Check out the third part of video artist Marco Brambilla’s trilogy that makes art from film

Exhibited at the Nicole Klagsburn gallery in New York, Creation (Megaplex) was inspired by the disposability of film and images, featuring a swirling helix of hundreds of movie clips, some familiar.

To bring it to life, Brambilla enlisted creative agency Ntropic. “I was introduced to Marco when we opened our New York office,” says Ntropic creative director Nate Robinson. “We discussed the possibility of working together, and hit it off after a couple of meetings.”

The Ntropic team gathered all the materials, rotoscoping the characters before using Flame and Maya to pre-viz, design, animate and composite the film.

“The stereo capabilities of Flame were instrumental in planning and execution,” says Robinson. “Flame enabled us to work in a 3D design space by incorporating 2D tools with 3D technology. Flame and Maya’s interoperability allowed us to move cameras and geometry seamlessly between them.”

What should be in the shortlist for the 3D World Hall of Fame Award in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Discover 15 inspiring examples of 3D art at our sister site, Creative Bloq.

Director Kohta Morie reveals the story behind his emotional music video for Express by Hidetake Takayama

When I listened to the songs of Hidetake Takayama on MySpace in 2009, I became a big fan. I wanted to see him, but he wasn’t doing any live shows at that time. I chased him down for a year, but never got any information about him except his name and music, then I heard that he was going to do a gig at a CD shop to promote his first album. I got really excited, and headed off to the venue. I had many questions, but I still didn’t know what he looked like or how old he was. Arriving at the shop, there he was, wearing an Adidas jersey, with short hair and a friendly smile. I got who he was at first glance. Once he touched the piano, the mood changed dramatically. The music took me away from the time, place and everything. I couldn’t even move.

A year passed. I had been working on many other CG projects, one of which won an award at a Japanese CG festival. Then, I received a message from Takayama that would change my future: “Do you want to make a music video with me?” Of course, I answered, “Yes!” in a flash.

This was the beginning of my struggle to make Express. I had been working as an animator for years, but never as a director. Everything was new to me. Not only as a director, but also, I was still working as a chief animator at Transistor Studio in Tokyo. My schedule was packed with other works, but my desire to work with Takayama was overwhelming.

What we did right

1. We made it from the very beginning

We had already made four other fully CG music videos, but every time there was a director. Even though we made every single shot, the work was all attributed to the directors in the end as far as the audience was concerned. Like VFX studios in LA, the studios that have their original contacts can survive, but the rest can’t. We could have stayed in this position, being used by other companies, even though we had enough skill to become independent.

When the project began, I was worried about whether we’d made the right decision to finish everything else in the studio, because there was no studio that had accomplished what we were doing before. At the same time, I really wanted to challenge the notion that there was a border between bigger studios and smaller studios, and to possibly get out of the situation where we felt we were being used. Everything is very easy if you just use the technology, but it’s harder to be more creative. We actually made it, and we showed the potential to be creative if you really have the desire, and you keep working hard.

The water men were animated in Maya and exported to Houdini to finish the simulation. The data was then imported back to Maya to render

2. The crew worked well together

On each of the music videos we’d worked on before, something would challenge us. For example, to finish one of the music videos, we stayed at the office and couldn’t go back home for about two months. It was very tough work for everyone, but we made stronger relationships of mutual trust through the circumstance. When I got the offer from Takayama, we had already built up the skills needed to make music videos.

Three of the major contributors to the project were Junichi Akimoto, Takahiro  Shibano and Hiroyuki Ito, who had been working together for the previous two years.

Akimoto specialises in Houdini. He started using it when he was 18 years old, and now writes about it every month in the Japanese CG magazine CG World. He did everything in Houdini that’s hard to accomplish in Maya such as the liquid orchestra (called ‘water men’) and made the processes smoother.

Houdini made dust, fine particles and smoke from the data

Shibano did most of the lighting, rendering and compositing. These jobs are better for methodical people who can pay careful attention all the time. Since the director tends to correct the details of the colour, light and shadow, Shibano’s job really needs to meet the director’s demands.

Ito worked on many different things such as building a pipeline tool to connect different software with Alembic, scheduling management of all the shots (of which there were 164), keeping track of progress, and being a great listener whenever I was struggling with something.

These three members of the team have great skills for certain, but more important than that, I’d already been working with them for two years, so I was easily able to maximise their skills. The other crew members also helped with secondary animation, error checking and compositing. When the deadline was looming, I was upset that the project would soon be over, and they asked me if there was something they could do to help me. That made me really happy. It was interesting to see their true characters come through this project in the end.

3. We made friends with the client

When I came up with the story for the video that would eventually be used (there had been other ideas), I hurried to see Takayama to show him the image board. My mind was full of worries about whether he would like it. Contrary to my fears, he had a boyish grin when I showed it to him. His eyes said, “Very good.” We were connected to each other at the time. Even though I didn’t give him any details of the story, he understood what I was proposing, and I knew what he wanted at the same time. Now that the project is over, we are still hanging out and drinking with him and his band members. It’s an honour to get such a great friend through this project, and this must be the most important evaluation.

We also made friends with Mr JFK, a capable producer that works in the same office as Takayama. He’s responsible for 2.5 million CD sales (selling over 10,000 CDs is said to be very good in Japan), and he gave our entire studio access to Takayama’s concerts any time, anywhere, forever.

The flying lights of snake gourds – one of the film’s many particle shots produced in Houdini

What we did wrong

1. We had to animate by hand

We used HIK in Maya 2012 for the character setup because it was supposed to be able to use the motion capture later. I visited Carnegie Mellon University’s website to check whether there was a free database of motion capture to use for a simple walk animation, but I found nothing. For that reason, all the animation in the video is animated with keys.

Also, I tried to use the hair system in Maya 2012 to automatically move the boy’s hair, but the result wasn’t what I wanted. So all the hair was animated by key animation too.

2. We struggled to synchronise the video

I synchronised the video to the music perfectly, down to the keys of the piano synchronising to the piano sound. I tried to make the synchronisation automatic using the CHOP tool in Houdini to move the keys to the sounds automatically, but I failed. The reason was the music video was 15fps, so it dropped frames, which spoiled the final look.

3. I couldn’t see my family

Abroad, in general, the first priority is family, but in Japan, work is the most important priority. I got married during the project, and I couldn’t see my wife for three weeks. To make it worse, we had the wedding one week before the deadline. I totally depended on my wife for the preparation, and met her just before heading to the wedding hall. She supported me a lot because she totally understood that this was my first and most valuable work as a director.

Each light worked as key light, shadow light and rim light so the characters could have complex shadows

Lessons learned

It was interesting to see unexpected things happen, and discover how much I could challenge myself. I think there are two kinds of people to accomplish one project. The one is the creator, the other is the audience. We can’t finish and evaluate the project without these two. I didn’t want to make a meaningless video.

Sadly, there are tons of videos that are by the creators for the creators, but this makes no sense to the audience. Pictures and images are tools to move people and let them think. They even have the possibility to change their lives. Think about when you play catch. What would happen if the players didn’t try to catch the ball? We used high technology for this project, but the audience still get the details. That’s why I focused on the story only.

On the other hand, the fundamental skills supported this project a lot too. I’ve been learning character animation, rigging, camera work and layout for six years. This combined knowledge made it possible to accomplish this one big project.

VITAL STATISTICS
Project: Express feat. Silla (múm)
Studio: Transistor Studio
Format: Music video
Client: GOON TRAX/ Media Factory, Inc.
Time taken: Six months
Team size: Two full-time, one support, 11 at peak
Software used: Maya, Houdini, ZBrush, Nuke, After Effects, Photoshop
Release date: 30 October 2012

Kohta Morie is a 3D animator and director, and the author of Animation Style +. He was born in Matsue, Japan, and graduated from the Tokyo Multimedia College. He loves his wife much more than CG

What should be in the shortlist for CG Animation Short of the Year in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Discover the best 3D movies of 2013 at our sister site, Creative Bloq.

Super-powered youngsters grow up to work for Virgin Atlantic in this spot from The Moving Picture Company

‘Flying in the face of the ordinary’ is the tagline for Virgin Atlantic’s recent tongue-in-cheek spot, created by The Moving Picture Company’s VFX department.

A girl with psychic powers, a boy who can catch a fish with his bare hands are two of the five superhumans who appear in the 90-second ad – and, naturally, as adults they become Virgin cabin crew, designers, ground staff and pilots. The sixth spot MPC have worked on for Virgin got started when the team ad agency RKCR/Y&R approached MPC with the brief.

“MPC was responsible for all the VFX work on the spot, from the CG plane to the hologram chair,” says 3D creative director Jim Radford. “The project involved a wide range of work, from multi-pass compositing, de-rigging, digital matte painting, hologram graphics and miniatures.”

To achieve such a range of intricate effects, the team used a combination of packages from multiple vendors, including Maya, Cinema 4D, After Effects and Photoshop.

“After Effects was vital to production on this project,” says Radford. “We carried out the look development for the hologram chair in AE, including using the Trapcode plug-in suite from Red Giant. “We built and rigged the chair in Maya, and used Alembic to pass it to Cinema 4D and then onto AE. The immediacy of After Effects really helped us create a range of looks quickly, which was important given our short schedule.”

What should be in the shortlist for CG Commercial Campaign of the Year in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Create a superhero! Learn how to give a 3D character super-powers at Creative Bloq.

Aardman combined 2D and 3D seamlessly in this promo for folk trio The Staves

Bristol-based animation studio Aardman recently collaborated with Atlantic and folk trio The Staves to create this music promo for the band’s ‘Winter Trees’ single.

“The brief came directly from the band and Atlantic Records,” says co-director Saul Freedman. “It was quite loose and only had a few references, so we were free to come up with a visual concept.”

With an opportunity to create something completely different to Aardman’s trademark styles, the team got to work, creating everything from scripting to the final delivery.

Mixing up techniques, the video used 2D and 3D software to achieve the desired look. “The entire pipeline from Flash to Maya was very technical,” explains lead technical director Nathan Guttridge. “We were able to create custom Python tools to bring Illustrator paths into Maya, then create geometry on the fly from these. Having the ability to automate large chunks of the character-build process was essential to completing on time.

“We also used Flash to do most of the animation. A useful process was having the ability to take the Flash files straight into After Effects, where we could edit them into loops and assemble shots quickly.”

What should be in the shortlist for CG Animation Short of the Year in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Discover what’s next for Augmented Reality at our sister site, Creative Bloq.

Disney used innovative CG techniques for its Oscar-winning short Paperman, animation supervisor Patrick Osborne tells Mark Ramshaw

If Paperman’s appearance in cinemas alongside Wreck-It Ralph mirrors an approach regularly used by Pixar, it’s notable that director John Kahrs originally began developing the idea while over at Disney’s sister studio. But it was only after returning to the Disney fold as a supervising animator for Tangled that Kahrs pitched the idea to John Lasseter for development at Disney instead.

“John had actually planned to use global illumination to make the film back when he was at Pixar because it was a rather fresh, new take on CG imagery,” reveals animation supervisor Patrick Osborne. “But then, by the time it got the green light, it wasn’t feeling so fresh to us.”

Instead, Osborne started experimenting with a new style of animation, making art with the CG medium in its native, realtime form, using only in-Maya hardware shaders. “The computer is a powerful tool for simulating physical phenomena like light, cloth and hair, but you sacrifice some control of animated design and shape by applying those effects after the animation is finished,” he explains.

“We never wanted to pursue toon shaders, because they’re generally ugly and unwieldy,” says Osborne

Initial tests used only flat-shaded colour, and lacked the line work that Kahrs was looking for. The next step came when VFX supervisor Kyle Odermatt suggested checking out what Disney software engineer Brian Whited had created down in R&D. “He had been tasked a few years earlier to streamline the tedious tight in-betweening process in hand-drawn 2D animation, and part of his project was Meander, a vector drawing tool built to stay fast and keep that connection between artist and work as tight as possible,” says Osborne. “He had even built in brushes to mimic the favourite brushes of Shiyoon Kim and Jin Kim, two of our fantastic character designers, so we knew it could handle the beautiful line work we were going to need. It fit in perfectly with what we were trying to do. I’d say that if Brian hadn’t already been there working on Meander, then Paperman would not have been possible.”

With Whited on board, several more months were spent adding further animation and motion-tracking capabilities to the software, and developing a completed test shot to present to Lasseter. Osborne says this one shot took animators Joel Reid and Hyun Min Lee nine months to bring to a level they were all happy with. “Lasseter didn’t know we were doing anything different until we showed the finished test,” he says. “We loved the look, and stood behind it, but we were nervous. It’s one of those looks where the description doesn’t do it justice, so there was some scepticism. That’s why we chose not to show it until it was complete.”

“The rigs were much simpler than the rigs in Tangled or Wreck-It Ralph, with far fewer pose space deformers,” says Osborne. “They did have to be very well made, however. It was important that nothing strange was going on in the rig or we would see it in the line later.”

As well as adding the ability to track motion vectors both forwards and backwards through time, and adding additional layering capability, Meander also crucially needed to allow the artists to draw outside the lines, to track brush textures, and also to cope with various pipeline plug-ins. “Brian was a champ at meeting our requests without rendering the product unstable at all,” says Osborne. “The control of line it gives is a testament to the polished level of the software. The artist doesn’t have to wait for very much, and so artistic flow is maintained, allowing detailed work to not feel too tedious.”

The CG curse

Speed is one of the virtues of the tool. Of the 15 months spent developing Paperman, the 2D and 3D animators were only required for around three. Osborne says that they were able to generate frames in the time it took to press the paste button – sometimes even faster, because the tool enables artists to paste several frames at one time. “As with most animation, the beginning is slow as people ‘find the characters’ and learn how to draw them in an appealing way,” says Osborne, “but we found that once an artist got the hang of it, they became rather fast at the technique. I’d say our process on Paperman was comparable, if not slightly lighter, than on a¬typical CG film.”

He says that the most difficult shots were those with dynamic cameras, where the challenge lay in creating backgrounds that looked sufficiently ‘hand-made’ to mesh with the characters. “When the camera starts to move, it shows all the CG tricks you’re using a little too much. We paid careful attention to the camera work to remove to take the ‘CG curse’ off it when possible, and also calculated motion vectors for the backgrounds, so that we could also paint on them using Meander.”

“An artist would open the Meander file, with the layers set up for them exactly as the comp artists wanted to receive them back,” says Osborne

Meander’s data requirements were also sometimes an issue, admits Osborne, particularly when dealing with long shots. Some had 50 layers at 2K each, with each layer featuring a motion vector file, a paper texture sequence of images, and the image sequence itself. “Combine that with painted layers on a shot, and that’s 600-plus frames – quite the memory and GPU hog. We had to borrow the heftiest machines in the building to get those shots done.”

Osborne says it’s been an incredible two years spent working on the short. “We didn’t expect to be able to get it done at all when we started, so to see it finished was very satisfying.”

The finished short has had an exceptional impact. An immediate critical darling, it received a further boost in March when it provided Disney with its first animated short Oscar since 1969. Internally, meanwhile, Paperman is no longer a oneoff – with Disney now committed to an ongoing shorts development programme. It also looks certain that we’ll see further use of the hybridised ink-and-CG tool created for the project. “The system is flexible, so the art style is completely up to the art director,” says Osborne. “We’re still playing around with the technique, but I can’t say any more than that at the moment.”

“It was up to the artists to paste as they saw fit. With Meander you can draw anywhere on a character and the paste will work,” says Osborne

VITAL STATISTICS
Title: Paperman
Released: 3 June (UK), 5 March (USA)
Formats: Blu-ray/DVD (on selected versions of Wreck-It Ralph)
Distributor: Walt Disney Video
Watch out for… The subtle performances in the first railway station meeting, and the aeroplane chase through the city streets

What should be in the shortlist for CG Animation Short of the Year in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Discover the best 3D movies of 2013 at our sister site, Creative Bloq.

Have you seen a great use of CGI animation? Let us know in the comments below…

WANT MORE ABOUT PAPERMAN?
See breakdown videos of the hand-drawn technique and an interview with John Kahrs

Barcelona-based creative agency Trizz stirs the senses with this fully CG spot for drinks brand Renuage

“We were contacted directly with the brief,” explains executive producer and Trizz co-founder Christopher Vulpi. “The company had a basic idea, which we were graciously allowed to develop.”

The client provided the Trizz team with a lot of technical detail on how the brandy is made, with the wish that the spot would put across a similar story. “These taste experts are able to sense every bit of the drink, so the creative idea was for us to find a way to produce similar sensations, only visually,” says Vulpi.

To create a spot full of complex fluid simulations, the team used a Maya, RealFlow and Maxwell Render pipeline. “With RealFlow, you are dealing with simulations that have their own kind of reactions,” says Trizz co-founder and creative director Oriol Puig.

“Knowing that, you aim for special moments, but it still doesn’t mean you get exactly what you’d envisioned. What Trizz do, especially with fluids, is design the look of them and predetermine the edit so that more of what we want will happen. There’s still a lot of redoing meshes, which is why R&D time is nice to have on these kinds of projects.”

What should be in the shortlist for CG Commercial Campaign of the Year in the 3D World CG Awards 2013?

Tell us now on Facebook or Twitter #3dwawards!

Discover 10 beautiful examples of spirit bottle labels at Creative Bloq.

Peer into three alternate realities through the magical CBS eye in this set of striking idents

The global media brand CBS has recently launched a series of new idents for its TV channels in Europe, the Middle East and Africa, managed by Chello Zone. In each, the instantly recognisable CBS eye logo acts as a portal, providing a glimpse into an alternate space.

The concept for the rebrand was developed by creative agency Studio Hansa, which then approached Analog to collaborate on three idents. “We were aware from the start of the project that two of our pieces were going to be live-action, and the third ident, ‘Hero’, was to be entirely CG,” says Analog’s VFX supervisor Mike Merron. “ ‘Hero’ was initially also intended to be shot, so for us a large part of the brief was to not make it look CG.”

The Analog team called on a diverse range of software to achieve this. “We always use a variety of 3D software, including Max, Maya and Softimage,” says Merron. “For us, it’s about picking the right tool for the job and not limiting our artists. We base most of our projects in 3ds Max, but ultimately it depends on the brief. nCloth is incredibly powerful, so this project required a sprinkle of Maya.

“Point-caching helped us with the separate elements so we could lay out everything in Max while simulations were being generated in Maya or Softimage. Almost everything was rendered through Max using V-Ray.”

 
What’s been your favourite commercial this year? Let us know who you think should win the CG Awards now via Twitter: #3dwawards

Ioannis Karathomas explains how he made this tribute to Konami’s mech antagonist

Artist: Ioannis Karathomas
Title: Anubis
Software: Maya, ZBrush, Photoshop

“I’m a 3D environment artist at Zenimax Online Studios, and Anubis was a project I did in my free time after work. It’s a fan-art redesign of a character from Konami’s PS2 mech game Zone of the Enders, and it took me about four months to finish. I used Maya for the base modelling and blocking, ZBrush for detailing and rendering, and Photoshop for texturing and to composite the final images.

“I didn’t have a concept to begin with, so everything was made by blocking out geometry and trying to create interesting shapes. On some occasions my imagination stuck, and I would put my mesh into ZBrush and start sketching, trying to figure out the cut lines’ proportions and which surface to extrude. As soon as I created something interesting, I’d import the mesh back to Maya, resurface it and try to create a new clean geometry.

“The thing that inspires me as an artist is other people’s work. It’s a big world out there, and you need to keep up with fresh ideas in order to make it. When I see the work of a master artist, all I want is for my work to be on that level. That’s why every piece of work I create is always a little more ambitious and challenging than the last.”

Stage 1: Modelling

01 Creating a base mesh

Creating the base mesh was a big challenge. I rebuilt the same parts again and again until I’d made them right and I was satisfied with the proportions. Once I got the big shapes right, I started adding the big details such as cables cutting the panels. After that, I moved on to smaller details like bolts, small cables and mechanical parts, to add any extra elements to the scene and composed them as necessary. Once I’m happy, it’s ready to render.

02 Building a mech

Building a mech can be difficult. You need a lot of references and imagination to make it look functional, not just parts stuck to each other. That takes time, and experience of the kind you gain by making mistakes and redesigning the same shapes over and over again. You can see from the image how I ended up shaping and evolving my mech from blocking cubes in Maya.

There are many ways to achieve similar results, and another method is by using ZBrush 4R5’s panel loops. Unfortunately, I had already built most of the model before the latest version of ZBrush was released.

I try to put as many details on my models I can before I enter ZBrush. I feel that by doing it that way my mesh will be cleaner, and the shapes will be sharper. I also spend a lot of time bevelling edges and making sure that the light catches nicely when I smooth the objects.

03 Adding details

Most of the time I have no idea what I’m aiming for, and I don’t have a concept, so I’m looking for pictures of machines on the internet, or other ideas.

In ZBrush 4R5 it’s really easy to create groups and extrude and polish edges, especially when it comes to hard surfaces.

To begin with I import my base mesh, and I play with these shapes. I divide my mesh a few times and mask the area I want to extrude. Before ZBrush 4R5 was released, I would either extrude the masked area or inflate the unmasked area, but with the new version of ZBrush, I have the option to extrude my surface, keep it as a separate group and polish the edges – something that was a huge problem for me before.

I will start by masking a flat area that I want to change. Pressing [Ctrl]+[W] will create a new poly group that you can view if you turn the wireframe on. I [Ctrl]+[Shift]-click to hide everything except the group I want to adjust, and then under Geometry Edge Loop I set one loop, turn off Double mode, and also adjust the thickness which will tell me how far the geometry will extrude. Then I click the Panel Loops button and end up with two new groups of geometry extracted from my mesh.

If my edges are deformed because of my lack of resolution, I can adjust that by going to the Deformation menu and using the Polish by Groups slider along with Polish Crisp Edges to sharpen and tighten my edges.

I continue using the same techniques to create different shapes throughout the mesh.

Stage 2: Surface detail

04 Brushes and alphas

When I finally achieved my desired result, I loaded all the mech alphas that you can download from the ZBrush alpha library, along with some custom-made ones. I used these alphas to add bolts and other details to my model.

Pixologic also added new Insert brushes like IMM Ind Parts by CgCarter, which are awesome for adding bolts, especially for close-ups; but when I want to add hundreds of bolts, the Insert method will increase my geometry, so it depends how close you will see those bolts. In the case of Anubis, it’s there as decoration that you will see from far away.

I selected the Standard brush then the DragDot stroke and the desired alpha, and started placing bolts wherever I thought it would create the impression that this is a man-made machine that’s been patched up.

Once I’d finished adding bolts, I selected the Clay brush on Zsub, set to a low intensity, then went around all the edges and started adding details. Because I wanted it to look like beaten-up metal, I also used the H Polish brush. I always make sure I save a morph target or add a new layer before I do this, in case I want to make any changes. One of my favourite costume brushes is Orb_ Cracks, as it allows me to create battlescars, damage and cracks easily. I repeated these steps all across my model to get the desired result.

05 Extra details

When it comes to details it’s hard to know what you need to add and when you need to stop. You can keep adding and adding until your model starts becoming too busy or your mesh too polygon-heavy. After I thought I was done with Anubis, I put a render next to a screenshot of Optimus Prime from the Transformers movie. This convinced me I needed more detail, so I kept adding more until I was satisfied.

06 Posing parts in Maya

Posing was a challenge, and I had to be smart about it because my model was really complicated. For the hand, the easiest method was to go back to Maya, combine each joint separately, and after that start to parent each joint by pressing [P] and moving the pivot point to the right place. Then I was able to really easily pose each finger however I wanted. Then I imported my hand into ZBrush and placed it in the right position.

07 Posing parts in ZBrush

To pose the arms and legs, I took a different approach. I combined all the parts into different subtools in a way that was logical for the joints to move and rotate. After that I masked and grouped each panel that I wanted to be able to move. Once I finished that, the only thing left was to put the subtools in the right place.

Stage 3: Lighting and rendering

08 Lighting

For the lighting I did a basic three-point lighting in ZBrush. Under Light > LightCap, I added three new lights: one key light from the top with 100% intensity and the shadows on; one soft light from the back with 40% intensity; and one more soft light from bottom-front with about 25% intensity. The use of secondary lights was so my model would not go completely dark in the shadows, and I could show off all the details.

09 Render passes

For starters, I did a BPR render with the MatCap Gray material. That gave me a grey base, a shadow pass, ambient occlusion, a mask and depth. Then I used the Chrome B material to render specular, and the Pearl material for cavity detail. From ZBrushCentral, I’d also downloaded the Silver material, which I used for a shiny metal overlay, and SC_Pewter, which gave me highlight edges and dirt detail.

10 Layer comp

After I finished the renders, I added all the layers in Photoshop. I started with the basics: I used the grey layer as a base, then I multiplied the shadow, depth and occlusion layers, colour-dodged the spec and overlaid the reflections. As soon as I finished, I started painting colours, then overlaying metal textures and painting rust. For the finishing touches, I added highlights and hotspots.

11 Final image

For the final image, I created a night sky background and used the mask that I got from the BPR render to crop the background and place it behind my model.

CG Awards – HAVE YOUR SAY! Make your nominations now!

* Closing date: 14 June 2013
[Twitter] @3DWorldMag #3dwawards
[Facebook] www.facebook.com/3dworldmagazine
[Website] www.3dworldmag.com/cgawards

This new 30-second animation from international media studio Kompost introduces Viggle

Viggle is a new smartphone app that collects redeemable points while customers watch their favourite television shows. Created for US digital TV provider DirecTV, Kompost came on board with the project after the company’s creative director approached the studio’s US representative, Duck.

“The guys at DirecTV wanted to work with us on this commercial,” says Kompost creative director Oliver Conrad. “They loved our style, and felt we were an exact match for the brief.” So the team went on to design, direct and animate all aspects of the ad, using Maya and Arnold to fully achieve this.

This was the first large project the team rendered completely in Arnold, and they weren’t disappointed by its capabilities. “No production depends on anything more than its renderer,” says Conrad. “We can put in weeks of time and energy into modelling, animation and surfacing, but all that goes down the drain if the final images are not produced on time and to a high standard. We chose Solid Angle’s Arnold because it simply gets the job done. Though this piece was not in a style usually associated with physical ray tracers, it was flexible enough to produce a highly stylised look without any problems.”

Discover 20 top free 3D models at our sister site, Creative Bloq.

In Oz the Great and Powerful, Sony Pictures Imageworks conjures a CGI vision of an iconic fantasy world. James Clarke discovers the magic behind the effects

The entertainment industry has been adapting and creatively mining L Frank Baum’s Oz books ever since a 1901 Broadway musical. The latest movie version, Oz The Great and Powerful, directed by fantasy-film veteran Sam Raimi, explores the origins of the enigmatic wizard from Baum’s original novel, The Wonderful Wizard of Oz. Sam Raimi’s prequel plays to his strengths as a director, with Oz portrayed as a rich, imaginary world that’s both bright and dark, full of fantastical characters and exciting events.

To help create this immersive world, Raimi brought in two-time Oscar winner Robert Stromberg, the production designer behind Avatar and Alice in Wonderland. Instead of relying on greenscreen technology for every scene, they decided to construct physical sets so the actors could have a visual reference. “Robert worked with artists to build things on set very early on as an art-directed stage aesthetic,” says visual effects supervisor Scott Stokdyk, who has worked with Raimi on several projects – most notably the Spider-Man trilogy, sharing an Oscar for his work on the second in the series.

To build the digital versions of all the locations and lands in the movie, the visual effects team had laser-scanned versions of the physical sets. For example, in the scene where Oz first sees the Yellow Brick Road in the company of the witch Theodora, the team scanned the environment and modified it. “The topography gives us great visual cues, which we cherry-pick and re-art direct,” says Stokdyk. The team accumulated digital assets, such as greenery, and then were able to re-dress the set digitally.

“It is harder to combine live-action and CG, but we resisted the trend to go all-CG,” says animation supervisor Troy Saliba. In all there were 1,105 shots, and Raimi demanded only the best for each. “One of the things I love about working with Sam Raimi is that he keeps you on your toes,” says Stokdyk. “He was constantly challenging us to incorporate everything he wanted, and to create a consistent tone.”

THE LANDS OF OZ

Principal photography on Oz The Great and Powerful took place in Detroit, Michigan in the second half of 2011. During this phase, Sony Pictures Imageworks gathered information and data on-set to eventually refer to in post-production, which ran for a year from January 2012.

The visual effects team laser-scanned the physical sets in order to build the digital versions of all the locations

Balancing the on-set lighting with CG lighting was an overall aesthetic goal for the movie. In collaboration with Peter Deming, the film’s director of photography, the team gathered all the necessary data for the CG lighting in post. Sony Pictures Imageworks used a Spheron HDR camera to gather its extensive library of 360-degree photographs of the set; from these they made detailed records of on-set light positions and light intensity values to translate into the CG realm.

As well as building the surreal world, the team was tasked with creating digital characters and landscapes, such as a city of teapots. “We were doing flying monkeys and bubbles flying through beautiful landscapes,” says Stokdyk. The team used the original books for reference to the surreal world, and took visual cues from the original illustrations by WW Denslow for the fantastic characters.

Stokdyk admits that they didn’t even scratch the surface in terms of creatures they could have featured. One wonderful character they decided to bring to the big screen was China Girl, but creating her was fraught with technical difficulties.

A SMASHING LITTLE DOLL

China Girl presented a unique animation challenge. After seeing the first pieces of animation for the character, animation supervisor Troy Saliba recalls how he had to tell a rather surprised animator that the animation was too articulated, and that it had to be dialled back in its expressivity, which is rather counterintuitive. Saliba explained how the creative trick with China Girl was in capturing “a childish preciousness in a little porcelain doll, with her face all one piece”. They had to address the character’s movement, but also how to best make her face move without drawing the audience away from the illusion that she was made of china. “Because of the limitations, China Girl is really unique – there’s no recognisable animation formula for her,” Saliba explains.

Animating her was not only technically challenging, but also extremely time-consuming. For a ten-second shot, such as the one of China Girl interacting with Oscar Diggs (played by James Franco), the crew would typically commit ten to 12 hours per day over four weeks to the process of solving the dynamics of movement. With work on this part of the character animation completed, animation would then be produced for the cloth simulation and the interaction of the characters.

The animators were asked to ‘dial back‘ the expressiveness of China Doll’s face to keep the illusion that she was made of china

To keep the interaction of the digital and live characters believable, Stokdyk devised the Puppetcam, a pole attached to a marionette that allowed the actors on the set to see the face of the voice actor. “Puppetcam really helped with ad-libbing and vital performance,” says Saliba. Unfortunately, China Girl was too small to use the ingenious Puppetcam, and so a marionette was used on the set, operated by talented manipulator Phillip Huber.

Getting eye levels was just one of the issues. Saliba recalls how the film’s animated character work also included significant efforts to generate a wide range of digital doubles for the human characters. Additionally, thousands of digi-doubles were generated as extras to populate the land of Oz. “The toughest ones were stunt doubles for hero characters, because they were closer to the camera than we expected,” says Saliba. The team found creating digi-doubles of the witches particularly troublesome, because they had to try to capture the attitude of each witch when they were in flight around Oz.

“For the 3D characters, we were infusing a bit of the actor into the performance,” says Saliba. “Once we have a model, we start rigging and doing the facial poses. There’s some back and forth on modelling, depending on the topology of the face.” Animation tests of each character were produced once the rigging was developed, which would then go to Raimi for approval.

STEREOSCOPIC MAGIC

By January 2012, edits of live-action footage began reaching Sony Pictures Imageworks. Simultaneously, work began on some of the film’s more complex digital environments. It was an evolving process, and part of this creative evolution was in terms of the film’s 3D demands. Scott Willman, the film’s stereography supervisor, came on board immediately after the principal photography was complete, for which the on-set 3D stereographers had been Ed Marsh and James Goldman. One of Wilman’s key responsibilities was to oversee how animated characters would tie in with the movie’s 3D aesthetic. Willman recalls that Raimi’s feedback upon reviewing the 3D footage was to “make it bigger”.

FLEXIBLE DESIGN DEPTH

Oz The Great and Powerful had always been intended as a stereoscopic film, and so it had been shot natively in 3D rather than being converted to 3D in post-production. As such, the stereo element of the film was essentially set in stone by the time the footage from principal photography reached Sony Pictures Imageworks.

Willman explains that across the film there was lots of flexibility to design depth, making sure that multiple camera depths still worked with the native photography.

“Shooting native on-set was James Goldman, and it was his call to set the interaxial for the on-set camera. We’re used to seeing everything in binocular, and it’s important that things have a basis in reality, so in matchmoving we make sure that the CG has the same distance from the camera as the live action. There’s a higher degree of fidelity in stereo,” he says.

The film was shot natively in stereoscopic 3D rather than being converted to 3D in post-production

The Sony Pictures Imageworks team was able to enhance the native footage with new visual effects elements and stereoscopic adjustments. Being able to direct various aesthetic choices after principal photography was a great boon.

As the crew grappled with the climax of the film, they went back to the source material. “There’s a big effects sequence at the end of the film with a giant projected holographic head,” explains Stokdyk. “We wanted it to be an original-looking creation, so we looked back at the books – and there’s a floating head over a throne graphic. That’s what we went with.”

So, in a bright new digital world the old-world inspiration endures, and as the animators and visual effects crew become more comfortable with the technology, so their creative choices expand and unfold like a Yellow Brick Road.

Discover the best 3D movies of 2013 at our sister site, Creative Bloq.