Design Design Survey - Zelda Style Games

Hello, GameMaker community! I am currently working on the design for my new game. This is another Zelda-like game, just like my previous title. (Read more here). As I am researching and considering some key elements of this game, I wanted to get some input from other developers on the primary aspects of the design. This is very general: I am interested in any input you have on the topic so that I can get new ideas and refine my project goals. I will start with a few specific questions and my answers. I want to start a conversation regarding the design of Zelda-style games.

What defines a Zelda game? This is a question--perhaps--concerning genre, because I don't think a Zelda game is quite the same as other action-adventure or puzzle games. I am looking for the key design elements that create the experience of a Zelda game. Here is what I have working at the moment (completely my opinion), once I am more confident in this definition I may write a blog post about it.
  • A Zelda game is primarily defined by the way that key items interact with the two main aspects of the gameplay: exploration and puzzle-solving (dungeons). Game progress is tied to the key items (bombs, boomerang, hookshot, etc). (This is not emphasized as much in open-world Zelda games, like BotW/LBW, however any progression that exists is still usually tied to obtaining new items).
  • Key items are required to traverse the explorable space. Progress in exploration is tied to obtaining key items, as new items allow you to explore new areas. This is the same design as is used in Metroidvanias, and so I would say that Zelda has Metroidvania world design, however Zelda games often have a much more linear structure than a Metroidvania like Hollow Knight does.
  • Key items are required to solve puzzles. These puzzles are especially prominent in the dungeons, and are the aspect which somewhat separate Zelda games from Metroidvanias. (In some ways, I might consider calling a Zelda game a puzzle-Metroidvania). One of my friends suggested then that the Metroid Prime series might be like a sci-fi Zelda and I would tend to agree, though I haven’t played those games.

How to design key items? Based on my definition above, it seems apparent that the design of the key items in a Zelda style game is critical. (I could be completely off-base though, let me know!) In regards to key items I have a few questions.
  • What are your favorite Zelda items? We can learn a lot from what is already well designed. My favorite Zelda item is the double-clawshot in TP because it is really fun to use, and has a lot of uses throughout the game. I also like the Dominion Rod (also TP) because of its unique function, but it isn’t utilized much outside of the one area. Some Zelda items are good because they are just fun to use, and others are good because they have good purpose or have well-designed puzzled. (Both is the best if possible!)
  • How can key items work together? This isn’t featured much in Zelda, so let me explain with an example from The Last Librarian. In my game, the bomb and the boomerang each have their own function, but the boomerang is also able to lift and move the bomb across gaps. This leads to more interesting puzzles as you are required to combine your abilities for new functions. Any other ideas about ways to design items that work together like this? I seem to recall Ittle Dew did this, but I haven’t played it myself.

What other elements aid the design? Besides my definition, there are other key elements that are important to get right for a Zelda game, even though I’d suggest they aren’t required for the genre. Here are some of them and my brief thoughts on each.
  • Combat. All Zelda games feature combat. I would suggest that combat in a Zelda game is primarily used as dressing to improve the exploration and puzzle solving, and to provide diversity in the gameplay. Notably, many Zelda enemies (especially bosses) require particular key items to efficiently defeat.
  • Soft progression. While I argue that hard game progression is usually tied to key items, there are other techniques used to encourage particular order of progression, especially in more open-world Zelda games. These include making more challenging areas further into the game, story progression, and by clearly explaining one valid path to the player, while leaving alternatives obscure. (New puzzle mechanics are introduced that must be learned, aside from just new key items).
  • Story? Story is usually not emphasized in Zelda games, but that is not an aspect of the genre. I’d like to see Zelda games focus on story more, and that is what I did myself in The Last Librarian. I argue that the story does improve the experience, but maybe it should just be kept simple?

Anyways, that’s a lot. Please do let me know what you think are key elements to designing a Zelda game. Especially regarding ways you might improve on the Zelda formula.
TLDR: Tell me your ideas about how to design Zelda like games.
When I read the title of the thread, I immediately had about a dozen ideas. You already mentioned 10 or 11 of those. So here's what I have left:

Dungeon Design: Most of the dungeons in Zelda games fit a formula. The first half of the dungeon is about getting to whatever item is stored within the dungeon. The second half is about teaching the player how to use that item both to defeat enemies and access new areas. The final boss of the dungeon is about the player showing that they've learned how to use that item. The tools at the player's disposal are the most important thing, with dungeon/world layout and enemy design reverse engineered to provide players with opportunities to use those tools. A big part of that is the classic Metroidvania thing of looping back to the beginning to access new areas that you couldn't have without one key item.

Items: Most items in Zelda games dating back to the very start have served two functions: combat and exploration. Boomerangs stun enemies but also grab far away items. Arrows kill foes but also light torches if aimed correctly. Hell, even the raft in the original game can give you a strategic advantage in dungeons, letting you stand in areas that enemies can't access. In the Game Boy game Final Fantasy Adventure, which tried as hard to be a Zelda game as possible, you had ice magic that would turn enemies into snowmen. You could then use them to solve pressure plate puzzles. Even your basic weapons had some use regarding exploration and traversal, with the ability to clear paths through forests with axes and cross gaps by wrapping your whip around a post. Some of the worst Zelda items (in my opinion) are the ones that are integral to the dungeon they're in but serve no real purpose beyond this other than being used to reach some optional collectibles. The player should never feel like "Well, I've cleared that dungeon. Never have to use that item again."

Story: You can make an interesting Zelda-like without having story as a main focus (sighs heavily) but I haven't kept as up-to-date on Zelda games as I perhaps should have in part because of the lack of emphasis on story. I'm one of those people that will play the same damn thing over and over if you change the story and keep it interesting. I'm sure that many Zelda titles I missed out on have great moments in terms of story, but the marketing made them look like "Here's Ocarina of Time but with a new gimmick!" My favorite Zelda game (not counting the first one which wins because it was, well, first) is probably A Link to the Past because it took what made the original great and layered an interesting story on top. There was motivation beyond just clearing dungeons for the sake of clearing dungeons. I personally like my games to provide motivation for defeating the end boss that isn't just reaching the credits.

Enemy Design: While enemies should be designed to either counter or encourage the use of certain items/weapons, its annoying as hell when enemies are designed with only one way to defeat them. Say for example you have an item that creates projectiles that ricochet when they hit walls. A good way for the player to learn to use this item is by introducing enemies that only take damage when attacked from behind. The projectiles would be the best way to deal with these enemies, but if the player moves quickly, they could still strike with their sword/other items. It may take longer, but it shouldn't feel impossible. The exception to this that Zelda games use is boss creatures. As I said before, boss creatures are more of a way for the player to show that they understand how to use the new item. Your final boss should do the same thing with as many of the items you've acquired along the way as is feasible. This is where having multiple forms could be a benefit, as it would be frustrating if, for example, Ganon had a single form that required you to constantly switch between arrows, bombs, hookshot, boomerang, etc, etc. But if you only had to use one or two of these before he switched to a new form that required a new combination of items, it could be interesting.

Hope any of this rambling helped. Even if most of it ended up just being what you said but reworded...
Excellent additions, @HeWhoShallNotBeNamed Thanks!

I especially like what you said about enemy design. I think it clarifies well the purpose that combat/enemies serve in Zelda games. I agree about enemies being able to be defeated with other weapons, not just the intended strategy. One cool way this can be done is to introduce those before you have the intended item, then when you get that item it feels more impactful because it allows you to combat enemies that were much more challenging before. Also, I hadn't considered the use of multiple phases of bosses to test different key items. That makes a lot of sense, especially for final bosses. (Zant does this, notably, in Twilight Princess).

In regards to dungeon design, I definitely enjoy the way that obtaining a new item recontextualizes the rest of the dungeon. I definitely think that's core to the way dungeons are made in Zelda. As an additional resource, here is Mark Brown's series called Boss Keys, for those who haven't seen it already: Youtube Playlist. This is a deep dive into the dungeon design in Zelda games (and he has even more on Metroidvanias). He talks about the impact of collecting key items. Also of note is his analysis of "Puzzle-Box" dungeons, which are dungeons that change their shape to modify the explorable space, like the Water Temple in OoT or Stone Tower in MM.


GMC Memer
GMC Elder
To add to @HeWhoShallNotBeNamed 's excellent suggestion about dungeon design: you should go watch Mark "GMTK" Brown's series Boss Keys. At the very least all of the first season (which goes through all Zelda games' dungeon design), preferably all of it. Here's a random episode:

Here's all the stuff that feels "Typically Zelda" off the top of my head...
  • The iconic Link character design, but you probably shouldn't steal that.
  • Some gimmick involving time travel. Link To The Past introduced it with the dark world, Ocarina of Time also had a ruined future world, Wind Waker had the OoT world frozen in time under the sea, Majora's Mask involves resetting a doomed 3-day loop, Oracle of Seasons has you jump around the year to freeze lakes and other shenanigans to make otherwise inaccessible paths accessible... all of those has the gameplay benefit of being able to reuse content with a new context, and if you did some Yooka-Laylee & The Impossible Lair shenanigans with dungeons having two separate forms you could get a cool unique twist on the Zelda formula.
  • Memorable characters. Usually there's a very low amount of NPCs in Zelda, but each of them are unique and memorable (and often with an exaggerated design). You're usually forced to interact with most of the named characters as you progress the story as well.
  • A 50-50 mix of combat and puzzles. Usually the combat involves heavy puzzle elements as well, like luring ghosts close to a light source so you can stun them and make them vulnerable, figuring out what an enemy's weak point is from its design, or using an enemy to press a button.
  • Very little inventory management. Items either are important tools you use to solve puzzles (or in combat), key items that are managed automatically, or collectibles/money/ammo that adds to a meter or counter somewhere. You never need to worry about equipment stats, inventory space, or having to throw away items to not get slowed down by the excess weight.
  • A world centered around a large open field, usually with separate areas for fire, forest, water and desert biomes at different sides. The world is mostly just a means to get to the dungeons, which is the real stars of the game, but designed full of little secrets to reward exploration and free enough that it makes you feel like you're an adventurer exploring uncharted territory.
  • Speaking of uncharted territory, the games usually have the area map - and the compass that points out items and other points of interest - be an unlockable item you need to work pretty hard for, making them rewards when you do get them (rather than the ubisoft "get a new checklist with busywork" approach).
  • A story about an unlikely hero gradually growing into the role and finding the courage within himself, which gameplaywise maps pretty well to how you gradually get strong enough that enemies that once were lethal threats now are pushovers - the earlier games had this in spades with stuff like the magic rod and the ring that halves / quarters damage, BotW revisits it.

As for key items... I feel like a good key item is one that you want to use all the time, not just something you get out when you need to use it to pass an obstacle and then put away again. The grapple hook in Windwaker is one of my favorite examples, it lets you steal items from enemies and it's much more snappy to use than the hookshot in OoT. The Dominion Rod in TP is on the other end of the spectrum, it is only used for one or two puzzles and has no secondary use. (The Spinner would be awesome if it was more practical for general traversal).

To combine items, there's the classic "use a stick to light a torch, then fire an arrow through it to light another torch that's too far away" puzzle. I think designing key items around an element rather than a purpose is the way to go here - if you have items designed to manpulate fire in multiple ways (like arrows and deku sticks) you could base puzzles around fun interactions with fire... utilities like burn things, melt ice, light up dark areas, and then different ways to achieve those utilities: arrows have longer reach but only in straight lines, sticks make you unable to climb ladders or use any other tool, lamps and candles drain MP and have short range. Combine constraints and level gimmicks to make different puzzles using that core element. For instance, you could have a puzzle where you need to get fire across a big watery chasm to burn a thing, and you need to jump across ice platforms that melt if they get heated up, so the trick is to figure out in what order you need to set things on fire by shooting arrows through a fire and across the water so you can get a shot lined up with the final target, trying to jump across with a stick will melt the ice and you'll fall down and extinguish the fire.
Last edited:
To add to [COLOR=rgb(3, 157, 91)][FONT=Open Sans]@HeWhoShallNotBeNamed[/FONT][/COLOR] 's excellent suggestion about dungeon design: you should go watch Mark "GMTK" Brown's series Boss Keys. At the very least all of the first season (which goes through all Zelda games' dungeon design), preferably all of it. Here's a random episode:
Yeah, I linked that too. I definitely recommend that series. It is sort of the foundation I used for the dungeon design in The Last Librarian.

I think designing key items around an element rather than a purpose is the way to go here
I really like this approach. I think that might work quite well. It reminds me of the BotW concept of their chemistry engine, which allows for puzzles kind of like that. I will have to think about this and see how I can use that for my key items.
One more update from me on this thread. I finished a blog post covering this topic on my development blog. Here is the link if anyone is interested: It's much of the same that I wrote at the beginning of this topic, just a lot more polished, and refined from the input provided on this thread. Particularly, I added the enemy design as a third core element of a Zelda-like, as I did not consider it as important before.

Any more input is always welcome. The design process does not end here, though I have enough now I think I can move onto further steps in the meantime.