[Tig] 16fps Film Transferred on Rank Cintel & Pulldown

Craig Leffel craig at optimus.com
Fri Oct 22 15:55:17 BST 2010

Hi Gary -

Isn't it true that the Meta-speed add on for Ranks delivered a locked transport anywhere between 2fps and 96 fps ?? I modified a few different machines I worked on and always found it to be the case... and in fact, a verified field sequence output. However, it's also true that you can't do a locked edit at those non-traditional speeds, so we typically crashed in on our TLC's - and then it's true that you cannot account for what field sequence you'll be starting on, in the case of 16 fps whether it's a 3 or a 4. That would be random, but shouldn't matter in a piece to be re-conformed. In an archival print show, multiple tries are the best course... I would expect. It can be done if someone seeks out the right machine.

Craig Leffel
Senior Colorist

----- Original Message -----
From: "Gary Adams" <garyada at ix.netcom.com>
To: "John A. Mozzer" <jamworks at earthlink.net>
Cc: tig at colorist.org
Sent: Friday, October 22, 2010 9:34:45 AM GMT -06:00 US/Canada Central
Subject: Re: [Tig] 16fps Film Transferred on Rank Cintel & Pulldown

Sohonet http://www.sohonet.co.uk sponsors the TIG.
IBC Focus Sheet 2010 at http://www.colorist.org/wiki/TIGIBCFS10

Hello John.  I can offer some information here.  As far as I know, most, if not all telecines are not capable of a "locked" sequence when transferring any speeds other than 23.98 and 29.97 in 525 worlds, as well as 25.00 in 625 worlds.  Later URSA telecines were designed with a "locked" sequence when running at 24 fps in a 25fps video standard.  In the URSA handbook, the sequences were described for various frame rates but they were not guaranteed.  So while most of any given sequence would indeed follow the exact sequence, there would be the occasional added or missing field to make up any long term variation of the linear film speed with respect to the video output.  

If you consider film running at 15fps, you would need exactly 2 video frames (4 fields) for a given film frame in a 30 frame video standard.  So with film running at 16 frames per second, we need room for one whole film frame every second.  This would mean dropping 4 video fields out of the sequence every second to make room for the extra frame.  Since there are 60 video fields in the video second, the sequence would look something like:

15fps second:  	4:4:4:4:4:4:4:4:4:4:4:4:4:4:4
16fps second:	3:4:4:4:3:4:4:4:3:4:4:4:3:4:4:4

So indeed the sequence would be 3:4:4:4 as you have found.  You can verify this by stepping through the frames of your video where there is obvious motion and count the fields of four successive film frames. 

I would say you will find this sequence to be fairly good for the most part but realize each time the telecine is started, the sequence will start randomly, and the exact speed of the telecine is not guaranteed to be 16fps.  In fact in a 59.94 world, it will be slightly lower than that anyway, but over the long term, the sequence will drift a bit.

Sorry to be so detailed here.  It's an important problem for archivists since they wish to recreate as best they can the original film speeds in a digital world and it is not easy.  If you are going to work on films that are not our traditional locked speeds, it is best to transfer in a 1:1 mode and post process the film speed.  Then we get into looking at all of this on a digital display which is greatly off topic.

Best Regards, Gary Adams
Former TLC Product Manager
Now Revival Product Manager
Blackmagic Design

More information about the Tig mailing list