MidJourney or Dreamer for UX designers

UX NATA
Bootcamp
Published in
6 min readNov 27, 2023
/blend midjourney and figma logos

Some words about MidJourney…

Well, artificial intelligence takes me up more and more. I admire current solutions and wonder how good and convenient they are. How they can make our lives easier.

Immediately make a reservation that they won’t replace us, no! Be sure. Artificial intelligence is an amazing helper, which we still need to guide. It’s like a baby who absorbs everything but is not ready yet to do research and design interfaces on his own.

And I’d like to present you a few more interesting tools, which you can use as a source of inspiration: Midjorney and Dreamer.

Midjorney is a comprehensive platform designed specifically for designers. It offers a range of tools and features to streamline the design process and enhance collaboration.

With Midjorney, designers can access a wide variety of design assets, including templates, icons, and illustrations, to kick-start their projects. The platform also provides a powerful design editor that allows designers to create and customize their designs with ease.

One of the key features of Midjorney is its collaboration capabilities. Designers can invite team members and stakeholders to collaborate on projects, share feedback, and make real-time edits. This promotes effective communication and ensures that everyone is on the same page throughout the design process.

Midjorney also offers version control and project management features, allowing designers to easily track changes, manage multiple iterations of a design, and stay organized. This helps streamline the workflow and improves efficiency.

Furthermore, Midjorney provides integrations with popular design tools, such as Sketch and Adobe Creative Cloud, allowing designers to seamlessly import and export their designs. This flexibility enables designers to work with their preferred tools while still benefiting from the collaborative features of Midjorney.

In summary, Midjorney is a comprehensive platform that empowers designers by providing a range of tools, collaboration features, and integrations. It enhances the design process, promotes effective communication, and improves efficiency, making it an invaluable tool for designers.

My cat (MidJourney). He is too 🤩

The site offers three Midjourney subscription options. The cost mainly depends on the time of access to quick requests processing and the number of simultaneously processed tasks.

The Basic tariff allows generating a picture in a bit more than 3 hours (and that’s speed-up mode). The Standard subscription plan will cost you $30 per month. This tariff allows loading the processor slowly and generating pictures for whatever time you need.

Midjourney, like any other neural net, is in the process of never-ending training, and this is contributed by users who evaluate generated pics. By the way, you can get extra time to use the increased capacity for your assessment of the images.

And what is Dreamer?

Link to Dreamer https://www.figma.com/community/plugin/1151245850609894407/dreamer

Dream studio is an innovative user service based on the Stable Diffusion neural net. With this service, you can create images by text request or based on a reference picture.

As distinct from Stable Diffusion, which you have to install on a powerful PC, Dream Studio makes it possible to generate pics directly on the site.

The interface of the Dream Studio site is very simple and user-friendly. Users can adjust the parameters of images, for example, increase width or height, select the number of images generated per request and set the number of steps the neural net will take to generate the image.

If we compare these two services, in my opinion, MidJourney performs better. Let’s compare the result on the “woodpecker Art Deco” request:

MidJourney
Dreamer

Actually, both neural nets did it well. But Dreamer is a free plugin, and MidJourney is a separate service.

By the way, if you think that I made up this request, you probably have too good an opinion of me! There are special helper services to create this kind of request, I’d like to recommend you this one: Helper. This service allows selecting the variants of keywords, as well as styles and sizes of images.

Generally speaking, everything depends on your goals and objectives, and if you often work in Figma, learning Dreamer will be quicker, and if you’d like to be more creative, MJ fits better.

Let’s discuss the functionality!

1. Familiar in a new style

The Witcher (designer)

So, the first function to pay attention to is a mix of styles or characters. Harry Potter, designer — easy! Witcher writes something on a laptop? No problem.

2. Тemplates for ideas and integrations

For example, you can ask to generate fonts and then look for similar existing options if you want something new but have no time to create your option. Also, I recommend this service WhatFonts to find the font from the image — it’s very convenient.

3. How to get components from a complex image

To divide the image into components use the /describe function, select a suitable image and break down it into parts. Next, you can find your images on your account on MidJourney.

An example of the decomposition of the meditation app interface:

And another service can help you with generating a text ;)

5 rules for requesting in MidJourney:

  1. The main parameter of the image: photo, film frame, illustration, reference image. You can upload your pics or generate new ones but you should understand that this service is not the best helper in creating accurate interfaces so far, however it may be a good inspirator and support when creating.
  2. Object. Your image should have a character — person, table, chair, bird, etc. Of course, if you type ‘nature…’, this will work too but be ready for surprises.
  3. Object’s details: description, action. Try to approach the request as a descriptive essay — what the object is doing, how it looks. Besides, the action is better when describing the object’s feelings.
  4. Stylistics: aesthetics, color palette, author, artist. This matters too. Specify style and palette to make your image as interesting as possible. You may experiment and mix totally different and even opposite styles.
  5. Additional parameters: settings of the camera, lighting, weather, time of day, schedule.
  6. Add-ons of the service: image aspect, style, resolution, chaos parameter. Use the functions of the service, they can help to get exactly what you want — zoom, selection of image versions, improved quality, etc.

I hope that the materials were helpful to you! Good luck!)

--

--

UX NATA
Bootcamp

Sr UX Researcher (sometimes designer and analyst)