https://bugs.kde.org/show_bug.cgi?id=395632
Benjamin Xiao changed:
What|Removed |Added
Status|REPORTED|RESOLVED
Resolution|---
https://bugs.kde.org/show_bug.cgi?id=395632
munzua...@gmail.com changed:
What|Removed |Added
CC|munzua...@gmail.com |
--
You are receiving this mail because:
https://bugs.kde.org/show_bug.cgi?id=395632
munzua...@gmail.com changed:
What|Removed |Added
CC||munzua...@gmail.com
--
You are receiving
https://bugs.kde.org/show_bug.cgi?id=395632
Evstifeev Roman changed:
What|Removed |Added
CC||someuniquen...@gmail.com
--
You are
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #62 from Florian Hänel ---
Created attachment 132786
--> https://bugs.kde.org/attachment.cgi?id=132786=edit
kde neon +nviodia supportinfo
if I disable vsync in ezquake fullscreen I can get the desired fps
--
You are receiving this
https://bugs.kde.org/show_bug.cgi?id=395632
Florian Hänel changed:
What|Removed |Added
CC||he...@gmx.de
--- Comment #61 from Florian
https://bugs.kde.org/show_bug.cgi?id=395632
edgbla changed:
What|Removed |Added
CC||edg...@yandex.ru
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=395632
King Kang Kong changed:
What|Removed |Added
CC||r2b2x3+kde...@gmail.com
--
You are receiving
https://bugs.kde.org/show_bug.cgi?id=395632
hoperidesal...@protonmail.ch changed:
What|Removed |Added
CC||hoperidesalone@protonmail.c
https://bugs.kde.org/show_bug.cgi?id=395632
Postix changed:
What|Removed |Added
CC||pos...@posteo.eu
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=395632
Sergey Kondakov changed:
What|Removed |Added
CC||virtuous...@gmail.com
--- Comment #60 from
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #59 from Kailash ---
(In reply to Benjamin Xiao from comment #58)
> Did the session start at 120fps and you turned it down to 100? That's one of
> the bugs I am experiencing, where Chrome seems to use the refresh rate value
> that was set
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #58 from Benjamin Xiao ---
Did the session start at 120fps and you turned it down to 100? That's one of
the bugs I am experiencing, where Chrome seems to use the refresh rate value
that was set at the beginning of the X session.
--
You
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #57 from Kailash ---
(In reply to Benjamin Xiao from comment #56)
> (In reply to Kailash from comment #54)
> > https://pastebin.com/Mn9Bxgxs
> >
> > Attached support information here.
>
> At least in the Screen 0 section, its reporting
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #56 from Benjamin Xiao ---
(In reply to Kailash from comment #54)
> https://pastebin.com/Mn9Bxgxs
>
> Attached support information here.
At least in the Screen 0 section, its reporting 100 for refresh rate. Is that
what your display is
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #55 from Kailash ---
Also, window transitions are smoother with KWIN_USE_INTEL_SWAP_EVENT set to 0.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #54 from Kailash ---
https://pastebin.com/Mn9Bxgxs
Attached support information here.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #53 from Kailash ---
Tested those patches on top of 5.16 with and without exporting that flag, still
running at a solid 60, support info seems to report refresh rate as 0 for some
reason.
--
You are receiving this mail because:
You are
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #52 from Kailash ---
(In reply to Roman Gilg from comment #51)
> There is
>
> https://phabricator.kde.org/D23105
>
> (as branch
> https://cgit.kde.org/clones/kwin/romangilg/kwin.git/log/?h=glxCleanup)
>
> which is my WIP experimental
https://bugs.kde.org/show_bug.cgi?id=395632
grizzlyuser changed:
What|Removed |Added
CC||grizzlyu...@protonmail.com
--
You are receiving
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #51 from Roman Gilg ---
There is
https://phabricator.kde.org/D23105
(as branch
https://cgit.kde.org/clones/kwin/romangilg/kwin.git/log/?h=glxCleanup)
which is my WIP experimental branch for complete overhaul of compositing. In
the
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #50 from Kailash ---
(In reply to Roman Gilg from comment #49)
> (In reply to Benjamin Xiao from comment #48)
> > (In reply to Roman Gilg from comment #45)
> > > What would help me is if you guys would be on KWin master branch (for
> > >
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #49 from Roman Gilg ---
(In reply to Benjamin Xiao from comment #48)
> (In reply to Roman Gilg from comment #45)
> > What would help me is if you guys would be on KWin master branch (for
> > example via KDE Neon dev unstable) to test
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #48 from Benjamin Xiao ---
(In reply to Roman Gilg from comment #45)
> What would help me is if you guys would be on KWin master branch (for
> example via KDE Neon dev unstable) to test upcoming patches.
>
> I have here an AMD RX470 and a
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #47 from Kailash ---
And yes, I am ready to do whatever I can to help you, testing wise :)
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #46 from Kailash ---
> > This issue isn't RX Vega 64 specific it seems. On my laptop with Intel
graphics, Painting blocks on vertical retrace is also returning No.
>
> Does this mean you noticed low refresh rates on a laptop with high
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #45 from Roman Gilg ---
I'm currently working on the compositing. There have been already several
patches landed as preparations: https://phabricator.kde.org/T11071
What would help me is if you guys would be on KWin master branch (for
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #44 from Kailash ---
(In reply to Martin Flöser from comment #6)
> (In reply to Benjamin Xiao from comment #4)
> > The monitor most definitely supports 144Hz. It's a Acer Predator XB271HU
> > bmiprz 27" WQHD (2560x1440) NVIDIA G-SYNC IPS
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #43 from Kailash ---
(In reply to Benjamin Xiao from comment #42)
> Created attachment 121974 [details]
> KWin info on Intel graphics
>
> Has there been any progress on this, Martin?
>
> This issue isn't RX Vega 64 specific it seems. On
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #42 from Benjamin Xiao ---
Created attachment 121974
--> https://bugs.kde.org/attachment.cgi?id=121974=edit
KWin info on Intel graphics
Has there been any progress on this, Martin?
This issue isn't RX Vega 64 specific it seems. On my
https://bugs.kde.org/show_bug.cgi?id=395632
Kailash changed:
What|Removed |Added
CC||kailash.sudha...@gmail.com
--- Comment #41 from
https://bugs.kde.org/show_bug.cgi?id=395632
petrk changed:
What|Removed |Added
CC||pepk...@gmail.com
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=395632
Nick changed:
What|Removed |Added
CC||futurepi...@gmx.us
--
You are receiving this mail
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #40 from Benjamin Xiao ---
Still running into this issue on Plasma 5.14.5 and Frameworks 5.55. Has there
been any progress on this? It's really ruining the out of the box KDE
experience on gaming displays.
Been doing some more research
https://bugs.kde.org/show_bug.cgi?id=395632
Albin Engström changed:
What|Removed |Added
CC||albin.e.engst...@gmail.com
--
You are
https://bugs.kde.org/show_bug.cgi?id=395632
Scott Deagan changed:
What|Removed |Added
CC||scott.dea...@gmail.com
--
You are receiving
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #39 from Benjamin Xiao ---
After a brief period of using KWIN_TRIPLE_BUFFER=0 to get proper FPS in Kwin,
I've noticed increased micro stuttering in a lot of apps particularly scrolling
in web browsers, which I guess makes sense since we're
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #38 from Benjamin Xiao ---
So it seems like if Kwin thinks that triple buffering is available then it
causes "Painting blocks for vertical retrace" to be no.
-
|
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #37 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #36)
> (In reply to Benjamin Xiao from comment #35)
> > Oh interesting! So when I try disabling KWIN_TRIPLE_BUFFER, Painting blocks
> > for vertical retrace becomes yes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #36 from Martin Flöser ---
(In reply to Benjamin Xiao from comment #35)
> Oh interesting! So when I try disabling KWIN_TRIPLE_BUFFER, Painting blocks
> for vertical retrace becomes yes. Kwin then correctly runs at 144 FPS!!!
>
> But why do
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #35 from Benjamin Xiao ---
Oh interesting! So when I try disabling KWIN_TRIPLE_BUFFER, Painting blocks for
vertical retrace becomes yes. Kwin then correctly runs at 144 FPS!!!
But why do I have to disable KWIN_TRIPLE_BUFFER? Should this be
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #34 from Benjamin Xiao ---
I even tried enabling TearFree in Xorg. It didn't make a difference.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #33 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #32)
> According to the debug output you have vsync set to "Only when cheap",
> please try the default "automatic".
I've never set vsync to anything else besides Automatic.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #32 from Martin Flöser ---
According to the debug output you have vsync set to "Only when cheap", please
try the default "automatic".
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #31 from Martin Flöser ---
Ok, I just re-read the relevant code section.
If you have:
Painting blocks for vertical retrace: no
then the refresh rate is ignored and the maxFps setting is honored, which is by
default 60. So the true
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #30 from Benjamin Xiao ---
Again, I am willing to rebuild kwin on my system with any additional code to
prove it, but I don't think simply printing out the refresh rate variable is
going to cut it. There's something else in play here.
--
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #29 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #28)
> If KWin correctly detects the refresh rate the maxfps setting has no
> influence at all. It's the same code.
If I set MaxFPS=144, there's definitely a difference. I
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #28 from Martin Flöser ---
If KWin correctly detects the refresh rate the maxfps setting has no influence
at all. It's the same code.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #27 from Benjamin Xiao ---
> I definitely feel a difference in fps just dragging the
> mouse around though.
Sorry I meant dragging windows around. Mouse itself is fine.
--
You are receiving this mail because:
You are watching all bug
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #26 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #25)
> Please ignore the showfps effect. It is not a Benchmark. It cannot handle
> 144 Hz. Using it with such a monitor is undefined behavior.
Which way would you recommend
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #25 from Martin Flöser ---
Please ignore the showfps effect. It is not a Benchmark. It cannot handle 144
Hz. Using it with such a monitor is undefined behavior.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #24 from Benjamin Xiao ---
The Show FPS desktop effect shows 75. Also dragging windows around is
significantly more laggy than if I forced 144Hz.
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #23 from Martin Flöser ---
(In reply to Benjamin Xiao from comment #22)
> HOWEVER, Kwin still seems to be running only at 75 FPS instead of 144.
How do you measure this?
--
You are receiving this mail because:
You are watching all bug
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #22 from Benjamin Xiao ---
Created attachment 113957
--> https://bugs.kde.org/attachment.cgi?id=113957=edit
New kwin support information with Xorg 1.20 and Plasma 5.13.3
So I switched to Arch Linux just to see if the issue still occurs
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #21 from Benjamin Xiao ---
(In reply to Christian Martinez from comment #20)
> Forcing KWin to 144 fps only goes to 100 fps.
Btw the Kwin "Show FPS" effect is capped at 100fps. It won't go higher than
that even if Kwin is running at 144.
https://bugs.kde.org/show_bug.cgi?id=395632
Christian Martinez changed:
What|Removed |Added
CC||cmartinez2...@gmail.com
--- Comment #20
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #19 from Benjamin Xiao ---
Let's ignore the Chrome issue for now. There's probably just something screwy
with Chrome itself, because on boot it will get stuck @60Hz on vsynctester.com
but then the moment I launch Steam, it will go to 144Hz.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #18 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #17)
> The fps effect is capped, so will never get the 140 fps. So it looks like on
> startup the resolution is misdetected. Could it be that kscreen changes the
>
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #17 from Martin Flöser ---
The fps effect is capped, so will never get the 140 fps. So it looks like on
startup the resolution is misdetected. Could it be that kscreen changes the
resolution?
--
You are receiving this mail because:
You
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #16 from Benjamin Xiao ---
Okay, this problem gets weirder. I am using the "Show FPS" desktop effect,
which I know isn't a totally accurate benchmark, but for some reason every time
I restart my machine and login, the FPS is capped at 60.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #15 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #13)
> First of all I would like to know whether KWin finds the refresh rate at
> all. And what the refresh rate is which KWin calculates.
At line 90, I logged the refresh
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #14 from Benjamin Xiao ---
(In reply to Christoph Feck from comment #12)
> Are you using X.org 1.20? If not, you probably see
> https://bugs.freedesktop.org/show_bug.cgi?id=103282
I am using 1.19. However that bug is related to Xwayland.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #13 from Martin Flöser ---
(In reply to Benjamin Xiao from comment #11)
> (In reply to Martin Flöser from comment #8)
> > Our calculation is here:
> > https://cgit.kde.org/kwin.git/tree/plugins/platforms/x11/standalone/
> >
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #12 from Christoph Feck ---
Are you using X.org 1.20? If not, you probably see
https://bugs.freedesktop.org/show_bug.cgi?id=103282
--
You are receiving this mail because:
You are watching all bug changes.
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #11 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #8)
> Our calculation is here:
> https://cgit.kde.org/kwin.git/tree/plugins/platforms/x11/standalone/
> screens_xrandr.cpp#n79
>
> I can only assume that your output is
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #10 from Benjamin Xiao ---
(In reply to Roman Gilg from comment #9)
> I have noticed that on my Xserver compile from master branch games run in
> the X session locked at 60 fps, while they run unlocked through Xwayland.
> This might be
https://bugs.kde.org/show_bug.cgi?id=395632
Roman Gilg changed:
What|Removed |Added
CC||subd...@gmail.com
--- Comment #9 from Roman Gilg
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #8 from Martin Flöser ---
Our calculation is here:
https://cgit.kde.org/kwin.git/tree/plugins/platforms/x11/standalone/screens_xrandr.cpp#n79
I can only assume that your output is doing something we don't cover yet. But
as you say: without
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #7 from Benjamin Xiao ---
(In reply to Martin Flöser from comment #6)
>
> Well we do exactly that. We use the xrandr API.
>
> It's btw. not that surprising that it works on Wayland. There we interact
> more directly with the API.
Thanks
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #6 from Martin Flöser ---
(In reply to Benjamin Xiao from comment #4)
> The monitor most definitely supports 144Hz. It's a Acer Predator XB271HU
> bmiprz 27" WQHD (2560x1440) NVIDIA G-SYNC IPS Display.
>
> xrandr also shows that I am on
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #5 from Benjamin Xiao ---
Created attachment 113524
--> https://bugs.kde.org/attachment.cgi?id=113524=edit
kwin wayland support info
Funny enough, if I run KDE with Wayland, it gets the correct refresh rate from
the monitor. I've
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #4 from Benjamin Xiao ---
The monitor most definitely supports 144Hz. It's a Acer Predator XB271HU bmiprz
27" WQHD (2560x1440) NVIDIA G-SYNC IPS Display.
xrandr also shows that I am on 144Hz as well. How come KWin doesn't just check
xrandr
https://bugs.kde.org/show_bug.cgi?id=395632
Martin Flöser changed:
What|Removed |Added
Component|compositing |platform-x11-standalone
Flags|
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #2 from Benjamin Xiao ---
Created attachment 113452
--> https://bugs.kde.org/attachment.cgi?id=113452=edit
kwin support information
The kwin support information seems to list my screen as 60Hz? This is strange
since I am clearly running
https://bugs.kde.org/show_bug.cgi?id=395632
--- Comment #1 from Benjamin Xiao ---
Created attachment 113451
--> https://bugs.kde.org/attachment.cgi?id=113451=edit
xrandr output
--
You are receiving this mail because:
You are watching all bug changes.
75 matches
Mail list logo
