Announcement

Collapse
No announcement yet.

AMD P-State v4 Linux Testing On A Ryzen 5 5500U

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • AMD P-State v4 Linux Testing On A Ryzen 5 5500U

    Phoronix: AMD P-State v4 Linux Testing On A Ryzen 5 5500U

    With AMD having published a new revision to their AMD P-State Linux CPU frequency scaling driver that they are working towards mainlining with a goal of better power efficiency on Linux, here are some initial benchmarks of that new patch series when using a Ryzen 5 5500U notebook.

    https://www.phoronix.com/vr.php?view=30709

  • #2
    Basically useless

    Comment


    • #3
      Originally posted by wooque View Post
      Basically useless
      Depends on how you interpret the results. What I see shows amd-pstate-performance being a great choice for desktops while their powersave and schedutil pstates need a bit of optimizing compared to the acpi-cpufreq versions...and that having both shedutil and ondemand seems redundant. For laptops and energy conscious they're not the best choice, but for desktops where you're not worried about power saving and/or battery life their performance option actually gives the best performance.

      Comment


      • #4
        I have this kind of laptop with the same processor and whatnot, and looking by the disassembled ACPI DSDT data it seems like Lenovo added a Linux mode for the device. I've enabled it myself (adding
        Code:
        acpi_osi=Linux
        to the kernel command line), but I'm not sure if it actually has an impact in the power management with this AMD P-State driver. Would be great if someone is be able to look into that by doing a measured test

        Comment


        • #5
          Feels like when trying to test system power efficiency it would make sense to benchmark a moderate load. Like if you fps limit to screen refresh rate when doing game tests.

          Comment


          • #6
            Originally posted by wooque View Post
            Basically useless
            Not quite, I noticed that with amd-pstate the max turbo is reached more often and longer.

            Comment


            • #7
              Originally posted by wooque View Post
              Basically useless
              ...unless you want to save power.

              Comment


              • #8
                Originally posted by tildearrow View Post

                ...unless you want to save power.
                -0.4W is not something you will notice

                Comment


                • #9
                  Originally posted by wooque View Post

                  -0.4W is not something you will notice
                  Well I can say in practice the average on a non benchmark load is more, often at least 1W less, sometimes ~2W

                  Also, anyway, let say you laptop draws about ~7W and the battery last up too 7 hours, so you get ~21 minutes more with -0.4W, that is noticable.

                  Comment


                  • #10
                    For the compilation tests it would be better to have the results in time and Joules of energy used so you can determine which is more efficient.

                    Comment

                    Working...
                    X