Midjourney v6 Tips and Tricks for Beginners and Advanced Prompters. Part 1

Thaeyne
26 Dec 202308:22

TLDRThis video offers an introduction to the new features and capabilities of Midjourney V6, an AI technology for image generation. The host, TH, explains how to activate and use the V6 model, focusing on the aspect ratio control, upscaling options, and the menu of buttons available for post-generation modifications. The video also discusses the limitations and challenges faced when compared to the previous version, such as the absence of pan and zoom features, and provides tips on how to prompt for desired image outcomes, like avoiding close-ups. The summary invites viewers to discover more about Midjourney V6 and share their own tips and findings.

Takeaways

  • 🔍 Midjourney V6 is a distinct model from its predecessors and requires different prompting strategies.
  • 🛠️ Check your settings with '/slash settings' to ensure V6 is active or append '--V6' or '--V6.0' to your prompts.
  • 🖼️ Use the 'D-AR' argument to control the aspect ratio of your image, which affects its shape and pixel dimensions.
  • 🔄 The 'U1-U4' buttons in the Discord menu allow for upscaling images from the grid, but do not change the image size.
  • 🔄 The 'refresh' button rerolls the same prompt, which can be redundant unless the seed is set differently.
  • 🔄 The 'V1-V4' buttons offer variations of a liked image, with options for subtle or strong variations.
  • 🎨 The 'remix' feature allows for complete prompt changes if enabled in settings.
  • 📈 V6 currently lacks some features of V5.2, such as pan, zoom, and VAR region options, but promises to add them gradually.
  • 📐 The 'very subtle' and 'very strong' buttons provide more variation options compared to the grid image's V1-V4 buttons.
  • ❤️ The 'heart' button sets the image rating to the highest on Midjourney's website, aiding in organization and retrieval.
  • 🔗 The 'web' button links to the image on Midjourney's website, a method for downloading or viewing images.
  • 👢 Negative prompting, such as '--no closeup', can help control the image composition, especially in the absence of pan and zoom features.

Q & A

  • What is the main topic of the video script?

    -The main topic of the video script is introducing the new features and changes in Midjourney v6, focusing on tips and tricks for both beginners and advanced prompters.

  • How can you check which AI model is active in Midjourney?

    -You can check which AI model is active by using the '/slash settings' command to see the active model, or by adding '--V6' or '--v6.0' at the end of your prompt.

  • What does the 'D-AR' argument stand for in Midjourney v6?

    -The 'D-AR' argument stands for 'Aspect Ratio', which allows users to control the shape of their image, making it wider or taller as desired.

  • What is the default aspect ratio in Midjourney v6?

    -The default aspect ratio in Midjourney v6 is 1:1, which produces an image that is 1,024 by 1,024 pixels.

  • How does changing the aspect ratio affect the image size in Midjourney v6?

    -Changing the aspect ratio makes the image elongated in one direction while reducing its size in the other direction, but it does not increase the total number of pixels.

  • What are the U1 to U4 buttons for in the Midjourney v6 image menu?

    -The U1 to U4 buttons are for upscaling an image from the grid. However, in v6, pressing these buttons does not upscale the image but rather separates the image with some additional buttons.

  • What does the refresh button in the image menu do in Midjourney v6?

    -The refresh button is for rerolling the same prompt, generating the same image again, which can waste generation hours without changing the seed in the prompt.

  • What are the V1 to V4 buttons used for in Midjourney v6?

    -The V1 to V4 buttons are used when you like an image from the grid but want to see variations of it. They provide images that look similar to the original.

  • What features are currently missing in Midjourney v6 compared to version 5.2?

    -Midjourney v6 currently lacks features like pan, zoom, and VAR region options, which were available in version 5.2.

  • How can you avoid close-ups in your images when using Midjourney v6?

    -To avoid close-ups in Midjourney v6, you can use negative prompting by specifying '--no closeup' in your prompt, which may help to zoom out in the image.

  • What is the author's suggestion for organizing and finding favorite images on Midjourney's website?

    -The author suggests using the heart button to set the rating of an image to the highest option for better organization and easier finding of favorite images later on.

Outlines

00:00

🤖 Introduction to Mid Journey V6 AI Prompting

The video script introduces the viewer to the new features and changes in Mid Journey V6, an AI technology. It explains how to check the active model settings and suggests using '--V6' or '--v6.0' to specify the model in prompts. The script details the use of the 'D-AR' argument for controlling the aspect ratio of images, which affects the shape and size of the output. It also discusses the menu options available under generated images in Discord, including upscaling, rerolling prompts, and variations of images. The importance of aspect ratios in previous versions and the new features in V6 are highlighted, along with the potential for remixing prompts and the limitations of the current version compared to 5.2.

05:01

🔍 Exploring Mid Journey V6 Features and Limitations

This paragraph delves into the current features of Mid Journey V6, noting the absence of pan, zoom, and VAR region options which are important for controlling image output. It discusses the challenges of using photography language in prompts and suggests using negative prompting to avoid close-ups. The script shares personal experiences and tips, such as specifying the kind of shoes to depict the entire person in the image. It also mentions the different styles of image generation, like photography and painted styles, influenced by the type of prompt used. The paragraph concludes with a teaser for the next video, which will explore advanced features of V6, and an invitation for viewers to share their own tips and subscribe to the channel for more content.

Mindmap

Keywords

Midjourney V6

Midjourney V6 refers to the sixth iteration of the AI model developed by the company Midjourney. It is a significant upgrade from its earlier versions and is designed to enhance the capabilities of AI in image generation. In the video, the host discusses the differences and new features of V6, making it central to the theme of exploring advancements in AI technology.

Aspect Ratio (D-AR)

The aspect ratio is a fundamental concept in image composition that defines the proportional relationship between the width and height of an image. In the context of Midjourney V6, D-AR allows users to control the shape of the generated image, whether they want a wider or taller image. The script explains that the default ratio is 1:1, but users can specify different ratios to achieve the desired image dimensions.

Upscaling

Upscaling in the video refers to the process of increasing the resolution of an image while maintaining or improving its quality. Midjourney V6 offers options for upscaling images to a larger size, such as 2x upscale, although it is noted that there is no 4x upscale option available yet. The script mentions that upscaling does not change the aspect ratio but increases the pixel count.

Rerolling

Rerolling is the action of generating a new image using the exact same prompt to see a different outcome based on a different seed or randomization. The script warns that rerolling with an explicit seed set in the prompt will result in the same image, which could be inefficient use of generation hours.

Variations (V1-V4)

Variations in Midjourney V6 are options that allow users to see different versions of an image they like, with subtle or more pronounced differences. The script explains that there are buttons for different levels of variation, from very subtle to very strong, providing users with a range of options to explore different visual outcomes from a single prompt.

Negative Prompting

Negative prompting is a technique used in AI image generation where users specify what they do not want to see in the generated image. In the script, the host mentions using negative prompting with the phrase 'dash dash no closeup' to avoid close-up shots and instead get a more full-body representation in the images.

Photography Language

Photography language refers to the specific terminology used in the context of photography to describe image composition, such as 'close-up' or 'full shot'. The script notes that Midjourney V6 seems less responsive to this language compared to previous versions, requiring users to find alternative ways to communicate their desired image composition.

Seed

In the context of AI image generation, a seed is a value used to initialize the random number generator, which influences the outcome of the image. The script mentions that rerolling an image can involve changing the seed to produce a different image from the same prompt, unless the seed is explicitly set in the prompt.

Pan and Zoom

Pan and zoom are features that allow users to control the field of view in an image, either by panning across a scene or zooming in and out. The script notes the absence of these features in Midjourney V6, which affects the ability to control the composition of the generated images, such as getting a full-body shot.

Remix Feature

The remix feature in Midjourney V6 is an option that allows users to completely change their prompt after generating an image. The script explains that this feature can be enabled in settings, and when active, it provides a popup with the prompt text for editing, offering flexibility in the creative process.

Image Resolution

Image resolution refers to the number of pixels in an image, which determines its size and quality. The script discusses the default resolution of images generated by Midjourney V6 and how upscaling affects the resolution, noting that the upscaled images are 248x248 pixels, which is a 2x increase from the original.

Highlights

Midjourney V6 is different from earlier models and offers new ways of prompting.

Check settings with /settings to see which model is active or add --v6 for specifying V6.

D-AR argument allows control over the aspect ratio of the image, affecting its shape and size.

Default image size is 1:1 ratio, producing a 1,024x1,024 pixel image.

Changing the aspect ratio elongates the image in one direction while reducing it in the other.

Upscaling options are available under the generated image menu in Discord.

U1 to U4 buttons are for selecting images from the grid without upscaling.

Reroll button recreates the same prompt with a different seed for variation.

V1 to V4 buttons offer variations of a liked image from the grid.

Variation buttons allow the use of the remix feature for complete prompt changes.

Upscaled images in V6 result in a 2x increase to 2,048x2,048 pixels, with subtle and creative options.

V6 currently lacks pan, zoom, and VAR region options, making some image controls more challenging.

Negative prompting, such as specifying --no closeup, can help control image focus.

Adjusting the aspect ratio can provide more room for body representation in images.

Specifying details like shoes can influence the style and composition of generated images.

The speaker will explore more advanced features of V6 in a follow-up video.

The community is encouraged to share their findings and tips on using Midjourney V6.

Subscribers are thanked for their support, and new viewers are invited to subscribe for more content.