Announcement

Collapse
No announcement yet.

Fedora 17: Beefier Than Ubuntu 12.04 LTS?

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

  • Fedora 17: Beefier Than Ubuntu 12.04 LTS?

    Phoronix: Fedora 17: Beefier Than Ubuntu 12.04 LTS?

    Proper and extensive benchmarks of Fedora 17 against Ubuntu 12.04 LTS are forthcoming, but here's a preview of some results...

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

  • #2
    Fedora 17 was quick to adopt GCC 4.7 where as Ubuntu 12.04 LTS took the conservative approach of living with GCC 4.7
    I think you meant that Ubuntu stuck with GCC 4.6, no? Anyway..
    Thanks for the benchmarks. It's good to see a performance boost with gcc 4.7 and newer kernels, even for Ubuntu users.

    Comment


    • #3
      Originally posted by DanL View Post
      I think you meant that Ubuntu stuck with GCC 4.6, no?
      Ja, fixed thanks.
      Michael Larabel
      http://www.michaellarabel.com/

      Comment


      • #4
        Does Fedora 17 comes with a software center now? I'm sorta attached to Ubuntu or Mint for this.

        Comment


        • #5
          Sorry, I didn't quite get that. The compile bench is faster on Fedora because it's GCC 4.6 vs. 4.7 or because disk access is faster in Beefy Miracle?

          Comment


          • #6
            Originally posted by Dukenukemx View Post
            I'm sorta attached to Ubuntu or Mint for this.
            The Ubuntu Software Center is being ported to PackageKit as GSOC project so even if there isn't one yet there soon will be.

            Comment


            • #7
              Michael,

              Please consider using Fedora (Or any other Linux distribution that comes in first) the next time you do a Windows / Solaris / BSD vs. Linux benchmark.
              Using the slow(er/est) Linux as yardstick is plain wrong.

              - Gilboa
              DEV: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB + 2x3TB, GTX780, F21/x86_64, Dell U2711.
              SRV: Intel S5520SC, 2xX5680, 36GB, 4x2TB, GTX550, F21/x86_64, Dell U2412..
              BACK: Tyan Tempest i5400XT, 2xE5335, 8GB, 3x1.5TB, 9800GTX, F21/x86-64.
              LAP: ASUS N56VJ, i7-3630QM, 16GB, 1TB, 635M, F21/x86_64.

              Comment


              • #8
                Originally posted by gilboa View Post
                Using the slow(er/est) Linux as yardstick is plain wrong.
                It's not plain wrong when Ubuntu Linux is the most widely used desktop Linux distribution out there and there's many derivatives of it... Sure with tweaking and custom package compiling, I'm sure Arch or LinuxFromScratch could be made the fastest, but then not many people would be able to easily and reliably achieve those same results, etc and it would just represent a small minority of individuals.

                Including both Ubuntu and xxxxxxx though in a future comparison against others? Perhaps.
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Originally posted by Michael View Post
                  It's not plain wrong when Ubuntu Linux is the most widely used desktop Linux distribution out there and there's many derivatives of it... Sure with tweaking and custom package compiling, I'm sure Arch or LinuxFromScratch could be made the fastest, but then not many people would be able to easily and reliably achieve those same results, etc and it would just represent a small minority of individuals.

                  Including both Ubuntu and xxxxxxx though in a future comparison against others? Perhaps.
                  Here's the thing, given the lack of credible information, the only two means to somehow measure market share is Google trends and Distrowatch, and both are severely broken.
                  The only *credible* market share information (in this case, information backed by cold hard cash) comes from either RedHat or SUSE, and neither is being used as the "Linux" yardstick.

                  Beyond that, Mint, OpenSUSE and Fedora (just to name a few) are just as easy to use as Ubunutu and both are used by millions and are just as worthy to be the Linux yardstick as Ubunutu. (Again, given the lack of credible usage numbers).

                  Having said all that, It's OK to say "I personally prefer Ubunutu, hence I use it", however, if this is indeed the case, I believe this should be stated openly.

                  - Gilboa
                  DEV: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB + 2x3TB, GTX780, F21/x86_64, Dell U2711.
                  SRV: Intel S5520SC, 2xX5680, 36GB, 4x2TB, GTX550, F21/x86_64, Dell U2412..
                  BACK: Tyan Tempest i5400XT, 2xE5335, 8GB, 3x1.5TB, 9800GTX, F21/x86-64.
                  LAP: ASUS N56VJ, i7-3630QM, 16GB, 1TB, 635M, F21/x86_64.

                  Comment


                  • #10
                    Originally posted by gilboa View Post
                    Having said all that, It's OK to say "I personally prefer Ubunutu, hence I use it", however, if this is indeed the case, I believe this should be stated openly.

                    That's not at all the case, I'm not an Ubuntu fan boy. I also have Fedora, SUSE, MacOSX, Solaris, etc systems around for PTS and other purposes. My main system is even still running Ubuntu 10.10 since I haven't been too fond of recent Ubuntu releases.

                    Edit: And perhaps the stat about usage I care the most about is the OS stats from OpenBenchmarking.org: http://openbenchmarking.org/s/OS since most PTS users are Phoronix readers. So yes, most of the people likely reading the content are using Ubuntu.
                    Michael Larabel
                    http://www.michaellarabel.com/

                    Comment


                    • #11
                      Originally posted by gilboa View Post
                      Again, given the lack of credible usage numbers
                      I think we have every reason to believe that Ubuntu is the most popular destop Linux out there:
                      -It's searched the most in Google.
                      -It's talked about the most in Twitter, blogs and news sites.
                      -Ubuntu.org is by far the most visited Linux distribution site (looks like it has a lot more visits than all other combined) according to Alexa.
                      -It has the biggest forums (8 times more members than both openSUSE and Fedora forums combined).
                      -It's has actual hardware parters and computers with Ubuntu preinstalled can be bough from Dell and some Chinese vendors.
                      -Canonical reports ~20 000 000 Ubuntu installs.
                      Last edited by Teho; 06-09-2012, 08:33 AM.

                      Comment


                      • #12
                        Originally posted by gilboa View Post
                        Here's the thing, given the lack of credible information, the only two means to somehow measure market share is Google trends and Distrowatch, and both are severely broken.
                        Maybe the best way to get a somewhat credible number would be to get the number of unique machines accessing each distro's repository, like what google does with Android. The problem is that it would require that all (or a very large percentage) of the available mirrors provide this info and tracking mechanism.

                        Originally posted by gilboa View Post
                        Beyond that, Mint, OpenSUSE and Fedora (just to name a few) are just as easy to use as Ubunutu and both are used by millions and are just as worthy to be the Linux yardstick as Ubunutu. (Again, given the lack of credible usage numbers).
                        I can imagine that Ubuntu is more used mainly because it has much more related sites and blogs than what we can find for other distros. This is a good indication that it's popular, although by how much is hard to tell.
                        On the "easy to use" argument I couldn't agree more. Before software center I could never understood why people said that Ubuntu was "easy to use" when AFAICT Fedora was exactly the same but in a blue hue. Nowadays Software Center does make it easier to discover and install applications, and Unity is different from what you get in other distros, but other than that there's not much difference. I use Ubuntu now, after a few years with Fedora and openSUSE, mainly because of Unity (yes, that's right: Unity) and software center.

                        Comment


                        • #13
                          Fedora 17 is an all around great release, I just switched from 12.04 to fedora the other day. Much snappier, and boots faster too.

                          Comment


                          • #14
                            These seem like extreme results... does a slightly newer kernel and a newer gcc really explain this? Is unity hogging resources? I've notice on my laptop Ubuntu feels much snappier under gnome 3 than unity.

                            Comment


                            • #15
                              I'm using Fedora 17 as well (KDE spin, of course), and it's snappier than most other distributions. I would like to see them compared to OpenSUSE (and if possible, Debian unstable) though, just to get an idea how the "Big 3" (or "Big 4") perform against each other. I'm sure Arch will probably win the race anytime, but like someone said before, it's not an easy distro to set up and since it's bleeding edge, it's prone to some packages misbehaving/crashing altogether. i remember KDE crashing quite a few times when I used Arch. Don't know if it got better since then.

                              Comment

                              Working...
                              X