Live data from Hacker News

Viewing profile — amarshall

amarshall

HN member
Joined
Mon, Sep 05, 2011, 7:58 PM UTC
HN karma
2,322
Public activity
605 items

About amarshall

https://johnandrewmarshall.com

Recent public activity

  1. comment
    Comment #49092408

    Use uBlock Origin and add custom filters to block scripts on http:// ; e.g. http://$inline-script,domain=~localhost http://$script,domain=~localhost

  2. comment
    Comment #48978137

    It’s possible to make a fake display to avoid that; both Linux and Windows have this ability. No dummy plug is required (though it is simpler than software solutions).

  3. comment
    Comment #48577947

    Pretty light on details, heavy on fluff. 9.8s to 3.1s was userfaultd + hugepages, 500ms was PS/2 mouse and… where is the rest of the time to get to 400ms?

  4. comment
    Comment #48564592

    No, and having that is not required to know output speed nor the effect of thinking, so I don’t see the point in such a superfluous, indirect question. As for the question you’re l…

  5. comment
    Comment #48547974

    Thinking doesn’t change output speed. Anthropic’s models are ~ 40–60 t/s median output speed.

  6. comment
    Comment #48547580

    Can take the geometric mean of total and active parameters of MoE to get approximate equivalent quality to dense model params. So sqrt(35*10)≈18.7. The trade-off of MoE is that it …

  7. comment
    Comment #48547519

    What context length and kv cache quant (if any) are you using? And MTP?

  8. comment
    Comment #48101187

    Probably want `kitty --single-instance` to reduce memory usage when opening multiple. https://sw.kovidgoyal.net/kitty/invocation/#cmdoption-kitty-...

  9. comment
    Comment #48009677

    The actual paper (linked in another comment here) is literally titled “Nonergodicity and Simpson’s paradox in neurocognitive dynamics of cognitive control”. Why they omit it from t…

  10. comment
    Comment #47996628

    Seems like a case of Simpson’s Paradox https://en.wikipedia.org/wiki/Simpson%27s_paradox

  11. comment
    Comment #47619724

    If you are running on 4090 and get 5 t/s, then you exceeded your VRAM and are offloading to the CPU (or there is some other serious perf. issue)

  12. comment
    Comment #47478249

    Oh I missed the "quad" before 3090.

  13. comment
    Comment #47473409

    You're almost certainly (definitely, in fact) confusing the 120b and 20b models.

  14. comment
    Comment #47237150

    Yes. Even 90 Hz is a noticeable improvement over 60 Hz. I wouldn’t pick it over high-DPI, though.

  15. comment
    Comment #46456144

    No, the connector is longer than it is tall.

  16. comment
    Comment #46455615

    Maybe, but Apple doesn’t make them thinner anyway so the argument is invalid. iPhone 6S with headphone jack: 7.1mm thick. iPhone 17 is 7.95mm thick.

  17. comment
    Comment #46172579

    It’s easier than that. Replace shorts with watch in the URL.

  18. comment
    Comment #45987563

    Non-Shorts can be vertical as well. It works fine, and without the swipe-encouraging UX. If they were “just making long/medium-form videos in portrait now” they don’t need YT Short…

  19. comment
    Comment #45819401

    DisplayPort 2.1 (which the monitor supports) provides sufficient bandwidth for 7680x4320@60 Hz 10-bit without DSC when using UHBR20. The press release unfortunately doesn’t clarify…

  20. comment
    Comment #45268537

    Redownloading everything isn’t a risk when the lock file contains a hash of the download on first update.

  21. comment
    Comment #45262192

    Enabling Reduce Transparency is a sure-fire way to find a dozen bugs within a few hours. It is always quite apparent no one who is empowered tests it at Apple. At least the padding…

  22. comment
    Comment #45256624

    > SO much padding No idea on macOS, but turn on Reduce Transparency on iOS and there’s tons of padding most of the time, but then sometimes zero padding. And I mean zero. The edges…

  23. comment
    Comment #45191721

    As an iPhone Mini owner, I have not experienced any poor or broken UX in apps, but ofc I am n=1.

  24. comment
    Comment #44864894

    > I'd have to spend thousands to get that setup Can be had for under US$1000 new https://pcpartpicker.com/list/WnDzTM . Used would be even less (and perhaps better, especially the …

  25. comment
    Comment #44806772

    Indeed a max-load trailer truck is equivalent to 10,000 cars in road wear. However the benefit of them (since they transport goods for many) is somewhat outweighed. Regardless, tha…