Announcement

Collapse
No announcement yet.

Benchmarks Of The Gentoo-Based Calculate Linux

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

  • Benchmarks Of The Gentoo-Based Calculate Linux

    Phoronix: Benchmarks Of The Gentoo-Based Calculate Linux

    Traditionally at Phoronix we have stayed away from publishing benchmarks of Gentoo and similar source-based distributions for the lack of them having a standard or "stock" configuration for which one can easily replicate our tested software stack due to all of the different variables that come into play so the value of these benchmarks are much less compared to those distributions providing pre-compiled binaries for a standardized set of packages. However, satisfying a number of requests, we are publishing such benchmarks today. Rather than using Gentoo itself for benchmarking, we are using Calculate Linux Desktop, which is Gentoo-based while providing a very nice "out of the box" experience, i686 and x86_64 binaries, and overall is a polished and user-friendly Gentoo experience.

    http://www.phoronix.com/vr.php?view=15170

  • #2
    First of all: thanks for the benchmarks.

    However, this is a slap in the face for all Gentoo users: you were,
    as I understood, asked to perform a comparison against Gentoo. Instead, you pick some random distribution with *binary* packages, which are also pretty outdated. Calculate Linux is not Gentoo, despite it is being based on.

    To clarify the point of Gentoo (IMHO, of course):
    - CPU-optimized compilation of source packages
    - availability of most recent packages (e.g. GCC, Glibc, X11, ...)

    Please: either work out a proper benchmark, or, if you can't, just do not mention "Gentoo" in your article titles!

    Thanks!
    paravoid


    Originally posted by phoronix View Post
    Phoronix: Benchmarks Of The Gentoo-Based Calculate Linux

    Traditionally at Phoronix we have stayed away from publishing benchmarks of Gentoo and similar source-based distributions for the lack of them having a standard or "stock" configuration for which one can easily replicate our tested software stack due to all of the different variables that come into play so the value of these benchmarks are much less compared to those distributions providing pre-compiled binaries for a standardized set of packages. However, satisfying a number of requests, we are publishing such benchmarks today. Rather than using Gentoo itself for benchmarking, we are using Calculate Linux Desktop, which is Gentoo-based while providing a very nice "out of the box" experience, i686 and x86_64 binaries, and overall is a polished and user-friendly Gentoo experience.

    http://www.phoronix.com/vr.php?view=15170

    Comment


    • #3
      graphs

      Please, for easier viewing consider doing the following:
      When you have multiple test-systems and are *not* comparing the systems, give them separate scales.

      Avoiding graphs like this should be a high priority, as one side of it is completely useless...
      [img]http://www.phoronix.com/data/img/results/calculate_gentoo_benchmarks/7.png[img]

      thanks for the tests, anyway, though what most people asking for gentoo benchmarks likely want is likely far from these benchmarks

      Comment


      • #4
        Interesting, will have a look into the bzip and gzip numbers.

        Comment


        • #5
          In before "Gentoo's not supposed to be faster until it's completely built for the exact cpu"

          Comment


          • #6
            Originally posted by curaga View Post
            In before "Gentoo's not supposed to be faster until it's completely built for the exact cpu"
            A common misconception, its usually faster due to USE flags keeping your system down to what you use only.

            This can have the reverse effect also, with a Gentoo system being more bloated because the user has enabled every USE under the sun.

            Gentoo is always going to have the "ricer" reputation I suppose


            [In before http://funroll-loops.info/]

            Comment


            • #7
              In general, performance improvements due to CPU-specific compiler optimizations are minimal, if at all visible. (If you need good performance, there is no way around custom intrinsic/asm code.)

              Comment


              • #8
                Yeah, but performance improvement due to, say, disabling Java in OpenOffice, or disabling PulseAudio systemwide is considerable.

                And you can't do this easily with a binary distro.

                Comment


                • #9
                  Sigh...

                  Really, why try to please the clueless part of the Gentoo users?

                  Everyone with a little bit of common sense knows that we can't benchmark Gentoo reliably because person A's Gentoo isn't person B's Gentoo. They know it, we know it, so why try?

                  Comment


                  • #10
                    Originally posted by MaestroMaus View Post
                    Sigh...

                    Really, why try to please the clueless part of the Gentoo users?

                    Everyone with a little bit of common sense knows that we can't benchmark Gentoo reliably because person A's Gentoo isn't person B's Gentoo. They know it, we know it, so why try?
                    *sigh* This is just the point. Gentoo is not a general purpose Linux distribution. Instead, one can tweak the O/S to the maximum, which you cannot do with any other distribution.

                    You say we can't benchmark it ? IMHO, it would be interesting to see the performance numbers from the Phoronix benchmark. But only, if the system is tuned properly. Using stock Gentoo, for example, won't gain much performance I believe.

                    Comment


                    • #11
                      BTW, I've never met a gentoo user who's into high performance tuning, and I know plenty.

                      Most of them simply appreciate the flexibility and the source-based approach. Chances are that stock gentoo will not perform worse than a good binary distro, and that's sufficient for the most.

                      What you get is flexibility. Disabling GNOME or PulseAudio is a one-liner. In Ubuntu, you cannot even do it.

                      Comment


                      • #12
                        Originally posted by pingufunkybeat View Post
                        BTW, I've never met a gentoo user who's into high performance tuning, and I know plenty.

                        Most of them simply appreciate the flexibility and the source-based approach. Chances are that stock gentoo will not perform worse than a good binary distro, and that's sufficient for the most.

                        What you get is flexibility. Disabling GNOME or PulseAudio is a one-liner. In Ubuntu, you cannot even do it.
                        This sums it up for me. I'm a very long time Gentoo user and I always skip through these benchmarks to the conclusion because I don't really care if one distro took 0.05 seconds longer to gzip a tarball than another. Being able to trim the unnecessary fat off my packages, however, does interest me very much.

                        Comment


                        • #13
                          So what's the point--given that performance isn't--hard drive space?

                          Comment


                          • #14
                            What's the point of what? Flexibility?

                            I can install Compiz without installing all of GNOME. You can't. I can install MPlayer with multithreading, but without PulseAudio support. I can install OpenOffice without Java, but with a Qt4 frontend. I can actually choose whether I want to use GStreamer or xinelib as a backend with most players. You simply turn on options you want, and turn off those you don't want. Most packages have a wide range of possible functionality, and with most distributions, you do not get to choose it, it is chosen for you because of the limitations of the binary package formats. Gentoo lets you have Linux the way YOU like it, not the way Shuttleworth likes it.

                            That's worth something.

                            Obviously, not all users want this, and this is the reason why there are different distributions.

                            Comment


                            • #15
                              I'm not saying I don't care about performance. USE flags can help performance, primarily saving RAM by not loading libraries you don't need. But these benchmarks don't measure RAM usage and will (probably) always compare packages built with roughly the same options and features otherwise some might not consider it a fair test. In other words, benchmarks are probably useless unless they are performed under your ideal setup.

                              Comment

                              Working...
                              X