Super Mario Bros. Special

“Do You Have What It Takes to Save the Mushroom Princess?”

“You are the greatest player.”
High contrast / quality art by @SarahSSowertty.

No doubt we’re all familiar with the story of the American version of Super Mario Bros. 2 and the ““secret”” that it’s a sprite swap of Doki Doki Panic and blah blah blah The Lost Levels et cetera et cetera et cetera. Look, I apologize, but if you’re genuinely not sure what I’m talking about here? I have to ask that you research it on your own time — just this one time. I made a solemn vow that after hearing this story from a hundred different other people, I’d never inflict it on anyone myself. Besides: We’ve got bigger Cheep Cheeps to fry today.

What if I told you that Super Mario Bros. 2 – in either its American or Japanese formats – weren’t actually the direct sequels to the original Super Mario Bros.? What if I said that Nintendo has more or less managed to successfully bury a Mario platformer game in the depths of obscurity, never to be officially acknowledged in the years since its release? What if I noted that this game marked some of the earliest working relations between Nintendo and Hudson Soft, who would go on to become one of Nintendo’s most trusted third-party developers? And what if I told you that these are all slightly exaggerated points I’m making right now, but that the story behind this particular game is still pretty interesting regardless?

Before Japan even had a chance to take a crack at the Famicom sequel the States were never meant to see, Sharp X1 and NEC PC-8801 computer owners were treated to their very own installment in the series. But with weaker hardware [in terms of games performance] and a third-party developer behind it, would it meet the lofty standards set by the original Super Mario Bros.? Probably not! But hey, we’re gonna play it anyway.** Prepare yourselves, paisanos: It’s time for the Super Mario Bros. Special Super Show!

** And by “it,” I primarily mean the Sharp X1 version, for reasons we will get into later in this article.

“A Host of Black Magic Traps That Only a Koopa King Can Devise.”

Before Hudson Soft became known as “the Mario Party team” and associated somewhat synonymously with Nintendo, they were a developer with a rich variety of different series across different platforms. They were probably best known initially for their multi-platform Bomberman series, and for their role in helping to provide much of the games catalogue for the TurboGrafx-16 / PC-Engine. They brought us such classics as Adventure Island, Bonk’s Adventure, Dungeon Explorer, and Faxanadu… along with such ““classics”” as Keith Courage in Alpha Zones and China Warrior.** What they are maybe less known for (outside of Japan, anyway) are a series of licensed Famicom-to-Japanese-PC conversions of some of Nintendo’s popular first-party titles; primarily targeting the popular NEC PC-88.

PC-88s were an early line of 8-bit Japanese computers, comparable to other home computers of the 80s such as the Commodore 64 and ZX Spectrum. However, at this point in time, different countries typically kept to their own“localized computers, meaning that computers like the C64 were as unlikely to be seen in Japan as the PC-88 was to be seen in the United States. Competition against the PC-88 within the Japanese market would include the the Fujitsu FM-7 and the Sharp X1; both of which Hudson Soft would develop at least some number of games for as well. By my completely uneducated estimation, the PC-88 was the least powerful of that lot, but still topped the sales charts regardless as it was the most affordable product. The first model of Microsoft’s MSX was apparently a distant fourth in this 8-bit race, though the release of the MSX2 would eventually turn fortune back in their favor.

To say that the PC-88 suffered from a “limited” color palette would be an understatement: A maximum of eight colors were the hardware’s lot – with several games only able to utilize four  – rendering many of these Famicom conversions as rather garish-looking affairs. Add to that the lack of smooth scrolling on PCs of that era (it would take id Software and Commander Keen to crack the code in 1990), and you can begin to imagine how several of these games would face difficulty in the move to the new hardware. Titles such as Balloon Fight and Golf would come out the other end more or less intact (if not a little less colorful and slightly slower), while conversions such as Excitebike, Donkey Kong 3, Ice Climbers, and Tennis lost a tad bit more in translation.

In addition to these attempts at straight conversions, Hudson Soft were also granted access to the Mario Bros. license to seemingly do with which they pleased. In 1984, this resulted in two unique variations on the classic arcade game: Mario Bros. Special and Punch Ball Mario Bros. These were released across the PC-88 and Sharp X1, as well as the FM-7.*** Brief aside, but I was genuinely really excited to try these games, as I am a huge fan of the original arcade Mario Bros. and will give any variation on the formula a shot.

Unfortunately, Mario Bros. Special strays maybe a little too far from said formula for my taste. The game takes place across a series of four repeating stages, à la Donkey Kong, each with their own unique layout and objective. The first stage has you jumping through scrolling holes in the girders that comprise the screen, where at the top you must activate a series of five switches in quick succession in order to escape. There are no enemies in this stage, and it doesn’t really feel much like Mario Bros. The second stage comes a little closer, giving you a series of trampoline platforms which you can jump on the tops of in order to flip over enemies currently standing on it (which you must then run into to knock them out). This at least feels like a proper variation on the Mario Bros. formula, but I can’t call it a change for the better. Stage three consists of a series of conveyor belts, moving platform, and an elevator, with the goal of reaching the top and snagging a diamond ring. And finally, the fourth stage is a slight variation on the classic Mario Bros. bonus stage where you must collect as many coins as you can, as well as an additional ring in order to actually collect on those bonus points.

Punch Ball Mario Bros. is actually a more straightforward take on the original arcade game, albeit with one major change. While it gives you the classic goal of clearing out all the enemies in each given phase, it removes the mechanic wherein hitting the platforms out from underneath enemies flips them over. Instead, you are given a ‘Punch Ball’ with which to carry around and toss at enemies, achieving the same flip-over effect. The problem is, you can’t toss the ball very far, and you can’t toss it while moving either (as it’s bound to the same key as jumping). This certainly serves as a “unique” take on Mario Bros., but is it a particularly fun one? Personally, I didn’t get much out of it either. If the punch ball had worked as an additional means of attack alongside the traditional platform-bumping, I feel like it would’ve served the game far better. But for whatever reason, Hudson was absolutely averse to including that mechanic in either of their variations on Mario Bros., thus removing the defining mechanic of the original game across both releases.

On top of that, the jumping and movement just don’t feel very good in either game. Neither game manages to quite recreate the all-important sense of control over the original, missing the “hang time” in mid-air and leaving the movement lacking momentum. Whether that’s a limitation of the hardware or a deliberate design decision to further set the games apart from their source material, I can’t say for sure. All I can say with certainty is, I didn’t much enjoy Mario Bros. Special or Punch Ball Mario Bros., despite my hopes for both. While the 1983 conversion of the game to the NES / Famicom (released in ’86 in the States) isn’t “arcade perfect” either, it’s still leaps and bounds ahead of these two games in terms of fundamentals. Even if it lacks new features and other assorted gimmickry, it’s still a more enjoyable game overall.

Hey, seeing as I might not ever get another chance to talk about it, can I tell all y’all about the best conversion of Mario Bros.? It’s another Japanese exclusive, this time on the Famicom Disk System and released in 1988. It goes by the name Kaettekita Mario Bros., translating roughly to ‘The Return of Mario Bros.’ And boy, what a return to form it is! It improves on the control of the original game – now allowing you to change direction mid-jump – as well as slightly improving the graphics and adding an additional game mode titled ‘Nagatanien World.’ See, Nagatanien is a Japanese food company, who sponsored the development and release of this game. As such, the game is actually littered with intermission advertisements for some of their various foodstuffs, as well as promotion for the then-upcoming Super Mario Bros. 3. The Nagatanien World mode adds a handful more features, including a slot machine to award you extra lives (à la Super Mario Bros. 2 USA) and incentives to send in postcards in order to potentially receive free Mario merchandise. It’s a wacky little package of a game, but it’s honestly just a ton of fun to play.

Getting back on track here: Hudson’s track record for re-imagining Mario wasn’t looking particularly great by this point, was it? Well, with the release of Super Mario Bros. on the NES / Famicom in 1985, Hudson were given one more chance to bring the joy of Mario to the Japanese computers of the time. With the traits and innovations that made Super Mario Bros. an instant classic being so easily definable, Hudson would wisely stay more true to the source material on this attempt. That being said, it wouldn’t be very “Special” if they just converted the original game as was, would it?

** Oh, don’t worry: We’ll be getting to both of those eventually on this site.

*** As long as we’re mentioning the Fujitsu FM-7 here, allow me to clear up the apparently widespread misconception that Super Mario Bros. Special found its way to this hardware. With the FM-7 being discontinued in 1984 and SMB Special being initially released in 1986, it’s unlikely Hudson would’ve bothered re-working the game for a two-years-dead console. Furthermore, the Japanese FM-7 fansite Oh! FM-7 has no entry for the game in its seemingly exhaustive database.

“Search These Characters!”

“We present you a new quest.”
Japanese PC-8801 box art.

Before we get started here, let me explain my decision to base my review on the Sharp X1 version instead of the PC-8801 version. While I had access to and played portions of both, the X1 version features a scrolling transition between screens that makes the game about a dozen times more playable, while the PC-88 simply flashes the screen black and back to the new screen on reaching the transition point — often giving you no time to react to enemies or holes near the edge of the screen. For me personally, this made the game damn near unplayable, and would’ve been reason enough on its own to ditch the PC-88. In addition, the X1 also boasts an expanded 3-bit RGB color palette, which makes the game far easier on the eyes (though it is still clearly outclassed by the Famicom palette).

Another thing that should be addressed straight away: Super Mario Bros. Special is not a straight conversion of Super Mario Bros. on the Famicom. In much the same way that the Japanese Super Mario Bros. 2 (which we’re gonna shorten to “SMB2 Japan” from this point forward) was more or less a standalone expansion on the original game, SMB Special also attempts to approximate the same style of gameplay and re-uses many of the same assets as the original game. It does, however, introduce a handful of “new” enemies and items (more on those later), and every level designed for the game is new as well. And so, I believe it’s fair to refer to it as a full-on sequel to the original Super Mario Bros. Hey, if they got away with calling SMB2 Japan a sequel, I’m gonna apply the same logic to this installment as well.

Also similar to SMB2 Japan, the game is not concerned with introducing a new story to players or abandoning the ‘World [1~8]-[1~4]’ progression structure. You still have a Bowser to battle on every fourth level, a Toad to tell you your Princess is in another castle, and a Toadstool at the end of the proverbial tunnel. You’ll alternate between the traditional varieties of stages which take place above ground, underground, atop mushroom platforms, underwater, and inside of castles. That being said, at least every level is built brand new by Hudson, so you’re not just subjecting yourself to inferior-playing versions of the well-worn Famicom levels. In fact, the levels feel more akin to the gimmick-riddled “Lost Levels” than the more forgiving / straightforward set from Super Mario Bros. classic.

Hudson’s approach to level design does its best to take the lack of screen-scrolling into account, mostly placing gaps in the center of screens so that you should have adequate time and space to jump them. This isn’t always the case, however, with some jumps closer to the edges of screens either catching you off-guard on the left side, or forcing you to take leaps of faith to the right. The screen transitions in the Sharp X1 version make these a little easier to stomach and adjust for, but the PC-88’s abrupt flash-to-blacks make rounding every corner feel like a round of Russian roulette. I really wish I could’ve recorded more footage from the PC-88 version – since the color scheme is absolutely hideous and it would’ve made for funnier-looking animated GIFs – but I honestly could not make it past World 1-4 due to the awful camera mechanic.

Aside from having to take the edges of the screen into account, the stages often present players with some odd platform geometry. Granted, it’s not as if the floating platforms and invisible blocks of the original game made much logical sense, but perhaps it’s just more noticeable in the context of static screens? Also, while I’m fully aware that most pipes in the original game were non-functional [in terms of being able to enter them and access secret zones], it feels like even less are accounted for in SMB Special. Funnily enough, the game actually forces you to manually enter the pipes taking you to the underground at the beginning of those stages, whereas you may recall the original Famicom game implemented a brief automated animation sequence for this. Again, I don’t know whether to chalk these up to programming limitations or deliberate design decisions on the part of Hudson.

What I would definitely contend to be deliberate on the part of Hudson are experiments with one-off gimmicks and easter eggs in individual levels. SMB2 Japan received well-deserved flak for failing to add a substantial variety of new items and enemies, and for generally failing to iterate on the established formula other than simply making the game more challenging. SMB Special flirts with the idea of new enemies and items, only to incorporate them once or twice in single stages across the 32 in total and swiftly forget about them. For example: The hammer from Donkey Kong makes an appearance in this game as a fully-functional weapon, for use again rolling barrels and anthropomorphic flames straight out of the same game. An altogether neat little throwback! The thing is, the hammer only appears twice in the entire game (in World 3-4 and 5-1), and is hidden inside invisible blocks in both cases. Now, that’s a bit weird, innit? Imagine if there had been hammers secretly hidden across more levels in the game, serving as alternative means for taking out chains of enemies. That honestly would’ve been pretty neat.

Also returning are a handful of other enemies plucked from Mario Bros.; including icicles, Sidesteppers, and Fighter Flies. Unfortunately, their appearances are also rare, feeling as if their potential is being wasted. I mean, you’d think that after going through all the trouble of re-drawing them and programming them to match their original behavior as well as possible, Hudson would try and make more use of them across their game? Imagine an entire stage set up like a side-scrolling take on Donkey Kong, with a handful of hammers and a barrage of barrels and flames to contend against. Or how about a stage populated entirely by those returning Mario Bros. enemies, where you’re given multiple opportunities to knock them out from underneath brick platforms? Surely, these ideas must have crossed the developers minds.

Another oddity includes a ‘Wing’ power-up allowing Mario to briefly swim in mid-air, trying to create the illusion of flying. Bear in mind, this predates the P-Wing from SMB3 by a couple of years. Perhaps Nintendo took note of Hudson’s idea and decided to improve on it? Or maybe it was just a logical step forward for this style of platformer. The world may never know. There’s also a ‘Clock’ tucked away in World 8-3 which puts an additional 100 units on the timer, a ‘Lucky Star’ in 4-1 that clears the screen of enemies, and a hidden Hu-Bee (Hudson’s bee mascot) in 1-1 that gives you some additional points. But again, the fact that I can only point to individual stages where these new elements appear is a testament to how under-utilized they are.

I mean, I can’t imagine this being due to some sort of programming limitation, right? The fact that these items and enemies are present in the game at all indicates that they had the functionality in place, and they probably could’ve just as easily repeated them across other stages without eating up any more disk space than any of the other “traditional” items in the game. The only reasons I can think of for their rarity over the course of the game are:

1. Hudson couldn’t figure out what the hell to actually do with the new elements after creating them. This would point to a massive lack of creativity and forward-thinking on their end.
2. Hudson believed that making them rarities and like hidden easter eggs would make them more special, and make their rare appearances that much more of a treat for observant players and “long-time” Mario fans.
3. There were grander ambitions in mind for these gameplay elements — probably similar to something along the lines of the novelty stages I suggested earlier. However, Nintendo may have caught wind of this development, felt threatened by how it might potentially overshadow their upcoming SMB2 Japan, and demanded that Hudson cut / reduce the presence of these elements in order to avoid looking like they were being one-upped by one of their own affiliates. If this was the case, that would certainly point to Nintendo being the dirtiest players in the game-making game, wouldn’t it?

Now that I reckon we’ve determined what the PC hardware was capable of, let’s go over what it wasn’t capable of. For one, I mentioned the lacking color palettes of the PC-88 and X1 earlier, which forced the two different computers to compensate in two different ways: The PC-88 version utilizes dithered coloration** across most of the sprites in the game to create the illusion of different colors not directly available in the palette. The closest approximation of green, for example, is a dark-dithered yellow. What must’ve been particularly frustrating for Hudson was the absence of a pure white in this color palette, having to use a base yellow in its place. Even more frustratingly, there are some particularly hideous issues whenever Mario crosses in front of a dithered background element, where the dithering will appear in front of Mario creating a further darkening effect on his sprites. Add to that further choppiness in movement thanks to weaker hardware, and you’re left with an absolute eyesore of a visual experience.

The Sharp X1 fares slightly better with an expanded palette, though the difference between it and the Famicom is still night and day. You’ll find the game utilizes a lot more vibrant, primary colors. And while the color yellow still sees a ton of use, there are now spots where the color white can shine bright. Far fewer sprites have to utilize dithering, with more colors being outright substituted rather than approximated. Perhaps the most obvious example of this is Mario himself, whose overalls are now yellow to match his new Simpsons-esque skin tone. And when he powers up with a Fire Flower, his red hat and undershirt now turn a garish hue of green, rather than turning his overalls white. Also due to hardware limitations: When Mario’s sprites collide with other objects on the same plane (items, enemies, and hazards), the colors will invert where they intersect. This is a far less distracting anomaly than the dithering on the PC-88, but it’s still noticeable nonetheless.

Other technical limitations directly affect gameplay, such as the aforementioned lack of screen scrolling. There are also some issues with collision detection and hitboxes, which manifest in a number of ways. In example, there is only one springboard object in the game (in World 2-1), and it is broken to the point of being nearly impossible to use reliably. In theory, it’s meant to propel you up in the air, over a wall and past the edge of the screen. In practice, it only dribbles you a few pixels up and down, unless you can manage some incredibly specific / unreliable timing that involves holding down the jump button. Realizing this, the developers also included some invisible blocks in this area which can be used as an alternative to the springboard, and the springboard does not appear to reappear elsewhere in the game. Why it even features once in its broken state is a question I would like to ask of the developers.

But alas, I again cannot say with certainty whether technical limitations are to blame for the most glaring issue with the game: The game just doesn’t feel like Super Mario Bros. Sure, it goes through all the motions, is structured largely the same way, and generally looks the part. But the all-important sense of control over Mario just doesn’t match what’s presented in the original Famicom game. Jumps don’t have the same hang time, the momentum of running is different, and even swimming feels like it requires far more rapid button pressing just to stay level. Naturally, all of these issues are only further multiplied when playing on the PC-88*** instead of the Sharp X1, which leaves Mario controlling stiffer than a Thwomp. The importance of fluid and predictable movement in platformers cannot be understated, and the steps forward that the original Super Mario Bros. made cannot go underappreciated.

Let me break it down like this: Super Mario Bros. Special could have had an identical or an even better set of levels compared to the original game – plus superior graphics somehow – and still fallen short due entirely to the inferior controls and lack of smooth screen scrolling. That’s how essential the controls are to the original game. In that regard, SMB Special is not a proper Super Mario Bros. experience by my estimation, even if it may look the part closely enough. Honestly, it’s kind of surprising Nintendo would be so willing to allow Mario to be misused in such a disappointing way, and so shortly after the release of perhaps the most instantly iconic video game of all time? Again, I can only think of a small handful of reasons for this:

1. Nintendo genuinely thought the game was “good enough” to bear the Super Mario label. This would indicate their own lack of full understanding of what made Super Mario Bros. so successful in the first place, or a lack of care for the reputation of their property at this point.
2. Nintendo initially overestimated what Hudson could achieve with the hardware at hand. When presented with the finished product, and seeing how much work had gone into it, they took pity on Hudson and allowed them to release what they knew full well to be a sub-par product.
3. Nintendo knew exactly what the hardware at hand was capable of, and knowingly gave Hudson an impossible task in trying to recreate their game. They knew the game would look, sound, and play in a fashion easily recognizable to even the most casual consumers as being “worse” than on the Famicom. The intent was simple: Demonstrate to owners of these computers that the Famicom was the superior piece of hardware for gaming, and that the days of their PC-88s and Sharp X1s were numbered. Dirtiest players in the game-making game, bay-beeee!

So, no: SMB Special is not quite an objective improvement on the classic game. It has a few novel ideas which go underutilized, stage design that is challenging for reasons perhaps unintended by the developer, and it can’t quite match the presentation of comparable titles. Oh, I didn’t even mention yet that Luigi is missing, along with the alternating two-player mode. I guess they didn’t wanna just call it “Super Mario Special.” You know what, though? That actually raises a potentially interesting question in my mind: What if Super Mario Bros. Special wasn’t explicitly marketed and presented as a Mario game? What if it had been a standalone platformer game, to be judged by its own merits? By that standard – compared against other computer games of the era – it’s actually quite the achievement for the hardware!

… Well, the Sharp X1 version is, anyway. That PC-88 version is hot garbage no matter how you slice it. But yeah, the controls aren’t really so terrible, per se: They just fail to meet an incredibly lofty standard, and going into SMB Special expecting the same controls as the original Famicom game is a recipe for disappointment. But compare it to something like a Manic Miner or Snokie (which I’m not saying are awful games by any stretch), and it’s certainly leaps and bounds ahead of them technically speaking. If previous paragraphs came across as being particularly harsh on the game, it’s only because the game stands in the shadow of a genuine masterpiece. Yes, the occasional frustrations with regards to the edges of the screen border on the side of inexcusable, but it’s not without its charm and merits.

** “Dithering” is an artistic effect where alternating dots are utilized to create the illusion of different shades and colors. For example: Alternating between standard green and black pixels in a checkerboard pattern makes the greens appear to be darker, almost like mixing paints. These effects are better realized on CRT displays where colors and pixels have a tendency to “bleed” into each other, creating a smoother effect.

*** Before anyone questions me on this: Yes, I manually set the clock speed in the PC-88 emulator to 4MHz instead of the default speed. Playing the game at a higher clock speed makes the game play exponentially faster and faster with every extra unit of megahertz. Many of the videos on YouTube which demonstrate the game being played in emulation fail to adjust the clock speed accordingly, resulting in what looks (and feels) like overly-sped-up gameplay. That said, the game does not control well even at the intended speed.

“Thank You Mario!”

Photo of a South Korean magazine advert, promoting a conversion of the game to SPC-1500 computers.

I don’t believe sales numbers for Super Mario Bros. Special were ever publicly released. Given that Super Mario Bros. on the NES / Famicom sold an approximate total of 40.24 million copies, whereas the combined number of PC-88s / Sharp X1s computers sold is likely dwarfed by that number, I’m gonna say it’s a pretty safe bet that SMB Special failed to outsell its source material. In fact, I’m gonna go as far as to guesstimate that it probably sold less than 100K units; given that the games reported to be the “top sellers” on Japanese computers of the time seemed to peak with lower six-digit figures, and that SMB Special is nowhere to be found on these lists. The fact of the matter was, the Nintendo era was upon us, and it seemed like it was past time to put these old computers out to pasture.

That being said, that didn’t stop the game from seeing one more conversion to the Samsung SPC-1500 hardware — as primarily intended for the South Korean market. Actual screenshots and release info for this version are difficult to track down, but a poster on the Hardcore Gaming 101 forums provided what appears to be a photo of a Korean magazine article advertising the title, apparently released in 1987. From what little else I can gather, the SPC-1500 may have been closer in specs to the X1 than the PC-88, so we can only hope that its version of SMB Special shared more similarities with the Sharp version of the game than NEC’s.

Finally, I’d be remiss not to mention the attempts by modders and ROM hackers to “bring SMB Special to the NES.” There are actually no less than three separate projects dedicated to this goal; all of which do their best to approximate the stages, but none of which can decide on how the game should be presented. Some mods match the color palette of the X1 version of the game, another retains all the original Famicom sprite work, and yet another uses some of the sprites from SMB2 Japan, inexplicably. Unfortunately, none of these releases seem to incorporate any of the new enemies or power-ups, which is kind of half the fun of the original. That being said, having the improved control over Mario and the smooth scrolling throughout the levels makes a major difference as you may expect. Still, if you’re gonna experience SMB Special for yourself, you should do it the right way and play the Sharp X1 version.

So, what should the legacy of SMB Special be? Is it a rightfully forgotten relic, or an obscure curio still worth some degree of respect? It’s certainly a fascinating interpretation of Super Mario Bros., to say the least. But what is perhaps most fascinating about it is the possibility that Nintendo might’ve intended for it to convert Japanese computer owners into Famicom console consumers. The idea of putting out a knowingly inferior version of your flagship release on a competitor’s hardware in order to try and drive sales of your own is just a totally wild idea to me! And the idea that Hudson may have well known they were being hung out to dry and still did the best they could with the hardware at hand is admirable.

Nintendo would like us all to believe that Mario was built by them, and them alone. But the fact of the matter is, it was a concomitant effort between creative minds at Nintendo and a number of other developers who helped to define and give structure to their concepts. They’d maybe like for us to forget that Ikegami Tsushinki had a big hand in making Donkey Kong a smash hit,** or that Hudson helped bring several Mario games to an audience outside of the Famicom. History may be written by the victors, but try as they might, it’s tough to try and wipe away years worth of well-documented physical evidence.

** For more on that fiasco, might I recommend our article on Radar Scope? It goes into some detail on Nintendo’s early forays into the video games industry, their working relations with Ikegami Tsushinki, and the arcade cabinet that would become Donkey Kong.