Olivier Houot wrote: > you don't even have to go MIMO... > > DVB-T2 is supposed to offer a 30% improvement over > standard DVB-T, as a minimum, while preserving > comaptibility with current user antennas (which implies > single polarization broadcasts). Well, that was one of the original design GOALS of DVB-T2. But the scheme in the BBC research paper does something different. It gives you up to 100 percent improvement, with a different antenna, or it can give less or no improvement in spectral efficiency, with the same receive antenna. For example, you can have a 32K mode plain and simple. Or if you have multiple transmitter antennas and just a single receive antenna, you will get no more bit rate than DVB-T, but you will get potentially easier reception. Which means you can maybe increase the COFDM mode some. But in short, you can't add a design *goal* on top of the performance of an actual design, and expect the sum of both. > The french TNT will squeeze 3 HD streams in a standard > 8Mhz channel (at 24 Mbps), using standard DVB-T. Right. I remember that the CSA did this on their second HDTV DTT trial. That is a different question. That is about compression algorithm efficiency. As far as I can tell, even that isn't out of the question with MPEG-2 (H.262). These numbers are not hard numbers. If you transmit 720 at 24p, the DTV Team of a few years ago was saying 9 Mb/s with MPEG-2. And in fact, it was reported on here that you could even crunch 720p down to as low as 6 Mb/s. No doubt, it wouldn't be as nice as 720p at a higher bit rate, but you can still pretend it's "HDTV." So if you can do three HDTV streams (e.g. prime time TV shows) with an average of 8 Mb/s each, using H.264, that does not seem like way out of line. That example I gave for WETA-DT, 3 SD and one HD in the 19.39 Mb/s channel, uses 1080i HDTV. And I figure, no more than 12 Mb/s, and probably less. The SD streams would have to be 2.5 Mb/s on average to leave almost 12 Mb/s for the 1080i stream. That's awfully tight! No matter what compression algorithm you have. And going beyond that ought to be possible with H.264. However, the way you achieve those higher efficiencies is by extending the GOP. As we have seen many times, though, in practice that isn't always going to be acceptable. The DTG even alluded to this recently, in their "debates" with Ofcom. Bert _________________________________________________________________ More immediate than e-mail? Get instant access with Windows Live Messenger. http://www.windowslive.com/messenger/overview.html?ocid=TXT_TAGLM_WL_Refresh_instantaccess_042008 ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.