[Tig] GraphicsMagick & SMPTE DPX

Martin Euredjian ecinema
Tue May 10 23:06:20 BST 2005

Dick Hobbs wrote:

> Arcane discussions about file formats are getting towards the edge of

Well, I think they are not as close to the edge as it may seem.  Relevant
they are.  You just have to have a thick skin to get into it publicly on the
TIG, 'cause you ain't gonna make everyone happy, no matter how much of an
effort you put forth.  Just try and count the number of public, in-depth,
information-ladden discussions found over the last five years or so.  One
hand might suffice.  You guys are a tough crowd.

Of course, there are various USENET newsgroups where such discussions are
significantly more relevant.  Some that come to mind:

...and more.

> when people get impatient because others are not dropping everything to do
their research for them, I think we are missing the point.

The best I can gather is that Bob wanted to see a version of Annex B showing
LSB-first order.  Having burned two hours on my very first post (where I dug
through six-year-old code and offered-up some of it), and another couple of
hours giving it another shot, I truly didn't have the time or patience to
give it another go.  What was needed, for my tiny little single-neuron
brain, was a better question.  

As Mr. Orton likes to say, two people divided by a common language.

Perhaps a study of the Cineon format would help here.  Also, there's one
very important historical difference.  Cineon allows for a much finer
definition of the data packing mechanism.  DPX does not, as it only allows
specification of packed or filled states.  I've seen files come out of
equipment where the authors choice fit neither option.  There are some
hybrid DPX-a-la-Cineon files out there. That's when you have to get creative
and allow for customization of the packing decoder.

Martin Euredjian
eCinema Systems, Inc.
voice: 661-305-9320
fax: 661-775-4876
martin at ecinemasys.com
ecinema at ieee.org

More information about the Tig mailing list