So the discussion begins in France on the future of DTT. Will it be squeezed another 100MHz (the 700MHz band this time), or will there be a UHD HDR-bright future ahead like the HD-forum is suggesting. MPEG-2 will most likely die: there is an acceptable legacy fallout of some (cheap) settop boxes; most newer iDTVs all have HD MPEG4 decoders on board through a foresightful mandate.

The author has had a long debate with some industry veterans on the topic on Linkedin, arguing that the era of DTT has passed, and that wired IP should take over the role of base medium for TV distribution, allowing society to poor all its money into one such network. Some were not convinced; and it has to be stated that this industry is very conservative and slow to change for not always clear but undeniable reasons.

France opens 700MHz consultation | Advanced Television.

Thomson is French, and the French love French technology. DTT in France is a Thomson thing. It’s future is under debate. Thomson is demonstrating that 5 HD MP4 channels on a DVB-T (not T2) mux @ 24.8Mb/s is possible in good quality with t heir new encoder. This would allow all 30 DTT channels in France to move to 6 muxes. Unfortunately there are SD-only decoders (but stll very few iDTVs) in the market which can only decode MP2. And introduction of DVB-T2 would also constitute a legacy break.

The Thomson proposal calls for one UHD mux to be introduced (which would be T2 and HEVC) on one of the 6 muxes (squeezing 6 HD MP4 channels per mux on the remaining 5 muxes). Maybe with 45Mbit/s 3..4 UHD channels are feasible in due course.

But one can wonder: how far should one push DTT. In the author’s opinion SD is underrated: good quality 576i or better 576p @ 25/50Hz is perfect for almost all content out there on a 42 inch screen. So unless we expect DTT to have to serve a large population of 50inch+ screens in perfect quality, this all makes little sense. Why does DTT have to remain an “attractive medium” (Thomson’s report of the CSA’s opinion): it should be the other way around. DTT is an attractive medium so we innovate. It there something wrong here?

Given where the French are they should at least innovate their DTT network once. The author agrees with Thomson that one can ditch the support for MP2 broadcasting, especially since few iDTVs seem to be affected and replacing one cheap STB for another after 10 years should be socially acceptable cost. Moving all channels to HD: not really necessary: 576p should do it for most channels. Innovating now to HEVC and T2: this only makes sense if you expect DTT to innovate again: say in 10 years from now and move to all T2/HEVC. That case is far less clear: see this article on this website and the associated discussion on linkedin, and considering the need for UHD on DTT.

Thanks to Videonet for this excellent report on the French situation:

Thomson HD demo: the first step to a UHDTV launch on French DTT by 2018? | VideoNet.

Plumb and Farncombe have produced a report on a change of infrastructure for the future of spectrum management for the European Commission. Their conclusions state that it is not clear that a move from broadcast (high power high tower) to cell based (Low Power Low Tower) transmission is beneficial.

The crux of the matter lies in the question posed and the scenario assumptions. The report works with scenarios to avoid making a choice on developments; which include the continued need for linear services over the air to TVs as well as mobile devices (see figure). The question is whether the trend can be predicted more accurately and/or be created using a little vision (we need the French for this).

Surely in the near future the whole of the European Population will need a (fixed location) broadband infrastructure. This is economically speaking probably more essential than today’s TV. This has been a priority for years in the EC, and for good reasons. Fiber and cable speeds, augmented by ADSL and wireless, at minimum 20Mbit/s, but as a rule 100Mbit/s and above.

Such a broadband infrastructure should be able to handle the “public” task of delivery of linear and VOD TV services to every home. It may not be completely free (at least the BB connection is not) but VOD and linear viewing can be seamless; which would be a huge step forward for all. Given where things are moving linear viewing may be out for all but a few main-public live-oriented channels. The rest of the content on today’s linear channels will be much better off being delivered through VOD. The freed up spectrum (possibly in higher frequency bands) might be partially given to Wifi local connections. If these start to run out of spectrum that may bring more value to society than even more bandwidth for cellular services.

Whether linear services should be offered for mobile applications is dubious. The demand for this (outside the home) today is very modest, market value very low; it is unclear whether it will grow. Wifi based on underlying BB infrastructure is the rule for TV on mobile. DVB-H was a total flop; is there a reason to repeat this. I.e. forget about the right two scenario in the figure, specifically the bottom-right. The emphasis (for TV) should be on the top left.

The case should be investigated to stop investing in DTT completely, certainly in UHD, and even in HEVC and DVB-T2. Instead the focus should be to create a BB based TV infrastructure as soon as possible that includes delivery of public services should be investigated with the highest priority. Of course one could offer a limited set of key channels on a modest quality via a DTT for an extended period to bridge the gap.

Should the conclusion of the Plumb and Farncombe report be revised if the above bears merit: no! There indeed is no reason to think that a combined network makes sense. The,  key business for it, mobile linear TV, is very weak, and probably better options exist that free up even more spectrum for new uses (cellular, Wifi).

Argument for converged broadcast platform unresolved

Plumb and Farncombe executive summary

Plumb and Farncombe full report

The UHF band is a key band for expanding mobile internet services in the future. It is used today in many countries for DTT services. The question is: how / when to migrate the services. The EU has opened a consultation on the topic.

The provision of a free TV medium may be important in many countries. But providing BB internet services to all homes may be more important these days. It is a reasonable hypothesis that the internet could carry basic TV services for free by 2020/2022 in most countries, certainly in western Europe. Would that be a basis for shutting down DTT? Is there a case for a LTE / 5G compatible broadcast mode for mobile reception that can also provide basic TV distribution for stationary purposes.

These are very interesting times. Hopefully the EBU will do some serious soul searching on more forward looking strategies to the occupation of spectrum and the distribution of public TV in the future. Put full emphasis on 2nd screen and VOD, do not get stuck in the past of linear channels:  these will serve a purpose in 10 years from now but a gradually decreasing one, and rather than more there will be fewer true broadcast channels. VOD will be more than 70% of all viewing.

The only other thing we should hope for is some more space for license free applications (wifi etc.).

BBTVnews: EC opens public consultation on UHF band.

The Dutch spectrum is used at the moment for the cheapest basic TV service in the country using a (partial) SFN network. The network (Digitenne) was created 10 years ago and is using DVB-T/MPEG-2 (Generation 1 DTT). The service is gradually loosing subscribers as its owner KPN is luring the customers to IPTV/Fiber services and has increased the price to a still the cheapest (cheaper than analog cable) but not all that cheap level of 10 euro/month. There may be a place for a cheap and cheerful TV service in the Netherlands, but the service will need a technology upgrade in the future. It is questionable whether its current owner KPN is willing to invest in new network and settop box technology (based on presumably HEVC and DVB-T2 or LTE-broadcast technology).

The Dutch government has decided that the 700MHz band should go to mobile applications by 2019; in the light of the above this seems in the general interest; misusing the precious bandwidth for a commercial service of low value is not a good return for the public. But on the other hand: there is a case for a cheap and cheerful TV distribution technology for a new era: LTE-broadcast or IP  “straight-up”?

CSI Magazine: Spectrum blow for Dutch DTT.