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

[Dotonbori Technologies]

Changing topics here to ask a question on AI Art:


Over the last few years, I’ve seen a wide range of opinions on AI art being used across different mediums. Most people seem to be either opposed or neutral, with very few expressing excitement about consuming AI-generated art. Some even go as far as to call it offensive, soulless, or “not real art.”


I can understand the anti–AI art perspective and even share some of the same feelings. There’s no denying that some AI art is low-effort and feels soulless.


However, as someone who has been building a game with AI art for around two years, I can’t agree that simply using AI art automatically invalidates the “art” in my work. Even if you write off the art assets, a game involves so much more—programming, story, design, and even animation (in my case, done manually in After Effects). And even when using AI, there’s still a lot of artistic direction involved—world building, scene composition, storyboarding, and guiding the tools toward a vision. Surely, there are multiple places where creativity can shine through?


Anyway, all of this brings me to my main question: What do you all think about AI art in games?

• Would you immediately write off a game if it used AI-generated art?
• Why or why not?

(As a side note - I know a lot of the old art I shared still included some bad AI (fragment, bad fingers, etc)- much been fixed in and will share updated versions soon)

For me, most AI art looks the same. I don't know why, but I like to play games where the art is drawn by an artist - I admire their craftmanship. I am not opposed to using AI for your art assets (it's a LOT cheaper than commissioning an artist), but it's just not for me.

Blindly hating everything made with AI is stupid, but I understand people who have a preference for non-generated art. I wish you the best of luck with your game!

 

[FROMSPACE]

Changing topics here to ask a question on AI Art:


Over the last few years, I’ve seen a wide range of opinions on AI art being used across different mediums. Most people seem to be either opposed or neutral, with very few expressing excitement about consuming AI-generated art. Some even go as far as to call it offensive, soulless, or “not real art.”


I can understand the anti–AI art perspective and even share some of the same feelings. There’s no denying that some AI art is low-effort and feels soulless.


However, as someone who has been building a game with AI art for around two years, I can’t agree that simply using AI art automatically invalidates the “art” in my work. Even if you write off the art assets, a game involves so much more—programming, story, design, and even animation (in my case, done manually in After Effects). And even when using AI, there’s still a lot of artistic direction involved—world building, scene composition, storyboarding, and guiding the tools toward a vision. Surely, there are multiple places where creativity can shine through?


Anyway, all of this brings me to my main question: What do you all think about AI art in games?

• Would you immediately write off a game if it used AI-generated art?
• Why or why not?

(As a side note - I know a lot of the old art I shared still included some bad AI (fragment, bad fingers, etc)- much been fixed in and will share updated versions soon)

This AI Hate trend is growing now because AI is more accessible, and soon, everyone will use it in some form. It’s inevitable that some people will still prefer doing things manually and avoid AI altogether. However, what truly matters isn’t the tools used but the quality of the final product and whether the end user or player enjoys it.

As you’ve seen, there are many ways to use this technology, and each approach impacts the end result differently. For example, players enjoy AVNs that use the same assets, such as Daz models, because ultimately, quality isn’t just about the assets themselves, it’s about the overall experience.

By the way, congratulations on your game! It looks very deep and well-made. I’ve noticed you’re using After Effects for the animations, and they’re really impressive. I’ll look into learning it for my own project. Thanks!

 

[OffWhite Studios]

For me, most AI art looks the same. I don't know why, but I like to play games where the art is drawn by an artist - I admire their craftmanship. I am not opposed to using AI for your art assets (it's a LOT cheaper than commissioning an artist), but it's just not for me.

Blindly hating everything made with AI is stupid, but I understand people who have a preference for non-generated art. I wish you the best of luck with your game!

Thanks and I can respect that opinion. I think I feel that same way, in that I enjoy art made by artist more than AI (most of the time). For me, and others I imagine, it lowered the barrier to entry and made it possible to jump into game making for the first time. That’s why when I see a game with AI assists don’t automatically think “this is replacing an artist” and rather “This game wouldn’t exist without AI”. I now have more options and there is more opportunity in the space.

I may never hire an artist for my game, but I could imagine (if it’s successful), hiring a VA, animator, etc. If not for AI I may not have had the chance to make this game and commission other down the line, etc.

 

[OffWhite Studios]

This AI Hate trend is growing now because AI is more accessible, and soon, everyone will use it in some form. It’s inevitable that some people will still prefer doing things manually and avoid AI altogether. However, what truly matters isn’t the tools used but the quality of the final product and whether the end user or player enjoys it.

As you’ve seen, there are many ways to use this technology, and each approach impacts the end result differently. For example, players enjoy AVNs that use the same assets, such as Daz models, because ultimately, quality isn’t just about the assets themselves, it’s about the overall experience.

By the way, congratulations on your game! It looks very deep and well-made. I’ve noticed you’re using After Effects for the animations, and they’re really impressive. I’ll look into learning it for my own project. Thanks!

Thank you for your comment!

As for AE, I find it perfect for making short animated loops. I basically only need to learn the puppet pin tool and some commands like “easy ease” to smooth out motion. If you’re looking to make what’s basically an animated GIF, I think it’s a great tool with a pretty easy learning curve (for that basic use case). For me, I just chop up my still image in photoshop, segment them into parts, paint in the missing background, and then use puppet pins to create the animated loop. Quick and easy and really elevates the look above a still image I think.

 

[OffWhite Studios]

Hey everyone,


In my last few posts I’ve been talking a lot about the work I’ve been doing to update and polish different scenes, and I finally got around to sharing the updated art here. I went ahead and replaced a lot of the images from my original post with new ones, and I’ve also added some gifs. Please check it out and let me know what you think!


As for my progress, I’m actually making good time for once — I’m now about 50% of the way through my first polish pass, and I think I can get the other 50% done within the next few weeks. After that, the second and third playthrough/polish rounds should go fairly quickly, and it’s looking like I can probably release the game this year.


I’m also in a bit of a dilemma regarding some of my older animations. Some of them can be updated and upscaled pretty easily, while others would take quite a bit of time and effort—possibly requiring me to redo them entirely.


Here’s where I’d love your input: if you were releasing a game, would you want every single animation at top consistent quality? Or, if some were maybe 70% of your best work (slightly lower resolution, a little messier), would you still keep them for the sake of content?


Personally, I can really see the line between my earlier animations—made with older hardware and less skill—and where I’m at now, with better tools and more experience. It does bother me a bit. At the same time, those older animations used to meet my standard back then, so maybe players will still find them fine today?


I just want to get this game out, but I also want it to look consistent. What would you do in this situation?


Thanks as always for the feedback!

 

[FROMSPACE]

Hey everyone,


In my last few posts I’ve been talking a lot about the work I’ve been doing to update and polish different scenes, and I finally got around to sharing the updated art here. I went ahead and replaced a lot of the images from my original post with new ones, and I’ve also added some gifs. Please check it out and let me know what you think!


As for my progress, I’m actually making good time for once — I’m now about 50% of the way through my first polish pass, and I think I can get the other 50% done within the next few weeks. After that, the second and third playthrough/polish rounds should go fairly quickly, and it’s looking like I can probably release the game this year.


I’m also in a bit of a dilemma regarding some of my older animations. Some of them can be updated and upscaled pretty easily, while others would take quite a bit of time and effort—possibly requiring me to redo them entirely.


Here’s where I’d love your input: if you were releasing a game, would you want every single animation at top consistent quality? Or, if some were maybe 70% of your best work (slightly lower resolution, a little messier), would you still keep them for the sake of content?


Personally, I can really see the line between my earlier animations—made with older hardware and less skill—and where I’m at now, with better tools and more experience. It does bother me a bit. At the same time, those older animations used to meet my standard back then, so maybe players will still find them fine today?


I just want to get this game out, but I also want it to look consistent. What would you do in this situation?


Thanks as always for the feedback!

They look great! get out there, listen to feedback, and take action based on it. You already have animations (and they're great from my perspective) so just release the game and start building your community.

 

[OffWhite Studios]

Progress Update


Hey everyone,


After what feels like forever, I’m finally getting close to releasing the first version of my game. I’m aiming to put it out before December 15th. Which is, uh… about eight months later than I originally planned But oh well, progress is progress!


This is actually my first time ever launching a game, so I’m really hoping to get some advice from more experienced devs here. I’ll be hosting the game on F95, and at the moment I’m not planning on Steam or Itch.io. Mostly because I don’t think the game would get approved there, and also because I’m not really doing this for money.


I will set up a Patreon (or other donation option) for anyone who wants to support the project, and that would definitely be appreciated, but monetization isn’t the main goal for me — I just want to make something fun and get it out into the world.


Since I basically have no idea what to expect, I’m hoping to hear any advice you’re willing to share:

• Best practices for launching a first version
• Where you recommend hosting / file sharing
• How to advertise without being annoying
• What kinds of screenshots or previews work best
• Whether trailers help engagement
• How to handle bug reports and feedback
• Anything you wish you had known before your first release

I’m going into this totally blind, so literally any tips (even obvious-sounding ones) would be super helpful. Thanks in advance, and I’m really excited to finally share the project soon!

 

[melantha]

i wish you good luck in the upcoming game! looking forward to it, love the characters!
i admire the perseverance of getting this far, finally being able to show after so long.
many would have given up, but you're pulling through

do you have a discord server by any chance?
that would be a good way to have a community that can provide feedback, ideas all in one place.

also just a quick question, those skill icons, they are reminiscent of warcraft 3 skill icons, is it inspired by them somehow? i know many modern games have this style now adays, but what caught my eye was the "frame" around it.

 

[OffWhite Studios]

i wish you good luck in the upcoming game! looking forward to it, love the characters!
i admire the perseverance of getting this far, finally being able to show after so long.
many would have given up, but you're pulling through

do you have a discord server by any chance?
that would be a good way to have a community that can provide feedback, ideas all in one place.

also just a quick question, those skill icons, they are reminiscent of warcraft 3 skill icons, is it inspired by them somehow? i know many modern games have this style now adays, but what caught my eye was the "frame" around it.

Thank you so much for the kind words — it really means a lot. This project has been a long journey, and hearing that people are excited for the characters and the world honestly gives me a ton of motivation to keep going. I’m really happy you’re looking forward to the upcoming release!

As for a Discord server, I don’t have one yet. I’m still figuring out how I want to handle community spaces, but it’s definitely something I’m considering. I agree it would be a great place for feedback and ideas. I see this as more of a passion project than a business, so a bit hesitant to commit sometimes to community stuff (since I don’t want to ignore people) - but I see the value.

And good eye on the skill icons! They’re definitely inspired by MMO skill icons!

 

[Mike145]

TLDR: Looks good.

Not sure how much time the animations are taking you in After Effects, but if you're using a specific sprite/artwork often, I think it may be worth rigging them in Blender or Spine 2D and making animations from there. Will give you more options and might be faster in the long run. It takes a new skill to learn, so I wouldn't try it until you're finished with at least the demo. It's more of a consideration for later.

You may also want to test and see if you can leverage bigger and better models like Google's Nano Banana. It won't help you with anything NSFW, but you can easily use the image for inpainting later and/or manually correcting the image. For example, to test things out, I generated a highly detailed stylized artwork and asked it to make a sprite out of it. Similar style, but less detailed. I gotta admit, maybe I was lucky, but it did it exactly how I imagined it. Getting more sprites with certain angles was a huge challenge tho, but I think if you've used any txt2img model, you can imagine. But it gave me a surprisingly good sprite in 10 seconds that I could then rig and make animations for. That's what impressed me.

As for AI art, it's good when it's consistent and doesn't have the typical AI lighting and other AI hurdles like deformed anatomy, but actually resembles an artist's style (not a specific one, what I mean is just a human artist). I'm pretty sure most of the hate comes from that cause there's a lot of low-effort AI junk around here. What I also think will happen in the future is that, eventually the artists who are able to leverage AI tools to speed up and perfect their creation workflow will thrive, and the ones stuck in the past will just get left behind. Cause even if you're not a real artist, with AI, if you can draw and correct a few things manually (not even the full part, but just a sketch for inpainting), that's a big advantage.

Good luck with the game.

 

[OffWhite Studios]

TLDR: Looks good.

Not sure how much time the animations are taking you in After Effects, but if you're using a specific sprite/artwork often, I think it may be worth rigging them in Blender or Spine 2D and making animations from there. Will give you more options and might be faster in the long run. It takes a new skill to learn, so I wouldn't try it until you're finished with at least the demo. It's more of a consideration for later.

You may also want to test and see if you can leverage bigger and better models like Google's Nano Banana. It won't help you with anything NSFW, but you can easily use the image for inpainting later and/or manually correcting the image. For example, to test things out, I generated a highly detailed stylized artwork and asked it to make a sprite out of it. Similar style, but less detailed. I gotta admit, maybe I was lucky, but it did it exactly how I imagined it. Getting more sprites with certain angles was a huge challenge tho, but I think if you've used any txt2img model, you can imagine. But it gave me a surprisingly good sprite in 10 seconds that I could then rig and make animations for. That's what impressed me.

As for AI art, it's good when it's consistent and doesn't have the typical AI lighting and other AI hurdles like deformed anatomy, but actually resembles an artist's style (not a specific one, what I mean is just a human artist). I'm pretty sure most of the hate comes from that cause there's a lot of low-effort AI junk around here. What I also think will happen in the future is that, eventually the artists who are able to leverage AI tools to speed up and perfect their creation workflow will thrive, and the ones stuck in the past will just get left behind. Cause even if you're not a real artist, with AI, if you can draw and correct a few things manually (not even the full part, but just a sketch for inpainting), that's a big advantage.

Good luck with the game.

Thanks a lot for the feedback and support — really appreciate it.

My current workflow is actually super basic: I generate an image, chop it up into parts in Photoshop, and then animate everything with puppet pins. It’s fast, which is why I stick with it, but the downside is the limited range of motion since the images aren’t actually 3D. If I could somehow keep the 2D look but gain 3D-style functionality, that’s definitely something I’d love to explore down the line.

For consistency, I use a specific model with an applied art style. It does a pretty good job keeping characters recognizable, but it’s not always perfect. And honestly, when I started this project I had zero skills — you can probably tell which scenes were made first and which ones came much later. I could redo the older stuff, but perfect is the enemy of done, so I’d rather release the demo, hear what the community points out, and improve the rough spots afterward.

And yeah, I completely agree with you on AI art. Most of my time actually goes into refining the faces — the expressions are what make characters feel human and handmade. At this point I’ve got thousands of image assets… I genuinely don’t know how solo devs made full games before AI. Even if I hired an artist, drawing this much material would take forever.

Thanks again for the advice and encouragement — means a lot

 

[OffWhite Studios]

TLDR: Looks good.

Not sure how much time the animations are taking you in After Effects, but if you're using a specific sprite/artwork often, I think it may be worth rigging them in Blender or Spine 2D and making animations from there. Will give you more options and might be faster in the long run. It takes a new skill to learn, so I wouldn't try it until you're finished with at least the demo. It's more of a consideration for later.

You may also want to test and see if you can leverage bigger and better models like Google's Nano Banana. It won't help you with anything NSFW, but you can easily use the image for inpainting later and/or manually correcting the image. For example, to test things out, I generated a highly detailed stylized artwork and asked it to make a sprite out of it. Similar style, but less detailed. I gotta admit, maybe I was lucky, but it did it exactly how I imagined it. Getting more sprites with certain angles was a huge challenge tho, but I think if you've used any txt2img model, you can imagine. But it gave me a surprisingly good sprite in 10 seconds that I could then rig and make animations for. That's what impressed me.

As for AI art, it's good when it's consistent and doesn't have the typical AI lighting and other AI hurdles like deformed anatomy, but actually resembles an artist's style (not a specific one, what I mean is just a human artist). I'm pretty sure most of the hate comes from that cause there's a lot of low-effort AI junk around here. What I also think will happen in the future is that, eventually the artists who are able to leverage AI tools to speed up and perfect their creation workflow will thrive, and the ones stuck in the past will just get left behind. Cause even if you're not a real artist, with AI, if you can draw and correct a few things manually (not even the full part, but just a sketch for inpainting), that's a big advantage.

Good luck with the game.

Follow up to say I tried Nano Banana and it’s very good. I think it can help me with more complex poses and scenes my current Pony based model does not like. Thanks for the tip! Will use this one

 

[Mike145]

Thanks a lot for the feedback and support — really appreciate it.

My current workflow is actually super basic: I generate an image, chop it up into parts in Photoshop, and then animate everything with puppet pins. It’s fast, which is why I stick with it, but the downside is the limited range of motion since the images aren’t actually 3D. If I could somehow keep the 2D look but gain 3D-style functionality, that’s definitely something I’d love to explore down the line.

I'm not exactly sure what you mean by "2D look but gain 3D-style functionality," but you can do something like this in Blender -

You must be registered to see the links

But like I said, it's more of a consideration for later, and I would probably only do it for some special moments or sprites that are reused often.

I'd also consider paying someone to proofread the game's script if you're not a native speaker, and especially if you've used AI for it.