The Stack

The very first time I heard about Blue Prince, it was from a social media post by Andrew Plotkin, in which he stated that he didn’t feel the need to write a blog post about it because everyone else was doing that. It’s basically this year’s Animal Well — and not just in the sense that it’s this year’s popular indie game, but in the sense that it’s got multiple layers of mystery to solve even after you’ve technically won. It makes me wonder if every hit puzzle game from now on is going to be like this. The canonical joke among the fans is that when the credits roll, that’s when the game begins. Reaching that point took me four days of real time, and 52 days of in-game time, which I understand to be well above average, but the RNG wouldn’t give me a break — in most runs before that, I’d either be able to reach an entrance to the Antechamber, or have the means to open it, but not both at once. But taking so long meant that I had already made significant progress on the post-game content that’s dominated my attention for the last couple of weeks.

But I get ahead of myself. Let’s cover the basics: Blue Prince is fundamentally a puzzle-based first-person adventure game, but it’s sort of embedded or intertwined with a tile-placing board game along the same lines as Betrayal at House on the Hill or the D&D Adventure System games. The premise is a classical freak will: your deceased great-uncle has bequeathed you his manor, but only if you can meet the challenge of finding its secret 46th room, a challenge made more challenging by the way tha the manor’s layout changes from day to day. Whenver you open a door, you get a choice of three rooms, drawn at random from a pool, that can be on the other side. When you’ve either filled up the grid or (more likely in the early stages) run out of doors to open or keys to open them with, you can call it a day and reset the estate. Most of the adventure-game puzzles rely on drawing specific rooms, or specific combinations of rooms, or specific rooms in combination with specific randomly-placed objects.

On the face of it, this sounds like a recipe for disaster. Don’t you get frustrated and annoyed waiting for the combinations necessary for progress to randomly come up? And yes, you do, somewhat. And there certainly exist people for whom this is enough to make them lose interest in the game completely and then make lengthy posts complaining about it on the Steam message boards. For me, it produced the opposite effect. Not being able to try out my ideas immediately makes me all the more eager to keep playing until I can. But why? That’s the interesting part.

I think part of it is that it gives you multiple aavenues of progress. I’ve noted this before about RPGs and adventure/RPG hybrids: when you’re stuck, you can always grind for XP. Blue Prince doesn’t have XP, but it does have other incrementally accumulating attributes that help to make you feel like you’re making progress even when you don’t accomplish your main goals. For example, raising your allowance, which is the amount of money you’re given at the start of each day to spend on special items. Eventually you get to the point where your allowance is large enough for any expense and you don’t need to increase it any more, but by that point you’ve probably discovered another incremental goal to take its place.

Moreover, though, the very fact that you can go through multiple runs without an opportunity to try out your intended solutions to puzzles means puzzles tend to stay alive longer than they would in a conventional adventure game. At any given moment, you have several back-burner goals that you’re ready to pounce on the moment the game deigns to give you the necessary resources. You’re probably thinking “Sure, until the end, when you start running out of goals”. But it maintains this state for a remarkably long time, just by unfolding its mysteries gradually and revealing new meaning in what you’ve already seen.

But I think the really crucial thing is that the tile-placing game is engaging. Picking rooms on the basis of their contents and constraints requires enough thought and attention that it could easily be made a decent game of its own, without the rest of the story and puzzles embedded in it. As a result, most of the time you spend playing this game isn’t actually spent thinking about the adventure puzzles. Minigames in adventures feel annoying when they feel like interruptions, disrupting the game that you were absorbed in playing. Here, it’s kind of the reverse: the adventure is embedded in the minigame.

Still, things do wind down eventually, and I’m well past the point of diminishing returns. As in Animal Well, you have to decide when you’re done. I’m not quite done — there’s still a thing or two I want to do before closing it for good — but I’m getting there.

So, yeah, I’ve dropped into a months-long silence once again. There is a reason for this: I have been making a game. An indie puzzle game, but one that I intend to fill out to the point where I can in good conscience ask people to give me money for it. I’ll have more to say about it later, when it’s closer to complete, but the title is Runecaster and the elevator pitch is “DROD with spellcasting”.

I’m well aware that this is not likely to be a profitable endeavor — the average indie game, to a very strong first approximation, sells zero copies. But I’ve committed to spending a year developing it anyway, because (A) I can afford to, and (B) it beats looking for work in the current job market. Moreover, this is a game that I’ve been thinking of making for many years — decades, even. The initial inspiration came not from DROD, but from an obscure 90s action-RPG called Four Crystals of Trazere (or simply Legend in its European release). Four Crystals had this magic system where you constructed spells from runes, each of which had a specific and deterministic effect, forming a sort of miniature programming language. There were a handful of points in the game where it made puzzles out of this, like a sealed chamber that can only be opened by a lever inside the chamber, so that you have to figure out a way to press the lever with magic. But it wasn’t the game’s focus, and I always felt like the idea could be taken a lot farther.

Over the years, I’ve made a few previous attempts at implementing my ideas — once in Unity, once in Javascript — but this current attempt, using Godot, is the first time I’ve gotten far enough to think I’m going to finish it. And I’ve contemplated repurposing this blog as a dev diary. But the unfortunate fact is: Blogging and working on this game seem to tax exactly the same mental resources. On any given day, I am capable of working on the game, or blogging, but not both. And development has taken priority. I’m going to try to establish a schedule of game development on weekdays, blogging on weekends. We’ll see how that goes.

But in all honesty, although this has been the thing occupying my attention for most of the last few months, it hasn’t been the thing occupying my attention for the last two weeks. Something else took me over pretty thoroughly, and is only just now letting up somewhat. I’ll describe that in my next post.

Speaking of titles that I played obsessively for a time, 2024 also saw the release of a new Creeper World game! But it occurs to me that I never posted about Creeper World IV here, so let’s talk a little about that first. Creeper World IV was the franchise’s foray into 3D, and it was fine. If you’re a fan of Creeper World, and you’ve wondered what it would be like with 3D models, it’s worth a look. But it’s nothing to write home about, especially after Creeper World 3: Arc Eternal, which is, to my mind, still the ultimate and definitive Creeper World experience. Being 3D adds the possibility of a first-person mode, and, while this wasn’t used in any of the campaign mission, it’s telling that basically all of the top-rated player-made levels use it. It’s like the fanbase decided it was more fun to use the engine to play a different game.

Now, the new one: Creeper World Ixe. (Or, according to some of the title graphics, Ixe Creeper World. “Ixe” is the name of an alien race in the game’s backstory, which I will not be describing any more than that.) This game abandons the 3D and in fact brings us back to the vertical 2D view that we haven’t seen since Creeper World 2, based on cave systems that constrain and pressurize your fluid enemy. But the game isn’t just Creeper World 2 brought up to modern standards. It’s Creeper World 2 hybridized with Noita.

This might seem strange. The Creeper World games are real-time strategy games in a sci-fi milieu, and Noita is a fantasy Roguelike. But they both have a lot to do with simulating fluids, and the main thing Ixe gets from Noita is its pixel-level simulation. CW2, in contrast, was fundamentally tile-based. The world was a grid where everything you could build occupied one square and the Creeper was essentially a cellular automaton. The pixelation of Ixe is notably coarse, but not tile-level coarse.

And the pixelation doesn’t just affect fluids. As in Particle Fleet by the same developers, your own ships take damage by having pixels eaten away. This isn’t the only thing it takes from Particle Fleet, either: some levels feature a similar particulate enemy, and, as in PF, the number of ships of any type you can have at a time is limited, making for smaller-scale battles. The very fact that I refer to your units as “ships” is a symptom of how Particle-Fleet-ish it is; Creeper World is usually about land battles. But the pixel-level simulation is stronger and weirder here: when you move your ships, they move by physically breaking apart into the pixels they were built from, which form a sort of snake-like chain, slithering its way around walls to reach its destination and reform.

But back to the fluids. In addition to Creeper and Anti-Creeper, there are several other fluids found in the environment, as well as substances with “sand physics”, pixels that form heaps when they fall. And some of them are useful: oil, sulfur, pixellium, etc. These can be sucked up and combined into other useful substances, like explosives or acid. And, as in Noita, you combine them by throwing them into a pit together. This is the single thing that makes me certain that Noita was a direct influence, rather than just something that hit on similar ideas independently. The system of alchemy here isn’t nearly as complex as Noita‘s, but I’ve seen player-made levels that extend it with secret combinations and new substances.

In the campaign’s final level, it makes a final turn towards Noita by stopping being a RTS and instead becoming a 2D metroidvania, with a single player character running around a complex, shooting at Creeper, picking up keycards, and mixing chemicals in vats. I feel like this might be a reaction to all the first-person levels made for CW4, a way to get ahead of the inevitable genre shift among the fans, to make it planned and deliberate. But I haven’t seen any player-made levels like it yet.

I’ll say it again: Creeper World 3 is the definitive Creeper World. This game isn’t even trying to be the Next Big Development of the series. It’s the quirky offshoot of the series, an experiment in what else you can do with the basic idea. And I kind of love it for that.

When I started this blog, I posted about every game I played. That hasn’t been the case for a long time, and some games that have eaten large portions of my life have gone completely unremarked on. I did make a solitary post about Train Valley 2 a few years back, but this does not even begin to cover my experiences with it.

To recap a little: In contrast to the original Train Valley, which is a scramble to meet unpredictable demands (a bit like Mini Metro, but with completely different mechanics), Train Valley 2 is about making a plan and then executing that plan. It’s essentially a crafting game, where the crafting is mediated by trains: a city might need, say, dozen Copper Ingots, which are made at that factory over there out of Copper Ore and Coal, each of which is produced at a mine somewhere else on the map using Workers. (Workers are, like everything else, a consumable resource.) You have multiple demands to meet within a time limit ¹That is, there isn’t a time limit per se. You can exceed the par time and still meet all your other goals. But only if you’re willing to settle for fewer than five stars., and thus have to prioritize, sending your trains where they’re most needed.

I think I found this structure most appealing at times when I felt blocked in other areas of my life. If I can’t make progress in my real plans, at least I can make a plan and execute it in Train Valley. The fact that it really does involve making a plan is important, I think. So many modern games tell you outright what you need to do at every moment, but TV2 just delivers a bunch of requirements and constraints and lets you figure out what needs to be done. The result is something that I found extremely compelling, to a perhaps unhealthy degree, downloading and playing user-made levels well past the point of enjoyment. I’ve uninstalled it several times, but then they’d come out with a new DLC pack²One of the DLC packs adds back the random passenger trains from the original Train Valley, letting people mix together the two play styles in a single level. My reaction to this is usually annoyance and a desire to get through the random stuff as quickly as possible so it stops interfering with the execution of my plans. and I’d begin the cycle anew.

Last year, an ad on the TV2 main menu announced the release of a third Train Valley game, Train Valley World. Despite a title containing a word suggestive of MMOs, this game turns out to still be essentially about levels playable by a single player. It does add multiplayer modes, but this is of not much use to me, as I don’t know anyone else interested in these games. It changes up the presentation and feel once more, basically going for something more like Civilization VI: levels are larger still than TV2, the graphics are finer and typically viewed at a greater distance, and, as in Civ, the cities all have names this time, these names being the names of real cities, even though the geography they’re placed in is nothing like reality. The tracks follow the same tile system as always, but the larger scale makes this fade in relevance. Instead of placing a route by dragging over every tile you want it to go through, you usually just click on a series of waypoints.

But the largest change to the feel of the thing is in how you give orders to trains. In the previous games, you send trains on single missions. You’d select a source of freight and a destination to bring it to, and a train would do that once, and then it would be done. If you wanted to send four trainloads of lumber to a sawmill, you’d give that order four times, perhaps using all of your trains in a convoy to send it all the lumber it will ever need at once. In TVW, you program a route into a train, potentially including multiple stops where it drops off one freight and picks up another, and then that train repeats that route until given new orders. Or attempts to, anyway; sometimes it can’t, like if a source of freight is empty, or a destination is full and can’t take any more. The overall gameplay, then, becomes less dominated by goals. It still has goals similar to TV2, with cities having specific demands, but you reach those goals by creating stable, balanced systems than can keep running indefinitely without your attention.

And for whatever reason, I find this a great deal less compelling than the get-it-done-and-then-stop approach of TV2. Perhaps it’s because it suggests endless labor, even if I’m not the one performing that labor. I’ve played through the TVW campaign, but I don’t feel drawn back to it. I’m hoping this breaks my Train Valley habit for good.