Page 5 of 5 FirstFirst ... 345
Results 41 to 45 of 45

Thread: Mir Now Allows Multi-Threaded Compositing

  1. #41
    Join Date
    Jan 2009
    Posts
    1,480

    Default

    Quote Originally Posted by nerdopolis View Post
    I tried out Mir today.

    Right now, the display server isn't even compatible with TTY switching, and it hijacks them all. I start Mir on TTY1, and I switch to TTY2, and the text for the TTY displays only for a moment, before the contents of Mir take over.

    The only test client that runs/displays is the unaccelerated one on my spare Intel laptop... all that does is show a color changing rectangle on the screen, at 0,0.

    There is no cursor, so it seems there's no mouse input handling yet, as far as I can test...

    There's no keyboard input yet. All the keyboard input goes right to the TTY. If I start Mir on TTY1 with the & to get the prompt, I can blindly type "killall mir" into the TTY's bash prompt to kill mir.

    Remember, this has been developed in secret for 9 months. I think in Wayland years, Kristian Høgsberg was already demonstrating higher capability even earlier in Wayland's development. See his post from late 2008? http://hoegsberg.blogspot.com/2008/1...-terminal.html
    To be fair, who is more knowledgable of X, it's extensions, and DRI than krh? That is why canonical will fail. I don't know that there are any other developers who have the kind of experience in this exact area that krh has that aren't already working somewhere (and, to my knowledge, canonical hasn't been a big contributor to the graphics stack they wouldn't have these people on staff).

  2. #42
    Join Date
    Oct 2012
    Posts
    299

    Default

    Quote Originally Posted by ворот93 View Post
    What's your problem with Python? Last I checked, portage in gentoo works a lot faster than its C++ rewrite.
    i can code a basic script running in a vm in a vm in a vm running faster than a coded c++ program.

    but this does not mean that the basic interpreted code is in general faster. it mean my c++ coded program has been coded VERY VERY BAD!

  3. #43
    Join Date
    May 2012
    Posts
    912

    Default

    Quote Originally Posted by nerdopolis View Post
    I tried out Mir today.

    Right now, the display server isn't even compatible with TTY switching, and it hijacks them all. I start Mir on TTY1, and I switch to TTY2, and the text for the TTY displays only for a moment, before the contents of Mir take over.

    The only test client that runs/displays is the unaccelerated one on my spare Intel laptop... all that does is show a color changing rectangle on the screen, at 0,0.

    There is no cursor, so it seems there's no mouse input handling yet, as far as I can test...

    There's no keyboard input yet. All the keyboard input goes right to the TTY. If I start Mir on TTY1 with the & to get the prompt, I can blindly type "killall mir" into the TTY's bash prompt to kill mir.

    Remember, this has been developed in secret for 9 months. I think in Wayland years, Kristian Høgsberg was already demonstrating higher capability even earlier in Wayland's development. See his post from late 2008? http://hoegsberg.blogspot.com/2008/1...-terminal.html
    Yeah, however imho Canonical has 3 major advantages due to starting in 2012 vs 2008 and Wayland having been around:
    1) Compared to 2008 the Linux graphics stack in 2012/2013 is much better and robust.
    2) The Mir team doesn't have to create XMir, because it's pretty much XWayland customized for Mir. Figuring out XWayland and creating it took a lot of time and effort.
    3) Wayland trunk is rather close to being fit for the desktop and the Mir devs can always borrow the ideas implemented in Wayland (some of which took a lot of time and effort to figure out properly) and implement them in Mir or take decisions based on the experience Wayland had implementing said ideas.

    The big question is whether these big bonuses are enough for Mir to succeed, I for one plan to use and tinker (only) with Wayland when it ships as the default display server.
    Btw, thanks to Rebecca Black Linux 13.04 I can finally test Wayland as the native display server for the 1st time, since it ships with Linux 3.8 which allows nouveau to work properly with my GeForce 560Ti, though it wouldn't install, probably due to this.
    Hope to see an amd64 version too.

  4. #44
    Join Date
    Feb 2011
    Posts
    1,299

    Default

    Quote Originally Posted by mark45 View Post
    Yeah, however imho Canonical has 3 major advantages due to starting in 2012 vs 2008 and Wayland having been around:
    1) Compared to 2008 the Linux graphics stack in 2012/2013 is much better and robust.
    2) The Mir team doesn't have to create XMir, because it's pretty much XWayland customized for Mir. Figuring out XWayland and creating it took a lot of time and effort.
    3) Wayland trunk is rather close to being fit for the desktop and the Mir devs can always borrow the ideas implemented in Wayland (some of which took a lot of time and effort to figure out properly) and implement them in Mir or take decisions based on the experience Wayland had implementing said ideas.
    But that will mean Mir will always behind Wayland.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •