• gregorum
      link
      fedilink
      English
      arrow-up
      5
      ·
      edit-2
      7 months ago

      I dunno… I saw another press release that said it only beat the M3 in multi-core by about 8%, which is hardly “blown out of the water”, especially for a 9-month-old processor, and Qualcomm didn’t release single-core benchmarks, probably because they didn’t outperform the M3 at all. And Apple will be announcing the M4 line soon, expected to release Q3 or Q4 2024 for M4/M4 Pro, with Max and Ultra in Q1/2 2025.

      But it is, by all measure, a very good competitor to Apple’s M-series chips which have, until now, been unavailable to anyone but them, which is crappy.

    • Midnitte@beehaw.org
      link
      fedilink
      English
      arrow-up
      4
      ·
      7 months ago

      For most people (for the time being), the most important benchmark will be how the emulator/translation layer performs for their most common programs.

      Chrome, Firefox, and more obviously already have ARM builds, but most programs do not, not to mention device drivers that will need to be recompiled for ARM.

      There’s an Arstechnica article that goes into it, hopefully it works well and both MacOS, Windows, and Linux users will be able to reap the resulting benefits

      • gregorum
        link
        fedilink
        English
        arrow-up
        2
        ·
        7 months ago

        This is something Apple has done very well: Rosetta 2 is highly performant. Broadly speaking, there’s very little performance hit when running x64 apps on Apple Silicon. Of course, there hasn’t been x32 support on macOS for years, so I can’t speak to that.