Announcement

Collapse
No announcement yet.

Benchmarks: AMD Ryzen 7000 Series Performance Boosted With Ubuntu 23.10

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

  • Benchmarks: AMD Ryzen 7000 Series Performance Boosted With Ubuntu 23.10

    Phoronix: Benchmarks: AMD Ryzen 7000 Series Performance Boosted With Ubuntu 23.10

    With Ubuntu 23.10 due for release on Thursday, I've been benchmarking a number of systems to look at the Ubuntu 23.10 performance against prior releases like Ubuntu 23.04 and Ubuntu 22.04 LTS. Besides the open-source graphics driver performance for Intel and AMD Radeon graphics always being a stand-out improvement, one area that is particularly exciting with Ubuntu 23.10 is for those with newer AMD processors where there are some nice performance gains to find with this new Ubuntu Linux release. Here are side-by-side benchmarks of an AMD Ryzen 9 7950X desktop along with an Intel Core i9 13900K desktop while testing Ubuntu 22.04.3 LTS / Ubuntu 23.04 / Ubuntu 23.10.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    I will be even better with 6.6.
    ## VGA ##
    AMD: X1950XTX, HD3870, HD5870
    Intel: GMA45, HD3000 (Core i5 2500K)

    Comment


    • #3
      Why can amd beat intel in everything that requires raw force (i.e. processing power), but suck so much at browser benchmarks? Javascript jit favors intel scheduling?

      Comment


      • #4
        Originally posted by mlau View Post
        Why can amd beat intel in everything that requires raw force (i.e. processing power), but suck so much at browser benchmarks? Javascript jit favors intel scheduling?
        Likely the result of aggressive single-core boosting behaviour in conjunction with higher clockrates.

        Comment


        • #5
          Originally posted by mlau View Post
          Why can amd beat intel in everything that requires raw force (i.e. processing power), but suck so much at browser benchmarks? Javascript jit favors intel scheduling?
          Not only this, but also: Photoshop, Lightroom, Premiere Pro, After effects, sometimes 7950X loses by more than 30% and consumes more energy.

          That's why, because not every filter or effect in these programs takes up all cores.

          Additionally, the hardware encoders in AMD are also much weaker than Intel Quick Sync.‚Äč

          Comment


          • #6
            More performance and less power consumption, it seems all good. You said similar gain should be obersvable for zen 2 and 3?

            Comment


            • #7
              After my kernel upgrade and using pstate my cpu finally can downclock to 550MHz.

              Tbh I consider that way more important than the few percentage of more performance

              Comment


              • #8
                Originally posted by HEL88 View Post

                Not only this, but also: Photoshop, Lightroom, Premiere Pro, After effects, sometimes 7950X loses by more than 30% and consumes more energy.
                if (cpu_vendor() == VNDR_INTEL)
                fast_path()
                else
                slow_path()

                Comment


                • #9
                  Originally posted by Errinwright View Post

                  Likely the result of aggressive single-core boosting behaviour in conjunction with higher clockrates.
                  The AMD cores consume also more power, i.e. they're doing _something_, more than the intel cores, and yet
                  the results show their work is for naught, which is surprising since recent Zen have very good power management features
                  and perf/watt in the "usual" phoronix benchmarks is above intels.
                  Yes intel cores can boost better, but their power consumption is usually (i.e. seen on other benchmarks in the past)
                  also significantly higher than amds when they do that.
                  Last edited by mlau; 10 October 2023, 03:44 AM.

                  Comment


                  • #10
                    using AMD P-State EPP (active mode) with the powersave governor rather than ACPI CPUFreq with the Schedutil governor
                    Sad that even powersave is better than schedutil, but Ubuntu has another improvement waiting if they switch to performance on the next release.

                    Has anyone noticed that with EPP active on powersave, singlecore load doesn't raise frequencies? If I load up all cores it even gets above base clock but a single core stays below 1 GHz. Or is this buggy behavior from my platform?

                    Comment

                    Working...
                    X