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

TK Transfer v Compensated Conversion

Jon at TVP UK wrote;

"At the risk of stirring up a hornet's nest, I would like opinions on the
following scenario:"

" A client has asked me whether to save money they can TK in 625 onto
DigBeta and then use a motion compensated converter with shot detection
correction to make a 525 converted master on DigBeta"

As Jew Dewolde mentioned standards converters are designed for video, not film.
But film originated can be run through a good standards converter and at times
produce acceptable results. You would find great variance between converter
brands, my guess is using Alchemist PhC the results will be relatively better.
This does not take into account the need for correcting colour space differences
between the two line formats, as they imply thay will be viewed by monitors with
different phosphors. 
However a better method (as suggested by Jeff) is mastering in 625/50 with TK
running the film at 25 frames, just as you would do in Europe (since it has more
spatial resolution than 525, e.g. more lines) then playing back over a VTR
capable of 48Hz (24 film frames) through equipment that simply adds the 3:2
field sequence (which are copies of identical frames) and then makes the more or
less transparent spatial transformation from the higher 625 to 525. Hence no
resolution is lost nor can there be any temporal artifacts (standard conversion
footprints) other than those originating from the film itself and would show up
no matter what. This process is not a standards conversion process as it does
not carry out  any temporal transformation, it simply does exactly what the
telecine does when running film at 24 frames and out putting in 525 format. By
the way, this is almost the inverse process of the DEFT and TK3:2 processes.

This is an excellent process but perhaps controversial as it requires mastering
and recording in 625 in a place where this is very seldom carried out ( the US).
Now if you are in the UK that makes the issue so much more interesting! I should
add that there are additional advantages in doing so in terms of timecode to
filmframe matching (no drop frames and a perfect film to videoframe match). As
you can see the audio falls right in place in terms of pitch rate for the 525
release. And it is even possible to manipulate field sequence (non-standard
then) to get equal runlength if a must. 

There is no hokus pokus in the process as such which conceptually has been
around for quite a while but it requires an entire working and greased system
and lots of experience if you want to deliver on time.

"Ahh! Sharp intake of breath from the TIG.  But wait - this very grainy
print needs considerable electronic noise reduction to be acceptable to
meet the distributor's requirements and therefore the effective
information bandwidth has already been compromised. "

Do not expect existing noise reduction technology to solve your problem without
disappointing side effects. This is not a case of noise reducers being good
enough it is simply that all these devices assume that film grain is equal to
random noise and that it has no doing in the image. This is of course wrong as
grain is the actual information carrier. Personally, I would rather see the
grain than the pain of NR smearing.... 

Mike Reichel
mikael.reichel at culinaire.se

      TIG subscriber count is 840 on Thu Aug 21 01:25:38 PDT 1997
   archives and much more at http://www.alegria.com/telecinehome.html
     mailinglist digest available.... unsubscribe via a message to
        'telecine-request at alegria.com' with Subject: unsubscribe