The Stack

Today is the end of the judging period of the 27th annual IF Comp. I have not been posting about it. What’s up with that?

Last year, after playing over a hundred entries, I said that I’d probably skip the Comp if it kept growing at the rate it had been. By September, I even had an alternative in mind, for people who want IF reviews: earlier in the year, the talented and prolific IF author Ryan Veeder announced “club wooby“, a metagame, rewards program, and attention-getting scheme where you earn “buttons” by playing and producing transcripts of his games, and trade them in for Veeder-branded tchotchkes and/or one-of-a-kind handmade felt dinosaur dolls. Although I ignored this at first, dedicating a month to it seemed like a good Comp substitute. I may still do that at some point.

As it turned out, the Comp didn’t grow this year: there were only about 70 entries, a number I would have considered huge a few years ago but which now seems modest and manageable. Aim at playing two a day and you’d easily get through them all within the deadline. So I had a choice to make — or I would have, if I hadn’t suffered a fairly severe wrist injury at the end of September that prevented me from typing or using a mouse with my dominant hand. Playing any form of IF, let alone writing reviews, became too difficult to consider.

I still haven’t fully recovered. Obviously I’m typing now (with both hands, even!), but my capacity for using a mouse right-handed is limited. This has hampered my ability ability to play games, but not eliminated it. I just have to be selective.

One thing I’ve been playing a lot, if you can call it “playing”, is the seminal idle game Cookie Clicker: I’ve made a few goes at it in the past, but its recent Steam release put it back on my radar, and it’s gotten a significant amount of new content since the last time I paid attention to it. The title of the game is a bit misleading: it is not, for the most part, about rapidly clicking on things, and never requires it. Mostly it’s about waiting to afford upgrades to your passive income. It can be played in a more active style, where you’re waiting to click on randomly-appearing “golden cookies” that only last a short time, or it can be played more passively — there are mechanics that reward choosing one play style or the other and sticking to it. And passive mode is basically perfect for satisfying one’s craving for numbers-go-up while other games are unplayable.

I also got back into A Monster’s Expedition, which has the virtue that it can be played perfectly adequately with just the left hand, requiring nothing more than WASD plus Z for undo and R for reset. I had already reached the ending, but I hadn’t found my way to all the optional islands, and in addition, there was an update that added even more islands. This was a little consternating, as I had no way to differentiate the new content from the old-but-unsolved. It’s all just mixed together in the same map. Now, I posted before about how the retreat of the clouds aids completion, showing exactly where the remaining islands are. The new content makes this less of an issue: islands are basically everywhere! I’ve actually found it easier to start over from scratch, carving the cloud cover out of only the immediate vicinity of where I’ve been, as this makes it easier to see which unvisited islands are close enough to bear consideration. Possibly re-solving everything has helped me relearn how the game works, too. Whatever the case, I got severely stuck when just trying to continue from my old save, and have easily outpaced it from the new.

To some extent, I’ve been able to use a controller: unlike keyboards and mice, you essentially operate controllers with your wrists in a fixed pose. In this way, I’ve managed to play most of Teslagrad, a rather good puzzle-oriented Metroidvania themed around magnetism in a Russian-ish steampunky setting. However, as your range of actions increases over the course of the game, eventually it gets to the point where you’re using chords of face buttons and shoulder buttons that would probably be a little awkward even when they’re not actively painful. I want to finish it at some point, but it’ll have to wait until I get my grip strength back.

At any rate, I’m getting better. In fact, over the last two weeks, I managed to do a two-part stream of Return of the Obra Dinn, a game that I had completed before, but which I wanted to solve better. The whole game is about figuring out the grisly fates of the crew of an abandoned Regency-era sailing vessel, using observation and deduction. The game has a way of letting you know you when you have enough information to know someone’s identity, and I wanted to see if I could find the necessary reasoning as soon as the game thought it was possible. (I mostly succeeded, but not entirely.) You can still see the recordings on Twitch if you’re interested. Anyway, this is a first-person game, controlled via mouse and keyboard, and I realized after the first stream that it had been a bad idea: even in a sedate non-action game where you can spend a lot of time standing still and going over your notes, and even using the mouse left-handed where I could, my hand was wrecked by the end. I went ahead with the second stream anyway, hoping that an extra week of healing would make it better, but I’m not doing it again soon.

I’ve found myself devoting more time to A Monster’s Expedition. Not as much as Steam thinks I have, mind you. This is a game without urgency, and that seems to make me willing to alt-tab out and leave it running in the background while I check my email, or watch a movie, or take a nap. But also, I’m still finding passages to new islands frequently enough to make mere wandering around appealing.

One big way the game encourages this behavior: After you’ve won, it inverts its cloud policy. At the start of the game, there’s cloud cover obscuring your view of the world everywhere except for the islands you’ve visited and the sea lanes you’ve passed, and a little margin around that. Sometimes, if an island is close enough, you can barely glimpse its edges through the mist. Sometimes they’re completely obscured until you find them. But after you’ve won the game, the rule changes from “Clouds everywhere, except known islands” to “Clouds only over unknown islands”. The map is now mostly clear, except in spots that, because of the clouds, you know must have islands in them. This gives the game a certain Pac-Man-like appeal as you try to clear the board of every last crumb of cloud, but it also provides guidance. If I think I might be able to build and launch a raft in some spot, I can trace where it would go and decide if it’s going to hit anything of interest or not before I spend time trying.

I’m at the point where most of the remaining puzzles are probably the larger sort, involving multiple islands together. I haven’t been keeping good track of this, but I’m pretty sure that some of the puzzles I’ve solved have required time travel, in the sense of solving an island one way, then resetting it and solving it another way that uses a log or raft or even just an alternate entrance that you used the first solution to obtain. Checking puzzle dependencies on a thing of this size must have been a large undertaking. But the clouds make it all seem just a little bit more approachable.

I’m still in what I believe to be the final stretch. It’s a long stretch, and the puzzles are getting monumentally difficult. Clearing just one per session feels like an accomplishment.

The DROD fandom uses the word “lynchpin” for a puzzle’s crucial insight, the non-obvious realization that enables you to solve the whole thing. A lot of DROD puzzles consist of a lynchpin plus a bunch of tactical maneuvering. I’m finding that a lot of the later puzzles in SSR have multiple lynchpins. You tinker with a puzzle for a while without getting anywhere, and then you realize “Wait, I can push this thing over by this ladder and stick my fork in this gap and lift the whole thing out of the water!” or whatever, and that opens up new possibilities, but it doesn’t solve the puzzle. It just gives you the tools you need to start thinking about the real puzzle. I’ve also seen a puzzle or two that are just outright multi-stage affairs, where you have to get a sausage from its starting position to the grill with a series of unrelated mini-puzzles along the way.

Once again, this stands in contrast to A Monster’s Expedition, where every island is small and elegant. Ah, but AME has larger puzzles that span multiple islands. I suppose the difference in feel has to do with the sense of what the smallest unit of puzzle is. In AME, where boundaries are fluid and every change persists, you can frequently think of the different parts of a compound puzzle as separate puzzles. SSR doesn’t allow that. Puzzles are sharply defined, with discrete conditions for entering the puzzle and leaving it, and if you leave without solving all of it, you haven’t solved any of it.

A Monster’s Expedition reminded me enough of Stephen’s Sausage Roll that I had to pull it out and give it another try, starting over from the beginning and hoping to finish it this time. And playing the one right after the other, it’s striking how vastly different their attitudes are. AME is the “good twin” here, all gentle and welcoming. SSR wants to make you uncomfortable. Suspiciously tinkly music that occasionally goes discordant, noisy and unclear textures, weird movement rules — in AME, pressing an arrow key always just attempts to move you in that direction, but in SSR, what it does depends on your orientation. Even after playing for most of a day, I still spend a lot of my time fruitlessly pivoting in place.

And when you do move, you have to be careful to avoid smacking sausages into the wrong place with your fork, because you take up two tiles and it’s awkward. The easiest way to avoid trouble is to walk backwards most of the time. You can’t just walk up to a ladder and climb it, because you can’t reach it past your fork; you have to back up to it, then turn and climb it sideways. Once you’re at the top of the ladder, the controls make it impossible to turn to face it, because pressing in that direction will just make you climb back down. An inadvertant flaw? No: there are puzzles based around it. I’d complain about the controls a lot more if I didn’t think they were deliberately alienating.

And when you make a mistake, you can explicitly fail. Burning a sausage, or knocking it off the island, results in a big failure message and the necessity of undoing or resetting the puzzle (although I notice that you can keep on playing while the message is up). AME never did that. That is, it’s certainly possible to render a puzzle unsolvable in AME, but it never tells you that you failed.

I’ve reached the end of the expedition, the point where the monster gets on a boat to go home and the credits roll, but I clearly haven’t solved all the puzzles. There are still islands that I can see but not reach. This game will probably last me a while, but unless it does something really unexpected, this is going to be my last post about it.

Overall, I’m really impressed by how pleasant it is, how it goes above and beyond to make sure your experience is a good one, even when you’re stuck and frustrated. Take the music. Lots of puzzle games play soothing ambient music in the background — so much so that I’ve joked about “Abstract Puzzle Game with Soothing Ambient Music” as a distinct genre. A Monster’s Expedition has gentle ambient chords, but then it adds musical sounds on top of it that reflect your actions: a few notes on a piano or guitar when you push a log, a little tap on a cymbal whenever you undo a turn. I half-suspect that there are hints encoded in this, that you get more pleasing notes when you’re on the right track.

The undo functionality is particularly sweet. Normal movement is animated, and can lag slightly behind your keypresses, but undo is instant, even when you’re in the middle of a lengthy raft trip. It took me a while to notice this, because other games have trained me to not expect to be able to undo in the middle of what is effectively a cutscene. But what really impresses me is the reset functionality. Consider the problem of what it means to be able to reset a puzzle in a game built entirely in a single coherent space. Reset applies to just your current island, the extents of which are a little vague, because you can be on a log in the water and still reset. But the islands aren’t self-contained. If you manage to roll a log across a raft onto a different island, what happens then? The answer is that when you reset an island, all logs originating there teleport back to their initial positions, no matter where they are in the world. That’s more state-tracking than I expected.

I suspect that some of the puzzles I haven’t solved yet exploit this. I’ve definitely solved a few puzzles by means of similar meta shenanigans, although I’m not sure if they were necessary or not.

I’m getting into the tougher parts of A Monster’s Expedition, and I’m noticing patterns. Islands are grouped into clusters, separated by long raft rides, each with its own distinct skin for the terrain and trees. And they function on the DROD model. That is, there don’t seem to be any new gameplay elements introduced after you start encountering double-length trees fairly early on, but each island cluster focuses on a particular interaction. You’ll have a group of puzzles where you need to position a log where it’ll stop another log from rolling too far, a group of puzzles where you have to push a log into the water in a place where it doesn’t make a complete bridge so you can climb onto it and push another log from an otherwise inaccessible direction, and so forth. Recognizing these themes can be a great help. Sometimes when I return to an incompletely-solved cluster, I re-solve some of the easier puzzles to remind myself what I should be looking for in the harder ones.

Another useful pattern: parity. When you push a single-length log lengthwise, it doesn’t shift, but rather, tumbles. If it’s positioned east-west, and you push it from the west side, it’ll pivot up and come to rest upright, one tile east of where it was. Once it’s upright, you can choose to push it down from the north or south, changing its direction. This scheme has the consequence that if you never roll a log from the side — and you frequently can’t roll logs without losing them over the edge of the island — then its possible positions and orientations are bound by parity in two directions. In any 2×2 square, you’ll have one tile where it can only be upright, one where it can only be east-west, one where it can only be north-south, and one where it can never be at all. Rolling the log, making it move until it hits an obstacle such as a rock or another log, can break this: if it rolls an odd number of spaces, its parity will be changed in the direction it rolled.

Trees always start off upright, so I find it useful to think of the grid of the world in terms of places where a particular tree can be upright without rolling. This is a pattern you can just perceive, if you try. If you know where a log needs to go, and what orientation it needs to have there, then you can scan to see if it has the right parities, and if not, what obstacles could change that. It takes a good long time to reach the point in the game where it’s at all useful to think in these terms, though. Mostly you can just assume rocks are useful because they’re there.

A Monster’s Expedition (Through Puzzling Exhibitions) is a new sort-of-Sokoban-like puzzle game from the creators of Cosmic Express, Sokobond, and, most particularly, A Good Snowman is Hard to Build, which it resembles in a number of ways, starting with the appearance of the player character. Where Snowman had you pushing around large snowballs in the chambers of a hedge maze, Expedition has you push logs around in a network of tiny islands. (The logs start off as trees, but turn into logs when you give them a shove.) The islands and the pushing of cylindrical objects remind me a fair bit of Stephen’s Sausage Roll (which I should get around to finishing sometime), as do the numerous informational plaques. But Expedition is infinitely gentler than SSR, both aesthetically and in terms of difficulty curve.

One of the key factors in this gentleness is that not every island is a puzzle. The islands all exist in a frankly enormous connected space, impressive when you pan around it in fast travel mode. And some of them are just connections between other islands, ways to control the flow. You get from island to island by pushing logs into the water to make bridges or, sometimes, rafts (about which more later). When you get to an island and all it has is a couple of trees in exactly the right positions to make bridges to other islands, it feels like a respite from puzzling, an easy task as a cushion between the hard tasks.

One thing I’m really liking about it is the elegance of the behaviors. Although the rules are complex enough to produce puzzles, they feel simple because they make sense, visually if not logically. Logs move one space if pushed lengthwise, but roll until they hit an obstacle if pushed from the side. You can stand on a stump and walk onto an adjacent log, but you can’t step onto a log directly unless it’s in the water. If a log is in the water, you can walk along it, but attempting to step off laterally just makes it spin in place. This stuff is very easy to internalize. Similarly, the goals are natural and implicit. You’re not trying to push things to demarcated destinations that arbitrarily make doors open. The way that logs bridge gaps to other islands follows naturally from their behavior.

There’s one moment in particular where I was in awe of the designer’s mastery of their craft. Mild spoilers for the game’s beginning follow.

When two logs fall into the same water space, they spontaneously turn into a raft. This is something you discover early on. You can walk on rafts like logs, but unlike logs, you can step off them in any direction, and this seems like the main thing distinguishing them. You reach an island that has a windmill on it — clearly an important landmark that you want to reach, but the way is blocked by a tree, and you can’t push the tree down because the spot you’d have to stand to push it from is water. With a little cleverness, however, you can use a couple of other trees to make a raft there! And so you stand on the raft, and you push the tree, expecting it to fall, but instead, the force of your push launches the raft, sending it drifting far away into unexplored waters, revealing a larger world. It’s at this point that the game displays the title, effectively telling you that you’ve left the prologue and entered the main part of the game, but more importantly reassuring you that what you just did was right. Even though you were tricked into guessing wrong about what was going to happen, the game isn’t going to punish you. Rather, it’s rewarding you with something better than you expected, because it wants you to keep trying things.

I’ve managed to find my way back to the windmill, but I haven’t finished the game yet. On the basis of precedent (particularly Snowman), I expect that eventually everything will feed into a semi-hidden metapuzzle that makes you re-solve old puzzles in new ways. The thing is, even this follows kind of naturally. Because the goals are implicit, they’re flexible. Sure, most islands are set up so that there’s only one useful thing you can do with the logs. But you inevitably consider other possibilities when you haven’t yet figured out what that one thing is.