Unity Looking for guidance on 3D graphics

[Laiga]

I am looking for guidance on how to approach the graphical side of a project I've been wanting to work on.

About me
I am a recently graduated game dev with some experience in Unity, coding, and design.
I've made a few games of small to medium scope. (15-30 min experiences)
I specialize in systems and monetization.
Studying monetization absolutely violated my innocent view on the games industry.

What I need help with
Finding a model to start with.
- I am looking for a base model that's similar in style to Koikatsu.

How to approach soft-body (jiggle) physics in Unity.
- Any starting points for this would be very helpful.
- Looking for a system that is interactive during play but won't fry a computer from having ~5 entities simulated.

How to make a model editing system.
- Similar to how games like Black Desert Online's character creator allows players to adjust meshes.
- How do I even start approaching this topic?
- Beyond mesh editing, I also want to know if there's a way edit the skeleton (rig) of a model in real-time.

I will likely ask more questions on things like decals and shadowing (huge).
For now I just want to make a simulated body in Unity that has interactive soft-body physics.

 

[Luderos]

I am looking for guidance on how to approach the graphical side of a project I've been wanting to work on.

About me
I am a recently graduated game dev with some experience in Unity, coding, and design.
I've made a few games of small to medium scope. (15-30 min experiences)
I specialize in systems and monetization.
Studying monetization absolutely violated my innocent view on the games industry.

What I need help with
Finding a model to start with.
- I am looking for a base model that's similar in style to Koikatsu.

How to approach soft-body (jiggle) physics in Unity.
- Any starting points for this would be very helpful.
- Looking for a system that is interactive during play but won't fry a computer from having ~5 entities simulated.

How to make a model editing system.
- Similar to how games like Black Desert Online's character creator allows players to adjust meshes.
- How do I even start approaching this topic?
- Beyond mesh editing, I also want to know if there's a way edit the skeleton (rig) of a model in real-time.

I will likely ask more questions on things like decals and shadowing (huge).
For now I just want to make a simulated body in Unity that has interactive soft-body physics.

You could check into UMA2 as a start. With a bunch of work, it can help you do most/all of these. However, be aware that these are all very big questions without simple answers.

 

[osanaiko]

There are a couple of developers who sometimes post updates in the https://f95zone.to/forums/programming-development-art.73/ subforum (the parent of this one) who have showed unity/unreal games with realtime soft body physics. Perhaps try to engage with them to talk about stuff.

Just my 2 cents: I'd consider being able to build out that stuff to be on the extreme limit of what an amatuer game dev could accomplish due to the sheer technical difficulty. And more than that, in my opinion, based on what I've seen so far of those sort of games (including semi-commercial stuff that pops up in the Latest Releases channel sometimes), it seems that they often have so much effort put into just getting the 3d graphics and physics working that there seems to be no energy left to make an enjoyable or interesting *game*. It's just tech demo stuff; no proper story, no character development, no eroticism beyond "wow look at the bouncy tits".

Is it due to the brain workings of such devs who can handle the complex coding for 3d engines lacking the ability to create engaging stories? Who knows...

 

[Laiga]

You could check into UMA2 as a start. With a bunch of work, it can help you do most/all of these. However, be aware that these are all very big questions without simple answers.

Yep the question I asked has a huge range of depth.
I am looking for guidance on the more nitty-gritty stuff though.
I've looked through guides like CatLikeCoding's mesh deformation tutorials, but it's difficult to figure out how to apply this to a human model.

The UMA2 has the features I'm looking for but not the model.
Perhaps I can look through their code and apply it differently.
I'll definitely take a look thanks.

 

[Luderos]

Yep the question I asked has a huge range of depth.
I am looking for guidance on the more nitty-gritty stuff though.
I've looked through guides like CatLikeCoding's mesh deformation tutorials, but it's difficult to figure out how to apply this to a human model.

The UMA2 has the features I'm looking for but not the model.
Perhaps I can look through their code and apply it differently.
I'll definitely take a look thanks.

I should have linked this before, but I would start by going through all of Secret Anorak's tutorials:

You must be registered to see the links

Also realize that the UMA stuff you see everywhere is just using the meshes that come with it. The real fun is when you're using your own assets with it instead. There are bugs though. Expect lots of debugging and reading the Unity forum thread and the discord 10x over.

 

[Luderos]

Is it due to the brain workings of such devs who can handle the complex coding for 3d engines lacking the ability to create engaging stories? Who knows...

I would guess that in general the better someone is at complex coding, the worse they are at storytelling. Also, no one's going to care about your story if your game is broken lol, not that I would know anything about that lol /sigh.

 

[Laiga]

I would guess that in general the better someone is at complex coding, the worse they are at storytelling. Also, no one's going to care about your story if your game is broken lol, not that I would know anything about that lol /sigh.

I am indeed quite dogwater at art and audio.
I can do basic vector art for icons and I can do an ok story. (Never told a NSFW one though)
I will look through the UMA tutorials and see how it works.

Just my 2 cents: I'd consider being able to build out that stuff to be on the extreme limit of what an amatuer game dev could accomplish due to the sheer technical difficulty. And more than that, in my opinion, based on what I've seen so far of those sort of games (including semi-commercial stuff that pops up in the Latest Releases channel sometimes), it seems that they often have so much effort put into just getting the 3d graphics and physics working that there seems to be no energy left to make an enjoyable or interesting *game*. It's just tech demo stuff; no proper story, no character development, no eroticism beyond "wow look at the bouncy tits".

Is it due to the brain workings of such devs who can handle the complex coding for 3d engines lacking the ability to create engaging stories? Who knows...

I have plans on what to do for story and interactivity. I'm going more for a management/simulation game.
My goal really is to have a solid system for 3D models/animations so that even if I start small I can build on it.

I am using the Unity DOTS system for better optimization in the "under the hood" systems. (like AI)
- It's super efficient with multi-threading jobs, but it is a pain to set up.
- Also made to be extended and built on easily.

Honestly if I can make a game with just a handful of animations using the 3D system and a very basic brothel sim around it, I would call it a win.


What do you think would be the better start for a project?
1) A highly interactive game (Little to no content) -> The Game Route
- NPCs with AI and banter on a 2.5D (3D) world.
- RPG style character system with real-time combat.
- Basic economy and progression. Get strong and take on challenges.
- Assets will be terribad and mostly placeholder.
- If this project does gain some momentum, I worry that the lull while I add graphics will have consequences.

2) High quality content. Animations & Tech Demos -> The NSFW Route
- I have to get to it eventually because I want it badly. It will take a while.
- I worry that adding this later will cause optimization issues. (DOTS is volatile)

 

[Luderos]

I am indeed quite dogwater at art and audio.
I can do basic vector art for icons and I can do an ok story. (Never told a NSFW one though)
I will look through the UMA tutorials and see how it works.



I have plans on what to do for story and interactivity. I'm going more for a management/simulation game.
My goal really is to have a solid system for 3D models/animations so that even if I start small I can build on it.

I am using the Unity DOTS system for better optimization in the "under the hood" systems. (like AI)
- It's super efficient with multi-threading jobs, but it is a pain to set up.
- Also made to be extended and built on easily.

Honestly if I can make a game with just a handful of animations using the 3D system and a very basic brothel sim around it, I would call it a win.


What do you think would be the better start for a project?
1) A highly interactive game (Little to no content) -> The Game Route
- NPCs with AI and banter on a 2.5D (3D) world.
- RPG style character system with real-time combat.
- Basic economy and progression. Get strong and take on challenges.
- Assets will be terribad and mostly placeholder.
- If this project does gain some momentum, I worry that the lull while I add graphics will have consequences.

2) High quality content. Animations & Tech Demos -> The NSFW Route
- I have to get to it eventually because I want it badly. It will take a while.
- I worry that adding this later will cause optimization issues. (DOTS is volatile)

Imo, do something MUCH, MUCH smaller than you'd like. Everything is going to take SO much longer than you think. Also, be sure to have a potato computer to test on. Unity will bite you performance-wise in ways you don't expect (careful of premature optimization and all that) and in ways that probably won't be apparent on your dev machine. Having said that, I would pick #2, with the caveat that it should be a tech demo that is carefully designed to not feel like a tech demo.

 

[osanaiko]

Honestly if I can make a game with just a handful of animations using the 3D system and a very basic brothel sim around it, I would call it a win.

You are absolutely right! Making anything complex and cool is very impressive, and you would have every right to be proud of it. The dark gods know that it would be greater than any thing I have produced for the community. I hope you enjoy and succeed at your self-assigned goals.

 

[The_Aesthetik]

I am indeed quite dogwater at art and audio.
I can do basic vector art for icons and I can do an ok story. (Never told a NSFW one though)
I will look through the UMA tutorials and see how it works.



I have plans on what to do for story and interactivity. I'm going more for a management/simulation game.
My goal really is to have a solid system for 3D models/animations so that even if I start small I can build on it.

I am using the Unity DOTS system for better optimization in the "under the hood" systems. (like AI)
- It's super efficient with multi-threading jobs, but it is a pain to set up.
- Also made to be extended and built on easily.

Honestly if I can make a game with just a handful of animations using the 3D system and a very basic brothel sim around it, I would call it a win.


What do you think would be the better start for a project?
1) A highly interactive game (Little to no content) -> The Game Route
- NPCs with AI and banter on a 2.5D (3D) world.
- RPG style character system with real-time combat.
- Basic economy and progression. Get strong and take on challenges.
- Assets will be terribad and mostly placeholder.
- If this project does gain some momentum, I worry that the lull while I add graphics will have consequences.

2) High quality content. Animations & Tech Demos -> The NSFW Route
- I have to get to it eventually because I want it badly. It will take a while.
- I worry that adding this later will cause optimization issues. (DOTS is volatile)

In my opinion some people on the forum are looking for something more than good animations and graphics related to sex, I think there are a lot of people around here who are also interested in the project of a complete game.
Undoubtedly, the two options take a lot of work, but I think you'll already know that

 

[aereton]

- I am looking for a base model that's similar in style to Koikatsu.

Well, there would be VRoid Studio. Doesn't have genitals though, you could go the route of adding them to the body mesh or look for a body base mesh (i.e. build-a-bod from smutba.se) and headhack the head onto it in the 3d modeling software of your choice.

How to approach soft-body (jiggle) physics in Unity.
- Any starting points for this would be very helpful.
- Looking for a system that is interactive during play but won't fry a computer from having ~5 entities simulated.

If you are seeking jiggle phys on only certain parts of a model, there are ways to to do them without involving the physics system via Runtime Animation Rigging in Unity. See minute 39:06 here

You must be registered to see the links

More on that below.
Otherwise it is kinda necessary to simulate multiple rigid bodies at once to give the impression of a soft body.

- Similar to how games like Black Desert Online's character creator allows players to adjust meshes.
- How do I even start approaching this topic?
- Beyond mesh editing

Forget advanced mesh editing. You don't want to edit your mesh at runtime, this would be a very messy approach. You'd want to use BlendShapes for that

You must be registered to see the links

You will have to set these up in your 3d modeling software, in Blender they are called ShapeKeys. They can be exported into the .FBX and then controlled at runtime in Unity through the SkinnedMeshRenderer component.
In the past I have thrown together a quick example of how that would work out in runtime. Please note, I did NOT put effort into it to make it look good, the ranges are quite bad and I did not fix the mesh normals. You can try that out here:

You must be registered to see the links

I also want to know if there's a way edit the skeleton (rig) of a model in real-time.

Well, there is a way to control and expand rigs in runtime. As mentioned above, this is done via the "Animation Rigging" package in Unity.
Unity Docs:

You must be registered to see the links

Good intro to this topic:

You must be registered to see the links

What do you think would be the better start for a project?

I'd scope way down. MVP in your case should include a simple character that has a few customization options (don't go overboard, 2 or 3 is enough) and all the necessary base gameplay stuff. You can always expand later, rework systems, etc. At the start, it is important to get across to people of what you are trying to do.

I hope this answered some of your questions. Unfortunately there is no simple answer to any of them as these are quite in-depth topics that usually depend on the scope of the game and available tools in your project.