Demo 2 Terrain and its Consequences. Also Bevy 0.20!

Overview

The demo 2 terrain and its consequences have been a major boon for Exofactory in depth, fun and visual impact. I am really excited to show it off in this month's dev post. Just the foliage and fauna really brings out the feel of the game. Exciting.

Additionally the demo 2 release date has been set. November 28th, 2026. More details below, but it should be playable most places, even those with less traditional support.

Lots to show and cover so without ado lots to show and see.

As always, if you find this post interesting, or if you are interested in the game, demo 2 or the full thing, I would recommend wishlisting it on Steam. It helps a ton.

Demo 2 Release

First off, the Demo 2 is officially scheduled to be released on November 28th 2026.

If you are on x86_64 Linux or Windows Demo 2 will be available on Steam.

If you are running an aarch64 or RISC-V system I will provide binaries on this site.

Ideally, I would like to produce *BSD binaries too, but that depends on difficulty, testers, and time.

Regardless it should be fun. I will be posting and publishing a blog post at demo 2's release with more details.

New Terrain Life & Limits

Exofactory went from a 2d top down game, to a voxel style game to a non-voxel 3d game over the course of its lifetime. Each of those steps have both majorly improved the game while leaving lasting traces.

One of those traces was the Demo 1 terrain. Because Demo 1 had used quite a bit of voxel logic it was virtually impossible to set up the rich terrain I dreamed of.

Even after the game was fully and properly de-voxelized the flat terrain was retained as a lasting trace.

We can see the richness and depth to the new deep savanna terrain.

I am very proud to present this as the replacement for that flat terrain.

The Demo 2 terrain was enabled by all the preexisting work done and in my opinion is beautiful and worth exploring.

As the global 'sensor fusion' camera we can't explore past where our sensors can reach.

This being said it is a demo map so the map is limited in size, so if the exploring gets too out of hand you may run into the edge of the map. Given that, in game, the edge is just the limit of what your in game sensor net can reach I thought a cool static effect might be fitting. The more I use it the more I find myself a fan of Bevy's ExtendedMaterial implementation of shaders and the recent WESL work

New Terrain Consequences

Balance

It can be harder to spot ore with just the naked camera in this richer terrain. Was that a rock or ore?

All of the foliage, and sloping, and terrain details do have a major effect on the ease of building factories. While there isn't a need to change any of the crafting recipes at this point some adjustments had to be made around:

I didn't expect to have to change so much just around terrain, but it makes sense given the simplicity of the last map.

Map

With a richer more dense environment it can be easy to lose track of where things are. Strangely enough it is shockingly hard to write a real time, performant map system. I did intend to include it with this update, but I couldn't get it into a place where I was happy.

Right now I am using a combination of the terrain height map and the color data to construct a passable "map" minecraft style but it isn't great.

My inspiration for the Exofactory map comes from the Cartograph Map Mod. I quite like how buildings, conveyer belts and the like are represented here in addition to the topography.

I do expect to clear things up by the time demo 2 is out so you should be able to see the full glory of the map then.

Environment Scanning & Compass

At the top of the Exoframe hud we now have a compass line. We can see the tier 1 core marker below the compass line.

Given that you interact with the world through a series of exoframe robots somewhere between the size of a large cat and a small pig it can be easy to lose track of where you are.

Given that it is much harder to find ore nodes in the wild given all the new details a scanning system has been added too.

The compass will show icons to represent any arbitrary point, building, resource, or exoframe you select in the map. You can manually add points to the map, or scan for points that will be added automatically. You can scan for things like specific ores, resources, or objects. Scanning will be done by exoframes, passively at a small range as they wander the world, or actively at a large range when a high power scan is manually started.

Bevy 0.20

Additionally in conjunction with the work listed (and not listed) above Exofactory has been migrated to Bevy 0.20 from 0.19.

The migration itself can be described in one word, this word being "churn". Almost 2000 lines of BSN changes were made alone. This being said the improvements to things like Enums, and better children handling in BSN are quite nice and make it cleaner. Better to fix this now than to live with less than ideal BSN later.

Other than that the game migration was relatively smooth.

I will reiterate two pieces of advice here:

  1. If you are actively working on a long term Bevy game project you should migrate to the latest Bevy version as soon as you can. To stay on an old version of Bevy is to risk your project stagnating.
  2. Be selective in which 3rd party crates you include as a core part of your game. If your game depends on a specific 3rd party bevy crate you should be confident that is well maintained with timely updates ideally spanning multiple Bevy releases. If a core to your game crate is abandoned or not updated that leaves you with the burden of maintaining the crate yourself, or your game being stuck on an old Bevy release.

On point two I have two additional comments.

First and most importantly, if the core of your game depends on a 3rd party bevy crate, and the creator of said crate accepts donations you should consider donating. Even just from a selfish perspective it makes sense, let alone from a community oriented perspective.

Secondly, I created a tool named BevyDex to browse and see details on 3rd party bevy crates. I created it for myself originally, but if you want to see how long a crate has been around or the like I recommend checking it out.

Conclusion

I am excited to polish off the Demo 2 features and to see all the feedback and fun from it. Exciting times ahead. Surely there are no big competing games that are being released in November, right?

And once again, if you found this enjoyable, informative, or want to play the demo, I would suggest giving the game a wishlist.