[tig] Subliminal motion in 16mm film tranfers

Robert Lund robert.lund
Sat Jun 19 23:01:33 BST 2004


Craig Leffel wrote:

>Are you all INSANE??  Seriously, what Kool-Aid have you guys been drinking 
>where you've convinced yourselves that 16mm film weave is a GOOD or Necessary 
>thing?
>
>I've spent the better part of my career hating it and trying to get rid of it. 
>It is a catagorical mechanical anomily. ( Camera OR Telecine ) WHY would 
>anyone want to see it? 
>  
>
This thread has strayed from what I think was the original intent of 
George's post. His key point was very thought-provoking:

>There is something about the eye needing to
>constantly "track" the image that seems to draw the viewer into the
>program that does not take place with a perfectly locked video frame.
>The 16mm film asks you to work a bit more... to become a participant if
>you will. Video says, "just sit there and shut up...we are doing all the
>work for you". 

I like steady pictures as much as the "next guy." I was extremely impressed by the steadiness of the digitally projected Star Wars film a couple of years ago compared to the film version I saw later the same week as the Ziegfeld Theater in NYC. And, struck as I initially was by the jitter in a film I saw at the Lake Placid Film Festival last week amidst many other video projections, I was no longer able to perceive it after several minutes into the film, as the eye-mind apparently adjusts. But the subliminal effect of film jitter on the viewer's attention might indeed have the effect hypothesized.

Key point:

>All this takes place without the viewer's knowledge. 

Hence no implication that anyone would "want to see it."

>That is, unless they work in our industry and understand what is happening.

The technical mind strives for "perfection", but that can be an elusive quality. An example which comes to mind is a project I worked on at MTV a few years ago, where Beavis & Butthead were to be controlled in real-time by data gloves (worn by Mike Judge). When models constructed in SoftImage were controlled by the data, the character motion looked smooth and perfect, but eerily unlike tha familiar characters. B&B animations are constructed from a set of pre-drawn images resulting in the familiar jerky motion of the characters. So, although the models performed "more realistically", it was decided to store a set of 100 or so fixed images of the boys, and use the glove data to select images from this random-access flipbook, effectively reproducing an artifact which arose out of the original technical constraints of producing the weekly episodes. The result, though far less "realistic" than the 3D models, produced characters which resembled what viewers were accustomed to seeing on their TVs. Just another case indicating that "better" isn't always "better."









More information about the Tig mailing list