Many master-control functions are controlled through three ordinary-looking PCs. Three PCs, a big plasma screen and a few backroom racks handle the main master control functions for six TV channels. Pictured: Pierce leads a demo.

PBS offer: automated control room in a box

Originally published in Current, Dec. 1, 2003
By Steve Behrens

PBS never has known quite what to call the thing. “Station in a box” didn’t sit right with some people. “Advanced Control Environment,” seen lately on signs, makes a nice acronym — better than EIOP, the Enhanced Interconnection Optimization Project.

EIOP amounts to a couple of new options that PBS is giving to stations. Most concretely, it’s a small roomful of computers that handle program scheduling, traffic, broadcast automation, digital storage, special effects and other functions that can automate and replace most of a station’s master control room.

Then there’s an ongoing PBS service that’s made feasible by having that standard hardware and software in the field: The station’s staff can go home or do other work while PBS computers and engineers monitor the station’s master control via the Internet from Alexandria, Va.

PBS techies have been showing off the prototype — most recently in a videoconference last week — and they described it for managers at last month’s regional round-robin meetings. In April, PBS will haul a whole EIOP control room to Las Vegas for a demonstration at the National Association of Broadcasters convention.

Andre Mendes, PBS’s chief technology integration officer, expects the equipment package to be available in March and the monitoring service to begin in July.

Though a brief description of the plan could distress pubcasters who aren’t keen on centralized control, it doesn’t threaten a system takeover by the forces of Alexandria. Stations will remain in charge of scheduling programs and will store them locally. Both the hardware/software package and the monitoring service are optional for stations, PBS says.

Though EIOP takes advantage of the new capabilities of PBS’s planned Next Generation Interconnection System, it’s a separate package for stations to consider. The mixed satellite-and-landline interconnection system, for which public TV is now seeking federal money, is planned for operation by late 2006, when PBS’s present satellite leases will end.

PBS hasn’t made final choices of hardware and software or set firm prices, but Mendes expects the EIOP package to cost stations about $1.2 million. Monitoring service will run less than $48,000 a year, he says, equal to or less than one technician’s salary.

The offer will have the strongest appeal to the dwindling number of stations that haven’t automated yet—where technicians still stand by to watch the gauges, record satellite feeds on videotape, and push buttons to play back the tapes along with interstitial material and local programs.
But PBS is expecting EIOP’s economics and features will still appeal to stations that already have automation equipment, even where the engineers go home at night and rely on a beeper to warn them of technical malfunctions.

Not “centralcasting”

Proponents stress that EIOP isn’t the same thing as centralcasting — the setup adopted by NBC’s owned-and-operated stations that centralizes program storage at the network hub and transmits separate program schedules to each station it serves.

“What we’re doing,” says Mendes, “is exactly the opposite.”
PBS Chief Engineer Jim Kutzner says the new interconnection system is better described as “edgecasting,” because the programs will be distributed as cheaply as possible, in Internet-style packets, usually by satellite file transfers and often at night, and will be stored on stations’ videoservers at the so-called edge of the network.

Storing duplicates of programs on a station’s server is cheaper than transmitting a whole broadcast schedule at broadcast time because the cost of digital storage capacity is falling rapidly, while satellite and fiber transmission costs are not, Mendes explains. NBC can afford the inefficiency, he adds.

EIOP leapfrogs over an earlier efficiency-seeking proposal for station master control rooms—the Advanced Digital Distribution Entity. Two years ago, Dennis Haarsager, who oversees Washington State University’s station in Pullman, Wash., was advocating that stations band together and create regional ADDEs that would store programs centrally and send out complete schedules to distant transmitters, as in centralcasting. But Haarsager says EIOP’s economics now make edgecasting more attractive.

Stations in several Northwest states once talked about creating a regional ADDE to handle their master control, but the only resulting shared facility was one without great distances between stations: It’s a shared control room for stations in Spokane, Wash., and nearby Pullman.

Mendes contends that the system will be 99.99 percent reliable, with enough redundancy to reduce problem broadcasts to 53 minutes a year. In case of calamity in the Washington, D.C., area, PBS plans to have one or more backup control rooms away from the city.

He acknowledges that human errors could hurt reliability, as in the case of programs that run longer than expected. “This demands accurate metadata upstream,” says Mendes. “If we sent content that was poorly timed, stations would have upcuts or on-air black.”

Sign of times: pink computer cables

EIOP looks nothing like the analog master control room of the past. Most of the visible wires are pink computer cables that converge on an Internet-style router. Three unassuming office PCs with colorful flat-panel screens replace expensive custom-made broadcast consoles. Hanging above them is a big plasma screen showing input and output program streams.

The package will be able to handle six channels—four channels for DTV multicasting, a high-definition channel and a spare channel for backup.

Marilyn Pierce, PBS’s director, digital asset management, who headed design for the system, demonstrates how a programmer could assemble a schedule with a mouse, dragging and dropping icons representing programs and station-break elements. The software alerts her if programs would overlap or come up short, and it keeps track of unsold underwriting time. She can assemble station breaks weeks ahead of time and preview them on screen, including fades, squeeze-backs and other special effects. Meanwhile, the system compiles text program schedules for newspapers, cable systems and digital TV receivers.

PBS expects to assemble the equipment package and ship it whole to stations, Mendes says. Station engineers who want to learn the system by building it can come in and help do the work, he says.

Across the room, Pierce demonstrates a different package—the small array of computers and screens where PBS technicians will monitor station malfunctions via the Internet. Across the country, sensors look deep into the stations’ digital hardware, watching for the failure of cooling fans and other signals of potential trouble.

Pierce simulates an equipment failure at a Midwestern station, using a real Internet connection. About 10 seconds after the sensor notices a motherboard heating up, an icon representing the station turns red on a map of the country. The PBS computer automatically switches the station’s programming to backup hardware at the Midwest site. Technicians follow up, checking the signal path to diagnose the failure. If on-site repairs are needed, PBS would call a local technician. In most cities PBS expects two-hour response time, Mendes says.

PBS’s prototype doesn’t yet have circuits to monitor the internal workings of transmitters, but its engineers are looking into ways to add that function, Mendes says. However, PBS technicians will be able to eyeball the picture broadcast by the transmitter via a video stream on the Internet.

As stations switch to all-digital master controls, they will face a new financial challenge. Many don’t set aside funds in their annual budgets to pay for future replacement of aging equipment, says Haarsager. Instead, they use what he calls the “hail Mary” method of depreciation, in which the stations trust that their prayers will be answered by the state legislature or a fundraising campaign before equipment dies.

“This is a real problem for us as we move into a more computer-centric environment,” Haarsager says. Servers wear out and become obsolete in three to four years, he says, much faster than much of the old analog equipment being put out to pasture.