Welcome to the Month of Balan! For one week in September, Source Gaming finally digs deep into the history of Balan Wonderworld. In our first part (of two), Cart Boy reviewed the game as part of the history of Sonic Team. Here’s our second.
Let’s go back to the Xbox Games Showcase of July 2020. Amid a teaser for the new Fable and Jack Black’s psychedelic singing, Square Enix had a trailer on hand announcing a peculiar project. It was a Japanese game, which wasn’t odd (despite having never made a foothold in Japan in twenty years, Microsoft has been breaking the bank for some time courting the market and its developers). It was an all-ages 3D platformer, which wasn’t odd (that’s the genre of the game Black was singing in). It was by JRPG giant Square, which wasn’t odd (they were also on hand to announce a Xbox port of Dragon Quest XI S: Echoes of an Elusive Age – Definitive Edition, though that one actually was weird). What was odd was all of it together: Microsoft cheering on the Square Enix-produced, hyper theatrical 3D platformer Balan Wonderworld—a game directed by Yuji Naka, no less.
That name carries weight. Naka is the co-creator of Sonic the Hedgehog, and like Stan Lee he tends to overshadow his partners. As SEGA’s golden boy for its first fifteen years, his work as a producer was integral to classics like Sonic the Hedgehog 2, Sonic Adventure 2, Billy Hatcher and the Giant Egg, and, um, several handheld Bleach games (hey, man’s gotta eat). His career slowed after going indie in 2006, as did his list of credits and games; the most interesting was the well regarded Wii version of Rodea the Sky Soldier. But he joined up with Square around 2018 and sold them on going outside their niche—prestige, cutting edge Japanese RPGs increasingly geared to an American audience—with a 3D platformer.
That was itself surprising. Square has a concrete brand by the game industry’s standards. There’s a general idea of what a “Square Enix game” is, no less than what a “Capcom game” or “Nintendo game” or “Sony game” are. You expect graphics that push a console to its limit, grand and cinematic dramas, fights with inventive RPG mechanics. Their stuff that falls outside that purview tends to come from the dozens of companies they’ve purchased over the years and barely care about. Gex, for instance. Family friendly jumping games don’t fit their brand; hell, they haven’t even been a mainstream genre for decades. But the publisher dived into Balan Wonderworld from 2018 onto its release three years later.
With its support, Naka formed Balan Company within Square Enix to make games outside that standard, and he partnered with an independent studio for their first. Arzest Corporation had a portfolio mostly comprising forgettable Nintendo spin-offs (and its predecessor Artoon, the studio behind the atrocious Blinx series, was no better), but it was consistent. It also had a notable president in Naoto Ohshima, one of Sonic’s other creators. Both men struck out from SEGA, but Arzest was far more successful than Naka’s company Prope, which had effectively shut down by this point. This would be the two artists’ first collaboration since the beloved Sonic Adventure—another 3D platformer—in 1998. Balan sold itself on this partnership and overtly harkened back to NiGHTS into Dreams… , their cult classic from 1996. Naka was a tad dramatic; he implied in interviews that this was his last chance to make a platformer and prove to Square it was a genre worth tapping. Given how the genre’s offerings today are mostly either indie nostalgia trips, remasters, or Mario, getting a major studio to jump in could be a godsend.
Unfortunately, Balan Wonderworld not only failed, but failed so spectacularly and dramatically that it earned more notoriety than it might have as a success. It’s helped set a standard of the worst of the worst for the 2020s. It was one step in the self-destruction of one of video games’ most storied artists. But it also does so in a way that’s fascinating. Works like Balan are rare these days, for many reasons, and I truly find them magical because of that. To paraphrase MST3K’s Mike Nelson, this is one of the most ambitiously bad games I’ve ever played.
The game revolves around dreams, music, and performance. There’s this magical, theatrical top hat of a man named Balan who spirits a child away to a magical land of the subconscious. Each of its twelve worlds, all divided into two levels and a boss fight (third levels are unlocked in the postgame, which I didn’t touch), goes inside the mind of one distraught person. A cutscene introducing them plays once you’ve beaten both levels; one showing a dance party and Balan having blithely solved their problem plays after finishing the boss. None of this is interesting, which isn’t necessarily a problem. This genre rarely does well with storytelling. At the same time, there’s so little onto which you can hold. It’s unclear what part you actually played in helping a competitive chess player, traumatized fiancée, or overworked famer. The dreamlike settings allow a lot of grandeur, though the art style—each area is populated by dozens of dancing people and giants, all of whom dance out of tune and vanish as you go near them—runs a bit thick. Salvadore Dali meets Kewpie Dolls.
There is an actual story, though it’s relegated to a tie-in novel. Balan Wonderworld: Maestro of Mystery, Theatre of Wonders gives context to the events and the voiceless, shallow supporting cast. It explains the deal with Lance, the evil Balan who’s relegated to the game’s worst mini-games and its miserable final boss fight. But it’s the characters you help who are apparently given the most attention, which makes that absence in the game frustrating. Without it, it’s hard to not make unflattering comparisons to Psychonauts 2; the singing Jack Black platformer came out five months later, has a similar premise of mental worlds and helping people through trauma, but also has incredible writing and storytelling (and, well, any writing and storytelling). Both games fill their levels with imagery that teaches you about the characters you’re exploring, but one’s far more inventive and satisfying with what it deploys.
Mechanically, Balan Wonderworld is a standard 3D platformer. Your child avatar—not Balan, who only shows up for cutscenes and quick time events that rank amongst the most bizarre things I’ve seen in a modern game—collects golden junk hidden across each level to open new levels and find more golden junk. The MacGuffins du jour are trophies of our mascot. There’s a great imagination to the setting, though it’s inconsistent and unfortunately, that’s not only a matter of aesthetics. Between the stiff camera, poor level design, and a framerate that chugs whenever anything is on screen, you’re left with an odd problem in that the most imaginative and dynamic levels are invariably the worst to actually play. World 10 is set in a cube of MC Escher staircases, but it’s dully lit and wrangling the camera is an exercise in futility. World 2’s ocean floors are at least navigable, even if they are utterly boring. Technical and environmental issues are everywhere; it’s just markedly worse when the game’s ambitious.
What defines the gameplay and continues that unfortunate theme is its main idea, a reliance on costumes and acting. The protagonist’s powers come from a variety of outfits, each of which possesses a power to overcome an obstacle. This is not itself unique, of course. Certainly Mario has used suits as powers for decades, and variations of this are standard fare for 3D platformers like Banjo-Kazooie. It’s the basis of the upcoming Princess Peach: Showtime! What makes this take unique is also what damns the game: that these transformations have not “powers” but exactly one power.
Balan Wonderworld is a “one button” game, meaning that with the exception of the shoulder bumpers (which let you switch between your three on-hand costumes) and pause buttons, every single button does the same thing. For the default character, that’s a boring, weightless jump. That’s true for the costumes, too, but the jumps are slightly different, maybe adding a tornado or Homing Attack or ground pound. Suits based around attacking, like a praying mantis with a boomerang, lack a jump—a jump, the most basic of video game moves. A few do two things, but rarely well: one gives you fire hair so as to set giant wicks aflame with each ungainly leap, one fires laser attacks but only if you aren’t moving, and two do slight variations on letting you swim through rectangular waterways. The least explicable of the lot is “Box Fox,” a suit that lets you jump… but also turns you into an invulnerable, uncontrollable metal box that slides off a cliff every five seconds. Madness, but that at least gives it a notoriety the others lack. This rule also applies to the menus, meaning there’s no dedicated “back” button.
There are eighty costumes, all of whom have one idea that’s never iterated upon even once in a meaningful way. Mostly, they’re keys: a bunny suit that lets you “freeze time” (i.e. it slows down a fast-moving pathway that you’re not allowed to even attempt without the outfit anyway), a conductor getup that summons minor prizes when you stand on special platforms, an asinine gear tutu that slowly turns giant cogs. Their mechanics are painfully uninteresting. Far too many have the same disappointing gimmick as another costume with only slight differences. And often they’re bad gimmicks, like an awful teleporter, hookshot, or teleporter that’s also a hookshot. The best hint at something special, like how the great Frost Fairy lets you bypass obstacles with an impromptu midair ice road paved with each step. Most, though, are miserable. The clown balloon whose only power is slowly building up an explosion that goes off with the tiniest “pop?” That’s the worst by my estimation, and indicative of the problem.
The experience of this game, such as it is, is close to the exact opposite of “emergent gameplay.” In many of the best games, features and mechanics regularly interact. Link fuses a flamethrower to his spear, strikes an enemy, and their wooden weapons burn as they jump in agony. The Dragonborn sticks a bucket on a guard’s head before stealing and sequence breaking in front of them. By firmly separating the already basic powers and forcing you to wait a few agonizing seconds as you jump from one suit to the next, Balan goes out of its way to not let you experiment or build on what you’ve learned. You use one, wait, then another, unless you lose it from a bad fall and go all the way back to collect it again.
Sonic the Hedgehog was also a one button game, of course. Famously so, and there was value there. Sonic moves fast, you want him to move fast, and tertiary mechanics would only slow that down. It also lends itself to an accessible design. But Sonic is based around doing one thing and doing it well. Balan is a collect-a-thon 3D platformer; by design it incorporates a hodgepodge of unrelated ideas. That’s why Mario has so many more actions in 3D than 2D. There’s more space, more room for going off the beaten path, so there needs to be more to do. A gimmick like this could be amazing for a tight puzzler platformer. You could limit players’ options and lead them to make decisions. But not here, and I suspect it’s the main mechanic mostly just to gesture at that classic Sonic energy. Nods to that are everywhere else.
It does feel natural as the health system: take a hit, lose a suit. Except you only get costumes through these floating crystals, you only unlock those with a generic nearby key, and both take minutes to respawn. The first time this happened I assumed it was just a tutorial, but it’s required for every costume, even in the boss fight where the terrain under the key literally explodes (one suit lets you bypass the requirement, which feels like an admission that it’s busy work). You are punished for failure through multi-step backtracking whose ease is dependent on, say, whether you’re allowed to jump. I think the idea is to make losses mean more, to make you attached to the outfits, which isn’t far from the character switching in Donkey Kong Country. But it’s so tiring in practice. This is why defenses by Balan fans that it’s made for kids or casual players don’t work for me, because these systems only exist to punish you unduly. You can summon a closet at checkpoints to pull out suits from other worlds—except you can only wear duplicates you’ve found, effectively adding farming to a genre that’s already castigated for its padding. You also can’t remove a suit other than by just taking damage or replacing it. I came into Chapter 9-1 with only one costume, one that couldn’t jump, and I had to borrow one that could from my supply because the first and only thing you’re allowed to do is jump up a flight of stairs and push a button.
When you aren’t hoping for a jump or going through random sports mini-games, you spend your time in a colorful but joyless hub. There are these creatures you bring from the levels, you can feed them, and they slowly push a wheel to aid in the construction of a clockwork tower. What these Tims are, what they’re building, what feeding them does, and why you want to take part in this goes, as per this game’s style, unsaid. It’s part of an elaborate breeding mechanic that gives you one optional costume. Ostensibly this is more “dream logic,” but it isn’t. It really isn’t. To me, dream logic is Dale Cooper getting smoky, sexy messages in a lodge of red and black wallpaper. It’s Link walking through a nightmare that’s not his own, and a quest to wake up that will kill his new friends. This is simply a refusal to explain basic mechanics in what’s ultimately a fairly simple video game plot, and for something opting for accessibility that’s as big a pitfall as the ones you constantly fall into.
On that note, this otherwise simple game is surprisingly hard at times. The myriad of technical issues futzes up your jumps, gameplay is finicky, and a secret dynamic difficulty goes totally unmentioned. Boss battles can be especially bad, though it’s good they have multiple solutions. Enemies respawn when you reenter a space, by which I mean they can come back if you scoot just a few inches away from the center of a room. One time in the tree world, I ignored them, but the game kept piping in their low key battle music until I came across and defeated another horde. It’s common to lose your only useful costume late into a level and run back to find it. Nothing feels satisfying, no reward well earned, and no discovery exciting. Balan’s Bouts, the forty-eight identical quick time events found in every level, are the worst; you have to do every prompt perfectly to get their trophy, and if you mess up you’re not even given the kindness to opt out of what are ultimately dull, three minute cutscenes. If I only got a “great” instead of “excellent” on one prompt, I’d just wait it out, pull out my phone, and catch up on the news while Balan struggled against a floating rock. The only other thing they give is more useless drops for the Tims.
Ulike many critics I’m actually not opposed to QTEs, but these ones feel like they come from some other universe, one where we learned nothing from the Nineties and Aughts. And that’s what people saw in this game: an instant relic of a bygone age. The difficulty spikes, the bad powers, the padding, the inscrutable collectibles that mean nothing… Games like Balan are part of what caused 3D platformers to die. It follows every stereotype the genre accrued in its peak. By the time you get to the levels that not only include Balan’s Bout, but two or even three versions of it, it’s beyond deflating. I didn’t have a visceral reaction to it the way I’ve had to stuff like Metroid: Other M or Yooka-Laylee; it’s probably not even the worst game of 2021 thanks to the inexplicably pro-incest Daisy Ridley / Willem Dafoe thriller 12 Minutes. But this is astonishingly bad, albeit in a way that’s unique and fascinating. You wonder what, if anything, Naka, Ohshima, et al took from the last twenty years.
Perhaps the biggest kick in the teeth is how derivative Balan Wonderworld is. At every turn, it reads like a ripoff of the Mario brand, particularly the joyous Super Mario Odyssey. The twisting worlds mimic Super Mario Galaxy’s spacescapes, but the parades of dancers don’t make them feel any more lively. The suits follow the classic power-ups and Captures, but they’re painfully required and feel terrible to use. One of the worst is even a blatant lift of a beloved Odyssey enemy, in a level whose music sounds like the Fossil Falls theme on valium. But while the plumber’s adventures are full of life and mystery, this one is insipid. It turns padding and recycled content, an unfortunate but at least avoidable part of Mario’s 2017 escapade, into the main course. And to be fair, the Captures are also shallow transformations that do only one thing, but each feels great, and Mario himself has a robust moveset. Sonic in his heyday fielded unfair accusations of being a Mario clone—a response, certainly, but not a ripoff—so Naka and Ohshima cribbing from their old rival is sad. For what faults the Sonic Adventure games had, they undeniably cut their own path from the trailblazing 3D platformers they followed.
When the game is at its best, which isn’t much, it’s channeling that theatrical energy. Balan has a creepy look (lazy jokes about him being a child abductor were legion in the runup to launch), but in a fun, “kid-friendly darkness” way. It’s one of the only memorable designs. The energetic, jazzier songs are the standouts of an otherwise bland soundtrack. I think the game would have done much better for itself had it leaned on a motif of performance where wearing costumes—cut, combined, and fleshed out—felt like playing a role. Perhaps Naka and Ohshima were attempting to recapture NiGHTS by focusing on mental conflicts? But these nothing characters suck the energy out of the room, as does playing as this lifeless avatar instead of the zany hat monster. As an aloof mentor who seems to do everything anyway Balan becomes a kind of albatross; you resent him for the stupid QTEs and being the model of the collectibles.
These weren’t concerns to players during development. People were excited to see Naka and Ohshima teaming up again for an entry in an underserved genre. Even if the graphics weren’t great, and they really weren’t even from the first trailer, the art direction seemed strong. Plenty of projects look poor at the outset and manage to turn around. Public reaction only soured on January 28, when Square released a demo. It was dire. People hated all of it: the graphics, the costumes, the gameplay. Producer Noriyoshi Fujimoto did eventually announce a standard day one patch to address complaints, one week before the March 26 launch, but he was also open that less than two months wouldn’t be enough time to address every one. The biggest thing they couldn’t fix? Everyone who played, saw, or read about the demo. Perhaps there wasn’t a huge appetite for Balan Wonderworld after all and this wasn’t the main reason for its failure, but the preview was an unbelievable unforced error.
Release was somehow worse. Without that day one patch, everyone who managed to get to the final boss would be hit with a series of flashes and strobe effects that triggered epileptic seizures (on my patched version, three of the last bosses had physically upsetting flashing patterns I had to look away from). The Nintendo Switch version, in theory the most receptive platform, enjoyed even more technical hiccups than its already bad counterparts. It couldn’t crack the sales charts as a high profile, full budget release title by one of the world’s biggest publishers. Despite that, a wave of somewhat suspicious 10/10 fan reviews popped up across Metacritic, which only served to keep the game in people’s minds as a joke. By the time Square finally yanked the demo off digital storefronts, the damage was done. Their charming new IP had become the industry’s newest, easiest, most captivating punching bag. Which is, to be clear, only part of why I’m reviewing it.
Balan’s failures were felt painfully by all. For Square Enix, it was one more bomb in a series over the past few years: this year’s Forspoken, their Marvel flops Avengers and Guardians of the Galaxy (the latter of which is a rarity for this list in having been well reviewed), Babylon’s Fall, the Final Fantasy I reimagining Stranger of Paradise, and that’s just the 2020s. They bore the financial brunt of their ill-fated Marvel collaborations. Free to play spin-offs of their series keep collapsing. The company’s been floundering, shedding studios like fur and throwing all their cooks into the NFT kitchen. They’re still on that with projects like the upcoming and awful-looking Symbiogenesis, even after replacing the president who announced in that press release that Square had to look beyond consumers who “play to have fun” by embracing cryptocurrency. Many of their successes, like the beloved Final Fantasy VII REMAKE, still cost a ton. Of the company’s current problems, Balan at least seems less ruinous, but it was still a blockbuster title. And of course, its failure has most likely killed whatever internal interest there may have been in making more platformers, if any was actually there to begin with.
As for Yuji Naka, he had his own trevails. He left Square a few months later over the staggering failure of Balan… publicly, anyway. He had actually been fired six months before launch according to a lawsuit he filed at the time that would be quietly dismissed or settled. He was more than upfront after announcing this to the public, alleging that the heads of marketing and sound, managing director, Human Resources, and producer—his old friend Naoto Ohshima— forced him out of development for demanding painfully needed and unanswered changes. He also, well, released an old NiGHTS press photo and blacked out Ohshima’s entire body, like a clumsier version of a Stalinist scapegoat being “unpersoned.” Naka’s story shows a game in desperate need of a lifeline whose very leadership, apparently, resented his demands for polish (and also to not use fan-made remixes in official trailers).
But there are cracks in his story corroborated somewhat via a claim by an anonymous staff member to Yahoo News Japan (an admittedly very incomplete, independent summary here). Apparently he was very abusive and demanding; the most shocking claim is that he wanted Arzest to do a month-long delay, and for free. This matches up with behavior of Naka’s from the past few decades, particularly during his time at SEGA; he helped kill Sonic X-treme to stop colleagues from making a 3D Sonic before him, and his plan to fire most of the Geist Force staff led to a mass studio exodus. He’s also straight-up lied about productions in the past, like hailing the classic Sonic Ring mechanic as a miracle of eleventh hour crunch despite it appearing in an early prototype. From the anonymous source and Naka’s claims—again, by his own admission a wide swath of the production turned against him—it’s reasonable to conclude that this likely was a case of a director antagonizing his own production, maybe while making some worthwhile points along the way. Of course, as the publisher Square’s also responsible for this (and him) having gone so far out of control. All of this and his lawsuit became somewhat moot, however, when Naka was arrested and convicted for insider trading. Twice. One of them involved Final Fantasy VII: The First Soldier, yet another failure on Square’s part. You know, I was on a podcast right when that first arrest was announced by English language news.
I’ve had Balan Wonderworld on the brain for years. I was suspicious initially, disliking Sonic that I do, but I do like 3D platformers and was won over by the theatricality. I feel like I was also one of relatively few people to gel with Balan’s vaguely off putting energy. It being a turkey wasn’t good for me or anyone. But how badly it got, and in so many ways, was undeniably mesmerizing. It was a natural subject for “The Forgotten, the Maligned,” which at the time hadn’t gotten an entry for two years (and still wouldn’t for another two). The problem was that every few months a new wrinkle—the firing, the lawsuit, the allegations, the arrest—came up, and I couldn’t do the story diligence without seeing where it went. Now, though, we’re at the end. Even if Naka finds another studio, as he did right after being fired from Square, his time in the industry mainstream has almost certainly passed. I’m striking when the iron’s cold, but still.
His former friend Ohshima, meanwhile, has been working on the upcoming Sonic Superstars. Sonic’s had quite a resurgence as of late, so getting his co-creator to design a new character feels special. In fact, it’s not just him—Arzest is a co-developer! That admittedly doesn’t give me confidence, but I’m honestly just relieved they weren’t shuttered by Balan’s reception. Studios in this industry can get destroyed if they “only” sell a few million copies.
Square, of course, seems to have more troubles with each passing week. The reception to the upcoming Final Fantasy VII REBIRTH and Super Mario RPG remake has been strong, but they keep making high profile failures and diving ever further into the putrid, tanking NFT market. That’s what drove them to sell off things like Tomb Raider. It’s sad to see this behavior from an industry pillar. So while a platformer didn’t fit their brand, this one’s high profile fate kind of did. There’s not much like Balan Wonderworld out there, and that’s not entirely a good thing. Hopefully we can look at this boondoggle as a warning about the dangers of uncontrolled directors, corporate and creative mismanagement, and compromised game design. But I doubt we will, so at least we can end on Invisible Man, the costume that makes you disappear, tricks enemies, and turns precision platforming into a nightmare. A bad idea? Almost certainly! But it’s the kind of bad idea I never get to enjoy in a game these days, and there’s merit in that.
…Then again, Mario Wonder’s got a power that does the same thing.
- SG Choice: The Gimmicky Halloween Tie-In Candies Our Games Deserve - October 31, 2024
- Gun Metal Gaming Chapter 10: Manufacturing Descent - October 30, 2024
- Holism: Echoes of Wisdom and the Tri Rod - October 18, 2024