Pro-bel automation at TV2 Norway.
Dave Collins, head of automation at Pro-Bel, has spent many years integrating complex automation systems. Today he gives "Automation Technology Update" his views on broadcast automation.
Automation Technology Update: What is the most important reason that a broadcaster should use automation?
Dave Collins: It depends on the type of broadcasting. For prime networks, it allows them to run complex interstitial packages that are just not possible with manual operation. This needs frame-accurate control of several devices such as DVEs and image stores. For the multichannel operator, automation allows a single operator to run 10 to 20 channels from a single control position.
ATU: What areas should a broadcaster automate first?
DC: From my experience, the first area is commercial playout. Playing back-to-back short segments is best done from the server controlled by automation. Wrapped up with commercial playout is ingest and the transmission of program content. Once the automation is installed, then the next area to turn to is asset management and archiving.
ATU: Does this mean the broadcaster can approach automation with a building-block approach? If so, doesn’t this mean that the broadcast is then forever locked into one vendor?
DC: Broadcasters start with traffic and sales systems. The playout automation comes second. We integrate with the traffic, and latterly with archive systems from third parties. As an example, a recent customer bought a single channel, and then later added media management and an archive.
ATU: What are some of the regional differences in how automation is implemented?
DC: There may once have been regional differences, but the move to file-based playout has led to a convergence of requirements. We have seen a demand in Europe for extensive alarm monitoring, whereas our North American customers have a higher staffing level to cope with gear failure.
ATU: What immediate benefits would a broadcaster see with the implementation of automation?
DC: The obvious benefit is a reduction in staff numbers. The on-air look can be more adventurous, as more complex junctions can be scheduled. The end-credit squeeze needs frame-accurate synchronization of several devices. Multicasting can mean channels splitting, each with different aspect ratio. Again it's about the control of devices such as ARCs and image stores.
ATU: There has been much discussion about moving control of playout from the master control area back into the traffic department, where programs and commercials are initially booked. What challenges does this place on an automation vendor? What differences does this make in the operation of the control room?
DC: The problem with passing control back to traffic is when do you allow late changes and when do you not? Is the new commercial on the playout servers? Traffic is more remote from the master control, and it would be easy to make mistakes. At the end of the day, master control is responsible for transmission, plus the split-second timing in a live environment.
ATU: What sort of interface is needed with traffic and sales?
DC: We are beginning to see rich media schedules from traffic systems based on XML. We also have a MOS interface that exposes the playlist to other applications. We use the MOS to drive interactive applications, and XML for playlists. We publish our data dictionary to traffic system vendors, although we have a schedule translator that can convert data fields in playlists and as-run logs as required.
ATU: What key features should a customer look for in an automation system?
DC: Resilience is key, plus flexibility for channel expansion. In transmission, it is crucial that commercials air reliably. We use hot failover for the primary controller to give resilience. There are many options for system design, you can use mirrored video servers or N+1 redundancy. It depends on budgets. Flexibility means that new channels can be added in a very short time. Before automation, it was a major capital project to add a new channel. With automation and file-based workflows, channels can be added with minimal disruption.
ATU: How can a customer estimate the ROI for an automation system?
DC: There are the obvious things such as saving in staff costs, not only for master control, but also running the tape library. File-based playout from video servers leads to savings in maintenance cost; there is not the same wear and tear on VTRs. The big plus is that there will be fewer lost commercials.
ATU: What do you think will be the next big idea in automation?
DC: I think the next big idea is full support for MXF file-systems. We have all talked about it, but there are still many issues to make it viable. There are workarounds, but they all need separate transcoding systems. Once metadata can be encapsulated with the essence in an interoperable system is going to really ease the media management challenges that we have with playout automation today.
For more information, visit www.pro-bel.com.