That's mostly what I'm seeibg online as well. I have seen a handful of "Im paying and its still taking 48hrs for it to fail most of the time", but knowing people, they are probably paying for credits and not a premium membership.

In 2 weeks I have gotten 2 5s videos. I just had 12 more videos fail on me again last night after waiting about 48 hours for them. The two videos I got were solid. But I would have really liked to be able to fiddle with it a LITTLE more before having to decide if I want to pay for a membership.

I just bit the bullet and paid for a membership. EVERYTHING works and a 5s video takes only about 5 or 10 minutes to render. None of them failed, other than prompt wording. I had used 'gasp' and it didn't like it, so I just changed it to 'deep breath' and all was fine.

I'm doing a new 'Slime Pit' game show series and already have lots of 5 and 10 second clips to edit into a video. I can even go to my 'Process Failed' movies and click 'Try Again' and within 10 minutes the video is ready. So much for waiting a day or two.

I guess I'm just going to have to do the same. Like I said, the 2 videos I got were great. One of them I didn't even prompt, I just uploaded the photo and hit generate. Im just so careful about paying for something I don't like in the end. Last night I saw some pretty great images from Adobe Firefly. I had never heard of it. I HAD used their previous image generator and did not like it. So I gave Firefly a try. I generated 12 images of women in pools. It did great with wetlook... but every one of the women looked like a 2 day old corpse. Im sure that it can do better. I have seen it. But Im not paying for what I got so far.

Kling may have to be an exception to that rule.

As if the powers that be at Kling AI were reading my messages here, 6 of my failed attempts suddenly appeared as working videos. 5 of the 6 turned out fantastic. The 6th isn't "bad" either, the clear mess just kind of becomes invisible once it starts to move, which I mostly expected. Considering how well the 8 videos I have been able to receive turned out, I am going to have to pull the trigger on a membership for sure.

My most ambitious AI video so far . . . I had done a fake trailer for 'Mudslide-The Movie' a while ago, but have created trailer number two for the same non-existent movie.

So without further ado, here is trailer number TWO for Mudslide-The Movie. (and a new movie poster)

https://umd.net/profile/i/bobographer/section/videos/video/mudslide--the-movie--trailer-2

I think this may be my favorite so far . . . a game show . . . where the loser falls into a pit of slime . . .

https://umd.net/profile/i/bobographer/section/videos/video/slime-pit--the-game-show

Wow, top stuff, that looks amazing.

'...and some woman have their own flotation devices... , classic line.

By now, you ought to know I have a 'thing' for Francesca, that rich, sexy, young, long-haired Italian bombshell in the red dress.

I've posted a number of photo sets featuring the AI creation and now I'm sharing a video of her at work.

She inherited a food fillings plant, and brought it to life by maintaining high quality, better staff, and marketing the company as one owned by a sex bomb. In fact, she even did promotional videos of her wearing some of the products she manufactures.

And here is that video, complete with a short introduction . . .

https://umd.net/profile/i/bobographer/section/videos/video/francesca-at-work

I'm working on a huge video project, the entire 'Tub in the Park' series. I created a very short trailer that you may want to see . . .

https://umd.net/profile/i/bobographer/section/videos/video/tub-in-the-park--trailer

Got very interesting result from kling was very surprised

Here is a nearly 25 minute AI video, a documentary about the 2024 'Tub in the Park' event.

Enjoy!

https://umd.net/profile/i/bobographer/section/videos/video/tub-in-the-park--full-documentary

I wasn't aware of that UMD stipulation. I'll try to keep future videos under ten minutes.

Cool mabye try reupload it if u can but split it in to two parts if its not gonna take a long time if it is dont worry.

I'm putting up a YouTube channel with the same video content, so length won't be a factor. But already I got a copyright hit on my 'Inflata Hammock' from a company that makes inflatable toys. (but nothing resembling the inflata-hammock)

I thought I'd put together a short compilation of my earlier AI video gungings, combining other fetish passions of mine by featuring attractive cougary redheads in satin and fur getting covered in foam and cream. Preview pic shown in my avatar image.

https://umd.net/videos/glamorous-ai-redheads-get-messy-1

Still working on getting pies to be thrown and landing, but gungings have been working nicely with this site/tool.

Good Pudding!

Those are nicely done! Thanks for sharing.

Couple i posted in one of my gunge vote threads

Normal speed and slomo

Here's the current playlist of Splat the Brat Episodes. It will probably come to an end after a couple more. Maybe.

https://youtube.com/playlist?list=PLbUVghNCfQHLlZKccN4z4RpPj6PXckScw&si=AM5JGLBTl4r166r7

Hope you're planning to continue making content though! Amazing stuff.

Almost certainly. I think as the tech improves, coming back in say, 6 months with another season is likely. I'd like to include voice and sound into the video, and I'd like to be able to vary the substances in the same way I could with still images (at the moment it's a bit samey). I'd also like it to be easier to generate the ever popular gunge dunk, at the moment these are a bit of a fudge. The goal is to make something that feels like full segments of an actual show, rather than just a 'best bits' montage as it currently is.

Another AI compilation video from me, with satin clad guests and waitresses at classy restaurants getting creamy. Hope you like.

https://umd.net/videos/ai-messy-restaurant

Good Pudding!

I had hoped that this section might have really taken off. Apparently, a lot fewer of us are interested in AI videos, even though the quality is much better than just a short while back.

For that and one other reason, I've been posting my videos to a YouTube channel. The other reason is that any video over 10 minutes in length is only viewable here by VIP members. Non VIP members cannot see them.

I already have a ton of content on YouTube, so please have a look. It's at https://www.youtube.com/@BobOGrapher

I may be late to the party here, but I've gotten some impressive results. Sure, it's still plagued by the odd disfigurement and some Bing face, but there's definitely potential.

I think that's the key point at least for me. AI video is improving but it's not there yet. It's like looking at dalle2 compared to dalle3 or Flux finetunes. It's getting better but with limited time it's not worth the effort and resources from my perspective yet to play around with it. It will get better though and at some point I'll take a longer look. It's like do you build a rocket now that takes 200 years to get to another star or wait 50 years and build one that takes 50 years.

To put it into perspective, the latest Gunge Lora I'm working on has taken $40 of compute, , 23hrs to train and countless more hours captioning images and testing the outputs.

I'm starting to get the hang of this, results are improving.

Another new AI video on my YouTube channel.

https://www.youtube.com/watch?v=G4ulp5gnhnA&t=339s

I agree, but the fact that we can create anything at all with video amazes me and I will keep toying with it. I use Kling, and have the option of the 'standard' fairly low-resolution or a 'professional' mode with higher res video quality. The problem boils down to cost. It's either 10 credits or 35 credits to create 5 seconds of video, depending on the quality level chosen. Those credits can get eaten up very fast, so I've been using the lower res option. I fully agree though that in time, everything will improve, and in order for the masses to accept it, the cost must be kept reasonable.

It's the learning curve that's burning through my credits, once I have everything worked out the daily ration may be adequate, but the $20 I paid for 1300 credits is what I'd normally spend on drinks. I've only created one clip in pro mode ( up-scaled my best result ) & extended two clips to 10 seconds - with mixed results. If I can get consistency I'll stick with it, if not I'm only down $30.

EDIT - Can anyone recommend the optimal Creativity - Relevance setting? I tend to keep it between 40-60.

I might make a full post on this in future. I use image to video exclusively. I'm fortunate in that my tastes are quite repetitive, and I can re-use the same prompt with different images to a good degree of success.

I set the relevance to 0.9. I want it to follow my prompt closely. I only prompt in pro mode, as there is no way of recreating a result from standard mode. I use version 1.5, which doesn't allow prompt extensions. If you've seen my stuff, you'll know I like gunge being poured overhead, with a lively reaction from the subject. Here are some things I've found:

- If I want the gunge to pour from above, I need to have to the gunge already pouring in the prompt image. Otherwise, there is a good chance it will materialise on the subject from thin air. There is an optimal 'thickness' of the gunge pour that leads to the best results.
- the gunge will follow the properties of the gunge in the starting image, but I can reinforce one aspect of that in the text prompt (e.g., translucent, or chunky, or sloppy). The viscosity tends to be determined by the image prompt.
- if I want the gunge to stick to the subject, I need to put this in the text prompt (e.g., "the gunge is sticky and sticks to her fair, face, body and legs as it flows down"). This works well, however, weirdly hydrophobic hair can still happen.
- the reaction described in the text prompt should be compatible with the image. In general, the more synergy between the text and the image, the better the result will be.
- synonyms for gunge are all possible, 'sludge' seems to work well for me.
- the more complex the image prompt or text prompt, the more there is to go wrong. The best results are from a single subject with no background people. The more realistic the image prompt, the better the result will be.
- if something in the periphery of the video ruins an otherwise good result, I can crop it out of the video using editing software (making sure to use the best interpolation option).

I'll buy some credits to experiment, then use the daily allowance to make 2 clips a day until it refreshes or I want to buy more. I hope this is helpful to you the next time you get credits.