Announcement

Collapse
No announcement yet.

Asahi Linux To Users: Please Stop Using X.Org

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • What this reminds me most of is the old debate over pulseaudio. And that debate died very quickly when pipewire came along.

    I wonder how long it will be until there is a kind of pipewire for a new display server protocol that will replace Wayland. Successful projects do not take 15+ years to achieve maturity and adoption, and Wayland appears to be headed to 30+ years at this pace.

    Comment


    • Originally posted by andyprough View Post
      What this reminds me most of is the old debate over pulseaudio. And that debate died very quickly when pipewire came along.

      I wonder how long it will be until there is a kind of pipewire for a new display server protocol that will replace Wayland. Successful projects do not take 15+ years to achieve maturity and adoption, and Wayland appears to be headed to 30+ years at this pace.
      Display server is way more complex than audio. And even pipewire took 6 years to be what it is today. And yes. Wayland is ready to replace old broken garbage that x11 is.

      Comment


      • Originally posted by Quackdoc View Post

        "works for me so it works for thee". gotta love phoronix, I don't have any breaking bugs so no one else does either!
        "I'm quite eager to hear how it doesn't work or is broken" Read it again without trying to shove words into my mouth. There's a difference between claiming something is "broken" and saying it has bugs. If it is "broken" it doesn't work for anyone by definition, if it has bugs then it might not be suitable for some people but work for others. I don't know what you got from my post but I never said anything about KDE not having bugs, I just don't agree with people that call something I've been using daily in a multitude of ways without issues "broken" without providing specifics which is why I asked in what ways it is "broken". Care to enlighten me or do you just have witty remarks that miss the point?

        Comment


        • Originally posted by osw89 View Post
          If it is "broken" it doesn't work for anyone by definition, if it has bugs then it might not be suitable for some people but work for others.
          this is wrong, very wrong, old car is broken, thing cant get past 2nd gear, power steering is blown out, high beams blown and one light angled wrong, that's a broken car, but could still be very usable depending on the person. and by the very same token, something could be completely and utterly broken for another person and work fine for another.

          KDE is broken in the idea that a lot of people experience bugs which completely break usability. for instance, until very recently KDE was completely broken for me, because fullscreen apps on my tablet were forced into native orientation, which left the tablet completely and utterly unusable for me.

          and loads of people experience a variety of issues that completely break for them. something doesn't need to be broken for everyone for it to be broken.

          Comment


          • Most comments seem to be from users...

            I'm both a developer and user.

            From a development point of view I need the following: global shortcuts, multiple monitor support, ability to traverse windows and control position and stack order from multiple programs.

            I find that X libraries satisfies the above requirements with little developer effort .

            I keep trying (7 years now) to move to Wayland as a developer but I'm stuck on how to implement the equivalent of; global shortcuts and how to traverse window information controlling the window positions and stack order. Can anyone point me to documentation that can help me to implement these with Wayland based libraries?

            Both of these could be considered as accessibility functionality. Windows has this functionality....

            From a normal user perspective I like the ability to use X over SSH to display a single program remotely. In fact from a maintenance point of view I need that or similar capability as my servers are 30 odd miles away. Bandwidth can be a problem with deployed systems as connections can be radio or sat-comms. TBH we've moved to using nomachine when there is sufficient bandwidth.

            Using a local desktop for general use I don't have any real issues with either, except I have NVIDIA GPUs on a number of laptops I use... (yes I know there is movement here but it's not there yet unless you want to bleed) so X on those.

            I use 4 monitors with X and I don't have any issues. In fact I've had very few issues since I started using multiple monitors back last century, so I don't get the Wayland user's saying X has a problem with multiple monitors.

            So from a user perspective I don't care which unless it does not work for me...

            From a developer perspective it has to work for my application's use cases.

            Comment


            • Originally posted by RejectModernity View Post
              Display server is way more complex than audio. And even pipewire took 6 years to be what it is today.
              Pipewire was a pretty big hit out of the box, mainly because the design ideas were correct originally. Clearly Wayland has had no such luck, and it will be many years before the majority of software can use it in even a very limited fashion. I have nothing against it, but after 15 years we have to face the reality that the entire project is clearly a failure and is unable to overcome insurmountable fundamental problems and it will eventually have to be scrapped or replaced.

              Comment


              • Originally posted by Damian View Post
                From a development point of view I need the following: global shortcuts, multiple monitor support, ability to traverse windows and control position and stack order from multiple programs.

                I find that X libraries satisfies the above requirements with little developer effort .
                Depending of the desktop environment, those implementations bar the global shortcuts are already available on GNOME Shell. The reason why global shortcuts are missing on Wayland protocol is mainly security related issue requiring root access. One of suggestion is handling portal support

                I keep trying (7 years now) to move to Wayland as a developer but I'm stuck on how to implement the equivalent of; global shortcuts and how to traverse window information controlling the window positions and stack order. Can anyone point me to documentation that can help me to implement these with Wayland based libraries?
                On which distributions? For development perspective, Fedora, possibly OpenSUSE and Arch in some extent have a better integration of Wayland protocol. Applications need to transition to full use of Wayland library otherwise XWayland is a better option.

                Using a local desktop for general use I don't have any real issues with either, except I have NVIDIA GPUs on a number of laptops I use... (yes I know there is movement here but it's not there yet unless you want to bleed) so X on those.

                I use 4 monitors with X and I don't have any issues. In fact I've had very few issues since I started using multiple monitors back last century, so I don't get the Wayland user's saying X has a problem with multiple monitors.​
                The issue is related on NVIDIA themselves who took longer to implement Wayland protocol while both AMD and Intel hardware run smoothly for years. As an example, I am able to run my laptop Dell Inspiron 14 7425 on a big monitor under GNOME Shell using Wayland protocol by default. Recently, the former slowly start to address their own problem with their newer driver.

                Comment


                • Originally posted by Myownfriend View Post

                  What does Redhat have to do with any of this? I have a feeling I know what you're gonna say but I'm asking anyway.
                  The Redhat comment was mostly poking fun at the earlier comment that said Redhat is deprecating X.org in favour of Wayland. I don't care what Redhat pushes on their users. I care that my systems remain functional until they can be migrated to Wayland without disrupting my workflow. Nothing I use is particularly tied to X.org, it just happens to work on X.org and not Wayland at the moment. I have no ill will at attempts to modernize the Linux desktop.

                  Comment


                  • Originally posted by osw89 View Post

                    "I'm quite eager to hear how it doesn't work or is broken" Read it again without trying to shove words into my mouth. There's a difference between claiming something is "broken" and saying it has bugs. If it is "broken" it doesn't work for anyone by definition, if it has bugs then it might not be suitable for some people but work for others. I don't know what you got from my post but I never said anything about KDE not having bugs, I just don't agree with people that call something I've been using daily in a multitude of ways without issues "broken" without providing specifics which is why I asked in what ways it is "broken". Care to enlighten me or do you just have witty remarks that miss the point?
                    Sure, Under Wayland the MATE panel is treated as a window, not as a panel. It moves around the screen seemingly randomly rather than staying at the top of the screen and doesn't interact in the way it works on X.org. It also isn't very stable and tends to crash. I haven't found a window manager replacement for Marco either. I tried a few options but couldn't find anything close enough to put up with. The Caja File manager also doesn't draw desktop icons on Wayland so at the moment, the Panel menu is broken, and desktop icons are broken. So the two main ways to launch/manage applications are broken in MATE on Wayland. Now the poster this article refers to is telling people to "just use Wayland" despite all the issues I mentioned.

                    Comment


                    • Originally posted by sophisticles View Post
                      I just don't get the point of Asahi Linux, other than as a working resume type project that the developers can show to a prospective employer as to why they should be hired.

                      Anyone desiring to run any Linux based distro, or even Windows, on M powered Macs completely misses the point of Apple's new chips.

                      From the point of view of pure CPU and GPU performance, they are not that great, both AMD's and intel's processors with iGPUs are capable of outperforming them, with the added benefit of being cheaper, more accessible and well supported by Linux and Windows.

                      The big advantage of the Apple M chips lies within the hardware media accelerators, that allow these processors to handle 8k ProRes and ProRes RAW in hardware, and the only application that currently supports ProRes RAW is Final Cut Pro.

                      So until the day Apple decides to release FC Pro for Linux and the day they open up the documentation to both ProRes RAW and the hardware accelerator within the M chips, then spending big bucks to buy an Apple computer just to install Linux on it is a complete waste of time and money because for a fraction of the cost I can buy either and AMD or Intel chip, install Linux on that and smoke Asahi on Mac.

                      Maybe they named this Asahi Linux because you have to be drunk to think it's a good idea.
                      It looks like you've been given the "single dumbest comment" award.
                      Attached: 1 image Okay, I've heard many ridiculous "Asahi is dumb" arguments, but this one takes the cake. Ladies and gentlemen, the only reason to use Apple Silicon is ProRes acceleration. Nothing else matters. That is literally the only benefit. This Phoronix commenter says so. Also lol, we already have a [ProRes decoding demo](https://github.com/AsahiLinux/m1n1/blob/main/proxyclient/experiments/prores.py). So much for Apple having to "open up the documentation"... they just keep repeating the same story over and over again as if we haven't reverse engineered a dozen other components already.

                      Comment

                      Working...
                      X