Page 2 of 2 FirstFirst 12
Results 11 to 20 of 20

Thread: Benchmarking The Intel P-State, CPUfreq Changes

  1. #11
    Join Date
    Aug 2012
    Location
    Pennsylvania, United States
    Posts
    1,875

    Default

    Quote Originally Posted by GreatEmerald View Post
    Or just make a systemd unit file that sets the kernel variables to needed values on each boot, if a static setting is desired.
    That works too, its just a personal preference for me to use the designated tool rather than hack together a unit file / shell script. But you are right, that would work

  2. #12
    Join Date
    Apr 2013
    Posts
    39

    Default

    Quote Originally Posted by Pontostroy View Post
    Problem with _RADEON_ driver and kernel 3.8+
    Is this with Intel or AMD cpu? Also, are you sure that it's a problem only with 3.8+? IIRC I observed reduced performance with ondemand governor long before 3.8, I'm used to run custom scripts to switch cpu governor and gpu profile between ondemand/low and performance/high depending on whether I need max performance (for games and benchmarks) or not.

    Probably we need more clever cpu governor that would take into account gpu activity to ensure best performance with OpenGL & OpenCL apps. Though I think the behavior of the ondemand governor can also be improved by tuning its parameters, e.g. the following seems to help on my system (though as I said I'm usually just switching to performance governor, so I didn't test it a lot):

    Code:
    echo 60 > /sys/devices/system/cpu/cpufreq/ondemand/up_threshold
    This parameter is cpu load (%) that triggers increase of cpu freq, default value is 95%, but according to "cpupower monitor" cpu load per core is simply not reaching this threshold with most opengl apps and r600g for me, due to load balancing between cores etc.

  3. #13
    Join Date
    Sep 2008
    Location
    Vilnius, Lithuania
    Posts
    2,537

    Default

    Quote Originally Posted by Ericg View Post
    That works too, its just a personal preference for me to use the designated tool rather than hack together a unit file / shell script. But you are right, that would work
    Well, when using Arch or Gentoo, such unit files shouldn't be much of a problem. You're configuring everything manually, anyway.

  4. #14
    Join Date
    Aug 2012
    Location
    Pennsylvania, United States
    Posts
    1,875

    Default

    Quote Originally Posted by GreatEmerald View Post
    Well, when using Arch or Gentoo, such unit files shouldn't be much of a problem. You're configuring everything manually, anyway.
    True, but I also like being able to go back to the wikis and run off them. So if I start deviating from the wiki's too much then I might forget what I ever did originally haha

  5. #15

    Default

    Quote Originally Posted by vadimg View Post
    Though I think the behavior of the ondemand governor can also be improved by tuning its parameters, e.g. the following seems to help on my system (though as I said I'm usually just switching to performance governor, so I didn't test it a lot):

    Code:
    echo 60 > /sys/devices/system/cpu/cpufreq/ondemand/up_threshold
    This parameter is cpu load (%) that triggers increase of cpu freq, default value is 95%, but according to "cpupower monitor" cpu load per core is simply not reaching this threshold with most opengl apps and r600g for me, due to load balancing between cores etc.
    95% does seem too high, honestly. Perhaps they designed it with the expectation that most CPU heavy loads would be hitting 100%.

  6. #16
    Join Date
    Mar 2013
    Posts
    21

    Default don't select governors.,...

    Quote Originally Posted by AnonymousCoward View Post
    95% does seem too high, honestly. Perhaps they designed it with the expectation that most CPU heavy loads would be hitting 100%.
    The native P state driver does NOT use cpufreq governors.... no point selecting or setting them (in fact, nothing good can come out that so please just don't do that)

  7. #17
    Join Date
    Mar 2013
    Posts
    21

    Default

    Quote Originally Posted by fenrus View Post
    The native P state driver does NOT use cpufreq governors.... no point selecting or setting them (in fact, nothing good can come out that so please just don't do that)
    also one general note: for benchmarks that run fully flat out, don't expect many changes; that's the easiest case where pretty much every governor will pick the highest performance point and stay there.
    the most interesting benchmarks for this sort of thing are cases where the cpu is not 100% busy.

  8. #18
    Join Date
    Oct 2009
    Posts
    2,081

    Default

    Wow!!! LMFAO! Somebody doesn't know the first thing about frequency governors or how to test them. Unless there is something particularly broken about them, when maximum CPU is demanded, they will ALL ramp the CPU up to full power. PERIOD. END OF STORY. That means that there WILL BE no difference in peak performance, which means that testing the peak performance of them is entirely absurd!

    What you need to test is POWER CONSUMPTION and SYSTEM RESPONSIVENESS at low to moderate loads.

  9. #19
    Join Date
    Sep 2008
    Location
    Vilnius, Lithuania
    Posts
    2,537

    Default

    Quote Originally Posted by droidhacker View Post
    Wow!!! LMFAO! Somebody doesn't know the first thing about frequency governors or how to test them. Unless there is something particularly broken about them, when maximum CPU is demanded, they will ALL ramp the CPU up to full power. PERIOD. END OF STORY. That means that there WILL BE no difference in peak performance, which means that testing the peak performance of them is entirely absurd!

    What you need to test is POWER CONSUMPTION and SYSTEM RESPONSIVENESS at low to moderate loads.
    Not the powersave ones, and conservative can affect the non-peak performance.

  10. #20
    Join Date
    Jun 2013
    Posts
    1

    Default Powersavings ?

    Hi,

    The pstate is around now for several weeks and I wonder if there are people, who have actually measured some energy savings.

    So, before I try my newest kernel compilation, are there any powertop measurements or similar which could show
    improvements in battery life. This is what concerns me most, I am not interested in playing high-cpu games but to hear
    experiences in normal programming/email/surfing usage.

    Thanks to all the others

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •