Interview:GDC March 24th 2004

From Zelda Dungeon Wiki
Jump to navigation Jump to search
Want an adless experience? Log in or Create an account.
This interview does not yet have standard formatting or is otherwise incomplete. It should follow the format established in other interviews.


Good Afternoon. My name is Eiji Aonuma and I work in the software production department of Entertainment Analysis and Development, or EAD, at Nintendo Company Ltd. For roughly eight years, I've been doing work related to a game known as Zelda. But, I never imagined an opportunity for me to talk to you about this experience would ever come. I was very surprised when the opportunity arose and at the same time I was incredibly honored. So, let me outline for you the topics I will be talking about today as I outline the evolution of the Zelda Franchise. I'll start by introducing to you, in order, the numerous titles in the series that have appeared in the series since the birth of the first Zelda game. Next, I will explain how Zelda has changed over the years. Then I'll talk about my main theme today, which is behind-the-scenes of the franchise's evolution. I'll talk about what it means to be Zelda-esque. or what is Zelda-ness. Finally, I'll talk about approaches to evolving the franchise. I'm sure that many of you that are here today are familiar with the Zelda franchise, but since some of you might not be, I'd like to take a look back at Zelda's history. Let's start by taking a look at each of the Zelda games in the order that they were released. [Shows video clips from all Zelda found on the Collector's Disc].

Now, since this movie won't allow you to completely comprehend the flow of the series, please take a look at this timeline. [Shows timeline]. The numbers represented under sales are worldwide sales, and the units are in thousands of units sold. This list starts with the release of the Nintendo Entertainment System in 1985, 19 years ago and two years after its launch in Japan. This marked the beginnings of all Nintendo videogame products, not just Zelda. Two years after that in 1987, we saw the now celebrated release of the first title in the Zelda series, The Legend of Zelda, unveiling the top down viewpoint. This was followed up a year later in 1988 with The Adventure of Link, a more action oriented Zelda with side scrolling areas. You may notice a red asterix by The Adventure of Link in this list. This marks the year I joined Nintendo, which technically means Zelda has seniority over me at Nintendo.

In 1991, we released the Super Nintendo Entertainment System and followed that a year later with The Legend of Zelda: A Link to the Past on that system. With this title the Zelda series once again returned to the top down, isometric view. But, it can probably be said if it were not for this title, the Zelda franchise would never have been developed. It established many of the conventions for Zelda games to come, including those that were refined in The Legend of Zelda: Link's Awakening released for the Game Boy the following year. Even now, Link's Awakening is lauded as a quintessential isometric Zelda game. It was remade in full color in 1998 for the Game Boy Color.

Finally, in 1998 came the game that revolutionized the Zelda series by taking the top down 2D series into full 3D. The Legend of Zelda: The Ocarina of Time refined the 3D camera of Super Mario 64, allowed players to target enemies, and created a smooth sword fighting experience -- strengthening the appeal of the Z button on the Nintendo 64 controller and selling 7.6 million units world wide.

It was after this 3D Zelda title that I was put in charge of the series. So, when I talk in detail about development shortly, I will be talking mainly about Majora's Mask, which was released two years after Ocarina, and the subsequent release of The Wind Waker when the platform changed to GameCube .

So in 2004, we've seen that the Zelda series has developed into top down styled games on the Game Boy Advance and into full 3D styled games on home consoles like the Nintendo GameCube. But, this year we will see the release of the multiplayer game Four Swords, a game that links the Game Boy Advance to the Nintendo GameCube for a top down game that takes advantage of a 3D engine on the GameCube allowing for "2 1/2D" artistic expression.

So with this new expressive quality and the upcoming release of the unique Nintendo DS system, I think we'll see more changes coming to the franchises --the likes of which we can talk about at this year's E3, when we also talk about development ofThe Wind Waker 2.

That covers my discussion of the history of the Zelda series. But, in addition to the games I've listed here are games that can't be left out when talking about the Zelda series. [Shows a clip of Super Smash Bros. Melee and SoulCalibur II].

What you just saw are two Zelda titles developed by different creators after my involvement with Zelda began. The first was Super Smash Bros. Melee and I'm sure you're all aware it was developed under the direction of Masahiro Sakurai (former director of HAL Laboratories), who is also speaking here at GDC this year. The second was SoulCalibur II, which was made in collaboration with Namco and featured Link as an exclusive playable character in the GameCube version.

While I was not directly involved in the development of these titles, both teams were extremely careful with how they handled the Zelda characters and universe. Their work led to the even further expansion and development of the Zelda franchise.

In planning collaborations like these and the Capcom developed Oracle games for Game Boy, which were listed in the timeline, there can be difficulties in fitting the direction the creators would take into the existing Zelda universe. We're always nervous that the collaboration will feel forced and worry that it will negatively affect the franchise, but allowing other talent to developer new possibilities for Zelda games is a very important development for the series.

That covers the major developments in the Zelda franchise from past to present, so with this as our background, I'll now begin with my involvement with Zelda and my work on the Zelda team.

My first encounter with Zelda occurred in 1988 shortly after I joined Nintendo. After studying design in college, I began work designing pixel characters. At the time, I didn't have much experience playing games, and I was particularly bad at playing games that required quick reflexes. So, immediately after I started playing the original Zelda, I failed to read the movements of the Octorok in the field and my game suddenly game to an end. Even after getting used to the controls, each time the screen rolled to a new area new Octorok appeared and I thought 'am I going to have to fight these things forever?' Eventually, I gave up getting any further in the game. the result was that I was under the impression that the Legend of Zelda was not a game that suited me. So what kind of games did suit me? Those would be text-based adventures. For someone like me who enjoyed reading stories, these were games that allowed you to participate in the story and letting you experience the joy of seeing your own thoughts and actions affect the progression of the story. Plus, these games don't require fast reflexes and don't require traditional gaming skills. So, I thought that if I were going to make games, I would like to make this type of game.

After that, I spent my days drawing pixel art of Mario & Peach and honing my design talents on a variety of projects. Then in 1991 I came in contact with a new Zelda game called A Link to the Past. Although I'd been frustrated with the original Legend of Zelda, since I knew the graphical improvements of the SNES, I knew that A Link to the Past was a game I had to play even if I quit half way through.

In planning A Link to the Past, I kept playing basic actions that were completely unrelated to battling the enemies -- things like cutting the grass, lifting stones to search beneath them, and using keys to open doors. In these, I discovered that I was proceeding through the game, and I got the same feeling I did when using command inputs to actively participate in the story of a text-based adventure. I realized that those same feelings, coupled with a sense of play control response, far exceeded what I could experience through command input alone.

So in realizing the types of control input response you could have while still pushing through the story, I realized that this was the kind of game I wanted to create. Unfortunately, at that time Nintendo still had need of my role as a designer, so my hope to create a Zelda-like game could not be immediately realized.

Two years later though, there was a project that gave me just that opportunity in 1996. The game was released in the overlap between the SNES and Nintendo 64 and for a variety of reasons it was never localized, so it did not make it to the worldwide market. But, this game called Marvelous was built upon the Zelda style of adventure events and was praised in Japan as an ambitious work that felt like a change to the Zelda-style gameplay. Now, I've never asked how Mr. Miyamoto viewed this game, so I can't really make any claims about his thoughts on it. But, it was after this game that he instructed me to join the team to create Zelda.

I wasn't involved with Ocarina of Time from the initial stages of development, but rather from the point at which the planning framework had already been finalized and work was beginning on building onto that framework. This project started off with multiple directors being responsible for individual portions of the game, which was a different style from the way EAD had developed software until then.

I became responsible for dungeon design and the design of enemy creatures in the dungeons. Of course, I felt it was strange that I, who was so terrible at fighting creatures in the original Zelda and decided that Zelda wasn't the game for me, ended up working on enemy design. But, the type of gameplay used in enemy battles becomes an extremely important mechanic in Ocarina of Time, so there was really no way for me to escape it.

In Ocarina of Time, in addition to doing the dungeon design, I also took up the challenge of incorporating adventure elements into dungeons. By which I mean, giving the dungeon some type of theme, such as rescuing trapped Goron or hunting down the Poe sisters.

After finishing work on Ocarina of Time, Mr. Miyamoto instructed us to use the Ocarina of Time engine to create Ura Zelda -- Ocarina's second version with rearranged dungeon gameplay. But, I felt that just changing around the puzzle solving without changing the overall structure was too limiting.

I turned down Mr. Miyamoto's offer and proposed that if anything I wanted to make a new Zelda game. Now, Ura Zelda was developed for the Nintendo 64 disc system that was released only in Japan, and since I turned down this project other staff members developed it. It was finally bundled in the GameCube version of Ocarina and released as The Master Quest. Even now, Mr. Miyamoto scolds me for being the lazy type of guy who jumps to conclusions before even giving something a try.

Now, despite this, when I proposed doing a new Zelda title, I didn't think I'd be the director. But, under Mr. Miyamoto's instruction, we were to create a new Zelda game with a focus on its game system and we were to complete the game in a short period of time using the Ocarina of Time engine. It was necessary for some people who had been responsible for bringing together all the details of Ocarina's game spec to handle the overall design of the new game. That role was given to me.

Mr. Miyamoto's goal of completing the game in a short period of time was a result of having spent so much time developing the 3D Ocarina engine. He wanted to make effective use of that engine in creating a new game. While we decided that this would be very important for future Zelda development, we were faced with the very difficult question of just what kind of game could follow Ocarina of Time and its worldwide sales of 7 million units. In response to this challenge, we came up with the idea that the solutions to puzzles would be found in a series of recurring events. We adopted a three-day time system. This three-day system in Majora's Mask introduced players to a variety of events that occurred at the same time over this period of three days that the player played through multiple times. Once all the puzzles were solved, then the hidden goal would appear. With this system, it was possible to make the game data more compact while still providing deep gameplay.

Designing all the matters and events that were governed by this three-day period had its challenges, and thus led us to a number of new discoveries. Before creating Majora's Mask, there was actually one new type of expression that we began considering but weren't able to accomplish on the N64. With the immergence of the GameCube platform, we had the power we needed. We went back in this direction with the toon-shaded graphical style of The Wind Waker, which we completed in 2002. We decided in advance that the story of the Wind Waker would begin with Link being young and unfold from there. But, we felt that there was an unnatural feeling to using a more evolved version of the realistic Ocarina model to tell the story of a child. That led us to adopt the toon-shading graphics technique. We all had great expectations for evoking new gameplay ideas out of this new style.

The result of this was to use the main character's impressive eyes in a new focus system and increase the game's action elements with improved sword fighting. In the end, I feel I can boast that with the new visual style and new game ideas we were able to push the franchise in a pioneering new direction and further expand the Zelda universe.

Now that explanation was a little long, but it covered my encounters with the franchise and some important points in game development. The result of this is that I now work as a producer overseeing all Zelda development. I have inherited this role from Mr. Miyamoto, but the situation is more one of me being a producer who is closer to the development team and Mr. Miyamoto remaining the ultimate producer who has final say over all things Zelda. That process has not changed.

This leads us to my next topic today, which is my main theme, is the all important question that Mr. Miyamoto always chides us on, which is 'what does it mean to be Zelda-esque?'

The first thing I want to tell you about in my discussion of Zelda-esqueness is an extremely important process that is incessantly connected to Zelda development and that is something I call the "Miyamoto Test," also known as "upending the tea table".

The tea table is a low dining table that you sit on the floor to eat at. But, you don't see them often now a days as most Japanese people eat at Western-styled dining tables. When I was growing up in the 1950s through the 1970s, every house had one. What it means to upend the tea table -- it actually comes from a scene in a famous Japanese manga called "Star of the Giants," in which the father seated in back there on the left hits his son in front of him so hard that the food on the table is knocked up into the air. The father being in no mood to eat what's been served, upended the table -- forcing his wife to cook a new meal. This action by the head of the household was absolute and it represents the action of an old fashioned Japanese father.

Of course now a days, if someone was to do that it would actually destroy the family and the father would be arrested for child abuse. So, all that remains today is the phrase 'upending the tea table.' So, long ago there was period in Japan where they thought this was proper and acceptable. Now a days with only the phrase remaining, it is generally the lead figure who takes on the role of the strong proper father. Whenever a game nears completion with only the final polish remaining, with no fail Mr. Miyamoto upends our tea table and the direction that we all thought we were going in suddenly changes dramatically.

Mr. Miyamoto doesn't just upend the tea table and send the team into utter confusion. He then sits down with us and together we rethink what we've done that has been affective and what we can do that will create a positive result for a Zelda game. Mr. Miyamoto always says that when it comes to upending the tea table, I always pick up my own plate. His test is very important, and it's something that we actually welcome. When he speaks, there is a phrase that Mr. Miyamoto always mentions that speaks directly to the very nature of the Zelda series. The phrase is, 'Zelda is a game that values reality over realism.' In the art world, realism is a movement to faithfully replicate the real world to whatever extent possible. Reality is not mimicking the real world, but rather making players feel like what they are experiencing is real. The big difference is that even exaggerated expression through toon-shading can be an effective means of making things feel more real.

So what exactly is this Zelda reality that Mr. Miyamoto talks about? The next image is one scene that represents that reality. [Shows clip of Link from The Wind Waker entering the Bomb Shop at night]. As you can see, this is a scene from the Bomb Shop in the Wind Waker. Now, I'll explain this scene in detail. First we see the Bomb Shop shopkeeper. Now, as you can see from the movie, he's got kind of a weird looking face -- other than that he's a normal guy. Next Link comes. In response to Link's arrival, the shopkeeper says, 'welcome,' or in this case 'this is the Bomb Shop.' So far, this is the general flow of shopping at any store and there's nothing special about it. But, I think there may be some who notice that this scene occurs in the middle of the night. Link is just a child. This means that technically this scene fails the Miyamoto Test.

What would the correct answer be? When we finally reach the point where Link talks to the shopkeeper, it's the shopkeeper's response that must change. 'Are you alone kid, where are your parents? This is a BOMB Shop. This is no place for kids to come in the middle of the night. You be a good kid now and run on home. Well, that's what I should say, but the thing is these pirates came and stole my bombs, so business hasn't been good lately. I tell you what, I'll sell you bombs if you promise not to cause any trouble for my shop.' This type of response from the shopkeeper would pass the Miyamoto test. Of course, when I tell you this some of you will think, 'It's Nintendo, of course they want to remind people of what's proper, but that's not the point.

The important point here is that people who have been playing the game for a long time tend to forget that Link is just a child on an adventure towards some sort of objective. You know, when playing through the game there's no need to be aware of Link's age or what his ultimate goal is. But, when this happens, the things that the player is doing tend to become typical game actions and the awareness that player is just playing a game becomes stronger. Players who need bombs to progress through the game, but don't have any, will by chance find themselves visiting the bomb shop in the middle of the night. When the shopkeeper says, 'Hey, you're just a boy!' the player who had not been consciously thinking that Link was just a boy realizes, 'Oh! That's right! I'm just a boy.' The player than reflects that he's walking around in the middle of the night and starts to feel the loneliness of the middle of the night. That leads the player to become one with the game world, and the player experiences reality.

Allow me to show you one more example. [Shows a scene from Ocarina of Time where Link is destroying a blocked passage with a bomb.] This is a scene of blowing a wall up in Ocarina of Time. In this case, Link is blowing open the entrance to the dungeon, but that's not what's important. It's the sound that's important. Now, allow me to explain this scene in detail. First, the player has an object that he thinks he might be able to destroy. He puts a bomb there and the bomb explodes. This results in the wall breaking and room ahead being revealed. Now, there is a start and an end to the explosion. There are several frames from the start of the KABOOM until the smoke blurrs your view. The player still does not know for sure what has happened. Once the smoke is cleared, the player can see the result. This is the time that we hear the now famous Zelda success chime. The purpose of this chime is to inform the player that he has solved the puzzle. So in the scene we just saw, the chime was timed at the start of the explosion. This also miserably fails the Miyamoto test. Please look at the correct answer. [Shows clip from The Wind Waker of Link destroying a stone pillar with a bomb.]

This is the same type of this scene, but this time it's taken from The Wind Waker. Were you able to catch the difference between this scene and the one from Ocarina? In the Ocarina scene, the success chime was heard before the explosion ended. Although this chime is supposed to indicate success, it occurs before the player realizes the result of his action. Thus, he does not feel the reality that his actions were the correct anwser. Instead, they feel like a mere game mechanic. In the Wind Waker scene, the success chime was moved to play at this point in time [end]. This way, the player has some idea of the results of his actions when a success chime is heard and as a result the player has a feeling that he accomplished something. This is reality.

Now, the Zelda success chime is one element of Zelda-esqueness that has been carried on throughout the entire series. If we think we can just insert it whenever, it becomes a negative rather than a positive. So, this is one of the examples of a mistake that developers accustomed to developing Zelda games can make. I think these examples might give you an understanding of the Zelda reality, but these are primarily what would be considered production techniques and are really quite trivial in relation to gameplay. But, Zelda reality is built by piling on these trivial elements. These are what draw players into the game world. This begs the question of whether we should apply these trivial details to all areas of the game, but the answer is no. There are countless areas where we must remove areas of reality for the sake of gameplay.

In considering implementing production techniques, what's important is not adding them everywhere, but instead adding them with effective timing and anticipating how players are going to play the game and how they are going to interpret the different moments is what determines this timing -- such as the Bomb Shop example above. Of course, trying to figure out how players will interpret what you are creating while you're creating it is extremely difficult. Whether game developers can do so objectively is also difficult to answer.

How then do we assure that Zelda reality is maintained in a Zelda-esque way? The answer to that is to allow outside developers to play the finished game and have the developers stand behind them and watch their reactions. Another option is to ask the opinion of the producer who is separated from the actual development team. This is why I think the Miyamoto Test will continue to be an important one for us. But at the same time I get frustrated that I did not notice sooner the thing that Mr. Miyamoto notices. So, it is my objective to be able to look at Zelda from the same perspective that Mr. Miyamoto does.

In the few Zelda titles that I've had my hand in, I've tried to establish a new theme that guides the gameplay. We feel that if we continue to do the same things simply because Zelda has succeeded in the marketplace as a franchise -- it will only be a matter of time before people grew tired of the series and Zelda would have its place stolen in the videogame world stolen by some other revolutionary new title.

Even more importantly, as creators it wouldn't be any fun to simply continue making the same thing over and over again. We have continued creating new Zelda titles over the last few years without changing our core team members. Creating the next installment with the same team as the last can lead to barriers caused by conformity of ideas, but being able to take the regrets of the last title and make them a theme for the next is extremely effective and leads us to decisions of change and continuity.

For instance, in the 3D Zelda games since Ocarina, we have continued to make use of the targeting camera that always focuses on the axis that links the player's character to the target. The reason we've made this decision is because this patently Zelda-esque camera allows players the smoothest control of their character in a 3D environment and thus should not be changed. Instead, we've continued to tune this system to ease gameplay even more with each new installment.

Also, since our objective in displaying the button control for the actions the player can perform via the action icons on screen is to give players guidance in solving puzzles, which have become more difficult in 3D, we have continued using this action icon system as well.

It's true that we changed the entire game system in Majora's Mask into a three-day time system. This idea actually stemmed from the thought that we could take the time passage system that we developed in Ocarina of Time and find a way to flesh it out more. Likewise, the idea of turning the main field of play in The Wind Waker into a vast ocean that players traverse by boat started with the idea of presenting players with a new type of movement to enjoy that would be different from the horse-based movement of Epona in Ocarina of Time. The changes that we implement often come from ideas for improving and expanding elements found in previous installments.

Probably the most representative example of a change made without consciously thinking about a past installment would be the toon-shading of The Wind Waker. But, as I explained before we thought this was the best style suited for a younger Link. Our objective was to develop this into new game ideas -- so it could be said that even this was changed in the pursuit of reality.

Basically, most of the things we try to change in new Zelda games come from experiences in developing improvements and expansions on past Zelda games. What I like to call inevitable changes. The question of whether what the creators see as being inevitable or necessary changes will be considered so by the consumer is a difficult one to answer. But, we think that as long as we're able to add new elements of fun without losing what was good about the last installment, then we believe the new games will continue to be games that Zelda fans are happy with.

This is a bit of tangent, but can everyone here cook? I personally am a bit of a glutton, so I like to both cook food and eat it. Cooking is interesting because I think it's something that everyone can try and immediately see the results of their efforts. It's a very friendly form of creative work. It may be difficult to understand, but when you stew together different ingredients you have to skim impurities off the top when stewing because if you don't the stew you are making won't be delicious. It's difficult to explain clearly, but generally if you're combining distinct flavors and textures, the stronger the taste of the flavor or texture -- the more impurities or aku will come out when you mix them into the stew. On a side note in Japan, people with very colorful personalities are described as people with a lot of aku. Even when you're making a game, mixing together different ingredients into one pot produces aku. In games, the impurities are the elements that feel unnatural or out of place. I think that making a good game or being a good chef means doing a good job of removing these impurities. By combining this slow and steady work of removing these impurities and adding the spice is what Mr. Miyamoto calls Zelda reality. Doing a good job of stirring in the occasional fresh ingredient keeps the deliciousness of Zelda and leads us to new innovations with the franchise.

I'm afraid that some of you here today were hoping to hear a more technical discussion, but the essence of the Zelda franchise is difficult to sum up in technical terms. So, instead I have tried to sum it up with my own open and honest impressions.

Since I've only been involved in three Zelda titles to date, I can't say I'm anywhere near having a complete and thorough knowledge of the Zelda franchise. Going forward as a producer, continuing to make the Zelda franchise an even more stimulating one, when a time comes that I can talk to you about Zelda in more technical terms, I would be happy to do so.

Finally, I want to show you something that was released in Japan last week, and you'll be seeing it at E3 this year, I'd like to show you The Legend of Zelda: Four swords. [Shows trailer.] That's it, thank you.