Songbringer Review

Who’s afraid of procedural generation? If you took anything away from Songbringer’s press releases and factsheets, it probably had to do with freedom. Freedom to explore, fight, and experience a vibrant, unique game-world millions of times over, if you were so inclined. And how you approach the daring conquest of that brand-new world is up to you. Nine dungeons, procedurally generated, that can be completed in any order; a crafting mechanic that allows players to personalize their combat-style; you can even finish the game without taking the much-touted nano-sword which sets the plot in motion. Oh well. It’s a little unfortunate that there’s such a premium placed on freedom in games – and especially the trend that links “freedom” to randomization. Yes, here’s one more lesson on the failings of procedural generation. It’s worth repeating and discussing, though; Songbringer could have been very special.

Songbringer is another Kickstarter-birthed, top-down, self-styled Zelda-like from Wizard Fu Games. You play as Roq, a roving space-musician (which impacts nothing in the game, as far as I can remember, and is hardly ever mentioned) who arrives on a planet called Ekzera aboard the game’s eponymous spacecraft. The protagonist fits readily the archetype of “lovable rogue,” and his motivation, in that tradition, is kind of unclear. He’s decided – after taking a strange, vibrating sword from a cave and awakening an ancient evil – to save the world. Only vaguely convincing, sure, but smooth enough to push you right into investigating. The world he’s dead set on rescuing is generated according to a six-letter seed entered at the beginning of each playthrough. My Ekzera (the seed was BILED, if anyone’s interested) was rainy and swampy. Wet or dry, densely or sparsely populated, brimming with “loot” or not, the planet is always the one-time home of a galactic civilization. There are overgrown ruins, cryo-tubes, and mysterious artifacts which will occasionally dispense cryptic information about Ekzera’s demise. Despite a grim background, the tone is pretty uniformly light-hearted. Roq runs around shirtless, in suspenders and a top-hat, uses the word “awesome” liberally, and cracks jokes at his drone-sidekick, Jib. Our protagonists are evidently meant to feel familiar in an unfamiliar setting. And they do, although I’d be hard-pressed to describe them in any more detail than “intrepid/free-wheeling explorer” (Roq) and “foil” (Jib). Aside from an ill-fitting remark here and a joke-missing-a-punchline there (there was one about a waterfall making Roq thirsty, I remember), Songbringer is competently scripted. If the plotting here isn’t what I’d call engrossing, it’s not what I’d call intrusive either. The story is there, but it’s not what you’re playing. Songbringer’s focus is firmly on fighting and exploring – which are, helpfully and respectively, the things it does mostly poorly and mostly well.

In my experience, procedural generation doesn’t do combat-heavy games too many favors. I imagine that knowing your dungeons will be randomly populated doesn’t leave much motivation to create levels that play to specific enemy and group strengths. It seems to follow, I think, that procedural generation makes designing “tight” combat more difficult, less appealing, and easier to put off. Playing Songbringer, I frequently found myself in rooms full of enemies that couldn’t reach me, because the room layout didn’t allow for them to flank me, and/or none of them could fire projectiles, and/or they weren’t fast enough to hit me between my attacks, and/or they stumbled when struck. On the very first boss fight, I just stood in an entrance hallway (where the enormous, goat-demon couldn’t reach me) and threw my hat (which works like a boomerang) until it died. The next boss, a massive ice dragon assisted by some minions, was the complete opposite. Every enemy in this scenario shoots beams or projectiles that freeze you in place, setting up the net-like paths of fire (or ice, I guess) typical of this sort of game. The problem here isn’t that the fight is hard, or that there’s a huge difficulty spike (which there is, probably attributable to the order of dungeons being variable). The problem is that you’re constantly losing control, waiting for your character to be unfrozen. The fast-paced battles are, if unpolished, almost the only time the game’s combat is fun. Emphasizing fast play and then repeatedly slowing it – sometimes unavoidably – slows the process of improving your playstyle as well and makes it more frustrating. More generally, because it’s not exactly strategic and methodical, Songbringer suffers whenever the flow of combat slows. These are the conspicuous design failures, and they led me to be more aware (and less forgiving) of other flaws, like lackluster pathing or unclear hitboxes, ranges, and feedback.

Despite a wide variety of enemies and a neat collection of weapons (you have your sword, bombs, a boomerang-hat, and a dash move called “blink”), the actual fighting in Songbringer – bosses or otherwise – can be categorized as frustratingly hard, frustratingly easy, or plain, harmless “forgettable.” What I’ll remember about this game, segue, is how it handles exploration. Here are some of the more salient questions you should ask yourself when “implementing exploration”: How does a player do it, what motivates a player to keep doing it, and how are they rewarded for doing it? Hopefully, the following example will show Songbringer’s answers. In one dungeon, I came upon a patch of water that I couldn’t cross, but I needed to cross it in order to get to the boss. I was stuck, so I left – the game does a good job of encouraging you to visit dungeons you can’t yet complete. In another area, I found a crafting ingredient called “ice.” After using it to upgrade my sword so that it could now fire a projectile, I discovered that shooting it over a lake froze the water. Immediately, I went back to the dungeon from earlier and crossed the patch of water.

Not a complicated puzzle and not an obstacle that introduced a fundamentally new kind of gameplay – like lock-picking or hacking mini-games do. It was a puzzle solved through the primary means of player interaction with the world (namely, swinging a sword) being applied in a previously unconventional way. Simple, yeah, but it was discovered spontaneously, had uses outside of combat, and led directly (even in the case of lock-picking minigames there is often a change in perspective or a mechanic-specific screen that separates the player’s direct action from its results) to progress. This is the part of the proverbial wall where Songbringer’s influences stuck. Exploration and discovery work naturally and directly – a more refreshing result of its debt to the “old-school” than, for instance, pixel-art-graphics. As soon as the water froze, I knew how to solve the puzzle, but I also thought of every other pond in the overworld I could now make into a bridge. The desire to explore, then, is motivated largely by the pleasure of discovery itself. Trite, I know, but those realizations are their own rewards.

Discovery should be satisfying in Songbringer because discovery is what the game’s about. It’s not subtle, by any means: hallucinogenic cactuses, mystical water, and meditation – prominent symbols of self-discovery/the middle class – are featured heavily, restoring health, revealing secrets, and activating ancient monuments. They’re also important plot points. Subtle, no, but they are themes well-integrated with gameplay and in ways that follow the logic the game has set up. They also represent, ironically, the more mundane side of discovery. Unlocking secrets, a lot of the time, comes down to meditating in every area until a musical cue plays (a satisfying bit of feedback that is too often left out). Slightly better is when meditating only clues you in to something strange about a location, even if all you have to do is then place a bomb and blow open an unseen passage. It’s worth noting, and maybe lamenting, that the game being built on procedural generation (and the player knowing that it’s procedurally generated) might encourage a more scattershot approach to exploration.

But I suppose “scattershot” is just in Songbringer’s blood. It’s perhaps counter-intuitive that a project where one person had complete creative control (Wizard Fu Games is, in actuality, one Nathanael Weiss, who wrote, designed, programmed, and scored the game) should feel so inconsistent; on the other hand, it makes it more likely that some things will get the short end of the stick or go unnoticed if you have to take care of everything. And anyway, there might be some things you simply don’t know about. Take the music, for instance. It’s completely serviceable, and despite the occasional incongruity with a scene’s tone, rarely gets in the way or stands out. Except for the “rimshot” sound (and a few others, if I’m being honest), which is of unbearably low quality. Any musician worth their salt could have brought that up and easily found an alternative. I’d say that conceit applies to Songbringer’s issues with level and combat design too. More palatable is the mish-mash in the art-department. The direction is somewhere between Ralph Bakshi, a Renaissance LARP, and Hyperlight Drifter. There are tri-corner hats, snake-people, flying motorcycles, imposing monoliths, suspenders, and amorphous blobs bent on murder. The thing is, whatever you call this style (and whether or not it’s your style), it’s hard not to admire the work on display. Although you might find yourself wondering if they couldn’t have toned down the cool lighting and particle effects from time to time. In any event, despite trying my absolute, daggum hardest not to be, I was charmed by the game. That’s not to say its flaws are endearing– they’re definitely not. When Songbringer gets something right, though, it feels pure. Simple, sincere, and enjoyable. Maybe it wouldn’t in a more solid game. But it probably would. What a shame: Songbringer has a lot of love to offer and not enough to love.

Developer: Wizard Fu Games

Publisher: Wizard Fu Games

Platform: Mac, PC, PS4, Xbox One

Release Date: 1st September 2017

Songbringer is a love-letter to older exploration-based adventure games and, like love-letters often do, fixates on one element at the expense of others. It’s kind on the eyes and mostly succeeds when it comes to recreating a pure, satisfying feeling of exploration. However, combat is too sloppily implemented for how much of a player’s time it’ll take up. Millions of different worlds may improve the game’s replayability, but it seems to me that the design attachment to procedural generation may have lowered its initial playability.
  • Exploration is well-executed in terms of design philosophy and in practice
  • Gameplay, themes, and story serve each other and are well-integrated
  • Pretty and fun to look at, with a great deal of attention to visual detail
  • Key aspects of the game, mainly combat, are carelessly implemented
  • Level-design is often poor, even within limits of procedurally generated worlds