Kodama and M2 on Phantasy Star, what development was like in 1987, SEGA Ages games on Switch, lots more
–After listening to everyone, I can’t help but think how remarkable it is that this game was made in a sort of “inelegant” work environment, but the incredible talent of those involved led not only to a very special game, but the first game that launched a long-running, successful series. I’d say it’s only natural to want to port this game for people to continue experiencing it.
Horii: Absolutely. I felt the same way before, but after breaking it down like that, I feel it 110%.
It took a lot of talent to fit all that content into a 4Mbit ROM
–So, to sum things up: it seems that the reason for the delay to “SEGA AGES Phantasy Star” is because the code was consistently giving everyone a tough time.
Horii: Yes. The keyword there is “consistently”. So in addition to the auto-mapping feature and the bestiary, we also included a feature where your characters’ statuses are always viewable outside of battle. And even that took more time than expected. If these three things weren’t such a headache, I think we could’ve released the game 3 to 4 months earlier.
It took a long time, but I really have to thank everyone at SEGA for giving us approval to release this as an entry in the “SEGA AGES” collection on Nintendo Switch. Without their support, this couldn’t have happened.
Kodama: Well, we didn’t think it was gonna take THIS long, though! (Laughs)
Horii: Right?! I’m so sorry. I feel so bad! I’ll make it up to you! … Oh my god, this is so intense!
Everyone: (Laughter)
–The situation gets worse the more I think about it. After getting the approval, you realized you misjudged how much work it’d be and thought, “Oh no, what do I do?” You must’ve felt so anxious from all of that.
Horii: Let me take this opportunity to say that I’m not trying to make excuses. Generally, you put together some forecasts and analyze how things will go, but it doesn’t always go according to plan. No matter how prepared we were beforehand, we couldn’t have predicted that each process for each monster was totally different. If we had any idea that that was the case, we would’ve figured out something else to do.
–Indeed. I’m still really interested about mysteries behind the compression processes, though.
Horii: So, for example, if an image uses a lot of tiles or if there’s a lot of the same color in a row, there are different ways to compress those patterns to get the data down to a smaller size. I think Naka took each one into account and did them individually to get the most efficient use of space.
–And then all the small amounts of saved space started to add up, right?
Horii: Right, right. It seems like they were able to add a few more monsters in thanks to all space they ended up saving.
Kodama: That kind of thing happened quite often back then. For example, not me but another designer told me that they were asked to use a certain selection of colors because it could be compressed better.
Horii: Interesting! So, sort of the opposite situation, but that kind of thing happened, too, huh? That designer had to think about efficient data compression in addition to a color pallete that would display appropriately on televisions at the time.
Kodama: Exactly. They had to choose a color scheme that matched criteria for both.
–The way developers today tackle a project is different, but designing is still done on a technical level.
Kodama: We used to refer to an 8×8 pixel as one cell because if you inverted the image, it would actually take up less space. If you did that liberally, you would end up saving a lot of data.
Horii: But if you did it too much, the image would end up looking kind of cheap, then it would be obvious that you were trying to save space. If the lighting drawn into the object was very precise and you inverted the image, then it would look off. So it took a very skilled hand to make that work.
Kodama: In a large image, you could invert a certain spot here or there and it wouldn’t be visible to the naked eye. Then it would take up less space. I think we made relatively generous use of this method for the cut-scenes in “Phantasy Star”.
There are also some beaches in the game where waves come and go along the shoreline. It looks like that’s the case, but the colors are actually just changing in order to give off that impression.
Horii: If you change the color gradation of the blue used on the waves, it looks kind of like sand, so you don’t have to animate any images. You can make it look like it’s being animated with a simple change like that.
–So the goal was to make it look like it’s being animated, all while still saving data.
Kodama: We used to use those kinds of techniques all the time.
–What was the feeling like back then as you discovered these kinds of methods one by one? I imagine it was along the lines of, “Oh, this is great! Let’s do it like this!” Quickly followed by, “Oh, this is good, too!”
Kodama: Yes, it was absolutely like that.
Horii: Once they figured out something new, then that was the cutting edge way to do things. I think that’s how it happened with a lot of games. They even took measures to reduce the data on the fonts and colors for text. They thought of everything!
During the original development of “Phantasy Star”, Kodama’s manager asked her to play “Dragon Quest”, and on company time, too!
–So far we’ve discussed the original release at length, but something we haven’t mentioned is that at the time in 1987, SEGA was well-known for their Action and Arcade games. “Phantasy Star” was really the first RPG made by SEGA. It’s quite impressive that SEGA came right out of the gate with a home run like “Phantasy Star”.
Horii: Yeah! It was like SEGA was breaking into the home console market with a huge game like this!
–I’d say it was an impressive starting point that couldn’t have come together without SEGA’s extreme determination to make an RPG, wouldn’t you?
Kodama: Absolutely. It was probably Yuji Naka and Kotaro Hayashida, from planning, who held that conviction. I didn’t start working on the team until after the project started. The flow of pre-production generally starts off with planners and programmers having a meeting about what they want to do with a game. Such as, “We should put 3D dungeons in it,” or, “We should have monsters with animations.” Afer that, they would come to the designer to discuss these ideas and go from there. So in our case we had a meeting after they talked and they asked if I could design 3D dungeons.
Actually, not long before that meeting—and this is kind of weird, but… The manager of my department told me to play “Dragon Quest”.
Horii: Wow! He told you to play it?
Kodama: Yeah, he was so excited about it that he brought it to me. He was like, “You’ve gotta check this out! I don’t care. Go ahead and play it during work hours.” And I was so surprised that I had to double-check with him. Like, “Are you kidding me?!” (Laughs)
Horii: That’s crazy. So even members of management were fans of “Dragon Quest” and wanted SEGA to make an RPG.
Kodama: I only know what I know, so you’re better off asking someone like Hayashida about that. I think the boss of my department had just played the game on his own, to be honest. I’m pretty sure that there were some staff members that liked playing western RPGs like “Ultima” and “Wizardry”, too. So there are some roots of “Phantasy Star” there.
*NOTE: “Dragon Quest II: Luminaries of the Legendary Line” was released on the Famicom in Japan on January 26th, 1987 on a 1Mbit ROM. “Phantasy Star” was released on December 20th of the same year on a 4Mbit ROM.
–They probably took inspiration from “Wizardry” with their idea for 3D dungeons.
Kodama: I think so, too.
Horii: So, during pre-production, did you guys already decide to make these smooth, 3D dungeons?
Kodama: Yes, we did set out to do that. We wanted to do even more with the dungeon environments, though. But we ended up running out of data, so we had to give it up.
–Such as, changing the aesthetic of each dungeon?
Kodama: That’s correct. I remember working on a design for a dungeon that was made of rock.
Horii: Did you get to the trial phase of the design?
Kodama: Yes, we even had it working. But it looked like it wasn’t going to be possible due to data constraints.
Horii: I see…
Kodama: Actually, when I was working on “Phantasy Star IV: The End of the Millennium”, I sort of challenged myself to make a bunch of varied 3D dungeon designs. Unfortunately, we couldn’t get them to transition smoothly. We had a hard time without the aid of the 3DCG software, so we ended up doing a top-down view for the dungeons.
Getting back to the first “Phantasy Star” game, movement throughout the 3D dungeons had to be drawn by hand, but it was actually hard to tell that you were moving around when we made our first attempt. So Naka created wire-frames for the dungeons and printed them out for me. Then, I used them as a reference and drew new designs. It really helped me figure out how to depict the walls when taking a step forward. There’s a different sensation when moving forward like that. I worked on all the small details for a long time and eventually came up with our animations.
Speaking of, Naka tried out speeding up the wire-frame dungeons he created, and he was really impressed with the result. He was like, “We could make a Shooter like this!” It felt a lot like that last scene from “Star Wars” when they go to the Death Star, so everyone was pretty into it. (Laughs)
