• Electricdoggo@lemmy.world
    link
    fedilink
    English
    arrow-up
    61
    arrow-down
    1
    ·
    8 months ago

    So a 80W 12 Performance core CPU could only manage +10% single core and 2x multi-core over a 20W 4+4, year old CPU? That doesn’t sound like a dunk. Performance per Watt looks to be worse than the M2 as well.

    • SpeziSuchtel@feddit.de
      link
      fedilink
      arrow-up
      41
      ·
      8 months ago

      „Best Snapdragon CPU beats Apples basic entry level Mac processor“ doesn‘t generate enough clicks and outrage to divide Apple haters and Apple fanboys further.

    • xep@kbin.social
      link
      fedilink
      arrow-up
      9
      ·
      edit-2
      8 months ago

      This is mentioned in the article:

      The critical thing to remember during all these benchmarks is that Qualcomm matches or beats the competition (as of today) at all these CPU and GPU tests, but at less power than the others, sometimes up to 70% less power than Apple or Intel.

      Even against the M2 Max from Apple, which will beat the Snapdragon X Elite on most benchmarks (except single-thread), the Snapdragon X Elite still consumes 30% less power when matching Apple’s single-threaded peak performance.

      Looks like a 30% efficiency improvement, although the article doesn’t detail the performance against M2 besides in writing. We’ll have to wait for more benchmarks.

      On the more familiar and widely used Geekbench 6, both configurations easily beat Razer’s Blade 14 (2023) powered by the AMD R9 7940HS. The MacBook Pro 13” with M2 processor came last (compared to our best gaming laptops) with 2,658 single-thread and 10,088 multi-thread. By comparison, Qualcomm pulled off 2,940 ST, 15,130 MT, 2,780 ST, and 14,000 MT at its lower TDP configuration.

      Cinebench 2024, which replaces Cinebench R23, hasn’t been used a lot by us yet as it’s brand new, but the new version, which is compiled to run ARM natively, still shows the Snapdragon X Elite way ahead of the competition with 132 ST and 1,220 MT for Config A. The MacBook Pro with M2 could only muster 121 ST and 572 MT and was still easily beaten by the Config B model with 122 ST and 950 MT.

      • fuckwit_mcbumcrumble@lemmy.world
        cake
        link
        fedilink
        arrow-up
        2
        ·
        8 months ago

        The article didn’t even make it a day before Apple announced the M3 CPUs which will probably dunk on this in both performance and efficiency.

        The M2 CPU was kind of a disappointment since it was mostly just an overclocked M1 with a few more GPU cores, but the same bad architecture. The M3 should actually be a newer architecture and fix the M2s PPW regression.

        • xep@kbin.social
          link
          fedilink
          arrow-up
          1
          ·
          edit-2
          8 months ago

          Apple says the M3 is 50% more efficient than the M1, and since it’s on a 3nm process it’s likely at least competitive, and probably more efficient than the Snapdragon at 4nm.

    • morrowindOP
      link
      fedilink
      arrow-up
      5
      arrow-down
      2
      ·
      8 months ago

      Apple just announced m3, we’ll see how it compares

  • AutoTL;DR@lemmings.worldB
    link
    fedilink
    English
    arrow-up
    4
    ·
    8 months ago

    This is the best summary I could come up with:


    Qualcomm caused quite a stir last week with its long-awaited announcement of its Snapdragon X Elite platform based on its new Oryon CPU, creating what some are calling the “Apple Mac Moment” for Windows.

    During Qualcomm’s keynote, the company went on stage with some fancy graphs and a few handpicked benchmarks, putting it up against Intel’s best 13th-generation Core laptop CPUs and Apple’s M2 (and even M2 Max in one scenario).

    More importantly, when we turned around, there were well over 20 Oryon-powered laptops with Geekbench 6, Cinebench 24, PCMark 10, Procyon AI, and 3Dmark WildLife Extreme and Aztec Ruins (pre-commercial builds).

    But, similar to Apple, that platform can range from low TDP (thermal design power; basically, how much wattage the chip draws) to very high, with or without fans.

    Each time you run a benchmark, the score fluctuates depending on external and internal thermal conditions or any Windows background processes that may temporarily be active.

    It is worth noting that by the time Qualcomm’s Snapdragon X Elite hits store shelves, Apple’s M3 line of CPUs (which are expected to be announced this week) and Intel’s next-gen Meteor Lake laptops processors with its beefy NPU and GPU, will be the new competition.


    The original article contains 1,296 words, the summary contains 202 words. Saved 84%. I’m a bot and I’m open source!

  • poddo@discuss.tchncs.de
    link
    fedilink
    arrow-up
    3
    ·
    8 months ago

    So when they drop uefi are we even going to be able to get windows off of these things? If not it sounds like a living hell.

    • irdc@derp.foo
      cake
      link
      fedilink
      arrow-up
      2
      ·
      8 months ago

      If anything you need UEFI to run Windows on one of these things.

    • Alex
      link
      fedilink
      arrow-up
      1
      ·
      8 months ago

      Why do you need to drop uefi?

      • poddo@discuss.tchncs.de
        link
        fedilink
        arrow-up
        1
        ·
        8 months ago

        True they could just keep uefi and drop support for non-certified software its their ecosystem after all