Common questions regarding ARRIRAW, MXF/ARRIRAW and High Density Encoding (HDE)
Common questions regarding ARRIRAW, MXF/ARRIRAW and High Density Encoding (HDE)
ARRIRAW is the raw format of ARRI's digital cameras. Its file extension is *.ari or *.mxf.
The sensor readout of our ALEV3 and ALEV4 sensor is a raw Bayer frame provided as uncompressed, 12 bit log ARRIRAW data. To turn ARRIRAW data into RGB images, the files or clips need to be processed (de-bayering and color processing).
MXF/ARRIRAW is ARRIRAW with a wrapper. All frames are wrapped within a MXF container. ALEXA Mini and AMIRA use this format to write ARRIRAW to CFast 2.0 media. For ALEXA Mini LF and ALEXA 35 we switched completely to MXF-wrapped ARRIRAW.
Good to know:
The resolution of ARRIRAW is not fixed to a specific number! It depends on the camera and its selected sensor mode. ALEXA 65 records the biggest ARRIRAW frames to date with a resolution of 6560x3100 pixels.
The document ARRI Formats and Resolutions Overview provides information on all recording modes (including our other recording formats ProRes, DNxHD and MPEG).
This, again, depends on the type of ARRIRAW you want to record. 16:9 2.8K ARRIRAW records at 1.34 Gigabit per second at 24fps; Open Gate on ALEXA 65 with 5.89 Gbit/s at 24fps.
See our Formats and Data Rate Calculator for more!
The ALEXA or AMIRA series cameras do not offer a special HDR mode. The camera's ALEV3 sensor provides an unrivalled 14+ stops of exposure latitude (measured with the ARRI Dynamic Range Test Chart) without any special capture mode.
Equally for ALEXA 35: The camera's ALEV4 sensor provides 17 stops of exposure latitude (measured with the Xyla chart and ARRI Analysis method) without any special capture mode."
That's "HDR since 2010" - when ALEXA with ALEV3 sensor was first introduced and continued with a greater dynamic range with ALEV4 in 2022.
Yes, it is. We've successfully submitted two SMPTE Registered Disclosure Documents (RDDs) on ARRIRAW:
RDD 30:2014: ARRIRAW Image File Structure and Interpretation Supporting Deferred Demosaicing to a Logarithmic Encoding
RDD 31:2014: Deferred Demosaicing of an ARRIRAW Image File to a Wide-Gamut Logarithmic Encoding.
These documents are available from the Society of Motion Picture & Television Engineers (SMPTE)'s website at https://www.smpte.org.
ALEXA 35, ALEXA Mini LF, ALEXA LF, ALEXA SXT, ALEXA XT models and ALEXA Classic cameras with an XR module upgrade as well as ALEXA 65, ALEXA Mini and AMIRA can record ARRIRAW footage internally, without the need for an external recorder.
ALEXA Classic cameras (with the SxS module) cannot record ARRIRAW internally, but use a special protocol called T-Link to output ARRIRAW via HD-SDI. ARRIRAW T-Link output requires an onboard recorder that was certified for ALEXA ARRIRAW (T-Link) recording:
The ARRIRAW T-Link (Transport Link) has been developed to utilize a standardized HD-SDI interface for sending raw data from the camera to an external recorder. The camera packs the ARRIRAW data into a 12 bit log encoded RGBA HD video stream, which is sent over a standard SMPTE 424M or SMPTE 372M dual link HD-SDI connection. Since the recorder then has to recreate the individual raw frames from the stream, only certified recording equipment can be used.
The camera always outputs an HD or UHD monitoring image with optional surround view (see what is around the frame), status information and configurable frame markers.
The ALEV4 sensor offers 17 stops and the ALEV3 sensor 14+ stops of dynamic range with a base sensitivity of EI 800. This dynamic range is available in both ARRIRAW and Log C encoded ProRes footage. Due to the image compression used in ProRes, however, higher compressed ProRes footage can start to exhibit encoding artifacts when the material is excessively pushed in grading, whereas ARRIRAW footage will remain clean.
ARRIRAW footage is wrapped into a MXF container or can be a straightforward file sequence. There are various software tools to generate a verified copy of the captured ARRIRAW data. We suggest to use checksums for all copy processes to ensure a perfect 1:1 copy of your data. More information about file copy and checksums you will find on our workflow pages under Copy and Backup of Camera Negatives.
The metadata contains all camera settings, production notes, optionally scene and take information, camera tilt and roll and lens metadata (if an LDS lens/camera was used). Tilt, roll and lens information is recorded for each frame individually, so it is possible to track the exact timing, e.g. when the focus pulls away from one actor to an object in the back.
That data can be accessed or extracted using ARRI Reference Tool (or our legacy ARRI Meta Extract and ARRIRAW Converter).
All up-to-date VFX tools have native support for ARRIRAW processing.
For applications without ARRIRAW support, the footage can be processed to e.g. 10 bit log *.dpx or 16 bit floating-point *.exr files using our free ARRI Reference Tool. Processing the files in native resolution usually provides a better basis for VFX work. Once the material has been composited, it can be scaled to the target format.
Provided your system can deliver the data at a sufficient speed, ARRIRAW can be edited in the latest versions of Adobe Premiere, Assimilate Scratch, Autodesk Smoke, Blackmagic Resolve and others.
Usually small-size dailies are being rendered for editorial from the ARRIRAW footage.
We strive to make the ARRIRAW format accessible to anyone. With our recent effort, to have the format specification published as SMPTE RDD (Registered Disclosure Document), we want to make sure that anyone who needs to process ARRIRAW archives in the future can rest assured that the instructions, how to do it, will be there.
The sensor in ALEXA and AMIRA cameras is covered with a mosaic RGB filter pattern, called the Bayer pattern. When an exposed frame is sampled by the sensor, it is captured as a one-channel image (as opposed to a regular RGB three-channel color image). The reconstruction of the missing color components for each pixel, based on the type and position on the Bayer pattern is called de-bayering.
With a Bayer pattern sensor, 50% of the sensors photosites are used to represent green, 25% of the photosites represent red and the remaining 25% represent blue.
For more information, please see our ARRIRAW page.
ARRI offers a software development kit (SDK) for ARRIRAW processing that software vendors can incorporate into their application. ARRI also supports vendors who wish to implement the ARRIRAW processing procedure on their own. Depending on the implementation, the following processing settings can be adjusted:
With the ARRIRAW SDK, you can also apply a custom ARRI Look File, which offers CDL and 3D LUT options.
The different ARRI De-Bayer Algorithms (ADA) control how the color channels are reconstructed.
The processing version is directly connected to the de-bayer. We’ve decided not to list it as a separate version (as we used to in the past) to avoid confusion.
Images with Log C encoding in the ALEXA/AMIRA-specific wide gamut color space provide the camera's full latitude in an unconfined color space. Log C grading is very similar to grading of scanned film negative. It is efficient and feels familiar to any colorist who is used to grading film.
Yes you can. The ARRIRAW SDK and ARRI Reference Tool can output Open EXR files with ACES color space. ARRI provided all partner companies with IDTs, so the ALEXA/AMIRA profiles are already available in most tools that support ARRIRAW data and the ACES color pipeline.
No. Going from ARRIRAW to 12 bit Log C encoded files means no information loss. ProRes is capable to hold all the information that is inside an ARRIRAW frame in ARRI Wide Gamut color space.
When viewed directly, Log C footage looks flat with desaturated colors, LogC4 might seem underexposed. To get an image with a grayscale characteristic and color reproduction that is visually correct, the Log C footage must be tone-mapped and transformed into the target color space. This conversion can be applied with a 3D Look Up Table (LUT). A one-dimensional LUT can take care of the tone-mapping, so the resulting image will at least have a grayscale characteristic suitable for display. The transform into the target color space, however, requires a 3D LUT. This type of LUT contains both, the grayscale and the color transformation.
In the grading application, the LUT is applied for the viewing/output path. The color correction is applied on the Log C image beforehand; the result of this operation is converted into the target color space using the corresponding LUT.
HDE stands for High Density Encoding. It is an encoding technique that is optimised for Bayer pattern images. ARRIRAW images encoded with HDE are approximately 60% of the original size. HDE encoding is completely lossless - when an HDE file is decoded, it is a bit-for-bit perfect match to the original file. For more information we also refer to the Codex HDE website.
HDE provides a lossless reduction of the storage size of ARRIRAW images. An uncompressed ARRIRAW image can be reduced to a HDE file size not much larger than a corresponding ProRes 4444 XQ file. Expect all the original pixel values being accessible with HDE. The reduction in size for ARRIRAW images drastically reduces storage costs and file transfer times, speeding up your workflow and saving you money.
When an ARRIRAW image is encoded as HDE, the file extension changes from .ari to .arx. The image essence is encoded, but the file header is otherwise identical.
HDE typically results in a file that is aproximately 60% of the original ARRIRAW file size. For example, ARRIRAW OpenGate 3.4K is normally 11.26 MB per frame and 972.5 GB per hour. When stored as HDE, this becomes 6.76 MB per frame and 583.5 GB per hour. HDE encoding example
HDE is designed to provide fast encoding and decoding speeds. For example, ARRIRAW Open Gate 4.5K can be encoded comfortably at 24fps on a modern MacBook Pro laptop.
HDE is resolution independent, it can be used on ARRIRAW images of any resolution.
HDE files are supported in industry standard software such as Filmlight’s Baselight and Daylight, ColorFront Transkoder, Pomfort Silverstack and YoYotta. The list of supporting software is growing rapidly.
The HDE SDK is available via our partner CODEX.
These copy tools that can be used to back up HDE footage:
A ‘cp’ command from Terminal will also copy the files correctly, but Finder drag-and-drop will not work.
Codex hosts a video that explains about HDE and the workflow options.
The zero-bytes size is expected for HDE (*.arx) files from a Capture Drive presented via the VFS mechanism. This is because the encoded size is dependent on the original image content and cannot be pre-determined.
Codex has been working with a variety of copy tool vendors to ensure they understand how to correctly copy HDE files so that they are transferred to the destination disk with their final size.