The constant struggle in broadcasting these days seems to be reducing costs. Centralized operations — centralcasting — has been touted for several years as the way broadcasters might reverse the downward spiral of their cash flow. Centralized operations come with benefits, risks and costs. The balance for the technical side of our business is to find ways to tie business needs up with a bow of the right size, without breaking the bank.
The problem arises from the seemingly inexorable decline in broadcast sales revenue and the desire to have increased income. More goes out and less comes in, a recipe for a dying business. The thought of reducing costs by increasing the efficiency of the delivery plant is attractive. The devil is in the details, though, because, in many smaller markets, replacing the average salary by increasing efficiency in the master control room will not move the equation the right direction. The idea is that you move master control to a central site, reduce head count and rely on automation to run the air operations without human fingers on buttons. In an average station, that might save five to seven salaries, but in smaller markets the trade-off might be several thousand dollars to more than $10,000 a month for an interconnection line to feed the station from a distant location. Clearly, the details have to be worked out in individual cases.
Variations on the theme certainly exist. In one approach, the station is simply run remotely from low-speed data lines carrying automation screens and low-bit-rate streaming-media confidence feeds. In this case, the interconnection cost can be low, in the range of less than $1000 per month for a guaranteed bandwidth circuit. Clearly, this looks attractive, but it comes with risk. If the system fails on the remote end, no data circuit will allow you to repair the air chain. Extending the control to include routing, proc amps and audio levels offers a bit more comfort. But, when a patch needs to be thrown, you have to hope the maintenance staff has not gone home. Several group broadcasters have implemented this model with success, so it is not one to ignore. The capital cost for start-up is quite low, especially if the stations involved already have automation.
Another model that has been used in real-world implementations is the exact opposite of the first model. In this case, a remote site runs the entire operation. Think of it as simply a remote transmitter. If local news is involved, it is returned to the central hub site, usually over the duplex return portion of a DS3 line connecting the two sites together. The greatest advantages in this model are obtained when several stations can be combined in one location and the total number of staff can be minimized using highly automated facilities, with one person supervising more than one channel. DBS and cable operations routinely have 10 or more streams monitored by a single operator. With modern monitor matrices such as those available from Barco, Evertz, Miranda and others, it is practical to have flexible monitoring with customized screen views. For instance, a standard view might show many small monitors with streaming media displays of return-confidence monitors. When a problem occurs, you can switch the monitor “wall” to give preference to the site with the issues while maintaining smaller monitors for the remaining sites with a single button push. Some manufacturers have customized the display and control circuitry to allow for intelligent interdependence between the displays and SMTP monitoring of the remote end. This offers flexibility and shows the operator the most relevant information, including the controls needed to fix the problem remotely.
This model has the highest interconnection cost, because when the lines go down, no signal makes it to the remote site at all. It also has the highest possible labor savings, but with higher capital cost compared to using existing facilities as is. If the current station or stations have to be rebuilt anyway, the capital in this case may be lower. As with all equations, however, you can't solve it without knowing the variables and the data involved. If a news operation exists at the station, the latency of the circuits must be considered. Returning the news to the hub for turnaround to the station means that it will likely traverse two codecs, along with networking hardware, frame syncs, etc. The cumulative delay will make mix-minus audio for journalists in the field a challenge. With DTV transmission as the ultimate goal, the latency issue becomes even more problematic. The possible economic return is good, the annual operations cost for interconnection is high, and the return on investment can vary widely.
The final model is one of distributed media and control, with both the local station and the hub having significant portions of the operational facility. I like to call this distributed broadcasting. Modern servers push or pull content to a server at the station under automation control. Content that can be effectively centralized and shared across many stations can be pushed simultaneously to many locations. If the content is live, it need not traverse long lines with serious consequences in the event of failure. PBS is beginning to roll out a system based on this model, which it calls ACE. The goal is to allow stations in public broadcasting to avoid the cost of building multiple stream control facilities that air largely the same programming in many markets. While some stations have embraced this model, others are not yet convinced. Watch the trade publications, including these pages, for results as ACE rolls out later this year.
John Luff is senior vice president of business development for AZCAR.
Send questions and comments to:firstname.lastname@example.org