KWinFT 5.20 With Aims For Better Wayland/X11 Experience Than KDE Plasma 5.20's KWin

[R41N3R]

Thank you very much for testing.

I'm somewhat ashamed there was still a crash when switching off the display with the power button. It would be great if you can open an issue ticket for that in KWinFT including a backtrace. I've written some guide recently how to get such a backtrace: Logging and Debugging KWinFT

I will try to get a backtrace soon. Thanks for the guide, I'm sure I will learn something new!

[dragon321]

I wonder how it compare with kwin-lowlatency. Also I hope it will back fullscreen undirect support because stopping whole compositor isn't prettiest solution.

[darkbasic]

So glad KwinFT is moving forward.

Me too, I'm finally getting some hope that Plasma will work decently on Wayland one day...

[HighValueWarrior]

So glad KwinFT is moving forward.
Add me to those who want to see it get merged.
Thank you for the work.

[romangg]

I've just compiled kwinft-git and I've expected that the subsurface issue would be worse (Firefox webrender) on Wayland. But yes, it needs some work before I can switch my main system to KwinFT. As KWin/Plasma is crashing when turning off my display with the power button, I tried that with KWinFT, but I was kicked back to SDDM as well. I will test KWinFT now a little more to see if I can spot some differences, next I want to try some games :-)

Anyway I'm glad to see all these low level improvements in KwinFT and I'm sure this is the right way. So thank you for your great work!

Thank you very much for testing.

I'm somewhat ashamed there was still a crash when switching off the display with the power button. It would be great if you can open an issue ticket for that in KWinFT including a backtrace. I've written some guide recently how to get such a backtrace: Logging and Debugging KWinFT

[Mario Junior]

I totally agree on this!

What are they waiting
for?

Maybe because some KDE developers are a bunch of ass*oles?

>NOOOOOO YOU CAN'T FORK KWIN AND MADE THIS BETTER
>WE DON'T MERGE YOUR CODE 😤😡

[R41N3R]

Yes, that is true. I did improve subsurface handling in the past already but there is still some stuff to do. I am currently looking into improvements to KWinFT's internal window representation so that might be an opportunity to check back on subsurfaces. But it might also be necessary to work on the rendering pipeline more for that, which is a project I have planned to do afterwards.

Sadly there is so much to do.

I've just compiled kwinft-git and I've expected that the subsurface issue would be worse (Firefox webrender) on Wayland. But yes, it needs some work before I can switch my main system to KwinFT. As KWin/Plasma is crashing when turning off my display with the power button, I tried that with KWinFT, but I was kicked back to SDDM as well. I will test KWinFT now a little more to see if I can spot some differences, next I want to try some games :-)

Anyway I'm glad to see all these low level improvements in KwinFT and I'm sure this is the right way. So thank you for your great work!

[romangg]

shmerl Code can be re-merged but the communities will most likely stay fragmented. There’s really no way back. Roman claims KDE now puts marketing above engineering. (Hello Weekly blog spam and littering features on top of a broken architecture).

I wouldn't call it "spam" as I know how much work goes into gathering the required data and formulating these blog posts. But yea, albeit I respect Nate's endurance and work ethic in regards to formulating these weekly posts, they are part of the problem as they over-emphasize short-term gains instead of long-term planning.

That being said short-term improvements can also be important, a balance must be found, but in KDE such things just happen without coordination.

[shmerl]

shmerl Code can be re-merged but the communities will most likely stay fragmented. There’s really no way back. Roman claims KDE now puts marketing above engineering. (Hello Weekly blog spam and littering features on top of a broken architecture).

I don't mind the fork. Kwin development itself always felt too slow to me.

[romangg]

I will try KwinFT soon. It is now available on ArchLinux in the chaotic-aur, so no need to compile it by myself.

I wonder how well it will work on Wayland. As far as I know the subsurface flickering issues are not addressed in KwinFT, but this was one of the main pain points of kwin for me (though even there it is not solved completely, Firefox basic renderer works terrible).

Yes, that is true. I did improve subsurface handling in the past already but there is still some stuff to do. I am currently looking into improvements to KWinFT's internal window representation so that might be an opportunity to check back on subsurfaces. But it might also be necessary to work on the rendering pipeline more for that, which is a project I have planned to do afterwards.

Sadly there is so much to do.