CLOSE
Original image

Will the Real "Super Mario Bros. 2" Please Stand Up?

Original image

This week, Nintendo released "New" Super Mario Bros. 2. But "Old" Super Mario Bros. 2 isn't what you think it is. Here's the odd story of how Nintendo crammed Mario into places he was never meant to be.

Super Mario Bros. 1 (1985)

To start this saga, we have to go all the way back to the original Super Mario Bros. (or SMB for short). SMB on the Nintendo Entertainment System (NES) defined a generation of side-scrolling arcade games -- its fluid graphics, fun soundtrack, and extensive gameplay gave NES fans a benchmark against which to measure all future games. And because it was bundled with the NES console in the US, tons of people owned copies and played them. (The "B side" of that bundle cartridge, Duck Hunt, didn't get nearly as much play.)


SMB's gameplay was not heavily plot-driven. While there was a general sense that Mario and his brother Luigi had to save Princess Toadstool from the evil dragon/turtle Bowser, it wasn't really explained why Bowser had kidnapped her in the first place, nor what he planned to gain from the crime. Regardless, gameplay was fun, and simple game mechanics (jumping and running) combined with clever level design made the game challenging. After playing a few rounds, I don't think I ever asked myself why I was fighting flying turtles, gobbling mushrooms to grow up, or eating flowers in order to throw fireballs -- the whole situation was sufficiently trippy that a sensible plot wasn't necessary.

Super Mario Bros. 2 (in Japan, 1986)

In Japan, Nintendo released a version of "Super Mario Bros. 2" that was a very clear sequel -- it looked very similar, using the same game engine and the same characters, along with exactly the same plot. This game is now known to US players as "SMB2j," and a version of it was eventually released in the US as "Super Mario Bros.: The Lost Levels." But as a kid with an NES in the States, I never knew about this one. Most US players have never heard of SMB2j, although we probably would have bought and played it if it had been available. So let's dig in.

SMB2j cart

In SMB2j, there is no 2-player mode. You must choose to play either Mario or Luigi from the start. The differences are subtle: Luigi can jump higher and has a little less traction on the ground. In many cases, the game mechanics are slightly different -- jumping off of bouncy objects (such as turtle shells) was tweaked, adding more lift to the jump. (Don't ask me why jumping off of a turtle should give you a boost in the first place -- just eat a mushroom and it will eventually become clear, man.) This extra turtle-lift was necessary because the level designs often used extremely long gaps, requiring a mega-jump to clear them. SMB2j is also much nastier with its Piranha Plants (those flowers, sometimes spitting fireballs, that emerged from pipes) -- in SMB2j they continue rising out of the pipe even if you're standing next to it, making your game life a lot more difficult. And remember the Hammer Bros., those weird helmet-wearing turtles who throw an infinite stream of hammers at you? Well, in SMB2j they advance towards you, making it harder to sidestep them. In general, SMB2j is a wicked-hard version of SMB, with all new levels.

The other major changes are the introduction of wind on some levels (which pushes the player left or right, even while standing -- making platform jumping a huge challenge) and the introduction of poison mushrooms that looked a lot like power-up and 1-up mushrooms. Touching a poison mushroom hurts Mario or Luigi much like touching an enemy, and further reinforced what camp counselors have been telling kids for years: there are good mushrooms and bad mushrooms.

To give you an idea of what SMB2j looked like, here's a roughly nine-minute speedrun. Note how similar it is to SMB, just way, way harder:

So why didn't Nintendo release SMB2j in the US? Nintendo felt it was too difficult for the US market. Rather than alienate casual players, Nintendo delayed releasing an SMB sequel for two years as it worked on Plan B.

Super Mario Bros. 2 (in the US, 1988)

SMB2 boxThe search for a US-friendly SMB sequel made a false start: initially, Nintendo designers developed a prototype vertical-scrolling SMB-style game in which two players cooperatively jumped up a long tunnel of death, rather than the SMB style of running left-to right and avoiding chasms of doom. The prototype was canned early on, for two reasons: the NES hardware couldn't really handle simultaneous two-player mode with all that jumping, and Mario creator Shigeru Miyamoto didn't feel that the vertical-scrolling gameplay was particularly fun (especially in one-player mode, where you couldn't bounce ideas -- and bodies -- off that other player).


So Nintendo turned to Doki Doki Panic, a game that was also designed by Miyamoto and developed by the original Mario team. "Doki doki" is Japanese onomatopoeia for the sound of a heart beating, hence the common translation of the game's Japanese title "Yume K?j?: Doki Doki Panikku" as "Dream Factory: Heart-Pounding Panic." Let's just call it DDP for short.

DDP, while clearly a Miyamoto joint, was a very different game from SMB. While it was a side-scroller like SMB, it eschewed the stomping-on-enemies aspect of SMB, focusing instead on pulling vegetables out of the ground and throwing them at enemies. There was also no "fire flower" equivalent, which had been a favorite part of SMB -- the fire flower had allowed you to become Fire Mario and throw fireballs at enemies. In DDP, no fireballs were allowed -- just vegetables and the occasional fruit. There was no Bowser; instead you fought various animal bosses, ending with the evil frog king Wart, who could only be defeated by force-feeding him a large number of vegetables. The graphics and enemies were vaguely similar to those in SMB, with one extremely notable exception: instead of throwing turtle shells at enemies, in DDP you threw blackface heads and (at least in concept art) African ceremonial masks. Lots of art changes were made when DDP was adapted for the US market, but removing this racially charged sprite was a particularly good choice.

On the bright side, DDP had a fascinating new game mechanic: the ability to choose among a family of four different players, all with different abilities. You could choose to play Imajin (later Mario), a well-rounded character with no special abilities and no special weaknesses. Or you could pick Mama (later Luigi), who could jump high and hover at the top of the jump just a bit. My favorite was Lina (later Princess Toadstool), who could jump and hover for long distances -- great for zipping by a bunch of complex jumps on the ground. Finally, you could be Papa (later Toad), who was really good at throwing things. Allowing the player to choose a character changed gameplay, and meant that each level could be defeated in many different ways (as a Princess Toadstool player, my main tactic was avoidance).

DDP boxAside from the gameplay and art differences, the overall vibe of DDP was super weird -- and a lot of that carried over to SMB2. The game was envisioned as occurring in an Arabian dreamworld, where many characters are masked (DDP even included masks looking suspiciously like the members of the rock group KISS), a lot of action occurs at night, and you finish each stage by stepping into the mouth of a creepy head embedded in a wall. Of course, it can be argued that this is no weirder than the mushrooms/castles/turtles trip of the original SMB. But the overall concept was less clear, partially because of the DDP-to-SMB2 adaptation -- in the original DDP, a family was trying to rescue two children held hostage by Wart. In SMB2, Wart was more of a generic bad guy, who created masked minions using a Dream Machine and generally just messed up the world of Subcon (it's no accident that the game takes place in the Subcon(scious), and at the end -- spoiler alert -- is revealed to be Mario's dream, possibly the result of eating too many turnips just before bed).

Coincidentally, the DDP-based SMB2 came out in the US the same month as Super Mario Bros. 3 came out in Japan. US players didn't get to enjoy SMB3 until two years later, though a sneak peek appeared in the Nintendo-themed movie The Wizard.

Here's a peek at what DDP looked like before it was SMB2-ified:

You can also read up on the differences between DDP and SMB2, including screen-by-screen comparisons.

Let the Circle Be Unbroken: Super Mario USA

After the US release of SMB2 (aka Doki Doki Panic with special sauce), the game was actually released back into the Japanese market as Super Mario USA. And that's how a Japanese game that was too hard (SMB2j) begat a remade Japanese game with Americanized art (SMB2), which finally became a Japanese game again...with an all-American name (SMUSA).

Further Reading

For more on Mario, check out IGN's History of Super Mario Bros.. Also relevant are the super-extensive Wikipedia pages on SMB and SMB2.

You can also buy SMB2j on a cartridge, assuming you have an NES to play it on.

Original image
iStock // Ekaterina Minaeva
technology
arrow
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
May 21, 2017
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
quiz
arrow
Name the Author Based on the Character
May 23, 2017
Original image
SECTIONS
BIG QUESTIONS
BIG QUESTIONS
WEATHER WATCH
BE THE CHANGE
JOB SECRETS
QUIZZES
WORLD WAR 1
SMART SHOPPING
STONES, BONES, & WRECKS
#TBT
THE PRESIDENTS
WORDS
RETROBITUARIES