‹ BackHN Continuity

Thread

M5 Ultra Mac Studio Review

269 points · 262 comments · piotrgrabowski

  1. simonw · · focus · HN ↗
    The numbers I was most interested in are tucked away in a chart towards the bottom - the speed comparison of the Mac Studios v.s. a RTX 5090:

      Qwen3.8 27B tokens/sec generation speed
    
      Prompt size    8K    64K   128K   256K
      RTX 5090 PC    59    51    44     n/a
      M5 Ultra       48    39    32     24
      M3 Ultra       31    23.5  20     15
    
    A whole bunch more comparison numbers in this section: <a href="https:&#x2F;&#x2F;www.macstories.net&#x2F;stories&#x2F;m5-ultra-mac-studio-review-the-dream-mac-for-local-ai-agents&#x2F;#mx-pc" rel="nofollow">https:&#x2F;&#x2F;www.macstories.net&#x2F;stories&#x2F;m5-ultra-mac-studio-revie...
    1. RationPhantoms · · focus · HN ↗
      Thank you for this. I wish Apple focused their silicon design on improving the TTFT metrics but coming from an M3 Pro, it still looks laggard compared to Nvidia&#x27;s TensorCores in the 5090.

      Maybe Apple is an acquisition away from changing that balance.

      1. wlesieutre · · focus · HN ↗
        The rumor on Apple&#x27;s processor roadmap is that they&#x27;re skipping other M6 variations (all previous generations had Pro and Max, a few had Ultra) in order to focus on the M7 generation for AI reasons. What exactly the M7 improvements are who knows.

        <a href="https:&#x2F;&#x2F;www.macrumors.com&#x2F;2026&#x2F;06&#x2F;25&#x2F;2027-macs-m7-chips&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.macrumors.com&#x2F;2026&#x2F;06&#x2F;25&#x2F;2027-macs-m7-chips&#x2F;

        1. kridsdale1 · · focus · HN ↗
          I think that comes down to TSMC. Nvidia apparently booked out the whole A18 or 16 node. Apple is on 2nm right now and M7 will jump right to A14. According to my quick AI research anyway.
          1. dagmx · · focus · HN ↗
            That sounds a lot like AI fantasy slop.

            Apple just shifted to N2. They’re not going to be doing another major shift right away.

            And TSMCs own roadmap would put your hallucination years away at best for a a product that follows a roughly annual cadence <a href="https:&#x2F;&#x2F;www.tomshardware.com&#x2F;tech-industry&#x2F;semiconductors&#x2F;tsmc-unveils-process-technology-roadmap-through-2029-a12-a13-n2u-announced-a16-slips-to-2027" rel="nofollow">https:&#x2F;&#x2F;www.tomshardware.com&#x2F;tech-industry&#x2F;semiconductors&#x2F;ts...

            1. smith7018 · · focus · HN ↗
              Yeah, Apple has spent 3 years each on the 5nm and 3nm nodes with TSMC. There are some reports [1] that it will jump to 1.4nm after 2 years due to AI but there&#x27;s no real proof. The source is Digitimes who are frequently wrong with their predictions and rumors.

              [1] <a href="https:&#x2F;&#x2F;wccftech.com&#x2F;apple-to-move-to-1-4nm-process-soon-to-secure-adequate-chip-supply&#x2F;" rel="nofollow">https:&#x2F;&#x2F;wccftech.com&#x2F;apple-to-move-to-1-4nm-process-soon-to-...

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.