One of the surprising messages that we received at the Qualcomm industry analyst briefing was that the company believes that MediaFLO is alive and well and could become global at any minute. We’re not quite sure how much credibility to give this message, but the company certainly continues to believe it, despite set backs.
One of the biggest problems that Qualcomm has for its DVB-H lookalike technology is the failure of DVB-H itself to make any major impact around the world, and the fact that in the USA, where the two major cellular networks, Verizon and AT&T, have signed up for the service, it appears to be going exceptionally slowly.
The Qualcomm version of the story throws a different light on matters ‘ it cannot light up all of its spectrum at full power for some time, until the spectrum it bought some years ago is fully vacated by other services, and it cannot light up its newly purchased spectrum at all, until February 2009. It is this that has persuaded Verizon to market the service under-aggressively and it is this lack of flat out commitment that has in turn prevented Qualcomm subsidiary MediaFLO USA, the organization that runs the network there, from spending on more expensive content.
So Qualcomm is quietly building out the quality of its service ‘ through the addition of addressable advertising services using its Xiam acquisition’s technology for selecting the right advert to show any given viewer, and through the build out in territories that it will not switch on until next February. It insists that it is also signing major TV shows that will run as catch up mobile TV versions (rather than simple simulcast) and it won’t switch these on until February because there are up front costs associated with them, and its two major cellular partners are not yet ready to push the service.
Also it is busy trying to convince AT&T and Verizon to switch on the more esoteric, but potentially highly successful datacasting capabilities, which includes clipcasting ‘ sending pre-digested themed broadcasts of goals or touchdowns or home runs or whatever, to individual clusters of phones.
The company gives the impression that it will flick a switch and motor into overdrive come February 2009. However we have always said that local broadcasters will do much the same, and they already have the content line ups within their own territory, but will use the ATSC mobile handheld (M/H) technology completing standardization this month and being installed throughout the fall, for the early adopters to be ready come February 2009.
The question that Qualcomm executives raised with us was ‘just who is going to bring the ATSC technology together into a single coherent proposition for the public,’ which is valid, but around the world, free to air mobile TV uptake happens around 3.5 times faster that even low paid subscription services ‘ so we reckon that someone is going to be tempted, even if initially it is just an MVNO like Virgin Mobile or a desperate operator like Germany’s T-Mobile, badly lagging in the US, or even Sprint.
Qualcomm may deliver the chips for around 70% of this market, but the handsets themselves come from Qualcomm partners like Samsung and LG, both of whom have vested interests in ensuring ATSC M/H is a success, since they have intellectual property in the technology. It is true that if AT&T and Verizon do not initially offer handsets that have this capability, then the local broadcasters will have few champions to turn to, and could only end up with a handful of specialist handsets, from smaller operators, along with PC dongles and specialist portable TV sets with which to stimulate the market. But if the uptake rate is 3.5 times the rate of paid services, in a segment squeezed to around 30% of the market, ATSC will still give MediaFLO a run for its money. If that happens, and we believe it will, it is only a matter of time before AT&T or Verizon or both, give in to LG and Samsung wishes to offer a handset that offers both MediaFLO and ATSC M/H ‘ both free to air and pay TV in one, if you like.
Given the close partnerships that Qualcomm has with these Korean handset makers, we cannot believe that technical talks have not happened with both, and also with partners such as Harris and Rohde and Schwarz, dominant in US transmitter sales, especially since Harris is a partner in MediaFLO as well as another inventor of the ATSC M/H standard.
The big problem that ATSC presents to MediaFLO is that it can be brought about with a $100,000 digital exciter for a transmitter upgrade plus another $100,000 or so of software and equipment additions for each TV transmitter station. This is a far cry from the $1 million to $2 million in equipment and transmitter costs for each new market that MediaFLO enters. The local broadcasters are also fairly desperate for new revenues and ATSC M/H has been seen as something of a saviour for the broadcast industry, so it will be keen to invest.
There is the idea that for each local TV channel that is transmitted, there is room not for just one mobile TV channel in ATSC M/H, but for several, some say as many as 4 or 6. That’s far too many for local stations to sensibly fill, and the outcome could be that someone comes up with the bright idea of aggregating spare transmitter capacity over wider territories and adding new, or even subscription content, countrywide. Once again who is it that is going to bring this content to market, since it is only going to create a hike in the local advertising market, because the services are free to air?
This is the type of play that would be right for a Clear Channel or MobiTV or Sirius Satellite Radio to take up (despite existing hostility with US broadcasters) or even Dish (which is expected to have its own mobile TV service come 2009). We argued to Qualcomm that surely MediaFLO is the natural partner for such stations, and that the two capabilities ‘ ATSC and MediaFLO could be created side by side on-chip ‘ to which we were told that ‘ATSC takes up a lot of our time,’ but that just might be carrying out scenario planning to defend against its launch.
Outside of the US Qualcomm still has high hopes for MediaFLO technology in the UK, Japan, Taiwan, Hong Kong and Malaysia. But in the UK Qualcomm will have to risk a further $250 million to build out a network in very unpromising 1.4 GHz spectrum; in most other territories it much duke it out with DVB-H, which is better rooted with stronger partners, with the exception of Japan where there are already 25 million plus, ISDB-T mobile TV viewers, using their handsets for TV for over an hour a day on average, using a free to air model.
Here Qualcomm has produced a special chip which offers both MediaFLO and ISDB-T and it talks about offering pay TV alongside free TV with its partner KDDI. This is exactly what we are expecting it to do in the US once the ATSC M/H service gets off the ground, so it is not such an alien idea for Qualcomm. Also its MediaFLO chip offers DVB-H, so it can attack Taiwan, Malaysia and Hong Kong in the same way ‘ with a dual chip strategy. There is a small problem that ATSC is the only system that does not use native OFDM for transmission.
So far no-one has taken the MBP1600 ISDB-T hybrid from Qualcomm and the company only sees that happening if MediaFLO gets a grip there.
We asked about the implementation of DVR capabilities into MediaFLO handsets and Qualcomm said that it was unlikely to happen, citing considerable upset when you switched on to watch a program that you had previously recorded, only to find at a crucial moment in the recording you were driving through a tunnel and the recording went blank. MediaFLO said that it would augment this with streamed file delivery over high speed mobile broadband or over wi-fi or use sideloading through a fixed broadband network. That might be true, and we see a big future for real DVRs sideloading content onto handsets, such as from U-Verse or from FiOS TV, but it seems pointless claiming that the MediaFLO system has better error correction and is therefore more effective than DVB-H, if you can’t rely on it to establish an accurate DVR record.