Design of Yamato by Makoto Kobayashi

An essay by Makoto Kobayashi, from Hyperweapon 2013.
(October 2013, Great Japan Pictures)

Hyperweapon is a semi-annual magazine that showcases the work of designer/director/modeler/manga artist Makoto Kobayashi. In recent years, it has been a major source for the art and design of Yamato Resurrection, and in the 2013 issue he wrote about his involvement with Yamato 2199 for the first time.

Design of Yamato

When I received the invitation from Mr. Nishizaki to participate in the planning of Space Battleship Yamato 2199, I knew that some work had already begun, such as design and scriptwriting, under an independent production unit. So I figured I’d go see what was up for now. I asked the guy in charge if something was wrong, and was told that the mecha wasn’t good.

He’d heard that a team with a good reputation had gathered to do mecha and art design, but in case there was some sort of problem, I went to a “concept meeting” at the studio doing the design work. When I heard from the general director that “the staff for 2199 is well-equipped and help is not needed,” I saw that it was so. When I was shown a script and a design book, I saw that six mecha had already advanced into 3D with no deficiencies. I thought it was wonderful when the general director said, “for example, Gundam Century gives a more detailed view of the world of Gundam, and delivers a greater sense of reality, and I want to depict a more detailed view of the world of Yamato in this production.”

[Translator’s note: Gundam Century was a book published by the makers of OUT magazine in 1981. one of the first of its kind, it investigated the history, technology, and science of Mobile Suit Gundam with fascinating results that captivated a generation of fans and pros alike.]

There was great love in the script and both the mecha and characters, but my concern was that the background designs were insufficient. Was this a good time? Say something. It would start production before long, so I submitted a report to the producer saying, “the mecha is very nice. Maybe it would be good if I did a comparison table poster?” This took place in the middle of the production stage for Yamato Resurrection.

Afterward came the non-hit of the movie, then the exceptionally big hit of the Resurrection DVD through desperate double-packaging, and I lost touch with the 2199 group. Then there was that unexpected death [of Yoshinobu Nishizaki].

Shortly after that, I got a request from the 2199 studio to look over the designs. Again, the backgrounds were insufficient. I forget who was previously in charge of backgrounds, but it had now been left up to Mr. Hirukawa, who did production support for Mr. Habara on Resurrection. However, I guessed the situation was that criticism was not welcome since the start of animation work was imminent, so I conveyed to Mr. Hirukawa what was missing and left the studio.

(In point of fact, there really were all sorts of things going on at the time, so I admire greatly the executive staff and the man in charge, who started out working to somehow keep Yamato going.)

So, you could say that the order to start on 2199 followed the production example of the Resurrection Director’s Cut.

I heard it was Episode 1. Had it come up short? While considering that it would increase my design workload, I said I’d be “happy to help” do art layouts for the production. In fact, a similar situation arose with Episode 2. Well, in that way, if I took up several scenes that nobody else could do, maybe they could help out with the end of the Director’s Cut in return. And so production of that and the Director’s Cut proceeded simultaneously, without much trouble.

However, one day I was suddenly asked, “will you draw Garmillas and Iscandar?” I was asked to submit a detailed topographic map, and I wondered, “Isn’t there someone else who wants to draw it?” Then came the Garmillas capital city. For Yamato, after they go into space, the only places you really see are the interiors of Yamato and Garmillas.

I studied images in the Complete Works book [translator’s note: the famed “silver books” published by Office Academy] from the original production. I heard from one person on the main visual staff that the circumstances were that meetings could no longer be held frequently because it was impossible for the general director to manage the workload.

“Since the Garmillas capital Balerus appears early on, can we get your help for design development?” That’s how the workload cumulatively increased; it meant time was moving quickly.

I thought this might get serious, since it came up during the time I was doing the Yamato Resurrection calendar. It also came up during the time of Last Exile. [Translator’s note: Kobayashi was a primary designer of the first Last Exile series; here he refers to the sequel, Fam the Silver Wing.]

The general director seemed to like the Resurrection calendar that I did, and put it up on the wall in the 2199 conference room. Not only that, it seems that it was used as an image source for guest-scene designs in 2199. The pictures of the Salvage Yamato in the dimensional Sargasso and the the Rainbow Star Cluster background behind Blue Noah continued to evolve, respectively. Also, there seemed to be interest in the poster for the Director’s Cut; “I guess, in a battle it doesn’t look right unless it’s covered with a lot of appropriate damage.”

I came to put damage processing into 3D models. By the way, it was like this when Nishizaki directed, too, but since nothing would ever get done if the people on the scene kept telling the director to make the film this way or that way, it would tend to go unspoken as a matter of sensitivity.

So, in the case of a long series, I’d just look over the script and storyboards. I wouldn’t see the footage with sound on it, because if I did see it, I’d speak up.

Now let’s move on to the design of the Garmillas capital city Balerus.

Creating Balerus

The picture in the top right corner was done as a freeform image since there was no policy at the time. They said they didn’t want to do a ceiling city so I did the two pictures in the center right of a huge presidential office tower building. The smaller one left of the upper right corner was drawn with the idea of seeing towers with a little more landscape.

An unabridged version of the original is seen on the facing page, where I wanted to try and preserve the basic silhouette of the Presidential Office tower from the previous work. Also, I thought I could make use the original shape of the old tower in a slimmer silhouette. This appears on the lower right of the facing page.

What appeared in the finished story was gathered up from the feeling of this area. As you know from the completed form, the concepts at central left and lower right were tried out based on the idea of “an underlying structure of mushrooms.” While I had a feeling it wouldn’t work, it was faster just to draw it than talk about it.

The base on the floating continent is at bottom left.

Creating Balerus

This original art was the background for a bird’s-eye view scene in the last chapter. While I didn’t have the time to pencil in the fine detail to properly reflect its being a background, producing this “rough sketch” allowed us to economize on the production of the background, ultimately having a direct influence on improving it in the end.

Since the camera rotates around the tower in this scene, it was necessary to draw the “unseen portion” in the inset below. a high-resolution texture had been applied to the 3D beforehand, because correcting for the detail loss you get when you push in too far on parts had been emphasized during the making of the original drawing. So this work is not as complicated as you might expect. As for the slope that extends out from the lower part of the tower, the 3D group mapped original art onto a model that was created separately and rendered it. As a result, if you consider the efficiency of enormous human effort poured into a scene that’s just a few seconds long, the time hurdle can be overcome to some extent.

Improvement of the Production for Movie Theaters

Now, although I talked about design development for Balerus, I will continue the story that came before that. Last Exile season 2 was a long-term animation series. With the first Last Exile series, I watched it on TV during the production, and I’m not saying it’s because I liked it. Since the production situation was that it might not meet the deadline of a TV broadcast schedule, I had the worry of checking that it could be safely broadcast every week. Therefore, I didn’t watch the completed film during the production period.

I went through that this time, too. In the outfield, there was no end of changes to the storyline, with people saying, “Let’s do it this way. No, let’s do it that way.” In 2008, I had to be connected by phone to the executive director right in front of Mr. Nishizaki, who said, “I don’t know the contact address” so as not to antagonize him.

Because the first chapter of 2199 was made for the purpose of TV broadcast according to the original production plan, it had restarted with the assumption of a limited theatrical release and its quality had to be raised so it would not look like an inferior feature film. It was evident that the specs and production mechanism would have to speed up by several times (the 3D in particular) or it was a foregone conclusion that it would derail in the middle stage. Aside from the content and sensitivity, I thought it should have a general director or production manager who understood that.

Divergence from the original and re-approach

Separately from the site of 2199, when the Resurrection Director’s Cut was completed, “sensitivity” and “compromise” was carried out by the general director as mentioned above. When Resurrection was initially released and there was a negative reaction from the audience, it was seen as proof that Yamato‘s original audience was not interested in new content, and it was decided to take the Director’s Cut closer to the original by using classic music and sound effects. There was a strong point of view that it should not end as a double disgrace. Also, the elements added near the end were considered nonsense; after completion, several fans were invited in-house for a screening, and if their reaction was bad, apparently they’d decided to proceed by dismissing the director and rebuilding it.

When you think about it, it’s understandable not to want to deepen a wound any further. This is a title that inspires a lot of people to argue non-canonical opinions, both inside and outside the production. That stuff may be hypothetical, but speaking as one who has experienced many remakes such as Giant Robo and Samurai Seven, because a remake is a remake, “tempo and how to show it now” are essential. The audience will be confused if the work offers a direction and tempo that are unsatisfying, and it is understood that one tries to find the reason for that confusion in the minor details.

Since the audience reaction in the wake of the Resurrection release proved that, it was hoped that getting [Nobuyoshi] Habara-kun (who directed the Broken Blade theatrical shorts at the time) to do the cutting would give the audience a better impression. Also, I thought that by introducing the music and sound effects of the original, it would be an important element to give the audience that would be consistent with the “Showa era atmosphere” left behind by Director Nishizaki.

But honestly, I was so surprised by the unusual rejection of the sound portion [in the first cut of Resurrection], it may have been unavoidable that the fear of rejection closely haunted elements close to the original work. When the General Director (who in fact reacted negatively to the initial version of Resurrection) watched the completed Director’s Cut, he said “I didn’t think the sound of the original work would blend in so well with a modern picture like this,” it became a favorable thing for 2199 and I thought it was very good.

It reached the state where a Resurrection calendar was hung in the studio meeting room and the Director’s Cut poster was hung in the studio entrance next to a poster for each chapter of 2199, and the studio would be able to continue making 2199 to the end.

Challenges pile up early in the voyage

Although the release schedule was decided and production advanced, the complex layout work based on insufficient designs needed more time than I thought. At first, the mecha-related parts of the angle composition were so complicated that I wasn’t able to draw them freehand, so my flow was to produce finished images directly from storyboards. [Translator’s note: the goal was that Kobayashi would only produce images that would appear in finished footage, nothing redundant.] Then, I built the sets that would appear frequently on the ship in 3D, such as the third hangar and elevator shaft, so as to shorten the drawing time for the background layouts that were needed by the character animation staff. These days, looking back at the production period for the theater screenings, I think it worked out.

Since the design work on the Pluto base had been too rough, I had to make a detailed design of the base layout, based on the general director’s indications of the type of scale he was going for, so I made a detailed ground plan, based on where fighters were located in each shot of the storyboard. In the end, you could say my work extended to the layout phase, and to me personally, it’s the first highpoint of the original show.

Balerus was the third color board I made, following the Pluto base and the ruined supply base on the moon of Saturn.

Dropping the live ammo of the original’s raw material to produce fine detail

Like Dessler’s crest on these pages, I began to work on production material intended to combine with backgrounds. I think the first one I made was for Planet Balun. Though the visual concept was far from the original work, the general director modified his instructions and made changes to the previously-completed art board. Since the art staff gets angry over proposed changes to drawings, the surface of Balun was created as a texture that could be rotated and deformed on a sphere through photographic processing, and it was brought to the finished scene.

At the time, an external studio was phased out, supposedly over difficult quality issues, and although I didn’t think the production mechanism to create theatrical-quality pictures was perfect, I did get the feeling that we reached a place where there would be no setbacks at the halfway point.

I did more voluntary 3D models: “Yamato analysis room” and “Domelus bridge” were completed. Based on an arrangement with each production manager, digital models were output prior to layout based on scenes in the storyboard, and after realizing a considerable time savings, the animation staff was asked to take a hit by carrying out freehand drawings without digital data. I somehow became able to predict the schedule and divisions.

While I was working on the freehand layout drawing of the Garmillas naval port for Mr. Habara on Episode 19, showing the night before the Rainbow Star Cluster battle, I heard that the storyboards for Episode 20 were not finished yet.

I knew from the script stage that the general director would be drawing this storyboard himself because diversions from the original were greater, but it seemed to be taking more time than expected. The problem was that the episode at the beginning of the next chapter would be in the hands of a live-action director (Makoto Bessho). As you know, the spatial orientation of a storyboard for live-action is much looser than anime. Although the general director and production staff were analyzing and revising the storyboard, it was postponed because of Episode 20 and time was running out. The presumption was that a huge number of layouts would be needed for events on Planet Garmillas. This was going to be a problem.

And the last chapter

The situation became that the storyboards for Episode 25 were significantly delayed. I thought this was going to be severe, but the general director is also an artist. After he finished his first check of the storyboard, despite the fact that color concepts and design started from there, and in relation to the unease he still had over Iscandar, he didn’t have me work on design development with him. He said to trust his abilities and sensitivities as an artist, leaving me to concentrate on the digital data design layouts for each episode, and the battle on the Garmillas homeworld itself.

There was a huge number of shots, but we managed it somehow. As a result, however, it was very unfortunate that the theatrical release had irregular content. As you know, the work continued after the release for the TV broadcast, and the full version was sold on home video.

Space Battleship Yamato 2199

After talking with Mr. Nishizaki for the first time in 1985 and the time that passed since then, I did not expect this much to develop. 2199, like 2220 Resurrection, represent a single viewpoint in regard to the great works of the era. I think it’s wrong to view it as a historical superscript. I think it’s interesting that Resurrection was inherited by the hands of others and that a new remake was created by the fan generation. Regarding the content, I feel that some of it was very impressive and I was dissatisfied with other parts, and it’s probably not so different for you.

And the new movie. Where will the ship go from here? One thing that never changes is that it depends on the support of the audience. As the story proceeds into the future, I’m grateful to the people who have continued supporting us.

Yamato III

1/330 scale model, 1988

I can’t remember how it was decided to do it in 1/330 scale. Like the reason for it flying, an undocumented fact eventually blows away like the wind. Please allow me to use the end of this book to write about a few of these “ships of my mind” that I don’t want to end up blowing away like that.

First of all, it is Yamato III. This design was not yet 2520. The mold that was manufactured in 1988 was called Yamato III. To explain why it was named Yamato III, the battleship of the Imperial Japanese Navy was I, the rebuilt space battleship was II, and this further rebuilt version was called III. The production period was three months. At 1/330 scale, the total length of the model is 1m, 20cm.

I can’t remember how the scale of 1/330 was decided, but it was derived from the size of the display stand in a “size that would not protrude from the base.” Maybe it had to do with copy reduction.

For the cross section, dozens of thinly-sliced sheets of balsa wood were lined up a glued together. After applying polyster resin to stiffen it, the finish was polished intently. It was outfitted with details and electronic components. Paint was applied and polished. I recall that I had no intention of stopping.

Further, since I had heard that the plamodel based on this original design was recently reissued, refined by Takashi Hashimoto to match the animation design, I thought it would be good to make it along those lines. Since the form distribution of the original is fundamentally different, I figured I shouldn’t modify it for the sake of my mental health.

Yamato III now rests in the warehouse of Voyager Entertainment. [Translator’s note: for a few months, it was on public display at the Yamato-themed Cafe Crew restaurant in Tokyo.]

The general director liked the special missile boat I drew in the cross section of Yamato from the Director’s Cut calendar, and for some reason Tsukuba appeared in 2199 for the maniacs. I asked the general director, “Seriously?” and he said, “I like it, so I want to use it.” To explain for the maniacs where it appeared, you can see it reflected on the window of the security room, on both sides of the frame. Please look for it.

It was scaled down from the calendar to fit into the fire boat hangar, and though it appears in 2199, it strangely still has the 2220 anchor mark. It appears in episodes 21 and 25.

Yamato III

Hand drawing & marker print, 1986

Last but not least, this is the Battleship Yamato that triggered my first discussion with Mr. Nishizaki in 1986. In the midst of Gundam, I was asked to “base it only on the battleship Yamato, and draw it in a different taste than the old [anime] design.” I got carried away and contained boats inside it like the Popynica toy. The Chrysanthemum crest is attached, and there is a turret at the bottom. Although it has a clumsy feeling when you look at it now, it was a surprise that Mr. Nishizaki liked the “small boats that go in.” This draft was drawn with extra-fine felt-tip pen on BB Kent board and colored with oil-based marker, a true freehand drawing from a time before the digital era.

The End

Special thanks to Neil Nadelman for translation support.

Read the Yamatour 2012 interview with Makoto Kobayashi here.

2 thoughts on “Design of Yamato by Makoto Kobayashi

  1. Hi,
    Where can I find this book? Please, I’ve been looking for an art book like this but it is very difficult for me to go through all the japanese sites since I have no idea of their contents 🙁

Leave a Reply

Your email address will not be published.