The Future of Ingest
The Future of Ingest
Is there any VTR here?
When did you last use a VTR to ingest content? Most of you will answer "a loooong time ago,” but they are still there, in a room corner, hidden by papers. VTRs are not used anymore for ingest but are still needed as players for legacy material. Think about your good old cassette tapes that you put in your Sony Walkman from time to time when you feel nostalgic; no one is using it anymore to record music.
Nowadays everybody is using some hardware box with BNC inputs to acquire video feeds. From a single channel video board in a workstation to a massive ingest factory with hundreds of ports connected to a SAN storage, ingesting live content has become a commodity. Formats have evolved with the time to support HD, then UHD, while maintaining the bitrates as low as possible. But nothing really impacting the day to day ingest operations.
But with the move to the cloud and IP streams progressively replacing baseband signals, ingest is back in the strategic discussions.
Moving to IP
The first step is to take the raw data of an SDI signal and load it into RTP/UDP/IP packets according to SMPTE-2022-6. Then let's replace all the BNC connectors by RJ-45 ones and use SMPTE-2022-6. Et voilà! Well, it may work for point to point/contribution scenarios, but for production it is not so optimal, as a lot of the transported data is blank (ancillary data space), 4K and other UHD variants do not fit in a 10 Gig Ethernet link, and individual streams (video, audio, metadata) cannot be processed or distributed easily. This looks like a short-term approach.
So let's take the individual essences (video, audio and metadata) instead of the raw SDI data and load them into RTP/UDP/IP packets. There is a community pushing in that direction, named ASPEN (Adaptive Sample Picture ENcapsulation), which has published SMPTE RDD-37 to complete an entire IP MPEG-2 TS ecosystem able to transport essences without abusing bandwidth and while offering signal routing flexibility. There is another community with a similar goal but different approach, named AIMS (Alliance for IP Media Solutions), which is not in favor of MPEG-2 -TS and instead promotes RFC 1475 and AES-67 as standards to carry video and audio respectively. This last approach is described in VSF TR-03.
Those two first steps are already a technological challenge for all the traditional video servers as it means dropping SDI which was standardized in...1989...But as such this will not significantly change how we do ingest.
Leveraging the transition to IP to offer new workflows
Moving all those signals to the IP world is a great opportunity to reduce costs and increase flexibility by using standard mainstream IP switches. It also opens new alternatives for ingest operations as any source becomes potentially available for anyone “connected” on a routed network. Consequently, an infrastructure management layer on top of all those streams is mandatory to organize IP streams lives.
Fortunately, a group of experts known as the Joint Task Force on Networked Media (JT-NM) has already produced a reference architecture describing a conceptual model for interoperability, whose ultimate goal is to ensure that devices and services can co-exist and offer new workflow possibilities. The Networked Media Incubator Project led by the Advanced Media Workflow Association (AMWA) is a pragmatic program to build and learn on aspects of the reference architecture while ensuring early interoperability. As for any significant technological change, interoperability is key to guarantee a large adoption rate and quick deployment.
So, what is the future of ingest?
We clearly see a great effort from the broadcast community to lead this revolution to success with standards and interoperability as clear targets. Beyond SMPTE-2022-6 and TR-04 things are not yet fully defined, but for sure moving to IP will facilitate software-only solutions, as there will be no more physical inputs required.
As an ingest operator, the most important thing to understand is that this change opens a whole new range of possible applications to manage ingest operations. And those are still left to the creativity of the media industry!
Go further with the Future Series
|The Future of Transcode||The Future of Media Asset Management|