Announcement

Collapse
No announcement yet.

Windows 10 vs. Ubuntu 20.10 Performance With Intel Tiger Lake, AMD Renoir

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

  • Windows 10 vs. Ubuntu 20.10 Performance With Intel Tiger Lake, AMD Renoir

    Phoronix: Windows 10 vs. Ubuntu 20.10 Performance With Intel Tiger Lake, AMD Renoir

    Stemming from our initial Intel Core i7 1165G7 "Tiger Lake" benchmarks on the Dell XPS 13 9310 last week and then also discovering better single-threaded performance on Ubuntu 20.10, one of the pressing questions was whether this is expected performance on Linux or if it's coming up short of Microsoft Windows for this first tier-one notebook to market with Intel Tiger Lake. So following those earlier tests I proceeded to do a Windows 10 Pro with all available updates comparison on Ubuntu 20.10 with the i7-1165G7. For added context, the same software stack and tests were repeated on an AMD Ryzen "Renoir" notebook.

    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
    But we need to mention, that AMD is still crippled
    Glibc-HWCAPS To Help With AMD Zen Optimizations, Other Per-CPU Performance Bits
    Written by Michael Larabel in GNU on 7 July 2020
    Stemming from Glibc semantics that effectively "cripple AMD"

    in just checking or Intel CPUs while AMD CPUs with Glibc are not even taking advantage of Haswell era CPU features,
    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

    Haswell launched June 4th, 2013




    And it is huge crippling
    Nov 18th, 2019 10:53 Discuss (67 Comments)
    MATLAB is a popular math computing environment in use by engineering firms, universities, and other research institutes. Some of its operations can be made to leverage Intel MKL (Math Kernel Library), which is poorly optimized for, and notoriously slow on AMD Ryzen processors. Reddit user Nedflanders1976 devised a way to restore anywhere between 20 to 300 percent performance on Ryzen and Ryzen Threadripper processors, by forcing MATLAB to use advanced instruction-sets such as AVX2. By default, MKL queries your processor's vendor ID string, and if it sees anything other than "GenuineIntel...," it falls back to SSE, posing a significant performance disadvantage to "AuthenticAMD" Ryzen processors that have a full IA SSE4, AVX, and AVX2 implementation
    MATLAB is a popular math computing environment in use by engineering firms, universities, and other research institutes. Some of its operations can be made to leverage Intel MKL (Math Kernel Library), which is poorly optimized for, and notoriously slow on AMD Ryzen processors. Reddit user Nedflander...



    Glibc-HWCAPS To Help With AMD Zen Optimizations, Other Per-CPU Performance Bits
    Written by Michael Larabel in GNU on 7 July 2020
    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


    GNU C Library 2.32 Released
    Written by Michael Larabel in GNU on 6 August 2020

    Sadly not making it for this release is the restartable sequences support. Additionally, the hardware capabilities discussion is also still ongoing.
    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




    • The current development version of glibc 2.33, releasing on or around February 1st, 2021.


    Latest News


    2020-08-05: glibc 2.32 released.

    2020-02-01: glibc 2.31 released.

    2019-08-01: glibc 2.30 released.

    2019-01-31: glibc 2.29 released.

    Comment


    • #3
      Originally posted by Peter Fodrek View Post
      But we need to mention, that AMD is still crippled
      Latest News


      2020-08-05: glibc 2.32 released.

      2020-02-01: glibc 2.31 released.

      2019-08-01: glibc 2.30 released.

      2019-01-31: glibc 2.29 released.
      I like this from the glibc 2.32 announcement:

      * The GNU C Library now provides the header file <sys/single_threaded.h> which declares the variable __libc_single_threaded. Applications are encouraged to use this variable for single-thread optimizations, instead of weak references to symbols historically defined in libpthread.

      Comment


      • #4
        Windows good for AMD
        Linux good for Intel

        Comment


        • #5
          Michael
          Speaking of Windows, Edge for Linux was launched in preview form. I took it for a spin to do what it is used for on windows and, surprise, it didn't started Chrome download on Google's website. Purposeful or only a innocent glitch? Firefox can be downloaded normally, though.

          Be the first to preview what's new by becoming a Microsoft Edge Insider and downloading the Insider channels.

          Comment


          • #6
            Mmmh, these benchmarks just add more confusion than else.

            I suspect Intel is pushing harder on the power drainage during the workload first seconds/minutes, then slows down to normal frequencies later.
            Short timed benchmarks or microbenchmarks are favourable to Intel or tied between the two, but benchmarks that lasts longer are much more favorable to AMD (look at Blender or Appleseed here )

            This is more evident for the Basis benchmark too, on the short one (~10 seconds) the chips are tied, on the longer ones the 4500u wins by a significant margin. On the same page the compression benchmarks, which are generally less than 60 seconds, are tied or favourable to Intel.

            Power figures would have been very useful, but are not provided

            Comment


            • #7
              Originally posted by blackshard View Post
              Power figures would have been very useful, but are not provided
              Especially idle and light workload power drain compared to windows. Those kind of workload is what most laptops do all day

              Comment


              • #8
                Originally posted by Mario Junior View Post
                Windows good for AMD
                Linux good for Intel
                AMD claims they support Linux (and they show it)

                At the end they support Windows better than Linux

                Comment


                • #9
                  Typo:

                  Originally posted by phoronix View Post
                  Ubuntu was much faster on each notebok.

                  Comment


                  • #10
                    Originally posted by tildearrow View Post

                    AMD claims they support Linux (and they show it)

                    At the end they support Windows better than Linux
                    Except we have no real data about Windows mitigations. Furthermore, it may be glibc crippling performance. Not for the first time..

                    Comment

                    Working...
                    X