Video services

How Will Cable Deliver IPTV?

Cable's big MSOs seem to feel IP video is the way of of the future, but how will they get there?

To use their networks to provide IPTV, cable MSOs can: 1) deliver IPTV directly through the cable modem termination system (CMTS); 2) bypass the CMTS and deliver video through edge QAMs; or 3) deploy a new class of gateways capable of encapsulating traditional QAM-based digital video and ensuring it can be understood by IP-based devices in the home. (See Bypassing the IPTV Hype and Cable IPTV Debate Brews )

Engineering execs from Cox Communications Inc. , Charter Communications Inc. , Time Warner Cable Inc. (NYSE: TWC), and BigBand Networks Inc. (Nasdaq: BBND) kicked around the pros and cons of each approach on a Multichannel News Webcast yesterday. Other than BigBand, a vendor favoring the bypass approach, the panelists weren't yet ready to bet all their chips on any one option. (See BigBand Lays Cable IPTV Groundwork.)

"There's a growing desire to get down to one network," said Louis Williamson, a senior engineering fellow at TWC.

That's one reason why Jay Rolls, Cox's SVP of technology architecture, is somewhat partial to the video-over-CMTS approach, which originates everything in IP from the get-go.

"It gives you some degree of nimbleness to hit other devices in the home," he said, noting that it would allow the MSO to take video "off-net," and send it, like a Slingbox, to customers that want to access video services when they're not connected to the operator's cable network.

Doug Jones, the chief cable architect at BigBand, replayed one of the economic arguments his company has made for the bypass approach: that the costs of edge QAMs greatly undercut the costs of core CMTS downstreams. [Ed note: Granted, it's taken cable about a dozen years to cost-reduce video QAMs to their present level. It probably won't take that long for them to do the same for a CMTS QAM.]

"I definitely see the appeal," Rolls said of the CMTS bypass economics.

Doug Ike, the VP of advanced video and applications engineering at Charter, said emerging "modular" CMTS designs, which physically separate the downstreams from the "brains" of the core CMTS, should make the video-over-CMTS economics more attractive.

"I think there needs to be a bit of an evolution in the CMTS platform to adapt to [IPTV]," Rolls said. But if the MSOs indicated they wanted to move in that direction rapidly, "I'm sure the CMTS community would respond," he added.

But can MSOs afford to wait for next-gen, video-optimized CMTS technology to show up? BigBand's Jones doesn't think so, noting that a CMTS bypass architecture can foist cable to the IPTV forefront today.

"Bypass technology certainly has its attractiveness in the near-term," Charter's Ike agreed.

But Rolls wondered if bypass represents merely an interim step of where cable wants to go, suggesting that it may behoove an MSO like Cox to hold off and then make the move to IP in one fell swoop.

TWC: Open to the gateway
The approach that could get MSOs closer to IPTV is a Docsis 3.0-capable, multi-tuner gateway/set-top that can speak the language of MPEG and IP video transport.

"We are looking at going down this path," Time Warner Cable's Williamson said. He didn't say how, but it's widely known that the MSO is pursuing such a device under the internal "Santa Monica" code name.

TWC is keen on the gateway because it leverages the MSO's existing networks and combines all of the signals into one device, allowing for multi-room DVRs and media sharing. "We think it will lower costs overall with homes that have multiple services... and multiple outlets," he said.

Others aren't so sure. Rolls suggested that such a device won't be had for cheap, and presents another potential bugaboo: If something goes wrong in that complex gateway, cable MSOs probably won't be able to avoid a $50 truck roll to fix it.

Rolls also questioned the device's ability to deliver video off-net. "It seems a little handicapped to me," he said.

Although this debate is far from solved, MSOs are already thinking about when they'll need to make these hard decisions. Rolls, for example, put it in the two-to-four year timeline. Ike said some trials using "interim technologies" should start showing up by 2010.

— Jeff Baumgartner, Site Editor, Cable Digital News

Interested in learning more on this topic? Then come to Cable Next-Gen Video Strategies: Competing in the Three-Screen World, a one-day conference that will take a comprehensive look at the cable industry's attempts to generate revenues and fend off its rivals by deploying next-generation video technologies. To be staged in Atlanta, June 25, admission is free for attendees meeting our prequalification criteria. For more information, or to register, click here.

Honestly 12/5/2012 | 4:04:20 PM
re: How Will Cable Deliver IPTV?

Jeff, discuss how Fios is delivered.  It is a hybrdd IPTV/cable play and they are kicking At&t's rear.  No reason to re-invent what can be copied.

paolo.franzoi 12/5/2012 | 4:04:20 PM
re: How Will Cable Deliver IPTV?

Except that FiOS has more Bandwidth than cable.

FiOS has a 870 MHz video + the BPON or GPON data streams.  Today, FiOS runs their VOD over the data streams in IP format with the normal broadcast in the RF domain.  Data and Voice run over the BPON or GPON wavelengths as well.  So, there is more RF spectrum free and the bandwidth for a PON system is divided over a split of 32 users not several hundred as it is in an HFC system.

So, it is a bit of rocket science when you have an inferior infrastructure.



Honestly 12/5/2012 | 4:04:20 PM
re: How Will Cable Deliver IPTV?

Thanks brook.  Great opportunity for infrastructure up-grade then, no?.  Could the OpX be kept down by just upgrading the cable infrastructure.?



Jeff Baumgartner 12/5/2012 | 4:04:18 PM
re: How Will Cable Deliver IPTV?

Based on what's been discussed by MSOs and the suppliers so far, I agree that it's fairly clear that adding IPTV to the cable mix won't be rocket science. It seems as though they are still grappling with which approach makes the most economic sense and addresses what they really want to do -- not just replicate their broadcast and vod content in IP, but bring in Web video and add place-shifting to the mix.

But it's also true that whatever they end up doing will have that QAM/IP hybrid approach that VZ is doing now, but minus the aid of FTTP.

johnhgn13 12/5/2012 | 4:04:02 PM
re: How Will Cable Deliver IPTV?

All of the MSO execs participating in the webinar agreed that the CMTS approach is the way they want to deliver all IP services, including IPTV, in the long term. i agree with Jeff's comment that some MSOs are grappling with the economics of IPTV in the near term, but there was little support among the panelists for Bypass as a near-term strategy. Doug Ike said Charter is "on the fence" about whether Bypass is an appropriate technology for the near term, and Jay Rolls of Cox described the Bypass solutions as "ideas" even though the concept has been talked about for several years. Jay added that Cox is looking for "standards and interoperability," which effectively eliminates the proprietary Bypass solutions from consideration for deployment.

While the CMTS approach for delivering IPTV may not be rocket science, there are many ways for MSOs to optimize the solution architecture to meet their technical and business objectives. Perhaps the LR editors can invite MSOs and CMTS vendors to share views on how best to architect a DOCSIS 3.0 network for IPTV and gracefully transition from current digital video systems.

Jeff Baumgartner 12/5/2012 | 4:04:01 PM
re: How Will Cable Deliver IPTV?

That's an interesting idea, John. I just wonder how many different answers we might get, since we seem to be falling way short of anything resembling uniform buy-in yet. 

I've found it interesting that some of the initial iptv bypass deployments (those made public, anyway) have been mostly in Korea (perhaps a product of the regulatory environment there?) , with a little less going on in Europe.  Then we see that Guangzhou Digital Media Group in China is doing iptv directly through the CMTS...all this makes it more difficult to identify any specific trends.   In the US, meanwhile, (and with Virgin in the UK, so we hear) there appears to be more interest in the souped up gateway approach...but I'd love to get a bunch of vendors and msos in the room with different motives, hash it out, and see what sticks when we're done.

Sign In