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

Re: BIT DEPTH



 Reply to:     RE>BIT DEPTH
Regarding the use of Dynamic rounding, 

On 01-17-95 David Catt responds...

>Quantel are probably correct to some extent however, we at
>da Vinci manufacture systems that are not items that are generally 
>used in a transparent state, unlike Quantel. We are often dealing 
>with picture information that requires much amplification, a sky 
>that needs much more blue, as an example.

>Henry and Harry are normally cutting and pasting images which have 
>been graded to provide the correct signal level and only require a
>slight balancing of that color.

>David Catt / da Vinci Systems.

On 01-17-95 BT responds...

Hmmm. Good point there. I guess my concern is mostly in the stages
starting at the output of the color corrector. For Telecine, the
more we can stuff into the front end (Renaissance) the better.

What do you suppose would be the effect of a D1-D1 Renaissance color 
correct on material that has already been dynamically rounded, such
as a final grading of a multilayer HENRY spot?

Also, I wonder if the 10-bit depth is OK for film, whereas it may
not be enough for computer generated images. For film, I would
imagine that grain and other physical issues may tend to "dither"
the signal enough so that normal photographic gradations will not
exhibit the quantization artifacts. And since more of our work
involves a mix of film and CGI, I thought this was worth some
batting around a bit.

bt at edny.univid.com