Gotta say: swirling vortexes .... everything is downright gorgeous! I fell in love with the stairs! I am going to have to try that in my prompts. Li-core ... lion is beautiful ... the mushrooms sort of remind me of fairy houses. Things like architecture and portraits are breath-taking. Honestly, I fell in love with aspects of each, so can't wait to try them. Thanks so much for these wonderful, informative videos!!
How are you getting so much variety like in your vortex wildlife? Is this one grid or did you choose your favorites from several grids? Mine just keeps giving me a fox.
It's just one grid. I guess it just prefers foxes and it varies across different seeds. You could try increasing the chaos value or if you have a certain animal in mind, you could try to combine the vortex with the name of that animal.
Hi ! I love the zoom intro and conclusion, but I always find new words in the other formula and I miss it. Do as you prefer, don't worry for me. Il always found something to try. Thx to you and GG for 8k. You deserve it (sorry for my bad english, I'm french)
Yep, I wanted to make a smoother zoom out though like in the intro, but wanted to do it with 2x zoom because it changes the image more 1.5x zoom does, but this needs some more experimenting.
Love your videos, very inspirational, I am intrigued by --seed 777, what does it represent, is it a particular image, a random image, it's very mysterious
Thank you. There is nothing mysterious about it, it's just a random number that I selected and specified for my prompts instead of relying on it being anything between 0 and over 4 billion. You don't have to set it at all like I have unless you want to get exactly the same images as I have from your results. Having it the same value when testing things out (like I have) makes it easier to compare different things as I know that all the differences in the images come from only the prompt text.
Don't be confused about it, you don't have to use it, it's just a random number I have selected so it is not any of the other 4 billion ones that is assigned to the grid generation. I wanted to make it possible for people to be able to generate the exact same images if they wished to so providing the seed number I used makes that possible under certain circumstances. Otherwise I would have to look it up separately each time somebody asked for it.And also because I like to compare things, this removes one layer of randomness from the prompt so I am able to compare just what the words of the prompt do.
@@thaeyne Ooh thank you for this explanation. So, if I understand, you can use any # (ie: 123) and if you use it for each generation, it will somehow match your previous prompts? Or, maybe you have to find the seed number from your original prompt, then keep using that #? Sorry if I'm being ignorant, trying to figure this seed thing out. Thank you for your time!
@@mocachip1 Yes, that's basically it. There is a number which is called the seed and usually Midjourney selects a random number out of over 4 billion numbers. The seed number combined with all of the text in your prompt should result in the exact same image grid if all conditions are met and the model has not been changed. You can look up the seed for each image grid later or you can set it in place during prompting so you know exactly what it is without looking it up. I hope this helps to clarify it. The most important thing is that you don't need to set the seed or have to look it up at all unless you want to create the exact same image as someone else and that someone has provided you with the exact prompt they used and the exact seed they used. Otherwise just ignore the seed number.
Thank you for your honest opinion. It is my hobby to experiment around with all things AI and I happen to love the surprises that come with it, but I understand that not everyone shares my enthusiasm for it.
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
Saw today this channel recommended on a Facebook post. So hopefully some more subs coming soon.
Great explorations. Thank you for sharing! I hope/trust you are having fun!!
Love the outro!
Gotta say: swirling vortexes .... everything is downright gorgeous! I fell in love with the stairs! I am going to have to try that in my prompts.
Li-core ... lion is beautiful ... the mushrooms sort of remind me of fairy houses. Things like architecture and portraits are breath-taking.
Honestly, I fell in love with aspects of each, so can't wait to try them. Thanks so much for these wonderful, informative videos!!
thank you for this! I’m a great fan of your channel btw🐁
Some of these were quite surprising. I like the way you use the same subjects. It makes it easier to compare
Li Tobler was H. R. Giger's girlfriend who was the model used in a lot of his pictures.
Beautiful, thanks!
Fantastic. Thank you! 💯
Thank you very much, your video’s are very good, I’am a huge fan of you❤❤
Thank you so much 😀
Still I don't recognise what is the seed 777 that always exists?😢
Just a random number I use so if you wanted, you could replicate the exact same image (depending on circumstances).
The last 30 Second is amazing, can you tell me please how you create those pictures?
1:26 I like the firefox logo on wildlife #1. 😏🦊
Awesome prompts and examples. Curious how you morphed the zoomed out images together.
Thanks!
For the zoomed out images I used image interpolation from RunwayML.
How are you getting so much variety like in your vortex wildlife? Is this one grid or did you choose your favorites from several grids? Mine just keeps giving me a fox.
It's just one grid. I guess it just prefers foxes and it varies across different seeds. You could try increasing the chaos value or if you have a certain animal in mind, you could try to combine the vortex with the name of that animal.
Hi ! I love the zoom intro and conclusion, but I always find new words in the other formula and I miss it. Do as you prefer, don't worry for me. Il always found something to try. Thx to you and GG for 8k. You deserve it (sorry for my bad english, I'm french)
Strong words again.. Loving this format with the shorter video's. the outro was made with zoom out over zoom out? 👍
Yep, I wanted to make a smoother zoom out though like in the intro, but wanted to do it with 2x zoom because it changes the image more 1.5x zoom does, but this needs some more experimenting.
Awesome tip$! Curious Thaeyne how you made the opening zoom out video images blend together??
Thanks! I used the RunwayML image interpolation for that, but I guess any other image interpolation tool will work for that.
Love your videos, very inspirational, I am intrigued by --seed 777, what does it represent, is it a particular image, a random image, it's very mysterious
Thank you. There is nothing mysterious about it, it's just a random number that I selected and specified for my prompts instead of relying on it being anything between 0 and over 4 billion. You don't have to set it at all like I have unless you want to get exactly the same images as I have from your results. Having it the same value when testing things out (like I have) makes it easier to compare different things as I know that all the differences in the images come from only the prompt text.
many thanks for your explanation @@thaeyne
Very cool channel - but, why always seed 777? I'm confused?
Don't be confused about it, you don't have to use it, it's just a random number I have selected so it is not any of the other 4 billion ones that is assigned to the grid generation. I wanted to make it possible for people to be able to generate the exact same images if they wished to so providing the seed number I used makes that possible under certain circumstances. Otherwise I would have to look it up separately each time somebody asked for it.And also because I like to compare things, this removes one layer of randomness from the prompt so I am able to compare just what the words of the prompt do.
@@thaeyne Ooh thank you for this explanation. So, if I understand, you can use any # (ie: 123) and if you use it for each generation, it will somehow match your previous prompts? Or, maybe you have to find the seed number from your original prompt, then keep using that #? Sorry if I'm being ignorant, trying to figure this seed thing out. Thank you for your time!
@@mocachip1 Yes, that's basically it. There is a number which is called the seed and usually Midjourney selects a random number out of over 4 billion numbers. The seed number combined with all of the text in your prompt should result in the exact same image grid if all conditions are met and the model has not been changed. You can look up the seed for each image grid later or you can set it in place during prompting so you know exactly what it is without looking it up. I hope this helps to clarify it. The most important thing is that you don't need to set the seed or have to look it up at all unless you want to create the exact same image as someone else and that someone has provided you with the exact prompt they used and the exact seed they used. Otherwise just ignore the seed number.
Salvador Dali thought of this many years ago
Van Gok made me laugh 😂
good vid but I can't stand the computer voice.
Thanks, but my own voice sounds worse.
@@thaeyneI love this computer voice!
@@thaeyne Which text-to-speech system or service are you using?
AI is garbage art period good luck with it. It's crap btw.
Thank you for your honest opinion. It is my hobby to experiment around with all things AI and I happen to love the surprises that come with it, but I understand that not everyone shares my enthusiasm for it.