Closed Bug 612603 Opened 15 years ago Closed 14 years ago

A lot slower scrolling in firefox4 than firefox 3.6

Categories

(Core :: Graphics, defect)

x86
Windows 7
defect
Not set
normal

Tracking

()

RESOLVED WORKSFORME
Tracking Status
blocking2.0 --- -

People

(Reporter: simon.strandman, Unassigned)

References

Details

(Keywords: regression)

Attachments

(2 files)

User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0b7) Gecko/20100101 Firefox/4.0b7 Build Identifier: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0b7) Gecko/20100101 Firefox/4.0b7 On my netbook, a samsung NF310 with a Atom N550 CPU and a Intel GMA 3150 graphics card, Firefox 4 is a lot slower compared to firefox 3.6 when scrolling on pages. I've tried both 4.0 beta 7 and a build of beta8pre from today (20101116) and both have the same problem. Enabling or disabling hardware acceleration doesn't seem to do much difference. This is with smooth scrolling enabled but even without it firefox 3.6 is faster. Reproducible: Always
Can you go to Troubleshooting Information from the Help menu, copy the part that says Graphics, and paste it here, please ?
Here it is: Graphics Adapter Description Intel(R) Graphics Media Accelerator 3150 Vendor ID 8086 Device ID a011 Adapter RAM Unknown Adapter Drivers igdumdx32 Driver Version 8.14.10.2117 Driver Date 4-19-2010 Direct2D Enabled false DirectWrite Enabled false GPU Accelerated Windows 0/1 Apparently acceleration isn't enabled. :/
Component: General → Graphics
Product: Firefox → Core
QA Contact: general → thebes
It's not enabled because of this: 538 IMPLEMENT_INTEL_DRIVER_BLOCKLIST(gfxWindowsPlatform::kWindowsXP, deviceFamilyIntelGMA3150, V(6,14,10,5260)) I believe that maps to 8.14.10.5260 for the "Driver Version" output in about:support for the Intel drivers. That's the minimal version that doesn't have serious bugs if you try to use hardware acceleration with it; yours is older than that. :( That said, it's odd that the scrolling is slower than 3.6.... I would expect it to be no better, but also no worse.
Okay, unfortunately I can't upgrade the driver since samsung only provides this version and the brightness fn-keys doesn't work with the driver from intel :/ "538 IMPLEMENT_INTEL_DRIVER_BLOCKLIST(gfxWindowsPlatform::kWindowsXP, deviceFamilyIntelGMA3150, V(6,14,10,5260))" Does this affect windows 7 as well? Because that's the OS I use (ie. not "WindowsXP"). Yea, I guess it's also a problem that an unaccelerated firefox4 is slower than 3.6.
> Does this affect windows 7 as well? Ah, good catch. The relevant Win7 line is: 552 IMPLEMENT_INTEL_DRIVER_BLOCKLIST(gfxWindowsPlatform::kWindows7, deviceFamilyIntelGMA3150, V(8,14,10,2117)) So your version should be ok, at first glance. bjacob, any idea what's going on there?
I might be wrong, but the X3150 looks like it might have a maximum texture size of 2048; we require 4096.
I think Joe nailed it! We should expose more information as to why a feature (here D3D layers) failed to initialize. It's at least the 3rd time that I see a whole discussion generated by this issue of graphics cards not supporting 4K textures.
I just tried 4.0b9pre vs. 3.6.14pre (most recent snapshot as of today) and the problem is still there. It's more visible when I enable aero.
Attached file Slow scroll example
Same using FFbeta10 on Thinkpad X61, hardware acceleration and smooth scrolling disabled. Attached example scrolls fast on FF 3.6 and also FF4beta5. Adapter Description: Mobile Intel(R) 965 Express Chipset Family Vendor: ID8086 Device ID: 2a02 Adapter RAM:Unknown Adapter Driver: sigdumd64 igd10umd64 igdumdx32 igd10umd32Driver Version8.15.10.1867Driver Date8-6-2009 Direct2D Enabled: Blocked on your graphics card because of unresolved driver issues. DirectWrite Enabled: true (6.1.7600.20830) WebGL Renderer(WebGL unavailable): GPU Accelerated Windows0/1
I've just found and used window.mozRequestAnimationFrame (implemented in FF4beta4+ by Boris) wich demonstrates slow scrolling in FF4beta11. Tro to open this page in FF4beta4pre (20100826041432) and then latest FF4beta. Try to scroll rapidly and see the window title with FPS info. Sometimes I get 0 frames per second on the latest FF and not more than 30-40 on the FF4beta4pre. Personally I'd mark this bug as major as scrolling is a significant interaction with users and can degrade overall speedup in FF4.
Status: UNCONFIRMED → NEW
blocking2.0: --- → ?
Ever confirmed: true
Keywords: regression
So this affects ThinkPads with the default set of drivers? bz: you nominated this, so I'm assuming you're asserting that the web content patterns that scroll slowly are pretty prevalent on the web?
(In reply to comment #11) > So this affects ThinkPads with the default set of drivers? > > bz: you nominated this, so I'm assuming you're asserting that the web content > patterns that scroll slowly are pretty prevalent on the web? As far as I understand it this is mainly happening without hardware acceleration. I.e. this would affect any machine that for one reason or another would not get hardware acceleration. However how prevalent the content is is something I don't know much about. (In reply to comment #9) > Created attachment 511291 [details] > Slow scroll example > > Same using FFbeta10 on Thinkpad X61, hardware acceleration and smooth scrolling > disabled. Attached example scrolls fast on FF 3.6 and also FF4beta5. > > Adapter Description: Mobile Intel(R) 965 Express Chipset Family > Vendor: ID8086 > Device ID: 2a02 > Adapter RAM:Unknown > Adapter Driver: sigdumd64 igd10umd64 igdumdx32 igd10umd32Driver > Version8.15.10.1867Driver Date8-6-2009 > Direct2D Enabled: Blocked on your graphics card because of unresolved driver > issues. > DirectWrite Enabled: true (6.1.7600.20830) > WebGL Renderer(WebGL unavailable): GPU Accelerated Windows0/1 Make sure you don't have DirectWrite enabled when hardware acceleration is off, this is not a default/supported configuration.
Mike, I don't know. I'm asserting that someone who knows something about graphics (not me) should triage this bug and decide the answer to your question....
I can't see us blocking on something this generic; please renominate if there are more details.
blocking2.0: ? → -
Simon or Jaroslav, are you still able to reproduce this in the latest Nightly?
I don't use that netbook anymore so I can't test it atm! I'll leave a comment if I can test it on similar hardware again.
I'm going to close this for now. Please do reopen if you're still able to reproduce.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: