Bug 356446 - [Meta] Enable Qt scaling on X11
Summary: [Meta] Enable Qt scaling on X11
Alias: None
Product: plasmashell
Classification: Plasma
Component: general (show other bugs)
Version: 5.5.0
Platform: Other Linux
: VHI normal
Target Milestone: 1.0
Assignee: Nate Graham
: 366088 391682 408055 420860 424465 445526 450032 451193 452643 457802 (view as bug list)
Depends on: 422559 423390 389448 407458 413673 422578 423563 428514 446627 450443 456168 456453 457446 457814 461022 471480
  Show dependency treegraph
Reported: 2015-12-09 19:09 UTC by Lukas Schneiderbauer
Modified: 2023-11-29 21:47 UTC (History)
45 users (show)

See Also:
Latest Commit:
Version Fixed In: 6.0

screenshot (215.60 KB, image/png)
2015-12-09 19:10 UTC, Lukas Schneiderbauer
Scaling works, a good eyecandy. (353.60 KB, image/png)
2016-12-12 10:55 UTC, Tatsuyuki Ishi
Option in KScreen.png (66.54 KB, image/png)
2022-02-10 03:10 UTC, Fushan Wen

Note You need to log in before you can comment on or make changes to this bug.
Description Lukas Schneiderbauer 2015-12-09 19:09:02 UTC
I'm using plasma with a 4k-montor.
The system tray settings dialog (reachable through right-clicking the system tray) seems to not adjust to the hidpi settings in contrast to other plasma apps.
I will attach an screenshot that shows this. In particular, the checkboxes have different sizes.

Reproducible: Always
Comment 1 Lukas Schneiderbauer 2015-12-09 19:10:26 UTC
Created attachment 95957 [details]
Comment 2 Kai Uwe Broulik 2015-12-21 23:34:22 UTC
Support for this is currently deliberately disabled as it causes some breakage in plasmashell. There's a bug about it but I cannot find it right now.
Comment 3 David Edmundson 2015-12-23 15:27:01 UTC
from main.cpp

//    Devive pixel ratio has some problems in plasmashell currently.
//     - dialog continually expands (347951)
//     - Text element text is screwed (QTBUG-42606)
//     - Panel struts (350614)
//  This variable should possibly be removed when all are fixed


The KDE ones are fixed. 
The Qt one is not listed as fixed.
Comment 4 Kai Uwe Broulik 2015-12-27 14:45:53 UTC
Text rendering looks fine for me on Qt 5.5.1? Only issues I can still see so far are: Task Manager label being blurry (its custom rendering doesn't take dpr into account) and minimize task geometry being off, wallpaper blurry. I think we should reconsider that for 5.6? Or is the Qt bug driver-dependent?
Comment 5 David Edmundson 2015-12-27 19:06:23 UTC
Text rendering seems fixed now. Thanks for checking.

Seems like we can fix this for Plasma 5.6 then \o/
Comment 6 Marco Martin 2016-02-04 13:03:11 UTC
updates on this?
Comment 7 David Edmundson 2016-07-07 13:57:55 UTC
there's a Qt bug in Text with nativeRendering where it will get the kerning completely mental.
Comment 8 Kai Uwe Broulik 2016-08-03 12:45:51 UTC
*** Bug 366088 has been marked as a duplicate of this bug. ***
Comment 9 Jan Wiele 2016-09-28 20:51:08 UTC
Is this bug also responsible for non-scaling icons (wifi, klipper, etc.) in the systray or should I open/search for a new/another one?
Comment 10 David Edmundson 2016-09-28 21:05:44 UTC
that's unrelated
Apparently it's intended - it's been discussed on here somewhere
Comment 11 David Edmundson 2016-11-03 13:07:14 UTC
*** Bug 372013 has been marked as a duplicate of this bug. ***
Comment 12 Tatsuyuki Ishi 2016-12-12 10:54:40 UTC
All bugs listed in the comment are resolved, there should be nothing blocking this. I have successfully rebuilt one that works.

Please re-enable the scaling ASAP.
Comment 13 Tatsuyuki Ishi 2016-12-12 10:55:29 UTC
Created attachment 102741 [details]
Scaling works, a good eyecandy.
Comment 14 David Edmundson 2016-12-12 11:45:29 UTC
If only it were that simple, there's still Qt bugs to do with text spacing. 
I even ended up disabling QT_DEVICE_PIXEL_RATIO in more Plamsa places: See 366451

Also I don't get what screenshot is meant to show, everything there should look exactly the same. It's "only" context menus and config modules that would change.
Comment 15 Tatsuyuki Ishi 2016-12-12 12:28:50 UTC
Poor, readable text is still better than unreadable tiny text.

As far there is no issues I can notice with plasmashell, except the bottommost pixel of the taskbar doesn't respond to click.
Comment 16 David Edmundson 2016-12-12 13:36:12 UTC
The text isn't currently tiny...we still have the font DPI set.

I think you have some sort of different issue/setup to the one this report is talking about.
Comment 17 Tatsuyuki Ishi 2016-12-13 02:51:37 UTC
Big text in 1x widget is worse: inconsistent, breaks more layout.

A big taskbar is quite needed for readability and comfortness; when it comes to the mouse, it's hard to focus on such a small area.
Comment 18 David Edmundson 2016-12-22 16:19:13 UTC
Your taskbar won't be small if your font DPI is also set in xrdb, which we set when you configure your screen in the kscreen KCM.
Comment 19 David Edmundson 2017-07-09 15:48:51 UTC
*** Bug 382146 has been marked as a duplicate of this bug. ***
Comment 20 David Edmundson 2017-08-07 18:54:45 UTC
Git commit bbf7aaf3448532b38819456fbe88ed866fb6fec0 by David Edmundson.
Committed on 07/08/2017 at 18:54.
Pushed by davidedmundson into branch 'master'.

Use Qt scaling in Plamsa

I've done this only when on Wayland because:

On X, our own scaling to font generally works well.
There are bugs on older Qt, especially when it comes to fractional
scaling, and we can't support per screen scales anyway.
There's advantages to disabling, and (other than small context menu
icons) few advantages in enabling

On wayland, it's different. Everything is simpler as all co-ordinates
are in the same co-ordinate system
we don't have fractional scaling on the client so don't hit most the
remaining bugs and
even if we don't use Qt scaling the compositor will try to scale us
anyway so we have no choice

Test Plan:
Had this for about a week on my laptop.
Haven't noticed a single problem.

Reviewers: #plasma, mart

Reviewed By: #plasma, mart

Subscribers: mart, plasma-devel

Tags: #plasma

Differential Revision: https://phabricator.kde.org/D7188

M  +12   -8    shell/main.cpp

Comment 21 Peter Eszlari 2017-12-12 23:09:50 UTC
(In reply to David Edmundson from comment #14)
> If only it were that simple, there's still Qt bugs to do with text spacing. 
> I even ended up disabling QT_DEVICE_PIXEL_RATIO in more Plamsa places: See
> 366451

What are the Qt upstream bug numbers? Are these fixed in Qt 5.10?

Maybe you can provide a environment variable for testing?
Comment 22 David Edmundson 2018-01-22 10:54:04 UTC
>What are the Qt upstream bug numbers? 

You can search as well as I can.

>Are these fixed in Qt 5.10?

Some things are. See other scaling threads on here.

>Maybe you can provide a environment variable for testing?

There is now in 5.12


on will repsect QT_DEVICE_PIXEL_RATIO, unset will be the current X behaviour.

I'm not doing anything else on this issue, closing this bug.
Comment 23 Jason A. Donenfeld 2018-01-22 11:00:47 UTC
Wow, setting this environment variable significantly improves the look and feel of Plasma as a whole. Thanks for that.
Comment 24 Peter Eszlari 2018-01-23 00:01:54 UTC
(In reply to David Edmundson from comment #22)
> I'm not doing anything else on this issue, closing this bug.

Does this mean, this will never become the default for X11? After some testing, I could find any problems with it and it improves the look of plasma greatly. (It seems not to apply to kwin's context menu icons, should I file new bug for that?)
Comment 25 David Edmundson 2018-01-23 10:03:42 UTC
>Does this mean, this will never become the default for X11? 

Not necessarily.

> should I file new bug for that?

No, that is simply fixable under X. 
On wayland that does works fine.

You can find an existing bug in kwin.
Comment 26 David Edmundson 2018-01-23 10:51:57 UTC
*simply not fixable
Comment 27 David Edmundson 2018-02-11 23:34:35 UTC
*** Bug 389813 has been marked as a duplicate of this bug. ***
Comment 28 David Edmundson 2018-03-10 23:25:46 UTC
*** Bug 391682 has been marked as a duplicate of this bug. ***
Comment 29 battaglia01 2018-03-11 04:46:10 UTC
I was sent here from Bug 391682 (https://bugs.kde.org/show_bug.cgi?id=391682)

Just want to chime in as well my support for setting PLASMA_USE_QT_SCALING=1 by default. FWIW, setting this thing improves the entire experience for me so much that I would not even consider using KDE at all on a HiDPI monitor without it -- tiny icons, weird rendering issues, etc. I am *very* happy to discover this environment variable as it fixes pretty much everything.

Earlier in the thread it was mentioned that this is not a default setting because "there's advantages to disabling" and "few advantages to enabling." I'm curious what sorts of things tend to "break" if this is enabled? I haven't run into anything yet, but I also haven't had this set for too long.
Comment 30 Nate Graham 2018-03-11 20:00:48 UTC
A major issue with setting PLASMA_USE_QT_SCALING=1 is that it regresses the appearance of text with non-integer scale factord, compared to not setting it. I have just submitted a patch that remedies that, though: https://phabricator.kde.org/D11244

If that lands, hopefully we will be one step closer to being able to turn on PLASMA_USE_QT_SCALING=1 by default for our X users.
Comment 31 Nate Graham 2018-03-16 22:27:22 UTC
Git commit a21bc11fe11651f6d211489ebfc8435cef877194 by Nathaniel Graham.
Committed on 16/03/2018 at 22:26.
Pushed by ngraham into branch 'master'.

Fix text scaling with non-integer scale factors when PLASMA_USE_QT_SCALING=1 is set

When `PLASMA_USE_QT_SCALING=1` is set, Plasma uses native Qt scaling. This works fine for integer scale factors, and fixes a lot of bugs (see [[https://bugs.kde.org/show_bug.cgi?id=356446|Bug 356446]]) but it introduces a new one: with non-integer scale factors, text becomes blurry and pixellated because of a bug in `Text.NativeRendering`: https://bugreports.qt.io/browse/QTBUG-67007

QQC2-desktop-style forces the use of `Text.QtRendering` rendering for non-integer scale factors, successfully working around the problem. But PlasmaComponents QML objects don't implement the same workaround, so we see the issue in Plasma. This patch fixes that, and gets us one step closer to being able to use Qt scaling in Plasmashell.

There is no effect when `PLASMA_USE_QT_SCALING=1` is not being used.

Related: bug 391691, bug 384031, bug 386216, bug 391695, bug 391694, bug 385547, bug 391692

Test Plan:
Before: `PLASMA_USE_QT_SCALING=1` set, 1.2 scale factor: Plasma text looks awful:

After: `PLASMA_USE_QT_SCALING=1` set, 1.2 scale factor: Plasma text looks amazing!

Note that we still get sub-pixel anti-aliasing and good kerning. There appear to be no layout regressions.

Without both `PLASMA_USE_QT_SCALING=1` and a non-integer scale factor set, there is no visual change compared to the status quo.

Reviewers: #plasma, davidedmundson

Reviewed By: #plasma, davidedmundson

Subscribers: mart, broulik, #frameworks

Tags: #frameworks

Differential Revision: https://phabricator.kde.org/D11244

M  +4    -2    examples/applets/testtheme/contents/ui/FontGizmo.qml
M  +5    -1    src/declarativeimports/plasmacomponents/qml/Label.qml
M  +5    -1    src/declarativeimports/plasmacomponents/qml/private/DualStateButton.qml
M  +4    -0    src/declarativeimports/plasmacomponents3/ComboBox.qml
M  +5    -1    src/declarativeimports/plasmacomponents3/Label.qml
M  +5    -1    src/declarativeimports/plasmacomponents3/TextArea.qml
M  +5    -1    src/declarativeimports/plasmacomponents3/TextField.qml
M  +5    -0    src/declarativeimports/plasmastyle/ComboBoxStyle.qml
M  +4    -1    src/declarativeimports/plasmastyle/SpinBoxStyle.qml
M  +4    -1    src/declarativeimports/plasmastyle/TextAreaStyle.qml
M  +4    -1    src/declarativeimports/plasmastyle/TextFieldStyle.qml

Comment 32 Nate Graham 2018-03-21 02:20:53 UTC
David, with your Qt fix that's landing in 5.11, what more will be needed to enable this behavior by default?
Comment 33 David Edmundson 2018-03-21 02:59:20 UTC
1) We need to depend on Qt 5.11, which is going to be ~Plamsa 5.14

2) blur rects in KwindowSystem. 

3) testing, as no-one has mentioned 2 yet!
Comment 34 Nate Graham 2018-03-21 03:33:46 UTC
Thanks David! What's the bug tracking #2?
Comment 35 Nate Graham 2018-04-14 14:49:00 UTC
*** Bug 393131 has been marked as a duplicate of this bug. ***
Comment 36 David Edmundson 2018-04-30 09:27:53 UTC
*** Bug 393666 has been marked as a duplicate of this bug. ***
Comment 37 David Edmundson 2018-05-29 12:25:27 UTC
See other blocking bug about minimise positions.
Comment 38 David Edmundson 2018-10-24 22:47:36 UTC
*** Bug 400269 has been marked as a duplicate of this bug. ***
Comment 39 Nate Graham 2018-12-16 19:43:28 UTC

Is it later yet? :) Can we re-open this, or else just close it INTENTIONAL if we don't intend to do it?
Comment 40 Nate Graham 2019-06-21 09:21:58 UTC
*** Bug 408929 has been marked as a duplicate of this bug. ***
Comment 41 Peter Eszlari 2019-06-21 15:00:50 UTC
I'm using PLASMA_USE_QT_SCALING=1 now for over a year on X11 and I don't see any problems with it. Maybe it's because I'm using an integer scale (2x)? Could it be made the default for this case?
Comment 42 Yan 2019-06-21 15:06:07 UTC
I'm using 1.2 scale
Comment 43 Nate Graham 2019-06-23 15:24:01 UTC
We discussed this at the Plasma sprint and came to the following conclusion:

We will re-investigate doing this, but first need to resolve the following blockers:
- Fix the Task Manager's minimize animation zooming into out of the wrong location
- A few other Task Manager-related positioning and screen-related issues

We have a path forward and the blocking issues will be marked appropriately.
Comment 44 Nate Graham 2019-06-23 16:52:23 UTC
*** Bug 381794 has been marked as a duplicate of this bug. ***
Comment 45 Pierpaolo Valerio 2019-08-06 08:48:31 UTC
Nate, you probably already know this, but I just wanted to report the bugs I found when enabling the PLASMA_USE_QT_SCALING variable.
It works fine for the most part, but the pager applet is wonky (at least visually, as the preview area represents a smaller area than the actual screen) and latte-dock's new feature to track the active window across screens completely breaks, as it starts considering only a tiny fraction of the screen and fills the rest with a "Plasma desktop" app that it shouldn't track.

I hope it helps checking whether you're ready to flip the switch on this feature.
Comment 46 Nate Graham 2019-08-06 17:19:58 UTC
*** Bug 410609 has been marked as a duplicate of this bug. ***
Comment 47 phrxmd 2020-06-14 15:29:41 UTC
If it helps, I just moved from Kubuntu 20.04 to OpenSUSE Tumbleweed and had to enable PLASMA_USE_QT_SCALING=1 manually to get the Plasma icon scaling right - this is with Plasma 5.19, Qt 5.15, Frameworks 5.70, on a Thinkpad X1 Yoga with a 2560x1440 resolution. On Ubuntu it looked right for some reason, but every distro seems to depend on its own little HiDPI hacks to make KDE look right.

Are there any showstoppers for making this the default? So far it seems to work fine. I don't use the Task Manager, but what does not work for me is the text in the virtual desktop switcher which is unscaled, but that might be an unrelated bug - I don't know how to take a screenshot of that (Spectacle refuses to capture the virtual desktop switcher), otherwise I'd open a separate bug for it.
Comment 48 Luca Carlon 2020-06-28 09:29:56 UTC
I wouldn't even use Plasma without PLASMA_USE_QT_SCALING=1. Is there an updated list with all the tasks blocking this?
Comment 49 Luca Carlon 2020-06-28 11:20:19 UTC
Stupid question. "Depends on" is probably what I was looking for, sorry.
Comment 50 Nate Graham 2020-06-28 15:09:17 UTC
Yep, I'm using that to track the remaining issues. I'm also trying to fix all of them. :)
Comment 51 Bug Janitor Service 2020-07-04 03:53:34 UTC
A possibly relevant merge request was started @ https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/139
Comment 52 Nate Graham 2020-07-23 19:04:45 UTC
*** Bug 424465 has been marked as a duplicate of this bug. ***
Comment 53 Nate Graham 2020-09-29 21:10:12 UTC
*** Bug 420132 has been marked as a duplicate of this bug. ***
Comment 54 Nate Graham 2020-10-26 15:25:10 UTC
*** Bug 428138 has been marked as a duplicate of this bug. ***
Comment 55 Nate Graham 2020-11-09 17:55:45 UTC
*** Bug 426978 has been marked as a duplicate of this bug. ***
Comment 56 Nate Graham 2020-11-13 18:32:51 UTC
*** Bug 429061 has been marked as a duplicate of this bug. ***
Comment 57 Nate Graham 2020-12-17 19:42:01 UTC
*** Bug 408055 has been marked as a duplicate of this bug. ***
Comment 58 Nate Graham 2021-01-08 21:05:25 UTC
*** Bug 420860 has been marked as a duplicate of this bug. ***
Comment 59 Nate Graham 2021-06-07 16:52:59 UTC
*** Bug 437243 has been marked as a duplicate of this bug. ***
Comment 60 Nate Graham 2021-07-26 18:33:25 UTC
The case of huge icons being too large will be fixed by https://invent.kde.org/frameworks/plasma-framework/-/merge_requests/300.
Comment 61 Nate Graham 2021-07-26 20:17:35 UTC
Git commit 32bfd1ed01f0cbd48071dadf89458edbb8edfd1e by Nate Graham.
Committed on 26/07/2021 at 18:31.
Pushed by ngraham into branch 'master'.

Handle huge icons in Units::roundToIconSize()

This function was missing a use case for icons that are between the
sizes of huge and enormous, causing huge-sized icons to appear too large
on X11 when not using Qt scaling. This commit fixes that, and addresses
one of the sources of complaints when using scaling on X11. Small icons
are still too small, but that must and will be handled separately.

M  +3    -0    src/declarativeimports/core/units.cpp

Comment 62 Nate Graham 2021-07-26 21:04:27 UTC
All of the issues that are possible to fix on X11 without turning on Qt scaling have been un-duped and I am in the process of fixing them. See the open bug reports in the See Also field for details. As a result, I don't believe we will need to ever use Qt scaling on X11 to achieve a pleasant high DPI experience on X11. Turning on Qt scaling on X11 simply causes too many bugs compared to the number that it fixes. Closing this.
Comment 63 Nate Graham 2021-07-28 16:38:24 UTC
*** Bug 424465 has been marked as a duplicate of this bug. ***
Comment 64 Nate Graham 2021-07-28 16:38:26 UTC
*** Bug 391682 has been marked as a duplicate of this bug. ***
Comment 65 Nate Graham 2021-07-28 16:38:59 UTC
*** Bug 366088 has been marked as a duplicate of this bug. ***
Comment 66 Nate Graham 2021-07-28 16:41:18 UTC
Re-opening since upon further investigation and consultation with colleagues, the duplicate bugs are impossible to fix unless we turn on Qt scaling. Our alternatives are:

1. Live with them being broken forever; not really desirable
2. Have everyone use Wayland; this is ultimately the goal, but it's not ready yet

So I think we need to find a way to do this in the meantime. I will adjust the blocking bugs to show the things we still need to do.
Comment 67 Nate Graham 2021-07-28 16:42:15 UTC
*** Bug 408055 has been marked as a duplicate of this bug. ***
Comment 68 Nate Graham 2021-07-28 16:58:00 UTC
OK, our two open blockers are Bug 422559 and Bug 440362.
Comment 69 postix 2021-11-23 19:48:54 UTC
*** Bug 445526 has been marked as a duplicate of this bug. ***
Comment 70 Nate Graham 2022-01-21 05:08:53 UTC
It's pretty clear that this isn't going to happen, unfortunately.

Breaking mixed-DPI screen arrangements is a hard blocker because we know people are using these on X11. There's no practical way to fix it on X11, and it already works on Wayland, so there is little chance it will ever get fixed on X11, which means the overall effort will never get unblocked, which means it can't be done.

Unfortunately I think we have to close this. Folks who want a decent hidpi experience will have to use Wayland and look forward to upcoming changes to improve the experience for XWayland apps in particular.

Sorry everyone. :(
Comment 71 Fushan Wen 2022-02-10 03:10:26 UTC
Created attachment 146517 [details]
Option in KScreen.png

Even if the bug is marked INTENTIONAL, KScreen can still provide an option and warn the user of the consequence of enabling the option.
Comment 72 Nate Graham 2022-02-10 03:25:28 UTC
You know, that's not a bad idea. Maybe we should do that.
Comment 73 Fushan Wen 2022-02-10 06:43:18 UTC
I sent a mail to plasma-devel and visual-design mailing lists. See https://mail.kde.org/pipermail/visual-design/2022-February/004315.html
Comment 74 Nate Graham 2022-02-15 20:30:56 UTC
*** Bug 450032 has been marked as a duplicate of this bug. ***
Comment 75 Fushan Wen 2022-02-20 10:35:48 UTC
See also https://invent.kde.org/plasma/plasma-desktop/-/issues/34
Comment 76 Nate Graham 2022-03-22 03:34:25 UTC
*** Bug 451193 has been marked as a duplicate of this bug. ***
Comment 77 Éric Brunet 2022-03-22 11:03:39 UTC
I am the reporter of bug 451193, which has just been marked as a duplicate of this bug. I have read all the comments of this bug, and I don't think this addresses the main message of 451193, which is:

plasma is behaving differently under Xorg depending on the low-level graphics driver (Intel or modesetting, in my case)

I think changing the low-level driver can change performance and/or stability, but not the overall look of the Desktop.

In particular, I am very happy with the look of plasma using the Intel driver and PLASMA_USE_QT_SCALING=1, and I wish that it would look as good with the modesetting driver, but unfortunately PLASMA_USE_QT_SCALING has no effect for me with the modesetting driver.

Should I reopen 451193 ?

BTW, I can't hide I had hoped a solution would be found for Xorg, but of course there is a limited amount of time available, and I wanted to thank you for all the work done.
Comment 78 Nate Graham 2022-03-23 22:57:06 UTC
If your issue was purely about differences in scaling relating to modesetting, feel free to un-dupe it. I might have gotten confused because you mentioned PLASMA_USE_QT_SCALING.
Comment 79 Nate Graham 2022-04-16 19:20:56 UTC
*** Bug 452643 has been marked as a duplicate of this bug. ***
Comment 80 Fushan Wen 2022-05-13 08:15:22 UTC
Git commit ecc8b628f9f7fe68f6f3bc67ffd6d474cf562a44 by Fushan Wen.
Committed on 13/05/2022 at 07:25.
Pushed by fusionfuture into branch 'master'.

containments/panel: bind currentApplet's position to mouse position

Do not use difference as the difference doesn't accurately represent the
relative moving distance when Qt scaling is enabled on X11.

Ref https://invent.kde.org/plasma/plasma-desktop/-/issues/34

M  +9    -6    containments/panel/contents/ui/ConfigOverlay.qml

Comment 81 Nate Graham 2022-07-13 17:23:33 UTC
As I see it, we have only three issues to fix before we can enable this by default:
- Bug 428514
- Bug 422559
- Bug 423390

I'd love to turn it on by default and then remove Plasma scaling entirely, so we're always using a single codepath for our scaling. It would get rid of a lot of complexity and weird bugs.
Comment 82 Nate Graham 2022-08-12 19:26:36 UTC
*** Bug 457802 has been marked as a duplicate of this bug. ***
Comment 83 pktriss03 2022-08-12 20:46:11 UTC
(In reply to Nate Graham from comment #81)

One issue I'd like to add is that the desktop editing toolbar that displays on the top (one that says "add desktop widgets", "Desktop settings", "Change global theme"...) is off-center with PLASMA_USE_QT_SCALING=1, sorry if that's not the best place to report this but it's relatively minor for what it is
Comment 84 Nate Graham 2022-08-12 21:08:17 UTC
Can you file another bug report for that, and mark this bug as a blocker?
Comment 85 pktriss03 2022-08-12 21:23:53 UTC
(In reply to Nate Graham from comment #84)
> Can you file another bug report for that, and mark this bug as a blocker?

Will do
Comment 86 Bug Janitor Service 2022-10-15 12:03:18 UTC
A possibly relevant merge request was started @ https://invent.kde.org/frameworks/kirigami/-/merge_requests/771
Comment 87 Nate Graham 2022-10-17 14:24:52 UTC
Git commit 0ac3033c1f688a593e64dd0213f1a1c0b67e2ea0 by Nate Graham, on behalf of Fushan Wen.
Committed on 17/10/2022 at 14:23.
Pushed by ngraham into branch 'master'.

Avatar: fix blurry avatar when using Qt scaling

sourceSize should be multiplied by devicePixelRatio to represent the
real size.

M  +3    -2    src/controls/Avatar.qml

Comment 88 Nate Graham 2023-06-22 12:21:41 UTC
Git commit 85014ec0aa0bc33470108da4366a8fcbd94b8f5d by Nate Graham, on behalf of Marco Martin.
Committed on 22/06/2023 at 12:21.
Pushed by ngraham into branch 'master'.

remove the force qt scaling env var

Now we use Qt scaling everywhere, to reduce the number of different code
paths and visual differences between the X11 and Wayland sessions. In the
past we didn't do this because there were too many bugs, but almost all
of them have been fixed now.

The one thing this is known to regress is the mixed-DPI multimonitor use
case. But that use case is already not supported on X11, so our story
remains the same: if you have a mixed-DPI multimonitor setup, use Wayland.

M  +0    -3    krunner/main.cpp
M  +0    -11   shell/main.cpp
M  +0    -4    shell/tests/screenpooltest.cpp
M  +0    -1    startkde/startplasma-wayland.cpp

Comment 89 Nate Graham 2023-11-29 21:47:04 UTC
*** Bug 477636 has been marked as a duplicate of this bug. ***