Ren'Py Serenia [Development Thread][NSFW Fantasy RPG][2D Art]

[OffWhite Studios]

Dev Blog: Serenia

Hi Everyone,

I’m Offwhite Studios, the developer behind an upcoming game I’m very excited to share with you: Serenia, an NSFW Dungeon Crawler. This is my first game, and I’m excited to finally be at a stage where I can connect with the community here and start sharing my progress. The first part of the game is nearly ready for initial release, and I’m aiming for the release in just a few weeks (aiming for March).

In this development blog, I’ll be sharing details about the game, its features, and how it’s shaping up. I’m also here to hear from you—your questions, comments, and suggestions are welcome as I work toward release.

Thanks for checking out my work!

About the Game
Serenia is a RPG Dungeon Crawler packed with NSFW scenes. You play as Anno, the Court Wizard of Serenia. The plot kicks off when a prophecy foretelling the rise of a mysterious dark lord and an army set to descend upon the city. Despite your warnings, the Queen dismisses your pleas to prepare, leaving you no choice but to take matters into your own hands. To save Serenia and those you care about, you must grow stronger. Your journey takes you into ancient, treacherous labyrinths, where forgotten magic lies buried, waiting to be claimed. But the road to salvation is fraught with danger. The land is as perilous as it is seductive, with threats that will test your strength, resolve, and morality—and plenty of NSFW situations to stumble into.

Current Status:
As of now, I’m thrilled to say that Chapter 1 of the game is 90% complete. Although it's just the first Chapter, there is hours of content here. Here’s a quick rundown of where things stand.

What’s Done:

• Core Gameplay Systems: Art (characters, locations, items, etc), Combat, leveling mechanics, item mechanics, and monster encounters are complete.
• Story and Dialogue: All conversations, scenes, and story content are finished.
• NSFW Content: The game features around 21 unique, animated scenes, all implemented except for one final sequence.

What’s Left:

• About half the animations (the art is done, just not the animation)
• Final Polishing: Cleaning up animations, refining a few late-game scenes, and tightening the presentation.
• Playtesting: Ensuring balance and smooth performance across gameplay scenarios.
• Dialog Choice Impact: Right now, most choice only change the dialog, not long-term relationships.

Content:

• 21 Animated NSFW Scenes: The game includes 21 unique, explicit encounters, each featuring its own animation.
• Diverse Cast of Characters: Meet and build relationships with dozens of characters, each with their own personalities and storylines.
• Interactive Dialog Choices: Shape your journey through meaningful dialog options that impact relationships and outcomes. (May not be in first release)
• Strategic Turn-Based Combat: Engage in battles with a variety of enemies, using items and leveling up to grow stronger.
• Dungeon Exploration: Navigate dungeons filled with traps, puzzles, and minigames that challenge your skills and wits.
• Unique Monsters: Encounter dozens of enemies, each with distinct resistances, attacks, and loot to discover.
• Scene-by-Scene Open World: Explore an interconnected world one scene at a time, uncovering secrets and progressing your story.

Tags and NSWF Themes:
Here is a list of what type of NSWF content you can expect in the game.

• Vanilla Sex / Romantic
• Non-Con / Forced
• Furry / Monster Girl
• Gangbang
• Pregnancy
• Minigirl (fairy, etc)
• Corruption
• Humiliation
• Groping
• Mind Control
• Big tits
• Big ass
• Virgin
• Harem

Release Plans
My goal is to release the first version of the game in March-April 2025, but it could be earlier - who knows?

The Road Ahead
This project has taken a year and a half to reach this point, but things will move faster from here. When I started, I had no experience with programming, AI, or animation, and I was building everything on outdated hardware (GTX 1070). Now:

• I know what I’m doing with Ren'Py, AI tools, and After Effects.
• The base code and character templates are complete.
• All the major systems are in place.
• I can focus on creating new content and refining what’s already there.

I plan to upgrade to the NVIDIA 5000 series in March, which should 50x performance. This means you can expect much better art and faster updates! After the initial release, I aim to deliver updates every 2-3 months, focusing on adding new chapters, refining animations, and expanding the game world.

Patreon and Monetization

This game is, and will remain, completely free. I plan to release it on F95Zone for everyone to enjoy, with the full experience available without needing to join Patreon. Creating this game has been a passion project for me, and I want as many people as possible to experience it without barriers.

That said, I’m setting up a Patreon for those who would like to support my work. Your support would mean a lot and help offset the costs of hardware upgrades, software tools, and the countless hours I’ve invested in this project.

For Patreon supporters, I’d like to offer some perks, such as:

• Taking requests for adding new content or features to the game.
• Hosting voting polls to help decide on scenes or creative elements.
• Sharing behind-the-scenes content, such as unused artwork, concepts, or cut ideas.

More perks may come later as I get more familiar with using Patreon.

Rest assured, the game itself will always remain free. Patreon is simply an optional way to support the project and become more involved in its development.

You must be registered to see the links

Thank you for considering it, and most importantly, for playing and supporting the game!

 

[OffWhite Studios]

Progress Update

After totaling up the remaining animation and completed animations. There is currently...

• 51 Animated Images
• 20 Unfinished Animated Images
• 6 Remaining Scenes (multiple animated images make up a scene)
• 8 Completed Scenes

Each image is usually a 6-10 second animated loop.

I'm considering posting a whole completed scene as a teaser trailer if people are interested. A scene has dialog, character side images (with matching facial expressions), and animated images. They are usually 2-10 images long.

Example:

 

[OffWhite Studios]

Progress Update


It's been a busy few weeks, and I'm currently deep into the animation process. It's taking longer than I initially expected, but I believe it's for a good reason.

At first, my plan was to animate only key images, expecting to complete around 30 animations. However, as my skills have improved over the past few weeks, I've found myself revisiting and enhancing all the animations. On top of that, I've shifted towards animating almost every image in my scenes rather than just select ones.

Right now, I have around 120 different animated images, and I anticipate creating at least 80 more. Given the current pace, I believe I should be able to release Chapter One of the game in late April, though the final timeline will depend on how long the polishing phase takes.

While I'm not happy about missing my original planned release date, I believe the delay will be worth it. The animations will be significantly more detailed and polished, and at the end of the day, I know that the scenes are a key part of what people play these games for.

I'll share another update in a few weeks—stay tuned!

 

[OffWhite Studios]

Progress Update

Hey everyone! Just wanted to share a quick update—it’s been about two months since my last post. The past few months have been pretty busy on my end, but I want to reassure you that progress is still being made, and I’m still aiming to launch the game in the next few months.


As of now, my total animation count is up to 173 (so about 50 in 2 months), and I’ve recently implemented a few new combat systems to support different play styles and make the gameplay feel more dynamic. I’ve also squashed a few bugs along the way.


At this point, I think I’ve got just a handful of scenes left—maybe around 20 images to go—before moving into polish and prep for the initial release. I’ve also been revisiting some older animations that I wasn’t fully satisfied with and giving them a fresh touch-up. You can see an example of that below.


New / Old

 

[OffWhite Studios]

Hey everyone,

Just wanted to drop a quick update on the game. I know it’s been delayed way longer than I originally planned, but I think that’s because my expectations for the quality have really changed since I started.

When I first began this project, it was just for fun—I had zero experience with things like Ren’Py, Stable Diffusion, After Effects, or really any of the tools I’m using now. It was a blast learning and building something for the first time, but as both the tech and my own skills improved, I kept finding myself going back and polishing older scenes and gameplay. Stuff I used to be super proud of now feels off or low quality, and I’ve ended up redoing things two or three times in some places.

I’m 100% committed to finishing this and getting it out, and honestly the game is basically done. I just have to accept that not everything will be perfect at launch—and that’s okay. That said, I really think the final quality is going to be way better than anything I imagined I could pull off when I started this journey.


Right now, I’m going scene by scene, updating any art that doesn’t match the visual tone I’m aiming for. AI art can be a bit wild with consistency, so that’s been a big challenge. But I think it’s coming together nicely into something that feels cohesive.

At this rate, I’m feeling good about a release sometime by the end of the year. Still need to figure out how I’ll package and launch it, but I’ll get there.

Thanks so much to anyone who’s still following the project—I really appreciate the patience. I promise it’s still coming!

 

[Kodacho]

Great work, the screen shots look clean, the AI art does look nice and consistent, and the premise is interesting. I love 2DCG games set in a high fantasy universe, so I cannot wait to see what your first release will look like.

Seems as if there will be more than enough content! If you are looking for a playtester, please do not hesitate to contact me.

Good luck on the development of the game!

 

[OffWhite Studios]

Great work, the screen shots look clean, the AI art does look nice and consistent, and the premise is interesting. I love 2DCG games set in a high fantasy universe, so I cannot wait to see what your first release will look like.

Seems as if there will be more than enough content! If you are looking for a playtester, please do not hesitate to contact me.

Good luck on the development of the game!

Thank you! I really appreciate that. I think I'm still a few weeks/months from a playtest, but I think it could be helpful. I'll let you know.

 

[umbronium]

Wow, I can tell a lot of effort and polish went into this, even if it is AI artwork. I'd love to know more about how you carefully craft these images in another post if you're willing to share.

 

[mibc9394]

I'm not attracted by the piggy girl but I really like the design of her!

 

[OffWhite Studios]

Wow, I can tell a lot of effort and polish went into this, even if it is AI artwork. I'd love to know more about how you carefully craft these images in another post if you're willing to share.

Thank you so much for the kind words — I really appreciate it! I’d be happy to share a brief overview of my process.


When I first started this game about two years ago, I was using SD 1.5 on a pretty outdated graphics card, so the way I worked back then was very different compared to now. As both the tech and my own skillset evolved, so did my workflow. These days, especially with the release of the Pony model, I feel like I’ve finally reached a stable process that will let me finish the game in the quality and style I’ve been aiming for.


I’ll admit, I wasn’t a huge fan of the “AI look” in games at first either. They often looked beautifully detailed but somehow lifeless. So one of my biggest goals has been to avoid that telltale AI feel as much as possible. I try to achieve that by using more dynamic poses, varied camera angles, expressive faces, and a flatter, block-color animated look that’s more forgiving of AI’s typical weirdness. Of course, I completely respect people who value the personal touch of hand-drawn art — that’s something special. But for me, realistically, I didn’t have the time or talent to create all the artwork needed for a game of this scale by hand. If I wanted to make this project happen at all, I needed help — and AI gave me a way to do it.


That said, if the game ever gains traction, I’d love to reinvest in it — maybe even hire voice actors or animators down the line. Ideally, this could become a project that contributes to the creative community rather than taking away from it.


As for the process itself:


It all starts with choosing one primary model — I’ve found that’s key to getting a consistent style across the game. I try to pick a model with a very distinctive aesthetic that can output similar results without needing a ton of prompt engineering. Then, I start from a super simple character description — something like “blonde, short hair, blue eyes, blue dress” plus an emotion or pose. I’ll generate a big batch and pick whichever one best matches my vision (or previously generated versions of that character). I’ve learned not to get too caught up in super specific prompts — things like a flower pattern on a dress just don’t translate well across multiple images, especially if the pose or angle changes.


Once I’ve got a base image I like, I’ll inpaint the face to get the right expression — this is usually the most time-consuming part since it’s where I focus most on avoiding that “AI face” look.


After that, I toss everything into Photoshop and use the color match feature to align the tones with previous images. This step is a recent addition to my workflow, but it’s helped a lot in keeping everything visually cohesive — especially when I’m reworking older assets. Actually, a lot of images I posted earlier have already been scrapped or redone. I’ll have to share some new examples soon.


Finally, I animate in After Effects using the puppet tool — stretching, resizing, and adding movement to the characters.


And that’s basically it! Still far from perfect, but I’m trying to resist the perfectionism trap and focus on finishing. I definitely fell into the feature/content bloat trap in the beginning, but now my goal is just to get this thing done!


The good news is, future updates should be way faster now that I’ve got a solid process, better tools, and more experience under my belt.

 

[mibc9394]

Thank you so much for the kind words — I really appreciate it! I’d be happy to share a brief overview of my process.


When I first started this game about two years ago, I was using SD 1.5 on a pretty outdated graphics card, so the way I worked back then was very different compared to now. As both the tech and my own skillset evolved, so did my workflow. These days, especially with the release of the Pony model, I feel like I’ve finally reached a stable process that will let me finish the game in the quality and style I’ve been aiming for.


I’ll admit, I wasn’t a huge fan of the “AI look” in games at first either. They often looked beautifully detailed but somehow lifeless. So one of my biggest goals has been to avoid that telltale AI feel as much as possible. I try to achieve that by using more dynamic poses, varied camera angles, expressive faces, and a flatter, block-color animated look that’s more forgiving of AI’s typical weirdness. Of course, I completely respect people who value the personal touch of hand-drawn art — that’s something special. But for me, realistically, I didn’t have the time or talent to create all the artwork needed for a game of this scale by hand. If I wanted to make this project happen at all, I needed help — and AI gave me a way to do it.


That said, if the game ever gains traction, I’d love to reinvest in it — maybe even hire voice actors or animators down the line. Ideally, this could become a project that contributes to the creative community rather than taking away from it.


As for the process itself:


It all starts with choosing one primary model — I’ve found that’s key to getting a consistent style across the game. I try to pick a model with a very distinctive aesthetic that can output similar results without needing a ton of prompt engineering. Then, I start from a super simple character description — something like “blonde, short hair, blue eyes, blue dress” plus an emotion or pose. I’ll generate a big batch and pick whichever one best matches my vision (or previously generated versions of that character). I’ve learned not to get too caught up in super specific prompts — things like a flower pattern on a dress just don’t translate well across multiple images, especially if the pose or angle changes.


Once I’ve got a base image I like, I’ll inpaint the face to get the right expression — this is usually the most time-consuming part since it’s where I focus most on avoiding that “AI face” look.


After that, I toss everything into Photoshop and use the color match feature to align the tones with previous images. This step is a recent addition to my workflow, but it’s helped a lot in keeping everything visually cohesive — especially when I’m reworking older assets. Actually, a lot of images I posted earlier have already been scrapped or redone. I’ll have to share some new examples soon.


Finally, I animate in After Effects using the puppet tool — stretching, resizing, and adding movement to the characters.


And that’s basically it! Still far from perfect, but I’m trying to resist the perfectionism trap and focus on finishing. I definitely fell into the feature/content bloat trap in the beginning, but now my goal is just to get this thing done!


The good news is, future updates should be way faster now that I’ve got a solid process, better tools, and more experience under my belt.

I found that in order to remove the AI look, it is better to mix different models together with different weighting, until you get the look you are looking for. And after that, it is needed to push the poses even further to breath life into the painting. AI generated results tend to be a few specific poses and angles, so if there is no further paint over, they will have a strong AI look.

 

[OffWhite Studios]

I found that in order to remove the AI look, it is better to mix different models together with different weighting, until you get the look you are looking for. And after that, it is needed to push the poses even further to breath life into the painting. AI generated results tend to be a few specific poses and angles, so if there is no further paint over, they will have a strong AI look.

Absolutely agree — and I think I probably glossed over just how much in-painting I do in my last reply . Especially when working with multiple characters or more complex scenes like fight sequences or dancing, there’s just no way to get a usable image straight out of the box. A single character facing forward? Sure, easy. But anything dynamic takes a lot of detailed cleanup and tweaking.


I also rely heavily on the sketch tool and similar features to help force more dynamic poses into place. Like you said, AI tends to default to the same few safe angles, so without intervention, it really starts to look repetitive or stiff.


As for model mixing — I’ve experimented with it, but haven’t had much luck. I often found that blending models messed with the overall art style too much. Like, I’d get a body with a more Western anime vibe but a face in a completely different (often more Japanese) style, and it just didn’t sit right. These days, I mostly stick to one base model and rely on LORAs for specific enhancements, especially to push facial expressiveness.


Then in After Effects, I can add micro-movements like eye shifts, winces, slight blinks — stuff that helps breathe a bit more life into the stills. It’s definitely a lot of manual work, but honestly I enjoy the process!


Thanks again for the input — always fun to hear how others are approaching the same challenge.