Which compositor is the most suitable for me?

Hello friends,

which compositor (or no compositor at all) should I use to get the best performance and/or lowest system load?

I use Ubuntu MATE of course. There is the AMD APU in my PC. The until-now used catalyst driver is no longer available for Ubuntu 16.04, the new amdgpu driver does not support my APU, therefore I have to use the radeon open source driver. The performance of the radeon driver is noticeably lower, the temperatures of the CPU/GPU are higher even for idle desktop with no applications running and rendering of any page in the web browser is quite demanding task for my PC. These high temperatures and whirling fan is the thing I’d like to get rid of. I was wondering if the using of the proper compositor, or not using compositor at all, could somehow lower the load for APU. I do not need any special effects like shadows or transparency in desktop UI, however there is a lot of all those CSS3 transparencies, transitions, animations, shadows etc. effects on present-day websites, where compositor might be helpful (or might be not, i don’t know really). The computer is not used for gaming or for watching videos via web browser (most resource-demanding tasks). And I am not looking for some additional effects, I want to lower the system (CPU/GPU, APU) load.

Which selection of window manager would be the best? There are following choices:

  • Marco (No compositor)
  • Marco (Software compositor)
  • Marco (Compton GPU compositor)
  • Metacity (No compositor)
  • Metacity (Software compositor)
  • Metacity (Compton GPU compositor)
  • Compiz (Advanced GPU accelerated desktop effects)

Marco (Software) is what you will probably want. It won’t tax your GPU, since the bulk of it happens on the CPU and its very light on resources and processing requirements, while still giving your desktop transparency, drag-able windows and shadows.

It’s what I personally use due to its tremendous stability and problem free experience. I have yet to see a compositor related crash on my machine.

Composition won’t affect your browser contents, video playback or anything of the kind. It will simply provide some visual effects to the Window Manager. Those effects being things like shadows on windows, drag windows while seeing their contents, or window transparency.

I would recommend Marco with Compton compositor. It’s available as an option in Mate Tweak. The reason I recommend it is because whilst Marco with software compositing is okay for most things, I find it does not sufficiently reduce tearing in streaming video. Apart from that it’s okay. So, only use Marco with software compositing if (a) you don’t need 100% tear-free streaming video and/or (b) Compton is demanding too much from your machine. I doubt that it will, though. I run UM with Marco/Compton on a tiny little Acer ONE 32 bit notepad with an atom processor and 2GB of Ram and it runs fine.

Actually Marfig, whilst it is true to say that software compositing (which uses the CPU) does not affect things such as tearing video, hardware (GPU-based) compositing has a huge beneficial effect on the amount of tearing in streaming video. Compton is, all other things being equal, a GPU-based hardware compositor.

Well, @some_user did mentioned that he’s not interested in videos. I’d think Compton would be unnecessary in this case.

Tearing affects people differently. I, for one, on this aged NVIDIA GTX 560 Ti, am not affected by it and can watch video feeds (local or streamed) without issues on a barebones software compositor. Including such demanding encoders as x265. The GPU is still being used for rendering onto the screen, not the CPU. Curiously enough many Compton claims about tearing have not been met by users of certain cards, who would still suffer through it despite having Compton installed and enabled. But I also know for a fact that it does its magic on many others. I just don’t think however that Compton should be a default choice before being properly tested by the user. It’s benefits are minimal if you don’t experience tearing or it can’t fix it.

Marco is a fork of Metacity for Mate. Metacity will load and run on mate, but I would opt for Marco. Its a natural fit and works well.

Tearing has been solved by opting for either Compton or Compiz over just the standard Marco compositor. Also some have seen better performance by switching.

On my system (GTX1070 proprietary drivers), I had best results with Metacity (No compositor) and Compiz. Though Compiz sometimes crashes and occasionally there are some weird artifacts with some graphics.

Is it weird that Marco does not work as well as Metacity? With Marco I have screen tearing all the time, compositor or not.

I use Marco and Compton. The in-built Compton configuration that comes with UM is simple but effective. However, if you want to know how I have set it up to get much more attractive compositing effects, let me know and I will explain how I have done it.

You should use Marco and Compton, they are light and best fit together.

Using software compositor would use CPU. Compton is basic, stable and it use your GPU instead of the CPU.

Regards,
BT

1 Like

hi steve,
When i look in the source code for selecting marco+compton in mate tweak, it takes me to here:

Which then gets installed to here:

/usr/bin/marco-compton

Therefore, it seems like the configuration is written as hard coded arguments to that system script. Therefor if you could please elaborate, how you setup your compton in mate … I would really appreciate that !!!

(on ubuntu 18.04 + mate-session 1.20.0 over here)
Thank you

Simple answer. Try them all. If one appears buggy, don’t use that one. If there are features you don’t like, such as tearing of videos or the way desktops are displayed when switching, or instability, move on. When you get to one that doesn’t bug you, stop there.

1 Like

Don’t activate Compton via the option in mate tweak. Instead, in mate tweak, select Marco with no compositor. Then, go to startup apps and add Compton as a startup app so that it loads independently. The command to start Compton is, simply, “compton”.

Finally do the following:

  1. copy and paste the Compton configuration, below, into a blank Pluma file

  2. Save the file as "compton.conf in the the “home/username/.config” folder. You will probably need to press the “CTRL/h” key combination to show hidden folders whilst in the save dialog box in order to see the “.config” folder

  3. log out and back in.

    #################################

    Backend

    #################################

    Backend to use: “xrender” or “glx”.

    GLX backend is typically much faster but depends on a sane driver.

    backend = “glx”;

    #################################

    GLX backend

    #################################

    glx-no-stencil = true;

    GLX backend: Copy unmodified regions from front buffer instead of redrawing them all.

    My tests with nvidia-drivers show a 10% decrease in performance when the whole screen is modified,

    but a 20% increase when only 1/4 is.

    My tests on nouveau show terrible slowdown.

    Useful with --glx-swap-method, as well.

    glx-copy-from-front = false;

    GLX backend: Use MESA_copy_sub_buffer to do partial screen update.

    My tests on nouveau shows a 200% performance boost when only 1/4 of the screen is updated.

    May break VSync and is not available on some drivers.

    Overrides --glx-copy-from-front.

    glx-use-copysubbuffermesa = true;

    GLX backend: Avoid rebinding pixmap on window damage.

    Probably could improve performance on rapid window content changes, but is known to break things on some drivers (LLVMpipe).

    Recommended if it works.

    glx-no-rebind-pixmap = true;

    GLX backend: GLX buffer swap method we assume.

    Could be undefined (0), copy (1), exchange (2), 3-6, or buffer-age (-1).

    undefined is the slowest and the safest, and the default value.

    copy is fastest, but may fail on some drivers,

    2-6 are gradually slower but safer (6 is still faster than 0).

    Usually, double buffer means 2, triple buffer means 3.

    buffer-age means auto-detect using GLX_EXT_buffer_age, supported by some drivers.

    Useless with --glx-use-copysubbuffermesa.

    Partially breaks --resize-damage.

    Defaults to undefined.

    glx-swap-method = “undefined”;

    #################################

    Shadows

    #################################

    Enabled client-side shadows on windows.

    shadow = true;

    Don’t draw shadows on DND windows.

    no-dnd-shadow = true;

    Avoid drawing shadows on dock/panel windows.

    #no-dock-shadow = true;

    Zero the part of the shadow’s mask behind the window. Fix some weirdness with ARGB windows.

    clear-shadow = true;

    The blur radius for shadows. (default 12)

    shadow-radius = 5;

    The left offset for shadows. (default -15)

    shadow-offset-x = -5;

    The top offset for shadows. (default -15)

    shadow-offset-y = -5;

    The translucency for shadows. (default .75)

    shadow-opacity = 0.5;

    Set if you want different colour shadows

    shadow-red = 0.0;

    shadow-green = 0.0;

    shadow-blue = 0.0;

    The shadow exclude options are helpful if you have shadows enabled. Due to the way compton draws its shadows, certain applications will have visual glitches

    (most applications are fine, only apps that do weird things with xshapes or argb are affected).

    This list includes all the affected apps I found in my testing. The “! name~=’’” part excludes shadows on any “Unknown” windows, this prevents a visual glitch with the XFWM alt tab switcher.

    shadow-exclude = [
    “! name~=’’”,
    “name = ‘Notification’”,
    “name = ‘Plank’”,
    “name = ‘Docky’”,
    “name = ‘Kupfer’”,
    “name = ‘xfce4-notifyd’”,
    “name *= ‘VLC’”,
    “name *= ‘compton’”,
    “name *= ‘Chromium’”,
    “name *= ‘Chrome’”,
    “name *= ‘Firefox’”,
    “class_g = ‘Conky’”,
    “class_g = ‘Kupfer’”,
    “class_g = ‘Synapse’”,
    “class_g ?= ‘Notify-osd’”,
    “class_g ?= ‘Cairo-dock’”,
    “class_g ?= ‘Xfce4-notifyd’”,
    “class_g ?= ‘Xfce4-power-manager’”
    ];

    Avoid drawing shadow on all shaped windows (see also: --detect-rounded-corners)

    shadow-ignore-shaped = false;

    #################################

    Opacity

    #################################

    menu-opacity = .9;
    inactive-opacity = .9;
    active-opacity = 1;
    frame-opacity = .9;
    inactive-opacity-override = false;
    alpha-step = 0.06;

    Dim inactive windows. (0.0 - 1.0)

    inactive-dim = 0.2;

    Do not let dimness adjust based on window opacity.

    inactive-dim-fixed = true;

    Blur background of transparent windows. Bad performance with X Render backend. GLX backend is preferred.

    blur-background = true;

    Blur background of opaque windows with transparent frames as well.

    blur-background-frame = true;

    Do not let blur radius adjust based on window opacity.

    blur-background-fixed = false;
    blur-background-exclude = [
    “window_type = ‘dock’”,
    “window_type = ‘desktop’”
    ];

    #################################

    Fading

    #################################

    Fade windows during opacity changes.

    #fading = true;

    The time between steps in a fade in milliseconds. (default 10).

    fade-delta = 4;

    Opacity change between steps while fading in. (default 0.028).

    fade-in-step = 0.03;

    Opacity change between steps while fading out. (default 0.03).

    fade-out-step = 0.03;

    Fade windows in/out when opening/closing

    no-fading-openclose = true;

    Specify a list of conditions of windows that should not be faded.

    fade-exclude = [ ];

    #################################

    Other

    #################################

    Try to detect WM windows and mark them as active.

    mark-wmwin-focused = true;

    Mark all non-WM but override-redirect windows active (e.g. menus).

    mark-ovredir-focused = true;

    Use EWMH _NET_WM_ACTIVE_WINDOW to determine which window is focused instead of using FocusIn/Out events.

    Usually more reliable but depends on a EWMH-compliant WM.

    use-ewmh-active-win = true;

    Detect rounded corners and treat them as rectangular when --shadow-ignore-shaped is on.

    detect-rounded-corners = true;

    Detect _NET_WM_OPACITY on client windows, useful for window managers not passing _NET_WM_OPACITY of client windows to frame windows.

    This prevents opacity being ignored for some apps.

    For example without this enabled my xfce4-notifyd is 100% opacity no matter what.

    detect-client-opacity = true;

    Specify refresh rate of the screen.

    If not specified or 0, compton will try detecting this with X RandR extension.

    refresh-rate = 0;

    Set VSync method. VSync methods currently available:

    none: No VSync

    drm: VSync with DRM_IOCTL_WAIT_VBLANK. May only work on some drivers.

    opengl: Try to VSync with SGI_video_sync OpenGL extension. Only work on some drivers.

    opengl-oml: Try to VSync with OML_sync_control OpenGL extension. Only work on some drivers.

    opengl-swc: Try to VSync with SGI_swap_control OpenGL extension. Only work on some drivers. Works only with GLX backend. Known to be most effective on many drivers. Does not actually control paint timing, only buffer swap is affected, so it doesn’t have the effect of --sw-opti unlike other methods. Experimental.

    opengl-mswc: Try to VSync with MESA_swap_control OpenGL extension. Basically the same as opengl-swc above, except the extension we use.

    (Note some VSync methods may not be enabled at compile time.)

    vsync = “opengl-swc”;

    Enable DBE painting mode, intended to use with VSync to (hopefully) eliminate tearing.

    Reported to have no effect, though.

    dbe = false;

    Painting on X Composite overlay window. Recommended.

    paint-on-overlay = true;

    Limit compton to repaint at most once every 1 / refresh_rate second to boost performance.

    This should not be used with --vsync drm/opengl/opengl-oml as they essentially does --sw-opti’s job already,

    unless you wish to specify a lower refresh rate than the actual value.

    sw-opti = false;

    Unredirect all windows if a full-screen opaque window is detected, to maximize performance for full-screen windows, like games.

    Known to cause flickering when redirecting/unredirecting windows.

    paint-on-overlay may make the flickering less obvious.

    unredir-if-possible = true;

    Specify a list of conditions of windows that should always be considered focused.

    focus-exclude = [ ];

    Use WM_TRANSIENT_FOR to group windows, and consider windows in the same group focused at the same time.

    detect-transient = true;

    Use WM_CLIENT_LEADER to group windows, and consider windows in the same group focused at the same time.

    WM_TRANSIENT_FOR has higher priority if --detect-transient is enabled, too.

    detect-client-leader = true;

    #################################

    Window type settings

    #################################

    wintypes:
    {
    tooltip =
    {
    # fade: Fade the particular type of windows.
    fade = true;
    # shadow: Give those windows shadow
    shadow = false;
    # opacity: Default opacity for the type of windows.
    opacity = 0.85;
    # focus: Whether to always consider windows of this type focused.
    focus = true;
    };
    };

If you inspect the compton configuration, you will see it is possible to change the size and colour of shadows or turn them off completely as well as a host of other settings including transparencies. If you decide to experiment with the settings in the compton.conf file, I suggest taking a backup of it first so you can drop it back in if things go wrong.

3 Likes