Let's try again with that earlier question from another angle - how do you make users feel uncomfortable but still keep them involved?

MT: Well, balance... How do I make them continue? It has to be good music, or more like as long as I continue to write music that users like; for example, to maybe motivate the user to continue by having them anticipate a new BGM if they reach the next boss battle.

And then how, within that framework of keeping them, do you make uncomfortable or jarring sounds that do not make them turn the system off?

MT: You must break through. Maybe you would call it a trial. Midsize bosses... how would you put it..? If you defeat him, you can move on to a comfortable place that just lies ahead. I want to do things like that. If you have nice music here, there are also uncomfortable moments, and then here you get good music again [gestures in an arc]. So in order to get here, you choose to move forward through a thorny path. As long as there's a fluctuation in emotions it's good.

For instance, in survival horror games, like Silent Hill, the sound design might make a player feel "Oh my god, I don't want to go over there!" but still you go, because you feel compelled. It seems a little more complicated than the emotion of hoping for something good in the future. Something about it makes you feel scared or uncomfortable but you have to figure out what that is, so you want to keep going.

Even in your own games it's more complicated than what you just said. It's a little deeper. And if it were an even pattern of bad, good, bad, good, that would be very predictable, but in Grasshopper games, it's like good, then surprise, and then maybe bad, then surprise, and then...

MT: That's true (laughs). If you have to name one example that's what it's like, but according to Suda's thinking itself, his main goal is to surprise users. So if you look at our project plans, gameplay is broken up by cutscenes and so forth - that's how Killer7 was made.

We, on the creating side, are often surprised ourselves. Because it's Suda's game there's that flow, and the flow of the audio also needs to... if there's a surprise here, you adhere here, sort of like that. That's what I think.

From the developer point of view, I know there's a surprise here, so the key is how to rouse up the experience and take it to the next level, so immediately before I would take the audio down a notch and stuff, to create a wave [gestures]. That's how I create the audio.

So it's like an algorithm of high and low points? I think maybe you don't give yourself enough credit by saying it's all Suda, because I've asked Suda about this, and he said to ask you!

MT: Is that so? It's true that writing music for Grasshopper games compared to writing music for titles under other companies like, God Hand, Earth Defense Force - and that goes for Beatmania too - is like using a completely different area of the brain.

With the audio design that you do for Grasshopper, with the wave and the algorithm and all, do you map it out? Do you make a chart of it as far as the user experience? I mean the figuring out of where to put things. And if so, do you make an audio map?

MT: Not at all. Not at all, but only for Grasshopper games, audio specifications aren't included in the project plan, so I don't know how many songs or what sound effects are needed until the very end. So until the game is complete, I have no idea how many songs I need to write.

Is all your sound designed to the game, then, or is the game sometimes designed for the music you've written?

MT: I'm shown the game in the beginning, during the concept stage when it's not yet confirmed for production. In response, an image of what I want to do and how I want to proceed is mostly formed during this stage.

You're talking about the design document stage?

MT: Yeah. After that I discuss it with Suda. I explain the relative idea and feeling, write a few pieces, try imposing them onto the game, and then Suda makes the final judgment. Recently, I've been doing whatever I want. First, I apply the audio to the game. Suda listens to the audio in the game and makes a call; whether it's a no-go or an OK, I have him decide.

Not often [do I make music before the game is created]. How do I put it... sometimes it's hard to determine whether the audio works, from the music alone. There are times when you can't tell if it matches or not, so I try to apply the audio to the game while it's running before making a decision.

It sort of reminds me of the relationship between Sergio Leone and Ennio Morricone. You know what I'm talking about?

MT: Yes, I see.

So then, with the sound effects, when you know, "Okay, this is a part we want to surprise people," and you're doing this to get to this point of that rise and fall of emotion for the player, once you have everything ready, is that something that you can chart? Because if you're working with a prototype, you have to know where this point is and how to get to that point right.

MT: About that, it's not just the sound effects, there's enemy placement, where the enemies are and stuff, these things also need to be consulted with and have the input of the level designer. So depending on the ability of the planner, a common pattern is where audio is also automatically placed in a similar fashion. So I don't [make a sound map]. I trying playing the sound and then determine what's needed where.