Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Freeview HD Lip Sync suddenly going way out of sync
2,461 views
Skip to first unread message
Dickie Mint
Nov 11, 2011, 11:51:34 AM11/11/11
to
Not seen any mention in these hallowed walls.
Visited the son in Sheffield Wednesday & Thursday. Saw a strange fault
where the sound would suddenly go way out of lip sync on HD channels.
Sometimes many seconds. I saw it on his new Sharp Freeview HD telly,
but he says it happens on another one in the bedroom and others have
mentioned seeing it to.
It can happen on any channel, we were watching both BBC 1 HD and BBC HD.
At first it seemed to coincide with opt breaks, but also happened on
BBC1 HD going from afternoon TV to CBBC. He doesn't think it happens on
SD channels, but he makes a point of watching HD mainly!
It can't be the TVs, and I noticed a thread (3rd June, 20011;
uk.tech.broadcast - Freeview HD lipsync) some while back where Glyn said
"In my experience anything that processes the MPEG transport stream has
the ability to screw-up the timing information carried in the service
Program Clock References (PCRs), component Presentation Time Stamps
(PTSs) & even possibly the video Decode Time Stamps (DTSs). At a Tx
station this could include the equipment that inserts the separately
delivered SI data or even just the conversion from 188 to 204 byte
packets to insert the Reed-Solomon error correction. Both of these
examples can involve shuffling the PID ordering around and adding delays
that will involve re-stamping the timing signals."
So as he gets Hull on BBC1 SD, and his aerial is horizontally polarised,
I'm assuming he's on Belmont.
Anyone else hear of this? Bill?
Richard
Visited the son in Sheffield Wednesday & Thursday. Saw a strange fault
where the sound would suddenly go way out of lip sync on HD channels.
Sometimes many seconds. I saw it on his new Sharp Freeview HD telly,
but he says it happens on another one in the bedroom and others have
mentioned seeing it to.
It can happen on any channel, we were watching both BBC 1 HD and BBC HD.
At first it seemed to coincide with opt breaks, but also happened on
BBC1 HD going from afternoon TV to CBBC. He doesn't think it happens on
SD channels, but he makes a point of watching HD mainly!
It can't be the TVs, and I noticed a thread (3rd June, 20011;
uk.tech.broadcast - Freeview HD lipsync) some while back where Glyn said
"In my experience anything that processes the MPEG transport stream has
the ability to screw-up the timing information carried in the service
Program Clock References (PCRs), component Presentation Time Stamps
(PTSs) & even possibly the video Decode Time Stamps (DTSs). At a Tx
station this could include the equipment that inserts the separately
delivered SI data or even just the conversion from 188 to 204 byte
packets to insert the Reed-Solomon error correction. Both of these
examples can involve shuffling the PID ordering around and adding delays
that will involve re-stamping the timing signals."
So as he gets Hull on BBC1 SD, and his aerial is horizontally polarised,
I'm assuming he's on Belmont.
Anyone else hear of this? Bill?
Richard
Bill Wright
Nov 11, 2011, 6:27:47 PM11/11/11
to
Dickie Mint wrote:
> So as he gets Hull on BBC1 SD, and his aerial is horizontally polarised,
> I'm assuming he's on Belmont.
No.
> So as he gets Hull on BBC1 SD, and his aerial is horizontally polarised,
> I'm assuming he's on Belmont.
Bill
Bill Wright
Nov 11, 2011, 8:49:19 PM11/11/11
to
Bill
Brian Gaff
Nov 12, 2011, 12:03:57 AM11/12/11
to
Dunno about hd, but a friend of mine said it seems to happen a lot on five
US recently, but ehen we all know how quality conscious they are over
there....
Brian
--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email: bri...@blueyonder.co.uk
______________________________________________________________________________________________________________
"Bill Wright" <bi...@invalid.com> wrote in message
news:j9kjb7$tvv$1...@speranza.aioe.org...
US recently, but ehen we all know how quality conscious they are over
there....
Brian
--
Brian Gaff....Note, this account does not accept Bcc: email.
graphics are great, but the blind can't hear them
Email: bri...@blueyonder.co.uk
______________________________________________________________________________________________________________
"Bill Wright" <bi...@invalid.com> wrote in message
news:j9kjb7$tvv$1...@speranza.aioe.org...
Steve Thackery
Nov 12, 2011, 4:46:33 AM11/12/11
to
Ah, I was about to ask the same thing.
Lip sync seems fine on my telly EXCEPT for "The Jury" on ITV1 HD. I
can't be sure, but I think the sound is leading the picture.
--
SteveT
Lip sync seems fine on my telly EXCEPT for "The Jury" on ITV1 HD. I
can't be sure, but I think the sound is leading the picture.
--
SteveT
Steve Thackery
Nov 12, 2011, 4:59:57 AM11/12/11
to
Steve Thackery wrote:
> Lip sync seems fine on my telly EXCEPT for "The Jury" on ITV1 HD. I
> can't be sure, but I think the sound is leading the picture.
Sorry, I mean I can't be sure from memory. When watching, it's very
> Lip sync seems fine on my telly EXCEPT for "The Jury" on ITV1 HD. I
> can't be sure, but I think the sound is leading the picture.
obvious.
--
SteveT
Dickie Mint
Nov 12, 2011, 6:45:42 AM11/12/11
to
Richard
Dickie Mint
Nov 12, 2011, 6:46:44 AM11/12/11
to
behind the picture.
Richard
Scott
Nov 12, 2011, 7:53:28 AM11/12/11
to
the TV from the wall for a few minutes and starting again? This seems
to be the recommended solution for all computer problems. I had sound
out of synchronisation (on SD, before HD) and this was cured by
starting again.
Ian
Nov 12, 2011, 8:20:00 AM11/12/11
to
In message <muqsb7hjs86u43uel...@4ax.com>, Scott
<newsg...@gefion.myzen.co.uk> writes
I've found in the past that changing channel and back again would cure
it.
--
Ian
<newsg...@gefion.myzen.co.uk> writes
it.
--
Ian
Geo
Nov 12, 2011, 1:53:15 PM11/12/11
to
On Sat, 12 Nov 2011 13:20:00 +0000, Ian <ne...@nospam.demon.co.uk>
wrote:
>I've found in the past that changing channel and back again would cure
>it.
Your method works fine for me - I just don't bother with the second
bit.
wrote:
>I've found in the past that changing channel and back again would cure
>it.
bit.
Dickie Mint
Nov 12, 2011, 6:13:38 PM11/12/11
to
Both tried and the sound soon goes out again.
Richard
Woody
Nov 13, 2011, 5:23:49 AM11/13/11
to
"Dickie Mint" <richard_ta...@trapyahoo.co.uk> wrote in
message news:9i8ct4...@mid.individual.net...
IME loss of sound sync is because the decoding is having to work
too hard getting what you want to watch out of the mux and error
correcting it.
What we used to see as only a very slight ghost on analogue and
which didn't bother us in the slightest can cause havoc with a
DTV signal. Ergo it raises the question of whether you have the
best possible signal quality? If you do have a very good signal
level it might be worth trying a bit (nay, even a lot) of
attenuation which could remove any phasing errors on the signal
and make it easier for the processing.
--
Woody
harrogate three at ntlworld dot com
message news:9i8ct4...@mid.individual.net...
too hard getting what you want to watch out of the mux and error
correcting it.
What we used to see as only a very slight ghost on analogue and
which didn't bother us in the slightest can cause havoc with a
DTV signal. Ergo it raises the question of whether you have the
best possible signal quality? If you do have a very good signal
level it might be worth trying a bit (nay, even a lot) of
attenuation which could remove any phasing errors on the signal
and make it easier for the processing.
--
Woody
harrogate three at ntlworld dot com
Steve Thackery
Nov 13, 2011, 7:02:13 AM11/13/11
to
Woody wrote:
> IME loss of sound sync is because the decoding is having to work too hard
> getting what you want to watch out of the mux and error correcting it.
>
> What we used to see as only a very slight ghost on analogue and which
> didn't bother us in the slightest can cause havoc with a DTV signal.
But is it likely that only the sound would be affected by phasing
> IME loss of sound sync is because the decoding is having to work too hard
> getting what you want to watch out of the mux and error correcting it.
>
> What we used to see as only a very slight ghost on analogue and which
> didn't bother us in the slightest can cause havoc with a DTV signal.
errors on the signal?
I don't know much about DVB, I must admit.
--
SteveT
Bill Wright
Nov 13, 2011, 8:15:35 AM11/13/11
to
Woody wrote:
> IME loss of sound sync is because the decoding is having to work
> too hard getting what you want to watch out of the mux and error
> correcting it.
>
> What we used to see as only a very slight ghost on analogue and
> which didn't bother us in the slightest can cause havoc with a
> DTV signal.
DVB-T is designed for high immunity to reflections. Many times I have
> IME loss of sound sync is because the decoding is having to work
> too hard getting what you want to watch out of the mux and error
> correcting it.
>
> What we used to see as only a very slight ghost on analogue and
> which didn't bother us in the slightest can cause havoc with a
> DTV signal.
obtained good DTT from an aerial that had unacceptable analogue
ghosting. This led to us having a number of communal systems that used,
say, Emley for DTT and Belmont for analogue.
>Ergo it raises the question of whether you have the
> best possible signal quality? If you do have a very good signal
> level it might be worth trying a bit (nay, even a lot) of
> attenuation which could remove any phasing errors on the signal
> and make it easier for the processing.
reduce phasing errors?
Bill
Woody
Nov 13, 2011, 11:57:45 AM11/13/11
to
"Bill Wright" <bi...@invalid.com> wrote in message
news:j9oftl$l8$1...@speranza.aioe.org...
What I meant is that if there is an incoming ghost then
attenuating the incoming signal may reduce the ghost enough that
it does not upset the decoder.
As you rightly say you can get a perfectly good picture from a
relatively weak signal provided the quality is good. If the
incoming quality is suffering for any reason then attenuating the
signal - assuming there is enough present - may remove whatever
is offending and leave a better quality signal which needs less
processing. Less processing, IME, means that sound has more
tendency to stay in sync.
Dickie Mint
Nov 13, 2011, 12:46:56 PM11/13/11
to
The relevant bitstreams in DVB like video and sound have "Presentation
Time Stamp" (PTS) tables which give the decoder the precise time at
which to present the stream. If vision is working OK, then so should
sound and be effectively synchronised to it.
Richard
Ian Jackson
Nov 13, 2011, 2:59:17 PM11/13/11
to
In message <dGSvq.6605$_W7....@newsfe27.ams2>, Woody
<harro...@ntlworld.spam.com> writes
ratio between the wanted and the interfering (the ghost) which is the
important (ie not the absolute level of the ghost)?
>
>
--
Ian
<harro...@ntlworld.spam.com> writes
>
>
>What I meant is that if there is an incoming ghost then
>attenuating the incoming signal may reduce the ghost enough that
>it does not upset the decoder.
>
As the signal will be AGCed somewhere in the STB works, surely it is the
>
>What I meant is that if there is an incoming ghost then
>attenuating the incoming signal may reduce the ghost enough that
>it does not upset the decoder.
>
ratio between the wanted and the interfering (the ghost) which is the
important (ie not the absolute level of the ghost)?
>
>
--
Ian
Scott
Nov 13, 2011, 4:42:38 PM11/13/11
to
sometimes out of sync? Are you saying it has to be a problem with the
broadcast not the receiver?
Glyn Morgan
Nov 13, 2011, 5:58:00 PM11/13/11
to
"Scott" wrote in message news:cae0c71aggq6mhv8d...@4ax.com...
anywhere along the broadcast chain. My experience is that the coding and
multiplexing are now fairly reliable and a lot of effort has been put in to
understand and check all parts of the play-out and transmission chain.
However it can certainly go wrong and I understand that my old post that
Richard pointed to earlier turned out to be just such an occurrence.
However if the problem appears to go away with a power cycle, or just a
channel reselection of the receiver, then I believe that this probably
indicates some sort of issue in the receiver although it might, in rare
cases, be the receiver performance being challenged by some poor
configuration at the broadcaster.
The synchronisation of different service components is quite complex. It
takes a lot longer to code or decode the video than, say, either the audio
and subtitles components. The coder should be configured to make all the
components available to the receiver just before they are needed but it is
not all that easy because some frames of the video are sent out of order.
These are bidirectional frames that reference earlier or later frames for
improved codec performance. This implies that the decoder has to perform
some complex buffering to store component elements until they need to be
output.
As Richard has indicated the synchronisation is handled with some time
stamping. Samples of a 33bit counter running at 90khz (the 27MHz video
program clock divided by 300), called the Program Clock Reference, or PCR,
are embedded in the digital stream and this is used to synchronise a clock
in the coder and decoders. Values of this clock, set a little in the future
to allow for transmission and decode delays, are sent with each component to
say when the decoder should output that component. These are the
Presentation Time Stamps, or PTS. For the video there is also a Decoding
Time Stamp, or DTS, which tell the decoder when it needs to decode and store
a frame to be used as reference for an out-of-sequence but earlier presented
frame.
Normally this works very well but it is relying on the receiver to maintain
synchronisation of its buffers. I'm no expert in the design of receivers
but it seems that something, possibly impulsive interference, can interrupt
the stream enough for the receiver to loose track of its buffer
synchronisation without forcing it to fully reacquire the signal and relock
its timing.
There are other scenarios. For example it is possible that the broadcaster
has not configured his coders correctly and, say, the audio is being coded
early without any compensation for the much longer video coding time, then
although the audio has the correct time-stamps it might require more
buffering than is available in the decoder. The decoder then has no way of
maintaining synchronisation and from experience will just fall back to
output the components as soon as they are available.
These problems are always exceedingly difficult to diagnose and fix and
usually involve getting knowledgeable people to test the entire end-to-end
chain which of course might cross a few contract boundaries.
Apologies for such a long explanation, which I hope makes sense. I also
ought to point out that I have not been involved in this for over three
years since I retired in 2008 :).
Glyn
>So if the system is perfect how come the sound and vision are
>sometimes out of sync? Are you saying it has to be a problem with the
>broadcast not the receiver?
I think no system is ever perfect and lip sync errors can creep in almost
>sometimes out of sync? Are you saying it has to be a problem with the
>broadcast not the receiver?
anywhere along the broadcast chain. My experience is that the coding and
multiplexing are now fairly reliable and a lot of effort has been put in to
understand and check all parts of the play-out and transmission chain.
However it can certainly go wrong and I understand that my old post that
Richard pointed to earlier turned out to be just such an occurrence.
However if the problem appears to go away with a power cycle, or just a
channel reselection of the receiver, then I believe that this probably
indicates some sort of issue in the receiver although it might, in rare
cases, be the receiver performance being challenged by some poor
configuration at the broadcaster.
The synchronisation of different service components is quite complex. It
takes a lot longer to code or decode the video than, say, either the audio
and subtitles components. The coder should be configured to make all the
components available to the receiver just before they are needed but it is
not all that easy because some frames of the video are sent out of order.
These are bidirectional frames that reference earlier or later frames for
improved codec performance. This implies that the decoder has to perform
some complex buffering to store component elements until they need to be
output.
As Richard has indicated the synchronisation is handled with some time
stamping. Samples of a 33bit counter running at 90khz (the 27MHz video
program clock divided by 300), called the Program Clock Reference, or PCR,
are embedded in the digital stream and this is used to synchronise a clock
in the coder and decoders. Values of this clock, set a little in the future
to allow for transmission and decode delays, are sent with each component to
say when the decoder should output that component. These are the
Presentation Time Stamps, or PTS. For the video there is also a Decoding
Time Stamp, or DTS, which tell the decoder when it needs to decode and store
a frame to be used as reference for an out-of-sequence but earlier presented
frame.
Normally this works very well but it is relying on the receiver to maintain
synchronisation of its buffers. I'm no expert in the design of receivers
but it seems that something, possibly impulsive interference, can interrupt
the stream enough for the receiver to loose track of its buffer
synchronisation without forcing it to fully reacquire the signal and relock
its timing.
There are other scenarios. For example it is possible that the broadcaster
has not configured his coders correctly and, say, the audio is being coded
early without any compensation for the much longer video coding time, then
although the audio has the correct time-stamps it might require more
buffering than is available in the decoder. The decoder then has no way of
maintaining synchronisation and from experience will just fall back to
output the components as soon as they are available.
These problems are always exceedingly difficult to diagnose and fix and
usually involve getting knowledgeable people to test the entire end-to-end
chain which of course might cross a few contract boundaries.
Apologies for such a long explanation, which I hope makes sense. I also
ought to point out that I have not been involved in this for over three
years since I retired in 2008 :).
Glyn
Ian Field
Nov 13, 2011, 6:08:21 PM11/13/11
to
news:9i524m...@mid.individual.net...
Americans who move here eventually get used to 50Hz flicker.
> Not seen any mention in these hallowed walls.
>
>
> Visited the son in Sheffield Wednesday & Thursday. Saw a strange fault
> where the sound would suddenly go way out of lip sync on HD channels.
> Sometimes many seconds. I saw it on his new Sharp Freeview HD telly, but
> he says it happens on another one in the bedroom and others have mentioned
> seeing it to.
Its fairly normal for freeview - we'll eventually get used to it just like
>
>
> Visited the son in Sheffield Wednesday & Thursday. Saw a strange fault
> where the sound would suddenly go way out of lip sync on HD channels.
> Sometimes many seconds. I saw it on his new Sharp Freeview HD telly, but
> he says it happens on another one in the bedroom and others have mentioned
> seeing it to.
Americans who move here eventually get used to 50Hz flicker.
Bill Wright
Nov 13, 2011, 8:41:46 PM11/13/11
to
Woody wrote:
> Sorry Bill, wrong use of expression.
>
> What I meant is that if there is an incoming ghost then
> attenuating the incoming signal may reduce the ghost enough that
> it does not upset the decoder.
The attenuator will reduce the wanted and unwanted signals by the same
> Sorry Bill, wrong use of expression.
>
> What I meant is that if there is an incoming ghost then
> attenuating the incoming signal may reduce the ghost enough that
> it does not upset the decoder.
amount, so the ratio between the two will stay the same.
> As you rightly say you can get a perfectly good picture from a
> relatively weak signal provided the quality is good. If the
> incoming quality is suffering for any reason then attenuating the
> signal - assuming there is enough present - may remove whatever
> is offending and leave a better quality signal which needs less
> processing.
Unwanted signals do not 'disappear below the noise' when attenuated;
they merely add to the noise.
If what you said was right we would have been able to remove ghosting by
the use of attenuators. That was not the case. We would also be able to
improve the BER of DVB-T when there is co-channel interference by the
use of attenuators. Again, that is not the case.
Seriously, and I don't want to be rude, but are you having a laugh?
Bill
Dickie Mint
Nov 14, 2011, 12:13:31 PM11/14/11
to
On 13/11/2011 22:58, Glyn Morgan wrote:
> snip ...
>Apologies for such a long explanation, which I hope makes sense. I also
> ought to point out that I have not been involved in this for over three
> years since I retired in 2008 :).
>
> Glyn
Thanks, Glyn, nice to hear from you!!
I've not only been retired for over 6 years now, but I never grasped the
more complex bits in the first place! And thank you for putting up with
me at the time.
PL is looking into it, as several people have reported it on different
transmitters, and with different sets, in case it is maybe a source problem.
Richard
> snip ...
>Apologies for such a long explanation, which I hope makes sense. I also
> ought to point out that I have not been involved in this for over three
> years since I retired in 2008 :).
>
> Glyn
I've not only been retired for over 6 years now, but I never grasped the
more complex bits in the first place! And thank you for putting up with
me at the time.
PL is looking into it, as several people have reported it on different
transmitters, and with different sets, in case it is maybe a source problem.
Richard
0 new messages