Yeah, I’d love to add 2D animation once I have enough money to hire an animator.
- Thread starter LocusTenebrarum_Rue
- Start date
The problem is that the AI struggles to do things that are highly detailed in a small area of the image. Hands and faces that take a lot of the image are more likely to be okay, while faraway ones tend to be a struggle.
Yeah, that’s on the roadmap. Unfortunately, I am not very good at animation. I’ve made some attempts, but they weren’t good enough to put in the game.
Once I have some more Patreon money coming in, I would love to hire an animator and make this game look amazing.
Wow,, 3 hot damn sexy exquisite beauty waiting to be deprave by old man. Old man.......we got new target. I will choose the black hair first. She's delicious. 
Where is that damn old man? Gangbang mode activated
Where is that damn old man? Gangbang mode activated
I’m not sure why your comment got removed, but I promise that wasn’t my doing.
You’re actually not the first person to ask for a Linux version, so it’s definitely on my radar. Right now the intro to the game works, but I just get a black screen once it gets to the 3D environments. I hoped that the resolution and windowing modes might fix it when I added them last night, but it didn’t.
I have no idea why it’s not working, and I have a ton of other stuff I need to add at the moment. Between my day job and this game, I’m working 70-90 hours per week. I would love to dig in and get this working, but there’s just so much to do, and no time to do it. I promise that this is important to me, and I hope to get it some day. I just can’t promise that it will happen soon.
Strangely enough, the Windows version seems to run better under Wine than the Linux version. I’m not sure what’s up with that. Apparently this black screen issue happens with Unity games sometimes, but I haven’t found a solution yet.
No, the report specifically said it was deleted for asking for a port/translation/etc. Which doesn't make any sense to me especially if no such comment had been made in the thread before. I did search for "linux" and there was nothing. The comment by the dev you posted is in fact a response to my asking for a linux port but the quote to my comment got deleted with my comment being deleted as well.
You can skip, though it’s pretty glitchy because of the way it works. You can hit the zero key on your keyboard to make it skip through a scene, but it often leaves the game in a bad state. I wouldn’t encourage players to use this because it behaves really badly a lot of the time.
Yeah, that continue button needs to be disabled if you don’t have a save. Sorry, the game is still new and I only added the start screen a few days ago, so it hasn’t had much dev time yet.
I’ll add that to the bug tracker and try to get it into the release today/tomorrow.
As for the slowdown, that’s odd. I’ve gone through that scene a bunch of times, as have other players and this is the first time I’ve seen it reported. I’ll keep an eye on it and see if more reports come in.
- Oct 31, 2017
- 2,016
- 3,516
Hi,
First and foremost, thank you for the share. Well done for the great work so far. All the best.
If I may suggest, I believe that instead of starting small or big, you start this (first?) Unity game of yours with what you are most comfortable with, at the pace you are most comfortable at and using the tools you are most competence at.
That being said, considering animation is not your forte (and you will depend on third party for that), I totally support your decision to hold off animation for the moment. No one requires your game (or any game) to have animation in the first place, but yeah it will be less cool without animation. However, worst if the animation is not good, it ruins the others that are great about this game.
Anyway, it is just my opinion and I am not a game developer myself so I don't know stuff
. Cheers.
I’ve experimented with a Mac build, but it’s not running correctly on my M1 Mac. It’s on my radar. I may even experiment with an Android build someday, though the performance requirements might be a little intense for low-powered mobile devices.
Yeah, the 3D lewd scenes are done. I’m only doing 2D ones from now on. Once I have enough Patreon earnings to hire an animator, I’d like to add 2D animations.
Firstly, Detail Per Pixel.
Basically the AI allocates a certain amount of processing power to a certain average of pixels. Since fingers are dense and high detail, it's basically like trying to draw the Mona Lisa in Minecraft. you're going to need a lot more pixels and a lot more power.
Secondly, Perspective Logic and Shape ID.
it's difficult to teach the AI perspective of abstract shapes. You can easily understand which way a Cube is facing based on the corners and shadows. But a squigly worm like finger... that's near impossible. Even people sometimes have difficulties identifying abstract shapes from Still images. It's the whole idea behind Optical Illusions.
For an AI that was born Yesterday, pretty much every abstract image is an Optical Illusion.
(abstract may not be the right word. But I'm trying to say it's not a simple shape)
if you want to know a little more about AI Images.
You must be registered to see the links
Last edited:
I’m strangely fond of the “chatter” sound, but as a compromise I’ve added a volume slider for the chatter so you can mute it without losing the rest of the SFX. I hope to ship it in a new build tonight or tomorrow.
I came here to praise your artist. Turns out it was an AI. Not sure how I feel about that, but it's still very impressive.
I'm curious, how does the AI handle multiple subjects, different environments/poses/expressions and genitalia? From the AI art I've seen until now, the program always struggled with these things. Genitalia looked completely wrong, different subjects tended to blend into one instead of remaining distinct and it had a hard time with varying poses.
I'm wondering how effectively it can give you a nice variety of images from text prompts alone.
I'm curious, how does the AI handle multiple subjects, different environments/poses/expressions and genitalia? From the AI art I've seen until now, the program always struggled with these things. Genitalia looked completely wrong, different subjects tended to blend into one instead of remaining distinct and it had a hard time with varying poses.
I'm wondering how effectively it can give you a nice variety of images from text prompts alone.
Thank you. It turns out that (as you said) the AI struggles with complicated images. For that reason, I end up doing a fair bit of work on most images going back and forth between Photoshop and the AI.
I estimate that I spend between 45 minutes to two hours on most images depending on complexity. Obviously that’s a lot faster than I could paint these images, but it’s not exactly low effort, either.
I’ve started to think of using AI generated art for lewd games as being a bit like using DAZ, just with a different set of trade offs. It still takes time and effort, but where DAZ can be a little aesthetically boring, AI struggles with consistency instead.
Is it possible to train the AI to be able to distinguish a figure or face and apply it to a specific named character as a prompt? Kind of a like a pregenerated seed it saves the distinguishing features of the character as a seed that you can use as a text prompt.
I imagine as an artist it would also be frustrating to get it to give you something close to the exact image you have in your head.
Yes, Dreambooth does what you’re describing with varying degrees of success.