Видео

Thanks!

This is a while back but it’s better now 😀👍

I get an email every day about how to do SEO ..my answer is the same SHOW ME. If you want to help with 3 videos first and it works then I will pay. I'm not paying first until you show me you can help.

Did you see minimax today ?

At some point I will need more people that’s would be cool to talk then!

Thanks Andrew !!!! I can't wait to share more with you!!!

Hey just Day one of a journey of creating a Science Fiction film! Thanks for asking!

Yeah its the room I film in the office terrible connection.

@phill_ai check connection and everything prior to fim for good results...

@phill_ai check connection and everything prior to fim for good results...

Greta feedback thanks

It has a very specific set of skills!!!

Yep we broke it !!!! I sent the CEO an email see what she says but me too!

Unfortunately this is common with Video gen. Always amazes me they try and promote themselves and do the opposite. I sent the CEO a email to say this is what happened last time and they are still running the offer that isn't working. Not cool! Thanks for letting me know.

@ 😁In the end it was 7 hours for the first, 🐇still hopping on the others😂…. I’ve been having some great results with Kling, but this one from Pikka did exactly what I’d wanted, enough that I’ll definitely have a little sub and see how the regular gen times are. Thanks for the heads up👍🏼

Nice love that when it comes out as you want!!! Thanks again!

Always the same … we still blowing up servers 😀🙏

Ahhh Steve there you go ask and you shall receive !!! Thank you for the rest of us! Cool Yep so much fun!!!

Hey Luna did you get many generations?

That is very cool 😎

@@phill_ai ☺️

I love Midjourney too! Is it a style you have or many different shots?

I usually edit wide shots to mid shots, then to close-ups, and repeat. In terms of style, I use --style raw and photorealistic prompts. Once I find the right style, I use -cref for the character and -sref for the style of the image to maintain consistency. I always try to finish shots with a hand movement or blinking eyes and start the next shot with the same, so transitions are seamless. I also use LuMa ai sometimes to extend my videos. A good trick is to join clips-one played normally and the other reversed-so they match, but only if the movement is subtle, like blinking, hair movement, or transitioning from a smile to a serious expression (and vice versa). Cheers!

@@Michf666 Roll on the one platform hey ! Are you enjoying the process ?

Hey thanks first of all Gerwald!! I'm thinking the same Adobe = TOAST I was at IBM ha ha. Yeah I believe there will be an enormous merger acquisition next year and could be them? You are thinking ahead I love that, yes a one for all is coming but then thats exactly what Microsoft did in software early on and what Google Amazon are doing now so I think so too. Great to talk!

I appreciate you commenting!!!! I know it’s like when cgi went on every film then it got too much but it will calm down next year into less ai 🤖 looking. Very polite point thank you 🙏

thanks @BlackFlame-Studio I would keep it fluid right now as soooo many more updates to come. Mastercard have a hitman looking for me I'm not the person to copy !

We will see the last parts I think of this next year so many I think this will bring loads of old shows to the future ... time machine stuff

I appreciate that... what do you think AI Editor Wars?

Hey thanks for coming back. Firstly I have done this hundreds of times. My first videos I made where with dreams and meditation in mind. So I see the result I want then aim for that. I am looking for a certain morph, and will generate multiple times OR improv as I go. I am going to run a free course on this as I have a few people want to do it and struggle. I think I'm going to start going live here to as I go live on TIkTok Every day.

runway hates those kind of characters as I guess it's trained on humans. I will try for you

@@phill_ai thanks.

My brain has an error code coming up 😀

Peter this is brilliant!!!! Do more of this excellent!

@@phill_ai Cheers Phill! I'm hoping to get my hands on Google's new Veo 2, it looks amazing. Then there's also a Video version of Flux coming. 2025 looks like being a good year for hobby filmmakers, though not so much for getting a good night's sleep ;)

Hi Peter! Yes after lots of looking at videos I tend to agree !!! Google won the year how the heck did they come from the back to pip everyone at the the finish line!!!

Yea that is cool. Which ai's did you use?

@@AA-ho5mg Vidu is this one ..is that what you meant?

Hey that is very kind @SkyShazad !!! Thank you !!!!!

Hi Fabien just talking about at my tiktok live Runway used to be the same thats why I cancelled too make terrible gens makes the price 3 x as expensive. I just did Vidu review same this but a few more features check the last video same price. Same quality really too.

Hi Aly Like this one?

@phill_ai yeap great job phill!

@@aly11zya Ahhh thanks

Couldn't agree more! Marc! It's going to be a bloodbath next year lets hoe the credits price comes down 😀

Thanks Peter!! I always appreciate your heads up !!! So your wish is my command...vidu review video just uploaded! Hard to tell I think on quality for me as only did a few but Vidu definitely got some key things right?

Yeah here you go:- give 15 - 25 images and it learns what it is then you can output more like it. Full version In the context of AI learning to create images, LoRA (Low-Rank Adaptation) works by adding small, specialized modules to an existing AI model. Instead of retraining the entire AI from scratch, LoRA allows the AI to learn specific aspects, like generating particular styles of images (e.g., landscapes, portraits, or specific art techniques), by adjusting only a small part of the model. Here’s a simple visual explanation: Original AI Model: Think of an AI that can generate basic images, like a simple drawing of a cat. Adding LoRA: By adding a small LoRA module, the AI now learns to create a specific type of image, like a realistic landscape. Result: Now, the AI can create more specialized images (e.g., landscapes, detailed portraits) without needing to relearn everything. This makes AI more efficient at learning new tasks, like creating specific types of images, without needing to retrain from scratch.

Lovin the feedback !! haha yes no time but next time I will do some...oh might be able to edit some in oh might kill my reach !! Thank you I appreciate it

Hey thanks for this ... I always love the feedback so I checked it out and did a quick comparison. Vidu got some great ideas and yes very similar ...quality hard to tell without using it more I think?

@@phill_ai Yeah actually, the render quality is not there yet, looks pixelated I spoke to them and they say they are working on it, I have not used Pika yet tho

Yes wont be long with the quality but amazing start!

The zero gravity is awesome! Not sure with multiple people but will try but I got 5 inputs which is crazy! Did you have a go?

Yeah 26 vids for 10 buck I got ?

Ahhh thanks 🙏 that is very kind !!! I’m not sleeping anymore too much going on @NEURAMos 😀

Ahhh thanks !!! Glad you loved it!

You have to watch it all or my engagement is down 😀 The clothes detail is what shocked me as I tried LoRa's for my thumbs and products but this does 5 items and they all look great. AND the detail is next level crazy good!

@@phill_ai Loras are way to versatile as they go anytime soon. What you might mean as it comes the closest to it is stuff like "ip adapter" and or "control net" since they also dont need extra training and can work with a few or even a single image.

@@DetSmike That what I said! Ha ha very well explained DetSmike. I get technical to a point never under the hood but I know a little about a lot of stuff. What I was doing was working on LoRas for people and products and this tech seems like we won't even need any training as you say. The quality is very cool at this stage but needs to get to 1080p to really be useful. Hey thanks I really appreciate it!

Well thank you !!!!!

I understand as a person who has a Mastercard hitman looking for him 😀 but the challenge is that they have is the AI Tourist burning 20 gazillion credits and then no cash to pay for the servers. Is there something you wanted to try and I'll generate it for you?

@phill_ai I don't mind their policies as long as they are open and transparent and clear. why do they have to use this clickbait or whatever. just the optics of it is very bad.

@phill_ai and by the way I do think that all this energy spent should be spent on something useful like scientific research and other things of similar significance. burning fuel just for fun is totally not ethical.

@@TheAptimn I agree! Free should be flippin free 😀 . Not a good start to a relationship tell them thats what I do as som eof these companies are so new they dont understand the basics

@@TheAptimn Yep Google ai just looking a nuclear power station! That's one way to do it! www.bbc.co.uk/news/articles/c748gn94k95o

Hey Intonova!!!!!!!!!!!! Yeah it's the clothes ....so spot on the detail is so impressive!

Yep Seb many moments like this but Actually I’m starting to think this is biggest thing to actually take us forward???

Maybe faster but yes 👍

Hi David ! No sleep for me again ha ha . I just did a peaky blinder one and it's awesome. I am in shock!

You are right!!!! I thought that but didn't double check . I put suit on one and it did a suit so the exchange between prompt and image is amazing! But Yes midjourney can gen clothes so will do that more ...I did a peaky blinders one on my live call so so good !!! I'll share it later

Yep I agree and great it's got a big prompt box !

Hey Steve thank you for watching first and hello !!!! Oh you are so welcome. So what do you think of Pika v2 then was it a bump as I said or not ready yet?

I had a massive schedule to day that video was done before jumping in the car so no time to ads Fabian and I appreciate the feedback !

You are becoming a regular SkyShazard! I really appreciate your support!!!!! I love this too ..this is massive not perfect but HUGE

@@phill_ai I may not comment always , but I watch all your videos sir

@@SkyShazad You are a star !!!!

Could just pop it in Runway but I think all the companies will be working on lipsync and this will be very high on there lists right now... this has shocked the other companies I'm sure !