Mastering Midjourney v6 : A Deep Dive into the New Version!

Theoretically Media
21 Dec 202312:06

TLDRThe video discusses the release of Mid Journey version 6, highlighting its significant updates over version 5.2 in aesthetics, prompt understanding, and new features. It emphasizes the need to relearn prompting for V6, which is more sensitive to keywords and offers better detail, especially in hair and skin textures. The video also explores creative upscaling, the ability to prompt for text, and the current limitations of certain commands in the beta release.

Takeaways

  • ๐Ÿš€ Mid Journey version 6 has been released, marking a significant update over version 5.2, especially in terms of aesthetics and prompt understanding.
  • ๐Ÿ”ฅ Version 6 introduces a few new features but is a beta release, meaning some features are missing.
  • ๐Ÿ“ To use version 6, users can either set their settings to Mid Journey model V6 or add 'D-V6' to the end of any prompt.
  • ๐Ÿ“ Prompting in V6 is notably different from V5, requiring users to relearn prompting techniques and avoid 'junk' keywords for better results.
  • ๐Ÿ“ธ Image quality, especially in terms of details like hair and skin, has improved with V6, demonstrating the model's sensitivity to keywords.
  • ๐Ÿ“š The update encourages experimentation with minimal and creative prompts, showing V6's broadened capability in generating diverse imagery.
  • ๐Ÿ“ฑ V6 enhances the depth and realism of images, as illustrated by examples comparing the new model's output to previous versions.
  • ๐Ÿ“ข Creative upscaling is a major feature of V6, offering the ability to enhance details in upscaled images.
  • ๐Ÿ’ฌ Text prompting has improved, allowing for text to appear more accurately within images, albeit with some limitations.
  • ๐Ÿ›  Some commands from previous versions are not yet supported in V6, but existing functionalities like aspect ratio adjustment remain.

Q & A

  • What is the main focus of the video?

    -The main focus of the video is to provide a detailed overview of the newly released Mid Journey version 6, its features, improvements, and changes in prompting compared to version 5.2.

  • How can one switch to Mid Journey version 6?

    -To switch to Mid Journey version 6, users can issue the command 'SL settings' and choose Mid Journey model V6. Alternatively, they can add 'D-V6' to the end of any prompt to render it with version 6.

  • What are some differences in prompting between version 5.2 and version 6?

    -Version 6 has significantly different prompting compared to version 5.2. Users need to relearn how to prompt for V6, as it is more sensitive to the prompt and requires more explicit instructions. Avoiding junk keywords and being clear about desired outcomes are recommended.

  • What is the significance of the 'styze' parameter in the new version?

    -The 'styze' parameter in version 6 allows users to adjust the level of stylization in the generated images. By increasing the styze value, such as to 1,000, and giving minimal prompts, users can achieve interesting and creative results.

  • How does version 6 handle text in images?

    -Version 6 introduces the ability to prompt for text within images. To include text, users should encapsulate the text within quotation marks. The new version seems to handle text better than its predecessors, with improved accuracy and integration into the image.

  • What are the two major new features introduced in version 6?

    -The two major new features introduced in version 6 are creative upscaling, which enhances image details and resolution, and the ability to prompt for text within images, improving the integration of textual elements into the generated scenes.

  • What are some commands that are not yet working in version 6?

    -As of the video, some commands that are not yet working in version 6 include pan, zoom, very region, in painting, tuning, and describe. However, describe is expected to be updated for V6 soon.

  • How does the 'very subtle' and 'very strong' parameters perform in version 6?

    -In version 6, the 'very subtle' and 'very strong' parameters have become more powerful, allowing for nuanced adjustments to the image based on the keywords used. For example, changing 'motorcycle' to 'van' in a prompt results in a detailed and accurate representation of the new keyword.

  • What is the speaker's opinion on the quality of the images generated by version 6?

    -The speaker believes that version 6 has improved significantly in terms of detail, realism, and understanding of prompts. The characters and scenes generated by version 6 are noted to have more personality and depth compared to previous versions.

  • How can users access the images and PDF worksheet from the video?

    -Users can access the images and PDF worksheet from the video by visiting the gumroad link provided in the video description. The content is offered for free, though donations are appreciated.

Outlines

00:00

๐Ÿš€ Introduction to Mid Journey Version 6

The script begins with an introduction to Mid Journey Version 6, highlighting its significant updates over Version 5.2 in terms of aesthetics and prompt understanding. It mentions new features and the fact that this is a Beta release, hence some features are missing. The speaker plans to dive deep into Version 6, discuss prompting, compare images, and share tips and tricks. The audience is advised on how to run Version 6 and the importance of relearning how to prompt for this new version, which is more sensitive to keywords and requires more explicit instructions.

05:02

๐ŸŽจ Detailed Comparison and Prompting Techniques

The speaker continues with a detailed comparison of images generated by different versions of Mid Journey, emphasizing the improved detail and realism in Version 6. They discuss the importance of using the right keywords and avoiding 'junk' words for better results. The speaker shares personal experiences with prompts and the impact of adjusting the 'style' parameter. They also mention the ability to include text in images and provide examples of how this can be done effectively, despite some commands not working in the current version of the software.

10:04

๐ŸŒŸ New Features and Upcoming Updates in Version 6

The final paragraph focuses on the new features of Version 6, such as creative upscaling and the ability to prompt for text. The speaker explains how creative upscaling enhances images and how text can be incorporated into prompts more effectively than in previous versions. They also discuss the limitations of the current version, noting which commands are not yet functional. The speaker expresses excitement for upcoming updates, particularly the new 'describe' function in Version 6, and shares their anticipation for future improvements.

Mindmap

Keywords

๐Ÿ’กMid Journey Version 6

Mid Journey Version 6 refers to the latest update of the AI image generation software, which is a significant leap from Version 5.2 in terms of aesthetics, prompt understanding, and new features. It is a beta release, meaning it is still in testing and may have some missing features. The video discusses the improvements and changes brought by this version.

๐Ÿ’กAesthetics

In the context of the video, aesthetics refers to the visual appeal and quality of the images generated by the AI software. The update to Version 6 is said to have improved aesthetics, meaning the images are more visually pleasing and realistic.

๐Ÿ’กPrompt Understanding

Prompt understanding is the ability of the AI to accurately interpret and respond to the user's input, or 'prompts', which guide the generation of images. The video suggests that Version 6 has significantly enhanced prompt understanding, requiring users to relearn how to effectively use prompts to get desired results.

๐Ÿ’กBeta Release

A beta release is a version of a software that is still in the testing phase and has not been fully finalized. It is made available to users to gather feedback and identify issues before the official release. In the video, Version 6 is described as a beta release, indicating that it may have some features missing or not fully polished.

๐Ÿ’กCreative Upscaling

Creative upscaling is a feature in Version 6 that allows users to enhance the resolution and detail of an image in a way that adds artistic elements and improvements beyond just increasing the pixel count. It is a more sophisticated process than simple enlargement.

๐Ÿ’กText Prompting

Text prompting is a new feature in Version 6 that enables users to include text within the generated images. This is done by encapsulating the text within quotation marks, allowing the AI to interpret and visualize the text as part of the image content.

๐Ÿ’กAspect Ratio

Aspect ratio refers to the proportional relationship between the width and height of an image. In the context of the video, aspect ratio is a command that users can adjust to change the shape and size of the generated images, affecting the composition and focus of the visual content.

๐Ÿ’กJumping Quality

Jumping quality refers to the significant improvement in the quality of the generated images from one version to the next. The term is used to describe the leap in visual fidelity and the ability of the AI to create more realistic and detailed images.

๐Ÿ’กForeground Objects

Foreground objects are elements in an image that appear closer to the viewer than the background elements. In the context of the video, Version 6 seems to favor placing objects in the foreground, which adds depth and a sense of presence to the images.

๐Ÿ’กCommands

Commands are specific instructions or requests given to the AI to control certain aspects of the image generation process. The video discusses which commands are supported and which are not in Version 6, affecting the user's ability to manipulate the output.

๐Ÿ’กVery Subtle and Very Strong

Very subtle and very strong are terms used to describe the level of change or intensity in the generated images. 'Very subtle' refers to minimal changes that maintain the essence of the original image, while 'very strong' indicates a more pronounced transformation. These terms are used to guide the AI in the degree of alteration from the base image.

Highlights

Mid Journey version 6 has been released, marking a significant update over version 5.2.

Version 6 introduces changes in aesthetics and prompt understanding, and includes new features despite being a Beta release.

Users can switch to version 6 by issuing the command 'SL settings' and selecting Mid Journey model V6, or by adding 'D-V6' to any prompt.

Prompting with V6 has changed, requiring users to relearn how to prompt effectively, with V6 being more sensitive to prompt keywords.

V6 advises against using junk keywords like 'award-winning', 'photo realistic 4K', and '8K', and encourages explicit descriptions.

The new version is less 'viby' but better at understanding what users are looking for, according to Mid Journey.

The presenter shares their personal mantra that there is no right or wrong way to prompt, only efficient and effective methods.

An example is given where a simple toast prompt can be transformed with the addition of 'style 1000', resulting in a detailed image.

V6 shows improvement in detail, especially in skin, eyes, and hair, as demonstrated by the image of a young couple in the rain.

The new model pays close attention to keywords, as changing 'Northeast' to 'Northwest' significantly alters the image's atmosphere.

V6 introduces creative upscaling, allowing users to choose between 'upscale subtle' for resolution increase or 'upscale creative' for detailed enhancements.

The presenter notes that V6 seems to favor putting objects in the foreground, adding depth to images.

Version 6 now supports prompting for text within images, with the text encapsulated within quotation marks.

Some commands from previous versions are not yet working in V6, including 'pan', 'zoom', 'very region', 'in painting', 'tuning', and 'describe'.

The presenter shares tips for getting better text results in images, such as waiting the text prompt with quotation marks and a weight, or using all caps for the text.

The presenter encourages experimentation with different prompting ideas and minimal prompts to discover the capabilities of the new model.

The presenter concludes by mentioning they will continue to monitor Mid Journey updates and shares a link to a previous video for further insights.