Random Recordings!

[Black Hole]

This morning I've caught my radio-dedicated 'Fox converting a recording I have no knowledge of to MP3. What's more, it is scheduled to make a recording on Monday 17th April with a different name, and it recorded yesterday (same name)!

The programmes in question are on BBC Radio 4 Extra:

Yesterday and today: "Wild Honey" by Chekhov (in two parts)
Next week: "Give Dotty a Chance" by Martyn Hesford

I presume this is a S-CRID thing, resulting from me scheduling a dramatisation of Live And Let Die last week, and being offered to "record series". I thought that might mean there were other 007s being offered, but it seems they've simply re-used the S-CRID (nothing remains in the schedule corresponding with Live And Let Die, not even in the dormant series links, so that's a smoking gun!).

Give Dotty A Chance has an S-CRID FP.BBC.CO.UK/B_2C31EN, but I can't figure out how to get the CRIDs for existing recordings.

 

[Luke]

It is normal for the BBC Radio dramas to have what appears (to me) to be random linkages to each other, but "Give Dotty a Chance" and "Wild Honey" do have different S-CRIDs for their entries yesterday and in the next 7 days, but you appear to be saying that they are linked.

Theoretically it is possible for the freeview epg to simultaneous assign two different S-CRIDs to each programme but I've never noticed the BBC doing that when it has been appropriate to do so.

 

[Black Hole]

you appear to be saying that they are linked

What other options are there?

Give Dotty A Chance appears as a pending reservation, so it can't be linked with any other existing entry in the pending list or the dormant list, and it isn't a non-series reservation, whereas Live And Let Die appears in my recordings but not in the current or dormant schedule list.

"Give Dotty a Chance" and "Wild Honey" do have different S-CRIDs for their entries yesterday and in the next 7 days

How do you find that out? Does that apply to the multiple same-day re-broadcasts?

 

[Luke]

How do you find that out?

Yesterday's Wild Honey was still in my epg when I looked earlier. It has cleared itself now, so I turned on my second active box and immediately had a look.
E.g.

How do you find that out? Does that apply to the multiple same-day re-broadcasts?

I looked at all the "Wild Honey" and "Give Dotty a Chance" epg entries covering yesterday until next Tuesday.

 

[EEPhil]

Does that apply to the multiple same-day re-broadcasts?

I've had the case where something is on 3 or 4 times in the day and it's on series link. Usually the earliest is recorded and the rest ignored. When I've listened (and deleted) the recording before the next repeat, the damned thing records again! Occasionally I've found a programme after the first broadcast and set it to record/series link. If I record today's at noon, the next episode may be picked up at 6am or even on a different station ( R4 or R4X). I'd like to pick up all episodes on the same channel at the same time but can't do it on series link.
And don't get me started on problems I've had with some of the drama strand on R4 (and maybe 4X). Even though I've series linked to a particular n-part drama it wants to record everything in that slot. On the 5000T the virtual folder for the series gets renamed to the name of the last added item - so finding what you're looking for is challenging!
The only solution I've found is to abandon series link. But that's messy as nothing gets nicely stored in folders. (And whereas you can create folders in a 2000T and move stuff into it [same for Fox?], no such possibility with the 5000T).

 

[gomezz]

Maybe they have had a new starter on the job? On Sunday I noticed the omnibus version of The Archers had been added to the daily episode podcast feed instead of the omnibus feed. And then the Sunday daily episode was the omnibus version until they noticed it and put up the right recording.

 

[Black Hole]

When I've listened (and deleted) the recording before the next repeat, the damned thing records again!

I guess that means they used the same series ID but a different programme ID (stupid!).

It would be useful if we had a way to extract the CRID from a recording (surely it must be in the non-programme data somewhere?), but as WebIF doesn't show it I guess we don't have the means. hmt offers nothing, unless "Event ID" is anything to do with it.

 

[Luke]

"Event ID" is anything to do with it.

Nope. Event ID is unique for a particular broadcast slot. So unique that have there is a 15 minute delay and the main epg is updated, when a programme is rescheduled by 15 minutes in the epg it will be allocated a new Event ID, despite having the same CRIDs.

 

[EEPhil]

It would be useful if we had a way to extract the CRID from a recording (surely it must be in the non-programme data somewhere?), but as WebIF doesn't show it I guess we don't have the means. hmt offers nothing, unless "Event ID" is anything to do with it.

It would be useful to have a ready access to CRIDs at all. With the 5000T the hidden menu's display CRID isn't sticky. Not sure if it vanishes on standby, but my device gives the impression of crashing in the middle of the night and the displayed CRID feature vanishes. Can't even access the database like you can on the 2000T.

You'd think a CRID for the existing recording might be stored in the recording, but for what purpose? They must be in the data stream somewhere. How else would the CRIDs find the way to the database? Probably stripped out by the Humax. Have to see if I can capture a short period of a whole multiplex and see if there are any clues

 

[prpr]

It would be useful if we had a way to extract the CRID from a recording (surely it must be in the non-programme data somewhere?)

It is, in the EIT data, so unless you have shrunk the .TS file it will be there. The tricky thing is getting at it as it's mixed in with loads of other stuff. You have to write a custom decoder effectively (and deal with the encryption if it's an HD service). I'm a fair way down the track with all this (but not on a T2) as you may have seen from snippets posted elsewhere e.g.

11 18:00:34 Event 17540:07969 4E:01 started 18:00:00 00:30:00 "The One Show" /m/VA5T /b/4ARMT
11 18:30:21 Event 17540:07970 4E:02 started 18:30:00 00:30:00 "EastEnders" /m/VA5Z /b/3N2E9
11 19:00:41 Event 17540:07971 4E:03 started 19:00:00 01:00:00 "Interior Design Masters with..." /m/VA65 /m/UKUV

but don't yet have a tool for extracting from a .ts file.

but as WebIF doesn't show it I guess we don't have the means.

True.

 

[MymsMan]

but don't yet have a tool for extracting from a .ts file.

Another entry for the over long wish list!

It would be nice to have something that extracted the identifying information from the EIT and stored it in some of the unused space in the .hmt for display by the webif, the tool could be run as part of the auto processing at recording end.

 

[Black Hole]

Is this any use?: https://mkrufky.github.io/libdvbtee/

 

[antipodean]

I've had the case where something is on 3 or 4 times in the day and it's on series link. Usually the earliest is recorded and the rest ignored. When I've listened (and deleted) the recording before the next repeat, the damned thing records again! Occasionally I've found a programme after the first broadcast and set it to record/series link. If I record today's at noon, the next episode may be picked up at 6am or even on a different station ( R4 or R4X). I'd like to pick up all episodes on the same channel at the same time but can't do it on series link.

I get frustrated with this too, especially when I am trying to set the programmes up so they don't clash/overlap but it insists on recording the first available instance of an episode, sometimes resulting in the pink bars appearing and repeated emails about impending recording clashes.

 

[EEPhil]

sometimes resulting in the pink bars appearing and repeated emails about impending recording clashes.

Eh? I take it this is a CF feature. Something not available on my kit. The original post was in the Freeview section, so forgive my confusion. Something definitely changed a while back. I used to be able to series link a particular time, now it's the earliest available. I wonder who (at the BBC, seeing that it's their programmes this usually applies to) thought this was a good idea. Or is it part of the standards?

 

[Black Hole]

I used to be able to series link a particular time, now it's the earliest available.

That relies on different transmission slots being given different S-CRIDs. If they all have the same S-CRID, then naturally the same episode has the same P-CRID and the recorder will grab the first one which comes along (and then try again if that one failed for some reason). It doesn't record multiple copies of the same episode if the P-CRID matches a P-CRID lodged in the "recently recorded" database.

 

[prpr]

It would be useful if we had a way to extract the CRID from a recording

Tool now working well enough for a proper test. If you can do this on the decrypted (but not shrunk) .ts file:
dvbsnoop -if input.ts -s ts -m2ts -n 10000 -b 18 >output.ts
obviously replacing input.ts and output.ts with appropriate names, and tell me the service ID and event ID from the hmt utility for the input.hmt file, then upload it to my Dropbox (will PM you a link), I will decode it and return you the CRIDs.
Here's an example from a short recording (renamed) I did the other day:

humax# hmt Junk_20230411_1925.hmt
Format:HD
Title:EastEnders
ITitle:The One Show
Channel:101 (BBC ONE HD)
Episode:0,0/0
Folder:/mnt/hd2/My Video/
Filename:Junk_20230411_1925
Genre:Entertainment (48)
EPG:Alex Jones and Emma Willis are joined by singers Jonas Brothers and Golda Rosheuvel and India Amarteifio, stars of Queen Charlotte: A Bridgerton Story. [S] [HD]

Recording Status: Valid (OK)
Flags: HD,New,Unlimited Copies,
Copy count:0

Scheduled start:1681236000 (Tue Apr 11 19:00:00 2023)
Scheduled duration:1800
Recording start:1681237511 (Tue Apr 11 19:25:11 2023)
Recording end:1681237630 (Tue Apr 11 19:27:10 2023)
Duration:119
Stored duration:119
Play resumes at:0 seconds in.
Timezone offset:3600

Service ID (SID):17540
Event ID:7969
Transport Stream ID (TSID):16514
Originating Network ID (ONID):9018
Programme Map Table PID (PMTPID):6600
Video PID:6601
Audio PID:6602
Bookmarks:0 =

generates this:

14 11:46:11 Event 17540:07969 4E:01 2023/04/11 18:00:00 00:30:00 "/m/VA5T" "/b/4ARMT" "The One Show"
14 11:46:11 Event 17540:07970 4E:01 2023/04/11 18:30:00 00:30:00 "/m/VA5Z" "/b/3N2E9" "EastEnders"

as Now'n'Next info.

Aside: Why does the crappy Humax software mis-name things inconsistently (title, ititle, filename) when you do an instant-record close to the end of a programme?

 

[antipodean]

Eh? I take it this is a CF feature. Something not available on my kit.

Sorry, yes this is a feature on the HDR-Fox-T2 CF, where conflicting recordings can be identified on the RS and WebIF.

 

[EEPhil]

Aside: Why does the crappy Humax software mis-name things inconsistently (title, ititle, filename) when you do an instant-record close to the end of a programme?

I don't know. Why does...? Oh, it wasn't a joke!
I quite often get the wrong Now/Next in my 2000T's hmt files. That's probably my fault for using padding rather than AR. Sometimes that translates to the previous programme being part of the filename of the one I want. When I instant record it often picks up the name of the next programme. Obviously the software wasn't tested on real users and their bad habits.

 

[MymsMan]

Aside: Why does the crappy Humax software mis-name things inconsistently (title, ititle, filename) when you do an instant-record close to the end of a programme?

Probably designed for use with padding so that programmes don't get the previous programmes name,
instant record near end should probably default to recording to end of next programme.

 

[Black Hole]

Tool now working well enough for a proper test. If you can do this on the decrypted (but not shrunk) .ts file:
dvbsnoop -if input.ts -s ts -m2ts -n 10000 -b 18 >output.ts
obviously replacing input.ts and output.ts with appropriate names, and tell me the service ID and event ID from the hmt utility for the input.hmt file, then upload it to my Dropbox (will PM you a link), I will decode it and return you the CRIDs.

I'm not in a position to do any of that until next week.