A Perfect Midjourney Prompt Formula (Great for Beginners or Advanced Users)
TLDRIn this informative video, the host delves into the art of crafting effective prompts for mid-journey image generation. He introduces a structured framework consisting of medium, style, composition, scene, modulate, and dash-dash parameters, emphasizing the importance of brevity in prompts. The host demonstrates how varying the medium can drastically alter the output, from photography to painting and even to comic book illustrations. He also discusses the impact of style and composition on the final image, using examples like Pixar and Tim Burton styles. The scene section is crucial, encompassing subject, action, props, and location. The modulate section allows for atmospheric effects, while the dash-dash section offers advanced commands like chaos for varied outputs. The video concludes with a teaser for a future video on the dash-dash commands and an invitation for viewers to join the conversation on Discord.
Takeaways
- 🎨 **Medium Matters**: Changing the medium (e.g., photo, painting, TV show) significantly alters the output image's style and mood.
- 🖌️ **Style & Composition**: Specifying a style (like Pixar or Tim Burton) or composition (various camera angles) can guide the image towards a desired aesthetic.
- 🚀 **Experimentation Encouraged**: There's no right or wrong way to prompt; experimentation with different mediums and styles is key to finding the perfect output.
- 📏 **Token Limitation**: Mid-journey prompts are limited to about 77 tokens, emphasizing the importance of brevity and concise prompts.
- ⛓️ **Cascade Information**: Structuring prompts in a cascading manner helps large language models parse information more effectively.
- 🌐 **Cross-Platform Usability**: The prompt framework is not limited to mid-journey but can be applied across various image generators.
- 👔 **Art Direction**: Small details like a blue color palette or era-appropriate clothing can add an artistic touch to the generated images.
- 🎭 **Emotional Actions**: Directing characters in emotive actions or describing their emotional state can prevent the common back-to-camera composition.
- 🌆 **Scene Manipulation**: Altering elements within the scene, such as the subject, action, props, or location, can drastically change the narrative and tone of the image.
- 🌡️ **Modulate for Atmosphere**: Adding atmospheric effects (lighting, fog, weather) or seasonal changes can create unique and unexpected visuals.
- 🔄 **Chaos for Creativity**: Using the chaos command (-C) introduces variability, which can be useful for world-building and generating diverse image sets.
- 📄 **Additional Resources**: Further information, including a list of camera angles and a PDF guide, is available for those interested in deepening their prompting skills.
Q & A
What is the significance of using a framework for prompting in mid-journey?
-The framework for prompting in mid-journey provides more direction and control over the output, ensuring that the generated images align more closely with the user's expectations.
Why is brevity important when creating prompts for mid-journey?
-Brevity is key because mid-journey prompts are limited to about 77 tokens, and shorter prompts allow for more effective communication with the language model.
How does the medium section of the prompt framework influence the generated image?
-The medium section determines the style of the image, such as a photograph, painting, or comic book illustration, which significantly affects the final output's appearance.
What is the role of the 'style' section in the prompt framework?
-The 'style' section allows users to specify a particular artistic style or the work of a specific artist, which can influence the overall look and feel of the generated image.
How can the 'composition' section of the prompt be used to direct the camera angle or shot in the generated image?
-By specifying terms like 'long shot', 'close-up', or 'satellite view', users can guide mid-journey to use specific camera angles or shots in the image generation process.
What is the purpose of the 'scene' section in the prompt framework?
-The 'scene' section encompasses the subject, action, props, and location, allowing users to describe the setting and actions within the generated image in detail.
How does the 'modulate' section affect the atmospheric effects of the generated image?
-The 'modulate' section introduces atmospheric effects such as lighting, fog, weather, or time of day, which can dramatically change the tone and mood of the image.
What is the 'chaos' command in the dash dash section, and how is it used?
-The 'chaos' command, invoked by 'dash dash C' followed by a number between 0 and 100, introduces variability to the initial seed images, creating a set of diverse images that can be useful for world-building and storytelling.
Why might a user need to experiment with different 'medium' keywords in the prompt?
-Experimenting with different 'medium' keywords allows users to explore a wide range of options and find the most suitable style for their desired output, even if the exact medium isn't traditionally recognized.
How does specifying an action or emotive state in the 'scene' section help to avoid the 'bullseye composition' issue?
-By directing the character in an emotive action or describing the character's eyes as 'sad' or 'happy', users can signal to mid-journey that they want to see the character's face, thus avoiding the common issue of the character having their back to the camera.
What is the significance of the 'aspect ratio' in the prompt, and how does it influence the generated image?
-The aspect ratio determines the proportional relationship between the width and height of the generated image, which can affect the composition and the way the scene is framed.
What is the 'coolish off effect' mentioned in the script, and how does it relate to the 'chaos' command?
-The 'coolish off effect' is a film editing technique that suggests our brains create meaning when we see two back-to-back shots. It relates to the 'chaos' command as the varied images generated can be seen in sequence, prompting the viewer to imagine a story.
Outlines
🎨 Mid-Journey Prompting Framework Overview
The video begins with an introduction to a framework for prompting in mid-journey, suitable for users of all levels. It emphasizes that there's no one-size-fits-all approach to prompting, but the framework aims to provide more direction and control over the output. The framework is structured into sections: medium, style, composition, scene, modulate, and parameters (dash-dash). The video explains that shorter prompts are preferable due to the token limit in mid-journey and how the prompt formula helps the AI parse the information more effectively. The medium section is explored first, demonstrating how changing the medium from a photograph to a painting or a 1960s TV show drastically alters the output.
📸 Exploring Style, Composition, and Scene Elements
The video delves into the style and composition section, discussing how style is closely linked to the medium and can help achieve a specific artistic look. It provides examples of how different styles, such as Pixar or Tim Burton, can influence the output, with the caveat that specifying an artist doesn't guarantee a particular outcome. The composition and shot section covers camera angles and how they direct the AI, with a cautionary note about the satellite view due to potential scale issues. The scene section is broader, encompassing the subject, action, props, and location, and demonstrates how manipulating these elements can yield vastly different images. The video also touches on directing character actions and poses to avoid common AI-generated compositions and the use of emotional descriptors to influence the character's expression.
🌐 Modulation and Chaotic Creativity
The video moves on to the modulate section, which involves atmospheric effects like lighting, fog, weather, and time of day. It shows how adding seasonal changes to a cyberpunk theme can produce unique and unexpected visuals. The dash-dash section is briefly mentioned, with a promise of a dedicated video to explore its many commands. However, the video highlights the 'chaos' command, which introduces variability to the images by breaking up the initial seed images. This chaos command, when set to 100, generates a wide range of images that can stimulate storytelling and world-building. The video concludes with an invitation to subscribe for more in-depth exploration of the dash-dash commands and an encouragement to experiment and have fun with mid-journey.
Mindmap
Keywords
Prompting
Mid-journey
Framework
Medium
Style
Composition
Scene
Modulate
Parameters
Chaos Command
Tokens
Highlights
A framework for prompting in mid-journey is introduced, suitable for both beginners and advanced users.
There is no right or wrong way to prompt in mid-journey, but the framework offers more direction and control.
The framework's structure includes medium, style, composition, scene, modulate, and dash-dash parameters.
Brevity is key in mid-journey prompting, with prompts limited to about 77 tokens.
The prompt formula cascades information to mid-journey, making it easier for the model to parse.
Medium can drastically change the output, such as switching from a photograph to a painting or a 1960s TV show.
Style is linked to medium and can help in achieving a specific look, like Pixar or Tim Burton styles.
Calling out an artist or style does not guarantee cooperation from mid-journey, as seen with the Tim Burton example.
Camera angles and shots can be directed using terms like 'long shot', 'close-up', and 'satellite view'.
The scene section includes subject, action, props, and location, which can be manipulated for different results.
Action and character poses can be directed to avoid the typical 'back to camera' composition.
Experimenting with the 'Style by' parameter can lead to imaginative and unique results.
Modulate section allows for atmospheric effects like lighting, fog, weather, or time of day to be added.
The chaos command (-C) can be used to create varied images, useful for world-building.
The dash-dash section contains many commands, with a separate video planned to explore them in detail.
The framework is not limited to mid-journey and can be applied across various image generators.
A PDF with camera angles and additional information is available for free on Gumroad, with donations appreciated.