Broadcast is starting to reap benefits from IT advances.
After a decade of talking about convergence in the broadcast industry, it is finally happening. The fruits of that convergence can be seen in some obvious areas and some that are not so obvious.
Standard computer platforms have been in the broadcast environment for a long time now, handling obvious IT tasks such as automation control, asset and database management, and systems monitoring and control. Recent advances in computer power give designers the chance to improve user interfaces.
For example, a central monitoring system used to only provide a list of events on a screen, each flagging an out-of-tolerance event on some device, with no real indication of its importance. Now such a system can contain a high degree of intelligence. It can interpret events, understanding which are important and which are routine. And it can present the results in different ways, for example as a dynamic schematic of the whole system being modelled, allowing engineers to immediately spot the root cause of the error.
A standard PC or Mac now has the ability to perform intensive broadcast-specific tasks. Editing is the obvious example. Desktop editors can deliver multilayer, multiformat edits, even in HD, in real time on a standard computer. With a simple plug-in accelerator board, it can meet any editing requirement. And in the newsroom, journalists can have sophisticated edit controls on the same workstation that they use for writing scripts.
This ability to tune the user interface to the needs of specific applications and groups of people is one of the benefits of working with a standard computer platform. With a hardware product, the buttons are set, and users have to find them. With software, the control layer can be mapped separately from the underlying application. If a function confuses the user by its presence, then it can be hidden by the interface. Journalists working close to time might be allowed cuts-only editing to make the user interface simple.
Aside from software applications running on standard computers, does the IT industry have a place in the broadcast world? The answer is yes. Embedding IT functionality into broadcast-specific products enables new power and new affordability.
But equally, it must be clear that the old principles of broadcast-quality products cannot be denied. Real time, in our terms, means a new picture every 40 milliseconds. Availability does not mean performance in the high 90th percentile, as some computer applications would accept, but rather 99.999 percent uptime — and even that means more than five minutes of downtime a year.
To address this, broadcast manufacturers are making products that take the best techniques and technologies the IT industry has to offer, while continuing to rely on our broadcast experience. This enables products to be developed quickly and cost-effectively.
Deep IT R&D pockets
The IT industry is an order of magnitude larger than broadcast. And its highly competitive nature means that major suppliers invest huge amounts on research and development. For example, Intel's R&D budget was reportedly $3,800M in 2005; Microsoft's was $5,300M.
To put that in context, the value of the entire broadcast industry's product sales is less than the R&D budgets of those two manufacturers. Besides just making processors faster, this continuing investment has reaped other benefits relevant to our industry, for example in storage and in networking.
Last year for the first time, it became cheaper to store video content on a spinning disk than on videotape. An hour of DV50 material needs around $8 worth of tape. The equivalent disk space costs $6. By 2009, thanks to the R&D investment of the big IT companies, the disk cost will be less than a euro.
Gigabit Ethernet is now commonplace in the IT world. Every computer in the Apple range — laptops, desktops and all-in-ones — has GigE as standard. So it makes sense to design a modern broadcast server using a RAID array out of all these low-cost commodity disks and connecting it to video ports using GigE. The cost is right, and the bandwidth is right. Installing and maintaining Ethernet is far simpler than Fibre Channel.
So anyone can now build a video server. Well, that's not quite right. Ethernet has the bandwidth, but it does not have the means to manage that bandwidth.
In a video server, that means less critical movements could divert critical content flows, such as playout feeds. This is where broadcast specialists have to step in to make it work for the particular constraints of our industry.
Clinging to proprietary
Despite the obvious attractions of low-cost commodity IT platforms and technology, there remains a strong sense in the broadcast world that video challenges need video solutions. We see this most obviously in the way many manufacturers are proposing to move from tape-based to file-based acquisition. Whether their plans are based on magnetic disk, optical disk or solid-state memory, they are proprietary solutions.
Why is this? If standard IT devices meet the required specifications, why not use them? Provided they meet the necessary standards, they certainly should be used. They offer the prospect of an affordable, open alternative.
The open solution
Until recently, IT technology has not reached the camera, but we now see digital media camcorders with magnetic disk subsystems and solid-state storage based on commodity products. Camcorders include GigE, FireWire and USB 2.0 ports to allow any other storage to be attached if required. This allows the user to select the recording medium best suited to the application.
Commodity disk media based on standard desktop products have been adapted for the special requirements of broadcasting. HD recording needs guaranteed sustained transfer rate over 100Mb/s. The disk housing must be protected against the ingress of moisture and dust — contaminants that are rarely a problem on the desktop but are a continuing bane for news photographers.
And these modifications do not skew the economics of the commodity IT proposition. Broadcast-grade disks are priced similarly to an equivalent digital videotape. More important, while there is a VTR equivalent available, in many applications, the disk will slot into a drive on the desktop or built in to the computer. And the drive costs less than $375.
But none of that would be relevant if the format was not open. However neat it looks to slot a disk into, say, a Mac running Final Cut Pro, if the editor cannot read the disk, the exercise is pointless.
Competition must cooperate
In order to ensure that the fundamental benefit of a commodity is realized, competing manufacturers must cooperate. Some call it “co-opetition” when companies that are otherwise competitors work together to develop open standards and integrated workflows. If file-based content handling is ever to become as ubiquitous as 1V video on a coax plug once was, then co-opetition is vital.
Broadcast's IT core
The broadcasters of tomorrow — and indeed some of today — will rely on content as files rather than as video. Material will only exist as a video format, with its real-time constraints, at the point of acquisition and the point of delivery. For IPTV broadcasters, it only becomes video again at the point of display — in the viewer's set-top box. File format conversion and transrating will become core technologies.
The same benefits transforming broadcast equipment are also apparent in adjacent markets. Concerts, conferences and places of worship now regularly expect high-quality presentations that freely mix prepackaged and live video with computer content. Digital signage is a whole new market, mixing simple graphics with complex imagery to catch the eye.
These are exciting times for our industry, as traditional expertise converges with the best of the IT industry, in a new spirit of co-opetition.
Jeff Rosica is senior vice president of marketing and technology for Grass Valley.