Qt 5.9.3 Released With Fixes & Performance Improvements

[phoronix]

Phoronix: Qt 5.9.3 Released With Fixes & Performance Improvements

The Qt Company has issued Qt 5.9.3 as the latest tool-kit update in the Qt 5.9 Long-Term Support series...

http://www.phoronix.com/scan.php?page=news_item&px=Qt-5.9.3-Released

[aufkrawall]

More than hundred fixes, but still butchered fonts in various UIs when scrolling:

https://bugreports.qt.io/browse/QTBUG-59261

Nice priorities...

[smartalgorithm]

interesting if this solves any crash issue in KDE Plasma... having a crash after clicking on an icon is horrible...

[cipri]

More than hundred fixes, but still butchered fonts in various UIs when scrolling:

https://bugreports.qt.io/browse/QTBUG-59261

Nice priorities...

I'm happy to see that also others are talking about the many bugs of QT. There are so many bugs, and I don't understand why not that many understand that QT is going to disappear if they continue like this. Sadly there are not yet good alternatives, but if they continue like this with introducing all the time new pre-alpha stuff with a tones of bugs, than I cannot imagine that QT has a bright future. And KDE should have left QT many years ago, like this they continue being hold back but the Qt bugs.

[R41N3R]

I'm happy to see that also others are talking about the many bugs of QT. There are so many bugs, and I don't understand why not that many understand that QT is going to disappear if they continue like this. Sadly there are not yet good alternatives, but if they continue like this with introducing all the time new pre-alpha stuff with a tones of bugs, than I cannot imagine that QT has a bright future. And KDE should have left QT many years ago, like this they continue being hold back but the Qt bugs.

KDE and Qt have a long and fruitful relationship and many things in Qt originated in KDE. What I dislike is that several bugs that affect KDE badly are not fixed with priority and often take many released to be addressed. Qt should focus again on its core, close bugs and assign more resources to critical areas.

[rtfazeberdee]

I'm happy to see that also others are talking about the many bugs of QT. There are so many bugs, and I don't understand why not that many understand that QT is going to disappear if they continue like this. Sadly there are not yet good alternatives, but if they continue like this with introducing all the time new pre-alpha stuff with a tones of bugs, than I cannot imagine that QT has a bright future. And KDE should have left QT many years ago, like this they continue being hold back but the Qt bugs.

Show us a toolkit that is bug free on first release or even in later releases, unfortunately utopia doesn't exist. My experience is fine with Plasma, i've not come across a show stopper and i'm not using the LTS releases. From the content of your post, I almost thought it was our favourite gnome troll was back stirring nonsense.

[R41N3R]

Show us a toolkit that is bug free on first release or even in later releases, unfortunately utopia doesn't exist. My experience is fine with Plasma, i've not come across a show stopper and i'm not using the LTS releases. From the content of your post, I almost thought it was our favourite gnome troll was back stirring nonsense.

I do agree, KDE and Plasma on X11 is stable. The Qt bugs affect especially Wayland and the Qt releases have been a big up and down. At least this is my personal experience as a Plasma-Wayland user.

[carewolf]

More than hundred fixes, but still butchered fonts in various UIs when scrolling:

https://bugreports.qt.io/browse/QTBUG-59261

Nice priorities...

That is a libinput bug, just use evdev drivers or stop relying on pixeldeltas that were completely platform specific in the first place.

[cipri]

KDE and Qt have a long and fruitful relationship and many things in Qt originated in KDE. What I dislike is that several bugs that affect KDE badly are not fixed with priority and often take many released to be addressed. Qt should focus again on its core, close bugs and assign more resources to critical areas.

exactly! The thing is, I also had hopes, but Qt has shown in the last years, that the chances to focus on the core and fixing important bugs, are very little. KDE is risking to be hold back by Qt for longer time, till in the end KDE might come to the same conclusion that they need to split up wit Qt. I guess for Qt are important the paying clients, and if they ask for QML improvements then that's it. Dealing with the bugs of a complex project, waiting for bug fixes, working with foreign undocumented code can be a lot more time intensive than writing it yourself from scratch, I can tell you this from my own experience. I spent a lot of time with Qt3D till i noticed it's simply too buggy, bad performance and badly documented. Nearly all of the examples where in QML and not c++. And then because of the lack of documentation I had always the problem not knowing if the crash/problem happened because of me not understanding completely well how it needs to be done, or it's really a bug. And after a lot of time I came to discover a lot of bugs, and many very so obvious that I could not explain myself how the developers of Qt3D didn't see them before. And Qt3D is developed, I think, for more than 5 years!
With less effort and time spent on Qt3D I will have soon my own Qt3D replacement based on Vulkan, and less buggy and better performance.
I wished I would have split with Qt a lot earlier.

[cipri]

Show us a toolkit that is bug free on first release.

I can show you a toolkit that has horrible bugs, flagged as Important or Critical, and which were not even touched in years. And bugs related to the Android Platform, where you would have expected that they might care about. If you find bugs like this related to the UI in Haiku, and you reported them, it's a matter of days till it's fixed and a new nightly build is available for testing.
Then look at this bug, reported like a year ago: "Qt3DExtras::Qt3DWindow crashes when deleted" and flagged as "Critical". It seems it's still not fixed.

https://bugreports.qt.io/browse/QTBUG-58257

But the big question is: how is it possible that you introduce a class that crashes the application on destruction of the object? Is this completely untested code or what?