Dear Zelda Diary,
I got all the shrines in Tears of the Kingdom. It turns out this is the best puzzle in the game. I recommend the task to anyone who has the game and the time to spare—and I recommend it over reading the spoilers below. Do all the shrines. It’s worth it.
I forget how hard it was to find all the shrines in Breath of the Wild. I assume I had to look up the locations of the last few. But I barely found any cause to look for outside help for Tears of the Kingdom.
The attentive player eventually learns that the area called the “Depths” mirrors the surface world in more ways than the game initially lets on. Maybe first you realize that each shrine on the surface corresponds to a lightroot at the same XY coordinates in the Depths. The game teaches you fairly directly that wherever there’s a town on the surface, there’s an abandoned mining operation in the Depths. After you find one or two treasure maps, you start to notice that the best treasures of the Depths are directly beneath major landmarks of the surface. Maybe a little later on, you realize that wherever there’s deep water on the surface, there’s an impassable barrier in the Depths. Or you may figure these things out in some other order. But I think many players will realize last of all that elevations are inverted in the Depths: Where there’s a deep valley on the surface, there’s a mountain range in the Depths. The tall, narrow pillar of Akkala Citadel on the surface corresponds to a deep, narrow pit in the Depths.
As you accumulate this knowledge, you solve the puzzle of how to navigate the Depths, which mostly lack the visibility and clear landmarks of the surface. Without activating nearby lightroots, your map of the Depths is useless—but you can still find your way around the underground by consulting your map of the overworld. When you figure this out, you feel like a genius. And it becomes easier to track down lightroots, which obviously contribute to your Depths map and the surrounding visibility.
After I beat the game, I saw a spoiler of the reward for finishing all 152 shrines. I wish I hadn’t seen it. I wish someone had just told me, “Do all the shrines. It’s worth it,” the way wise people will tell you not to find all the Koroks. (I also was spoiled on the fact that there are 152 shrines, but that turned out to be indispensible information—the game really should tell you this after you beat it. It is against my better judgment that I spoiler tag it here.)
As soon as I saw that spoiler, I started working on finding all the shrines. I had skipped a bunch of them, only registering them as fast travel points before running off to continue more urgent quests. When I finished those, I had logged roughly 75 shrines. About half. I said to myself: Here comes the hard part.
I pulled up the map of the Depths and checked all the lightroots I had activated to see if I had found their corresponding shrines. There were also a few lightroots I had marked from afar, but hadn’t personally visited; I was able to mark their corresponding locations aboveground as shrines as well, and hunt those down. It wasn’t frustrating or tedious; I was just checking off a list of navigational tasks that I didn’t know I had been accruing during the main game.
I want to say “I logged another 30 shrines this way, and then I tried a new strategy” or something like that, but really the stages of this process overlapped quite a bit. I found several shrine locations that needed to be uncovered with puzzles I didn’t feel like solving just then, or they were underground and I couldn’t find a cave entrance right away. We all know the great thing about these games is that there’s always something else to do.
But eventually I started running out of shrines whose locations I could deduce from map data I’d already gathered. I said to myself: Here comes the hard part.
This was when I started really leveraging my understanding of the Depths. Activating a lightroot fills in the map and illuminates the surrounding area up to a distance where it would be overlapping the “zone of darkness” around another lightroot. Put another way, when you see the map area revealed by a lightroot, you know there are no other lightroots in that area. That means there are no more shrines in the area above!
This answers/solves/dovetails with a gameplay problem/question/issue/technique from Breath of the Wild, something that arises in many similar games: When you’re searching for shrines, one of your most valuable tools is your metagamey knowledge that the shrines should be spread out to some degree over the map. There shouldn’t be too big an area with no shrines in it, and no two shrines should be too close together. The reasoning is intuitive, but not exactly reliable, since no game is really obligated to obey these unspoken laws of design.
But the way lightroots behave brings this notion of “distribution of interestingness” to the surface (!!!so to speak!!!) and let me hunt for shrines in a way that made me feel incredibly efficient and smart. Now that my map of the Depths was mostly filled in, each remaining area of darkness stuck out as a lead on a new lightroot. When I found it, not only did I have the coordinates for its corresponding shrine, but I got to wipe away more of the darkness and possibly reveal more probable lightroot locations. Consulting my map of the surface was the best way to navigate the Depths, and now a full map of the Depths was the best way to locate shrines on the surface.
I got to the point where I had only eight or nine shrines left. I said to myself: Here comes the hard part. Then I identified some likely-looking blank spots on my map of the Depths, found their lightroots, and found their shrines. It wasn’t hard. It was really satisfying.
Now I was missing three shrines. I had lit up almost all the lightroots, and all remaining areas of darkness in the Depths were accounted for by shrines I’ve already found on the surface. If it wasn’t a matter of something evading my notice in the map screen, I had to be missing a few shrines up in the sky. This would probably be the hard part.
Well, I applied the “distribution of interestingness” principle to the sky, and found two more shrines without too much trouble. One shrine left.
This was in fact the hard part. I couldn’t see anywhere in the sky that a shrine could be. I couldn’t see any lightroot I hadn’t matched to a shrine on the surface.
It was at this point that I got kind of desperate. I started chasing down leads that I really knew had nothing to do with shrines. I started activating all the lightroots, just to see if I’d somehow reveal a new one, anomalously close to a known location. When I activated the last one, the game gave me a very stupid little prize, as if to confirm that I had wasted my time.
I should have been paying closer attention. I should have sat back and thought about it harder. When the game told me all the lightroots were activated, I was able to see that the total number of lightroots (and thus the number of shrines on the surface) was 120. The game probably should have given me that information more directly somewhere earlier on, but I also should have puzzled it out for myself without actually visiting all the lightroots.
That meant that there must be 32 shrines in the sky. This was based on my knowledge from outside the game that there were 152 shrines in all, which again is something the game should have told me at some point. Knowing the total number of shrines was absolutely vital to how much fun I had with this whole quest.
I counted the shrines I had found in the sky: 32. I counted them again to be sure. I must have missed one on the surface.
I checked my surface map against my Depths map, as I had so many times during this quest. If I had been a little more careful and observant, I would have seen it days ago: The one lightroot without a corresponding shrine. I got it.
I solved the big puzzle! The very last step wasn’t a transcendent moment of catharsis, but the overall process was incredibly satisfying.
And it taught me a lesson about design. When I had those first few realizations about the Depths, I wasn’t excited about it. It seemed a bit disappointing—uncreative, “unrealistic”—for the underworld to match the surface in so many ways. But soon I noticed that understanding these correspondences was really enriching my experience. Like the intersecting entries of a crossword, everything I knew about one map was a clue to something on the other map. And it gave me a feeling of proficiency with the world of Hyrule that was extremely, sorry I don’t have a better word, satisfying.
I’ve often tried to avoid that sort of matchiness/symmetry/patterning in things I’ve designed, because I worry the audience will see through it and find out how “lazy” I’m being. But I should have known that being lazy would never let me down.