Відео

Thx for the reality check!

Can you please let me know how to lip sync when there are 2 characters in the same scene? is it possible? Runway keeps detecting the wrong face. (I also have a robot in the scene which isnt getting detected at all)

I think you gonna run into problem when more than one characters.

I guess it’s a reflection of the real cost to Runway of running the servers, so might have to wait a little longer for the tech to get more efficient and cheaper to run.

When I first tried Runway, before Gen-3 was released, I was shocked at how many credits you got for your money. Now, with Gen-3, it makes even less sense. The price might have been justified if each generation perfectly matched the prompt, but that’s not the case. Additionally you have to upscale and enhance each clip in another software. The 2250 credits in the included plan equals 450 seconds of Gen-2 ! That's not much at all unless you are trying to make 15 seconds UA-cam shorts. Has anyone tried the unlimited plan and the explore mode at the relaxed rate? This might be a solution if the generation time isn't too long, at least you get unlimited generations.

Like you, ive been with Runway nearly since day 1. I think Runway gas gotten a bit arrogant, and the quality just isnt worth the price.

Very well made video, and a very good review! 👏 👍

yes agree - it is extreamly expansive - additionaly only.about 10-20% of output is good quality -)

This video is so damn good... 😅... Great info and presentation 👏🏽👏🏽👏🏽

You are so wright, I hope runway crew will see and hear this... the pricing is crazy

Can we make a full movie with Gen 3 with real artists

Hey apple! What?!?! ..:

really great vid, love the gen3 prompt guide that was a cool tip. i like how you looked at it from a neutral perspective and described your personal experience with the iterations you actually get vs price points.very pricey, but thats not to hate on the amazing technology its just like how you said, we need iterations to make the complete videos thats just how the process works since the video gens can be so random you need to keep rerolling the same prompt or tweaking it or whatever so there no way around it, and thats part of the fun is making a buncha variations and tweaking it or getting something you didnt expect that looks weird and making it work, just very pricey if you actually wanted to spam generate theres no way unless your like the high tier plan which is more than my car insurance cost, but supply and demand i guess, they have the supply, the demand is there so its fair business

Good grief - you are so right! The influencers ignore all the fails and praise it to the skies for one random shot (a simple campfire when God only knows what their prompt was about (a simple hammock?) Good for you for calling 'em out on their ridiculous pricing.

Excellent video. Thank you. You are correct that AI video us getting more and more expensive to use. Gen-3 is great but not for me.

I subbed to Runway for 2 months then cancelled due to costs, it's just not affordable for the average content creator, especially when you take into account all the other AI subscriptions, I might be able to create GREAT content with it, but, I won't be able to afford to live, or even EAT the way prices are atm ha ha

I have to agree with you on the costs. Even the most basic AI sites are now pumping up the price and i don't think its because the server costs are getting more expensive, they are just taking advantage of more people wanting to try it, therefore they are thinking short term gains but long term it becomes off putting. I see the amount of companies offering these tools dropping as people start to walk away from them because of cost. I watched your Wonder Dynamics video the other day and for the price $25 a month for that quality of output, no thanks.

You are right about pricing and who gets to use this fantastic new medium for creative expression. Those prices aren't worth it. I did a few tries and couldn't use anything I generated, so went back to Gen-2. Because one needs a lot of iterations and experimentation -- rolling the dice -- I have to have an unlimited plan. $95 is a lot, and I think Dream Machine was around $600!! I use Haiper for $30, and I get outstanding results overall. I only use Runway when Haiper can't do what I want, and for lip-syncing. My $35 Runway plan runs out fast, and I had to do a $10 booster last month, and that's just for selective generations. I won't be using Gen-3 unless it has more controls. And even if we can make video using this technology, it's hard to get it seen. Some of my best work can't break 1k views. Check out my channel for some next level AI creations.

Excellent presentation and thank you for keeping it real. I completely agree with what you’ve said considering pricing and restrictions. It’s for this very reason that I prefer to stick to my free tools that somehow work quite well for me. Once this stuff eventually becomes easier accessible, I might start working with it. When paying top price, I expect to have very few limitations and plenty of options.

Beta version was wayyy better to create funny and surreal videos. Sadly you don't have access to it since they dropped the new model

I wonder when I can expect this type of technologie to become free to use. Maybe in a decade or so all AI will be much more advanced and this isnt special anymore.

Blue got 2 wifes

Ai cats before: 🤨 Ai cats now: 💀

I wonder if thru understanding this version of ai theres a possibility of finding out whats going on in the brain during a drug induced hallucination...Ive done a substantial amount of mushrooms one time and have never seen the things i saw in any simulated way until these videos

Super interesting and helpful 🙏

New stack: - MJ - CapCut or Canva to cut out character - Animate background separately with Pixaverse (better more stable motion brush) or Stable Video - or just camera move with... - Lip Sync with Pika or Sync Labs - Optional cleanup step with Topaz or some of these ComfyUI steps

Great content bro! I just subscribed!

this is how i normally see cats, hideus creatures.

0:01 HIS HEAD CAME OFF💀

Nice... 😍

Hope the fix the blur around the robot aoon before its useless to put into a movie

still can't do hands for shit...lol

so coool haha🤛

Such a touching story :") Excellent job on the sound editing 👍👍 Love the way the monster sounds.

awesome, very nice sound editing

nice one!

how long is his snout

What prompt did you start wirb

This video somehow reminds me of the following song: ua-cam.com/video/QjKJ8kffCAY/v-deo.html

A new emerging field: improv AI generation. This is quite impressive. Is this pure Luma or was some post-processing involved (lip sync etc)?

lol what the hell is this nightmare??

Voiced by Brendan Fraser apparently

Nice... Did you prompt each extension? Or craft the story around whatever the hell Luma decided to do? 😅

Gotta love a happy ending ❤

Fascinating. I mean, I think at this stage one could do better with image to video and compiling a dozen short animations, but this was done with one text prompt. Give it another few months or a year! [P.S. check out some of my AI vids, which tend to do themselves in the more original they are and the more I deviate from templates and tutorials].

Amen❤

This is fantastic, Mike! I hadn’t even considered using Luma this way!!!!

this is interesting but in the end, like many of the clips I have tried to generate, they become "cartoony." When you extended your clip, the astronaut at the end became a cartoon character. How can you get around Luma making them cartoons?

Horrifying...people are going to abuse this. 😢

I can't wait to see what Haydn does with Runway Gen-3 alpha. In the demo video where the wig falls on the bald guy, he goes through three stages of emotion (sad, surprised, happy). This capability literally didn't exist one week ago.