By James Gilbert
As HD makes its debut in mainstream broadcasting, MD of Pixel Power James Gilbert explores the opportunity for smaller and niche broadcasters to differentiate their offerings and add to their revenue streams affordably.
|~||~||~|At the end of 2005 in regions outside the USA and Japan where it was already prevalent, HD was still something of a cloud concealing a silver lining. With the lure of prosperous new business from forward-thinking producers, audiences and advertisers, many broadcasters were making the leap from standard definition (SD) in preparation for the boom that we now see coming to fruition.
Herein lay the problem. While the investment, in many cases, had been made in HD-ready systems, the challenge was an immediate scarcity of HD-ready business. The majority of post-production houses in the UK for example cited at the time that just 5-10% of their business was in HD, though they maintained hopes these figures would double over the following twelve months.
Now that HD has made its debut in the Middle East, with ArabSat’s first HD broadcasts in the region at MEB2005 last December, these are sobering statistics. The average cost of launching a TV channel in the Middle East is $5 million even before HD bumps up the costs, yet broadcasters appear undeterred: in the Middle East, the number of music and religious channels rose 60% this year and there are now 250 channels in the Arab world.
In terms of high definition, there are 14-17 HD channels in most major TV markets around the world. In addition, the USA anticipates 40% of its homes will have HDTV by the end of next year. Where the USA TV market leads, typically the rest of the world follows.
The pioneers of HD in this and other regions are conspicuously major broadcasters. Successful proof-of-concept trials throughout 2005 provided the reassurance many broadcasters of all sizes needed to make the leap to HD.
What was less clear, however, was how smaller, new and niche broadcasters could do it cost-effectively. The equipment used in the trials tended to be high-end, high-performance technology, often including features that would be superfluous to the majority of broadcasters.
Now that technology companies are paying attention to the market need for entry-level and mid-range HD equipment, the choice of equipment for acquisition, editing and transmission is broadening. Yet this is, literally, not the whole picture.
Graphics are now recognised as essential. From lower-third tickers to animated logos and squeeze-back capabilities, they help to improve branding, convey additional information and solve challenges of time restrictions.
This should be no different with HD, which also brings additional benefits and can even increase revenue. There are, of course, considerations. The challenge is to make the switch affordably. Some broadcasters attempt to up-res their existing SD graphics. However, we will see that this does not produce the highest quality result and the resulting work in getting it right makes it a costly option.
The key to keeping the costs down is to separate the essential functionality from the optional features so you understand what to ask your supplier. Beyond this, the actual features should be the broadcaster’s own choice. Ensure the system does what you need without providing additional, expensive features you don’t need.
The first consideration is resolution. Simply stretching existing standard definition capabilities and up-resing SD content to HD is not sufficient. While this can work to an extent, PC-based graphics equipment has only limited bandwidth capabilities such as the PCI bus that connects the various technology boards using.
Ideally, broadcasters should choose a graphics solution with its own dedicated hardware, which delivers performance many times that of PCs. This means graphics creators can try ideas in real time without the delays they would suffer using a software-only solution.
The ideal frame-store pixel processors have at least 4Gbytes/s of memory bandwidth and a combined instruction rate of more than 5 GIPS. A dedicated character generator should also have a video backplane which carries multiple streams of HD video between frame-stores and I/O modules. The PCI bus can then operate in parallel for file transfers and network based I/O.
HD graphics also place higher demand on bandwidth, software programmability and pixel-processing requirements so a multiple processor solution is essential. Each pixel processor needs around 500Mbyte/s of memory bandwidth and achieve five billion operations per second (BOPS). To help keep the costs down in the face of the increased bandwidth and processing-speed requirements, very long instruction word (VLIW) technology is an excellent option for extracting performance through architecture rather than by increasing the clock rate.
The second consideration is the aspect ratio of both overall picture and the pixels that make it. Where 4:3 is the dominant format for SD, all HD uses 16:9.
A major consideration for graphics creators is that the pixels used for HD are always square. This is not a problem on a standard computer screen, as the pixels are also square.
On SD TV, such as PAL and NTSC formats, they are not. This is further evidence that up-resing SD graphics to HD is not a practical solution and actually entails more work and expense than it might first appear. However, during the change-over period when HD and SD graphics are produced, there must be the architecture to support this.
To save time and costs, broadcasters can design graphics in one format but with a ‘safe zone’ so that when they are converted, they still appear without missing sections. The alternative is to have functionality within a graphics solution that converts automatically.
Even so, the graphics will not display properly when converted. The better process is to design graphics separately in HD and SD. This is a further driver for low-cost HD graphics solutions.
The right solution will not create more work as it will allow the creation of templates that can be easily and quickly manipulated. Provided the character generator simultaneously feeds metadata, such as names or scores, to both HD and SD templates, you only need create two templates instead of one. An added bonus is that HD graphics use more pixels and therefore remain readable when reduced in size. For example, a typical logo appears 20 lines high in SD. In HD, it would be proportionally twice the size but could be reduced by around 25%.
This frees up screen real estate for either more main picture or further graphics from sponsors, which would add a revenue stream. This is impossible through simple up- and -down conversion and would also result in ‘empty’ space on the graphic.
Of the various high definition formats in existence, including 1080i60, 1080i50, 720p60, 1080sf48, 1080p24 and 1080p25, some formats suit different types of programme. In addition to planning which you wish to use you need to ensure your graphics solution can accommodate your choices.
This is particularly important if, for example, a studio produces a drama in 1080p24 but transmits it in 1080i. The character generator must automatically adjust the graphics accordingly.
For example, p24 cannot accommodate the rich visual features of fast-action pictures such as sport and tend to make the picture appear ‘jerky’. In this case i50 would be better. However, p24 remains sufficient for slower action subjects or situations with less camera movement such as studio-based magazine shows or news.
The big story at IBC2006 will be the availability of graphics, and other, broadcast technology solutions for high definition designed with smaller, niche and start-up broadcasters in mind. Significantly, the most attractive offerings for this market will be those that offer features as options and a choice of packaged HD/SD solutions for all broadcasting needs, from pro-AV to high-end.
All broadcasters face increased competition from new and existing channels and this competition extends to the technology at the root of their operations. It has never been more important to stay at the leading edge of technological innovation. Consequently, technology providers are expanding their attention to the HD needs of broadcasters outside the category of large corporations. High definition is intended to deliver improved choice to broadcasters and the creation of lower-cost versions is the only way to keep with this spirit.||**||