIgs "Mouse-erasing" ---- is it a virus??



shack@onyx.southwind.net (Randy Shackelford) wrote:
>David Empson (dempson@atlantis.actrix.gen.nz) wrote:
>: In article <4mm54k$2o8@gap.cco.caltech.edu>,
>: Nathan Mates <nathan@cco.caltech.edu> wrote:
>: > In article <4mln3b$4c8@lynx.unm.edu>,
>: > stephen e buggie <buggie@pegasus.unm.edu> wrote:
>: > >On some of my software, I notice that scrolling the mouse around the 
>: > >screen erases the portion of screen image....
>: > >Is this a virus or what?   Teh software continues to work.
>: > 
>: >    If this is on a Rom 3, the cause is buggy software. Apparently, the
>: > way things get drawn on the 3 vs the 01 changed, mostly affecting
>: > mouse drawing. [BeatBox is one program that suffers from this]
>
>: I don't recall this problem.  The only one I know of regarding the
>: mouse on the ROM 3 is that the cursor is invisible while moving the
>: mouse up the screen if you have a ZIP GS, and this is fixed by the
>: ZipFix INIT.
>[snip]
>
>If you have the shotdown reminder extension installed, you can see this problem
>in action. It works properly on a ROM 01 but on a ROM 3, after the screen goes
>black right before you get the "you can turn off your IIgs safely" box, you
>can move the mouse around and wherever you move it, it changes the screen back
>to the desktop color.

Although I've not touched my GS in over 1.5 years, I do seem to remember
this stupid ROM 03 cursor problem is caused by the shadow SHR page at bank
$01. While working on Beatbox, I'd disassembled a good potion of the toolbox's
cursor handling code and I believe on the ROM 03, the cursor routine makes
too many assumptions and it continues to read from SHR bank $01 in shadow mode
even after you turn it off via $E1C035. On the ROM 01, the cursor routine
either checks to see if the shadow SHR page is active or not or it has been
hardcoded to read off bank $E1 (can't remember which). In any case, this only
happens when you mess with $E1C035 directly without telling the toolbox on the
ROM 03. This is really not a bug, but rather something Apple had warned us
about, but the problem is, depending on what you do, doing this the "right"
way can bring about a major performance hit. What a freaking dilemma!
I guess the best way to avoid this is to write your own cursor handling
routine, which is most likely faster than Apple's ugly crap anyway ;)


>--
>Randy Shackelford                                 I was internet
>shack@southwind.net                               when internet wasn't cool


--