Super Mario Galaxy: Q&A with Yoshiaki Koizumi on the Finished Game

We chat with Galaxy's game director about the making of the game and the future of Mario.


Games don't get much bigger than Super Mario Galaxy for Nintendo Wii, a game that has not only brought a Mario platformer to the console for the first time but has also rocketed the plucky little plumber into, of all places, outer space. With its release basically a month behind us, we took time to sit down with Galaxy game director Yoshiaki Koizumi to get his thoughts on the making of his masterpiece, the challenges of bringing Mario to the Wii, and the future of Mario's never-ending video game quest.

Mario's latest adventure takes him to the stars.
Mario's latest adventure takes him to the stars.

GameSpot: Did you have any expectations on the reception of the game when you were working on it? Or were you just so focused on getting it done and making it right that you didn't really have time to anticipate that response?

Yoshiaki Koizumi: So, of course we wanted to make something that people like, but during development…we were very much focused on the game. Yes, of course, [I was] just very happy; it's only been a month since the game has been out, but I've had the chance to see a few of the reviews and [I have] been very, very happy with what I've seen

GS: Can you give us a sense of the team's background? Because it seems like a lot of the team had worked on Donkey Kong Jungle Beat.

YK: The EAD Tokyo office was started in 2003, and their purpose was to go there and try to hire some of the talent that exists in the Tokyo area--because those people might not want to move otherwise, I suppose--and with those new hires, to start making games. So one of the first things they had to do was to teach them how to make games the Nintendo way. So, that meant bringing a lot of people from Kyoto over to Tokyo and spending quite a bit of time with them. The first game that was produced by EAD Tokyo was Jungle Beat and then… [Super Mario] Galaxy.

GS: When the team got tapped to do Galaxy--a Mario game--it's kind of like getting the keys to dad's car. Did they feel any pressure when they were working on it? Because not only was it the new proper console Mario game, but there were a lot of expectations on what a Wii Mario game would be like, so can you shed some light on how they all wrapped their heads around that?

YK: It wasn't so much pressure that [we were] making a Mario game, specifically, but rather just making a game for Wii I found to be rather daunting because the possibilities are so broad, it's hard to know where to start and where to stop. And not only that, but I felt like the 3D action genre itself is kind of a daunting one to jump into. So, much more than worrying about making a Mario game, those were the more prevalent sources of pressure on my mind.

GS: Let's talk about how you approach a Wii action game. Can you give us some insight into the design process? Obviously, if it's a Mario game, you already kind of have a framework. But how do you fill the framework? How did you brainstorm on how to use the Wii controller?

YK: The initial concept for Galaxy was in place actually, before we knew anything about the Wii, and what we did was, we created a prototype of an early version of the game with a small team and began to show that around. And as we were starting to get feedback at that point, a lot of us began to incorporate some ideas for what it meant to fully use the possibilities for Wii.

GS: As far as concept and gameplay, is the system that we see in the final game exactly what they were thinking of initially, or was there something that over the process of development they decided to take out because it wasn't working out?

YK: Of course a lot of fine points changed over time. But when I think about the entire game, I feel like it's not a very terribly different sort of a thing at all. Of course, all of the ideas were in place, like the playing surfaces were going to be flipped over and turned around, you can walk on the undersides. And I had this idea for like a really big boss robot that you could climb all over with enemies coming out of it. And so as development went on there were a few touches here and there that were added, but the broad strokes were all certainly in place.

GS: I have a question from a Luigi and a Yoshi enthusiast on staff. Why is Luigi only playable once you've already cleared all the game content? And where is Yoshi?

YK: OK, so, in regards to Luigi. Certain members of my staff asked me about it also, like, "Why can't we bring out Luigi a little bit earlier?" And my answer at that time [was], "People tend to think of the 3D Mario games as something you have to get 120 stars, that really makes up the main body of the game." So, if we were going to bring in Luigi earlier than that, we felt like it would... break up the experience going up to 120 stars. Plus, we needed something to motivate people to get 120 stars in the first place, and this seemed like the right way to do it. Not only that, but the one last factor that you can't leave out is, our concept of the character of Luigi is that he always plays second fiddle to Mario, and he has like a complete neurosis about it. So, we really have to feed that story back into it every single time.

Before Galaxy, the EAD Tokyo development team worked on Donkey Kong Jungle Beat.
Before Galaxy, the EAD Tokyo development team worked on Donkey Kong Jungle Beat.

As for Yoshi, we certainly thought about including like, a rideable Yoshi character for Mario, but there were already so many elements that we were putting into the game, we didn't want to dilute everything we had already done by including too much in that sense. Plus, the main functional quality that Yoshi brings to a game from a developer's standpoint is that he allows for that delay of hover jump, where he can do his little swoop right at the end there. And that functionality was already in the game in the form of the bee suit. So, from a functional standpoint, we didn't feel it was necessary.

GS: An excellent segue. Speaking of the bee suit, a lot of people were very excited to see the suit's return, and they were wondering were there more suits that were going to be in the game? Also what was the thinking behind making the ice and the fire suits timed?

YK: There were a couple of Mario abilities or transformations that didn't make it into the final game. Most of these were for balance reasons at the time. We didn't want to take out the difficulty of certain stages. But because there's a really high chance that we'll end up using these eventually in another game, I don't want to say exactly what they are right now.

Mario's different power-ups were carefully balanced in the game.
Mario's different power-ups were carefully balanced in the game.

In answer to the next question about ice and fire, these were felt to be a little bit too powerful. They're very strong power-ups. That's certainly true for fire, because it's so easy to take out enemies. But for ice, it's sort of a slightly different matter because you didn't want to take out all of the risk. [For example], if you could stay in ice form the whole time, and walk over frozen water, it just wasn't quite as exciting as when you have the danger of that timer eventually running out and you falling into cold water.

So, unlike the other suits, ice and fire are considered to be power-ups, whereas the others have certain demerits or disadvantages to them [to force the player] to make a little bit of a judgment call. Making the player undertake a kind of nuanced judgment is very interesting. Whereas for the power-ups, you just can't leave them with that much power for too long.

For the bee Mario suit, that was in there, of course, to give the hover jump, like the functionality we were talking about earlier, and we ended up implementing the bee Mario and the boo Mario. I implemented the functionality, but the ideas of its appearance and things like that came from various members of the staff. For example, there was a female member of the staff who said, "You really need to appeal to female gamers more, and this bee illustration that we came up with is really cute. So, you should go with that." Whereas, similarly, the boo Mario, there was a staff member who took a boo toy and then stuck a nose and a hat on it, and was like, "You got to make this." And so, I was like, "OK, OK, I'll do it."

GS: Regarding the two-player game, had any thought been given--especially in the wake of something like the GBA and DS Mario and Luigi games--to some kind of actual two-player co-op to have Mario and Luigi in the game? Or perhaps why that wouldn't work.

YK: On the screen at the same time?

GS: Yes.

YK: Of course I'd like both Mario and Luigi characters onscreen, but I never had a chance for two people to control each different character on the screen at the same time. And this has always been sort of a big problem that I've tried to think of ways to implement this. When you have two people onscreen at the same time, they want to go in different directions. They have different ideas about what to do next at any given moment. You know, if one goes to grab a bunch of coins, then those are coins that the other player can't grab. And so sometimes both wanting to do the same thing, it might hold the other back, even though they want to do the correct action from a play standpoint. But having two people onscreen at the same time, you can't--it would be a big problem in terms of screen movement as well. You have the problem of, if you have a scrolling game, what happens when one player gets scrolled out of the screen? It's very difficult to deal with.

But when you can find a way that the screen is not a problem and the differing wills of the players are not a problem, then you've more or less come to a solution. And that's what I feel happened here where you have like a breakdown of roles where one person is controlling movement and the other person gets to pick things up and stomp enemies and shoot at them and stuff like that.

In this case, because each person has a slightly different role, they need each other, and they have to work together in a complementary sense. And that feels much better in terms of overall gameplay.

GS: This is your first time working on the Wii and doing a Mario game. What did you learn in the process? And based on how the game turned out and what you learned, how do you think the Mario series' core gameplay mechanics will evolve in future games?

YK: Things that I learned from making the first Mario on Wii, I mean, there's just so many it's hard to really start to pin down a few of them. But in's especially important to think about the fact that new hardware always has new challenges. And in this particular case, the challenge of Wii controls, was the fact that it had so few buttons. A lot of the people were actually anticipating that you'd have difficulty finding the right type of control scheme for Mario-style gameplay using a controller with limited button layout like that.

But because I had experience making Donkey Kong Jungle Beat, you know, that felt like a very good place to have learned a lot of lessons for implementing a simple control scheme into like very engaging gameplay, and doing so in a way that feels very new. So, I was very happy about that.

As for where will Mario series gameplay go next, I feel like I've just finished this journey with Galaxy, so there's still a lot of thinking to do about it at this point. But certainly the next project will involve some of the things that I had wanted to do on this project but for one reason or another wasn't able to include. Those things always get re-used eventually, and those ideas become incorporated.

He's been into outer space, so where does Mario head next?
He's been into outer space, so where does Mario head next?

But another thing is, I'm still in the period of evaluating Galaxy and how it's being accepted by players. Of course, the reviews are good and the sales are very good right now, but there's something to be said for looking at how people's impressions of the game change over time. [That kind of] calm, cool, sort of dispassionate look at the game is, you know, maybe a year or two out. And not only that, but I don't only work on Mario games. There's other franchises that I get involved with. But of course, as soon as I have access to a new technology, new ideas about different functions, everything starts to change layers at a time as they go forward.

GS: That's a perfect closer. Thanks for your time.

Got a news tip or want to contact us directly? Email

Join the conversation
There are 124 comments about this story