The number of Midjourney parameters and command settings increases with each upgrade and the addition of new features. Each parameter may also have multiple options, providing more tweaking opportunities to expand creativity.
Using multiple parameters in a prompt to get the desired images has become the norm.
By default, the Discord and web do not require the use of any parameters. The bot will automatically set the aspect ratio to 1:1 and use the latest Midjourney version for you.
However, explicitly writing down the parameters (for example, aspect ratio and version) can serve as the placeholder or reminder to explore additional image generation options to achieve desired visuals.
For instance, consider the following two prompts:
Prompt 1: /imagine prompt: photograph portrait of a lady cyborg, fine art
Prompt 2: /imagine prompt: photograph portrait of a lady cyborg, fine art --ar 3:4 --v 6.1
Pro tip: Adding the style-related keyword "fine art" at the end of the prompt will give your images a more refined appearance than without.
Prompt 1 used the default settings of aspect ratio 1:1 and the current Midjourney V6.1 even when the users did not explicitly specify them.
Prompt 2 generated an additional set of visual options.
It is not the point here to compare which prompt generates nicer aesthetic, because they are identical!
Except for Prompt 2, which uses a different aspect ratio, 3:4 is generally a better choice for portrait images than 1:1.
Prompt 1 might generate an image that you prefer on the first try. If you did not specify the aspect ratio when creating the images, you may accept it without thinking any further.
Prompt 2, on the other hand, has a clear layout of the parameter settings and could serve as a reminder to explore other available options.
For portrait images, consider using a vertical format such as 3:4 or 9:16.
(Of course, if you prefer the close-up images focusing on the faces in Prompt 1, that's perfectly fine.)
What other creative options are there?
Have you thought about using Midjourney V5.2 to create cyborgs? Its aesthetic differs from V6.1 and often more creative.
Would you like to bump up the Stylize (--s) value to see how the image will look?
Do you want to create greater variety of images in the image grid by using the Chaos (--c) (or the Weird parameter)
On the Web, you can explore additional options by clicking on the various buttons.
It is okay if you only want to create a few images with simple parameters.
For more advanced and complex prompting, use Discord.
Explicitly writing out all parameters, including the default settings, will remind you of the creative possibilities you have yet to explore.
Taming the complex prompts
Let's now look at an example of a complex prompt.
Prompt 3: /imagine prompt: end scene photography shot of Emily was rescued by the helicopter team, extreme wide camera angle --cref https://s.mj.run/v-bkWDK3plU --cw 80 --sref https://s.mj.run/v-bkWDK3plU --sw 40 --ar 16:9 --v 6.1
When you want to be creative and start adding Sref and Personalization codes, adjust the stylize value, try style raw, chaos, multi-prompt with weights, and so on, the number of parameters quickly piles up.
Oh my, things can get out of hand.
Because there are so many interconnected parameters at work, you may be hesitant to adjust or touch the prompt again for fear of breaking it.
But that hesitancy limits your creativity.
So, how do you manage all of those parameters and options while still being able to explore creative possibilities without breaking a sweat?
That's a secret.
But I'd like to share it with you because you subscribed to the Geeky Curiosity Newsletter!
PS: My cat gave me an eye roll for my marketing pitch :-)
Keep reading with a 7-day free trial
Subscribe to Geeky Curiosity to keep reading this post and get 7 days of free access to the full post archives.