Midjourney V6 vs DALLE-3 Text Prompt COMPARISON (Print on Demand Ai Art Faceoff)

Philip Anders
22 Dec 202319:19

TLDRIn a detailed comparison, the video presents a face-off between Midjourney V6 and DALLE-3, two AI art generators, using nine different prompts for print-on-demand designs. The evaluation criteria include the quality of graphics, suitability for print, and accuracy of text representation in the images. Midjourney V6 is noted for its artistic style and vibrant colors, but sometimes strays from the prompt's instructions and struggles with text clarity and spelling. DALLE-3, on the other hand, adheres more closely to the prompts, particularly in maintaining a flat design style suitable for print-on-demand, and demonstrates better text accuracy. The video concludes with a personal preference for DALLE-3 due to its simplicity and suitability for the print-on-demand market, despite acknowledging Midjourney V6's quick generation capabilities and unique artistic outputs. Viewers are encouraged to keep score and share their preferences in the comments.

Takeaways

  • 🎨 **Midjourney V6 vs DALLE-3 Comparison**: A face-off between two AI art generators for print on demand prompts.
  • 🖌️ **Graphic Quality**: Both tools were evaluated on the quality of graphics, suitability for print on demand, and text accuracy.
  • 🧀 **Cheese Prompt**: Midjourney favored a more artistic style but deviated from the 'cartoon' prompt, whereas DALLE-3 adhered closer to the prompt but had some spelling errors.
  • 🏋️‍♂️ **Koala Prompt**: Midjourney produced a more detailed image, while DALLE-3's cartoonish and flatter design was more aligned with the prompt but had spelling mistakes.
  • 👩‍🦱 **Messy Bun Prompt**: DALLE-3 provided a flatter and more cartoonish style, with correct spelling, which was preferred for print on demand.
  • 🍕 **Pizza and Beer Prompt**: Midjourney's design was more aesthetically pleasing and correctly spelled, despite not following the 'cartoon' style as closely as DALLE-3.
  • 🦉 **Owl Prompt**: Midjourney captured a vintage feel, but DALLE-3's colorful and symmetrical design was deemed more suitable for print on demand with correct spelling.
  • 🌊 **Surfboard Prompt**: DALLE-3's vintage distressed graphic was more in line with popular print on demand styles, despite some spelling inaccuracies.
  • 🥑 **Avocado Prompt**: DALLE-3's design was more realistic and suitable for print on demand, with correct spelling and easier background removal.
  • 🌈 **Vaporwave Prompt**: Midjourney's detailed raccoon was hard to work with for print on demand, while DALLE-3's simpler design was preferred.
  • ✍️ **Text Handling**: DALLE-3 was more accurate with text, making it easier to read and better suited for print on demand applications.

Q & A

  • What was the purpose of the face-off between Midjourney V6 and DALLE-3?

    -The face-off was conducted to compare the two AI tools based on their generated graphics for print on demand, focusing on the quality of the graphics, their suitability for print on demand, and the accuracy of the text within the images.

  • How did the reviewer conduct the face-off?

    -The reviewer gave both Midjourney V6 and DALLE-3 the same prompt, ran it multiple times, and then selected the top two results from each tool. These results were then compared side by side in Adobe Illustrator.

  • What was the first prompt given to both AI tools?

    -The first prompt was to create a cartoon-style image with various types of cheese characters and the phrase 'cheese the day, carp brium', with the cheeses depicted in a lively and playful manner on a wide, isolated background.

  • Which tool followed the prompt more closely in terms of style for the first round?

    -DALLE-3 stayed closer to the cartoon style prompt, while Midjourney V6's style was more artistic and less cartoon-like.

  • What was the issue with the text spelling in the first round for Midjourney V6?

    -Midjourney V6 had issues with spelling the phrase correctly. The phrase 'cheese the day' was misspelled as 'cheese the da' and 'cheese the day with three e', and additional unrelated words were included.

  • What was the reviewer's opinion on the text spelling accuracy for DALLE-3 in the first round?

    -The reviewer was impressed with DALLE-3's text spelling accuracy, as it correctly spelled out the long and challenging phrase.

  • Which tool won the first round according to the reviewer's opinion?

    -DALLE-3 won the first round because it followed the prompt better and had correct text spelling.

  • What was the main difference between the graphics generated by Midjourney V6 and DALLE-3 in the second round?

    -In the second round, Midjourney V6 generated graphics that were more detailed and less flat, while DALLE-3 produced a more stylized and flat design that better matched the prompt.

  • What was the issue with the text in the second round for DALLE-3?

    -DALLE-3 failed to spell the text 'quality gains' correctly in the second round, which was a downside despite the better adherence to the flat design style.

  • Which tool won the second round according to the reviewer?

    -Midjourney V6 won the second round because it managed to spell the text correctly and produced appealing graphics, despite not fully adhering to the flat design prompt.

  • What was the reviewer's suggestion for viewers to do while watching the video?

    -The reviewer encouraged viewers to keep score as they watch the video, and to share their opinions on which tool performed better after each round by commenting at the end of the video.

  • What was the main takeaway from the face-off according to the reviewer?

    -The main takeaway was that while both tools performed well, DALLE-3 was slightly more suitable for print on demand due to its better adherence to prompts and text accuracy, although Midjourney V6 also has its strengths and can still be used effectively.

Outlines

00:00

🎨 Graphic Design Tool Comparison: Mid Journey vs. D E3

The video script details a comparison between two graphic design tools, Mid Journey V6 and D E3, across nine different prompts. The evaluation criteria include the quality of graphics, suitability for print on demand, and accuracy of text spelling. The author conducted tests by giving both tools the same prompts and selecting the top two results from each. These were then compared side by side in Adobe Illustrator. The first prompt involved creating cartoon-style images with various types of cheese characters and a specific phrase. Mid Journey's results were more artistic but less adherent to the cartoon style prompt, while D E3's results were more playful and closer to the prompt, with better text spelling accuracy. The author encourages viewers to keep score and share their preferences in the comments.

05:00

📈 Results Analysis and Viewer Engagement

The video script continues with a detailed analysis of the graphic results from both tools against various prompts. It discusses the adherence to prompt instructions, the aesthetic appeal of the graphics, and the accuracy of text spelling. The author highlights the strengths and weaknesses of each tool in different rounds, such as Mid Journey's detailed and artistic style versus D E3's flatter and more playful approach. The script emphasizes the importance of text accuracy and legibility, especially for print on demand designs. The author also encourages viewer participation by asking them to keep score and share their opinions on which tool performed better in each round.

10:01

🔍 Close Examination of Design Details

The script delves into the specifics of each design, noting the differences in style, color, and detail that set the tools apart. It points out instances where Mid Journey provided more detailed and textured graphics, while D E3 offered cleaner, more symmetrical designs that may be easier to work with for print on demand. The author discusses the challenges of text spelling and legibility, noting that while Mid Journey sometimes produced smaller and harder-to-read text, D E3 generally provided more accurate and readable text. The script also touches on the practical aspects of print on demand, such as the ease of removing backgrounds and the suitability of designs for t-shirt printing.

15:02

🏆 Final Round and Overall Conclusion

The final paragraph of the script wraps up the comparison with the ninth prompt, which asks for a vintage distressed graphic design featuring a surfboard. The author discusses the different interpretations of the prompt by each tool and evaluates the results based on suitability for print on demand. While both tools have their merits, the author concludes that D E3 is slightly more suitable for this purpose due to its style and text accuracy. However, the author acknowledges that Mid Journey has improved text capabilities and can generate graphics quickly, despite some issues with text size and legibility. The script ends with a call to action for viewers to share their scores and preferences in the comments, and to check out additional resources for using Mid Journey V6.

Mindmap

Keywords

Midjourney V6

Midjourney V6 refers to a specific version of an AI art generation tool that is being compared in the video. It is one of the two primary subjects of the face-off, where its performance in generating graphics based on text prompts is evaluated. The video discusses its output quality, adherence to prompts, and suitability for print on demand.

DALLE-3

DALLE-3 is another AI art generation tool that is compared alongside Midjourney V6 in the video. It represents the second party in the 'face-off,' where its graphic outputs are analyzed for quality, prompt adherence, and print suitability. The tool's performance is critiqued based on the same criteria as Midjourney V6.

Print on Demand (Print onand)

Print on Demand (referred to as 'Print onand' in the transcript, likely due to a typographical error) is a business model where products are not printed until an order has been received, reducing the need for inventory. In the context of the video, it is the intended application for the graphics generated by the AI tools, which influences the evaluation criteria.

Text Prompt

A text prompt is a textual description or request given to an AI art generation tool to guide the creation of an image. In the video, text prompts are used to instruct both Midjourney V6 and DALLE-3 to generate specific graphics, and the accuracy and creativity of the resulting images are compared.

Graphic Quality

Graphic quality refers to the visual appeal, detail, and overall aesthetic of the images generated by the AI tools. The video scrutinizes how well each tool's graphics look, particularly in the context of print on demand, where factors like color, detail, and style are critical.

Adherence to Prompt

Adherence to prompt describes how closely the AI-generated graphics follow the instructions provided in the text prompt. The video assesses whether the tools create images that match the description and style requested, which is a key aspect of their performance evaluation.

Spelling Accuracy

Spelling accuracy is the correctness of the text within the generated graphics. The video highlights the importance of spelling, especially for print on demand applications, where a mistake can diminish the product's appeal. Both AI tools are judged on their ability to spell words and phrases accurately.

Illustrator

Adobe Illustrator is a vector graphics editing software mentioned in the video where the top results from both AI tools are pulled and compared side by side. It is used for creating and editing vector images, which is relevant to the process of evaluating and preparing graphics for print on demand.

Cartoon Style

Cartoon style is a specific aesthetic mentioned in one of the text prompts, where the AI is asked to generate images with a playful and lively approach typical of cartoons. The video discusses whether the AI tools successfully capture this style in their graphics.

Vintage Sunset

Vintage Sunset refers to a design theme that is popular in print on demand, characterized by a retro, sundown aesthetic. The video compares how well each AI tool interprets and generates graphics that fit this theme, which is important for their suitability in the print on demand market.

Text Layout

Text layout pertains to the arrangement and presentation of text within a graphic. The video evaluates how effectively the AI tools incorporate text into their images, considering factors like legibility, size, and visual appeal, which are important for the print on demand context.

Highlights

Midjourney V6 and DALLE-3 were compared in a face-off using nine different print on demand prompts.

The comparison focused on graphic quality, suitability for print on demand, and text accuracy within the images.

Both tools were given the same prompts and the top two results from each were selected for comparison.

Midjourney V6 favored a more artistic style, while DALLE-3 adhered closer to the prompts, especially regarding cartoon styles.

DALLE-3 demonstrated better text spelling accuracy, particularly with longer and challenging phrases.

Midjourney V6 sometimes resulted in smaller and harder to read text, whereas DALLE-3 provided clearer and more readable fonts.

For print on demand suitability, DALLE-3's simpler and cleaner designs were often more favored.

Midjourney V6 had a tendency to produce more detailed graphics, which were not always ideal for print on demand applications.

The video encourages viewers to keep score and share their preferences in the comments for a more interactive comparison.

In the final round, DALLE-3 was slightly favored for its vintage distressed graphic style suitable for print on demand.

The reviewer noted that while DALLE-3 was more suitable for print on demand, Midjourney V6 had the advantage of generating graphics quickly.

Midjourney V6's detailed and artistic style was appreciated but could be less effective for certain print on demand markets.

DALLE-3's performance was slightly better in terms of adhering to the prompts and producing suitable print on demand designs.

The reviewer concluded that DALLE-3 won the comparison, but encouraged viewers to share their opinions for a broader perspective.

The video also provides a prompting guide for using Midjourney V6 to generate AI graphics with text.