Building a Full Ecology and Physics Simulator With Luanti

[Wranglyph]

Hey all! I recently posted a dev call for a fairly big project, where we are going to try and get a full ecology simulation working in Luanti.
(We'll be passing some data to an external tool, so I was advised to post here, rather than the regular game forum.)

The dream, in addition to being a fun game, is to make something that will really hook people and hopefully get some new FOSS converts. But alas, we only have the one dev right now, and are going to need at least a handful more.

So if that sounds interesting to you, then please check out the dev call, where I cover:

-overall game outline
-key technical features
-gameplay those features will enable
-why I'm building it
-why I think it CAN be built (it's fairly ambitious 😅)
-complete roadmap, from "just watched" to "just launched."

I hope to see you folks there. 🫡

https://www.youtube.com/watch?v=ZJSYKODQsY4

[Wranglyph]

Update: I have now made a *much* shorter (3 min) version of the pitch:

https://www.youtube.com/shorts/H51aqnxelyM

[proller]

One fork has some working parts of your ideas
tree grow https://github.com/freeminer/freeminer/ ... w_tree.cpp
real liquid flow https://github.com/freeminer/freeminer/ ... liquid.cpp
liquid freeze-melt https://github.com/freeminer/freeminer/ ... nt_abm.cpp

This features enabled on all "fm: *" servers.

Free advice: Do not even try make complex systems on lua, if you want something working with acceptable speeds - do it in core c++ (or better from scratch with godot voxel)

[Wranglyph]

Yea based on the notes I saw it looks like AnotherCatGirl (current lead dev) is planning to do the cell calculations in Rust, which is fairly performant. Later down the road, if the project is ultra-mega-successful, we might refactor the base engine as well. But for now I think that just getting something to work is the higher priority.

You're not the first person to mention freeminer either; I wonder if it might be a better fit for the project? 🤔
I'll ask on their discord if anyone is interested.

[proller]

What you need and it missing in luanti:
Larger world size (solved in freeminer)
Custom node params (instead of param1,param2 should use unlimited named values (light,temperature,humidity, and many others) )
Multithreaded world processing (partially solved in freeminer)
For good mobs system you need global objects with permanent global id's and global access api

Also useful feature in freeminer - abm_world - permanent abm runner for whole generated world

[Blockhead]

You're not the first person to mention freeminer either; I wonder if it might be a better fit for the project? 🤔
I'll ask on their discord if anyone is interested.

You're talking to the Freeminer main dev, proller :)

[Wranglyph]

You're talking to the Freeminer main dev, proller :)

Yea, I pieced that together once we started talking on discord. 😅

Good news for the project btw: proller seems willing to help with engine-level integration, which will allow us to move to production much faster than I anticipated. (Though it does mean forking almost instantly 💔)

Also: it seems that AnotherCatGirl (lead dev) has made some kind of cell-calculator prototype. Getting it to run is a little over my head at the moment (lots of Rust commands involved), but hopefully I'll be able to show some screenshots within a few weeks.

[Walker]

What you need and it missing in luanti:
Larger world size (solved in freeminer)
Custom node params (instead of param1,param2 should use unlimited named values (light,temperature,humidity, and many others) )
Multithreaded world processing (partially solved in freeminer)
For good mobs system you need global objects with permanent global id's and global access api

Also useful feature in freeminer - abm_world - permanent abm runner for whole generated world

Perhaps you should mention that you are the current main developer of Freeminer and that this is therefore advertising?

Nobody else would recommend freeminer, since the project is practically dead... most of the recent major commits are just "Merge tag '5.13.0' into wip5.13.0" or "Merge tag '5.12.0' into wip5.12.0" or "Merge tag '5.11.0' into wip5.11.0" ... yeah

[Wranglyph]

Perhaps you should mention that you are the current main developer of Freeminer and that this is therefore advertising?

I mean... an introduction would have been helpful, but it's all good as far as I care. What's the point of FOSS, if we can't share what we love?

As far as which engine is a better fit, that will be ultimately be up to AnotherCatgirl, as the de facto lead dev.
She's kinda busy irl, and the current prototype doesn't require engine work yet, so that decision may be a ways off yet.

[proller]

Nobody else would recommend freeminer, since the project is practically dead... most of the recent major commits are just "Merge tag '5.13.0' into wip5.13.0" or "Merge tag '5.12.0' into wip5.12.0" or "Merge tag '5.11.0' into wip5.11.0" ... yeah

You are terribly wrong! Most of them is "fix" and "wip"

[Walker]

Perhaps you should mention that you are the current main developer of Freeminer and that this is therefore advertising?

I mean... an introduction would have been helpful, but it's all good as far as I care. What's the point of FOSS, if we can't share what we love?

i didnt say he cant share ... i just said that he should mention that it is his project that he is praising so highly.

People do this simply because it leaves a strange aftertaste to praise their own project (even if it's justified) without saying that it's their own.

[Wranglyph]

Ok, here's a "miner" update:

I've been sick, so not very productive. (Kind of just making this to feel useful, honestly.)

Catgirl, on the other hand, has managed to make a rudimentary cellular automata tool, and tested it with a simple gaseous diffusion simulation. It's too jank to be worth sharing right now, but is very promising.

It seems like her and proller are planning to cut Lua scripting/tables out for performance, & store all of the cell entries in some kind of distributed database. This might mean servers could perform *better* than single-player? Idk. 🤷

As for me: sometime in the next week or two, I should have some "vision docs" ready, which will hopefully break all of the design work into tasks that can be done by 1-10 people. They'll mostly be fun, like "figure out how many flowers we need." (Although not quite that specific.)

Once we have some people who want to help design the game-play, I'll start handing them out. (Got one or two people already, but may want more.)

In light of discord's decision to jump off the deep end, I'll probably be organizing that on Zulip, which is a FOSS tool for organizations to coordinate with. Once we're at work, I'll mostly use this thread for updates.