Is there a 2D acceleration support on Linux?

rebootl · 2015-02-16 19:01:40

What DE/Windowmanager are you using ?
Have you tried a different one ? Maybe just a simple WM (like e.g. openbox) without compositing, to test.
And then maybe with a simple compositor (e.g. compton) added.

Maybe there are different problems and not all can be adressed by this, but I'd give it a try. I think compositing was already mentioned.
Maybe some components, e.g. the compositor from the DE and libreoffice/firefox + nvidia, are not playing well together - pure guess.

Tearing _is_ a problem in X but probably not that much... Wayland should solve this somewhen... poor help.

frank604 · 2015-02-16 22:03:20

As stated twice already, please do try compton and without compton with below or similar

compton --backend glx --paint-on-overlay --vsync opengl-swc

Also, I did some reading around and one thing to try is playing with backing store option.

startx -- -bs

Can you start x with that?

pypi · 2015-02-16 22:06:24

Again, I don't really know what I'm talking about - I've only used the nouveau drivers.

However, the wiki mentions a possible tearing issue with 600 or 700 series cards. Perhaps the fix mentioned might help you?

I'd also remove the VESA driver, unless you need it. Although it doesn't appear to be being used, it would be good to make sure...

drcouzelis · 2015-02-17 01:50:18

I think the next simple test is to load up a live Linux disc (Ubuntu, Linux Mint, whatever) and see if the problem still exists.

If the live Linux environment works correctly, then my next guess is that your fiddling with Nvidia and nouveau drivers has caused a mix up. I think a fresh Arch Linux installation with nothing but "xf86-video-nouveau" and the default (non-existent) X.org configuration files would be nice and smooth.

mkkot · 2015-02-19 22:19:28

Sorry for such a long delay but my real life gets in way with my linux life :>
Anyway, I though it is best to test live distro to get instant comparison but I'm having troubles booting Mint or Fedora from usb (failed to load ldlinux.c52 in both cases, tried unetbootin). I'll get back once I get some optical disc or more time.

// edit

okay, I managed to boot Mint. After a quick test I saw that excel file works better (but not perfectly) and firefox scrolls better as well. Mint uses nouveau. I returned to my archlinux, reinstalled xorg, mesa, nvidia but effect was the same. Finally I switched to nouveau and now it works same way as Mint. Google also maps work now better. Youtube HTML5 1080p works with problems, but works. I can even play Amnesia but with lower performance than with nvidia driver. Next tests are undergoing... anyway, it is possible that nouveau is better in 2D? If anyone knows a distro which loads nvidia driver by default as live medium, please tell me. There was games knoppix once, long time ago...

Last edited by mkkot (2015-02-20 00:32:51)

mkkot · 2015-02-20 10:14:04

Okay, I run Sabayon Linux as it provided nvidia binary driver out of the box as live medium. Unfortunately package manager didn't work as expected so I finally installed firefox form the site. I was unable to install LibreOffice, though

Well, on Firefox youtube doesn't work in 1080p mode when HTML5 is enabled, google maps work simmilar to my Archlinux system, CSS3 animations still use a lot of CPU (both on chrome and Firefox). I decided to run a benchmark and compare results: http://peacekeeper.futuremark.com/

Sabayon nvidia stock firefox

Sabayon nvidia stock firefox

Firefox
2600 Points

    Version info
Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
    Comment

Suite
Result
Rendering
40,01
renderGrid01
132,34 fps
renderGrid02
65,15 fps
renderGrid03
5,06 fps
renderPhysics
58,74 fps
HTML5 Capabilities
6/7
webglSphere
Yes (58,52 fps)
videoPosterSupport
Yes
videoCodecH264
N/A
videoCodecTheora
Yes
videoCodecWebM
Yes
workerContrast01
Yes (4319,41 ops)
workerContrast02
Yes (4448,66 ops)
gamingSpitfire
Yes (51,02 fps)
HTML5 Canvas
33,19
experimentalRipple01
47,78 fps
experimentalRipple02
23,05 fps
Data
35893,35
arrayCombined
7283,86 ops
arrayWeighted
176875,00 ops
DOM operations
12849,27
domGetElements
739017,00 ops
domDynamicCreationCreateElement
22501,50 ops
domDynamicCreationInnerHTML
25871,50 ops
domJQueryAttributeFilters
5560,00 ops
domJQueryBasicFilters
1096,00 ops
domJQueryBasics
3989,00 ops
domJQueryContentFilters
2641,50 ops
domJQueryHierarchy
8229,50 ops
domQueryselector
41999,50 ops
Text parsing
194023,71
stringChat
41926,50 ops
stringDetectBrowser
406450,00 ops
stringFilter
57612,00 ops
stringValidateForm
402951,50 ops
stringWeighted
695044,50 ops

Sabayon nvidia preinstalled chromium:

Chrome
2342 Points
Version infoMozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.111 Safari/537.36Comment
Suite
Result
Rendering
70.13
renderGrid01
180.10 fps
renderGrid02
148.72 fps
renderGrid03
15.60 fps
renderPhysics
57.89 fps
HTML5 Capabilities
7/7
webglSphere
Yes (59.28 fps)
videoPosterSupport
Yes
videoCodecH264
Yes
videoCodecTheora
Yes
videoCodecWebM
Yes
workerContrast01
Yes (7024.60 ops)
workerContrast02
Yes (8263.41 ops)
gamingSpitfire
Yes (51.20 fps)
HTML5 Canvas
24.99
experimentalRipple01
39.80 fps
experimentalRipple02
15.69 fps
Data
34918.01
arrayCombined
7883.00 ops
arrayWeighted
154670.50 ops
DOM operations
8489.68
domGetElements
593167.00 ops
domDynamicCreationCreateElement
17309.50 ops
domDynamicCreationInnerHTML
19304.50 ops
domJQueryAttributeFilters
2629.50 ops
domJQueryBasicFilters
1063.50 ops
domJQueryBasics
2254.50 ops
domJQueryContentFilters
1335.50 ops
domJQueryHierarchy
4329.00 ops
domQueryselector
31711.00 ops
Text parsing
135632.99
stringChat
57606.00 ops
stringDetectBrowser
295396.00 ops
stringFilter
36386.50 ops
stringValidateForm
450493.50 ops
stringWeighted
164560.50 ops

My Archlinux with nouveau, chrome packaged:

Chrome
2477 Points
Version infoMozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.111 Safari/537.36Comment
Suite
Result
Rendering
72,20
renderGrid01
179,19 fps
renderGrid02
148,88 fps
renderGrid03
16,00 fps
renderPhysics
63,67 fps
HTML5 Capabilities
7/7
webglSphere
Yes (59,31 fps)
videoPosterSupport
Yes
videoCodecH264
Yes
videoCodecTheora
Yes
videoCodecWebM
Yes
workerContrast01
Yes (7474,08 ops)
workerContrast02
Yes (7911,60 ops)
gamingSpitfire
Yes (51,40 fps)
HTML5 Canvas
27,93
experimentalRipple01
38,27 fps
experimentalRipple02
20,39 fps
Data
37256,39
arrayCombined
8399,50 ops
arrayWeighted
165252,50 ops
DOM operations
8937,71
domGetElements
610525,00 ops
domDynamicCreationCreateElement
18039,00 ops
domDynamicCreationInnerHTML
20464,00 ops
domJQueryAttributeFilters
3373,00 ops
domJQueryBasicFilters
1082,50 ops
domJQueryBasics
2188,00 ops
domJQueryContentFilters
1455,00 ops
domJQueryHierarchy
4299,50 ops
domQueryselector
32311,50 ops
Text parsing
138793,24
stringChat
56112,00 ops
stringDetectBrowser
323256,50 ops
stringFilter
37079,00 ops
stringValidateForm
457590,50 ops
stringWeighted
167353,00 ops

My Archlinux, nouveau Firefox packaged (my old everyday profile):

Firefox
2659 Points

    Version info
Mozilla/5.0 (X11; Linux x86_64; rv:35.0) Gecko/20100101 Firefox/35.0
    Comment

Suite
Result
Rendering
36,98
renderGrid01
136,95 fps
renderGrid02
61,51 fps
renderGrid03
4,65 fps
renderPhysics
47,69 fps
HTML5 Capabilities
7/7
webglSphere
Yes (41,78 fps)
videoPosterSupport
Yes
videoCodecH264
Yes
videoCodecTheora
Yes
videoCodecWebM
Yes
workerContrast01
Yes (3497,28 ops)
workerContrast02
Yes (3612,39 ops)
gamingSpitfire
Yes (36,92 fps)
HTML5 Canvas
32,68
experimentalRipple01
46,19 fps
experimentalRipple02
23,12 fps
Data
40619,92
arrayCombined
8363,32 ops
arrayWeighted
197287,50 ops
DOM operations
12824,08
domGetElements
734227,50 ops
domDynamicCreationCreateElement
22799,50 ops
domDynamicCreationInnerHTML
26564,50 ops
domJQueryAttributeFilters
5201,50 ops
domJQueryBasicFilters
1226,50 ops
domJQueryBasics
3780,00 ops
domJQueryContentFilters
2747,50 ops
domJQueryHierarchy
7302,00 ops
domQueryselector
43602,50 ops
Text parsing
211003,46
stringChat
67986,00 ops
stringDetectBrowser
406762,00 ops
stringFilter
59677,50 ops
stringValidateForm
381248,00 ops
stringWeighted
664767,50 ops

Chrome was force-enabled in both cases to use hardware acceleration.

drcouzelis · 2015-02-20 12:35:44

mkkot wrote:

I returned to my archlinux, reinstalled xorg, mesa, nvidia but effect was the same.

You mean, with "pacman -S ..."? For you information, that doesn't really do anything, so if there's a problem with your Nvidia / nouveau driver settings colliding with each other then reinstalling the packages wouldn't have changed anything.

onslow77 · 2015-02-20 13:34:15

Hello,

I asked before (see post 9) if you lacked a composite manager, and others have suggested the same - have you tried that yet? If you have not, it can not hurt to give it a try: https://wiki.archlinux.org/index.php/Compton

Regards
Martin

Last edited by onslow77 (2015-02-20 13:34:58)

mkkot · 2015-02-20 16:39:42

Thank you for your responses. I just thought that it would be a more definite test to start a fresh system. I'm using xfce with Arch and even though I have composition enabled, disabling it doesn't change much. I have also tried with startx -- -bs but couldn't see the difference. Hence I tried with live images.

Drcouzelis why pacman -S won't help? It helped me in some cases. How should I reinstall then?

olive · 2015-02-20 17:23:27

I experience some lags too with the xls file (but it remains usable) with a Corei3 (3.40 Ghz) and an AMD 6570 with the opensource radeon driver. I do not believe that the lags are caused by acceleration not functioning properly. I thinks it is Libreoffice that for some reason take some time to calculate what it should display.

teateawhy · 2015-02-20 17:51:52

Firefox youtube doesn't work in 1080p mode when HTML5 is enabled,

Probably not related to video drivers.
https://bbs.archlinux.org/viewtopic.php?id=178920

drcouzelis · 2015-02-20 18:16:27

mkkot wrote:

why pacman -S won't help? It helped me in some cases. How should I reinstall then?

When pacman first installed those packages, it downloaded the files, verified their integrity, then installed them. Unless you have been changing files in your filesystem yourself (without using pacman) then reinstalling will just copy the same files onto your computer that are already there.

onslow77 · 2015-02-20 23:58:24

mkkot wrote:

Thank you for your responses. I just thought that it would be a more definite test to start a fresh system. I'm using xfce with Arch and even though I have composition enabled, disabling it doesn't change much. ...

Hello,

I tried xfce a few weeks ago and suffered from screen teering - similar to what I understand that you are experience, and it all went away when I installed and configured compton.

Regards
Martin

Roken · 2015-02-21 12:29:14

compton --backend glx --paint-on-overlay --vsync opengl-swc

fixed it for me, (I used to run compton with only the -b switch), and I have no problem with 1080 video and HTML5 (but then, I didn't before, either).

pro_noob · 2015-02-21 15:45:30

I recently tried with a newly bought lowest end nvidia card (Geforce 210). I did not do good with the nvidia proprietery driver. It also did not do super with the open source driver, but it is somewhat usable.
The best results I got with the nvidia-340xx drive much better than the standard nvidia package.

So maybe your card is legacy even if you just bought it and you should try with the nvidia-340xx package.

mkkot · 2015-02-26 23:24:50

I have tried compton. Well, no tearing but laggy mouse instead (when moving window I can see that my mouse cursor is faster than the window which follows in sluggish way). That's when using nvidia driver. While using nouveau I take the fastest option meaning:

glx-use-copysubbuffermesa = true;

which of course is fast but tears as well

Here's my config with nvidia drivers:

#################################
#
# 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 = "3";

#################################
#
# 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 = 0.98;
#inactive-opacity = 1;
#active-opacity = 1;
#frame-opacity = 1;
#inactive-opacity-override = true;
#alpha-step = 0.001;

# 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 = false;
# The time between steps in a fade in milliseconds. (default 10).
fade-delta = 10;
# Opacity change between steps while fading in. (default 0.028).
fade-in-step = 0.2;
# Opacity change between steps while fading out. (default 0.03).
fade-out-step = 0.3;
# Fade windows in/out when opening/closing
 no-fading-openclose = false;

# 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;
    };
};

I can observe positive influence upon scrolling webpages in firefox. It's not a huge difference but it gave something compared to no composition enabled at all. Not sure how this is possible. With scrolling excel file I can see no difference. With compton I switched back to nvidia binary drivers counting that they will be faster with glx backend. Well, I can observe no much difference between them. I have also uninstalled xorg-server, removed nvidia and nouveau and installed back nvidia. I switched firefox back to flash on youtube because it works 3 times better than html5 unfortunately.

I run futuremark tests on the website http://peacekeeper.futuremark.com/ with both nouveau and nvidia but I can see no big difference between the drivers and compton on/off. It's always about 2500 - 2600 points with some differences between particular tests on different drivers.

pro_noob I know that my GPU is based on Fermi architecture and it's the same card as GT 520: http://www.game-debate.com/gpu/index.ph … rce-gt-520
However, they are both supported by newest driver: http://www.nvidia.com/Download/driverRe … 2252/en-us
I can see no reason why I should go into legacy. However, when I put this card into my PC I had been using nvidia-304 legacy drivers and they were working noticeable worse. I'll try the 340 version out of curiosity, though.

To sum up, I gained a lot switching to flash on yt, gained a little and lost a little switching to compton, and gained nothing for reinstalling drivers. I would say that nouveau is a bit better in 2D and nvidia is better in 3D games. When I'll be having some time to play around, I'm going to install Windows and do some more tests.

Arch Linux

#26 2015-02-16 19:01:40

Re: Is there a 2D acceleration support on Linux?

#27 2015-02-16 22:03:20

Re: Is there a 2D acceleration support on Linux?

#28 2015-02-16 22:06:24

Re: Is there a 2D acceleration support on Linux?

#29 2015-02-17 01:50:18

Re: Is there a 2D acceleration support on Linux?

#30 2015-02-19 22:19:28

Re: Is there a 2D acceleration support on Linux?

#31 2015-02-20 10:14:04

Re: Is there a 2D acceleration support on Linux?

#32 2015-02-20 12:35:44

Re: Is there a 2D acceleration support on Linux?

#33 2015-02-20 13:34:15

Re: Is there a 2D acceleration support on Linux?

#34 2015-02-20 16:39:42

Re: Is there a 2D acceleration support on Linux?

#35 2015-02-20 17:23:27

Re: Is there a 2D acceleration support on Linux?

#36 2015-02-20 17:51:52

Re: Is there a 2D acceleration support on Linux?

#37 2015-02-20 18:16:27

Re: Is there a 2D acceleration support on Linux?

#38 2015-02-20 23:58:24

Re: Is there a 2D acceleration support on Linux?

#39 2015-02-21 12:29:14

Re: Is there a 2D acceleration support on Linux?

#40 2015-02-21 15:45:30

Re: Is there a 2D acceleration support on Linux?

#41 2015-02-26 23:24:50

Re: Is there a 2D acceleration support on Linux?

Board footer