for in house purpose

DCP - learning guide -

DATA COMPILATION - various sources

MILE STUDIOS- DUBAI
Adriana Craciunescu

Monday, September 19, 2016

DCP

What Is DCP?

A Digital Cinema Package is a specially designed collection of file types de-

fined by the Digital Cinema Initiatives  organisation which was created by the major
Hollywood Studios to standardise the distribution and playout formats for digital
cinemas around the world.

Can be also characterised as:

- a digital equivalent of a 35mm film print.
- what you give to a commercial theater so that they can screen your movie on a
digital (also known as "D-Cinema") projector.

Background:
Due to the emergence of Digital Cinema, many theatres do not have any 35mm
equipment anymore. All new theatres being built are digital only but from the late 1800s up until
about 2011 if you wanted to show your movie in a theatre you had to make a 35mm film print.  
Even if you shot on video tape you’d have to transfer it to 35mm in a process called a “filmout”,
and once you had your film on 35mm you still had to make prints and send them to every the-
atre.
The utility and cheapens of DCP shows when thinking of the following example : A stu-
dio releasing a movie to 4000 screens might spend $6 million just in film prints alone; in contrast,
making a feature DCP typically costs 90-95% less than a “filmout”

What is it made of ?
A DCP contains video, audio and subtitles (if needed) along with instructions
for how to play them. The list you will see contains:

1. MXF File(s) containing the pictures in JPEG 2000 Codec & XYZ Colorspace.
2. MXF File(s) containing the sound channels.
3. A number of XML files identifying the elements of the film and how they should
be played.

Having separate picture and sound files is very practical for films, it means that different
language versions can be produced without having to re-encode all of the pictures.
 for in house purpose
DCP - learning guide -

DCP’s can be delivered to a theater on a

standard computer hard drive, either USB
or “CRU”.

Steps in creating a DCP:

1. Mastering this is the actual process of converting

your video/audio files into the format recognized by D-cinema systems.

2. A quality Check or 'QC' is where the final product is checked for glitches, dropout,
sync problems, gamma, color, etc. by an experienced technician.

3. Transfer to USB or CRU drive. This is the final step when the mastered files (collec-
tively called the DCP) are transferred to an EXT 2/3 formatted Linux hard drive.  
The actual drive can be a standard portable USB available in any computer store, or a
professional "DX115" drive carrier, which is called a CRU.  Again, both USB and CRU
have the exact same information on them, so there’s no difference in quality.  Some
theatres and film festivals demand a CRU however most are perfectly happy to use the
much less expensive USB drives.

MXF Wrapper Format and JPEG 2000 encoding:

First of all there are file containers, sometimes called wrappers, that wrap
around a number of video and audio tracks. Each of those tracks will have an appropri-
ate video or audio codec. A codec is a concatenation of “coder – decoder”. (Think of a
shipping container. There’s this standard “wrapper” (the container) which tells us noth-
ing. Inside could be a car, computer or a million wrist watches. Like the shipping con-
tainer, file containers can carry many different types of content – the video and audio
tracks. These tracks are encoded with some sort of codec. Most codecs compress the
video to reduce file size and time to download )

So why use a wrapper at all? if you think about a feature film having around 150,000 in-
dividual frames in it, then transporting and playing 150,000 individual files can be a real
pain so putting them in an appropriate wrapper format is really useful.
 for in house purpose
DCP - learning guide -

One of the big advantages of the MXF (Material Exchange Format) is that it
isn’t tied to any one manufacturer the way that something like Quicktime is tied Apple.
While there’s lot’s of formats capable of doing the job, MXF is one that’s been developed
by the industry and won’t suddenly leave the industry with licensing problems in a few
years time. It’s also pretty agnostic about about platform and codecs, so it’s very flexible
and it was designed from the beginning to be very good at handling the additional infor-
mation that goes with a lot of content these days – Metadata.

Within the Picture MXF the images are encoded with intra-frame com-
pression in the JPEG 2000 codec. This is very different to the everyday JPEG format
that everyone uses with digital cameras an a vast proportion of the images in the web.
( for this case JPEG uses a type of  DCT  compression which breaks the image down into blocks
of pixels near each other that can be grouped together and colour values averaged.  This works
on the principle that pixels near each other are often similar colours; the bigger the blocks, the
bigger the reduction in the amount of data used.  JPEG & DCT are great at getting good looking
images down to file sizes that are a fraction of the original. The mathematics behind the process,
therefore the processing power and time needed to do the job are relatively modest. The down-
side is that when the compression does become visible it is in the form of those blocks.  This is
most noticeable in big areas of similar colours light the sky or a white wall, where one block av-
erages one way and the adjacent ones go a different way.)
On the other hand, JPEG 2000 is different. It uses Wavelet compression
which is much more complicated both as a concept and to implement. The upsides are
that the quality is much better and on the rare occasions when it does degrade the im-
age, the effects look much more natural and are therefore much less noticeable to the
audience.

XYZ Color space:

A Color Space is a mathematical model that maps the colors that can be re-
produced by a device to a standard color model.
The eye is as good as any reference when building a color space. Why? Even
though we have displays that claim to show a billion colors, the eye can never see all of
those colors. One of the first mathematically determined color spaces is the CIE XYZ
1931 color space, created by CIE (International Commission on Illumination) in 1931.
A color space is actually a 3D image.The two-dimensional image is only a
cross-section of the entire color space.

It is called a Chromacity diagram.

The diagram represents all of the chromaticities visible to the average person.
The color region is called the Gamut of human vision.
The curved edge is called the spectral locus and corresponds to monochromatic
light, with wavelengths listed in nanometers.
The straight edge is called the line of purples. These colors have no counterpart in
monochromatic light.
White is at the center (E).
There are no three points within the gamut that form a triangle that includes the entire
gamut of human vision; or more simply, the gamut of human vision is not a triangle.
 for in house purpose
DCP - learning guide -

-example of the color space representa-

tion we are referring to in the previous
paragraphs.

Other color spaces are also represented with similar chromacity dia-
grams that make it slightly easier to visualize them in comparison with each other.
There are color spaces ‘smaller’ and ‘larger‘ in volume than CIE XYZ. Therefore, a
two-dimensional diagram shouldn’t be used exclusively to compare various color spa-
ces. It is possible for one color space to have lesser colors than another but still have
colors that the “larger” color space doesn’t have.

XYZ colorspace essentially takes RGB space and then refines it dra-
matically to match the way the various Rod & Cone “pixels” in our eyes actually combine
to perceive colour. So while RGB is a very electronic friendly way of capturing, control-
ling and manipulating colours, XYZ space is a very efficient way of presenting them so
that they best match the human visual system.
 for in house purpose
DCP - learning guide -

Mastering the DCP. INTEROP & SMPTE

INTEROP:
- a packaging format that was agreed between some of the early manufacturers
of digital cinema equipment to try to ensure that content was interoperable between all
of their systems. The SMPTE DCP is based upon the Interop DCP but with some fur-
ther enhancements;
- The interop DCP format cannot carry soft 3D subtitles. Instead they have
to be rendered into the image track of the film which means that every subtitled lan-
guage version has to be mastered and distributed individually. An InterOP DCP can
have encrypted picture and audio but not subtitles. This can be a risk particularly for dis-
tribution of previews of much anticipated blockbuster titles where the distributor wants to
keep the details about the plot secret.

SMPTE - Society of Motion Picture and Television Engineers.

- represents a standard.
- supports Auxiliary Data the ability to include additional data in the DCP other
than the picture, sound and subtitles. ( For example,Dolby use this facility to transport Atmos
soundtracks within the DCP – this data is then passed on to the audio processor from the playback server
for onward processing and playback. The same mechanism can also carry other payloads, such as the
data necessary to drive moving seats, or lighting effect systems and synchronise them with the playback
of the film. )
- SMPTE DCP supports ‘markers’ – these are points in time within the com-
position representing places where exhibitors might want other things to happen.
( For example, attaching an automation cue to bring up the houselights when the credit roll starts. With
markers it will be possible to program theatre management systems to insert these cues completely auto-
matically, reducing effort, eliminating the possibility of human error and working towards total automation.)
- supports ‘Extended CPL Metadata’ ( the ability to carry more information about the
content than is possible at the moment, such as the name of the distributor, the audio format (5.1/7.1/At-
mos etc), the aspect ratio, the light level at which it was mastered. Right now with the Interop DCP the
only way to carry this data is in the title of the content, following the “DCP naming convention” )

There is a small difference in how the binary data is organised between InterOp and
SMPTE *.mxf track files and therefore image and audio track are NOT interchangeable between
InterOp and SMPTE DCPs. It is also important to note that if the essence data is encrypted, you
must decrypt and re-encrypt essence (image and audio) data when converting from InterOp to
SMPTE.

DCP IN DAVINCI RESOLVE:

The fully functional version of easyDCP operates via licensing modules
purchased from http://www.easyDCP.com (info@easyDCP. com) and every new DaVinci
Resolve system (server) needs its own license and specific certificates for DCP and
KDM generation and for playback of DCPs.
The Server Certificate Set generated for each DaVinci Resolve will contain
files based on the purchased modules and specific Resolve server hardware.
 for in house purpose
DCP - learning guide -

Example:

> License:
The License is used to activate the purchased modules on a specific hardware server.

>Server Certificate:
Each DCP render (referred in the industry as an ‘Instance’) using encryption or decryp-
tion has an individual server certificate. This certificate is required to be able to receive
Key Delivery Messages (KDMs), which unlock encrypted DCPs.

> Signer Certificate:

A Signer Certificate is used to sign certain files within a DCP package and/or Key Deliv-
ery Message (KDM) to verify which authority generated the DCP instance.

EasyDCP Color Management :

The Color Management panel of the Project Settings has a Timeline Color-
space pop-up menu that is enabled for EasyDCP encoding regardless of whether or not
Resolve Color Management is used for the current project. You should set this to the
color space used by your current Resolve timeline. If, for example, you are grading us-
ing a Rec. 709 monitor for television deliverables but also wish to make a DCP, select
Rec. 709 Gamma 2.4 and Resolve will render the DCP with the correct Rec. 709 to XYZ
matrix.
 for in house purpose
DCP - learning guide -

EasyDCP Output in the Deliver Page:

1. Set “Render timeline as” to Single clip.


2. Choose easyDCP from the Video Format pop-up.


3. Choose the appropriate option from the Codec pop-up that corresponds to the
type (DCP or IMF) resolution (2K or 4K), and aspect ratio (native, scope, or flat) of your
intended output.


4. Set the Composition Name. This field is intended to hold a standardised name
for the DCP being encoded. You can either type a name into this field directly, or you can
press the “...” button to open the easyDCP Composition Name Generator window. An
editable Film Title field appears, along with a number of pop-up menus that let you select
various DCP attributes such as content type, aspect ratio, language of audio and subti-
tles, and so forth. As you populate each attribute, the name being generated appears at
the top of the window, and clicking OK copies the resulting Composition Name into the
Composition Name field of the Render Settings.


5. If necessary, set the desired “Maximum DCP bit rate” by either typing or drag-
ging within the field (the range is 50 to 250 Mbit/sec). If you’re not sure what data rate to
use, consult the client or distributor to whom you’re delivering the DCP.


6. There are two DCP package types you can output, determined by the “Use In-
terop packaging” checkbox: The standard package conforms to the “Interop” specifica-
tions for DCPs, which is turned on by default. With “Use Interop packaging” turned on,
however, the frame rate of your output is limited to either 24fps or 48fps, so you need to
make sure that your timeline conforms to these frame rates. 

If you want to generate DCP packages with other frame rates to match your timeline,
you need to turn “Use Interop packaging” off to generate a SMPTE-standard DCP.
This supports additional frame rates including 25, 30, 50 and 60 fps. However, SMPTE-
Standard-DCPs are not supported on all JPEG2000-based playback systems so it’s
generally recommended to use the Interop standard unless you know the player sup-
ports the SMPTE-Standard DCPs.


7. Turn on the “Encrypt package” checkbox to encode an encrypted DCP. This
sets the encoder to generate a Digest containing the keys used during encryption. This
Digest will allow you to play the resulting DCP on your system, and to generate KDMs to
allow that DCP to be played on other servers. 

Note: If you do not encrypt the DCP it can be played on any DCP player/decoder without
restriction. 


8. Set the Subtitles Path. If you have a properly formatted subtitle file, click the
Browse button to link to it.


 for in house purpose
DCP - learning guide -

9. If you’re including an audio mix in the DCP, go to the Audio section, turn on the
Render audio checkbox, and choose the number of channels in the “Render channels of
audio” pop- up menu that corresponds to the number of Audio Mixer output channels de-
fined in the Edit page.


10. Click the Browse button under the “Render to” field, and choose a location for
the resulting DCP. Make sure you pick a drive with enough room for the estimated size
of the final DCP.

KDM Generation and Management

Key Delivery Messages (KDMs) are required to allow an encrypted DCP play
on a designated projector at a particular theatre at a specified time. DaVinci Resolve is
capable of creating KDMs, which is convenient for exporting KDMs for select screen-
ings, but commercial distributors may require thousands of KDMs. Fortunately, easyDCP
allows you to use external Distribution KDM (DKDM) utilities to generate KDMs for your
clients, so you don’t have to tie up your Resolve workstation with this task.

Publishing Your Encrypted Digital Cinema Package

While you can play your encrypted DCP on the same DaVinci Resolve system
that generated it, if you wish to publish the DCP so other players can decode and play
you need to generate a KDM to send to the player. The user of the other player, or play-
ers, will need to generate a Server Certificate for each of their players and send this to
you so when you generate the KDM it will be just for those players.

Select the DCP in the Media page Library. Right-click and select Generate
KDMs. From the pop-up select the location of the Server Certificate file if the KDM
is for one player, or folder for multiple players. Set the start and end dates that the
KDM will be valid for, an output folder to place the KDM, and then select Generate.
You can now send your DCP and the KDMs to the player you authorised. The user
there will import the KDM and the DCP will play between the start and end dates.

Playing Your Digital Cinema Package

To play a DCP you’ve output from Resolve, use the Media page to add it to the
Media Pool and edit it into a timeline like any other clip.
Decoding the JPEG2000 images embedded within the DCP in real time is com-
putationally intensive. If your system is underpowered you can reduce the decoded reso-
lution of the files by selecting Half or Quarter Resolution Decode from the File > easyD-
CP menu. A smaller, less bandwidth-intensive version of the JPEG2000 files will be de-
coded by discarding some levels of the wavelet stage inside the decoder, which will di-
rectly increase the playback performance.
 for in house purpose
DCP - learning guide -

Playing Third Party Digital Cinema Packages

To play a non-encrypted DCP simply select the DCP in the Media page like any
other clip. To play an encrypted DCP from a third party you first must publish your Server
Certificate. They use the certificate to generate KDMs for their DCP to play on your
DaVinci Resolve system. From the File menu select easyDCP, then select Export Server
Certificate, and on the pop-up menu choose a location to save the file. Send this to the
third party for KDM generation.

When you receive a KDM or a Digest for an encrypted DCP you must first import
the file into your DaVinci Resolve system. Using the File, easyDCP menu select Import
KDM/Digest, and then select the file. Then simply select the encrypted DCP in the Media
Page Library to play.

KDM and TKR:

Part 1 - How do Digital Cinema Keys work?

The goal is to send a DCP to a theatre and get it there in such a way that no bad
person can pirate it. So we want to scramble or encrypt the digital DCP bits so you need
a special decoder ring (a key) to enable the movie to play it. It is secure enough that we
can freely send encrypted DCP’s to anyone and guarantee that they can’t read them or
pirate them. In fact, we can even include the keys in the same package and still guaran-
tee that they can’t read them. How does that work?

SIMPLE DCP ENCRYPTION

A DCP key (about 1Kbytes) is used to encrypt or scramble the DCP using magic math.
The security industry believes that if you get an encrypted file you will NOT be able to decrypt it
without the key. But if you do have the DCP key you can easily decrypt the file and recover a
copy of the original DCP. The DCP Key(s) are symmetrical – the SAME key that is used to en-
crypt the DCP is also used to decrypt the DCP at the theatre. Much like your home key, it locks
the front door when you leave and the same key unlocks the front door when you return home.
 for in house purpose
DCP - learning guide -

Public Key Infrastructure (PKI)

Each screen (or server/ projector) has two special keys - a Screen Public Key
and a Screen Private key. The public keys are just that - public! They can be freely given
out, but are useless without the private keys - which are well protected and safe inside
the server. The screens share their public keys by giving out the “screen CERT” which is
their Public key. Which is why it is necessary for a screen to provide its CERT when it
wants a KDM. So the key distributor uses the target Screen Public Key and the
DCP Key to create a KDM. The screen server takes the KDM using its own Screen Pri-
vate key to decode the DCP Key to enable play of the movie. The server does this inside
very well protected electronics called the IMB - Integrated Media Block - so that no one
can get to the unencrypted DCP Key or the DCP itself or the Screen Private Key.
A screen shares its Screen Public Key to a key distributor. The key distributor
sends the encrypted DCP out to the screen along with a KDM for that screen. The
screen uses the KDM and the Screen Private key to decrypt the KDM to get the DCP
Key and the DCP movie comes magically out for making a wonderful movie on the
screen. For DCP protection it also means each screen/server needs its own private key
so it gives better control of distribution. ANYONE can get the encrypted DCP and ANY-
ONE can get the KDMs, but only those with authorised private keys, i.e. the screen/
server, can play it.
 for in house purpose
DCP - learning guide -

We send the DCP encrypted on hard drives and over satellite to anyone that
MIGHT want to play or book the movie and the KDM is customised for each screen
server and sent to that screen server.

Part 2 Delivery of KDMs:

A KDM facility (a facility that makes and delivers KDM messages) is a very trust-
ed entity. They hold the unprotected DCP key. Something that should be very, very well
protected. The KDM facility needs to keep a good list of all the possible screens and
locations. They need the CERTs for each screen server, the owner, and the capabil-
ities of each screen (5.1, 7.1, screen brightness, 3D 2D, HFR, etc.) This list is a
Trusted Device List for the KDM facility. They need to have high confidence that these
are valid CERTs that go to a real authorised screens.

Next comes the booking. The studio builds a list of which screens / locations are
authorised to play the movie and for how long. And the KDM facility needs to know all
the screens in a complex since studios send KDMs for all screens in a complex.
Getting the right keys to the right theaters is a bit more difficult. The KDM facility
determines which sites need to play a movie and generates all the keys/KDMs for all the
authorised CPLs//screens in the complex. A twenty-plex could get 20 to 60 keys or more!
The most popular way of sending KDMs is through the email system. The KDM facility
takes all the keys for a particular site and “zips” them into an attachment to send to the
theatre manager. The theatre manager opens the attachment, unzips the keys and puts
them on a USB thumb drive and “sneaker net” the drive to the TMS (Theatre Manager
System). The TMS “ingests” the keys and appropriately distributes them to the appropri-
ate servers. Or the theatre personnel has to walk the USB thumb drive to each screen
(playback server) and ingest the KDM for the specific CPL version (5.1, 7.1, open cap-
tions ….)

Part 3 - A better way of delivering KDMs - TKR

The first version of the DCI specification required a Fax Modem line into each
server! This was for KDM delivery - it did provide direct access for delivery of KDMs, but
there were so many problems - getting phone lines into theatre booths, unplugging the
lines, etc. It has a failure rate of over 15%. But today every booth has a network be-
tween all devices and at some place a bridge to the web. A group of very smart (and
nice) folks devised a method of delivery called “Theatre Key Retrieval” (TKR) that
meets a set of very strong criteria for automated key delivery.

The KDM facility makes a “hidden” website (something like http://www.greatstu-

dio.com/ KDM/ 1c326ea0-77fb-11e2-b92a-0800200c9a66 - an address that is not possi-
ble to guess). The KDM facility provides this address to the maker of the CPL and the
address is included in the CPL. The KDM facility puts ALL the keys for that movie in the
hidden website with identification of which KDM goes with which server.
 for in house purpose
DCP - learning guide -

The Theatre gets the CPL and finds the address of the hidden website and goes
out to find the hidden website and “gets” the KDMs that it needs for its servers.

So the advantages of this approach:

1. The KDMs can be automatically generated (they already are automatically generated) and posted to the
hidden website. If you need to “send” a new key, just post a new one and the theatre will look for it.
2. The theatre is well protected against viruses or someone breaking in the security wall - it is a “get” sys-
tem and not a “push” to the theatre solution.
3. It meets all the requirements specified above.
CPLs need to include the hidden website address. Both Interop-DCP and SMPTE-DCP have been tested
and been shown to work (and not cause a problem for software in the field).

ABOUT DCP PROJECTION: (from forums)

All digital cinema projectors have one preset for Flat (1998×1080) and Scope (2048×858).
In a constant height theatre, the resolution, the side masking and zoom changes when projecting
the two presets. The flat preset has 1080 pixels in height, the Scope preset has 858 pixels in
height, so when going from Flat to Scope the lens zooms and the masking widens.The pixels
beyond 858 would have hit above the screen if they were not masked in the projector.The flat
preset has 1998 pixels in length, the Scope preset has 2048 pixels in length.When screening
2048×858 in Flat the pixels beyond 1998 is masked in the projector and the image is letter-
boxed. That is why you have to scale 2048×1024 down to fit the Flat preset or crop it
to fit the Scope preset even though 2K is 2048×1080.You should always avoid black padding/
letterboxing, it kills the contrast.That is why 1998×080 or 2048×858 should be your target.
 for in house purpose
DCP - learning guide -

DCI Requirements for Digital Cinema Packages -short:

4:4:4 , 12 bits per sample,

X’Y’Z’ color space,
DCI-P3 Color gamut

Image sizes:
• 2K scope: 2048 x 858
• 2K flat: 1998 x 1080
• 4K scope: 4096 x1716
• 4K flat: 3996 x 2160

DCP Audio Format:

• Files must be in reels matching the image file reels
• The sampling clock rate shall be exactly 48,000.000 Hz or 48 KHz at exactly 24 fps •
Reference level shall be -20dbfs

DCP Format
The DCP files shall be wrapped using the MXF Interop/ SMPTE for DCI. The DCP shall
consist of the following types of files:
• Assetmap
• Vol Index
• Packing List (PKL)
• Composition Playlist (CPL)s
• MXF Wrapped image track file(s) • MXF Wrapped audio track file(s)

The image and audio files shall be in sync.

Image files shall be wrapped from the JPEG2000 compressed image files or equivalent.
Audio files shall be wrapped for 5.1 sound with the following channel assignments:

• Channel 1—Left
• Channel 3—Center
• Channel 5—Left Surround
• Channel 2—Right
• Channel 4—LFE
• Channel 6—Right Surround
If two (2) channels of sound (Stereo—LtRt) are required, a separate CPL shall be used
within the same DCP.