Universal Pictures and MRC’s blockbuster Ted made us believe that the world’s most outrageous Teddy bear was real. For Ted 2, the new comedy from writer/director and voice star Seth MacFarlane, the bear is all grown up… in an amusingly immature kind of way.
The film’s eponymous hero is getting married, considering fatherhood – and even attempting to persuade the powers that be that he is legally a person. There is even a big musical number!
Bringing the bear to life seems highly complex, but the film’s motion capture was not the product of some multimillion-dollar facility. Instead it was produced live on set by Universal and MRC using Xsens' lightweight MVN inertial motion-capture suits.
CGSociety took the opportunity to sit down and chat with a true master of the art of mocap, Webster Colcord, who was integral in bringing Ted to life.
Where were you born and raised?
I was born and raised in Eugene, Oregon. My mother moved from Hong Kong when she was in her teens and my father’s family is from Michigan – Colcord is an old English name.
What sparked your interest in animation/SFX?
It sounds sort of cliché, but it was King Kong I think. When I was nine or so, the local TV station showed King Kong, Son of Kong, and Mighty Joe Young on subsequent nights. That and Seventh Voyage of Sinbad made a big impression. My dad was a big part of it too; he took me to the drive-in and would show super 8 movies for birthday parties. That was when you could check out films from the Public Library, and my mother was a Librarian. Over the years, I checked out every book on animation and SFX available.
I have to say as well that Ralph Bakshi’s Lord of the Rings – which was mostly rotoscope – really had a big impact on me. It was the weirdness of the motion, the hi-con processed footage of the Orcs, and the general psychedelic feel of it.
You started your professional journey as a stop motion animator working on animated gems like the Emmy Award winning Claymation Christmas Celebration and Michael Jackson's Moonwalker. Can you tell us a little about your professional beginnings? What was your big break? What was it like working at Vinton Studios?
My big break was with the Vinton studio in 1987. I was just out of High School and the studio happened to be doing its first big expansion to do a series of animated specials for CBS. All of us did “audition” sculptures and I was lucky to be hired despite my young age.
It was a very exciting time to be there. With today’s CG, we take dimensional animation for granted. But back then, clay was the only way to do that kind of expressive 3D cartoon; and their caricatures were stylistically like Mad magazine, which I had grown up on. I started fabricating characters but quickly moved into an animation role under the guidance of several mentors. The studio was loosely organized, not like a Hollywood shop, so everyone got to do a bit of everything. In that way it was a tremendous place to learn and grow.
I was on staff at the Vinton studio for three years, and then freelanced for them and a few other studios for four years, and then I had my own little studio for another four years doing commercials. During the time I had my own shop, I took a break from commercials to go to San Francisco to work on Henry Sellick’s James and the Giant Peach.
Your work so clearly and effortlessly shows core animation skills and principles, which breathe life into your incredibly diverse creations. Is this thanks to the countless hours spent in stop motion? Would you recommend other animators spending time doing this or other creative disciplines to hone their abilities?
Thanks for the compliment!
I don’t think that it’s particular to stop-motion. In fact, sometimes the physical challenges of stop-motion hamper your freedom in animation terms. I mean, you can do anything in stop-motion, but when you’re just starting out you may get caught up in some of the esoteric rigging challenges when you should just be concentrating on learning your basics. I wouldn't recommend it above other animation techniques for learning how to perform a character.Cel, CG, stop-motion – they’re all valuable and each has its own unique benefits as a learning tool. Stop-motion is uniquely good at helping one think about staging in three dimensions, and learning to block a scene like live action, because you’re kind of on a live action set in miniature.
After awhile, working with many different animators, directors, and animation supervisors, you pick up different formulas and techniques that expand your animation vocabulary. Richard Williams’ book and lectures illustrates this well. You end up with a pretty big toolbox and selection of tricks to choose from, and it’s the choices you pull from that toolbox that gives your work a certain style.
When, why and how did you break into the digital realm? What tools did you use back then?
Stop-motion as a sub-section of the industry, in the late 90s, was sort of where VFX is now; you had to travel to work on the interesting projects.
I was doing fine, with my own shop creating interstitials and commercials. But after Toy Story, there was this brief boom in recruiting for animators who were willing to learn the CG tools of the time. I had just become a father and wanted more stability, and frankly was sort of tired of being alone most of the time in my studio. There was a desire to learn something new and work with a bigger team.
After speaking with the various studios, I chose PDI/Dreamworks and moved to the Bay Area in 1997 to work on their first feature, Antz. It was quite a shock to go from puppets at my own shop, to working in a corporate environment on an SGI and learning Unix!
Almost 20 years later, you have now made us believe a Teddy Bear can talk, sing, dance and get outrageously stoned. Can you tell us about the software and hardware you used for a movie like Ted 2? Can you give us a rundown on the process you used?
We have a slightly older Boxx tower (purchased for Ted 1) that was our main mocap machine. At the time we needed a tower to house a PCI timecode card, which feeds a signal to MVN studio, but in the future we could do it all off a laptop with a PCI expansion chassis housing the card, in theory.There are three streams of data that we generate: video for performance reference, the actual mocap data, and a recording off the video card of the live mocap streaming into the character
in Motionbuilder. Timecode is what holds the three streams together, generated by three industry-standard “Lockit” boxes from Ambient Recording.
MVN studio records the mocap data with embedded timecode, and we use Motionbuilder to pass through the data, retargeting it onto the character model. After a typical session, editorial cuts the video elements in as picture-in-picture with timecode burns. I developed my own method for syncing up the data in Maya, based on timecode, and that has been pretty successful. Most of the time, I deliver the synced data as-is and leave any motion stitching to the animation team downstream.
We have a couple of backup laptops available for use as mocap recording machines in case we either get in extremely tight quarters on location (it happens!) or if our main system goes down. But most of the time, we use those machines for on-set previs and postvis. I prefer an all-in-one gamer-friendly solution, NVIDIA graphics cards, and I swear by the Asus “Republic of Gamers” line of laptops. In post-production, we have a little NAS setup and we used Syntheyes, Matchmover, Maya, and After Effects -pretty standard stuff.
For postvis, we contracted a talented former student of mine, Dan Babcock, to rig a Ted “Offset Rig” for the first movie. Though updated with new blendshapes, we're still using the same rig in our postvis on Ted 2. It’s a layered rig that works like the ones we used at Image Movers Digital (before the advent of Animation Layers in Maya) wherein the mocap lives underneath your animation controls. You’re just adjusting on top of the mocap, or “offsetting”. That’s probably the most important thing in the arsenal other than the Xsens MVN mocap suit itself.
What lessons have you learned from the Ted movies?
This is the most time that I've ever spent on a live action set, between the two movies. I've spent years of my life on set for miniature shoots, so I know my way around the gear, but there’s lots of unwritten etiquette on a live action set that you need to be aware of. And there are countless challenges involved with hauling the portable motion capture and previs gear around on location.
One of the things I've come away with is that you can’t possibly predict what is going to go wrong. You can try, but won’t be able to anticipate the weird problems that will come up. It’s just a leap of faith, and you have to keep your cool and be adaptable. That’s when experience really helps.
Between the first and second movies we spent some time to make the portable MVN unit more streamlined in the way it is applied to a person. We have a slightly different methodology than on the first film, on this one we keep the suit “live” and hard-wired to the computer all the time – and we don’t disconnect the individual sensors when we put it on. It’s always ready-to-go and we can get an actor in the suit and ready to mocap in about four minutes. It’s just an incremental improvement in the system that shaves a little time off of each mocap session.
How different is it to work on a project like Ted compared to other projects like X-Men: Days of Future Past?
I only did a little postvis work on X-Men, for The Third Floor. It was on the Fox lot and it’s always a thrill to walk through a studio lot because of the history.
The Marvel movies are very big shows, obviously, and at every stage of production there is a version of the VFX shots in some form or another – be it animatic, previs, or postvis. But for Ted we don’t do as much visualizing because Seth sees the bear in his head; he knows what he wants. The previs is more about camera and the flow of the scene, and the postvis is to help the preview audiences.
Also, there are challenges unique to working in comedy. For instance, any time there’s a cell phone screen replacement or a photograph added in post - which is usually an image that is key to a joke – we found that if there’s any wobble in the element it distracts the viewer and kills the laugh. We end up doing really precise corner pins for those types of shots.
One of the first things I do when I go to a previs house to work on a show is test drive their various ready-made particle rigs. Every shop has cool little tools for debris, explosions, and camera shake. But I don’t think I've done one single explosion for anything on Ted. That's a big difference from your typical VFX-heavy action sequence previs!
What is the secret of a successful mocap session?
One of the funny things about mocap is that it becomes very apparent when someone is not a skilled physical actor. It’s important to get someone who has that particular set of skills, who acts through their entire body. Then in the session, it’s easy to forget the context of the acting because you’re usually not in the actual environment that the character will be integrated into. Actors are embarrassed to be in the suits, which sometimes look and feel kind of silly. It’s important that everyone has the context of the scene in their imagination, and that you give the actors something to play off of if possible – set pieces, props, or stand-in actors.
The portability of the Xsens MVN system in many ways solves the context problem. Without having to set-up an entire volume, and without having to worry about marker occlusion issues, the inertial system enables easy motion capture on set or on location. In that way we get genuine interaction between the actors. With inertial mocap we’re not confined to a stage or an office, nor are we restricted by the expense of setting up a full optical mocap volume on location.
What movie from the past do you wish you had worked on and why?
There are actually a couple of very specific projects, which I tried to get on, but it just didn't work out. One of them was Army of Darkness. The main animator (stop-motion) on that was the late Peter Kleinow, who I became friends with years later.
And back in the 90s, just as I was starting up at PDI/Dreamworks, I was corresponding quite a bit with the late Fred Stuhr, who had directed a really great string of music videos including the famous “Sober” for the band Tool. We had done some work for the same ad agency, for Converse, so we were talking quite a bit and he wanted me to come to L.A. to work with him. I just couldn't do it at the time and a few months later he was killed in a car accident.
But more to the spirit of the question, I would say that it would have been a dream to work on any of the Godzilla movies at Toho, or Ultraman at Tsuburaya productions.
Do you have a motto?
I’ve been thinking a lot lately about “Gall’s Law” which states that all complex systems that work evolved from simpler systems that worked and that you can’t design a complex system from scratch and expect it to work. It reminds me of something that I learned the hard way early in my career, and that I applied to subsequent projects - simply that your ratio of skill to technology has got to favor the skill side. You can’t rely on tech to get the job done, but a little bit of technology in the right hands can give you a huge advantage. It’s not exactly a motto, but I always keep it in mind when starting a project.
What will you be working on after Ted 2?
I'm going to be Animation Supervising at a Bay Area VFX house; Atomic Fiction.
Thanks Webster for being so generous with your time! We look forward to seeing more of your work in the future.
The CGSociety is the most respected and accessible global organization for creative digital artists. The CGS supports artists at every level by offering a range of services to connect, inform, educate and promote digital artists worldwide. More about us on TheArtSociety.com