🖌️ AI Art Tips Thursday #5: Let's plant some seeds...

💡 AI Art Tips for the Day

By popular demand, we are going to explore the seed parameters within Midjourney this week.

In my personal understanding before the exploration, I thought the seed is one of the key things, or even maybe the main variable, that drives the result. If you use the same seed with the same prompt, you are most likely to get the same result. But I am not too sure about how a similar prompt with the same seed could work.

As always, if you are after a detailed explanation of this and some examples about Seed, you can check Midjourney’s documents.

First of all, you might have the question, how do I get the seed of the output as that is nowhere can be seen in the output message delivered by Midjourney. Well, the simple trick is an emoji reaction. You need to react to the output image with an ✉️ emoji.

Once that is done, you would receive a private message from the Mdijourney bot that shows you the seed from the output.

Alright, so now we have the seed, let’s try some examples with it and sow this seed of creativity 😉

I must say I can see the style and composition across the different images are quite similar. Especially the first pic from both Spring and Summer looks like a good blend between photo and oil painting which is nice.

Now, let’s try with some faces.

When faces are introduced, I can see if the prompt is not exactly the same, you are probably not going to see some differences. Although, I feel the emotion is somehow quite consistent. This is actually a question I have, why do they all look a bit sad? Is it some sort of bias from midjourney towards listening to music or teenagers? 🤔 

Anyway, I then also tested how the seed work if the prompt is the same. Well, the result is pretty aligned with my expectation. See them below 👇️ 

As you can see the two images, are almost identical. If it is not because of that easily-ignored white line on that teenager’s hand in the second image, I would assume both images would be the same. I have enlarged it so that you can check as well, but you can see the wonky texts are also different. So I guess the same prompt with the same seed would probably give you things that are close but not the same.

Cool, so in conclusion, this is what I found with playing with the same seed:

• Different prompts with the same seed: similar style and image composition, but the details won’t be the same

• The same prompts with the same seed: style, composition and details would almost be the same, just some really light differences in the details

• Is seed the “secret weapon” to create consistent character across different settings or scenarios?: Not really, you can find some close images but not exactly the same though

Try this in your next prompt

{your prompt describing the photo} --seed {whole number between 0 and 4294967295}

🔤 Word of the day

Double exposure

n. is a photographic technique in which two images are captured and combined into a single image, according to Wikipedia.

This is also commonly used in AI art creation as well. See an example below:

🖼️ Image of the day

I don’t know, just really like this image here. May even want to print it out and hang it in my room. Also, you can just see how a simple prompt sometimes really works wonders.

💕 Share the love, get some love!

I am doing a weekly draw to get people referring the newsletter to others with some cool avatars.

If you can refer at least 3 of your friends, you will be in this weekly draw of 3 to get 3 cool avatars using your chosen portrait with the styles of your choice.

So if you enjoy this newsletter and want to get some cool avatars for your socials, here is your link to share your love of the Prompt Craftsman with your friends. You are the best! ❤️