• Stumblinbear
    link
    fedilink
    English
    arrow-up
    21
    arrow-down
    4
    ·
    1 year ago

    It can often take longer to get what you want out of it than it would’ve to have just drawn it. I’ve spent 8 or 9 hours fiddling with inputs and settings for a piece and it still didn’t come out as good as it would have if I had commissioned an artist.

    I’ve been using it to get “close” then using it as a reference when commissioning things

    • rufus@discuss.tchncs.de
      link
      fedilink
      English
      arrow-up
      4
      ·
      edit-2
      1 year ago

      Yes. I also think that’s how it is. If you want to generate something meaningful, something that contributes something deep, it is quite a lot of effort. You need to do the prompt engineering, generate a few hundred images. Skim through them and find the most promising ones, then edit them. Maybe combine more than one or put it back into the AI to get the right amount of limbs and fingers. And the lighting, background etc right.

      You can just do one-shot, generate anything and upload it to the internet. But it wouldn’t be of the same value. But it works like this for anything. I can take a photo of something. Somebody else can have their photos printed on a magazine or do an exhibition. It’s a difference in skills and effort. Taking artistic photographs probably also takes some time and effort. You can ask the same question with that. Are photographs art? It depends. For other meanings of ‘OC’: Sure. The generated output is unique and you created it.