[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Some Queries About DVNR

Amid Amidi <junyer at concentric.net> asked:

>I was hoping to get a small explanation of how DVNR works in terms that
>readers without much knowledge of technical terms could understand.  The
>article is being written for artists and not technical folk so I can't make
>it too academic. Secondly, I was hoping somebody could maybe provide some
>info. as to why DVNR occassionally messes up cartoons - can this be
>attributed to the process itself or the editors who work with the
>technology or some other factors?

I've done hundreds (more like thousands) of projects processed through DVNR
over the last four or five years, and there's no need for any "mess up,"
provided the telecine operators and QC people do their jobs correctly.

Keep in mind that Digital Vision's DVNR does several different things, each
of which can help _or_ hurt, depending on how it's used:

1) noise-reduction (including adaptive NR) to reduce noise and grain;
overused, the picture can tend to "lag" and smear under some circumstances

2) image enhancement to help sharpen the overall picture; overused, you can
get an excessively "edgy" picture, adding a kind of ringing effect to
bright white transients

3) dirt/scratch reduction, which I suspect is the primary problem you're
talking about.  Used carefully, it can almost completely remove (white)
negative dirt from film transfers; if it's cranked too high, pieces of the
image can momentarily disappear and reappear.  In animation, I've seen
cases where characters' eyeballs go away, leaving the character looking
like Little Orphan Annie.  :-)

But the reality is that modern telecine equipment allows us to adjust each
of these modes on or off on a frame-by-frame basis.  I routinely will run
negative with little or no dirt reduction, and then turn it up _only_
during scenes with a lot of noise.  The moment any of the image is
adversely affected, we bypass it and make a note that the client may need
to do a dirt-fix elsewhere, in a Paintbox-type bay or in an on-line room.

Because of tighter post schedules, more and more animated series these days
are transferring from rough takes, then editing on videotape and
color-correcting the final videotape master.  In these cases, it's even
easier to control and predict the DNVR's effects.  Naturally, the original
camera negative is transferred directly to digital tape without processing;
only the final edited master is processed (color-corrected, noise-reduced,

The bottom line:  there's nothing inherent in the DVNR box itself that will
necessarily hurt animation or any other kind of programming.  The DVNR is
just one of many tools we have that can be used correctly, or used badly.
It's the _telecine colorist_ that makes it work or not work.  Sort of like
the "guns don't kill people... people kill people" philosophy.  Of course,
all bets are off if the transfer facility is trying to use the DVNR as a
standalone unit without any direct control.  It's not an automatic box that
can make its own decisions; it needs a human being to keep an eye on it
100% of the time.  And all of us have learned over time that a little
noise-reduction/dirt-removal often goes a long way.  Too high of a setting
is often worse than not enough, because of the potential for artifacts.

--Marc Wielage
  Complete Post / Hollywood, USA

Thanks to Video Post and Transfer for supporting the TIG in 1998..
No product marketing allowed on the main TIG.  Contact rob at alegria.com
979 subscribers in 36 countries on Sun May 10 21:46:04 PDT 1998 
subscribe/unsubscribe with that Subject: to telecine-request at alegria.com
complete information on the TIG website http://www.alegria.com/tig3/