Skip to main content

HD won't strain most existing file-based infrastructure

Last week, the Competitive Television Summit ended in Orlando, FL, after an intensive program of keynotes and panel discussions covering a variety of technologies that can help keep broadcasters competitive in today's shifting media landscape.

The event, co-produced by Broadcast Engineering and Broadcasting & Cable magazines, brought together about 150 high-level station, group and network engineers, managers and vendors. Among the vendors present was Omneon Video Networks, which has helped propel the transition from linear production methods to a server-based IT approach where content files, not video and audio signals, are shared, edited, scheduled and played to air.

HD Technology Update spoke with Paul Turner, Omneon VP of product marketing, about HD, the impact it's having on viewers and the industry in general and its integration into IT workflows.

HD Technology Update: Recent In-Stat research indicates that there are just fewer than 2 million HDTV over-the-air viewers nationwide and that the number isn't likely to grow appreciably. Is OTA HD transmission a viable business, or is it just something local broadcasters have to do to stay in the game to gain access to other distribution avenues like cable and satellite?

Paul Turner: HDTV is going to be the norm in a couple of years. So, your point is, do TV stations still need to do HD so they can produce programming that will go over cable? I don't know that's going to be their reason, but there are going to be a lot of households that will start watching HD.

In San Francisco, we now have three stations doing their news in HDTV. They're doing it just to stay competitive. These stations will be getting syndicated material that's coming in to them in HD. Most of them will get it in some form of HD, and if they pass it through without some processing, there will be all sorts of aspect ratio problems, audio problems and branding problems.

HDTU: Could you give me an example?

PT: On a recent trip to the East Coast, I saw a station switching between a sports program and some local advertising. It was horrific because they were switching backward and forward between upconversion, downconversion, widescreen and center cut. All of the graphics were going out of the viewable area and then snapping back in as the graphics devices figured out which aspect ratio they were working with — and that's just the video.

The fades between ads or to and from black caused all sorts of audio problems. That's probably the most disturbing of all of them. Huge level differences, dialogue jumping from center front to L/R and back again really disrupts the viewing experience. These issues are resolved to a great degree if the local station is also broadcasting native HD for both OTA and cable.

HDTU: At this point, what is the premium broadcasters can expect to pay for HD-capable acquisition, production and playout technology vs. SD? How has that equation changed during the past few years?

PT: The premium is pretty minimal these days. Again, there are two approaches. The first approach is to purchase an SD product that has a very clear upgrade program so that the same infrastructure stays in place, but maybe some I/O components have to physically be swapped out. The second approach is to purchase equipment that will switch between HD and SD with software key. Both are viable. Both are absolutely manageable.

The pricing differences between HD and SD equipment are tending to go down, and the reason is pretty simple. The components used to make these products are heading into the commodity market. As they do that, the prices of your hardware invariably go down. We've seen that over and over again, and I don't see any reason why that's going to change in the short term.

Just take a look at HD MPEG-2 decoding. Not so long ago, this was a two or three rack unit piece of equipment that you could only afford to have at your network satellite receivers. Now, it's available in a chip. This commoditization will happen in the HD arena also.

The only area where people do have to consider a cost burden is in the storage, simply because the files can be larger. But even there, the cost of bulk storage is also dropping, which again reduces the price differential.

HDTU: Will most early adopters of IT-based workflows at stations be able to accommodate HD with their existing infrastructure, or will they need to start over from scratch? I'm specifically wondering about network cable, routers, hard drives, etc.

PT: The great news is that once you adopt a file-based workflow, a file is a file. The only thing that varies is how big it is and the bit rate you feel you need to sustain to move it around your network. HD, strangely enough, in some circumstances is no larger than SD.

Here's an example: I would say most of our HD customers are dealing with contribution quality material in the 40Mb/s-60Mb/s range. But 50Mb/s DV and I-frame MPEG has been available for SD material for several years. So, the actual increment in networking strain and even storage strain is negligible.

The increment really comes to customers who run DV25 or 25Mb I-frame MPEG as their house format and decide because of competitive pressures to go to 70Mbit HD. Now the file sizes have double or tripled. Let's not forget that generally with HD there are more audio channels per video channel, so that increase comes into play as well.

In a news environment, a similar phenomenon is seen. If you're looking at XDCAM, that's 30Mb/s up to 50Mb/s. XDCAM HD uses very similar bit rates. Panasonic DVCPRO HD records at 100Mb/s. That's a fairly substantial bit load, but Panasonic has just announced AVC-I, which they claim gives you comparable picture quality at 50Mbits to DVCPRO HD at 100Mb/s. Once again, we see that there is very little difference between SD and HD in a file-based ingest/playout workflow

Tell us what you think!
HDTU invites response from our readers. Please submit your comments to We'll follow up with your comments in an upcoming issue.