D&D General DALL·E 3 does amazing D&D art
- Thread starter M.T. Black
- Start date
ezo
Hero
Some sacrifices must be made.
Kichwas
Half-breed
So yesterday I wanted to make a pic for a token for an upcoming one-shot.
I fed the same prompt into ChatGPT, Microsoft Designer, and a local copy of Stable Diffusion using the 'DynaVision XL' SDXL model checkpoint (essentially the base training model).
Got three extremely different results:
ChatGPT:
Microsoft Designer:
Local Stable Diffusion:
"A vibrant 1970s fantasy style RPG illustration of a freckled dark brown-skinned dwarf woman, age 29, in a fullbody shot. She wears a platemail breastplate, and her black wavy hair is tied into a ponytail, adding to her fierce and majestic appearance. She holds a giant fantasy warhammer in both hands, standing confidently in a bustling underground dwarven cavern fortress. The cobblestone streets are lined with lively market stalls, filled with various goods from weapons to food. The underground town is filled with dwarves in medieval merchant attire, engaging in various activities, creating a lively and dynamic atmosphere. The lighting is moody, casting a deep tone over the scene. The dwarves are seen bartering and socializing, with the overall atmosphere being one of camaraderie and industriousness. The camera captures her full body."
I've been trying to get the style seen in Microsoft Designer within my local Stable Diffusion - and thus far I've yet to 'find' the right checkpoint model to be similar to theirs. They use DallE 3 which I suspect Stable Diffusion can't mimic.
Microsoft Designer locks you to some 15 images per month, and only a little more if you're paid - and lacks the editing tools of stable diffusion. That last image from Stable Diffusion went through my hand drawing a new hammer in Gimp (open source alternative to photoshop) and some other edits, and multiple rounds of upscaling and going back and forth between hand drawn items from myself and the AI tool.
ChatGPT seems to really want to do Ghibli style, and the style it gave here is what I get if I don't do ghibli. It also sat on it for an hour in 'thinking' mode. But at least I could download things, edit them in Gimp or Paint or Illustrator, and then hand them back to ChatGPT - but I've found it has a very short limit per day also. Stable Diffusion is running on my own machine in offline, so I can make hundreds of edits a day back and forth between it and my own hand drawing tools.
I fed the same prompt into ChatGPT, Microsoft Designer, and a local copy of Stable Diffusion using the 'DynaVision XL' SDXL model checkpoint (essentially the base training model).
Got three extremely different results:
ChatGPT:
Microsoft Designer:
Local Stable Diffusion:
"A vibrant 1970s fantasy style RPG illustration of a freckled dark brown-skinned dwarf woman, age 29, in a fullbody shot. She wears a platemail breastplate, and her black wavy hair is tied into a ponytail, adding to her fierce and majestic appearance. She holds a giant fantasy warhammer in both hands, standing confidently in a bustling underground dwarven cavern fortress. The cobblestone streets are lined with lively market stalls, filled with various goods from weapons to food. The underground town is filled with dwarves in medieval merchant attire, engaging in various activities, creating a lively and dynamic atmosphere. The lighting is moody, casting a deep tone over the scene. The dwarves are seen bartering and socializing, with the overall atmosphere being one of camaraderie and industriousness. The camera captures her full body."
I've been trying to get the style seen in Microsoft Designer within my local Stable Diffusion - and thus far I've yet to 'find' the right checkpoint model to be similar to theirs. They use DallE 3 which I suspect Stable Diffusion can't mimic.
Microsoft Designer locks you to some 15 images per month, and only a little more if you're paid - and lacks the editing tools of stable diffusion. That last image from Stable Diffusion went through my hand drawing a new hammer in Gimp (open source alternative to photoshop) and some other edits, and multiple rounds of upscaling and going back and forth between hand drawn items from myself and the AI tool.
ChatGPT seems to really want to do Ghibli style, and the style it gave here is what I get if I don't do ghibli. It also sat on it for an hour in 'thinking' mode. But at least I could download things, edit them in Gimp or Paint or Illustrator, and then hand them back to ChatGPT - but I've found it has a very short limit per day also. Stable Diffusion is running on my own machine in offline, so I can make hundreds of edits a day back and forth between it and my own hand drawing tools.
Scribe
Legend
Is that the same thing as Bing/Dall-E?
Kichwas
Half-breed
I think so: Microsoft Designer - Stunning designs in a flash
- You get 15 uses a month. So be careful. I ran all of them out last month in about 5 minutes just trying to get started with figuring out the menus... which you can't even see until after you've burned one charge to get 'in' - and those menus go away if you come back later so you've got to burn a charge to get back to them again each session...
Twiggly the Gnome
Legend
Did ChatGPT see the word "socializing" and turn the dwarves into Communists?
Lanefan
Victoria Rules
The first one is nice but makes me think Gnome rather than Dwarf, as our Gnomes are generally darker-skinned and genetically somewhat related to Dwarves (though on average about a foot and a half shorter).
What's with the stupid-big weapons in the second and third images there?
Scribe
Legend
overgeeked
Open-World Sandbox
Got Pirate Borg on my mind.
overgeeked
Open-World Sandbox
In my head I'm calling these neon pirates. For some reason I'm getting a surprising amount of punch daggers.
