1

The main question is what the time for PCR error in DVB streaming ?

I am asking it because according to DVB standards (see additional information) PCR error retention period > 100 ms. But there is a lot of hardware which is logging PCR errors with PCR ~50 ms.

Aditional information

You can skip it, if you already know the answer, this information presents just as additional info about standards that I was found

ETSI TR 101 290 V1.2.1 (2001-05) - aka true DVB http://www.etsi.org/deliver/etsi_tr/101200_101299/101290/01.02.01_60/tr_101290v010201p.pdf

A PCR _accuracy_error occurs when a transmitted PCR value differs from what is expected by more >than 500 nanoseconds. The expected PCR value is calculated using an extremely stable internal clock >n the test device and previous PCR values. The calculated PCR is then compared to the transmitted >PCR values to check for accuracy. It is important to note that most receivers do not contain very >accurate clocks and therefore can be severely affected by this error.

ETSI TS 101 154 V1.9.1 (2009-09) http://www.etsi.org/deliver/etsi_ts/101100_101199/101154/01.09.01_60/ts_101154v010901p.pdf

Program Clock Reference (PCR) Encoding: The time interval between two consecutive PCR values of the same program shall not >exceed 100 ms as specified in clause 2.7.2 of ITU-T Recommendation H.222.0 / ISO/IEC 13818-1 [1]. Decoding: The IRD shall operate correctly with PCRs for a program arriving at intervals not exceeding 100 ms

ISO/IEC 13818-1 (Information technology — Generic coding of moving pictures and associated audio information: Systems) https://forums.xilinx.com/xlnx/attachments/xlnx/DSPTOOL/15095/1/iso13818-1.pdf

Thank you.

user1722669
  • 488
  • 2
  • 6
  • 22
  • Does this have anything to do with programming? – Carcigenicate May 24 '17 at 16:36
  • Well, I just want to know why DVB standards tell use about 100ms but many hardware manufacturers logged PCR packets > 50 ms as an error. – user1722669 May 24 '17 at 16:40
  • I'm going to vote to close as this isn't a programming question. Unfortunately, I don't know if there's a site where this is appropriate but it doesn't seem appropriate here. – Carcigenicate May 24 '17 at 16:41
  • Yes, this question didn't have any code snippets, but it's relative to programming, actually, I am asking it because we are working on integration DVB encoder in one of our products. DVB encoder that we are used DVB compatible, but first test shows that we have such PCR problem on the 50% of stream loggers. That's why I was asked a question, may be we have missed something during the integration. – user1722669 May 24 '17 at 17:24

1 Answers1

2

It's actually 40 ms, not 50 ms. But generally, both 40 ms and 100 ms are correct.

TR 101 290, section 5.2.2, note 2 to table 5.0b says:

NOTE 2: The limitation to 40 ms in the 'Preconditions' of 2.3 PCR_error and 2.3a PCR_repetition_error was removed from TS 101 154 [i.30] in 2005. The respective clause there now refers only to the 100 ms limitation in [i.1] which is recommended to be applied generally.

So I would conclude that the 40 ms has been relaxed to 100 ms, and that tools complaining about more than 40 ms have simply been implemented against an older version of TR 101 290.

  • Thank you for the answer, yes it make sence, but I actually still don't know how to solve current problem. Because a lot of people using hardware which is still logging PCR > 40 ms as an error. If there wouldn't be other answers I will mark your answer as right one. THank you again for your reply. – user1722669 Sep 27 '17 at 12:09
  • Well @user1722669, as the quote states, the limit value was changed back in 2005. If people still cling to old tools, there is probably little you can do about it other than (a) telling them to RTFM, and (b) telling them that they should have updated their tools 12 years ago. –  Oct 03 '17 at 19:57