Closed Bug 593565 Opened 14 years ago Closed 13 years ago

Less smooth tab close animation with Direct2D enabled

Categories

(Core :: Graphics, defect)

All
Windows 7
defect
Not set
major

Tracking

()

VERIFIED DUPLICATE of bug 633157

People

(Reporter: Virtual, Unassigned)

Details

(Keywords: nightly-community)

User-Agent:       Mozilla/5.0 (Windows NT 5.1; rv:2.0b6pre) Gecko/20100902 Firefox/4.0b6pre
Build Identifier: Mozilla/5.0 (Windows NT 5.1; rv:2.0b6pre) Gecko/20100902 Firefox/4.0b6pre

Tab close animation is no smooth in last part, when close button is removed animation is frozen for some amount of time (a few ms).
This should be fixed (like in Chromium), or simply close button should be removed first, before animation start to simply prevent this hungs (like in Opera).

Reproducible: Always

Steps to Reproduce:
1. Just close some tabs to see close tab animation
It's more visible with many "CPU usage" tabs
Actual Results:  
Animation is frozen for a few ms when close button is removed in the last part of the close animation.

Expected Results:  
Smooth animation like in Chromium or Opera
Blocks: 380960
Whiteboard: [dupeme]
Dao, related or fixed/dupe of bug 579323?
Dupe not, because it still occur, and patch for bug 579323 didn't' fix this
It's on Win7, too.
Can we finally mark this as NEW, because dupe for this bug not exist
OS: Windows XP → All
Hardware: x86 → All
I just tried in fresh profile again.

New tab/close tab animation was smooth until i opened some heavy JS site... That makes animations sluggish as hell...
In my case it's even on blank new tabs, close animation freeze for some time when close button is removed
Dupe not found, so how about marking this as NEW bug
My animations are generally pretty laggy, especially closing. Not even close to as smooth as chrome/opera.
http://www.youtube.com/watch?v=oCSJP9nG7Q8
the end half of the video is the best example of the laginess I usually see, at the beginning ti decided to behave just to spite me.
win7 64 bit
2GHz core2 duo
4gb ram
512MB Gddr3 mobility hd2600
Can someone with privileges mark this finally as NEW bug... ?
Severity: normal → major
blocking2.0: --- → ?
Read the comments of Bug Bug 599711.

Anyway these are duplicates. All of tab animations are sluggish especially when Fx does something other major task.

BTW, I would like to see a solution if is it achievable, too.
Similar bug, but for me it's reproducible always on blank pages, not need any specific JS or Flash hog sites
What happens if you disable hardware acceleration? Please also post the Graphic details from the Troubleshooting page under the help menu.
Adapter Description: Radeon HD 6800 Series
Vendor ID: 1002
Device ID: 6739
Adapter RAM: 1024
Driver Version: 8.782.1.0
Direct2D Enabled: true
DirectWrite Enabled: true
GPU Accelerated Windows: 1/1 Direct3D 10

Disabled hardware accel and tested. With hardware accel disabled opening and closing blank tabs the animation is far smoother, but animations during page loading/other heavy tasks got significantly worse, especially closing animations.
(In reply to comment #13)
> Disabled hardware accel and tested. With hardware accel disabled opening and
> closing blank tabs the animation is far smoother, but animations during page
> loading/other heavy tasks got significantly worse, especially closing
> animations.

Moving this bug to Core:Graphics. The second part should be covered by bug 599711.
Component: Tabbed Browser → Graphics
Product: Firefox → Core
QA Contact: tabbed.browser → thebes
Summary: Not smooth tab close animation → Not smooth tab close animation with hardware acceleration enabled
Version: Trunk → Other Branch
Summary: Not smooth tab close animation with hardware acceleration enabled → Not smooth tab close animation
Rechanging title as OP, because animation is frozen for a few ms when close button is removed in the last part independently of HW settings
Which last part?
(In reply to comment #17)
> > Which last part?
> Video posted by Brandon Watkins shows this perfectly
> http://www.youtube.com/watch?v=oCSJP9nG7Q8&hd=1

And an explanation would even be better. So you refer to your comment 13?

> but animations during page loading/other heavy tasks got
> significantly worse, especially closing animations.
Read my first post, I described this bug fully there...
I think I have found a way to even further showcase the tab lag. I installed a stylish theme with far more transparent tabs which makes the problem much more visible to the eye. The animations just aren't smooth. This particular video was taken on my desktop machine in my second post (radeon 6850, win 7 64 bit, catalyst 10.10, 2.6GHZ phenom II quad, hardware acceleration enabled)
I'll get on my laptop later and make another video maybe as it lags a lot more on the laptop.
http://www.youtube.com/watch?v=3GUT26LFqfk
Please renom if & when this bug gets confirmed.
blocking2.0: ? → ---
I can confirm this, if it counts.

But Bug 599711 is much more noticeable for me.
Marcia, could you have a look at one of our machines in the QA lab? I don't have a real Windows machine, to test this.
Not dupe found, so can we make this a NEW bug ?
Also this is visible on 64bit versions of Win 7.
Nothing new was there unfortunatly
Also it's dupe of this bug ;)
Nothing new? There is an important information in that bug:

the devs now the problem, but simply ignore it...

It's not a dupe, new tab opening is also not smooth. Anyway duping doesn't matter - the devs wont care about this bug, too. (or cant help the problem...)
Yep, but nothing new for me ;p

If they don't care (probably not) or can't fix this on Firefox 4 release, this bugged feature should be disabled by default because it looks bad and it's performance hog
blocking2.0: --- → ?
Please renom if & when this bug gets confirmed, and only then.
blocking2.0: ? → ---
It's been confirmed many times... it only lacks a status change to NEW
I can confirm this. I did a side-by-side comparison in beta 10 with two fresh profiles, one of which had gfx.direct2d.disabled set to true. Both windows had a blank tab and I repeatedly opened and closed one or two more blank tabs.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Keywords: perf
Summary: Not smooth tab close animation → Less smooth tab close animation with Direct2D enabled
Whiteboard: [dupeme]
Same with the latest nightly and updated drivers, ATI Radeon HD 4650.
OS: All → Windows 7
I think that this bug is most important to fix
Same with firefoc 4 beta 12 pre
Asking for blocking, now that we know what is causing the problem and the bug got confirmed.
blocking2.0: --- → ?
Changing to XP, because I firstly see this bug on this system
OS: Windows 7 → Windows XP
Depends on what you consider "this bug". There's no Direct2D on Windows XP. But since you just added the dependency on bug 599711, it seems like you're really just seeing that bug.
No longer depends on: 599711
(I intentionally removed the bug 599711 dependency, and I meant to change this back to Windows 7 as well.)
OS: Windows XP → Windows 7
OK, and I mean poor animation is also visible on Windows XP as I firstly reported this bug on this system
Right, based on that we could mark this bug a duplicate of bug 599711, but it seems more useful to keep this open and focused on the bad impact of hardware acceleration as reported in comment 13 and bug 593680 comment 3.
(In reply to comment #41)
> Right, based on that we could mark this bug a duplicate of bug 599711, but it
> seems more useful to keep this open and focused on the bad impact of hardware
> acceleration as reported in comment 13 and bug 593680 comment 3.

Are you seeing this with D3D9 layers enabled and Direct2D disabled as well? (Just go into about:config and set gfx.direct2d.disabled)
(In reply to comment #42)
> (In reply to comment #41)
> > Right, based on that we could mark this bug a duplicate of bug 599711, but it
> > seems more useful to keep this open and focused on the bad impact of hardware
> > acceleration as reported in comment 13 and bug 593680 comment 3.
> 
> Are you seeing this with D3D9 layers enabled and Direct2D disabled as well?
> (Just go into about:config and set gfx.direct2d.disabled)

Yes, that's exactly what I did, see comment 32.
Won't hold the release on this issue.
blocking2.0: ? → -
This bug is a mess. Can someone refile with a summary of the problem?
Version: Other Branch → Trunk
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → DUPLICATE
No longer blocks: 596954, 380960
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.