A day after Qualcomm delivered its seismic shock to the mobile TV industry, by convincing Cingular Wireless to agree to take service from its MediaFLO subsidiary, the DVB-H community was asking itself if this was a total US shut out for the promising, and so far globally dominant, mobile TV technology.
Those that answered this question for us at the 3GSM show in Barcelona, where Qualcomm sprung its announcement, gave a definitive ‘no,’ offering a variety of hopes for DVB-H in the US, from landing the Sprint mobile TV contract, to a potential merger between the two opposing DVB-H factions HiWire and Modeo, to the idea that Verizon Wireless and AT&T’s Cingular would be ‘mad’ to put all their eggs on one basket.
But to us the writing was on the wall when Qualcomm stole away the first hope for a DVB-H victory in the US, Verizon Wireless, announcing some 9 months ago it’s allegiance to its existing chip and technology supplier in return for what most people say were simple discounts from Qualcomm.
Put yourself in Cingular’s place and it is easy to see, as we have suggested in the past in Faultline, that it’s real mission is to ensure that Verizon doesn’t overtake it, with a weapon in its arsenal that Cingular doesn’t also have.
We can hear the doubts in the collective mind of Cingular, ‘What if MediaFLO really IS better than DVB-H,’ or ‘What if it takes too long to build a second network?’ or worse still, ‘What if the second network runs out of money and we have to prop it up?’ Just as many other operators the world over have chosen DVB-H simply to ensure they were in the pack, and shared the same collective fate of all the cellular operators in an existing territory, so too Cingular was happiest if in mobile TV it shared the same quality of programming, technology and network build, good or bad, with its closest rival.
Cingular has the iPhone AND the same mobile TV technology as Verizon, and even has a strategy for differentiation from Verizon. Surely that combination is good enough to keep it an inch ahead of Verizon with its 59 million subscribers to Cingular’s 61 million.
The hope that Sprint will somehow save the day for the DVB-H community is pretty forlorn. Sprint has around 90 MHz of 2.5 GHz spectrum, and insists that it will build out a multicast in WiMAX for TV, and it may well do that. But it also has the same issues as Cingular. There is as much chance that Sprint will cave in and buy into MediaFLO, to ensure it is not left behind in mobile TV, as there is that it will change tack and endorse DVB-H ‘ either could happen, neither are likely.
The truth is that Sprint has always made it clear it wants to be master of its own destiny. If it uses MediaFLO in 700 MHz, it is just copying the two other big cellcos in the US. If it tries to build out a MedaiFLO on its own, in 2.5 GHz, that’s a very different and far more expensive option, and that just flat out won’t happen.
No, either Sprint will take its partnership with the four biggest cable operators in the US, and build out a joint unicast-multicast service, partly in EV-DO and partly in WiMAX, OR it could take up either the HiWire or the Modeo service, because it might dominate that exclusively.
But let’s look at that for a moment. In the same way that Cingular has been able to grab a piece of MediaFLO from Verizon, promising to launch before the end of 2007, then either Verizon or Cingular might grab a piece of any DVB-H environment that Sprint backs, and it would be back to square one, in an equal face off with its two large US rivals.
So either Sprint has to have a 5 year exclusive, or buy the network outright. It makes far more sense to build out video capabilities, which are a little pioneering in WiMAX, but which once built would be more than a match for MediaFLO and belong totally to Sprint, so they can’t be shared.
So if the Sprint door gets well and truly slammed in the face of DVB-H, then any hope of getting two separate fresh networks funded in the US, go right out of the window. Modeo’s parent Crown Castle does not have enough cash to build a network, without borrowing money on the back of a deal with a major cellco. HiWire, a joint venture between spectrum speculator Aloha and SES Americom, has spent virtually nothing announcing its intent to build in 700 MHz, and paying for a solitary trial.
Aloha has enough spectrum to really be competitive. Modeo is looking for a ruling from the FCC to use extremely high power on its transmitters, in the hope of halving its costs to build a network in 1.6 GHz, which would still make it around double the network cost of MediaFLO in 700 MHz.
A merger may seem the only sensible thing to happen, but now seems less likely that ever. Both networks needed to treat Cingular like royalty and land it as a customer, proving they had the nerve to risk everything and build out so that Cingular was never at financial risk. To Cingular the nightmare was the chance of being left high and dry and needing to reach into its own pockets to pay for the network build. By plowing their own furrow Modeo and Aloha must have appeared greedy and unco-operative to Cingular and not worth the risk.
But if these two networks can come together, where will they get the money from to build a network which will rival MediaFLO. They could raise debt, but only with a guaranteed customer base. They could ask the Mobile DTV Alliance members in Intel, TI and Nokia and others, but they should have contributed funding before now if they were ever going to interest Cingular. Intel might conceivable confirm that it will add DVB-H chips to laptop motherboards, something that it has threatened to do, but always stopped short of promising. That might at least mean that there was a ready made market in DVB-H for non-handsets, a market that so far Qualcomm has overlooked.
But even Intel has to look to its margins and see that although it makes lot of sense to put DVB-H chips into laptops it is shipping outside of the US, to offer them in the US, it would just sow the seeds of confusion. Surely Intel will swallow its pride, take a trip to Qualcomm HQ and license or buy MediaFLO chips at a decent discount, to help Qualcomm shore up its mobile TV defense of the USA. Outside the US Intel can ship DVB-H dongles and motherboards, inside the US it will have to be MediaFLO or AMD will steal the limelight and go after a PC exclusive at Intel’s expense.
This is much the same for DVB-H players and MP3 players that include DVB-H chips. There will be a plentiful supply of Chinese and Taiwanese devices ready to wing their way around the world as DVB-H mobile TV begins to sell in earnest. But in the US there will be a black hole unless those same devices can negotiate with Qualcomm for MediaFLO chips at reasonable rates. Qualcomm will have to ask itself if it is worth making these companies pay through the nose for joining its US mobile TV stranglehold or whether they can contribute to the MediaFLO cause around the rest of the world and ensure there is no door left ajar for DVB-H in the US, in which case its pricing may be reasonable.
One DVB-H chipmaker that wants to remain nameless told us, ‘Cingular has made a big mistake. Instead of buying into a market that has real competitors and a standard, they will be blackmailed with high chip prices and a monopoly supplier.’ It’s clear however that the same ‘high’ chip prices were thought of as ‘discounts’ bartered with Verizon and they can’t have it both ways. Either Qualcomm gave away the Verizon deal or it charges high prices. In the end Qualcomm may charge handset suppliers high prices for the chips, but not the operators.
In fact MediaFLO insists that it has the rights to license the technology as if it invented every layer of it, and despite the fact that MediaFLO is largely a new packeting system set on time honored radio techniques ‘ OFDM, time and frequency interleaving, Forward Error Correction ‘ Qualcomm says it will license MediaFLO for free to anyone that has already agreed to license its CDMA technologies.
That’s a real problem in that much of the underlying technology will need other patent licenses, and the idea of ‘bundling’ patents makes it less than equitable to its main rivals. As the split between Nokia and Qualcomm becomes more violent, there will be court battles. Nokia might suggest that MediaFLO is little more than a DVB-H copy, and that Qualcomm has to answer to all the DVB-H patent pool.
Other mobile TV suppliers like DibCom and Siano and Frontier have no chance of paying for a CDMA license, so they will have to pay Qualcomm a license fee ‘ perhaps over 10% of the bill of materials for MediaFLO ‘ while Samsung and LG and others have no extra to pay. That kind of licensing formula is almost certain to be considered anti-trust and will take a court to unravel, but to us it sounds downright illegal.
Perhaps the DVB-H community could hide behind the legal actions of Broadcom and offer their skills to help it make MediaFLO chips, under some of the immunity its recent court actions offer it, but Broadcom would need to make a lot of chips to make a move like that worth its while.
MediaFLO says that it has already licensed 5 other chip suppliers to make MediaFLO chips, but the only one made public is Newport Media, which is a minnow compared to Qualcomm, and it could underprice Newport and take it out of the market at a moment’s notice, except of course it prefers to have competitors in the market to continue the suggestion that MediaFLO is a standard.
Another idea we heard at the show is that Cingular and Verizon are planning to offer both MediaFLO and DVB-H in the future. Well that’s possible, why choose, why not let the market choose ‘ except that without their blessing it is unlikely that any DVB-H network will be built in the first place, and they know that and there have been no blessing forthcoming. If they gave their assent now it would just create another competitor to dilute their efforts.
Also the next tier cellcos in the US look no more hopeful and both Alltel and T-Mobile are known to be at least looking at MediaFLO, in order to remain competitive.
When we originally talked to Qualcomm about MediaFLO it made it clear that operators would be able to differentiate by having some channels allocated to their customers only, and they might negotiate the content for these channels themselves. In the 6 MHz that MediaFLO has available to it, it has said publicly it can offer 20 TV channels, and many clipcast services, perhaps as many as 80 or so, and some audio channels.
A clipcast is a file that is squeezed into spare bandwidth alongside the broadcast signals, as the codec and statistical multiplexer finds it does not need all of the bandwidth allocated to it to render a particular scene. Clipcasts signals can be sent as best effort, like internet html pages, and only signal their arrival to a customer by raising an icon once they have been fully sent. News and sports carousels of video snippets can be sent in this way, with only small parts of it updated throughout the day, and many subject channels can be sent in this way. This is over and above audio and TV channels.
So the bouquets of Verizon and Cingular are likely to be 5 TV channels of their own, and perhaps 10 common TV channels, plus a huge variety of clipcasts, maybe as many as 20 clipcast channels each and 40 common channels.
It would be hard to allocate another bouquet to Sprint or to T-Mobile and so any cellco wanting to join the battle for mobile TV in the US may have to wait until 2009 when more spectrum becomes available through the digital switchover of broadcast TV to digital, which will free up huge amounts of UHF spectrum. Of course Aloha Partners will be itching to extract value from its own spectrum before there is a Tsunami of new UHF spectrum from the digital switchover.
That means it has two years to make some headway in using its two chunks of UHF spectrum which are capable of offering well over 40 mobile TV channels. Or Aloha will find that the only way for it to exit the investment is to make the spectrum available to MediaFLO when it looks to expand its channel base and perhaps open the door on the services for Sprint, T-Mobile, and Alltel or even perhaps offer the cable operators an alternative to Sprint. The MediaFLO network would need little further investment to exploit that spectrum, and in 2008 Qualcomm might be the only knock that Aloha hears at the door.
In this way Qualcomm has re-drawn the map that it helped to draw in cellular voice services, with great chunks of the US and the Far East taking CDMA despite the efforts of the traditional cellular network providers, but this time it has simply erased the CDMA sign over America and replaced it with a MediaFLO sign.
If the shut-out of DVB-H is maintained, then it will take a tiny amount of time before the US financial community realizes that MediaFLO is worth more than the two satellite radio networks combined, and makes Qualcomm an offer it can’t refuse for hiving it off and collecting a check at least ten to 20 times bigger than its investment to date in MediaFLO, currently cited at $800 million.