Hi! I took a week off because this is my website and I can do that. And I have bad news: I’m told a server move might be happening this week, which might cause some downtime. And then later this month I’ll be gone for two weeks for dinosaur riding lessons. I might find something to queue up during that time, I might not. It’s a “wait and see” thing.
I don’t know if/when I will return to the “daily updates at any cost” schedule. It’s not that important, really; tonight I want to talk about RHEM 4. Specifically, I want to walk you through one of its finest puzzles (so far) in fine detail. This certainly constitutes a spoiler, but I think it’s excusable in this context for three reasons:
1) You probably weren’t going to play RHEM 4 anyway.
2) On the off chance you were to play it, this puzzle only constitutes about 2% of the game.
3) I can describe the puzzle in detail without giving away the solution, and knowing the process won’t ruin your enjoyment of the puzzle if you’re the type of crazyheaded freak who is into games like RHEM.
I want to talk about this puzzle to illustrate a point: folks just aren’t making games like this. If I had my druthers there would be a RHEM-class puzzle game on my plate at all times, and there would be a precarious stack of graph paper meticulously littered with notes forever sitting upon my desk. Alas, circumstances are such that I only am able to really indulge myself in this kind of game once every few years. In fact, I believe the last one was RHEM 3, which I played in 2008.
Are there more games out there like this? If so, I don’t know about them. It’s not like this is the type of game that generates any buzz, and the publisher can’t afford to spring for a marketing blitz. Even asking around for recommendations fails to produce results, because the typical response to “Anyone got a good puzzle/adventure game?” tends to produce responses like Monkey Island or Sam & Max. Which, while fine games, don’t scratch the same itch as does RHEM. It’s like comparing Street Fighter to Mortal Kombat; fundamentally the games are identical, in that you use button input to hit, but the details are pronounced enough that a player can prefer one to the exclusion of the other.
It’s always about those little details, isn’t it? So what are the details of this puzzle?
The first puzzle clue was hidden inside a cabinet locked with an unrelated puzzle. This unrelated puzzle was a pretty minor affair; nothing but light observation (and decent memory, if you don’t take notes) required. Even a casual explorer would be able to get that cabinet open simply by noticing that the design on the locking mechanism is absurdly similar to the design on some framed artwork elsewhere in the game.
That framed artwork is, of course, itself locked up behind a much more devious mechanism, which a casual explorer may not be able to get open. They’re not all gimmes.
So I found the artwork, and opened the cabinet, and was thus rewarded with my first clue: a ten by ten grid of squares. Each of the four corners contained a number; 0 in the top-right and bottom-left, 9 in the top-left and bottom-right. The rest of the grid was empty save for five alphabetical four-letter strings, AAAA through EEEE, strewn throughout as though it were a half-finished crossword puzzle. I thought for sure that I would find an overlay of some kind later in the game which would help me transform the letter strings into numerical codes of some kind.
Not long thereafter I found an elevator with a diagram on the wall, exactly of the sort I was expecting: another ten-by-ten grid, only this time every square was occupied in a very regular way. The top row and left column counted down from 9 to 0, the bottom row and right column up from 0 to 9, and every row and column in between followed suit. By noting the position of my four-letter strings on this new grid I was able to translate each string into a numerical code; for example, AAAA = 0123.
I still hadn’t seen an input device asking for these kinds of codes; there were several devices that seemed to ask for some combination of letters and numbers, including some in the very elevator where I’d seen the grid diagram. Finding a clue in a particular spot in RHEM, however, doesn’t necessarily mean you’ll be applying the data contained in that clue anywhere nearby. What I was expecting was a button-box with five four-digit inputs, labeled A-E, connected to a door or a switch of some kind.
The inputs in the elevator seemed to correlate with the information I had, but not in a very strong way. Each input had two buttons; one changed the attached letter (A-E) and one changed the attached number (0-9). I could, therefore, enter something like “A5” into each of three consoles. If the grids I’d found were related, they would have to somehow reduce five four-digit codes down to a single letter/number pair. And they’d have to do it in such a way that I could extract three such pairs.
That’s where I left the puzzle for a while. Either the clues I had were related to the input consoles and I was still missing the middle piece of the puzzle, or they were unrelated and only grouped together like that to trap me into overthinking their proximity. RHEM employs both of these techniques constantly.
After a little more wandering I managed to find some other neat toys to play with, a few more clues, and lots of input boxes. Eventually I stumbled upon a metal wall which flashed a strange symbol when I moused over it. The symbol was a collection of tiny squares arranged in a snakelike pattern. The squares were contiguous, but they were empty, except for the top-left corner, which contained a 9. When I sketched the pattern of these squares over the super-imposed grid from before, I found that it crossed only one letter in the five strings. Thus was I able to translate the grid into five definitive letter/number pairs; for example, A0, B5, C3, D7, E1.
I got excited and ran back to the input consoles near the elevator. Only upon arriving did I remember there were three consoles, and that each one could only house a single letter and number. I decided I didn’t have enough information to solve the puzzle, but I did have enough to try and gain a better understanding of the consoles themselves. I picked one and started plugging my combination in. Nothing happened, of course, but I did learn that I had been interpreting the console’s feedback incorrectly. I thought I was plugging a letter into the top button and a number into the bottom (i.e.: one pair per console). However, I quickly realized that the top button was changing both letter and number, and was doing so in a predictable way. In fact, the letter button on each console was simply cycling through the five possibilities, remembering the state of whichever number I had previously entered. Thus did I discover that each box could actually hold five letter/number pairs; the box would remember what number I plugged in for A after I pressed the button and moved on to B.
So I plugged my combination into all three consoles. Two of them buzzed at me; one caused the nearby metal panel to open up with a “whoosh!” In fact, it wasn’t until just then that I realized that the pattern of squares I’d seen and the input console I’d just used were on different sides of the same metal panel.
I took stock of the physical construction of the elevator. One side was completely demolished; just a hole looking out over a broken catwalk. The other three sides all had metal panels and button consoles. I had successfully opened one; it stood to reason that I could open the others by checking their backsides and notating two more square patterns. And this turned out to be the case.
However, neither of these square patterns had corners. Without a 0 or a 9 to orient them, I had no way of knowing precisely how they laid atop my number grid. If these new patterns could result in more than one combination the puzzle would quickly dissolve into trial-and-error, so I decided (for the time being) to discount that possibility. That left the problem of finding the one (and only one!) configuration each pattern could sit on the grid — and I had no easy way of figuring that out.
I tried for a long time to solve the problem by eyeballing it, but the seemingly random patterns of squares were too large to imagine as I was staring at the number grid. (Each grid was 100 squares, remember.) What I needed was a physical, see-through copy of the patterns I could slide around my notes page. I cut some squares from an empty corner of my notes, colored the square patterns onto them in dark ink, and held them up to the light. After some positioning the solutions leapt straight off the page. My intuition was right; there was indeed only one way each pattern could go, while overlapping each letter string exactly once. Thus did I end up with two more collections of five letter/number pairs.
I had forgotten to notate which direction I was facing at each metal panel, so I wasn’t certain which code went with which door, but there were only two possibilities. I lucked out and got them my first try, completely opening this section of the maze. And somewhere, during all of this, I found my way down into an underground tunnel where the missing wall of the elevator had fallen. It, too, had a square pattern and a button console.
Bright readers will of course ask: if the crucial clue to opening each door was on the back of that door, what good is opening them? If I could already reach all four metal panels from both sides, what did I gain? The answer to that involves a physical object elsewhere in the maze that can be moved in a such a way that it opens one path while obstructing another. Opening the metal panels allowed me to reach areas I could get to, but could not interact with when that object was in the way. My prize for reaching each of these areas was the ability to move the object into three new configurations, each of which led to a completely new section of the maze. Each of these sections, of course, came with its own collection of diagrams, clues, numbers, buttons, locked doors and input consoles. None of these looked even remotely like anything I already had in my notes.
I’m also betting I haven’t seen the last of this elevator. Through all my explorations I made careful note that it was the sole passage into and out of the area. It’s a cert that I will eventually find a new path that will allow me to get to the bottom of the elevator shaft while the car itself is still situated at the top. There’ll be a hidden hatch or diagram down there that will help crack some distant puzzle wide open.
And that’s the story of one single puzzle in RHEM 4 — and likely not even one of the larger or more devious ones, at that. This is the kind of puzzle I want more of. This is the type of game I wish more people were making, outside of the context of tiny European development studios. And this sense of puzzle design — these intensely satisfying problems that can be attacked with logic and observation alone — they are dreadfully hard to design. So hard, in fact, that a lot of games merely trip and fall in the attempt. RHEM gets it, and no one else does. Myst sort of gets it. I could play the rest of those, I imagine, but I hear the fourth and fifth ones are really wordy.
Kind of like this post. Okay, I’ll stop now.