Open Bug 1650499 Opened 5 years ago Updated 4 years ago

Win7: Wrong Hz display detection

Categories

(Core :: Graphics, defect, P3)

80 Branch
x86_64
Windows 7
defect

Tracking

()

Tracking Status
firefox77 --- wontfix
firefox78 --- wontfix
firefox79 --- wontfix
firefox80 --- wontfix
firefox81 --- affected
firefox82 --- affected
firefox83 --- affected
firefox86 --- affected
firefox95 --- affected

People

(Reporter: cholismo, Unassigned)

References

(Blocks 1 open bug)

Details

Attachments

(5 files, 2 obsolete files)

User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:78.0) Gecko/20100101 Firefox/78.0

Steps to reproduce:

Load this site: https://www.displayhz.com/ or this one: https://www.vsynctester.com/

Actual results:

It shows 64 Hz

Expected results:

It would be 60 Hz instead (my actual display refresh rate).

Attached file about:support output
Attached file about:support output (obsolete) —
Attached file about_support.txt (obsolete) —

clean profile

Attached image vsynctester
Attached image displayhz

I have tried with Brave (Chrome) and IE 11 and the refresh rate is properly detected (60 Hz).

[Tracking Requested - why for this release]:

Component: Untriaged → Graphics
OS: Unspecified → Windows 7
Product: Firefox → Core
Hardware: Unspecified → x86_64
Attachment #9161355 - Attachment is obsolete: true
Attachment #9161354 - Attachment is obsolete: true

What happens if you flip the pref gfx.vsync.force-disable-waitforvblank?

Flags: needinfo?(cholismo)

(In reply to Timothy Nikkel (:tnikkel) from comment #8)

What happens if you flip the pref gfx.vsync.force-disable-waitforvblank?

It doesn't change anything, I knew about this switch: https://bugzilla.mozilla.org/show_bug.cgi?id=1628137

I already tried several versions pre-76 and it happens the same. I think that pref is only valid for Windows 8.1 and 10 because on Windows 7 it should be disabled by default: https://bugzilla.mozilla.org/show_bug.cgi?id=1628137#c2

Thanks.

Flags: needinfo?(cholismo)

I have just tested Firefox 68.10 ESR and the behavior is similar but slightly different; At first it detects the display Hz correctly but if you let the test run for a while it turns to 64 Hz eventually.

The severity field is not set for this bug.
:jbonisteel, could you have a look please?

For more information, please visit auto_nag documentation.

Flags: needinfo?(jbonisteel)
Severity: -- → S2
Flags: needinfo?(jessiebonisteel) → needinfo?(ktaeleman)

@Jamie: you've looked into refresh rate / vsync issues lately. I can't seem to reproduce this on Win 10, so maybe it's a windows 7 only issue. Could you give some insight into what the impact of this is.

Severity: S2 → S3
Flags: needinfo?(ktaeleman) → needinfo?(jnicol)
Priority: -- → P3

I also cannot reproduce on Windows 10. I'm really not familiar at all with the vsync code, but I'm struggling to see where 64 could come from!

@cholismo, would you be able to capture a profile (using https://profiler.firefox.com) when running https://www.vsynctester.com? Thanks.

I have just tested Firefox 68.10 ESR and the behavior is similar but slightly different; At first it detects the display Hz correctly but if you let the test run for a while it turns to 64 Hz eventually.

If you could run mozregression and find out when it changed from working at least for a bit to not working at all, that might give us some useful information!

Flags: needinfo?(jnicol) → needinfo?(cholismo)

(In reply to Jamie Nicol [:jnicol] from comment #14)

I also cannot reproduce on Windows 10. I'm really not familiar at all with the vsync code, but I'm struggling to see where 64 could come from!

@cholismo, would you be able to capture a profile (using https://profiler.firefox.com) when running https://www.vsynctester.com? Thanks.

Sure:
https://share.firefox.dev/3aqHmuS

Sorry for the delay, I haven't had access to this Win7 computer for a few days.

Flags: needinfo?(cholismo)

(In reply to Jamie Nicol [:jnicol] from comment #14)

If you could run mozregression and find out when it changed from working at least for a bit to not working at all, that might give us some useful information!

I don't know how to properly run mozregression to troubleshoot this bug but I found out version 69 runs bad from the beginning so probably there was a critical change between v68 and v69.

comment 4: The complaining website has an article about Firefox: https://www.vsynctester.com/firefoxisbroken.html
It sounds like bug 1630389 fixed vsync for Win10, but not for previous Windows versions.

Blocks: vsync
Summary: Wrong Hz display detection → Win7: Wrong Hz display detection

Animation Timing Deviation is a mess too because of this bug, almost 16 ms. Test here:
https://www.testufo.com/animation-time-graph#scale=20&ppf=2&measure=animation&busywait=0

You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: