Writer and directors Kevin Adams and Joe Ksander discuss the making of the Chinese-Canadian animated co-production about the bittersweet power of memories.
Set to debut on Netflix on Friday, September 7, CG-animated feature Next Gen is a funny, exciting and heartwarming adventure that explores the bittersweet nature of memory and celebrates the value of friendship. Written and directed by Kevin Adams and Joe Ksander, the offbeat action-comedy tells the story of the unlikely bond between an angry 12-year-old girl named Mai Su and a runaway combat robot, numbered 7723, as they team up to defeat a madman’s plans for world domination.
Next Gen is produced by Javier Zhang, Jeff Bell, Ken Zorniak, Patricia Hicks, Charlene Kelly and Olivia Hao. The Chinese-Canadian co-production was produced by China’s Baozou Manhua, with the animation provided by Toronto-based Tangent Animation. Netflix picked up worldwide rights (excluding China) to the film at the Cannes Film Festival earlier this year in a much buzzed-about $30 million deal.
Based on an online comic about a damaged robot by anonymous Chinese television personality Wang Nima -- who is best known for wearing a radish head mask -- and produced by Baozou, Next Gen examines memory and loss amid complicated family relationships. The film stars Charlyne Yi as Mai, John Krasinski as 7723, David Cross as the loveable Steve Wozniak-inspired Dr. Rice, Michael Peña as a foul-mouthed French bulldog named Momo, Jason Sudeikis as corporate megalomaniac Justin Pin, and Constance Wu as Mai’s robot-addicted mother, Molly.
Next Gen is set in a high-tech city of the future called Grainland, a vibrant, buzzing world endowed with a level of detail usually only seen in live-action movies. Production designer Craig Sellars, who had previously worked on live-action films such as Marvel’s Guardians of the Galaxy and Robert Zemeckis’s forthcoming Marewncol, was brought on board early during pre-production in order to help create the look of Grainland, which combines a retro mid-century vibe with a futuristic aesthetic from the 1970s and 80s.
Adams and Ksander have collaborated since 2009, when they both worked on the animated feature 9, and discovered a shared love of Kung-Fu movies and The Abyss. A Toronto native, Adams began his career at Disney Feature Animation, where he worked on now-classic films like Hercules and The Emperor's New Groove. He went on to production design at DreamWorks Animation and Reel FX Animation, and was the creative director of ARC Animation. Ksander grew up in Silicon Valley, where his obsession with monsters and spaceships drove him into a career in visual effects. He animated characters for movies like The Chronicles of Narnia, Pacific Rim and Transformers: Age of Extinction, and also worked as an animation director at Walt Disney Imagineering, where he animated the full-scale Na'vi Shaman figure for Pandora – The World of Avatar.
AWN had a chance to catch up with the writer-director duo to discuss the making of Next Gen, including the genesis of the story and characters and the development of the look of the film. read the full Q&A, which has been edited for length and clarity, below:
AWN: How long have the two of you been working on this project?
Joe Ksander: I started about three years ago. It was an earlier iteration of this movie where I was just helping out as a friend art directing for somebody. But we both came on as writers in January of 2016, and then, about a month later, Olivia Hao, our producer, asked us to also come on as directors. So really, we’ve been on as writer directors since January of 2016.
AWN: And where were the two of you based while you were working on the film?
Kevin Adams: Originally in Burbank, and then we spent most of the production in Canada, because the bulk of production was done at Tangent Animation in Toronto.
JK: Yeah, it was about a year in Burbank doing editorial and story. We were also working with a Toronto storyboard studio called House of Cool during that time, but we were based in Burbank until January of 2017. Then we both moved back to Toronto full-time, although we still both came back to California quite a bit because the voice actors were there, and stuff like that.
AWN: And were you also working with a team at Baozou, in China?
KA: In the beginning, we did story sessions with them to try to find which elements they liked from the earlier versions of the film, and then we put that stuff away. Then there was some correspondence, mostly with Olivia Hao, the EP on this project.
JK: Baozou is a Chinese multimedia company that produces all kinds of online and TV content. But they are also an online comedy group. Early on, in the outline phase, we were working with their writing team to hash out some of the big story issues and come up with gags and stuff. We worked with them pretty closely, and actually traveled to China, in January 2016, to meet with Olivia and the rest of the team. She showed us around parts of China like Guangzhou and Shenzhen, places she thought would be inspirational for the film, and a lot of that showed up in the design of Grainland.
AWN: Describe Grainland for us, and tell us some of the thinking that went into the development of the look of the city.
KA: Grainland is a fictional location, roughly based on, as Joe says, Chinese cities like Guangzhou, but also with a heavy mix of American cities like Los Angeles and New York. There’s also a touch of Amblin influence to a lot of things as well. We can’t help but reference the stuff we loved while growing up, so Mai’s neighborhood, for example, has a little bit of ET in it, as well as a little bit of China.
JK: We often refer to the city as Happy Blade Runner.
KA: When we were talking to our production designer, Craig Sellers, he was very excited about 1970s neo noir, because we were really trying to push the idea of not referencing animation, and referencing more live-action movies and old neo-noir films.
AWN: Sellers has background primarily in live-action, right?
KA: Craig is an old friend of mine from my Disney days, but since then he’s gone on to design Guardians of the Galaxy and stuff like that. But we got him specifically because he didn’t really do animation. In fact, when we had our first meeting and started talking to him, he said, “Well, you know, guys, I don’t really do animation.” And it was like, “Well, that’s exactly why we’re here to talk to you!”
JK: We were talking about things like the French Connection, The Parallax View, 70s neo-noir techniques for shooting a city that were unexpected, especially in animation, using a different kind of film language. Craig was really excited by that, and he brought a lot of influence to it, including a heavy dose of Canadian brutalist architecture from that era. So it’s a mix of us visiting Shanghai, us watching Blade Runner, and Craig taking his own architectural inspiration from the stuff around him, visiting universities and stuff in the areas around Ontario that had that look to it.
KA: We actually used a lot of 70s design references, but we tried to be selective about it. Mostly people do bell bottoms and afros, but there’s also a lot of really cool architectural and industrial design from that era, and a lot of the elements in our movie are influenced by that.
AWN: You also mentioned Amblin and Blade Runner as influences. Are there other cinematic references?
JK: We definitely captured ET in there, but there’s also movies we grew up with like Akira and Macross, and a lot of Japanese animation from the 80s like Cowboy Bebop. In the 70s things were a little more cartoony, but in the 80s things were given more of a live-action feel, but in a pushed animation kind of way. Because we were doing animation that was inspired by live-action, we looked at a lot of the tools that they used, and plus there’s just all cool robot stuff that we loved from that era that we referenced as well.
KA: In fact, if you go through the movie, it’s just chock full of references. You can call them an homage if you want, but really we loved the stuff that we loved, and so instead of hiding from it, as long as we’re not copying it, we’re happy to reference to it.
JK: Yeah, especially when you’re doing a modern movie about robots. There’s a whole history of movies about robots and you can’t ignore that. We imagined a world where it’s like that future that takes place maybe on a different timeline. It’s much more influenced by the 70s and early 80s, and so there’s definitely a thread of that in the design.
AWN: Going back to your overall approach for the film, how did tools like virtual cameras and lighting to help achieve the look of Next Gen?
KA: We played with some virtual camera stuff, but really it was a meticulous, really deep understanding of what happens with live-action cameras. For example, you’re doing a pan across, in live-action, it will tend to cut when the composition falls apart and starts to look ugly. That’s when the editor will go, “Okay, now I have to find a new piece of footage.” We did things like that. We would build in the problems inherent in live-action cinematography. Same thing in animation, if you establish a city, it’s a city shot, you tend to do a one big, huge sweeping camera through the city, because that’s actually cheaper than doing a whole bunch of different ones.
JK: But in live-action, sometimes they’ll shoot b-roll or montage footage. You look at the opening of Dog Day Afternoon, and they introduce New York by just showing shots of the city, seemingly at random, but they cut together nicely. We tried to do things like that, which in live-action are fairly easy to do, in our film. You just steal shots or you pick up what you can. But animation is incredibly complicated, because you have to build this new location from scratch, just to make it look like it’s something we caught on the off chance that we just happened to have a camera there that day.
We wanted to expand the borders of the frame a little bit -- even though everything ends one pixel outside the frame, the idea is to make it feel like it goes on forever and we’re just happy to catch these pieces and make the world feel a little bit bigger. Mai is an isolated girl who doesn’t have a lot of connection in the world, so isolating her in the city this way was a great tool for storytelling purposes.
KA: We also had a camera finaling crew that was very attuned, adding tiny, almost imperceptible moves to the animation that a cameraman would do while following action.
JK: We had another pass to actually make the camera feel like it was operated.
KA: For the lighting, we lit it very much like a live-action set. In animation, there’s a tendency to light each shot or each section a little bit differently, and then leave more of the polishing to the compositing. In this one, we lit actual sets like you would light a live-action set, and then kept the adjustments per shot to moving bounce cards and things around. Then we just let the happy accidents come with the rendering, which gave it a much more live-action feeling.
AWN: Were there any rules or other constraints that you used in your approach to designing this world?
JK: There were some cinematic techniques that we tried to hold onto. Kevin likes to call them “long term cinematic payoffs.” For example, in the city, Mai is isolated. She’s lonely in a big, crowded world. So we tried to make sure that we never showed a clean band of sky across the top of the screen. There’s always something cropping out the sky, either a character, a robot, or a building, even if we have to make these giant mega buildings in the distance that get cropped off. There’s always something there, kind of penning her in. Once things resolve themselves, near the end of the movie, then you finally start to see a crack of the sky across the top. As an audience member you might not necessarily notice that’s happening, but hopefully there’s a subtle psychological effect that happens over the course of the film that lets you know, “Oh, now she feels differently than she did at the beginning.”
AWN: Let’s talk about Mai. She evolves quite a bit over the course of the film. Tell us about her development as a character.
KA: This gets into a broader topic that I’ll just touch on lightly here, which is that the film was considered to be a little bit challenging by a lot of North American audiences because the character was damaged and aggressive and she had problems. Although if something were challenging in a Marvel film, for example, that’s okay, but it’s not exactly what people are used to in an animated film.
So there was some tension when they were watching Mai and we got comments like, “She feels a little bit angry,” and “Maybe you want to make her softer and not so violent,” and stuff like that. But what we believed was that if we really knew what her damage was and where it came from, and she was acting consistently based on that emotional damage, so instead of taking away those hard edges, we wanted to make sure that we really understood why she was feeling them, and then tried to give the audience some insight into that, because then you’d have the more emotional reaction and she would feel less false.
JK: I was going to say, sometimes people throw around the phrase, “Your character should be likable.” We didn’t need you to like Mai, we just wanted you to care what happened to her. I think being true to her emotional baggage made it easier to relate to her; either you feel that way or you know somebody who feels that way, and it feels authentic.
So even though Mai can be challenging, she’s hopefully like a real person. None of us are perfect. We all have these problems that we have to deal with and we deal with them constructively and non-constructively. We wanted to be honest about that. We didn’t want to make anything magic, even though we have robots punching each other into space. We wanted the emotional story to remain true and feel authentic, and that’s where Mai came from.
AWN: What were some of the challenging of depicting anger in a character that wasn’t necessarily meant to be likable, but still relatable?
KA: You have to be very consistent about how she acts relative to anything. You have to know when she’s lying to herself, so that you can show that in the acting. For example, one thing we tried to do is if Mai’s angry, there’s a second where you see that she’s a little bit sad first.
Her thing is that she turns sadness into anger, and we wanted that to be consistent throughout all the animation. So if she’s going to yell at someone, first she thinks about it, she’s sad, doesn’t want to deal with that, and then turns it into anger. Depicting these tiny reactions consistently lets audiences see that she’s reacting to things in a way that’s based on the damage inside her.
JK: Even during the writing process, when we would run into a situation we needed to resolve, a conflict or something, for story reasons, sometimes there would be an easy answer, or a fun or clever way to end a scene. But then we’d stop and we’d ask ourselves, “What would she really do in that situation? What would it really be like if I was in that situation, and I felt this way?”
Sometimes that led us down different paths than we expected, because we could answer the question but it would feel a little bit false, even if it was funny or fun or exciting. But if we found anything truthful in what she did, that was almost always the right answer.
KA: Yeah, which got us to our most challenging moment, which is where Mai basically suggests that a robot kill a kid. It’s a pretty hard thing to put into an animated film. But when we were talking about it, it was like, “Okay, if we’re going to be really honest, what’s the worst reaction that she could have?” And we said it out loud and we said, “Yeah, that’s really tough, but we have to run with that,” because that’s the one that felt most honest.
AWN: Let’s talk about 7723, who also evolves in a number of ways over the course of the film. How do his physical changes mirror his emotional states and moral compass?
JK: 7 has this design language in his action and his posing that matches his character, which we also did in the sound design, with the help of sound designer David Acord and his team at Skywalker Sound. When 7 first gets built by Dr. Rice, he’s a robot, he’s very robotic and he’s designed to learn from humans. In fact, he’s designed to learn morality from humans the way we learn it from each other, to be the good guy, but he gets hijacked by this angry little girl and learns from her instead.
When we start the film, he’s very symmetrical. His action is very robotic and very mime-like. He’s very controlled, and he moves around. He’s on wheels. As he spends more time with Mai, he starts to physically relax a little bit. He starts walking on two legs like she does, and we see him start to become more asymmetrical. His eye shapes start out very round, but then eventually they become more and more expressive as the film goes on. At one point, he even gets a mouth and starts talking with it. In the beginning, he has a very robot-y effect to his voice, very digital, and over the course of the film, that lessens -- he’s very human, he’s very relaxed. He’s got no robot affectation on his voice. Then he loses all that because of the big sacrifice he makes at the end of the movie.
Again, as an audience member you might not notice these changes, but -- like the proverbial frog in boiling water -- you can tell that it just feels different. Then we take all that away at the very end and make him robotic and stiff again. There’s an emotional reaction to make you realize something is wrong. Something isn’t right here.
AWN: Let’s talk about Momo, the adorable foul-mouthed French bulldog. How was that character developed, and how did you handle animating all the bleeps?
KA: Very early on with Momo, it was like, OK, we’ll let him be as outrageous as he needs to be, because we knew we could tone it down later if we needed to. But the fact that he’s so consistently aggressive in defense of Mai made him even more loveable. He can be as horrible as he wants if he’s always doing it for a morally good reason.
JK: Yeah, exactly. The animators did have a bit of a challenge with Momo, especially early on, because we had Michael Peña, who’s an amazing actor and equally amazing swear-smith. He was really great, because he added a lot of fun to stuff that could otherwise just be seen as aggressive.
Animators -- good animators -- will try to communicate visually what’s happening in the soundtrack, which meant that there was a lot of mouth shake with Momo that we had to dial back on, because even if you couldn’t hear the actual word, you could see it in the animator’s work. They ended up having to blur the lines a little bit so viewers can’t pick out what he’s saying just by watching.
AWN: So the bleeping isn’t just an audio thing, it’s visual.
KA: We often put M shapes in for F shapes, because they still read, but they’re a tiny bit more ambiguous. We also found that the few extra frames of bleep in front and behind things makes a huge difference, how much of the “F” sound you hear before a bleep means that more of it is left up to your imagination. Our editor is now an expert at exactly how to bleep things.
AWN: Talk about the casting. I know you made an effort to avoid whitewashing, as far as the Asian characters in the film are concerned, but can you talk a little more about how you found your cast?
JK: A lot of that -- not being forced to whitewash the characters -- came from the support of Olivia Hao and her team at Baozou. They had no interest in doing a movie specifically about white people. They just wanted us to tell the story that we needed to tell, and these are Asian or Asian-American characters.
KA: We were cognizant of the whitewashing because, especially when we started, there were a few movies that came out where it annoyed us, so we wanted to at least be fighting the good fight. Allyson Bosch was our casting agent, and she helped us find amazing actors and actresses, which I think should always be the main goal. We went for a small number of people who we were really excited about and we got most of them, which is great.
JK: One of the things we were looking forward to is working with people who play comedy, so folks like Constance Wu, Jason Sudeikis and Charlyne Yi, who are great actors but they also understand comedy, because there’s a natural sense of timing, even when they’re doing drama or big action moments, there’s a natural rhythm and entertainment to the way they can play against each other.
There’s a sequence near the end of the film where Molly, Mai’s mom, has an emotional confrontation with her daughter. She lets out all of her emotional angst, and they come together, and it’s a big reunion moment. But we had to get it done fast, so we leaned into the fastness of it, and in the early track record we actually sped up the dialog. But Constance was like, “No, let me try it. Let me try it.” She hit it at 100 miles an hour, and it’s one of the funniest bits in the movie. Mom just spilling her guts, the entire emotional arc spilling out in something like 15 seconds. It’s pretty spectacular.
AWN: The soundtrack is very striking as well, especially the opening track, Bikini Kill’s “Rebel Girl.” Tell us about the music in the film and what influenced those choices.
KA: The composers, [Los Angeles-based songwriting duo] Alexis & Sam, were a large part of that. We’ve worked with them and their indie group DYAN before on other projects, but they’re equally at home doing big orchestral stuff. They’ll do that giant orchestral piece at the end for a 60-piece orchestra, but they can also write a Punk Rock song or an indie song. All the humming and singing in Mai’s bedroom is all Alexis and Sam. That’s very much their voice, which makes it very specific.
The other part of it was that, in the movie, we wanted to express Mai’s inner angst in two ways. One, that there’s an angry voice there. That’s the person that she wants to be, the rage person. Then there’s an inner, more innocent side of it. Thematically, we said, “Okay, there’s an acoustic, raw version of that,” which is ukuleles and so on, and then there’s the heavy version, which is Punk Rock songs.
JK: It’s worth mentioning that all the songs that represent Mai are all female voices, female vocals, even the Heavy Metal and Punk Rock and stuff like that. For one, it’s a fresh sound. But it speaks to what’s going on inside Mai. There are a couple of background songs that are male voices, but those don’t represent May; they represent other things in the world. If you can track her emotions listening to the songs that she’s listening to, you can put the audience in Mai’s headspace.