Little China Doll: How Spicy Horse Imagined Alice: Madness Returns

This is the latest in a series of articles about the art technology of games, in collaboration with the particularly handsome Dead End Thrills.

Alice: Madness Returns is a very special action game, a piece of Lewis Carroll fan fiction conceived by an American (named American, naturally) and illustrated largely by the Chinese, a people not known for their absurdism. Unfortunately it's also a classic victim of checklist game criticism and marketing, not to mention it's the worst possible thing to reviewers working under deadline: long. Make no mistake, the business of game reviewing is a lot like the business of eating large amounts of Jacob's Cream Crackers in hot and rowdy East End basements. To put it another way, all games get tiring when you play them past your bedtime.

The sequel to American McGee’s Alice isn't great (8/10), solid (7/10), flawed (6/10) or average (5/10) - it’s none of that nonsense. It's a living art book, an Alice novel with the ratio of words to pictures spun around. Inspired by such wild and offbeat things as Burning Man, Dave McKean, Zdzislaw Beksinski, The NeverEnding Story, The Dark Crystal and the Brothers Quay, more than anything it's a tribute to the mechanical and the handmade. Dolls, miniatures and puppets crowd the dreams and nightmares of its Wonderland, while its 'real world' Victorian London would be right at home in Sondheim's Sweeney Todd.

You'll notice that none of these is Chinese - though there’s plenty of paper folding and porcelain in there, too. Before the first level was even white-boxed, Spicy Horse's crash course in surrealism had spawned one of the most extraordinary and diverse concept art collections a game has ever had, much of which adorns the 184 pages of The Art Of Alice: Madness Returns. The words in that book come largely from art director Ken Wong, who continues the story here.

DET: What exactly does an 'American McGee game' involve? What does he bring?

Ken Wong: American started off at id doing a bit of everything. He started off doing QA and went into level design and sound design. He’s still like that, he dabbles in a bit of everything. On Alice he started in kind of a game design role but it was collaborative. We had a combat designer, myself as the art director, and level designers all chipping in and designing various systems together. It's not like the entire game is his design, but he is sort of the leader of the studio and shepherded the project.

I think his primary contribution is creating the work culture. The fact that he set up this studio in Shanghai with some foreigners and some Chinese staff, in this environment of, 'Hey, let's create something that's maybe not the smart thing to do but it's going to be amazing, something outrageous.' That was a great environment to be in, and kind of the only environment I've worked in, these daring and stupid projects. Sometimes you fail and you learn, and I guess that's been the other side of my experience working with American. Every project we've worked on together was kind of out there, and a lot of times it didn't work out.

I did a piece of fan art for the first [Alice] which he saw, and he said, 'Would you like to try doing some work for my new game?' At the time I'd never done any professional work before, I barely knew what I was doing. Maybe I was hesitant to pack up my life and move to Hong Kong; I'd just graduated in Australia. He was working with the team that was making Bad Day LA and said he needed someone on the ground there to help out with the art. Literally a few days after he got me that job, he’s like: 'You know what? I've had enough of this country, I'm moving to Hong Kong as well.’

American and I have very different ways of working. He used to work very organically: he'd build something, try something, write something down and then iterate on it. I wanted to build a very strong foundation and get things to a high enough quality before moving on. That led to some interesting meetings where we didn't see eye-to-eye on the status of certain things.

DET: How did you avoid the 'brown and bevelled' look so prevalent in Unreal games at the time?

Wong: This was a big conversation when we were making Alice. There was always this agenda to inject a lot of colour into the scenes, to have very vivid hallucinatory colours wherever we could. Unreal is actually great at that, it can be very colourful. It’s just that everybody at the time was going for gritty realism and for some reason that meant desaturation. Since then you've seen Gears Of War add more colour, Assassin's Creed get more colourful, even Killzone.

Grimm was cel-shaded and looks very much like Tearaway does now, only they've got a much better look. So, coming from that point of view, we tried lots of different things. Grimm looks nothing like Unreal and it really prepared us for Alice where it's like, 'Okay, this is just a toolbox, what can we do with it?' The main influence for us was trying to make it look handmade, like a stop-motion film or a doll house. What we had in our toolbox was normal maps, lighting, specularity, bloom, depth of field - all these sorts of things. So we looked at what stop-motion films looked like and tried to emulate that as well as we could.

DET: The mix of Chinese, American and English sensitivities is a fascinating one, but was it deliberate?

Wong: I don't know if that was an approach we took, more like the challenge. Hopefully the result doesn't look Chinese, but perhaps it has an aspect of that outsider's point of view. When you have a generation of artists who grew up watching Aliens and Starship Troopers, that's their education and that's what they turn out. On the fantasy side a generation has now grown up watching Lord Of The Rings. Madness Returns has a certain naivety to it because the artists who worked on it, they didn't grow up with this common education. That forced them to always be learning; they could never fall back and 'riff on that thing I saw in a movie when I was a kid.' They were always having to go back to the source material for their references. That was a big thing that I imparted on them because I think concept artists anywhere can get a little complacent, especially when working in genre stuff, and especially in videogames. There’s a tendency to recycle a lot.

DET: That can be a very formalised process when big publishers are concerned, in which marketing has a very loud voice. Did you have to filter that input?

Wong: I've been sent style sheets from EA and they're scary. 'Let’s just blend A and C and there you go.' There is an aspect of that to what we did, but it tends to be crazy and off-the-wall stuff that you've never seen combined.

To answer your question on filtering the marketing machine: American himself was the filter most of the time. He would be in contact with those marketing guys from EA - and sometimes I’d be on the calls but usually not - and he would decide what was worth passing on to us and what was best ignored. It was established quite early on that he was this punk or rogue element, and it was assumed that he'd be doing things his way. He was encouraged to. There was no sense in commissioning American and his studio to make a game by the numbers, and so to EA's credit they gave us quite a long leash, and only really pulled on it when they felt they weren't getting what they wanted, which kind of became increasingly common towards the end of the project.

The conception of the game, we had a lot of leeway and room to be creative. There was sort of a change in the guard at EA Partners partway through the project. I believe what happened is that Brutal Legend came out and didn't get the numbers they were hoping for, and that sort of had an effect on where our project was heading. That was one aspect of it anyway.

There was a lot of tension related to the logo and the box cover, they were very controversial. Design-by-marketing can work, actually: you look at a lot of EA releases and they look great. But for whatever reason-- I kept saying, 'They're giving us the work experience kid,' because we were just shocked at the quality of the stuff we were getting. I don't really know what the true story was but we were just never satisfied with the quality.

DET: And yet, for all their input, the actual marketing seemed to diminish at launch. What were the numbers like?

Wong: The numbers for [American McGee's] Alice were just good enough that, ten years later, EA pulled it out and made Madness Returns happen. The numbers for Madness Returns weren't that great. You know how important marketing is, and I do point fingers at EA just not giving us any marketing support at the final hour.

American and I were in LA for the week of the release of Alice. We swung by E3, and I think the game was coming out two or three weeks after. It was released in the same week as Duke Nukem Forever, if I remember. Duke Nukem had a huge presence at E3: a booth, ads everywhere, TV, billboards. Alice was nowhere to be seen at E3. We got that through social media as well: people were saying it was hard to find copies, that there was very low buzz. The only explanation I can think of is that EA just got cold feet and weren't going to spend more money on this because they'd already written it off as a failure. Which is such a shame because you know how important that push is in the last few weeks.

DET: How did the Chinese team get on with classic English surrealism? What were its early attempts like?

Wong: I guess surreal is this very subtle and disturbing thing, very psychological. You're taking something people know and understand and representing it in a way where something’s not quite right, or where something people take for granted isn't quite the same.

To begin with [the team] drew very obviously violent things in a very straight way - gory, spiky, dripping with blood. Or quite comical, whimsical things that were strange but a bit random, like blending two animals together. There's a way that you can do that. It's almost a very English thing when you have something that's very silly but you take it very seriously; or something serious that you then make fun of. That's something we kind of had to develop because the Chinese culture maybe doesn't make fun of itself as much, or isn't as self-aware and self-critical. In British culture they may self-critique the classes or the school system, and in America you’ll make fun of Thanksgiving and how consumer culture has taken over. Chinese culture is only now graduating to this.

There have been aspects of it historically, but the young artists I was working with weren't used to videogame art being that involved. Remember, these guys didn't grow up with videogames at all. Nintendo systems weren't sold here legally until very recently and the PCs were quite primitive. Games have come late to China, and a lot of it's in the form of MMO's and mobile games. They're kind of just awakening to the possibilities of what games and game art can be. And that was exciting. Here we were with this huge project from the biggest publisher in the US, and we're like: 'We can make something amazing. This studio can be put on the map and you guys can be part of that. You can do something no one else in your country is doing.'

DET: Action games with a strong artistic component often get accused of being in the wrong genre. Was there ever a question of adventure versus combat?

Wong: We looked at the first game, American McGee's Alice, and just wanted to make a smarter version of that. We had our work cut out for us, really. This was going to be quite a big undertaking and we didn't want to reinvent the wheel. We thought we could do it all: a great story, a great environment and great action gameplay. In the beginning there was also stealth gameplay that didn't make it in. It was really ambitious. I think the premise of Alice was always one of a violent Wonderland that's always trying to kill her, so she's fighting back. That's why it does work as an action game.

But that does exclude people. I feel excluded from action-heavy games now because I spent my twenties and teenage years perfecting shooting skills and killing, and thematically it's boring now. It's just a very simple way of conflict resolution, just killing everybody. It's so great to see a lot of games growing up and finding other ways of telling stories and exploring interaction. Alice is what it is. We picked our battles.

DET: It's interesting that you looked at fan art during the early stages - though maybe not given your own background. Some might dismiss fan art as amateurish or derivative.

Wong: We wanted to put our finger on the pulse of why that first game had such resonance. Why were people still talking about Alice and making fan art of it ten years later? We looked at what people were producing and that's what helped the concept artists get a hold of what made this game special. She's such an unusual heroine for a game. She's not physically strong, she's not even mentally strong, she's mentally unstable. She's not even really an anti-hero; she doesn't want to destroy stuff or get revenge, she's just trying to make sense of her own head. That sort of thing comes across in the fan art as well. The fan art emphasises character and strangeness and all this stuff. I don't remember us ever feeling weird about looking at it, it seemed like the natural thing to do.

DET: How do you design an action game Alice for an audience often seen, especially by the bigger publishers, as a bunch of randy young men?

Wong: I'm very interested in that issue now because it's discussed by a lot of social media and a lot of people are writing articles about it, and that's fantastic. With Alice, I think and I hope we did the correct thing, which is that it wasn't really an issue. From the beginning American told us that half of the people who played Alice were females - probably more than half of the hardcore fans. There were people who didn't really play videogames but who loved Alice, and that means a lot to me. We didn't really set out to design Alice for girls or for guys. Obviously most of the development team were guys, but we just wanted to make a cool character.

We wanted her to be good-looking but not in a way that's condescending or objectifying. There's this aspect of: is she posing for the camera? Is she in a position of submission? And that's just not language we ever entered into. We always, I think, tried to depict Alice as being the main character. She's not framed in a relationship with a male, she's not the sidekick of a male, she's not a daughter, she's not the love interest. She's her own person. She's the star. It was really as simple as that.

But occasionally there were conversations... When we were trying to decide the look of Alice, this comment came from EA that, 'Oh, maybe she could look a bit more like Angelina Jolie.' That kind of became a joke for us.

There was only one instance where I remember us not being on the same page, and that was when the art team was designing the alternate costumes, the DLC costumes. One of them was the costume where she's wearing the Mad Hatter's hat, and she looks a bit like the Mad Hatter. Each domain has its own list of materials that it uses, and one of them is leather. So I gave her these sort of leather hot pants. She still had a skirt on but it was a very short one. American was like: 'No. Just no.' So we drew a line there. But every other costume we did got it right in that it was never designed to look sexy for men. It was never, 'Please have sex with me.' It was about making Alice look cool or look different.

For some people this whole debate is really silly. But it's just about: don't reduce yourself to making soft porn.

DET: Any other 'curious' requests from the publisher?

Wong: There was this really weird thing where towards the end of the project, some feedback came from EA saying the game wasn't 'M enough'. We weren't 'fulfilling the potential' of our M rating. So they were like: 'Can you make it more fucked up? Can you put more sex and violence into it?' Which is such a strange position to be in as it's normally the other way around. And we resisted. We were violent and sexy enough, thank you. Some genius in EA marketing went, 'Well, maybe the White Rabbit could go off into the bushes and you'd see him rutting away over this female rabbit.' It gave us a laugh, at least. But I think that everyone, including EA, got that Alice was not meant to be depicted like a supermodel. I don't remember anybody trying to force us in that way.

The difference between American and myself was kind of inverted when it came to Alice herself. I was prepared to iterate and thought it'd be a very long journey, whereas American was expecting it to be quite straightforward. I'm not sure what he had in his head but I wanted to really try everything with Alice. There were things I would have changed, but on the whole she was well received.

DET: Alice's hair is great, and arguably less incongruous than the more sophisticated TressFX.

Wong: We got this programmer on called Milo [Yip] and asked him, 'Can you do dynamic hair?' And he's just this wizard with this kind of stuff. He made this demo with this amazing-looking hair, and then we had to get it into the engine. From the start I was a sceptic. I wasn't prepared to sacrifice the rest of the game just to get brilliant hair. It was actually quite a battle to get our framerate back, so actually it did impact the game. For example, we can't have the camera too close to Alice while showing the background because the hair and all of its overdraw eats up so much framerate. I'm glad it was well received because we were never quite sure we were pulling it off.

DET: One of the early concepts was for an Escher-themed level you later scrapped for technical reasons.

Wong: It's funny you bring that up now after the Monument Valley trailer [Wong's new, very Escher-esque iPad game]. That particular bit of art, that level, didn't come from me, it was a concept by Tyler Lockett. That level was never really on our shortlist for the game because, I don't know, it didn't really inspire character moments or gameplay. I don't think anybody really thought about how we could make interesting story or gameplay out of that. But there was so much on the cutting room floor, which is why I'm glad we have that art book, because a lot of my favourite ideas are in there.

DET: Speaking of concept art, is there a danger when releasing those early bits of promo key art when the realtime art's still in flux?

Wong: There isn't really much discussion about those things. American is very keen to release stuff to the public, whereas I tend to be very picky about what goes out because I'm conscious about how the game is forming in people's heads. The first image we put out wasn't meant to be a literal representation of what is in the game. I don't think anybody was really disappointed that they didn't fight a giant snail on a lighthouse. What that image was saying was that here's a weird, surreal game where we're going to take something silly, a girl fighting a big snail on a lighthouse, and we're going to take it very seriously. This is going to be an epic battle.

It did have an effect on production because it helped anchor the project, that and several other images. Especially in terms of disgusting or disturbing images, when you show someone and their reaction is, 'Urgh, that's so wrong!' that's a really great feeling. And I always tell artists I'm coaching or teaching that one of the roles of concept art is to bind the team together. When you're talking about game designs in very abstract terms, a single image or a few can really pull it all together.

DET: The game takes a real turn during the Asylum level. Was there any hand-wringing over that?

Wong: It's interesting that you highlight that level because I don't think it was ever meant to up the ante, if you will. It was just a character moment, and also an art direction moment, of: is she in London or is she in Wonderland? Let's create this zone that's a little bit in-between. We had concept art of a bald Alice very early on, and American was just like, 'We've gotta have that in the game.' Dropping out all the colour and just making it white and black and red was something we were very excited to do. And again this environment of a Victorian asylum was just very exciting.

We took a lot of reference from The Elephant Man, and we had so much fun doing all the little details. I’ll tell you something: I was not an experienced art director. To this day I still haven't worked under an art director, so I don't know what 'proper' art direction is. One thing that I learned on Alice was that when you play videogames, your attention is concentrated on certain time spaces and areas of the screen, especially in a 3D world. People will spend a lot of time looking at the floors and the walls, and maybe not be looking at the leaves, the trees or the skybox. There's so much detail we created that's not very obvious. For example, the doctors in the Asylum have scalpels and other medical tools on their fingers, and the nurses have no eyes and seven fingers or something. It was so much fun to design all those little things.

So, to answer your question, if anything the Asylum was a break from the formula of what a level would be. You had Wonderland sections and London sections which were quite Mario-esque in their formula, and here you had the Asylum which was like a cabinet of curiosities where you're peering into certain rooms with all sorts of weird shit going on.

DET: Some critics described the London levels as 'film sets', insinuating a certain lack of interactivity. I'm not sure that's a bad thing in such a theatrical game.

Wong: Tim Burton's movies all have a very theatrical look. That comes from him growing up in Burbank, where they roll out these bunches of houses that are all same but just painted different colours. That was the inspiration for the neighbourhood in Beetlejuice. So, I instructed the guys that the whole game would be lit in a theatrical style. Not exactly realistic but exaggerated and to highlight certain things. Also, we just maybe weren't very good at what we were doing and it ended up looking a little fake. I'll tell you what: it’s really strange now, having moved to London myself, walking through those areas and seeing the real thing. It heartens me that our representation wasn't too bad.

Alice: Madness Returns is currently £14.99 on Steam and Origin. The Art Of Alice: Madness Returns is the much curiouser price of £16.89 £20.99 [they just put the price up, the scamps!] on Amazon. Ken Wong's new game, Monument Valley, is in development for iPad and other mobile platforms.