Surprise, it's another update for Iron Village! Iron Village got a big bump in sales recently, which initially looked like an end-of-sale spike, but turned out to be driven by this video by the Geek Cupboard:

It's awesome to see Iron Village popping up still, but of course more players means more opportunities for bug reports! There were two bugs reported last week:

1. On Android, the foundation for the Magick Transformer was missing, so the player couldn't build it.

2. On the Steam Deck (but also PC), the controller would stop responding when going back to the menu, even though the menu was in focus.

I'm still not 100% sure what happened for #1, especially since I couldn't reproduce it, but it did highlight the difficulty of debugging on someone else's Android device. Actually accessing saves isn't possible right now (each app's storage is kept secure from any other app accessing it, but that apparently includes the user too), and in order to get any logs you have to ask the user to install adb, plug their phone in, reproduce the bug, and copy/paste the relevant logs. (Hope the player understands what's relevant!) As you'll see below, the changes made here are mainly band-aid fixes, UI to better surface issues, and unrelated bugs I encountered.

For #2, the issue was specific to how I implemented the Steam Input API. Whenever an action came in, it would let the [c]ControllerHintManager [/c]know - that way, the controls get shown when the player is actively using a controller. I have [c]ControllerHintManager [/c]implemented as a singleton: it's a type of object in the game, but it's designed so that only one exists at a time. This singleton "pattern" can be implemented a few different ways, but in Iron Village I just use a static variable Instance, and set that when the first instance gets created. The singleton is an unfairly maligned pattern IMO - for a lot of software it does end up encouraging bad design, but there are quite a few places where it can be done more responsibly in game development, especially in smaller projects. Is it a good idea in this instance? IDK, it's not great, but not terrible.

Anyway, the issue here is that [c]ControllerHintManager [/c]only exists in the game itself, not the main menu. The first time you load the menu, this is fine - there is code that checks if Instance is null, and skips talking to the [c]ControllerHintManager[/c]. It works fine within the game of course, since the manager actually exists there. When you go back to the main menu though, that deletes the [c]ControllerHintManager[/c], and that's where the problem starts.

Warning, discussing about programming languages and other details ahead!

The follow up question is this: what does it mean to delete something? In most cases, you're not actually changing anything about the data in memory or on a hard drive, you just tell the operating system that you don't need to reserve that space anymore. Eventually something else will probably use that space, but the underlying bits don't immediately change. That's the basis of how a lot of data recovery tools work - the "empty" part of the hard drive can still hold data that was previously written.

If Iron Village were using a "lower level" language like C or C++, we might get away with this issue for a little bit: the Instance variable would be pointing to a "deleted" [c]ControllerHintManager[/c], but as long as it hasn't been overwritten in memory yet, it might work. (This is bad code because it's undefined behaviour, and could end with the game fully crashing out instead.)

In C# (and a lot of other "higher level" languages), there's an added layer of memory management and garbage collection. It's keeping track of what's been deleted, so if you try and access deleted memory, it throws an error that you're trying to access a disposed object. Crucially though, our Instance variable isn't null - it's pointing to that deleted object! So now when input comes in from the Steam Input API, it tries to talk to the [c]ControllerHintManager[/c], and throws an error. Because of how error throwing works, it skips a lot of code that comes later - including that part that actually acts on that input. (Without any additional code, this would crash the whole game, but the Godot engine tends to catch most exceptions once it's no longer your code. Sometimes this isolates the problem and lets the rest of the game continue working, other times it just makes things weird.) So every time you try and scroll down, the code tries to contact [c]ControllerHintManager[/c], fails, and gives up.

The solution was just to set the variable back to null whenever the [c]ControllerHintManager[/c] gets deleted (i.e. in Godot's [c]_ExitTree()[/c] method), a small change that probably should've been done from the beginning. 🙃

Anyway, if you've made it this far, here's the full release notes:

• Fixed an issue where corn fields (and only corn fields) were being drawn above buildings in front of them.

• Added an error dialog to the game to display errors that would normally only be seen in the logs. (Note that it doesn't capture all errors, only those in select sensitive areas of the code, like loading.)

• Fixed an issue where the Maximum Production Rate button wouldn't appear in the Railway Status Window immediately after building the Railway HQ.

• Added code to build the Magick Transformer Site on loading a game if it should be there.

• Fixed an issue where Steam Input would no longer respond after exiting to the main menu.

I know, I know, you’re looking at that title and thinking, why on earth would I want to look back on this year?! And yes, the world is uh, not doing so well, but we’re at least going to take a look back at the year from the perspective of Lunar Chippy Games.

But first, if you’re reading this and haven’t bought Iron Village yet – it’s 30% off for the rest of the year!

Iron Village at BostonFIG back in December 2024

The biggest date of course was March 24th, but I’ll backtrack a little bit first. Iron Village was at BostonFIG on December 15th, so after that and holiday shenanigans, the year started off with implementing the crazy amount of feedback from then. (Not all of it of course, feedback is super useful but not 100% correct.) Levels 4 & 5 hadn’t even been fully implemented yet, so there was still a big chunk of work to cram into two and a half months.

Somehow I pulled it off, all while holding a full time day job at the same time, launching to mostly positive reviews. Was it a huge launch? No, but Iron Village was well received, contrary to what part of my brain had tricked itself into believing, and it turns out I could in fact make and finish a game!

An image from Iron Village 1.1, showing off some of the (then) new features.

A lot of initial feedback from the game came to make the basis of version 1.1[lunarchippy.games]: Additional rotations of buildings, a Royal Ceremony to give the game a soft ending, as well as some other goodies and quality of life fixes. 1.1 released in June, and along with the corresponding sale, another big boost of sales came along.

Meanwhile, at some point during the early summer, the “Voluntary Exit Program” was announced at my day job (at least for the department which I worked in). I signed up for that so fast lol, the only delay was me getting lunch and calling my wife to double check. (She was more excited than I was!) Based on what other big tech companies were doing, getting to actually voluntary leave with severance was huge, and most likely thanks to our union. (If you can, join a union!) I’m tempted to rant about my manager here as well, buuut that can wait for another time, if ever.

The theme for update 1.2. Original logo ©3Division.

After signing up for that, but before actually ending my job, the 1.2 release[lunarchippy.games] came out, a.k.a. the Workers and Resources[lunarchippy.games] update. This was another hefty update, and included some charts for deeper analysis, some new buildings, and a bunch of other improvements.

Since then it’s been full speed ahead on Project OY, the unannounced new game I’m working on. It is a wee bit terrifying going full time right now (in this economy?!), and it’ll be awhile until anything actually comes out, but the way things are planned it should work for at least the duration of development. (If you’re interested, there is a Patreon that gets you some sneak peeks, but at this early stage I also totally get not wanting to drop cash on that!)

Another sneak peek at Project OY.

Finally, I wanted to drop some unsorted statistics, some of which I found pretty interesting:

Net Sales on Steam: 1,643

Users on Google Play: 113

Of 1679 total units on Steam, 72 are from Mac and 46 from Linux (I’m not sure how precise the OS assignment is though)

5,608 outstanding wishlists

Median time played: 1 hour 23 minutes

56 reviews on Steam, 91% Positive

Top Countries:

• 438 units (26%) in the US

• 347 (21%) in Germany

• 102 (6%) in the UK

• 89 (5%) in China (PRC)

• 83 (5%) in Japan

• 58 (3%) in Canada

• 57 (3%) in France

• 55 (3%) in Taiwan

• 50 (3%) in Australia

• 44 (3%) in Russia

• 356 (21%) Other

That just about sums up 2025 for Lunar Chippy Games, hope you all have a Happy New Year! At the very least, I hope that 2026 ends up being at least slightly better? 🤞