Announcement

Collapse
No announcement yet.

Quad-Core ODROID-X Battles NVIDIA Tegra 3

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

  • Quad-Core ODROID-X Battles NVIDIA Tegra 3

    Phoronix: Quad-Core ODROID-X Battles NVIDIA Tegra 3

    While not as popular as NVIDIA's Tegra 3 ARM SoC, the Samsung Exynos 4412 quad-core ARM Cortex-A9 found on cheaply priced ODROID-X can actually outperform the quad-core NVIDIA ARM processor. Here are benchmarks of the $129 USD ODROID-X benchmarked against the NVIDIA Tegra 3 reference tablet and a PandaBoard ES running the Texas Instruments OMAP4460.

    http://www.phoronix.com/vr.php?view=17793

  • #2
    Hey Michael, next time you bench these boards can you please test the USB speed by doing a simple file copy test?

    Comment


    • #3
      Originally posted by deanjo View Post
      Hey Michael, next time you bench these boards can you please test the USB speed by doing a simple file copy test?
      What use that would be, unless its to check to see if the USB ports are USB2 compatible.

      Nice board, so would love to see this as a nettop or even a micro-desktop box. With the performance this board exhibits it should do well for most audio/video stuff

      Comment


      • #4
        Originally posted by DeepDayze View Post
        What use that would be, unless its to check to see if the USB ports are USB2 compatible.

        Nice board, so would love to see this as a nettop or even a micro-desktop box. With the performance this board exhibits it should do well for most audio/video stuff
        maybe OUYA might consider this?

        anyway i would love a future mk802 to be based on this.

        Comment


        • #5
          Originally posted by DeepDayze View Post
          What use that would be, unless its to check to see if the USB ports are USB2 compatible.

          Nice board, so would love to see this as a nettop or even a micro-desktop box. With the performance this board exhibits it should do well for most audio/video stuff
          Why? Because many ARM boards have horrible USB throughput and well below their USB rated spec throughput and varies a lot from solution to solution.

          Comment


          • #6
            Ethernet testing would be good too, because on many of these the ethernet is actually via USB, which causes huge cpu usage and also limits it to below even 100 Mbps. The Pi has it via USB, for example.

            Comment


            • #7
              Nice tests!

              When will we see an Atom/Medfield vs Cortex-A9 battle again? I'm tired reading people claims that Medfield is faster even though I know it isn't

              Comment


              • #8
                Well, Atom often *is* faster than Cortex-A9, mostly because of high clock speed and good single-threaded performance. Cortex-A15 changes the game, though.

                Comment


                • #9
                  Originally posted by brent View Post
                  Well, Atom often *is* faster than Cortex-A9, mostly because of high clock speed and good single-threaded performance. Cortex-A15 changes the game, though.
                  No its not!!!

                  Comment


                  • #10
                    Originally posted by deanjo View Post
                    Hey Michael, next time you bench these boards can you please test the USB speed by doing a simple file copy test?
                    Same here, I can't seem to get a direct answer if each USB port is its own host or if every one is shared by a hub. This is a big deal with boards like these.

                    Comment

                    Working...
                    X