Announcement

Collapse
No announcement yet.

AMD Athlon 5350 APU On Linux

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

  • #11
    I am a bit suprised, even on the benchamrks it scaled good it was mostly only double as fast.

    But on the kernel-compile time it was only 10% faster. Even if you did not setup it properly that it uses 5 threads for kernel compiling (core-amount +1) it should be more than 10% faster.

    Do you use a harddisk and does it bottleneck it here hard? I think in generall the cpu should be 4x faster than a E350, 4 core instead of 2, higher clockspeed + better architecture.

    If I would just eat that benchmark, this cpu would not be much faster than a E-350, what would be horrific.

    Comment


    • #12
      Originally posted by dungeon View Post
      And E-350 with 2x1.6GHz bobcat cores for 16 minutes... something went wrong there is it Michael .
      Well, whatever the system is doing, it's completely reproducible... Each kernel build happened three times, all using standard config, etc. You can see the SE on the bars that in fact each time it was done it was doing the same.
      Michael Larabel
      https://www.michaellarabel.com/

      Comment


      • #13
        So many low cost options now it's hard to decide. I was looking at the Minnowboard Max but now I don't know. I'm guessing if your program can leverage OpenCL then it would perform better on the AMD APU than an Intel chip.

        Comment


        • #14
          Originally posted by Michael View Post
          Well, whatever the system is doing, it's completely reproducible... Each kernel build happened three times, all using standard config, etc. You can see the SE on the bars that in fact each time it was done it was doing the same.
          OK but those are unexpected results , it must be 3 times faster then e-350... i compare right now with result form anandtech for example x264. Here it is only 50% faster. but it is expected to be 300% faster .

          Comment


          • #15
            Originally posted by shaurz View Post
            Why would anyone buy this junk? Crap performance and it needs a big heatsink and fan.
            For the price point, you can't really complain. I feel like this would make a decent library computer or even a HTPC. I'm sure the heatsink they used is overkill, but they used it because it's likely cheaper than designing and manufacturing a new model. At least the fan should always run at its slowest speed, even when the APU is under full load. That being said, I'm sure this will be nearly silent.

            Comment


            • #16
              Benchmarks on the three other Sempron/Athlons under Linux will come next week.
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #17
                Originally posted by Michael View Post
                As said in the article, it's coming in a few days... Only got the 5350 a few days ago and had limited amount of time to run initial tests for this morning's launch.
                I'm sorry, I only saw the mention about upcoming Catalyst tests, my bad!

                Comment


                • #18
                  Originally posted by Michael View Post
                  Problem with that is lack of free, well-mirrored content in all the different bitrates, etc. If there is, feel free to add it to PTS and it will then be tested.
                  Pretty much anything that has come out of the Blender Project?

                  Comment


                  • #19
                    I would pull that article, or at least update it to make VERY clear mention that something in the tests went horribly wrong. As others have shown, those CPU results are nowhere near what they should be.

                    Comment


                    • #20
                      Originally posted by LinuxID10T View Post
                      Pretty much anything that has come out of the Blender Project?
                      http://samplemedia.linaro.org/ ?

                      Comment

                      Working...
                      X