One of the hot topics at AnimfxNZ was 3-D stereo and digital cinema. Dave Scammell, the President of SohoNet, ran a panel on the issue with Steve Schklair from 3ality Digital, Patrick von Sychowski from Adlabs/Reliance India, and Habib Zargarpour from Electronic Arts Los Angeles.

It’s obvious that the move towards 3-D is slow and painful. Patrick has been in digital cinema for 10 years, and says 3-D is one of those developments that’s always five years away. It’s a bit of a chicken and egg problem — stereoscopic 3-D in cinema is a function of the theaters’ conversion from 35 mm cinema to digital cinema. He likens the changeover process to the process of switching from left-hand drive to right-hand drive.

Steve Schklair

Steve is more optimistic. At one point, when asked if there will come a day when everything is shot in 3-D, he said, “I’d be branded a lunatic if I said everything is going to 3-D, so I’ll just say Jeffrey Katzenberg says everything’s going to 3-D, and he’s way too powerful and important to be a lunatic.”

He points out that when it’s as easy to shoot in 3-D and when it costs the same as 2-D, everything will be shot in 3-D. After all why not? Thanks to the new technology, you can always shoot something in 3-D and play it in 2-D. And, as the audience keeps getting more used to it, 2-D will grow to seem incredibly flat.

And, in an encouraging development for theaters looking for ROI from their screen conversions, there’s a heap of content coming down the pipe. Steve says there are 16 Hollywood features slated for release next year and his company is experimenting with live sports broadcasts in 3-D, to be shown in theaters.

Patrick von Sychowski

Patrick concurs; Disney and Pixar are doing all their animation in 3-D, and some people are going back and re-rendering 2-D films in stereoscopic, like The Nightmare Before Christmas. For some purists, this is horrible, but there is an upside: George Lucas has found yet another way to milk the Star Wars cow!

The game industry, represented on the panel by Habib, has a totally different perspective on the matter. The nature of games is that there is no camera – you could also say that all cameras exist simultaneously. Because a game engine has to render the total car and the total street and the total world, 3-D in gaming is just a question of flicking a switch.

In fact, NVIDIA has a 3-D enabled card that will allow you to see EA games in 3-D right now today, as long as you have a 3-D-enabled monitor. The problem is that all the cheats they do to make the game look great in 2-D fall away in 3-D and everything looks really simple.

So it is kind of ironic that the studios are pushing for 3-D to get people into the cinema, when the technology’s really possible at home. In fact, as Steve points out, there are now polarized televisions that allow you to take the same glasses you saw Journey to the Center of the Earth with and watch 3-D content at home, and the pictures are amazing.

Back to Habib and the games industry, for whom the biggest 3-D issue is the format and how it’s going to be supported, since they obviously aren’t going to ship a TV with every game.

Habib Zargarpour

That problem is a far cry from Habib’s early days. He started in digital effects for film in a company called Mr. Film, working on IMAX movies in 3-D in 1991. There was no industry standard because there was no industry. They would do match moves to make the digital stuff match what they shot on set, but they were doing the match moves on a 19 inch monitor to be shown on an IMAX screen. As you can imagine, they were frequently dismayed by the results.

Frankly, it’s amazing that analog 3-D worked at all. As Steve says, you couldn’t shoot it with a regular budget, cameras weighed a ton, you couldn’t change lenses easily, and you’d have dailies using two projectors that would be jittering independently so you were guaranteed to get a headache. That’s why they kept shoving something out of the screen at you every few minutes, because it was the reward for putting up with the headache.

Those problems have been mostly solved with digital. It’s okay for the projectors to be jittery in 2-D because at least your eyes are moving together, but in 3-D your projection needs to be stable and steady. If your eyes are forced to move in unnatural ways, you get a headache from using muscles you don’t normally use. Colors need to match. Vertically, the pictures need to match. Geometry needs to match.

The other issue in 3-D is the edits – your eye has to jump from focusing on something that’s 10 feet in front of the screen to something that’s 30 feet behind the screen. So your eyes are doing calisthenics in your head. No wonder you get a headache.

Steve is also bemused by the almost fanatical approach some people take to the question of whether to shoot converged or non-converged. Some people think you should converge at the point of focus, but you don’t always want that. The example he gives is a big scene with an actor far away in front of a tree. You want him to feel far away, so for that scene you might converge right up close to the camera.

Bottom line? These things take time and commitment from a variety of partners (think of the shift from DVD to Blu-Ray), and the credit crunch hasn’t made financing digital screen conversions any easier. Nonetheless, according to our expert panel, progress towards digital 3-D is inevitable.

Better dust off those glasses.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

This site is a member of Animation blogspot, part of the Animation World Network

Kaila Colbin reports from AnimfxNZ.

Tim Willits is the Creative Director at id Software, where he’s been since 1995. One of the interesting things in the gaming industry is how so many of their issues predate identical issues in other arenas.

One example is the current UGC / open source / community ownership explosion online, which is having a dramatic impact on music, film, and television. id has always allowed people to modify their games, which has the added benefit of allowing them to identify skilled and passionate talent. All of the designers they’ve got in software now are from the community; young people out of college with no experience can download games, make changes, add levels, create new art, and use that for their resumes to get into the industry.

id Software has made some incredible contributions to gaming and the broader world of technology. Their game Doom legitimized the shareware / demo model, where a portion of a game gets distributed for free and then people have to pay to get more involved. Quake was the first 3D action game and the first server model game: the precursor to online social communities.

Tim’s focus now is on video game storytelling, and the story in a video game, if done correctly, can be as great as any novel or movie.

Take Doom, for example. Yeah, there’s a ’story’ about demons, saving the Universe, blah blah blah. But if somebody with a bit of skill wrote down the experience — the adrenaline, the wonderment of discovering new worlds, the fear, the thrill of the battle — you’d have a truly compelling story.

The fact is the narrative is only part of the story. It’s all the stuff in the middle that makes the difference: the immersion, the feelings, the memories.

So experiences are at the cornerstone of id Software game design. They concentrate on what the experience is like for the gamer, using lighting, music, effects, design and immersion to show a story rather than tell it.

The narrative is then wrapped around those experiences. They look for a solid, straightforward plot that people can ‘get’ immediately, which is why Good vs. Evil works so well. They look for a dynamic story, which allows for the inevitable evolutions, adaptations, and budget fluctuations that arise.

Most importantly, they don’t commit everything to the narrative. After all, there’s nothing worse then a boring game with a great story.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

This site is a member of Animation blogspot, part of the Animation World Network

Kaila Colbin reports from AnimfxNZ.

Joseph Olin is the President of the Academy of Interactive Arts & Sciences. Just like that Other Academy, on which it’s modeled, the AIAS exists to recognize and reward outstanding accomplishments in video games. The Academy is 12 years old, has 15,000 members, and acknowledges the best of the gaming best every year through the Interactive Achievement Awards.

Last year, the Game of the Year award went to Call of Duty 4: Modern Warfare, which also received awards for Console Game of the Year, Action Game of the Year, and Outstanding Achievement in Online Game Play. The public obviously agrees: at any given moment, approximately two million people are online playing CoD4 through Xbox Live.

And, just like that Other Academy, the AIAs does more than just awards. They’re affiliated with the International Game Developers Association. They run Into the Pixel, an exhibition, exploration, and celebration of the art of the video game.

Joseph is feeling pretty positive about the games industry these days. Business is booming. The ‘retail games’ industry is running at about $19 billion annually, nearly twice as much as the $10 billion theater industry.

The women’s segment is growing, thanks in no small part to the Nintendo Wii. Interestingly, the kids’ segment is also pretty significant. After all, as Joseph says, we need the kids’ games to legitimize needing a new rig for our homes.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

This site is a member of Animation blogspot, part of the Animation World Network

Kaila Colbin reports from AnimfxNZ.

As Chief Visual Officer for EA Black Box, Henry LaBounta works with art directors and studios to define their visual goals in preproduction and help them execute.

Game Art Direction has three main focus areas – the look of the game, the characters and animation, and the graphic design (menus, user interface, and fonts) – and most of the unsolved problems are with the characters and animation.

One of the interesting things about this conference, and something I discussed with Henry after his presentation, is that video games have come so far in the past 14 years that we can actually have a conference like this one, where the issues facing movie animators and the issues facing game developers can overlap. In fact, when Henry first got into gaming from his TV, movie visual effects and feature animation background, he was surprised at how many similarities there were.

So if that much has changed in the past decade and a half, what can we expect for the next era of gaming? That’s what keeps him going: the pace of change in this industry is relentless. Just when we think we know what we’re doing, a new console comes out.

When it comes to art direction in games, there are four Golden Objectives:
•    Create a distinctive look and style you can own that is culturally relevant for that game
•    Understand what you’re trying to communicate visually and what’s visually most important
•    Create an immersive experience in which all visual aspects work well together; don’t break suspension of disbelief
•    Provide visuals that support and enhance game play

So when he started looking at driving games, he first had to understand how his audience looks at cars. They see cars in car commercials, and car commercials have a very distinctive look, so Henry looked at those commercials for reference.

You’ve got to focus on what’s important to the audience. Some developers get hung up on not having enough texture resolution, when basic things like the look, style and cultural relevance are off. Basic exposure control with balanced black and white levels are also a common problem in games. The eye is attracted to bright areas, especially during game play, but some games miss the mark on this fundamental issue– cars blend in with backgrounds, and the focus of attention is off or unclear. Asphalt isn’t black. For these games, the problem is not texture resolution!

Of course, games are far more difficult than film. They’ve got specific challenges that don’t come up in movies, such as open worlds where the camera can look anywhere. In addition, you’ve got to create a dynamic look for the game, otherwise the eyes get used to it and it gets boring visually.

But the Holy Grail of unsolved game problems is with dynamic characters: creating a believable character that performs and moves the way you’d expect it to. Although they’ve gotten pretty good at rendering static characters, they’re still working on:
•    Believable motion
•    Foot planting
•    Sense of weight and momentum
•    Responsiveness
•    Non-cyclic motion
•    Player contact
•    Player individuality, unique motions
•    Intelligent motion

In the past there was always a tradeoff between responsiveness and natural motion. The focus now is on having both.

So what do they work on? They want the guy catching the football to look towards the football, even though he’s running in the other direction. They want the guy playing basketball to see the telltale weight shift that presages a move to the right. They want the soccer player to start to slump as he gets more tired.

All of these things have something in common: they’re critical to the experience of playing the game. Get them wrong, and they distract from the experience and generate frustration. Get them right, and the game gets a lot more fun to play.

In short, they want a better experience.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

Check out this video interview with Henry LaBounta.

This site is a member of Animation blogspot, part of the Animation World Network

Carolyn Soper is a Vice President of Disney Animation Studios, and has been working on the upcoming Disney feature Bolt. She walked AnimfxNZ attendees through the process of building a hamster named Rhino.

Bolt is the story of a dog (played by John Travolta) who’s the star of an action show. He believes anything he can do in the show, he can do in real life. He thinks he can knock down stuntmen and jump across 16 lanes of traffic. He thinks he’s got laser vision. Unfortunately, he doesn’t realize it’s all fake. Bolt is owned by Penny (Miley Cyrus), whom he loves deeply and would do anything to protect.

The fun starts when Bolt gets shipped to New York, still believing he’s got powers, and now believing Penny was kidnapped by the Man With Green Eye. So he does the only logical thing: he abducts a cat. Because all cats are evil and therefore agents of the Man With Green Eye, this one, Mittens (Susie Essman) will surely know where Penny is.

Partway along their quest, they find Rhino the hamster, Bolt’s ‘biggest fan.’ So now there’s this poor sane cat tied to a delusional dog and hounded by an equally delusional hamster.

Bolt's biggest fan, Rhino, proved to be a major animating challenge. All Bolt images © Disney Enterprises.

Incidentally, Rhino’s voice comes from an unlikely source: Mark Walton, who had done the scratch dialogue. Although normally that original scratch voice gets replaced by a big-name actor, Mark was so compelling as the hamster that the animators petitioned for him to stay on, and this counts as his big-screen debut.

Animating the hamster presented a unique and complex set of challenges. First, the hamster has quadruped locomotion – he moves around on four legs. But he acts on two legs, so he’s got to be able to shift seamlessly from one to the other. He’s also got a fat body covered in fur. And finally, he exists in a hamster ball, the rolling movement of which has to correspond to the frantic scrabbling of Rhino himself.

Software Engineer Evan Goldberg solved the Hamster Ball Problem, but those other issues were significant. The computer skeleton of the creature’s body, known as the ‘rig’, was built to favor the quadruped, but when Rhino stood up he looked all wrong. So every time they brought him to a standing position, they had to make 31 adjustments to the model: the legs, the belly, the neck, the head, etc.

Even after all that tweaking, there were still some weird artifacts left over. If they tried to rotate the head 90 degrees the neck got destroyed. Each corrective tool they applied slowed the rig, and these difficulties led to conservative animation – it became easier to limit the hamster’s movements and make him less interesting than it was to fix all the problems. Sometimes the animators would go ahead with the movements they wanted, and then pass unappealing shapes downstream for others to clean up.

Conquering the Bolt model was the first task for the Disney team.

To address all these issues, the team created ‘paloozas’, starting with Bolt-a-palooza. Byron Howard challenged all the animators to create single-frame poses of Bolt that were the most appealing that they could be, and focused on what was attractive about these poses. This process led to the 2D, single-frame idealized version of Bolt; the rig now needed to be adjusted to match that.

They first started Bolt-a-palooza mid-December, and by the beginning of January they had a new Bolt rig. So it was natural for them to have confidence that they could do the same for Mittens, Penny and Rhino.

But when they were building Rhino’s palooza poses, the commensurate rig had collapses and penetrations everywhere. So they were faced with a decision: keep adding transformers and corrections to the rig, or make a drastic change and completely overhaul it.

The radical change they went for required a complete change in workflow. Traditionally, animation involved a linear, three-step process: first the character would be modeled, then the approved model would go to rigging, and then the approved rig would go to animation. If it didn’t work in animation, it would go back and forth with the rigging department until the problem was fixed.

So the linear walls were broken down, and instead a Character Team was created, where all three specialties could work together to create a character that worked.

When it came to creating the new rig, they had a few restrictions. They couldn’t touch the existing quad rig, because it worked. They also couldn’t build a totally separate rig (which would have required the director to send the hamster behind a bush every time it needed to go from four legs to two and vice versa).

Rhino's face was the first part of the rig that was mastered.

The facial part of the rig was already built, and was working well. Fortunately, they had built it using a bolt-on facial system, which made it easy to rip off the body and put it on another skeleton.

So that’s exactly what they ended up with: one skeleton, two skins, and one face, with a few blend shapes to transition from one to the other.

The next challenge was in the research and development of a new system for corrective shapes. If you bend your arm in real life, the flesh bends and squishes in certain ways. When you create an arm on a computer, though, and then instruct it to bend, it doesn’t know how to fill in that flesh.

The solution to this problem is in something called a Post Space Deformer, or PSD. What it does is take the desired, properly squished end shape, and then reverse calculate what the original shape would have to look like to get there. This new shape – called a ‘delta shape’ – gets blended in as the original shape is transitioning.

PSDs were a challenge on all the characters in Bolt.

The issue with PSDs is that every time you have a new end position, you have to build a new delta shape. And when it came to Rhino, every movement created a shape that needed correcting. He was fat and fluffy. He was potato-shaped. He had short arms and legs. All these problem areas overlapped, so if you solved one problem, like the neck, you might create more problems with another area.

As if things weren’t complicated enough, some Disney animators use a broken rig, which means they can position each piece of the body independently.

So the folks at Disney created a totally new solution: instead of creating the PSDs by joint rotation and final position, they created them based on the relationships between the body parts. So the PSD created by dropping Rhino’s right ear towards his shoulder was the same as the one created by him lifting his arm.

Using Lorez mesh-based distance locators, they created distance values for the PSD drivers that would work regardless of whether the animator had rotated, translated or scaled the character. Clay Kaytis, the Supervising Animator of Rhino, was also given control over how much he wanted to use the PSDs.

As an added bonus from all this development, poor Rhino was given something most of us would rather avoid: granny arms. Based on the trailer we saw, though, you won’t be focused on the PSDs or the cutely swinging granny arms. You’ll just be enjoying a great story.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

This site is a member of Animation blogspot, part of the Animation World Network

AnimfxNZ 2008 began with a song.

A contingent of Maori gathered to perform a ‘powhiri’, or New Zealand traditional welcome. The foreign visitors and speakers lined up so the kaumatua could determine if they were friends or foes. Thankfully, there were no enemies among them, and the conference was able to proceed.

Aimee McCammon, the General Manager of Park Road Post Production, explored some of the challenges faced by PRPP and post-production houses in general. It’s a commodity business. The pace of change is phenomenal. High entry costs for the really cool new equipment coexist with the plummeting price of downstream technology, generating a tricky balance between early adoption and ROI.

Consider the Quantel Pablo they bought in late 2006. They started the kit with a chargeout rate of $1600/hr. 18 months later the rate was down to $800/hr, and under continuing and relentless price pressure from competitors. Nonetheless, Aimee is proud of and will continue the company’s policy of being early adopters of technology.

Cheaper technology has a couple of other effects. Anybody can make a film these days with a MacBook and a copy of FinalCut Pro, which means there are a lot more crap films — and crap filmmakers — out there. Aimee, however, is confident that out of a thousand crummy ones there will be one or two that are exceptional. A million more rocks coming through the pipe will reveal exciting product and talent that might otherwise never have seen the light of day.

The term ‘world-class’ gets bandied about too easily, in my view, but Park Road Post Production clearly deserves to be in that category. They’ve never failed a QC in New Zealand or abroad. Their shelves are lined with awards, including an Innovation Award for a Red Post Workflow, a Qantas Media Award and two Oscars. Their vision is to be among the top five post-production houses in the world by reputation, and Aimee in particular would like to see the shop win an Oscar for a film by somebody other than Peter Jackson — James Cameron, are you listening?

On the technical side, they’ve got two DI suites with a total of 32 terabytes, meaning they can work on two films simultaneously if need be, and they’re rocketing into 3D & D.Cinema. Fancy gadgetry notwithstanding, Aimee closed her presentation with a firm admonition: you can’t win in this race with technology alone. Without passionate and driven people, you’re nowhere.

Kaila Colbin, the founder of Missing Link, is a frequent contributor to a variety of magazines.

This site is a member of Animation blogspot, part of the Animation World Network