A few hours in
When we left off Songbringer on the previous page, it was coming across as, while functional, a bit of a confusing mess. I had two personal goals during my play time this week: I wanted to get to the end of Songbringer, or as far as feasibly possible, to see where the story went, and I wanted to see at what point Songbringer narrative and mechanical systems came together in a satisfying way to create an interesting and engaging ludonarrative experience.
Technically I achieved both of those goals.
Good games are easy to review. I just list what I like about a game and then explain why I like those things, bam, done, time for lunch. Bad games are easy to review too, if a little more draining on the mental fortitude. There’s a lot to learn about why well-considered design choices may or may not land as intended. Mediocre games, now, those are challenging, because it’s hard to structure and sustain an entire review when your principal emotional reaction was ‘eh’.
And then there’s Songbringer, which… isn’t so much mediocre as it is confusing. The simplest way I can put it is that Songbringer is a blend of design approaches, part old-school styles and inspirations and part modern sensibilities and technologies. Except there’s no rhyme or reason to when it does what: It’s an old-school game except when it’s more convenient to be a modern game, which lasts until it’s funnier to be an old-school game again. And the fact that I just told you this is ‘the simplest way I can put it’ should tell what writing trouble I’ve been having.
Maybe something more stream-of-consciousness will help. My first major issue with Songbringer, which you’ve probably noticed on the previous page, is that it doesn’t really offer any guidance. At all. It doesn’t tell you what to do or where to go, it barely tells you how to control Roq, it hardly ever teaches you how you can and cannot interact with the world. If Songbringer was an emoji it would be /shrug: You go figure it out, player.
From a modern game design standpoint, this bugs me: I understand not wanting excessive tutorials, but at least do some player on-boarding.
But then Songbringer is mostly designed after the classics. Or rather classic, singular, as its major immediate influence is The Legend of Zelda. The OG Legend of Zelda, NES edition, no subtitles. The similarities are sometimes subtle, and oftentimes less subtle: The starting cave with the unexplained sword. The overworld map with eight dungeons, in each of which you find items and health upgrades. The Quarter Demon Tooth / Full Demon Tooth thing, which is literally a find/replace on ‘Heart Container’ in a Zelda design doc. There’s even a gameplay mechanic where you run around with a lighter, torching random bushes in the hope of finding secret underground treasuries.
So, okay, maybe Songbringer wants to be a new Legend of Zelda. That’s fine, fair enough, that could work. Legend of Zelda didn’t have a whole lot of player guidance from the outset either, getting to discover this whole new world was a major part of the draw of that game. It certainly wasn’t a functional impediment to me playing that game, and it didn’t stop me from completing Songbringer either.
But as little direct player guidance as Legend of Zelda, it did have from the outset a very clear player goal. ‘Collect the pieces of the Triforce to defeat Ganon and save Zelda’. That little bit of steering may not seem like much, but it’s enough to get players interested in striking out and exploring. And once players complete the first dungeon and discover that that’s the way to find the Triforce pieces, the rest of the game sort of structures itself. A complete arc of goal, method, progress, and eventual resolution is set up almost organically, aided by superb game design that all but forces players to get exposed to this.
Songbringer has no comparable arc. I… would honestly be hard-pressed to tell you what the overall goal of Songbringer was, at any given time. Especially at the start, you just… walk around for the sake of it. Narratively, there’s no real reason to get into that cave, or to fight monsters, or to go into a dungeon. Why doesn’t Roq just chill out at the crash site, waiting to get rescued? That’s what the character Smalls does, and they seem to have a better time of it. Mechanically, ludically, I understand why players immediately start moving and exploring, because that’s what you do in these games. But without narrative justification, without the Why, it’s hard to intuit what you’re supposed to be doing. Explore some more squares, I guess?
I understand that not everyone’s likely to consider this a major issue. Even with poor guidance and pacing, the game’s still playable to completion, right? I’ll posit that it’s a weightier problem than you’d maybe think: Especially in the latter parts of the game, I was often bored and confused, pushing through empty areas or running dungeons not because I wanted to or needed to, but because what else was I supposed to be doing? From a player motivation standpoint ‘I’m doing this because I guess I have to’ is something you want to avoid. But fine, fair enough, the guidance issue itself isn’t a gamebreaker, or particularly super damaging to the overall experience. But what is damaging is the related issue: Songbringer‘s lack of guidance dovetails into a lack of structure.
Quick, gaming pop quiz: How many dungeons do you generally run in a Zelda game? And in what order do you run them? Answers after the needless screenshot.
‘Eight’, and ‘whatever order the game presents them in, that’s 1 to 8’, yeah? Later games have done away with the strictness of that structure, but the idea of ‘there’s a number of dungeons you do in a particular order’ is enduring. It generally works out that you either complete dungeons to find items you need to explore more of the world, of completing the dungeon kicks some sort of plot thread forward.
(Side note: Not everyone knows this, but you can do the dungeons in the original Legend of Zelda out of order. I completely missed dungeon… 2 or 3, I think, in my playthrough, all the way until I relented after dungeon 7 and started looking for the hidden entrance.)
I wouldn’t know how to answer this question for Songbringer. There are eight dungeons marked on the map, called ‘levels’, except there’s secretly more like nine or ten? They have a numeric order, but that doesn’t seem to matter too much. After dungeon 1, a wrong turn while following a horn-headed ghost (long story) put me at the entrance to the dungeon I figured would be the next right one: I could get in, it wasn’t locked, that’s what matters, right? It wasn’t until I got stuck on a seemingly-impassable environmental obstacle that I backtracked, got out into the overworld, and noticed on the map that this was dungeon 4.
I needed dungeon 3 to get into dungeon 4, through a circumlocutious process I only partially remember. I don’t think I needed dungeon 2 to get into dungeon 3? Maybe I did. Dungeon 2 is important for the story, though, and you’re probably not gonna get far without the healing items you unlock there. I fully completed dungeon 6 before even setting eyes on dungeon 5. And dungeons 7 and 8…
The thing I’m trying to establish here is that Songbringer plays at being like The Legend of Zelda, but doesn’t incorporate most of the narrative goal-setting and mechanical structure that makes Zelda games work. There’s incredibly little connection between the actions you take as the player, and the when and how of the story proceeding. Why wasn’t the titular Songbringer introduced until I was halfway into dungeon 3? Why wasn’t I given a vague sense of what to until I found the teleport cube? Was I actually supposed to find my crewmates, did the story progress after hitting that marker, or did it have something to do with one of the dungeons I did? Or the map markers I hit? Or the upgrades I chose? I’m genuinely asking because I have no idea. Things just… happened in the Songbringer story, cutscenes and story beats coming out of nowhere, with little to do in-between for me except run around the map and hope I somehow trigger something.
Then, all of a sudden, you receive what appears to be a side mission on the Songbringer, which is in no way clearly-enough marked as a Point of No Return. Do that, and after you’re back on the planet, you are suddenly in the end-game, with a new NPC friend insisting you go to ‘the Tower’. This happens regardless of the dungeons you’ve completed. In fact, if — like me — you decide to go check out the lore and items in dungeons 7 and 8, which aren’t plot-critical but do help in combat, said NPC will snarl at you every step of the way. The game is explicitly yelling at you for doing what it implicitly suggests you should want to do.
These parts, I can explain easier if we go back to that earlier modern game design perspective. Part of Songbringer is, after all, its fairly impressive procedural world generation technology. Remember how I praised A Robot Named Fight earlier for making procedural generation work for Metroidvania games? Songbringer does this for Zelda games. This is the degree to which I will praise this: I completely forgot that the world was procedurally generated until I got into writing this review. It works. And from that perspective, it’s easy to explain some of the odd idiosyncrasies: Of course the dungeons are scattered around weirdly, and of course the hidden access tunnels are much longer than you’d think, and of course the map suggests you can walk here even though you can’t — procedural generation at play.
Two caveats, though. One, there’s one area where the procedural generation is very keenly noticed, and that’s inside the dungeons, which are… basically all the same rooms in different combinations, each with a different-colour background. With two obvious exceptions, most dungeons are in no way unique, or interesting, or related to the items you find in them. Don’t expect Zelda-level hookshot puzzles here. And two, the technology works, but it doesn’t add. Procedural world generation for a Metroidvania roguelike is applicable, because I’ll be seeing a lot of different worlds in short order. Procedural world generation in a Zelda game is something I’m probably not going to notice, ever. No, scratch that — I did notice when looking up walkthroughs for hard-to-find secrets. Because those by definition can’t exist. If I want to find the secret underground lore statue in my world, I’m gonna have to scour every single bush with fire — it’s definitely not in this other player’s [0,0,2].
Similarly, one modern design sensibility that resonates through Songbringer is giving players non-critical preference choices. You see this a little in the dungeons, which are partially optional. But you see this moreso in the fact that almost every item you get from dungeons, you can get somewhere else: I’ve seen bombs, bags, hat upgrades, and the teleport orb in stores — for exorbitant prices — way before getting them for free. This seems confusingly game-breaking, and it probably is — there’s an achievement for completing the entire game without ever picking up the sword, so. But that’s the sort of freedom Songbringer aims for.
And then there’s the upgrade system. You have five items that can be upgrades, and five elemental cubes: Fire, Ice, Lightning, Acid, and Woo Fear Woo. You can combine these in any way you want! So exciting! In theory, that alone nets Songbringer 120 different replays: Do I want Fire Sword, Ice Hat, Lightning Bomb, Acid Jib, Fear Teleport, or Fire Hat, Lightning Jib, Ice Bomb, Fear Sword, Acid Teleport? In practice it hardly matters, because it can’t matter: You still need to be able to use ice to cross water, or acid to break spines, regardless of the upgrades you pick. But it’s nice, I guess.
Real talk: In a game with more structure and better narrative grounding, I would probably praise freeform adventure and upgrade systems like these. They’re extremely my jam. In Songbringer, though, it just… falls flat, I suppose. There’s no narrative structure to give these choices weight, no unique dungeons or interesting enemies to make the tradeoff between different items and elements meaningful. All combat blends together after a while, or at least it did for me: Most of what I remember is mashing sword attacks into anything that got close, dropping bombs and teleporting away, spamming cactus pieces to heal, and sometimes throwing a hat. It got particularly bad in later dungeons, when Songbringer would get into the habit of throwing screenfuls of enemies at me. There’s really no way to meaningfully engage with that: Very literally all you can do is swing away and hope for the best. And then meditate and slowly regain health after that, because why risk losing all that progress?
You might argue that a combat system that rewards thoughtless mashing and waiting to heal, coupled with a functionally meaningless upgrade system, set in optional dungeons in service of an inscrutable narrative, wouldn’t make for a particularly great experience.
Yeah. Me too.
Oh, and one final thing: While I personally love this type of art style, Songbringer‘s graphics are just hard to parse sometimes. I’ve lost count of the number of times I got blindsided by those automated spiky blocks, just because they blend into the background a little bit and there’s generally enough going on on-screen that I can’t afford to scan everything.
I wish I could be kinder to Songbringer. It’s pretty, and it’s ambitious: A one-man project with a three-year running time, it’s impressive that Songbringer is as much as it is to begin with. And, again, the procedural generation tech on display works pretty much as well as you could ask for. But all the same, I can’t escape the conclusion that Songbringer needed a little bit more: Maybe more time, to get all its disparate ideas to work together well, or more focus, to bring it fully in line with either its Legend of Zelda inspiration or its procedural exploration game peers. The game as it stands is technically impressive, and generally a perfectly cromulent Zelda-like if you have a powerful hunger for Game Boy-era Zelda games, but otherwise not something I’d very quickly recommend. If you do think this sort of thing is your jam, the €20 asking price is perfectly fair for the amount of work and polish on display. But do make sure it’s actually your jam. If you go in expecting either jelly or peanut butter — you might be disappointed, that’s all I’m saying.
Jarenth should probably get new glasses, so it’s fair if you choose to ignore the visual fidelity complaints this time around. Suggest frames to him on Twitter or hang out with him on Steam. And if you dig Indie Wonderland and Ninja Blues in general, why not consider supporting our Patreon campaign?