1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Clock time on 921

Discussion in 'Other HD Receiver Support Forum (811, 921, 942)' started by Ken Green, Oct 18, 2005.

  1. Ken Green

    Ken Green Hall Of Fame

    1,105
    0
    Oct 6, 2005
    The processor clock time on my 921 is 30 seconds behind the actual time. I realize it's not much, but it can cause some issues with back-to-back timers. (9-10, 10-11, 11-12, etc)
    As example, if the actual time is 9:00:30am (taken from other consistent sources--SBC, Cingular, Navel Clock, etc) the EPG and channel information ribbons show the time to be 9:00:00am. I can be watching the beginning of any program, on any OTA or SAT channel, and the channel information ribbon says 8:59 for 32 seconds into the beginning of the program.
    As a result, if a timer is set to fire with a 0 pad at 9:00am, it fires 30 seconds into the event being recorded.
    We've found event recording failure(timers not firing correctly) is less likely with timer pads set at 0, most noticeably when simultaneously firing timers for OTA and SAT events.

    As example:
    9:00-10:00 - OTA event - Timer 1 - Channel 4-01 - Pad: 0/0
    9:00-10:00 - SAT event - Timer 2 - Channel 140 - Pad: 0/0
    10:00-11:00 - OTA event - Timer 1 - Channel 2-01 - Pad: 0/0
    10:00-11:00 - SAT event - Timer 2 - Channel 137 - Pad: 0/0

    In the example above, if timer pads are used, usually one, or all of the timer(s) will fail to fire, though it will show the record bubble(s) on the EPG, and it will show the event(s) being recorded in the DVR menu, but the red LED on the receiver will not be lit, and the event(s) not recorded. Usually leading to the software becoming unstable, not responding to remote commands, eventually requiring a soft-boot. One of the few things that is consistent on my 921 is if the red LED on the receiver is not lit, the event is not being recorded.

    Reading through the threads, I've become aware of the 921 and L270 issues, much the same as described above, but I've not found anything regarding an issue with the processor clock time.
    Is there a way to set the processor clock time on the 921?
    Very much appreciate any assistance anyone can offer.
     
  2. CCarncross

    CCarncross Hall Of Fame

    7,058
    60
    Jul 19, 2005
    Jackson
    If it works like every other sat based system I have heard of, the time comes from the sat and hence DISH...meaning you have no control over it...
     
  3. SimpleSimon

    SimpleSimon Hall Of Fame

    5,468
    0
    Jan 15, 2004
    "Processor Clock Time" is a misnomer. The time actually comes from the satellite feed itself.
     
  4. Ken Green

    Ken Green Hall Of Fame

    1,105
    0
    Oct 6, 2005
    OK, I understand that, but why then does my 921, 811, and 3 301's, which are all showing identical time fed by the SAT, have times which are 30+ seconds slower than the beginning of any broadcast, be it OTA, SAT, LOCAL, or any other channel, and 30+ seconds slower than any other accountable time source?
     
  5. mgcsooner

    mgcsooner Legend

    221
    0
    Dec 18, 2004
    be a good question for one who has validated this to pose to the next technical chat.
     
  6. Michael P

    Michael P Hall Of Fame

    3,117
    22
    Oct 27, 2004
    This could be caused by the latency of satellite-delivered data. You have to factor in the delay caused by the data being sent 22,500 miles up into space and then an equal amount of distance back to earth. There is also latency in the data itself. We run an i.p. based video conferencing system where I work. There is as much as a two second delay from the time a person speaks until the receiver hears the person speaking: this is over T1 lines less than 30 miles apart!

    What Dish should consider is setting their master clock ahead to take the latency issue into account. OTOH E*'s clock should be set to the programming that they are transmitting. I have noticed that the IRD time matched the "top of the hour" ID's on networks I watch over E*. You don't want to mess with that timing. Sure the time is off 30 seconds, however it's in sync with the programming they are sending. The only problem I see happening is if you are recording an OTA signal as opposed to a satellilte signal. There is some latency with OTA digital signals (compaired with the analog OTA signal) but it's not as long as the satellite delivered signals.
     
  7. langlin

    langlin Godfather

    369
    0
    Apr 8, 2005
    My 921 and 510 are both exactly 10 sec slower than the atomic clock and that's close. You should know that all major OTA networks start the Prime time out 1 minute early and make it up through the evening with an additional 60 seconds of commercials. Also affiliate stations syncronize their clocks with their networks and not "real" time in order to match local programming to it so the "real" clock time is not very meaningful.
     
  8. Bogney

    Bogney Legend

    160
    0
    Jul 11, 2003
    I doubt it. I work for one of the big four networks and our prime time starts at exactly 8:00 eastern time.
     
  9. Michael P

    Michael P Hall Of Fame

    3,117
    22
    Oct 27, 2004
    I used to sync my watch at 6:30 PM when "World News Tonight" came on the air. That seemed to be right on using an analog OTA broadcast, but via digital OTA there is latency. Satellite LIL is even later.
     
  10. Kagato

    Kagato Godfather

    414
    1
    Jul 1, 2002
    Good luck on that Michael. I've been playing phone tag for the 921 team for a couple weeks now. I have no idea why I need to confirm this with some people onthe east coast. Hard Drive is dead. The 921 knows it's dead. Adv Support knows it's dead. But I have to play phone tag with the 921 team.
     
  11. Michael P

    Michael P Hall Of Fame

    3,117
    22
    Oct 27, 2004
    The 921 "team" will issue you an "RA" once they talk with you. They just want to confirm some things with you before they ship out a replacement. Trust me, once you talk with them your replacement will be on it's way. The only thing to worry about is how UPS will handle the package. "Handle with care" apparently is not in UPS' vocabulary.

    BTW: It appears that you replied in the wrong thread. This thread is about the difference between "Dish time" and actual time.
     
  12. SimpleSimon

    SimpleSimon Hall Of Fame

    5,468
    0
    Jan 15, 2004
    No you don't - not for human-level clocks. Do the math.

    The propogation delay DOES affect online gamig, VPN, etc. when uing satellite internet, but that's a MUCH different thing.
     
  13. Michael P

    Michael P Hall Of Fame

    3,117
    22
    Oct 27, 2004
    The time is set at the uplink center. By the time the data reaches your receiver significant time has gone by. The delay is in both "travel time" and data latency (the processing of the streaming data takes time). Try watching an analog OTA station and simultaneously the same station off the sattellite LIL's. There is an extreme time lag. In the case of the LIL's add the time it takes for the programming to reach the uplink center form the local POP (actually that time far shorter in compairison to the space travel 45,000 miles round trip).

    If you were allowed to set your receiver to "human-level" time, the EPG (and PVR) would start early and end too soon for the programming streaming on the satellite.
     
  14. SimpleSimon

    SimpleSimon Hall Of Fame

    5,468
    0
    Jan 15, 2004
    Well, beings you seem unable to do your own math, the speed of light is approximately 186,000 miles per second. The round-trip distance is approximately 45,000 miles.

    That comes out to be approximately 1/4 second. Put another way, about 4 frames of video.

    But in NO case is IT any factor in why timers are "off". The problem is the networks are screwing with us.
     
  15. Ken Green

    Ken Green Hall Of Fame

    1,105
    0
    Oct 6, 2005
    With 2.72, the time on my 921 is now precisely synchronized with the USN Master clock. Before it was 31 seconds slow.
     
  16. Michael P

    Michael P Hall Of Fame

    3,117
    22
    Oct 27, 2004
    The only problem here is the signal is not traveling at the speed of light. I know for a fact that there is significant latency in the order of seconds when a signal is "bounced" off a geosynchronus satellite. Add to that additional latency issues when dealing with MPEG-2 datastreams (remember one transponder contains up to 12 channels that have to be encoded and decoded). All that processing takes time. We run a teleconferencing system where I work. It used to be analog via fiber optic lines. That signal was virtually in real time (fiber optics are at the speed of light). One day the phone company informed us that they would no longer be making the analog lines available. We now use T-1 lines on an i.p. system. There is significant latency, up to 5 seconds round trip, and that is just within one county.

    Why do you blame the "networks"? I guess somebody could figure out the average total delay and compensate by running their "network" early. Don't hold your breath waithing for that to happen.
     
  17. DonLandis

    DonLandis Hall Of Fame

    3,363
    0
    Dec 17, 2003
    MichaelP
    You are correct in that latency is an issue. Simple Simon has the math correct for approximate speed of light through space but the latency issues are concerned within process circuit delays not through free space. I have as much as .3 seconds of latency in just my digital capture off of the firewire here between what plays live from the computer and what plays direct off the VCR.
    Both Leo Laporte and his band of tech "TWITS" have spoken about this issue that they have been dealing with to enable their podcasts where different members are actually 500 miles apart but when they do them it sounds like they are all in the same room. His discussion on the How to of these conferences exposes the latency problems and has claimed they tested 5 different services and discovered that Skype has the least amount of latency of them all. Mayby you could look into this further for your company. There was no explanation on why but just that they have tested several and Skype was the best.

    Also, in satellite uplink and downlink and receivers use what are called Frame Store syncronizers that are required to sync lock the video audio to a time base standard. This aids in a multiple of issues but primarily keeps the video from going out of sync, ie picture roll both horizontal and vertical as well as locking audio to video time bases. Each process delays the video by 1 full Frame (1/30 second). A FSS could be used at any point in the process where a break in the sync lock (genlock) path is interupted. This is only one source of time delay but there are many others in the entire source to destination path. Several seconds delay does not surprise me. But 30 seconds does. I think there may be a time standard flaw here. Maybe someone's asleep at the WWV NBS receiver at the uplink center. :)
     
  18. CCarncross

    CCarncross Hall Of Fame

    7,058
    60
    Jul 19, 2005
    Jackson
    31 seconds cannot be accounted for in latency and round trip travel time...the worst case it may be a few seconds...if you have ever compared an OTA broadcast to the sat local feed, there is a little delay, but in all the tests I have done it was never more than a sec or so...
     
  19. SimpleSimon

    SimpleSimon Hall Of Fame

    5,468
    0
    Jan 15, 2004
    Yup - although I have seen slightly more delay than 1 second.

    Don has it right - it's NOT the satellite latency that could possibly cause a "human-detectable" delay, but ground processing.
     
  20. KKlare

    KKlare Godfather

    319
    0
    Sep 24, 2004
    I think a much worse offense of the Networks is starting programs before the scheduled time and extending them after. This makes multiple show recording a mess. I have had some more that a minute before and a minute or so later too.

    They are just pushing too many commercials, which I'll skip on my 921 anyway.
    -Ken

    Simon: it's 30 or 60 frames per second so 1/4 sec is not 4 frames. The signal goes up and down several times to get to us for many channels.

    BTW, I hate the fairly-new 1/2 sec paused jump of the 921, when you use to be able to inspect frame by frame. I know of key frames needed in reverse.
     

Share This Page