Tomohiro Nishikado – 2000 Developer Interview
As the creator of the revolutionary Space Invaders, Tomohiro Nishikado could rightly be called the Father of Japanese video games. This long interview from the jp book “game maestro” covers Nishikado’s early years at Taito as an engineer developing sophisticated (for their time) electro-mechanical games before getting into the details and challenges of designing Space Invaders. As a first-hand account of those earliest years of gaming, it’s a precious historical document.
—I’d like to start by asking some general questions about your early days. What kind of company was Taito Boueki back then, around the time you joined them?
Nishikado: Jukeboxes were their main business, but they had a small subsidiary company that did game development. They made what we now call “electro-mechanical” games. But almost all of them were copies of overseas games… not literal copies in the strict sense, but in those days, they made their own versions of whatever was popular overseas. No one was making originals yet.
—You had worked as an audio engineer up to then. What made you switch gears and join Taito Boueki (Taito Trading Company)?
Nishikado: After I quit my job, I loafed around for about a month. I had this acquaintance who I’d sometimes meet up with at a nearby train station. He was my senior colleague from my old audio company. One day I asked him what he was up to lately, and he told me he was now working at this company called Taito, making games. I knew about those game machines—I had seen them installed on the top floor of department stores and malls when I was a kid. They were simple driving games that used a belt track for the course, and you’d steer your car, trying not to go off-track.
This work sounded interesting, so I asked him more about it, and he ended up inviting me to come work there. He told me they really needed engineers. I had actually just accepted another offer at a communications company, so I wasn’t sure what to do. But my friend was insistent that they needed help, so I went in for an interview and ended up accepting Taito’s offer instead.
—What was the first game you worked on at Taito?
Nishikado: The first work I did was on a game we imported from America, a gun-target game called “Ghost Gun.”
—Ah, a lightgun game.
Nishikado: Right, right. But it actually didn’t use lightgun technology. Nowadays we’d just use a beam of light to shoot the target, but back then it was much simpler. (laughs) When you moved the gun, there was a circuit attached inside the cabinet that would also move. The target also had a circuit on it, and when you had the aiming perfectly aligned, the gun and target sensors would touch, forming a complete circuit. If you then pulled the trigger, electricity would flow through the circuit and the target would light up.
—Yeah, that is very simple.
Nishikado: I remember being really surprised at first. In college, and at my old job, I had worked at a very high level of research and technology, but what I was seeing here was more like child’s play… there were some moments when I thought I’d made a huge mistake in joining this company! It was still the era of transistors then, but when I’d suggest making a game with that level of technology, I was told by management that it was too expensive. I just couldn’t get through to them. Thankfully they let me use relays, at least…
—Relays, those are on/off switches using electrical current, right? Wow, that really is very basic! What kind of electro-mechanical games did you make with that technology?
Nishikado: The games market in Japan then was all overseas imports, so I felt the need to create something with more individuality. I made a number of different styles of games. Two years after joining Taito, I made an airplane shooting game called Sky Fighter. It was a big hit. It used an acrylic dome, and inside that dome a model airplane flew around that you tried to shoot. When you hit it, it exploded. I know it doesn’t sound like a big deal when I explain it now, but at the time it was a groundbreaking game. The planes really looked like they were floating in midair—and I didn’t use any strings or wires either! Of the electro-mechanical games of the time, I personally think it was the most well-made.
—The planes weren’t connected by strings or wires? How did you make them appear to be flying then?
Nishikado: Heh, I used mirrors! Beneath the playing field, where players couldn’t see, I placed the model planes, then I used angled mirrors to reflect them onto the dome. For the scrolling background, I painted images of the sky onto a film canister that I then revolved on a drum. That was also reflected in the mirror and it looked like real clouds and sky. Finally, the trajectories of the bullets were reflected in a mirror too, so that in total, I used three different mirrors.
—Using all those mirrors, it seems like some kind of magic trick.
Nishikado: Actually, when I was in school I was in the Magic club, and I had been thinking about how I could use some of those tricks in a game. People who saw Sky Fighter in operation were taken aback: “Wow, how did he make them fly like that?!”
—It’s almost like a “magic machine”: insert a coin, see a magic trick! But I can’t help but wonder at how difficult it must have been to produce such a “hand-crafted” game like Sky Fighter. Wasn’t it extremely time-consuming to make even a single machine?
Nishikado: Yeah, I think it took almost half a year to make one. Nowadays you can just compile your program and check it for errors instantaneously. But back then we were engineering everything off of hand-drawn blueprints, and it would take the sheet metal manufacturers and other craftsman anywhere from 2 weeks to 2 months (if it went poorly) to finish their part. Then, if there are revisions to be made, that added several more months.
—How many people did it take to make one game?
Nishikado: I did all the mechanical engineering by myself. Then we had someone working on the electrical side, and someone working on the cabinet hardware (things like the front panel, the logo and seal, etc)… so about 3 people total.
—So you made all the blueprints yourself?
Nishikado: I did. It was my first time too. Well, actually, I had learned a little bit in school, but I didn’t really do any substantial blueprint designing until I joined Taito.
—I’d like to ask some questions about Taito now. When did they start making video games?
Nishikado: In 1972, Atari released Pong, which was imported into Japan by both Sega and Taito. At the time, when our staff saw this newfangled “video game”, most of us didn’t think it would do too well. The reason was that the price was too high. A single machine cost anywhere between 600,000 to 800,000 yen (approx 6000-8000 USD). And to us, it looked like just a TV inside a box, with a dial on the front… we thought there’s no way this was going to sell! So we only bought one.
We set up the unit we bought in a storefront to do a test-run, and to our astonishment, it was really popular. Everyone’s opinion changed overnight: “the next era belongs to video games!” Eventually management started thinking that we could make our own games at Taito, too. To do that, they would need someone who knew about electrical circuits, and I was the only person at Taito with that knowledge—naturally, then, the job was given to me.
—And circuit boards were still quite rare at this time.
Nishikado: Yeah. Pong was made using ICs (integrated circuits) linked together like relays. But the thing is, ICs hadn’t made it to Japan yet, so looking at Pong, I didn’t immediately understand how it was put together. It was a huge challenge to disassemble and analyze it. I had no choice but to take out a parts catalog of chips, and shining my flashlight on the board, try to identify which chips were connected to which. After 3 or 4 months of this work, I was able to create a blueprint of the board. But somewhere along the line I had made a mistake, which I had to spend even more time analyzing and fixing! Ultimately it took over half a year. Luckily, I liked doing this kind of work, and I worked on it day and night without rest.
—So just to understand how the ball moved in Pong, and how the paddle moved, it took you half a year?
Nishikado: That’s correct. However, because I had spent all this time analyzing and understanding it, I didn’t want to just make a direct copy of Pong. Call it pride, but I wanted to improve it a little. The game I made there was “Soccer.” I added another paddle and added a soccer goal to the screen. It was simple, but it had a nice sense of speed, and was pretty fun. At the time, I was probably the only person in Japan who had figured out Pong and made an original game based off it. Sega had imported Pong, but they didn’t make their own game.
—Was Soccer installed in game centers?
Nishikado: Yeah. It was Japan’s first domestic-produced video game. Later, we also released a 4-player version of Pong called Derby Cup.
—Were these games exported from Japan?
Nishikado: They were! The first to be exported was Speed Race, in 1974. Midway, an American company, bought it. It’s the one I personally think is the most well-made. It’s the origin of driving video games. The screen is still monochrome, but we were able to use more realistic sprites for the car instead of just abstract squares and circles. At the time, Atari was the top game maker in America, and they had released a driving game of their own called Gran Trak 10, but it was far more complicated. Speed Race had a better sense of speed, and I think its simplicity made it more fun.
—In the course of Taito’s export business, did you ever visit America yourself?
Nishikado: I did, and I believe it was my first time visiting America. We exhibited our games at the AMOA Show, which was the biggest game trade fair in the world then. Having seen what the show had to offer, I remember thinking our Taito games were better made!
—America was the birthplace of video games though, wasn’t it? Did you visit Atari, or any other companies?
Nishikado: Ah, yes, I did meet with the President of Atari. He asked me, “Why don’t you come work for us at Atari?” And when I jokingly asked the Vice President, who was sitting there, “What’s the salary?”, I learned it was extremely good! It was 5 or 6 times what Taito was paying me. Had I not been married, I honestly might have accepted. (laughs)
Right around that time, Atari released Breakout. It had very simple graphics but it was fun. While I had been trying to do something with sprites and more impressive graphics, the designers of Breakout had approached things from a whole other angle. It made me realize that graphics weren’t the only important thing in a computer game. One of the managers at Taito said half-jokingly, “What the hell happened! America beat us!” But it turned out to be really true.
—This was around the time that video games started using microprocessors, if I recall.
Nishikado: Yeah. At Atari, they pioneered the use of computers to make video games. The Taito game Western Gun was our big break into the American market, but Midway paid the licensing fee and revised the game to use a microprocessor. As a game, I think the Taito version was more fun, but because the Midway version used a computer, the movement was a lot smoother and game looked better. As an engineer, this was the moment when I knew Taito needed to start using microprocessors from here out. And the first game that I made using a microprocessor was Space Invaders.
—You mentioned Breakout as an inspiration, but what aspects of Breakout did you find interesting?
Nishikado: I liked how you couldn’t advance to the next stage until you’d destroyed every block. Previous games didn’t have that “all clear” concept, you see. I also liked how the game got harder as it went on—the last block was really hard to hit. And I liked how the ball speed got faster. That gameplay was really good. I wanted to try making a game with those elements, but better visuals.
—There was also a sci-fi boom going on then: the movie Star Wars had just been released in America, and Space Battleship Yamato was being shown in Japanese theaters. Is that where you got the inspiration for the alien invader enemies?
Nishikado: People often say there’s a Star Wars influence in Space Invaders, but to be honest, they aren’t really related. It was of course famous in Japan, and there was a novelization out too. I also loved movies set in outer space. However, originally I had wanted to use tanks for the enemies. But the thing about tanks is, if their cannons aren’t facing forward, I don’t think they look cool. In the game I wanted to make, the enemies moved horizontally (side-to-side), which meant the cannons would have to be angled 45 degrees to the side. So tanks were out.
Well then, I thought, if I can’t do tanks, I’ll do planes! But the current technology couldn’t render sprite animation smoothly. With that limitation, I thought aliens would be a good match, since they wouldn’t look weird if their movement was all herky-jerky and choppy, nor would it look weird when they suddenly advanced down a row towards you. And at the time, there were a lot of depictions of martians as looking like octopuses.
—Yeah, that kind of octopus/squid-like alien design comes more from H.G. Wells’ War of the Worlds than from Star Wars.
Nishikado: The name “Space Invaders” wasn’t created by me either, actually. The title I originally had was “Space Monster”. However, after the game was completed, management told me they wanted to change it to “Invader”. I don’t really understand the reason why, but there was nothing I could do about it. “MONSTER” and “INVADER” have the same number of letters, so it wasn’t a difficult thing to switch out in the programming, but then from overseas we were asked to make it plural, “INVADERS.” Adding that extra letter meant changing the programming, and it was a real pain.
—I imagine that the development of Space Invaders must have taken a long time, since it was your first time creating a computer game in this way.
Nishikado: The actual game programming went by surprisingly quickly, taking only 3-4 months to finish. But the work before that, of setting up the programming/development environment, took about 6 months. It was the very first game developed in Japan with a microprocessor. There were no “personal computers” or anything back then, so I had to create the entire development environment myself from scratch. There were workstations and equipment in America, but they cost tens of thousands of dollars, so we couldn’t buy them… I bought the LSI chips myself, then soldered them to a board, and programmed directly to them in assembly. Looking at conversion charts for hexadecimal, I slowly programmed everything in, memorizing the hex conversions as I went.
—You did the programming, graphics, and sound for Space Invaders all by yourself?
Nishikado: Yeah. The ability to handle numbers in a more sophisticated way—having the speed of the invaders increase, or being able to shoot down individual invaders—that was all owed to the microprocessor. You couldn’t do things like that in games without a computer.
—I’ve heard that when Space Invaders was first completed, the response at Taito was initially rather tepid.
Nishikado: Yeah. Those who were involved in the development (including management) thought it was good. Some of the staff would say “Excuse me, I’ve got to go to the bathroom”, and sneak off and play Space Invaders, getting so absorbed in it that they’d never come back—that made me really happy. However, others at Taito didn’t like it, complaining that they couldn’t get to the end. Later we had a private screening of the game for game center operators, and their reception was also unfavorable. They said it was too difficult.
Indeed, for its time, it was a very hard game; the invaders aimed and shot directly at the player, and if one of them crossed the line into your territory, it was automatically game over. With older games, even if you were bad, you’d still be able to play for 3 minutes or so. But in Space Invaders, if you just let the enemies shoot at you, you might not last even 5 seconds. So it wasn’t well-received, no. We showed it to the general public at an arcade expo later that year, but we had another game called Blue Shark (a Midway import), and they showed that as our main game, with Space Invaders reduced to a supporting role.
—How long did it take for the “Invaders Boom” to get underway?
Nishikado: Management at Taito didn’t think the game was very good, so they weren’t really expecting much from it. However, a month or two after it was installed in game centers, the President of Taito himself remarked “That ‘Space Invaders’ game is really getting popular! Everyone is talking about it.” After that the real madness began.
—And with the boom came all those unique player strategies, like the Nagoya-Uchi (“Nagoya Shot or “Nagoya Attack”) technique.
Nishikado: Until the Nagoya Attack was discovered by a player, no one at Taito had ever thought of a technique like that. The fact that the invaders’ shots wouldn’t hurt you at the bottom of the screen like that—that was a programming bug. One day, I saw a really good player putting up some high scores around 150k. When I looked closely at what he was doing, I saw that the very bottom row of invaders’ shots seemed to pass right through his ship. It’s because I programmed it so that their shots would come out just a little bit in front of the invaders.
I had also wanted to randomize when the UFOs appeared. But it looked like using random numbers was going to be a pain, so I abandoned the idea and made the UFOs appear according to the number of times the player shot. Unfortunately, players discovered that rule very quickly. I was surprised at how fast they figured it out!
—There was also the Rainbow technique, where players leave only the bottom row of invaders alive.
Nishikado: Despite all the testplaying we did, I never imagined players would leave only the front row of invaders alive like that, so I didn’t check for it. If you play normally, you start by shooting the close, larger invaders first, and then proceed down the line. It’s interesting—the rainbow technique is one of those gameplay phenomenons borne of what you didn’t plan or predict.
—Yeah, those oversights spawned an interesting phenomenon of rumors and secret techniques being exchanged among players. Did all those strategies ever get compiled somewhere in a book or anything?
Nishikado: Hmm… good question. I’m not sure.
—Well, there was that “Game Center Arashi” comic about Space Invaders which introduced the public to the Nagoya Shot. How popular was Space Invaders in the US, by the way?
Nishikado: Very popular—I think it might have even been more popular in America than in Japan. It was ported by Atari, also.
—I understand there was a big copyright problem, with both clones and imitations.
Nishikado: Even Nintendo made one, now that you mention it. Though theirs was not an exact copy.
—Sega made a Space Invaders knock-off too.
Nishikado: I think I played that somewhere. There were three varieties I believe?
—There was ultimately a lawsuit about them, and that lawsuit set a precedent in the courts with regard to recognizing copyrights for games. What did you do after Space Invaders?
Nishikado: I made Space Invaders II in 1979. I don’t have a lot of good memories of the period after that, though. Namco released Galaxian later in 1979, and that game featured full-color sprites, but Taito had all these Space Invaders boards stockpiled (that could only do black and white), and they wanted me to make another game with them. I made Balloon Bomber in 1980, and some other games, with that stock. They had good gameplay, but graphically they looked very primitive. It was a sad sight compared to the colorful screens of Galaxian.
—You had a lot of overstock, I see. I guess that’s the down side to having a game that sells too well.
Nishikado: During that time they also added a lot of new people at Taito, and I took on an official managerial position, which made it harder for me to create anything truly new. I worked in games for about 2 more years, but after that I was tired of it. I transferred to another division and developed other projects.
I made an amusement machine robot that played guitar, and I worked on a card system that allowed game centers to use prepaid cards. This was before telephone cards, of course. The idea was a little ahead of its time, though, so it never got completed. I also worked on karaoke arrangements. I bought an Electone sound pack, and made electronic versions of songs for karaoke.
—Wow, you did karaoke too! Since you have that audio background, it makes sense that a lot of your projects involved music.
Nishikado: Nowadays karaoke is all done through the internet, but back then it was the era of the acoustic coupler. We still tried to figure out a way to transmit the data for karaoke. We conducted a lot of different tests.
Also, before the Famicom was released, I was working on designing a home video game console. I was using an American microchip–the same chip as the Sega SC-3000, actually. Unfortunately the project was dropped.
—Did you not make any more games after that, at Taito?
Nishikado: In 1989, I made some games for the Super Famicom. I was section chief then though, so I didn’t do any of the actual programming or other work. I made games in the Jinsei Gekijou and Kyuukyoku Harikiri Stadium series. I worked like up until the Playstation. Because I was a producer, though, I didn’t do any of the actual work, but as time went by I started to feel more and more like I wanted to make games myself again. So I quit Taito.
—When did you start your new development company, Dreams?
Nishikado: 1997. I planned everything myself and presented Taito with my new company. Pop’n Pop is one game I made where I did the programming and planning all by myself. It took about 2 years. I’m at the limit of what I can do now, though, so we’ve hired some programmers, and we’ve got a staff of 15 people in total. Unfortunately, as a result, I’m back to doing more managerial work, and I haven’t been able to get hands-on with projects like I wanted. And I have to manage the financial side of the company too, which is even harder! But the advantage of having my own company is having complete control over the final product, and I feel a renewed vigor for this work.
—Are you able to make any of your own independent creations now?
Nishikado: Right now, for fun, I’m working on a children’s game. It’s an electro-mechanical game called “Magic Bell.” It’s sort of a science experiment for me. It’s a quiz-style game where you have to choose which of three bells is hiding a piece of fruit. It uses half-mirrors, too. Right now I’m thinking of a sequel called “Magic Hat.” I’ve taken out a patent request for it.
—It sounds a lot like your old games: the handmade aspect, and the use of mirrors and optical illusions…
Nishikado: Yeah, I still love all that stuff. From a business standpoint, it’s not something I could really imagine making.
—By the way, what do you think of video games today?
Nishikado: There’s a lot of hardcore gamers and game fans today, and it feels like games have to be made for that audience. Personally, I love STGs, but I can’t get into recent STGs. I can’t handle them, so I just watch. (laughs) Lately I’ve been asked to speak about retro games. I think in a lot of new games, you find the core idea in an older game. Many games have their roots in Breakout, for example.
—Yeah, and even a relatively early dot-eater game like Pac-Man has its roots in Sega’s 1979 game Head On.
Nishikado: We’re slowly losing those origins as time goes by, so I think doing research on the history of those older games is really fun. I think there’s still fresh ideas and new directions waiting to be found there!
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!