Rayforce – 2016 Developer Interview
originally featured in game gene vol. 1
Tatsuo Nakamura – Director/Designer/Programmer
Tomohisa Yamashita – Planner/Designer
Hideyuki Katou – Graphic Designer
—To begin, please set the scene for us when the Rayforce development began. Rayforce was released in 1994, which was dead center in the heart of the versus fighting game boom. And Taito was developing their own FTG too, with Kaiser Knuckle.
Yamashita: At that time, Taito’s development teams were split into three main groups. Gun Frontier and Metal Black were made by the Chuou R&D group in Yokohama. We, on the other hand, belonged to the Kumagaya Research Group in Saitama. Although many of us in Kumagaya wanted to make a STG, the upper management told us it wasn’t feasible, and for a long time we weren’t able to get anything started.
—Darius, which can rightly be called Taito’s most well-known STG, and Night Striker too, were both made by the Chuou team. Was Taito’s attitude something like, “well, sorry, but Chuou is our STG team”…?
Katou: I wouldn’t say that. After all, Grid Seeker had been made by the Osaka team.
Yamashita: The Rayforce development has its genesis in Nakamura’s long, abiding quest to develop a STG game at Taito. It was really because his Rayforce planning documents were successful that we were finally able to develop a STG at Kumagaya. I was working at Kumagaya and happened to really love STGs myself, so that was a bit of good timing, too.
Nakamura: Unfortunately, Taito had a master business plan at the time, and it actually did say the Kumagaya team would not be making any STGs.
—No STG games for Kumagaya?
Nakamura: Nope. Management had already planned out the different genres that each development team would work on, and STGs were not listed for Kumagaya. For each department, it provided a general outline for what genre they would work on, the development timeline, and even the projected sales.
—With something like that, I can see how it would be hard to accommodate your requests, no matter how fervent.
Nakamura: Yeah, there was no room to negotiate. We didn’t give up though. I made sure they heard my complaints: “Chuo and Osaka get to make a STG, so why won’t you let Kumagaya?!” (laughs) I had experience working on the Darius II development in Kumagaya, and I personally loved STGs, so I was confident that if we just had a good idea for a game, we could make a successful pitch. (laughs)
—So you started working on the planning for Rayforce, then.
Nakamura: My thoughts kept returning to the 1989 game Master of Weapon, which came out before Grid Seeker. It has an aerial, top-down perspective, and I thought it would be awesome if that screen depth could be done in real 3D.
—Master of Weapon was the inspiration? I don’t believe it. (laughs)
Nakamura: Xevious had already done the targeting reticule for bombing ground enemies, and I didn’t want to just mindlessly repeat that. It was then that I thought of the lock-on laser: with one button press you’d be able to take out a bunch of enemies in sequence.
I took these basic ideas, and before going to Taito, I showed them to the planners and character designers to get their feedback and flesh things out. Then I had the project head at Kumagaya polish it further, and we presented that to management at Taito HQ. It was a success, and thus the Rayforce development officially got underway.
—But isn’t it true that, at that time, most of the arcade developers were caught up in the mania for versus fighting games? There weren’t a lot of new STGs coming out. Was Taito consciously trying to go against the grain by releasing a STG at this time…?
Nakamura: Honestly, we were out in the boonies of Kumagaya, so we weren’t privy to the going-ons of the wider world… (laughs)
Yamashita: Yeah, it was like that. (laughs)
Nakamura: We just wanted to make the game we wanted to make. And naturally, since we were putting our time into this, we wanted to try and make the very best STG we possibly could.
Rayforce looked back to Xevious’ targeting reticule and bomb-less game design, while drawing visual inspiration from Master of Weapon.
Game Design, Planning, World-Building
—Two aspects of Rayforce’s game design stand out to me: first, the lock-on laser, and second, the fact that there are no bombs (which was unusual for vertical STGs of that time). Were those features decided on from the very beginning of the game’s conception?
Nakamura: Yeah, they were some of my early ideas. It was my sense that bombs are most appropriate for STGs where the player’s movement will generally be confined to the lower half of the screen—the kind of games where you’re mostly just dodging left and right near the bottom of the screen, and you use a bomb when things get too dangerous. I don’t have any problem with that kind of game, but I wanted to try and return to an earlier style.
In Xevious you have to move all over the map to take enemies out, partly because of the targeting reticule. That was an inspiration for Rayforce… which got me thinking, “well, maybe it doesn’t need bombs?” Nor did I think bombs would be necessary as an emergency last resort, so yeah, from the get-go they were not included.
—Bombs were such a fixture of vertical STGs back then though. Were there any requests from management to put them in?
Nakamura: Maybe it’s because we were out in Kumagaya, but… if there were, we didn’t hear about it.
Yamashita: Yeah, the Kumagaya Wall of Silence kept those voices from reaching us. (laughs)
Nakamura: And also, in our location tests and Q&A feedback sessions, very few people objected to the lack of bombs.
—I think it’s because the gameplay was so tight otherwise. Even when I play Rayforce today, I never find myself wishing I had bombs. It’s a basic system, but it feels implemented to completion.
Nakamura: The way the lock-on laser works today is that, each time you lock-on to an enemy and press the fire button lasers comes out, but in the beginning it worked differently. For example, say you locked onto 4 enemies, and pressed fire: well, until all four of those lasers had landed, you wouldn’t be able to lock-on to any additional enemies. We kind of had tunnel-vision with the laser system so it seemed ok to us. But one of the programmers came to me and shared his opinion that this would be too difficult. I realized then he was right, and we changed it to how it is today.
Tomohisa Yamashita – Planner/Designer.
Yamashita: Yeah, we made many tweaks to the gameplay systems along the way. One example: originally when you didn’t have an enemy locked-on, and you pressed the laser button, nothing happened. But we figured it would be confusing if there was no response at all, so for a visual cue, we made it so a single laser shot out even if you weren’t locked onto anything.
—What I think works with Rayforce, though, is that those ideas weren’t just minor, one-off adjustments: it feels like you carried each idea to its conclusion and integrated it into a conceptual whole.
Nakamura: Maybe, though I do remember that when Katou, our graphics artist, asked what kind of world we were going for in Rayforce, I’m pretty sure I gave him the impression that we didn’t really know yet. (laughs)
Katou: I didn’t know what we were doing, and all Nakamura had told me was “inorganic.”
Nakamura: That was the keyword, yeah. The vision in my mind was a world with no organic or living enemies. I hadn’t thought it through much though at that point, so in the meantime I would just tell people “inorganic” and leave it at that. (laughs) Actually, I had hoped to participate in the worldbuilding later in the development, but unfortunately my boss told me “You need to stay focused on the programming”, and after that I stopped sharing my ideas.
Katou: Gradius came out in 1985, and it had organic-style enemies, right? The idea to break with that tradition and go for a completely mecha world came from Nakamura.
Nakamura: From the beginning, the idea was something like, everything you see in this world, even the nature—is actually an artificially constructed, “fake” lifeform.
—Did you hash out and develop your ideas together, as a team? Or was it the work of one planner?
Nakamura: In my early brainstorming drafts for Rayforce, I drew several sketches of what I thought the actual game screen would look like. Katou then took those and really improved/beautified them.
Katou: The story idea of descending into the planet’s core was also found in the earliest drafts, and I did my best to faithfully reproduce what was written there.
Nakamura: Yeah, the central ideas for the game were written out in the planning documents by myself and the project leader. It was chock full of neat stuff. However, there’s never enough time in a development to reproduce every idea in your planning document. We had all these detailed ideas we wanted to include for every stage, but many had to be cut for the final version.
Yamashita: Generally speaking we used the planning docs as our blueprints for constructing the stages, but partly because we didn’t have experience making STG games in Kumagaya, and partly because none of us were particularly good at STGs ourselves, everyone just kept throwing in their ideas until it was a real mess.
As far as game balance goes, the reality was that only Nakamura and the others who had worked on Darius II had any experience with balancing a STG, and they handled all of that work on Rayforce.
Color and Graphic Design
—As this was mostly everyone’s first time making a STG, then, what areas did you pour your efforts into?
Katou: Our mission was to make a “Kumagaya”-style STG game, not merely an imitation of Gun Frontier or Metal Black. Right around the time that Rayforce started, for each dev team Taito actually created a new graphics position, whose job was to unify the graphics and visual style for each game. I was one of the people picked for this position. I remember my first thought was, “huh? me?” (laughs)
Hideyuki Katou, Graphic Designer. He also worked on Ikaruga, Border Down, and Under Defeat.
The first thing I was asked to do was figure out what kind of graphics style we were going for. I spent a lot of time thinking about that, but all I could think of, when it came to Taito STG games, was Darius: a base of metallic grey, and a combination of different primary colors. Rayforce uses that same color scheme. There aren’t a lot of very colorful mechs in Rayforce… we intentionally made everything mostly grey, with highlights of red or blue, but mostly using combinations of just a couple colors.
Management got mad at me too, because they said I wasn’t using the typical “Taito” colors, that my colors were boring, I had bad taste, it was ugly, etc. (laughs) I had never worked like this before, but for Rayforce I spent a lot of time fiddling with the color palettes.
—How did you go about doing that?
Katou: Nakamura created some software tools for me to use. I would put all the graphic assets (backgrounds, enemies, etc) on-screen together, then the tool would show the pallette numbers, which I could cycle through using a joystick. Then I took down notes of which ones looked good, handed them back to Nakamura, and he manually entered the raw data back in. A very “analog” process. (laughs)
You know how in Rayforce, there are two sets of sprites for the enemies: one for when they’re in the distance, and one for when they ascend to the foreground level? We actually used different color palettes for those sprites, to help players distinguish at which “altitude” you can and cannot lock-on to the enemies. We wanted to show all that visually, and before I knew it, I got stuck with that job. I had a good time, of course, but even today I can still remember how tedious and annoying that work was. (laughs)
Yamashita: When the enemies are in the distance, we tried to use mostly white, so that they wouldn’t get lost in the background colors. Then when they come up to the foreground, they’re given more color to make them easy to see. The difference in color when the enemies reach your level is very distinct, so you can clearly tell when enemies have reached the same height as the player. We did the same thing for Raystorm, too.
Stage Design in Rayforce
—Changing hardware is always a challenge for game developers. Did you encounter any difficulties with Rayforce, which used the new F3 system pcbs?
Katou: Those in-the-know are already aware of this, but the F3 system had weird scaling functions. You could only do enlargement/zooming on the entire screen layer, while shrinking was limited to objects and sprites.
—Wait a moment… what do you mean by that?
Yamashita: The best example in Rayforce might be stage 6. In that stage your ship keeps on descending deeper and deeper into the enemy base. It’s a shining testament to the difficulties we had with the hardware. (laughs) Because zooming-up (enlarging) could only be done on the entire screen, it made the graphics look really ugly and pixellated as they got closer to you. We had no other way to do a stage like that, though, so Katou really labored to create graphic assets for the stage that would still look OK when enlarged like that. It was twice as much work for him. (laughs)
Katou: The enemies were always time-consuming to make, since they had to be done pixel-by-pixel, but many of the stage backgrounds had to be redrawn twice. I took great care in trying to draw backgrounds that wouldn’t look weird, given the limitations of the scaling.
The background graphics of Rayforce relied heavily on cylindrical and simple geometrical shapes to accommodate the F3 scaling hardware. Similar tricks were used in Space Harrier.
Yamashita: When you enlarged graphics with a lot of diagonal lines, the pixels would look very jagged and ugly. As such, we tried to avoid diagonals wherever we could.
Practically speaking, this meant drawing a lot of cylindrical objects, or using simple vertical or horizontal lines, which wouldn’t look bad when zoomed-up. I think we pulled off the “geometric” look pretty well. (laughs)
—I see now. And how about the stage design itself?
Katou: That was done by Nakamura and Yamashita, as well as the project leader. They would hand me these huge, interconnected maps that they’d created (on paper!) for each stage. We used those as a basis to talk about how the stages would play out: this enemy appears here, this happens here, etc. (laughs)
Then Nakamura would ask me the height of the different background elements… as Rayforce used parallax scrolling, I would explain how this was the foreground, this was the mid-ground, this was the background, and so forth. After that I would ask him specifics, like how many seconds it would take to scroll through a certain section. Then I just tried to somehow keep all that in my head as I worked. (laughs)
Yamashita: The very top layer of the map (the foreground) was the easiest to figure out, and in the planning documents we wrote out precisely how long the sections would take to play through… but for the layer beneath that, we only wrote some loose approximations.
Katou: Nakamura gave me very detailed instructions, like “For this next part, we’re going to use scaling for the background, so please make it this size.” But midway through I would start panicking, because I couldn’t remember exactly what he said. (laughs) If I had to say what the hardest part of Rayforce was for me, it would be that.
—Were there ever any requests from the planners where you went, “that’s impossible!”
Katou: Not really, not for me. If Nakamura said it was doable, it usually was. (laughs)
Nakamura: From my time programming, I already knew what the F3 graphics hardware was capable of. I knew how many layers you needed to give something the illusion of depth, and so forth. For that reason I didn’t write too much in the planning document that wasn’t possible for the hardware.
—In stage 2, the dust cloud effect from the Poseidon enemy reminds me of Metal Black.
Nakamura: Yeah, you mean where the huge, midboss-ish battleship enters from the right and kicks up all that dirt… we cribbed that from Metal Black.
Yamashita: Well, let’s say it was “inspired.” (laughs) The first time someone showed me Metal Black, I thought that effect was awesome, and we directly imitated it for Rayforce. We’d never done anything like that before, you see, so we had to model something. Besides the sprites being similar, we weren’t intending to make any other connection. But I do remember explicitly asking the programmers to make it look like that scene in Metal Black, where the boss bursts onto the screen. (laughs)
Aoki: I don’t know if you remember this, Nakamura, I seem to recall that you were initially unsure about whether to show points for the enemies you shoot down with the lock-on laser…?
Nakamura: Yeah, that’s right. When it came to the question of how to display the score on-screen, my first instinct was not to display it at all. With Gun Frontier and Metal Black, you see, it felt like we had reached the point where “games can be immersive and cinematic experiences now too”, and I wanted Rayforce to feel like a movie. To that end, I didn’t want the screen to look too “game-like”, and I really struggled over whether traditional gameplay elements like scoring should be displayed. Ultimately we did though, and I’m glad we made that decision.
Aoki: I remember you didn’t like it at first. You asked me point-blank, “Do you really think it looks better this way?”, and I said, as strongly as I could, “Absolutely, yes.” After that it was clinched!
Nakamura: Yes, yes, now I remember that.
The “dust cloud” effect was inspired by Metal Black.
—Did the entire development team really respect Gun Frontier and Metal Black, then?
Nakamura: It was more of a reference point. Not something we worshiped.
Yamashita: I wouldn’t call it respect either.
—More of a rivalry, then?
Yamashita: That might be closer to the truth, yeah. As we were talking about earlier, Chuou had been designated as the “STG team” for Taito, and we were the young upstarts trying to break off a piece for ourselves.
—I also wanted to ask about a “Rayforce secret” you mentioned on Twitter, Nakamura… about stages 7 and 8, and how they were connected?
Nakamura: In the planning docs, stage 7 had you hurtling through this tube at high-speed, towards the planet’s core, where stage 8 took place.
—So they were separate stages at first, but how did they end up getting combined?
Nakamura: It was a question of time, or the lack of it, with the development schedule.
Yamashita: That stage did end up feeling like we stuck two different scenes together. The part where you’re flying through the tube at high-speed is visually very static. Then you emerge in that area with all the buildings, and you start ascending above them. Those were actually separate stages.
Aoki: The way the BGM changes at that moment is a remnant of the original plan, right?
Yamashita: Yeah, it might be. I think we requested that the music change there, when you reach the interior.
Aoki: It’s the only stage where the BGM changes midway like that.
Yamashita: Up to that point, things were visually quite dark and gloomy, and when you finally reached the core, I wanted it to suddenly brighten up. The composer was asked (not directly by me) to write something that would capture that feeling musically.
—Where did the idea come from to have all the stages linked together without interruption?
Nakamura: Arcade hardware then didn’t have any loading times (like CD-ROM systems), so I don’t think we ever thought about doing it differently. It was normal for games then to be seamless in their stage transitions. I think games that put interludes/transitions in between stages did it deliberately, for other reasons.
—It seems like it would be quite a challenge to connect EVERY stage the way you did in Rayforce, though…
Yamashita: It was very difficult, yeah. Especially when we were in the planning stages. The transition from stage 2 (above the atmosphere) to stage 3 (the sky above the planet), for instance… it was like, how in the world are we supposed to connect those two? Our starting point was trying to figure out how to get the player down into the planet. The problem was, if we followed the planning docs as they were written, it was impossible. (laughs)
A “happy new year” card from the Rayforce development team.
Sometimes we had to employ tricks to keep things seamless, like whiting out the screen for an instant… I think every stage has moments like that actually.
In Nakamura’s early planning documents, the title of the game was “Layer Section.” The idea was that each stage, or “section”, would be composed of different layers that you descend through. The way we envisioned it was, while everything would be connected seamlessly, the sections would still be delineated by “floors” you reach at the end of each stage. We designated those “floors” with different colors, which you can see in the stage names (ie RED POWER TO PIERCE THROUGH, THE GRAVITY OF BLUE SIDE, etc). The “Layer Section” title, therefore, aptly reflects our ambitions: seamless, but with clearly defined areas.
The Rayforce location tests
—In arcade games, the first chance players have to see your game usually comes in the form of a location test or a game show. Do you remember which was first for Rayforce?
Yamashita: The location test was first for Rayforce. We did one at the game center that was downstairs of our office, in Kumagaya.
—Was that the only location test you did?
Yamashita: At first, yeah. We only had two stages of the game completed then.
Aoki: One of the nice things about working in Kumagaya was the access we had to the game center down there for location tests. We took advantage of that by putting that early, 2-stage version of Rayforce for people to try out.
Yamashita: It wasn’t the usual process for location tests.
Katou: Yeah, officially Taito had a two-tiered system for location tests. There was a “midway” location test where, the main point of which was to gauge how interesting the game was to players. That happened fairly early on, and was meant to guide the further direction of the development.
Yamashita: Then there was “real” location test, which was mainly for business purposes, to see how much income the game would make.
Aoki: And before both of those tests, and only in Kumagaya, we had our own little “local” location test.
—Did a lot of players come to check it out?
Katou: That game center wasn’t really aimed at hardcore arcade gamers. It was more of a hybrid arcade/karaoke place, with people stopping by to play a game or two before heading to karaoke.
Aoki: Which meant it gave us a lot more realistic picture, actually. (laughs)
—Were people taken aback by that game being there, then?
Aoki: Honestly I don’t think most people even realized it was a location test.
—Did it not make much of an impression, then?
Yamashita: The version of the game we had on display then, for that test, was very much aimed at general audiences. There were no bombs, and I seem to recall a lot of people came away feeling it was very hard.
—Was “Layer Section” the title you put up for the location test?
Aoki: Until we get the trademark registered, we don’t reveal titles. During the location tests we just wrote “NEW GAME” above the cabinet.
Nakamura: Yeah, so just before the final/business location test, we solicited three different departments at Taito for titles. What came back was “Gunlock” and “Rayforce”. Then we sent a questionnaire around to the developers, to see which title they liked more, and Rayforce was the winner, so we went with that. “Gunlock”, by the way, was the project leader’s favorite. He pushed for it.
—So after many twists and turns, you finally debuted Rayforce at the 1993 Jamma show. How did that go?
The asteroids in Rayforce were painstakingly animated by hand, despite the growing enthusiasm for pre-rendered CG at the time.
Katou: It was crunch time then, and we were spending nights at the company dorm. Go to work, back to the dorm, go to work, back to the dorm… I kind of lost my sense of time, and don’t really remember that period too well, including the show. About all I remember is that I was pissed about NebulasRay. (laughs)
Yamashita: Same here. The only impression I have of it is NebulasRay. (laughs)
Katou: Pre-rendered CG was pretty new then, and we were thinking we’d like to try using it, but Namco beat us to the punch.
Yamashita: At the time, during the Rayforce development, our department wasn’t doing any CG rendering. For example, the asteroids in stage 1, those were all rotated manually. Katou drew the animation frames by hand.
Katou: I digitized an image of a rock, then slowly rotated it and warped the image, making corrections by hand.
Yamashita: Yeah, it kind of looks like CG, but it was all drawn by hand. We labored to make the power-ups in Rayforce rotate correctly, and I was mad at NebulasRay when I saw how cleanly they rotated their objects with CG.
—In the liner notes of the Rayforce soundtrack album, the story says that the pilot is an android, but in the official game backstory, the pilot is a human with cybernetic implants (a cyborg).
Yamashita: I thought you might ask about that. It had always been a cyborg, right Nakamura, in your original planning documents?
Nakamura: No, actually the pilot was an android in the beginning.
Nakamura: Katou pointed out that an android would necessitate creating too much backstory, so I changed it to a cyborg.
Yamashita: Yeah, I do remember Katou telling me the same thing. All of the fine details of Rayforce’s setting and backstory like that were created by Katou.
Katou: The story behind that android thing with Zuntata is, honestly, kind of embarrassing, so I haven’t told anyone. (laughs) After the development was completely over, and it came time to create the soundtrack, Tamayo of Zuntata asked me for reference materials, so I hastily I wrote-up a story and background summary for her that mentioned the android. It wasn’t one of the original planning docs I sent over, in other words.
Yamashita: Ah, so that explains the discrepancy.1 Tamayo seems like the kind of composer who really values the world and setting of the game.
The bigger worry for us, with the Rayforce music, was that we didn’t receive the final BGM from Zuntata until right before the final deadline. (laughs)
Nakamura: It was delivered by bike messenger.
Yamashita: It was the first time I saw a bike messenger in Kumagaya. (laughs) Before that, we hadn’t heard a single note of the finished music—we’d been using demos up to then. Everyone was really worried: “Are they going to make it? Are they going to make it…?”
—Wouldn’t that leave no time for editing the music to fit the stages?
Yamashita: Outside of sound effects, we didn’t try to synchronize the music with the stages in any special way. It’s one long, seamless experience, so dropping the tracks in at the end like that wasn’t particularly problematic. The issue was more one of worry… as the deadline approached, it was like, “wtf?! are we really going to have to release the game with these demo tracks?!” They pushed it to the very very last minute, it was crazy. (laughs)
—At the time Rayforce was released, STGs couldn’t really compete with versus fighting games at the arcades in terms of income earned. In such hard times, I think the games that ended up surviving were the ones with relatively simple, “light” gameplay that could appeal to salarymen and the like. Raiden would be an example. I think it was an especially difficult era for games that were geared more towards hardcore players. They just didn’t make enough money. Given that environment, how did the R-Gear project get started? Was it management’s idea, initially?
Yamashita: Hmm, let me think… no, I don’t think it was official. It was more that we just started talking about a sequel amongst ourselves, and started working up planning documents on our own.
—Normally you wouldn’t make a sequel for a game if the game center income wasn’t very good, though.
Unfortunately, the only footage we have of R-GEAR is this video from the Taito Ray-Ray Collection. A prototype has allegedly been found, although I believe it is still waiting to be dumped.
Aoki: I don’t remember getting the official GO sign either. It was just something they started working on in the meantime, in the event a sequel did get greenlit.
Yamashita: Was it like that? Yeah, I think you might be right. No one had given their approval, we just started.
Aoki: And you started on it pretty early, I recall.
Yamashita: Yeah. We didn’t have many people available to dedicate to it, though, so it was worked on in a somewhat limited capacity.
—But you intended it to be a straight sequel, a “Rayforce 2”, correct?
Yamashita: Mainly we wanted to improve on every aspect of Rayforce, starting with the graphics. Also, when we were developing Rayforce, management told us that because the first stage started in space, it lacked visual impact to draw players in. To change that, we decided that R-Gear would start on a ground-based Space Colony. That atmosphere-entry scene in Rayforce had been really hard to make too, so this time we wanted to create environments that would be easier to stage events and scenes in. We wanted to include more stuff that would immediately appeal to players, too.
—So R-Gear’s development began as an outgrowth of things you wanted to improve in Rayforce, then.
Yamashita: Yeah. Basically we just wanted to make it look better. We wanted to add a little more gameplay stuff too, so we created another player ship to choose from, giving you more attack options.