Alright, mastering prompts is a quite a deep subject, but it's very likely going to be the most important skill you master going forward into a future of AI-Art, hope this helps :)
the amount of time I have typed in lense sizes,"trending on artstation/pixiv" has become almost second nature.Also I find layering the subject last and the context first with medium "3d,photograph commercial"etc in h middle works best for iterating . I also find mixing lights used in 3d engines octane,redhshift in regular photograph promts give them better overall lighting especially when you add "realistic,ultra detail,4-8k".Its strange I have gone from word salad to pretty much abandoning pinterest .Even if you do not like ai art not learning how to source better or not integrating it into your workflow is career suicide at this point.
I'm gonna try out some of your tips, thanks for sharing. Also I understand what you mean by subject last and medium in the middle but can you give an example of what you mean by "context first"? Is context the setting in which your subject is in?
@@BaoNguyen-iv1zx sure,it depends on which service you are using I have tried this mostly in midjourney,here is an example lets say I want a girl in a cyberpunk world for a film poster.so context first,then mediam in he middle,then subject it would look like this: #1 context(were you describe as much as you can)midnight,cold empty world with brazil shanty houses in the distance a futuristic slum with a little girl center of the composition #2nd medium(here you narrow down how you want it too overall look,you ad what medium you want it,but add context to the medium you want it so if you want it in 3d you use terms like "blender,made in maya,substance painter "etc)- vertical film poster,bordless ,printed on old paper,silhouettes of characters,action movie poster,realistic characters and environment,high resolution,4k,cyberpunk title at top,film credits at bottom of poster 28 pts #3 subject-slender girl standing in the middle of poster,yellow jacket,back tuned to camera,full body visible,center of composition,,shnty houses in the far distance,shanty houses are well lit. from here you render and see what works,takes a bit of tweaking,but I find with this formula you can easily read what limitations or data is needed to get a better picture and adjust each part of the 3 part chain to refine your search.I have found it usually reads the front ,forms how it will overall look with the front and second lines and uses the remaining words for detail.hope this helps
It's like being a librarian. Sure, you go to the library and the knowledge is in the books, but a good librarian can tell you how to find what you want. (Am I dating myself? Has search AI replaced librarians yet?) Now, having watched it, I must admit this is hugely dense information, and I'm laughing at how much you can change the images just by knowing what to ask for. Now I'm going to drop your entire playlist on everyone I know who does visual advertising as a business.
Man, watching as a casual digital image-maker (usually draw, trying to get into 3d-hence finding this channel!), I think this stuff is fascinating. Started watching these as something of a morbid curiosity, initially not thinking of AI ~art (AI-generated imagery, whatever you want to call it) as more than a novelty, but there’s clearly SOME degree of skill involved in making it-even if the type of skill is more akin to an image search than physical dexterity. And I /would/ argue there’s skill, or at least specialized knowledge, involved in database- or in this case algorithm-searching; for instance, finding references for odd poses or objects can be a /huge/ time-suck-and even outside the realm of images, compare “how do i become an artist quickly” to “beginner drawing tutorial fundamentals [with a time filter for, say, under 10 minutes]”. Like, look at all the words in the final query here; you need to do SOME research to know the camera specifics (30mm?), /why/ it’s important to specify it… /I/ can discern why the “trending on Nike website” part might be important, because I know a bit about commercial web design; web designers need to account for mobile devices, so the images need to be even bolder and simpler than even a poster (size is set, can be huge) or commercial (implies video, things can move in/out of focus). But I have no clue why 30mm is the camera setting of choice! You’d need, I assume, at /least/ cursory know-how of advertising photography for that. Anyway, especially after reading some of the comment backlash (which has more to do with the mere /concept/ of AI than anything actually being presented in a How To Use AI Tutorial-valid concerns, but imo kiiind of besides the point?), just wanted to pop in and say it’s really cool to see how this stuff actually works. You’ve definitely got me reconsidering my initial dismissal of AI-generated imagery and the ~nature of art~ and all that philosophical stuff, even if that’s /also/ besides the point :P Genuinely curious to see what the rest of the series has in store, and thanks for making such concise tutorials in general!
Your pro prompt words and attributes you use to describe your target - it might be that the system doesn't even contain those terms and words, so some might have simply a placebo effect. It would be good to be able to search all the contained keywords, labels, attributes and such (anything stored in text format by the system).
PLEASE, can you add descriptions for each part of the formula? Like, how to define lense, source, setting, destination in reference to this situation. Thanks!
Yes, I will be going over SD in later videos comparing it to Dall-E. I'm just going through Dall-E first, to make it easier to see the real differences between the two software -
Is there a way to make small adjustments to a piece to DALL-E art work? Like Color, Text Size, or Image Aspects? It always wants to generate completely different images with no consideration to the prompts to leave something alone. I might like 95% of the image but it is spelled something and I need to correct it.
My brain working overtime and dusting off all the English books I have stored in the baaack part of my memory trying to figure out how to be incredibly specific and write in complete sentences
Hi, I saw this video as I was struggling to come up with prompts. My thing is I have a skincare brand and I want great images for my products but this program doesn’t make my products look the same or even have the right name on there. Basically not useable and I’m just here wasting credits. Any help would be great full
fuuu how much did they pay him to promote it depends :) this Ai they should make him do something usefull like make retopology, so we dont have to and leave the rest to profesionals. fraking joke. Waste of time for total noobies who think they jump themselves into industry by "prompting" XD
i dont think its fair to call them an ai artist, they can be an ai art director, or appropriately an ai prompt engineer- but they did not create any art they just organized the prompt to the customers concept.
Instead of making AI for important tasks, like putting one to try and come up with an affordable cure for cancer they instead decided to ruin artist's careers. Geniuses
What do you mean "instead of"? AI for earlier detection of cancer existed long before this, AI for protein folding (which makes better medicine) already exist, AI for controlling fusion reactors (cleaner, safer, cheaper energy) are already being used to get better performance than we've seen in decades. Communicating with AI is going to become important over the next 20 years, and these image generators help people see how current AI interpret everyday language. Yeah, it'll scare some artists the same way the camera did - but those who aren't scared of it will learn the strengths and weaknesses of both techniques, know when to use each, and be better artists for it.
what do you mean "AI artist" ? you are typing on a search bar and adding words till you get what you want and unless you made the AI yourself, you are not "AI artist" unless thats like a slang i dont know about and why would people pay for haver thing other people do a google search and then pay for someone else to make the actual poster, why not just pay the guy that makes the actual poster? if the guy wants to use the AI as a tool you shouldn't pay extra for that
You saw step one, and assumed it was the whole process. That's like assuming all a photographer does is press the shutter, ignoring their skill in picking the lighting before that or retouching pictures afterwards.
@@jnjairo thats what i say if the artist use it as a tool, why would you pay more for that? thats its the artist choosing to use a tool, you are supose to pay for the concept art by the artist, not the ai generated art, also, this its free ? cuz if it is, why is people paying for other people using a searchbar?
@@remarit Just a few minutes of looking at the Stable Diffusion Reddit is enough to see that many of the pics the AI generates go through different clean up or post processes, many using other AI. Look closely at this video and you'll notice how the text on the posters is complete nonsense. AI Art is just like using a camera. You can be naive and say photographers aren't artists because anyone can push a button, but you're just cheating yourself out of understanding what's needed to make good images by doing so.
@@travissmith5994 i dont know what its "the Stable Diffusion" but it looks more like people trying to program an AI than just using one. and no i will never compare having to learn composition, ligthing, contrast and more, to using a searchbar
Royal Skies, I have a bit of constructive criticism. I love you man, but you are not an artist, you are a tool user. You are always more fascinated with what tools can do for you then understanding the actual art itself. You wanted to get into making characters. If you were a true artist, you would have gotten into human anatomy and made a series about that. Thorough knowledge of human anatomy is a difference between a pro character artist and a noob. Instead, you made a videos about Z-Brush (a tool), about buying props (tools), about AI generated art (tool). Forget about tools for a bit and start thinking more about what makes good art good. What do good character artists know (besides using tools) that makes them such good artists. A good character artist can make better character in Microsoft Paint then a noob can in Z-Brush or on DALL·E. This whole AI series is ridiculous, I don't want to be AI user, I want to be an artist.
cool bro but the world doesn't work like this... if you take 5 hours to paint something that i can make in 30 min using AI them i will win in the market artists will still be there but in a different way. you need to understand that using a tool for art is not cheating, because there are a billion ways to create art and am ai which doesn't steal, is one of them
@@FM_GOBi lol you seem to know me well ... Also again... Not my fault that things change and I'd rather be on top of it than to rant about the damn technology stealing my job in 20 years
Hi Gobi, Sure no problem - It really looks like you put genuine effort and thought into your comment, so I'd like to give you a genuine response. I think the true answer is we simply have different definitions of artist. For me, an artist is anyone who brings an idea into the real world. Whether it's a poem, painting, invention, music, animation - Art, fundamentally, is just an idea brought into reality, and recorded with color, sound, rhyme, and rhythm. The tools used to achieve this always change, but the result is what I would call the art. There's nothing wrong with doing everything by hand. But, that simply isn't an option for most of us who create art for a living. Your boss pays you by the hour. When you chose not to use tools that can save you time, you are wasting your boss's time and money. And, I firmly believe wasting people's resources is a bad thing. When you're doing art professionally, not using all the best tools available simply isn't an option - If you can't produce as much art as your competitors, you end up fired, homeless, and hungry. And, I don't want to see anyone in my community suffer that fate. I want to make sure you guys are armed with this information early, and have as much opportunity as possible to use it to your advantage - That's my mission, and I hope I was able to clear where I stand on the matter. I really do appreciate you taking the time to watch and leave comments like this - Even if you disagree with me, you're always welcome here on the channel - Hope you have a Fantastic Day Gobi - -RS-
@@TheRoyalSkies Fantastic day to you as well. Disagreements are a good thing. But I also stand by what I said. And of course, I'll keep watching the channel, it's a great source of information.
@@FitnessSheriff Just with humble 4Gbytes of RAM, I can make images of 1024 x 576 (proportional to a wallpaper) or 768 x 768. However, I can easily upscale these with IA tools as well.
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
Alright, mastering prompts is a quite a deep subject, but it's very likely going to be the most important skill you master going forward into a future of AI-Art, hope this helps :)
just ask chatGPT to write the prompt
Meanwhile here I am creating images of animals wearing Fedoras and trench coats.
this is the best use of ai
Me with a "Frog wearing a business coat"
Are you planning on covering any alternatives to Dall-E in this series? Like Stable Diffusion ?
Yes, Ill be covering them in later videos :)
the amount of time I have typed in lense sizes,"trending on artstation/pixiv" has become almost second nature.Also I find layering the subject last and the context first with medium "3d,photograph commercial"etc in h middle works best for iterating . I also find mixing lights used in 3d engines octane,redhshift in regular photograph promts give them better overall lighting especially when you add "realistic,ultra detail,4-8k".Its strange I have gone from word salad to pretty much abandoning pinterest .Even if you do not like ai art not learning how to source better or not integrating it into your workflow is career suicide at this point.
I'm gonna try out some of your tips, thanks for sharing. Also I understand what you mean by subject last and medium in the middle but can you give an example of what you mean by "context first"? Is context the setting in which your subject is in?
@@BaoNguyen-iv1zx sure,it depends on which service you are using I have tried this mostly in midjourney,here is an example lets say I want a girl in a cyberpunk world for a film poster.so context first,then mediam in he middle,then subject it would look like this: #1 context(were you describe as much as you can)midnight,cold empty world with brazil shanty houses in the distance a futuristic slum with a little girl center of the composition #2nd medium(here you narrow down how you want it too overall look,you ad what medium you want it,but add context to the medium you want it so if you want it in 3d you use terms like "blender,made in maya,substance painter "etc)- vertical film poster,bordless ,printed on old paper,silhouettes of characters,action movie poster,realistic characters and environment,high resolution,4k,cyberpunk title at top,film credits at bottom of poster 28 pts #3 subject-slender girl standing in the middle of poster,yellow jacket,back tuned to camera,full body visible,center of composition,,shnty houses in the far distance,shanty houses are well lit. from here you render and see what works,takes a bit of tweaking,but I find with this formula you can easily read what limitations or data is needed to get a better picture and adjust each part of the 3 part chain to refine your search.I have found it usually reads the front ,forms how it will overall look with the front and second lines and uses the remaining words for detail.hope this helps
I’m literally brand new to this and you couldn’t have explained this better. Much appreciated
Man thank you so much for keeping up with thia series. This is such a game changer dude.
It's like being a librarian. Sure, you go to the library and the knowledge is in the books, but a good librarian can tell you how to find what you want. (Am I dating myself? Has search AI replaced librarians yet?)
Now, having watched it, I must admit this is hugely dense information, and I'm laughing at how much you can change the images just by knowing what to ask for. Now I'm going to drop your entire playlist on everyone I know who does visual advertising as a business.
Man, watching as a casual digital image-maker (usually draw, trying to get into 3d-hence finding this channel!), I think this stuff is fascinating. Started watching these as something of a morbid curiosity, initially not thinking of AI ~art (AI-generated imagery, whatever you want to call it) as more than a novelty, but there’s clearly SOME degree of skill involved in making it-even if the type of skill is more akin to an image search than physical dexterity. And I /would/ argue there’s skill, or at least specialized knowledge, involved in database- or in this case algorithm-searching; for instance, finding references for odd poses or objects can be a /huge/ time-suck-and even outside the realm of images, compare “how do i become an artist quickly” to “beginner drawing tutorial fundamentals [with a time filter for, say, under 10 minutes]”.
Like, look at all the words in the final query here; you need to do SOME research to know the camera specifics (30mm?), /why/ it’s important to specify it… /I/ can discern why the “trending on Nike website” part might be important, because I know a bit about commercial web design; web designers need to account for mobile devices, so the images need to be even bolder and simpler than even a poster (size is set, can be huge) or commercial (implies video, things can move in/out of focus). But I have no clue why 30mm is the camera setting of choice! You’d need, I assume, at /least/ cursory know-how of advertising photography for that.
Anyway, especially after reading some of the comment backlash (which has more to do with the mere /concept/ of AI than anything actually being presented in a How To Use AI Tutorial-valid concerns, but imo kiiind of besides the point?), just wanted to pop in and say it’s really cool to see how this stuff actually works. You’ve definitely got me reconsidering my initial dismissal of AI-generated imagery and the ~nature of art~ and all that philosophical stuff, even if that’s /also/ besides the point :P Genuinely curious to see what the rest of the series has in store, and thanks for making such concise tutorials in general!
Your pro prompt words and attributes you use to describe your target - it might be that the system doesn't even contain those terms and words, so some might have simply a placebo effect. It would be good to be able to search all the contained keywords, labels, attributes and such (anything stored in text format by the system).
Bro you saved my butt again! You are the real MVP! 😻😻😻
I'm shocked, this is awesome.
PLEASE, can you add descriptions for each part of the formula? Like, how to define lense, source, setting, destination in reference to this situation. Thanks!
Thanks man. That was very useful 👍🏻
Finally, now I too can have nice shoe!
1:58 Do you have a variety of what lens size to put in?
Will you cover Stable Diffusion at some point? I mean it's open source, that alone is a guarantee watch for 80% of Blender users already.
Yes, I will be going over SD in later videos comparing it to Dall-E. I'm just going through Dall-E first, to make it easier to see the real differences between the two software -
I found this by accident right now and THIS changes SO MUCH omg thank you - can you recomend more books/ blogs to this topic ?
Stable diffusion !
Merci pour cet aperçu assez complet 😊
Great video :)
soo? What are we supposed to do with the messed up text on the ad? Use another AI to remove the junk?
damn, you are the best so far to explain what I need! big thanks
Great video. How do you overcome the spelling mistakes?
THANK YOU!!!!!!!!
how do you utilize negative prompts?
like, how do you activate them?
Nice! More prompt videos please! also later doing some Midjourney stuff would be nice
Gotta break out my old jr. college thesaurus😂
Best tutorials on RUclips
Is there a way to make small adjustments to a piece to DALL-E art work? Like Color, Text Size, or Image Aspects? It always wants to generate completely different images with no consideration to the prompts to leave something alone. I might like 95% of the image but it is spelled something and I need to correct it.
Is it possible to change product colors for variations with Dalle E 2?
Is it the longer prompt the better ?
Usually, unless you just want something realistic, or something cute -
How do you fix the Text into something you want?
niceee !!! thanks !!
omg great video
I literally can’t tell if he changed back to the “made up” voice or it’s just his normal voice is so similar to the one used for the past years 😂
Incredible Video! thank you so much for opening up my understanding!
Finally finished my PhD in prompt engineering. Wished I had seen this video 10 years ago, would have saved me 180k. #forgivestudentloans
I want to learn from you. Teacher!
My brain working overtime and dusting off all the English books I have stored in the baaack part of my memory trying to figure out how to be incredibly specific and write in complete sentences
Hi, I saw this video as I was struggling to come up with prompts. My thing is I have a skincare brand and I want great images for my products but this program doesn’t make my products look the same or even have the right name on there. Basically not useable and I’m just here wasting credits. Any help would be great full
I wonder if they'll make dalle for music
there are ai music programs but i couldn't find one that didn't cost money :/
@@TheCatMurgatroyd well son, nothing is free in this world unless you think like Bernie Sanders..
@@barabimbaraboom7830 well... There's blender, some AI generators and so on.. Also yes I'm a socialist ofc
there is some AI youtube music channel that has been streaming like 10 years of non stop made up music / death metal etc...its not very good.
What can we do about the fact the writing comes out in Gibberish?
Can I somehow enter my company logo in so I can create content for social media?
How do you get rid of the crap words that make no sense
bro how many dall e videos are you gonna make?
fuuu how much did they pay him to promote it depends :) this Ai they should make him do something usefull like make retopology, so we dont have to and leave the rest to profesionals. fraking joke. Waste of time for total noobies who think they jump themselves into industry by "prompting" XD
This is a fucking great video
I typed exactly the same prompts and mine looks like trash, like how?
*Bro sounded sarcastic*
I would love to see a video that explains how to make violent or racist AI art with DALL3
i dont think its fair to call them an ai artist, they can be an ai art director, or appropriately an ai prompt engineer- but they did not create any art they just organized the prompt to the customers concept.
few word make simple pic
many word make better pic
beautifully said 🥲
So many "intelligent" people in the comments today
That's great, but how to resolve spelling mistake. That's the major problem 😔
Self-proclaimed "pro"...
Amazing advetisement, can't even read the words lmao
Instead of making AI for important tasks, like putting one to try and come up with an affordable cure for cancer they instead decided to ruin artist's careers. Geniuses
What do you mean "instead of"? AI for earlier detection of cancer existed long before this, AI for protein folding (which makes better medicine) already exist, AI for controlling fusion reactors (cleaner, safer, cheaper energy) are already being used to get better performance than we've seen in decades.
Communicating with AI is going to become important over the next 20 years, and these image generators help people see how current AI interpret everyday language. Yeah, it'll scare some artists the same way the camera did - but those who aren't scared of it will learn the strengths and weaknesses of both techniques, know when to use each, and be better artists for it.
It’s all the same AI tech being applied to different fields
Algorithm
what do you mean "AI artist" ? you are typing on a search bar and adding words till you get what you want and unless you made the AI yourself, you are not "AI artist"
unless thats like a slang i dont know about
and why would people pay for haver thing other people do a google search and then pay for someone else to make the actual poster, why not just pay the guy that makes the actual poster? if the guy wants to use the AI as a tool you shouldn't pay extra for that
You saw step one, and assumed it was the whole process. That's like assuming all a photographer does is press the shutter, ignoring their skill in picking the lighting before that or retouching pictures afterwards.
@@travissmith5994 if you sell "Ai art" you are not making any retouch, and you know that
@@jnjairo thats what i say if the artist use it as a tool, why would you pay more for that? thats its the artist choosing to use a tool, you are supose to pay for the concept art by the artist, not the ai generated art, also, this its free ? cuz if it is, why is people paying for other people using a searchbar?
@@remarit Just a few minutes of looking at the Stable Diffusion Reddit is enough to see that many of the pics the AI generates go through different clean up or post processes, many using other AI. Look closely at this video and you'll notice how the text on the posters is complete nonsense.
AI Art is just like using a camera. You can be naive and say photographers aren't artists because anyone can push a button, but you're just cheating yourself out of understanding what's needed to make good images by doing so.
@@travissmith5994 i dont know what its "the Stable Diffusion" but it looks more like people trying to program an AI than just using one.
and no i will never compare having to learn composition, ligthing, contrast and more, to using a searchbar
🤣🤣
is this satire?
Royal Skies, I have a bit of constructive criticism. I love you man, but you are not an artist, you are a tool user. You are always more fascinated with what tools can do for you then understanding the actual art itself.
You wanted to get into making characters. If you were a true artist, you would have gotten into human anatomy and made a series about that. Thorough knowledge of human anatomy is a difference between a pro character artist and a noob. Instead, you made a videos about Z-Brush (a tool), about buying props (tools), about AI generated art (tool).
Forget about tools for a bit and start thinking more about what makes good art good. What do good character artists know (besides using tools) that makes them such good artists. A good character artist can make better character in Microsoft Paint then a noob can in Z-Brush or on DALL·E. This whole AI series is ridiculous, I don't want to be AI user, I want to be an artist.
cool bro but the world doesn't work like this... if you take 5 hours to paint something that i can make in 30 min using AI them i will win in the market
artists will still be there but in a different way. you need to understand that using a tool for art is not cheating, because there are a billion ways to create art and am ai which doesn't steal, is one of them
You have the exact same problem like Royal Skies. You are not an artist. You are a tool user. Artists enjoy making art, not acquiring art.
@@FM_GOBi lol you seem to know me well ... Also again... Not my fault that things change and I'd rather be on top of it than to rant about the damn technology stealing my job in 20 years
Hi Gobi, Sure no problem - It really looks like you put genuine effort and thought into your comment, so I'd like to give you a genuine response. I think the true answer is we simply have different definitions of artist. For me, an artist is anyone who brings an idea into the real world. Whether it's a poem, painting, invention, music, animation - Art, fundamentally, is just an idea brought into reality, and recorded with color, sound, rhyme, and rhythm. The tools used to achieve this always change, but the result is what I would call the art.
There's nothing wrong with doing everything by hand. But, that simply isn't an option for most of us who create art for a living. Your boss pays you by the hour. When you chose not to use tools that can save you time, you are wasting your boss's time and money. And, I firmly believe wasting people's resources is a bad thing.
When you're doing art professionally, not using all the best tools available simply isn't an option - If you can't produce as much art as your competitors, you end up fired, homeless, and hungry. And, I don't want to see anyone in my community suffer that fate. I want to make sure you guys are armed with this information early, and have as much opportunity as possible to use it to your advantage -
That's my mission, and I hope I was able to clear where I stand on the matter. I really do appreciate you taking the time to watch and leave comments like this - Even if you disagree with me, you're always welcome here on the channel -
Hope you have a Fantastic Day Gobi -
-RS-
@@TheRoyalSkies Fantastic day to you as well. Disagreements are a good thing. But I also stand by what I said. And of course, I'll keep watching the channel, it's a great source of information.
CRINGE
bro id wish you would made this video sooner... already lost all my credits 🥲
I lost all my credits, so I gave up on Dall-E and started playing with the AWESOME Stable Diffusion UI. :)
@@leandrodfcorreia2 yeah that is SO good too, I like Stable diffusion
@@leandrodfcorreia2 how much RAM you running for how big of a dimension image? (pixels)
@@FitnessSheriff Just with humble 4Gbytes of RAM, I can make images of 1024 x 576 (proportional to a wallpaper) or 768 x 768.
However, I can easily upscale these with IA tools as well.
Prompt Engineer => lexica.art