QuickTime and Final Cut Pro do not properly take advantage of the partial resolution decode capability of CineForm's Wavelet codec. To work around this, using a new option in ReMaster, it is possible to force QuickTime and Final Cut to use the partial decode capability for editing, yet maintain the full 4K source resolution for final output.
This new feature is important because Final Cut Pro - even 7.0 - will also not handle sources larger than 4000 pixels wide. Although Color 1.5 will generate 4K output, most theatrical finishing is still done at 2K. By using this technique, 4096-wide CineForm files are interpreted by QT and FCP as being half that width - in this case 2048.
This technique does require a 4K RAW source. In this example we will be using 4K files from a RED camera. that will be converted into 4K CineForm RAW files.
Another important part of the color workflow is to use FirstLight (part of Neo 4K) and the controls in its preference pane to achieve non-destructive control over color adjustments and debayer algorithms up until the final sequence is rendered. This means that all color adjustments applied in First Light, including white balance, saturation, lift/gamma/gain, even 3D LUTs, are all non-destructive, and all are applied to your clips in real time, until you choose to perform a final render.
Conversion to 4K CineForm RAW
Open the source clips. We started with the 4K RAW preset, and then selected the “Processing” tab and unchecked the “Enabled” box for the Image Size Change. This will keep the clips in their original 4K size.
In the “Output” pane, check the box next to “Wrap RAW as half size”. This will keep the source in the original 4K RAW format, but the movie declaration in the QuickTime wrapper will be half (in each dimension) that size. For our 4096x2048 clips, QT will interpret them as 2048x1024. For clips in 4096x2304 you would end up with the 16:9 format 2048x1152.
Convert the clips to CineForm. Although this does currently require the RED SDK to perform a full debayer, ReMaster extracts the original bayer pattern data from the decoded stream. This RAW stream is then compressed as a CineForm RAW stream and written out in a QuickTime movie. Since no extra processing such as color changes or image resizing is done, the full dynamic range of the original data is present. The conversion process is slightly faster than generating a 2K 444 format file from the RED material.
Sidebar on back-sampling to CineForm RAW from R3D Output: We'll get slightly geeky on you for a moment, but it's important to realize that back-sampling to uncompressed RAW extracts the original Bayer samples from your R3D recording prior to the Red DeBayer process. To understand this, you must first remember that a Bayer sensor is a grid of R, G, and B sensor sites. At each cell site there is only one pixel - either R, G, or B. When a DeBayer algorithm is performed it interpolates missing color information at each pixel location so the resulting pixel site has all three R, G, and B values. So for each R cell site new G and B values are computed (interpolated) by the DeBayer algorithm. Similarly for each G cell site new R and B values are computed. The output of the Red SDK provides traditional uncompressed RGB as computed by the Red DeBayer algorithm. Second, and this is important, a properly implementated DeBayer algorithm never changes the value of the actual pixel recorded at each cell site, it only calculates the missing values. This means that if you back-sample an RGB image to RAW (that was created by a DeBayer process) you are throwing away the computed values, but you retain the originally recorded pixel values (because the DeBayer algorithm did not alter them). This is what is being performed by CineForm's back-sampling algorithm to restore the original uncompressed RAW data as recorded on the Red sensor. With the original uncompressed RAW data restored it is now possible to reencode into CineForm RAW. End Sidebar....
Kona and Final Cut Pro Setup
Before editing using FCP, the Kona control panel must be set up to properly display the 2K material. Open the KONA Control Panel and click on the “Setup” tab. Enable the “2K Geometries” and disable the geometries and frame rates you are not using.
Start up FCP and select the Audio/Video Settings from the “Final Cut Pro” menu. In the Summary dialog, select the 2Kx1080 output.
Import the movies into the project. When you drag the first clip to the timeline, FCP will ask if you want to change the clip settings to match the clip, click on “OK”.
Right click on the sequence in the Project Browser window, and select the “Settings...” from the menu.
The sequence settings should look like the following image.
Click on the “Advanced...” button in “QuickTime Video Settings” to bring up the compressor settings dialog box. In the compressor settings, select “Millions of Colors+” from the depth menu, and the encode quality you need.
Click on the “Options” button to bring up the compressor options, and enable the 444 encoding.
Once you start editing, you can use FirstLight to adjust the color temperature, saturation, gain and gamma. You can also apply 3D 64x64x64 LUTs to clips. Normally you'll run First Light side-by-side with FCP.
A 2008 8-core MacPro with the Kona card will play back a 4K RAW CineForm file at 2K spatial resolution at 24 fps even with a 3D LUT applied.
Final Render from Final Cut Pro
When you are ready to render the final sequence from Final Cut Pro, within First Light you can first choose the BeBayer algorithm to be applied on a clip-by-clip basis. Ignoring the Bilinear DeBayer option, which is only used for Preview, CineForm offers 5 choices of DeBayer algorithms. The primary visual difference between the DeBayer algorithms is sharpness, which is lowest for "Matrix 5x5 Enhanced" and sharpest for CF Advanced Detail 3. The Default setting is "CF Advanced Detail 1".
Or globally using the Preference Pane
You may decide a scene or clip needs more or less detail, and this can be done at final render time.
Sidebar About Using Different DeBayer Algorithms. When discussing workflow with the post team that worked on Slumdog Millionaire, we discovered they used different CineForm DeBayer settings depending on varying sharpness of the source material. Recall in Slumdog that the early part of the movie has a lot of running through the slums of Mumbai. Not all scenes were perfectly in focus. The post team experimented with varying sharpening strategies including post-sharpening after DeBayer. They ultimately determined that sharpness applied using different CineForm DeBayer algorithms provided superior results than post sharpening. You might experiment with this on your own projects. End Sidebar....
Visual Examples Using different DeBayer settings.
As a visual reference, the 4K Red source clip below (upper image) was converted to 2K CineForm 444. This was performed using the 2K 444 CineForm preset in ReMaster which does a full 4K resolution decode and DeBayer from the RED SDK followed by CineForm software performing a 3-lobe Lanczos scaler to resize the image to 2K. The image is then encoded into CineForm 444 at 2K resolution.
All other images were encoded into a 4K CineForm RAW file by backsampling the RGB image extracted from the Red SDK. The 4K CineForm RAW images used the DeBayer algorithms mentioned below with the resulting image again scaled to 2K. Images below are shown at 1:1 pixel as extracted from the resampled 2K output.
In our experience the Red DeBayer is similar to CineForm's Default "CF Advanced Detail 1" setting, whereas CF Advanced Detail 2 (and 3) are sharper. Your choice of DeBayer algorithm will vary based on your source material and your preferred sharpness.