#Procedural generation

[prokopetz]

Like, I'm not gonna argue that AI art in its present form doesn't have numerous ethical issues, but it strikes me that a big chunk of the debate about it seems to be drifting further and further toward an argument against procedurally generated art in general, which probably isn't a productive approach, if only because it's vulnerable to having its legs kicked out from under it any time anybody thinks to point out how broad that brush is. If the criteria you're setting forth for the ethical use of procedurally generated art would, when applied with an even hand, establish that the existence of Dwarf Fortress is unethical, you probably need to rethink your premises!

[wauzmons]

Procedural tree trunk, making use of the new spline component in Elysian Mesh. Huge step towards a proper plant editor.

[squids-among-stars]

Squidolus [Day:2000 Hour:12]

[blake447]

Procedural Dungeon Generation

Alright~ After much work we have finally completed our dungeon generation algorithm. First, we generate the rooms, separate them using a separation steering algorithm.

Second, we generate a graph that represents connections between the various rooms

Third, we perform what is known as Delaunay Triangulation, in which we evaluate pairs of triangles and swap their shared edge if it fails a test boiled down to a simple matrix determinant evaluation Fourth, we construct a Graph of connectivity from the Delaunay triangulation

Fifth, we perform an A* pathfinding algorithm to connect points on the delauney triangulation through the neighborhood graph

Finally, if we didn't traverse through a room, we can remove it to simplify the dungeon into something a little less dense

There are some extra things we can do here and there, and bugs and edge cases to iron out here and there, but for now, we have a pretty neat little dungeon generator. The next step is to actually procedurally generate the rooms and then use the connectivity graphs to implement game logic and whatnot, but that we will save for another time.

[auranym]

Silt Spawn

an exploration-based survival platformer set within twisting, dark, dusty caverns.

demo releasing in late December! follow for updates :D

[sketchshroom]

I added a player character The player literally just uses the default player script the Godot engine gives you by default

You can move, jump, and reload the level

Trust me, it will look better soon

[pad-wubbo]

Good [crackalackin] news, my [Homeslices].

I have authored a [way to the promised land].

The [promised land] is available when you click on this [Hyperlink]:

https://github.com/wubmush/woomyband

It will [slice and dice] new combinations of [biosolids] that your eyes have never sizzled befizzle, fo' sure daaaawg!

[heartmachinez]

Pangea: Creating A Multiplayer Rogue-lite With Endless Open Worlds

We’ve shared extensively about our tech art strategies and proc gen processes in both a recent Heart to Heart with Len White and Christian Sparks as well as on our dev blog. We’ve also discussed our environment art works in progress in a different Heart to Heart with Will Tate and on our blog as well.

That was all months ago; often, a few months can mean a lifetime in game development.

Original Vision

Years ago now, when Alx was ideating on the design pillars of the game, the question that came to mind was “what would you do in an open world you’d never see again if you die?”

With that in mind, we made decisions in the early days of our game to try for a more reasonable approach of this idea as we built our systems, since it seemed insurmountable. So, we created an adjacent version, something that captured parts of this design ideal. We had a large, open biomes, but they were segmented in a stage-by-stage format to make it more feasible for us to build.

Over the course of development, we found that, as we continued to build the technology needed for these smaller open-biomes, that we could actually leverage the tools to make the original vision a reality. Thus, we shifted away from the more limiting and (ironically) more complex version of a stage-to-stage progression, and started on a “Pangaea Shift”.

The Shift

Pangaea is used as a code name, as we were essentially merging all of our stages into one larger map to create an open world.

This shift meant that we would mean that we lose some time up front reconfiguring some parts of the game to function in the ne structure, but gain time on the backend and a much more exciting game format to dive into. We were excited and scared, all at once.

This shift yields us:

Highly differentiated points of interest on a global scale, resulting in entirely new biomes to explore instead of sub-variations of the same biomes

Reduced per-level workload for Houdini, focusing on simpler, bolder biome elements since the context of other biomes being present shifted the dynamics of play so significantly

The ability to generate dynamic, global components that affect the whole run / playthrough, rather than just stage or biome-specific elements, opening up tons of exciting mechanics

A truly open world, procedurally generated, with biomes juxtaposed seamlessly on the same map

An open world you’ll only see once

It’s a thought that leads to a lot of questions and exciting ideas. How much do I explore this world? How much time do I invest, knowing I could die at any turn? What are the pressures driving me forward in this world? What’s new, exciting, different this time? What’s coming next?

These are all questions we ask and answer for development, and ones we are excited for you all to see the conclusions of for yourselves in Early Access and beyond!

Wrap Up

What do you think of our process shift? Share your feedback!

See you next time, Breakers!

-The Heart Machine Team

[justthatdailyrender]

This video counts today as 2 renders right... Please

This took me 2 days of trial and error

[bonicle]

I Just released my new game, Islander 95!

Chill out to the sounds of early internet and calming waves, and generate yourself some procedural islands! (comes with a free postcard maker too)

This is a cute little toy I made for a university assignment. You can probably guess what it does by now, so give it a shot if it sounds interesting!

oh also the names are completely random and there is some very funky stuff that can happen lol

have fun!!

[feathersalwaysmakepplattractive]

[midautumngame]

Introducing Chaos: Let's Build Some Procedural Narrative Systems! #GDoCExpo

Here's our project lead, Sherveen's talk from last year's GDoCExpo about building procedural narrative systems, which is a big part of the development of games like Hades, Wildermyth and Midautumn!

[wauzmons]

Elysian Mesh now supports "layouts", making it possible to place objects along circles, splines or other mathematical shapes!

[squids-among-stars]

Squidolus [Day:2140 Hour:0]

[blake447]

Alright! Delaunay triangulation is officially, properly implemented. Notable roadblocks: Some boolean logic was terminating early because I one-lined it improperly A random test case was running on update constantly messing up the triangulation, causing me to question my sanity as I desperately figured out why it was flipping things back and forth Swizzling and figuring out the order the elements had to go into the matrix. At the very least, they have to be ordered counterclock wise, and just for good measure I matched the order to this diagram (again, courtesy of wikipedia)

And finally, we no longer needed a test for convexity as the Delaunay process handles that case as well! meaning it was actively getting in the way of the algorithm and had to be removed! Super excited to move on from this nightmare

[auranym]

Silt Spawn is really coming together! :D

I added so much since the last update: a "new game" animation (seen above), save files with autosave, hunger mechanics and animations, and more!