Building An Empire: Koei's Generals Talk Strategy

Though its name is not on the tip of most people's tongues, Koei is an important force in game development. Its Dynasty Warriors series -- which has split off in a multitude of different directions across a nearly uncountable number of SKUs -- is one of the most popular in Japan, and a massive sleeper success in the rest of the world.

The company has also opened up a new studio in Canada, with its first Unreal Engine 3 title, Fatal Inertia, recently released on the Xbox 360. The company also has its first Wii title, Samurai Warriors Katana, coming in December. We talked to three senior staff at the publisher to learn more about its fortunes in the next-gen wars, how multi-title, multiplatform Japanese studios function, and about opening a new studio in Canada with an international team.

Gamasutra first presents a conversation with Takazumi Tomoike, managing executive officer and deputy general manager of Koei's software division, about the development of Koei's first internationally-focused title, Fatal Inertia, at its new Canadian studio.

You've probably been asked this many times but is Fatal Inertia -- which was originally planned for the PS3 but moved to the 360 -- still coming to the PS3?

Takazumi Tomoike: Yes.

Will you use the SIXAXIS controller at all?

TT: I do want to promote the 360 for this event, but yes, we are going to use the SIXAXIS, and I think the controller is well-suited for the game.

How is the Koei Canada studio going so far?

TT: They are still a young company, but now that we have arrived at the point of creating a [gold] master, it seems like going through this process has given them strength and they have grown quite a bit.

Koei Canada's Fatal Inertia

Have you been basically teaching the young team by making this game?

TT: Yes, that was my intention, to show them how games are built, but that has to do mostly with the methodology or the way to create games. In terms of content and a sensibility and concept behind the games, these were really left up to them. But what I showed mainly was common methods for building a game. But as it turns out, even though my intention was to show them how to build a game, they actually learned a lot of things on their own by trying things out. They are still very young, but they played a major role in putting this thing together.

Were you leading the project mostly or was it someone that you picked from the studio itself?

TT: In terms of the content of the game, I completely trusted the lead designer, called Michael Bond. He was in charge of the design aspect of the game.

Are most people in the studio Canadian, and do they have to speak Japanese as well?

TT: Most people actually don't speak any Japanese at all. We do have however six or seven Japanese people in Canada, helping out with the development. But the Fatal team had only one Japanese person. In that sense, it was pretty much a Canadian effort.

Why tackle the racing genre right now?

TT: As I mentioned earlier, Michael Bond came up with this idea around the same time Koei was interested in making titles that would appeal more to the American and European markets, and even though it's not as popular as FPS, racing is definitely a popular genre in those areas. So there was the company's intention to try out something new, and Michael came up with this idea at the same time, so these two things matched, and that's why we decided to go for it.

How many people are working on the game right now?

TT: There are about 20 people who are dedicated to developing this title. And of course we have a lot more people involved with the CG. We have a couple of members in Lithuania, as well as people in Japan who helped out with the graphics, but there are about 20 people involved with development.

That's quite small for many next-gen games, in fact.

TT: That's true, but this has to do with the fact that it's a racing game. With next-gen platforms, CG is really where the effort needs to go to. When we have a lot of human motion and events and stories happen in that title, then it can be quite labor-intensive. But with a racing there is no complicated story, so we were able to do it with a small group of people.

Is this also an experiment in using cross-platform middleware?

TT: Yes that is true. Until recently Koei has always used its original engine, but we decided to use someone else's technology. Unreal Engine is just one of them -- there are others out there as well. But using technology not developed by ourselves was a great learning experience, and it was an experiment on our part to see what we could learn.

Do you think that you will want to do that going forward? Or is it too hard to make the middleware fit your needs?

TT: I personally think that there's still a lot that can be learned by using a third-party engine. However, Koei does have a team that is a dedicated to developing our own engine. So the long-term goal is to develop something all our own that is quite high in quality, but in the meantime, I think we could still learn from using technology developed by someone else.

Recently, a lot of Japanese creators have said that Japan is falling behind in terms of technology, because there's not as much asset and idea sharing within the industry. It seems like that is why middleware is being used. Do you think you will be able to keep up for a long time?

TT: Actually, even within Koei, we are not always able to share technical information within the company itself. And it is true that we are not as good in Japan as the U.S. at sharing information and technology, to use the latest and greatest. The United States is better at doing that I believe, and I think this is reflected in the titles, the hit titles that you see worldwide. You used to see a whole bunch of Japanese titles and products and now there are only a few that you can count very easily. So I think you're right, Japan may be falling slightly behind in that sense.

Why do you think, speaking within Koei perhaps, that information cannot be shared even within its own teams? What is the logic behind that?

TT: I think this has to do with history in our industry. With PS2 and Xbox, what developers in Japan were trying to do was bring up the performance of each platform to their maximum potential. They were not so interested in using or creating an engine that could be generally applied.

But in the U.S. that was already happening from the time of PS2. And now the industry is moving in a direction where it's not so much about bringing out the maximum potential for each platform, but to have more of a common multiplatform way of creating titles, and using an engine isn't necessarily the best way to bring out the most of each platform, but it is useful in other ways, and I believe Japan is falling slightly behind with that transition as we shift to the next generation.

Koei Canada's Fatal Inertia

Why do you think that technology cannot be shared even within one company -- that within Koei, teams can't share technology?

TT: Actually we do have a group for that specific purpose of sharing information within the company in place, and we use it. However what I was referring to was when we were creating titles for PS2, each title would use a different engine, so for something like Dynasty Warriors, we would have a specific engine that would work best for that and we would have Kessen... which would have a separate engine, and these were used independently without a commonality between titles. So that's what I was really trying to say earlier, when I said we weren't able to share information.

It seems to be happening in many other companies -- another company was making two wrestling games, and couldn't share the same engine. They had to be completely different. It seems to possibly go back even further to the Japanese PC days, the early PC days, with NEC PCs and the Fujitsu PCs. You had to maximize the graphics for each platform, and so people would keep that very secret. There was famously one company that lost the source code for their own RPG, because they hid it so well from themselves.

TT: I understand what you're saying... that seems to be a little extreme. But there are companies that are like that. And in our case, we do share information within our company. We somehow managed to do that, but when it comes to applying information effectively to make things more common, we are not always successful at that. And it is true that we struggle with not having consistent data throughout the company.

But we're really working on creating a common base for that right now, and the biggest benefit of that comes from the speed at which we can build our know-how related to building CG resources. When we share this information our efforts become more coordinated and we can speed our building up our skills and knowledge.

It seems like Fatal Inertia went through many incarnations and is no longer as much about physics. Is that true, and if so, why is that?

TT: Actually, the basic concept has not changed in that the physics engine plays a major role still, and that's how were able to create very realistic behavior of the weapons and whether or not that has a huge impact on the game or not depends on how you look at it, but we pretty much achieved what we aimed to achieve at the outset. One thing I can say is that even with the next-generation platforms, the physics engine is still heavy.

Though Dynasty Warriors: Gundam may not mean a great deal to Western audiences, the match-up could be compared with Lego Star Wars to Japanese fans. Gamasutra spoke to Hisashi Koinuma, executive officer of software dept. 2, software division, about this crossover title, and about the company's first Wii efforts.

Did the idea for Dynasty Warriors: Gundam come from Koei's side or Namco Bandai's side?

Hisashi Koinuma: Koei side.

And how did that concept come together?

HK: Initially, there was a request made by another company about using a different type of content and applying it to the Dynasty Warriors series, and that didn't hammer out. But after that, we came up with this idea to use Gundam content, because it has a very big market in Japan.

Koei and Namco Bandai's Dynasty Warriors: Gundam

Switching topics, how is working with the Wii so far, on Samurai Warriors Katana?

HK: Well, as you know, Wii is a very unique platform with interesting controlling devices, and as a creator, I really wanted to play with it, I wanted to touch the remote and try out what I could do with it. And it took me a year and a half, but we finally managed to create a commercial product. And you can say this with any new platform, but we made a lot of trial and error, it wasn't easy.

Did you find it easier to develop for than other, larger next-gen systems? In terms of learning curve for a new console to be on?

HK: You can say this about any hardware platform, but unless you're trying to squeeze out the maximum potential from the platform, it's not all that difficult to develop a title for a new platform, a next-gen platform. But when you really try to take it to the max, then that's when the real big challenge comes in. And I believe it usually takes a couple of years before we reach that state, and I think we've yet to see the best titles come out for each platform. In terms of learning the platform, it wasn't that much more challenging than any other platform. Rather, the real challenge had to do with figuring out how to use the Wiimote.

So it was a challenge more from a design standpoint.

HK: Yes, figuring out how to use the remote within the context of the game, how to play with it.

It seems like you can't do many sophisticated attacks with the Wii remote. Is that a problem for you? Or does that fit with the simple play mechanic of the Warriors games?

HK: This title was suited for Wii in some ways and it was challenging to use with Wii in other ways. But I think Wii itself can do complicated, sophisticated things, but we made the decision to keep it relatively simple because we think that Wii users overall prefer games that are easy to operate, so that's why we made that decision. But I believe Wii in itself can handle more complicated, sophisticated types of commands.

Games like this seem a natural fit for the Wii, but so far there's only been this and Red Steel. Is it possible to make, for instance, a first-person version of Bushido Blade, where very precise sword movement and realistic swordsmanship would come into play?

HK: You really need a precise match between the motions that one makes, and what happens on-screen. And unless that is precise, it will be a very frustrating and stressful game for the user to play, and I believe there are still some hurdles for before that can happen, at this point. I believe it would be a very complicated [gameplay] system... to take in that direction.

Koei's Samurai Warriors Katana

Will the sensor technology need to get more advanced before something like that is possible?

HK: That may be the case, but we're actually not sure about what needs to happen for that to occur.

Many people having been saying that a number of companies in Japan are making fewer next-gen, Xbox 360 and PS3 games and many more Nintendo games for Wii and DS. Do you think that's going to happen, going forward, in general?

HK: It is true that it's easier to create titles for certain platforms, and it makes sense that you'll have more titles for platforms that are more simple in terms of developing new titles. However, from Koei's point of view, or I should say from a creator's point of view, I am more interested in finding the right platform to express my ideas, to create the most fun out of the concepts that I come up with. So, I don't think creators are necessarily interested in developing only for the best-selling consoles, but for finding the best platform for expressing their ideas. So I don't necessarily think that the growth will continue at its own pace for those two platforms.

Koei has traditionally not been a huge proponent of Nintendo consoles, much more the alternatives like PlayStation or Xbox. Any idea why that might be?

HK: I am often asked "when did we start developing for Wii", and actually, we were shown the Wii platform about a year before it was actually released on the market, and at that time we thought it was an interesting platform, and we thought it was unique, and could do things that could not be done on other platforms. So we were always curious about it, and in fact did develop a title, which has been released [in Japan], which is what I worked on. But again, it goes back to this idea about what platform is more suited for an idea, because not every platform can do all the same things. So it depends on the idea, and if we have the right idea for the right platform, then that's the platform that will be chosen.

As a final question, what were the difficulties or challenges in taking a third-person game and bringing it first-person? What were the considerations that you had in order to make that work for the player?

HK: This time we have a rail for the movements of the character, and then there's action that happens along that rail. Both games, the third person and first person, I think they're both action games, but the treatment of the action had to be changed. That was the most challenging part. Another point I wanted to make was we didn't try to bring Warriors to Wii, we started with Wii, and we looked at the Wiimote and we figured out what would be the best title to bring to Wii to make good use of the Wiimote.

One criticism frequently levied at the Dynasty Warriors series by the western press is that the installments are near-indistinguishable. Here, Gamasutra spoke to Akihiro Suzuki, executive officer, software dept. 1, software division, about that issue and the series' future.

The Dynasty Warriors game series has been very consistent in terms of gameplay for a long time. So what goes into making a new version different from all the others?

Akihiro Suzuki: The basic series concept is the same throughout the series, which is "one against a thousand", that sort of a theme. And the games are all based on that, but aside from that, the challenge is to give each title the same... flavor. We always work toward that.

The difference often seems very subtle. Is that enough for you? Or do you want to take a bigger step?

AS: So... for the major releases with numbers, like two, three, and four... for these it's mostly about increasing the volume and also adding more stories. But this may not seem like much of a difference to North Americans or Europeans. For people who understand the history, people in Japan specifically, for them it is a pretty big difference between each release. That said, for the titles without the numbers like Empires and Orochi, which we are displaying this time, is one such example. They have a different concept from the major releases. So we do try out different things with those types of titles. But we are going to try to create more of a difference between titles, even with our major numbered releases, as we go on into the future.

Koei's Dynasty Warriors 6

Do you think that these games could be used to teach actual history of the events they're based on?

AS: I don't think it can be used as an educational tool, because we arrange or rearrange the actual historical events. However, it can inspire a person to become curious about that historical period, so that person goes on his or her own search to study that period.

The "one against a thousand" game mechanic has been consistent throughout. Why do you think it has been so popular? It has been incredibly popular -- even to the point where it inspires other people to copy it.

AS: I think it has a lot to do with the fact that the operation is quite simple. It's easy for anyone to get used to the game and play well, and the speed of playing and immediate satisfaction that comes out of seeing your action on screen, I believe is the key for its success. Because all types of action games, everything you do is immediately reflected in the game. So it's very important to have an immediate sense of fun and satisfaction, and I think when you achieve that you get wide acceptance.

You're not a designer, but from a design standpoint, surely you're familiar with the series. How do you keep the flow of the game going throughout the game? Because it's very important that the action is constant.

AS: Well actually it is not that difficult to have continuous action. Of course, things have improved and it's easier than ever to create a long smooth sequence of action, but it has more to do with making it fun. It needs to continue, but it also has to be fun and enjoyable for the player. And that's where the challenge is.

How long do you think the series can continue as it is without making a major change?

AS: As long as [development studio] Omega Force wants to continue with it, I think it will continue.