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

Re: Windows, layers and masks

 >> So this now begs the question, why has windowing taken so long to get a
> hold?  
> My impression is that it HAS taken hold and those that have the capability

Good, at last!

> >> I don't follow the  difference, on a 2d screen what would the approach
> difference be?
> Let try to explain it another way. The inside the window and the outside
> the window corrections both get the same source. As I understand it, in a
> layering system a higher layer becomes the source of a lower layer 

> Until now although the daVinci window system has many advantages,   

Okay, I shall call you on that, what are they?  And I mean, what are they above Pogles?

> situation where the thing you wish to window passes/is behind something
> else was more difficult deal with than on those systems that user layers.
> da Vinci windows are at present transparent, so using 2 windows is like
> filtering 2 lights on the set. Where the 2 lights meet, both filters act.
> There are no barn doors. The transparency is often desireable and a
> strength of windows, but the "barn doors", masking or whatever is also
> desireable hence EDWIN

Now I understand.....  RSQ does not take the information from the higer layer, it only takes the 
information that the higher layer left behind.  Thus if you selected the whole image to be primary 
corrected on layer one....  Well, that's it, you can't modify anything else.  So for example you 
want to have an overall correction and modify just the blues....  Then you grab the blues on layer 
one and then have the rest of the picture selected on the lower layer.  This goes for as many 
colours/objects as you want.  So taking this further...  Say a red object layer 1.....  Blue sky 
layer  2.....  green spill layer  4....  Rest of image layer  5.....   In reality I work 
backwards...  I take the whole picture, do an overall correction then work on the individual 
colours/objects if necessary...  So in reality not much different than working on an 888.

> >>>Edwin, the new da Vinci hardware option...is there anything new
> there...?
> It is very new to da Vinci Windows. It means for starters that the problem
> above of tracking a window behind a foreground object is no longer an
> issue. Now windows can be applied as transparent or as a mask. Also the
> application of windows is further simplified, and although the interface
> was not on show at NAB, I believe that Edwin will save still more time in
> sessions - for both the colorist and other players in the post chain as you
> say.Generating windows as vectors and being able to track each point is
> also new. It means shapes can change dynamically from anything to anything.

So really no different from Pogle or RSQ.....

> >>frankly saw no great difference in speed of operation of either a Pogle
> or anything >>else ONCE you got used to it
> It is true - actual speed depends on the operator. But I stick by my point
> that simpler is inherently faster, and that windows are simpler than
> layers. But this is a personal opinion - to which I am entitled :-)

And where is the complication exactly on either Pogle or RSQ???? 
Kevin, I saw you do a demo in HK of a fully configured 888 machine and I was extremely impressed, 
but it still looked complicated (not for you I know!).  I have also seen jobs done in Sao Paulo, 
both on an 888 and RSQ and both seem as fast as each other in the right hands.  In fact a "look" 
was generated at one facility house which the other then had to catch up with, as Brazil's "look" 
is very different!  Both places had different Colour correctors.

And my opinion....  I think the fundamental difference between the dV and Pogle/RSQ is the ability 
to select an object by luminance value and or Chromance value......  


Ken Robinson
Imagen Transfer
Av. Cristobal Colon 4733
Santiago de Chile

Tel: 56 2 207 9515
Fax: 56 2 228 5871

186,000 miles per second.  Not just a good idea.....  It's the LAW!

mailinglist digest available......posting guidelines on the webpage

Follow-Ups: References: