Announcement

Collapse
No announcement yet.

100+ Linux Benchmarks Between The AMD Ryzen 7 4700U vs. Intel Core i7 1065G7

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

  • #21
    Originally posted by mppix View Post

    I cannot find this comparison article.. Can you link it? Thx
    Originally posted by JustinTurdeau View Post

    Or maybe these results are normal and the review you're talking about is just another case of Benchmarks Rigged^W Sponsored by Intel™...?

    https://www.anandtech.com/show/15762...ith-ryzen-4000

    Comment


    • #22
      Here is an interesting bench by lab501. They put a 4800U and 10710U in the same chassis head to head at different TDP caps. The 4800U doesn't seem to run that much faster at 25 W TDP than at 15 W. It also pretty nothing but nukes the 10710U.

      https://next.lab501.ro/notebook/engl...core-i7-10710u

      Comment


      • #23
        Originally posted by JustinTurdeau View Post

        Or maybe these results are normal and the review you're talking about is just another case of Benchmarks Rigged^W Sponsored by Intel™...?
        If they would have had spend their bribe budget on node development they mostlikely would have 5nm++ already....but this happens if you give too much power to bs guys in the marketing team.

        Comment


        • #24
          Originally posted by JustinTurdeau View Post

          Or maybe these results are normal and the review you're talking about is just another case of Benchmarks Rigged^W Sponsored by Intel™...?
          If it's not in AMD favour it's rigged. Yeah, right.

          Comment


          • #25
            Originally posted by birdie View Post

            If it's not in AMD favour it's rigged. Yeah, right.
            No of course it is not. Fair comparsion but at the moment AMD Ryzen are far superior in almost all aspects. The only point at the moment where Intel can win is single core + high clock freq and this is almost not a real world scenario anymore. Each Review focusing on that... lets say can be considered as biased.

            Comment


            • #26
              Hmm, by default on my dell xps 7390 2in1 all cores working at 2500mhz with 15 watt set to /sys/class/powercap/intel-rapl-mmio\:0/constraint_0_power_limit_uw
              and when i set 35watt all cores running on 3500mhz as in specs.

              This is my highest result on geekbecnh 5 with this settings https://browser.geekbench.com/v5/cpu/2146033
              1455 for Single-Core Score and 5205 for Multi-Core Score and there are results for 4700U https://browser.geekbench.com/v5/cpu...page=1&q=4700U
              with highest score multi-core score 6361 and single core at 1228


              Comment


              • #27
                Originally posted by CochainComplex View Post

                No of course it is not. Fair comparsion but at the moment AMD Ryzen are far superior in almost all aspects. The only point at the moment where Intel can win is single core + high clock freq and this is almost not a real world scenario anymore. Each Review focusing on that... lets say can be considered as biased.
                An exaggeration on top of an exaggeration on top of an exaggeration. AMD sect never fails to disappoint. I mean Intel still has undisputably the fastest single core CPUs using an arch from 2015 (Sky Lake) and AMD somehow "wins in almost all aspects". Bonkers!

                95% of average people out there are more interested in single-core performance unlike what AMD fans have been touting since the multicore failure called AMD FX.

                Originally posted by CochainComplex View Post

                No of course it is not. Fair comparsion but at the moment AMD Ryzen are far superior in almost all aspects. The only point at the moment where Intel can win is single core + high clock freq and this is almost not a real world scenario anymore. Each Review focusing on that... lets say can be considered as biased.
                What???? Absolute most tasks are limited by single-threaded performance even the ones which scale perfectly because at the end of the day you need to synchronize threads or wait for the last of them to complete.

                And this is what matters for absolute most people out there. Even among Phoronix readers barely anyone compiles tons of code daily or (re)encodes videos.

                Why is it always multithreaded performance BS?

                Biased are you, sir. Not only biased you don't understand the basic principles of desktop computing where single threaded performance is the end all.
                Last edited by birdie; 17 May 2020, 07:54 AM.

                Comment


                • #28
                  Originally posted by birdie View Post
                  An exaggeration on top of an exaggeration on top of an exaggeration. AMD sect never fails to disappoint.
                  I have no particularly loyalty to any CPU vendor, but I always prefer a company that competes and innovates vs. one that constantly gets caught rigging benchmarks and has a long history of antitrust violations.

                  Originally posted by birdie View Post
                  Why is it always multithreaded performance BS?
                  Because it's by far the most relevant metric for workloads where performance actually matters.

                  Originally posted by birdie View Post
                  you don't understand the basic principles of desktop computing where single threaded performance is the end all.
                  Who said anything about "desktop computing"? Apparently you care about that. Good for you. Everyone else who cares about CPU performance couldn't care less about some GUI app that spends 99% of it's time blocked in main_loop().
                  Last edited by JustinTurdeau; 17 May 2020, 11:36 AM.

                  Comment


                  • #29
                    For what it's worth, I ran these benchmarks on my laptop (same model) on Fedora 32 w/updates.

                    The results are here: https://openbenchmarking.org/result/...NI-2005169NI43

                    Comment


                    • #30
                      Originally posted by rogowskys View Post
                      For what it's worth, I ran these benchmarks on my laptop (same model) on Fedora 32 w/updates.

                      The results are here: https://openbenchmarking.org/result/...NI-2005169NI43
                      Looks like they are in line with my results, great! Has Renoir been working out well for you on Fedora 32? Haven't had a chance to test there yet.
                      Michael Larabel
                      http://www.michaellarabel.com/

                      Comment

                      Working...
                      X