I recently hit a milestone in Blue Prince, which I suspect will end up being a natural stopping point for me. Even though I suspect I’ll never do everything there is to do in this game, it’s definitely been one of my favourites, and it ended up capturing my interest far more than I initially expected. For anyone who’s not aware, Blue Prince is a puzzle game from last year that has been a major critical (and moderate commercial) success. At a high level, it combines first-person exploration of a mansion involving random tile placement that resets in a rogue-like fashion, with many mysteries and puzzles to be discovered, against the background of a larger narrative.
Why am I mentioning this here? I think there are a few things about this game specifically, that make for some interesting points of discussion.
The first fascinating thing is the story of the game’s creation. After devoting himself to making movies in LA, including working as a cinematographer for many years, the games creator, Tonda Ros, decided to try a first experiment in game design.1 After making a demo of a game (apparently a quite complete prototype of what would eventually become Blue Prince), he decided to quit taking on paid work so that he could focus full time on making his game for real.2
Unsurprisingly, things did not go completely as planned, and Ros ended up spending eight years making the game, basically without working on anything else during that time. According to what he says in interviews, working on the game was basically his entire life during that time. Apparently his routine was to get up, get to work immediately, and basically work through until he was exhausted, excepting food breaks. That would often be thirteen or fourteen hours a day, leaving just enough time to watch a movie or a play a game before going to bed, and then doing it all the next day.3 By his own admission, Ros basically did not socialize at all during the years of the game’s development (although he does mention seeing friends for extended retreats at various times of the year).
There is a lot here, in terms of the willingness to commit to a vision, the ability to work with such single-minded determination, and the fact that someone could make a game like this as essentially a first real attempt, without any prior experience, all of which I find tremendously impressive and fascinating. It’s not totally clear to me how many people worked on the game altogether, but it definitely sounds like Ros provided the guiding vision, and was the sole contributor for large parts of the game’s development.
Ros’s influences have been widely discussed, and indeed he allusively cites some of them in the credits, including Martin Gardner and Douglas Hofstadter. Having played the game, these make a ton of sense. Although there are some math and logic puzzles in the game, more of them involve word play, riddles, and interpretation.4 In addition, as many have pointed out, a key inspiration is the 1985 puzzle book, Maze, by Christopher Manson, which I remember having and loving as a kid, although I of course never fully solved it.5
The second remarkable thing about Blue Prince is how much depth there is, despite some seemingly superficial mechanics. The primary way one engages with the game is by drafting the rooms, each of which is (minimally) just a matter of choosing between three options. But of course there are many things to do in those rooms, including interacting with various elements, collecting resources, creating inventions, and unlocking permanent changes that will carry over to the next day.
Part of the depth comes from the layers of story, which only gradually reveal themselves. But part of it comes from the mechanics themselves, which in turn support a huge range of possible variations. Without any major spoilers, if the first part of the game is figuring out what kind of thing one is dealing with, and how to operate it, a later part of the game becomes about finely optimizing that thing to achieve particular outcomes, and yet more modes encourage one to perfect one’s mastery and understanding of those mechanics.
The biggest shock for me is that it all basically works, despite being so layered. This did stop being true beyond some point, but the extent to which information gradually and progressively begins to fall into place happened so naturally for me, that it almost never felt that things were unfair or impossible, at least not until very far into the game.6 Although I certainly got stumped on certain puzzles for a while, for most of my playthrough, there was almost never a run where I didn’t make some kind of progress, whether solving a puzzle, or finding a new clue, or putting things together, or simply being able to test a theory. Unlike some similar games, Blue Prince does not do a lot of information management for you, and so it’s basically essential to be taking notes externally, as you gradually piece together and develop theories about what matters.
Admittedly, towards the very end, the puzzles did become too hard for me to solve alone, at least with any kind of enthusiasm. The clues are certainly there, but when it gets to the point where it’s not even clear what the remaining mysteries might be, or where to look for them, it is much harder to make progress, which feels less satisfying. However, the beautiful thing about this kind of game is that it seems extremely well optimized for collective effort, while still preserving the option for true lone puzzlers to go as far as they want.
Ros has said that all puzzles in the game were designed to be solved by a single individual, without community help, although I have a hard time believing that very many people would actually be able to put together everything on their own. In any event, both the fact that a community exists around the game, and the fact that the puzzles are so gradually layered, with many different ways in, means that the whole thing is much more accessible (and also much more addictive) than something like Maze. The latter always seemed intriguing, but it was very hard to know if you were actually making any real progress. Collective engagement with puzzles obviously pre-existed online communities, but they have clearly made it much easier for something like a puzzle or a game to give rise to a whole cottage industry of wikis, guides, streamers, data miners, and so forth.
Having played this game slowly over the course of a year, it is shocking to me how quickly the community around it basically seems to have figured more or less everything out. The game officially released on April 10, 2025. By April 15th, GameSpot had a guide to solving what is generally assumed to be the game’s final puzzle.7 It’s certainly possible they could have had advanced access, but that would imply a remarkable solo effort. There was also previously a demo available since June 2024, which did allow for getting pretty far into the game, and would have allowed the community a big head start, but still!
Given the excitement lately around the capacity of AI to create social simulations or interactive worlds, (as well as to produce charting singles in music), it’s interesting to speculate on what would be required for whole-cloth generation of a game as compelling as this one. Part of the magic (at least for me) comes from how gradually things are discovered, such that there is always something to follow up on, but never to the point of being overwhelming. Obviously that is partly the result of extensive playtesting, and an AI-generated equivalent would presumably also need to go through many iterations. Still, the number of interlocking pieces creates the need for everything to fit, and it’s somehow harder to imagine that not coming from a deliberate authorial effort.
Similarly, one could certainly quibble with how satisfying, coherent, or original the story here is, but there’s no denying it at least acts as a compelling MacGuffin to help drive the player forward. Sometimes theme, mechanics, and mystery all simply come together in a perfect package that is hard to anticipate, especially when that something is a creator’s first work in a medium!
In any event, much like any other artistic medium, part of the joy with games is in recognizing influences, references, and homages.8 As mentioned, there isn’t much in this game that necessarily made me think of Hofstadter in particular (other than perhaps a certain room with a bit of an Escher feel), but it was not at all surprising to learn that the creator was a fan. It similarly made total sense to me when Ros talked in one of his interviews about the challenge of translating this game into other languages, and likened it to the challenge of translating poetry. This is another nice nod to Hofstadter, who not only wrote an entire book about translation, but who also has collaborated extensively on translations of his similarly incredibly-hard-to-translate book, Gödel, Escher, Bach: an Eternal Golden Braid.9
The one sad thing with puzzle games is that, compared to some other genres, it’s harder to begin again, and less rewarding to dive back into that world after some remove. As with one of my other favourite games of all time, Outer Wilds, I dearly wish there was some way to selectively wipe all memory of the game from my mind, to allow me to discover it anew. Thankfully, a close second-best is that remarkable individuals like Ros are willing to go to obsessive lengths to create the occasional masterpiece. I can’t wait to see what he comes up with next.
The information in this post comes from my own experience playing the game and from interviews with Ros, especially his appearance on Post Games. ↩︎
As a backstop that further enabled that, Ros had also created the website Mythic Spoiler, which provided a supplementary income during this time. If I understand correctly, the Mythic Spoiler website just displays card images for all cards in Magic: the Gathering. This seems like a hilariously clear copyright violation, but I guess he’s been able to get away with it! ↩︎
Perhaps the most remarkable part of Ros’ story (to me), is that he apparently does a thing that I once experimented with in grad school, which is to allow oneself to break out of the constraints of a 24 hour day. In Ros’ case, he apparently lives on something more like a 25 hour cycle, gradually shifting his sleep by an hour or so a day, and eventually coming full circle. As someone who also tends to always want to stay up a little later, I highly recommend giving it a try, if your schedule allows such a thing! ↩︎
As an easy example, the game’s title is a homonym for blueprints, which refers both to the plan of the mansion, and a mechanism in the game. ↩︎
The original Maze book is now out of print, and copies are selling for hundreds of dollars, thanks to Blue Prince. I’m not sure about the intellectual property situation here, but someone has made an online version of Maze that is free for anyone to explore. ↩︎
One challenge right on the threshold of too hard for me, and around the time I started taking occasional hints, was the Gallery, which involves four similar puzzles. I basically felt like I had no idea how to approach them, and had not made any progress on it after many attempts. Amazingly, all it took was for someone to tell me which of the four was easiest. By focusing on that one to start, I was able to figure it out, and the others immediately fell into place. ↩︎
The Wayback Machine confirms that this was not just a placeholder page, but that the guide was actually online then, at least as of April 16th. Ros has said he will never reveal whether everything has been figured out or fully understood, but it seems plausible that data miners have correctly identified the lack of any content that has not already been discovered. ↩︎
Although I don’t think there is necessarily a direct connection, it was a similarly delightful and eerie experience to have read Piranesi, by Susanna Clarke, during the time I was playing Blue Prince. With the right mindset, the parallels seem almost uncanny, with both featuring an extensive, shape-shifting, sometimes flooded mansion, populated by a handful of other characters who are not necessarily what they seem. While reading, I was thinking this was probably just a coincidence, until I came across a mention of the “AQUARIUM” (in all caps), immediately in the context of a description of several colours… ↩︎
Although it’s outside of the game, it’s also worth highlighting a video from nine years ago on Ros’ Vimeo channel, which does a beautiful job illustrating the delightful concept of ambigrams, also invented by Hofstadter. ↩︎