Re: ROM 1 better than ROM 3?

[ballwebr@mycroft.ECE.ORST.EDU (brian ballweber)]

David A Lyons (dlyons@apple.com) wrote:
: In article <3b6p6d$2rd@news.service.uci.edu> Trangdaithi Hoang,
: thoang@orion.oac.uci.edu writes:
: > DOCVu 2.0 works fine on a ROM 3 as long as you don't move the mouse.
: > Moving the mouse messes up the screen.  I've been told by a Digital
: > Exodus member that DOCVu 2.0 doesn't access the mouse and that it
: > might be caused by the differences in how the ROM 1 and ROM 3 shadow 
: > memory.  There's also at least one new program in the works that has
: > a similar graphics problem.  It also has a mouse problem with DOCVu
: > 2.0 loaded.  The ROM 1 has no problems running these programs.

: It sounds like the application is assuming that the Super Hi-Res
: screen lives in bank $E1, which is not the case if shadowing is
: enabled.  There's at least one GS technical note that talks about
: this...the one that talks about CloseView compatibility...might have
: "Universal Access" in the title.

: You might or might not be able to get the app to work better by
: removing CloseView from your system.  If that doesn't do it, the app
: could be -asking- for shadowing in QDStartUp (in which case it would
: be surprising if they were writing directly to the $E1 screen).

: Yes, there are differences in how ROM 1 vs ROM 3 QuickDraw track the
: mouse cursor.  It's not a problem with properly-written applications. 
: ROM 1 always draws the cursor in bank $E1, but ROM 3 notices whether
: Super Hi-Res shadowing is enabled, and if so it draws the cursor in
: bank 1.

I wasnt until recently aware of this problem, however I was always under
the impression that the desk manager took care of things such as this when
it entered the cda menu, apparently not.  DOCVu does write directly to the
$e1 screen, and offhand i dont remember what it does about shadowing <i plan
to look into it>.  However since DOCVu does not use quickdraw, and since it
already patches out quite a lot of stuff including the heartbeat stuff and
any scanline interrupts <possibly more i forget>, would taking out the
mouse interrupt effectively kill this problem?  If not I will work out
something with the shadow register to "fix" things...

I'll try to find that technote too...

- Brian
  Digital Exodus