Announcement

Collapse
No announcement yet.

AMD Ryzen 9 9950X & Ryzen 9 9900X Deliver Excellent Linux Performance

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

  • #11
    As expected, 9950X and 9900X have extremely severe memory bottlenecks, with many AVX512 workloads performing at the same level as even 9700X.

    Comment


    • #12
      Why every CPU company always has to make the worst decisions when it comes to what to focus on CPU perf?

      Comment


      • #13
        What's with the hate? This CPU has a pretty decent performance uplift in most tests (often beating Intel where they used to be #1) while consuming less power than the 7950X. That sounds like a solid generational uplift to me.
        I wouldn't be recommending a 7950X[3D] user to upgrade to this but I would see this as a worthwhile upgrade from DDR4.

        Comment


        • #14
          When taking the geometric mean of all the raw performance results, the Ryzen 9 9950X came out to being 17.8% faster than the Ryzen 9 7950X. The Ryzen 9 9900X meanwhile was 21.5% faster than the Ryzen 9 7950X across this wide mix of workloads.
          I think this is a typo? If the 9950x is 17.8% faster then the 7950x, then the 9900x cannot be 21.5% faster then the same 7950x. I guess it should be the 7900x instead of the 7950x in the last comparison?
          Last edited by peterdk; 14 August 2024, 10:33 AM.

          Comment


          • #15
            Looking really good to me, specially for production/compiling workloads.

            With some memory tuning, like 6400 1:1 CL28 (which should be possible with Hynix A Dies), then the memory bottleneck shouldnt kick too much anymore.
            AVX512 Performance is freaking awesome.
            Tomorrow, I will order one. Did some benchmarks on my CachyOS with my 7950X3D and will see how much it benefits.

            Before someone complains:
            This is already my second 7950X3D, since the other one randomly died. The current 7950X3D shows the same behaviour, as my recently broken 7950X3D.
            On our buildserver, hosted by Hetzner (no overclocking, expo or equal) also already died 2 7950X3D. I really can not trust this CPU anymore - might im unlucky

            Comment


            • #16
              Originally posted by peterdk View Post

              I think this is a typo? If the 9950x is 17.8% faster then the 7950x, then the 9900x cannot be 21.5% faster then the same 7950x. I guess it should be the 7900x instead of the 7950x in the last comparison?
              Yes meant 7900X, thanks. Fixed.
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #17
                Originally posted by duby229 View Post

                How can you call it botched? They reused the IO die from the last generation. It was a tock release not a tick release. Intel.spent nearly 15 years releasing every other product generation in this same fashion.

                EDIT: The thing that happened why you consider it botched is that they reused the last gens IO die... So does that mean you also think the last gen is botched too??? And do you consider every Intel release for a nearly 15 year period or so also botched???




                Nowhere to be seen.

                Comment


                • #18
                  Originally posted by avis View Post





                  Nowhere to be seen.
                  Those aren't comparing AMDs last gen, those are comparing Intel's current gen. As far as I can tell that's a pretty good representation.

                  Comment


                  • #19
                    For a proud owner of 5950X the new 9950X looks really impressive. In some charts it is faster up to 3 times. As a reasonable consumer I skipped 7xxx lineup as being the first in the lineup of new socket, new memory etc to avoid problems so it's performance doesn't matter to me.

                    Comment


                    • #20
                      Why does the 7-zip from HW unboxed show a regression and we have a ~20% gain here ?

                      Comment

                      Working...
                      X