[tig] Fw: 4K for DI - Unappealing image quality

mhafner at imdb.com mhafner
Wed Apr 14 08:39:43 BST 2004

#Marc Wielage wrote
#Go rent the DVD of OPEN RANGE, and you tell me.  We had zero complaints from
#Disney's QC department, and their DVD people tell me it went through MPEG
#compression just fine.  It also got very good reviews from all the major
#U.S. video mags -- not they they're necessarily bright, when it comes to
#spotting problems like NR lag.

I'll check out the 35mm print when it's playing here and report back.
"Wide Screen Review" reported minor smearing and some 'pixelisation'
for the DVD. But who knows if additional DNR was applied to the DVD master.

#Don't forget, there is such a thing as grain-reduction algorithms that are
#non-real-time.  These have to be rendered -- sometimes overnight -- and have
#zero artifacts because they don't rely on temporal information in the

Are you saying these algorithms do NOT use information from more than one
film frame to reduce the grain? If so, can you point me to a document that
describes the theoretical framework for such an algorithm? 

#So the answer is no:  not all grain reduction is created alike.

Yes, but effective grain removal without temporal filtering and general
loss of fine detail is news to me.

Michel Hafner

More information about the Tig mailing list