- Thread starter Deleted member 503379
- Start date
Honestly I can't predict that either, part of the reason I said I would prefer the other mod to be continued is because with that one you just transform the world into freeuse from the start, no convoluted planning or initial plot devices needed. With the new one I feel there's much more risk of the train wrecking just because of too many things to develop piling up. Unless the MC gets teleported to sexworld as soon as the cult cutscene ends.
Edit: I also personally prefer just the freeuse/sexworld as it is, without the 'normal' world getting in its way in any shape or form, just discard the 'normal' and go full sexworld.
Last edited:
- Aug 19, 2019
- 78
- 80
Im so confused, ive been reading comments and my version still has clothes on and stuff or have i just not progressed enough? I think i have 0.1
At some point the old version was replaced with this new one, I guess because the mod was restarted with another focus in mind. Go to post #257 if you want to find the original mod, I hope that mega link is still alive.
- May 6, 2021
- 29
- 27
maybe im fucking retarded but the link of download is not the game is an old version of less than 500 mbts
- Aug 9, 2016
- 498
- 953
We are still working on it but because of some of my real life obligations I don't have as much time to write. I understand the expectations being placed on us, and please understand that neither ipreg or I have given up on the mod.
Thanks for the status update, very glad this isn't dead.
- Apr 1, 2019
- 63
- 36
Merry Christmas lads, hope you guys have a wonderful time and an excellent new year
Hey there. Hope you’ve had a good start into the new year!
Recently, I got to mess around with stable diffusion. Apparently, you can now simply download it as a free program and run it as you wish, which I did.
It reminded me that a year ago, you tried to get AI-generated art to work for the mod. Since there have been, as far as I can tell, absolutely massive improvements in usability and function, have you given the idea a second thought, ipreg? I think you mentioned that the sample size was too small for a training set, but operations like in- and outpainting seem to work decently well for copying a style if it does not change too much, from what I’ve seen.
Personally, my computer is just too weak to do much with the program. The client I tried out (InvokeAI) also does not yet have training featured enabled in the simple interface. The few images, variations, outpaintings I *did* make to test stuff were rather easy do, though, aside from having to wait ages for each operation.
Recently, I got to mess around with stable diffusion. Apparently, you can now simply download it as a free program and run it as you wish, which I did.
It reminded me that a year ago, you tried to get AI-generated art to work for the mod. Since there have been, as far as I can tell, absolutely massive improvements in usability and function, have you given the idea a second thought, ipreg? I think you mentioned that the sample size was too small for a training set, but operations like in- and outpainting seem to work decently well for copying a style if it does not change too much, from what I’ve seen.
Personally, my computer is just too weak to do much with the program. The client I tried out (InvokeAI) also does not yet have training featured enabled in the simple interface. The few images, variations, outpaintings I *did* make to test stuff were rather easy do, though, aside from having to wait ages for each operation.
- Oct 6, 2020
- 19
- 64
First sorry about the long absence, my IRL life kinda went nuts and I didn't have much time for the mod. I have been able to convert it to Renpy 8/Python 3. Outside of that I mostly failed at customizing Ren'Py itself in various ways and failed at fixing/installing the many different python enviroments I've had to use.
I also tried various AI-related stuff outside of/before Stable Diffusion:
-
-
- img ->
- Using
-
-
-
-
At least this past year I actually did stuff beyond
P.S: It's actually really awesome that the authors of Stable Diffusion made it
I also tried various AI-related stuff outside of/before Stable Diffusion:
-
You must be registered to see the links
+ ? (couldn't get it to work well enough)-
You must be registered to see the links
+ ? (I really wanted the
You must be registered to see the links
)- img ->
You must be registered to see the links
-> Stable Diffusion- Using
You must be registered to see the links
for something?-
You must be registered to see the links
(Fuck this one so hard - impossible to get running)-
You must be registered to see the links
(Don't have a character sheet)-
You must be registered to see the links
(Just portraits)-
You must be registered to see the links
(I never actually got around to trying this one, just putting it here as a reminder)At least this past year I actually did stuff beyond
You must be registered to see the links
because
You must be registered to see the links
has been a godsend now that things like models, their weights, and datasets are available. Fuck every single computer scientist who did something with the
You must be registered to see the links
dataset, beause it felt like when I would find a potentially useful AI it couldn't be used due to being trained on that fucking dataset.
You must be registered to see the links
and while it's good for a literal first attempt, the fact that it took me like 2 months to make means that I am not going to be our artist any time soon. The place where we want to take the story is going to need custom art, and quite a bit of it I feel, so until some kind of art workflow is created - whether by a volunteer or AI - it'll be very difficult to make progress.I agree! I seem to have accidentally been a few years early in my AI art efforts as this past year has been absolutely crazy. I actually spent some time and money training a stable diffusion style in the style of the backgrounds:
You must be registered to see the links
,
You must be registered to see the links
,
You must be registered to see the links
,
You must be registered to see the links
. In fact, literally this week
You must be registered to see the links
was released and that might actually be close to what we need for automatic art generation and animation. I am still installing it/getting it set up so I don't know how well it'll work. I also think at one point I complained about the lack of NSFW AI models; well that
You must be registered to see the links
- so I am really hoping that somewhere in this unbelievable explosion in AI art there will be tools that can help us remix existing assets effectively enough to be of use in the game. Of course I am just one person so if there seems to be a useful tool, let us know about it.P.S: It's actually really awesome that the authors of Stable Diffusion made it
You must be registered to see the links
and didn't horde it like OpenAI and Google did. We were very close to not having this capability available to the public.
Last edited:
- Oct 2, 2017
- 227
- 417
- Oct 6, 2020
- 19
- 64
Quickest update I'll do as I am going to be very busy this spring again 
But I kinda got sucked into trying to get stable diffusion to do what I want and find the results pretty funny. So here is the results of haphazardly blending the ChilloutMix SD Checkpoint, the prompt "Man in a Suit", a kurzgesagt style, an image of a man raising his right arm generated from the prompt "A man in a suit with his right arm raised" with the openpose preprocessor, and the POV character standing in a black void. I find it funny that the only thing which survived seem to be the general face, so it's like the POV character from an even stranger AU.
But I kinda got sucked into trying to get stable diffusion to do what I want and find the results pretty funny. So here is the results of haphazardly blending the ChilloutMix SD Checkpoint, the prompt "Man in a Suit", a kurzgesagt style, an image of a man raising his right arm generated from the prompt "A man in a suit with his right arm raised" with the openpose preprocessor, and the POV character standing in a black void. I find it funny that the only thing which survived seem to be the general face, so it's like the POV character from an even stranger AU.Thanks for the update, and yeah "Open"AI’ projects seem to have become really closed really fast. It was a huge surprise to learn about stable diffusion being freely available in various implementations. Even more, that some are optimized enough to only need a few gigabytes of RAM to (barely) work.
It is a mixed blessing that it’s so slow for me. On the one hand, I’d love to mess around with it a lot more. On the other, this seems like it could get addicting really fast.
"strange" is still the thing these AIs seem to be best at, so that makes sense…
- Mar 8, 2019
- 517
- 599
I am happy to hear this is still alive if on life sup[port
it has so much potential.
it has so much potential.