Ren'Py Unreal Engine UE or RenPy Game... Which is Most Fun to Play?

[FredFappy]

Hey Peeps!

First off, you'll have to trust me that I can make a simple game in either engine. I have lots of resources and some decent skill (20+ years in AAA). I'm retired and bored and I love the thought of making something naughty.


Now.. I have extensive knowledge of Unreal Engine and could make a dialogue based game quite easily with some simple tasks (find the key to get in the door kinda stuff). I think that's within scope. I can make it pretty but it will still be 3d cg stuff.

OR...

I could make something in RenPy (easy to dev with) and move my 3d art toward photo-real using AI... which I've been playing with lately and although it's hard (hard to get consistent) it's pretty fun and the results are pretty amazing imo. The only thing is, there won't be much animation... probably just some of the more fun sex stuff. Not too different from what I see from other RenPy games.. not that I'm complaining. I love those games.

The big question (as they are both equally as hard and fun to do), which one do you think players will like better? I don't see many Unreal games here.... probably because it takes a decent gaming PC to run that stuff.

This would be around the level of quality I could produce in Unreal... (again, players would need a reasonable gaming PC):


BUT...

I can do these cool CG to AI images and get some pretty fun looking images also. They are surprisingly hard to accomplish so I'm thinking I could do a few of these a day along with some animation in the same vein. Here are my tests for getting CG (Daz and Blender stuff) to AI...


It's pretty hard to accomplish though and mostly, I get a kind of in-between look... like this:


I'm leaning toward just making something that I really know how to do (the Unreal game) but damn, when the AI works, it's kinda amazing!

I'm torn so I decided to just post here and see what you guys thought. It's not like a game in UE will be "easy" but at least it's a known for me... or I could make a RenPy game with shit like this (well if it will behave)...
View attachment CuteHooker.mp4

 

[Winterfire]

You'll like Ren'Py better and so will the players. You're retired, enjoy little programming enough to edit the custom stuff and add a bunch of new system, and focus on writing/graphics.

I'm a little confused to your connection Renpy -> Lack of animation. You can have just as much animation in Ren'Py.
There's Live2D support, and you can even have your own custom shaders (wavy effects, etc.), or simply unite the power of Unreal with Ren'Py... Create a scene in Unreal, record it, then use it on Ren'Py as a webm.
Of course that'd be time consuming, that's why people mostly focus on having animations for sex scenes only.

 

[FredFappy]

I'm a little confused to your connection Renpy -> Lack of animation. You can have just as much animation in Ren'Py.

I meant animating the AI stuff.. I have to do the naughty stuff locally and I'm not sure how hard that is.

Hey and thanks so much for your advice. The bit I've played with RenPy, it was super fun! I'll have to see how long it takes me to make the images. What is the average for a Daz3D guy working on a project? Like I wonder how many images they crank out a week?

 

[Noah Neim]

I mean you'd be better off prodcing models yourself. Views on ai are split and the ai cg tag is likely one on many peoples blacklist.

I agree with Winter though, you're retired, considering how this is just a project for fun, you should prioritize your own enjoyment. Renpy is easier and so is using ai. Though if you do opt for that route you're gonna have to be a good writer.

Good luck man.

 

[defnotalt]

Honestly, if you're set on a dialogue-based game with 3DCG style, RenPy will save both your time and the player's CPU/GPU resources. I don't think UE would be a good choice for a VN, I've seen it done a few times, but it doesn't add much substance when the main strength of the game is the story and dialogue. Obviously just my opinion though.

Regarding AI, I think it's fine if you have a solid base input image and are willing to spend some effort into fine-tuning it. Never really thought about giving Daz model CGs to AI, but your examples look promising. Just make sure the AI stays in its lane, and doesn't transform the input into generic and easily identifiable AI imagery.

 

[FredFappy]

I mean you'd be better off prodcing models yourself. Views on ai are split and the ai cg tag is likely one on many peoples blacklist.

I agree with Winter though, you're retired, considering how this is just a project for fun, you should prioritize your own enjoyment. Renpy is easier and so is using ai. Though if you do opt for that route you're gonna have to be a good writer.

Good luck man.

Thanks!

Yeah.. I get the AI thing but in my case, I am rendering the scene and then using AI as a "filter" of sorts. It's not like I'm just doing a txt2img thing. It's more AI enhancement really... but I could do straight CG.. hell, that would be a breeze.

I was just working on the second image some more and getting better results now (more real and less cg looking)...

Top is mine.. I made the char in Daz3D, posed and rendered her in Blender and the bottom image is the AI enhanced version. It's still cheating.. but not too too much.

 

[kinrean]

the first Q you should ask is what you want to forces or sell , if you want to forces or sell the story then you should go for renpy. but if you want to forces on game system or action type of thing, renpy can't do any run time thing.

and it seem that you just want to look good, so it seem renpy /vn is your best choice.

 

[TheExordick]

Thanks!

Yeah.. I get the AI thing but in my case, I am rendering the scene and then using AI as a "filter" of sorts. It's not like I'm just doing a txt2img thing. It's more AI enhancement really... but I could do straight CG.. hell, that would be a breeze.

I was just working on the second image some more and getting better results now (more real and less cg looking)...

Top is mine.. I made the char in Daz3D, posed and rendered her in Blender and the bottom image is the AI enhanced version. It's still cheating.. but not too too much.
View attachment 4730121

If i may offer my 2cents:

- on the engine choice, while i'm a big fan of UE, it makes little sense to use it for a dialogue/story based game where you have still images or pre-rendered animations. Let's use the right tool for the right job --> Renpy

- on the images: you can get the same results as your "AI-fitered/enhaced" image in blender or Daz as well. Your base render is pretty and it is a good starting point but it also leaves a lot to be desired. The image is washed out, it lacks contrast and depths which can be fixed with curated lighting; the materials feel plasticky/rubbery/silicon-like so improve them; facial expression feels a bit empty (also in ai image) and eyes are "void", choose nice expression from Daz not the cheap ones and tweak them with additional morphs (use references); color composition is off, the hair, wallpaper, bookcover and skin are basically the same color in different shades and the only other elements are neutral colors, black and white. Your blue light from the side is not helping either, side lights should be very very subtle, take a look at color theory and the different types of contrast (i suggest the book The Art of Color from Itten).

i feel the ai filter is helping you fixing these kind of problems instead of enhancing the realism of the person.

Don't get me wrong, these things are very difficult to achieve in renders especially when you have soft surfaces interacting (bed and body, clothes and body especially tight fits, body and body, penetration and so on) and in these instances AI can help, but keeping your intentional look can be very difficult.
I'm not a big fan of AI, not because of the look of it, but as a tool it is too unpredictable, so you are better off spending your time honing your skill in something you can predict the result of (like 3d rendering and drawing)

I hope i did not sound harsh, i think you are off to a very good start.

 

[FredFappy]

If i may offer my 2cents:

- on the engine choice, while i'm a big fan of UE, it makes little sense to use it for a dialogue/story based game where you have still images or pre-rendered animations. Let's use the right tool for the right job --> Renpy

- on the images: you can get the same results as your "AI-fitered/enhaced" image in blender or Daz as well. Your base render is pretty and it is a good starting point but it also leaves a lot to be desired. The image is washed out, it lacks contrast and depths which can be fixed with curated lighting; the materials feel plasticky/rubbery/silicon-like so improve them; facial expression feels a bit empty (also in ai image) and eyes are "void", choose nice expression from Daz not the cheap ones and tweak them with additional morphs (use references); color composition is off, the hair, wallpaper, bookcover and skin are basically the same color in different shades and the only other elements are neutral colors, black and white. Your blue light from the side is not helping either, side lights should be very very subtle, take a look at color theory and the different types of contrast (i suggest the book The Art of Color from Itten).

i feel the ai filter is helping you fixing these kind of problems instead of enhancing the realism of the person.

Don't get me wrong, these things are very difficult to achieve in renders especially when you have soft surfaces interacting (bed and body, clothes and body especially tight fits, body and body, penetration and so on) and in these instances AI can help, but keeping your intentional look can be very difficult.
I'm not a big fan of AI, not because of the look of it, but as a tool it is too unpredictable, so you are better off spending your time honing your skill in something you can predict the result of (like 3d rendering and drawing)

I hope i did not sound harsh, i think you are off to a very good start.

Not harsh at all.. remember, this used to be my job for a living so I got the critiques all the time. Thanks for taking the time.. honestly!

Yeah, that initial shot leaves a lot to be desired. I'm not as skilled as I used to be but I could have definitely spent more time with it. Knowing that I was going run it through AI made me lazy with it I think. The materials are just ass on that character as I did very little to her in Blender as she was supposed to go into UE. Here's a better example of where I spent time setting up the materials in Blender (not happy with her teeth but everything else looks pretty good). This is also my facial mocap capabilities...
View attachment MA_FaceTest01_0001-0754.mp4


But as far as UE, I can do a lot more than a RenPy type story. I could make something like that Demon game.. can't remember the name. The game where you fought guys and took the elf girl back to your layer and banged her. It was kinda a prototype but I honestly thought about making a comedic take on that game because the scope is fairly doable. "The Son of Demonater (or whatever he was called)". Mind you I'm just an artist but I know enough Blueprints and between great UE plugins and ChatGPT now, I can really get things done.

I need to show off more UE stuff so you get the idea.. lemme whip up a prototype in the next few days here.

 

[FredFappy]

I could also do a shit ton of animation like the redhead up there in RenPy instead of just a bunch of stills. I think that's what you meant though about using the 3d instead of the AI. I do like the way Red there turned out. She's my intern btw. Here's a shot of the studio...

 

[FredFappy]

Oh.. and here's my Production Supervisor, Mr. Gimpstein. He's quiet but he gets things done!
View attachment MrGimpy_Test01_0001-0547.mp4

 

[kinrean]

the problem of non vn game is the game system, or another way of saying is choosing what type of game to make. and as a artist you should know, too many choice is a bad thing. as you will have to choose many type of game play, like 3rd person, first person, topdown, side scoller, and other, and that haven't talk about the game system.

so i will always go down to the basic, what game you love to play the most, and then change base on the game you love.

a bit of into of my self, i a 2d/3d artist, and use unreal blueprint.

 

[FredFappy]

I'm leaning toward RenPy as I think it will make the choices easier as far as design. I mainly just want to tell a simple story and entertain someone. I also think I'm going to dump the AI idea even though it looks cool, it's just an extra layer of hassle. I might explore it again later but for now I'll stick with 3d animation stuff.

I've been working on this girl's materials today and I have her pretty well where I want her now..

 

[FredFappy]

Damn the AI stuff is cool though...
View attachment Samie_Happy_in_Bed.mp4

 

[unocrus]

Damn the AI stuff is cool though...
View attachment 4733316

The thing with content generated by AI is the more that you look at it, the worse it gets. This character's feet, for example. You're better off developing other skills, in my opinion.

As for your UE vs RenRy question, it's really a matter of do you want gameplay or not? If you want gameplay, how central is it to the story you're trying to tell? There are plenty of games that actually transitioned from RPGM to RenPy because the gameplay only got in the way of the story. UE is like that, but it's even moreso on you to develop interesting gameplay systems.

Most UE games I've seen have neglected their gameplay, ending up as sophisticated sex sandboxes and/or animation viewers.

 

[FredFappy]

unocrus I agree.. so it's decided then. I go RenPy with CG. I'm on it! I'll crack the whip on my 'team' and get something going so we can get a proper dev thread going.