No announcement yet.

NVIDIA Tegra 3 Makes For Fast Ubuntu On ARM

  • Filter
  • Time
  • Show
Clear All
new posts

  • NVIDIA Tegra 3 Makes For Fast Ubuntu On ARM

    Phoronix: NVIDIA Tegra 3 Makes For Fast Ubuntu On ARM

    Here are the first set of benchmarks of the quad-core NVIDIA Tegra 3. Needless to say, four Cortex-A9s combined with NVIDIA graphics leads to a fairly fast ARMv7 experience when running Ubuntu Linux...

  • #2
    Phoronix & Android phones

    Michael, could you please tell your ad-company to get rid of the "igirls"-ad. I can no longer use your site with my android tablet. Every second or third visit this "igirl"-ad pops up in fullscreen and wants me to subscribe to their service for 3.99/week. The ad itself does not provide any way to close the popup and it seems to be some kind of CSS overlay, so I can't just click the back-button.

    This ad actually prevents users from visiting your site, as it covers the full screen.


    • #3
      AFAIK the 5th core is invisible to the OS and only kicks in when the other cores are sleeping.


      • #4
        Great to see benches of the of the nvidia offering. Sad situation on nvidia drivers, is there any chance of seeing gains from benching on 12.04?


        • #5
          Upgrade this to 12.04 and I think it will be competitive with real laptops in terms of performance.

          Is the Ubuntu that Nvidia use provided anywhere? With their customisations. I'd love to flash it on to a Transformer Prime. Maybe update it to 12.04.


          • #6
            Originally posted by Del_ View Post
            Great to see benches of the of the nvidia offering. Sad situation on nvidia drivers, is there any chance of seeing gains from benching on 12.04?
            Seeing as they're running a custom Linux 3.1 kernel already, and that their stock file-system is very small, the only possible gains I could envision out of 12.04 packages would be the updated compiler... but I'll be looking at that soon.
            Michael Larabel


            • #7
              So, can I actually buy a tablet and install Ubuntu on it without voodoo and other related hackery?
              To me it all still seems at the experimental stage, but I'm really looking forward to having as full OS on a tablet.


              • #8
                There are any nvidia tegra board that can I buy?



                • #9
                  Acer Iconia Tab A510 has been listed at some e-tailers but no stock yet.
                  Asus Eee Pad Transformer Prime is already on sale in several markets.


                  • #10
                    Ok but Acer and Asus is a final product.

                    I'm looking for a tegra 3 motherboard like Panda and Beagleboard.



                    • #11
                      There will be a development board called CARMA, launching in 2012Q2. Dunno about the price and whether they will sell it to everybody or you have to be special friends with NVidia.


                      • #12
                        Cachebench is only testing the compiler

                        Good to see tests of Tegra 3 and other ARM SoC's!

                        However, the current use of Cachebench in PTS is of little use. It is only a test of how bad the compiler optimize when compiling with -O. The result differs a lot if better optimization is used.

                        The result from the Cachebench read-test I get this on an Athlon 64 X2 computer using Debian testing gcc-4.6:

                        PTS result, using the default -O: 1308 MB/s
                        PTS result, using -Ofast: 5487 MB/s
                        PTS result, using -Ofast -fprefetch-loop-arrays: 8440 MB/s

                        Cachebench is very sensitive to compiler optimizations, and just using -O is more of a test of the included optimizations when using -O and when tuning for the default CPU for the given compiler. Different versions of gcc include different optimizations, and different distributions of Linux set different default tuning options.

                        From a hardware test point of view, you should either use the same binary, or you should find the best results.

                        Now we see results from different compilers with non-optimal optimization and tuning flags, and even worse,
                        you don't state the compiler used, including default tuning options.

                        It is quite obvious that 1308 MB/s is a useless result when better optimization give you 8440 MB/s on the same hardware.


                        • #13
                          Needs more analysis...

                          we ran our Panda against this benchmark... differences from the reference Panda was PTS 3.8, GCC 4.7, and kernel ver 1409...


                          so the Tegra only wins a couple of these comparisons... we probably need to widen the search space. It is interesting how much the Panda performance varied. We couldn't run Crafty because of a dependency failure (libnuma-dev)...