Monthly Archives: September 2007

The Bandwidth Math of Dual Must-Carry

by Leslie Ellis // September 17 2007

If your company’s growth depends on the cable industry’s available bandwidth, then you’re probably wondering what all this crazy double-talk about the “digital TV transition” means. Especially the part about “dual must-carry.”

You’re not alone. The “DTV transition” is a dense and trippy subject. For that reason, this week’s translation shows how to do the bandwidth math of a cable system’s carrying capacity.

Let’s put a finer point on it. Say you’re a program network. You’re pitching three new HD channels to the person at the cable company who decides what goes on, and what doesn’t.

For the past few months, you’ve been hearing variations of “love the idea, Bob, but for each of your new HD channels, I need to remove four standard definition (SD) channels.”

You cross your arms. Nod. Blink slowly. And silently wonder: How come?

The math goes like this: Cable systems built to 750 MHz have about 33 digital “channels,” each of which is 6 MHz wide and has a total carrying capacity of 38.8 Mbps. (Cable systems built to 860 MHz have 51 digital channels.)

One digital, SD stream uses 3.75 Mbps of bandwidth. About 10 can fit comfortably into a 6 MHz channel. The math: 38.8 divided by 3.75.

Likewise, one HD stream, using conventional compression, uses 15-ish Mbps of bandwidth. About two can fit into a 6 MHz channel, with some wiggle room. The math: 38.8 divided by 15.

So, for an operator whose shelves are full, adding one HD stream (15 Mbps) could well mean removing four SD streams (3.75 x 4).

In both cases, SD and HD, operators often apply a method interchangeably known as “rate shaping,” “grooming,” and “statistical multiplexing” to squeeze, say, two more SD streams, or one more HD stream, into that 6 MHz “container.”

Refresher on statmuxing: It’s like driving at rush hour when you’re in a hurry. You seek the blank spaces between the cars in the other lanes as your way to dart ahead. Same with rate shaping — it’s a way of organizing the bits more efficiently for the ride.

But all of this is before the Federal Communications Commission decided that cable must carry broadcasters in digital, and in analog, until 2012. Take a market like LA, or New York, each of which supports a couple dozen over-the-air networks. What’s the worst that can happen?

In a dual must-carry environment, for a market with, say, 20 broadcasters seeking that treatment, and they’re transmitting in HD, cable operators would be forced to clear off ten 6 MHz channels (assuming those networks aren’t already carried).

The good news is, the FCC didn’t mandate must-carry of “all content bits.” Had that passed, operators would’ve been stymied to statmux, re-compress, or otherwise handle the incoming broadcast signal, for the sake of transmission efficiency.

But back to you. At this point, you’re probably wondering: If dual must-carry is such a squeeze on bandwidth that’s already pinched, then why, oh why, would the National Cable Television Association be glad that its constituents are now “allowed” to dual-carry broadcasters for three years after the transition? Isn’t the point to take back that analog bandwidth (to make more room for your HD stuff?)

It’s not that operators planned to yank the broadcast networks as soon as their analog signal went dark. Most realize the power of being the only guy in town who can serve all the TVs in the house — even that junky one in the back bedroom — with the wire that comes out of the wall. Meaning, without having to put a box on every set.

More, the good of the 9/11 FCC decision is that its dual carry obligation lasts three years, and not “perpetually,” as had been proposed. Plus, it is any trade association’s preference that matters like this be handled by the businesses involved, not by the government.

And, you wonder: What happens, then, 2012, after the sunset? A bandwidth glut? Answer: Probably not. Probably, operators will continue to reclaim analog channels (yours included, as negotiated), at a measured and steady rate — which is hopefully fast enough to accommodate all the “more HD” that’s coming.

This column originally appeared in the Technology section of Multichannel News.

A Reader’s Guide to the Latest in the Two-Way Talks

by Leslie Ellis // September 03 2007

In case you’re out of late summer reading, or were wondering whatever happened with those “two-way negotiations” between cable and the consumer electronics industry, there’s always those 500+ pages of collective snark sent to the Federal Communications Commission on August 24.

The comments — submitted by the lawyers of AT&T, Comcast, DirecTV, EchoStar, Intel, Microsoft, Samsung, Sony, TiVo, and Verizon (among others), as well as by trade groups like the Consumer Electronics Association and the National Cable Television Association — were filed in response to the FCC’s Third Further Notice of Proposed Rulemaking, issued in late June.

That’s the one that seeks the next chapter of the one-way plug-and-play rules, established in 2002. It was the one-way rules that gave us those instantly obsolete TVs with built-in CableCard slots.

(To find the reply comment mother lode, go to www.fcc.gov, click on the “search” button at the top right of the screen, scroll down to “Search for Filed Comments — ECFS” and enter in proceeding number 97-80. Otherwise, read on. This week’s translation begins to summarize 16 of the 100+ filings.)

It really isn’t as dull as it sounds. This is five years of pent-up frustration (and God knows how many aggregate air miles), trapped inside a gag order. If you’ve been following it, it’s one of those scuffles that forever elicits the same tense retort: “I just can’t talk about it.”

After a while, we all quit asking. That’s why it feels nearly voyeuristic to dip into the pages and pages of comments.

If your time is limited, or if you’re just not into reading that many FCC filings, start with the 80-page seethe from the National Cable Television Association. It’s deliciously pointed. The gloves are off. Even the footnotes are juicy.

Cable’s irritation is directed at the collective bunch of respondents who believe, with casual fervor, that cable services — specifically, the guide, on-demand, pay-per-view, and switched digital video — should be physically decoupled from the cable plant, and handed over to the devices that would play them.

At the helm of that idea is the Consumer Electronics Industry (filing: 163 pages), with its “DCR+” proposal, where DCR stands for Digital Cable Ready, with that plus sign.

NCTA calls DCR+ “consumer minus,” and rails that “the CEA proposal would be the most intrusive regulatory regime ever established.” (Ever.)

To do what CE side wants, NCTA argues, would require a massive do-over of the protocols currently used by cable to do VOD, the guide, pay-per view, and switched video. It would require new multi-stream cards, and new versions of leased set-tops. But, as NCTA puts it, “CEA is indifferent to engineering realities in cable.”

Beyond CEA, though, it’s nearly impossible to clump out who’s on which side. Most of the respondents support parts of the cable proposal, parts of the CEA’s proposal, and parts of their own proposals.

Industrial discord, even within the constituent camps, isn’t a big surprise. Everybody wants to protect their babies. Finding useful agreement amongst the multichannel video providers (“MVPD”) who filed, however, is somewhat surprising.

The point of accord, loosely paraphrased: It’s a bad idea for government to specify technologies in a rule. Different platforms (cable, satellite, telephone) use different technologies to reach consumers, and that’s as it should be. Let networks be networks. Encourage some kind of common interface, to let consumer devices attach to network services.

Naturally, the non-cable MVPDs argued loudly that they don’t want that common interface to come from their cable competitors. AT&T played the “we’re not a cable service” card; DirecTV and EchoStar, exempt so far from any navigation device rules, asked to keep that status.

Bogus, said cable (repeatedly). The existing “negotiations” are bounded to an FCC ruling that’s more than 10 years old.  Back then, satellite providers hadn’t made much of a dent in subscription television. Video wasn’t yet a twinkle in telco budgets. DSL was the name of their game.

The NCTA’s filing should be required reading for everyone who reads this newspaper — especially if you’ve ever wondered, but didn’t want to ask, what the heck this “OpenCable Platform” thing is really all about.

This column originally appeared in the Technology section of Multichannel News.