Slackware This Forum is for the discussion of Slackware Linux.
|
Notices |
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
Are you new to LinuxQuestions.org? Visit the following links:
Site Howto |
Site FAQ |
Sitemap |
Register Now
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
|
|
|
03-25-2014, 11:04 AM
|
#1
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Rep:
|
Pitty performance on OpenGL benchmark
Hi, I am running Slackware64 14.1 with 3.13.6 kernel, radeon 6670, latest mesa 64 bits (forgot the number, I am not on my system now) and alien's bob mesa compat32 version. I decided to run the Phoronix benchmark tool and I tried the OpenGL test and I was getting 3 fps on it! Is is normal? I mean, for the free radeon driver? I wonder if I ddi something wrong while compiling mesa.
|
|
|
03-25-2014, 11:18 AM
|
#2
|
Senior Member
Registered: May 2006
Posts: 3,091
|
What is the output of 'glxinfo | grep -i render'?
Adam
|
|
|
03-25-2014, 11:53 AM
|
#3
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
pedro@darkstar:~$ glxinfo | grep -i render
direct rendering: Yes
GLX_MESA_multithread_makecurrent, GLX_MESA_query_renderer,
OpenGL renderer string: Gallium 0.4 on AMD TURKS
GL_MESA_texture_signed_rgba, GL_NV_conditional_render, GL_NV_depth_clamp,
GL_NV_blend_square, GL_NV_conditional_render, GL_NV_depth_clamp,
|
|
|
03-25-2014, 11:58 AM
|
#4
|
Senior Member
Registered: May 2006
Posts: 3,091
|
Well your drivers (at least the 64-bit version of them) are installed properly. Have you tried playing any native OpenGL games to see how they work?
|
|
|
03-25-2014, 12:11 PM
|
#5
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
Still didn't try because I don't have Linux games, the benchmark the Phoronix Suite was from a game tho.
|
|
|
03-25-2014, 12:15 PM
|
#6
|
Senior Member
Registered: May 2006
Posts: 3,091
|
Was it a 32-bit or 64-bit app?
Adam
e
|
|
|
03-25-2014, 12:17 PM
|
#7
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
Pretty sure it was 64bits and, if it was 32bits, we can infer there is something wrong with alien's bob package?
|
|
|
03-25-2014, 12:32 PM
|
#8
|
Senior Member
Registered: May 2006
Posts: 3,091
|
Not necessarily. The app could be 32-bits, your drivers could be just fine, but the benchmark might simply want functionality that isn't present in the free radeon driver.
Adam
|
|
|
03-25-2014, 12:47 PM
|
#9
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
Hmm, was thinking about that too. Yes, it might be something not implemented by the free driver.
|
|
|
03-25-2014, 12:49 PM
|
#10
|
LQ Veteran
Registered: May 2008
Posts: 7,053
|
If you've not got any games installed and want to put OpenGL through it's paces then you could give the screensaver hacks shipped with xscreensaver a go.
/usr/libexec/xscreensaver/glschool -fps -fog -nfish 250 -delay 1
|
|
|
03-25-2014, 01:38 PM
|
#11
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
Quote:
Originally Posted by GazL
If you've not got any games installed and want to put OpenGL through it's paces then you could give the screensaver hacks shipped with xscreensaver a go.
/usr/libexec/xscreensaver/glschool -fps -fog -nfish 250 -delay 1
|
I am running mesa stock version now, will give it a try then upgrade to latest mesa and see if there is any difference.
EDIT: Oh, nvm, thought it was a more intensive task. Getting 30FPS and 72000 polys on it. Is that any good?
EDIT2: Slightly better perfomance with latest mesa, a solid 30FPS while the stock version was between 25 and 30. Not much but it is better than nothing.
Last edited by moisespedro; 03-25-2014 at 01:43 PM.
|
|
|
03-25-2014, 02:13 PM
|
#12
|
LQ Veteran
Registered: May 2008
Posts: 7,053
|
If you want to see the true speed you'll need to deactivate SYNC to VBlank. Otherwise you'll get a multiple of 30fps.
__GL_SYNC_TO_VBLANK=0 /usr/libexec/xscreensaver/glschool -fps -delay 1 -nfish 250
That gives me 75fps with the nvidia blob, but that's probably limited by the fact that the glschool process is pegging a CPU at 100%. If I start a second instance, they both run at that speed so I suspect the graphics processor still has some headroom and my dual core cpu is the bottleneck.
It's not really intended as a benchmark, just thought it'd give you something to compare.
|
|
|
03-25-2014, 02:49 PM
|
#13
|
Senior Member
Registered: Nov 2013
Location: Brazil
Distribution: Slackware
Posts: 1,223
Original Poster
Rep:
|
Quote:
Originally Posted by GazL
If you want to see the true speed you'll need to deactivate SYNC to VBlank. Otherwise you'll get a multiple of 30fps.
__GL_SYNC_TO_VBLANK=0 /usr/libexec/xscreensaver/glschool -fps -delay 1 -nfish 250
That gives me 75fps with the nvidia blob, but that's probably limited by the fact that the glschool process is pegging a CPU at 100%. If I start a second instance, they both run at that speed so I suspect the graphics processor still has some headroom and my dual core cpu is the bottleneck.
It's not really intended as a benchmark, just thought it'd give you something to compare.
|
What video card you have? Because 30 vs 75 is a huge difference
|
|
|
03-25-2014, 03:06 PM
|
#14
|
Member
Registered: Sep 2008
Location: The Netherlands
Distribution: Slackware64 current
Posts: 594
Rep:
|
I have a Radeon HD5750 and I reach between 24 and 28 fps.
Code:
bash-4.2$ glxinfo | grep -i render
direct rendering: Yes
OpenGL renderer string: Gallium 0.4 on AMD JUNIPER
GL_MESA_window_pos, GL_NV_blend_square, GL_NV_conditional_render,
You can achieve more performance with the amd catalyst driver.
Last edited by whizje; 03-25-2014 at 03:09 PM.
Reason: extra info
|
|
|
03-25-2014, 03:31 PM
|
#15
|
Member
Registered: Nov 2013
Posts: 744
Rep:
|
Quote:
Originally Posted by GazL
If you want to see the true speed you'll need to deactivate SYNC to VBlank. Otherwise you'll get a multiple of 30fps.
__GL_SYNC_TO_VBLANK=0 /usr/libexec/xscreensaver/glschool -fps -delay 1 -nfish 250
...
It's not really intended as a benchmark, just thought it'd give you something to compare.
|
checked this with perf
it is very cpu intensive (it calls pow() a lot)
i usually test with xonotic (tool comes with it)
but even that is more towards cpu bottleneck
Last edited by genss; 03-25-2014 at 03:33 PM.
|
|
|
All times are GMT -5. The time now is 06:59 AM.
|
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
|
Latest Threads
LQ News
|
|