LinuxQuestions.org
Visit Jeremy's Blog.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Hardware
User Name
Password
Linux - Hardware This forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?

Notices


Reply
  Search this Thread
Old 12-07-2022, 05:56 PM   #1
borgward
Member
 
Registered: Apr 2005
Location: Austin, Texas
Distribution: Feather, Darn Small Linux
Posts: 518

Rep: Reputation: 21
Stress Test


I stress tested Radeon HD 6970M on 27" mid 2011 iMac. How do Interpret the output?

mint@mint:~$ glmark2
=======================================================
glmark2 2014.03+git20150611.fa71af2d
=======================================================
OpenGL Information
GL_VENDOR: X.Org
GL_RENDERER: AMD TURKS (DRM 2.50.0 / 5.4.0-58-generic, LLVM 10.0.0)
GL_VERSION: 3.1 Mesa 20.0.8
=======================================================
[build] use-vbo=false: FPS: 2323 FrameTime: 0.430 ms
[build] use-vbo=true: FPS: 2876 FrameTime: 0.348 ms
[texture] texture-filter=nearest: FPS: 2805 FrameTime: 0.357 ms
[texture] texture-filter=linear: FPS: 2850 FrameTime: 0.351 ms
[texture] texture-filter=mipmap: FPS: 2773 FrameTime: 0.361 ms
[shading] shading=gouraud: FPS: 2482 FrameTime: 0.403 ms
[shading] shading=blinn-phong-inf: FPS: 2567 FrameTime: 0.390 ms
[shading] shading=phong: FPS: 2442 FrameTime: 0.410 ms
[shading] shading=cel: FPS: 2494 FrameTime: 0.401 ms
[bump] bump-render=high-poly: FPS: 1713 FrameTime: 0.584 ms
[bump] bump-render=normals: FPS: 3054 FrameTime: 0.327 ms
[bump] bump-render=height: FPS: 3013 FrameTime: 0.332 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 2435 FrameTime: 0.411 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 1338 FrameTime: 0.747 ms
[pulsar] light=false:quads=5:texture=false: FPS: 2838 FrameTime: 0.352 ms
[desktop] blur-radius=5:effect=blurasses=1:separable=true:windows=4: FPS: 1158 FrameTime: 0.864 ms
[desktop] effect=shadow:windows=4: FPS: 1903 FrameTime: 0.525 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 550 FrameTime: 1.818 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 776 FrameTime: 1.289 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 612 FrameTime: 1.634 ms
[ideas] speed=duration: FPS: 1326 FrameTime: 0.754 ms
[jellyfish] <default>: FPS: 1855 FrameTime: 0.539 ms
[terrain] <default>: FPS: 221 FrameTime: 4.525 ms
[shadow] <default>: FPS: 845 FrameTime: 1.183 ms
[refract] <default>: FPS: 311 FrameTime: 3.215 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 2234 FrameTime: 0.448 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 2319 FrameTime: 0.431 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 2239 FrameTime: 0.447 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 2280 FrameTime: 0.439 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 2339 FrameTime: 0.428 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 2313 FrameTime: 0.432 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 2320 FrameTime: 0.431 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 1318 FrameTime: 0.759 ms
=======================================================
glmark2 Score: 1967
=======================================================

I do not know the score indicates. I've seen some posted that are much higher. Is the score relative to the individual GPU?
Would a perfectly good low range GPU have a lower score than perfectly good super dooper GPU?

mint@mint:~$ glxgears
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
308 frames in 5.0 seconds = 61.583 FPS
300 frames in 5.0 seconds = 59.835 FPS
300 frames in 5.0 seconds = 59.782 FPS
300 frames in 5.0 seconds = 59.900 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.843 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.843 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.779 FPS
300 frames in 5.0 seconds = 59.899 FPS
300 frames in 5.0 seconds = 59.835 FPS
300 frames in 5.0 seconds = 59.840 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.842 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.842 FPS
300 frames in 5.0 seconds = 59.838 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.840 FPS
300 frames in 5.0 seconds = 59.840 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.838 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.838 FPS
300 frames in 5.0 seconds = 59.838 FPS
299 frames in 5.0 seconds = 59.787 FPS
300 frames in 5.0 seconds = 59.890 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.842 FPS
300 frames in 5.0 seconds = 59.838 FPS
300 frames in 5.0 seconds = 59.840 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.841 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.839 FPS
300 frames in 5.0 seconds = 59.806 FPS
300 frames in 5.0 seconds = 59.873 FPS
300 frames in 5.0 seconds = 59.837 FPS
300 frames in 5.0 seconds = 59.840 FPS
300 frames in 5.0 seconds = 59.842 FPS
300 frames in 5.0 seconds = 59.836 FPS
300 frames in 5.0 seconds = 59.842 FPS
300 frames in 5.0 seconds = 59.836 FPS
^C
mint@mint:~$

My understanding of this test is that if the numbers do not go down very much, that is good. How long should this test be run?
 
Old 12-08-2022, 01:55 AM   #2
pan64
LQ Addict
 
Registered: Mar 2012
Location: Hungary
Distribution: debian/ubuntu/suse ...
Posts: 23,673

Rep: Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816Reputation: 7816
You know, computers cannot display anything but pictures. When you see something moving on the monitor, it means that many still images (called frames) are displayed on it one after the other. FPS means (frame per second) how many pictures were shown within a second. https://en.wikipedia.org/wiki/Frame_rate. From this point of view 60 FPS is good enough for games and movies, higher values mean that the system is able to operate smoothly, lower values usually mean flickering (below 20-30).
Otherwise see here: https://linuxreviews.org/Glmark2
 
Old 12-08-2022, 09:05 AM   #3
borgward
Member
 
Registered: Apr 2005
Location: Austin, Texas
Distribution: Feather, Darn Small Linux
Posts: 518

Original Poster
Rep: Reputation: 21
Anyway to monitor GPU temperature while running the stress test?
 
Old 12-08-2022, 09:38 AM   #4
boughtonp
Senior Member
 
Registered: Feb 2007
Location: UK
Distribution: Debian
Posts: 3,866

Rep: Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722Reputation: 2722

First, use "[code]..[/code]" tags for command output.

Second, you're not running a stress test you're running benchmarking software. Benchmarking can be used for stress testing, but there's a subtle difference between "how fast can X go" and "how long can X go at its fastest for".


What's your reason for doing this?


You can see sensor information with the command sensors - install lm-sensors if you don't have it. (On newer hardware, upgrading kernel/firmware/etc can be necessary to see all sensors; I guess it shouldn't be an issue with something from 2011.)

If that returns the information you want, the poorly named "s-tui" aka "Stress Terminal UI" can show you the output over time on a command line, or there's psensor for a GTK application, and Xfce and KDE both have sensor utilities (not sure they can do graphing).

Of course, you should be including system information in a thread like this - at least "inxi -GS"


Last edited by boughtonp; 12-08-2022 at 09:40 AM.
 
Old 12-08-2022, 02:13 PM   #5
borgward
Member
 
Registered: Apr 2005
Location: Austin, Texas
Distribution: Feather, Darn Small Linux
Posts: 518

Original Poster
Rep: Reputation: 21
Imac OS stalls booting to login. That indicates bad GPU. Linux Live session boots no problem. Mac forum said the GPU may be on it's way out even if Linux boots it. The problem is either GPU or Mac OS and or driver issue. I want to determine if the GPU is good or not. If good I will then try to restore the Mac OS and then dual boot Linux.

I am not going to deal with replacing the GPU as that's a real slog. Used GPU costs $20 to $100 and New from China is $200. Used will probably last 3 months. I think the new is probably NOS and won't be reliable either.
 
Old 12-10-2022, 11:45 AM   #6
Arnulf
Member
 
Registered: Jan 2022
Location: Hanover, Germany
Distribution: Slackware
Posts: 306

Rep: Reputation: 111Reputation: 111
Quote:
Originally Posted by borgward View Post
I stress tested Radeon HD 6970M on 27" mid 2011 iMac. How do Interpret the output?
[…]
mint@mint:~$ glxgears
Running synchronized to the vertical refresh. The framerate should be
approximately the same as the monitor refresh rate.
308 frames in 5.0 seconds = 61.583 FPS
300 frames in 5.0 seconds = 59.835 FPS
300 frames in 5.0 seconds = 59.782 FPS
  1. glxgears is not a benchmark.
  2. Running glxgears in that way is only useful to test the monitor.
  3. For GPU test, glxgears should be started after executing: export vblank_mode=0.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
write stress to file? run vmstat/iostat simultaneously with stress commands rmcateer Programming 5 02-27-2015 08:27 PM
Im am running a stress autocorrelation function,to calculate stress correlation funct lehloks Programming 4 07-08-2013 06:59 AM
Load test, boundary test & stress test for USB EHCI/xHCI driver rama_toshiba Linux - Kernel 5 02-29-2012 03:43 PM

LinuxQuestions.org > Forums > Linux Forums > Linux - Hardware

All times are GMT -5. The time now is 09:24 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Open Source Consulting | Domain Registration