Closed Bug 142139 Opened 23 years ago Closed 23 years ago

GIF Images horked when scrolling

Categories

(Core Graveyard :: Image: Painting, defect)

Product:
Core Graveyard
Component:
Image: Painting
Platform:
x86
Windows 98
Type:
defect
Priority:
Not set
Severity:
critical

Tracking

(Not tracked)

Status:
RESOLVED DUPLICATE of bug 104992

People

(Reporter: selmer, Assigned: pavlov)

References

(
URL
)

Details

Attachments

(1 file)

Note munging of "People.com" gif image next to "Beautiful Covergirl"
250.44 KB, image/jpeg
Details
5/3 06 branch build. There are a bunch of bugs about PNGs having problems with scrolling. I am also seeing munged GIF images with scrolling in this build and at least one other build from earlier this week. Sorry for the NS internal link, but the page where I'm seeing this right now is http://client/machv/project_wide/project_overview.htm. I'll try to find another page that's not behind the firewall. Anyway, go to this page and scroll up and down and see that the gif image of the magnifying glass gets drawn all stupid almost everywhere it shows up. (I checked the source of the page and it *is* GIF rather than PNG.)
Adding nsbeta1 and Gagan to see if this can be reproduced by anyone else and maybe fixed before we ship this all over the place.
Keywords: nsbeta1
OK, home.netscape.com is completely gif based and also exhibits this problem in today's build. Just scroll up and down and watch the people.com image near the top-center turn into a bunch of random lines.
URL: home.netscape.com
Severity: normal → critical
any ideas when this regressed?
I'm the most likely candidate for having regressed it... checking
just checked a build on 98 from the day after my changes, and it looks fine. May not be me, or we may need more info.
using a 5/3 06 1.0 branch build on win2k; I cannot reproduce the home.netscape.com case, or the internal case.
I can't reproduce it with the 5/3 build either. My money is on video card/setup/driver funkyness. We're going to need info on the machines this reproduces on and probably access to them for testing on.
correcting component.. I can't reproduce this either. There have been no recent imagelib changes checked in to the branch. The only things I can think of would be dcone or saari's recent GFX changes (but I'm not sure if those went on to the branch or not).
Component: ImageLib → Image: GFX
mine are on the branch, but I'm not seeing the bug. It is conceivable that this could be driver or hardware acceleration related
Steve, what is your resolution, video setup, etc? Shrirang and I are unable to reproduce this as well.
I've twiddled my machine's video hardware acceleration to no effect. Still don't see it.
Singular idea poped into my head... perhaps Steve has finally found a system that really does need SetBrushOrgEx called after SetStretchBltMode... a screen shot of the error would help my guessing in the dark (a nice thing about graphics progragramming is that you can literally see the bugs quite often)
Chris, here's a screen shot. I'm on 1024x768, True Color (32 bit), LXA572W on 3D RAGE LT PRO (English) (DirectX) according to the settings dialog. This is a Mitsubishi 15" flat panel display.
Weird, I definitely don't see that. Steve, could you download the build from 4/22 and the build from 4/24 and see if the change occured in that timeframe? I'm trying to determine if it was my changes. If so, I guess I'll have to start feeding you builds until I lock it down as you're the only one who sees this.
I already have 4/21 build on my machine and I see it there. I have some older builds, I'll try those as well. Is this related to backgrounds and transparency?
Also happens in 4/9 build, the oldest one I still have.
3/1 trunk build also exhibits this.
Sigh, also exhibits on 1/1 03 trunk build :-( I'm not going back any further without some prompting.
My quick search for other ATI bugs pulled up 104992, although I seem to recall another one being around, but I can't seem to find it right now. Are you using the latest win98 drivers for your video card?
In the system info utility, the display category shows drivers that mostly have the letters ATI in the front, so mine must be an ATI. Oddly enough, my win98 drivers are all dated in 1998. Is there a place where I can get updated drivers that will fix the problem?
http://www.ati.com/support/driver.html would be the place to look.
Keywords: nsbeta1 → nsbeta1+
Whiteboard: [adt1]
Jaime, I believe this is going to be duped against 104992. The keyword foo probably should go on that bug rather than this one. It also seems that this is a driver problem and doesn't require a fix in the client. Pav, can you clarify that?
I believe that is a driver problem. There should not be any client fixes needed.
right on! changing to nsbeta1-.
Keywords: nsbeta1+ → nsbeta1-
Whiteboard: [adt1]
I have been unable to reproduce on any Windows system. I've tried with build 2002-05-05-09-1.0.0 on win XP, win 2k and win ME, and do not see the bug.
duping to 104992 instead of marking invalid since it is related to ati driver problems *** This bug has been marked as a duplicate of 104992 ***
Status: NEW → RESOLVED
Closed: 23 years ago
Resolution: --- → DUPLICATE
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: