Knut Erik Evensen – KESE

With Easydcp 2.0 you can make DCPs from Quicktime files like 1920×1080 ProRes 422 HQ. If you do not have a fast RAID setup this can speed up the process. But if the source files need deinterlacing and scaling you still need to deinterlace/upscale the source to either 1998×1080, 1920×1080 or 2048×858 in Adobe After Effects or MPEG streamclip. To make a ProRes 422 Quicktime you can use use Adobe Premiere Pro CS5.5 on a mac like they do in these Adobe Blog posts, you also need to buy Apple Motion or Final Cut Studio Pro X in the Apple App store to get the ProRes 422 codec.

When you make a 25 FPS SMPTE DCP from a Quicktime file Easydcp stores the audiofiles in a directory of ypur choice and you can convert these files to 24 FPS audio files and use them to make a 24 FPS INTEROP DCP that will play on older hardware.
If you copy the sound files from the directory to a new directory and open them with Wave Agent you can slow the sound down to 46080 hz (24 FPS). When you open these in software like Audacity 1.3 in a 48 khz project you will resample them to 48000 when you save them to 24 bit 48000 wav files. These files can then be used together with the 25 fps Quicktime in a 24 FPS INTEROP Easydcp project to make a 24 FPS INTEROP DCP. With a 24 FPS INTEROP Easydcp project you can also add Interop subtitles. You can also use this trick to convert 23.98 fps sound files to 24 fps, just use 48048 as the sample rate in Wave Agent.

You can play back DCPs with Stereoscopic player and jpeg2000 decoder or with Easydcp Player.

These can be used to check the DCP for some common problems or to screen the DCP. You could use the trial version of Easydcp Player (It plays only the first 15 seconds of a DCP) to check if the DCP loads correctly and even do a file hash check. With stereoscopic player you can check for audio sync and see if the colors seem OK (The trial version plays DCPs for 5 minutes at a time). But if you have offsets on audio and video, interop subtitles or the DCP is encrypted you need Easydcp player.

Some limitations:
-These players only have 8 bit color, not 12 bit.
-They need fast computers, Stereoscopic player can play a DCP on a Intel i7-2600 CPU PC. Easydcp player need a faster computer but seems to have higher quality decoding and it can use a dedicated Geforce 570 or better NVvdia CUDA graphic card as a jpeg2000 decoder.
-Your DCP will look and sound different in a movie theatre than on your typical PC.

Easydcp player could be used to play DCPs originating from 8 bit HDCAM since the source is 8 bit. For smaller venues at film festivals without digital cinema equipment this could be an alternative to HDCAM player rental. It could also be used to play 25 and 30 fps SMPTE DCPs made from video sources on older digital cinema equipment that do not support these. The money a festival saves on HDCAM player rental for some years could pay for a non profit license bundle of easydcp creator and easydcp player and computers. When transferring HDCAM to DCP you could replace the stereo mix with a 5.1 mix. Or add subtitles. Easydcp creator does not yet support subtitles on SMPTE DCPs, but it is possible to do it with third party tools like opendcp. Or the subtitles could be added directly to the movie. But there is the added logistics of asking for permission to transfer each movie, and if the transfer is done incorrectly you could loose quality.

You can use Cinecanvas XML subtitles in a 24 fps Interop DCP in Fraunhofer Easydcp.
Example test.xml file:
test.xml (right click and use save as)

Some tips:
Remember to save as UTF-8.
Use a text editor like textpad.
http://www.textpad.com/
or notepad ++ http://notepad-plus-plus.org/
You need a font that is not bigger than 640 kb according to the Texas Instruments Cinecanvas Specification posted here:
http://reduser.net/forum/showthread.php?33118-Digital-Cinema-Package&p=709332&viewfull=1#post709332

Place the font with the xml subtitle file.
Easydcp will place it in a special subtitle directory on the DCP.
You need your subtitles in the cinecanvas XML format with 24 fps timecode.
VAlign=”bottom” values of 14 and 8 works with Flat and Scope movies, (10 and 15 are also often used) but not movies which uses top and bottom black padding (letterboxed full container movies in Scope, use 2048×858 for Scope).
If there is only one line it is at 8.
You need to generate a UUID for the subtitleID
http://www.famkruithof.net/uuid/uuidgen

You can not use offsets on subtitles, if your sound and video has an offset
you either have to sync the timecode to that offset or delete frames from the video and sound.
(or choose a later frame as the starting frame, see comments)

A good idea is to split the DCP in reels.

Some subtitles tools:
http://www.belle-nuit.com/subtitler/
Can open many formats and output Cinecanvas XML, use Find and replace in a text editor to
change the output so it looks like the example XML. If it puts single lines at a middle height, change it to 8.

Michael Cinquin Subtitle Tools
Lots of tools, example 1: export Final Cut XML sequences to Belle Nuit subtitles and vice versa.
Example 2. Export Final Cut XML sequences directly to Cinecanvas XML.

Doing versioning.
When working with subtitles you may need to make fixes or different languages.
To do this faster I extract the jpeg2000 image sequence from the video MXF from the first DCP with asdcp-test
asdcp-test -x picture xxx.j2c.mxf
You can also use Fraunhofer Easydcp Player to extract the image sequence.
In Franhofer Easydcp Creator choose this extracted j2c image sequence as your picture trackfile, right click it, and choose “Color processing” – “bypassed”. And untick Options – Image Processing – Use JPEG2000 Codestream (*.j2c) Reencoding.
Also remember to untick “Automatic letterboxing/pillarboxing” in Options if your movie is Scope, or use larger number than 8 and 14.

This is a good resource for info on 5.1 mixes: http://duc.avid.com/showthread.php?t=87830

Among his links is this “5.1-Channel Production Guidelines” from Dolby:
http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/L.mn.0002.5.1guide.pdf

For film work, pink noise at reference level should produce a sound pressure level (SPL) of 85 dBC for each of the front channels (left, center, and right). Each surround channel should produce a sound pressure level of 82 dBC (the lower surround level is specific to film-style mixing rooms).
For television work, pink noise at reference level is typically set to produce an SPL ranging from 79 dBC to 82 dBC for each of the main five channels. The lower reference level for television is due to the lower average listening levels used by the consumer (typically 70ñ75 dBC). Since the reference listening level used can dramatically effect the balance and intelligibility of the mix, it is important to consider both the level at which a program was mixed, as well as the typical listening level in the home for the same program.

Often a stereo sound mix for TV uses limiters and compressors and has less dynamics than a theatre mix. This is because it is mixed at a 79 dBC per channel instead of 85 dBC per channel.
A theatre mix is also mixed with 20 db of headroom.
This makes it hard to convert a TV mix to a 5.1 theatre mix. The best option is to go back to the stems/sources that was used to make the TV mix and mix it again at 85 dBC per channel.
I recommend having all theatre mixes done at professional film mixers that uses the 85 dBC per channel standard.
In Norway I have worked with Storyline Studios and Aurora filmlyd which both uses this standard.

Updated for clarity.
This is an example of a best common practice VFX color space workflow used at Sony Dreamworks:
http://opencolorio.org/workflow/spi_vfx.html

Basics:
Color Space (Wikipedia)
3D LUT (Wikipedia)

This book has some examples of DCP color space workflows:
Color and Mastering for Digital Cinema (Digital Cinema Industry Handbook Series)

In this book the author Glenn Kennel says that the standard DCP workflow is to retain the film look on the digital version of the film instead of making it more saturated. A similar 3D lookup table (LUT) (film stock emulation 3D LUT) that was used to grade the film on a P3 projector can be used when making the DCP.

Arri Alexa whitepaper:
In this Arri Alexa whitepaper Arri have some suggestions for color space workflows using their Alexa Digital Video Camera. They suggest you can use a bleach bypass LUT on dailies if that is the look you intend to use on the final product.
They say it is possible to use a film stock 3D LUT for grading Alexa footage, but they warn that each channel of a film stock has a different gamma so you need to compensate for that.

Afer Effects Color Managment white paper:
Another resource is the color management workflow whitepaper for Adobe After Effects, Adobe suggest and describes how you can add a adjustment layer with a theater preview when making a image sequence intended for DCPs.

Some thoughts on low budget DCP color space workflows:
In “Color and Mastering for Digital Cinema” the author explains how the 2.6 gamma 12 bit, DCI XYZ color space standard of DCPs was made for the standard Digital Intermediate workflow of 2004 which uses P3 projectors for grading.

In 2011 you have more options when choosing a workflow: You can order LUTs from http://www.lightillusion.com/ cheaply and use them on digital intermediate files with software like Fraunhofer Easydcp +, After Effects and Davinci.

You can shoot video with lenses on SLR with settings that mimic 35mm Log curve (Technicolor Cinestyle)

Or use easy to use film look software (Magic Bullet Looks 2.0) that can emulate different film looks.

It is now cheaper to use similar workflows to the ones big budget productions used in 2004. But big budget post production workflows have access to a P3 reference grading projector. This is still expensive. Even if low budget productions intend to directly convert to DCI XYZ from Rec 709 and retain it´s white point, colorspace and gamma the greys could seem darker and your picture could seem more washed out because of less luminance (14 ftl) in the cinema than on your monitor. A computer monitor or TV could have more than double the luminance of cinema projection. That is why what could look like a film look on your monitor could look less like film when watched in the cinema.

In the Alexa whitepaper they suggested watching material shot with the Alexa with a bleach bypass LUT and this attraction to the desaturated look from 35 mm seems to be common in DCP color space workflows.

This attraction to the look of 35 mm is also seen in the choice of low budget SLRs that have film camera optics.

For a good looking DCP that looks more like Hollywood films shot on expensive cameras and graded on p3 projectors it seems low budget filmmakers should also try to mimic the projector based grading workflow.

Even though they can not afford to view the material in the full dynamics of 12 bit or in P3 color space, a projector that is calbriated for 14 ftl could help. http://www.lightillusion.com/ has examples of calibrated grading rooms. Other suggestion could be to grade in LOG or applying a S Curve. The goal is to get contrast rich pictures but also keep lots of details in the greys.

Digital has the possibility for much more saturated colors than 35 mm and if you want to see what DCI P3 is capable of you can go for a distinct non 35 mm look and use the saturated colors from an Arri Alexa. You could do this even though the footage is seen on displays with less saturated colors most of the time in the production and post production of the film by using 3D LUTs. This may be the future of DCP color space worflows, even though most worfklows now mimics the look and feel of desaturated 35 mm.

UPDATE: In Easydcp 2.0 you can make DCPs from (Prores 422) Quicktime files.

These DPX sequences can be used with Fraunhofer Easydcp to make DCPs.
Thanks to colorist Christian Berg-Nielsen (christian at sementbetong.no) at Sement & Betong Post Production.
If you need to have your film graded for digital cinema send him an email.

Step by step guide to make DPX sequence from a progressive REC.709 source. If your source is deinterlaced a quick solution is to deinterlace the source with MPEG Streamclip

Drag a clip to a project.

Change the project settings to 16 bit.

Right Click and interpret footage. Choose correct frame rate. Seperate Fields = off, The Audio needs to be corrected also.

Drag the clip to a new composition.

Right click the new composition, change resolution to 2048×858 for scope and 1998×1080 for flat.

Double click composition, right click on layer, choose transform and Fit to Comp Width (or Height).

Choose Composition – Add to Render Queue

Change the Output Module to “Format: DPX”

Choose Color Management , Cineon Settings, Full Range DPX

and choose a output directory with enough free space. (approx. 800 GB per hour)

Then Render.

The best common practice is to deliver the DCP in a Cru Dataport DX-115 harddrive caddy. These can be loaded directly in to the Doremi DCP-2K4.

You also need to support USB 2 so you have to get the Cru Dataport Docking Station DX115 USB/sata:

and a harddrive like the Western Digital Caviar Blue 500 GB 3.5 Inch Hard Drive :

To transport these safely you could use a Pelican 1400 Case:

Together these could look something like this DCP I made for a special screening of 11 short films.

Or like this:

If the DCP will not be transported much a ordinary USB 2 harddrive with a power supply will do. Like this one:

The physical medium specification for DCPs is a USB 2 external harddrive formatted as MBR Ext 3. Doremi and Sony support NTFS, but the correct format is the linux format Ext 3 with a inode size of 128 that is read and executable only. Source: ISDCF Disc File Format Note: Even though there is a standard that says that there should only be one Assetmap on each harddrive, it is common practice to include different DCPs in different folder on the harddrive. But do not use folders inside folders like this: /DCP/MYFILM. The name of the folder will not be shown when ingesting the DCP on a Doremi.

To do this I use a Ubuntu Linux computer with a carrier for DX115 harddrives.
¨

Or USB 2 with USB 2 harddrives.

You can copy the DCP to a inode 128 read and executable only EXT3 harddrive in many ways. This is one way:

Install gparted .
sudo apt-get install gparted

Run gparted
sudo gparted

Make a dos partition table.

Format the drive as ext3.

Note the name of the partition (like sda1) and format it again in terminal with -1 (128 inode). The -j is for journal (ext3).
mkfs.ext2 -j -I 128 /dev/sd..

Mount the drive with sudo mount /dev/sda1 /mnt/disk

Make a directory for the DCP sudo mkdir /mnt/disk/dcp

Then use sudo chmod -R 777 /mnt/disk/dcp so everyone can write to it

Then find the dcp on the local network from a shared windows folder or on a harddrive and copy it to the directory.

Then make it read only and executable. chmod -R 755 /mnt/disk/dcp

Before you check your DCP in a movie theatre, you need to know the specification of the movie theatre you are using.

Here are some question you could ask:
-Are the scope and flat presets without cropping?
35 mm cinemas often cropped much of the picture to mask the fact that the projector is tilted. Even with lens shift, some digital cinemas still has tilted projectors that need presets for scope and flat with cropping to fit the masking on the screen. This hides the fact that the shape of the picture often has a trapezoid shape. For a test screening of a DCP you could ask for a preset on the projector without cropping, this way you could watch for artifacts on the edges of the picture. You should also consider that the film will be shown in theatres with cropping, so you need to use safe zones. Also consider that a 1920×1080 DCP could reveal that the projector is tilted and the picture has a trapezoid shape, so scaling and cropping 1920×1080 material to 1998×1080 (Flat) when you make the DCP should be an option. If you want to watch the movie with better perceived contrast you should use the standard presets.

-Is the peak white luminance 14 Ftl on the screen?
35 mm projector lamps are often not adjusted for age by the cinemas or just run beyond their warranty which give you a too dark picture. With digital cinema there are other problems, some cinemas has the Sony 3D lens attached all the time to their Sony 4K projectors. When the 3d lens is attached the resolution is less than 2K and light output is lowered. But if they took it off they would have a same problem that other 3D cinemas has and that is that the picture is too bright in 2D because 3D needs lots of light. Sometimes this could just be that the lamp is set too bright by mistake, but it could also be a result of the high gain of the 3D silver screen. A quick fix in a too bright theatre is to zoom the lens out and scale the picture in. You will loose resolution but have less light. Another quick fix in a too bright silver screen theatre is to move away from the center seats which has the most light until you get seats which has approx. 14 Ftl. If the screen is curved there will not be as much light loss away from the center, but all silver screens will have a hotspot right in front of where you are sitting. Another option could be to install a lamp with lower strength. Or make two different version of the film, one in flat and one in scope (like movie trailers are made), often the scope preset has less light.

To get more info on a what specification too look for in a digital cinema theatre this book could help.

In book the author says that there should be readings of the light in different locations in the cinema when the projector was calibrated. If the technical manager of the cinema where you want to test screen your DCP do not know if the screen is 14 ftl you could ask for the readings that was done when the projector was calibrated. Hopefully with a fresh lamp these reading will still be accurate.

Another tips from the book is to look for a projector with a better than DCI spec contrast ratio, if the screen has a worse contrast ratio than the spec you will not notice some artifacts that can be seen in better projectors. You should also look for a projector with a good intra frame contrast ratio, if the contrast seems muddy a quick fix is to clean the port glass.

This year I converted all films that was not 35 mm or DCPs at the Films from the North 2011 section at Tromsø International Film Festival 2011 to DCPs. In total I made 33 DCPs (about 10 hours) in two weeks. I was also responsible for screening the DCPs at the film festival. I checked the sound level on each DCP in the cinema theatre and wrote down what level they should be played back at. I also wrote down what audio preset should be used (5.1 or 2.0 with Pro Logic decoding). If you are converting to the DCP format for screenings where you will not be present I recommend using standard cinema level 5.1 sound mixes.

Here are some tips if you are thinking about converting a film festival programme to DCPs:

I recommend a speedy jpeg 2000 encoder like the one included in Fraunhofer Easydcp.

I recommend using batch processesing when making DCPs of short films, I used Fraunhofer Easydcp + to make 1-5 DCPs each night.

Have a tested worfklow before the festival films arrives. It must cover both interlaced and progressive material. Standard definition and high definition. Stereo and 5,1 sound. Material that has subtitles outside the safe zone and material that does not use a standard resolution. . .

Use different quality control stages, I recommend two computer screens so you can look at the material fullscreen 1:1 . I recommend using the RGB parade scope in Apple Color to check how the source is color corrected/graded. I recommend checking the DCPs in a movie theatre, preferably the screen you will screen the film at or one that is suited.

Find out how much the picture is cropped at the theatre where the films are screened. Also check for a 1920×1080 1.77 preset.

If using 25 fps or 30 fps DCPs, check if they work in the theatre that will be screening the films.

If using DCPs with stereo sound make sure the theatre has Dolby Pro Logic 1 or 2 decoding or something similar to get the speech in the center channel. If a stereo soundtrack is played in stereo everyone on the right side of the screen will hear speech coming from the right and those one the left will hear it coming from the left.

In total I made 33 DCPs (about 10 hours) in two weeks. I always had 1-5 DCPs being made each night that I tested the next day. I could convert 1,5 hours from one day to the next with easydcp + on a quad core computer. Some people uploaded their films as Prores quicktime to a FTP server or used Filemail Corporate instead of delivering on Digibeta or HDCam. This helped speed things up.

When converting a 4444 or 422 YCbCr prores to a DPX sequence in Apple Color blacks below 0 and whites above 100 in Apple Color is lost. This is described in the Using Red Media with Final Cut Studio 3 Whitepaper. This information can be seen (if present) on default RGB Parade Scope (with the broadcast safe setting in project settings set to off). It can be preserved with the Highlight and Shadow sliders in “Primary in” or with changing the Luma curve.

Note: Check that the DPX sequence from Color matches your monitor. A possibility is to export RGB safe material from Color in the Prores 4444 format and then use some other software to make the DPX sequence .

Update: To export a DPX sequence from After Effects follow this step by step guide

The possibility to adjust the YCbCr to RGB conversion so no blacks are crushed and no whites are clipped is great. This is not possible with software that does the YCbCr to RGB conversion automatically.

With both Easydcp and Cineasset the safe choice is to only accept 422 and 4444 YCbCr material for conversion to DCPs that stays between 0-100 in Apple Color or RGB material.

For more technical info on converting YCbCr material to RGB:

5dtorgb tries to improve on converting the YCbCr video files from DSLRs to Prores or DPX sequences. Interesting stuff from their website: “…Many programs use QuickTime internally to perform YCbCr to RGB conversion which, according to our testing, does only a mediocre job….” “…It also recognizes Canon’s full range 8 bit YCbCr values (0-255), avoiding clipping and the resulting loss of picture information… ”

From the Final Cut Pro User manual
”
…
About Illegal RGB Levels

Most video formats are recorded, edited, and transmitted in Y′CBCR
(component) color space. Although you may never intentionally convert
your footage to RGB color space, it is fairly certain that it will
become RGB at some point—most likely during display on a television or
monitor. Video is almost always converted to RGB color space within
display devices.

When you work in Y′CBCR color space, levels are either described in
their native color space or they are referred to as RGB-equivalent
values, meaning the values of your Y′CBCR video signal when it is
eventually converted to RGB.

Illegal RGB levels are generally caused when certain combinations of
luma and chroma levels (in Y′CBCR color space) are converted to RGB
color space. Even if both luma and chroma levels are legal in Y′CBCR
color space, the combination of these two may cause illegal RGB
levels.
…
”

Quote from the wikipedia YCbCr article :
“…Since the equations defining YCbCr are formed in a way that rotates the entire nominal RGB color cube and scales it to fit within a (larger) YCbCr color cube, there are some points within the YCbCr color cube that cannot be represented in the corresponding RGB domain (at least not within the nominal RGB range). This causes some difficulty in determining how to correctly interpret and display some YCbCr signals…”