Editor’s note: The following is a guest contribution from writer Meleah Maynard.
What would the DNA of a one-of-a-kind being look like if it could be extracted, examined and repaired? How do you construct a visual representation of someone’s life being decoded and taken from a single disc? And what does it mean to design an interface for a world that already exists only in digital form? These are just a few of the challenges facing the GFX team Bradley Munkowitz (a.k.a. GMUNK) assembled at Digital Domain to work on Disney’s “Tron: Legacy.”
Bradley Munkowitz, who served as lead animated graphics artist for a team of as many as seven artists working on rotation, was originally asked by director Joseph Kosinski and visual effects supervisor Eric Barba to create eight minutes of content. But the project quickly expanded to twelve minutes—not including the making of the opening titles, which they were eventually asked to do.
David (“dlew”) Lewandowski was the lead animator on the team. In a recent interview he described some of the challenges the artists faced and how they solved them using MAXON’s CINEMA 4D and VRAYforC4D. He also shared some of his frustrations with code art.
How did you get involved with “Tron: Legacy?”
It was really flattering to get called by Bradley (GMUNK) to do this. I’ve built up a reputation for my style and C4D skills. They knew I would be ecstatic to get to work on this. Bradley contacted me and Jake Sargent (also a lead designer) to start with. The three of us did the bulk of the work and Adam Swaab (C4D/Houdini artist), Josh Nimoy (code artist) and Joseph Chan (designer/animator) came in at different points to help. I’d say Bradley, Jake and I worked for a solid twelve months and everybody else anywhere from three weeks to five months.
Why did you choose to use CINEMA 4D on this project?
I think it’s really interesting to note that C4D played a key role in three of the most graphics-heavy films to come out recently: “Avatar,” “Iron Man 2” and now “Tron: Legacy.” These are huge, big-budget films and you see a pattern of C4D being used, which was not always the case. I think that’s because we now have BodyPaint, the MoGraph Module and other tools, and C4D has really become a go-to package for graphics. I’m hearing more and more Maya and Houdini artists saying, ‘Hey, maybe I need to learn CINEMA 4D.’
There are a couple of reasons for this. The evolution of the tools and the quantity of training has gone up in the last couple of years. Nick Cambell has a lot of tutorials, and there’s Cineversity and a lot of guys on YouTube and Vimeo putting out free training. There has always been a very DIY mentality to motion graphics, at least here in Los Angeles, and particularly with CINEMA 4D. The Maya community is not like that.
Can you talk a little bit about a sequence your team found particularly challenging?
On a technical level, and perhaps creatively, the Solar Sailor sequence was incredibly challenging. Sam and Flynn have to repair some damaged code in Quorra’s DNA. Joe wanted her DNA to be beautiful and unique looking. But we weren’t on set to do any of the creative development, so we had no idea how this scene was going to look or what was going to happen.
We did some research and showed Joe several ideas before going with what we called a ‘HexSphere.’ It was a huge achievement for us and we did it with the MoGraph Module’s polyFX and cloners. That was the first sequence we used V-Ray on, too, so there was a lot of learning and understanding going on to figure out our workflow on those shots.
The perception seems to be that global illumination with V-Ray is too time intensive. You don’t feel this way?
I don’t. I remember when V-Ray first came out for CINEMA and there was a big fuss about it. Finally we were going to get real motion blur and depth of field. But just as fast as they were celebrating it, people started to point out all the limitations: It was very slow and crashy. That was years ago, but that perception remains. It’s not always a perfect solution, and it can be slow if you don’t understand the settings. But if you’re judicious and you don’t mindlessly click around, it works beautifully. The user community is quite friendly as well.
We tried V-Ray because our work was composited in Nuke and they wanted 32-bit depth passes to simulate depth of field. Advanced Render doesn’t lend itself to floating depth passes, so we had to render out elsewhere. We looked at RenderMAN and Mental Ray for C4D, but for one reason or another, those things didn’t quite have the solutions we needed. We tried VRAYforC4D and realized things had changed. I used VRAYforC4D for every shot I did, including the opening titles, and I learned to use it quickly thanks to wizards John Niehuss and Doug Wilkinson at Digital Domain who sat behind me, and helped translate their Maya/Max V-Ray knowledge into V-Ray for CINEMA.
On your website, when you talked about “Tron: Legacy,” you mentioned that computer-generated code would not allow you the control you needed for shots, so you used C4D instead. What did you mean by that?
This is such an important thing to talk about because there is some press about the code used in the film out there. The truth is, Jake Sargent and GMUNK both had worked with code artists in the past. They felt it was important to use code to create a completely different design aesthetic for the project. I felt from day one that code art would not be reliable in production. It can be beautiful and run at very high frame rates in real time, but there is very little modularity when it comes time to fly something five pixels over, or generate a depth pass. When you have to do that last 10 percent, you can forget it. Code art is interesting, but it offers no control for an artist, and any tool you want to use has to be built form the ground up every time.
In the end, code art was an important design tool on “Tron: Legacy.” But a lot of that art had to ported over to Houdini and executed by a artist for production. I also had to go back and rebuild a lot of the created code art later in CINEMA because there was no method for rendering passes, or any of what was needed in compositing. I feel it’s overly ambitious to use standalone code art applications beyond the design phase.
To create the detailed graphics for Flynn’s disc download Lewandowski used both C4D’s MoGraph Module and hand animation to create a nuanced and complex patter of movement for interlocking rings.
How did your team get the chance to do the opening titles?
They came out of nowhere, really. Being commissioned to do twelve minutes of motion graphics was already a tall order. We were wrapping stuff up and we were really crispy when Joe said to me and Bradley, ‘Hey, do you want to do the titles?’ Of course we jumped at the chance. I think he offered us the job because we worked well together and had some great jam sessions earlier on in production. Bradley also had a relationship with Joe from back in their commercial days. You don’t usually get that kind of time with a director and it was great. Joe had complete faith in us and Bradley had complete faith in me, so I was able to create the most visible work of my career, possibly more purely visible than anything else I could create.
Meleah Maynard is a Minneapolis-based writer and editor. Contact her at her website: www.slowdog.com