i think it did the exact right thing when it put it on her left cheek. it is her left cheek. it's not "her right cheek from our perspective". that is not a thing. her left cheek will always be her left cheek.
I tried similar prompts way more times than I showed in this video, and the results were pretty inconsistent - so I wasn't confident in declaring any sort of definitive perspective
Not really. When you're looking at someone, right is left. In a mirror, left is still left. With a camera, some cameras will mirror the image, some won't. So it's really not that simple. When I say left in an image, I mean left.
@@99xara99 but it's still HER RIGHT CHEEK. just because you're looking at someone who is right-handed, that doesn't make them left-handed from your point of view. just because you're looking at someone wearing an engagement ring, that doesn't make them married from your point of view. HER right cheek doesn't magically turn into her left cheek just because you're taking a picture of her.
@ElevenEvilExes, I think that’s spot on. Maybe that explains why some surgeons have lost their license when they amputated a patient’s wrong leg!! Bad joke, I know, but kind of illustrative of the point. When your friend tells you on the phone, “Gee, I have this huge pimple on my left cheek…”, which cheek do you think they’re referring to? THEIR left cheek, of course - not their right facial cheek, and neither their right NOR left butt cheek (sorry, I had to 🤭). But apart from all that, EXCELLENT video, thank you so much!!
Such a great tip about how to phrase the prompting in V6. I had been just copying and pasting in the same prompts as I was using in V5.2, I just changed it to the new phrasing you recommended and immediately got better results
Thank you so much for this amazing tutorial, it cleared up tons of my concerns prior spending money in -what I thought before- just another AI image generator, now feeling more optimistic about using it.
How amazing! Just subscribed to Midjourney Pro today, did my very first proms - not that successful. But then I ran into you and this video. Helped me so much to get an understanding of this new word. I’m excited and can’t wait to try everything. You rock >:^) All the best from Copenhagen - Anne Mette
I'm a MJ super user going on 1.5 years now and I didn't know about the prefer option set and suffix 🤯 Thank you for teaching an old dog some new tricks x
What an excellent guide for Midjourney beginners! I'm excited to see how Midjourney has evolved in 2024. I'm interested in how it approximates other AI image years media regarding ease of use and output marker. I'm looking forward to levelling up my Midjourney skills.
haha wow, what a good idea to use a style tuner as an image prompt. Thanks a lot for that tip!! And thanks for the kind words and feedback. Happy New Year!
Great video. I would love to see you do a series on creating characters and then developing the same character in different poses. Where Midjourney remembers the character than allows you to build on it and the various perspectives. You should also sell workflow prompt packs.
hmm .. cool ideas, thank you. I've resisted a consistent character video because I know they're working on a feature to accomplish that - and any tricks I share now will be obsolete sooner than later. But I have material for Poses and maybe I can prioritize that. Workflow prompt packs are a cool idea too!
Dearest Nolan, your video is perfectly awesome. Thank you for getting right into it and expressing the nature of the functions so fluidly as you go along. There is no fluff and not additives. Just pure information. Thank you for your good work and continue to do so. I subscribe. Glad you feel better.
Thank you. Just signed up for your newsletter. Some FANTASTIC articles. Thanks, so much for sharing your experience and knowledge. V6 sooooo exciting!!
Don’t mind me, just taking notes: - chaos - Weird - Blend images - Combine images - Stylized (how much creative to give Midjourney) - Describe function to describe the image, then drag the reference image in to Midjourney to have two references V5 - zoom feature - In paint changes
Fantastic video to start 2024 with. Thank you for this. Glad you're feeling better. Yes, I think the style tuner was the most powerful feature in the last year. Once I switched from MJ v5.2 'language' to more 'Chat-GPT' based input, the results are _much_ better. Look forward to more fantastic videos. 👍
Great work! You´re the one whose posts I appreciate and await most. Always filled with information and enthusiasm. About V6: Yes, it is nice somehow. But I really struggle with style features. To me creating a real piece of art begins with a "lucky error". Something that MJ evolves with intended surprise. That´s why I like to use paradox prompts. Just to see what comes out. I find 5.2 does that much better than 6.0. But who knows where this is going. Enjoy 2024!
Very well said about 'paradox prompts' - I love using those as well, and they were definitely better in 5.2 😂 but 6.1 might be even more impressive haha
haha happy new year to you as well! And yes, my goal is absolutely to never make a sponsored video. I'll know within a couple of months if that is a feasible future for me!
from the pov of a beginner to midjourney, this doesn't explain how i actually use it, where do i go to use it? is there a website? a discord? a software i need to download? an app? how do i actually access this? otherwise the content seems great. just maybe aimed more at intermediates than actual beginners.
Sorry! I definitely assumed if someone clicked on this video, they would have already looked into using it. You can sign up through www.midjourney.com - and start generating on Discord. (the main midjourney website will have creation enabled soon) You can check out this video on which plan might be best for you - ua-cam.com/video/X8l7UJYO4u4/v-deo.html
@@FutureTechPilot Just a suggestion, maybe you should put the beginner link under all your videos to help people navigate your videos from the beginning. You have an awesome channel, and I have learned so much from you. Thank you.
I tried over and over to get MJ6 to produce the Scottish flag. Even used the technical description, but it never even got close. I mentioned it in the forums & nobody else could get a simple Scottish flag.
Maybe I should have mentioned that in this video - if something isn't in the Midjourney dataset, then MJ will have a difficult time creating it. If it wasn't trained on it, it really doesn't know what you're asking for. But you can check out this video - ua-cam.com/video/4FQgkyMGdVY/v-deo.html - maybe you can use that technique to create it
the best video so far I saw, amazing needs more advanced prompt technics to get exactly what we imagine. random is fun, but I wish to create specific art. I have played with Bing for the last 2 weeks and more friendly, and pretty fast, but don't fulfill all I need, hope Midjorney will be able. Last night I tried a few things and found limits, for example, midjorney doesn't know what is a 8 track tape, i know I could "show him" some pictures as references, I will try that. Another thing is in photography and cinema we have different types of way to frame a portrait, close up, shoulder and up, waist, full length body, in Midjorney have difficulties recognizing those instructions, can't find the right way to prompt it yet. another missing right now is a complet good cheat sheet. good work
V6 is fantastic! Great video, your are always the easiest to follow.. Weirdly I find the "old" style prompt method better suited to me. I copied an older prompt into V6 and got an "upgrade" on my old image, but when I arranged it to a more conversational style, it took me further away. BETA is obviosly evolving as a couple of days ago, female charactors had a slightly Asian look, but now look more Western. (this was without prompting for ethnicity)
Hi, thanks for the video, it was easy to understand for this new midjourney user. Seems tho we need to double down on word use? I subscribed yesterday to MJ after finally hearing it has a website. Imagine my surprise when I still need to use discord…. Geesh, now two programs I need to learn. Not easy for this old girl! Cheers from Australia 🇦🇺❤️
favorite thing in entire video was that you didn't know that "stacked stones in a field" were called 'Carins" ...(pronounced 'karen'). they exist to keep hikers on the trail when no trail is obvious/visable :) you're welcome. Thanks for the kickass video guy.
thank you , i really appreciate it , i learned alot from the video .. its just so good thank you man and please tell me where can i get the course that you talked about ? the course on using midjourney
I'm currently filming it as we speak! Lots of work to get it all filmed, and then I still have to edit it. But I hope to have it available very soon!! Maybe by the weekend? Please check back on this channel in a few days
I'm new to midjourney, and this is a fantastic tutorial. May I ask you: I found out that upscaling in v6 seems to be different than in v5.2. Is that correct? I seem to get a higher resolution in v5.2 by using upscale 4x. Thanks again! I'll definitely subscribe to your newsletter.
Yeah it's different than 5.2 for sure. In 5.2 we had 2x and 4x, which were replaced by Subtle and Creative. I get the sense that the 5.2 upscalers were more pixel based, and probably more difficult to develop which is why we haven't seen them yet in Version 6
The ULTIMATE Beginners Guide to Midjourney in 2024 I love this video, it makes it sound much more achievable. But if you're paying so many credits for each time you regenerate an image, you literally would burn through your monthly credits in one day! How do you get around this?
Excellent videos! Is it possible to generate prompt images without modifying their overall composition, shape, or size, such as to stylize prompt images or render 3D models without changing whole picture? Is there going to be any updates for this? 😅 thanks
Right now, no you can't do that unfortunately. Every change to the prompt will change the picture. In-painting will eventually sort of help with that, but that's probably a week or two away
From the Ai perspective, maybe it 'sees' from within the codes(?) thus "left" and "right " from it's perspective is natural, as is "right" and "left" for us looking from the outside(?). Just musing. ;)
I am havign trouble creating a All Over Prints. I have told the program repeated print, all over print, --tile everything i can think of. Do you have any recomandations?
--tile is definitely going to be the thing you're looking for. But I've heard --tile might be wonky in V6. You can try --v 5.2 and I think you'll be happy with that!
My favourite parameter is --weird. Especially when you play with MJ. If you working on something specific it's better to stay away from weird entirely.
No not right now. It would change your furniture in each shot. You could generate the backgrounds and then place the furniture inside with something like photoshop
Oof, I gotta say it's a lot of new features to absorb! Very green newbie frightening even. I already get "nice" results with vanilla prompts... Gotta practice it, no other way to master MJ! Thx!
at the 3:00 minute part when you are talking about adding prompts one at a time, how are you doing this? Are you writing out the whole prompt again then adding an extra prompt and doing a whole new search each time? or are you selecting one of the 4 images and upscaling it then adding another prompt from there?
Thank yo. I am very new to this, so I appreciate your help.However, I find that even the most basic prompts are very random. As an example, if enter "street scene, Woman with dark hair wearing black tights", at first I might get 4 correct images, but after more promps out of the 4 images, only one them will have black tights. And even if I elaborate by saying "opaque black tights" in every prompt, the promps always seem to be random. Some of them have no tights at all and others have black socks or lots of variations, but not the same as the original correct promp. Almost as if the A.I is getting worse each generation. Is there any way to keep an look that is correct and then add more promprts without altering the origi nal look?@@FutureTechPilot
Is there a way I can contact you in private? I would really appreciate your help in being able to create something specific that is reliably repeatable. It is nothing illegal and if you could help me, I would happily pay you for your help. Thank you.@@FutureTechPilot
I am trying to get pictures for a coloring book of a duck. the pictures I get are too detailed for coloring book andI've indicated this to bot. Mentioned no shade, no detail. Not getting anywhere,
Hey, nice content btw, in your video you get confused indeed, I just wanna specify, so on you prompt u typed “on her left side” the v6, so the bot does it correctly. From your perspective, it’s your right side, but from the girl in the pic, it’s her left side.
hmmm .. maybe create the War photos first, and then use Vary Region on their weapons and try to replace them with WW2 weapons? - that's probably what I would try
i think it did the exact right thing when it put it on her left cheek. it is her left cheek. it's not "her right cheek from our perspective". that is not a thing. her left cheek will always be her left cheek.
I tried similar prompts way more times than I showed in this video, and the results were pretty inconsistent - so I wasn't confident in declaring any sort of definitive perspective
Not really. When you're looking at someone, right is left. In a mirror, left is still left. With a camera, some cameras will mirror the image, some won't. So it's really not that simple. When I say left in an image, I mean left.
@@99xara99 but it's still HER RIGHT CHEEK. just because you're looking at someone who is right-handed, that doesn't make them left-handed from your point of view. just because you're looking at someone wearing an engagement ring, that doesn't make them married from your point of view. HER right cheek doesn't magically turn into her left cheek just because you're taking a picture of her.
@ElevenEvilExes, I think that’s spot on. Maybe that explains why some surgeons have lost their license when they amputated a patient’s wrong leg!! Bad joke, I know, but kind of illustrative of the point. When your friend tells you on the phone, “Gee, I have this huge pimple on my left cheek…”, which cheek do you think they’re referring to? THEIR left cheek, of course - not their right facial cheek, and neither their right NOR left butt cheek (sorry, I had to 🤭).
But apart from all that, EXCELLENT video, thank you so much!!
Discussing left and right - gee, is that how far down the decline of humanity has come?
Thank you A LOT! I was afraid to use Midjourney and felt overwelmed, but your video opened a new world for me! I am waiting for your course!
That makes me happy to hear that! I'm working on the course as we speak, I think I'll have something to show next week
Fantastic synopsis for those unfamiliar with Midjourney! This guide's depth and clarity make learning a whole lot simpler.
Thanks for the kind words!
Great one Nolan, haven't seen any better vlogger for Midjourney prompting guidance than your FTP grtz
Hey I'm glad I could help out, and thanks a lot for the feedback!
Thanks. I am a total beginner. Have not begun Midjourneying yet but this gave me an idea of how I might begin. Thank you.
And thank you for the comment!
Such a great tip about how to phrase the prompting in V6. I had been just copying and pasting in the same prompts as I was using in V5.2, I just changed it to the new phrasing you recommended and immediately got better results
Well that makes me happy to hear!
Just starting out and I've learned so much from this video.
Hey I'm glad to hear that!
Thank you so much for this amazing tutorial, it cleared up tons of my concerns prior spending money in -what I thought before- just another AI image generator, now feeling more optimistic about using it.
I'm really glad to hear that! Thanks a lot for your feedback
How amazing! Just subscribed to Midjourney Pro today, did my very first proms - not that successful. But then I ran into you and this video. Helped me so much to get an understanding of this new word. I’m excited and can’t wait to try everything. You rock >:^) All the best from Copenhagen - Anne Mette
I'm so glad to hear that! Cheers 😊
I'm a MJ super user going on 1.5 years now and I didn't know about the prefer option set and suffix 🤯 Thank you for teaching an old dog some new tricks x
What would you type if you wanted your character to face forward in the picture? How would you describe this? I am a newbie.
Happy I could help!
Very helpful. Thank you!
Cheers!!
What an excellent guide for Midjourney beginners! I'm excited to see how Midjourney has evolved in 2024. I'm interested in how it approximates other AI image years media regarding ease of use and output marker.
I'm looking forward to levelling up my Midjourney skills.
I'm happy to hear that!
Ayyy been waiting for this! Your last beginner's guide was my favorite on UA-cam.
haha wow, what a good idea to use a style tuner as an image prompt. Thanks a lot for that tip!! And thanks for the kind words and feedback. Happy New Year!
very helpful and nice structured, thank you
And thank you for the comment!
Great video. I would love to see you do a series on creating characters and then developing the same character in different poses. Where Midjourney remembers the character than allows you to build on it and the various perspectives. You should also sell workflow prompt packs.
hmm .. cool ideas, thank you. I've resisted a consistent character video because I know they're working on a feature to accomplish that - and any tricks I share now will be obsolete sooner than later.
But I have material for Poses and maybe I can prioritize that.
Workflow prompt packs are a cool idea too!
yeah, right now its pretty shit tbh, its hard to get anything consistent
Thank you so much for the help learning Midjourney. You're a great teacher.
And thank you very much for the kind words, I really appreciate the feedback!
Dearest Nolan, your video is perfectly awesome. Thank you for getting right into it and expressing the nature of the functions so fluidly as you go along. There is no fluff and not additives. Just pure information. Thank you for your good work and continue to do so. I subscribe. Glad you feel better.
Thanks so much for the kind words and the support!
Very useful. Just want to point out you can have more than one parameter in the suffix e.g. --ar 3:2 -- chaos 10 --no words, text
Yah that's a very good point!
amazing summary, the best that I've seen so far
I'm happy to hear that!!
Thank you. Just signed up for your newsletter. Some FANTASTIC articles. Thanks, so much for sharing your experience and knowledge. V6 sooooo exciting!!
Don’t mind me, just taking notes:
- chaos
- Weird
- Blend images
- Combine images
- Stylized (how much creative to give Midjourney)
- Describe function to describe the image, then drag the reference image in to Midjourney to have two references
V5
- zoom feature
- In paint changes
homie doesn't have a phone
Amazing!
Cheers!
This is probably the best and most detailed guide I've seen. Thank you so much.
Hey thanks a lot for the feedback! I'm glad I could help
This is an extremely informative guide 👍 thank you for putting out a valuable free resource
And thanks for the feedback!
I enjoyed it very much
I'm glad to hear that!
Fantastic details! Keep it up!. Bravo!❤
Cheers!!
Such amazing tutorial. Thank you for sharing Nolan!
And thanks a lot for taking the time to comment!
Fantastic video to start 2024 with. Thank you for this. Glad you're feeling better. Yes, I think the style tuner was the most powerful feature in the last year. Once I switched from MJ v5.2 'language' to more 'Chat-GPT' based input, the results are _much_ better. Look forward to more fantastic videos. 👍
And as always, thanks for your feedback!!
Great class, man! Helped a lot. Thanks!
I'm glad to hear that! Cheers pal
Thanks for taking time to do this video - much appreciated 👏
Cheers pal, hope it helped!
Great work! You´re the one whose posts I appreciate and await most. Always filled with information and enthusiasm. About V6: Yes, it is nice somehow. But I really struggle with style features. To me creating a real piece of art begins with a "lucky error". Something that MJ evolves with intended surprise. That´s why I like to use paradox prompts. Just to see what comes out. I find 5.2 does that much better than 6.0. But who knows where this is going. Enjoy 2024!
Very well said about 'paradox prompts' - I love using those as well, and they were definitely better in 5.2 😂 but 6.1 might be even more impressive haha
You are amazing man. Thank you 🎉
Thanks a lot for the kind words!
Amazing Video... Thank You!
Cheers!
Happy new year man. I hope in 2024 you get a lot more new subscribers -and supporters- to keep the channel clean and add free.
haha happy new year to you as well! And yes, my goal is absolutely to never make a sponsored video. I'll know within a couple of months if that is a feasible future for me!
This is a fantastic video, thanks so much. Lovely and clear and motivational. I can't wait to get trying all your suggestions.
And I really appreciate the feedback!
Thanks for the videos, its easy to understand . Keep up with the good work, and good luck with your channel
Thanks for the feedback and kind words!
I'm Not dealing with discord... hope the site is up soon...
Yah, definitely soon! If you have made over 10K images, you have access to it right now - alpha.midjourney.com/explore
Amazing video!! Thank you!!
And thanks for the comment!
Excellent details! Love the side-by-side comparisons with the prompting. Really helpful information, thank you so much! Peace to you!
Thanks a lot for the feedback! I'm glad I could help
Awesome work bro. I wouldn't call myself a MJ beginner, but I'm always learning something from your vids. Many thanks. Well wishes for 2024. 🎉
haha well I'm glad the video makes sense from other seasoned users. Thanks for all the support!
Learnt a lot. Great video
I'm glad to hear that!
@@FutureTechPilot 🙏🏼
Awesome video man!!!
Thanks a lot!
Very useful tutorial. Thank you for sharing.
And thank you for the feedback!
Thank you so much for this, it was so very helpful and I will use this as a resource as I continue to learn how it works.
And thank you so much for the feedback! I'm happy I could help
from the pov of a beginner to midjourney, this doesn't explain how i actually use it, where do i go to use it? is there a website? a discord? a software i need to download? an app? how do i actually access this? otherwise the content seems great. just maybe aimed more at intermediates than actual beginners.
Sorry! I definitely assumed if someone clicked on this video, they would have already looked into using it. You can sign up through www.midjourney.com - and start generating on Discord. (the main midjourney website will have creation enabled soon)
You can check out this video on which plan might be best for you - ua-cam.com/video/X8l7UJYO4u4/v-deo.html
@@FutureTechPilotnah . karmah is right. way to fix it though and thanks for the video !
@@FutureTechPilot Just a suggestion, maybe you should put the beginner link under all your videos to help people navigate your videos from the beginning. You have an awesome channel, and I have learned so much from you. Thank you.
I tried over and over to get MJ6 to produce the Scottish flag. Even used the technical description, but it never even got close. I mentioned it in the forums & nobody else could get a simple Scottish flag.
Maybe I should have mentioned that in this video - if something isn't in the Midjourney dataset, then MJ will have a difficult time creating it. If it wasn't trained on it, it really doesn't know what you're asking for. But you can check out this video - ua-cam.com/video/4FQgkyMGdVY/v-deo.html - maybe you can use that technique to create it
Great tutorial thanks Nolan.
And thank you for taking the time to comment!
Very detailed and helpful Thanks
And thank you for the comment!
Very helpful video. Showed me a couple things I was wondering about. Concise and to the point. Subbed.
Thanks a lot for the feedback!
Thanks for the Patreon heads up. I’ll join today!
Genius video, very helpful. Thank you, Nolan! 🙌🏽
And thank you so much for taking the time to comment. Cheers!
Thank you very much. Very cohesive and complete view of the functions. Very helpful!
And thanks for the feedback!
Fantastic overview
Thanks a lot!
the best video so far I saw, amazing needs more advanced prompt technics to get exactly what we imagine. random is fun, but I wish to create specific art. I have played with Bing for the last 2 weeks and more friendly, and pretty fast, but don't fulfill all I need, hope Midjorney will be able. Last night I tried a few things and found limits, for example, midjorney doesn't know what is a 8 track tape, i know I could "show him" some pictures as references, I will try that. Another thing is in photography and cinema we have different types of way to frame a portrait, close up, shoulder and up, waist, full length body, in Midjorney have difficulties recognizing those instructions, can't find the right way to prompt it yet. another missing right now is a complet good cheat sheet. good work
I've noticed it's harder to get the framing right in Version 6, I'm working on figuring out better ways of triggering what we want!
wow thank you so much for sharing, love it!😀😀😀
thank you for incredible video
and thank you for the comment!
Regarding the tattoo...you mentioned "her right cheek", so it's hers. Quite smart...
💯
Great MJ Youtu. Truly Inspiring experiments Thanks a lot.
I'm happy to help!
V6 is fantastic! Great video, your are always the easiest to follow.. Weirdly I find the "old" style prompt method better suited to me. I copied an older prompt into V6 and got an "upgrade" on my old image, but when I arranged it to a more conversational style, it took me further away. BETA is obviosly evolving as a couple of days ago, female charactors had a slightly Asian look, but now look more Western. (this was without prompting for ethnicity)
haha I agree with you! I like the old way of prompting more - I'm working on a process to translate the old style into the new style
Hi, thanks for the video, it was easy to understand for this new midjourney user. Seems tho we need to double down on word use? I subscribed yesterday to MJ after finally hearing it has a website. Imagine my surprise when I still need to use discord…. Geesh, now two programs I need to learn. Not easy for this old girl! Cheers from Australia 🇦🇺❤️
haha take deep breaths! This tech isn't going anywhere. You'll learn at your own pace!
Thanks mate 😀
Thank you so much! 😊
And thanks a lot for taking the time to comment!
Fantastic video, thank you.
Happy to help!
12:45 Midjourney upscales can be run in relax mode. I rarely use fastmode for upscale, particularly x 4
Yeah that's a great tip, I should have mentioned that at. Thanks!
@@FutureTechPilot I put some random times that, lol.
So in Version 6 MJ you add a lot of "The" at being of sentences I looking for keywords or word structures Thanks for all You do Sir
I'm not sure if it's super necessary, but it helps my brain plan the scene! thanks for the comment
Great video!
Cheers!
Awesome instructions
Glad I could help!
How do you download all the images created?
Maybe this video will help - ua-cam.com/video/usQl1nd4avk/v-deo.html
Thanks Bro, Really good video
Cheers! Thanks for taking the time to comment
I'm trying to change the size of my images to get the ar 16:9 and its not working
That sounds frustrating! May I suggest asking for help in one of the 'prompt chat' rooms on discord? There are lots of helpful people there
favorite thing in entire video was that you didn't know that "stacked stones in a field" were called 'Carins" ...(pronounced 'karen'). they exist to keep hikers on the trail when no trail is obvious/visable :) you're welcome. Thanks for the kickass video guy.
hahaha and thank you for the lesson!
hi, thanks for your valuable information. Could teach us how to create constant characters?
Maybe this video will help you! ua-cam.com/video/zgeI_ffCudY/v-deo.html
Thank you!
Cheers!
thank you, great video
Cheers!
thank you , i really appreciate it , i learned alot from the video .. its just so good
thank you man
and please tell me where can i get the course that you talked about ?
the course on using midjourney
I'm currently filming it as we speak! Lots of work to get it all filmed, and then I still have to edit it. But I hope to have it available very soon!! Maybe by the weekend? Please check back on this channel in a few days
@@FutureTechPilot i will for sure , thank you , god bless you
thank you very useful indeed
I'm glad I could help!
Very helpful😊
I'm glad to hear that!
I'm new to midjourney, and this is a fantastic tutorial. May I ask you: I found out that upscaling in v6 seems to be different than in v5.2. Is that correct? I seem to get a higher resolution in v5.2 by using upscale 4x. Thanks again! I'll definitely subscribe to your newsletter.
Yeah it's different than 5.2 for sure. In 5.2 we had 2x and 4x, which were replaced by Subtle and Creative. I get the sense that the 5.2 upscalers were more pixel based, and probably more difficult to develop which is why we haven't seen them yet in Version 6
Thank you! Let's hope we'll get a higher resolution soon.
Wish i hopped on this during the free trial 😢
The ULTIMATE Beginners Guide to Midjourney in 2024
I love this video, it makes it sound much more achievable. But if you're paying so many credits for each time you regenerate an image, you literally would burn through your monthly credits in one day! How do you get around this?
haha it's not the cheapest hobby in the world! I pay $50 a month for 30 fast hours - which is like 1800 prompts per month
Excellent videos! Is it possible to generate prompt images without modifying their overall composition, shape, or size, such as to stylize prompt images or render 3D models without changing whole picture? Is there going to be any updates for this? 😅 thanks
Right now, no you can't do that unfortunately. Every change to the prompt will change the picture. In-painting will eventually sort of help with that, but that's probably a week or two away
Thanks a lot!!!!!!!!
Cheers!
From the Ai perspective, maybe it 'sees' from within the codes(?) thus "left" and "right " from it's perspective is natural, as is "right" and "left" for us looking from the outside(?). Just musing. ;)
haha it's fascinating to think about, eh?
I am havign trouble creating a All Over Prints. I have told the program repeated print, all over print, --tile everything i can think of. Do you have any recomandations?
--tile is definitely going to be the thing you're looking for. But I've heard --tile might be wonky in V6. You can try --v 5.2 and I think you'll be happy with that!
My favourite parameter is --weird. Especially when you play with MJ. If you working on something specific it's better to stay away from weird entirely.
haha yes I think I agree
Other question.. I sell furniture online... do you think MidJourney can help to provide new ambiance photos of my furniture product?
No not right now. It would change your furniture in each shot. You could generate the backgrounds and then place the furniture inside with something like photoshop
Oof, I gotta say it's a lot of new features to absorb! Very green newbie frightening even. I already get "nice" results with vanilla prompts...
Gotta practice it, no other way to master MJ!
Thx!
Take your time! Deep breathes, you got this!!
Hi. Maybe you know how to integrate a logo into an existing photo?
I think Photoshop will be your best bet!
at the 3:00 minute part when you are talking about adding prompts one at a time, how are you doing this? Are you writing out the whole prompt again then adding an extra prompt and doing a whole new search each time? or are you selecting one of the 4 images and upscaling it then adding another prompt from there?
Generate one at a time! See what the prompt creates, and then use the same prompt but add some more words to it.
Hope that makes a little more sense!
Thank yo. I am very new to this, so I appreciate your help.However, I find that even the most basic prompts are very random. As an example, if enter "street scene, Woman with dark hair wearing black tights", at first I might get 4 correct images, but after more promps out of the 4 images, only one them will have black tights. And even if I elaborate by saying "opaque black tights" in every prompt, the promps always seem to be random. Some of them have no tights at all and others have black socks or lots of variations, but not the same as the original correct promp. Almost as if the A.I is getting worse each generation. Is there any way to keep an look that is correct and then add more promprts without altering the origi
nal look?@@FutureTechPilot
Is there a way I can contact you in private? I would really appreciate your help in being able to create something specific that is reliably repeatable. It is nothing illegal and if you could help me, I would happily pay you for your help. Thank you.@@FutureTechPilot
13:30 the oversharpening is a known issue with the v6 alpha, I would expect to see this fixed in the upcoming weeks.
Yah you're definitely right about that!
I am trying to get pictures for a coloring book of a duck. the pictures I get are too detailed for coloring book andI've indicated this to bot. Mentioned no shade, no detail. Not getting anywhere,
try lowering the stylize value to --s 40 or maybe even --s 0 ... and try --style raw
Fantastic review. Love your videos! ☺😉
Thanks a lot for the feedback and kind words!
Hey, nice content btw, in your video you get confused indeed, I just wanna specify, so on you prompt u typed “on her left side” the v6, so the bot does it correctly. From your perspective, it’s your right side, but from the girl in the pic, it’s her left side.
Yah, but it was hard to reproduce those results! Oh well, there will be an update to text soon enough
☺Your videos are amazing! Thank you so much!
And I really appreciate you taking the time to comment!
Super informative! While Midjourney is decent, Stylar’s AI is my favorite. Looking forward to your thoughts if you review it.
🤝
Why don't the upscale and vary region buttons appear in midjourney?
They haven't added it to V6 yet!
Any tips on how to get fingers and hands looking real in v6 if it generates 6 fingers, etc?
in-painting might work! Or you just have to re-roll the prompt. It's an art slot machine lol
I'm having a hard time getting a cross-over. I am trying to get WW2 era weapons into the hands of American Revolutionary War Redcoats. Any tips?
hmmm .. maybe create the War photos first, and then use Vary Region on their weapons and try to replace them with WW2 weapons? - that's probably what I would try
Grazie.
Thank you!
I'm down for a course with you bro. Looking forward to it. Consider me as your first student. 🤜🤛
haha I appreciate the support!!
A great video! Amazing what AI can do, and that's just the beginning
💯