Mysterious S16mm unsteadiness

[Kar Wai Ng]

I have been working as the sole camera assistant on a documentary shooting on S16 for the past couple of months, and we are running into a problem regarding image stability. A large percentage of our footage, when transferred, is showing image unsteadiness, side to side, up and down, and the image movement oscillates in elliptical or figure-8 movements, enough to be noticeable in wide, locked off shots.

 

Reg tests show both passes completely in line with one another, with no breathing whatsoever, but the entire image shows unsteadiness. When reg tests were watched on a projector in prep, the only thing I could evaluate was registration...there is so much gate weave from a projector that nothing else can be evaluated. Only very recently when we were trying to troubleshoot this problem, did we put the reg tests up on telecine and saw the same unsteadiness here as in our footage. Camera is an Aaton LTR 54 that has been serviced recently. There does not seem to be a correlation of unsteadiness to certain mags we are using.

 

We have also pretty much ruled out telecine as the factor. Footage has been transferred on two separate Spirits on two coasts; we have also put footage on a Spirit 2 and it was also brought to another post facility here in Toronto. When we are in the suite, other productions' footage looks steady, and in fact, the lab leader on our footage is steady, but once our footage comes up, some unsteadiness presents itself. On a Spirit we can't zoom out enough to see the perf entirely, but the edge of the perf that is visible does seem to move in sync with the image. I've been told the Spirit does not have EPR (electronic pin registration) so it cannot be dialed out on the scanning stage. Because the movement is both horizontal and vertical, it apparently doesn't have anything to do with film tensioning or rollers on the telecine.

 

At the moment we're currently suspecting film. The vast majority of the film has been shot on 7219; so far we've shot 36 rolls, 32 of which have been 7219 (same batch, roll, cut), 3 have been another batch of 7219, and one has been 7201. It would be very easy to narrow it all down to this particular batch of 7219 we have been using, except not all of our footage manifests this unsteadiness, and in fact there's a possibility that the material we shot on a different batch of 7219 may be unsteady (hard to tell, because very little of this material is lockoff.)

 

We have gotten many colourists, technicians, etc to look at the situation, and it is perplexing everybody. We showed some footage to some Kodak representatives, and we floated the idea that perhaps the perfing or some other tolerance of the film is out, causing the film transport to place the frame in the gate exactly the same way in both passes of the reg test (hence good registration between passes), but temporally speaking, an unsteady perf location would cause the image as a whole to swim around. We have gotten given Kodak a reg test shot with our 7219 batch and a rack leader that was shot on 7217 (which looked steadier) which will go to Rochester for evaluation.

 

Until then, we have all this material that needs to be stabilized. No one noticed any of this in dailies and none of the dailies colourists reported problems...so only now in editorial when some simple compositing is being done that we notice the unsteadiness.

 

We go to camera again in mid-July, and need to isolate the problem as soon as possible. Post stabilization seems to be the only option for our existing material (or is it?), but for when we go to camera again in a month, we cannot continue to stabilize all of that footage.

 

I'm putting this out there to the cinematography.com community....thoughts/experiences/advice?

 

Thanks!

[Kar Wai Ng]

Sorry, this post should probably be in the 16mm forum.

[Simon Wyss]

After your description it appears to me that you'd encounter the very rare possibility of sabrage, as the technical term is.

 

It is film strips having edges of a snake form due to not perfectly straight running circular knives. It cannot be the perforation because hole irregularities almost always provoke fast movement such as jitter. A figure-8 movement leads me to imagine that the camera's film side guide was overstrained by wavy cut stock, all the more apparent since the image is offset.

 

You can instantly verify that with a projector that positions in conformity with ISO 5768 on which you run some feet of the suspected negative. You observe the film directly how it behaves in the gate. There are camera, printer, and projector constructions which undergo the concept of reference edge and spring-loaded lateral pressure on the opposite edge like the Aaton Penelope ( ! ). In the 16-mm field geometric issues are even more critical.

[Kar Wai Ng]

A bit of an update: Kodak doesn't yet have a determination what the cause of the problem is, but they have noticed the unsteadiness as well when they ran the film through their Spirit in Rochester. However, they also ran it through an older Shadow telecine...and it was rock solid.

 

I know the Shadow was the precursor to the Spirit, but beyond that, I'm not familiar with the technical differences. The Shadow doesn't have EPR, does it? Does the film transport differ in any way?

[Freya Black]

A bit of an update: Kodak doesn't yet have a determination what the cause of the problem is, but they have noticed the unsteadiness as well when they ran the film through their Spirit in Rochester. However, they also ran it through an older Shadow telecine...and it was rock solid.

 

I know the Shadow was the precursor to the Spirit, but beyond that, I'm not familiar with the technical differences. The Shadow doesn't have EPR, does it? Does the film transport differ in any way?

 

I could be wrong but my understanding was that the shadow was released after the spirit as a lower cost alternative type thing.

 

love

 

Freya

[Bernie O'Doherty]

I have dealt with this problem quite a few times now. I believe the problem is twofold. One, imperfectly cut rawstock, which, over the years, widens the aperture-plate rails. Two, aperture-plate rails which are incorrectly positioned. Your problem may even be a combo of both One and Two. Have a trusted technician check on the width of the rails.

Another issue is the age of the filmstock. Film rawstock shrinks an amazing amount over time, thus the development of spring-loaded side rails, which try to cover all the bases.

[Kar Wai Ng]

I have dealt with this problem quite a few times now. I believe the problem is twofold. One, imperfectly cut rawstock, which, over the years, widens the aperture-plate rails. Two, aperture-plate rails which are incorrectly positioned. Your problem may even be a combo of both One and Two. Have a trusted technician check on the width of the rails.

Another issue is the age of the filmstock. Film rawstock shrinks an amazing amount over time, thus the development of spring-loaded side rails, which try to cover all the bases.

 

Hmm. Thing is, the camera was just serviced last year, and another project that was shot with the same camera had no problems (different stock though-7218), so I don't think it could be the rails. And the film stock we're using is fresh stock directly from Kodak, not something that's been sitting around for a while in various storage environments or humidity levels.

[Simon Wyss]

I begin to gain the impression that there is virtually nobody of the telecine world who'd know anything about mechanical subjects. A Shadow functions on the continuous film transport principle like a Spirit, so when there is one which produces steady images the others appear to be out of trim. I am now believing that the films are okay and some telecine rollers revolve like the cock on the fire.

[John Salim]

All film, be it stills or MP is cut and perforated to an extremely tight tolerance.

Processing does change the physical dimensions by a few microns but nothing would cause that amount of movement in projection.

 

My gut feeling is, to get cyclic horizontal and vertical movement, it has to be shot like that - in the gate ( especially if it repeats over and over ).

 

I'd suggest a short test with the same film batch through a different camera.

 

John S :(

Edited July 16, 2009 by John Salim

[Robert Houllahan]

The Shadow and the Spirit share the same transport, gates (basically) and the servo it is the optics, light source and video processing electronics that separate them.

 

Have you thought about or tried a Pin-Registered scan? We have a beta 2K P+S Steady-Frame scanner at Cinelab now which does pin registration based on machine vision of the perforation I would be happy to run a scan test for you.

 

-Rob-

[Dominic Case]

At the moment we're currently suspecting film. The vast majority of the film has been shot on 7219; so far we've shot 36 rolls, 32 of which have been 7219 (same batch, roll, cut), 3 have been another batch of 7219, and one has been 7201. It would be very easy to narrow it all down to this particular batch of 7219 we have been using, except not all of our footage manifests this unsteadiness, and in fact there's a possibility that the material we shot on a different batch of 7219 may be unsteady (hard to tell, because very little of this material is lockoff.)

You've cearly illustrated the difficulty of diagnosing this type of problem here - simple comparison tests are rarely conclusive.

 

However, what you need to do if you haven't already, is shoot a comparison test on your camera, and on another similar one. A double-run registration chart test would be ideal. And you should use the same batch of stock that you've been using (if any is left) in both cameras. Be sure to get them processed together (excessive temperature in the drying cabinet can cause warping of the film, which could lead to the problem you describe - and that could affect one type of emulsion more than another, as they have slightly different water-loads, so dry at different rates).

 

Simon's theory is also relevant: if the camera uses edge-guides, then any sort of pin-registering in the telecine will show up any weave in the edge slitting. I haven't heard the term sabrage - thank you Simon, for that new piece of information.

[Stephen Williams]

Hi,

 

Historically 35mm with double pin registration was always used for shots requiring any optical work or title backgrounds. 16mm was not considered steady enough. With the ability to stabilize footage, the rules went out of the window.

 

I know a few foolish producers who tried to save money using 16mm for motion control against advice, always a disaster costing way more to fix than the few dollars saved.

 

Stephen