Mastering Midjourney v5.1 : A Deep Dive into the New Version! (RAW, Seeds, Tips and Tricks!)

Theoretically Media
9 May 202314:41

TLDRIn this video, Tim takes a comprehensive look at the new features and improvements in Midjourney version 5.1, a significant update that introduces two distinct modes within the version. The base 5.1 mode is more stylistically opinionated, resembling the shorter prompting style of version 4, while the Raw mode is a unique feature that offers a more realistic approach to image generation. Tim demonstrates the differences between these modes and previous versions using the same prompts and seeds, highlighting the enhanced sharpness and improved handling of complex subjects like fingers playing instruments. He also discusses the potential of the Permutations feature, which allows users to run the same prompt through different versions simultaneously. Tim concludes by sharing his excitement for the upcoming version 6 and encourages viewers to share their experiences with the new features in the comments section.

Takeaways

  • ๐Ÿ“ˆ **Version 5.1 Update**: Midjourney introduces a significant update with two distinct modes, each offering different stylistic approaches to image generation.
  • ๐Ÿ”„ **Modes within 5.1**: The update includes a more stylistically opinionated model resembling V4's shorter prompting style and a RAW mode that is unique and not simply V5.
  • โš™๏ธ **Settings Control**: Users can switch between different versions, including 5.1 and RAW mode, through the forward slash settings.
  • ๐ŸŒฑ **Understanding Seeds**: Seeds are numerical identifiers that influence the initial pattern of an image, ensuring no two images are identical but related.
  • ๐Ÿ” **Seed Number Retrieval**: The seed number can now be retrieved by adding a reaction to any image using the Envelope Filter, resolving a previous glitch.
  • ๐ŸŽจ **Stylistic Differences**: V5.1 produces warmer, brighter images with less shadow and more volumetric lighting compared to V5.
  • ๐ŸŒŸ **RAW Mode**: Offers a more realistic look with subtle differences in skin tones and is considered a good compromise between stylization and realism.
  • ๐Ÿ“ธ **Photographic vs. Artistic**: V5.1 is more imaginative and expressive, suitable for basic prompts and artistic styles, while RAW mode is better for photographic or cinematic images.
  • ๐ŸŽธ **Improved Finger and Instrument Representation**: Version 5.1 shows more accurate depictions of fingers playing instruments, particularly guitars.
  • ๐Ÿ“š **Sharpness Improvement**: The sharpness of images has improved in V5.1, addressing issues from V5 where images appeared softer.
  • ๐Ÿ”ข **Permutations Feature**: V5.1 enhances the use of permutations, allowing for experimentation with different versions off of the same prompt.
  • ๐ŸŽญ **Illustrative Styles**: For illustrative or artistic styles, V5.1 is preferred, while for more photographic outputs, V5.1 RAW is recommended.

Q & A

  • What is the main focus of the video?

    -The video provides a deep dive into the new Midjourney version 5.1, discussing its features, differences from previous versions, and offering tips and tricks for using it effectively.

  • What are the two different modes within Midjourney version 5.1?

    -The two different modes within Midjourney version 5.1 are the standard 5.1 mode, which is more stylistically opinionated and resembles the prompting style of version four, and the RAW mode, which is distinct and not just a continuation of version five.

  • How can users switch between different versions of Midjourney, including 5.1 and RAW mode?

    -Users can switch between different versions by accessing the forward slash settings, where they can control any of the previous versions as well as 5.1 or RAW mode.

  • What is a seed number in the context of Midjourney?

    -A seed number in Midjourney is a unique identifier for each generated image, functioning as an initial pattern that forms the image in response to a prompt. It ensures that the same image is never generated exactly twice, but variations will be within a similar theme or 'ballpark'.

  • How can users find out their seed number for an image?

    -Users can find out their seed number by adding a reaction to any image using the Envelope Filter, which will then generate a message containing that seed number.

  • What is RAW mode in Midjourney version 5.1?

    -RAW mode in Midjourney version 5.1 is a separate mode that tends to produce more realistic skin tones and less stylized images compared to the standard 5.1 mode. It is considered a good compromise for those who want the creativity of 5.1 without the heavy layer of stylization.

  • Why might a user choose to use version 5.1 over version 5 or RAW mode?

    -A user might choose version 5.1 for its more imaginative and expressive outputs, especially noticeable with basic prompts. It is better suited for those looking for illustrative or playful styles, whereas RAW mode is more suitable for photographic or cinematic images.

  • What is the significance of the sharpness improvement in version 5.1?

    -The improvement in sharpness in version 5.1 is significant because it addresses a weakness in version 5 where images appeared softer. Sharpness is crucial for the quality of the generated images, especially for those seeking a more photographic look.

  • What is permutations and how is it used in Midjourney?

    -Permutations is a feature in Midjourney that allows users to run the same prompt through different versions simultaneously. It's a way to compare and leverage the strengths of each version for a given prompt.

  • How does the speaker suggest using permutations?

    -The speaker suggests using permutations by adding a specific syntax at the end of a prompt, enclosed in curly brackets, to run the same prompt through different versions of Midjourney, such as V4, V5, V5.1, and V5.1 RAW mode.

  • What aesthetic improvements have been noted in version 5.1?

    -Version 5.1 has shown improvements in sharpness and a more accurate representation of fingers playing instruments, like guitars. It also has a better ability to handle basic prompts, producing more stylized and expressive images.

  • What is the speaker's opinion on the future of Midjourney versions?

    -The speaker anticipates that version 6 will be discussed soon and is excited about the prospect of further advancements and improvements in the Midjourney platform.

Outlines

00:00

๐ŸŒŸ Introduction to Mid Journey Version 5.1

The video begins with an introduction to Mid Journey version 5.1, highlighting it as a significant update with two distinct modes: a stylistically opinionated model resembling version four's shorter prompting style and a 'raw mode' that is unique and not simply a continuation of version five. The presenter plans to compare these versions using the same prompt and seed for consistency, explaining the concept of 'seed' as an initial pattern that influences the generated image. The video also addresses a recent update allowing users to retrieve seed numbers through an Envelope Filter reaction on images.

05:00

๐ŸŽจ Comparative Analysis of Mid Journey Versions

The presenter conducts a comparative analysis of Mid Journey versions 5, 5.1, and 5.1 raw using the same prompt and seed. Notably, version 5.1 exhibits warmer skin tones and more vibrant backgrounds, while 5.1 raw offers a more realistic skin tone and less glossiness. The video emphasizes version 5.1's increased creativity and expressiveness, especially with basic prompts, contrasting it with version 5's more standard and photographic output. The discussion also touches on the aesthetic choices between version 4 and 5.1, suggesting that the choice depends on the desired outcome.

10:00

๐Ÿ“ˆ Improvements in Version 5.1 and Exploring Permutations

The video discusses the improved sharpness in version 5.1, a feature that was lacking in version 5. The presenter then explores the least used and often misunderstood feature of Mid Journey: permutations. By using a complex prompt involving different versions (V4, V5, V5.1, and V5.1 raw), the video demonstrates how each version interprets the prompt differently, showcasing their unique strengths. The presenter also shares a personal anecdote about the improved rendering of fingers playing instruments in version 5.1 and mentions an experiment emulating the style of photographer Charles Peterson, finding that version 4 captured a grainy, black and white chaotic look more effectively.

Mindmap

Keywords

๐Ÿ’กMidjourney v5.1

Midjourney v5.1 refers to the latest update of a software or tool that is used for image generation. In the video, it is described as a significant turning point in the tool's history, introducing two different modes within the version and offering more stylistic and expressive outputs compared to previous versions. It is central to the video's theme as the host discusses its features, improvements, and how it compares to other versions.

๐Ÿ’กSeed

A seed in the context of the video is a unique identifier used by the Midjourney bot to generate images. It is likened to television static that forms the initial pattern from which an image evolves. The seed ensures that each generated image is unique, even if the same prompt is used, thus contributing to the diversity and creativity of the image outputs. The host provides an example of how to find the seed number using the Envelope Filter.

๐Ÿ’กRAW Mode

RAW Mode is a specific mode within Midjourney v5.1 that is distinguished for its more realistic and less stylized image outputs. It is presented as a middle ground between the heavily stylized version 5.1 and the more basic version 5, offering a balance between creativity and realism. The host uses RAW Mode to generate images that align more closely with the desired aesthetic, such as in the 'Cabin in the Woods' example.

๐Ÿ’กPermutations

Permutations, in the context of the video, is a feature of Midjourney that allows users to run the same prompt through different versions of the software simultaneously. This feature is considered underused and misunderstood, but the host demonstrates its utility by showcasing how different versions interpret the same prompt, revealing the strengths of each version in handling specific image generation tasks.

๐Ÿ’กIllustrative Styles

Illustrative Styles refer to the artistic and playful image outputs that version 5.1 of Midjourney is capable of producing. The host contrasts these styles with the more photographic or cinematic outputs of other versions. The term is used to highlight the versatility of the tool in generating images that are not just realistic but also imaginative and expressive, as demonstrated in the 'Dungeons and Dragons' example.

๐Ÿ’กCinematic Still

A cinematic still is a single, static image that captures the essence of a scene from a film or has the quality of a film still. In the video, the host uses the term to describe the type of image they are trying to generate with Midjourney, specifically referencing a fictional film by Stanley Kubrick as an example. The cinematic still is an important concept as it sets the creative goal for the image generation process.

๐Ÿ’กVersion 4

Version 4, or V4, is an earlier iteration of the Midjourney tool that is mentioned in the video for comparison purposes. The host notes that V4 was better at capturing a grainy, black and white chaotic look, which was useful for emulating the style of Charles Peterson, a photographer known for documenting the Seattle grunge scene. This comparison helps to illustrate the evolution and differences in the capabilities of various versions of the tool.

๐Ÿ’กSharpness

Sharpness in the context of the video refers to the clarity and detail of the images generated by Midjourney. The host mentions an improvement in sharpness with version 5.1, which was a noticeable issue in version 5. The term is important as it relates to the quality and realism of the generated images, with the host expressing satisfaction that the sharpness has returned to a level comparable to version 4.

๐Ÿ’กCharles Peterson

Charles Peterson is a photographer known for his work in documenting the Seattle grunge scene in the early 1990s. His style is characterized by chaos, movement, motion blur, and a dirty grain, which the host attempts to emulate using Midjourney. The mention of Charles Peterson serves to illustrate the host's experimentation with different styles and the tool's ability to capture various aesthetic qualities.

๐Ÿ’กGuitar Fingers

Guitar Fingers refers to the depiction of fingers on a guitar, which the host notes has been a challenge for Midjourney in previous versions. The host observes an improvement in the accuracy of guitar fingers in version 5.1, suggesting that the tool has become better at rendering detailed and realistic images of hands playing instruments. This term is used to highlight the advancements in the tool's ability to generate more complex and accurate images.

๐Ÿ’กAesthetic Choices

Aesthetic Choices in the video relate to the subjective preferences for different styles of image generation. The host discusses how choosing between version 4 and 5.1 has become more about personal taste in aesthetics rather than objective quality, as both versions produce solid results. This concept is important as it emphasizes the individual creative goals of users and how different versions of Midjourney can cater to those goals.

Highlights

Midjourney version 5.1 is a significant update with two different modes, offering more stylistic variety and control.

Version 5.1 introduces a more stylistic and imaginative approach compared to version 5, with warmer skin tones and more volumetric lighting.

Raw mode in 5.1 is distinct and not just a continuation of version 5, offering a unique style.

Seed numbers in Midjourney determine the initial pattern for image generation, ensuring unique outputs.

The seed number can now be retrieved for upscaled images using the Envelope Filter reaction.

Version 5.1 raw mode provides a more realistic skin tone compared to the smoother, glossier look of version 5.

Version 5.1 is more expressive and imaginative, especially with basic prompts like 'donut', producing vibrant and playful images.

The basic prompt 'happy kids playing with toy robots' resulted in a stock photo-like image from version 5 and a stylized, warmer image from version 5.1.

Raw mode is a good compromise between the heavy stylization of 5.1 and the more photographic look of version 5.

Version 5.1 raw mode is best for specific image ideas and longer prompts, balancing creativity with realism.

Additional training in 5.1 has improved the rendering of fingers and instruments, particularly guitars.

Midjourney V4 captured a grainy, black and white chaotic look better than V5, making it a good choice for certain aesthetic styles.

Sharpness has improved in version 5.1, particularly noticeable in paper quilling type images.

Permutations, a lesser-used feature of Midjourney, allows running the same prompt through different versions simultaneously.

Version 5.1 is ideal for illustrative or artistic styles, while 5.1 raw is better for photographic or cinematic images.

Experimentation with version 5.1 and 5.1 raw continues, with user feedback and discoveries expected to shape future updates.

The upcoming version 6 of Midjourney is anticipated to bring further enhancements and new features.