• boredsquirrel@slrpnk.net
    link
    fedilink
    arrow-up
    12
    ·
    3 months ago

    Minetest gets scary when you look at the code and see that the engine they use is basically abandoned and will never get Wayland support. Afaik?

    • rotopenguin@infosec.pub
      link
      fedilink
      English
      arrow-up
      16
      ·
      3 months ago

      It’s not like they’re stuck on some outdated proprietary engine like RPG Maker. Minetest is under active development, with a small list of dependencies that are also under active development. It is under no particular rush to get off of X11/Xwayland.

    • MachineFab812@discuss.tchncs.de
      link
      fedilink
      arrow-up
      5
      ·
      edit-2
      3 months ago

      Individual apps, particularly full-screen games, shouldn’t need “Wayland support”(quotes because what that means will vary between implimentations).

      Now, if you have to install xorg on a system that doesn’t have it in order to play a game? Yeah that would suck, although games are on my personal shortlist of application categories that should always be run from a flat-pack/equivalent and/or containerized wherever possible.

      Now I think about it, why don’t (anti-cheat)games just run their own VM’s and “calibrate” those versus any weird system variables? Seems like a better anti-cheat than hacking-my-kernel-to-make-sure-I’m-not-hacking-the-game…

      • boredsquirrel@slrpnk.net
        link
        fedilink
        arrow-up
        1
        ·
        3 months ago

        Even if you use Flatpak, you need XOrg / XWayland on the host system.

        Fedora Kinoite/KDE and the KDE Plasma desktop on its own are especially annoying, as I have no idea how to turn off those legacy support services from constantly running, like XWaylandVideoBridge (never used) or XWayland entirely.

        I think Windows is just too bloated to also use Containers. With WSL they found a good way and apps should totally run in containers, but this is simply not yet done.

        VMs would suck for efficiency as they rely on CPU virtualization and GPU passthrough. The former will never give native performance

        • MachineFab812@discuss.tchncs.de
          link
          fedilink
          arrow-up
          1
          ·
          3 months ago

          The point of flatpack is supposed to be that it takes care of ALL dependencies. So you’re saying it doesn’t deliver on that promise?

          • boredsquirrel@slrpnk.net
            link
            fedilink
            arrow-up
            1
            ·
            3 months ago

            Not all dependencies. Flatpak is an application, and a display server is outside of an application.

            Closing an app should not result in a black screen XD

            • MachineFab812@discuss.tchncs.de
              link
              fedilink
              arrow-up
              1
              ·
              3 months ago

              Not that hard to stop wayland or xorg at the launch of a given application and restart it at that application’s exit. Of course, I only did it on the Raspberry Pi because the hardware lagged horribly running such apps with a gui/compositer/desktop the app wasn’t using in the background, but it wasn’t hard for me to get working, and its exactly how we did things with DOS apps and even some Windows games back in the WFWG 3.11 days.

              Basically, there’s no technical reason the host operating system should have to be providing say X, KDE, Plasma, Gnome, Gk, Wayland, whatever, to a flatpack app that needs those things. Yes, the result is a larger flatpack, but that’s why flatpack’s do dependency consolidation.

              Unless … Unless, you just really want to to run your games windowed with smooth window-resizing, minimization, maximization, etc.