Announcement

Collapse
No announcement yet.

AMD FX-8350 Linux Performance-Per-Watt

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

  • AMD FX-8350 Linux Performance-Per-Watt

    Phoronix: AMD FX-8350 Linux Performance-Per-Watt

    The latest Phoronix benchmarks to share of the AMD FX-8350 "Vishera" processor are performance-per-Watt results for the Piledriver eight-core processor compared to the previous-generation Bulldozer FX-8150. Tests were conducted when running at stock speeds as well as overclocked settings.

    http://www.phoronix.com/vr.php?view=18074

  • #2
    Ok, it's nice to see that the 8350 is a improvement over the 8150, but can we please get the same Bench in comparison with the Core i7 3770K, please? This would add real value to the article.

    Comment


    • #3
      Originally posted by Lynxeye View Post
      Ok, it's nice to see that the 8350 is a improvement over the 8150, but can we please get the same Bench in comparison with the Core i7 3770K, please? This would add real value to the article.
      Yup, the onlt thing you can tell from this article is that the new architecture is a slight improvement over the old one. It would've benn rather puzzling if it weren't. But including the competition makes the whole thing pretty useless.

      Comment


      • #4
        I agree. And performance/watt/price statistics for 5 years would be nice too.

        Comment


        • #5
          Wow, 8 pages "Vishera" advertisement!

          Comment


          • #6
            FX-8350@4.6 compiles Linux kernel in 72.98 seconds, consumes 207 Watts on average.
            FX-8150@4.6 compiles Linux kernel in 79.25 seconds, consumes 235 Watts on average.

            How is it possible for FX-8150@4.6 to have better performance-per-watt? Results in performance-per-watt graph seems to be wrong...

            Comment


            • #7
              Originally posted by abacaba View Post
              FX-8350@4.6 compiles Linux kernel in 72.98 seconds, consumes 207 Watts on average.
              FX-8150@4.6 compiles Linux kernel in 79.25 seconds, consumes 235 Watts on average.

              How is it possible for FX-8150@4.6 to have better performance-per-watt? Results in performance-per-watt graph seems to be wrong...
              That also seemed strange to me. I think there's something wrong with the formula used to come up with those numbers. It's actually calculating how many watts are needed for each second, which obviously reverses the results. That formula works when more is better, but not when less is better. That's why all the graphs involving less is better are wrong.

              Comment


              • #8
                Well it depends how much overvoltage is needed to get 4.6 ghz. If you need less vcore for the older cpu because it oc better then you see this. oc values are never compareable as those are single part specific.

                Comment


                • #9
                  Originally posted by Kano View Post
                  Well it depends how much overvoltage is needed to get 4.6 ghz. If you need less vcore for the older cpu because it oc better then you see this. oc values are never compareable as those are single part specific.
                  It doesn't matter what voltage is set to run a core, what current does it takes etc. Even clock rate is not important. Only performance (here measured in seconds) and power consumption (in watts) do count. We can multiply time by power and get the number of joules required to do the job (compile kernel) with given setup, and that number will be better for 8350@4.6

                  Comment


                  • #10
                    Very cool testing!

                    I agree to spice the test with some good Intel rival... However anyone can do this by running the test on own system and submitting. So please, if anyone has Ivy and time do this

                    --
                    Very cool CPU indeed! The 20% AMD has talked about do appear in tests!

                    Also, its scales like godlike so it could be good idea to overclock the CPU for serious tasks and to underclock it for daily tasks. This way the CPU will be as energy efficient as Intel for daily work and can be turned into a beast for serious load. Ofc, one needs to pull 24hr stability test on it, but the possibility to use ECC makes this feature even more tempting.. ))

                    Btw, 85 Watts idle and 150Watts load for vicera - compare to my Athlon II x4 630 - 90W idle and 140W full load when in stock - 80W idle and 120W when undervolted to 1.2v (runs at stock Hz, rocksolid).
                    This is great!

                    Comment


                    • #11
                      Very curious results, having the older cpu beat the newer one in perf/W in a number of scenarios. And also having lower idle and max power use!

                      That's a different result from Windows reviews, which is why it's good we have Phoronix around.

                      Comment


                      • #12
                        All those recent tests on FX-8350 had mtune=generic (at least that's what has been written under the config table). So they don't show the real potential (if any). No AVX, no AVX2, no AES, ...
                        All those gathered numbers don't have any value...

                        Comment


                        • #13
                          Average watt usage of fx 8150 @ 4,6ghz is 297,8
                          time to compile is 79,25
                          297,8 / 79,25 = 3,76 - which is correct

                          Average watt usage of the fx 8350 @ 4,6ghz is 264,3
                          time to compile is 72,98
                          264,3 / 72,98 = 3,62

                          -> the calculation is correct but I dont see any sense in it.
                          shoulnd't it be more like:

                          fx 8150@4,6ghz: 79,25 / 297,8 = 0,266
                          fx 8350@4,6ghz: 72,98 / 264,3 = 0,276

                          => the fx 8350@4,6ghz gives 3,8% more performance per watt compared to the fx 8150@4,6ghz


                          also I'd calculate
                          297,8 * 79,25 = 23601 kJ for the fx 8150 @ 4,6ghz
                          264,3 * 72,98 = 19288 kJ for the fx 8350 @ 4,6 ghz

                          19288 / 23601 = 0,817 .. means that the fx 8350 almost needs 18% less power than the fx 8150 @ 4,6ghz - for compiling the kernel - which is definitely a better job.

                          correct me if I am wrong of course



                          mcgreg
                          Last edited by mcgreg; 10-26-2012, 10:12 AM.

                          Comment


                          • #14
                            -> the calculation is correct but I dont see any sense in it.
                            shoulnd't it be more like:

                            fx 8150@4,6ghz: 79,25 / 297,8 = 0,266
                            fx 8350@4,6ghz: 72,98 / 264,3 = 0,276

                            => the fx 8350@4,6ghz gives 3,8% more performance per watt compared to the fx 8150@4,6ghz
                            I think you got that wrong - that's seconds per watt, and the 8350 is taking _longer_ per watt. This would mean it's worse, not better?

                            Comment


                            • #15
                              Originally posted by curaga View Post
                              I think you got that wrong - that's seconds per watt, and the 8350 is taking _longer_ per watt. This would mean it's worse, not better?
                              No, it means that if the 8350 would only get 1 watt it can "compute" for 0,276s and the 8150 would only "compute" for 0,266s. So more would be better. But this number alone says nothing because longer "computing" does not generally mean it does more things.

                              /edit:
                              The only meaningful number is indeed this:
                              Originally posted by mcgreg View Post
                              297,8 * 79,25 = 23601 kJ for the fx 8150 @ 4,6ghz
                              264,3 * 72,98 = 19288 kJ for the fx 8350 @ 4,6 ghz
                              Last edited by droste; 10-26-2012, 11:12 AM.

                              Comment

                              Working...
                              X