• 0 Posts
  • 63 Comments
Joined 1 year ago
cake
Cake day: June 18th, 2023

help-circle
  • I’m sorry; AI was trained on the sole sum of human knowledge… if the perfect human being is by nature some variant of a psychopath, then perhaps the bias exists in the training data, and not the machine?

    How can we create a perfect, moral human being out of the soup we currently have? I personally think it’s a miracle that sociopathy is the lowest of the neurological disorders our thinking machines have developed.






  • That seems to require a level of foresight and planning that most corporations don’t have. That’s almost like a blueprint for failure when some middle manager changes the scope of a project with a hard coded time limit, IMO.

    Anyone interested in not-agile development? Maybe we can call it “Ship it when it’s ready” lol




  • Naz@sh.itjust.workstoMemes@lemmy.mlGet rich quick
    link
    fedilink
    arrow-up
    20
    arrow-down
    2
    ·
    edit-2
    18 days ago

    I’m an AI Developer.

    TLDR: CUDA.

    Getting ROCM to work properly is like herding cats.

    You need a custom implementation for the specific operating system, the driver version must be locked and compatible, especially with a Workstation / WRX card, the Pro drivers are especially prone to breaking, you need the specific dependencies to be compiled for your variant of HIPBlas, or zLUDA, if that doesn’t work, you need ONNX transition graphs, but then find out PyTorch doesn’t support ONNX unless it’s 1.2.0 which breaks another dependency of X-Transformers, which then breaks because the version of HIPBlas is incompatible with that older version of Python and …

    Inhales

    And THEN MAYBE it’ll work at 85% of the speed of CUDA. If it doesn’t crash first due to an arbitrary error such as CUDA_UNIMPEMENTED_FUNCTION_HALF

    You get the picture. On Nvidia, it’s click, open, CUDA working? Yes?, done. You don’t spend 120 hours fucking around and recompiling for your specific usecase.





  • Sure, I’ll try OpenSUSE!

    Tumbleweed is a bit of a spooky name for a distro implying that a gentle breeze sends it, but y’know

    Linux Mint as someone suggested, I’ve ran a long time ago for college on an ancient laptop, and it’s an extreme stable OS, similar to Windows 2000 Pro. I can’t remember it crashing or freezing even once on me, and the Thinkpad T42 has an anemic processor., which I ran with the Conservative Governor


  • I’m actually a little scared of running Linux on modern, fast hardware.

    How is multi-GPU driver support?

    My main machine is a 900 TFlops compute monster (4 GPUs) running ROCM on Windows, and the last time I’d tried Manjaro on Desktop, it seized up for unknown reasons.

    I’ve got asynchronous monitors - 1440p@165Hz main display and 4K@85Hz flipped vertical for a side monitor. Occasionally, I plug in a projector which is 1080p, mirrored to the 4K, but flipped horizontal.

    I’m not sure what I’d done wrong because it works perfectly on my 11 year old Z575 (Debian+KDE there).

    What distro would you recommend for an extremely fast/high RAM machine? I’ve got 128GB of main system memory, and 4TB of M.2 for a system disk running at 7.6 gigabytes/second actual/real-world RW I/O.


  • I had something like this happen at a corp I once worked at. The CTO said they were going to outsource their entire datacenter and support staff to India.

    I literally laughed in his face and obviously, got fired (always have 6-8 months of salary as an emergency fund, ahem-).

    I won’t name the company but when half the Internet went down and a few major services? Yeah, it was that asshat driving and running between the datacenters realizing people in Bangladesh can’t do shit for you physically.

    It’s like that graph: “Say we want to fuck around at a level 8, we follow this axis, and we’re going to find out at around a level 7 or 8”