Home Creators Posts Import Register Favorites Logout

Videos

  • Brinnia_Bathing_Blur.mp4
  • Comp_v2.mp4
  • NEW randomized patients.mp4

Downloads

Content

So much stuff going on, I need a mini-section prior to the main post:

TLS: The last story in the postlude set was posted this week. I'm glad that people have enjoyed spending a little more time in this world with me. ^-^

Four Liars: Initial pixel artist search seemed to go well, I hope. This will be going on in the background for a little while and I hope to have interesting results before too long.

DStP: Made another monthly post, completing the roadmap. This project is now in a weird place where there's not that much work to be done, yet the final release might be delayed. We'll see what happens.

ToT: Man, Throne of Time has done numbers of new patrons when the higher stakes arc hit. Probably my single week record for patron growth. Readers complain about cliffhangers, but they work.

All of those things having been said, I want to make this post primarily about Slave Rehabilitator. I haven't spoken about it very frequently, but it has been a continuing project all throughout the year. Now we've finally completed one of the major goals that was holding me back, so I want to talk about it and check in with everyone.

I can't embed NSFW animations here, but if you want to get the intended experience of this post, please go look at the attached animations (especially the new one) now.

One of the things my NSFW games haven't done is animations, which are increasingly being considered the standard. I'm going to be making two attempts at this: Don't Save the Princess with part-based animation and Slave Rehabilitator with hand-drawn animation.

(For those who haven't seen or don't remember, the concept of SR is that you run a medical clinic in Hentai Land, where various fantasy characters become addicted to sex. Instead of taking advantage, you help them recover. The concept is a mix of management game and VN-style plot, with a side serving of the sorts of investments and plot-impacting decisions you all know I love.)

If you've been following this project, you've already seen Kali's custom animations for Brinnia (the female lead) which I attached as well. Hand-drawn animation is slow and expensive, however, and a game with just those would be very short - and more importantly for my interests, it would be hard to integrate such animations with more complex game mechanics.

So one of the things we've been working on for a while is a "randomized patient" system. In addition to the main characters like Brinnia, who will have custom animations and major plot segments, your clinic also gets a stream of patients created via a randomized system. Curing these patients of their addictions is part of the core gameplay loop and grants the money and resources you'll need to upgrade the clinic.

This is where we come to the new animation, which is composed of over two dozen layers so that it can be altered for any randomized patient. We started with a test set of variants, which mathematically can produce over 50,000,000 different patients. Obviously many of those would be similar, but it would be easy to increase this number with more variants without significantly inflating file size or complexity, since this was basically a proof of concept.

Getting here was actually a bit of a trial, with various missteps along the way. We were originally going to combine animations, but it turns out that due to how Ren'Py handles the alpha channel, combining animation layers was fundamentally impossible. Instead, the animations are built by the engine itself by displaying the core animation frames, a method that thankfully seems pretty light resource-wise.

So my big question: do you think this is worth it? It might sound like I'm asking for affirmation after all this work, but I'm not. Making randomized animations like this is very expensive and it's possible that the results aren't worth it. We can add a lot of variations, but they won't ultimately disguise the same underlying animation. Decisions made now would be locked in for a lot of future work, so this would be the time to pivot.

(For the record, future animations of this type would not be as expensive or slow, now that we have figured out better methods. The idea would be that we have a different random patient animation for each monster type and this is just the tentacle one. In theory, if we keep our whole system organized, in the end of the game you could get a new randomized patient and be able to see at least half a dozen different animations of them looking consistent across animations.)

With this established, there's no technical limitation to doing more of them. My problems figuring this out had to do with dumb syntax mistakes and making the file structure, not the logistical side: it'd be a fair amount of work to arrange all the traits so they only create logically-consistent characters, but that's within my zone of competence. In my ambitious moments, I think I can even include personality variables so that even the randomized patients feel like they have some substance to them. If we went with this, we'd also get multiple different voice actresses to record different sounds, just multiplying the variations further.

(If you're wondering: it wouldn't be too hard to set up a mode where players can pick traits and create their own character. This doesn't interest me much personally, but I've written before about how I think a lot of "adult games" are more "interactive experiences" and there might be interest in this sort of thing. Enough to justify the project? Unknown.)

So yes, this is the post I've been waiting to make for a while and one of the things I've been working on without mentioning too much. How do people feel about the idea of randomized animations to flesh out the management portion of the game?

Files

Comments

Desertopa

The randomized patient system looks pretty impressive to me; I really appreciate how they have significantly different expressions between them, not just different features pasted onto essentially the same framework. I think it's difficult to say at this point whether animation is a worthwhile investment for other, larger projects though. I think one of your major strengths is as a creative game designer, and you have a history of coming up with interesting ideas other people just aren't using- I think the randomized patient system is a good example. But I think your biggest strength has always been as skilled storyteller and polished writer, and I think that people who're looking for games mainly based on what have the most impressive sex graphics might have limited overlap with your core audience. That's not to say there isn't plenty of value in good art; it's pretty much always the first thing that'll grab people's eyes when they see a project, and I think when people see that a really obvious feature of a game like that is conspicuously high quality, they adjust their expectations of its other features upwards as well. But I think that with the NSFW games sphere the way it is, leading with porn animations will tend to adjust people's expectations of a lot of the strongest features of your game work downwards. How many porn-forward games are really aiming for good story, character writing, or game balance? On the other hand, for players who are drawn significantly by sexual content, whether a game has animation or not might be a make or break factor, so it's possible it'll make a substantial difference. I'll be interested to see how it affects the reception of Slave Rehabilitator and DStP, but if it doesn't seem like it'd justify itself for larger projects, I don't think it's that surprising if the best way forward for you isn't the same as what other 18+ developers are taking.

Sierra Lee

I will look forward to what various people think of the feature. There have been some patrons opposed to the randomized patients from its announcement, and I worry it's one of those features that's technically challenging for limited actual effect. Different faces were a feature I emphasized! Expressions aren't too difficult to make more of, though I'm not sure how many players can really distinguish. The thing we need to do more of but didn't is hair: hair variants are much more expensive to animate than other features. (Strangely, Kali said the breast variations were surprisingly fast, so it'd be easy to add more.) I think we're on the same page when it comes to animations and NSFW. But I don't want anyone to be able to say I didn't give it my best shot, and I hope the result will at least produce a few games that a niche audience appreciates.

Makenshi

The comp_v2 was super hot so if you're adding that quality of animation to your usual writing and gameplay mechanics, sign me the fuck up! The randomized patients were good, I liked the variety of bodies and hair (more women with pubic hair!), and while the tentacle animation was repeated the change up was enough to not feel like everything was recycled.

Sierra Lee

Thanks for your feedback! Comp_v2 shows what we can do for custom/unique scenes, whereas I hoped the randomized patients would add variety and integrate the scenes into core gameplay.

AnythingAtAll

I think Desertopa has handled the technical side very well, so I'm not commenting on that. Personally, the problem I have with randomized systems is that there's not enough 'intentionality' in them. They tend to collapse into lifeless vessels over the course of playing the game. Monster Girl 1000 tried this, and it sucked, and I think the creator regretted doing it. It was one sultry archetype copy-pasted onto various NPCs with differing personalities. It was cool for the first few iterations, then it instantly crashed hard once I realized the trick, if you will. But that being said, I think you've taken steps to mitigate that. The 50 million variations was cool, but I definitely support adding variations to the personalities in particular. Voice acting is just a cherry-on-top bonus to me. And like Desertopa said, your biggest strength is writing, in that specific brand of expansive but granular. I assume 'expansive' takes a bit of a backseat here, but if you maintain your standard attention to detail, then the problem of that 'lacking intentionality' feeling shouldn't feel prominent until waaay into the game, if it does at all. The main question for me is what the purpose of these randomized patients are. Mechanically, you've already answered, but how about narratively? To give an example, I've been playing Stardew Valley lately. There comes a point where the daily dialogue starts reading like just randomized mundane things. And technically, they are, right? But I think what the game did well is add enough variations to the dialogue pool based specifically on the seasons changing and unique events that it still feels like a living world. And, of course, it helps that 'daily mundane conversations with neighbors' perfectly fits the vibe of the game. So my suggestions are to really hone in on what the randomized patients do for the plot, then fine-tune the variations based on those. Ultimately, we're never gonna be attached to these bit characters, but if you can associate their presence to specific events, then the charm of all those variations kick in. Did a bunch of feline people fuck up a sex dungeon run? That day's gonna be 80% catgirls. Did a massive slime attack a village? Slime animation on a litany of races. Some other hentai monster attacks? Closest related animation to said monster on a litany of races. And, of course, a bunch of other long-term plot-based events that you can't spoil yet. I think that's the single best thing you guys can focus on in regards to this. The production quality is actually really high, but it can only retain its charm if there's a micro-'story' behind every variant, however brief they are.

Sierra Lee

The idea of modifying the random character generation for specific events is a really good one, actually! Should definitely be feasible. I've been thinking about how much story/personality I can give each patient and I think it's a tricky balancing act. The more substance I give them, the harder it will be to make that content seem even somewhat unique. I feel I could manage a little more than expected, but I still think of them as patients first.

Abso Haram

Hmm, I think animating randoms is a bit of a waste. Go bespoke and not broad stroke. What you could do is something fun like treating X number of race/symptoms/hometown, having certain facilities/investments or story characters unlocks Boss Randoms who can be mayors or clergy or knights or whatever that challenge you, who you can use the random generator on. Give them a little backstory so there's a reason you'd be interested to watch but you're not spending a lot of time on character writing nor do they need custom animations, use the generator. Maybe give them a progression system so after enough you can treat a queen or a saint or a guild leader etc. and then once you cap it out you get a little bonus or something, like an in game achievement system. A fun minigame you can grind toward if you are inclined to do so, and the animation would be fine which since you're not doing it for every patient stays relatively fresh. I think it could work.

Austin Cook

I think the random animations like shown are essential for other nsfw management games of this nature. That said, I have not played SR. I am simply thinking of other brothel management games with random customers/ girls. Those games feature basically only the random patient loop described here and the animations add a lot to it. However you are not only watching the animation. The animation simply plays out while you juggle your other tasks. If that is the hole these are trying to fill, then I believe this is a great way to do it. If I have to stop playing the game to view it, I would likely mash out of the cutscene.

Helm

Wall of text warning, I felt strongly about this, sorry about volume ^^ Looks impressive, and I'd certainly enjoy this system giving some form of substance to filler characters. But, if we want to go this way, i'd point two other parts that (to me at least) are just as necessary for this substance to be perceptible and hold attention: - The first you mentioned: personality. Until something *happens* as a result of an individual peculiarity, it may as well not exist. This could be as simple as 3 line blurbs interactions linked to char traits (or being ambitious, *combinations* of traits), firing rarely enough to avoid repetition. - Second: sound design, including voice, becomes way more critical with animations. Some layer of mixing audios to *match* the animation + personality would really be the icing on the cake. Now to the original question that was 'is it worth it ?', it of course depends on cost, and you are the only one able to evaluate it against returns. However please include in your arbitration the fact that such a system, once invested in, can be reused. If you address *in general* the problem of 'how to use animation+audio+text to give a character substance ?', you have something that you can apply in pretty much *all* subsequent project ! The first take on it does not have to be perfect to allow building on. Last, I'd like to advocate use of AI in a task where in my personnal experience it revealed well suited, NOT content (or even code) generation, but as help in *designing complex systems*. If like me you enjoy building systems by laying out axioms and defining evaluable objective *before* actually considering solutions, you may, like I did, find current llm agents an invaluable help in challenging your logic for inconsistency and incompleteness. Replicate the rubber duck situation where you explain your whole reasoning to find holes in it, but with the duck running an actual sanity check. This along with agents excellent ability in building state-of-the-art-s (aka 'how have others addressed problem X ?') shaved weeks off some personnal projects of mine. Using those tools is a skill in itself though, and I only really felt a gain in productivity after roughly a month of consistent use and refining my usage. Still, you feel to me *rational enough* to reach positive roi fast (and actually learning something *really new* was all kinds of fun). Sorry for the side-rant, but I feel it applies and I'm getting invested ^^'

John Greymore

I think you can have your cake and eat it too. I agree with Austin in that if it's animation only I'm gonna skip it. Even as much as I liked TLS, I still skipped some of the scenes (sorry!) I think you can use the repeat animations to further subvert genre expectations. Even if everyone ends up liking tentacles, I imagine they might like it for different reasons. Since you already have different facial features, adding some dialogue or description (and variations of that) pointing to what each customer appreciates or enjoys can keep things fresh for longer. Or, you could subvert player expectations of when they expect the animations and hit the player the complete tentacle treatment before "regressing" from there to various flavours of vanilla. I think this might be a good way to entice mainstream/new players who know the genre but aren't familiar with your work. I imagine the animation for the complete treatment having little variation (every customer is the same), and then the animations get somewhat more complex when their own personality and preferences shine through.

AnythingAtAll

Also, is the setting literally named Hentai Land lol.

Sierra Lee

This system is going to fall short of your expectations. We can make a simple loop that accommodates endless patients, but they aren't going to get animated variations.

Sierra Lee

Ha, no. Just called it that for demonstrative purposes.

John Greymore

That's ok! I don't play this genre, and I have no idea about game development. All I know is that familiarity breeds indifference if not contempt, and looped animations tend to get familiar relatively quickly. I think ultimately, my suggestions are to limit how many times the player will see the animations (shorter progression for example) or giving flavour to the animations in terms of story like what AnythingAtAll suggested or dialogue/description via text would be my go to.

Atashkada

Some of those random patients are very cute, and I recently played a game called "Imperial Gatekeeper" which also featured random characters as the bulk of people you process. There is a late-game item (not really a spoiler I think) that you can use to "lock in" one of those random people if you like them, which makes them come back repeatedly. That seems like a feature that would be good to include, if you do go with having random patients set up like that.

Mycroft

For me, randomized patient does not work very well, because it is only a reskin with the same position/expression. So it feel like always the same scene, even if the character is different. But I come here for the game part, not for the NSFW part. I like that the game include NSFW part because it make the character more relatable, more alive, but your games are not, for me, NSFW games, but games with NSFW parts. The NSFW scene are often more "wholesome" than very hot, for me. Therefore, animation will not make the game really "better" for me, but I'm not necessarily representative of your player.

BlueTemplar En Taro Brian

Layer-composed variations indeed do not necessarily have to be (fully) random. I'm reminded of Sid Meier's, err, Brian Reynolds' Alpha Centauri, which is very strong at both 'management game' and 'writing' sides. There's at least 3 places where a (mostly non-random) "combined parts" system is used there : music, unit design, factional traits. (Though it should be noted that all of these are very far from the human body, which humans are *extremely* sensitive for errors in representation.)