• 0 Posts
  • 193 Comments
Joined 3 years ago
cake
Cake day: June 18th, 2023

help-circle










  • I was sliding it back and forth as fast as I could, because I didn’t understand the puzzle at first, but then I saw there’s an sorting optimization puzzle and thought that I was clever only to find out it’s tedium.

    Halfway through (or so I thought at move 80), I was just curious to see how ridiculously long the puzzle could be, shifting pieces back and forth

    So the answer was: I thought the puzzle was easy, and I was humbled, and stubborn.

    I solved it and wasted time, and used that time to warn others to not bother with the puzzle






  • Naz@sh.itjust.workstoComic Strips@lemmy.worldPost-truth society.
    link
    fedilink
    arrow-up
    9
    arrow-down
    1
    ·
    2 months ago

    Deconstructing the premise from first principles:

    except the healthy and truthful facts stand doesn’t and never did exist.

    False: Bodies of recorded objective facts such as encyclopedia and scientific papers exist.

    If it did it would be simple to train a healthy and truthful facts AI, you would simply train it off from the truthful and healthy facts.

    Logical fallacy: Assuming the antecedent, the first premise must be proved before a conclusion can be drawn.

    but everyone makes up their own narrative until all written word appear relative to the AI, making it impossible for the ai to tell fact from fiction.

    Opinion. Artificial intelligence is designed to reproduce patterns. If you presume your prior premise was true, a “facts trained” AI should in theory have been able to distinguish truth from falsehood.

    When you read “slop” you are simply reading the words of someone who doesn’t live in in your constructed narrative.

    Conflation: “Slop” is a pejorative term for all content generated by artificial intelligence, but particularly low quality, sloppy, one.

    And now to prove it I will have plenty of down votes on this comment.

    Logical fallacy: Strawman. Constructing a poor argument in order to prove it is, itself false, at best, a waste of time.

    Overall truth value: 16.17% (Mostly misleading/inflammatory comment)
    



  • I just need AV2 to not blow chunks on VR/Streaming.

    AV1 is currently hamstrung by decode compute costs, I’m sending 6K video @ 120 FPS and at 200 Mbits (25 MB/s) it looks like pea soup.

    At 400-500 Mbps (62.5 MB/s), you have twice as much network overhead but somehow H.264 still decodes faster even though it’s twice as much raw data per second.

    For straight up video file compression, AV1 seems to be absolutely unmatched if you use “peak framerate” (say 120) and then VFQ / variable quality set to like 35.

    I was able to convert a 900 MB (!!) H264 video of VR gameplay (6K across, 3K/eye!) into 96 MB, and slide right under Discord’s idiotic 100 MB limit.