I'll Walk You Through The Amazing World Of Artificial Intelligence
-
-
Hi, I'm Nolan, and it's my goal to be the best Teacher you've ever had. I think A.I is a lot of fun, and I think it can be easy to learn. From Midjourney to GPT and beyond, I aim to make content that explores the possibilities of all the new magic we have access to.
Cheers :)
Пікірлер
awesome video, does anyone know how to get consistent composition??
THIS IS ACTUALLY PERFECT OMG, IT WORKED FOR ME! Lmao, thanks, I'm excited 'bout this. I just got midjourney and the prompts I made don't always work... So this helps me out a ton. Thank you.
I'm really surprised you have to write bits of code to use this - forward slashes and all that nonsense! Dead clunky. Ridiculous in this day and age and rather ironic for an AI program. They need to develop it a bit more.
Im glad to be on the "not complicated enough side"
you`ve got to be an effing poet to use v6 to its max 😅
😂 just enjoy the journey!
Absolutely amazing. I'm relatively new to Midjourney and your videos are immensely helpful. Cheers!
I'm really glad to hear that!
Hey man ! Thanks ! How is your advanced training course ? I’m looking forward pay for that one !
Thanks for asking! You actually made me organize myself better and I'll tell you what's going on. So I have about 7 Units/Chapters completed - a total of 536 Presentation Slides There are 8 Chapters listed as 'High Priority' that I will begin to work on next. And after that, there are 18 more Chapters that I will include at some point. Now, I don't know if I will wait until I'm finished with more lessons or if I will record what I have and release it early at a discounted price, with updates coming over time. I think I'm leaning toward some-sort of early access. Thanks for your question!
This is me from Bahrain. Good job bro best Midjourney videos in your channel honestly
I really appreciate the support!
"Discord feature parody" sounds fun 😁
😂
I prefer to use both discord and the website until all the parameters have been incorporated into the user interface on the website. There are valid use cases for both platforms.
I completely agree!
New user confusion can be addressed if they spend more time improving their user documentation, and adding video tutorials. The program isn't too complex, it just needs more/better ways for new users to learn.
I agree! That's definitely one of the reasons I started my youtube channel
I learned from channels like this. If you want to be advanced, you just have to try.
Thanks for the summary! Midjourney programmers spent so much time on Rooms and it has turned out to be a dud. Every time I look in them, there will be one person generating the same prompt over and over, and monopolizing the room. This could have been predicted because you see the same thing going on in the Daily Theme room on Discord. There is very little interaction between users. Also, spending time aiming for the lowest skill user is a bad idea. If you make it too simple, everyone will soon get bored with it and move on to another program that allows more creativity.
I am going through your videos because.. I am not sure but I think in one of them I saw a girl with red hair in Disney pixar style and I wonder if there was a prompt there xD because I try to prompt it and it looks worse
I'll be honest - that doesn't sound familiar to me! But maybe you could try something like 'disney pixar 3d animation' and raise the stylize value like --s 400 !
@@FutureTechPilot yup, it wasn't yours :D but I found it and had a chance to rewatch your videos :D
I'm waiting v7 😴😴😴
I hope it will be worth the wait!
this is what happens when you make everything about Your comfort level in dev work, hiring teams scattered randomly across the globe with people going on leave in the middle of things. This is a one man show at the end of the day and his value system and work ethic and his sense of perfectionism. MJ has been a great product and the genius of the man is reflected in it, but the development pace, release cycle and features are also bearing the brunt of the very same genius. Developing great software and Tech should not be a one person affair however inspired he or she might be. You need dedicated people in situ, in office to push things along and inspire each other.
I can't imagine how difficult it would be to run one of these companies and keep your ethics/morals as the highest priority!
wait wait wait.....discord bad??? I tried the website creation thing and the result was blurry and undefined. Discord is the only place that I could get a workable generation.
It uses the same model as on Discord. I had never problem with this and always the results are nice in terms of quality of the images
I'm 99% sure it's just a perception thing. Like, the images on the Website are larger on the screen so it's easier to see the flaws - where on Discord, they are shown to you smaller so they look more crisp
Its interesting to see how many of us like the actual chaos of the machines !
haha there are dozens of us! 😂
Nolan, I’m another person that is not looking to MidJourney to produce exactly what I see in my mind’s eye. Since 2012 I’ve been involved in the mobile art community, writing tutorials on over 150 different iOS apps. So if I have something specific, I can get very close by fitting the pieces together myself. I subscribe to MidJourney for the surprise, the delight, the WOW factor. Not for monetizing. Right now, my main problem is that I do 95% of my interfacing on mobile devices. I infinitely prefer the website to Discord, but the website works (except for /describe) on the iPhone, but I can’t see the results well. On the iPad, I can see the results, but the website is really only about 50% workable, with very little support for cref and sref. A change to the website in the past day or two cripples it even further on the iPad, since there’s no “hover” capability. I know that it’s tough to do; Photoshop took more than a decade to produce a working mobile version. I just wish they’d realize there are a LOT of us mobile users who rarely get on desktop.
A W E S O M E….. UGLY is beautiful
yes!
Thank you!
Cheers!
What interface do you prefer? I was hesitant to lose my greatly organized channels on my discord server for different categories of prompts, but i use web 99% of the time.
Discord! But for a bunch of small reasons
Love the art on this one. Admit it: I'm inspiring you. 😂
hahah inspiration is all around us!
I go back and forth using the website and Discord. I hope they eventually get it to where I can do everything I want on the website but until then I will use both. I am learning new things all the time and improving my images all the time. I am patient.
I like your attitude!
So another 1 year they focous on Mostly site improvment rather than image quality🥲🥲
They should definitely have a free tier to start getting new users to try it out. Without that, they won't build users very fast at all.
They had one before but people found a way to abuse it - so I think we'll see a free tier again sometime in the future
Right now there are advantages to each,. but by creating in Discord is the only place you can create an image and it be fully editable and findable on both platforms.
💯 the both places thing is a big deal to me for some reason 😂
👋
🤝
I think they should focus more on the model. v6 is not great for me.
He seems quite excited for the next update!
@@FutureTechPilot I am too.. it sounds amazing actually. Cant wait to try it out. For some reason... Ive not been getting the best results in v6. These guys are amazing and I still rate MJ above all others. I have faith. Tho I have to say.... the Leonardo Character Regerence is way better imo. The outputs actually look like the original even from uploaded shots. Id like to see improvements to this system in MJ
I'm no longer subscribed. Meanwhile Leonardo released some crazy consistent character referencing and a new image generator. Mate, you should go beyond Midjourney, there's just not enough content to talk about in MJ.
I think it's great that their are quality alternatives! I just have a long list of MJ videos to make haha
I love that spider-man image!
Yess a bunch came out really well
Honestly, they are focusing on the wrong things, the more professional contestant users are getting tired of them dragging their feet worrying about things that are not what their users want. It's not even exciting anymore. I don't even stay on top of my subscription anymore. Once the Image generator drops along with SORA ( ITS A THING) WHY WOULD I COME BACK TO THE MORE EXPENSIVE MID JOURNEY.
To each their own!
Am I the only person who could not care less about their website?
I think it's really smooth and fun to use - but I've always been comfy on Discord!
I am not a fan of the website. I thought the rooms might be interesting to try out but haven't found them of much interest.
I use it occasionally but mostly prefer Discord.
I'm just glad i don't have to use freakin' Discord anymore.
😂 it's never bothered me but I know it doesn't click for some people
They should work on the promptunderstanding and describe tool instead. How can it be that free AI apps can do it a hundred times better? I'm still a subscriber and pay my 60 a month, but for how long? The filters are still too strict and even images that midjourney describes are blocked. V6 has trouble with human anatomy, especially in a sitting position. Thanks to the sref and cref, a lot can be adjusted. But u nolan do a great job thanks for the recap❤
Cheers pal!
Thanks for the updates, I enjoyed your images ! And thanks for giving the artist a shout out.
Yah it's the least I can do!
Would you please make a video on how we can use the images we create for business use. Eg: social media posts or website images.
Yes I am always taking notes. And I have a long list of videos I need to get to, but maybe I think about making that one a higher priority
I hope we see some improvements soon. I hope we haven't leveled off. Stable Diffusion looks like they may be down and out SD 3 may never be released to the public.
He seems pretty excited for the next update haha I can't wait to see if it was worth the way
Thanks for the info as always! Helen is so cool! And can you tell me as an aside, what are you using for your blue subtitles, they flow wonderfully!
I use Premiere Pro to edit my videos and they have a pretty good transcription -> caption feature. It's accurate and fairly customizable 👍👍
@@FutureTechPilot Thanks for the tip!
I kept my membership exactly because the images are not specific and controllable and thus adds to my inspiration. Rather than resulting in a product, it is entertainment to me, and that entertainment is well worth the money. I get to create(direct midjourney) fantastic images that are really inspiring me to make more. Dalle 3 does little to inspire and there is not alot of entertainment value in it. I agree that midjourney is fairly uncontrollable, but it's great fun and very inspiring. It creates stuff I would have never thought of, and it teaches me that there is always a different way to do stuff or a different technique that can be learned
I completely agree! Seeing what the bot creates is 100x more fun for me than prompting for something extremely specific. I think it's cool that we share the same sentiment toward the entertainment side of it
Uncontrollable can be good when you looking for ideas to expand on a basic prompt. Dumbing down the program will make it just a fancy version of Paint!
Absolutely agreed with you :)
V6 kept getting delayed forever remember that? And that was the version I was most excited about. But I think midjourney is so good at the moment that these delays don't bother me anymore. Sref random is such a good feature to spend time on. As long as I have that everything else can wait.
it takes a weathered soldier to find patience with technology haha I appreciate you spreading that vibe because I think it's important for people to enjoy the process of growth!
the website is a mess, i had to learn it so i ploughed thru, too many hidden features, to be fair its not just their website its so many. I wonder if having a different website which is just geared up to show you everything, like everytime you hover over something it just explains everything. I think if your artistic then things are gonna click, but if your new then even words like stylize, AR, blend, can be confusing
Dang that's some really good feedback. I honestly think you should write that as a message in one of the Discord rooms. I personally think the Website should have like a 'beginner' and 'advanced' mode - where the beginner is super simple, maybe it has like 3 sliders and the aspect ratio - kinda like how it is now. But then you should be able to click 'advanced' mode where it shows you every other option available. all of them haha
I let my membership expire because although the pictures are beautiful they are rarely specific enough for my needs. Eg a client asks for a 40's photograph of a naval officer shaking hands with an engineer in front of docked submarines. Dalle3 is pretty close to accurate and costs nothing.
there are a few hacks, you can take them dalle pictures and try to emulate them in midjourney by using image prompt and style prompt set --iw 3 and use --style raw. i find ideogram is even better for prompts, i sometimes take them and use them in MJ. I left chat gtp because of the ridiculous censorship.Useless if your into horror
💯 obviously that will improve in the future. But in terms of business, I could see Dalle being way more helpful at the moment
I'm feeling like the 910 or so artist styles that now come with stable diffusion 1.5 and also work with Excel might be generally Superior to these random seeming seed based styles but it still adds a lot of versatility to the model. Anyone have any thoughts on this? It would be interesting to use SDXL generations with these styles as style references for MJ.😊
I personally like any 'random' feature! It's just fun to see what the bot comes up with haha
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!
can i change my picture to 1990 art?
WOAH, nice! Now i can more efficiently make AI images (Not art, dont call it that thats so fucking disrespectful) and steal and sell fake art! (It is stealing, AI trains off of other peoples art WITHOUT THEIR CONSENT, its not inspired it LITERALLY copys actual art, fuck off)
Do you offer paid support? I'm looking for help with a specific midjourney project, which is to produce images in a given style using a reference image.
I have a couple of hours each week dedicated to consulting - check out my calendar and see if my schedule works for you! calendly.com/futuretechpilot/1-on-1-session - use that link if you'd like to pay with PayPal calendly.com/futuretechpilot/1-on-1-stripe - use that link if you'd like to pay with Stripe
Thanks for this review! While Midjourney is good, I think Stylar offers better AI tools. Would love your perspective on it!
maaaybe
Great stuff here! I switched to Stylar recently because their AI features felt more advanced. Maybe you can check it out too?
Your videos are always so informative! I'm curious how Stylar would hold up in your tests since I've switched and found it quite impressive.
🤝