Xorg development effort slowing in favour of Wayland
SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Ultimately this conversation was moot before it began. As I've pointed out many times before X is simply better than wayland. Let's hold the discussions about deprecating X until wayland based desktops can boast parity on both features AND performance.
I would much rather discuss the development process of kwin_wayland from the perspective of its flaws when compared to kwin_x11. Complaining about how old and crappy X is counter productive and doesn't benefit wayland in any way, shape, or form.
X is only better than wayland because at this point it just has been around longer - thats about it. If X were indeed even better, its codebase would have been much more maintainable, and like you pointed out moot and wayland would not be needed. Wayland is the new kid on the block, and from what I heard it still doesn't support multi-monitor yet , and again I put that down to the devs squabbling and almost every little detail. Which is also why I am rather dismayed that after 15 years, wayland hasn't quite progressed as it should have, leaving some basic features like multi-moonitor support as something to be added later.
Is X better than wayland? Yea sure, if you also rely on the network-aware aspect, me I do not; and like I stated earlier while even running a compositor for NVIDIA and KDE - I still experience screen tearing.
Again, i don't want network aware anything - I want a modern display - X can no longer keep up imo.
I don't get why people keep saying it "can't keep up". I've got 99 problems on my linux box and X11 isn't one. The only significant change I can think of for X in the time since I start using it is the automagical config system in xorg.conf.d. It's doing a much better job than wayland without any major development effort. That's 20 years of nothingness, and yet I can run Doom 2016 at 200FPS. I'm on a 1440p monitor with options for 60, 100, 120, and 144hz refresh rates. X11 is plenty modern.
When I run any of the wayland desktops I get eyestrain. There are noticeable render delays that hurt my eyes and as someone who is prone to migraines I'm not going to be forgiving when it comes to wayland's performance. X11 is fine. It works, and unless a developer for one of the widget toolkits wants to chime in and provide a specific example of something that they were unable to do because of an X11 limitation then I'm really not all that interested. Everything other than that is political garbage.
I don't get why people keep saying it "can't keep up". I've got 99 problems on my linux box and X11 isn't one. The only significant change I can think of for X in the time since I start using it is the automagical config system in xorg.conf.d. It's doing a much better job than wayland without any major development effort. That's 20 years of nothingness, and yet I can run Doom 2016 at 200FPS. I'm on a 1440p monitor with options for 60, 100, 120, and 144hz refresh rates. X11 is plenty modern.
When I run any of the wayland desktops I get eyestrain. There are noticeable render delays that hurt my eyes and as someone who is prone to migraines I'm not going to be forgiving when it comes to wayland's performance. X11 is fine. It works, and unless a developer for one of the widget toolkits wants to chime in and provide a specific example of something that they were unable to do because of an X11 limitation then I'm really not all that interested. Everything other than that is political garbage.
My expectations are less, I don't run games that are even current from 10 years ago - my eyesight is bad so for me a game running at 30-60fps is fine for me. If you also consider this to be political garbage, then I would look at the history of x some more, the codebase has been overextended and something got to give. Also again even I stated that I don't wayland yet, but sooner down the line I will be. The reason why I say X can't keep up, is because again X was never designed for modern-day usage, it had to be essentially worked / hacked to do so, and again the codebase has become untenable.
My expectations are less, I don't run games that are even current from 10 years ago - my eyesight is bad so for me a game running at 30-60fps is fine for me. If you also consider this to be political garbage, then I would look at the history of x some more, the codebase has been overextended and something got to give. Also again even I stated that I don't wayland yet, but sooner down the line I will be. The reason why I say X can't keep up, is because again X was never designed for modern-day usage, it had to be essentially worked / hacked to do so, and again the codebase has become untenable.
Whether or not X can keep up depends entirely on whether a suitable alternative is created to replace it. We both agree, but due to the nature of internet-based communication we are just arguing back and forth.
Give me a practical example of how X is failing and I'll consider an alternative. "OMG X is so hard" is not a valid reason to completely replace an entire display protocol. TBH if people don't get their shit together with these types of discussions then wayland is just gonna end up becoming the same bloated whale that X is. When history repeats itself I will laugh my ass off.
Crying on social media about how unmaintainable X is won't have any positive impact on Wayland's progress. Honestly that might be the reason it still hasn't gone mainstream after 15 years of development.
Thanks for the screenshot! I wonder why your Firefox render the page with much better contrast than my Firefox. I use Firefox version 122.0.1 and have to my knowledge not done any special configurations.
Thanks for the screenshot! I wonder why your Firefox render the page with much better contrast than my Firefox. I use Firefox version 122.0.1 and have to my knowledge not done any special configurations.
regards Henrik
Ditto here
The web page looks the same on Wayland or Xorg
It's also identical whatever laptop I use
When someone says to me, "Look, here is a Wayland compositor built for Slackware with a simple user interface similar to fluxbox or ice. You install it much like you would Xorg and it will run most graphical applications natively (and you can use Xwindow for the rest)"
then I will use Wayland. Not before.
Edit: Oh yes, and it has to be able to run on an Intel Bay Trail SOC.
Whether or not X can keep up depends entirely on whether a suitable alternative is created to replace it. We both agree, but due to the nature of internet-based communication we are just arguing back and forth.
Give me a practical example of how X is failing and I'll consider an alternative. "OMG X is so hard" is not a valid reason to completely replace an entire display protocol. TBH if people don't get their shit together with these types of discussions then wayland is just gonna end up becoming the same bloated whale that X is. When history repeats itself I will laugh my ass off.
Crying on social media about how unmaintainable X is won't have any positive impact on Wayland's progress. Honestly that might be the reason it still hasn't gone mainstream after 15 years of development.
Well the "omg its so hard" argument is technically valid - have you see the code to xorg? If you think it isn't hard, join the team, start contributing fixes and new features. And I am still experiencing screen tearing no matter what I do; so fix it for me please - as well as I stated the main point - x was never designed for this initially. If I have to give up network awareness to get a more efficient display system, then I'll do that gladly. Again, almost nobody on here is running a setup where they need to render an app over the network - and if there are - we already have RDP and VNC for that. Also, proof again is just go to x.org's site , see the last update - it was around 2013 - so yea doesn't instill confidence in me in regards to development.
Also I'm going to bet that there is a change 99.999999999% you yourself aren't mainstaining or contributing code to x at all. As for politics, again look at the history of x as a whole then, and wonder why are we even still at 7.x - not just because 'it just works'.
When someone says to me, "Look, here is a Wayland compositor built for Slackware with a simple user interface similar to fluxbox or ice. You install it much like you would Xorg and it will run most graphical applications natively (and you can use Xwindow for the rest)"
then I will use Wayland. Not before.
Edit: Oh yes, and it has to be able to run on an Intel Bay Trail SOC.
Thats the rub isn't it? It looks like fluxbox or IceWM will just be left behind - oh well. While I don't use a window manager, I am sure someone will pick up the slack no pun intended. I haven't used sway or hyprland either. I'll probably end up still with a DE. MATE most likely will be left out, unless there is serious efforts on that dev team. So what do we have so far, KDE Cinnamon, GNOME, then finally XFCE somewhat. Also again I stated previously that while I expect wayland to replace x as it is now the only alternative, until NVIDIA gets their shit together, I'll also stay with xorg until I can't.
If we end up with only big integrated DEs, we might as well be using Windows. Surely one of the great strengths of Linux has always been that you can mix and match. If you are not interested in eye candy, you can just use a window manager and still run all the applications you want.
DEs need a great deal of processing, so a great deal of cpu and graphics power. We'll all end up having to buy a new computer every two years to keep up, like Windows users do. I'd rather save my cpu cycles for applications, not eye candy on the desktop.
I think people are missing the value of history here.
Alsa replaced OSS. OSS was a limited format, but alsa sucked for 2 years at least. ACPI replaced APM on laptops, and boy did acpi suck for a while, too.
Selinux tried to take over a poorly defined set of features but failed, and has been rejected. Systemd has tried to replace sysVinit, and the jury is still out on that one. If every software package needs systemd, it may take over. But it's not there yet.
I 70% predict that Wayland will suck while we all get used to it, and when the fuss dies down, Wayland will be there and X will be deprecated.
And as for doing things in Xorg, I don't. I use startx, xrandr occasionally and Xorg.0.log. I use programs. I put funny files in odd places to humour things and proceed in blind ignorance like most users. I've used Wayland the same way. I just don't really know the rules there yet, so just accept defaults.
If we end up with only big integrated DEs, we might as well be using Windows. Surely one of the great strengths of Linux has always been that you can mix and match. If you are not interested in eye candy, you can just use a window manager and still run all the applications you want.
DEs need a great deal of processing, so a great deal of cpu and graphics power. We'll all end up having to buy a new computer every two years to keep up, like Windows users do. I'd rather save my cpu cycles for applications, not eye candy on the desktop.
Well then, thats up to the writers of fluxbox and icewm etc then to adapt to wayland isn't it? I mean don't get me wrong it will still work as long as xorg is still around, but what about when the time arrives when xorg is indeed 100% abandoned, and all distros have abandoned xorg? Granted thats not for years away, but also time does fly.
I am also not interested in eye candy, I am just more lazy as I get older. I can't be bothered to work in a tiling window manager; and i can't be bothered to take the time to learn keyboard only shortcuts, or manually configure a wm to automount and replicate DE features.
As for DE's being heavy , that isn't necessarily the case - while I haven't run GNOME since Slackware dropped it years ago - I have tried KDE5 for Slackware15.0 and decided to stick with it - and my computer is at this point reaching its 10 years mark, with a shit 2GB NVIDIA GT710.... And yea granted I just run the bare minimum of effects, just some transparency, and dragging windows with their contents showing while doing so (also I can't seem to figure out how to do that in a WM), and for me thats enough eyecandy. I like being able to see the contents of my windows while a drag windows around. Ever since that was introduced even back in Windows 9x, I was like "omg finally!" , because I just can't stand an outline dragging.
I digress though as I am going into another tangent again. Point is, even a fairly "heavy" DE on my system, doesn't seem that heavy to me; and like I stated the only issue I am still experiencing is a vertical window tear when scrolling text (web pages , or text file in a text editor).
Quote:
Originally Posted by business_kid
I think people are missing the value of history here.
Alsa replaced OSS. OSS was a limited format, but alsa sucked for 2 years at least. ACPI replaced APM on laptops, and boy did acpi suck for a while, too.
Selinux tried to take over a poorly defined set of features but failed, and has been rejected. Systemd has tried to replace sysVinit, and the jury is still out on that one. If every software package needs systemd, it may take over. But it's not there yet.
I 70% predict that Wayland will suck while we all get used to it, and when the fuss dies down, Wayland will be there and X will be deprecated.
And as for doing things in Xorg, I don't. I use startx, xrandr occasionally and Xorg.0.log. I use programs. I put funny files in odd places to humour things and proceed in blind ignorance like most users. I've used Wayland the same way. I just don't really know the rules there yet, so just accept defaults.
Wayland sucks because it has been 15 years; and from what they have to show for it is surprisingly little. It is the same shit that also hindered xorg development, infighting and politics. I want wayland to succeed. I also remember OSS , and yea I was so glad that ALSA came along - because omg I could finally have multiple sources of audio playing back at once; couldn't as far as I know under OSS without probably some weird work arounds or whatever. Why I also I bring up the dragging of windows with contents showing. I was there also when OSS was more or less the default, and it felt a step back to me, it was like windows 3.1 again - only being able to play one source of sound at a time. ALSA fixed this.
So for me, I want wayland to finally get rid of all that screen tearing.
Well then, thats up to the writers of fluxbox and icewm etc then to adapt to wayland isn't it?
No, it isn't. It's up to people who write wayland compositors to write some that don't require a complete KDE-style desktop environment in order to function.
Quote:
I am also not interested in eye candy, I am just more lazy as I get older. I can't be bothered to work in a tiling window manager; and i can't be bothered to take the time to learn keyboard only shortcuts, or manually configure a wm to automount and replicate DE features.
Ditto here. I'm not talking about niche products like ratpoison. I'm talking about lightweight stacking WMs that can run autonomously.
Quote:
And yea granted I just run the bare minimum of effects, just some transparency, and dragging windows with their contents showing while doing so (also I can't seem to figure out how to do that in a WM), and for me thats enough eyecandy. I like being able to see the contents of my windows while a drag windows around. Ever since that was introduced even back in Windows 9x, I was like "omg finally!" , because I just can't stand an outline dragging.
I wonder what WM gave you such a bad experience. Fluxbox lets you drag full windows in a completely normal way and you don't need to configure it specially to do so.
I just run the bare minimum of effects, just some transparency, and dragging windows with their contents showing while doing so (also I can't seem to figure out how to do that in a WM)
Opaque move/resize is not uncommon in window managers. Some window manager will rubber-band move/resize operations by default, some will do Opaque move/resize by default and some will provide a config option to allow the user to choose.
Well then, thats up to the writers of fluxbox and icewm etc then to adapt to wayland isn't it?
In the case of Window Maker (where I can automount and drag windows with contents showing) that is not to be expected "anytime soon" as the task is daunting.
I have used Window Maker for many years and it still receives support. It runs happily on hardware that chokes using KDE. I am still looking for a potential replacement using Wayland that is not prissy. I might have to look at sway, much as I am not keen on the tiling paradigm.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.