After publishing our E3 2010 preview of Ghost Trick: Phantom Detective, we worried that we hadn't adequately captured and conveyed the game's bizarre concept. We're not the only ones left a bit speechless by the amazing Ghost in the Rube Goldberg Machine adventure -- ooh, that's a good way to put it! -- as even the game's director, Shu Takumi, thought it a challenge to explain his new mystery game to the "bigwigs" at Capcom.

Joystiq: When you first had the idea for Ghost Trick, you had to go to someone and say, "Hey, we should make this game." I want to hear what that first day was like, and what their reaction was.

Shu Takumi: Since I made the Ace Attorney series for so long, I was like, "Please let me do something new!" So I went up to the bigwigs and said this, and they said, "You go ahead and do what you want, we'll see what happens." The concept was really difficult to explain, so they were like, "Okay okay, why don't you make something and we'll see how it goes." So, that's what happened when I first presented the idea to make a new mystery game.

Now that you've come up with this unique and crazy idea, do you think the people in the marketing department might hate you? Because they have to, you know, sell this.

[Laughs] It's as you said, the promotion team and the marketing team are all like, "How are we supposed to promote this?!" I'm a very kind of pure person, and just made what I wanted to make since I have all these ideas, so I just kind of made it. "This is the idea that I had. I'm going to make this." And then, yeah, basically all the marketing people are like, "How do you expect us to promote this? Way to go." And I'm like, "Oh, I'm sorry." [Laughs]

I noticed that the animation in the game is of a very high quality, There are lots of unique animations with no repetition. For example, in the opening scene, a lady backs up against a fence and quickly glances back at it. Why devote so much effort to the animation, and how are you creating it?

First of all, in terms of how we made the motions, we did not use any motion capturing whatsoever. So the modeler actually went in there and made everything from scratch with polygons and from the ground up. We were able to put in a lot of those little nuances, a lot of those little details, because we were in full control of the actual motions and animations of the characters. One of the reasons we did that is -- not to disparage Western games or anything -- but a lot of the motions in Western games are very broad, and Japanese people are a little bit more sensitive to nuances and to little details. And so we paid really close attention to details, because that's what they find appealing about the characters, about the motions and being able to express that, and just as a user being able to view that kind of detailed, nuanced motion.



Looking at the Phoenix Wright games, you can say they're interactive novels with graphics and sound. Do you feel like Ghost Trick can be described in the same way, with its heavy emphasis on story?

In some ways, it's still interactive novel-ish, and a heavy emphasis on the scenario as you pointed out, but in a way, you really can't call it an interactive novel because we have so many more movements, the characters are very lively. And on top of that we have the puzzle aspect, you're not necessarily just reading a story anymore, you're actually interacting with the environment, you're interacting with the stages and solving these complex puzzles. We really can't say it's exactly the same, but you still have that great story element in there.

How did you go about designing the puzzles? Did you have a dictionary open and look at inanimate objects that you would like to place in the game?

[Laughs] The most interesting part is saving the different characters' lives or helping them out, but as you said, [when deciding] what objects to place in there, there was a foreign book that we looked at. It was almost like a picture book, it had pictures of objects and then had the name of them in there and we just picked them out. Whatever caught my eye. The title of the book is "Wordsworth" or something like that, I can't remember exactly. But it was through that each time, flipping through it and getting inspirations from that.

[Interviewers note: Wordsworth's Book of Words, perhaps?]

What's the weirdest object in the game?

[Takumi ponders for a few seconds.] The object I think is the most interesting is unfortunately that we cannot say, because it's a big spoiler for the game. But, we'll give you a hint: It's in the latter half of the game and it's ... small. So we hope players will look for that!

Is it a Ghost Trick game cart?

[Laughs] No, it's not a game cart, unfortunately.

The important thing is that you know you're going to fail somewhere along the way, so instead of failing at the end of your development cycle, the good thing is to fail at the beginning of your development cycle.- Shu Takumi

When you make these puzzles, you're very reliant on someone being able to figure out that they need to connect objects and use them in certain ways. How do you determine when a puzzle is too difficult? How can you predict that someone might not be able to make the right connection and be unable to proceed?

Balancing this game is obviously a difficult task. Generally, we have a bunch of other people play it -- people on the team, people on other teams, people in the company -- so we have a big sample and variety of people playing it and we get feedback from them. The ultimate test is the producer. Mr. Takeshita is like a super light casual user, he doesn't play that many games at all. [Hironobu Takeshita, who's also sitting at the table, smiles and shrugs.] So, when he gets stuck on a puzzle for more than two hours, that's when you know it's too difficult.

One time we had him play it on a different floor, so that nobody could give him hints, and all of a sudden he calls up and says, "This is too hard. I can't solve this!" So, when he gives that phone call, you know it's time to redo it. But, you know, we don't want to make it too easy either, for those really hardcore fans who love a good puzzle. We've made a good system out of it where there are hints throughout the game, but if you don't want to read it, you don't have to read it. So, it's up to the player. If they find it's too hard and they want to read a hint, they're more than welcome to read the hint. If it's too easy, don't read any hints. Don't bother clicking on it. We've made it so that it's a good balance between difficult and easy, and for people of all different levels, hardcore users and casual players.

And your number's going to be in the manual so I can phone you and say, "This is too hard, I can't solve this!"

[Laughs] No, we'll fix it again.

Are there any plans to release a demo on DSiWare?

Right now, we don't have any plans to make a DSiWare or a downloadable version yet, but we're going to wait and see how it goes. On the Japanese website, we have a Flash version and, when the time comes, we'll probably have a Flash version for the official English site. So, maybe players can play the Flash version.

After so many Phoenix Wright games, what have you learned from those and incorporated specifically into Ghost Trick? Something you might look back to now and say, "Ah, I learned my lesson back then, I'm changing this for Ghost Trick."

One thing I learned when I was making the Ace Attorney series is that it's really difficult to make something new. Like to make a new title or to make a new game. The important thing is that you know you're going to fail somewhere along the way, so instead of failing at the end of your development cycle, the good thing is to fail at the beginning of your development cycle. So, we made sure to work through as many concepts as possible and then we were like, "Okay, maybe this didn't work." So we went through as much as we could first, and had our big failure early on instead of at the end. So that was important. And the other thing was: if you fail more than twice, your team morale is going to be pretty low. So we tried to make sure while we were developing the game to keep everyone's spirits up and just to make the development cycle run smoothly.

Was there a big failure at the beginning of Ghost Trick development?

[Laughs] It was kind of like a wave, a little bit of failing here and a little bit of failing there, but we moved along pretty well.

How do you think it's going to be received by the U.S. audience?

One of the big things we learned coming to E3 is that with any new title, we've got to do a better job of getting it out there. It's so hard to tell people what the game is like, especially if it's a completely new concept and something unique like Ghost Trick. So, we'll definitely try our best to get the word out, do the promotion and things like that. As far as how it will be received, we think that obviously Ace Attorney fans will love this game, and we think that even if you're not an Ace Attorney fan, you should definitely give it a try. This game, because it's so puzzle-oriented, people who like puzzles will definitely find something interesting about this game. People who like mysteries will find something interesting about this game. There's no backstory so you don't need to have played the previous games or anything like that. We think this game should appeal to a variety of players.

Are we going to sit here, five years from now, talking about the new game after you've made a bunch of Ghost Trick games?

[Laughs] Well, we'll see about [Ghost Trick] two, three and four, but that depends on how well this game is received. We'll wait and see on that one, but we definitely hope that people will enjoy the game enough. Maybe we will make a two, three and four.

Ghost Trick: Phantom Detective is due on DS this Winter.

[Image: Ace Attorney Wiki]

This article was originally published on Joystiq.

Joystiq Podcast 143 - Mr. Scratch edition