you are here [x]: Scarlet Star Studios > the Scarlet Letters > coraline's technical innovations in historical context
<< before coraline puppet technology - with photos!
after >> learning ceramics with sara swink

February 10, 2009

coraline's technical innovations in historical context

by sven at 7:00 am

In my opinion, three major technical innovations in the art of stopmo have made Coraline possible: digital cameras/framegrabbers, CNC machining, and the 3D printer.

These relatively new tools are being used to try to solve stopmo's biggest challenge: how do we create puppets with expressive faces?

Whoever comes up with the best answer, leads the art form forward. There is no perfect solution yet -- but I think that Coraline presents the most plausible solution to date.

From King Kong to Jack Skellington

The influence of King Kong (1933) cannot be overestimated. The work of Willis O'Brien lead to the era of grand master Harryhausen, which led to the "hyper-Harryhausen" innovations of Phil Tippett... Three generations of "Dynamation," using stopmo to put photorealistic monsters onscreen with live action.

That lineage effectively came to an end with the release of Jurassic Park in 1993. Hollywood was overzealous in adopting CG monsters; Dynamation almost died out. Stopmo technology has now advanced to a point where a maverick director could make excellent non-ironic use of puppets in a feature monster film. Yet, there are big advantages to doing effects films in an all-digital environment, and one of stopmo's weaknesses is doing swarms of monsters... So it's implausible that the golden age of stopmo monsters (if one ever existed) will return.

However, 1993 was also the year that The Nightmare Before Christmas came out. Technically, it has roots that go back to folks like George Pal and Rankin & Bass -- yet, it was truly a seminal film. Rather than going for photorealism, TNBC embodies visually stylized storytelling. The aesthetic is one of translating 2D illustrations into 3D media.

Along with this new aesthetic has come new technical challenges to conquer. We know a lot about how greenscreen monsters into a live action context (for example) -- but do we know enough about how to make humanoid puppets that can give a full range of emotional expression?

Technical innovations in stopmo since 1993

Corpse Bride and Coraline have both been attempts to build upon and best the innovations of TNBC. Technically, TNBC had puppets with astonishingly tiny joints and it made remarkable use of replacement heads. Corpse Bride was able to take the art further by making use of digital cameras, framegrabbers, and clockwork head technology. Coraline further innovates by using stereoscopic photography, faces produced on a rapid prototyping 3D printer, and armature joints smaller than anything I've ever seen.

What do these innovations mean?

Well, framegrabbers and digital cameras have revolutionized stopmo as a whole. Our ability to create subtle movements, check our work, and delete a frame if a mistake was made... Well, that's huge.

Stereoscopic photography, though? It was beautifully accomplished in Coraline, no question -- but I'm going to feel this technology is a gimmick until there's even the slightest chance that it could become the default for live action filmmakers. I just don't see that ever happening.

Tiny joints are also here to stay, I believe. I can't confirm this, but I think that CNC machining has come a long way since TNBC came out. From photos, it appears to me that armature makers then had to make the majority of their joints one at a time. Now, the standard practice is to outsource making joints to a small parts manufacturing company. Having large supplies of generic joints on hand changes the armaturist's job. It's only on a very special character -- like Oogie Boogie, for instance -- that you're going to see things like universal joints, collet joints, etc. Armaturing now is, for the most part, more about assembly than design.

The big challenge: expressive faces

Where the real "arms race" lies, I believe, is in the technology required to make expressive faces. It's very difficult to do, and I don't think the perfect solution has been found yet.

Corpse Bride's clockwork heads were astonishing... But also profoundly complicated -- far out of the reach of any indy filmmaker. (Though Ron Cole's cable puppets offer a plausible alternative.)

By contrast, creating heads like those in Coraline is not overly complex. There are two main strategies: paddles attached to tiny step-block joints inside the puppet's skull, or resin faces that snap into place with magnets.

Paddles are somewhat prohibitive due to the difficulty of making tiny joints, which are more easily accomplished on a CNC machine. Snap-on faces are most easily created using a 3D printer, but could still be done on a small scale by resculpting a set of identical clay heads and making multiple resin castings. If you had the right tools, neither strategy would be too difficult; yet even without CNC and 3D printing, you can hobble along making things by hand. The only thing that's really out of reach for the solo animator is making thousands of snap-on facial expressions -- and that's really just a function of time/money.

[In an odd way, I find this all very heartening. It was depressing that I couldn't even understand how the Corpse Bride heads worked. As an artist, I felt defeated by the "experts" in my field. With the Coraline puppets, though, I've been excited to realize that I understand what I'm looking at. And I could imagine how one might use this set of strategies to solve specific puppetmaking problems. It's encouraging to think that I could maybe hold my own alongside other puppet fabricators, were I to ever land a job at someplace like Laika.]

Which technology will win out?

I think the Corpse Bride clockwork head technology will be abandoned because it's overly complex. There will only ever be a tiny number of people who understand how to make those heads. It won't be passed on to the next generation... And research and development costs at the outset of a film would always remain high.

Snap-on faces and paddles, on the other hand, are not that complicated if you have access to CNC and 3D printers. It seems likely to me that this technology will catch on and become the future of puppet faces in stopmo feature films.

Yet, snap-on faces are not a perfect solution to the problem. In contrast to paddles, you get exquisite control -- freedom to work out the faces in advance, so that "acting on the head" becomes as mechanical for the animator as following an X sheet for lipsync. But what about those seams between the forehead and the lower face? The folks working on Coraline did a great job of digitally erasing them (in 3D no less!) -- but what a hassle! In some shots you can see that they just decided to leave the seams in (notably on Bobinski).

So, there you have it... Coraline represents the state of the art -- but our art form still has a significant technical problem to resolve: how to best make expressive puppet faces?


Maybe use of 3D printers isn't beyond solo filmmakers after all! I just remembered a link that Shelley Noble sent a few months back, about an online service that does 3D printing called You send them the file, they send you a 3D plastic object. Price? $2.89 per cubic centimeter of material. 3D printing demo

posted by sven | February 10, 2009 7:00 AM | categories: stopmo