DIRECTV Satellite Discussion D-14 @99W

Discussion in 'DIRECTV General Discussion' started by Sixto, Jun 17, 2010.

  1. slice1900

    slice1900 Well-Known Member

    10,533
    1,494
    Feb 14, 2013
    Iowa
    As Diana pointed out, the network is just fine. Gigabit ethernet is super cheap, wireless supporting hundreds of megabits is reasonably priced (though harder for a home user to install properly so not a very good solution for streaming video, be it HEVC or MPEG4 encoded) and MoCA's 170 Mbps top rate has plenty of headroom. Directv can always update to MoCA 2.0, which doubles the bandwidth and makes up for the doubling imposed by HEVC.

    HEVC encoding is totally irrelevant to what you were responding to. Encoding happens at Directv (or before they receive the content) not in the customer's home. It is totally irrelevant to RVU.
     
  2. slice1900

    slice1900 Well-Known Member

    10,533
    1,494
    Feb 14, 2013
    Iowa
    I think it wouldn't be any problem to fit two very high quality 60 fps 4K channels in a Ka tpn. The specs I've seen for HEVC encoders are all around a 15 Mbps average bit rate.

    They might prefer however to put a single high quality 4K channel in with some less important HD channels, so they can allow the 4K channel to demand quite a bit higher instantaneous peak rate, at the momentary expense of some less watched HD channels.
     
  3. Diana C

    Diana C Hall Of Fame DBSTalk Club

    2,117
    294
    Mar 30, 2007
    New Jersey
    Mixing differently encoded video (MPEG 4 and HEVC) is problematic. The statmux would have to understand both bit streams and be able to provide feedback to both compression engines to keep the aggregate data rate under control. It may not be impossible, but I don't see such a system being built. AFAIK, no one ever built a MPEG 2/MPEG 4 hybrid statmux, so a MPEG4/HEVC one is not too likely.
     
  4. HoTat2

    HoTat2 Hall Of Fame

    7,863
    256
    Nov 16, 2005
    Los...

    You mean in the above comments using the current MPEG-4/AVC for 4K?

    Since HEVC isn't anywhere near ready yet.
     
  5. egakagoc2xi

    egakagoc2xi Cool Member

    53
    0
    Jul 1, 2010
    DuTGGFTYrtrffg
     
  6. HoTat2

    HoTat2 Hall Of Fame

    7,863
    256
    Nov 16, 2005
    Los...
    :confused:

    Pardon my lack of knowledge of internet jargon, if that is what this is.

    But, what does this mean egakagoc2xi?
     
  7. P Smith

    P Smith Mr. FixAnything

    25,966
    510
    Jul 25, 2002
    W.Mdtrn Sea
    yeah... H.264 if tell you precisely... H.265 is new compression algo standard aka HEVC
     
  8. egakagoc2xi

    egakagoc2xi Cool Member

    53
    0
    Jul 1, 2010
    Sorry, it was my two year old son who took the iPad.

    About the nick is to lame to let you guys know.
     
  9. RAD

    RAD Well-Known Member

    16,621
    185
    Aug 5, 2002
    Dripping...
    DuTGGFTYrtrffg


    Actually what was one of the better posts then I've seen around here :hurah:
     
    1 person likes this.
  10. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
    Most DIRECTV installers aren't licensed for CAT5 so Gigabit would have to be left to someone else. Any technology that is based on RF is subject to interference and other bad things that may prevent the user from getting the full measure of theoretical throughput.
    Will the typical DBS coax network (including splitters and BSFs) support support MoCA 2.0? That MoCA 2.0 adds to the MoCA 1.1 frequencies used will add another level of complication when it comes to serving both older and newer equipment from the same server. It is notable that the frequencies in the sub-GHz range didn't change with the introduction of MoCA 2.0 so unless they do something quite magical, bonded MoCA 2.0 may not be possible with the DBS compatible versions.

    http://www.mocalliance.org/news/pr_100615_MoCA_Announces_MoCA_2.0.php

    Finally, as Wi-fi spreads and people start adding proprietary systems using the same frequencies (like the wireless Genie), the Wi-fi spectrum is likely to get jammed in more and more situations.
    The question is bandwidth for the content as opposed to the RUI. There's little point in doing UHD RVU if there's not enough local bandwidth to stream the content. My point was that real time encoding probably isn't as efficient as encoding that can take its sweet time so it is conceivable that the 2:1 ratio will be a distant goal in terms of real time compression. This would take up more satellite bandwidth as well as local bandwidth.
     
  11. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
    By the same token, mixing pre-compressed video of a particular format with on-the-fly compressed video of that same form may also be complicated.

    This is part of my real time compression of UHD argument.
     
  12. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
    The specifications for DIRECTV HD (AVC encoded) are around 7Mbps, but VOS has documented instantaneous rates as high as triple that.

    It is imperative to consider whether the quoted rates are for real-time or more intensive (much slower than real time) compression.
     
  13. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
    You seem to be assuming that the network in question might well be dedicated to this single load. Get some other traffic on there and things will likely change. Half duplex performance goes down rapidly with multiple streams of traffic.
     
  14. inkahauts

    inkahauts Well-Known Member

    24,580
    1,425
    Nov 13, 2006
    Vos found those rates in mpeg2 ota. Never seen anyone see 21 on a sat fee from DIRECTV for anything.

    I've started to wonder if the three streams maybe isn't a tech limit but a future limit for UHD.
     
  15. Diana C

    Diana C Hall Of Fame DBSTalk Club

    2,117
    294
    Mar 30, 2007
    New Jersey
    DirecTV installed systems (which, as you are often quick to point out, do not use Ethernet) run on a dedicated coaxial network, with at least enough bandwidth to handle a UHD stream along with at least 4 normal HD streams simultaneously. As I said, the ONLY home LAN configurations that would be bandwidth constrained are old 10Mbit Ethernet and 802.11b wireless. But these wouldn't work very well for regular HD either.

    As far as "realtime" versus "take your sweet time" compression, let me just point out that MPEG4 AVC originally needed a minimum of two passes to compress video significantly more than was possible with MPEG2. If HEVC realtime compression can't get a UHD bitstream down to around 20mbit/sec then UHD will only be used for off-line content like movies and other pre-recorded content.

    Keep in mind that DirecTV and Dish had to invest in MPEG4 compression equipment only because most sources were being delivered in MPEG2, since that was the (ATSC) broadcast standard. If UHD ever breaks out of BluRay or on-demand content then the content originators (HBO, Showtime, etc.) will have to compress UHD down to fit on a satellite transponder themselves. It therefore very likely that should linear UHD ever become a reality it will already be compressed using HEVC by the time it arrives at the cable and satellite operators' facilities.
     
  16. slice1900

    slice1900 Well-Known Member

    10,533
    1,494
    Feb 14, 2013
    Iowa
    Your argument doesn't even make sense, as usual. First you try to imply that existing splitters and BSFs won't support MoCA 2.0. Then you go on to say that MoCA 2.0 uses the same frequency range as MoCA 1.1. You're wrong, as usual. MoCA doubles the frequency range used. Partly for the bandwidth increase (>400 Mbps in a 16 node network versus only 140 Mbps on a 16 node MoCA 1.1 network, with some help from more complex modulation) and also to allow MoCA 2.0 to simultaneously operate over the same coax network as MoCA 1.1. 400 Mbps is enough for the worst case of the MoCA 2.0 maximum of 16 nodes, all serving 4K streams.

    Bonded MoCA 2.0 (which would double speed again to >800 Mbps) is simply not necessary, but it works just fine in MoCA 2.0 for DBS since the E band has double the bandwidth range (300 MHz versus 150 MHz) As for your objection that splitters and BSFs won't be compatible with MoCA 2.0, the former is laughable, and the latter would only apply if for some reason Directv designed the BSF to block only the MoCA 1.1 frequencies. It is probably more of a 'gross' block than that, likely using the same filtering used on the SAT output of a typical diplexer.
     
  17. slice1900

    slice1900 Well-Known Member

    10,533
    1,494
    Feb 14, 2013
    Iowa
    How could he find 21 Mbps in OTA when ATSC is capable of carrying only 19 Mbps plus change (minus error correction, framing, etc.)
     
  18. slice1900

    slice1900 Well-Known Member

    10,533
    1,494
    Feb 14, 2013
    Iowa
    The 15 Mbps was for a hardware encoder intended for broadcast use. I assume that's real time because otherwise there's no reason to use an expensive and less flexible hardware encoder.
     
  19. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
  20. harsh

    harsh Beware the Attack Basset

    21,192
    187
    Jun 14, 2003
    Salem, OR
    Band E is not capable of supporting the MoCA 2 bitrates if DECA is already there. E band is only 200MHz and DECA burns half of that with its 100MHz band. Assuming DECA remains to serve existing equipment, MoCA 2.0 will need F Band (675-850 center frequency). I'm pretty sure your assertion that bonding isn't needed is a pretty wild guess at this point.
    Its only laughable if you don't understand resonant circuit quality (Q). Diplexers work at the spectrum level but BSFs work in narrow bands so its not really fair to lump them together.
     

Share This Page

spam firewall