It's amazing to me that I spent a semester in the late 90s learning how to use marching ants, fill, select, magic wands, and... at the end of that time, I was able to create mind-blowing magic that looked as good as any professional's work in just a few hours' time.
Now, instead of a semester of learning (followed by years of practice), you can learn how to do prompt engineering in a day or two, then create stuff that would have been possible with Photoshop, but which would have taken you a month to do, in seconds.
One correction: With Midjourney, you don't need to spend more than minutes to learn how to get good images. It's literally as simple as describing what you want.
Sure, you can dive deeper to get more control and find out how to work with consistent characeters, multiprompts, etc. But just typing "photo of a person" will get you a realistic image of a person that doesn't exist.
I've been playing with Stable Diffusion and Midjourney since September last year, and it still feels like absolute magic to watch those pictures emerge from latent space. It's wild.
That's fair. I guess I meant getting the images you want- that takes a little practice, but creating something that you simply couldn't have made previously in seconds is very much a thing.
I need to dive into Stable Diffusion some time soon... it's just not as high of a priority as I thought it would be!
Oh yeah, in that sense it does require a bit more getting used to. I'd say anyone with Photoshop skills or proper knowledge of graphic design is still leaps and bounds ahead of noobs like me who can get a pretty image out of Midjourney but can't truly manipulate it afterwards.
And Stable Diffusion is definitely a stronger tool in the right hands as it has many far more advanced features you can nerd out with. Give it a whirl!
It's amazing to me that I spent a semester in the late 90s learning how to use marching ants, fill, select, magic wands, and... at the end of that time, I was able to create mind-blowing magic that looked as good as any professional's work in just a few hours' time.
Now, instead of a semester of learning (followed by years of practice), you can learn how to do prompt engineering in a day or two, then create stuff that would have been possible with Photoshop, but which would have taken you a month to do, in seconds.
Seconds!
True!
One correction: With Midjourney, you don't need to spend more than minutes to learn how to get good images. It's literally as simple as describing what you want.
Sure, you can dive deeper to get more control and find out how to work with consistent characeters, multiprompts, etc. But just typing "photo of a person" will get you a realistic image of a person that doesn't exist.
I've been playing with Stable Diffusion and Midjourney since September last year, and it still feels like absolute magic to watch those pictures emerge from latent space. It's wild.
That's fair. I guess I meant getting the images you want- that takes a little practice, but creating something that you simply couldn't have made previously in seconds is very much a thing.
I need to dive into Stable Diffusion some time soon... it's just not as high of a priority as I thought it would be!
Oh yeah, in that sense it does require a bit more getting used to. I'd say anyone with Photoshop skills or proper knowledge of graphic design is still leaps and bounds ahead of noobs like me who can get a pretty image out of Midjourney but can't truly manipulate it afterwards.
And Stable Diffusion is definitely a stronger tool in the right hands as it has many far more advanced features you can nerd out with. Give it a whirl!
Thanks, Daniel! I may reach out with dumb newbie questions in a week or so, if that's all right.