It's hard to think of a broadcast technology that has changed more in the last decade than multi-image display systems. Even several years ago, the most common question was, “What's wrong with my good old CRT stack?”
Within a few short years, the multiviewer became a standard piece of equipment in the broadcast workflow. The multiviewer today has evolved into a platform that comes in varying sizes and capabilities, from simple quad-split systems to large-scale, full-facility systems that consolidate routing and monitoring functions. Making the right choice can quickly become overwhelming.
There are many angles to consider when evaluating the choices. The most popular multiviewer requirement is calculation of the number of monitored sources and the required amount of displays. The number of inputs and outputs becomes the key driver. Future I/O expansion, redundancy levels and ease of installation, operation, and maintenance are significant parts of the picture.
Architectural approach
There are two architectural approaches to multiviewer deployments: single frame and distributed. Both approaches are valid, each having specific benefits and challenges.
The integrated single-frame system supports migration to one box associated with routing capabilities that support every input and output. These frames consume less space, offer simplified control and reduce cabling and installation costs. They typically also have fewer points of failure. (See Figure 1.)
A larger up-front investment in the architecture may be required. This is partially related to the need for discrete I/O stages in the signal flow. This requires additional flexibility to scale for future requirements with an understanding of the cost to add more inputs or outputs.
The distributed architecture is an alternative approach. This system uses discrete multiviewer systems to drive a small number of displays, spreading the load to lessen the possibility of a single point of failure. (See Figure 2.)
The distributed approach supports the assignments of individual multiviewers to specific control room functions, minimizing interference from other operators. Audio control, for example, might be assigned to the production switcher output or the graphics system. This naturally separates the control and operation of these systems to the controllers in that area.
The addition of more inputs or outputs is normally as simple as adding another unit. However, control challenges may increase with an expanded architecture, and costs may be higher compared with an integrated approach.
Application overview
The intended applications will likely influence the choice of an integrated or distributed architecture. Although use cases vary from site to site, and unique requirements exist for each application, the majority of broadcast applications fall under four categories:
- Production control rooms;
- Outside broadcast vehicles;
- Master control rooms; and
- Transmission/cable headend monitoring facilities.
Production control room
The studio director in the production control room uses the multiviewer to create live content. The director, always the demanding artistic individual, is tasked with split-second decisions on camera angles and shots. There must be confidence that what is seen on the multiviewer will be transmitted correctly to the viewer.
This has a large influence in how the production gallery operates. Picture quality must be pristine enough for obvious recognition of fine detail, and the director will normally insist on fewer sources per display to maintain resolution. This is usually accomplished by adding more displays instead of more PiPs to a display. The real-time nature of live production means that the processing delay (or latency) through the multiviewer and display is a key metric.
Picture quality is often assumed and not evaluated, but quality can vary hugely across models. Video sources must endure two main processing stages — scaling and deinterlacing — to become part of a multi-image display. Both stages will reduce picture quality if not correctly designed.
Scaling decreases the picture size to fit on the multi-image display. This process requires intelligent scaling software to understand what pixels maintain the image quality at any size. It also uses large dynamic horizontal and vertical filters to remove scaling artifacts without reducing picture detail.
Deinterlacing combines the two fields into a single, progressive frame for interlaced sources. If performed incorrectly, artifacts will appear due to interfield motion or a reduction in picture resolution. Side-by-side comparisons of the two techniques will reveal the quality variations.
Simple control and configuration software is a must as different directors and shows require a variety of setups. Engineers will frequently make last-minute changes to the setup. This can include the incorporation of more cameras, size changes to PiPs, and variations in the UMD and tally styles — all to meet an individual's preference.
Outside broadcast vehicles
The outside broadcast environment echoes many of the same requirements as the production control room, while introducing some unique considerations. The most obvious are the space, weight and power limitations of OB trucks.
OB environments benefit greatly from the integrated approach, combining the multiviewer with the router and other signal processing equipment. This translates to enormous reductions in footprint and power use. The single-frame design also reduces the number of spare parts and redundancy options within the truck.
Quad-split solutions are a popular choice in the OB environment. The limited number of inputs typically reduces power consumption, and resolution is maintained by requiring four or fewer sources per display. This also easily accommodates the monitoring of just one source at full screen — a common request for outside production events.
Master control rooms
The master control room is the last stop before transmission. Operators here are responsible for monitoring the outgoing signals and ensuring they conform to all regulations, including decency ratings, teletext/captioning requirements and station ID insertion. The process also ensures that audio tracks, aspect ratio data and channel branding are correctly encoded.
Multiviewers in the master control center must offer options to decode and display the relevant data. This includes decoding the closed-caption data and its subsequent display on the video as it would appear to viewers. Automatic display of the source aspect ratio data in its native format is also helpful.
A sophisticated monitoring system will automate much of the monitoring, flagging errors on screen. The same system will likely generate messages over SNMP to control software. Some top-end devices will send e-mail and/or SMS text messages to alert engineers of potential problems.
Transmission and cable headends
Cable and other transmission headend environments typically require monitoring of a large number of sources. The purpose is to ensure that the signals are valid and transmission is taking place without error. It also confirms that audio signals and required metadata are being transmitted.
The multi-image display in this application focuses on high PiP density as presence is more important than picture quality, which is monitored prior to transmission. These facilities are highly automated, so alarming options and rules-based responses are extremely desirable features.
The transmission chain has evolved to mirror the changes in how consumers access content today. IP-based transmission has become as important as traditional RF transmission, and most broadcasters are dealing with hybrid signal environments for baseband and broadband.
The monitoring of IP signals represents a new shift in multiviewer architecture design. The majority of applications require IP-to-baseband conversion for signal monitoring. This adds cost and introduces unnecessary processing and another potential point of failure. New discrete IP and hybrid monitoring systems have appeared on the market, offering direct IP monitoring, simplifying the workflow, and reducing cost and complexity.
The various requirements and demands in signal monitoring and processing stipulate that different architectures are necessary to match the environment. No single system will cover everything. Therefore, it's pertinent to prioritize requirements when evaluating monitoring options and remain open to different approaches to finding the optimum system. Vendors should offer a range of options to suit the architecture chosen, with an application-tailored system to best match the requirements of today and tomorrow.
Kevin Jackson is product manager, multiviewers, at Harris Broadcast Communications.
