Good to see a huge range of the same thing in so many styles. I've done similar at a smaller (free) scale to explore art styles and avoid the clichés that the image models prefer, with some success.
My feelings, when presented with this big grid, is that if it looks simple[0], or that what complexity it contains is the sort of thing I expect in either a clip art[1] or stock photo[2] library, I prefer it.
Where it appears complex[3], that's where my gut says ~"nobody would put this much effort in, they must have asked an AI to do it for them".
[0] e.g. CoBrA, Vorticism, Minimalism
[1] e.g. Memphis / 80s Postmodern, Wayfinding / Signage
[2] e.g. Streamline Moderne, Constructivist, Dada Collage
[3] e.g. Woodcut, Mexican Calavera, Victorian Broadside
https://news.ycombinator.com/item?id=49764791#49767382 had a lot of healthy discussion, and many people pointed out what I already knew, the posters were too cluttered.
Claude steered me into improving the prompts with actual evaluations... which cost me actual money...
Thanks!
Good to see a huge range of the same thing in so many styles. I've done similar at a smaller (free) scale to explore art styles and avoid the clichés that the image models prefer, with some success.
My feelings, when presented with this big grid, is that if it looks simple[0], or that what complexity it contains is the sort of thing I expect in either a clip art[1] or stock photo[2] library, I prefer it.
Where it appears complex[3], that's where my gut says ~"nobody would put this much effort in, they must have asked an AI to do it for them".
[0] e.g. CoBrA, Vorticism, Minimalism
[1] e.g. Memphis / 80s Postmodern, Wayfinding / Signage
[2] e.g. Streamline Moderne, Constructivist, Dada Collage
[3] e.g. Woodcut, Mexican Calavera, Victorian Broadside
Sorry, this still looks like slop.