Dragon Quest VI – 1995 Developer Interview

Yuji Horii - Producer/Writer

—When did you start making Dragon Quest VI?

Horii: After DQV was finished, that October we held our first meeting for DQVI, so it was about 3 years ago exactly.

—What did you discuss at that meeting?

Horii: It was a chance for everyone to share what they thought was lacking about DQV. We also talked about our ideas for a new gameplay system. We decided at this meeting that we'd bring back the job system from DQIII once again.

—What led you to that decision?

Horii: I personally felt that in DQV, we really emphasized the drama, and as a result, the player had less freedom. I thought we should have let the player go to more locations, and do more things. We had just ported DQI&II to the SFC, and for the first time in quite awhile, I got to play DQII again. I thought it was fun how many different places you have access to. That's why, for DQVI I suggested we craft a world that, while still retaining a dramatic story, allowed higher degree of player freedom and let you do things in the order you wanted.

—That sounds... very difficult to design.

Horii: Yeah, trying to avoid contradictions with these two opposing goals was tough. So early on, we went on a company retreat, and everyone shared their ideas. My tentative plan was to have two continents and a job system with some freedom in how you develop your character. I worked out the details with our programmer Yamane. After a lot more discussion, once it seemed like we had something interesting on our hands, I announced the official start of the development.

—At that time, you planned to release it one year later, correct?

Horii: That's right. But it ended up taking two and a half years. (laughs)

—What was it that caused the delay?

Horii: Partly it was the overlapping schedules of Chrono Trigger and the DQI&II port, but ultimately we just needed a lot more time to refine our ideas. The required memory kept expanding too. At first we talked about it being a 24Mbit game, but when we started running out of space that turned into 32Mbit. Even then we were short on space, so there were events we had to cut. I'm personally proud of how much we managed to fit in there though.

—How much written material do you generate when you're planning a game?

Horii: The dialogue alone takes up 9 volumes worth of 8cm-thick binders. When you add the map and data notes, it's probably more like 20 volumes. I can't believe we jammed all of that into one cartridge. Anyway, there were times I thought it would never end. The scenario alone took about 5000 pages.

—Do you ever end up confusing yourself, with all you've written?

Horii: Confusion, yes, in the sense that I will often forget things I've written. One time a debugger came to me saying "hey, if you remember this villager said this over here, so is this part ok…?", and I ended up having to write more so it all makes sense. (laughs)

—The scenario for DQVI is gargantuan.

Horii: Yeah, and we put a lot into the events this time around. Each time you clear an event, the town villagers' dialogue changes. Depending on the town it might change ten times or more! The reason we did that, is that we expect people will get lost in this game a lot. And when you're lost, your first recourse is to go back to the town and talk to the people there. So we tried to prepare a lot of dialogue for that eventuality, otherwise players would get bored, we thought. We want players to enjoy themselves even when they're stuck.

—What else did you pay special attention to for DQVI?

Horii: The yes/no dialogue choices. In previous DQ games, even if you kept saying "no" over and over, it would eventually force you to say yes. This time if you reject someone, they'll act quite surprised, but the game won't force you. In order to feel like you're really the one playing this game, your choices have to be as free as possible.

—Do your dialogue choices make a big impact on events down the road?

Horii: No, not to any large extent. We write the endings so that everything wraps up tidily in the end, you see. The choices make a small impact—maybe making things a little easier, or a little harder. However, there is one really big choice to make. We did our best to make it as big as possible. (laughs) I think it will get players' chests pounding. Anyway, we want players to empathize with what's happening on-screen, and feel like they're immersed in this world.

—The game world as a virtual reality space, in other words.

Horii: That's right. I think we did a good job there. Lately there's been a lot of automated events in RPGs. You control the battles, but during the events the characters act of their own accord. We wanted to get rid of that as much as possible here.

There is one thing I'm worried about though. For DQVI, if you don't use your head, you won't be able to advance or beat the game… so I'm concerned it will be too difficult for players who have become used to stories that move everything along on their own. If you can't enjoy that mental challenge here, of thinking on your own, it could be rough.

—In what other ways has DQVI become a harder game?

Horii: The bosses are noticeably stronger. At first, we gave the enemies about the same strength as the ones in DQV. However, DQVI has over 150 special moves, and enemies were dying right away whenever a player used a skill like seikenzuki or mawashigeri. We'd gone to the trouble to create all these nice battle animations, but they were dying before the animations could even complete! I knew this wasn't gonna fly, so I changed the monster stats.

—How long will it take to beat DQVI?

Horii: Around 60 hours. DQV had about 30 hours of playtime, so this is double. Depending on the person it could take even longer. There's also some fun things awaiting you after you beat it.

—Will there be a hidden dungeon like DQV had?

Horii: Like that, but powered-up. We've prepared something quite large for the post-game, and if you include that, the playtime could reach to over 100 hours. It's chock full of cool ideas, to the point that I wondered if it wasn't too much. Aside from that, if you really try to max things out, like getting 3 metal babbles on your team, or making ever character a Hero, you could spend a whole year on DQVI.

—What parts of DQVI are you most excited for players to see?

Horii: There's so many I'm hard-pressed to pick a few, but I think the hidden stuff is really good this time. It feels weird for me to say it like that though, "I want you to see what's hidden." (laughs)

—You must mean the post-game content. That makes me want to rush to the end.

Horii: Yeah, I don't think DQVI is the kind of game you can rush through. If you try to hurry, that means you're skipping over a lot of the dialogue. If you do that I think you'll end up stuck. I want players to take their time, knowing there's more awaiting them at the end.

—Is the story itself more difficult to play through then?

Horii: Yes, because one of the themes of DQVI is "discovery." We want you to find things out on your own. In a sense we've merely prepared a big world: what the player discovers there is up to them. Up to Dharma Temple, the rules are laid out clearly for the player, but beyond that point there's no rules. You're completely free. You could say that's where Dragon Quest VI really begins.

Manabu Yamana - Programmer

—Yamana, what was the hardest part of the programming for you?

Yamana: Probably the window system and party display. When we display a status or message window over the map and sprites in the customary Dragon Quest style, all the on-screen sprite data has to be temporarily stored in memory. Stuffing that all in was very difficult.

—Really? That's not the kind of answer I was expecting at all.

Yamana: If we could switch the entire screen (a la Final Fantasy), then it wouldn't require so much memory. That would make my life a lot easier, but it wouldn't feel like Dragon Quest then. That's why managing the display memory was the thing I struggled with the most.

—What are some of the "wow" moments of DQVI?

Yamana: There's a festival scene in the opening prologue. From this scene, I think players can get a strong premonition about what the rest of the story will contain.

—How are you hoping players experience DQVI?

Yamana: DQVI has been developed with freedom in mind; we didn't want to push a particular playstyle on people. Every choice is up to you… and every person who buys it will have their own unique experience with DQVI.

Koichi Sugiyama - Composer

—How would you describe your composition process for DQVI?

Sugiyama: For DQVI, I took a different approach from the other games. I first created a motif for one song, and then I used that as a core around which I wrote other pieces, a kind of repertoire built around one idea.

—What exactly do you mean by motif...?

Sugiyama: The simplest way to explain it is just a small fragment of a melody. For a concrete example, you know the famous part from Beethoven's 5th, "da da da dannnnn"? That's a motif. He uses it throughout the first through fourth movements, getting faster and more layered.

—So DQVI employs a similar kind of motif construction?

Sugiyama: Yeah. For example, there's an "evil" melody motif in DQVI that goes "ra~ so ra fa~". I made sure to use that in every different cave scene, since a cave is a den of evil. I use the "ra~ so ra fa~" motif in different ways, sometimes layering it, or dispersing the notes differently, or making it longer or shorter.

—Do you use it in battle scenes too?

Sugiyama: Yeah, since there's enemies, it appears in a modified form there.

—Is there a counter motif to the evil one, something bright and cheerful?

Sugiyama: There's a motif that plays in town scenes, which is a bright, peaceful melody. You hear it again in the casino, this time arranged in a swing jazz form, and in scenes with folk dancing, it becomes folk music. Also, at the beginning of the game, there's an ocarina melody that causes something miraculous to happen, and that melody is the "miracle" motif.

—How many songs did you make for DQVI?

Sugiyama: There's 29 long-form songs. There's a couple others which change the effects or tone slightly, so around 30 total. Those thirty songs are just the survivors, though: I actually created two or three times the amount of music for VI, songs that didn't get used.

—It sounds like it wasn't easy whittling it down to those 30.

Sugiyama: I narrowed them down myself, but it was very challenging. I had to think about how to arrange things so there weren't too many songs. Main programmer Yamana and I worked together to find the most effective use of a single song.

—Why did you need to remove so many?

Sugiyama: The more songs there are, the less of an impression each song leaves in the player's memory, and each song is made less meaningful. That's why I really wanted to make it like the musical Cats, where it's all based on one single melody. If you use the same melody, a person would only need to hear it once to remember it (even if they didn't want to). Well, that might be a little too extreme.

—What are the other key features of the DQVI music?

Sugiyama: We strived to draw out every bit of power and expression from the Super Famicom's music chip, and I hope players can sense that. If you just playback the samples straight on the SFC, it doesn't sound like real music. That's why, rather than fiddling around with technical stuff, I need real musicians to record actual performances, and we had a sound team assembled for DQVI to do just that.

—That's quite extravagant.

Sugiyama: The members are Tsukasa Tawada, who wrote the music for Ihatovo Monogatari, and Hitoshi Sakimoto, who composed the music for Tactics Ogre. These two are composers but they can both program, too. The three of us formed a sound team, and we set to creating sounds and samples to use. Yamana also created a specialized sound driver for DQVI at my request.

—Wow, you even made a custom sound driver?

Sugiyama: Yes. As a sound team we had all these requests for musical things we wanted to do. The sound driver was built to those specs. It produces a really nice tone. While DQVI still uses the Super Famicom's native sound chip, I think you'll enjoy how closely it approaches a real orchestra.

—Wow, I bet fans can't wait to play it now.

Sugiyama: I truly feel like we've achieved the best possible sound on the SFC. With these tones and textures, I've finally accomplished what I always wanted to do on this system.

—Then you must really want players to hear DQVI's music in stereo, then.

Sugiyama: Absolutely, yeah. If you can pump up the volume on a nice system, it'll be just like having a symphonic orchestra playing inside your Super Famicom.

—Please offer a final message to Dragon Quest fans everywhere.

Sugiyama: Writing good music that players enjoy listening to while they play, that's my ambition as a composer. Please, if at all possible, don't turn the music down low on your TV. (laughs) Also, try listening for the different motifs while you play. I want you to find the "discoveries" within the music of DQVI too! I'm sure it will enhance the experience for you.

Dragon Quest VI - 1996 Post-Release Interview

with Yuji Horii and Manabu Yamana

—Happy New Year! It's been one month since the release of Dragon Quest VI, and once again we saw people lined up in those mammoth "Dragon Quest lines" that wrap around the block. Many stores were sold out, too.

Horii: Yeah, I saw them out there! Nothing makes me happier than that sight. I wonder if everyone has finished the game by now…?

Yamana: Well, we hid quite a bit this time.

Horii: We made it so these secrets could be found, but yeah, after having put so much work into them… we want people to find them! (laughs)

Yamana: The stuff on the ocean floor is very hard to find though, isn't it? I think there's a few things that may be like that… like the place where you can change names.

Horii: You could change character names in DQII also, but when you press reset they would revert to the original names. In DQVI there's a temple where you can change the names of your party members, your monsters, even the bag… and they'll stay the same. However, if you input a dirty word for a name, the Naming God will get mad at you. (laughs) For those who want a good scolding, try it out and you'll see. It was embarrassing for the programmers to input all those checks. (laughs)

Horii: We're seeing a lot of variation between players, in terms of how much they explored. That goes for the legendary equipment too. Whether they found them all, or which one they couldn't find… that's varied from player to player.

Yamana: That goes for the events too. It really varies from person to person. When people would talk to me about the game after they beat it, I could tell which parts they did and didn't do.

—The AI is very good in DQVI, and it helped me out in the boss fights.

Horii: That's funny, because midway through the development we had people complaining that the AI was too smart. (laughs)

Yamana: Some of the staff asked us to make the AI more stupid, but I said that wasn't really possible. (laughs)

Horii: The issue is that the AI quickly identifies the enemy's weakpoint. So we added an "AI Resistance" parameter on the backend. The AI is great in a 1v1 scenario, but when teamwork or coordination is required, it's a bit weak.

Yamana: Once we added that AI resistance parameter, relying purely on the AI became impossible. There's also bosses with 3 parts, and the AI always tries to take down the strongest looking enemey, but that can be a problem in boss fights, where if you don't strategically kill the other parts first, you're in for a bad time. (laughs)

Horii: Yeah, in those cases relying on the more broad-minded human intelligence is the decidedly better option. I do think it's pretty sweet though, that you can use the AI to find the bosses weakness, and then turn it off and devise your own better strategies. We just wanted to make sure we avoided a situation where, for example, the human player can't beat the last boss until level 40, but the AI can do it at level 35. (laughs) Anyway, it's really cool how the AI can accurately choose the right skill and attack from among so many options.

Yamana: Yeah, it'll use mawashigeri (spinning kick) on enemies that are likely to dodge seikenzuki (knuckle sandwich), for example. Normally that decision-making takes the CPU a very long time.

Horii: It happens so fast in DQ6, you can't even notice it. It's very well-done, honestly.

Yamana: Up to now, there's only been about 30 skills in a Dragon Quest game. But DQVI has over 150. At first, it took about 6 seconds of time for the CPU to calculate what the AI will do, which meant 24 seconds total for all four players… it was completely untenable. By the end, we sped it up immensely.

Horii: We were discussing a hacky solution that would have prioritized your current job skills in the AI, but eventually Yamana optimized the AI routine so that wasn't necessary.

Yamana: I remember that period, the path ahead looked completely dark. We had no idea what to do. (laughs)

Horii: We faced a similar dire situation with the memory… four or five times, actually. (laughs)

Yamana: I remember being at one of Sugiyama's concerts in late August, and just sitting there worrying about the memory situation the entire time. (laughs)

Horii: We fit an awful lot in there though.

Yamana: We really did. When people actually play the game, I think they're going to be impressed by how packed it is.

—Well, it did take three years.

Horii: Dragon Quest is very much a "handmade" kind of game. You could even call it a homemade craftwork. (laughs) We have a small staff too compared with other companies, there's only two main people working on the graphics. For the music, it's just Sugiyama and his two assistants.

Yamana: The programming is five or six people, and the scenario is written by Horii and a few other members. A 100-person team would be a lot, but for Dragon Quest we've only got about 20 people.

Horii: I think the smaller teams mean that everyone gets to see everything, which lends a sense of overall unity. If we split it up into multiple smaller teams it would fragment that unity. The tradeoff is that it always takes us a long time, everytime.

—As we'd hoped, the dungeons in DQVI are chock full of inventive tricks and traps.

Yamana: It was huge pain for me as a programmer. The Moonmirror Tower gave me a lot of trouble. The reflections change depending on the party composition, and there's some that don't reflect too, it was all very confusing. (laughs)

Horii: That part is very well done. That was an idea we had since the Famicom days, a "mirror, mirror" scene like that, and we finally realized it.

Yamana: I think the best dresser contest came out great too. That part ate up a ton of memory too.

Horii: The drum sound there is perfect.

Yamana: We added that drum sound at the very last minute, but it came out better than I'd expected.

—In the pantheon of Dragon Quest games, where do you see DQVI?

Horii: It's the one that's got it all! Drama, job changes, recruitable monsters… everything from the previous games. Fans of the job system in DQIII, fans of the nail-biting challenge of DQII… it's all here in DQVI. Everytime we finish a game there's always some regrets, but DQVI probably has the fewest of all the Dragon Quest games so far. I'd say we accomplished 95% of what we set out to do. Partly that's explained by the amount of time we spent on it, of course, but it's the most complete Dragon Quest to date.

If you've enjoyed reading this interview and would like to be able to vote each month on what I translate, please consider supporting me on Patreon! I can't do it without your help!