[Tig] manual framing

Craig Leffel craig
Fri Mar 21 17:56:39 GMT 2003


Martin Parsons wrote;
>
> You can see this effect clearly on 2K scans. If the Datacine has
transported
> the film backwards, the sprocket tooth will have worked its way to the
> "back" of the perf. If the film is now run forwards, the tooth gradually
> shifts forwards, towards the "front" of the perf because of the "slight
> slop" mentioned above.
> This will affect the vertical framing of the 2K scans during this
positional
> transition of the sprocket tooth within the perf. I assume it will affect
> any image, SD, HD etc. - I mention 2K scans as that's where I've
personally
> seen it).
>


We see this everyday in commercial HD/SD transfers as well. It is
particularily fun when you are trying to line up a position against a 3/4"
( or any tape for that matter ) rough cut. You very quickly learn to only
step into the frame you're looking for while going forward. Reverse yeilds
an incorrect position much of the time, and takes more pre-roll to
accurately line up to the "correct" position  -- when stepping or running
the film.

As to repeatability, I have found that if I need to re-transfer a scene or
end up with a new forground for a comp, I usually try and retransfer all the
scenes in question, to make sure they all end up with as closely the same
position as possible. I don't manually frame much of anything... I will use
Cam Ap  and Tilt but not the manual framing keys. This seems to avoid
trouble when we do Spirit upgrades or multiple layoffs from different rolls.
At least, all I can say is that our compositors rarely complain about
multiple lay-offs not lining up. I was trained early on to frame up to an
academy countdown, but I have abandoned that idea long ago in search of a
single "correct" position I can rely on ( per format ).  I can't really see
a point where I would need to manually frame against my configs ( super,
standard, 2 perf, 3 perf, vista, etc ) outside of occasional 16mm jobs. In
the case of 16mm I would assume everyone would agree that all best are off
on repeatability.

We see the same thing happen in our 2K scans if we do not leave enough
pre-roll time.


Hope all is well out there in tele-color land. These are strange times we
are living in...


Craig Leffel
Optimus






More information about the Tig mailing list