Q&A: Capcom's Kujawa On Revisiting Classics, Bullet Hell
That experience has carried over to his current role as Capcom's design director for the United States and the UK, where he is overseeing a number of projects that should be familiar to nostalgic gamers.
Kujawa sat down with Gamasutra to discuss Capcom's current third-party development structure, the challenges in revisiting franchises such as Commando and 1942 via PlayStation Network and Xbox Live Arcade, and how the company is trying to straddle the line between accessible shooter and brutal bullet hell.
Keeping Communication Strong at the (Third) Party
Kraig Kujawa: Iím the director of design for the U.S. and UK Capcom CEI. Technically Iím watching over the design and direction of all of our titles. I am heavily involved with [Wolf Of The Battlefield: Commando 3 and 1942: Joint Strike XBLA/PSN updates] right now, working with developer Backbone to just get the direction.
What we have here is an interesting development model. What we have is an internal team who came from development. We have a direction of production (thatís Adam Boyes), a director of design (thatís me), weíre getting a director of art, and what we do is we work with design teams to drive the directions of the Capcom U.S. and UK projects.
Every development team has a lot of strengths, and they have weaknesses, so what we try to do is we try to supplement any weaknesses they might have, and then let them concentrate more on their strengths. Thatís kind of our development model.
Iím pretty involved with Commando and 1942 right now. Iím also involved with Dark Void that was announced back in October, and, to a lesser extent, on [piratical strategy title] Plunder, because youíve got Certain Affinity, with ex-Bungie guys, so theyíre going to make a kick-ass product. You almost just stand and watch them work their magic. Thatís my involvement on our projects right now.
I actually heard from Christian Svensson about the developer summit you had just before GDC, where you had all the developers in here and played some of the games. It sounds like youíre trying to approach this differently than other publishers with external studios Ė what is the different mindset, would you say?
KK: Yes, actually that happened in this very room. We feel like, we came from first-party development, everybody in our team has been in development, so we know what itís like. We also know the advantages of being able to talk to one another in different teams in order to share tech and solve certain problems, so what we wanted to do is different from most third-party development.
We wanted to treat our developers as partners. We want to do multiple games with them. We want them to talk to one another. We want them to show each other what theyíre working on. Conceptually, if one developerís having problems with this, they can talk to other developers and work out problems.
You can imagine what kind of breadth of developers Ė we have Backbone, we have ex-Bungie guys, we have another really, really good team that we canít talk about, and we have some guys in Sweden, some guys in Scotland. Really talented development partners, where they can actually talk to one another and try to overcome challenges.
Most of the time in third-party development? Nobody in third-parties talks to one another. Theyíre out on their own island, and if they have problems theyíre kind of just stuck there and they have to work through it. Here at Capcom what we want to do is act as the cohesive glue to keep these guys talking to one another and almost create a third-party family. So far we have a lot of examples where thatís worked really well.
Do you have any examples that you can mention?
KK: We have A.C.R.O.N.Y.M. that developed Rocketman: Axis of Evil for us, they have a lot of PS3 tech, and they also have a lot of experience with the PS3 headset support.
Certain Affinity, of course, have a lot of 360 experience, so when we put together the Plunder deal, what we ended up doing was have A.C.R.O.N.Y.M put together the PS3 SKU, and also do the PS3 headset support, so theyíre sort of working together to do these multi-SKUs, and I really donít think thatís done very often.
Revisiting the Past
The style for Commando 3, the illustrations are pretty cartoony, and the in-game stuff is Team Fortress 2-looking. What was the mindset for the aesthetics?
KK: Thatís a good analogy. What we wanted to do was make it more casual and friendly for all players, as it is an Xbox Live game. So I think when you make the graphics a little more cartoony you can get away with more.
You can use flamethrowers to set guys on fire, you can run over guys, hit them with grenades and you wonít get into that weird space where it feels a little more mature. So you can just have fun with the cartoony graphics. We think that because itís over the top violence itís sort of a good way to approach it.
A good comparison would be the Rambo movie that just came out, right? The violence is very over the top, but the characters feel very real, so it kind of feels a little uncomfortable. We wanted to take cues from that over the top violence but balance it with the cartoony graphics. We felt [that] was the appropriate way to approach it.
For both of these games, how much was borrowed and how much did you feel needed to be changed? There are a lot Ė and will continue to be a lot Ė of instances of Capcom taking older titles and revisiting them. I assume thereís not a methodology and itís a case-by-case basis, but how do you determine whatís good about a series and what needs to be taken out?
KK: We do a number of things. In some cases the producers from Capcom are still working at Capcom Japan, so we actually throw some ideas back and forth with them and get their sign-off on it so far as, ĎThese are the features that we think are cool, what do you think about it?í And they give some feedback. They donít drive it, but will give us feedback on what they think.
The cool thing is, Iím a huge arcade fan. The guys at Backbone, they have a huge arcade there, and they know what the essence of these games are. We donít want to disappoint any fans, so itís a balance of keeping enough things that are the same, and tweaking it enough, adding enough new things to keep it fresh for everyone.
So with Commando 3, we very much kept the action the same but added co-op play and co-op vehicles. The co-op vehicles add a really cool twist: when you get three guys in that vehicle, and are going over stuff, it adds a really cool piece of depth that wasnít in the game before.
With 1942, the joint strike attacks completely change the game Ė the spatial game mechanics we have with the two planes, trying to keep far apart so that you can get the maximum impact with the weapons, that adds a ton of depth. But we still lift a ton of art from the original game.
We still look at 1942, 1943, 194X, we watch the long plays that are all online, where you can actually watch people play through the game. Backbone actually has 1942 and 1943 arcade machines. So you do a ton of research, you do a feedback loop with Capcom Japan, and I think weíve hit the sweet spot with the games.
There hasnít been a Western-developed shooter that was a major release for a long time. The last one I can think of that wasnít on a handheld or something was Project X-2 from Team 17/Ocean. The expertise doesnít necessarily exist natively, so how did you go about managing that?
KK: Thatís a good question. When I was in development, I would never have thought that Iíd end up contributing on a shooter. You would think that genre was gone and buried, so thank God it isnít anymore. I think the first answer lies with Backbone. They have a lot of good arcade DNA. Theyíve done a lot of conversions to Xbox Live Arcade. They have their own arcade and Michael, who runs the studio, is a huge arcade fan. These guys just get casual play and pick-up-and-play.
But, honestly, a lot of it is just a learning curve. The best thing about these games is that we got them up and running very early. Once you get things running early you can start playing them, seeing what does work and what doesnít, and between that and studying the classic games you can see, ĎOh, thatís why they did that patterní, ĎOh, thatís how they got the ebb and flow from moment to moment.í
I think itís a learning experience, and maybe thatís why we put more time into these games than was originally slated, because we wanted to do the originals justice and we had to nail what a shooter was all about.
Cooling Down Bullet Hell
The 194X series never seemed to be very bullet-oriented, it was always more that the planes were coming to get you. But with the new game, it seems to be more of a combination of the old-school stuff and the more modern, curtain fire, bullet hell approach?
KK: You totally nailed it. The trick is that most people remember the earlier ones, because thatís what they grew up on: 1942 and 1943. And then the hardcore players who really know the history Ė like yourself Ė know 194X.
What we wanted to do is make sure that the people who remember it in the beginning still have some things to latch on to, but that the hardcore notice some stuff as well, so we take grab-bag of the best bits, that we think fit together, and thatís the first layer. The second layer is the co-op attacks that fit on top of it. So thatís how we develop it. We want to take the best of each one, and I think thereís a little bit of each game in there.
Because there have been two quite excellent shooting games on Xbox Live Arcade Ė Triggerheart Exelica and Omega Five.
KK: How good are you at those, by the way?
How good am I?
KK: Yeah, because I get owned a lot.
Iím bad Ė I like them a lot, but Iím bad.
KK: Yeah, Iím in the same boat as you.
I play Cave shooters a lot like Do Donpachi and Ketsui, which are also coming to 360.
KK: Ikaruga? Have you played that as well?
Yes, I try to play all the shooters. But when I was in Japan I met with a guy from Milestone whose job at Milestone is just programming bullets. Milestone did a bunch of games, none of which made it here except for Chaos Field. Thereís also one called Radirgy. How much attention is being paid to those specific shooter mechanics?
KK: We played Ikaruga quite a bit. I think itís an awesome game, but itís really, really hard, and just to interject into the conversation we just had - we love the games but we canít get through them.
I want to love the game and be able to get through it. And thatís kind of the mindset weíre developing these games with. We want the casual audience, and we want people to be able to enjoy every boss. Whatís the point of doing five or six bosses if only ten percent of people are going to see the fifth boss?
So we do have difficulty settings, but the achievements and everything are tied into the normal one and the slightly harder setting. Rey Jimenez, whoís the producer on the project, is the biggest die-hard shooter fan. We were taking screenshots of bullet patterns from Ikaruga, and boss reference art, and saying, ďHey, Rey, what do you think of this?Ē So weíre very cognitive of everything. Weíre diehard.
I downloaded Exelica and Omega Five Ė itís kind of weird and quirky. But weíre huge shooter fans. One of the things in 1942 with the tanks is that you can blow of the turrets and they keep rolling, thatís from Tiger Heli.
One of my fondest memories from Tiger Heli is watching the really dumb tanks creep after you blow off their turrets, and I felt really sorry for them. So I said, ďWeíve got to do that in í42.Ē So we do it in level two. Weíre all super diehard shooter fans, so thatís why I think itís going to come out well, because weíre fans of the genre.
When I was watching Commando, I assumed it was going to be very Cannon Spike oriented, but it reminds me more of Ikari Warriors now. Where Commando 2 was very vertically scrolling, itís now both directions.
KK: I think a lot of that is that we want to use the whole screen, and thatís a feature thing. It was kind of sad to lose the vertical screen, but thereís a lot of shooters like that on Live Arcade, so we chose to use the whole screen and that gives you more room for the co-op.
It is more like Ikari Warriors meets Commando in the way itís a four-way shooter, but I think weíre intuitively programmed to expect that now in a way, when you have that sort of perspective, so we want to embrace what youíve learned already and then layer Commando on top of it. So there is a bit of leeway, which weíve given ourselves there, for sure.
Moving a fully vertical series to also incorporate a horizontal plane Ė what are the logistics of that?
KK: I can give you some of the benefits. We have some explorable areas Ė you can blow up areas and move into them. But logistically you have to accommodate for players being able to dodge more of the projectiles. Because before you had them move up and down, and usually what would stop you from going up is the boss, right? Well now, the boss is still here, but you have this much more room.
So it really comes down to iterating: get a first pass of it. Here at Capcom we try to constantly engage the community, have them come in every two weeks and playtest, so you put down your regular bullet patterns, have the community and focus testers come in and play, and think, okay, thereís too much leeway. Letís add another bullet pattern here, letís tone these guys back.
What weíre having to do is different patterns and more different patterns, as theyíre able to move around so easily and dodge. Honestly what youíd see a lot is curlicue bullets, which are easier to dodge when you have more horizontal space, and then youíd see a lot of the two-way.
You donít have enough of those to make you move around normally, so you have to put in more of them. You can do a little relay stuff where you can actually predict where itís going to go, and then get closer to that rather than just firing where you used to be. So thatís how we accommodate it: iterate, iterate, iterate, and then watch how people react to it.
Do you find that youíre putting more low-level AI in versus straight patterns?
KK: To an extent. AI is difficult because you donít have much dev time to spend on it. Especially with 2D engines, you donít have as much out-of-the-box AI as you would working on Unreal [Engine 3]. So we do have some special case AI, for example, if the player does this, do this Ė very special case Ė and then layered in between we have some pattern stuff, which is typical of arcade games.
It seems that itís slightly less pattern-based than before, so it varies up the experience more that rote-memory.
KK: Exactly, it will predict where youíre going and try to intercept.
It seems that both of these games are being made for the middle-range casual who remembers arcades Ė why focus on that instead of going for the hardcore guys that want to one-credit the whole game?
KK: Spot on. Yes, as Iím sure you know, itís more a functionable audience on Live Arcade, and some of our own personal biases Ė we want to make sure that people can play this game.
Casual people want to get in and get out Ė they donít want to get frustrated. Some of those classic shooters might do some of that for them, so we just want a shooter thatís really easy to play and enjoy casually. Just hunker down with a couple of beers and your friends and get through it and not get frustrated.
For the people who really like that diehard feeling we have the harder difficulty levels Ė we have four of them in fact, and we will have some achievements tied to it to show that youíre a badass with shooters. But we donít want to exclude everyone, because we feel a lot of shooters do that. God bless, thereís a lot of shooters I absolutely love that do that.
It seems with 1942 you could have a hardcore one-credit mode where you couldnít get those kind of scores elsewhere.
KK: I would love to do that. Youíre the second person to mention that, and if we had time weíd love to do something like that.