Cambria 5.8
Release Notes Updated (06/30/2026)
Release Build 5.8.0.31576 Release (and 5.8.0.x Branch Builds)
Release notes cover Cluster, FTC, FTC API Packager, Cluster/FTC Linux & Kubernetes
Notice: Older NVIDIA GPU Architecture Deprecation Notice
Support for older NVIDIA GPU architectures will be deprecated in a future release. Future releases will require Turing or newer GPUs as the minimum supported architecture.
Notice: Removal of Cloud Extend Storage
As MinIO has discontinued support for the filesystem backend functionality leveraged by Cluster’s CloudExtend features, the CloudExtend Storage feature (used for simplified on-premise storage mapping) will be removed from Cluster. Version 5.6 will be the last release to support this functionality.
Notice: EZTitles plugin has reached the end of its life-cycle
-
The Capella EZTitles plug-in will no longer be in active development because the plug-in product has come to the end of its life-cycle.
-
EZTitles will no longer provide customer support for the plug-in, although existing installations are expected to continue to work at this time.
-
Please contact Capella with your caption/subtitle requests.
Notice: Removal of non-HTTPS APIs
-
The methods of the API must be executed using HTTPS. HTTP has been removed and will no longer work. To use HTTPS, simply use the “https://” prefix instead of the “http://” prefix for the URIs. For FTC the port number used for HTTPS methods is 8648. For Cluster the port number used for HTTPS is 8650.
-
SNMP support for Cluster/FTC was using non-HTTPS API. For the 5.1.0.12651 build of Cluster/FTC, SNMP has been disabled. We will look into adding support back in for a future build.
Important notes about the PostgreSQL upgrade
-
Database migration between Postgres 9.3 and 14.3 is not automatic
-
Installing Cluster/FTC ver. 5.0 will come with a clean Database (Postgres 14.3)
-
If you are upgrading from FTC versions 4.8 (or lower) to 5.0 then Postgres 9.3 will still be installed on the system, which will be used to migrate all of the information to Postgres 14.3
-
Only migration from Postgres 9.3 (FTC versions 4.8 or lower) to Postgres 14.3 (FTC 5.0 and above) is supported.
-
Please read the migration document below on how to migrate the database: https://www.dropbox.com/s/otias8vdolo6215/Postgres%2014.3%20Migration%20Document.pdf?dl=0 Downgrading note: Downgrading from FTC 4.5 (or higher) to FTC 4.0 is not supported
Important Licensing notes
-
If you are using a build older than Floating License Server 4.5 or Cluster/FTC 4.5, there is a good chance the online activation license has stopped working sometime after May 2021.
-
If you are using Cluster/FTC without the Floating License Server, then all you will need to do is update Cluster/FTC to v4.5 or higher.
-
If you are using Floating License Server then you will need to update to v4.5 or higher and Capella will need to issue you new Activation Keys.
-
Offline licenses will continue to work.
-
Please contact sales@capellasystems.net for instructions for additional instructions.
System Requirements
Minimum Requirements
Operating System: Windows 10 (64 bit), Windows 11, Windows Server (64 bit) 2016 / 2019 / 2022 /
2025, Linux (Ubuntu 24.04 recommended)
Processor: Intel 2.8GHz (Quad-Core) or faster
Memory: 8GB or more (4K/8K encoding requires more memory, 32GB or more suggested)
Video Card: Supports Direct3D acceleration
OS Installation note: File Convert and Cluster installation requires addition system files included in recent Windows Update. If you encounter issues during installation, please conduct "Windows Update", reboot, and try again.
Warning for machines with more than 64 cores or machines that have multiple processor groups
Cluster/FTC support of multiple processor groups requires the Capella License option “Windows Multi- Processor Group Support (more than 64 logical processors)”. Without this license option, FTC will only utilize only one of the processor groups for all FTC Jobs run on the machine. If the license option is enabled, FTC will distribute multiple Jobs to each processor group. However, a single job will still only use one processor group at a time.
Key Points (Behavior Clarified 10/2025)
Cambria FTC limits the number of logical CPU cores that can be used. Extra licenses (options) are available to support more CPU cores. Supported OS are Windows 11, Server 2022, Server 2025.
Recommended
Operating System: Windows 10 (64 bit), Windows 11, Windows Server 2016/2019/2022, Linux (Ubuntu
20.04.3)
Platform: HP Z4
CPU: 1x Intel Core i9-10980XE @ 3.00GHz (18-core Cascade Lake)
Memory: 64 GB (4x16 GB) DDR4-3200
Video Card: Supports Direct3D acceleration
Network Adapter: Gigabit Ethernet (Wired Connection) or 10 Gigabit Ethernet
System Drive: 1TB Samsung M.2 NVMe SSD
Windows System Settings
Machine Administrator for Installation
It is required that machine administrator account be used for the Cambria software installation. After installation, a non-administrator account can be used for normal File Convert operation. However, for Cluster it is required that the logon user has administrator privilege. Cluster service by default is logon with LOCALSYSTEM, which has administrator privilege. If it is being logon with other users, that user must be a local administrator as well. Otherwise, the Cluster may not start.
To add a user to local administrator: Run lusrmgr.msc, go to Groups -> Administrators, Add that user to this group.
Service Credentials
After installation check the Capella CpServiceManager (File Convert) or CpClusterManager (Cluster) service! These services will need to have the appropriate access rights to read from and write to network file locations. Here are the instructions on how to modify ‘Capella CpServiceManager’ service credentials.
a. Open the Services panel (Control Panel>Administrative Tools>Services) b. Open Properties panel for Capella CpServiceManager c. In the Log On tab, select “This account:” d. Enter the full domain account and password. (ie. Yourname@company.com) e. This domain account must have access rights to the locations f. Restart the service as prompted
Windows Virtual Memory Setting
To prevent some memory allocation errors that can occur, we recommend changing a Windows Virtual Memory setting to allow for “System managed size”.
a. Open up to the Advanced tab in the System Properties panel (Control Panel>System and Security>System>Advanced system settings)
b. Select Settings under Performance c. From the Advanced tab select Change under Virtual Memory d. Uncheck “Automatically manage paging file size for all drives”, select “System managed size”
Dolby Licensee Notice
Cambria FTC contains products licensed from Dolby Laboratories, including:
- Dolby Digital Professional Decoder
- Dolby Digital Plus Professional Decoder
- Dolby E Decoder
- Dolby Digital Professional Encoder
- Dolby Digital Plus Professional Encoder
- Dolby E Encoder
- Dolby Dialogue Intelligence.
- Dolby Encoding Engine
- Dolby Vision Live Distribution Processing Manufactured under license from Dolby Laboratories. Dolby and the double-D symbol are trademarks of Dolby Laboratories
New Features
New Features added for 5.8
Features listed include all of the new features added for 4.x, 5.x up to 5.8. Please note that some of
these features are in a beta state.
Module Feature Capella
Reference #
Filter Expanded Subtitle Inject Input Formats for 608 Captions 21250
The Subtitle Inject video filter now supports additional caption file
formats when generating embedded 608 captions.
Supported input formats for creating 608 output: SCC, GLA,
TTML, XML, SRT, STL, PAC, and VTT.
Filter SCTE-35 Inject Option to Preserve GOP Timing 21626
Support added for controlling whether the SCTE-35 Inject filter
starts a new GOP at SCTE-35 marker positions. This helps
workflows that require fixed keyframe timing, such as
maintaining IDR frames at regular 2-second intervals while still
inserting SCTE-35 ad signaling.
Usage: In job XML, set SCTE35StartsNewGOP for the SCTE35
Injection filter to control whether SCTE-35 markers start a new
GOP.
Notes: This option is API/job XML only and is not exposed in the
UI.
Filter Subtitle Burn-In Font Styling 22062
The subtitle burn-in filter now supports font styling, including font
color, edge color, bold, underline, italic, and transparency. This
makes it possible to control subtitle appearance directly from the
FTC subtitle filter settings.
Notes: For DVB Subtitle Burn-in, the styling applies when Burn-in
Mode is set to Version 2. The related DVB Subtitle Inject filter
also exposes these styling options in the FTC tab.
JobXML Audio Normalizer Loudness Log Output 21642
Submission
JobXML only feature that enables loudness logging option for the
Audio Normalizer filter. When LoudnessLog="1" is added to the
Audio Normalizer filter settings in the job XML, FTC generates a
CSV loudness log in the output location.
The log includes loudness information for the original and
converted audio over time, including original and converted
loudness values. Generated LoudnessLog CSV filenames now also
handle Korean characters correctly.
Usage: Add LoudnessLog="1" to the Audio Filter - Normalizer
filter settings in the job XML.
Note: Some fields from the customer’s reference log format are
not available from the normalizer filter, including source/output
dialnorm level and source/target codec information.
JobXML XML File Operation Jobs 21798
Submission
FTC now supports file operation jobs in XML for Copy and Move.
Use <Job Type="FileOperation"> with <Operation Type="Copy"
/> or <Operation Type="Move" />, then define the source and
destination locations in the job XML.
Notes: This is XML-only and has no UI.
WebUI Queue Jobs to Remote Machine 21321
Cambria FTC now includes WebUI options to queue all jobs or
selected jobs to a remote FTC/Cluster machine. You can enter a
machine name, hostname, IP address, or a full API endpoint such
as https://10.0.0.24:8648/CambriaFC/v1/Jobs or https://api.myh
ost.com/CambriaFC/v1/Jobs/?usertoken=myusertoken.
Manager / Web UI S3 Browser 21944
WebUI
A new S3 browser is available from the Job Creator folder icon,
allowing users to browse an S3 location using the entered Access
Key and Secret Key.
Notes: Credentials are used for browsing and are not saved.
Manager / Web UI Simulated Jobs for Testing 21682
WebUI
The Job Creator page now includes options to queue 1-minute
and 10-minute simulated jobs from the FTC/Cluster Web UI.
These fake jobs are useful for troubleshooting and quick testing.
Notes: The simulated jobs appear in the preset list as 1-Minute
Simulated Job and 10-Minute Simulated Job.
Manager / Web UI Filters / Fields 21679,
WebUI 21675
Job ID Filter
The FTC/Cluster WebUI search box can now filter jobs by exact
job ID. Enter the full job ID to narrow the list to that job.
Machine RAM Display
The Machines list now shows the reported RAM value for each
machine. This makes it easier to review system memory
information directly in the WebUI.
Linux / Updated Kubernetes Diagnostic Script for Manager and 21672
Kubernetes Ingress Changes
Updated support for the current Kubernetes deployment layout.
The script now checks the capella-manager namespace for
Manager components and uses traefik for ingress-related
New Features added for 5.6
information instead of the previously used ingress-nginx.
Improvements also include AWS EKS reporting by attempting to
resolve an ingress hostname to an IP address when the ingress
exposes a hostname instead of a direct IP.
Windows Windows Smart App Control-Compatible Signed Installers 21945
Release builds are signed Windows installer builds using Microsoft
Azure Artifact Signing, also known as Microsoft Trusted Signing.
Signed builds help prevent Windows 11 Smart App Control from
blocking Capella installers and application binaries with warnings
such as “Smart App Control blocked an app that may be unsafe.”
Signed builds include Capella executables, DLLs, redistributable
files, and MSI installers required by the application package. This
is especially important on new Windows 11 installations where
Smart App Control may block unsigned installers or dependent
binaries.
Note: Capella cannot retroactively sign previously released
builds; older builds must be rebuilt and redistributed to include
signing. We recommend updating to the latest release version
first.
Module Feature Capella
Reference #
Filter Subtitle Inject Filter 19954
Added a new DVB Subtitle Inject filter to allow DVB subtitle
streams to be injected into the output from subtitle files.
Target (H.264 NETINT VPU test module (Akamai Cloud) 20998
-VPU)
If you’d like to test the new VPU module, please contact us at:
https://www.capellasystems.net/contact-us
In FTC the VPU is used for hardware acceleration of H.264 video
encoding. Testing may require a new build of the software.
Target (AV1) Added support for Film Grain Denoise in MP4 AV1 encoding with 21092
two new settings that reduce film grain and allow adjustment of
levels.
(Linux / On-prem Kubernetes Support 20593
Kuberenetes)
Canonical Kubernetes for on-premise is now supported for
Cambria FTC. All machines that are part of an on-premise
Kubernetes deployment must use Ubuntu as the Linux
distribution. The tested Linux distribution is Ubuntu 24.04. For
high availability, at least 3 machines are required and must
New Features added for 5.5
be designated as control plane nodes. For Cambria Cluster
manager redundancy, it is recommended to have at least 3
dedicated worker machines separate from
the control plane nodes.
There are some notable currently unsupported features: FTC
Autoscaler, GPU and VPU support.
Document link:
https://www.dropbox.com/scl/fi/b4hofqgc8r3yp1plftmwo/Cambri
a_Cluster_and_FTC_5_6_0_on_On_Premise_Kubernetes.pdf?rlke
y=raeaiu9tnpmwlh942kfa3kz14&st=prmaws1c&dl=0
Windows Use External PostgreSQL 14 installation method 21239
Please email support@capellasystems.net if you would like
instructions for this.
Module Feature Capella
Reference #
Source Support Blackmagic RAW (braw) sources 19232
(Blackmagic
RAW)
Target GPU support for AWS EKS machines 19770
(NVENC)
Target 8K support added to HEVC (NVENC) for ABRv2 18506
(ABRv2)
Target Add XAVC-Intra encoding to Elementary Stream exporter 20245
(Elementary
Stream)
REST API Setting Tag based slots at boot 18756
Added to API documentation in GET/SET CambriaManagerOptions
(Linux / Oracle Kubernetes (OKE) supported 20351,
Kuberenetes) 19817
(Linux / Oracle Object Storage support (through S3 interface) 19971
Kuberenetes)
(Linux / Installation via Terraform added 19947
Kuberenetes)
(Linux / IAM role used for Auto-scaler permissions and Loki on 20469
Kuberenetes) AWS
New Features added for 5.4
Module Feature Capella
Reference #
Linux (Docker Base image for docker container updated to use Ubuntu 24.04 19613
Installation)
Source (R3D) RD3 source decode support 19327
Cambria FTC can now decode R3D files (Windows only)
Notes:
- Windows only
- Due to limitations with the R3D SDK, these files are not
supported via cloud storage platforms like S3
Target (DV5) Encoding to Dolby Vision Profile v5 (DV5) is now 18956
supported in TS, MP4, ABRv2 (DASH, HLS/fmp4, CMAF)
containers
Note: Sources are required to be HDR (HLG, PQ, HDR10 color
spec) to encode to DV5. Jobs with Non-HDR sources will fail
when encoding to DV5.
Target (WAV) WAV Exporter: CART Chunk Metadata support 19528
API job submission only. CART Chunk metadata can be added to
‘wav’ XML job. Here are the attributes supported in the
MuxerSettings for this:
"WriteCart"
"CartTitle"
"CartArtist"
"CartCutID"
"CartClientID"
"CartCategory"
"CartClassification"
"CartOutCue"
"CartStartDate"
"CartStartTime"
"CartEndDate"
"CartEndTime"
"CartProducerAppID"
"CartProducerAppVersion"
"CartUserDef"
"CartLevelReference"
"CartURL"
"CartTagText"
For dates, format is yyyy/mm/dd
For times, format is hh:mm:ss
Target (audio) Audio Language Code passthrough 18445
Audio Language Code (ISO 639-2) can now be passed through to
the output by setting the audio language to 'src'.
Note: Only certain containers allow for this to work (TS, MP4,
MOV).
Note: Known limitations / unsupported cases:
- audio passthrough
- audio remapping filter
- audio mapping
- using additional audio sources (eg separate .wav files)
Target Specifying Multiple DVB Subtitle PIDs 19014
(Subtitle –
DVB) There is now a way to specify multiple DVB subtitles by their own
PID instead of the PID being set in sequential order. This can only
be set by editing
a job XML directly for use with Cambria FTC API. In the
<MuxerSettings>, set the attribute 'DVBSubtitlesPIDX' to the PID
that the output track should be. See the following example,
<Job><Settings><MuxerSettings DVBSubtitlesPID="650"
DVBSubtitlesPID2="660" DVBSubtitlesPID3="670"
></MuxerSettings>
Note: If no DVBSubtitlesPID is specified for a track, that output
track's PID will be: FIRST TRACK PID + TRACK NUMBER – 1
Target WEBVTT alignment setting added 19014
(Subtitle –
WEBVTT) Overriding the default alignment of WEBVTT subtitles is now
possible.
"align:center"
"align:start"
"position:30% align:start" - position moved 30%
Setting added to Subtitle Stream Settings.
Target (GPU Machines with multiple NVENC GPU cards supported 19362
acceleration)
Target OCR enhancements 19654,
(Analysis 19507
Exporter - - Tooltip added to ‘Language’ field to indicate what language
OCR) codes to use
- Mixed English and Japanese text support
Filter (Speech- Speech-to-text enhancements/fixes 19640,
to-text) 19629,
New Features added for 5.3
- .vtt and .srt output fixed to work with VLC playback 19628,
- Filter removed from target side filters 19627,
- AVX2 is required to use this feature, improved error added 19433
- String replacement supported for output filename
- input field "Language" was added to allow the user to specify
the expected language to detect in the audio
Filter (Blur Blur strength increased 19309
Face)
S3 / Post Task New XML attribute added: ContentTypeOverride 19269
Upload
API job submission only. The new attribute
"ContentTypeOverride" allows you to set the content-type for S3
uploads.
WebGUI WebUI Preset Editor enhanced 19029
Target Preset Editor for the WebUI updated to be more in-line
with the Preset Editor for the Application UI.
Note: Functions such as video/audio filters, scriptable workflow,
post task commands cannot be configured through the WebUI
Preset Editor. If these functions are needed, we recommend using
the Windows application UI to create JobXML templates for API
submission.
Module Feature Capella
Reference #
Source (Dolby Dolby Vision Module Updated 18478,
Vision) 18457
The updated version supports using Dolby Vision MOV as sources.
Source Read sources directly from Azure Blob storage (JobXML, 18991
(Azure) API only)
To use this feature, the following is needed:
1. Url to the source (Eg.
https://myuserstorageaccount.blob.core.windows.net/my_contain
er/my_source.mp4)
2. SAS credentials for the source
Example of Azure Blob storage url in JobXML (Note: replace the
'https://' in your url with '[azure]'):
<Source
Location="[azure]myuserstorageaccount.blob.core.windows.net/
my_container/my_source.mp4" Name="Src1"/>
The SAS token will go in the <JobDescr /> tag like this:
<JobDescr Description="FTC UI (Preset: DASH)"
azure_blob_sas="<my-azure-blob-sas>">
Notes:
- The SAS token must have at least Read and List permissions
How to get URL and SAS Credentials:
To get the url, click on the source in Azure Blob storage and copy
the URL property in the "Overview" section.
To get SAS credentials, click on the source in Azure Blob storage
and go to the "Generate SAS" section.
Target (Post Upload to Azure Blob Storage 19040
Task Upload,
Azure) In FTC Post Task, there is now an Azure Blob storage upload
option that allows you to specify the following:
Storage Account Name: The Azure Blob Storage account name
(under the 'Storage accounts' section of Azure)
Container: The name of the container to output to (under
'containers' in the storage account page)
SAS Token: access token to the container (can be generated
through the container's option 'Generate SAS')
Block Size: the upload chunk size (in kilobytes). Azure limits this
to 50 GB per file
Notes:
- SAS token must have write permissions to the container
- The Storage account needs to have "Allow storage account key
access" enabled. This setting can be found in the Storage
account's Settings > Configuration section
- Currently, writing directly to Azure Blob storage is not
supported
Target (GPU) Supported added for using multiple NVENC cards on a 19030
single machine
Target (GPU) 8K support enabled for NVENC jobs 18506
Target Split and Stitch 19321,
14850
A job/source can be split into multiple video segments and one
audio file. The job segments are distributed to multiple machines
to encode. Once the segments are complete, they are stitched
together to create the final output, as if it was transcoded in one
machine.
The temporary files are written to a user-configurable location
called the WorkFolder.
In Cambria FTC UI (not the Cambria Manager UI) you can specify
this location in Settings > Options > Split And Stitch Options.
For Windows, you can run regular jobs as "split and stitch" jobs
by selecting the "Queue All Jobs" dropdown and selecting either
"Queue All / Split And Stitch" or "Queue Selected / Split And
Stitch".
In JobXMLs, to specify that a job should be in "split and stitch"
mode, set the Type attribute in the <Job></Job> tag to
"SplitAndStitch". Also, set the attribute SegmentDurationSec to
the amount of seconds
to split each segment and WorkFolder to a shared location where
the segments will be written to.
Sample Job Tag:
<Job SegmentDurationSec="30" Type="SplitAndStitch"
WorkFolder="C:\Temp"></Job>
For Cloud/S3:
<Job SegmentDurationSec="30" Type="SplitAndStitch"
WorkFolder="[s3]mybucket@us-east-
1:SplitAndStitchTest/Segments"></Job>
Known Limitations (does not work with Split and Stitch):
- Frame Rate Conversion
- Adaptive Streaming Output
- Audio mapping jobs
- Jobs with more than one segment configured
- Stitched sources
- Multi-Target jobs
- Not All Video Filters work
- Watch Folder
- Job Tags (Cambria Cluster)
- Audio Only Jobs
Target Analysis Exporter Enhancements 18768,
(Analysis) 19007,
- Detect Video Cadence 18959
- Find Faces
- Improved OCR (Text Recognition)
Note: The new OCR feature requires additional files to be
installed into C:/Program Files (x86)/Capella/Cambria/cpx64
Please contact support@capellasystems.net to request for the
additional files.
Filter (Blur Blur Faces Filter 19007
Faces)
There is a new video filter "Blur Faces" which places a rectangle
around faces and basically blurs them. Users can adjust how
much the faces are blurred (strength) and the confidence level
threshold that determines whether something in the video is a
face or not.
Notes:
- The lower the threshold, the more likely faces will be blurred.
However, currently there is a higher chance other objects will also
get blurred with lower thresholds.
- Works well with faces further out in the frame. Faces may still
recognizable when close to the camera.
Filter (Speech Speech to Text Filter 18932
to Text)
New filter added that allows for text to speech in the following
formats: .xml, .vtt, and .srt.
Filter Nexguard SDK version update to 1.18.0. With that, Linux now 19134
(Nexguard) supports Nexguard video filter.
Locations to add the Nexguard license:
Windows (FTC): C:\Program Files (x86)\Capella\Cambria\cpx64\NexGuardV2
Windows (Cluster): C:\Program Files (x86)\Capella\CambriaCluster\cpx64\NexGuardV2
Linux (FTC): /opt/capella/Cambria/bin
Linux (Cluster): /opt/capella/CambriaCluster/bin
WebGUI WebUI Preset Editor (Beta) 18711
The preset editor was added to the Cambria FTC/Cluster WebUI.
This only includes encoding configuration and not all encoding
target/settings are included. Functionality such as filters, post
task delivery, notifications cannot be configured through the
preset editor at this time.
Licensing Enhancement for Cryptlex error handling 18499
Reactivation retry implemented for unexpected fail case.
Licensing Fully Redundant Licensing for Hosted (and Hosted Floating 19313,
Server) Licenses 19292,
18514
This feature is only supported for Hosted Licenses and Hosted
Floating Server Clients. Requirement is to whitelist this IP:
54.241.85.49
If the machine loses connection to cryptlex (i.e. cryptlex server
goes down, etc) then backup licensing will keep the machines
license active. Other scenarios of sudden invalidation or grace
period is over are also covered by the backup licensing server.
Hosted Floating Server Clients
Please note that this feature is ONLY for the clients and not the
Floating License Server itself.
In the case when the floating server license unexpectedly
invalidates, then currently connected and leased client machines
will continue to be operational because of the backup licensing
support.
Notes:
- This feature works for Hosted Floating Server Clients but not the
Floating License Server itself.
New Features added for 5.2
- Testing to make sure 54.241.85.49 I reachable.
ping cpfs.capellasystems.net
or
wget https://cpfs.capellasystems.net:8481/ ===> should get
xml (it will states an error but enough to check connectivity)
Security Support for TLS 1.2 added, TLS 1.1 and 1.0 is disabled 18828
TLS 1.2 is the minimum version needed to negotiate SSL.
Module Feature Capella
Reference #
Filters Added Video Denoiser Filter 18321
Filter added to remove grain/noise from video.
Filters DFXP Support added for Burn-in Filter 18241
Scriptable Security Enhancements affecting executables used in Jobs 18370,
Workflow/Exec and Perl scriptable workflow scripts 17658
utables
Starting in 5.2 release, any executables (post-taks, notifications)
or Perl scripts (scriptable workflow) used as part of a Job must be
trusted by the machine executing the Job.
Have your IT department review this document to determine if
you want to use this feature, and if so to assist with usage of this
feature
The reason why executables and scripts need to be trusted is to
ensure that only authorized programs are executed on a machine.
When a script or an executable is registered as trusted, it means
that the user or the system administrator has explicitly allowed it
to run on the machine. Without this trust, any script or
executable could potentially run on a machine, including those
that are malicious or harmful.
By registering trusted executables and scripts, organizations can
ensure that only authorized programs are executed on their
machines, thereby reducing the risk of malware infections, data
theft, and other security breaches. Additionally, the trusted list
can help prevent unauthorized modifications to scripts or
executables, as any change to the program will change its
SHA256 hash, requiring the trusted list to be updated.
Overall, the registration of trusted executables and scripts is an
important security measure that helps to mitigate the risk of
security incidents caused by malicious programs, and
organizations should take this process seriously to ensure the
safety of their systems and data.
Here are some instructions for using the features described:
Registering Trusted Executables:
To register a trusted executable on a machine, follow these steps:
a. Generate the SHA256 hash for the executable file you want to
register.
b. Open the registry editor by typing "regedit" in the search bar
or the Run dialog box (press Win+R).
c. Navigate to the following registry key:
HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\CAPELLA\CA
MBRIA.
d. Create a new string value called "TrustedAppSHA256".
e. Set the value of the "TrustedAppSHA256" to be the SHA256 of
the executable, separated by commas if there are multiple
executables.
f. Repeat these steps for each machine where you want to
register the trusted executable.
Alternatively, you can create a .reg file with the SHA256 of the
trusted executable and run it as administrator on each machine
where you want to register the executable.
Registering Trusted Perl Scripts:
To register a trusted Perl script, follow the same steps as above,
but use the "TrustedScriptSHA256" registry key instead of
"TrustedAppSHA256". Note that any changes made to a Perl
script will change its SHA256, so the trusted list will need to be
updated accordingly.
Using Python Scripts:
Python scripts are executed in a sandbox, so they don't need to
be registered as trusted. However, they will not have access to
the network, which can be necessary for some workflows.
Security Considerations:
It's important to note that any trusted executable will be allowed
to be called with different arguments, so users need to make sure
that trusted executables can't cause any damage with other
arguments. For example, if a trusted executable takes a file name
as an argument, it should check that the file is of the expected
type and doesn't contain any malicious code. Similarly, if a Perl
script takes user input, it should validate the input to prevent
injection attacks. If network access is not needed, it's
recommended to use Python scripts instead of Perl scripts to
reduce the risk of security issues.
Please contact your IT department to assist with properly
identifying which executables and scripts may be risky.
Kubernetes AWS Kubernetes Support 18280
Link to installation guide:
https://www.dropbox.com/scl/fi/xt7vj917tg9bejo0r6x8d/Cambria
New Features added for 5.1
_Cluster_and_FTC_RC1_5_2_0_on_AWS_Kubernetes.pdf?rlkey=z
r11ja7gk7fkh37ucw7nw4dsl
Kubernetes Argo Events to FTC Job 17939,
18136,
Argo Events can be used to trigger FTC encoding jobs. 18270
Kubernetes Akamai NetStorage Watcher 18275,
18220
NetStorage Watcher can be used with Argo Events (webhook) to
trigger FTC jobs. This adds “Watch Folder” like functionality.
Linux Upgrade to PostgreSQL 15 18280
Module Feature Capella
Reference #
Source HEIC source support added 17273
Source 12 bit DPX source support added 17165
Source Preset Creation function expanded to support MXF, DNxHD, 17417
Analysis ProRES, IMX, XDCAM HD
Target MPEG-2 Transport Stream supports writing 17524
(Transport AU_information into output file/frames
Stream)
This setting can be turned on for the preset through the
Container Format Settings for the Target Preset. Required to
support DVB-AU spec. May be needed for some parsers/players.
Target (MXF) Closed Captions settings added to MXF exporter 17432
Both the Video Settings and Container Settings area for a MXF
Target Preset now contain a Closed Captions setting which will
allow CEA-708 to be used.
Target (HLS) TTML support for MSS added 17219
TTML Support for MSS was added, both for Sidecar and
Embedded cases. Currently, we only support embedding / sidecar
with an external Subtitle file.
Target (AVID XAVC 300 and 480 support was added to the AVID 15407
AAF) exporter
Target Segmenter Exporter 17287
(Segmenter)
This exporter allows the user to partition their video into multiple
files based on a user configured Segment Duration.
Filter (Subtitle SCC caption file format supported by Subtitle Inject filter 17589
Inject)
Watch Folder Smart Watch Folder Enhancements 16443,
16546,
In the Watch Folder Source Acceptance section, ‘Source is a 16545,
group of files’ feature has been updated with 3 new workflow 16544,
functions. 17440,
17195,
- Use Workflow Job 17016
- Dynamic Grouping (Parse XML to find associated files)
- Passthrough those files to create an ABR output
Use Workflow Job
Enables the ‘Scripts Settings’ function on the Encoding Tab that
will allow for Jobs to be created via a Workflow Script. This will
enable a Job to include functions for decision making and various
tasks (encoding and administrative) to be part of the Workflow
Job.
Dynamic Grouping (Parse XML to find associated files)
Allows for the Dynamic Grouping Script to be used instead of
Filename Matching Criteria for groups of files.
Passthrough those files to create an ABR output
Sets the Filename Matching Criteria to look for multiple video files
that will be used as sources for an Adaptive Streaming
(Packaging Only) Job.
Cambria Status and Summary section added to Watch Folder tab 17444
Manager
(Watch These sections allow for users to get more information about the
Folder) current status of a watch folder, what process is currently
running, etc.
Cluster User configuration for ‘Number of Urgent Jobs’ in Cluster Manager 17045
Manager
Web UI Add/Remove Machines from Cluster function added 17597
Linux Cluster / FTC Linux Document 17738
Linux startup guide has been updated with a new Vulnerabilities
Information section.
Link to Cluster/FTC 5.1 Linux startup guides:
https://www.dropbox.com/s/236y5m30jde54s0/Linux_FTC_Guide
_5.1.0.pdf?dl=0
https://www.dropbox.com/s/crprvyzl2etlabo/Linux_Cluster_Guide
_5.1.0.pdf?dl=0
Latest Version (9/08/2023):
https://www.dropbox.com/scl/fi/mf9h58uip3nalt42591j5/Linux_C
luster_Guide_RC1_5.2.0.pdf?rlkey=1a46fg2hok6n1smw5pa2n842
e
https://www.dropbox.com/scl/fi/0msy7rj48ktmmmcou5h23/Linux
_FTC_Guide_RC1_5.2.0.pdf?rlkey=5lbxzpvcfcrsseg6j9jq69j7u
New Features added for 5.0
Kubernetes Akamai Kubernetes Support 17607,
Support 17612,
Cambria Cluster / FTC can be run in an Akamai Kubernetes 17615,
environment. Currently functionality is limited to API only for Job 17617,
submissions. API or WebUI can be used to monitor Jobs. 17618,
Not all features in the Windows version of FTC are available in 17620,
Linux. Please reference the latest Linux startup guide for a list of 17621,
known Linux limitations. 17783
Link to Cambria Cluster / FTC 5.1 installation and help guide for
Akamai Kubernetes:
https://www.dropbox.com/s/0uxi0o0tnq9zs66/Cambria_Cluster_a
nd_FTC_on_Linode_Kubernetes_5.1.0.pdf?dl=0
Latest Version (9/08/2023):
https://www.dropbox.com/scl/fi/x1vdb97zac7e2tmyeq2om/Camb
ria_Cluster_and_FTC_RC1_5_2_0_on_Akamai_Kubernetes.pdf?rlk
ey=db1sn20pm577zei8lxfslvqow
Module Feature Capella
Reference #
Exporter Fraunhofer FDK encoder 16730
(Audio)
Added as optional configuration in the audio encoder settings.
Exporter Adaptive Streaming v2 exporter supports teletext to 16788
(Adaptive WebVTT conversion
Streaming v2)
FTC now supports teletext input for ABRv2 subtitle streams.
Steps:
- Users will need to first add a subtitle track to the ABRv2.
- Inside the subtitle track, add individual subtitle streams.
- Inside the subtitle stream, specify input type as Teletext.
- Specify Teletext Page number for the input.
Note: Users can find the Teletext Page number by analyzing the
source either in FTC source analysis feature or using third party
apps such as MediaInfo.
Filter Color Conversion filter performance improvement 16848
Color Conversion processing improved multi-threading taking
advantage of systems with more threads (particularly 46 threads
or more).
Filter Nexguard A/B Watermarking (requires Nexguard license) 16350
Nexguard A/B Watermarking support added to FTCs NexGuard
filter. Also, a checkbox option can be found in Adaptive
Streaming v2 exporter that can also apply the A/B Watermarking.
Filter (Beta) Teletext Extractor filter 15921
Enables workflows for converting VBI Teletext from the source to
OP47 in the target.
Source (Beta) S3 Requester Pays buckets supported 16488
Supported in JobXML through API submission.
Please contact support@capellasystems.net for info
Cluster Cluster/FTC network multicast dependency has been 16589
removed
Cluster can use network multicast to automatically detect FTC
clients. If multicast is disabled after the initial detection,
Cluster/FTC will still continue to work.
Cluster / CloudExtend Improvements 15884,
CloudExtend 15895
- AWS EC2 Credentials and AQS Instance Control settings can be
exported to save file and reimported.
- AWS Operations such as (stopping and terminating) are now
disabled for non-AWS machines.
FLS API functionality added to register and deregister Licenses 16589,
15867
Please contact support@capellasystems.net for updated API
documentation.
Module / PostgreSQL version updated to 14.3 16347,
components 16972
Postgres is used by FTC and Cluster has been updated to 14.3
from 9.3.2.
Notes:
- Installing Postgres 14.3 does not replace/remove old 9.3.2.
Users can remove old version once they migrated. However,
make sure that old database information is migrated properly.
- New Postgres 14.3 is not a seamless upgrade and require user
actions. User is expected to run "PostgreSQLUpdater" to clone old
database to new database (migration).
Migration doc link:
https://www.dropbox.com/s/otias8vdolo6215/Postgres%2014.3
%20Migration%20Document.pdf?dl=0
- User may not setup Redundancy across different Postgres
versions.
Linux FTC (Linux) Improvements 16977
- Job Tag column added to WebUI
New Features added for 4.8
- Cluster (running on Windows) can distribute jobs to FTC (Linux)
Link to FTC 5.0 Linux startup guide:
https://www.dropbox.com/s/vzyg64o1ec7axld/Linux_FTC_Guide_1.
1.0.pdf?dl=0
Module Feature Capella
Reference #
Linux OS Linux Support 14892
FTC can now be installed onto Linux.
Currently functionality is limited to API only for Job submissions.
API or WebUI can be used to monitor Jobs.
Not all features in the Windows version of FTC are available in
Linux. Please reference the startup guide of a list of known
limitations.
Link to FTC 4.8 Linux startup guide:
https://www.dropbox.com/s/my7m1f4dosvq9ir/Linux_FTC_Guide_
1.0.pdf?dl=0
Link to FTC 5.0 Linux startup guide:
https://www.dropbox.com/s/vzyg64o1ec7axld/Linux_FTC_Guide_1.
1.0.pdf?dl=0
Target GPU (NVENC) Acceleration for HEVC Jobs 16321
(NVENC Video
Encoder) New video encoder option added: HEVC (NVENC)
Enables FTC to use specific NVIDIA cards to do accelerated HEVC
encoding. This encoder option is included as part of the HEVC
(purchase option).
Video Settings:
Profile: main
Levels: 2.0, 2.1, 3.0, 3.1, 4.0, 4.1, 5.0, 5.2, 6.0, 6.1, 6.2
Tier: main, high
Frame Size: 224x126, 320x180, 640x360, 864x486, 1280x720,
1920x1080, 3840x2160
Frame Rate (fps): 23.98, 24, 25, 29.97, 30, 50, 59.94, 60
Interlacing: Progressive, Upper Field First, Lower Field First
Display Aspect Ratio: Same as Source, 4:3, 16:9
Rate Control Mode: CBR, VBR
Bitrate: User specified: max is 240,000 kbps
Tuning: High Quality, Low Latency, Ultra Low Latency
Preset: 1-7: 1 is Low Quality/Fast: 7 is High Quality/Slow
GOP Structure: I only, IPP, IBP, IBBP
(B frames supported in *Qualified* GPUs only)
Maximum CPB Size: 1, 2
IDR Period (GOP Count): User specified: default is 120
Performance:
Quality, Speed, and Number of Concurrent encodes are limited by
the type of Nvidia GPU used. List of features that are influenced
by GPU type:
1: Number of GPUs supported on a single machine
2: Number of concurrent NVENC encodings
3: HEVC B Frame support (B frame in GOP structure)
4: Processing speed of a NVENC encode
7th NVENC Generation (or newer) *Qualified* GPUs are
recommended for better performance and least limitations. For a
list of GPUs, please visit https://developer.nvidia.com/video-
encode-and-decode-gpu-support-matrix-new
The only officially supported card (Capella tested):
*Qualified* GPU
Benchmark:
PNY NVIDIA Quadro RTX 4000
(Machine Spec: 12th Gen i7-12700K, Windows 11)
4K → 4K
Resolution/FPS same-as-source (3840x2160@29.97fps)
Tuning = High Quality
GOP = IBP
Preset (quality/speed) = 3 (default value)
Bitrate=10,000kbps CBR
Result:
1 job at 2.89x RT
4 jobs at 1.13x RT
HD -> HD
Resolution/FPS same as source (1920x1080@29.97fps)
Tuning = High Quality
GOP = IBP
Preset (quality/speed) = 3 (default value)
Bitrate= 5,000kbps CBR
Result:
1 job at 8.84x RT
16 jobs at 1.02x RT
Benchmark Note: Different machine/storage environments may
affect the results. Changes in encoding settings will also affect
results (Ex: 4K@60fps runs 2 jobs at 1.02x RT)
Additional Notes:
- For qualified cards, there's unlimited number of jobs you can
concurrently run.
- For unqualified cards, the limit is 3 (concurrent jobs)
- Unqualified cards we've tested with is Nvidia Quadro P400 and
NVIDIA GeForce GTX 960. These cards cannot do B frames
- GPU is only used for encoding. CPU is used for everything else,
such as decoding, scaling, frame rate conversion, color format
conversion, and any other video filters
- Multiple cards is not supported
- Higher end card does not directly imply better performance. For
details, please consult Capella Support.
Source (DPX) DPX files can now be loaded without needing the XML file 15994
descriptor. You can load any of the dpx files in the list and the
same behavior as the xml file should apply.
Source (Y4M) Support added for Y4M sources 16132
Target CMAF output added to Adaptive Streaming v2 exporter 16006
(Adaptive
Streaming v2) The Adaptive Streaming v2 container now has an option for CMAF
output. Similar to some of the other Adaptive Streaming v2
options, CMAF has support for DRM with Fairplay as the HLS
option and Widevine/PlayReady as the DASH option. You will see
this as "Fairplay/Widevine" or "Fairplay/Playready".
Target AV1 AV1 video codec option has been added to MP4 exporter 16343,
15430
FTP Upload Create subdirectories for FTP Upload with String 16355
Replacement Variables
String replacement variables can now be used in the 'Server' field
of the Post Task FTP uploaded settings to create subdirectories.
These string replacements can be used in Post Task FTP config:
%presetName%
%sourcePath%
%sourceName%
%sourcePath0%, %sourcePath1%, etc
%projectName%
If the preset is used in a WatchFolder, these variables will not
work:
%presetName%
%projectName%
Retrieval / NetStorage support enhancements 16342,
Post Task 16341
Upload 1. Watch Folders can now process a source file directly from
NetStorage without downloading the entire file to the Watch
Folder first.
In the NetStorage retrieval option for Watch Folders, there is a
new "Process without downloading as a file" checkbox. Selecting
this setting allows for a local Watch Folder to also "watch" a
NetStorage location and use sources directly without having to
transfer it first.
2. For API users. Added support for specifying the NetStorage
credentials in FTC JobXMLs.
The attribute "netstorage_secret_key" in the
<JobDescr></JobDescr> tag which allows the user to specify
their NetStorage HTTP API Key for reading files from and
uploading files to NetStorage, and the format for reading file from
storage: [netstorage]<netstorage username>@<netstorage
domain>:<path to the source in netstorage>
Ex:
<JobDescr netstorage_secret_key="adfsd">
...
<Source
Location="[netstorage]netstorage_username@netstorage_domai
n:pathtomyfile/subfolder/myFile.mp4" Name="Src1" />
...
</JobDescr>
Audio Track Track Name Label added to the Audio Track Mapping dialog box 16155
Mapping to allow for user description for the various tracks. JobXMLs
created from these jobs will also contain the description.
Filter (VITC) Filters added to support VITC generation and extraction 15958
There are two new filters added:
- VITC Generator
- VITC Extractor
The generator filter adds VITC specific timecode information into
the output. The VITC Extractor takes VITC timecode information
stored in the source and uses it in some way (currently only to be
reused with the burn-in filter).
Notes:
1. In order to use VITC generator, the source needs to have a
frame width of 720 (if source filter) or target (if target filter)
2. In the FTC UI, both of the VITC filters don't have any
configurable options for the filters themselves
3. For VITC Generator, the source needs to either already have
timecode embedded or a timecode overwrite filter needs to be
used alongside it. If using timecode filter, the timecode filter
needs to activate first. This means the order of these filters
matters.
4. For VITC Extractor, the filter needs to be used as a source filter
(whether it is used directly as a source filter or on the preset
side).
5. The only ways to get VITC Extractor works as a target side
filter is when same as source is used for encoding settings such
as frame size, frame rate, and interlacing or VITC generator +
VITC extractor are used
in the same job in that order (Eg. Job with Timecode overwrite +
VITC generator + VITC extractor + Timecode burn-in)
Filter NexGuard v2 filter enhancement 15989
(NexGuard)
In our NexGuard v2 filter, we have now included a way to do
ClipMark watermarking, along with G2 (which was already there
before). In order to do ClipMark watermarking, you need an
appropriate license from NexGuard.
Licensing note: NexGuard Watermarking V2 For FTC 4.8 has
changed licenses due to NexGuard (NAGRA) requirements.
Hence, your existing Nexguard watermarking licenses will not
work. Please obtain new licenses from NexGuard.
Windows OS Windows 11 and Windows Server 2022 Supported 16483
New Features added for 4.7
Module Feature Capella
Reference #
Watch Folder Support encoding directly from S3 15841
When setting up a Watch Folder for Retrieval from S3, you can
check the ‘Retrieve as S3 shortcut’ option. When you configure
this setting, instead of moving the entire source file down from
S3, FTC will create just the shortcut in the Watch Folder. The Job
that is generated will encode the source directly from S3.
Watch Folder Support retrieval from S3 15739
Retrieval from S3 has been added as an option that can be
configured from the ‘Retrievals’ tab for Watch Folders.
Source Support for AV1 codec files 15387
Source Support for S3 pre-signed URL source 15785,
(S3) 16072
"Pre-signed URLs" HTTP-based urls that a user can generate via
the AWS command-line tool for files on an S3 bucket. These
URLs can now be added to a FTC job as a replacement to the
source path.
Steps:
1. Get a pre-signed url from the aws command-line or other tool.
See
https://docs.aws.amazon.com/AmazonS3/latest/userguide/Share
ObjectPreSignedURL.html
2. Edit an existing FTC job or project and add the pre-signed url
to the location you'd like to use the resource (such as Source,
Logo Filter source, etc).
3. If a JobXML, queue the job and check that the job runs
successfully. If an FTC project, see if the source loads into FTC
and that it acts similar to what the source would be like in local
storage.
Limitation: For Pre-Signed URLs, images cannot be used as file
sources into FTC because FTC image sources in FTC are decoded
as a sequence (Limitation).
Therefore, FTC expects a sequence from the pre-signed URL
source and it cannot find / import.
Target Output ATS files 15748,
15812
This feature is implemented only for H.264 inside TS.
Transport Stream container now has a new checkbox for "Write
EBP Markers" in the Container Format Settings section.
When the checkbox enabled, users are able to set EBP GOPS per
segment and fragment.
ATS metadata, called Encoder Boundary Points (EBP), are
injected and carried in the transport stream layer to provide
adaptive boundary information to downstream processing tasks.
Target Output HEV1 files 15835
For x265 HEVC and NTT's HEVC (only for MP4) we can specify
HEVC Codec Tag for the output file (either HEV1 or HVC1)
Filters Normalize at True Peak audio sample 13902
(Normalizer)
‘Normalize to True Peak audio sample’ option added to the
normalizer audio filter.
Filters EZTitles is supported for Passthrough jobs 13134
(EZTitles)
Retrieval / Akamai NetStorage Retrieval and Post Task Upload 15923,
Post Task (purchase option) 15864
Upload
Limitation:
- Japanese character handling is not currently supported for
NetStorage upload or retrieval.
- Progress bar does not complete in a linear way, so may not be
an accurate representation of upload progress.
Scriptable Allow user-installed Perl packages and executable (Beta) 15791
Workflow for use with Scriptable Workflow.
Please contact support@capellasystems.net for instructions.
Cluster / Launch function for AWS EC2 instances 15465
CloudExtend
(AWS) The CloudExtend tab in Cambria Cluster Manager has this
function that allows users to launch AWS EC2 instances.
Instances launched by this function will automatically be
connected to the Cluster Manager.
Please contact support@capellasystems.net for instructions.
Cluster / Automatically Shutdown AWS instances 15468
CloudExtend
(AWS) CloudExtend setting allows EC2 instances that are launched by
Cluster to auto stop or terminated when idle
Cluster / Tool to dynamically manage number of AWS instances 15469
CloudExtend
(AWS) CloudExtend setting called ‘Dynamic Instance Controller’ allows
users to run a tool that can manage automatic launching of AWS
instances based on:
- The number of slots specified per machine
- The number of queued jobs in the Cluster queue list
- The maximum number of machines specified
Notes:
- Users can also change the AwsDynInstCtrl tool to any other tool
that works with AWS instances.
- For best results, 30 minutes should be the minimum time used
for the Controller interval.
Cluster / Configure On Premise storage to become accessible to 15467,
CloudExtend AWS EC2 15512
(Storage)
When this is configured, a MinIO server will run on the Cluster
machine to allow On Premise storage to become accessible to EC2
instances.
Limitation: Scriptable workflow does not work for CloudExtend
where the jobs run on a machine that does not have direct access
to the sources, even if the sources are located in a location
mapped via the "CloudExtend Storage" feature.
Cluster / Pay As You Go (PAYG) feature has been removed 15470,
CloudExtend (10/2025) 13391
Show Pay As You Go (PAYG) Balance
The PAYG balance is shown in the "Pay As You Go licensing"
section on the CloudExtend tab.
The balance shown is the lowest balance of any machine in the
Cluster machine list that has PAYG enabled.
Manager Automatic Job Cleanup 13261
Options
This setting can be found in the Manager, under the ‘File’
dropdown. You can configure Manager to periodically clean up
jobs. You can specify what job statuses to clean up and how old
jobs need to be to be removed.
FTC Packager Packaging Jobs can be configured and launched from 15738
Watch Folder (Beta)
- There is a new check box to specify "if the incoming group files
are a part of ABR sources" under Source Acceptance tab. It will
be only shown when "Group Files" check box is ON.
- When this check box is ON, group files will be changed to
"%match%_1.*" ~ "%match%_4.*", and Encoding tab will be
changed to only accept Packaging action.
- Packaging action setting allows users to associate files in the
group with layers of output ABR streams. (when the watch folder
has files with something like ABCD_1.ts ABCD_2.ts ABCD_3.ts
and ABCD_4.ts)
- Once a group of files are dropped to the Watch Folder,
Packaging action will be kicked in and ABR stream will be
generated (using passthrough) based on the configuration.
Notes:
Since it is passthrough packaging, please make sure each source
is using different bitrates to avoid conflict between layers. Each
New Features added for 4.6
file needs to be GOP synced/aligned. Current packager does not
enforce this limitation (it does not show up as an error).
Licensing Pay As You Go (PAYG) support added to Hosted License 15777
method
Module Feature Capella
Reference #
Target (Closed Closed Caption exporter added 15245,
Caption) 15229
The new exporter in FTC called "Closed Captions" allows you to
extract closed captions from a source (such as 608, 708, and
Teletext) and output those captions to a specific caption file.
Currently the exporter only outputs to WebVTT format (.vtt).
Limitations:
- The exporter has an option for "Teletext Page Number" which
needs to be used to extract Teletext caption information.
This option is ignored for all other source caption types.
- DVB subtitles are not currently supported
- Japanese characters are not supported for WebVTT filename
output, this will be fixed for 4.8.
Target (MKV) VP9 codec added to MKV exporter 15305
Target SMIL file creation option for Adaptive Streaming v2 15461
(Adaptive exporter
Streaming v2)
The ‘Type’ setting in the ‘Container Format Settings’ section of a
Adaptive Streaming v2 target preset can be set to SMIL (MP4).
Similar to the other ABRv2 targets, you can customize the
bitrate configurations for both video and audio, and customize the
naming conventions of the manifest, media, and subtitle
folder/files. You can also add special metadata to the manifest
which is used on the player side.
Target AV1 codec added to Elementary Stream exporter 15038,
11717
Limitation: Currently AV1 is restricted to use only through the
Elementary Stream exporter. We plan to add it as an option to
the MP4 exporter in the future.
Target (AAC Dual Mono encoding support 15231
Audio)
FTC now supports Dual Mono in AAC. Users have an option to
encode each channel independently from each other in AAC.
To enable this there is a new checkbox under Audio settings:
"Use Independent Channels".
Filter (Color Custom LUT Tables can be used with Color Conversion 15267
Conversion) Filter
In the Color Space Conversion filter, the user can now specify
custom look up tables that the user provides via
config files (in this case .cube files).
In order to do this, select the ‘via 3D-LUT’ option in the filter’s
‘Conversion Mode’ dropdown. There will be a file input field ‘LUT
file’ to specify the custom LUT.
Filter (Color Color Channels Remapper Filter added 15352
Channels
Remapper) The Color Channels Remapper filter can remap the Alpha value to
the Y value of the output. For example if source is YUV-Alpha
then target will be Alpha-UV. If the A to Y mapping value is set
to 128 then output will be grayscale.
Limitations:
- The Color Channels Remapper filter will not work for outputs
with 10 bit depth. Currently only works for 8 bit outputs.
- The Color Channels Remapper filter will not show preview in the
UI in following cases:
a) any source that is 10 bits with alpha
b) any source without alpha
c) if preset editor is not configured yet
Filter EZTitles filter can be used in MPEG-2 TS passthrough jobs to 15052
(EZTitles) insert captions without transcoding.
Post Task Harmonic VOS 360 / VOS CNS integration (purchase option 15322,
Upload / Filter required) 15306,
15084
> Support for VOS 360 Upload. You can access the upload
settings through the Target Preset ‘Post Task’ tab.
> Support for VOS CNS Upload. You can access the upload
settings through the Target Preset ‘Post Task’ tab.
> SCTE 35 inject filter can be used to inject SCTE-35 markers
(via ESAM XML) into the output.
Help document for using the VOS Uploader:
https://www.dropbox.com/s/9wdvmc7c5qs7ee6/FTC%20Harmoni
c%20VOS%20Upload%20Guide.docx.pdf?dl=0
Manager Database Maintenance Manager feature 14811
FTC and Cluster now support "Database Maintenance" upon
service reboot only when scheduled through UI (access the
setting through Cambria Manager → File menu dropdown.
New Features added for 4.5
Maintenance is done through a "vacuum" feature in postgres
where DB size will be reduced (i.e. deleted rows are cleaned up
from the database)
Notes:
- Vacuum will be performed with "best effort" upon next service
start only when DB Maintenance is scheduled.
- Vacuum will not be performed (skipped) when database is not
connectible, or read only, or when POSTGRES or CLUSTER/FTC
services are stopped.
- Vacuum will be waited to be performed, before
servicerunner/webserver starts.
- If user sees error in the UI (such as "unable to obtain info")
then UI should be restarted.
Cluster Multicast Independence for Cluster Redundancy 15341
In previous versions of Cluster the network needed to allow
Multicast in order for Cluster redundancy to work. This
requirement is now removed for this Cluster 4.6.
Floating New tool Floating Server Monitor 15025
License Server
There is a new tool that can be used with Floating Server
Manager. The tool can be installed on any machine, it will watch
the Floating License Server and can send out notification alerts
based on various conditions:
- Floating Server Manager does not respond error
- License Expires soon (10 Days)
- Activation status becomes invalid
- Number of Agent Machines less than number indicated
Dashboard Additional metrics and Installer for Prometheus/Grafana 15266,
15265
Documentation:
https://www.dropbox.com/s/b3n9yvkd7wfytpu/Capella%20Monitorin
g%20System.pdf?dl=0
Module Feature Capella
Reference #
Target VP9 codec added to MP4 and Elementary Stream exporters 11889
(MP4,
Elementary
Stream)
Target Dolby Audio support added 14484
(Adaptive
Streaming v2)
Target 12328
(Adaptive I-frame only playlist support for HLS
Streaming v2)
New Features added for 4.4
A checkbox has been added to enable this for HLS. It can be
found in the Adaptive Streaming v2 Exporter in the ‘Container
Format Settings’.
Target Black Segment Detection added to analysis exporter 14901
(Analysis)
This setting can be found in the Analysis Exporter in the
‘Container Format Settings’ section under ‘Video Contents’.
Source/Target FTPS retrieval and post task delivery support added 14319
Filter Color Space Conversion filter enhanced 14448
BT.709 to HLG (and HLG to BT.709) can be done in one step.
The new option has been added to Color Space Conversion filter
in the ‘Conversion Mode’ options.
Filter Nexguard V2 added to Target Filter list 14636
This filter uses Nexguard SDK version 1.7.1.
Documentation:
https://www.dropbox.com/s/5qevmx0lay8f4ky/NexGuard%20FTC
%20Guide.docx
Job AWS credentials for JobXMLs retrieved from Manager are now 14373
submission encrypted.
WebUI Web UI added 14738
Cambria Manager includes a selection to quick launch into the
WebUI under the ‘Help’ dropdown.
You can also launch it by going to this link on the Cluster/FTC
machine:
https://localhost:8161/
(The first time you run it, it will show the location of the
instructions to set up user accounts)
Licensing Upgraded to use latest version of Cryptlex API 14883,
14913,
Please refer to the licensing note at the start of the Cluster/FTC 14924
4.5 release notes.
Module Feature Capella
Reference #
Source DivX input supported 14233,
14144
Limitation: Cannot seek DivX files.
Source (FTP SFTP retrieval supported 14140
retrieval)
Target (MXF) Added official support for ARD-ZDF SDF 01/02, HDF 13194
01/02/03 and adhere to corresponding specifications
Target (MXF) XAVC Long GOP supported 14072
FTC now supports XAVC LONG GOP 4:2:0 8 bit, 23.98, 25, 29.97
FPS, will be 100Mbps, 50, 59.94 will be 150Mbps.
Target (MXF) XAVC Intra updated with Class 50 option 14061
Target (AAF) AVCIntra100 and AVCIntra50 support added to AAF 14114
exporter
Target Dolby Digital and Dolby Digital Plus audio has been added 13516
(Adaptive to Adaptive Streaming Exporter
Streaming)
The Dolby setting requires applying a config file. Here are the
steps to generate Dolby Encoding Setting XML for Adaptive
Streaming:
1. Open FTC and go to the "Encoding" tab
2. Create a new encoding preset with any container that has
Dolby Audio available. For example, "MPEG-2 Transport Stream"
3. In the "Audio/Track 1" option, select "Dolby Digital" or "Dolby
Digital Plus"
4. Scroll down to "Audio Settings / Track 1" and modify the audio
settings to your liking
5. Click on the "Save..." button in the Preset Editor and save
the .cen file
6. Rename the .cen file to have a .xml extension
7. Open the file in a text editor and look for the audio
<EncodingSettings> tag
8. Delete everything else in the file except for the audio encoding
settings and all of the contents inside it
9. Save the modified file
10. This file can be used in the ‘Load Config’ button when Dolby is
selected for audio in the Adaptive Streaming Exporter.
Here is an example of a Dolby Digital configuration:
<EncoderSettings
EncoderName="Audio Encoder - AC3"
Type="Audio"
NbOfChannels="6"
BitsPerSample="Same as Source"
BitrateKbps="384"
AudioCodingMode="7"
LFEEnabled="1"
Dialnorm="31"
BitstreamMode="0"
EvolutionFrameworkEnable="1"
SurroundMode="0"
SurroundExMode="1"
LineModeProfile="1"
RFModeProfile="1"
StereoDownmixMode="0"
LTRTCenterMixLevel="4"
LTRTSurroundMixLevel="4"
LOROCenterMixLevel="4"
LOROSurroundMixLevel="4"
UseHDCDConverter="0"
UseLFEFilter="0"
UseSurroundPhaseShift="0"
UseSurroundAttenuation="0"
CopyrightFlag="1"
OriginalFlag="1"
PeakMixingLevel="105"
RoomType="2"
LanguageCode="eng"
Name="AudioEnc1">
<FilterSettings Type="AFilter"
FilterName="Audio Filter - DPLC" Mode="1770-
2+DI" SpeechThreshold="20"
UseTruePeakDCBlock="0" UseTruePeakEmphasis="0"
UseScaling="0" />
</EncoderSettings>
Target (Beta) Adaptive Streaming V2 exporter 13539,
(Adaptive 14484,
Streaming v2 The beta version of the ‘Adaptive Streaming V2’ exporter is 14457,
Beta) available to test and use. This exporter includes a purchase 14373
option to have CPIX/Irdeto integration for DRM. And has more
user controls for folder and file naming schemes, compared to V1.
Also included are DASH Conformance Mode for HbbTV 2.0/DVB-
DASH 2014.
Limitations:
- Dolby Audio output does not currently work with the V2
exporter. Jobs will error with a “MultiBitrate Encoding:
Unexpected Error”. This will be resolved in a future build.
- VBR jobs will fail with “Unexpected Error”. This will be resolved
in the next version.
- JobXMLs extracted from a job with CPIX and/or AWS S3 will
show the credentials in plain text. This will be resolved in a
future build.
Source and Dolby to Dolby transcoding supported 12330,
Target (Dolby 14347,
Audio) Limitations: Audio Delay and Audio Normalizer filters cannot be 14362
used in a Dolby to Dolby transcoding job.
Filter (XML XML Titler support for EUDC and Carbon Coder XMLs added 14380
Titler)
A help guide is available. Please contact support
(support@capellasytems.net) for more details.
Filter Teletrax Watermarking Filter (target side) support 14361
(Teletrax) Requires a license from Teletrax.
New Features added for 4.3
Documentation:
https://www.dropbox.com/s/zxt62k4s932i8c3/How%20to%20Tel
etrax%20Watermarking%20.pdf?dl=0
Post Task Retry attempt value and delay value added to Post Task 13421
network delivery
FTC List of various modules/components that have been 13933,
Module/compo upgraded: 13928,
nents 13492,
- ffmpeg upgraded to v4.2 12942,
13105,
- x265 upgraded to v3.2.1 14317,
14358,
- mp3 encoder updated 14281,
14105,
- Dolby Vision upgraded to SDK v 2.2.0 13071,
14260,
- MXF XDCAM reading from S3 optimization 14193,
14377
- Avid AAF exporter enhancements: Auto delete AAF after check-
in option, limit # of files in folder, additional logging
- Color Primaries/Transfer Characteristics/Matrix Coefficients
options added to DNxHD/DNxHR and ProRes configuration
- ProRes optimization to increase processing speed
- Logo Filter position accuracy increased
- Specify Channel Layout option added to MOV container settings
- Whitelist option added for Watch Folder Source Acceptance tab
- UI option added to switch UI language (English / Japanese /
Chinese)
API Packager (Purchase Option) CPIX and Irdeto Integration for DRM 14175,
PlayReady/Widevine Support 13835
Module Feature Capella
Reference #
Source Growing source files can now be segmented by Timecode 13540
(Growing
source)
Target (Avid Avid AAF Exporter enhancements: 13730,
AAF Exporter) - Added Custom Metadata (Tagged Values) interface 13682
- String replacement variable can be used for the Material Name,
so source name can be maintained
Audio Audio enhancements: 13808,
- Analysis exporter supports loudness measurements for multiple 13329
audio tracks
- Audio description values for Voice Over and Control Track are
used during transcoding with "Audio Voice Over Control Renderer"
filter
UI UI updates: 13242,
- UHD/4K/8K selections added to UI dropdown for x264, x265, 12068,
ProRes, HEVC(NTT) codecs 13571
- The column sorting in Manager occurs across all pages
- New field added for modifying filenames for Adaptive Streaming
MPEG-Dash video streams
API Different credentials can be used for access to files in 13472,
(Credentials) / various locations 15318,
Credential 15345
Manager Credentials Storage
- Credentials are stored into database
- Credentials will hence be persistent across builds, and persistent
across Cluster Redundancy
Credential Manager
- Credentials are entered via Credential Manager
- Credential Manager is identical, and can be found in FTC and in
Cambria Manager
- Credentials will then be written to JobXML (and assigned to
Manager) when jobs are created.
- Removing entries in Credential Manager DOES NOT invalidate
the credentials
- In other words, users who get hold of your JobXML can then
modify the JobXML to R/W files on the path, given that they know
the path
- Passwords are encrypted, but does not prevent impersonation.
UI workflow (limited to reading/writing to location that UI have
access to)
1: FTC -> Settings -> Credential Manager (Or, for WF, Cambria
Manager -> File -> Credential Manager)
2: Put in the path (folder path), user name, password and
domain.
Password and domain is optional, depending on user
environment.
3: Create jobs and run accordingly.
API workflow (craft and submit as JobXML):
1: Then, you can work with any path, as long as you know it
2: FTC -> Settings -> Credential Manager (Or, Cambria Manager
-> File -> Credential Manager)
New Features added for 4.2
3: Put in the path (folder path), user name, password and
domain.
4: Export -> Export All -> Save the ccf file
5: Open the ccf file, you will see lines of "Credentials"
6: When crafting JobXML, Insert "Credentials" as child element of
"Job", and put those lines of "Credential" as child of "Credentials",
eg:
<Job>
....
<Credentials>
<Credential Domain="mycompany.net"
Password="foobar1##" Path="\\network-nas1"
Username="nasuser1" />
<Credential Domain="mycompany.net"
Password="foobar2##" Path="\\network-nas2"
Username="nasuser2" />
</Credentials>
</Job>
Limitations:
Lists of functions that do not work with Credential Manager:
- Retrievals (Network, FTP, etc)
- Uploads (Network, FTP, etc)
- Scripts for Scriptable Workflow
- Third-Party Plugins
API Return TotalItems value that represents the number of 13792
items in the Jobs List
http://localhost:8647/CambriaFC/v1/Jobs/
http://localhost:8649/CambriaFC/v1/Jobs/
Scripts MultiTargetBitrateLadderBasedOnVideoComplexity.pl 13788
script added
It allows multiple individual targets (MP4 or TS) to use source
video complexity values to set target bitrates.
Module Feature Capella
Reference #
Adaptive Adaptive Streaming Repackager (API only) (Separate 13202,
Streaming Purchase Option) 13595
Repackager
Packager test sample link:
https://www.dropbox.com/s/oefcbw7fujhsctd/PackagerTest.zip
PackagerTest notes: The package includes instructions for
creating the MP4 elementary streams in the FTC UI, then using
the API for repackaging.
Source IMF ProRes import 13587
(IMF)
Source Support for lossless codecs: FFv1 and FFvHuff 13379
(AVI and MKV)
Source Enhancement: S3 performance optimization with MOV and 13083,
MXF XDCAM sources 13105
Source and Enhancement: Improved credential handling for S3 to 13081
Target allow different credentials to be used per job.
Please contact Capella at support@capellasystems.net for
instructions.
Target Option added to output fragmented MP4 with byte range 13338
(MP4) external file
Target AVID AAF Exporter creates AVID MXF (MPEG-2 IMX, 13222,
(Avid MXF) XDCAM HD, or DNxHD) and handles check-in to Interplay 13638,
14822
Limitation: When the AVID AAF exported is used in combination
with the FTC setting to “overwrite” target files that already exist,
a crash may occur. Please do not overwrite AVID AAF targets
that exist.
Target Added support for ProRes to MXF exporter 13574
(MXF)
Target Added support for DV (DVCPRO25, DVCPRO50, 11118,
(MXF) DVCPROHD) to MXF exporter 13486
Target Added Long GOP support to MXF (XAVC) 13617
(MXF XAVC)
Target Enhancement: MXF output is usable in Premiere as it 13294
(MXF) grows
Target Analysis Exporter enhancement to allow video matching 13496
(Analysis) detection
Video matching detection setting added Analysis Exporter. This
setting is exposed when setting the Analysis Exporter video
setting to ‘No Encoder’ and checking the ‘Video matching
detection’ checkbox in the Video Contents section. You point to a
video file that us used to match against the source.
The output XML will indicate if a match was found and the
position (in frames) of the matched section. Example output:
<AnalysisInfo>
<VideoMatch DetectionMode="initial" FrameEnd="2834"
FrameStart="2697" Matched="1" LowestPSNR="40.6390"
PSNRThreshold="35" SourceFrameRateDen="1001"
SourceFrameRateNum="30000"/>
</AnalysisInfo>
This feature can be combined with the Scritable Workflow to
identify if the source has a match, then remove (or replace) that
section of video.
Filter Kantar SNAP Audio Watermarking 10338
The watermarking filter is now officially certified. It requires
Capella purchase option as well as license from Kantar.
This can be tested upon request. Please contact Capella at
support@capellasystems.net if you are interested in testing.
Link to documentation:
https://www.dropbox.com/s/no05ii0vkiibn9x/Audio%20watermar
king%20integration%20with%20%27Kantar%27%20guide.docx
Filter Logo filter enhancement for motion logos 7648
Source Ending Handling (Repeat Last Frame / Disappear / Loop)
setting added to Logo filter. This setting is exposed when using a
video (.mov) file as the filter source.
Dashboard Web-based Dashboard/Monitoring/Alerting 13423
Prometheus data output for Cambria is added. Instructions for
setup of a dashboard for monitoring and alerting are included in
the document.
Link to documentation:
https://www.dropbox.com/s/yn3slyqxbzt14is/Cambria%20FTC%2
0Feature_%20Prometheus.pdf
Windows Windows Processor Groups support (Beta) 13313,
Component 13407,
Warning: Cluster/FTC support of multiple processor groups 13690
is limited and we do not guarantee that a multiple
processor group machine configuration is optimal. We do
not recommend that machines with more than 64-cores
are used in production due to performance issues
Multiple Processor Groups are defined by Windows when systems
have more than 64 cores.
A single FTC job can only use 1 processor group.
We have enhanced FTC to distribute individual jobs into different
processor groups to utilize machines that have more than 64
cores.
Virtual Automated AWS clients for Cluster 13365,
Machines 13299
(Cluster/FTC) This is part of our ongoing development for improving Cluster on
premise/cloud hybrid handling.
This can be tested upon request. Please contact Capella at
support@capellasystems.net if you are interested in testing.
New Features added for 4.1
License ‘Pay As You Go’ Licensing Option (Beta) 13365,
(Cluster/FTC) 13299,
This licensing feature can be used to allow customers to pay 13484
based on extra encoding volume instead of adding full perpetual
licenses.
Can be used with Automated AWS clients for a more cloud centric
workflow or to support unanticipated spikes in encoding volume.
Also added ‘Pay As You Go’ balance to Prometheus metrics.
This can be tested upon request. Please contact Capella at
support@capellasystems.net if you are interested in testing.
Module Feature Capella
Reference #
Source S3 optimization for source handling 13030
Performance is improved for sources being read from S3.
Target MXF with Dolby E 13035,
(MXF) 13027
Improved MXF muxer. Also, passthrough optimized to allow for
various Dolby E workflows to be supported:
In the global Transcoding Preferences (File [Symbol] Transcoding
Preferences), there is now "Dolby E Decode" option. This can be
set to "Decode (if licensed)" or "Do not decode".
Target Presets by default will use the global Trancoding
Preferences. But target presets can be modified to not use the
Application Defaults and so this setting can be changed on the
Target Preset also.
Based on the global Transcoding Preferences, Dolby E files will
load in as all PCM if the setting is set to "Do not Decode".
Otherwise, Dolby E will load in normally, which means two source
channels get decoded to one 5.1 Dolby E track.
Target ProRes Target Enhancements: 13042,
(ProRes) 13031
- Custom PAR option added.
- Improved ProRes backwards compatibility with older players.
Target (XAVC) XAVC Encoder Performance Improved 13055
Transcoding speed should be 2 to 3 times faster than the
previous version.
Filter Preroll / Postroll Filter 13041
Allows for preroll and postroll to be added. Type can be black,
image, or freeze frame. Duration can also be set.
Filter BT.709 to HLG Conversion Improvement 12690
The color correction filter has been updated to have a more
accurate conversion algorithm.
Filter Option added to Audio Remapper filter to allow 12868
automapping to match output tracks
Allows for linear mapping of channels from source to target until
all the tracks have a source channel assigned or until there are no
source channels left.
Transcoding Audio Track Handling option allows for silent audio to be 12832
Preferences added to target tracks automatically
(Source File
Handling) In Transcoding Preferences (Source File Handling), there is a new
option for Audio Track Handling. The setting is called “Create
Silent track if used Source Track does not Exist”.
Transcoding Adaptive deinterlacer option 12932
Preferences
(Standards Transcoding Preferences Standards Conversion option added for
Conversion) “Deinterlacer Mode” to all for per frame interlacing detection.
This enables FTC to adaptively handle video that has both
interlaced and progressive segments.
Watch Folder Watch Folder enhancements: 13080,
12913
- Subfolder depth limit increased to unlimited. Please note
that this feature allows for an unlimited depth, however,
other limitations such as Windows maximum path length
may prevent a really long watch folder path from working.
- FTP upload and S3 upload added as Watch Folder Actions.
Notification Notification enhancements: 12940
- New string replacement variables added, hover over the help
(?) for the list of supported variables.
Log Automatically delete logs for Cluster/FTC 12890
Logs are now automatically deleted for Cluster/FTC if it is past 45
days.
Source and Dolby Vision Support (Beta) 12483
Target
This can be tested upon request. Please contact Capella at
support@capellasystems.net if you are interested in testing.
API Role Based API Access (Beta) 11930,
21406
This feature allows for the administrator of Cambria Cluster to set
role based permissions for API access. This can be unlocked for
beta testing for those who request it. Please contact Capella at
Known Issues
support@capellasystems.net if you are interested in testing.
Limitation: When the API is used in role-based mode, GET
/CambriaFC/v1/Jobs/ returns only jobs owned by the
authenticated user unless that user has the Administrator role.
Custom roles with full permissions may still see an empty job list.
This affects the job ID listing API, not the Administrator role.
Here is a list of currently known important issues. Some of which will be fixed in future releases of
Cambria.
Known Issues
Module Issue Capella
Reference
#
File Convert / Cambria currently uses decimal points instead of decimal 9039
Cluster commas. This is the case even in OS languages that
normally uses decimal commas.
Watch Folder Remote Retrieval 7160,
14910
Unless the network retrieval location is shared as Public with
no login necessary, the username and password fields must
be filled in. By doing so, you can ensure that the network
retrieval will still work even after the user logs out.
When doing FTP / Network Retrieval for a Watch Folder, use
the "Subfolders to watch" dropdown to select
the depth of subfolders to retrieve. Otherwise, no content
inside of subfolders will be retrieved by default.
Watch Folder Growing file support 5003
Allows transcoding to begin on a source file that is still being
transferred into the watch folder.
The Allow growing files options can be accessed and modified
through the Watch Folder Source Acceptance (tab).
Limitations: The growing files feature should be used for
files that are growing at a rate equivalent to their bitrate.
This feature is not to be used as a workaround for a slow
network transfer rate. Also, since the full size of the source
file can be unknown, the completion percentage can be
inaccurate.
Watch Folder Window’s folder shortcuts are not a supported input to 6060,
trigger Watch folder encoding jobs. Also, shortcuts created 5987
from files in Windows Libraries folders also are not supported
Known Issues
by Watch Folders.
Watch Folder Maximum number of subtasks for a single job is 8. This 1888
limitation will only apply when ‘Perform actions in parallel’ is
unchecked.
Watch Folder If ‘Submit as Job XML is selected as a Watch Folder action, 13951
no other actions are performed after this selection.
Watch Folder (S3 S3 Retrieval in Watch Folders does not honor the "Retrieve 16383
Retrieval) as S3 shortcut" checkbox. The current behavior is that it will
always retrieve as S3 shortcut regardless of the option being
enabled / disabled.
Note: This issue will be fixed in a future build.
Source (Import Reporting tool for source files that fail to import 10667
error reporting tool)
When source files fail to import into the File Convert Sources
tab, there will be a link at the bottom of the application,
Show File Import Error Log. You can click on this link to
access the ‘Create Report’ function. This tool will create a
folder that contains information about the file and also will
extract the information at the start and end of the file. This
folder can then be sent to Capella for
troubleshooting/support purposes.
Source Import support for IMF files 10283
The source can be loaded by pointing File Convert to the
playlist XML (CPL.xml)
Source Decode source file once when encoding to multiple 9361,
targets 9591,
9592,
When encoding to multiple targets using the same source 9727,
file, you can now configure File Convert to decode the source 9718,
once. To do this use the option on the ‘Encoding’ tab in the 9587
main UI to “Encode Targets as Single Job”. This allows us
the amount resources spent on decoding/deinterlacing to be
reduced and increases overall transcoding speed for the
target group.
1. Conditional Audio Mapping cannot be used with the
‘Perform encoding tasks as one job’ watch folder option.
2. Source side filters cannot be used with the ‘Perform
encoding tasks as one job’ watch folder option.
3. HTTP and Adaptive streaming targets cannot be used with
‘Perform encoding tasks as one job’ watch folder option or
‘Encode targets as single job’ File Convert UI option.
Source Analyze Source Function 8221,
8133,
The ‘Analyze Source’ button can be found in the main UI in 10302
the source tab. This function allows the application to 10196
display more information about the source. (muxer used,
video format, audio format, resolution, frame rate,
interlacing, presence of closed caption, presence of
timecode, bitrate, GOP size, number of B-frames, closed
GOP/Open GOP, and many file-specific things like TS PIDs,
H.264 settings (CAVLC/CABAC), etc.)
Analyze source also has an option to allow an Encoding
Preset to be quickly created based on the analysis.
Limitations:
- Create Preset option does not work on all sources. MXF
(XDCAM) sources cannot be used with this function.
- If a source is detected as a High-10 profile and 4:2:2 10-
bit, the create preset function will create a 4:2:0 10-bit
preset. This occurs because High-10 does not support
4:2:2.
- The MPEG-2 decoder will always decode the file with the
color format YUV 4:2:2. Because of this, the video
properties will always show YUV 4:2:2 for all MPEG-2 videos.
- Some files, especially longer files may take a long time to
analyze. You can modify the ‘Analysis Duration’ for the
options tab in the main UI. Setting this to ‘Partial’ will limit
the analysis to the first 2 minutes of the file. However,
setting this to a shorter analysis segment may reduce the
accuracy of the detection for the bitrate setting.
Source Reported duration for HEVC (TS) source files can be off by 3 7499
to 5 seconds.
Source HEVC Decoder 6885
Limitation: HEVC sources may not be seekable, therefore
scrubbing through the Segments Editor timeline will not
work.
Source Sources with timecode that does not increase linearly can 6085
cause the progress of the job to be reported inaccurately.
Source 4k support 5748
Limitations: 4k output is only supported with HEVC and
x264 targets. In order to use 4k resolution, the Level in the
encoder UI must be set to 5.2 or higher. 4k input formats
that have been tested include HEVC, Grass Valley HQX,
Grass Valley Lossless, ProRES 422, and ProRES 4444.
Recommendations: 4k conversions are memory intensive.
Here are some recommendations to avoid problems related
to Windows 32-bit memory limits.
- It is recommended to use the 64-bit FTC UI for 4k
conversions. (C:\Program Files
(x86)\Capella\Cambria\cpx64)
- When using the File Convert UI, it is recommended to use
Queue instead of Convert to start a job.
Source Sources with multiple video streams are unsupported and 4183
will now fail with this error: “Configuration error: Audio-
Video Source. Unsupported source –contains multiple video
streams”.
Source Sources with a mix of interlaced and progressive pictures are 14905
not supported. The job will fail with an “unexpected internal
error” message.
Source (ProRes) ProRes Source Error Tolerance 11814
Source Error Tolerance now works for ProRes sources. The
user can specify how many source frames to ignore (per
minute) and continue decoding when FTC encounters certain
errors in the source file. This enables encoding to continue,
but may result in duplicated or skipped source frames.
The Sources Error Tolerance configuration can be found in
‘Preset Editor’ → ‘Transcoding Preferences’ → ‘Other
Settings’.
Source Uncompressed ‘raw’ RGBA MOV files show artifacts 10858
(uncompressed) during decoding
Note: We are looking into fixing this for a future release.
Source (IFO - Source IFO (PS/DVD) files usage and limitations: 3492
PS/DVD)
How to use:
Go to Video_TS folder (which contains the IFO/VOB files),
and
import the IFO that starts with "VTS". Using VIDEO_TS.IFO
will not work. Users should not import all VOBs manually, the
IFO loader will concatenate all of the VOBs automatically.
Limitations:
1: IFO: The accuracy of source duration is relied on IFO. In
our observations, it is often a few frames off of actual source
duration through frame counting.
2: VOB/PS: The accuracy of source duration is relied on the
accuracy of PTS (Presentation Time Stamp). If the PTS is
inaccurate, detected source duration will be different than
actual source duration through frame counting.
3: Does not support files that do not have system header
(Spec required to have at least 1).
4: IFO Demuxer does not support accurate seeking for
23.976 Pulldown sources. There may be a few frames off
when we seek/segments.
5: IFO Demuxer does not support segmenting by chapters. It
will demux all chapters found in the main title in first PGC.
6: When demuxing IFO, we do not have "FBI Warning"
sections (VTS_0X_0.VOB) since that is not included as
valid/used sectors in first program chain. By we only demux
sectors that are referred in first program chain.
7: Does not support DVD that have multiple main program
chains, we always pick first one.
8: Does not support Closed Captioning subtitles (CC). We
can only support bitmaps subtitles (indicated by IFOs).
Source (MPEG-2 If the timestamps of the source are continuous and accurate, 3513
TS/PS) and the file is seekable, then we are frame accurate. If the
timestamps are non-linear and/or non-accurate, then our
duration will be accurate within an average of 2-10 seconds.
Source (WMV) Windows Server 2012 R2 requires Windows Media 12313
Foundation
and Desktop Experience to be installed for WMV support.
Source Seeking in the Segments Editor can be off by 1 frame for 6108
(Segments Editor) DirectShow sources.
Source When Map Audio Tracks is used. The duration of the audio 9292
(Map Audio Tracks) for the source will be modified to match the duration of the
shortest audio source. This can result in the loss of audio for
longer audio assets used in the same Map Audio Tracks
configuration.
Source (Analysis) For files we load using 3rd party SDKs, like Windows Media 13913
or DirectShow, the reported properties of the file are what
our decoder receives, which can be different from the native
properties of the source.
Ex:
GV codec - Audio sample is 24-bit but 16-bit is displayed.
WMV - Color format is YUV4:2:0 but 4:2:2 is displayed.
Source (Audio When encoding from a source file with different length audio 12693
Duration) tracks, the longer audio duration will be used.
Source (Watch Watch Folder Support for XML Job Submission 8323
Folder)
Watch Folders can accept job XML files as long as the Watch
folder Action is set to “Submit as job XML”.
Requirement:
Job XMLs that are submitted must have <JobDescr> as the
root element.
Example:
<JobDescr Priority="5" NumberOfRetries="3"
Description="File to file conversion 123"
Submitter="10.12.0.155">
…
</JobDescr>
Target/Source Read and write to Amazon S3 12724,
(Amazon S3)
Using Amazon IAM user authentication, FTC can access
Amazon S3 buckets. Source files on S3 can be imported into
FTC via JobAPI submissions and FTC can write directly to S3
while encoding.
Limitations:
- Some source formats cannot be read directly from S3.
- Some output formats cannot be written to S3 while
encoding, these outputs will be written locally first then
transferred to S3.
Source and Target 8K Support (purchase option) 8811
Limitations: 8k output is only supported with HEVC targets.
In order to use 8k resolution, the Level in the encoder UI
must be set to 6.0 or higher. 8k input formats that have
been tested include HEVC, TIFF, DPX, Uncompressed YUV.
Recommendations: 8k conversions are memory intensive.
Here are some recommendations to avoid problems related
to Windows 32-bit memory limits.
- It is recommended to use the 64-bit FTC UI for 8k
conversions.
- When using the File Convert UI, it is recommended to use
Queue instead of Convert to start a job.
Source and Target 10-bit/12-bit/16-bit video Support 7361,
7127,
10-bit supported formats: 8389
8526,
Source and Target Formats: 8539
ProRes422 (QuickTime MOV)
ProRes4444 (QuickTime MOV)
Uncompressed (QuickTime MOV)
Uncompressed (AVI) * source only
HEVC (Elementary, TS, MP4)
H.264 (Elementary, TS, MP4)
DNxHD (MXF, MOV)
AVCIntra (MXF) *source only
XAVC (MXF) *source only
12-bit supported Sources and Targets:
TIFF *source only
HEVC (Elementary, TS, MP4)
Conversion Modules:
Scaler
Interlacer (*Deinterlacing is only 8-bit)
Frame rate converter
Pulldown
10-bit/12-bit/16-bit Source and Target Filters:
Crop, Logo, Text Burn In, Timecode Burn In
Note: Video data being processed in 8-bit if any module
used is 8-bit and the resulting output file will be 8-bit. *For
Example: if Deinterlacing is used in transcoding process,
output will be 8-bit.
Source and Target DVB and Teletext Support 8659,
8621,
DVB subtitles and Teletext from TS sources can be passed 8474,
through from source to target. In the TS Exporter, Container 8471,
Format Settings (Preset Editor → Encode [tab]), check the 8249
appropriate setting. Write DVB Subtitles or Write Teletext.
DVB subtitles and Teletext from TS sources can also be
burned into the video prior to encoding. You can do this
through a ‘DVB Burn-in’ or ‘Teletext Burn-in’ source filters.
Known Issues / Limitations:
- DVB subtitles with multiple languages on one DVB stream
are not supported. Only the first language will be used.
- Some subtitles may appear in incorrect positions with the
burn-in filter.
Target (H.264) Encoding parameters for H.264 are clamped to the range 1047
allowable for the specified Profile / Level and Conformance
Mode. Some of these restrictions on parameters are not
visible to the user and will occur without warning.
Target (MP4) MP4 files created at 29.97fps will display 59.94 for the frame 4226
rate field in VLC’s codec information section. VLC is showing
field rate information instead of frame rate.
Target (MP4) MP4 files output files that are longer the 7 hours may be 3494
created with a file duration that is incorrect. This could
cause the files not to open up in certain media players or
cause other unwanted effects.
Target (MP4) When using the ‘Place moov at start’ option, File Convert will 3194
create a temporary file, which gets deleted when the job
ends.
Target (MOV) Write timecode track in MOV 5877
Requirement: ‘Write Timecode Track’ checkbox in the MOV
target encoding settings must be checked for the timecode
track to be written.
Target (MXF) MXF Muxer requires all channels to be contained in the same 6664
track. Use the Map Audio Tracks filter on the source file,
when needed, to map source audio channels to 1 track to
use for a MXF target.
Target (MXF) MXF files that are created in exactly this setting (1280x720, 5938
4:2:0, 50P, 35Mpbs) will not always transfer to the XDCAM
deck.
Target (MXF) Some combinations of settings for our XDCAM exporter can 3119,
produce output files that do not playback in Sony XDCAM 11777
Viewer. Our MXF exporter allows for users to configure files
that are out of MXF specification. It is up to the user to
configure “in-spec” outputs.
Target (JPEG2000) JPEG2000 Export (purchase option) 9759,
9756
Limitations:
1. RGB444 color format option is not officially supported.
Using this setting may result in a problematic output file.
2. 4K JPEG2000 targets are not officially supported by FTC.
If a JPEG2000 job is configured to a 4K resolution, the job
will attempt to execute, however, it is possible that the job
will run out of memory. If this occurs, you can try changing
the ‘Encoding Threads’ value to 4 and requeue the job.
Please contact Capella if you are interested in 4K JPEG2000
encodes.
Target Restrictions removed for number of audio channels 10301,
(XDCAM) per track for MPEG-2 XDCAM MXF 9563
The exporter also allows users to configure any number of
audio channels between 1-8. This can lead to creation of
MXF XDCAM HD files that do not conform to any
specification.
Known Issue: Previous restrictions have been removed,
File Convert no longer automatically makes audio settings
adjustments to conform to the Sony XDCAM spec. Please
refer to the chart below the recommended settings for
certain configurations:
1. XDCAM MPEG-2 HD 420
1440 x 1080 - 2ch 16bps/ 4ch 16bps
1280 x 720 - 4ch 16bps
2. XDCAM MPEG-2 HD 422
1920x1080/ 1280x720 - 8ch 24bps
3. XDCAM MPEG-2 IMX
8ch 16bps
4ch 24bps
4. XDCAM DVCAM
4ch 16bps
Target (MPEG-2) Timecode MPEG-2 video (GOP headers) usage and 3588
limitations:
As long as File Convert can read the timecode from a
particular source, file convert will pass source timecode
information to the target given that the source and target
frame rates match. If frame rate differs, File Convert will
pass the starting timecode (minus the frame number info)
from the source to the target and automatically make the
timecode increase linearly from that point.
There is also filter function that allows users specify the start
time code manually.
Target (WMV) WMV encoding is slower in 64-bit than in 32-bit. Since FTC 11939
now defaults to submitting jobs in 64-bit. Encoding may be
slower than before.
Note: This is scheduled to be investigated for a future
release.
Target (WMV) Modifying file extensions for the Windows Media Exporter 8691
Changing the file extension through the preset editor to an
extension not recognized by the encoder may cause the
transcoding job to fail.
Extensions that have been verified to work: .wmv, .wm, .asf
Target (WMV) When using a custom PowerToy .reg file, if b-frames or 5412
lookahead is used, the first frame of the output is likely to be
duplicated. For maximum quality, both b-frames and
lookahead should be used.
Target (AVI) Canopus HQ AVI export support 5922
Limitation: Output files don’t have interlacing and aspect
ratio information, because the AVI standard doesn’t handle
it. Different tools like Edius and Premiere use their own
made up extensions to store this, so users may have to
overwrite these values when they import the files, for
example into Edius.
Target (AVI) Canopus DV and DVCPro50 outputs only support 720x480 5542
and 720x576 output resolutions. If you output to any other
frame sizes you will get an “Error multiplexing streams …”
error.
Target Adjusting audio speed for audio only output 11042
(Audio only)
Audio only targets do not have encoding parameters for
frame rate. However, the source speed can be adjusted with
the ‘speed adjustment’ source filter.
Example: If you a source video at 24fps and would like to
create audio only file to use with a 25fps video of the same
content. You can apply the ‘speed adjustment’ filter to the
source and set the speed percentage to 104.2%
Target Video passthrough support and limitations 5170,
(Passthrough) 15519
Video essence can be passed from source to target without
decoding for certain containers and codecs.
Containers that are supported:
MPEG-2 Transport Stream
MPEG-1/MPEG-2 Program Stream
MP4
MOV
MXF
Elementary Stream
Video streams/codecs we can pass through:
MPEG-2
H.264 (including x264)
DV
DVCPRO
DVCPROHD
ProRES
VC-3 (ie DNxHD)
Limitations:
- Cannot apply video filters.
- If the sources files are I-frame only then segment output
will be frame accurate. If source files are long-GOP and the
segment in point is not a key frame, the in point will be set
to the previous key frame.
- There is no video preview while transcoding.
- XAVC Intra VBR source cannot be used in MXF Passthrough
Target The Transcoding Preference: Source Speed Adjustment 8678
(Passthrough) cannot be allowed when using audio passthrough. The
Source Speed Adjustment setting must be changed to
‘disallowed’ for the Job to complete without error.
Target NTT HEVC quality improvement parameter files 10950,
(HEVC)
Three configuration files have been added that can be used
to improve the quality of HEVC output.
These configuration files can be loaded through the
‘Encoding Parameter File’ setting found in ‘Video Settings’
section of a HEVC target.
The configuration files can be found here:
C:\Users\Public\Documents\Capella\Cambria\Config
HEVC_CBR_Type100.cfg: Used to avoid underflow errors
HEVC_CBR_Type101.cfg: Used with CBR encodes
HEVC_VBR_Type100.cfg: Used with VBR encodes
Target Known Adaptive Streaming v1 limitations 14444,
(Adaptive 14493
Streaming) - Smooth streaming sub-jobs will stall at 99%. The
workaround is to use ‘run as a single job’ option.
- Cluster may not work well with Adaptive Streaming jobs.
As a workaround please enable ‘Encode Layers as a Single
Job’ or use ‘Adaptive Streaming v2”’ exporter.
- An “Unexpected Internal Error” will occur if Black Segment
Remover Filter is used with an Adaptive Streaming target.
- Apple HTTP Streaming validator will show errors for HLS
target.
- Adaptive Streaming v1 with Dolby audio should disable the
default checkbox “Encode Layers as Single Job” to avoid a
configuration error.
Target Encode all Adaptive Streaming layers as single job 10663
(Adaptive
Streaming) ‘Encode Layers as Single Job’ setting allows Adaptive
Streaming jobs to be optimized for source decoding,
interlacing/deinterlacing, and framerate conversion
operations. The setting can be found in the ‘Container
Format Settings’ section in the Preset Editor for Adaptive
Streaming targets. It is enabled by default.
Target Adaptive Streaming layers can be configured to have 10630
(Adaptive different frame rates
Streaming)
Limitation: It is recommended that layers with the lower
frame rates are set to use the H.264 encoder. If all layers
are using HEVC and you are also using mixed frame rates,
there may be restrictions on the GOP sizes for your layers.
Please contact Capella for more info.
Target Adaptive Streaming (HLS) Video and Audio layers are 10888
(Adaptive linked
Streaming)
You can have multiple video and audio layers. For HLS we
automatically link the audio layers to the video layers. The
lowest audio layer (by bitrate) is linked to the lowest video
layer. The 2nd lowest audio layer is linked to the next video
layer and so on. Once the highest bitrate audio layer is
reached, it is linked with the remaining video layers.
Target (Mpeg-DASH, Adaptive Streaming Packaging 12327
HLS)
FTC can be used to package MPEG-DASH and HLS, without
re-encoding the source.
- This feature can only be used via JobXML API submission.
- Sources must be MP4 segments with GOP sync and same
framerate.
- Job Type must be set to "MultiBitrateMux".
- Sources must contain 'Source' child elements, with
'Location' set to the different source layers
- SegmentNamingScheme should not have “_bitrate”, but
needs to have %i.
We can send sample JobXMLs to you, please email your
request:
support@capellasystems.net
Target (Transcoding When using the Transcoding Preferences option to move or 6031
Preferences) copy source files when transcoding failure occurs, it will
always overwrite any file in the move/copy target location.
Target (Post The Post Conversion Task (Command Line) occurs before 12682
Conversion any Upload tasks
Command Line)
Target (Post If the post-task upload is used for a Adaptive Streaming 11037
Conversion Task / (HLS ) target with AES encryption, the AES keys are also
HLS AES keys) uploaded. Be cautious of uploading them to a web server
that is not secure.
Target (Analysis Video Text Recognition (OCR) Enhancements 12694
Exporter)
The Optical Character Reading (OCR) feature added to the
Analysis exporter since FTC version 3.5. OCR allows text
found in videos (Ex: scrolling end credits) to be identified
and written as text to an output XML. This XML is searchable
and can be repurposed/integrated into other workflows.
Text translation is not expected to be perfect.
Enhancements for version 4.0:
• Improved detection speed.
• Additional tuning parameters added.
• Output XML now contains x an y position information for
the text.
Basic Use Instructions :
1) Load the sources into FTC that you would like to analyze.
2) Use the source Segment Editor to mark and In/Out point
for the segment you would like to analyze.
(this step helps to reduce the time it takes to produce the
XML output)
3) From the Target tab select the Analysis Exporter.
4) Uncheck the options found in the Video Complexity
Measurement
(these settings are not related to OCR, leaving them checked
can slow down your analysis processing)
5) Check the ‘Use OCR’ box.
6) Specify the ‘Data’ folder.
(this files in this folder are used in the text analysis
processes. The Data Folder can be found in FTC installation
package. Point the setting to the “…\Data” folder)
7) Set the 3-digit Language code. (eng for English, fra for
French, deu for German, spa for Spanish, ita for Italian).
8) Once done with the setup of the preset, you can queue
the job as normal.
9) The output XML will contain the text recognition results.
Target (Still Image Scene Detection export requires x264 to be licensed. 13691
Exporter)
Target (Post YouTube Upload 3917
Conversion Task)
Any target preset can be setup to upload automatically to
YouTube. To enable the YouTube upload, you will need to
add YouTube as a destination address through the Post Task
(tab). Twitter and Facebook notifications can also be
configured as part of the post task to tweet or post upon
completed delivery.
Target preset recommendations:
1) It is recommended that your File Convert encoding preset
is configured to a high quality, since the file will be re-
encoded by
YouTube upon receipt. Keep in mind that there is a 128GB
limit (per source).
2) Use Progressive.
3) Do not add black bars, keep original aspect ratios.
4) Use Closed GOP.
Security:
When configuring YouTube upload as a Post Task option,
users will be required to enter their YouTube account
credentials. YouTube will return an authorized credential that
allows our application to act on behalf of users. We do not
store YouTube username and password in File Convert
presets or project, but we will store the returned authorized
credential in encrypted form.
Limitations:
1) When we attempt to upload identical video, YouTube
returns 201 Created but will show "Rejected (duplicate
upload)" when checked in browser on the user's channel.
2) Does not support upload resume.
Target (Post Facebook Upload 14630
Conversion Task)
Video Guidelines:
https://www.facebook.com/help/218673814818907
Recommended video dimensions is 1280 x 720 for
Landscape and Portrait.
Minimum width is 600 pixels (length depends on aspect
ratio) for Landscape and Portrait.
Landscape aspect ratio is 16:9.
Portrait aspect ratio is 9:16 (if video includes link, aspect
ratio is 16:9).
Mobile renders both video types to aspect ratio 2:3.
Max file size is 4GB.
Recommended video formats are .MP4 and .MOV.
Video length max is 120 minutes.
Video max frames 30fps.
Limitations:
With Cluster/FTC 5.1, there is a known issue where users
may run into an “Authorization Error” when credentials are
first entered. The workaround for this issue is to click “ok”
on the message and then go through the authentication
process again. Authorization should work on the subsequent
attempt.
Target Gmail mail account settings change to allow for File 10087
(Notifications) Convert and Cluster notifications to work with account
Gmail accounts need to have account settings modified to all
for less secure apps to access your account.
Here is a link to the Google webpage for instructions on how
to change the setting:
https://support.google.com/accounts/answer/6010255
(Follow the steps detailed for Option 2)
Target HTTP notification 7543
(Notifications) http:// is a required prefix for the ‘destination address’ field
for http notifications to work.
Target Create XML log for each job completion or transcoding 8339,
(Notifications) error / string substitutions list 10258
In Notification Settings (Preset Editor → Notification [tab]),
you can now select ‘Create Log’ for the Notification Type.
Steps to Setup:
1. From the Notification Settings dialog, enter in a Name.
(Ex: Job Completed!)
2. Choose “Create Log” for the ‘Notification Type’.
3. Choose a ‘Notification Event Type’. (This settings
specifies the trigger that will create the XML log)
4. Specify a ‘Location’ of where the logs will be written to.
5. Enter in the ‘Content’ of what the log will contain.
If you are writing XML logs then here is an example of an
entry that can be used:
<?xml version="1.0" encoding="utf-8"?>
<Log>
<Info OutputFilename = "%outputfilename%" Source =
"%sourcefilename%" EndTme="%jobendtime%"
SubmitTime="%jobsubmissiontime%"/>
</Log>
If you are writing text logs, here is an example:
Job Completed!
JobStatus: %jobstatus%
JobType: %jobtype%
SourceFilename: %sourcefilename%
OutputFilename: %outputfilename%
JobSubmissionTime: %jobsubmissiontime%
JobStartTime: %jobstarttime%
JobEndTime: %jobendtime%
JobRealtimeSpeed: %realtimespeed%
Here is a full list of string substitutions that can be used in
the content section:
%jobid%
%errorid%
%errormessage%
%ftpurl%
%jobstatus%
%jobtype%
%sourcefilename%
%outputfilename%
%jobsubmissiontime%
%jobstarttime%
%jobendtime%
%realtimespeed%
%numberoferror%
%numberofretriesleft%
%lasterror%
%progressatlasterror%
%description%
6. Make sure you check the checkbox for ‘Is Content XML’ to
specify that you want the output to have an .xml extension.
Unchecked means the output will be .txt.
Known Limitations: String substitutions will not work
when using the conversion button in the File Convert GUI for
transcoding. It is recommended that you “Queue” the job or
use Watch Folders. Also, when the ‘Notification Event Type’
is set to ‘On Job Start’, the duration shown in the log will be
00:00:00 because the database does not have the updated
source duration at that point.
Target Job stalls due to scripts 12591
(Scriptable
Workflow) During script phase, our job's progress does not increase,
this is detected as 'no progress' internally. This means if this
time exceeds the Job Timeout value, we will stop the job as
stalled. You can increase your job timeout to a longer
duration to avoid this issue.
Target Limitation for Single jobs with multiple targets using 11983
(Scriptable scripts
Workflow)
Only the scripts from the first target will be used. All scripts
in all other targets will be ignored.
Target (audio only) Audio only targets may have audio sync issues since there is 14779
no video to synchronize with. If the audio only output is part
of set of targets the sync issue can be avoided by encoding
the targets using the “encode targets as single job” option.
Filter (Target Side) Target side video filters are not applied with ABR v2 targets. 15933
This will be fixed in a future release.
Filter ITU-R BS. 1770-3 Loudness option added to Audio 9396
(Audio) Normalizer
This option allows for audio to achieve EBU R128 compliance.
Known Issue: Audio bitrates at 128 kbps or lower can
cause the true peak to exceed the maximum level that is
specified by the filter.
Filter Subtitle burn-in filter does not support all languages 9517
English, Traditional Chinese, Indonesian, Korean, Malaysian
subtitles are all confirmed to work. However due to a
limitation we cannot correctly display characters for
Vietnamese(VIETN4), Arabic(ARABI3) and Thai(THAI01). In
these cases, we will now show the error “Unsupported
language in subtitle file”.
Note: File Convert supports a 3rd party solution for handling
a larger set to caption formats, EZtitles plugin:
Here is the link to the product info:
http://www.eztitles.com/index.php?page=ezp-cambria
Free trial for their File Convert plugin:
http://www.eztitles.com/index.php?page=download-free-
trials
(EZtitles can be used to workaround many of the limitations
of File Convert’s native subtitle format handling)
Filter 608/708 closed caption burn-in 6976
The Closed Caption Burn-in filter will read the 608/708
captioning data from the source and burn the captions onto
the video picture of the output.
Limitation: When used as a filter on the target side, the
frame rate of the source and target must match, otherwise
the captions will not be burned into the target. This is not a
limitation when the filter is used on the source side (or when
adding a source side filter to the target).
Filter DVD Subtitle Burn-in (from a DVD IFO source file) 4134
Available settings:
1. Subtitle Language - DVD's can contain multiple subtitles.
There can be English, Japanese, and/or Spanish subtitles in a
DVD.
Choose the subtitle language you want displayed in this
section.
2. Subtitle Language Index - Each language may contain
different subtitles. For example, if Chinese is picked from the
Subtitle Language there can be two ways to write the
characters. Index 0 is chosen by default and will always
contain subtitle information if the Subtitle Language is
supported by the .IFO. To choose the second subtitle style,
choose Index 1.
Limitations:
1. If the IFO contains lots of CHG_COLCON subtitles in a
"burst", we may consume lots of memory.
2. We currently do not support previews of subtitle
information.
Filter Logo filter preview does not display logo when ‘apply to 8899
(Preview) segment’ is used.
Filter Filter Preview Enhancement 6601
(Preview)
Filters can now be previewed all at once. By default, all
filters are previewed. You can right click on the filter
preview and select or deselect filters that are included in the
preview. Preview for different filters can also be
enabled/disabled during playback. Some filters cannot be
previewed and are not included in preview. A “*” is shown
next to the current filter being edited.
Note: Source filters and target filters are not previewed at
the same time. When editing a source filter, the preview will
only show source filters. When editing target filters, the
preview will only show target filters.
Filter (Timecode Multiple timecode sources 15991
Burn-in)
In the case that there are multiple timecodes in a source, the
timecode burn-in filter will now choose the most significant
one.
For example, the user has a source with MPEG-2 GOP
timecode and VITC timecode. In our case, the VITC timecode
takes precedence over the MPEG-2 GOP timecode and so the
Timecode Burn-In Filter will use the VITC timecode.
The ordering of the filters matter in this case as the burn-in
filter chooses the most significant one.
For example, if the user wants to burn-in the VITC timecode,
the VITC extractor filter should come before the timecode
burn-in. If you want the MPEG-2 GOP timecode and VITC
timecode burned in for example, you'll want to first have the
timecode burn-in filter for the MPEG-2 GOP timecode, then
the VITC extractor, and then the timecode burn-in.
Filter (Overwrite Timecode overwrite filter can make timecode modifications 14446
Timecode) that are applied through the muxer, so timecode that are
inside of the video stream will not be affected by this filter
for a Passthrough job.
Manager Removing a job in the “Starting” state will delete the job 9598
from the Job Manager. However, the CPJobExec for the job
may still be running in Windows Task Manager. Since the
“Starting” state occurs only for a few seconds at the start of
a job, this is a rare issue.
Manager When there is multiple Cambria Manager user interfaces 6565
open on the same machine, the changes made in one UI
may not be reflected in other UI. This may lead to some
confusion.
Workaround: You can close the Cambria Manager UI that
is not updated. Newly launched Cambria Manager will
contain all the updated changes.
Manager Option to control job timeout 5982
A timeout setting has been added to the Cambria Manager
Option (tab) to allow for increasing the timeout duration to
avoid job failure caused by slow network speed.
Cluster Cluster communication uses HTTPS by default as of 11803
FTC 3.5
While HTTP connection is still possible now, HTTP connection
may be deprecated/rejected in a future release.
Recommendation: It is recommended that external
applications that are interfacing with FTC/Cluster web
servers via HTTP are retooled to use HTTPS.
Cluster Machines that are connected “online” to the Cluster with 9738
transcoding slots set to 0 may still accept transcoding jobs.
These jobs will remain stuck in a queuing state. In order to
avoid this problem, any machines with 0 transcoding slots
set should be disconnected (set to “offline”) from the
Cluster.
Cluster Converting a Cluster machine to a FTC machine 9667
When a Cluster machine is being used in a Backup
redundancy role, the role must be set to “no backup” before
the Cluster is uninstalled. If this is not done, then when FTC
is installed on that machine, the database will be in read only
mode and jobs will not function properly on the machine.
Cluster Cambria not compatible with VirtualBox / VMware 9010,
6599
Cluster and File Convert instances are not compatible with
(VirtualBox / VMware Workstation / VMware Player).
Machines that have (VirtualBox / VMware Workstation /
VMware Player) installed may cause File Convert not to show
up in Cluster properly, in addition, File Convert may not be
able to receive jobs.
Note: To avoid this issue FC/Cluster should not be installed
on the same machine that (VirtualBox / VMware Workstation
/ VMware Player) is installed on.
Cluster Cambria Cluster and Cambria Broadcast Manager 8635
cannot be installed on the same machine.
Cluster It is generally recommended for Cluster installations that the 6517
same build version for both Cluster and File Convert are
used. If you need to use them in different versions, please
contact Capella.
Cluster If the HTTP response from the client machines takes longer 6039
than 1 second (client machines are located far away or have
non-optimal internet connection), the Cluster manager may
unable to set client's number of encoding slots.
Cluster When a machine is added through IP manually to the 6038
Cluster, the connection is maintained off of the exact IP
entered. If the client machine changes its IP address, then
you will need to manually delete the machine from the
machines list and add it again.
Cluster When Cluster Manager shows a lot of "Cluster could not 3341
(Jobs) connect to FileConvert machine to Add new job "errors for a
particular machine, the causes could be:
1)temporarily network issues (disconnected)
2) machine is rebooting
Recommendation: If you are not running into case 1 or
case 2. Rebooting the client machine can resolve the
problem.
Cluster (Job Special Job Tags to prevent over distribution of Jobs to 11601
Management) a single FTC node
Dynamic slots does a good job of distributing tranconding
Jobs to the client FTC machines. However, there can be
cases where FTC can be oversaturated with lots of Jobs.
This can occur when jobs use less CPU at the beginning, and
then increases CPU use substantially later. During this
period of low CPU use, Cluster may assign too many jobs to
the FTC client. A new special Job Tag can be used to prevent
this from occurring.
The Job Tag ‘CPUBaseline=X’ can be used, where X is a user
specified number from 1 and 100. When the Job is running,
the X value will be used as the minimum reported CPU
utilization value for that Job. So, value X or the actual CPU
utilized for the job is used, whichever is higher. Cluster will
use this reported CPU utilization of all the Jobs running on a
FTC machine to decide whether or not to distribute any
additional Jobs to the machine.
Ex: CPUBaseline=80
(When a Job with this tag is running, it will report to Cluster
that the Job is using 80% CPU utilization at the very least. If
the Job is using more than 80% CPU, the higher value will be
reported)
In addition to using the CPUBaseline tag, a new special
Machine Tag has been introduced to allow CPUBaseline to be
dynamically adjusted depending on what machine is running
the Job.
The Machine Tag ‘CPUBaselineAdjPct=X’ can be set through
the Cluster Manager for any Machine in the Machines list. X
can be any number from 0 to 100. CPUBaselineAdjPct is
used as a scaling factor for CPUBaseline. This allows the
user control over the CPUBaseline value across machines of
different processing capability.
Ex: When a CPUBaseline=80 Job is sent to a Machine with
CPUBaselineAdjPct=50, then the actual baseline value is
scaled (80 * .5) = 40.
Cluster (Redunancy) Cluster Redundancy 8426,
8804,
A second Cluster Manager can be configured as a Backup 8945,
Cluster. When the active Cluster becomes unavailable the 9042,
Backup machine will take over and assume the Cluster 9046
Manager’s role. Watch Folders will continue to be monitored
and jobs distribution will continue. This feature adds
redundancy and eliminates having a single point of failure for
the Cluster.
Limitations / Known Issues:
When Redundancy is triggered by rebooting the Redundancy
Primary machine, the Redundancy Backup machine becomes
the triggered Primary. Upon bootup, the original Primary
may not be properly stopped. In this case the Watch
processes of the original Primary will still be active. This
may lead to duplicate unwanted jobs to be created. Please
set the status of the original Primary to ‘Stopped’ if you
suspect this is the case.
If Primary and Backup machines are disconnected from the
network at the same time for more six hours, the
Redundancy replication will stop (health indicator will be
red). If this occurs, the Backup must be configured to point
to the Primary IP again.
Disabling the active network adapters through Windows
Network Connections interface can cause the
Primary/Backup the Redundancy replication to stop (health
indicator will be red). Re-establish the machine roles again
to establish a healthy Redundancy connection.
Redundancy role may become unset after a machine reboot.
Re-establish the machine roles again to establish a healthy
Redundancy connection.
Backup machine cannot be used for encoding when it is in
Backup role. However if redundancy is triggered and the
Backup is promoted to Primary, the machine can be used as
an encoding node.
Due to technical limitation of PostgreSQL database
replication, it cannot handle IPchanges. So, both Primary and
Backup should be assigned with staticIP. Please consult
network administrators for such settings.
Cluster (REST API) When using the API to queue jobs to the Cluster for normal 4109
Cluster job distribution, you should be using port 8649 for
HTTP or 8650 for HTTPS.
REST API When using the API, the output filenames will show the 7826
requested filename. It will only show the true output
filenames for a job when the jobs are completed. This
behavior is important to note in the case where the Job is
writing to an output location where the filename already
exists.
Cluster Cluster Manager may stop working if the machines system 6111
(License) time is rolled back. It is recommended that restart the
CpClusterServiceRunner service from (Windows - Service
Manager) after the clock has been rolled back.
Changing the computer clock forwards beyond actual real
world time can affect the duration of the license validity
when you clock is rolled back. If you are running into an
issue with the dongle license, please contact Capella.
Cluster (License) When updating a dongle license for Cambria Cluster, you 5173
should restart the CpClusterServiceRunner service from
(Windows - Service Manager) for the license to be detected
properly.
Licensing Floating License Server / Online Activation 11682,
(Alternate method) 11819,
Floating License Servers can be used to lease Capella 11826,
product licenses to physical machines and virtual servers 12275,
over the network. This feature supports both “on-premise” 14409
machines as well as virtual servers on the IBM, Google, and
AWS cloud.
The Floating License Server software is activated using
Capella’s new Online Activation licensing. Online Activation
is an alternative to Capella’s USB license dongle. Online
Activation allows Capella Products to be unlocked on a
machine through the internet using a 25-digit activation key.
Notes:
• Floating License Server must be installed on a physical
machine. The Windows OS must also be configured to
prevent the machine from going to sleep.
• Updates/Extensions to Licenses by Capella will apply
automatically if the license is online. However, it is
recommended that users re-activates after Capella makes
the change so that the Licenses can be updated immediately.
• Machines that have Capella Products that are Online
Activated should remain connected to the internet. If
disconnected, the license will continue to work until the end
of the Grace Period (2-3 days), during this time users can
reestablish their internet connection.
• If the machine is rebooted while in Grace Period, Capella
Product licenses no longer work. The licenses will need to be
re-activated via the Online Activation procedure.
• For Cluster and FTC that are installed onto Virtual Servers,
the virtual machines should have a single network adapter
and TCP traffic must be enabled (allowed through security
group setting/firewall) between the Cluster and FTC.
• Uninstalling the Floating License Server will automatically
deactivate the license. Capella products that were clients of
the Floating License Server will automatically go into a 7-day
grace period. Installing and reactivating the Floating License
Server license again should allow clients to re-establish the
connection for the licensing link.
• Server sync may not update the number of clients/license
a floating server has. To workaround this, you will need to
manually re-activate the key on the floating server.
License We support grace period of 10 days after license has 16012
(Activation/Cryptlex) suddenly invalidated or fingerprint changed.
But, if the machine reboots during grace period (GP) state,
then upon boot the License Manager UI will not show the GP
state. However the license is in GP state internally.
License (USB) Changing the computer clock forwards beyond actual real 6111
world time can affect the duration of the license validity
when you clock is rolled back. If you are running into an
issue with the dongle license, please contact Capella.
License (USB) After USB dongles are remotely updated, it is recommended 5173
that the Cambria machine be rebooted so that the new
license features and job queuing functions can apply.
License (USB / If the hardware or software license is not plugged in, File 3874
Software) Convert will fail to encode when the “Convert All Jobs”
button is clicked. Plugging in the license at that time will not
allow you to bypass the error.
Workaround: Save the project and close and reopen the
project with the license plugged in.
Migration Issues Cambria Manager information such as Watch Folder, Job, 8885
and Settings are not guaranteed to remain after upgrading
from an older build to a newer build. It is recommended to
save these settings before updating Cambria.
Windows Disable Microsoft Window’s Hibernate/Sleep mode 8957,
Component 9046
Cambria Cluster does not support Microsoft Window’s
Hibernate/Sleep mode. If a computer goes into
Hibernate/Sleep mode while Cambria Cluster, some functions
of the application may not work properly.
Recommendation: Disable Hibernate/Sleep on the
Cambria machine:
1. Go to Control Panel>Power Options
2. For the selected preferred plan, select ‘Change plan
settings’
3. Select ‘Change advanced power settings’
4. Go to Sleep>Allow hybrid sleep
5. Change the setting to “On” (this change will disable the
Hibernate option in the start menu)
6. Go to Sleep>Hibernate after
7. Change the setting to “Never” (this change will keep the
system from Hibernation when it sleeps)
IMF Editor Tool IMF Editor Tool (Purchase Option) 10852
Generate Interoperable Master Format Packages with this
tool.
The IMF tool allows for Clip Lists to be created from
video/audio files. A Clip can be used to create Application
#2 compliant output.
Limitations:
- No Metadata entry (language, title, description, etc)
- No compliance checking is done to make sure that video
and audio durations match. (It is recommended that this is
checked by the user)
- No automated workflows (watch folder, script/template)
- no upload, post-task, script, notification, filters, etc
- Does not support re-import/modifying existing IMF
- Does not support captions
- Does not support source trimming
- Does not support video previews
Cluster Settings
Watch folder input and output locations should be a shared network location so that all cluster
machines can access the folder location through a shared network path.
Please refer to the ClusterInfo&Troubleshooting.pdf file for more information on Cambria
Cluster. This document is included in the Cambria Cluster installation package.
REST_API_(File Convert/Cluster)
Please refer to the REST_API_FileConvert.pdf file for more information on API for Cambria File
Convert and Cluster.
Please also note that API Samples are included in the FTC_APISamples package.
Here is the link:
https://www.dropbox.com/sh/lc5y7mzwp73vclf/AAAGgOdY_JyxOf8LCjHtSNPXa?dl=0
REST_API_(Cluster Replication)
Please refer to the REST_API_Replication.pdf file for more information on API for Cambria
Cluster Replication.
Here is the link:
https://www.dropbox.com/sh/lc5y7mzwp73vclf/AAAGgOdY_JyxOf8LCjHtSNPXa?dl=0
Scripting Guide_(File Convert/Cluster)
Please refer to the Scriptable Workflow Guide.pdf file for more information on how to develop
and test Target Preset scripts.
Here is the link:
https://www.dropbox.com/sh/lc5y7mzwp73vclf/AAAGgOdY_JyxOf8LCjHtSNPXa?dl=0
REST_API_(Floating Server Manager)
The Floating Server Manager API is available upon request.
Email your request here:
support@capellasystems.net