Zelda Maker Wouldn’t Work; Here’s Why

With Super Mario Maker quickly turning into one of the Wii U’s biggest games, and the popularity of level editors as a whole getting greater and greater, it was only inevitable that people would start asking for a Legend of Zelda version.

But while a lot of people have wondered about such a game, and came up with ‘questionable’ ideas about how a Zelda Maker might work, we’re somewhat sceptical of the idea. And so with attention paid to everything from the gameplay of the series to the different art styles, here’s why a Zelda Maker game just wouldn’t work.

The Art Styles are too different

When it comes to Mario, art styles are… fairly easy to translate across. After all, here’s the blueprint for pretty much every 2D Mario tileset ever made (lack of slopes not withstanding).


Using the SMW graphics to illustrate, you can see that the design has nine tiles; four corner tiles, four wall tiles and one general dirt tile in the middle to fill up blank space. You can then basically translate this to every single Mario game ever made, like as follows:


Above: Lacking NSMBU, since no one ripped its tile graphics yet.

As a result, building levels is pretty easy, as is working out what tiles to use to fill gaps and how to translate levels between art styles. It’s why you can press a single button in Mario Maker to turn your level from a SMB 1 stage to a New Super Mario Bros U one.

This would not work for Zelda. Why? Well, here’s a Zelda 1 tileset:


It’s very stripped down, but it’s already far more complex than a Mario one.

Here on the other hand, is a Link to the Past style tileset:

SNES - The Legend of Zelda A Link to the Past - BG Kit 1

Holy crap that’s complicated to assemble. Your average Joe would struggle to figure out where to even start with this one…

And that’s nothing compared to the Minish Cap or the other GBA games, whose graphics are a royal pain in the backside to use even for experienced fan game or ROM hack developers like myself. That level based on Minish Cap in my SMW ROM hack literally took me around six hours to assemble, without even placing the enemies or blocks. Think you can do better? Well, here’s an example tileset:


So that’s the first art issue. The tiles are complex to assemble, let alone make usable in a simplified editor like Super Mario Maker. But there’s another problem.

Put simply, the tiles don’t run on the same sort of grid. Remember, all Mario tilesets use a simple 9 x 9 block formula. As a result, you can translate the art styles easily, like in Mario Maker.

Zelda games, don’t do this for their graphics. You simply cannot easily translate between Zelda 1, A Link to the Past, Minish Cap and Link Between Worlds, since the graphics run on entirely different grid setups and the amount and placement required is so different between styles. As a result, without a ridiculous amount of complex programming (read, generating tiles from thin air based on what avoids cut off), the graphics just wouldn’t work.

So what’s the alternative here? Come up with four new art styles that all sort of match the official ones but not really yet somehow tile better? That’s going to be way, way too much work for a glorified level editor on a disc. Make it Zelda 2 style? Well, it’s been suggested, but then you get the equally awkward issue that only one Zelda game is mostly in side scrolling 2D, and the art styles for the few 2D sections that do exist in later games are difficult as hell to translate from/to.

It doesn’t work. A Zelda Maker would have to come up with a whole new art style, then offer nothing else.

But graphics aren’t the only issue…

How do you structure a Zelda level anyway?

Because when it comes to Mario, this is an easy, solved problem. Levels are simple blocks of gameplay that don’t require anything from other levels to complete, have no real necessary connection to the ‘storyline’ and only loosely tie together in any thematic way.

As a result, they’re perfect for a level editor. People can make them, people can share them, and they can all be tackled independently of one another through modes like the 100 Mario Challenge.

Zelda does not have this ‘luxury’. Zelda does not have ‘levels’.

It has dungeons, but they’re dependent on things the player has done elsewhere in the game. For example, to tackle the Shadow Temple in Ocarina of Time, you need the Hookshot from Dampe’s grave, the Lens of Truth from the Bottom of the Well, Din’s Fire from seven years ago, the Ocarina of Time from seven years ago, the Nocturne of Shadow that Sheik teaches you in Kakariko Village… the list just goes on and on.

How do you work around this in a level editor?

You can’t require people to beat certain dungeons to play other dungeons, not unless you want to come up with some sort of insanely complicated grouping/tagging system where each fan made dungeon is chosen to introduce an item required by the next one.

So right off the bat, you’ll lose much of what makes a Zelda game a Zelda game, your ‘Zelda Maker’ will become more akin to ‘Four Swords Maker’. But this then brings us to another issue.

Namely, what IS the level you’re supposedly making? What counts as a ‘level’ or ‘main’ gameplay in a Zelda game? Because while the dungeons are cool, some might say it’s actually the big open world you explore that’s the main ‘game’ or ‘structure’ so to speak. But how do you translate that to a level editor?

You can’t just give people a big empty world to explore with nothing in it. So you need NPCs, dungeons, towns, items, enemies, bosses… you name it. But then you’ve gone from a ‘level editor’ to a game maker, with the inherent jump in complexity required from such. Some fans might want to make a forty hour adventure experience with Link, but many people just won’t have the time or inclination to either make or play such a game.

Above: If you want that, there’s always Zelda Classic.

Besides, think Mario Maker is bad enough on ‘Expert’ difficulty? Yeah, good luck beating a forty hour adventure game made by a sociopath with no idea what they’re doing. You may as well go and play Kaizo Mario World at that point.

So perhaps the basic gameplay of the Zelda series isn’t great for a Mario Maker esque experience. But there’s also one last issue the idea has…

Amiibo/Costume Forms aren’t practical

Because the concept just doesn’t translate well to the Zelda game format. It works in Super Mario Bros, because the only sprites a Mario character needs (in a platformer) are running, jumping, flagpole and swimming sprites.

That’s not the case in a Zelda game. In Zelda 1 alone, you need sprites for Link (or whoever) going in every possible direction and using his items. In later games? The amount of necessary sprites starts getting into the hundreds. Look at this:


That is all the poses Link has in A Link to the Past, taken from a sprite sheet by daemoth on the Spriter’s Resource.

That’s not gonna be possible to duplicate for around a hundred Amiibo type characters. Heck, even basing them off the Zelda 1 sprites would require a lot of frames:


Which itself fails to take into consideration that Zelda items change throughout the series. Changing Mario to those costume forms was easy, they’re basically Super Mario with no extra powers. Making sprites for Link forms that can use the sword, Hookshot, bow and arrows, bombs, roc’s feather, etc? Now that’s going to be difficult.

Need more proof? Well, the above Zelda 1 sprite sheet has 22 frames for Link. Add the needed ones for other items, that rises to 30 or 40 frames. Per character (times 100), and you need 3000 to 4000 brand new graphics just to replicate what everyone expects from the Amiibo/costume/mystery mushroom functionality.

Go to more advanced styles (which may easily have 200 or 300 hundred frames per character), and you need around 20,000 to 30,000 new sprites.

That’s a lot of work for a simple feature, isn’t it? And a good example of how a non-platformer doesn’t really lend itself to the Mario Maker formula.

And the other resources…

Yeah, think the necessary sprites and tilesets are overkill?

Just wait til you try and get your head around how many enemy graphics are needed, or how much new music will need to be composed.

The enemy graphics for example, would need much more work, since very few Zelda enemies appear in every single game in the franchise. Mario has Goombas, Koopas, Cheep Cheeps, Bowser, etc in most instalments. Zelda’s own main games sometimes lack common enemies like Octoroks or Stalfos.

So that’s a ton of new sprites and models, since enemies will need to be standardised across generations.

But then, we get to the music. Mario fortunately had this thing called ‘themes’ and ‘levels’, where each level theme had its own song in each instalment. So water levels could use the water songs from each Mario game.

Zelda… doesn’t do this. At least not all the time. Zelda 1 for example, had about three songs:

That’s not going to be enough music for that game style, is it? So off that bat, Nintendo would need to compose an eight bit forest theme, an eight bit water theme, an eight bit cave theme, an eight bit boss theme, an eight bit ice theme, an eight bit desert theme… ad nauseum.

This would likely need to also happen with the other games too, since it’s kind of inconsistent which area themes get their own songs in each title. So Nintendo would need to come up with say, ten level themes, then compose original songs for most of the style variations.

So that’s why a Zelda Maker probably wouldn’t work. It’d be a potentially popular idea, and it seems logical enough on the surface, but the amount of resources and work required would just make the idea too impractical, especially if the budget is meant to be lower than that for a ‘real’ Zelda title like the upcoming Wii U one.

But what do you think? Do you have any ideas for how a Zelda Maker game could solve these issues?


Zelda Maker Mockup by tonicodrea on Reddit


Leave a Reply

Notify of