Workflow engine Puns

Traditionally we do 3D sculpting/modeling/retopo in DCC, UV unwrap, texturing in Substance, and then we get game-ready assets to use in Unreal.

Now we have Houdini Engine, which allows us to use HDA directly in Unity/Unreal. But how does Substance fit in this workflow? As far as I know there is no way to use HDA in Substance, right? So do I still need to export .obj/.fbx first? But if I would export .obj/.fbx anyway, what's the point of Houdini Engine? Is it only for assets that can be procedurally textured?

I recently started working in Love2D in preparation for the Playdate, and I quickly found I wanted a similar workflow that I have in Unity3D. I spent some time reading the (entire) Lua manual, and felt I was in a pretty good spot to make a light weight unity style engine. I also learned to really love Lua. Such a cool language.

I was after these things:

• A scene graph (game objects and parents/children, full transform support with local/global position and transform caching)
• Game object events and lifecycle management (Enable/Disable, Messaging eg Update/LateUpdate)
• Components (multiple components attached to game objects)
• Unity2D style layer management and auto scene graph matrix pushes
• Portability - make stuff in Love2D, but easily move it over to Playdate by wrapping up some core engine features
• Robust error handling - component errors shouldn't break the engine
• Not to recreate unity, but bring over core workflow features

I spent a little time putting together little library called "Lunaty", partially because it was madness I was doing it in the first place.

I thought I'd ask in here if that's something the community may be interested in, because if so I could flesh it out, smooth off the edges, and offer it as an open source library for anyone else who wanted to use it.

As an example - this is how you create a new GameObject and add a couple of components

https://preview.redd.it/vysnntubw0381.png?width=920&format=png&auto=webp&s=0135358e66ddf22344f77d00a22d1bf8a7f16c7b

And this is how you define a component

https://preview.redd.it/6mcxy2fow0381.png?width=731&format=png&auto=webp&s=35606ed57ea73bbabda565bb780c4d52fd606bb4

The system will manage the lifetime of the object, unity style, so you don't need to track object. Just create new ones and they'll be alive until destroyed.

To set it up in a project, you just create a game object like this, and call it from the Love2D update loop

https://preview.redd.it/abkb4axrx0381.png?width=804&format=png&auto=webp&s=25bcab59f0872b0634fe0d2641eb8f308191bda6

https://preview.redd.it/0pdmlaawx0381.png?width=222&format=png&auto=webp&s=b2f3cc9e7a6cf3a2528def12fd5bc2b336cfa9e4

So yeah - thought I'd just test the waters to see if this is something that others may want to use. If not, then I can just keep using it as my internal tool, but maybe others would be keen on something like this too.

EDIT: I did just discover there is another proj

I've been using Blender for quite a while, and I'm used to it's texture painting tools, creating procedural materials and baking the textures, etc. for game development. I optimize the mesh on every single face, always taking a lot of care with polygon count. For that, I use the Edge Split modifier a lot, instead of adding extra edge loops to sharpen edges. Also, no subdivision surface.

Then, I decided to give Quixel Mixer a go. The thing is, quixel mixer requires a lot more geometry to work with smart materials. SM simply don't recognize edges when they're split (and yes, it makes sense, if you imagine an algorithm guessing edges under the hood). An alternative is to work with curvature maps. The problem with this is it requires one to model an extremely detailed model, and then bake it into a low res mesh together with the required maps (curvature, normals, AO...), and this seems like an extremely time consuming task. There is also UV unwrapping, rigging, and all the good stuff to debug until ending up with the final game-ready asset. Considering I'm working right now on a character, which in game is a skinned mesh, just makes things worse on the polygon count problem.

Just because of this tiny little problem with 3d edge detection, I have to take this whole process, which is very indie-unfriendly to me. Seems like I'm better off (I mean, faster and simpler) just modeling the final game-ready mesh once, and making the quixel mixer's work on my own, hand painting masks, edges, imperfections and stuff all in blender with nodes + cycles.

My question is: Do you know a nice workflow from Blender to Quixel Mixer to UE4 that could speed up this process?

I don't have experience modeling and baking extremely detailed meshes, so I'm a little bit afraid of the problems I might encounter, like UV unwrapping, rigging, etc. Tell me about your experiences, if you faced the same problem and managed to figure out a way to do that efficiently. I hope I'm just being too attached to what I'm used to in blender, and the whole modeling + low res baking process is fine and efficient.

Maybe, there is a way to quixel mixer recognize Edge Split modifier in .blend files? idk.

Does bridge for blender overcome this situation? I never tried
I need help. Being a solo dev under burnout on the spare time sucks quite a bit.

I'm keen to understand where and how they are used. Thank you in advance for any help you can provide.

I downloaded unreal engine about a month ago and the render times are absolutely unbeatable, but blender still tops it for animating and creating custom assets. Does anyone have any resources to get the best of both worlds here? I’m thinking something along the lines of material baking -> fbx scene export -> import to unreal? Any help would be great, ty

Github link

C++ Workflow is a light weighted Asynchronous Network Framework and its elegant design can satisfy most C++ back-end development requirements including search services, HTTP applications, online advertisements, etc.

To quickly build an HTTP server:

#include <stdio.h>
#include "workflow/WFHttpServer.h"

int main()
{
    WFHttpServer server([](WFHttpTask *task) {
        task->get_resp()->append_output_body("<html>Hello World!</html>");
    });

    if (server.start(8888) == 0) { // start server on port 8888
        getchar(); // press "Enter" to end.
        server.stop();
    }

    return 0;
}

• It currently supports HTTP, Redis, MySQL, and Kafka protocols.
• It has built-in service governance and load balancing features.
• ...

For more information, you can see the github.

Any feedback would be greatly appreciated!

Hey, I recently interned at a small startup and they used something they called a workflow engine, which used to generalize the process of building user workflows. It would have been so complicated to write business logic for every new flow, but this piece just made it super easy to develop new flows. It changed the way I thought about design.
Q 1. Is this something that is used very heavily in the industry?

When I googled I got Form Builder websites. I am thinking of building one (I have a use-case), but I am interested in knowing if there already exists something like this.
Q 2. Are there other ways of developing complicated user flows where multiple users, multiple steps, etc might be involved?

Thank you in advance!

For me it's been Q to straighten the connections between nodes as well as "open the last level" option on the engine startup.
I think I've already saved myself hours just by knowing these and I was wondering what else I'm missing out. Do you have any tips?

I bought a 256gb Flash card and I'm wondering how to setup my library on the flash card. I read the manual but it still wasn't clear to me.

Do I put my music on my flash card and then point Engine DJ software on the PC to that card? Or do I keep the music on my PC and then load it into Engine DJ and use the sync button to get my music onto the flash card?

Shipt released Plinko - a golang library for building light weight statemachines and workflow engine for go. It's licensed under the MIT license. The number of states and transitions are limited only by memory, and speed of transitions happen at over 25 million transitions per second per thread/go-routine. Business processes during transitions will take much more time, but this demonstrates the minimal overhead in using the library.

Check it out. https://github.com/shipt/plinko

Hey, I recently interned at a small startup and they used something they called a workflow engine, which used to generalize the process of building user workflows. It would have been so complicated to write business logic for every new flow, but this piece just made it super easy to develop new flows. It changed the way I thought about design. Is this something that is used very heavily in the industry? When I googled I got Form Builder websites.
Are there other ways of developing complicated user flows where multiple users, multiple steps, etc might be involved?