Thanks for all the news this whole year and keeping the entire community updated. Your work has been great to follow! Huge thanks and happy festive season!
Only 625 likes? Unbelievable! Thank you for this huge amount of information and multiply the likes with 1000. Best regards from Germany and have a good Christmas time.
I can't invest my time using generative AI to realize scenes from my script until I see a productive level of creative control. I call it the "shot list" test. Generate a shot list for a one-minute scene with dialog. Sort the shots into master and coverage. Generate AI prompts from the shot list. Upload character headshots for at least two characters in a composite scene at various angles. Render a one-minute master shot with reasonable blocking. This doesn't have to be a fight or other intense action shot but should be more than a static two-shot with no camera eye contact. Generate coverage shots from master shot frames of a specific duration. Assemble into a complete scene. Demonstrate consistent characters, costumes, lighting, set design, and color grading across all shots. Add lip-synch to the script dialog. When I can see someone do this in a reasonable amount of time without excessive regeneration, I'll be interested in working with whatever tool(s) works.
I advise you to invest now. When it's finally at production level, you'll need to invest a lot of time learning the tool or pay someone else to use it.
The Xmas Coca-cola commercial backlash is a great example of people’s rejection of carefully cherry-picked AI shots looking just strange and fake enough to cause a mass negative reaction by viewers. Humans notice subtle weirdnesses because our brains are trained from infancy to perceive the specific physics properties of different phenomena.
@@eSKAone- Directors, producers and the Hollywood suits who greenlight productions and VFX budgets notice. I predict that your prediction with be wrong, because the real cost of AI will still be more expensive that the proven older methods... and not as reliable.
It also brought attention to itself with calling itself AI. Plenty of hated commercials out there but mentioning AI can be risky for PR purposes rightn ow.
@@curiousrefuge i think the interesting thing is that as the technology progresses, humans will adapt slowly, without realizing. The disruption this will cause and is already causing to the value of 'traditional content' in the Hollywood model, in my opinion is going to rebirth the industry. Why would anyone 'greenlight' a $60m feature film if you can make something comparative digitally for less than half? Why would 'buyers' continue to pay content license fees at the same rate? Investment in traditional content will continue to drop off because returns at traditional rates will become near impossible.
This is the channel I follow for AI video news. So good and informative style, showing important compatisons. However I start to worry because no new content for 3 weeks 🤔🙂 Fyi: 3 days ago minimax introduced character reference (which is not the best imho)
You missed some very important points about Hunyuan. Firstly that it's open source with the ability to run on consumer hardware. Secondly that it supports LoRAs, allowing consistent characters. With these two aspects taken into account, I think in some aspects it's the best AI video model available right now.
@pixelpilfer What? 300k? Did you even read the "consumer hardware" part? At a minimum it can run on the 12GB RTX 3060. You can get that for a few hundred dollars. The annual cost of Midjourney's standard plan would pay for most of it.
As of making that video, it wasn't very accessible (even though it was open source). But we are aware that this has since bee nchanged and is now more accessible.
So fab I kept losing track of the actual content! Even matches the moustache. Spotless hair + no frills crew neck to top it off. So perfect I doubt it's even real!
I really hope they update veo allow you to import your own images soon. It would just be more beneficial, for example, taking myself and implementing into AI films, but I also hope they learn from the safeguards of ChatGPT and make it less strict but still protective of copyright and the prevention of “bad intentions”
Please have a couple of _tried & tested_ "Image To Video" prompts ready when testing these platforms. It is for the sake of consistency. Otherwise, it's just confusing. We want to see what each platform produces with standardised tests. 🙏
10:42 That's a really cool barista, gets some illusionist tricks going! Spoon just disappears, and then a napkin materalizes right there in his hand. How does he do it?
Why do people think Kling is good? The physics were good but the shots look like crap. The veo 2 generations look like a literal commercial for coffee.
people in the film industry are obviously going to explore AI just like they did digital FX, they always have been at the cutting edge of technology, Actors have always been replaceable as many have found when they start demanding too much, it just means the unknown ones willing to do it for less can be edited to look more the part, I can see closer historical resemblances, better aliens etc with less time attaching uncomfortable prosthetics, makeup etc. Screenwriters who refuse to use AI won't be replaced with AI but with screenwriters who use AI.
HI!!! Thanks again for the great video. I wanted to try Sora and YES I have a PAID sub to GPT, but when I went to sign up, I did not see where I can login if I already have an account with them. Can you help with that please?
I like the video that other AI TH-camr did where he tested out VEO versus SORA and they were both kinda' disappointing a lot of the time. That said, I prefer reality to hype.
@@curiousrefuge I think it was Matt Wolfe. Just plugged in a bunch of prompts from requests on social and compared VEO and SORA. Often both flopped. That more parallels the reality one experiences when trying to make an AI film. More often than not you don't get what you want, or what's usable.
For those who can generate infinite videos, SORA may indeed be good at something, since it is necessary to generate millions of videos. But SORA needs to improve urgently, and content creators need to talk about it.
I got a ton of access to AI tools through the Odyssey competition, but every video generator is still taking 10+ hours to render and I'm not getting any results yet. So frustrating.
Why are people so full of this shit...? Just about anyone now can be a movie producer, director, cinematographer, graphical designer... you name it all... Just from their lazy bedroom. How over saturated the movie industry will be in the near future... Bye bye hollywood and good luck to the movie noobs trying to capitalize their (sh*tty) movies beneath the heap of millions.
Then again, how oversaturated are things now even without mass adoption of these tools? These tools will can certainly be used to create more 'meh', but it will also weed out lots of people.
@@curiousrefuge It's certainly devalued many of the main industries even if they are still essential. I produce historic scenes and animations for museums etc. I have been quite excited to integrate Ai and speed up certain processes and got some pretty decent results. However, my clients think I now just press a big red button and the whole thing is done and therefore quibble the fees. I am still spending just as long on a project I just use Ai to do some of the grunt work. Its going to be messy times as the new phase of this industry finds its feet.
The unwanted „zoom in“ is actually a tracking shot, or a push in handheld. The „multicamera“ is just another shot which could have been done with the same camera. 35mm is a 35mm lens or a film shot with a 35mm film camera or stills or digital?The wording in terms of camera technique and equipment is giving me a hard time, not just in terms to hear from others but also when using myself, because what are terms on an actual set, or the camera department with very clear meaning is not understood well by the ai tools out there, so it’s very hard to tell the tool what to do, what would be in the real world an easy thing to communicate. We would need more people from the film world to teach the tools. On the other hand, do we want to hand out that knowledge this easy?
Sorry, but the generations from Veo 2 with the barista, that you call a "good job", are actually a pretty sloppy job 😂 It is basically unusable, even in a non commercial project. Needless to say the rest. It may be better than Sora in general, but from here to say "good job" for this clip.. I'm sorry. We need to stop the BS and the pink glasses effect when we use AI.
i think we need to keep in mind that ai will never be worse than it is now, this tech is in its infancy and moving quickly. 6 months ago we were fighting for 3s of 'cinematic' footage. Now we've got action sequences and character performance.
200$ a month for surreal sora is not going to fly. Never mind that it's pretty much useless, if you are going to charge real money it better be their best model, not the model they hope can make them a lot of money or is convenient for them. Just charge per usage and offer their best. Then people with real use might use it. 200$ a month pays for a proper computer in only a few of month and that's a lot of compute
The fact that in VEO 2 we can't upload our own images, is a turn off for me at least. this will make it useless, unless you want to create some sporadic videos that have no continuity.
Sora is such a let down after all the hype. Like when it was announced, it seemed amazing, but the industry has since completely blown it out of the water and Sora doesn't seem to have made any progress.
There are problems in every shot example. Small snippets may pass on a phone but every shot shows glaring oddness or at least uncanny valley-like off-putting elements on a large screen.
Sora should only charge credits for videos you download or share. Charging for the gibberish this platform is giving me is absurd when you consider that we are training the model in the process.
make a video on free open source ai video generator, bcoz i m individual freelancer , and i dont have much money to spend , try to get free ai video and it should be free and unlimited
No more than 1.5 seconds of usable footage in all your test examples. AI video is too unpredictable, hallucinationatory and mix different physics speeds in the same footage (example: the dog in the pool’s slo mo is mixed with near realtime speed of the water ripples).
Thanks for all the news this whole year and keeping the entire community updated. Your work has been great to follow! Huge thanks and happy festive season!
Thanks, you too!
Only 625 likes? Unbelievable! Thank you for this huge amount of information and multiply the likes with 1000. Best regards from Germany and have a good Christmas time.
Veo is current clear winner. Can't wait to use.
I had a go at the image generator, its so so censored, i think you'd struggle to make horror or action film.
Incorrect. MiniMax allows for much more freedom, as Veo and the others censor too much.
@@armondtanz Yes, most of them are too censored, except for MiniMax. MiniMax is the winner when it comes to having much less censorship.
@@FilmSpook yea. Had a go of Google image creator. Wouldn't even make an action scene with character and machine gun.
Useless.
Once it has true Image-to-Video it will be ultra powerful
Awesome videos! Thanks for always keeping the community updated!
You bet!
I can't invest my time using generative AI to realize scenes from my script until I see a productive level of creative control. I call it the "shot list" test. Generate a shot list for a one-minute scene with dialog. Sort the shots into master and coverage. Generate AI prompts from the shot list. Upload character headshots for at least two characters in a composite scene at various angles. Render a one-minute master shot with reasonable blocking. This doesn't have to be a fight or other intense action shot but should be more than a static two-shot with no camera eye contact. Generate coverage shots from master shot frames of a specific duration. Assemble into a complete scene. Demonstrate consistent characters, costumes, lighting, set design, and color grading across all shots. Add lip-synch to the script dialog.
When I can see someone do this in a reasonable amount of time without excessive regeneration, I'll be interested in working with whatever tool(s) works.
Wait a few months
this!
Yep
I advise you to invest now. When it's finally at production level, you'll need to invest a lot of time learning the tool or pay someone else to use it.
Its working for me and takes no time at all. I'm Anti-social so Im not making any video to prove it to anyone lol
The Xmas Coca-cola commercial backlash is a great example of people’s rejection of carefully cherry-picked AI shots looking just strange and fake enough to cause a mass negative reaction by viewers. Humans notice subtle weirdnesses because our brains are trained from infancy to perceive the specific physics properties of different phenomena.
" carefully cherry-picked AI shots looking just strange" 99% of Ai crap on the net
In a few years you won't notice. Most people don't notice good CGI neither.
@@eSKAone- Directors, producers and the Hollywood suits who greenlight productions and VFX budgets notice. I predict that your prediction with be wrong, because the real cost of AI will still be more expensive that the proven older methods... and not as reliable.
It also brought attention to itself with calling itself AI. Plenty of hated commercials out there but mentioning AI can be risky for PR purposes rightn ow.
@@curiousrefuge i think the interesting thing is that as the technology progresses, humans will adapt slowly, without realizing. The disruption this will cause and is already causing to the value of 'traditional content' in the Hollywood model, in my opinion is going to rebirth the industry. Why would anyone 'greenlight' a $60m feature film if you can make something comparative digitally for less than half? Why would 'buyers' continue to pay content license fees at the same rate? Investment in traditional content will continue to drop off because returns at traditional rates will become near impossible.
This is the channel I follow for AI video news. So good and informative style, showing important compatisons. However I start to worry because no new content for 3 weeks 🤔🙂
Fyi: 3 days ago minimax introduced character reference (which is not the best imho)
Hi there! Most of our team is in LA and the holidays + fires have pushed some of our content. We will be back soon!
R.I.P. SORA
Don't count them out yet
This’s huge. Thanks for sharing🎉
You’re welcome 😊
You missed some very important points about Hunyuan. Firstly that it's open source with the ability to run on consumer hardware. Secondly that it supports LoRAs, allowing consistent characters. With these two aspects taken into account, I think in some aspects it's the best AI video model available right now.
You would need a high end GPU which is financially beyond the range of the everyday consumer. (from 5k up to 300k)
@pixelpilfer What? 300k? Did you even read the "consumer hardware" part? At a minimum it can run on the 12GB RTX 3060. You can get that for a few hundred dollars. The annual cost of Midjourney's standard plan would pay for most of it.
As of making that video, it wasn't very accessible (even though it was open source). But we are aware that this has since bee nchanged and is now more accessible.
Thank you for sharing this!
You're so welcome!
Pika is back! Was worried for a minute there
Pika 2.0 is awesome!
That coat tho. :D U can tell it's the holidays now :))) JK, you're doing great! Keep it up.
So fab I kept losing track of the actual content! Even matches the moustache. Spotless hair + no frills crew neck to top it off. So perfect I doubt it's even real!
Happy holidays!
Nice informative video about AI advances in cinema, thanks
Thanks for watching!
42:15 Didn't expected Itzy would be here lmaooo especially their "LOCO" dance practice
haha glad you enjoyed that!
Has there been any suggestion of a Portrait-version of Act One? I can't find anything. Thanks! Great content
Not that I know of
I really hope they update veo allow you to import your own images soon. It would just be more beneficial, for example, taking myself and implementing into AI films, but I also hope they learn from the safeguards of ChatGPT and make it less strict but still protective of copyright and the prevention of “bad intentions”
same!
agree!
Please have a couple of _tried & tested_ "Image To Video" prompts ready when testing these platforms.
It is for the sake of consistency. Otherwise, it's just confusing. We want to see what each platform produces with standardised tests. 🙏
Good point!
Would love to know what editor or template you use for the Picture in Picture and transitions. Looks super clean
I used a few different ones but largely done in Adobe Premiere, but I can make a whole video on that later!
Love Caleb as Lil Yachty!!
b-roll ads seem likely to be the first to use AI video. huge difference in production cost
very true!
10:42 That's a really cool barista, gets some illusionist tricks going! Spoon just disappears, and then a napkin materalizes right there in his hand. How does he do it?
thanks for watching!
Why do people think Kling is good? The physics were good but the shots look like crap. The veo 2 generations look like a literal commercial for coffee.
I suppose just agree to disagree :)
❤thanks !
Glad you found it useful!
Amazing Video, and very informative, would really love to know more about Google's Veo 2 and music generated by AI is it legal to use or not.
We'll do some research!
요즘 영상이 안 올라와서 걱정돼서 글 남겨요. 건강하시고 괜찮으신거죠? 바쁘시겠지만 건강 챙기면서 하시길 기원합니다. 늘 응원해요.😢🥲
Appreciate you watching!
@@curiousrefuge It's a relief. I was worried because no videos were uploaded for a while, wondering if something had happened. 다행이어요😊
Hey . Do you any videos that can help me with consistent characters with 1 image?
Yes, check our video with midjrouney and about training loras
0:04 how can we expand ? Is it open for public?
Not yet! Still a select few
Veo 2 is clearly better than sora
It's a good tool!
4:51. There are two plans, you can use Sora on the $20/m plan.
Yeah but not enough to do anything substantial with it given the number of attempts required per shot.
@user-on6uf6om7s OAI should consider a Sora only plan or two.and. Amos tier $50-70 plan with some relax hours.
Very true!
36:21 The AI correctly showed not only that boats can appear on water, but also that not all Fathers who have no legs are incapable of levitation. :D
hahah love the comment!
people in the film industry are obviously going to explore AI just like they did digital FX, they always have been at the cutting edge of technology, Actors have always been replaceable as many have found when they start demanding too much, it just means the unknown ones willing to do it for less can be edited to look more the part, I can see closer historical resemblances, better aliens etc with less time attaching uncomfortable prosthetics, makeup etc. Screenwriters who refuse to use AI won't be replaced with AI but with screenwriters who use AI.
They certainly are!
Google has developed a new competitor to Sora, enhancing video editing capabilities by allowing expansion into any aspect ratio.
yup!
The second syllable of raccoon is a definite no-no. I had a similar problem 18 months ago when asking Midjourney for an image of Milford Sound🤣🤣🤣
haha!
HI!!! Thanks again for the great video. I wanted to try Sora and YES I have a PAID sub to GPT, but when I went to sign up, I did not see where I can login if I already have an account with them. Can you help with that please?
You may not have the right subscription model!
@@curiousrefuge Hi! Thanks for answering. I have the 22.00 a month sub. What Sub is needed?
I like this
Glad you enjoyed this !
🎉🎉🎉🎉
Thx for watching!
34:46 There's something strange with the eyes too.
Thanks for watching!
Still have not gotten access even doing the waitlist for veo 2
It's taking some extra time to rollout to the masses
Dude just saw the flying spaghetti monster and registered nothing, how far have we strayed as species... Ramen.
What do you mean?
This is really going to change the game for creators trying to make poor-quality five-second clips that can't be used for anything.
Comment of the YEAR!! 💜
haha, agree to disagree here!
Anyone knows how to access to Veo 2? I've subscribe to the waiting list but still no response :(
It's unfortunately has a strict waiting list.
I like the video that other AI TH-camr did where he tested out VEO versus SORA and they were both kinda' disappointing a lot of the time. That said, I prefer reality to hype.
Which video?
@@curiousrefuge I think it was Matt Wolfe. Just plugged in a bunch of prompts from requests on social and compared VEO and SORA. Often both flopped. That more parallels the reality one experiences when trying to make an AI film. More often than not you don't get what you want, or what's usable.
For those who can generate infinite videos, SORA may indeed be good at something, since it is necessary to generate millions of videos. But SORA needs to improve urgently, and content creators need to talk about it.
True!
i was so disappointed with Sora.
I got a ton of access to AI tools through the Odyssey competition, but every video generator is still taking 10+ hours to render and I'm not getting any results yet. So frustrating.
Very true and common frustration
Why are people so full of this shit...? Just about anyone now can be a movie producer, director, cinematographer, graphical designer... you name it all... Just from their lazy bedroom. How over saturated the movie industry will be in the near future... Bye bye hollywood and good luck to the movie noobs trying to capitalize their (sh*tty) movies beneath the heap of millions.
Then again, how oversaturated are things now even without mass adoption of these tools? These tools will can certainly be used to create more 'meh', but it will also weed out lots of people.
@@curiousrefuge It's certainly devalued many of the main industries even if they are still essential.
I produce historic scenes and animations for museums etc. I have been quite excited to integrate Ai and speed up certain processes and got some pretty decent results. However, my clients think I now just press a big red button and the whole thing is done and therefore quibble the fees. I am still spending just as long on a project I just use Ai to do some of the grunt work. Its going to be messy times as the new phase of this industry finds its feet.
Si tomé mi ponche :)
Thx for watching!
The unwanted „zoom in“ is actually a tracking shot, or a push in handheld. The „multicamera“ is just another shot which could have been done with the same camera. 35mm is a 35mm lens or a film shot with a 35mm film camera or stills or digital?The wording in terms of camera technique and equipment is giving me a hard time, not just in terms to hear from others but also when using myself, because what are terms on an actual set, or the camera department with very clear meaning is not understood well by the ai tools out there, so it’s very hard to tell the tool what to do, what would be in the real world an easy thing to communicate. We would need more people from the film world to teach the tools. On the other hand, do we want to hand out that knowledge this easy?
We appreciate you watching and commenting, Shinola!
It wasn't happy with the coffee you brought
Appears not!
Sorry, but the generations from Veo 2 with the barista, that you call a "good job", are actually a pretty sloppy job 😂 It is basically unusable, even in a non commercial project. Needless to say the rest.
It may be better than Sora in general, but from here to say "good job" for this clip.. I'm sorry.
We need to stop the BS and the pink glasses effect when we use AI.
Fair points :)
i think we need to keep in mind that ai will never be worse than it is now, this tech is in its infancy and moving quickly. 6 months ago we were fighting for 3s of 'cinematic' footage. Now we've got action sequences and character performance.
Genesis Project Physics Engine tutorial please ❤❤😢😢
That's a great idea - I'll add it to the list!
11:12 Hey Mr Barista! We saw that. Your mom does NOT work here.
That's true...
200$ a month for surreal sora is not going to fly. Never mind that it's pretty much useless, if you are going to charge real money it better be their best model, not the model they hope can make them a lot of money or is convenient for them. Just charge per usage and offer their best. Then people with real use might use it. 200$ a month pays for a proper computer in only a few of month and that's a lot of compute
True, that's very steep for a single person.
How robust is Suno Personas in creating original tracks from existing copyrighted instrumental? There's no way to demo this without a subscription.
We'll consider making a video on it! :)
The fact that in VEO 2 we can't upload our own images, is a turn off for me at least. this will make it useless, unless you want to create some sporadic videos that have no continuity.
That's a good point. It would be great to have that option very soon!
11:20 AGAIN? What do you think, that you're somehow above everyone else working here? What a jerk.
Right?!
Sora is such a let down after all the hype. Like when it was announced, it seemed amazing, but the industry has since completely blown it out of the water and Sora doesn't seem to have made any progress.
It has taken quite a hit!
There are problems in every shot example. Small snippets may pass on a phone but every shot shows glaring oddness or at least uncanny valley-like off-putting elements on a large screen.
It's hard but doable to find the right shots :)
Sora should only charge credits for videos you download or share. Charging for the gibberish this platform is giving me is absurd when you consider that we are training the model in the process.
Totally understandable perspective. It takes lot of credits to get what you need.
Lol, ethically sourced just means - it wont be as good as the chinese AIs because they use TH-cam/netflix/hollywood as their biggest training models
Those models will certainily take time to catch up!
@curiousrefuge we all know veo2 is trained on youtube... for sure ;)
make a video on free open source ai video generator, bcoz i m individual freelancer , and i dont have much money to spend , try to get free ai video and it should be free and unlimited
We'll consider it!
SORA is a disappointment. The videos don't follow the commands. The physics are ridiculous. It's expensive. SORA is a very poorly made joke.
It is quite expensive that's true...
:(
Thanks for watching!
No more than 1.5 seconds of usable footage in all your test examples. AI video is too unpredictable, hallucinationatory and mix different physics speeds in the same footage (example: the dog in the pool’s slo mo is mixed with near realtime speed of the water ripples).
It certainly takes a lot of rolls sometimes
it's really crazy how nobody is talking about the book The Hidden Path to Manifesting Financial Power, it changed my life
Interesting..
Lol
Or you can hire a barista, meet new people and have fun in the process. Waste of time.
We disagree! :)
@@curiousrefuge Yeah, who likes baristas anyway? Their cool tattoos and beautiful smug faces. Yuck.
Sora is weak.
Lots of air out of the balloon