mir performance with glxgears and other info

**ventrical** · April 25th, 2014

Just testing unity-system-co performance.

with-mir

Code:

ventrical@ventrical-desktop:~$ glxgears
Running synchronized to the vertical refresh.  The framerate should be
approximately the same as the monitor refresh rate.
2753 frames in 5.0 seconds = 550.537 FPS
2745 frames in 5.0 seconds = 548.931 FPS
2727 frames in 5.0 seconds = 545.074 FPS
2715 frames in 5.0 seconds = 542.526 FPS
2732 frames in 5.0 seconds = 546.348 FPS
2706 frames in 5.0 seconds = 540.632 FPS
2750 frames in 5.0 seconds = 549.898 FPS
2731 frames in 5.0 seconds = 545.880 FPS
2729 frames in 5.0 seconds = 545.703 FPS
2669 frames in 5.0 seconds = 533.750 FPS
2651 frames in 5.0 seconds = 529.565 FPS

with-xorg

Code:

ventrical@ventrical-desktop:~$ glxgears
Running synchronized to the vertical refresh.  The framerate should be
approximately the same as the monitor refresh rate.
259 frames in 5.0 seconds = 51.742 FPS
246 frames in 5.0 seconds = 49.108 FPS
250 frames in 5.0 seconds = 49.717 FPS
261 frames in 5.0 seconds = 52.103 FPS
248 frames in 5.0 seconds = 49.179 FPS
247 frames in 5.0 seconds = 48.981 FPS
250 frames in 5.0 seconds = 49.908 FPS
249 frames in 5.0 seconds = 49.375 FPS
246 frames in 5.0 seconds = 49.108 FPS
246 frames in 5.0 seconds = 49.106 FPS
248 frames in 5.0 seconds = 49.504 FPS

Code:

ventrical@ventrical-desktop:~$ lspci
00:00.0 Host bridge: Intel Corporation 82Q963/Q965 Memory Controller Hub (rev 02)
00:01.0 PCI bridge: Intel Corporation 82Q963/Q965 PCI Express Root Port (rev 02)
00:03.0 Communication controller: Intel Corporation 82Q963/Q965 HECI Controller (rev 02)
00:19.0 Ethernet controller: Intel Corporation 82566DM Gigabit Network Connection (rev 02)
00:1a.0 USB controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #4 (rev 02)
00:1a.1 USB controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #5 (rev 02)
00:1a.7 USB controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #2 (rev 02)
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 02)
00:1c.0 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 1 (rev 02)
00:1c.1 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 2 (rev 02)
00:1c.2 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 3 (rev 02)
00:1c.3 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 4 (rev 02)
00:1c.4 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 5 (rev 02)
00:1d.0 USB controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #1 (rev 02)
00:1d.1 USB controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #2 (rev 02)
00:1d.2 USB controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #3 (rev 02)
00:1d.7 USB controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev f2)
00:1f.0 ISA bridge: Intel Corporation 82801HO (ICH8DO) LPC Interface Controller (rev 02)
00:1f.2 IDE interface: Intel Corporation 82801H (ICH8 Family) 4 port SATA Controller [IDE mode] (rev 02)
00:1f.3 SMBus: Intel Corporation 82801H (ICH8 Family) SMBus Controller (rev 02)
00:1f.5 IDE interface: Intel Corporation 82801HR/HO/HH (ICH8R/DO/DH) 2 port SATA Controller [IDE mode] (rev 02)
01:00.0 VGA compatible controller: NVIDIA Corporation GT218 [GeForce 210] (rev a2)
01:00.1 Audio device: NVIDIA Corporation High Definition Audio Controller (rev a1)
03:00.0 IDE interface: Marvell Technology Group Ltd. 88SE6101/6102 single-port PATA133 interface (rev b1)
07:03.0 FireWire (IEEE 1394): Texas Instruments TSB43AB22A IEEE-1394a-2000 Controller (PHY/Link) [iOHCI-Lynx]
ventrical@ventrical-desktop:~$

Code:

ventrical@ventrical-desktop:~$ lsb_release -a
No LSB modules are available.
Distributor ID:    Ubuntu
Description:    Ubuntu Utopic Unicorn (development branch)
Release:    14.10
Codename:    utopic
ventrical@ventrical-desktop:~$

**grahammechanical** · April 25th, 2014

@ventrical, if you do not mind me saying so, there is something very wrong with the results for with-mir. This is what I get with xserver.

graham@Sdb8-Roll-Dev:~$ glxgears
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
270 frames in 5.0 seconds = 53.764 FPS
265 frames in 5.0 seconds = 52.990 FPS
248 frames in 5.0 seconds = 49.566 FPS
250 frames in 5.0 seconds = 49.991 FPS
248 frames in 5.0 seconds = 49.412 FPS
249 frames in 5.0 seconds = 49.524 FPS
249 frames in 5.0 seconds = 49.609 FPS
249 frames in 5.0 seconds = 49.608 FPS
248 frames in 5.0 seconds = 49.599 FPS
249 frames in 5.0 seconds = 49.758 FPS
251 frames in 5.0 seconds = 50.198 FPS
252 frames in 5.0 seconds = 50.330 FPS

I am not running xmir so I cannot compare but it sure cannot speed things up by more than 10 times. My guess is that mesa-utils is not compatible with mir. I found this the other day. It is old but somewhere there might be something that points to a better method of testing performance.

http://samohtv.wordpress.com/

glmark2 is in the trusty/utopic software centre.

https://launchpad.net/glmark2

Regards.

**ventrical** · April 25th, 2014

Thanks for your reply. Actually I was quite surprised. I did notice however that with-xorg the gears are rather chunky but with-xmir it is extremely smooth. I'll look for other tests.

**grahammechanical** · April 25th, 2014

I have just installed OpenGL (ES) 2.0 benchmark. It is run in a terminal with glmark2 and Wow! Is that a unicorn I see before me? Just let it run and run. It will eventually stop and this is what I get with xserver.

graham@Sdb8-Roll-Dev:~$ glmark2
================================================== =====
glmark2 2012.08
================================================== =====
OpenGL Information
GL_VENDOR: nouveau
GL_RENDERER: Gallium 0.4 on NVA5
GL_VERSION: 3.0 Mesa 10.1.0
================================================== =====
[build] use-vbo=false: FPS: 113 FrameTime: 8.850 ms
[build] use-vbo=true: FPS: 115 FrameTime: 8.696 ms
[texture] texture-filter=nearest: FPS: 113 FrameTime: 8.850 ms
[texture] texture-filter=linear: FPS: 118 FrameTime: 8.475 ms
[texture] texture-filter=mipmap: FPS: 114 FrameTime: 8.772 ms
[shading] shading=gouraud: FPS: 112 FrameTime: 8.929 ms
[shading] shading=blinn-phong-inf: FPS: 116 FrameTime: 8.621 ms
[shading] shading=phong: FPS: 114 FrameTime: 8.772 ms
[bump] bump-render=high-poly: FPS: 113 FrameTime: 8.850 ms
[bump] bump-render=normals: FPS: 112 FrameTime: 8.929 ms
[bump] bump-render=height: FPS: 114 FrameTime: 8.772 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 113 FrameTime: 8.850 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 114 FrameTime: 8.772 ms
[pulsar] light=false:quads=5:texture=false: FPS: 111 FrameTime: 9.009 ms
[desktop] blur-radius=5:effect=blur: passes=1:separable=true:windows=4: FPS: 108 FrameTime: 9.259 ms
[desktop] effect=shadow:windows=4: FPS: 97 FrameTime: 10.309 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 34 FrameTime: 29.412 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 35 FrameTime: 28.571 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 35 FrameTime: 28.571 ms
[ideas] speed=duration: FPS: 70 FrameTime: 14.286 ms
[jellyfish] <default>: FPS: 111 FrameTime: 9.009 ms
[terrain] <default>: FPS: 56 FrameTime: 17.857 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 111 FrameTime: 9.009 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 110 FrameTime: 9.091 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 111 FrameTime: 9.009 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 112 FrameTime: 8.929 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 115 FrameTime: 8.696 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 113 FrameTime: 8.850 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 114 FrameTime: 8.772 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 112 FrameTime: 8.929 ms
================================================== =====
glmark2 Score: 101
================================================== =====

If only I knew what it meant.

**ventrical** · April 25th, 2014

I installed a program from the Software Center called Blobs. It measures fps. I am still getting 3 to 4 times faster than with xorg.

**ventrical** · April 25th, 2014

I think it is an average score.

With mir.

Code:

ventrical@ventrical-desktop:~$ glmark2
=======================================================
    glmark2 2012.08
=======================================================
    OpenGL Information
    GL_VENDOR:     nouveau
    GL_RENDERER:   Gallium 0.4 on NVA8
    GL_VERSION:    3.0 Mesa 10.1.0
=======================================================
[build] use-vbo=false: FPS: 211 FrameTime: 4.739 ms
[build] use-vbo=true: FPS: 236 FrameTime: 4.237 ms
[texture] texture-filter=nearest: FPS: 219 FrameTime: 4.566 ms
[texture] texture-filter=linear: FPS: 216 FrameTime: 4.630 ms
[texture] texture-filter=mipmap: FPS: 225 FrameTime: 4.444 ms
[shading] shading=gouraud: FPS: 211 FrameTime: 4.739 ms
[shading] shading=blinn-phong-inf: FPS: 212 FrameTime: 4.717 ms
[shading] shading=phong: FPS: 203 FrameTime: 4.926 ms
[bump] bump-render=high-poly: FPS: 195 FrameTime: 5.128 ms
[bump] bump-render=normals: FPS: 233 FrameTime: 4.292 ms
[bump] bump-render=height: FPS: 217 FrameTime: 4.608 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 194 FrameTime: 5.155 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 119 FrameTime: 8.403 ms
[pulsar] light=false:quads=5:texture=false: FPS: 203 FrameTime: 4.926 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 210 FrameTime: 4.762 ms
[desktop] effect=shadow:windows=4: FPS: 226 FrameTime: 4.425 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 179 FrameTime: 5.587 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 191 FrameTime: 5.236 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 175 FrameTime: 5.714 ms
[ideas] speed=duration: FPS: 221 FrameTime: 4.525 ms
[jellyfish] <default>: FPS: 139 FrameTime: 7.194 ms
[terrain] <default>: FPS: 27 FrameTime: 37.037 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 253 FrameTime: 3.953 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 192 FrameTime: 5.208 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 253 FrameTime: 3.953 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 256 FrameTime: 3.906 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 199 FrameTime: 5.025 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 237 FrameTime: 4.219 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 260 FrameTime: 3.846 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 213 FrameTime: 4.695 ms
=======================================================
                                  glmark2 Score: 204 
=======================================================
ventrical@ventrical-desktop:~$

So ..according to this test it can be assumed that with-mir is twice as fast as xorg and I think we are running the same graphics adapter cards (nVidia GeForce 218).

There is the old mir bug, though, that old bug that makes typing in from the keyboard show a significant delay.

**grahammechanical** · April 25th, 2014

I have just run glmark2 on my 14.04 install that I was using as a fallback OS. I installed ubuntu-desktop-mir and unity-system-compositor and I made sure that it was running on xmir and this is what I got

graham@sda2-fallback:~$ glmark2=========================================== ============
glmark2 2012.08
================================================== =====
OpenGL Information
GL_VENDOR: nouveau
GL_RENDERER: Gallium 0.4 on NVA5
GL_VERSION: 3.0 Mesa 10.1.0
================================================== =====
[build] use-vbo=false: FPS: 287 FrameTime: 3.484 ms
[build] use-vbo=true: FPS: 324 FrameTime: 3.086 ms
[texture] texture-filter=nearest: FPS: 309 FrameTime: 3.236 ms
[texture] texture-filter=linear: FPS: 310 FrameTime: 3.226 ms
[texture] texture-filter=mipmap: FPS: 319 FrameTime: 3.135 ms
[shading] shading=gouraud: FPS: 304 FrameTime: 3.289 ms
[shading] shading=blinn-phong-inf: FPS: 303 FrameTime: 3.300 ms
[shading] shading=phong: FPS: 301 FrameTime: 3.322 ms
[bump] bump-render=high-poly: FPS: 250 FrameTime: 4.000 ms
[bump] bump-render=normals: FPS: 324 FrameTime: 3.086 ms
[bump] bump-render=height: FPS: 317 FrameTime: 3.155 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 277 FrameTime: 3.610 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 191 FrameTime: 5.236 ms
[pulsar] light=false:quads=5:texture=false: FPS: 283 FrameTime: 3.534 ms
[desktop] blur-radius=5:effect=blur

asses=1:separable=true:windows=4: FPS: 299 FrameTime: 3.344 ms
[desktop] effect=shadow:windows=4: FPS: 301 FrameTime: 3.322 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 211 FrameTime: 4.739 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 220 FrameTime: 4.545 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 214 FrameTime: 4.673 ms
[ideas] speed=duration: FPS: 285 FrameTime: 3.509 ms
[jellyfish] <default>: FPS: 196 FrameTime: 5.102 ms
[terrain] <default>: FPS: 48 FrameTime: 20.833 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 312 FrameTime: 3.205 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 296 FrameTime: 3.378 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 311 FrameTime: 3.215 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 312 FrameTime: 3.205 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 306 FrameTime: 3.268 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 312 FrameTime: 3.205 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 311 FrameTime: 3.215 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 289 FrameTime: 3.460 ms
================================================== =====
glmark2 Score: 277
================================================== =====

I have xserver glmark2 score = 101 and xmir glmark2 score = 277. Yes, similar Nvidia card. GT220 but the important point is not that we are comparing the results from two different machines but we are comparing two results on the same machine. The difference being the video server, x compared to mir.

Regards.

**ventrical** · April 25th, 2014

I have been doing some research on testing graphics adapters.. etc...and some say that fps rates are not an accurate way to test. To each their own. By your results and the results I have also put up I would say it is safe to assume that these test prove that mir runs significantly faster than xorg.

Regards..

**Mateusz Stachowski** · April 26th, 2014

First of all glxgears is not a benchmark. Then you have this:

Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.

The results should be around the refresh of your monitor which in most cases this would be 60 FPS. You get much more FPS under XMir because of this bug:

[regression] GL(X) apps don't have vsync when running under XMir

**ventrical** · April 26th, 2014

Originally Posted by Mateusz Stachowski

First of all glxgears is not a benchmark. Then you have this:

The results should be around the refresh of your monitor which in most cases this would be 60 FPS. You get much more FPS under XMir because of this bug:

[regression] GL(X) apps don't have vsync when running under XMir

How about the Blob(s) test? <from USC> They both cannot be wrong , can they? or is this mir bug making a run across the board?

regards..

Thread: mir performance with glxgears and other info

Thread Tools

Display

mir performance with glxgears and other info

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Re: mir performance with glxgears

Bookmarks

Bookmarks

Posting Permissions