I took the prompt and made a Poe bot for it. Now i can use chatGPT or any other LLM to compare results. Great video. Thank you for your thorough walkthroughs.
I love using both apps thank you for connecting the two. One of my biggest hurdles was prompting. I am a visual thinker. Emphasis on the visual, so ,effective communication is not a strength of mine. Thanks to your tutorial I can now effectively prompt the AI to bring my vision from imaginary to reality in stunning accuracy. Thank you bro!
Nuance happens in the narrative, in the frame accurate cutting. Where one frame too early or too late will make or a break a scene. Then there is sound. You want to be AI movie star? Learn Photography. Learn lighting. Then use those tools. Stop convincing yourself that AI brings you creative talent. Everyone has a pretty high quality camera in their phone for the last 5-10 years. What have you done with that? Worked on your composition? Found that exposure lock button on your screen?
Bro , I wanted to a short film, after a few phone calls I give up , emails and enquiries. Before I was even onto a script I was looking at 10k. Went from 10k (5 years ago) to $100 with AI subscriptions. You cant knock it.
Thank you so much for this - especially the prompt guide for ChatGPT. I'm an author, and my dream is to make my fantasy worlds come to life. This has really helped me find the right path.
Congratulations on this excellent walk through. One question though : when I upload the image, there is no need anymore to describe in detail the character eg. Age, hair color, clothing etc etc for consistency in every scene?
Thank you so much for the tutorial! It was really helpful and detailed in explaining Runway’s features. Just a bit of additional insight - while Runway is excellent for several aspects, especially commercial products, there’s still some room for improvement in terms of dynamic motion, which sometimes feels a bit artificial, particularly when applied to portraits. Compared to platforms like Kling or Minimax, Runway’s motion still lacks a bit of the natural flow needed for more dynamic results. Hopefully, this is something that will improve in the future. Thanks again for the great tutorial!
🎯 Key points for quick navigation: 00:00 *🎥 Introduction to Combining AI Tools* - The video presents the integration of ChatGPT and Runway for creating cinematic videos. - Breaking down effective prompt writing using ChatGPT, - Emphasizing the versatility of the approach across various AI video generators, - Importance of detail in shooting prompts for realistic outputs. 00:54 *✏️ Crafting Effective Prompts* - Discussing the elements needed to write detailed prompts for video generation. - Key components include subject, scene, camera motion, and aesthetic intent, - The significance of cinematic elements like lighting and camera specifications, - Availability of a free prompting guide to assist users. 05:05 *📈 Enhancing Productivity with AI* - Introduction of HubSpot's AI guide for implementing AI in business workflows. - Features include AI integration techniques and productivity enhancement tips, - Offers a decision flowchart for AI usage in business contexts, - Importance of keeping AI-generated content sounding natural. 12:04 *🎨 Advanced Techniques for Unified Aesthetic Prompts* - Exploring methods to maintain stylistic consistency across video shots. - Techniques involve choosing aesthetic elements like lighting and color palettes, - Ability to generate multiple prompts from a single concept for cohesive film production, - Importance of combining AI art and video prompts for enhanced outputs. 15:16 *🌟 Examples and Applications of AI Video Creation* - Showcasing advanced AI-generated content in marketing and cinema. - Highlighting specific visual techniques such as color palettes and camera movement, - Demonstrating contrasts in visual storytelling through AI-generated scenes, - Encouraging exploration of various AI art and video models for creative projects.
Bro your link requires my personal information like my phone number to download the pdf, is not cool at all. I don't mind giving an email but the rest is a no.
Thank you for your video ♥ The renderings are fantastic! But one thing is perhaps missing: the "memory" of the characters created for example. If we want to do something more developed than a clip or a trailer, if we want to keep "the same characters" for this time all the time of a series episode or a film, it is currently not yet possible right?
Is it possible to achieve character consistency throughout a sequence of shots without using img2video? Like uploads a photo reference of the likeness of a chacter or something of the sort? Thanks!
Al Samson has mastered the medium in its current state. These samples are examples of the best that AI can produce at this point in time. Yet, they all look very contrived and are immediately recognizable as AI. There is a very long way to go before any AI generated material is actually usable in a commercial setting.
@@phadrus I have no idea. That's why I used "very long" instead of X number of years. But, it will be years for a couple of reasons. Realistic visual reproduction needs to reach a point where it deals convincingly with the fractal elements of both nature and man made objects. And, the on screen "talent" in AI will have to be able to take direction and react to that direction to the same degree as present day live actors.
This video does an amazing job explaining how to use ChatGPT and Runway together to create truly cinematic video prompts. I love the detailed breakdown of how to define subjects, scenes, and camera settings for realistic, immersive visuals. The free prompting guide is a fantastic bonus for anyone looking to improve their AI-generated videos!
A bit off-topic, but for me, when I see all these "perfect", "ideal", photorealistic people being generated (7:43, 1:35, etc.), it doesn't engender any emotion or feeling in me. I just feel, meh. Yet there have been outright _cartoons_ (that are not at _all_ photorealistic) which have actually brought me to _tears._ Not sure of all the psychology behind it, but I know that's how these things make me feel (or not feel).
is there anyway we can get same characters in all prompt outputs, this randomness is not allowing to create anything serious yet but random experiments.
Thank you for this very interesting and usefull Video! There is one thing that i alway wondered: Why do these generators almost always genereate Footage in Slow Motion? Is there any posibility to let them generate in "normal" 24fps-velocity?
📊Hai Ai Samson, in the future, could you please use 👉 “god mode” ‘n’ not “God Mode” The capital letter G is appropriately utilize for the Almighty God , the Grand Creator of the universe ❤🎉 Please take note 📝🙏
All this power, technical prowess, forward thinking and ability to have AI develop and/or analyze scripts, storyboards, prompts and final editing goals; but no (0) 👌"goose egg" ability to add CHAPTERS to your own 19+ minute video. (Hint: feed an .MP3 copy of the audio of your final edit to Notebook LM and have it cough up your timecode chapters.)🙄🤘 ...Unless the goal is to just intentionally wear out the viewer🥱 and fail to allow them to more efficiently use the video as a "quick reference" when they want to review or repeat any given advice or section. (Such oversights; weather intentional or not; definitely restrict your potential subscriber base opportunities.)
First thank you, Samson, for this awesome tutorial, second the number one problem for text to video is consistent characters which you didn't mention at all? I hope you try to fix it because text to video is much better than image to video, at least for minimax.
![Chris Langan - The Interview THEY Didn't Want You To See - CTMU [Full Version; Timestamps]](/img/n.gif)
I took the prompt and made a Poe bot for it. Now i can use chatGPT or any other LLM to compare results. Great video. Thank you for your thorough walkthroughs.
Thanks for commenting this. I hadn’t come across Poe and it looks very interesting.
Poe bot? Interesting
I love using both apps thank you for connecting the two. One of my biggest hurdles was prompting. I am a visual thinker. Emphasis on the visual, so ,effective communication is not a strength of mine. Thanks to your tutorial I can now effectively prompt the AI to bring my vision from imaginary to reality in stunning accuracy. Thank you bro!
Nuance happens in the narrative, in the frame accurate cutting. Where one frame too early or too late will make or a break a scene. Then there is sound. You want to be AI movie star? Learn Photography. Learn lighting. Then use those tools. Stop convincing yourself that AI brings you creative talent. Everyone has a pretty high quality camera in their phone for the last 5-10 years. What have you done with that? Worked on your composition? Found that exposure lock button on your screen?
Bro , I wanted to a short film, after a few phone calls I give up , emails and enquiries.
Before I was even onto a script I was looking at 10k.
Went from 10k (5 years ago) to $100 with AI subscriptions.
You cant knock it.
Thank you so much for this - especially the prompt guide for ChatGPT. I'm an author, and my dream is to make my fantasy worlds come to life. This has really helped me find the right path.
Wow... this was like a mini course for free with so much detail... thanks for sharing Al including your downloads... awesome!!
Thank you so much for this - especially the prompt guide for ChatGPT . This has really helped me find the right path.
Congratulations on this excellent walk through. One question though : when I upload the image, there is no need anymore to describe in detail the character eg. Age, hair color, clothing etc etc for consistency in every scene?
If the character in the image remains consistent, then no
Love the channel. My question is where is Sora? Why are we still crossing over between the LLMs?
Thank you so much for the tutorial! It was really helpful and detailed in explaining Runway’s features. Just a bit of additional insight - while Runway is excellent for several aspects, especially commercial products, there’s still some room for improvement in terms of dynamic motion, which sometimes feels a bit artificial, particularly when applied to portraits. Compared to platforms like Kling or Minimax, Runway’s motion still lacks a bit of the natural flow needed for more dynamic results. Hopefully, this is something that will improve in the future. Thanks again for the great tutorial!
🎯 Key points for quick navigation:
00:00 *🎥 Introduction to Combining AI Tools*
- The video presents the integration of ChatGPT and Runway for creating cinematic videos.
- Breaking down effective prompt writing using ChatGPT,
- Emphasizing the versatility of the approach across various AI video generators,
- Importance of detail in shooting prompts for realistic outputs.
00:54 *✏️ Crafting Effective Prompts*
- Discussing the elements needed to write detailed prompts for video generation.
- Key components include subject, scene, camera motion, and aesthetic intent,
- The significance of cinematic elements like lighting and camera specifications,
- Availability of a free prompting guide to assist users.
05:05 *📈 Enhancing Productivity with AI*
- Introduction of HubSpot's AI guide for implementing AI in business workflows.
- Features include AI integration techniques and productivity enhancement tips,
- Offers a decision flowchart for AI usage in business contexts,
- Importance of keeping AI-generated content sounding natural.
12:04 *🎨 Advanced Techniques for Unified Aesthetic Prompts*
- Exploring methods to maintain stylistic consistency across video shots.
- Techniques involve choosing aesthetic elements like lighting and color palettes,
- Ability to generate multiple prompts from a single concept for cohesive film production,
- Importance of combining AI art and video prompts for enhanced outputs.
15:16 *🌟 Examples and Applications of AI Video Creation*
- Showcasing advanced AI-generated content in marketing and cinema.
- Highlighting specific visual techniques such as color palettes and camera movement,
- Demonstrating contrasts in visual storytelling through AI-generated scenes,
- Encouraging exploration of various AI art and video models for creative projects.
What with all the curvaceous, beautiful creations I think Al’s hard drive needs *checking!!!
*Joking obviously.
Or am I?
Peace and love
1:28 that’s cool
thanks!
Thank you so much for sharing this whole process!! You ROCK :)
Thank you Samson. Great video and walk through. Great tempo and informative.
Thanks for this great and useful tutorial, and I learned new things.🙂🤝
Bro your link requires my personal information like my phone number to download the pdf, is not cool at all. I don't mind giving an email but the rest is a no.
"...no worries, mate. Your mother's maiden name, and your SS number will work."
That stopped me cold, too. Email and first name. That's it. Switch to this and the mailing list will grow.
Just put in a fake number
@@donnelllowery4376agreenwith this dude. Use textnow. Get a free number. Use that number.
Also your name is Frederico
This is perfect explenation on how to combine ChatGPT for creating a very good and usefull prompt. Keep up your good work Mr. Samson.
Thank you for your video ♥ The renderings are fantastic! But one thing is perhaps missing: the "memory" of the characters created for example. If we want to do something more developed than a clip or a trailer, if we want to keep "the same characters" for this time all the time of a series episode or a film, it is currently not yet possible right?
Bro bringing the magic to us for free! Thanks!
TY! now to add matched voices
Awesome guide
Very helpful. Thanks man! 💯💥
Happy to help!
Is it possible to achieve character consistency throughout a sequence of shots without using img2video? Like uploads a photo reference of the likeness of a chacter or something of the sort? Thanks!
Wow this is so unbelievable so amazing and great quality video
Great video. Really inspiring to try this out. Thanks for pushing the limits.
Beautiful content!
My Master! ❤
Thanks
Are you AI generated? (sic) ;) Great presentation.
Seems good but... runaway, pika, and luma dreams (to mention some of them) just take a life to create 4s videos (at least for me).
Genius ❤
It's amazing - but not currently amazing enough for me to use it in a production environment or final product.
Al Samson has mastered the medium in its current state. These samples are examples of the best that AI can produce at this point in time. Yet, they all look very contrived and are immediately recognizable as AI. There is a very long way to go before any AI generated material is actually usable in a commercial setting.
How long would you say is “very long”, just curious?
@@phadrus I have no idea. That's why I used "very long" instead of X number of years. But, it will be years for a couple of reasons. Realistic visual reproduction needs to reach a point where it deals convincingly with the fractal elements of both nature and man made objects. And, the on screen "talent" in AI will have to be able to take direction and react to that direction to the same degree as present day live actors.
Very informative and good examples! Can you keep a character consistent in sequence of shots in a scene to another scene?
...so, 6 mths or so?
Is anyone else having problems copying the prompt guide from notion?
The perfect tutorial
This video does an amazing job explaining how to use ChatGPT and Runway together to create truly cinematic video prompts. I love the detailed breakdown of how to define subjects, scenes, and camera settings for realistic, immersive visuals. The free prompting guide is a fantastic bonus for anyone looking to improve their AI-generated videos!
A bit off-topic, but for me, when I see all these "perfect", "ideal", photorealistic people being generated (7:43, 1:35, etc.), it doesn't engender any emotion or feeling in me. I just feel, meh. Yet there have been outright _cartoons_ (that are not at _all_ photorealistic) which have actually brought me to _tears._ Not sure of all the psychology behind it, but I know that's how these things make me feel (or not feel).
is there anyway we can get same characters in all prompt outputs, this randomness is not allowing to create anything serious yet but random experiments.
There are extensive Midjourney tutorials about creating consistent characters.
Must watch on 1.6 speed 🥵
Amazing
Amazing, thank you
My boy thanks !!!🎉🎉
Thanks for watching
Thank you
Thank you for this very interesting and usefull Video! There is one thing that i alway wondered: Why do these generators almost always genereate Footage in Slow Motion? Is there any posibility to let them generate in "normal" 24fps-velocity?
bravo bravo bravo !!!!!!!!!!!!!!!!
📊Hai Ai Samson, in the future, could you please use 👉 “god mode” ‘n’ not “God Mode”
The capital letter G is appropriately utilize for the Almighty God , the Grand Creator of the universe ❤🎉
Please take note 📝🙏
👍
i heart ai
All this power, technical prowess, forward thinking and ability to have AI develop and/or analyze scripts, storyboards, prompts and final editing goals; but no (0) 👌"goose egg" ability to add CHAPTERS to your own 19+ minute video.
(Hint: feed an .MP3 copy of the audio of your final edit to Notebook LM and have it cough up your timecode chapters.)🙄🤘
...Unless the goal is to just intentionally wear out the viewer🥱 and fail to allow them to more efficiently use the video as a "quick reference" when they want to review or repeat any given advice or section. (Such oversights; weather intentional or not; definitely restrict your potential subscriber base opportunities.)
i love it
0:36
The Best part ?
How to get consistent image in text to video prompts?
Can I add my own art to this?
Are you able to use text to video where you can have your own face image?
Plz make fruits turn into animal type trend in it plz
❤❤❤❤ awesome
Just use AI to do all the prompting after feeding it similar videos.
im gettin an error. cant download the guide (((((((
First thank you, Samson, for this awesome tutorial, second the number one problem for text to video is consistent characters which you didn't mention at all?
I hope you try to fix it because text to video is much better than image to video, at least for minimax.
Thanks for the tip! I think using Midjourney for character consistency then img2video is the best flow for this
@@aisamsonreal thanks Samson, Is there any way to make consistent characters in minimax text to video?
why slow motion is it few images in 30frames per second i would like to create fast moving 3d cartoons
rất tuyệt vời
👍
Hey samson olz help
💯 ♥
Those ladies were not life like, more like wish like
My wife looks like that and shes 50
@clifftonicstudios7469 that's because she keep her skin youthful with lots of protein
chatgpt = lame mode gpt sucks compaired to llama
there is ni adidas ad in the link. wrong link
@thatbleachedblonde