Star Ocean 3 – 2003 Developer Interview
originally featured in the official Star Ocean 3 strategy guide
Yoshiharu Gotanda – Director
Hiroshi Ogawa – Planning Director
Masayasu Nishida – Planner
Shigeru Ueki – Programmer (Field)
Yuichiro Kitao – Programmer (Battle)
—I’d like to start with some general questions. —At tri-Ace, what things are important to you when you create a game?
Gotanda: In the past it wasn’t like this, but today fans can reach out to us much more directly and tell us what they like/dislike, what they want and don’t want. But if we are too deferential to their ideas, it can sometimes be paralyzing to us as developers. To take difficulty as one example: there’s people who say our games are too easy, and there’s people who say our games are too hard. If you don’t have confidence in yourself as a developer, and what you are doing, then it’s easy to lose your way trying to accomodate players’ different needs. You end up with a game that is neither fish nor fowl, a compromised work. So we try to make a game that, first and foremost, we find fun and interesting ourselves. It’s kind of obvious, I know, but that is our basic stance.
—This may just be my personal conjecture, but with SO3, I get the strong sense that tri-Ace was attempting to make a game for the hardcore fans. How has the response been from the different segments of your fans: the backstory/character fans, and the ones who like to grind and really get deep on the gameplay system?
Kitao: We didn’t start out with that intention, of catering to the hardcore… (laughs)
Ogawa: Yeah, I would say it’s more a result of the developers themselves, and our tendency to fuss over the details. We stuff so much in there, it can’t help but appeal to some hardcore fans. (laughs) But no one will remember your game if it’s too difficult to even complete, so we try not to create a gameplay system that’s too complex or shuts out more casual players. We try to balance it so that as long as you level up, you should be able to advance.
Nishida: By the latter half of the development everyone is eager to add this and that, and it often happens that we’ve gone and added too much.
The Star Ocean 3 developers: (L-R) Masayasu Nishida (planner), Hiroshi Ogawa (planning director), Yuichiro Kitao (battle programmer), and Shigeru Ueki (field programmer).
—From what you’re saying, it sounds like each of the individual developers at tri-Ace is given a lot of freedom.
Ogawa: It’s true that, if you give people a certain amount of freedom, they’ll do their best to add whatever they can. We’re pretty lenient, and you don’t hear a lot of outright “No!”s at tri-Ace. If there’s an appeal in what you want to add, we’ll try and find a way to add it. We don’t have a lot of preset restrictions. We’re not that kind of company.
—The battle trophies, the terminology glossary, the way NPCs dialogue changes throughout the game… you can really feel the spirit of fanservice in your work. Why do you think that level of detail is necessary for your games?
Nishida: Personally, I just have an easier time coming up with dialogue for NPCs, rather than trying to create some personality for them out of thin air. I think this approach also allows for more change in the characters: instead of having a set personality, they can evolve over time with the story.
—How do you actually go about giving them personalities, then?
Nishida: Basically, we figure out how old they are, who their family members are… I also have a habit of including pointless details. (laughs)
—I like how casually you just described something that, frankly, sounds like an insane amount of work. (laughs) Do the rest of you feel the same way, with regard to your part?
Kitao: More or less. (laughs)
Ueki: The same. (laughs)
Ogawa: We have no expectation that the typical player is going to complete all the Battle Trophies. We don’t add all this content with the idea that everyone has to be a completionist. It’s there for those players who do want to play a little more… though I’m not always sure whether that’s a blessing or a curse we’re giving them. (laughs)
—Well, before I get down to the questions about SO3, I wanted to confirm something… the story was written by Gotanda, correct?
Ogawa: That’s right. For the Star Ocean games, Gotanda always comes up with the basic story himself. Likewise with SO3, Gotanda first came up with the rough outline of an idea, then there were revisions, and finally it was worked into a single story. Basically, Gotanda thinks up the general world/backstory info, then Nishida works it over and fleshes it out.
—Is Gotanda responsible for the strong sci-fi influence?
Kitao: Yeah. He’s something of a trekkie. (laughs)
Nishida: And to that end, I had to do a lot of studying up myself! There’s a whole lexicon and vocabulary to sci-fi, you know? Special physics concepts and such. I had never read much sci-fi before, so I had to do my due diligence and learn all about warp drive, laser weapons, futuristic energy sources….
—It sounds like a lot of work goes into a Star Ocean game behind-the-scenes.
Nishida: I really like that work, learning about the details of all this sci-fi physics. But… I’m sure there are players out there who know way more about it than me, and will spot mistakes!
Kitao: I don’t know. This is all fantasy, so in a sense, you can just say “it’s all magic” and be done with it.
Ogawa: Yeah. After all, if you do try to explain everything, there will always be those fans who find a bone of contention “Oh, that’s not how warp works, it’s actually like this.” I don’t mind those kinds of debates over the minutiae, but I am afraid of making any big blunders!
Nishida: Yeah. Even with things like symbology magic and the like, there’s a danger in overexplaining it. (laughs)
Reality and the Fourth Dimension
—I’d like to turn to the story of SO3 now. What are your feelings about it?
Ogawa: …the very first time it was shown to me, to be honest, I felt like there were a lot of things I would personally do differently. The Fourth-Dimensional Space was one aspect of the plot where opinions were particularly divided at tri-Ace.
Kitao: As a programmer, I didn’t have a very strong opinion about the story one way or the other. It’s more like, “ok, so this time it’s going to be like this.” I did talk about certain concerns with Ogawa, though.
Ueki: I’m… also a programmer. (laughs) My first job was the programming for the message system, and focusing too much on the content of the messages would have been a hindrance to my work. We programmers must stay focused on the programming! (laughs) So I basically kept my head down and worked on making dungeon traps, lots and lots of dungeon traps. (laughs)
—In SO3, we learn that the universe the characters live in is really just a “net game” simulation. Was that plot point something you had from the beginning?
Ogawa: That was one of Gotanda’s initial ideas, yes. I worked with him on it, offering suggestions and fleshing it out. Other ideas we had from the beginning were the Fourth-Dimensional Space and Beings, as well as the roles of Maria and Fayt.
—Was the “net game” idea included because of the current boom in MMOs?
Nishida: No, I don’t think they’re related at all. I think it came more from Gotanda’s interest in science fiction.
Ogawa: Gotanda started thinking about the story for SO3 three years ago, actually. Back then, online games weren’t very popular yet in Japan.
Kitao: I’ve also never heard that Gotanda plays any MMOs or online games.
—So this wasn’t some pilot idea for a Star Ocean MMO? Like getting people used to the idea?
Nishida: Not at all. Net games are really annoying, anyway. (laughs)
—Don’t you think that SO3, by declaring the universe to be just a simulation of reality, kind of upends and trivializes everything that the previous games had built? Did any of you feel, “what a waste!”
Nishida: No, not at all.
Ogawa: I think a lot of it is up your interpretation, really. From the start, we were concerned that some players might be hurt by these revelations. That’s why we deliberately tried to leave it a little ambiguous, and leave room for players to take it how they like.
—Well then… let me ask some questions about it! First, is the world of SO3 in the same universe/continuum/timeline as SO1 and SO2?
Nishida: Yes, it is.
—Then, is everything that happens in SO1 and SO2 also part of this net-game simulation?
Ogawa: Well, there’s the rub. From the Fourth-Dimensional beings’ perspective, and what they say, yes—but to the people of Eternal Sphere, it’s their reality. Their individual feelings and experiences are real to them.
—Is that connected to the ending? In the ending, the Creator vanishes, but the people, who as mere data should also have disappeared along with him, somehow remain in existence. Isn’t there something weird about the whole world continuing to exist…?
Ogawa: That’s one way of looking at it. But if we make a sequel, we won’t follow that interpretation of the ending. Roughly speaking, if one establishes an individuality and selfhood, then a world for that selfhood must also exist. That’s kind of what we were going for.
Yoshiharu Gotanda, director.
—Gotanda, if I may ask you directly, why did you choose “net games” as a theme for SO3?
Gotanda: Actually, the important idea in SO3 wasn’t “net games” per se. Rather, it was the idea of beings from another dimension who control us. In religious terms, it asks the question: what would you do if an entity who called itself your Creator suddenly appeared before you? And what if our world was nothing more than the technological creation of another civilization? You don’t even need to call it a “net game”, really—a “world simulator” would do just as well. But since “net game” is an easy term for players to understand in today’s world, we called it that.
—What gave you this idea? After all, the worlds of SO1 and SO2 were already complete…
Gotanda: The idea came from thoughts I’ve had about our own universe, how it might actually be like that. It’s not something we can prove with our current level of science, though… I guess it’s just one of my many theories about the nature of our universe. And I based the universe of Star Ocean, at least in part, on the real world we live in.
—I’m curious about what you were trying to say, with this idea.
Gotanda: As I said, in my mind, the universe of Star Ocean is partly an extension of the real world, but with some fantasy elements mixed in. That’s one reason why the Star Ocean universe includes Earth, to help strengthen those connections. Though it seems like I didn’t really succeed there. (laughs)
Star Ocean is a place that exists halfway between Fantasy and Reality… so it allows me to play with interesting cosmological ideas. If this world really were just a game, just data in a computer, how would that change the meaning and value of our own existence? I don’t know that it would. It’s not the container that is important, but the thing within, right?
Anyway, I’m not sure I have specific answers that I wanted to convey to everyone with this idea. It’s more like I wanted to present questions, and a certain way of thinking about it. (laughs)
The Battle System
—Thank you for the explanation. Well, I guess there’s not much more to say in this interview now! Hah… actually, if you don’t mind, I’d like to move on to part 2 of my questions, about the game systems. Regarding the battle system, it seems you were very cognizant of the FTG game genre when you made SO3?
Kitao: For better or worse, tri-Ace has become known for our real-time battle systems. SO1 and SO2 had them, and while Valkyrie Profile was a rare turn-based game, it too featured a timing-dependent special move system. In the beginning, the combat system for SO3 was more strategic—the action wasn’t as strong. That’s because we were basing a lot of it on the system from SO2. We weren’t initially planning for it feel like a FTG game, but as we developed it, with the addition of air combos and the quickened pacing, it ended up turning out that way.
Ueki: I think the fact that it was full 3D now had a lot to do with it.
—The way you can dodge enemy attacks and fire off special moves feels a lot like a FTG game to me.
Ueki: The person who handled the battle systems loves versus FTG, and he told me he wanted to try making a game like that.
—Having all these different attacks: quick attacks, strong attacks, the fury guard system… were you worried it would be hard for players to understand?
Kitao: Yeah. When we first exhibited SO3 at a game show, players only used the quick attacks, and when their guts reached 100% and they were shielded, they were confused, “huh, I’m not getting hit?” We didn’t add a tutorial until later in the development, you see.
—You’re referring to the Battle Simulator?
Kitao: Yeah. Another difference is that, in the beginning, you couldn’t see the enemies’ Guts gauge. It made it very difficult. And the way the fury guard worked wasn’t very clear either. But as a remnant of that early version, for the difficult Fourth Dimensional enemies, we deliberately made their guts gauge invisible to players.
There was also no Bonus Battle Gauge in the beginning, either. It was a different gauge there.
Kitao: It was called the “Special Gauge.” You built it up by using special skills, then when it was full you could unleash powerful attacks and abilities that would turn the tide of a battle. We cut it halfway into the development though. After that, we couldn’t decide on what else to do for awhile, until someone came up with the idea of adding a bonus system.
—I remember playing and thinking how fiendish the Bonus Battle Gauge system was. (laughs) Whenever you reloaded your game the gauge would be reset, so when it was full, you felt like you shouldn’t quit. And if it breaks, then you lose all will to play: “…screw this, I’m going to bed.” (laughs)
Kitao: Yes, it’s very convenient, but it can piss you off just as easily. (laughs)
—Was that all as you planned?
Kitao: Let’s just say, we were all grinning to each other when we came up with the idea. (laughs) We had to do it this way, or it would be too easy to reach bosses with a full bonus gauge. And if we allowed the bonus gauge to carry over between saves, it would again suck away all the tension and excitement. I mean, in a sense that’s what it’s supposed to be… an “excitement” gauge.
—The way elemental attacks work is also much easier to understand in SO3 than it was in SO2.
Kitao: Yeah, it’s just earth, water, fire, wind, and void. In the Aquios Library you hear about a “light” and “darkness” element, but it’s not a part of the actual game system.
Ueki: Early in the development, we had them. But at one point, we were actually going to remove the entire elemental system…
Kitao: SO2 had all these different elements, plus you had to factor in the absorb and null mechanics. Most players just overlooked it. We talked about removing the elements altogether for SO3, but ultimately decided it would be best to keep them.
—What about the environmental obstacles/hazards in the battle fields? What happened to those?
Kitao: In SO2, you had pillars that would fall, and handcarts that you could move. We had originally planned to include those things in SO3, too—using the mine carts in the Becquerel Mines, for instance. Unfortunately, due to time constraints, those ideas were abandoned. (bitter laugh) Boulders on the battlefield was the only thing we did.
The Role of the Programmers
—Ueki, I understand you programmed all the dungeon traps and gimmicks… were there any especially challenging parts of that work?
Ueki: It was all a challenge. (laughs)
—Are there any parts you’re particularly proud of, then?
Ueki: Hmm, for dungeon gimmicks, I like the mechanics of the disintegration equipment (the ring/bomb/hammer), how they can remove obstacles in the dungeons and open up new paths. Creating and mapping out all that took half a year.
—When you remove something with those, the way you handled the physics, they break apart differently everytime.
Ueki: In reality you have other factors like air pressure, but adding those would have been much more difficult, so I just added a little randomness into the way things crumbled. Originally I wanted to have more complex physics, though. Of course, had I done more accurate and precise calculations there, to the player it would have appeared that they were breaking apart the same way every time, and that would have been boring.
—I see you spent a lot of time on this mechanic, even though it’s something players will never directly see or notice.
Ueki: Indeed. I really worked hard on it.
Kitao: It took half a year just to make it look natural. When we handed it over to the planners, we said, “ok, here’s how the disintegration mechanic works. Now place the treasure chests accordingly.” But we weren’t really on the same page there—their way of thinking about dungeon design is different from ours—so at first they kept putting the chests in these absurd places… like inside boulders. Their designs surprised us.
Ueki: Yeah, for example, in the Cave of Trials, there was a place where you disintegrated an obstacle and all these treasures chests would come tumbling down. But depending on how they fell, some of them might be unrecoverable… that wouldn’t be any fun for players, so a number of redos were required.
—Another thing I’ve noticed with tri-Ace’s games: they all have very fast loading times.
Kitao: You can’t have players waiting for every fight. Our basic approach is to do all the data reading and loading while the memory loads.
Ueki: We definitely paid careful attention to loading times. When we first exhibited SO3 at game shows, we went around and compared our loading times to other companies’ games… (laughs)
Kitao: We all feel it’s important, but to Gotanda it’s a top priority. It’s the number one thing that stresses players out, and his goal is a stress-free experience.
Ueki: Yeah. We try to avoid having the screen go all black during loading, since psychologically, that makes it feel all the longer. We also try to program it so the loading is done in advance. For SO3 we’ve got both DVD and hard disk loading. But the loading speed isn’t really all the different, even if you don’t have a hard disk. It’s all very fast.
Kitao: Yeah, we pushed the PS2 hardware to its limits there.
—Other than the load times, what other aspects of SO3 did you focus on, as programmers?
Ueki: I think each of us has different things, but for me, it was just making sure the source code was neat and organized. (laughs) But really, having a logical and efficient code is central to everything. Without that, there’s no chance of anything else being fast. We aimed for a 60FPS framerate for SO3, so we worked very hard to make sure there was no slowdown… that was the biggest challenge of all, actually.
—Yeah, it’s kind of amazing how little slowdown there is in SO3.
Ueki: Naturally, if there’s too many special effects onscreen, the game will still slow down…
Kitao: We tried to make it so it didn’t stand out to players, at least. We were ok with dropping frames here and there, but we worked hard to make sure there would be no actual loss of speed.
—I’d like to ask some questions about the characters now. Starting with Fayt, what was your concept for him?
Ogawa: He’s an ordinary college student. We didn’t want to make a “superman” character. Fayt is an overachiever, and we worked hard to distinguish his character from the SO2 protagonist, Claude. Whereas Claude was somewhat of a hothead, we tried to make Fayt different when we wrote his dialogue.
Nishida: If you look at specific things, he has a lot of similarities to Claude. But we had to make him different somehow… it was very hard. (laughs)
—It only happens once, but in his awakening scene, Fayt has those angelic wings… what was the meaning behind that?
Nishida: Robert Leingod reasoned that if Fayt was going to fight a god, he would need power similar to a god. An angel is close to his image of what a god is. Fayt acts as the “man-made” god, in other words.
Ogawa: We sure use that wings motif a lot in tri-Ace games, don’t we? (laughs)
—Players’ opinions seem to be really split on the character of Sophia. Some think she is cute, and others find her annoying.
Ogawa: Even at tri-Ace, there were always people who didn’t like her. That was the background we intentionally gave her though: Fayt’s clingy, somewhat needy childhood friend. (laughs)
—Forging ahead despite their concerns… I like that. (laughs) One thing though: while Sophia has the important plot position of being the main character’s childhood friend, as to her own personality… I have to admit it seemed kind of vague to me.
Ogawa: She does have her passions, but she is essentially just an ordinary girl. I think “vague” is, in one sense, a fair description. That was part of our intention: as a character, we didn’t want her to have any special or unusual things that stood out.
—Then what would you say makes Sophia, Sophia?
Nishida: Basically, she remains a normal person to the end. We didn’t want a character who, in the course of all the fighting, gradually becomes numb and used to it.
—At the beginning Cliff seems very apathetic and irresponsible, but he turns out to be quite the opposite.
Ogawa: He’s actually an excellent person. He did start Quark, after all.
In a survey of 2500+ players, Albel was voted favorite character.
Sophia finished with an unpopular 5%, for reasons alluded to above.
—I thought it was interesting that Cliff acts and dresses very young, but he’s actually 36!
Ogawa: I can’t speak for his designer, but we didn’t want to give him the personality of your typical old guy. He doesn’t look or act his age, but that’s fine. He’s young at heart.
—It’s pretty unusual to see video game characters that age, isn’t it? You do see the “fatherly, experienced” trope pretty often, but characters over 35 still seem very rare.
Nishida: We handle him more like he’s in his late 20s, though. A lot of players might see him as the mentor figure for the heroes, and that’s cool.
—Now to the character I’m most curious about… Elena. Is she from the Fourth Dimension? She clearly seems to be holding back, and not sharing all she knows. Is she a player controlled by someone in the Fourth Dimension? Or is she an employee of Sphere, like a special character who manages the gameworld? Or is she an in-game persona from another world?
Nishida: Um… those are all correct in a way. (laughs) She is connected to Sphere, but she’s not working for them at the moment. She’s doing her own thing. Like Luther, she isn’t a character created by someone: that’s her actual self, but she just lives in the Eternal Sphere 24/7.
—So, let me get this straight then… there really was an actual person named Elena?
Nishida: Yes. But if I say more, well, I’ll be telling you my personal opinion… not what we wrote into the game.
—By all means, please share.
Nishida: There was a woman named Elena in Sphere. She was a developer. There’s dialogue in the game that alludes to this. Due to an unfortunate accident, the mental projection system somehow failed and she was trapped in the Eternal Sphere. So she has no way of really using all she’s learned in her real life, and she lives out a kind of meandering, purposeless existence.
—The other mystery character I wanted to ask about is Welch… where in the world is she from?
Ogawa: Ah… I don’t think we can talk about Welch and the Guild Master. For now, let’s just say they’re a mystery. (laughs)
—Luther is ultimately the game’s final boss, but can we also consider him a normal human being?
Ogawa: Yes, he’s just an exceptionally talented programmer. He’s a pillar of the Sphere Corp., and its President.
“It’s not the container that is
important, but the thing within.”
—What was Luther trying to accomplish?
Ogawa: All he wanted was to remove the foreign elements from the world he and his corporation had created. He had no idea what effect those elements would have on the system as a whole and the other players. Plus, if some characters can change their own parameters, wouldn’t that ruin the balance of the game? So what he was doing was natural as a game/systems administrator.
—At the end of the game, was Luther actually inside the Eternal Sphere?
—If Luther had so much control over everything, why couldn’t he delete Fayt and the others?
Nishida: To the Eternal Sphere, Fayt and the others were like a very bad virus: self-replicating, self-preserving. Luther designed programs to attack this virus, but that wasn’t the correct way to get rid of it.
Ogawa: This is a bit of my personal interpretation, but I think Fayt, Maria, and Sophia were like special “bugs”—changes to the program itself—created by Robert Leingod. If you don’t find and fix a bug at the source, it won’t matter how many times you try to delete it. Luther didn’t try to fix it, he simply tried to quarrantine and cordon it off from the rest of the program.
—Unlike Fayt, did Luther disappear for real…?
Nishida: Yes, you can think of it that way, that he vanished along with the Eternal Sphere. There’s no chance of him being revived.
—Gotanda, what parts of SO3 are you really satisfied with, or that you think came out really well?
Gotanda: How about I tell you some of the things I’m unsatisfied with instead? (laughs) Due to various circumstances, I think we didn’t give enough explanation for all the different science fiction concepts and vocabulary in the latter half of the game. We thought it would be too pedantic and verbose to explain everything in-game… it was difficult. And from the beginning, we thought players might be used to science fiction settings now, so maybe we didn’t need to explain everything. But I think the reality is that they’re used to space settings, but not necessarily familiar with all this science fiction terminology, so that was a failing on our part.
This was also our first time on the PS2 hardware, our first time doing a full-3D game… many firsts. We aren’t completely satisfied with a number of things, which is too bad for this game, but we’re committed to improving on them in the next installment.