Announcement

Collapse
No announcement yet.

Mac OS X 10.6.2 vs. Ubuntu 9.10 Benchmarks

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

  • Mac OS X 10.6.2 vs. Ubuntu 9.10 Benchmarks

    Phoronix: Mac OS X 10.6.2 vs. Ubuntu 9.10 Benchmarks

    Back in August upon the launch of Apple's Snow Leopard we delivered benchmarks comparing Mac OS X 10.5 and Mac OS X 10.6 along with initial benchmarks of how Ubuntu 9.10 was running against Mac OS X 10.6. Since that time though Ubuntu 9.10 has been officially released with various changes since last August and Apple has issued two point releases for Snow Leopard, now putting it at version 10.6.2. As we await the release of FreeBSD 8.0 to deliver a larger operating system comparison, we have carried out a fresh round of tests comparing Mac OS X 10.6.2 and Ubuntu 9.10 (both x86 and x86_64 editions) under a variety of tests.

    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
    you forgot to run the disk IO benchmarks

    Comment


    • #3
      Nice!

      It looks like Mac OS/X's parallel processing agility is more than a bit of marketing if the ray tracing performance is to be believed. It would be very interesting to see how GCD compares with MPI when more software becomes available.

      The nice thing is you can't really go wrong with either OS.


      Dave

      Comment


      • #4
        You should add more multimedia-encoding benchmarks as well. I love to see Ubuntu beat this so called multimedia station in it's own home

        Comment


        • #5
          Mac OS X 10.6.2 though had a particularly strong advantage in the heavy computational tests like Crafty,...
          Wrong (about Crafty). If you'd compiled 32 bit craft on OS X, it would have been slow there. I think you're being fooled by the lack of a 32bit OS X benchmark. Crafty heavily depends on 64bit operations. Chess boards have 64 squares, and crafty uses 64 bit ints as "bit board" bitmaps. It probably has to AND, OR, and XOR 64 bit ints all over the place, so no wonder it's more than 3 times slower on a register-poor 32 bit architecture. I haven't looked up other (non PTS) crafty benchmarks, but there are lots published. I wouldn't be at all surprised if this 32 vs 64 slowdown is typical.

          The other wins for OS X are interesting, and some are either better filesystem performance, or performance regressions from gcc 4.2.1 to 4.4.1 (if those were the respective compiler versions). Some probably are actually better multi-thread handling by the OS, which is of course most interesting, since that's one thing you can't solve just by trying your code with different gcc versions to find which one makes the fastest binary.

          Which benchmarks are compiled from source, and which aren't? I expect most/all of the computational ones are, but what about e.g. Nexuiz? They distribute handy binaries...

          Comment


          • #6
            Originally posted by llama View Post
            Wrong (about Crafty). If you'd compiled 32 bit craft on OS X, it would have been slow there. I think you're being fooled by the lack of a 32bit OS X benchmark. Crafty heavily depends on 64bit operations. Chess boards have 64 squares, and crafty uses 64 bit ints as "bit board" bitmaps. It probably has to AND, OR, and XOR 64 bit ints all over the place, so no wonder it's more than 3 times slower on a register-poor 32 bit architecture. I haven't looked up other (non PTS) crafty benchmarks, but there are lots published. I wouldn't be at all surprised if this 32 vs 64 slowdown is typical.

            The other wins for OS X are interesting, and some are either better filesystem performance, or performance regressions from gcc 4.2.1 to 4.4.1 (if those were the respective compiler versions). Some probably are actually better multi-thread handling by the OS, which is of course most interesting, since that's one thing you can't solve just by trying your code with different gcc versions to find which one makes the fastest binary.

            Which benchmarks are compiled from source, and which aren't? I expect most/all of the computational ones are, but what about e.g. Nexuiz? They distribute handy binaries...
            I've had my MBP (unibody late 2008) tri booting with a 500gb hard drive. One of the main purposes was to see if i could gain a performance advantage while still maintaining software compatibility. The last time i had used ubuntu before this was jaunty, so i was looking forward to it.

            While i was able to install Wow, with wine, most modern games are not supported. I got RTCW running as well and i fail to see ubuntu spank OS X in frame rates.

            RTCW is Q3 engine, so i should see it but i don't, The Frame rates are generally within 10fps.

            I don't get how you guys alone have found this open GL bug in OS 10, I don't have it, can't prove it, and no one else has said anything for the months 10.6 has been around. (aka, i personally think your full of shit)

            Comment


            • #7
              Originally posted by Durandal View Post
              While i was able to install Wow, with wine, most modern games are not supported.
              On Ubuntu, you may want to try the latest beta of wine from
              https://launchpad.net/~ubuntu-wine/+archive/ppa

              Mainstream Ubuntu ships the 1.0 stable release to avoid regressions in dev versions, but the PPA has v1.1.33. (and the packages are called wine1.2)

              No comment on your other results; I don't have an OS X machine, and have never gamed on one.

              Comment

              Working...
              X