@ed-71
- 564
Forum Replies Created
-
Posts
-
It’s so annoying when you get the resulting clip almost perfect on the first or second go but then you’re still trying to get rid of messed-up speech or weird animation (stuff I can’t edit away) after another 6 attempts.
It all depends on what AI model you are using on Pollo. They have a huge range of them. Most are content-restricted and expensive but some of the cheaper ones are pretty good. The complication is that things keep varying so what works one month may not give the same results or offer the same options a month later. Wan 2.5 on Pollo used to work very well, but I haven’t used it in a few weeks. Pollo 2.5 was total cr*p last time I used it – bizarre distortion, random nonsense and completely inconsistent results. I’m hoping they’ve improved that.
I like the models which allow you to set a start frame and an end frame – that usually gives a LOT more control. You often need to turn the inference up too, and most of the models don’t give you control over that. I’ve found wording is SO important and you often need to specify when something is not doing anything or a person is not speaking just as much as the things or people who are active. Qwen isn’t bad and Grok is very good, but the nudity moderation is pretty strict on both of them (and about to get stricter, I imagine).
Of course, the addition of audio has complicated the entire field, but I’m not giving up yet. I swear all of them have good days and bad days too. Sometimes I can’t get a result on Qwen, but do on Grok or vice versa. Sometimes I end up saving a prompt and using it a day or two later when the AI is in a better mood.
I didn’t once have such a bad day getting videos out of Qwen that I used the prompt “A woman smiling at the camera” with no start frame image and the result got moderated.
Sometimes it feels like a video generator is trying to be infuriating. I tried a clip of a London cabbie telling a nudity-related joke:
- The first go was fine except the joke was too long and the last words were cut off. I reworded it and tried again.
- Go 2 was fine except, for some weird reason the London cab was bouncing all the way through as if someone heavy was repeatedly getting into the back. Added an instruction to keep the cab still.
- Go 3 – saw the back door slide open and then partially shut. That made no sense so I added an instruction to keep the doors shut. It was distracting and the cabbie’s performance wasn’t great.
- Go 4 – the driver’s door opened at the start and the cabbie was (unlike the 1st 4) facing the wrong way. Stipulated which way the cabbie should look and tried again.
- Go 5 – the cab moved perfectly, no doors opened, the cabbie looked at the camera and spoke total gibberish. No idea what he said.
- Go 6 – the cab rocked a little too much but not as badly as 2. Also, the driver’s door opened slightly, but the cabbie’s vocal performance was perfect. I try really hard not to spend too many credits on a single clip so I’m calling it a day there.
I feel that video could easily have been perfect on attempt 2. I’m sure a lot of people think you just tell it what you want and you get a perfect result first time. As I’m trying for more precise output, it’s taking a lot more work (and credits) to get there.
I’ll need to pay £100-200 per month to do these videos with proper uncensored nudity. We need a lot more people to become NE patrons to make that viable.
I’ve got an NSFW one on the way but any nudity is fleeting or minor and snuck through the moderation. If I directly prompt nudity, it gets reliably rejected.
Please post a comment if you’d like to see more of these.
Hmmm… odd. Wan 2.2 seems to be open to nudity and sensible movement again, but the image quality has dropped markedly. I wish I could get more affordable access to Wan 2.5/2.6. And end frames on it. That would be great.
Thank you!
I’m beginning to feel pressured to get a VPN.
And I certainly won’t be getting a subscription if the UK government is threatening to block X from the country.
I have to feel like I’m achieving something otherwise I’d go mad. After spending 10 days in hospital in November, I’ve been a lot less able to let time just slip by.
I’m trying to get a straight answer on whether Grok Spicy does or doesn’t allow NSFW content. Even their website says both.
Speech synchronisation is still a little wonky, and it varies from video to video. Some of it might be manually correctable with intensive video editing but I’m not going there.
Another advantage of these short clips is that the lack of consistency with voices isn’t a problem. If I create two video clips with the same characters in them, the AIs cannot be relied upon to be consistent with the voices produced.
Grok seems much more reliable on lip-sync than Qwen, but models are constantly being upgraded so that can change quickly.
All completely AI creations. Very detailed prompts, cherry-picking the results (I picked the best 20 videos out of 56 created), and an element of luck combined to generate these.
Unlike the previous couple of videos I worked on, these clips were created by Grok and Qwen, mostly via my mobile in spare moments. As I’ve spent a lot of time in hospital waiting rooms lately, it seemed like a good use of time when I don’t have my laptop available.
As I’ve (temporarily) given up on Pollo (trying to rebuild some credits), I had a mess around again with Grok imagine and Qwen. They’re both pretty well locked down when it comes to nudity, so I went down a different route (but still on topic):
It would be good if more people would consider sharing their stories.
My aim now is to get the momentum back up after the Christmas/New Year’s lull.
No problem!
I’m giving up for a while. I’m just getting hilarious results across several generators. Even using HuggingFace with the old reliable Wan 2.2 (with start + end frames) is acting up, largely ignoring prompts and doing its own mad things which don’t even end on the “end” frame. I don’t know what’s going on.
I just spent my last credits trying to get Pollo 2.5 to produce a scene from a starting picture. Yet again it seems to be deliberately buggering it up. I got dialogue produced perfectly except both times the opposite person said each line, which makes zero sense, and an item of clothing which I prompted as being pushed down her legs both times was stretched comedically upwards leaving an identical item of clothing in its place.
Also, I’m finding it really hard to work without being able to set an end frame. It’s SO much more cost-effective and reliable to be able to set a start frame and an end frame.
Pollo 2.5 seems good with female nudity but I’m having huge problems getting to follow instructions. It feels like it deliberately finds a new and small way to completely ruin each clip.
I’m trying to build a longer video with 2 female TV presenters and an audience in a studio. First it insisted on adding huge headsets to the presenters, then it would make the audience look anywhere except at the presenters. For one clip the presenters suddenly developed west country accents, then one of them took of a bra and had it immediately turn into a large sheet as she dropped it on the stage. An attempt to make the audience clap in response to something the presenter said resulted in the audience repeating the same words in unison. I’ve also had one presenter morph into two, one the presenters change hair colour mid 5-second clip and remove a bra to reveal a denim t-shirt “underneath”. It’s been very costly on credits.
I’ve also had the “third party service report inappropriate content” error when the presenters were simply introducing themselves, but not when clothes were removed. It’s very inconsistent and I’ve now run out of credits.
It seems that the resistance to nudity jumped upwards significantly in the past few days. It’s hard to be sure (I could just be unlucky recently) but I’m definitely wasting my limited credits on failed attempts. One 6-second clip has so far gone 11 attempts without success, leaving me with insufficient credits to complete the whole video this month (and probably not next month). The most common outcome is extra clothing appearing (that hasn’t happened in months) or bizarre unrealistic animation where limbs or spines rotate impossibly.
I’m going to give it a break – see if things improve.
Sorry for the delay. Had a bit of a messed up evening and didn’t get anything done.
Just checked. I don’t have a record of the name but it did involve a photo with your partner. No photos were submitted.
No one should ever have to apologise for their body.
Wasn’t it naked with your partner?
It caused a right kerfuffle. It didn’t matter what assurances I made about consent, a few people were convinced consent wasn’t possible.
Don’t forget to take some photos or record some video!
Who’s going to this year’s WNBR?
Cool. Lots of ideas. Now I need to generate a lot more videos.
I’ll see if I can focus on stuff that would not be possible for us to achieve in reality.
Has anyone else got WNBR photos they have taken themselves?
That would be awesome. It could be good to compare WNBRs from different cities.
We should probably have a WNBR gallery or two.
I know some people disagree with the use of systems like this but videos that simply would not be possible otherwise, simply won’t exist without this.
Those of you familiar with my second generated video might appreciate this, to give a taster of what may be possible very soon:
Here’s a few more examples:
The last one didn’t work properly. The gameshow host sounds a little rough (but acceptible) but one of the women is supposed to be saying “This is so embarrassing” and another should have said “I can’t believe it!” but we got an unintelligible babble.
Sorry I didn’t have time to post preview images for these, but in this case I think that works better.
And one last one:
I know the dress motion is rubbish but I ran out of credits before I could try a few potential corrections.
I should point out that Qwen firmly blocks attempts at nudity or mature topics. Some amount of underwear slips through but you have to be very careful how you phrase things. Also, only about 1 in 3 videos is coherent and makes sense.
Cool. I’d appreciate it if you e-mailed them to me, if you don’t mind then being used on the site, to consider for the section.
I’ve been meaning to watch the Cardiff WNBR for the past few years but every time it’s landed on a really bad weekend for me. On one occasion I had flu that weekend and could barely stand up let alone travel into the city. Fingers crossed I can see the 2026 one. It would be great to have a section on the site discussing the WNBR with some photos to illustrate it.
That link won’t play for me. None of the videos on that site play in Chrome, Firefox or Edge for me.
Does the copy I posted in this forum work for everyone? I shouldn’t leave it up as it needs to be in the age-restricted area.
Lots as far as I know. I’ve posted loads and got positive feedback for most of those.
Any difference with the one on the link? It’s the same script running both.
I’ll try posting it here in the forum instead:
I don’t know why the page version doesn’t work.
I can’t find the article, but I’ve read the Cardiff event has a screened off area at the start for people to undress, with signs up saying please don’t use cameras in this area.
I would have thought it would encourage more people to take part. If you want to be recorded undressing, just do it outside the screened area.
I’ve heard people say that photography isn’t allowed in the “undressing area” at most of the events.
Thank you everyone who has begun to use the “Like” button!
I forgot to transfer that to this new site, but I’ve done that now:
Obviously, it’s behind the age-verification wall (got to comply with the law).
The more I think about it, the more I like this idea. I’m going to begin planning something along that line. Hope everyone likes the idea as I’m not getting any feedback…
A pretend documentary could be good as a spoken narrator could be applied as a voice-over, thus getting around the lip-sync issue. That solves a HUGE cost issue for me.
I was hoping that the somewhat more prominent “Like” button would encourage more clicks. It’s only a tiny interaction (compared to a reply or initial post) but I figure it will help encourage more posting and anything which increases the rate of interaction in the forums here is good.
Anyone else got stories to share about those missed opportunities? I’m sure there must be many examples of this from many of our members. I think we can all use it as a sort of learning experience about being more confident to grab some opportunities.
My experience with most things in life if that things usually aren’t as bad/scary as you expect them to be.
Is there any interest in another video along similar lines to the one above? Possibly a gameshow or dare thing, or possibly some sort of psuedo documentary.
Before I reactivate my AI account (I let it lapse do to a health issue), I need to plan something out in some detail. Creating the videos takes a LOT of work but there’s no point paying for the account until I know what I’m doing with it.
Yes, it’s going to take time to get the community back up to speed and, while I will be around daily from now on, it’s going to take me a few weeks to get back up to speed myself. Please don’t expect a sudden return to normality from me. My stamina is terrible but improving, my vision is affected but that should return to normal in the next week or so (a side effect of the anaesthetics etc.) and my concentration is sorely messed up.
When I’m a little more with it, I’ll see if I can find the e-mail addresses of some of the more active old members and get in touch with them.
THanks for your patience, everyone. 🙂
We’ve lost a few which is disappointing. Visitor numbers remain good but it’s hard to keep any momentum going in the interactions.
All was not well for me but I’ll be home from hospital soon. All is well now but I’ve got months of recovery to do. Apologies for not wanting to share details on a public forum. I survived and will be better than before once I recover.
I’m not back just yet but, but I’m hopeful I can get age-verification up and running by about Friday. I’m expecting dozens backed up but I can’t see the NE Teams account on my mobile.
Thanks for bearing with me. I’m trying to get back as soon as I can.
Sorry everyone, I’m going to be offline for a few days. Back soon!
What does everyone think of these new age-verification laws?
Having spent a few hours over the last 4 days working on another video idea, I think I’m going to drop it and come back to the presentation style used above. With the loss of two of the AI tools I was using and a significant quality drop in another one, the results just aren’t good enough. I’ll keep looking for replacements but ultimately I need to save up and buy a LOT of credits on Pollo and give it another go there instead (or find a way to get hundreds of people to sign up for a free account there instead).
I’ve just now regained access to the ed@nakedexperiment.com e-mail address. Seems I have some 400+ emails to go through…
Might take a while. I’m certainly not going to get through it all tonight.Thank you!
Everything is going slow this week. My internet has been poor (at best) and I’m still without my main laptop (it’s dead). It’s going to take another couple of days to restore access to my e-mail, so I’m feeling a little disconnected and restricted. There are a few sites I can’t get into (like Patreon), but that should sort itself out as soon as I get into my e-mail.
I need a LOT more credits. Many things would be possible then.
Really the main issue is the rate of wastage. I’m still “wasting” around half of my credits on producing video clips which don’t come out usable. It’s the nature of the beast but it’s very annoying.
The built-in screen is set as the primary display.
I’m slowly putting together a longer more-involved version of the “Show Me Everything” video but credits are in short supply, I’ve been ill and there’s an issue keeping the subtitles in sync with the video. It seems the longer the video the more they slip – which is really annoying.
Male nudity is tricky currently. There may be some AIs out there that will do it, but my experiments so far have shown them to be MUCH more reliable with female nudity. Men tend to get depicted with female genitals or no genitals at all. The only reliable way around that seems to be starting out with a nude photo as your starting point, but that restricts story possibilities. Also, if you start with nudity, many of the AI photo editors won’t work with it.
The ban on adult material may be loosening:
https://www.msn.com/en-us/news/technology/openai-to-allow-mature-content-on-chatgpt-for-adult-verified-users-starting-december/ar-AA1OsNUkOf course, that could just be a way of tapping into the adult market (you can produce your own bespoke adult content, we just need to charge your credit card $10 to prove you are over 18). Am I being overly suspicious?
Cool – I’ll add you on the next ranking table update.
Cool. What was your username on the old site? I haven’t got a folder for “David” or anything similar.
2 people yesterday too. It only takes a few seconds.
Unless you can’t remember your username, then it takes a little longer.
I don’t see why not. Most of it’s just a case of forward planning. Each system has its strengths and weaknesses. Pollo doesn’t reject nudity. GetImg is very good at people and scenery and passable at photoediting. I find Qwen is pretty good at combining stuff from different photos into one scene. The main problem I experience is picture resolution. Qwen’s output doesn’t go high enough. Pollo goes to FullHD (Although it will upscale to 4K) but it’s expensive on credits to do that. I’ve done all my experimenting at minimal resolutions and I’m constantly running out of credits anyway.
I reckon I waste about 50-60% of my credits producing nonsense output, which I end up deleting. My success rate used to be MUCH lower but I think it’s plateaued now. The AI will usually come up with some unexpected way to ruin the output half the time. You really must try to specify every detail. Otherwise someone will be removing a bra to reveal another bra underneath, or someone will turn their body all the way around but leave their head facing forwards. In the example above, the first go saw her remove her towel to reveal pubic hair going all the way up to her chest.
For those of you who prefer nudes to be shaved, that’s borderline impossible with the AIs I’ve tried so far. I suspect that may be a protection for other reasons. Trying to achieve it chews through a colossal amount of credits and I don’t want to get flagged for something dodgy.
If you post your stories into Notepad first, then copy them into the editor here, that will ensure you don’t get any formatting commands included.
Did you and two other people today. Looks like I narrowly missed a fourth.
Yes, totally artificial. I created the group in chairs (they were supposed to be sitting on the floor BTW) on Qwen, then created to performed on GetImg and combined the two with Qwen Image Edit. By using the group without the woman in the towel as the start frame and the group with her inserted as the end frame, I was able to get the AI to walk her across from out of view. Then I saved the final frame of the view (and messed up slightly – thus the glitch) and had the AI remove her towel. It’s a quick and cheerful lowres go, but it gives you an idea of what’s possible. I’m assuming you haven’t seen the video I posted in the age-restricted forum?
There’s a lot of it about. I’ve still got a terrible cough and blocked nose – haven’t slept properly in days.
I’m pleased that this didn’t turn out to be contentious (No one threatening to never visit the site again etc.) but it would be nice to get some sort of feedback.
Whatever this cold is I’ve caught, I’m still trying to interact on here several times per day. Anyone else been hit by the plague that’s going around? It’s like a heavy cold with a horrendous chesty cough which just lingers.
Weirdly, since posting that, I’ve seen no one on Teams at all. It’s been dead.
Is something going on that I’ve missed?
Here’s a quick example of the sort of thing I’m talking about. In some cases, I’d use photos, in others, it would be videos (probably less than 30 seconds) like this one:
This is just a low-resolution example.
Also, although I try to be online every weekday morning, please message me on Teams if other times are more convenient. I’ve age verified some people in evenings and weekends. Just send me a message on Teams and I’ll see what I can do.
I’m on Teams most mornings. I should be on tomorrow morning, and probably Friday too.
I hope so!
It’s an annoying waste of credits when things go wrong, but some of the off results are crazy. I might post some of the videos which went completely off the rails if anyone’s interested.
Assuming I don’t die…
You’re right. It was that whole clash of systems thing, and I can confirm they have not fixed it.
I’m left with recommending either use the onsite editor or use “Paste as plain text” from the Windows copy buffer. I’m not sure if Android has the same option (I don’t use Android for web browsing).
Also, if you paste as plain text text (right click menu) then you avoid such problems in the first place.
You should be able to edit your own messages. You don’t get the “edit” link in the text menu? A few weeks back there was a huge discussion about that and we managed to get it fixed. It still works as far as I know. My test user can definitely edit his own messages.
Creating the new table has been a little delayed. I seem to have caught plague over the weekend and am in no fit state to work on anything technical. I will get back to it ASAP. Sorry.
Visitors are definitely up, around 20-25%. No increase in activity though 🙁
Sadly the table looks quite thin with SO few participants listed. I’m aiming to post it in the next few days.
I wonder if any of our current or former members have caused a fuss by being seen naked somewhere unexpected.
I have to ask: Where is the video?
Okay, construction of the new ranking table is underway. I’m probably going to do it in two stages.
I need to find some way of reaching or persuading more MDL participants to let me republish their ranking table entries.
Generally I try to keep AI generation (or photo editing) down to 3-4 levels of change. There are techniques to get around that, which I’m slowly learning, but it definitely requires creative planning to get the best out of it.
Do you think it’s worth doing some sort of non-nude video so I can show it in the public area of the community? I know non-nude kind of avoids the topic of this community, but this age-verification thing is stifling discussion.
Sadly I would have to be offshore too. Sorry to see you go.
One more Qwen picture. None of them seem to do reliable male nudity unless you start out with a naked photo as a video source. Qwen is pretty good at changing the text on signs and posters.
You can see in this photo that too many tweaks and changes and you clearly lose image quality. Looks like 5-6 is fine, but much beyond that can cause problems. If you want to make the effort you can use a photo editor to neatly clone just the changes into the original photo, thus prserving the original quality.It’s still very much the case that the more effort and planning, the better the results.
There’s a lot to learn, and the technology keeps changing and advancing. What you knew yesterday may be different today. I don’t know if you’ve seen the video I posted yesterday (age-verified forum), but I’m sure I would not have been able to create that just a year ago.
HuggingFace can remove the towel. Setting up scenes in GetImg or Qwen saves my HuggingFace credits. I find it works best to use the strengths of each.