HOME | DD

Description
I posted this not to show off the neat Stable Diffusion art I generated in a whopping seven seconds, but to maybe open a discussion about how artists deal with this. I just spent the last three days tinkering in DAZ Studio, learning more about Camera Raw's regional adjustments, and making notes about how to improve textures in Substance Painter and export high detail normals from Zbrush - and then I cut and paste a prompt from a Midjourney image into my local Stable Diffusion, change a couple of the words, and get this - something I COULD NEVER GET out of all of those programs put together if I spent a year trying. I didn't even do any postwork here, I just shrunk it and posted.

It's both exciting and depressing. To potentially make art with my characters or ideas in amazing and beautiful styles that I could never do otherwise is really inspiring and fun. The fact that it's still extremely random and very unlikely to be as accurate to my ideas as I'd like, as well as the resulting figures not being able to interact with other figures properly, is quite frustrating. AI technology is advancing rapidly, and it's possible that all of the problems with crazy hands and wonky faces could be solved in just a few months time. Maybe the one feature I would most like to see - to be able to simply take an image you give it and recreate it in a different artist's style without changing the content - will be implemented, and we'll be able to make amazing art like this while still giving it the human touch and attention to detail that gives it personality and originality.

But at the moment, all this makes me do is wonder why I continue struggling with finding and purchasing content I want, getting complex 3D programs to work with each other, and learning painting techniques that take a lifetime to master when I can get this in seven seconds. Does it discourage you too?

Generated in NMKD Stable Diffusion using a prompt from a Midjourney gallery entry with minor changes. This was the second image it made out of five. No postwork, no work from me at all besides resizing and a slight sharpen filter in Irfanview. :sigh: Thank you very much for taking a look!

Description
I posted this not to show off the neat Stable Diffusion art I generated in a whopping seven seconds, but to maybe open a discussion about how artists deal with this. I just spent the last three days tinkering in DAZ Studio, learning more about Camera Raw's regional adjustments, and making notes about how to improve textures in Substance Painter and export high detail normals from Zbrush - and then I cut and paste a prompt from a Midjourney image into my local Stable Diffusion, change a couple of the words, and get this - something I COULD NEVER GET out of all of those programs put together if I spent a year trying. I didn't even do any postwork here, I just shrunk it and posted.

It's both exciting and depressing. To potentially make art with my characters or ideas in amazing and beautiful styles that I could never do otherwise is really inspiring and fun. The fact that it's still extremely random and very unlikely to be as accurate to my ideas as I'd like, as well as the resulting figures not being able to interact with other figures properly, is quite frustrating. AI technology is advancing rapidly, and it's possible that all of the problems with crazy hands and wonky faces could be solved in just a few months time. Maybe the one feature I would most like to see - to be able to simply take an image you give it and recreate it in a different artist's style without changing the content - will be implemented, and we'll be able to make amazing art like this while still giving it the human touch and attention to detail that gives it personality and originality.

But at the moment, all this makes me do is wonder why I continue struggling with finding and purchasing content I want, getting complex 3D programs to work with each other, and learning painting techniques that take a lifetime to master when I can get this in seven seconds. Does it discourage you too?

Generated in NMKD Stable Diffusion using a prompt from a Midjourney gallery entry with minor changes. This was the second image it made out of five. No postwork, no work from me at all besides resizing and a slight sharpen filter in Irfanview. :sigh: Thank you very much for taking a look!