
New stuff








Well i'm impressed...like i said it's a learning curve.
SD 1.5 with my style is much worse compared to MJ, but sometimes not so bad



Mot mine, but nice SD with custom mixed models


Some of my Midjourney V4 art








Was playing just a little but guess it's a learning curve. I'm not impressed by AI...
I've been playing with SD quite a bit, and have to say it's probably one of the best tools for mood boards, I'll explain why.
Due to the data dragnet style of training, it tends to generate imagery that is stereotypical. So it kinda forces you to be creative with prompts, and to think about what image you really want, and what constitutes that image.
"Person sitting on chair" quickly turns into: "middle aged man sitting on an old wooden chair, in a small room with wall paper, lit through an open window."
Then you can get an infinite amount of permutations of this. Just brilliant for mood board generation!
That's a very good point indeed! Very useful indeed for mood boards. Maybe with a bit more development and tailored suggestions it could even be handy for quick and dirt storyboards as well.
Shishkin style























I use hosted thing.
But better you need 11GB or better VRAM and some high speed GPU with fast single precision CUDA.
So, 3080, 3090, 3090 Ti.
Or new 4090.
note that Stable Diffusion v1.5 is still not public for download and local use (and this is that is used here)/
@Vitaliy Wow! How much RAM do you need for your cool experiments?
On lotus flowers









Metals and colored neodymium magnet balls





Experiments with mosaic and materials
















Mathematically it looks like this

Where peaks are visually pleasing images (as it thinks of them). As people trained NN to provide them that they want (most probably using back propagation learning).
It looks like you're new here. If you want to get involved, click one of these buttons!