• 0 Posts
  • 26 Comments
Joined 1 year ago
cake
Cake day: June 12th, 2023

help-circle



  • Ran Asahi for several months, tried it out again recently. It’s good/fine, I just don’t love fedora.

    There’s some funkiness with the more complicated install, the AI acceleration doesn’t work, no thunderbolt / docking station.

    MacBooks are great hardware but I don’t think they’re the best option for Linux right now. If you’re never going to boot into macOS then I’d look for x13, new Qualcomm, isn’t there a framework arm64 option now or was that a RISC module?

    I’m also assuming you’re not looking to do any gaming? Because gaming on ARM is not really a thing right now and doesn’t feel like it will be for a long while.



  • Taking ollama for instance, either the whole model runs in vram and compute is done on the gpu, or it runs in system ram and compute is done on the cpu. Running models on CPU is horribly slow. You won’t want to do it for large models

    LM studio and others allow you to run part of the model on GPU and part on CPU, splitting memory requirements but still pretty slow.

    Even the smaller 7B parameter models run pretty slow in CPU and the huge models are orders of magnitude slower

    So technically more system ram will let you run some larger models but you will quickly figure out you just don’t want to do it.






  • Just a note, the orange pi drivers are not in great shape. It’s getting better but I have a cluster of raspberry pi’s for development, bought an orange pi without first checking out much about them and it’s rough. Rockchip CPUs are great, and the driver / firmware situation is getting better, but something I’d read up on before buying one.

    I’d still look at the N100, it’s about 2.5x the performance of raspberry pi 5, and being x86 you have more options than arm.



  • I’m far from an expert in init systems, but there are some benefits to declarative approaches for configuration. It’s one of the main reasons yaml and toml are as popular as they are. The short version is, declarative configuration tends to be less verbose, and the declarative contract defines what state you want things to be in, not how to get there which makes it easier on the person writing the unit file, and on the implementers of systemd in that there’s a smaller surface-area to test

    Generally declarative:

    • requires less verbose configuration files, less room for error
    • is easier to document and easier to understand
    • leaves the implementers more freedom to improve their system as long as they live up to the agreed-upon contract
    • is easier for implementers to test/validate. They don’t need to support a scripting language and every single crazy thing someone might try with one but still consider valid







  • One really annoying problem, in firefox the mouse cursor position is wrong. Clicking in firefox clicks on the thing 30-ish pixels above my mouse. I noticed that it only happens when firefox is snapped to the left or right half of the screen (of course that’s how I almost always use it). I can fix it if I maximize firefox then snap it left or right. 100% scaling on the monitor, nothing funky, reset theme/appearance, reset my firefox profile, etc…

    Hopefully it gets sorted soon