By Barbara Robertson

Two things to know about Sony Pictures Animation’s Hotel Transylvania. First, it’s a CG monster movie unlike any you’ve seen before. Second, that’s because director Genndy Tartakovsky insisted on a 2D style.

“The thing about this show is that it wasn’t about naturalistic acting performances,” says Patrick Danaher, a senior animator at Sony Pictures Imageworks, and a mentor at Animation Mentor. “This show was about physical comedy, appealing wacky poses, and a wild range of motion rather than natural and emotionally powerful performance.”

Patrick Danaher

The story centers on Count Dracula who has turned his castle into a resort hotel for monsters, a place for a little rest and relaxation away from those pesky humans. Dracula has a second agenda, though, too. The castle hideaway is keeping his daughter Mavis, about to celebrate her 118th birthday, safe and sound. And then, along came Jonathan, a goofy backpacker, a human, who stumbles onto the scene.

Dracula, Mavis, and Jonathan have a supporting cast that includes Frankenstein, a Werewolf family, Quasimodo, the Mummy, and dozens of others, characters that are perfect fodder for Tartakovsky’s graphic style.

“Genndy comes from a cartoony background,” Danaher says. “He had never directed a CG movie before. So, he was worried that when he moved into CG he would constantly be told, ‘I’m sorry, you can’t do that.’ But the [Imageworks] animation director and crew were willing. We ended up copying his wacky ideas based in the world of 2D into 3D.”

Willing was one thing. Execution was another. “The 2D style was very difficult,” Danaher says. “There were so many situations where you would do your first pass of animation on a shot and think, ‘I’m in decent shape here.’ You’d show it to Genndy and he would make it three times wackier and more designed.”

Tartakovsky would do that using a Wacom tablet and in-house software to draw on top of the keyframed poses and illustrate what he wanted.

“Many times our first reaction was, silently if not vocally, ‘I don’t know how I can do that,’” Danaher says. “How can I get that 2D drawing across in 3D animation? But, in almost every case, we managed to make it work. It was a massively educational and growth experience for most of the animators on the show. It forced us to really look at our posing and our animation from a 2D perspective.”

Making the task more difficult for the animators technically was that most of the characters had rigs fitted for a more natural performance. The show had been in pre-production for years. Most of characters had rigs designed and built before Tartakovsky agreed to direct the film.

For example, Danaher animated a shot that had a Gremlin in it, a supporting character that moved using a rig designed three or four years earlier. “Because he was only in a handful of shots, there was no time to bring the rig up to date,” Danaher says. “But Genndy didn’t like the way the character looked out of the box. He didn’t like the default pose. He didn’t like the design.”

To stretch the rig beyond its initial purpose, Danaher used blend shapes, clusters, and SpWrinklefree, Maya® tools more typically found in the hands of modelers than animators. And this was true for other characters, as well.

“We had to sculpt and even rig characters as we went, on a shot-by-shot basis,” he says. “We had to design poses that worked exclusively from the shot camera view. It was tedious.”

For example: “On a lot of other shows, there is a pose library,” Danaher explains. “So when you get into your shot, you can bring a pose in for a given character from the library. That speeds up animation and keeps the character on model. We had rough pose libraries, too. But, they weren’t very useful. Genndy designed each pose for the shot camera.”

Danaher says that the starting pose Tartakovsky designed for the Gremlin wasn’t especially wacky, but the action that happened next was wild. “The Gremlin eats a scooter, a little kid’s toy that is about as big as the character,” he says. “It was physically impossible. The character had, in the span of four or five frames to do this explosive chomp and devour the scooter in one bite. I scaled up the head two or three times normal. His teeth got massively huge to emphasize the chomping motion. I scaled the scooter down to nothing to fit down the throat. There was a flurry of activity. During those two or three frames, the character looks way off model. It was a wildly exaggerated version of the character.”

Animation Mentor graduate Jason Malinowsky, who joined the crew in January, soon after production started, found the idea of animating in this way refreshing. “They had some scenes done and they were amazing,” he says. “The style of animation and the rendering — I had never done that style of animation. It was an exciting moment to get to try that.”

Jason Malinowsky

Like Danaher and other animators on the crew, Malinowsky pushed rigs beyond their design. “A few rigs — Dracula, Mavis, and Jonathan — were built more for pushing. But the other characters … we would tear them apart. If you couldn’t hit a pose exactly, you’d have to go frame by frame, pulling the geometry, sculpting, shaping the pose to what you needed. I was pushing past what I knew, which is always a good thing. I learned so much trying something new.”

All the shots weren’t exaggerated cartoon poses, though. Malinowsky describes a scene in which Mavis is sitting on a rooftop and is sad. For that shot, he worked from the dialog and he filmed himself acting out the scene to discover elements he thought would work.

“I used my cell phone,” he says. “I stood my iPhone up on my desk and emailed a movie to myself. I could look at it frame by frame on my workstation or export it as a JPEG sequence and throw it into a shot in Maya.”

For a scene with Frankenstein jumping off a diving board into a pool, though, it was back to cartoony animation. “You can’t film reference, so you just have to put yourself in that place,” he says. “What would I do if I wanted to go crazy and show off?”

Danaher was one of the first animators to create performances for Frankenstein; in fact, his early shots became a reference for the character. “Frank was interesting,” he says. “In a way, he has a very Genndy kind of design. He’s shaped like a big tombstone. A large hump with massive feet. His legs go straight down. He has a relatively small head buried in this tombstone silhouette. One thing I talked about with the animation director is that maybe his limbs become slightly disconnected. If he tries to gesture with a hand, it goes limp and he has to jam it back in the socket. Working with him was really fun.”

Both animators feel that this film taught them new ways of working. For Danaher, the film became a learning experience in ways he didn’t anticipate. “One thing I will definitely incorporate into my workflow is to do drawings of what I want my poses to look like,” he says. “Even if I have a clear image in my head of what I want, I’ve learned to go ahead and draw it, bring the drawing in to Maya or a window nearby, and compare the pose directly with the drawing. I always ended up pushing the pose farther than I thought I could.”

“There’s a tendency to pose and think I’ve got the parts roughly rotated and translated to match the pose,” he continues. “But when you look at the drawing, there’s so much more exaggeration you can get out of a character than you think you can. Having that loose, impressionistic, flat drawing reminds you how much farther you can go.”

Danaher says that he would draw now even if he were animating more natural human characters. “Something happens when you do a drawing,” he says. “It has more energy, life, action. A more organic quality. Even if I were animating a body double or a creature and wouldn’t be pushing, pulling, sculpting the character as far as on this film, I think it would help me inject extra life or expressiveness into the pose.”

For Malinowsky, the lesson learned was how to animate on the edge. A former engineer, Malinowsky’s choice to study with Animation Mentor changed his life. “Film was always an important part of my life,” he says. “Today, I am where I am because of Animation Mentor.” His first job was at Luma Pictures where he did simple, non-character animation. From there, he entered Rhythm & Hues Studios’ apprentice program and rolled onto Alvin and the Chipmunks. Now at Imageworks, he animated shots with Dracula and Jonathan as well as Mavis and Frankenstein.

“Generally, animating a character is kind of nerve wracking,” he says. “You don’t know if you’re supposed to push it. In this show, it was go for it. Do crazy stuff. I definitely learned how to push a character beyond what it’s designed for and that was something I hadn’t done before; taking a rig and pulling it to its max. Of course you have consequences, but it’s important. You want to try to do things people maybe haven’t done before. Being at that cusp, that edge, trying things people haven’t tried before … It’s exciting.”