RE: [AV Media Matters] Video to film for preservation-another perspective.

[James Lindner <jim@vidipax.com>]

Sometimes you KNOW when you hit the send key that you should have
waited. That was the case here with my comment on video to film for
presentation. Eddie Zwaneveld's email was so good that I posted it twice
(not really - that was a mistake) but seriously I should have not been
so Grumpy with my reply - call it post NAB Grumpy Syndrome.

Eddie did a nice job going over the history of video to film transfer
technology. There is quite a bit of it - and I agree that there is some
very interesting work being done around the world - Swiss Effects in
particular has done some fine work in video to film transfer - but to me
it seems that the point of the discussion was whether film is a viable
media type for video PRESERVATION. To me the issue is not so much
whether one can record video images on film (we have been doing that
with different levels of success since BEFORE videotape), but whether
doing this is in fact preserving the video imagery in a way that truly
represents what it was in the first place. I think that the answer to
this issue is still - no - video recorded on film DOES undergo a very
significant change and does look dramatically different upon playback -
and at the moment I am referring to use film as an image holder as
opposed to a digital data recording which is another subject entirely.

While 24P to film is very interesting to consider as a preservation
format, - note that the assumption here is that the footage was shot at
24P VIDEO in the first place. Most NTSC video is shot at the 29.97 frame
rate, and PAL at 25 (closer but no cigar). So then one must consider how
one get's down to the 24 frame rate for PRESERVATION - clearly the
standards conversion approach (either motion vectors or field/frame
interpolation or field removal) are not appropriate, and then one has
the rather difficult issue of converting BACK to video from the film -
the 3:2 pulldown creation process.

Could one use film more as a "carrier" in the intermediate sense are
record with a single frame to single frame relationship. Yes you can -
and indeed I tried this once. The issue there is again reversing the
process - but it is not that tough on a modern telecine (of course you
are using MORE film in this process then the normal 24 fps frame rate).
But when all is said and done what do you end up with - and is it an
accurate representation of what the video looked like in the first
place.

The answer is no. It is interesting, it is different, but it is
definately does NOT look like the NTSC image that it started out as -
and to me this is the ultimate point of it all. While film does indeed
have more range then video - the response curves are totally different -
indeed they differ from each stock type - and the basic color system is
totally different. That is not to say it looks worse (although often I
think it does) - but it certainly looks very different then the
original. IF the maker of the original content wanted it to look that
way as an aesthetic decision it is an entirely different matter - but
from a preservation point of view - I don't think that this is the way
to do it.

Now using film as a carrier to store data is an entirely different
matter - and it could be any type of data. To my way of thinking this
has a much better chance of success - but there are competing
technologies that simply do it much better and MUCH cheaper.

jim lindner