Musttaffa

Newbie
Aug 9, 2025
19
43
38
You worked on the game, right? Or maybe you’re working on it again—I’m not entirely sure. And instead of actually addressing the criticism, you’d rather defend a user who created something like 20 accounts just to like their own posts and react with laugh emojis to anyone who argues against them.

Instead of protecting someone like that and attacking someone else who, in my view, is making reasonable arguments, why don’t you actually engage with the criticism? You were part of the team, or maybe still are—I don’t know—but why is it so hard to respond constructively to the criticism you’re receiving, especially regarding the use of AI?

I haven’t read a single solid argument that actually defends this AI slop in Amnesia. Not one. Aside from SuperAlex saying “AI is the future and you have to use it immediately or you’ll be too late,” which in my opinion is complete nonsense. But I still haven’t seen a single well-thought-out discussion explaining what exactly is better about this AI slop in Amnesia compared to the earlier handcrafted scenes, which, in my view, looked a thousand times better. So go ahead—say something about that.
зачем мне что-то говорить по этому поводу? Мне нравится AI, так же мне нравится 3D. Когда я показываю своему знакомому игры, которые мне нравятся - он отвечает, что предпочитает традиционное порно, а вот эти "3D куклы" ему непонятны и кажутся странными... Когда-то мне были, например, омерзительны мысли об инцесте, а потом моё мнение изменилось. Зачем мне переубеждать вас, навязывать своё мнение и тд? Это из разряда, если я сейчас пойду на форум фанатов серии соулс, или вов, и начну доказывать там, что их линейка игр по-моему мнению мусор, и буду требовать, чтобы они как-то оправдались передо мной. Просто каждому своё. О вкусах не спорят. Если мне что-то не нравится, то я просто отказываюсь от этого в пользу того, что мне нравится.

Why would I even say anything about that? I like AI, and I also like 3D. When I show my friend the games I’m into, he says he prefers “traditional” porn, and these “3D dolls” don’t make sense to him and seem weird... At one point, for example, even the idea of incest disgusted me, and then my opinion changed. So why would I try to convince you, push my views on you, etc.? That’s like if I went to a forum for Souls fans or WoW fans and started arguing that their whole game lineup is trash in my opinion, and then demanded they justify themselves to me. Everyone likes what they like. You don’t argue about taste. If I don’t like something, I just drop it and stick to what I do like.
— translated by ChatGPT
 

Sony676

Engaged Member
Mar 14, 2021
2,154
5,707
698
зачем мне что-то говорить по этому поводу? Мне нравится AI, так же мне нравится 3D. Когда я показываю своему знакомому игры, которые мне нравятся - он отвечает, что предпочитает традиционное порно, а вот эти "3D куклы" ему непонятны и кажутся странными... Когда-то мне были, например, омерзительны мысли об инцесте, а потом моё мнение изменилось. Зачем мне переубеждать вас, навязывать своё мнение и тд? Это из разряда, если я сейчас пойду на форум фанатов серии соулс, или вов, и начну доказывать там, что их линейка игр по-моему мнению мусор, и буду требовать, чтобы они как-то оправдались передо мной. Просто каждому своё. О вкусах не спорят. Если мне что-то не нравится, то я просто отказываюсь от этого в пользу того, что мне нравится.

Why would I even say anything about that? I like AI, and I also like 3D. When I show my friend the games I’m into, he says he prefers “traditional” porn, and these “3D dolls” don’t make sense to him and seem weird... At one point, for example, even the idea of incest disgusted me, and then my opinion changed. So why would I try to convince you, push my views on you, etc.? That’s like if I went to a forum for Souls fans or WoW fans and started arguing that their whole game lineup is trash in my opinion, and then demanded they justify themselves to me. Everyone likes what they like. You don’t argue about taste. If I don’t like something, I just drop it and stick to what I do like.
— translated by ChatGPT
Then I think I’m well within my rights to point out that Amnesia stood for a long time for its wonderful visual aesthetic with handcrafted animations. That’s the reason I fell in love with this game in the first place. Yes, it has always been a slow burner and it never really picked up full momentum, but now that the course has suddenly been changed toward this AI slop, it also loses, in my opinion, the very last bit of quality it still had.

And I think that’s a completely fair point to make, because visually this game hit my taste perfectly. And it’s obvious that many people don’t like the direction this is now taking with AI-heavy content. So my question is simply: why take this step? Is it just to make the work easier and to move faster after years of very little real progress? Or what exactly is the reason?

Because in my view, this isn’t just a matter of personal taste. It’s something that can be backed up by facts: AI, in this context, is simply not at a level yet where it can fully replace handcrafted animations in terms of aesthetics. Especially not when you’ve previously built a framework that was entirely based on handcrafted renders and animations, and then suddenly decide to slap AI-generated content on top of it.

It doesn’t feel coherent. It doesn’t feel authentic. It feels like two things being forced together that simply don’t fit. And I think it’s more than fair to raise that question without being told that this is “just subjective” or “just personal preference.” That’s nonsense. I’m laying out concrete arguments here, and I’d honestly like to get a fact-based, logical answer that an average user can understand—instead of having this dismissed outright without any real response.

Because the way it’s handled right now comes across as if there’s no clear plan, and things are just being justified however it happens to suit the moment. And that’s exactly what a discussion forum is for: engaging in discourse and responding to constructive criticism in a meaningful way. Unfortunately, that’s what I feel is missing here.
 

muppetman61

New Member
May 13, 2023
1
0
43
Are the remaster downloads bugged? I have downloaded it with 3 different links and have extracted them all and when i click on the applications in the game they won't run
 

Ashira13

Engaged Member
Jan 31, 2024
2,243
6,263
641
Are the remaster downloads bugged? I have downloaded it with 3 different links and have extracted them all and when i click on the applications in the game they won't run
try the split mega or pixeldrain download mirrors. i personally don't trust any of the ones in the op.
 

Musttaffa

Newbie
Aug 9, 2025
19
43
38
Then I think I’m well within my rights to point out that Amnesia stood for a long time for its wonderful visual aesthetic with handcrafted animations. That’s the reason I fell in love with this game in the first place. Yes, it has always been a slow burner and it never really picked up full momentum, but now that the course has suddenly been changed toward this AI slop, it also loses, in my opinion, the very last bit of quality it still had.

And I think that’s a completely fair point to make, because visually this game hit my taste perfectly. And it’s obvious that many people don’t like the direction this is now taking with AI-heavy content. So my question is simply: why take this step? Is it just to make the work easier and to move faster after years of very little real progress? Or what exactly is the reason?

Because in my view, this isn’t just a matter of personal taste. It’s something that can be backed up by facts: AI, in this context, is simply not at a level yet where it can fully replace handcrafted animations in terms of aesthetics. Especially not when you’ve previously built a framework that was entirely based on handcrafted renders and animations, and then suddenly decide to slap AI-generated content on top of it.

It doesn’t feel coherent. It doesn’t feel authentic. It feels like two things being forced together that simply don’t fit. And I think it’s more than fair to raise that question without being told that this is “just subjective” or “just personal preference.” That’s nonsense. I’m laying out concrete arguments here, and I’d honestly like to get a fact-based, logical answer that an average user can understand—instead of having this dismissed outright without any real response.

Because the way it’s handled right now comes across as if there’s no clear plan, and things are just being justified however it happens to suit the moment. And that’s exactly what a discussion forum is for: engaging in discourse and responding to constructive criticism in a meaningful way. Unfortunately, that’s what I feel is missing here.

Окей. Я постараюсь ответить. Супер Алекс решил идти в этом направлении сам, ещё когда только появился Stable Diffusion. Он просто пришёл и поставил задачу в этом направлении (изучить - я хочу вот это и это...). Уже тогда он питал надежды, что с помощью этого можно будет делать анимации. Так же было и когда появился WAN. Поэтому нет. Это не решение "с бухты-барахты". Как он сам говорил : "чтобы сделать такое же количество анимаций "вручную" потребовалось бы десяток ТОП аниматоров и несколько месяцев работы, что несопоставимо". В общем. Он хочет много анимаций в игре, но покупать одну анимацию с 4 ракурсов за 300$ - он не может (точнее, он может себе позволить одну или две такие анимации, но это не сопоставимо с сотнями ИИ анимаций). Так что, думаю, что я внёс ясность в этом вопросе. Что касается качества ИИ. Теперь анимации в ИИ он делает в основном сам. Вот что он ответил под одним из постов : "ИИ тоже набирает обороты, эта сцена была можно сказать первая серьезная проба пера. Пока что ИИ еще не дотягивает до тех идей которые я хочу реализовать) но это, уверен, временно". Так что, надеюсь, у него всё-таки получится вернуть ваше расположение.

Okay. I’ll try to answer.
Super Alex decided to go in this direction on his own back when Stable Diffusion had just appeared. He basically showed up and set himself a goal in that area (to explore it — “I want to do this and this…”). Even back then he was hoping it could be used to make animations. Same thing happened when WAN came out. So no, this wasn’t some random spur-of-the-moment decision.
Like he said himself: “To make the same number of animations manually would take a dozen top-tier animators and several months of work — it’s not even comparable.”
Bottom line: he wants a lot of animations in the game, but he can’t afford buying a single animation from 4 angles for $300 (or rather, he can afford one or two, but that’s nothing compared to hundreds of AI animations). So I think that clears it up.
As for AI quality: now he mostly makes the AI animations himself. Here’s what he replied under one of the posts: “AI is gaining momentum too — that scene was basically my first serious attempt. For now AI still doesn’t quite reach the ideas I want to pull off, but I’m sure it’s temporary.”
So hopefully he’ll still manage to win you back.
— translated by ChatGPT
 

Sony676

Engaged Member
Mar 14, 2021
2,154
5,707
698
Окей. Я постараюсь ответить. Супер Алекс решил идти в этом направлении сам, ещё когда только появился Stable Diffusion. Он просто пришёл и поставил задачу в этом направлении (изучить - я хочу вот это и это...). Уже тогда он питал надежды, что с помощью этого можно будет делать анимации. Так же было и когда появился WAN. Поэтому нет. Это не решение "с бухты-барахты". Как он сам говорил : "чтобы сделать такое же количество анимаций "вручную" потребовалось бы десяток ТОП аниматоров и несколько месяцев работы, что несопоставимо". В общем. Он хочет много анимаций в игре, но покупать одну анимацию с 4 ракурсов за 300$ - он не может (точнее, он может себе позволить одну или две такие анимации, но это не сопоставимо с сотнями ИИ анимаций). Так что, думаю, что я внёс ясность в этом вопросе. Что касается качества ИИ. Теперь анимации в ИИ он делает в основном сам. Вот что он ответил под одним из постов : "ИИ тоже набирает обороты, эта сцена была можно сказать первая серьезная проба пера. Пока что ИИ еще не дотягивает до тех идей которые я хочу реализовать) но это, уверен, временно". Так что, надеюсь, у него всё-таки получится вернуть ваше расположение.

Okay. I’ll try to answer.
Super Alex decided to go in this direction on his own back when Stable Diffusion had just appeared. He basically showed up and set himself a goal in that area (to explore it — “I want to do this and this…”). Even back then he was hoping it could be used to make animations. Same thing happened when WAN came out. So no, this wasn’t some random spur-of-the-moment decision.
Like he said himself: “To make the same number of animations manually would take a dozen top-tier animators and several months of work — it’s not even comparable.”
Bottom line: he wants a lot of animations in the game, but he can’t afford buying a single animation from 4 angles for $300 (or rather, he can afford one or two, but that’s nothing compared to hundreds of AI animations). So I think that clears it up.
As for AI quality: now he mostly makes the AI animations himself. Here’s what he replied under one of the posts: “AI is gaining momentum too — that scene was basically my first serious attempt. For now AI still doesn’t quite reach the ideas I want to pull off, but I’m sure it’s temporary.”
So hopefully he’ll still manage to win you back.
— translated by ChatGPT
I really appreciate your response, because it’s much more transparent and actually helps fans understand the thought process behind it. That’s honestly all I need.

Even if I personally don’t like the direction the development has taken (so far), I can at least logically understand Super Alex’s mindset now. I think a lot of devs fail exactly at this point — being open and transparent like this — even though it would save them a lot of criticism and unanswered questions in the long run.

So yeah, thanks a lot for taking the time to give such a detailed answer. I genuinely appreciate it.
 
May 6, 2024
57
77
155
Bazen44 left a comment on , and they gave this recommendation for a pipeline to improve the video quality. I only have a 3090 here, nor do I have the assets or the time/skills to rough out animation keyframes for the game. But it's food for thought for others if they want to experiment with this pipeline:

Here is how to eliminate mutations using freely available tools:
1. Hybrid method: EBsynth + Keyframes This is the most reliable way to maintain 100% identity. Instead of AI generating each frame anew, you use the EBsynth software.

Procedure: You render a 3D scene. You select only a few keyframes that you adjust to perfection in AI (e.g. in Stable Diffusion). EBsynth then “spreads” these adjusted pixels to the other frames according to the movement from the original render. This way, Kate’s face does not change between frames because it just deforms according to the movement, instead of being redrawn.

2. ComfyUI and advanced nodes If you want to stay pure in AI, you have to leave simple web generators and start using ComfyUI. There are specific nodes designed for this:

PuLID-Flux or IP-Adapter FaceID: These models can fix facial features with an accuracy of over 90%. AnimateDiff with ControlNet: AnimateDiff ensures that pixels between frames seamlessly connect, while ControlNet (Depth or Canny) forces the AI to stick to the exact contours of your 3D character in every millisecond of movement.

3. Custom training (LoRA) The surest way to prevent mutations is to teach the AI what Kate looks like.

You create a so-called LoRA model (a small data set) by “feeding” the AI 20-50 high-quality static renders of Kate from different angles. You then activate this model when generating the video. The AI will have her exact cheekbone and eye structure in “memory”, so even if it needs to calculate something, it will stick to the learned template.

The hardware (RTX 5090) will give you the necessary power to perform these complex calculations locally and quickly.
If Super Alex can actually nail down that precise and consistent facial and body geometry by using a LoRA of each of the girls, so that it actually does end up superior to his current manual 3D model pipeline, then we might really be looking at an actual increase in quality for the game, where he'll be able to crank out videos faster and in higher quality, and leave himself more time to develop the game over-all.

We can dream. ;)
 

FineUser123

Member
Apr 23, 2025
110
558
154
If Super Alex can actually nail down that precise and consistent facial and body geometry by using a LoRA of each of the girls, so that it actually does end up superior to his current manual 3D model pipeline, then we might really be looking at an actual increase in quality for the game, where he'll be able to crank out videos faster and in higher quality, and leave himself more time to develop the game over-all.
Dev shamelessly used in his game the first generated results from the first googled free AI app, knowing that his mindless herd still will eat his lame, lazy bullshit, and guys here still think and speculate about how dev could make this lazy trash look better, while he didn't even intend to make quality AI work in the first place :WeSmart:

P.S. Even more than half of the typical AI slop from the AI segment of R34 looks way better than whatever Super Scammex "created."
 
  • Like
Reactions: misterx2999

WxpCOMan

Newbie
Aug 6, 2020
16
3
110
Does anyone has remake version guide? I unlock the achievement except Kate's socks, Elena's panty, all Iris achievements and Eva's lolipop, how to solve these achievements?
 
Last edited:

cxx

Message Maestro
Nov 14, 2017
78,198
40,581
1,570
Does anyone has remake version guide? I unlock the achievement except Kate's socks, Elena's panty, all Iris achievements and Eva's lolipop, how to solve these achievements?
check wt.pdf of old version.
 

Gurdas3d

Newbie
Mar 28, 2025
37
29
62
There would be no problem if AI could create images that were indistinguishable from those created in Daz Studio. But, at this stage, AI-generated content ruins the game's immersion and does not match the quality previously achieved by Alex's manual creation. If Alex wants to use AI, that's fine — sit down, study AI and improve your skills. Only when you can produce decent results that are indistinguishable from Daz images should you use AI and add it to your project. Right now, AI produces mediocre content. The main problem with most developers today is that they create AI content while only just beginning to learn about it. First, master the tool and gain complete control over it. Then there won't be any discussions about the character's face, body shape or twitching changing.
 
  • Like
Reactions: misterx2999
3.40 star(s) 247 Votes