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

Re: Custom Curves



Just to return to the gamma curve thread that ran a few days ago
I was trying to reconcile my own understanding of the comments
made with the practical use made of the current tools.

Once a base grading has been set for a given exposure then the 
process of balancing or, more typically,  creating and maintaining
a  "look" starts.  Now the original gamma curve, which most likely
will have been set with full black detail and uncompressed whites,
is subject to stretching. This will happen using a set of primary
controls, ie lift and gain in conjunction with gamma, which may be
in the telecine colour channel or further downstream.

As soon as you introduce another set of primary controls contained
within the first then it becomes quite possible to have a bulge
to the original curve that can maintain the details at either black
or white levels. In multilayered systems the layers of primary
control can be used to protect, enhance or obscure, original details
at various levels on the original gamma curve based on either
luminance levels or areas. The biggest problem with such a system 
is compressing either black or white levels at too early a stage and 
having to undo a near correct grading in order to change it. 

As these density changes contained within the overall primary grade
can be made with just the luminance of a secondary then we can 
currently get six layers involved although the RGB differential will
not be present on all involved layers.

Does the ability to play even two sets of primary controls against each 
other lead to a custom curve?

The other interesting problem is translating your multi gamma curved
grade for STV gamma response into a film print, a process that happens
very often, albeit now usually via an upres'ed D1. What seemed to be
a wonderfuly edgy look can quickly fall apart on projection with areas
of the frame crushed into mush and objectionable noise caused by
over stretched gamma in parts of the frame. The forthoming high
resolution grading systems will enable us all to produce some 
interesting results when these abilities can be incorporated into 
the Data outputs of the telecines.

Regards
Seamus O'Kane
VTR. Ltd. London. 

---
Thanks to Complete Post L.A. for supporting the TIG in 1998..
No product marketing allowed on the main TIG.  Contact rob at alegria.com
957 subscribers in 36 countries on Tue Mar 17 16:49:46 PST 1998
complete information on the TIG website http://www.alegria.com/tig3/