Multi-camera production has been around since the dawn of professional video. While the overall workflow is well known, there are some nuances to be aware of if you are producing a multicam shoot for NETFLIX. All productions are unique so please think of this information as “best-practices” rather than requirements. The Netflix Post and Production Technology teams are available to assist with designing a workflow that is efficient and preserves your creative intent. Reach out to your Netflix Post Manager for further guidance.
We will break things down into the following categories:
- Data Management
- Post Workflows
The NETFLIX camera-native capture workflow | Minimum image processing and maximum quality during capture.
At its core, the camera-native workflow encompasses the following principles:
- Try to select cameras from our approved list.
- Capture with a Log transfer function, using the cameras native colorspace.
- Do not "bake in" any looks.
- RAW is preferred but not required.
RAW | Record in-camera.
All of the RAW options provide outstanding image quality. Be mindful of storage needs, especially when dealing with uncompressed RAW formats. Some cameras require a separate, dockable recording module in order to capture RAW.
Compressed Baseband-Video | record in-camera or with an external recorder.
The compressed formats listed below may not be available on some cameras. An external recorder will need to be used in these situations.
CAPTURE | Best Practices
- Recording with approved cameras.
- Recording in the cameras' native color-space.
- Ex: Sony S-Log3 / SGamut3.cine, Panasonic V-Log/V-Gamut, etc.
- On cameras that don’t offer a true “Log” transfer function, you can try using an alternative. Ex: Sony Hypergamma, Panasonic Cine-like D, etc.
- If using "broadcast" cameras, try to avoid using too many paint functions.
- In order to provide more flexibility for the color grade, use camera shading/painting functions to achieve a matching, consistent image across all cameras while being mindful to not apply any specific look that will limit what can be done in the color grading session.
- Try to select a camera that will perform optimally in your particular shooting scenario.
- If the set or venue you are working in has dynamic lighting, a “cinema” style camera may perform better and give a cleaner image than a “broadcast” style camera.
- Watch out for certain types of LED lighting. Blue LED lighting in particular can create out-of-gamut colors. Some cameras can deal with this better than others - testing with your Lighting Designer and/or Director of Photography is recommended.
- Bear in mind that a well lit area may look fine but when the camera swings around to a dimly lit area noise will become more apparent.
- For shows that are going to finish in Dolby Vision HDR, extra care should be taken to keep noise in the image to a minimum.
- Black balancing, APR correction - can be used to eliminate "dead pixels".
When possible, at least one UHD reference monitor on-set | Image QC is critical.
The use of professional UHD monitors allows for confidence during capture by giving a pixel-to-pixel representation of what is being recorded.
- Able to identify dead pixels that may not be visible on an HD monitor.
- Noise due to low light becomes more apparent.
- Focus can be more accurately judged.
LUT boxes for applying looks on-set | Using CDL's for non-destructive look application.
The use of LUT boxes and live-grading software can give creative freedom while not compromising the captured image.
- A better alternative to using the cameras’ built-in paint functions.
- More control without baking-in a particular look.
- Using a CDL or ACES workflow unifies the color pipeline from capture to Post.
MONITORING | Best Practices
- Proper monitoring should represent what the camera is capturing as well as allow for testing different looks non-destructively
- Monitoring should be setup with the capture transfer function and colorspace in mind.
- All monitors should be calibrated in advance of the shoot.
- Multiviewers should be used with large, UHD monitors. Consumer OLED TV's can be used in this instance as well but should not be used for critical viewing.
- Dead pixels should be identified and corrected prior to capture.
- Noise in the image should be within acceptable levels.
A robust data-management process may require extra resources in multicam production compared to typical single-camera shoots. The number of cameras to be backed up and shorter turnarounds that are common in multicam production can strain available resources but with a bit of pre-planning it can be a smooth process. For further reference, see our Production Assets: Data Management article.
DATA MANAGEMENT | Best Practices
A few questions to ask when planning a reliable data management process:
- When will the initial backup happen?
- On-set or in the truck.
- Camera cards will travel and be backed up offsite.
- What type of media will the backups happen on?
- RAID 1/5/6/10 drives
- LTO tapes
- Netflix Content Hub (cloud storage).
- Who will be responsible for backups?
- A Data Manager or DIT is highly recommended due to the volume of material that typically needs to be backed up.
- Personnel at a post facility can be designated if original camera media will be backed-up offsite.
- How will backups be generated?
- Using a proven, professional software solution for backups is very important.
- Checksums should be verified every time a backup is generated.
- Checksum manifests, such as MHL, should accompany all original camera files: from the initial on-set backups to final OCF delivery.
- Visual QC of all original camera files should be performed prior to wiping primary camera cards.
One of the pain-points we hear about on multicam shoots is the short turnaround time for getting footage into the editorial system. Working natively with the source material isn’t always possible with current workflow paradigms and offline editorial proxies are typically created to fit within these current workflow paradigms. Whether generating proxies or working from native source it’s important to think about everything downstream of editorial.
Working with Native Source or Generating Editorial Proxies | Think forward…
- The Post Production ecosystem often includes multiple different toolsets and passing material between these toolsets. Whether you’re working with source files in your offline edit or generating editorial proxies it’s important that the offline editorial clips going into your timeline have a meaningful relationship with the original camera files for finishing and any other required interchange. The method you decide should be compatible with all of the tools involved in the ecosystem of your project.
In-Camera Proxies | Concerns
- Due to extremely short turnarounds, some shows record in-camera proxies and use them for actual editorial work. This is a practice that should be avoided because there is significant QC benefit in having the editorial team looking at footage that was derived from the OCF that Post will be finishing from.
- If working with in-camera proxies cannot be avoided due to time constraints, productions must be mindful of replacing the in-camera proxies with media transcoded directly from the OCF during the conform phase.
- In instances where in-camera proxies are required there should be a Visual QC of all OCF and a vetted meaningful relationship between the editorial proxies and the OCF that is compatible with all of the tools involved in the ecosystem of your project.
All of the cameras on our approved list are capable of capturing an image with enough latitude for a Dolby Vision HDR finish. Netflix does not require HDR monitors on-set during capture.
- Proper exposure is key.
- Use a Log gamma/transfer function.
- If there is a lot of contrast in your shooting environment, a digital cinema type camera may perform better than a broadcast camera.
- HDR finishing tends to exaggerate noise in an image. If shooting in dimly lit environments, a camera with good low-light performance can help.
- Perform tests prior to the shoot if you plan to use HDR monitors on-set. Verify monitors are calibrated and setup to properly interpret the SDI signal from the cameras in the correct colorspace.
A FEW OTHER THINGS...
Cameras should receive an identical timecode feed generated by a master clock.
Workflow drawings should be provided to NETFLIX in advance of the shoot for alignment and support purposes. Here is a basic template to give you an idea of the info you may provide so we can better support your production. This template assumes a Dolby Vision HDR finish:
Whenever possible, the following technical personnel should be present at a NETFLIX multicam production in order to minimize the risk of technical issues.
- Technical Manager / Technical Producer
- Supervises all technical aspects of the shoot.
- Directs crew (sometimes hires crew).
- Oversees show setup and strike.
- EIC - Engineer-in-Charge
- Technical support lead.
- Typically employed by production facility or OB provider.
- VC - Video Controller
- Camera setup, exposure and color.
* For sitcom production and other scripted types of shows, all of the above personnel may not be required. Please consult with your Netflix Post Manager for context.