Announcement

Collapse
No announcement yet.

Ubuntu 24.04 Boosts Performance, Outperforming Windows 11 On The AMD Ryzen Framework 16 Laptop

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

  • Ubuntu 24.04 Boosts Performance, Outperforming Windows 11 On The AMD Ryzen Framework 16 Laptop

    Phoronix: Ubuntu 24.04 Boosts Performance, Outperforming Windows 11 On The AMD Ryzen Framework 16 Laptop

    With the Framework 16 laptop one of the performance pieces I've been meaning to carry out has been seeing out Linux performs against Microsoft Windows 11 for this AMD Ryzen 7 7840HS powered modular/upgradeable laptop. Recently getting around to it in my benchmarking queue, I also compared the performance of Ubuntu 23.10 to the near final Ubuntu 24.04 LTS on this laptop up against a fully-updated Microsoft Windows 11 installation.

    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 must think the CPU draws (quite a bit) less power on Windows

    Comment


    • #3
      There is no question that Windows 11 is a step back from Windows 10, anyone with a modicum of intellectual honesty has to admit that to himself.

      In many ways, Win 7 was the peak of Windows perfection, Win 2k was way better than any of the Win 9x clan, Win XP didn't come into it's own until XP64, Vista sucked, and Win 7 may have been Microsoft's magnum opus.

      Win 10, especially after updating it comes close to Win 7, but Win 11 has regressed in many ways, in the context menu, responsiveness, now that Microsoft is supposedly thinking about showing ads in the start menu, it's a shame.

      The one thing these benchmarks are not capable of measuring is latency i.e. the amount of time between when a user initiates an action and when it happens.

      Think of it like throttle response, the benchmarks show acceleration times like 0-60 and quarter mile, but they don't capture the driving experience of how quickly an engine revs up and down or how flat the power band is which is what matters to the driver most in every day driving, how responsive the car is to throttle changes.

      This is on place where Manjaro really shines, i haven't dived into what they are doing, but if you install Manjaro Mate, Ubuntu Mate, Fedora Mater and Win 11 on the same system the difference in responsiveness, from a picture opening up when you click on the icon, to when all the files in a massive folder show up, to when new windows open, is quite stark.

      With Manjaro KDE the responsiveness is so great you would think you upgraded you computer but there just so many stupid annoying bugs that it can't be used long term.

      Comment


      • #4
        I presume all these measurements done on balanced with AC adapter that came with machine plugged in right?

        It would certainly be interesting to compare without the AC adapter because PPD 0.21 does different tuning in this case.

        Also to see how "performance" and "power saver" compare between OSes.

        Comment


        • #5
          I'd kinda like an analysis of what these performance differences mean. Is it filesystem or disk or scheduler? Or is it background processes? A lot of these benchmarks are highly CPU bound where Linux is quite a bit faster than Windows, but being CPU bound many of these results should be nearly identical. I'm having trouble interpreting where exactly the performance difference is coming from.

          Comment


          • #6
            Ars Tecnicha was bashing Framework a few days ago for failure to keep up with software and firmware updates - https://arstechnica.com/gadgets/2024...rking-on-them/

            Framework responded and said they're trying to improve their processes - https://frame.work/au/en/blog/enabli...ware-longevity

            Comment


            • #7
              Originally posted by superm1 View Post
              I presume all these measurements done on balanced with AC adapter that came with machine plugged in right?
              Yep, that's the
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #8
                Trillions of dollars, thousands of “top engineers” for 30+ years: yet bloated perf, crappy unconventional foundations from msdos heritage, still packed with unchanged UIs from vista/xp/older.

                I would feel ashamed by the result if I was them.

                Comment


                • #9
                  Originally posted by rmfx View Post
                  Trillions of dollars, thousands of “top engineers” for 30+ years: yet bloated perf, crappy unconventional foundations from msdos heritage, still packed with unchanged UIs from vista/xp/older.

                  I would feel ashamed by the result if I was them.
                  Not trillions of dollars, billions yes, but not trillions.

                  I think the Windows code base is suffering from the same thing projects like Gnome, KDE, and the Linux kernel are, namely lots of legacy cruft.

                  I've said this before, what i would love to see MS do is start over with a relatively clean code base, start with BSD as the core, use the Lumina desktop, both of these are under the commercially friendly BSD style licenses, layer .NET on top of it, which already works on BSD, layer DX on top of it, and call it a day.

                  Release two versions, one a legally free version that people can download and install as they desire, this one will have no updates and will not be able to run MS like games or MS Office and a premium or pro version that can run MS games and MS Office.

                  This version would have product activation and enforce a 1 seat per license model.

                  I think that would be great for everyone.

                  Comment


                  • #10
                    How does the battery life compare on the 3 different OSs ???

                    Comment

                    Working...
                    X