The best tutorial I've seen yet. Too many Tubers are focused on themselves and shoving their face in the camera for likes and subs... They are promoting themselves more than the subject of the video. I dislike them as in Thumbs down... This video is very to the point with Very helpful info.... Thumbs up and +1 sub....
This was hands down the best video I've seen for MJ and cinematic styles. I've been at ai for a time now but am just diving into the cinematic side. Thanks again!
Thanks for this great tutorial! I'm working with a company that sells watches and wallets. We use midjourney for some shots and i struggled a bit to create good images. After watching your tutorials, the results are much more realistic. i can't thank you enough man
That other guy said it, and I agree enough to repeat it... best delivery of info, examples and straight forward explanation I've seen in the genre thus far, many thanks.
Your video just popped up to me while scrolling, and Dudeee, your content is amazing. You just didn’t waste time as other youtubers usually do! Every minute i spent watching the vid, i learned something to add to my knowledge. Thank you man and wish you the best❤ Edit: and i subbed of course!
Como eu não achei o seu canal antes!? Eu quase choro com esse vídeo porque o que estou "apanhando" para fazer exatamente a cena que eu quero e de repente você entrega tudo isso assim, "mastigado", para facilitar a nossa compreensão. Me senti agora sendo "alfabetizada"😅 apta a descobrir um mundo totalmente novo. Muito obrigada mesmo 🙌
Wow. I was playing with the prompt for Midjourney but was never able to really get the result that i wanted. These simple and easy tricks that you are giving are incredible in how effective they are. Thank you so much ! ❤
You're welcome! Midjourney can be tricky to prompt in sometimes, so it's always good to know the types of prompt keys you need to get what you have in mind.
Excellent; one of the more high quality prompt instructions. Interesting way to weight each characteristic (by description order flow). This is also achieved by numeric weight descriptors after a prompt word, like ::0.5 or . I think Mid Journey recognizes the former.
Thanks! Weight descriptors will definitely work as well. It seems that in Midjourney version 6, it can be hard to get all the exact elements, especially when it comes to certain camera angles since the model likes to focus on faces quite a bit. It's important to try and control each element of the image separately.
Simply incredible! Thank you very much for your dedicated work. It made my understanding of prompt construction so much easier! I will keep an eye on all your content! I wish you all the success in the world!
Thank you so very much for creating and sharing this video. It’s given me more clarity on the photographic possibilities than any other Midjourney video I’ve watched - and I watch A LOT. 🙏🏼
Thank you for this great tutorial. I watched it about 10 times now. Can you make tutorial about consistent characters in mid journey because this is something I am still struggling a lot.
btw the thing with the order of words, for example naming the colors, has changed with 6.1!i got the same results and right colors, no matter which order i described the outfit. cool thing though!
@@taoprompts they do, indeed! i am recording a biiiiiig tutorial for a marketing agency and mentioned this thing, too, but then noticed afterwards, that this problem doesn´t exist anymore, haha. but btw, even if some tutorials of you are a bit older, most of it didn´t lose it´s meaning in mj :D
Thank you! For short films you can try Pika Labs or RunwayML. Here's an old tutorial I made, I will post an updated version soon. th-cam.com/video/8awNnoot_qg/w-d-xo.html
As important as camera position, it's the LIGHT placement, please tell me how to place the Sun on an image. how to tell that I want the Sun at my back to use Sunlight without seeing the Sun. Example Hard Test: Try to do an image of a face illuminated by a Sunset, and we are NOT seeing the Sun, nor the sunset, the Sun it's the source of light at that time of the day at the beach for example, but I don't want to see the Sun, just the person with frontal light coming from the Sun.
This channel has been so helpful in my own growth with ai art! I am actually using Tensor Art but the information you give naturally translates across all these platforms with some minor tweaking. The particulars of prompt structure itself is the most important, along with camera angles etc. Thankyou and I can see your content has blossomed over time into these very well crafted presentations and so I not only thank but must also congratulate you on your own growth too!
Thanks I appreciate that! I haven't tried Tensor Art but I have used SDWebui which is similar and the control you get is really nice. The different platforms do seem to have a lot of similarities when it comes to prompting.
Great guide. Thanks to it, I'm doing better and better things. Thanks a lot 🙏I'm starting my journey with midjurney. I work as a tattoo artist and generate references in it. I'm creating better characters, but I still struggle with generating realistically looking skulls. They all seem toy-like, even when following your tips. Do you have any advice?
Thank you. Maybe try a different location where you expect to see a real skull: "photo of a human skull in a museum" "Stock photo of a human skull with a plain white background" I not sure exactly what a skull is supposed to look like, but they look realistic to me.
Hey, I tried your exact prompt at 3:47 and got nowhere near the girl with the different angles and photo quality you got with the 6 images split. Was there more to your prompt or image references? Thank you
Your work is amazing, can you please simplify the free PDF access. As your link is taking me to website and I've to pay something. I dont have card. Thanks for your support, I've already subscribed your channel.
I haven't tried that yet. In my experience the /shorten command frequently misinterprets your original prompt. It will shorten phrases, but then the shortened keywords mean something different to the Ai from your original prompt.
How do you make the background not clear in closeup photos of a person. For instance if I had a prompt like "cute girl hiking in the mountains," the mountains come out blurry in the background. how do you fix the blurry background?
That's a tricky one. You could try using the words "deep focus photography" which in real life will sharpen the background, but it doesn't always work well. The opposite would be "shallow focus" example: "deep focus photo, a woman at the arches national park, noon, insta x"
@@taoprompts thank you. I figured it out. I used *"high depth of field, every detail of the valley is crystal clear, shot with a DSLR 4k, Camera Lens 24mm, f/16"*
❤❤ hi. ❤MJ releasing model personalization. Get at least 200 image ratings and then add --p after your prompt. Personalization completely changes the way system interprets your prompts.
i want a better way to create image to image with 100% same face. I tried with several ways but failed. I used my own photo instead, but cannot. My purpose is to create 64 or many photos with same face or image of different post or style in order to produce a video later. Please advice, tks.
Hey, there's a new consistent characters feature which makes that possible. I have provided an update tutorial here: th-cam.com/video/Vi5KQUZSKkM/w-d-xo.html
amazing tutorial. I have some prompts I just cant figure out. Hoping you or the community can help me out. How do I replicate a realistic photo of a person that makes it look like it was taken on an iphone? I also want to be able to do this with the only lighting coming from the camera flash. I can't seem to get it right
Ahhhhh! You cracked the code. Some of my problems stem from word order. The other topics were awesome reminders and gave me a fresh perspective on my prompting. This video rocks. Thank you once again!
You're welcome! I've had a ton of issues with word order too. Idk if that's something they can even fix with the ai model, but you can always work around it. Always happy to help!
@@stylewithaudra5570 The Midjourney reddit forum is actually a decent place to get some exposure. Instagram is more saturated, but I've seen accounts use Ai images combined with an Ai motion tool like leiapix, with simple capcut effects to create reels which seems to do pretty well. exnihilo.creations is an instagram account like that.
thanks for this Tao, I find that MJ really doesn't like venturing away from portraits even when first prompt is for full body/full length - time and time again it just brings those portraits which is incredibly frustrating. Tried putting full length/full body after the subject and described a full cinematic scene but it wants to do its own thing.
Hey, did you try starting with a vertical aspect ratio like --ar 9:16 and then panning/custom zoom out horizontally? Also if you include clues for the lower body, like types of shoes that can also help with full body prompts.
I hope this was helpful! Here's a FREE ebook pdf guide:
taoprompts.gumroad.com/l/cinematicprompts
👋
Your tips are simple and easy,tricks that you are giving are incredible @@LouisGedo .But the free ebook link cannot work!?
THANK YOU!! 🙏🤝
Thanks Bro!
puchased, owesome!
This is by far the best TH-cam tutorial for midjourney prompting. Thank you!!!❤
Great to hear you liked this guide! Midjourney is the best for realistic & cinematic photography imo
The best tutorial I've seen yet. Too many Tubers are focused on themselves and shoving their face in the camera for likes and subs... They are promoting themselves more than the subject of the video. I dislike them as in Thumbs down... This video is very to the point with Very helpful info.... Thumbs up and +1 sub....
I'm glad you liked this guide and thanks for subbing!
Bro your videos are simply amazing.. Man i wish more people made videos like you. Def going on my favorite Crators list...
Thanks man, I appreciate that! It's always awesome to hear these guides are helpful 👍
This was hands down the best video I've seen for MJ and cinematic styles. I've been at ai for a time now but am just diving into the cinematic side. Thanks again!
Thanks I appreciate that. Midjourney is probably the best tool for cinematic and photorealistic images.
The best channel I've ever visited among all mid-journey-related channels here on youtube!!
Thanks! I appreciate that, I've got many more vids planned!
Thanks for this great tutorial! I'm working with a company that sells watches and wallets. We use midjourney for some shots and i struggled a bit to create good images. After watching your tutorials, the results are much more realistic. i can't thank you enough man
Glad to help man! It's always cool to hear more and more companies using these technologies.
That other guy said it, and I agree enough to repeat it... best delivery of info, examples and straight forward explanation I've seen in the genre thus far, many thanks.
Thanks, I really appreciate that!
Your photos in the ebook pdf are SOOOOOOOOOO beautiful!
Thanks! Those ebooks are fun to make, I'm planning to make a couple new ones for my recent videos.
Your video just popped up to me while scrolling, and Dudeee, your content is amazing. You just didn’t waste time as other youtubers usually do! Every minute i spent watching the vid, i learned something to add to my knowledge. Thank you man and wish you the best❤
Edit: and i subbed of course!
Thanks for subbing! Happy to hear you found this video helpful, I tried to include as many tips in here as possible.
Como eu não achei o seu canal antes!? Eu quase choro com esse vídeo porque o que estou "apanhando" para fazer exatamente a cena que eu quero e de repente você entrega tudo isso assim, "mastigado", para facilitar a nossa compreensão. Me senti agora sendo "alfabetizada"😅 apta a descobrir um mundo totalmente novo. Muito obrigada mesmo 🙌
Fico feliz em ajudar você! "Midjourney" é complicado de usar e também tive problemas com ele antes. Então eu gosto de simplificar
Wow. I was playing with the prompt for Midjourney but was never able to really get the result that i wanted. These simple and easy tricks that you are giving are incredible in how effective they are. Thank you so much ! ❤
You're welcome! Midjourney can be tricky to prompt in sometimes, so it's always good to know the types of prompt keys you need to get what you have in mind.
Excellent; one of the more high quality prompt instructions. Interesting way to weight each characteristic (by description order flow). This is also achieved by numeric weight descriptors after a prompt word, like ::0.5 or . I think Mid Journey recognizes the former.
Thanks! Weight descriptors will definitely work as well. It seems that in Midjourney version 6, it can be hard to get all the exact elements, especially when it comes to certain camera angles since the model likes to focus on faces quite a bit. It's important to try and control each element of the image separately.
5:24 that imp detail, for prompt, your explantion is very very nice 5:34
Word order was really important for version 5.2! Now that we have version 6 it does not seem to matter as much any more.
These simple and easy tricks that you are giving are incredible ,thanks!
Thanks, I'm glad these were helpful!
Priceless tips and MJ tricks bro, you should def have a cheat sheet or PDF u sell w this gold!
Thanks man! I'll work on a pdf guide for this tutorial, it will be something new for me to try.
@@taopromptsawesome to hear Tao bro! 🎉
Simply incredible! Thank you very much for your dedicated work. It made my understanding of prompt construction so much easier! I will keep an eye on all your content! I wish you all the success in the world!
Glad to help! I'll make updates on the prompts for Midjourney version 6.
Thank you so very much for creating and sharing this video. It’s given me more clarity on the photographic possibilities than any other Midjourney video I’ve watched - and I watch A LOT. 🙏🏼
You're welcome! It took a lot of testing and image generations to compile these tips, I'm glad it was helpful to you.
Amazing video on lighting, camera angles and prompt structure. Really good. Thank you.
Thanks! There's a lot you can do with Midjourney.
this is an incredible video, thank you
Thanks, happy to help!
one of the best mj videos out there! Thanks Tao
Thanks Carlos! More guides coming soon.
Bro, extremly helpfulk thanks.
I'm always happy to help 👍
Thank you for this great tutorial. I watched it about 10 times now. Can you make tutorial about consistent characters in mid journey because this is something I am still struggling a lot.
Hey, I have a couple tutorials on consistent characters, here's a link to one of them: th-cam.com/video/RPxEg7irwAA/w-d-xo.html
Another excellent Midjourney tutorial!
amazing tutorial, Tao! 🤩🤩
Thank you 🙏, I will have to make an updated version of this for the newest Midjourney version soon!
Love this, thanks for the clear and concise guide
Thank you, glad this helped!
btw the thing with the order of words, for example naming the colors, has changed with 6.1!i got the same results and right colors, no matter which order i described the outfit. cool thing though!
Yup, the newer Midjourney models are much better with understanding prompts. Pretty much all parts of the prompts are included properly now 👍
@@taoprompts they do, indeed! i am recording a biiiiiig tutorial for a marketing agency and mentioned this thing, too, but then noticed afterwards, that this problem doesn´t exist anymore, haha. but btw, even if some tutorials of you are a bit older, most of it didn´t lose it´s meaning in mj :D
awesome awesome awesome
muy bueno gracias!
de nada! estoy feliz de ayudar
INCREDIBLE!
Excellent stuff! Thank you very much!
Thanks and you're welcome!
Thanks a lot for this amazing helpful video!
You're welcome! Happy to hear you enjoyed this.
This is absolutely amazing tips thanks bro 👏😁
Thanks and you're welcome! I'm happy to hear this was helpful for you.
Thanks for such a great knowledge
You're welcome, I'm happy to help.
Great video, thank you!
Thank yoy
How you are able to make so realistic ai short film like lighting so realistic and camera moment.❤❤❤❤
Love your way.
Thank you! For short films you can try Pika Labs or RunwayML. Here's an old tutorial I made, I will post an updated version soon. th-cam.com/video/8awNnoot_qg/w-d-xo.html
@@taoprompts thanks
Thx very much for video
Nice images dude!
Thank you!
As important as camera position, it's the LIGHT placement, please tell me how to place the Sun on an image. how to tell that I want the Sun at my back to use Sunlight without seeing the Sun. Example Hard Test: Try to do an image of a face illuminated by a Sunset, and we are NOT seeing the Sun, nor the sunset, the Sun it's the source of light at that time of the day at the beach for example, but I don't want to see the Sun, just the person with frontal light coming from the Sun.
Man, I'm debating if I sure invest in MJ. Ty for the tips.
It's worth trying, at least for a month or so. I think it does make the best looking images.
Thanks a lot. It was very helpful! Is there any way to use real photo as a background and add the character on it?
Not in Midjourney, I do plan some making some videos about photoshops generative fill though which would be able to do that.
This channel has been so helpful in my own growth with ai art! I am actually using Tensor Art but the information you give naturally translates across all these platforms with some minor tweaking. The particulars of prompt structure itself is the most important, along with camera angles etc. Thankyou and I can see your content has blossomed over time into these very well crafted presentations and so I not only thank but must also congratulate you on your own growth too!
Thanks I appreciate that! I haven't tried Tensor Art but I have used SDWebui which is similar and the control you get is really nice. The different platforms do seem to have a lot of similarities when it comes to prompting.
Excellent video and excellent guide, thank you very much, emmmm these cinematic propmts work for Midjourney 6?
Thank you, I appreciate that. These should definitely work in version 6.
Great guide. Thanks to it, I'm doing better and better things. Thanks a lot 🙏I'm starting my journey with midjurney. I work as a tattoo artist and generate references in it. I'm creating better characters, but I still struggle with generating realistically looking skulls. They all seem toy-like, even when following your tips. Do you have any advice?
Thank you.
Maybe try a different location where you expect to see a real skull:
"photo of a human skull in a museum"
"Stock photo of a human skull with a plain white background"
I not sure exactly what a skull is supposed to look like, but they look realistic to me.
@@taoprompts thanks for the answer. Love your channel. Keep going bro! 😁
Thanks
Hey, I tried your exact prompt at 3:47 and got nowhere near the girl with the different angles and photo quality you got with the 6 images split. Was there more to your prompt or image references? Thank you
Hey, this video was made with Midjourney Version 5.2 so that may be the difference there.
can I use this prompts in Leonardo?
You can use the basic ideas, but the results won't be as good for photorealistic images.
Any idea of these prompts would also work for SD?
They keywords should mostly translate over and have a visible effect, but the quality and style will be slightly different.
Your work is amazing, can you please simplify the free PDF access. As your link is taking me to website and I've to pay something. I dont have card. Thanks for your support, I've already subscribed your channel.
In the payment area, you can just enter "0" and get it for free.
@@taoprompts thanks a lot, your information is very helpful. Downloaded PDF. 👍👍👍
Have you tried these prompts with /shorten? I wonder how many of these long prompt keywords matter..
I haven't tried that yet. In my experience the /shorten command frequently misinterprets your original prompt. It will shorten phrases, but then the shortened keywords mean something different to the Ai from your original prompt.
How do you make the background not clear in closeup photos of a person. For instance if I had a prompt like "cute girl hiking in the mountains," the mountains come out blurry in the background.
how do you fix the blurry background?
That's a tricky one. You could try using the words "deep focus photography" which in real life will sharpen the background, but it doesn't always work well. The opposite would be "shallow focus"
example: "deep focus photo, a woman at the arches national park, noon, insta x"
@@taoprompts thank you. I figured it out. I used *"high depth of field, every detail of the valley is crystal clear, shot with a DSLR 4k, Camera Lens 24mm, f/16"*
Oh wow, that sounds complicated. Good to hear you got it working tho.@@MikeBNumba6
Please write this valuable information in the PDF file to download
I'm working on it!
@@taoprompts We are waiting for that. But please, this service will be free because we do not have enough money. Thank you.
I never get a hybrid to work in Midjourney. I get 2 images of the one and 2 images of the other. No hybrids…. What is your trick?
This guide was made for version 5.2.
In version 6, try a prompt like "Animal fusion hybrid of a tiger and an ox in natural habitat --style raw"
Create on the Leonard
These prompts tips also work on leonard ai
Too difficult to understand ai image generators 😂
It is overwhelming at first, but once you get some practice, it's not so bad 👍
❤❤ hi. ❤MJ releasing model personalization. Get at least 200 image ratings and then add --p after your prompt. Personalization completely changes the way system interprets your prompts.
thanku. awesome and great! thanks for PDF
This information is gold, thanks for sharing...i've used two of your tutorials to inspire me to make covers for my music videos
That's awesome! It's cool to hear so many different ways to use Midjourney, you can make some really nice album art with it.
i want a better way to create image to image with 100% same face. I tried with several ways but failed. I used my own photo instead, but cannot. My purpose is to create 64 or many photos with same face or image of different post or style in order to produce a video later. Please advice, tks.
Hey, there's a new consistent characters feature which makes that possible. I have provided an update tutorial here: th-cam.com/video/Vi5KQUZSKkM/w-d-xo.html
amazing tutorial. I have some prompts I just cant figure out. Hoping you or the community can help me out. How do I replicate a realistic photo of a person that makes it look like it was taken on an iphone? I also want to be able to do this with the only lighting coming from the camera flash. I can't seem to get it right
What happens if you find a real life photo with the flash photography style you like and use it as a style reference?
Damn...... You so good..... Thank you so much. The best video
Thanks man, glad to hear you enjoyed this guide 👍
Liked and Subscribed. Solid.
Thanks for subbing! A lot more is on the way.
Ahhhhh! You cracked the code. Some of my problems stem from word order. The other topics were awesome reminders and gave me a fresh perspective on my prompting. This video rocks. Thank you once again!
You're welcome! I've had a ton of issues with word order too. Idk if that's something they can even fix with the ai model, but you can always work around it. Always happy to help!
u are so cool
Thanks! I hope this guide helped you out a bit
great job bro ..thx
Thanks man
You are the GOAT
Thanks, I appreciate that
Yay! When I clicked thumbs up, it jumped to 2K. Tao, you deserve more than that! Thank you for the time and effort going into this guide.
Thanks Suzie! That's a lot of likes, it's cool to know this guide has helped a lot of people like you 👍.
Awesome Tutorial. Thank you.
Glad you liked it 👍. I'm working on another similar guide for my next video.
@@taoprompts Keep up the good work. Do you have any pointers for a good place to share unique AI art?
@@stylewithaudra5570 The Midjourney reddit forum is actually a decent place to get some exposure.
Instagram is more saturated, but I've seen accounts use Ai images combined with an Ai motion tool like leiapix, with simple capcut effects to create reels which seems to do pretty well.
exnihilo.creations is an instagram account like that.
@@taoprompts Great advice! Thank you so much. Waiting for your next tutorial drop. 👏
Great video Tao!!...you always over give...thanks!
Thanks man, Glad to be of service!
This is absolutely amazing tips,thanks bro,I spent the whole night learning it.
That's awesome man, great to hear this helped.
You must be a master photographer! Thank you so much for sharing such an amazing work!
Thanks, I'm still learning just like everybody else. I hope this helped!
very good tutorial, thnx !
thanks for this Tao, I find that MJ really doesn't like venturing away from portraits even when first prompt is for full body/full length - time and time again it just brings those portraits which is incredibly frustrating. Tried putting full length/full body after the subject and described a full cinematic scene but it wants to do its own thing.
Hey, did you try starting with a vertical aspect ratio like --ar 9:16 and then panning/custom zoom out horizontally? Also if you include clues for the lower body, like types of shoes that can also help with full body prompts.
@@taopromptsthanks I'll give that a try
So much useful informations , thanks ❤
You're welcome! Happy to hear this helps you.
You sir are a genius. Thx for sharing your insights
Thanks! I'm happy to help.
Very useful, thank you! ❤
You're welcome!
thank you
Amazing tutorial!
Thanks!
This was really helpful! I am just starting my ai journeys and really appreciate an honest tutorial that doesn't feel rushed.