147845 - [Matrox video card] Graphics corruption when scrolling

Status: RESOLVED EXPIRED

Description

[Amir Khan]

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.0rc3) Gecko/20020523
BuildID:    2002052306

When you scroll down the page there are horizonal "blocks" than appear across
the page, so you cant see the content below.  Looks like the memory where the
page is stored gets ncorrupted.

I can supply a screen grab if required. No problem with Netscape 4.7 or I.e. 5.5

Reproducible: Sometimes
Steps to Reproduce:
1. go to URL http://news.bbc.co.uk/hi/english/uk_politics/newsid_2014000/2014220.stm
2. scroll up and down the whole page
3. see the problem ;-)

Randomly surft the site to make it "appear" ..

Actual Results:  screen corruption as explained.

Expected Results:  no screen corruption

I use NT 4 
SP6a
compaq Ap550 dual 866 P3
512MB RAM
Maxtrox Video card with 2 LCD (1525X) displays
Matrox powerdesk 4.51
Matrox driver version 4.00.1381.4510 (4.51.021)
Matrox Model : G200 MMS Gfx chip MGA-2000 8MB RAM BIOS 2.5.19

Comment 1

[Amir Khan]

Attachment: example 1 / 2

here is an example of the problem.

Comment 2

[Amir Khan]

Attachment: second example of the problem

here is another example (I scrolled up and down a few times).

Comment 3

[Amir Khan]

I just discovered that this bug happens ONLY when you use the up and down arrows
to scroll in the html paege.  If I use page up and down then its OK.

Comment 4

[Michael Lefevre]

WFM with 2002101612 on win2k.

reporter (Amir): can you reproduce this with a recent build of mozilla? if so,
please comment again with details. if not, please resolve this bug as
WORKSFORME. thanks.

Comment 5

[Daniel Wang]

*** Bug 153864 has been marked as a duplicate of this bug. ***

Comment 6

[Daniel Wang]

*** Bug 168390 has been marked as a duplicate of this bug. ***

Comment 7

[Dymaxion]

I'm still seeing this with 1.2final on win2k, matrox millenium II, driver
5.0.2144.1. It takes some time to show up, and I haven't found a more reliable
way to trigger it than browsing through large images for a while. If you keep
browsing after the behaviour starts up, parts of the browser window will
eventually get corrupted, too.  IME, the corruption generally shows up as a
copying other parts of the page, and failing to redraw the page and/or UI
elements when they change.  Scrolling using the scrollbar and mouse will make
the corruption go away, but only for page areas moved out and back into the
window, selecting items will cause those items to show up (but not, for example,
backgrounds underneath them), and moving the entire window offscreen will also
get rid of it, however once the corruption has started, things will not start
redrawing correctly until a complete browser restart.

Comment 8

[Stefan Moebius]

According to Matrox this is a bug in the driver (Search for 'smearing of web
pages' in their forums). 
You should upgrade to the most recent driver:
http://www.matrox.com/mga/support/drivers/latest/home.cfm

Users of some older cards, like Millenium II on Win2k are somewhat out of luck -
the page lists powerdesk 5.82.018 as final for this card and the release notes
still list the bug.
http://www.matrox.com/mga/support/drivers/files/notes/w2k582notes.txt

This bug should be INVALID.

Comment 9

[Fred Goldstein]

Before you write this off as a user problem, I'll note that I've already  upgraded Win98 to the latest "beta" Matrox driver, 6.83.17, which has been on  their web site since last July.  It didn't help.  (I've also "updated" Win98 to SE.)  There's nothing in its included release notes about this bug.      I'd check the Matrox forums but they seem to be down right now.   No applications other than Mozilla builds, newer than a certain vintage, trigger this video card behavior, so there's obviously something very, very unusual in the way Mozilla uses the MGA drivers.  It impacts both page rendering and widgets (buttons, menus, etc.).  And there's no corresponding problem in Linux, with the X Windows Mozilla (which works fine). 

Comment 10

[Stefan Moebius]

Fred:
You're correct in that it seems to have something to do with the way webpages
are displayed. However the problem is not restricted to mozilla:

http://forum.matrox.com/mgaforum/Forum9/HTML/009502.html
http://forum.matrox.com/mgaforum/Forum9/HTML/010024.html
(The latter thread indicates that the problem _is_ persitant in 6.83)

However, if you ask in the forums (and there are some other tech support forums
around the net where you'll get the same information), you will be told that the
problem is one of Windows+Graphics driver.

[ironic]
If the problem does not occur on Linux, I think there's quite an easy
work-around ;-)
[/ironic]

Comment 11

[Fred Goldstein]

I don't believe the forum reference to "smearing", as it occurs in Opera, is 
the same problem at all!  Smearing seems to be a degraded picture quality 
within the scrolling rendered-web-page window.  Also, it impacted browsers 
other than Mozilla, including IE, on dual-head systems.  Mine's a 
single-headed system and IE renders perfectly (I've had to make it the Windows 
default); I don't have Opera. 
 
The error I'm getting is that the images in the window render perfectly, WHEN 
they render.  But when the bug erupts, nothing renders unless the area is 
scrolled in.  So if I load a small (no vertical scrollbar) window, it doens't 
show at all.  If I try to edit say a bugzilla text box when this problem 
erupts (this actually happened a few days ago), I stop seeing my input and 
lose track of what I was typing!  But if I can scroll it in and out, then I 
see the page.  Also, the buttons on top no longer show up until I mouse over 
them, and other Windows graphical elements (not the rendered page) are 
similarly not shown.  
 
The basic idea is that the application (Mozilla) apparently thinks that the 
material has already been displayed by the GDI, when it hasn't, so it only 
becomes visible when Windows tells it to repaint.  On Win98, this problem is 
probably not in the application per se, since it requires a reboot, but in the 
way the application talks, or perhaps more accurataly listens?, to the display 
subsystem. 
 
So yes, in a sense it's a Matrox problem, but finger-pointing isn't getting 
anybody anyplace.  The bug I'm dealing with is Mozilla-specific. 

Comment 12

[Lothar Leidner]

I do have graphics corruption in linux (Mozilla 1.5, SuSE 8.1, Matrox G450DH 
with only one monitor). When I scroll down, there are always horiztontal lines 
which cover the original content. The extent of corruption seems to depend on 
scroll velocity. The corruption disappears when I mimimize and then restore the 
window.
I have updated from Mozilla 1.3 (which worked fine) to 1.5. Other X-
applications work fine.
There are also artefacts along the scrollbar and toolbar.

Comment 13

[Gervase Markham [:gerv] (not reading bugmail)]

This is an automated message, with ID "auto-resolve01".

This bug has had no comments for a long time. Statistically, we have found that
bug reports that have not been confirmed by a second user after three months are
highly unlikely to be the source of a fix to the code.

While your input is very important to us, our resources are limited and so we
are asking for your help in focussing our efforts. If you can still reproduce
this problem in the latest version of the product (see below for how to obtain a
copy) or, for feature requests, if it's not present in the latest version and
you still believe we should implement it, please visit the URL of this bug
(given at the top of this mail) and add a comment to that effect, giving more
reproduction information if you have it.

If it is not a problem any longer, you need take no action. If this bug is not
changed in any way in the next two weeks, it will be automatically resolved.
Thank you for your help in this matter.

The latest beta releases can be obtained from:
Firefox:     http://www.mozilla.org/projects/firefox/
Thunderbird: http://www.mozilla.org/products/thunderbird/releases/1.5beta1.html
Seamonkey:   http://www.mozilla.org/projects/seamonkey/

Comment 14

[Gervase Markham [:gerv] (not reading bugmail)]

This bug has been automatically resolved after a period of inactivity (see above
comment). If anyone thinks this is incorrect, they should feel free to reopen it.

Comment 15

[arch4non]

I'm getting this error when I scroll up and down a page.

Comment 16

[Philip Chee]

> I'm getting this error when I scroll up and down a page.
Whatever your problem is, it isn't this ancient bug. Please file a new bug for your issue.