Yuji Naka – 2000 Developer Interview

—Naka, how did you get hired at Sega?

Naka: It was the middle of the Xevious boom then, and I had actually wanted to join to Namco. They were my top choice. But Namco didn't accept high school graduates. I went to an industrial arts high school so there were a lot of inquiries from various companies, but of game companies, only Sega and Taito solicited our school.

—Taito is a big one. Space Invaders and all...

Naka: I never got to play Space Invaders when I was a kid. When it came out, there were scary PTA Moms standing in front of and patrolling the game centers. I was too scared to go in! I was 23 the first time I finally played it. Before that I'd only played the PC version.

—Why did you decide on Sega, then?

Naka: Kazuya Nishioka, the writer of the One Key Progam corner of ASCII magazine, is the one who taugt me how to program. Nishioka was heavily involved in the promotion of PC sales in Japan, and is famous for creating over 500 computer games. He was a key figure in Japan's PC boom.

I was part of a manga fan club for Rumiko Takahashi and Hikaru Yuzuki in Kansai called the YTFC, where I met Nishioka. I used to go hang out at his house, and that was where I learned to program. Influenced by my friendship with Nishioka, I also started hanging out at this PC shop in Osaka. I made a lot of friends there. One of them was an older adult who always showed up in a suit and tie and briefcase. I asked him where he worked and he said that he worked in the computer data center of Mitsubishi Heavy Industries! I thought that was so cool, and I decided I wanted to work in a computer data center too. My first choice was CSK, and my second was NEC. Sharp was my third choice. Unfortunately, all three of those companies were based in Osaka. But I wanted to move to Tokyo, and only Sega had offices in Tokyo.

—You wanted to live in Tokyo?

Naka: I had a glamorous vision of Tokyo. I liked anime, and I wanted to experience living there. I had gone to Tokyo many times in the past to meet manga artists. I'd met Hikaru Yuzuki there, of course, and I'd visited Sunrise's production studio for Gundam and met the staff there. I also visited Studio Live and met Toyoo Ashida. My favorite animator was Yoshinori Kanada, who I'd also been able to meet in Tokyo.

—Wow, you got to meet so many artists! No wonder you loved it there.

Naka: I'd bought a lot of anime cels too. I had a big collection. You could only buy those in Tokyo. To an anime otaku like me, Tokyo was the promised land. I figured that if I applied to Sega, I could take their hiring exam in Tokyo, thus giving me another chance to visit.

—And did that happen?

Naka: No, they administered the hiring exam in Osaka. (laughs) Eventually I got a notice in the mail that I'd passed, and I was given a spot in Sega's Tokyo dormitory. Today Sega's offices are in a big building, but back then it was a hiraya [a long rectangular one-story structure]. Haneda looked a lot different then, too.

—Where we you initially placed within Sega?

Naka: Well, this is also embarassing. (laughs) I'd been bragging to my friends, "That's right guys. You're looking at a Sega developer now. When you go to the game center, don't forget to put a 100-yen coin in my games." But as it turned out, I'd been assigned to the personal computer department. When I saw that my eyes bulged, "WHAT!"

—What was that department? Were you working on Sega's PCs?

Naka: Sega did have PCs, yeah. Remember the SC-3000, with its attached keyboard? There was also the SG-1000. I believe the SC-3000 had just been released when I joined the company. The hiring ceremony was on April 1st, and by April 3rd they'd sat me down in front of a computer screen. They didn't teach me anything, it was just… "Here you are, now make a game. You said you could, right?" (laughs) That was when I started making games.

—Back then there was a stark difference between the capabilies of arcade games and computers, both in terms of what they could do, and the development work itself.

Naka: Yeah. The computers back then were way too primitive for arcade game ports. Arcade games used the latest cutting edge technology, so I was pretty jealous.

—You went on to create Girl's Garden at Sega. Was that your debut?

Naka: They told me, "As a training exercise we want you to make a game. If it's good, we'll sell it." I didn't think they really would, though. In any event, in a bid to increase the number of girl game fans, I created Girl's Garden. I believe it sold about 80,000 copies.

—Wow, for a debut, that's a massive hit!

Naka: I didn't have any training before I made it, so looking back now, the programming is really embarassing to me. I did have access to a programming manual, but I didn't even know how to compile programs then. I understood assembler, but I only knew about display and sound routines, nothing else… I didn't have the skills to program an entire game loop.

—After Girl's Garden, you worked on console games for a long time.

Naka: When the Sega Mark III came out, I created Great Baseball for it. This time my senior colleague taught me more about programming, and I was able to do the main programming myself. After that I worked on ports of F16 Fighting Falcon (1986) and Space Harrier (1985). I also made Hokuto no Ken (1986). They had said Space Harrier was unportable, but I persevered and did it.

—Now I'd like to ask some questions about the classic RPG Phantasy Star, which you helped create. Phantasy Star came out around the same time as Dragon Quest. The 3D dungeons, a result of your ingenious programming, are still talked about today.

Naka: The challenge was to overcome what we had not been able to do before due to low data storage capacity.

—But Phantasy Star used a 4MB cartridge. It seemed like a big undertaking for Sega?

Naka: Hmm, I'm not sure I would call it that. The Phantasy Star team only had five people, after all. That's tiny. I wanted Phantasy Star to be the first game with smooth 3D dungeons. I had played Wizardry, you see. I also wanted to make sure that when you turned a corner in the dungeon, the screen didn't abruptly switch, but instead smoothly rotated to the left or right. When I play Phantasy Star today it still does feel smooth, but it's also way too slow for me. The prototype I'd created actually went a lot faster. You could zoom through those caves, like a roller coaster! I remember showing it to people, "Get a load of this!" and their response was "I think I'm gonna be sick." (laughs)

—Ah, 3D motion sickness.

Naka: "Why is everyone getting dizzy?" It was my first time learning about 3D motion sickness. (laughs) Then I thought OK, maybe this would work better in an arcade game? So I went and showed it to the arcade development staff. (laughs)

—What do you think it was about you as a programmer that allowed you to succeed in programming this new technology, where others had failed?

Naka: Hmm, I'm not sure. Perhaps it was that I had studied a lot of other people's code very closely. Back then optimizing the speed of your code was a very important thing, but there's a limit to what you can learn independently by yourself. That's why I studied other people's code. I used to ask Yu Suzuki to show me his code all the time. Everyone has their own style or way of programming. I looked at a lot of different people's work.

Nowadays programmers have access to a lot of CPU power and richly featured development environments, and I think a lot of programmers today have become somewhat naive or complacent with how they code. That's why you have these consoles with massively powerful CPUs compared to the past, but games still have slowdown. They can trim things down and cut things out but their skill at optimization, to make it go faster, is lacking. Things were different back in the day, and maybe that was one of my strengths.

—Now that you mention it, a sense of speed might be the commonality in the console games you created.

Naka: Yeah. A feeling of dashing, rushing. That might be at the heart of what makes my games interesting.

—Eventually the Megadrive is released, and the turning point for you, of course, is Sonic the Hedgehog.

Naka: Yes, but a big part of what made Sonic possible was the work I did before that on the Daimakaimura (Ghouls and Ghosts) port.

—Daimakaimura was originally a Capcom game.

Naka: I first encountered it at an AM Show. I thought to myself, "This is amazing! How did they program this?" The map terrain is all uneven and bumpy, but Arthur somehow walks over it smoothly. Capcom's programmers are incredible. I went right away to our supervisor and told him, "I want to port this!" That AM Show was in November, and the next month, in December, we met Capcom to discuss porting it. I received their data and started programming, but it was 45MB. I had to somehow fit that into a 4MB cart. I remember muttering to myself "there's no way… this is impossible…" throughout the development, but I gave it everything I had. It was like a duel to the death! And by May of the following year, I had finished it.

—In only five months?! Incredible.

Naka: I had to simplify the color data, but I think the gameplay compares favorably with the original arcade version. It's something close to a "complete port", and it taught me a lot. And Sonic was an evolution of the programming I learned porting Daimakaimura. Like Super Mario Bros, in Daimakaimura you can only move 2 pixels per 1 interrupt. I didn't like the slow movement in Mario, so for Sonic, I wanted to find a way to raise the speed and allow you to start moving quickly from standing still. So I started doing proof-of-concept programming on how to move fast.

—You wanted the technical aspects of the programming in place before making the game.

Naka: Yeah. That's true for any game, but even if you have a neat idea, you can't just immediately start making it. You have to verify the necessary technical aspects, and once those are feasible, that's when you can say "OK, let's make this game." Likewise, for Sonic I spent several months creating a program with smooth fast scrolling, and only after that did we begin making the game proper.

—Sonic became a huge hit for Sega, but I understand you temporarily quit the company at one point around that time...?

Naka: Yeah. We made Sonic and finished the development around May, and I quit immediately after.

—Why?

Naka: I thought Sonic would be a huge hit, but Sega didn't have any faith in it. Also, my [monthly] wage at that time was only 160,000 yen [approximately 1600 USD]. It had been 9 years since I graduated high school! I felt like I had to do something.

—That is rough... at the time, there was a rumor going around that you'd been talking with Miyamoto Shigeru.

Naka: When I was heading home that day, I drove by the front of Nintendo's offices. I wanted to measure how far it would take to commute from my home to Nintendo. (laughs) When I got up to the gate, the security guard was eyeing me suspiciously. (laughs) I went home feeling kind of dejected. If I'd had more bravery that day, who knows what might have become.

—Your whole fate might have changed.

Naka: So I went home, and as I was heading back to Tokyo, I was contacted by Mark Cerny, who went on to produce Crash Bandicoot, asking me if I wanted to come work in America at a new company he was starting. Mark had worked at Sega in America for a long time. I promised I would, provided that I didn't have to make Sonic 2, but in the 2-month interim that I was waiting for my visa, Sonic had its big breakthrough in America.

—I've heard that Sonic was extremely popular in America then, on the Genesis.

Naka: Once I went overseas, it started to become clear that I was going to have to make Sonic 2. Even though I pointed out that my contract didn't say anything about that, I still ended up going. Mark was my boss, and 12 or 13 of us worked for a year to create Sonic 2.

—Wasn't the language difference an impediment?

Naka: Yeah, very much so. (laughs) I can't communicate nuances in English. Things that would be difficult for me to explain even in Japanese, were even more impossible in English. Ultimately, the development centralized around five core Japanese developers, and that's how the game was made.

—You worked in America for awhile after that, correct?

Naka: Yeah. We made three games in three years: Sonic 2, Sonic 3, and Sonic and Knuckles. I finally returned to Japan in August 1994, just before the release of the Saturn.

—You formed Sonic Team after returning to Japan, then.

Naka: The development team that made the first Sonic had also been called Sonic Team, actually. The name definitely existed then, for what it's worth. We had planned to make a Sonic game on the Sega Saturn, actually, but they told us "let's make something simpler first", and that's how Sonic Team began our first game NiGHTS (1996). It ended up not being "simple" at all though! It took two whole years. (laughs)

—Considering the experience you'd racked up and the crazy pace you'd been going at making games up to then, I can only imagine it took so long because the pre-development proof-of-concept work must have been very difficult.

Naka: I tried out many things, again and again, seeing how far I could push the Sega Saturn hardware. I don't know though. That was my last time working as a main programmer. I created all of the code for the player actions and the enemy movements, but it was not easy. And I kept getting sucked into meetings during the day, meaning I could only program at night.

—Now you have changed positions to that of producer, but I bet your approach to making games hasn't changed. First comes the technical experimenting...

Naka: That's correct. I spent a lot of time researching networking for Phantasy Star Online. Only after I felt, "OK, this could work" did we actually begin the project. I'd always wanted to make a game with online networking. I'd chatted with people on the computer over 15 years before, using a 2400 baud modem.

—Recently, Sonic Team was spun off from Sega and is now an independent company. The sense of responsibility must be even greater now.

Naka: We have to figure out what the market needs are. Recently game developers have dubbed themselves "game creators", and the people who create games are being treated like they're authors… but truth be told, I think our work has a stricly commercial base, and our games have to survive as commercial products. It's a bit different from someone who does oil painting as their hobby. That's why Sonic Team has a goal, at least for a time, of conquering every genre we can. I want people to trust the Sonic Team brand, to trust that we can create good games.

—When I play your games, it's strange, but I think of Shigeru Miyamoto of Nintendo. In response to Super Mario Sega created Sonic, and in response to Super Mario 64 they created NiGHTS, and in response to Ocarina of Time there's Sonic Adventure. Somehow, the direction of your respective games feels similar to me.

Naka: I've only ever talked with Miyamoto once, but I am jealous of him. He has 10 or 20 people that he's been working together with for 15 years… in my case, I have hardly anyone that I've worked with that long. We've had such a high turnover rate at Sega, it's felt a bit like I'm going it alone... I've had to believe in myself because there was no one else there.

—If you were told you could freely make games like an amateur, without any corporate or other responsibilities, what would you do?

Naka: Hmm, I don't know. Games are fairly censored, there's a lot of prohibited words and terms under the law, for example. I've never had someone tell me you can ignore all that, and hey by the way it doesn't matter if it sells or not.

—Ok then, what is the one game your heart truly wants to make.

Naka: I don't think there's anything like that in my mind right now. If anything, I'm extremely conflicted about my work right now. If someone asked me if, right now, I'm really making what I want to be making, I'm not sure I could say yes. It's hard to say. What I'd like to make right now is a race game. I love cars.

—Perhaps it wouldn't be any different than what you're doing with Sonic Team right now.

—Really?

Naka: I had always hoped to stay active as a programmer until I was 55. I recently heard that Bill Gates is both the CEO of his company, but also a technical programmer. I'm so jealous of that! I wish I could have a dual role like that on my resume… CEO / Programmer, or Executive Producer / Programmer.

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!