3G or Not 3G, That Is the Question
August 11, 2009
With file-based ‘tapeless’ being the way forward, why would you need to send baseband at 3G (SMPTE 425M) around your system?
If you weren’t a broadcaster, the question would probably not arise, but once you realise that you might have to take in a live video feed or two, add some extra content and then send the finished (live) video stream to air, you quickly become aware that the ‘tapeless’ model was never really going to work in all but a few specialised operations.
The current HD standards for broadcasting only need to cater to 1080i, or 1.5 Gb/s, but given the progress made in recent years on compression for home delivery and possible enhancements to the home viewing experience, it very much looks like 3G should be the minimum size of infrastructure being considered. I’m not sure of the business model for 3D or higher frame rates and resolution to consumers, but future-proofing is a must unless you really want to tear out your infrastructure in just a few years. TV stations in particular have a long shelf life. It never ceases to amaze me just how long TV technology lasts these days. I would suggest, for cabling, that 10 years is the minimum, and who could possibly predict what we’ll all be watching by then?
The pace of change in recent years has been nothing less than astounding. The first HD demos I saw were at least 20 years ago, but as new delivery technologies have emerged and been taken up by consumers, conferences such as IBC2009 are awash with what could possibly be the next enhancements: 4k, 3D and frame rates up to 300. (Check out the BBC white paper WHP169 in which they discuss the pretty obvious idea that more frames equals nicer pictures.) So, as is usual, the things that engineers invent will find a use sometime. But 20-odd years ago we were all still living in a PAL world, even a component infrastructure was unusual. Lucky were the organisations that went from PAL to SDI — most of the cabling would do the job. For 3G, however, there is little that can be repurposed.
I’m actually working on a new facility for a major broadcaster, and I can categorically say that “yes, 3G should certainly be considered,” as it most definitely is on this project. Whether there will be enough take-up by the consumer of services that need this kind of internal infrastructure, the fact remains that a new facility ought to have the capacity to grow with whatever producers and consumers demand. And with some TV business models being a little strained, that growth should be cost effective as well.
According to the EBU, very few European broadcasters have moved to all-HD production. In a survey by the EBU earlier this year of 53 broadcasters, just one of 37 who answered the question stated that it had migrated its production facility to HDTV. A further 20 (54 percent) have partially migrated, while 12 (32 percent) plan to start in the next couple of years, and four (11 percent) currently have no plans. This means there is still the opportunity to move directly to a 1080p50 infrastructure.
Not surprisingly, a lot of the equipment required to make a full 1080p TV station is not quite here. The cabling is easy — good quality co-ax can handle 3G over a reasonable distance and then there is fibre — but some of the ‘glue’ is only on road maps at the moment, although I suspect IBC2009 will feature a few announcements from the usual suspects about that one.
The main blocks of gear are pretty well there. Vision mixers, cameras and editing systems are all deliverable now or jolly soon — but PC-based items and graphics in particular, are lagging behind a tad. This is due to the lack of 3G-capable cards and the extra processing power required to render the stuff in real time. According to Gennum (purveyors of 3G chips to go inside these units), demand for 3G equipment is rising. About 30 percent of what Gennum sells now is already 3G, so the future looks bright.
Is 3G going to be future-proof enough? In the foreseeable future, yes, especially given that the next couple of years will see all the parts required come on stream. The SMPTE have even got a standard for 10G (SMPTE435), which currently is for transporting multiple 1.5 and 3G streams, but it could easily apply to more enhanced services in the future. And it would certainly be daft not to at least consider the upgrade path.