Announcement

Collapse
No announcement yet.

Canonical Is Planning For Vulkan Support In Mir By Ubuntu 16.04

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

  • Canonical Is Planning For Vulkan Support In Mir By Ubuntu 16.04

    Phoronix: Canonical Is Planning For Vulkan Support In Mir By Ubuntu 16.04

    With Vulkan now being public, Canonical developer Stephen Webb has confirmed they are planning to have Vulkan platform support ready for Mir by Ubuntu 16.04...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Why bother? To this day I still don't understand why the hell they're making Mir. I'm all for alternatives, but Mir doesn't seem to be very different from wayland at all (on any level; except that Wayland supports Vulkan and Mir doesn't at this moment that's like the biggest difference between them right now).

    I mean if all they wanted was full control over their Display Server code, then why not just work with wayland devs and develop wayland on the wayland devs terms, while having an ubuntu specific variant/fork of it that changes all the things canonical doesn't agree with. Mostly (if not completely) backwards compatible with wayland itself. For this project that approach would just make a lot more sense.. Don't go reinventing the wheel when it's already being reinvented by your more experienced neighbors, get in on their fun instead!

    Comment


    • #3
      Since Canonical broke up with Wayland I always had a minor suspicion that there was a manor of control in the display server, bickering turned out to be major arguments in the long run til' Canonical said "eff it" and forked it and made their own. Besides Canonical has their own environment to think of as well (with Unity). This is all assumption though, so none of this are facts or anything like it.

      Besides.. Mir is good for Wayland regardless, the competition makes Wayland a better Display server (in theory).

      Comment


      • #4
        Originally posted by Sethox View Post
        Besides.. Mir is good for Wayland regardless, the competition makes Wayland a better Display server (in theory).
        At stated many times before, this did NOT help at all. Canonical promised to work on Wayland, changed their minds, kept it a secret for 6+ months while people waited on them. If Canonical would have helped Wayland, Wayland could've taken their needs into account. Basically the protocol discussions you see happening now to turn Wayland from just "ready" into something with enough standards to replace existing desktop environments.

        Now Canonical have said that applications don't need to worry; everything is hidden behind that toolkit layer. But in practice, things aren't that nicely separated. Mir is an Ubuntu only thing while most developers aren't on Ubuntu. This makes this way more complicated than it needs to be.

        I went to FOSDEM and attended the convergence demo. They weren't able to demonstrate it. So many years, still never seen it in practice.

        Comment


        • #5
          Originally posted by rabcor View Post
          Why bother? To this day I still don't understand why the hell they're making Mir. I'm all for alternatives, but Mir doesn't seem to be very different from wayland at all (on any level; except that Wayland supports Vulkan and Mir doesn't at this moment that's like the biggest difference between them right now).

          I mean if all they wanted was full control over their Display Server code, then why not just work with wayland devs and develop wayland on the wayland devs terms, while having an ubuntu specific variant/fork of it that changes all the things canonical doesn't agree with. Mostly (if not completely) backwards compatible with wayland itself. For this project that approach would just make a lot more sense.. Don't go reinventing the wheel when it's already being reinvented by your more experienced neighbors, get in on their fun instead!

          I can sort of understand why they are doing it, for the mobile as well as to have more control over it. But it still doesn't seem worth it to me.
          I think however that it will be bad for everyone if canonical keep on using it while no one else wants to support it, much worse than having to different display protocols.

          Comment


          • #6
            Originally posted by bkor View Post



            I went to FOSDEM and attended the convergence demo. They weren't able to demonstrate it. So many years, still never seen it in practice.
            Really? There is a bunch of Youtube videos of people using Nexus 4 connected to a monitor, the new tablet is specifically designed for convergence and they will demonstrate it on MWC. Can you elaborate exactly how they werent able to demonstrate it?

            Comment


            • #7
              Back on topic, this is good news. Ubuntu 14.04 saw proper OpenGL 3.x series support thanks to the updates by Intel to Mesa. If Canonical can sneak Vulkan support in to 16.04, that's big, since it means we wont have to wait until 2018 for a mainline LTS Ubuntu with access to a modern graphics API.

              Comment


              • #8
                Originally posted by PoVRAZOR View Post
                Back on topic, this is good news. Ubuntu 14.04 saw proper OpenGL 3.x series support thanks to the updates by Intel to Mesa. If Canonical can sneak Vulkan support in to 16.04, that's big, since it means we wont have to wait until 2018 for a mainline LTS Ubuntu with access to a modern graphics API.
                I agree, having Vulkan ready and waiting for drivers is important for an LTS release, this one will be probably one of the biggest in years, Vulkan, ZFS support, containers, online search off by default, LTS kernel, probably optional experimental Unity 8 build, lots of stuff, mostly invisible to the average user though.

                Comment


                • #9
                  Planning, I bet.. but being ready... nahh.
                  Regardless, the first step is to plan. Let's just wait and see what the execution will lead us.

                  Comment


                  • #10
                    Originally posted by Cerberus View Post

                    Really? There is a bunch of Youtube videos of people using Nexus 4 connected to a monitor, the new tablet is specifically designed for convergence and they will demonstrate it on MWC. Can you elaborate exactly how they werent able to demonstrate it?
                    No one had an HDMI-to-VGA converter box. Convergence requires 21st-century hardware and the venue only supplied 20th-century displays. Convergence was demonstrated, but only to those in the front two rows.

                    Comment

                    Working...
                    X