Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Hi and welcome.
See Digital Cinema Tools Distribution for an easy-to-use Setup script. It will install everything required (batteries included) to use dcp_inspect, cinemaslides and a bunch of other useful tools related to digital cinema.
Please consider contributing to Testing encrypted DCPs and KDMs.
Read Mystery Meat to learn where the beef is. Frequently asked questions lists a number of issues that seem to come up again and again. See Links for some excellent articles related to digital cinema. DC Auth Blog is my notebook about digital signature, essence encryption and KDM generation, mostly. Couple of useful things in there. Digital Cinema Authentication is a gritty, bare-faced collection of technical notes, all about certificates used in Digital Cinema, Key Delivery Messages, server tests etc. Also feel free to check out progress reports about Dietrich, a digital cinema authoring tools suite.
If you find errors or would like to contribute please get in touch.
I work at the Munich Filmmuseum, Munich’s municipal cinematheque, which houses an extensive film collection. In addition we restore and publish forgotten, so-called lost gems of film history and provide daily screenings — spanning film history in all of its fascinating diversity.
What we take away from tracking down and screening thousands of film prints throughout the years, are, from a technical point of view, two major trends:
- Significant deterioration of most prints (not only old prints)
- Serious and, in many cases, scientific effort and investment to restore and secure select historic works and produce new negatives and screening prints
In other words: While some select works enjoy proper attention and care, large parts of our film heritage are literally doomed, mostly due to the lack of incentive and the significant costs of re-releases to screening prints. It became clear that digital tools would offer a chance, at least, to improve that situation. So we were thrilled when, in the summer of 2009, we could muster the budget to take our first steps into this promising terrain.
To establish an open platform for digital cinema production and distribution was the general idea of the Digital Cinema Initiative and their specification. Much like the 35mm platform (specified in 1916) which has had a lot of breath.
The DCI1 had realized that in order to create a platform good for decades to come they needed to adhere to tried and tested standards. As a result you can satisfy the requirements for DCDM2/DCP3 mastering and authoring with a set of freely available (open source) tools. Which is the laudable kind of Good Thing geeks feel irresistibly attracted to, like bears to the honey pot.
Film production, on the technical side of the imaging part, is all about workflows. Parties involved include cinematography, editorial, color correction (or, rather, authoring), archiving, delivery (film prints, digital cinema packages, BluRay/DVD, TV deliverables). They need to get on the same track in order to deliver photons to the silver screen. When film production was a film-only process (up until the early 1990s) the steps and techniques involved were, for the most part, clearly defined4.
This changed dramatically with the advent of digital tools and processes. Connecting the dots (which is what intra- and inter-facility workflows must accomplish) in the age of digital became more flexible and less defined at the same time.
Because DCI was aiming for a well-defined digital cinema standard the engineers came up with a gatekeeper mechanism (my words) that would precede the authoring of digital cinema packages (digital equivalent of screening prints): It is called Digital Cinema Distribution Master (DCDM) and consists of image, sound and subtitle/caption files with, for the most part, clearly defined properties:
- Black and white levels, bits per color channel, transfer function, colorspace and aspect ratio for images
- Bit depth, sample rate, channel count and reference level for sound
A DCDM is transformed — via a meticulously defined process — into a DCP.
See Open source tools for a digital cinema pipeline for commandline examples (keeping in mind that, due to the nature of these ad-hoc examples, the border between DCDM and DCP can be rather blurry).
See Cinemaslides for a DIY glue tool to create slideshow (or motion picture, for that matter) DCPs. It can, in a proof-of-concept kind of way, build encrypted DCPs and generate KDMs (Key delivery messages).
The required information and knowledge is, while public, somewhat buried in man-pages, specs and SMPTE standards. However, there is no reason DCDM/DCP production should remain more complicated and opaque than, say, the creation of DVDs or BluRay media.
Feedback, corrections, insights, code etc. very welcomed and much appreciated. Get in touch.
1 Founding DCI members (2002): Disney, Fox, MGM, Paramount, Sony, Universal, Warner
2 DCDM: Digital Cinema Distribution Master, collection of image, sound and subtitle/caption files ready (in compliance with SMPTE standards 428-1, 428-2, 428-3, 428-7, 428-9 and 428-10) for packaging.
3 DCP: Digital Cinema Package, collection of, at least, one packing list and all its referenced assets. Possible assets are MXF containers with compressed (JPEG 2000, MPEG2 elementary stream) and optionally encrypted image and sound essence, subtitle resources and composition playlists.
4 A voice, coming from an unidentified lab, says Define defined. Right. Color timers were actually considered wizard magicians. It is outright stunning what they could achieve with the limited tools (in comparison to today’s digital color kits) they had. Also, alas, I’m not aware of extensive collections of the work of color timers. Which means that although we’d have the technical means to reproduce and create any kind of color work in restoration, in many cases we simply don’t know how the original releases were graded. Therefore color restoration projects have to — in the best case — rely on the memories of production members, or come up with — in most cases — a lot of guesswork decisions.