I’m a software developer with a platform-independent stack (java / postgre / mysql / intellij / docker), I use a Linux distro. I have a workstation, but would like to be able to work away from home. Good battery life, small size, staying cool under load are the priorities; I don’t need a lot of power. So I thought maybe I should try ARM?

My first idea was to get a [refurbished] MacBook Air and learn how to use MacOS, although I’d love to support something… less proprietary and more open. I’ve never used an ARM Linux distro or ARM laptops, and I’m not sure how good they are for my application.

What is your experience?

  • recursive_recursion [they/them]@programming.dev
    link
    fedilink
    English
    arrow-up
    6
    arrow-down
    1
    ·
    2 months ago

    I’ve heard that used/refurb Thinkpads are pretty solid for this usecase

    maybe the older Frameworks might also be a good option as I’ve seen the older models go on sale

    Macsbooks I’m hesistent to recommend as you’d be SOL for repairs unless you’re near a good repairshop like Louis Rossmann but Macs could still be good depending on what sales you find


    the closest exp I’ve had with ARM is Arch Linux on my Raspberry Pi 4 which runs but is definitely hampered on performance and functional apps in comparison to any other laptops

      • sloppy_diffuser@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        2
        ·
        2 months ago

        Brew or Nix do a pretty good job of closing the gap.

        I’m the only Linux user on my team and was getting pissed at Apple’s old bash (4.x something) and some of the tools like find having slightly different flags causing some scripts to not work locally.

        Added a nix flake + devshell + direnv and now they run the same version the servers do of everything, but only when in the project.

  • Pasta Dental@sh.itjust.works
    link
    fedilink
    arrow-up
    4
    ·
    edit-2
    2 months ago

    I think in general ARM support is better on Linux than on windows, because ARM servers have been a thing for a few years now. I think basically the entire Ubuntu repository is compiled for ARM now.

    For the battery life, I’m not even sure an arm laptop is what you need considering the state of driver support for power efficiency features vs Intel or amd chips. The new Intel lunar lake chips especially score exceptionally well in terms of efficiency. Better battery life with less noise as the AMD HX Zen5 and about as good as the Apple M3 levels of good

  • lkdm@programming.dev
    link
    fedilink
    arrow-up
    3
    ·
    edit-2
    2 months ago

    Refurbished ARM-based MacBook Air will be suitable as long as you get 16 or 32 GB configuration. Anything under 16 is unusable with docker, a browser, an IDE, and LSP running at the same time.

    ARM-based Macs can run Linux thanks to Asahi Linux.

    But I love and use Mac OS. It’s really good for development work.

  • fubarx@lemmy.ml
    link
    fedilink
    arrow-up
    2
    arrow-down
    1
    ·
    2 months ago

    Have been a longtime MacOS user. Have used Parallels/VMWare to run Windows and multiple flavors of Linux (Ubuntu, Debian, Kali, Arch).

    Have also used it for professional development targeting multiple platforms, including web, iOS, Android apps, AOSP, Windows, Linux, and hardware like Arduino, FreeRTOS, and Raspian. Docker desktop and K8 lets you build, test, and deploy just about anything anywhere. If you plan to run ML and do training stuff, I’d invest in something beefier than a MBA.

    If hung-up and religious on open-source OS, then would avoid. But if you just want to get shit done, it works really well.

  • sloppy_diffuser@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    1
    arrow-down
    1
    ·
    edit-2
    2 months ago

    Do you even linter bro? lol. I do TypeScript, but even my i9 can get bogged down by the language server from time to time.

    The whole hard capped RAM on the SoC is not for me also. I do a few things in python. If I want to load up a 32G pandas data frame on my laptop, I want the option of 64G RAM without having to replace the whole thing.

    I get MAYBE 2h battery life (Dell Precision) when I’m hacking away, so my use case may not match yours. Vroom vroom…

    That said, ARM itself is generally fine. We have some graviton EC2s hosting our application. I’d just make sure the docker images you expect to use have ARM builds. That bit us a couple times.