Announcement

Collapse
No announcement yet.

Chromium OS, Moblin, Ubuntu Netbook Remix Benchmarks

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

  • #16
    The correct why would be:

    a) Check in the initrd for Intel VGA, that's definitely trivial. If not there do:

    chmod 4755 /usr/bin/Xorg

    then X will start.

    b) Use a bigger virtual filesystem, why not 3 gb? currently it is fixed to 2 gb with 131 mb free, even with more ram available it is impossible to use more.

    Comment


    • #17
      Originally posted by Kano View Post
      @sofar

      Also i would like to know how to stop Moblin from tring to start X as user. That only works with Intel onboard.
      ATI is just about there as well with KMS etc...

      are you really wanting to run MORE stuff as setuid root and consider it a good idea??

      Comment


      • #18
        As long as vesa requires it - or use just a fallback that works. You can do that on the fly in the initrd for live mode. It is definitely a very bad thing when you use vbox and see just a black screen. What about increasing the filesystem for live mode?

        Comment


        • #19
          Originally posted by Kano View Post
          The correct why would be:

          a) Check in the initrd .

          what initrd??

          Comment


          • #20
            Originally posted by Jimbo View Post
            I think the review is totally unfair!!!!

            All OS accept ext4 or ext3 FS, so, why not to run all test on the same filesystem, this way results are not filesystem dependant.

            Yeah, i know that on some distributions you can't choose ext4 on installation process. But i think this is a feature not implemented yet, with a little work you can convert ext3 to ext4 and run the test.
            The tests represent an "out of the box" experience.
            Michael Larabel
            http://www.michaellarabel.com/

            Comment


            • #21
              @arjan_intel

              I only test the img file and NO hd install! And of course that uses an initrd to boot! Would not hurt to call it iso, as it uses just isolinux hybrid mode.

              Comment


              • #22
                Originally posted by Michael View Post
                The tests represent an "out of the box" experience.
                Ok, but doing so, the benchmarks are totally filesystem dependant. So the title must say ext3 and ext4 benchmarks , instead of Chromium OS, Moblin, Ubuntu Netbook Remix Benchmarks.

                It has no sense to me comparing different distros under different FS, it is not representative.

                Comment


                • #23
                  Originally posted by Jimbo View Post
                  It has no sense to me comparing different distros under different FS, it is not representative.
                  It's representative of what the average user will see if they install the OS on their netbook using the default settings: which is what 90+% of people will do.

                  Personally I was reluctant to use the default ext4 on my UNR install, but so far it hasn't truncated all my files to zero bytes.

                  Comment


                  • #24
                    Sorry i donít want to be unpolite, but i totally disagree.

                    For example, if u want to compare 2 sportive cars, you have to make the effort to transport them to the same circuit and to put them under the same professional driver. Then do the benchmark . You have to make a previous effort to put them under the same conditions.

                    In engineering if u want to compare one implementations against other/others and write and article u have to create the same conditions, and do the testing.

                    Phoronix is a referent to linux, lot of journalists take you as a referent, they can write that X distro is better than Y distro without mentioning that the benchmark has been realised under different filesystem, this way, a lot of disinformation is created.

                    All kernels > 2.6.28 support ext4 FS, it is easy to setup a ext4 root partition.

                    Comment


                    • #25
                      Originally posted by Jimbo View Post
                      Sorry i donít want to be unpolite, but i totally disagree.

                      For example, if u want to compare 2 sportive cars, you have to make the effort to transport them to the same circuit and to put them under the same professional driver. Then do the benchmark . You have to make a previous effort to put them under the same conditions.

                      In engineering if u want to compare one implementations against other/others and write and article u have to create the same conditions, and do the testing.

                      Phoronix is a referent to linux, lot of journalists take you as a referent, they can write that X distro is better than Y distro without mentioning that the benchmark has been realised under different filesystem, this way, a lot of disinformation is created.

                      All kernels > 2.6.28 support ext4 FS, it is easy to setup a ext4 root partition.
                      Maintaining the same track and driver would be like maintaining the same set of benchmarks and settings. However, changing out the file-system would be like changing out the tires on the vehicle. The default file-system is part of the decision that distribution vendors make as part of their process, just as automobile makers recommend a specific set of tires. As a reader pointed out, a far majority of Linux users will go with the default file-system choice.
                      Michael Larabel
                      http://www.michaellarabel.com/

                      Comment


                      • #26
                        Why screenshots of only 3 out of 5 tested distros? Certainly would have liked to see Chromium's shot.

                        Comment

                        Working...
                        X