Interview:GDC Keynote March 11th 2007

This presentation can be found over at <a href="http://www.nintendoworldreport.com/feature/13085" target="new">Nintendo World Report</a>. The original presentation was seperated into 15-parts.

Eiji Aonuma, director of the recent Zelda games, gave a talk at GDC 2007 entitled "The Legend of Zelda: Twilight Reflections in the Hourglass." As the title suggests, Aonuma describes the development process behind both Twilight Princess and Phantom Hourglass as well as other recent Zelda games. He also reflects on the changing markets and design of innovative ideas and intuitive control schemes and how they even touched his own family. The presentation was given in Japanese with live English and Korean translations made available through headsets.

NOA translator Bill Trinen introduced Aonuma and mentioned that he was the dungeon designer for Ocarina of Time and designed the Water Temple, which elicited cheers. He was promoted to director for Majora's Mask in 2000 and is now manager/producer of Nintendo EAD Software Development Group No. 3. While the translator did an amazing job with the translation, please note that since the translation was provided live during the talk, it may contain slight errors. The full text of the hour talk is found below.

---

Eiji Aonuma:<br

The game industry has undergone a dramatic shift over the past few years. This transformation has significantly affected the Zelda franchise. It started in 2002, just as I completed work on The Wind Waker, and I'd like to speak to you about how this shift affected the work that we were doing.

We completed work on the Japanese version of Wind Waker at the end of 2002, and at E3 2003, we announced that the North American version, which was released at the beginning of the same year, had sold up to a million copies. But the truth of the matter was that the game did not fare as well in Japan. I won't go into exact numbers, but it was enough to show us the difference in the degree of enthusiasm between the Japanese and North American markets.

It had become clear that the Japanese market was moving away from games, in what we have come to call "gamer drift." The upper management at Nintendo knew that it had to come up with a solution to this phenomenon. This gave birth to the Nintendo DS, introduced in 2004 and was intimately tied to Nintendo's Touch Generations initiative, which specifically targets those who have never played games before. I am sure you are all familiar with it. At this time, I had still not experienced this first-hand. For this reason, I was still convinced the reason the Wind Waker did not perform well was because of its toon-shaded graphics style. It was something that you either loved or hated, and there was nothing that we could have done about it.

I was thinking about what was happening to the market and what it could potentially mean. We hadn't been able to add and truly new ideas to the core Zelda gameplay since the series made the jump to 3-D. This resulted in some seasoned gamers growing tired of the formula. In contrast, those who had never played a Zelda game were intimidated because they felt these games were too complicated. These, Miyamoto felt, were the real reasons that the game did not sell well, so we started on a project that utilized Wind Waker assets and was based on the theme of creating a new style of gameplay. We called this project Wind Waker 2, and it was eventually reborn as Twilight Princess, but I will explain that in more detail later.

First, allow me to speak on the theme of innovation in gameplay. The need to innovate gameplay was not only limited to Zelda. Nintendo recognized the problem of gamer drift, and our philosophy was that a new style of gameplay was needed into order to breathe new life into the market, and our answer to this was the invention of a certain system. As you are all aware, the system we called "connectivity," linked the Game Boy Advance to the GameCube, allowing the Game Boy Advance to be used as a controller with its own screen. We implemented this system of connectivity in The Wind Waker as the Tingle Tuner, and several other titles took advantage of connectivity as well. However, there wasn't any one title that used connectivity as a central game mechanic, and Miyamoto that no one had conveyed to gamers just how much fun connectivity could be.

We began working again on a multiplayer Zelda game that used connectivity as its main gameplay system, and I was made producer of this title. This project which was released as Four Swords Adventures, was based on Four Swords, a game developed for the Game Boy Advance as the very first multiplayer Zelda game. Adventures took place in a 3-D world where four players run around the field on a main monitor, but when players enter dungeons or rooms play field shifts to the players' own Game Boy Advance screens. There they can play in an environment where their opponents cannot see them or what they are doing. Ultimately, this game introduced a completely new style of play.

At E3 2003, the response to this game from attendees was very positive, and I was very hopeful for the game's release at the beginning of 2004, but the results were not very good and I felt very disappointed by the outcome. I believe this result stemmed from the need for each player to have a Game Boy Advance and the need for each player to also have a cable to connect that Game Boy Advance to a GameCube. I thought that requirements like these prevented it from doing as well as we hoped, but there was another problem. I think you might have noticed this as I was explaining the game, but it suffered from seeming too complicated. It was too difficult to convince the consumer that they wanted to play the game. Simply put, it was a challenge to give users incentive to play because it was difficult to show how much fun the game was even through television commercials and other advertisements. Based on the negative result we got with connectivity, Nintendo learned that no matter how innovative the gameplay is, unless we have something that we can convey directly and intuitively, people will not be interested in it.

Let me backtrack a little. As I was busy working on the connectivity project, it wasn't as though the Wind Waker 2 project that I spoke of earlier came to a halt. Not at all. As some of you know, at E3 2004, we unveiled the game that would become Twilight Princess, the realistic Zelda game, and we announced that it was developed by the team that had been developing Wind Waker 2. Actually, there was a reason that that decision was made at the time. At one point, I had heard that even Wind Waker, which had reached the million mark in sales, had become sluggish in North America, where the market was much healthier than in Japan. I asked NOA why this was. What I was told was that the toon-shading technique was, in fact, giving the impression that this Zelda was for a younger audience and that, for this reason, it alienated the upper teen audience that had represented the typical Zelda player. Having heard that, I began to worry about whether Wind Waker 2, which used a similar presentation, was something that would actually sell. In addition, because we knew how difficult it would be to create an innovative way of playing using existing GameCube hardware, we knew what a challenge it would be to develop something that would do well in the Japanese market, where gamer drift was happening.

That's when I decided that if we didn't have an effective and immediate solution, the only thing we could do was to give the healthy North American market the Zelda that they wanted. So, at the end of 2003, I went to Miyamoto and said, "I want to make a realistic Zelda." Miyamoto was skeptical at first. I was so focused on changing the look of the game as being the solution we were looking for without coming up with a breakthrough game idea, and he advised me that "If you really want to make a realistic Zelda, you should start by doing what you couldn't in the Ocarina of Time. Make it so that Link can attack enemies while riding on his horse using the Wind Waker engine, and make your decision based on how that feels." This is something that went against everything that the staff had been working on and I expected to come as quite a shock to the team. Surprisingly, my entire staff was enthusiastic about this change, and the project on which progress had slowed was given a much-needed jumpstart.

Four months later, development had progressed to a point where Link could swing his sword in battle against enemies while riding on his horse in a realistic looking environment. When it was announced as a surprise trailer at the 2004 E3, it received a standing ovation by the media audience. This was a very exciting moment for us, but we were still very much in the early stages of converting the game into something more realistic. We knew that we had to create a Zelda game that would live up to expectations of fans in North America, and that if we didn't, it could mean the end of the franchise. But I also trusted the ability of the team, which was able to bring the game so far in just four short months, and Miyamoto and I announced that this Zelda would be released in the fall of 2005.

The 2004 E3 was a bit of a turning point for the Zelda franchise. It was also a very important show for Nintendo as it was the year that the Nintendo DS was introduced. As I mentioned earlier with connectivity, we learned that even if we come up with an innovated type of gameplay, unless we can convey that experience clearly, people will not be interested in it. The DS, with its direct control, touch-screen operation, and two screens, immediately intrigued people. They were asking, "What's that? What does it do?" It was clear from your response that developers were also excited about the possibilities with this new hardware. The change in the gaming environment that brought about the creation of the DS hardware was not clearly evident at the time of the 2004 show.

I was so deeply involved with the realistic Zelda project, that I still hadn't realized how important it was for a DS Zelda to be developed. However, Nintendo was already moving in that direction and Miyamoto felt that it was time for Zelda to move in that direction as well. The team that had developed Four Swords Adventures, had already begun work on DS Zelda. Upon returning from E3, I heard from my staff that the DS was hardware was capable of supporting toon-shading. I immediately asked my staff to implement toon-shading on the DS because I was disappointed that the toon-shaded Zelda was not received as well as I had hoped. In a short amount of time, they were able to show me 2-D Link moving in a 3-D world on the top screen, and on the lower screen, a Link shown on a map that was controlled entirely by touch. I talked about this with a member of my staff the other day, and we dug up the prototype version. Here's a movie from it:

While I was surprised to see the toon-shaded Link from the GameCube version moving around on the DS screen, the controls were not intuitive. I asked my staff to display the 3-D world on the bottom screen and change it so that players would control Link by touching him directly. The style of playing where the player touches the screen directly, connects perfectly with the gameplay in Zelda, where the player is encouraged to interact with various objects in the environment. When I thought about what the overriding theme of the Zelda DS game that took advantage of that would be, I knew that we had to make it so that standard control would have the player touching Link and his surroundings to interact with them.

At first my staff was bewildered. Once they understood that the traditional controls were replaced by something more direct, new playstyles emerged. Players lock onto enemies and attack them by tapping, draw the path a boomerang would take, and make notes on a map. The stylus made all these interactions possible. Based on this, we were able to create the first major innovation to Zelda's gameplay in 3-D since the Nintendo 64. In addition, we felt that although past versions of Zelda may have intimidated new uses with their complicated controls, the direct control of the stylus meant that it would be easy to entice new users even in a market experiencing gamer drift. In order to ensure that the game would appeal to seasoned Zelda players as well as new players, we added the online battle feature using WiFi, a first for the Zelda franchise. Direct control functioned well on that level as well, and I will speak more about that later.

At the same time that we were working on taking Zelda DS in a new direction, we had to make a decision about the direction of the realistic Zelda game. We knew what a challenge it would be to innovative gameplay on the GameCube, so we had to come up with something new from another direction. Try though we did, we couldn't come up with any good ideas. We were afraid that the game was no longer selling as it once did in Japan, Zelda's basic gameplay had been received well by many users in the past. If we changed it just to change it, we worried that longtime Zelda gamers might not appreciate the new direction, and rather than draw new users, we thought that we would end up alienating everyone. I think that this is something that all developers working on a long-running series confront. It certainly wasn't easy for me to come up with a viable solution. This went on and 2005 was quickly approaching.

Knowing that I had raised the expectations of end-users by promising that a realistic Zelda was on the way, I knew that I had to come up with something brilliant, something that would take advantage of the look of the game. My staff suggested that changing the environment would change the gameplay and looking to past Zelda games which use the contrast between two different worlds, dark and light, past and future, we thought to incorporate this idea again. This is when the idea that Link would transform into a wolf was proposed. The idea of Link's transformation into a wolf came from The Legend of Zelda: A Link to the Past, in which Link transformed into a rabbit in the Dark World. We wanted to change Link's environment, but we also wanted to make a completely different skill set available to him in this new environment and thus create a completely different style of gameplay. We wanted him to become an animal, combining the ideas of both the wild and heroic into one. That's how we came up with the idea of the wolf. I was also scolded by Miyamoto who said, "It's hard enough to get a two-legged Link to move around in a 3-D world convincingly. To consider a four-legged wolf is something an amateur would do." I knew that what he was saying was right and would eventually regret having even tried it, but even if it was a challenge at the time, I thought that this kind of disruptive breakthrough was just what the staff needed to change their way of thinking.

I had no doubt that this was the right direction to pursue, but having set the framework of this new direction, my work was required on another project. I left the realistic Zelda team to work details and began work as director of The Minish Cap for the Game Boy Advance, which Capcom developed. Even in the weakening Japanese market, the Game Boy Advance install-base continued to grow. The market seemed to have stabilized and as one of the titles that would help sustain it, Minish Cap was very important to the franchise. Also, because this game involved Link moving freely between the normal world and the microscopic world of the Minish, which was very close to the idea that we were shooting for with the realistic Zelda, I was very passionate about my work on this title. I felt certain that the way the gameplay changed as the environment changed in Minish Cap would have a positive effect on the development of the realistic Zelda. In hindsight, by immersing myself in the Minish Cap project, I gave myself a way to escape from not being able to find a breakthrough on the realistic Zelda project. I regret that I might have been too selfish, thinking that I could just leave my staff to come up with the solution, but my mistake in thinking caught up with me.

Having released Minish Cap without a hitch, I started to think about what content to include on the E3 2005 playable version of the realistic Zelda. However, development on the project that I hadn't been able to support was struggling. Although there were various small sections of gameplay that would create the framework for the game, there wasn't one single distinctive model of play or a timeline that would connect each of the events. I really didn't think that at that rate, we would create that would generate positive buzz on the E3 show floor. What concerned me most was at this stage of development, as a result of having put priority on the idea of the two worlds and the wolf, there was still nothing special about the most important aspect of the game; there was nothing special about the movement of the realistic Link.

In addition to this, after the 2004 E3 announcement, Miyamoto heard the response from end-users about the realistic Zelda. Miyamoto gave us a mission: our goal must be that if we were going to do it, it had to be 120% Zelda, and that meant making a Zelda that exceeded the Ocarina of Time. But where we were at the time did not exceed Ocarina, and regardless of the result, I knew that we had to focus on a new play feeling and also on creating alluring Link actions in preparation for E3. Once we got past E3, only then could we really work toward Miyamoto's mission. From that point on, I worked not as a producer, but as the game's director.

At E3 2005, though the presentation might have had something to do with it, response was huge. Some attendees waited up to three hours in a long line to play the playable version of the realistic Zelda called Twilight Princess. Ultimately, we were able to deliver something that lived up to their expectations. However, I knew that what we had at that point, thought the art style was new and its own appeal, still didn't have the innovation in gameplay that DS Zelda had, and that was something we had to overcome. Could we provide Twilight Princess with a feeling of something that DS Zelda had? As though he could see through my concerns upon returning from E3, Miyamoto approached me and said, "It's as though the Revolution pointer was designed specifically for the arrow control in Zelda. Why don't you consider making a Zelda that uses this?"

Control that you experience though direct pointing device and motion sensor: Upon its conception, I was curious to see the Revolution, Nintendo's new platform announced at E3 2005, would innovate Zelda's gameplay. I was eager to try it, but I thought that that could happen only after we completed work on Twilight Princess. I was surprised at first to hear Miyamoto's suggestion that we try it right away, and I didn't know how it would impact the Zelda that at that time was still missing a key ingredient. But we had to start by testing what Miyamoto had suggested and that was getting the pointer to work as the control for the bow and arrow. Though the pointer control that we developed as a trial still needed tweaking, and as imperfect as it was, the feeling of being able to aim directly at the target on the screen was, as with the DS version, exactly what was needed to give Zelda a completely new feel. Having experienced it, I was convinced that that was the only direction we could take. "This direct control is exactly what Twilight Princess needs in order to penetrate a market experiencing gamer drift." Miyamoto knew that this would be my response, and he himself felt strongly about it.

However, if we put Twilight Princess on the Revolution, we ran the risk of those looking forward to playing it on the GameCube. For this reason, we knew that we had to release it for the GameCube as well. But doing so meant developing two versions, a GameCube version and Revolution version, and we knew that it was impossible to create both before the 2005 release as promised. Miyamoto negotiated with the higher-ups, and in the end, while it would disappoint those users who were looking forward to the 2005 release, if we really were able to create a 120% Zelda game that could be played both on the GameCube and on the Revolution, then even if we did release at the time of the launch of the Revolution in 2006, we could make even those users happy. This was Iwata's response and the decision to create Twilight Princess for both the Revolution and the GameCube.

Creating a Zelda title for launch was a first in the history of the Zelda franchise. This was the first launch title that I had worked on, and I had no way of knowing how much of a challenge it was going to be. In addition, because we were making the GameCube and Wii versions (I'll be calling it Wii from now on), we had to be particularly careful about the schedule, and being inexperienced, I had no idea how long it should all take. Thankfully, the Wii hardware supports compatibility with the GameCube, so I heard that transferring data that we had developed for the GameCube to the Wii wasn't that much of a challenge in terms of coding. I decided that once the game content had proven to be stable enough to run on the GameCube, we would work on the Wii-compatible version. I set our first goal of having the GameCube version complete through the middle while parallel to this work, Miyamoto and I began thinking about what controls would be changed in the Wii version, deciding what to do first about the pointing.

As I mentioned earlier, after the pointing experiment, I thought the feeling of aiming at the screen with the bow and arrow was convincing, so we started polishing this control. What Miyamoto wanted to accomplish first of all with this control was to eliminate the 3-D stick as the camera control in the GameCube version and make it into something simpler where the camera could be moved using only the pointer. I think those of you who have worked on an FPS will understand, but it was a reoccurring issue with existing camera controls where there are two styles of control. There was the joystick style, where tilting the stick forward would move the camera down, and tilting it back would move the camera up, and cursor controls, where the camera would move in the same direction that the stick was tilted. Neither control style was really acceptable. It was our goal to create a solution for this and to this problem.

Please watch this movie where the boomerang is being used. When the boomerang is being held, in the center you can see a circular mask appears. If the pointer is moved within this area, the camera will not move, but if the pointer is moved outside of that area, then the camera will move in that direction. A camera that rotates in the pointer's direction is unlike anything that had been used before and took some time to get used to. But because we were able to point in the direction that we wanted to look and cause the camera to move, we felt that this was a very direct and effective control style.

Next, the motion sensor. I had been thinking about direct controls like swinging a remote to have Link swing his sword and wanted to push the experience of playing even further. When we tried changing the camera perspective from one that normally is in third person to one that switches to Link's first-person perspective in battle. Please watch this movie. For this prototype, all we did was change this perspective to the first-person view we currently had. We haven't tweaked things like the spacing between Link and enemies, so it's really awkward, but having played this version in first person, we didn't think this was the most effective way of presenting battles. Link normally has a variety of movements, but when he enters into battle with his enemies, which has come to be known as core scene in Zelda, if that variety is lost, it feels very strange.

So, we brought the camera back to the third-person view and swung the Wii remote like a sword. This time we saw problems that we hadn't noticed when the camera was in the first-person perspective, and we made the decision to abandon using the motion sensor to swing the sword. The reason for this is that the player character of Link is left handed. You see, when a player had the Wii remote in his right hand and swung it, it felt strange because on the screen Link would swing the sword that was in his left hand, and this was something that couldn't be avoided. This might seem trivial, but the act of shaking the remote is a control that the player physically experiences, and if it doesn't match up with what is happening on the screen, then it becomes something that has no effect at all, and ultimately this becomes something that is considered extraneous control.

Because the sword control was the most direct in Zelda, it was the one that we most wanted to implement, so we were very disappointed that we could not use it. We settled upon the assigning sword swinging to the B trigger played in the past. In addition, because there are no X or Y buttons on the Wii, which were used as the item buttons on the GameCube, though it takes a little time to get used to, we assigned left, right, and down on the plus control pad above the A button for this. Consequently, we added one more button than in the GameCube version, and by making decisions like this, we were confident that we had created something that was not inferior to the GameCube version for the Wii. Feeling that the structure of the Wii version was finalized, we put a playable version on display at E3 2006 and awaited the response in Japan.

In daily reports that I received from various staff members who were at E3, there were things like, "Feedback on the Wii has been very positive, but there is evidence that people think the control is too hard." I was concerned, but this was made even clearer after Miyamoto returned to Japan. Many said that the plus control pad item controls were too easily confused. A number of people said that the camera would move around on its own in the first-person item view, making it confusing, and people were left not knowing what they were supposed to do.

So, including the extension to develop the Wii version, we had only four months left and there was still so much of the game data that we had to come up with. Now, to receive negative feedback like this, what would you do in my shoes? "Maybe it's because on the show floor, there isn't enough time to get used to the Wii controls and that's why we're getting this response. Or, it's been switched over from the GameCube version and people are used to the GameCube controls, so of course they're going to have a hard time with it." These were the kind of thoughts that were running through my head as I was trying to get away from having to revisit the controls. I had to come up with a response to the negative reactions, and so while playing the other E3 playable launch titles for reference, I discovered that the number one thing missing from Zelda at the time.

The controls we as developers had decided to implement in Zelda were ones that we had forced into the game in order to take advantage of the Wii controls. I had realized that I hadn't worried about making it easy to understand and intuitive and making people want to play the game; all those things relate. The feeling that I had that people would have to get used to the controls when working on switching controls over to the Wii was evidence of this, and I realized that I was forcing these controls onto the user. It wasn't a Zelda that would have been received well by the end user. I felt as though Zelda had been left behind by the other launch titles, and having hit rock-bottom, Miyamoto and I began rethinking the Wii controls completely.

If you analyze all the negative feedback from E3, it all seemed to point to the same thing. It wasn't something that could be controlled with ease. When a person tries something new, they are simply looking for something that feels comfortable, and when it doesn't, they start to feel like they need to get used to it. But for users who aren't able to get used to it, it just becomes something complicated and difficult. I thought that making the controls direct was the answer to the game being too difficult, but ultimately, it was the complete opposite. That's when I was forced to realize that in order to resolve the issue, the most important thing to do was to provide the user with a sense of comfort and ease.

First, we looked at the one thing that received the most negative feedback, and that was the movement of the camera with the point control. We tried fine tuning the rotation speed and the sensor area, but we weren't able to get it to a point where anyone was simply able to pick up the game and control it comfortably. In addition, the fundamentals of the gameplay are such that the player does not actively have to think about pointing. Pointing is key when the player is choosing an item. So if when that transition happens, the pointer is in a camera rotation area, the camera would rotate all of the sudden, and we couldn't avoid confusing the user in this way. Please watch this movie of the pointer control after we changed it. I think you can see the icons marking the right and left movement of the control stick. The result was that we decided to just kept the vertical camera control using the pointer because it doesn't rotate and returning lateral movement to the control stick. When we switched to this mode, we decided to add text that told the player when the pointer was outside the detectable range, making it so that the player does not perform any action until the pointer is back in range. In the end, we were not able to fulfill Miyamoto's wish for uniform camera control completely because a game like Zelda involves switching between first-person and third-person camera perspectives and so it's something that we cannot get around. But for our traditional FPS, which primarily uses the first-person view, it should be more intuitive for the user. He would be ecstatic if developers creating games like that would adopt Miyamoto's vision.

Next, we addressed the plus control pad item controls. As many of the item controls in Zelda require that the player hold and release, we thought that while a player was pressing and holding in one direction, they would inevitably press another, causing a faulty button press. This problem of key assignment and control was one that we had a really hard time resolving, but we knew the only way we could have players swap items was to use the plus control pad. If we were to change the press-and-hold item control to a switch, it would negatively affect the gameplay and make the game less fun. We felt like we were stuck. Have you ever experienced anything like this? We could have made the decision to give up one thing or another, but we decided to separate the controls to alleviate the problem. In other words, I assigned the hold function to the B trigger button, and used the plus control pad as a place where players could queue up other items and swap that item out with the one assigned to the B button just by pressing the corresponding direction. These changes minimized the mis-entered data and made it easy and comfortable for users to point while holding down the B trigger button to control the item.

Key assignment doesn't seem like a big deal if you look at it objectively, but it was a great challenge to assign the same number of functions with significantly fewer keys to work with, and once we assigned one, we had to reassign another. That thing happened when we decided to assign items to the B trigger button. Until then, we had been using the B trigger button for the sword control, but actually, ever since we started reconsidering the key assignment for the sword at E3, we already decided that we had to go with another control, the control that users wanted to perform. I'm talking about the motion sensor swing. When we were developing the E3 version, we knew that it felt strange for Link to be left-handed, so we removed that, but if we still wanted to implement that control, we would have to make him right-handed. However, in order to do this, we would have had to redo Link's character in the game, which, with only four months left in development, would have been impossible.

It was at that time that we thought if we flipped the entire world laterally, so that left and right were reversed, Link would be right-handed, and that decision to use this bold solution was made. Some of the staff members criticized the idea, saying that it wasn't the final course layout that they designed and that the composition would be thrown off. They had a hard time with it at first, but after personally playing the game, the feeling of strangeness was gone after only a week, and in contrast, the original GameCube version started to feel strange. I knew that end users would feel the same way and I convinced my staff of this as well. Of course, the effect of these changes was immense, and giving the player the ability to swing the remote to make Link swing his sword was imperative to this game.

In the beginning, we thought about making it so that the player could swing the remote vertically to have Link swing his sword vertically, and horizontally to have Link swing his sword horizontally. But in the end, we decided to make it so that no matter which way the remote was swung, Link would swing his sword. If the sensor is made to detect direction, the response is inevitably effective. We made our decision because whenever a player is immersed in a game, they really aren't thinking about whether they are moving the remote vertically or horizontally and really wasn't a benefit to the experience unless they are making a conscious effort to think about what they were doing. But rather than adding that distinction, we decided to make ease and comfort our priority so that players knew when they swung their remote, Link would swing his sword. In order to test this new version and see how playable it really was, we held an internal playtest. When I saw some of our female employees defeating a giant boss, I was confident that as with the DS Zelda, the realistic Zelda had been reborn.

Initially, I was quite bitter about the negative response that we received at E3, but in the end, I believe that it was critical for us to get that feedback at the time that we did. In many ways, creating is about suffering. "You must learn from your pain. Those who do not will never become what they are truly meant to be." This is something that Miyamoto has been saying a lot lately, and though I have been making games for a long time, for someone like me to struggle to create something true without some amount of suffering is near impossible. There have been many occasions where in the final stages of development, Miyamoto has said to me, "You aren't thinking about the end user," what's come to be known as "upending the tea-table," and I felt that Twilight Princess had officially been christened. For those of you who are curious, please feel free to check the Zelda section of "Iwata Asks" on the Nintendo homepage to read more about the anguish I experienced after being reprimanded by Miyamoto.

The Twilight Princess was completed as Miyamoto and I had envisioned, and having heard that in North America and Europe, the majority of users who bought the Wii also purchased a copy of Twilight Princess, I thought that all of our hardship and hard work had paid off. However, in comparison, sales figures were not what I had been hoping for in Japan, and so I believe that many users still have the impression that Zelda was too complicated and therefore too hard to play. Based on this and realizing that Zelda needed to change even more, I was well aware that getting more people interested in the title was going to be difficult. DS Zelda, which we are planning to release this year, is full of concessions that we made based on this understanding, and I am looking forward to seeing what kind of results our efforts bring about.

I mentioned earlier that in the new DS Zelda, we undertook something totally new to the franchise, a WiFi enabled battle mode. I will explain exactly what that will be like here. The name battle mode probably conjures up images to some of you of multiple Links appearing on the screen and swinging their swords, but this battle mode has a nickname, "WiFi hide and go seek," and it's a completely new way of playing. The number of players is simple: it's a one-on-one game. Both players on the same field taking turns as the offense, whose goal it is to take items to their base, and the defense. On offense, one player plays as Link and the other, the defense, move three, what we are calling "Phantoms," whose job it is to stop Link. How will one player control three characters? Well, this is what's key to playing this game, and it's the same control that in DS Zelda, the player can use to control Link. Please take a look at some footage of actual gameplay. Now here, I am playing as the red Link against my opponent, the blue Link, who just happens to be the planner who designed this battle mode. I certainly hope that you'll cut the red Link some slack for playing so poorly.

First please focus on the red Link who is playing on offense. I'm going to stop here for a second. The top screen: this game is a battle so normally you would check to see the enemy, but not on this map. On this map, all you see is Link. Now you can see the icons of the three Phantoms that I mentioned earlier. This is because red Link has picked up an item, a Force Gem. You can see the Phantoms while carrying these. The red Link switches between carrying and lifting the Force Gem because even though I have the advantage of seeing where the Phantoms are while I'm carrying them, they also slow me down, so if I carry a Force Gem for too long, it increases the risk of the Phantoms catching up to me. This is how the game is meant to be played, and not just because I am bad at it. The red Link just placed the Force Gem in his base and earned a point. You might not have noticed the red Link's speed drop that much while he was carrying a Force Gem, but the bigger the Force Gem, the more it will slow Link down. However, while the risk is higher, the points that the player can earn are also bigger with these Force Gems, and that's a strategic decision that the player must make. The red Link was attacked by a Phantom. In this game, though Link cannot attack the Phantoms, the Phantoms will automatically attack Link if he comes near one of them. When a Phantom comes near Link, he must run without a moment's delay, but with my reflexes, I was unable to escape that attack. With this, the players switch sides, and the player controlling the red Link will now play defense and control the Phantoms.

Let's take a look at that same game, this time from the perspective on defense. I think you can see the difference from when the red Link was being controlled. Lines are being drawn in order to control the Phantoms and the map is being displayed on the lower screen. Also, with red Link, you start the game without being able to see the Phantoms, but the Phantoms can always see where Link is. Link has disappeared from the map screen. Do you see the light blue areas on screen? If Link enters these spaces, he cannot be seen by the Phantoms. Also, Phantoms cannot enter these spaces, so these serve as safe zones for Link. Link has reappeared on the screen. Red Link was in a safe zone for a while, but then the player controlling the blue Link predicted my movements and placed the Phantoms around the safe zone. It was too dangerous for me to leave it. It's obvious that the blue Link is clearly the better player. Link has disappeared again. Link will also disappear when he is on his base; now this is a safe zone too.

…And the red Link is down again. The red Link is surrounded by the Phantoms, and this happens quite often, especially when Link cannot see where the Phantoms are. Because of this, the player controlling Link must check the location of the Phantoms while carrying a Force Gem and predict where they will go next and move Link accordingly. That's the key to playing this game. In this way, each player takes turns trying to get as many items to their base in the given amount of time without being attacked by a Phantom. After the set number of turns is taken, the winner is determined. Each game is short, the longest possible being twelve minutes, but with games where you are forced to try to read your opponents next move, the more you play, the better you get at reading their habits, and so this game is very addictive. You will understand this game as soon as you experience it, so I certainly hope that you will head to the Nintendo booth and try the game out for yourself.

As an example of this, and in conclusion, I would like to tell you about something that recently happened very close to me. At home, I have a little boy who will soon turn 5 and my wife, who is about the same age as I am. I'll tell you more about my son in a little bit. But my wife, who has lived with me, her husband, who works at Nintendo for over 15 years, has absolutely no interest in games and takes the stance that they are not necessary at all. In that way, I suppose my wife is kind of similar to Miyamoto's. And because she feels this way, she treats me like the happy-go-lucky who plays the games that he loves every day at work without understanding my day-to-day hardships at all. I feel like I'm saying "at all" too much; I'll change that to "very much."

Back to my home life: I had never even considered taking a game console into a household with a small child and wife like that until, one day, my son said, "I want a Wii Remote." I think he probably saw one in a TV commercial one day. What really struck me was he didn't say "I want a Wii." He said he wanted a Wii Remote. To his eyes, that remote looked interesting, and he really wanted to hold it. He had no idea at all that it was connected to a game machine. It was such a surprise to me. Next, after placating my wife's concerns that five years old was too early to start playing games, I brought a Wii console home along with Wii Sports because they were first-time gamers, and although I thought it might be too much, Twilight Princess.

I started by having my son play Wii Sports. I don't mean to praise my own company's product, but even my five year-old son was able to play Wii Sports. Take boxing for example: without understanding the rules, he knew that all he had to do was move the Wii Remotes relative to the character on the screen, and it looked like he was having a great time doing it. I thought that this would be the case, so I wasn't all that surprised. Then, when he saw the copy of Twilight Princess on the table and asked, "What's this? I want to try this too." I thought it would be too much for him, but because he asked, I started the game up and handed him the Wii Remote and Nunchuk. Because the controls aren't as simple as with Wii Sports, my son was a little lost in the beginning. After I showed him how the controls work, he was able to fish and call the hawk in Ordon village at the beginning of the game. He slowly got used to the controls, and was finally able to point and shoot targets accurately with the slingshot. After my wife made him stop, saying that playing games too long was bad for his eyes, he asked, "Can I play Zelda again tomorrow?" A five year-old was able to play Zelda; this was a huge surprise.

The next day, I came across a more shocking scene. I got home earlier than usual, and in our living room was my wife, fighting a monster in one of the dungeons, and next to her, as though he was cheering her on, sat my son shaking his arms in front of him in small movements. I asked my wife what was going on, and she said, "He was playing Zelda and got to a part in the game where a bunch of monsters come out. Because he got so scared, he couldn't make it any further. At first I was just playing to get him further along, but as we were playing, he started giving me advice, and we were having so much fun playing together. He just gets so sad when he sees that big "Game Over," so I can't help but keep playing to make sure that that doesn't happen." That's how she explained it, and ever since, I've been getting one question after another about, "What's going on here? Am I supposed to do this here?"

When I see the two of them getting so excited about it, I can't help but think that I should have introduced my wife to Zelda games sooner. To that, my wife said, "Old games, if I was watching someone play, I never felt like I wanted to try it myself, but with the Wii, after seeing our son play it, I wanted to try it too." Upon hearing that, well, I knew that the fact that our now having a son also played a part in it, but that feeling that she had of seeing someone play and wanting to play herself reminded me of when I first played Mario on the Famicom. I realized that the Wii, which was reviving this feeling in the end user, was reminding us as developers of what is important to a game.

Watching my son, who refuses to leave Ordon village because he is scared of monsters and he spends every day collecting Rupees and getting excited screaming, "Hey, I found a new place," it reminds me of noticing even small details. But for now, as a follow-up to Twilight Princess and in hopes that many of you will try the DS Zelda, Phantom Hourglass, which will be released later this year as the latest evolution of the Zelda franchise. Please enjoy this movie as I conclude my presentation.