Jump to content


Photo

Freeview Vs. Dvb


  • Please log in to reply
143 replies to this topic

#51 Aurora8

Aurora8

    AV Forums Member

  • Member
  • PipPipPipPip
  • 711 posts

Posted 28 June 2011 - 03:01 PM

So I guess Freeview aren't updating the program information at the same time as the program titles on their EPG in timeslots where there have been changes made. The examples you've outlined with Drew Carey/Jim and Mother/NOF are both cases where amendments have been made from one program to the other.

#52 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 05 July 2011 - 12:33 PM

WOW! Pete said 'freeview' was working hard and that they sure are. At screwing up. Good God. Guys, is it really this hard to provide a stable useful functional 'freeview EPG'.

<nope, not going to point it out, more fun to see if the local Freeview rep can>

#53 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 06 July 2011 - 09:50 AM

In what has to be possibly the most telling occurence to date ... Freeview has removed the 'EPG feedback' links in their "Questions about Freeview" web page. Previously there was a link that one could use to submit issues about the Freeview EPG in the "Experiencing Delays" and "Why Can't I See" topics but it seems that even Freeview no longer wants to hear about it.

"Every day, in every way, I'm getting better and better". :lol:

#54 pgdownload

pgdownload

    AV Forums Member

  • Senior Member
  • 11,466 posts

Posted 06 July 2011 - 11:50 AM

Previously there was a link that one could use to submit issues about the Freeview EPG in the "Experiencing Delays" and "Why Can't I See" topics but it seems that even Freeview no longer wants to hear about it.

In light of this thread, I think they probably figured it redundant. :)

Regards

Peter Gillespie

#55 digitalj

digitalj

    AV Forums Member

  • Senior Member
  • 12,929 posts

Posted 06 July 2011 - 12:10 PM

In light of this thread, I think they probably figured it redundant. :)

Regards

Peter Gillespie


more like they're sick of the negativity over it.

#56 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 06 July 2011 - 12:41 PM

more like they're sick of the negativity over it.

That could be resolved by fixing it but all the only movement seems to involve an increase in the number of issues, not a decrease. You'd think that 'freeview' would want to know all about the problems, but by removing the links it appears not.

Just as well its 'free'.

#57 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 12 October 2011 - 06:42 AM

... and just to show that Freeview still couldn't tell if its arse was on fire, it seems that the repeats of new shows that are appearing in the schedule are in fact not repeats, they are totally different episodes. At least according to the freeview EPG that is.

#58 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 13 November 2011 - 07:44 AM

nate over at dvbforums has been working on a handy tool that has an unintentional feature in that it will let people easily see how accurate (or inaccurate) the information that a Freeview PVR uses to record a show is in their area. If you find the accuracy is lacking, contact the broadcaster. It is the only way it is going to get fixed this far along as clearly the networks either aren't aware of issues or don't care.

tsnowandnext

The accuracy has been less than stellar here (regional QLD). 7mate timing means that between 3 and 5 mins would be missed from either end of a show (toss a coin as to which end); a random test from GO! just now shows the kids would have missed the first 6 mins of Camp Lazlo.

Sunday on SC10. Well you didn't want to see the beginning of Junior Mastercheck anyway because its not your type of show. You really want to see SCIFI. Which is why you'd be really annoyed that the beginning of two mins of Terra Nova would have been missing.

But wait, there's always other networks to watch. 7 QLD, whoops, just as well the start of Royal Pains wasn't that important, eh?

Edited by DrP, 14 November 2011 - 05:54 AM.


#59 prl

prl

    AV Forums Member

  • Senior Member
  • 2,620 posts

Posted 16 November 2011 - 03:21 PM

Any idea why ACMA is so far behind it posting their reporting on now/next accuracy compliance (report links in right-hand sidebar)? The most recent reports are for Mar 2011.

#60 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 16 November 2011 - 05:41 PM

It probably takes so long as they have to write a work of fiction because, from my own observations, those statistics are a lot of crap. How crap? Well suppose you were a Xena fan. In regional QLD you'd be glad to know that the last 7 mins 30 seconds of the show were apparently Hercules... at least according to the EIT.

Here's the answer: 'The ACMA has ceased its formal monitoring and reporting program but will still respond to complaints regarding EPG inaccuracies. Contact details for reporting inaccuracies are available on this page.'

So having determined that, from a seemingly fictional batch of data, that there were no problems with 'N&N' accuracy, ACMA decided that it no longer had to monitor things. Well guess what ACMA? At the time you were saying its all ok, it wasn't and it still isn't!

Edited by DrP, 16 November 2011 - 06:20 PM.


#61 pgdownload

pgdownload

    AV Forums Member

  • Senior Member
  • 11,466 posts

Posted 16 November 2011 - 06:32 PM

Any idea why ACMA is so far behind it posting? The most recent reports are for Mar 2011.

The answer would be in the report :)

For these reasons, the ACMA has decided to discontinue the regular monthly and quarterly reports on performance against the EPG principles, effective from 31 March 2011. To ensure that full compliance with the EPG principles continues, and any future lapses are promptly resolved, the ACMA will continue to monitor (but not report publicly on) EPG performance by conducting random checks of broadcasters’ services and responding to consumer complaints submitted via the ‘EPG principles’ page on the ACMA website (http://www.acma.gov....RD/pc=PC_311870).


Its interesting the 2, SBS and 10 only get random unpublished checks due to their previous good results, yet the checks on the other channels amount to little more than the same (basically reads like some secretary checked her STB once every three nights for each channel).

Although perhaps most telling is that the report states that the N&N info is distinct from the EPG info (including the N&N items in the EPG). Since most PVRs / timers / people use the EPG info then its perhaps not surprising that the in house experience of EPG accuracy differs from the ACMA results...

Regards

Peter Gillespie

#62 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 16 November 2011 - 06:52 PM

ACMA's reports supposedly reflects the accuracy of the EIT 'N&N' events (current / following). This is not just monitoring EPG accuracy for general public facility, this is also for classification restriction compliance.

Simple observation of the specific part of the transmission in concern demonstrates that one would have more accurate results by throwing a dart at a board, at least in my part of the world. This is not concering the times detailed in the N&N events nor the times in the EPG in general. This is soley examining the 'N&N' events themselves to see a) what event is in the 'Now' position, what event is in the 'Next' position and what event has the running status set as 'running'. nate's tool makes it easy enough for anyone to determine the quality of the data in their region but it's not telling me anything that I wasn't already going about the quality here based on my linux efforts.

How ACMA's observation method could be so flawed as to produce data that is so demonstrably wrong is beyond me. If ACMA is continuing to use the same methodology then the random tests may as well cease as they would have no relation with reality at all.

#63 pgdownload

pgdownload

    AV Forums Member

  • Senior Member
  • 11,466 posts

Posted 17 November 2011 - 08:01 AM

Not sure why there would be a 'running' status flag? Surely whatever is "Now" is equal to "Running" and if it wasn't then force them to always be?

But it shouldn't be all that hard to replicate what ACMA is doing. Bring up the N&N info box on the TV and note when the show changes (be interesting to see how spot on it would be on those shows the networks merge into each other (eg Desperate Housewives ends and Brothers and Sisters starts without interruption). If the discrepancy is as big and consistent as DrP says then ACMA have said they'll respond to viewer complaints on the matter.

Regards

Peter Gillespie

#64 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 17 November 2011 - 09:35 AM

Ah stuff it. Its pretty clear that pgdownload doesn't grasp the finer points and I'm not going to explain it to him.

Anyone that is interested in getting things such as the so called Accurate Recording feature of Freeview working properly will perform their own tests and instead of contacting the toothless and blind dog that is ACMA they will contact the local offices of the broadcaster involved.

Edited by DrP, 17 November 2011 - 10:06 AM.


#65 miata

miata

    AV Forums Member

  • Member
  • PipPipPipPipPip
  • 1,633 posts

Posted 18 November 2011 - 06:32 AM

EPG fpr GEM and NBN GEM shows TODAY SHOW from 0530 - 0900. The show isn't on either channel. FOXTEL Guide shows Arabian Adventure, so clearly the HD broadcast of TODAY has ended. Nice to know that FOXTEL is more accurate than the FTA EPG!

#66 pgdownload

pgdownload

    AV Forums Member

  • Senior Member
  • 11,466 posts

Posted 18 November 2011 - 09:35 AM

EPG fpr GEM and NBN GEM shows TODAY SHOW from 0530 - 0900. The show isn't on either channel. FOXTEL Guide shows Arabian Adventure, so clearly the HD broadcast of TODAY has ended. Nice to know that FOXTEL is more accurate than the FTA EPG!

The networks / ACMA don't have to provide an accurate EPG, they have to provide an accurate N&N. It would be interesting to see what the N&N indicated for the Today Show. Although it would seem likely that some, many? STB might just use the EPG entries in displaying their N&N information anyway so its not a great test.

Regards

Peter Gillespie

#67 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 23 November 2011 - 04:48 AM

The last 8 minutes 30 seconds of The Equalizer? 7Maaaaate, you didn't want to see that anyway!

Dad's Army? Who do you think you are kidding Mr Freeview if you think this guide is right.

Edited by DrP, 23 November 2011 - 05:00 AM.


#68 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 24 November 2011 - 05:12 AM

CRIDs. HAH!

Depending on the network, channel and some random factor the 'show' component of the CRIDs indicates:

a) a season of a show - Adam-12 (7mate)
B) an instance of show during the week where there is a repeat or different scheduling of the show - The Killian Curse (ABC2)
c) nothing in particular as it changes for no apparent reason - American Horror Story (ELEVEN)

Using the show text title is a more reliable way of recording something and we all know the troubles of that method.

#69 Puss in Books

Puss in Books

    AV Forums Member

  • Member
  • PipPipPip
  • 371 posts

Posted 25 November 2011 - 08:02 AM

...
Using the show text title is a more reliable way of recording something and we all know the troubles of that method.

What troubles are you refering to? The obvious ones I can think of are not being able to pick repeats and the occassional name mangling (eg. "new episode", "includes crimestoppers"). I haven't figured out a way around the former, but the latter is easily solved (using a topfield and taps).

#70 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 25 November 2011 - 09:11 AM

That's it - consistent program naming. The whole concept of CRIDs is to give a consistent ID to the content so that machines don't have to deal with the vagaries of us fleshy beings.

#71 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 29 November 2011 - 05:58 PM

WHEEEEEEEEEEEEE! American Horror Story has yet another CRID. I guess these CRID thingys are going out of style and the networks have a large store of them that they need to use up before they are no longer valid. :lol:

#72 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 10 March 2012 - 09:16 AM

How many CRIDs does it take to describe a program with, so far, a single season? I'm not sure, but Ringer looks like the front runner so far with 5 since about December 2011.

Is this a result of mismatches due to Ringer being dropped into another time slot thereby temporarily being tagged with another show's CRID in error? Nope. The CRIDs Ringer has been assigned so far have been unique (not only within the carrying network which is where it matters, but also other networks).

#73 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 19 March 2012 - 05:34 AM

Yet another CRID for Ringer.

#74 digitalj

digitalj

    AV Forums Member

  • Senior Member
  • 12,929 posts

Posted 23 March 2012 - 05:26 PM

http://www.smh.com.a...0314-1uyzq.html

#75 DrP

DrP

    AV Forums Member

  • Senior Member
  • 19,883 posts

Posted 23 March 2012 - 05:32 PM

HAH! The EIT N&N doesn't necessarily tick anywhere near the minute accurate point. The CRIDs can change, seemingly at random, as per my previous posts.

The CRIDs are carried in the N&N information and a freeview PVR appears to use this information to determine when a program ends. Since the Australian freeview spec is not public the method has to be determined by educated guessing. The ETSI standard provides several possible methods. However examining the full broadcast transport stream the CRID only appears in the freeview daily EPG data files and the EIT N&N (depending on the network CRIDs can also appear in the general EIT EPG data). There does not appear to be any trigger in the freeview data to say a program is running 'now' and even if there was the file transmission time is too long, the duration of transmission of a file is variable, the data itself is compressed requiring the freeview applet to unpack it using the stored huffman table leaving the N&N data as the only apparent source of recording triggering.

Any PVR that snoops N&N looking for a pattern match in the program name will have equally as useful results when it comes to a program that has 'run over time' as a freeview PVR monitoring the CRIDs in this respect. Arguably, at least in my part of the world, given the flex-o change-o CRIDs a PVR that ignores the CRIDs instead using pattern matching against the program name will have more success in recording the desired program!

Edited by DrP, 23 March 2012 - 06:22 PM.