PMT Pid wasn't found on the PAT (2 Viewers)

[doveman]

I understand how this is a problem, I just wonder how Freeview STB manage it (or were you not referring to Freeview and in fact Freeview channels all do broadcast all the time, although I don't think this is the case)?

Freeview STBs cope with this by only having to cope with Freeview. They do not need to worry about DVB-S, Sky, cable providers in The Netherlands, ATSC providers in America .... which probably all have their own way of doing this (this is the main thing that gives things Freeview / Freesat certification)

It would be simple enough to make MP deal with Freeview as a STB does but this would then cause issues with other providers

So you're saying that Freeview channels all do broadcast all the time, so when doing a scan on a Freeview STB there's never an issue of a channel not being detected because it's not currently broadcasting?

 

[jameson_uk]

So you're saying that Freeview channels all do broadcast all the time, so when doing a scan on a Freeview STB there's never an issue of a channel not being detected because it's not currently broadcasting?

Not directly (although this is the case)

What I am saying is that a Freeview STB can use specific Freeview logic to identify channels (which IIRC contains some custom IDs which are unique to Freeview). So they are coded to only deal with a single spec (if you took the STB to another DVB-T provider elsewhere in Europe it might not pickup channels correctly)

The issue with channels not broadcasting 24x7 is kind of down to how MP has been built in the past but if you trawl through the DVB specs you will see there are several different tables but a lot of options as to how these tables are populated. Freeview sets them up one way and Sky another. Due to this MP picks up things like BBC3 on Freeview when it is off air (Freeview does appear to be transmitting a channel by all accounts) but not on Freesat / Sky (where details are missing in one table when the channel is off air so it appears it does not exist if you look at this table). Now this table with missing data is normally needed to do things like pick up the channel numbers and some other important data but it is only recommended and Sky transmit this data in a completely different way.

 

[doveman]

So you're saying that Freeview channels all do broadcast all the time, so when doing a scan on a Freeview STB there's never an issue of a channel not being detected because it's not currently broadcasting?

Not directly (although this is the case)

What I am saying is that a Freeview STB can use specific Freeview logic to identify channels (which IIRC contains some custom IDs which are unique to Freeview). So they are coded to only deal with a single spec (if you took the STB to another DVB-T provider elsewhere in Europe it might not pickup channels correctly)

Well I asked elliottmc how STB cope with channels not broadcasting all the time and you chipped in with "Freeview STB's cope with this by only having to cope with Freeview", which wasn't particularly helpful. I don't know why you feel the need to take every opportunity to bang on about how complicated it is making MP work with all the different providers, rather than try and be constructive and think of ways to make things easier for users, like elliotmc has done.

Then take it to the next level and what you have (if as many UK users do) combined Freeview and Freesat channels. If a channel stops broadcasting on Freeview but continues on Satellite (as CNN did recently) then MP can not delete the channel after scanning DVB-T as the same channel is shared across DVB-T and DVB-S.

How have you identified that "many" UK users have both Freeview and Freesat? I would have thought there'd be many more users with just Freeview sticks/cards than those who've also installed a dish and bought a Freesat card. If a channel stops broadcasting on Freeview but MP doesn't delete it as it's still on Satellite, doesn't this cause problems when MP tries to play/record the Freeview channel?

Quite simply as discussed many times before, there will be no Freeview tick box. Freeview is probably one of 150+ different providers and if you consider how many people have multiple providers the number of combinations is pretty large.

So who's made this decision that MP is never going to be made more user-friendly for Freeview users? As I say, I imagine most UK PC TV tuner owners just have Freeview sticks/cards and if they're introduced to MP as it currently is, I think they'd be distinctly unimpressed with how complicated it is to manage channels at the moment.

If the scanning/channel management is purely done in TV Server config, perhaps the best way to encourage more UK users to use MP would be to make a Freeview-only version of that, so that channel management works more like a STB. Ideally this could be done from MP, but I understand that won't be possible before MP2, so perhaps a separate, channel management app (in both Freeview-only and multiple provider versions, with a bigger, clearer interface than TV Server config, that could maybe be operated with just a remote) could be made to make it easier for users to do channel scanning/management on their TV screens.

Also, I don't know the details/differences but whilst Freeview is UK-specific, I know that other countries use DVB-T, so if a version of TV Server config, or a separate channel management app, could be made that worked well for channel management for all DVB-T users, rather than just Freeview, obviously that extends the amount of users who'd benefit from it.

 

[jameson_uk]

Well I asked elliottmc how STB cope with channels not broadcasting all the time and you chipped in with "Freeview STB's cope with this by only having to cope with Freeview", which wasn't particularly helpful. I don't know why you feel the need to take every opportunity to bang on about how complicated it is making MP work with all the different providers, rather than try and be constructive and think of ways to make things easier for users, like elliotmc has done.

I can give you a million and one ways in which MP could be improved and be constructive in terms of ideas but I am just trying to be realistic and set expectations.

Even what you might consider a simple tweak is going to take someone many weeks (or more likely months) to develop this.

In terms of Mantis there is already an entry for this
0002130: Give the user the opportunity to delete old channels after a complete scan - MediaPortal Bugtracker
which was submitted in April 2009 and because of the complexity of this is still outstanding

I totally agree that scanning is not great and needs improving but until a dev is willing to take on this big bit of work all the ideas in the world are not going to improve things. This is why I was talking about DJBlu's work that he has put in to do exactly what you want to do but is limited to Sky. This can develop into something that is extensible and could be plugin driven so there could be a Freeview plugin etc but that is still a fair way off from being part of the core MP code.

 

[doveman]

I can give you a million and one ways in which MP could be improved and be constructive in terms of ideas but I am just trying to be realistic and set expectations.

Even what you might consider a simple tweak is going to take someone many weeks (or more likely months) to develop this.

In terms of Mantis there is already an entry for this
0002130: Give the user the opportunity to delete old channels after a complete scan - MediaPortal Bugtracker
which was submitted in April 2009 and because of the complexity of this is still outstanding

I totally agree that scanning is not great and needs improving but until a dev is willing to take on this big bit of work all the ideas in the world are not going to improve things. This is why I was talking about DJBlu's work that he has put in to do exactly what you want to do but is limited to Sky. This can develop into something that is extensible and could be plugin driven so there could be a Freeview plugin etc but that is still a fair way off from being part of the core MP code.

Well it still seems more useful to encourage ideas and discussion rather than go on about how complicated it is. Someone might have a great idea that simplifies things or appeals to a dev. You know that you've already told me about the complications with multiple providers, so there's really no need to keep doing so each time you see a post by me about scanning.

It's a shame that there's no devs available to work on this, as I think TV is the main thing that sets MP apart (I'm sure there's other benefits as well, but I haven't tried anything else for a while to compare), although NextPVR has this now (I haven't tried it though, so don't know if it's any good).

I don't know if DJBlu has any plans to develop a Freeview-plugin, but he seems to be working on a USALS plugin at the moment, and has very little time even for that.

Perhaps another possibility until something better can be done would be to use another free app that can already properly scan for Freeview channels, and then write an app to import the list of channels it generates into MP's database.

 

[mm1352000]

Hi folks

As always, the lack of developers is the limitation! Speaking as somebody who has investigated these things I can confirm that what jameson_uk has said is accurate. I get the impression that he and I approach the "problem" from a similar perspective...

To elaborate: personally I like the idea of an automatic channel management approach - ie. automatic handling of adding, renaming, deleting, moving of channels - over manual approaches like the one suggested by elliottmc. From a user perspective an automated approach seems far more "elegant". Of course there are dangers like not allowing enough flexibility for manual intervention. However on the whole I still favour that approach.

Having said that, I'm also fully aware of the complexity and potential conflicts in implementing such a system. elliottmc's suggestion is probably a far more realistic one for software like MediaPortal that is used in many corners of the world. That is not to say that both approaches couldn't co-exist - just that I would estimate the time involved in implementing the automated approach even for one provider exceeds the time that it would take to implement a reasonably streamlined manual approach that would be more than adequate for most people's needs.

If mm agrees with this approach (present user with a list of non-detected channels with the option to keep or delete) then I will mantis it.

I'm no dictator! You're more than welcome to mantis the idea. I can say now that I will not be able to implement it as my hands are totally full with other things and I don't anticipate having any room in my list for the next 6 months or even longer. Further, I'm aware of other devs working on other changes that are close enough (in terms of code not function) that would make my life very miserable if I decided to dive in an make changes there. As great as Git is turning out to be, I still favour avoiding stepping on other devs' toes when I make changes.

doveman:

Well it still seems more useful to encourage ideas and discussion rather than go on about how complicated it is. Someone might have a great idea that simplifies things or appeals to a dev. You know that you've already told me about the complications with multiple providers, so there's really no need to keep doing so each time you see a post by me about scanning.

You may or may not be aware that we discussed scanning at exhaustive length during the 1.2.x release cycle on our internal forums. To the point where I almost cringe when I think about it now. We made some changes that we thought were fairly minor and safe and in my opinion ended up delaying the release of MP 1.2.0 final by a good 2 months. The moral for me is nothing is ever as simple as it seems. There is nothing wrong with discussion but if you want to do that then please start another thread in a more general section of the forum and ask a mod to move the relevant posts.

It's a shame that there's no devs available to work on this, as I think TV is the main thing that sets MP apart (I'm sure there's other benefits as well, but I haven't tried anything else for a while to compare), although NextPVR has this now (I haven't tried it though, so don't know if it's any good).

I don't know if DJBlu has any plans to develop a Freeview-plugin, but he seems to be working on a USALS plugin at the moment, and has very little time even for that.

I agree. Even if we had 100 TV Server devs it wouldn't be enough as I wager I could give each one a task on the spot that would keep them busy for a good month or more!
In reality I estimate we have more like 4 or 5 active TV Server devs and they're all occupied with other things at present. Your best bet might be to try and entice one of them with bribes and promises of beer and chocolate.

Perhaps another possibility until something better can be done would be to use another free app that can already properly scan for Freeview channels, and then write an app to import the list of channels it generates into MP's database.

If you can find somebody who is willing and able to do that then all power to you. In the big scheme of things for TV Server I think there are bigger issues to deal with that will take precedence for me.

mm

 

[doveman]

To elaborate: personally I like the idea of an automatic channel management approach - ie. automatic handling of adding, renaming, deleting, moving of channels - over manual approaches like the one suggested by elliottmc. From a user perspective an automated approach seems far more "elegant". Of course there are dangers like not allowing enough flexibility for manual intervention. However on the whole I still favour that approach.

Having said that, I'm also fully aware of the complexity and potential conflicts in implementing such a system. elliottmc's suggestion is probably a far more realistic one for software like MediaPortal that is used in many corners of the world. That is not to say that both approaches couldn't co-exist - just that I would estimate the time involved in implementing the automated approach even for one provider exceeds the time that it would take to implement a reasonably streamlined manual approach that would be more than adequate for most people's needs.

Yes, I think everyone would prefer an automatic approach, but if a manual approach is a lot easier to implement and that's all that someone's willing to do at the moment, it would still be an improvement. I have to disagree that any manual approach would be "more than adequate for most people's needs" however, as this still requires the user to know or go and research which channels actually have been taken off air (rather than just not currently broadcasting), although it seems this wouldn't be an issue for Freeview as the channels are always broadcasting, so perhaps an automated method for that wouldn't be as hard to code compared to some of the other providers.

You may or may not be aware that we discussed scanning at exhaustive length during the 1.2.x release cycle on our internal forums. To the point where I almost cringe when I think about it now. We made some changes that we thought were fairly minor and safe and in my opinion ended up delaying the release of MP 1.2.0 final by a good 2 months. The moral for me is nothing is ever as simple as it seems. There is nothing wrong with discussion but if you want to do that then please start another thread in a more general section of the forum and ask a mod to move the relevant posts.

I wasn't aware of that but obviously a lot of users (including myself) aren't on the internal forums, so I think we should be encouraged to try and think of ideas, in case one of us thinks of something that hasn't already been thought of or an approach that catches a dev's imagination. It's true, this discussion has got OT and I should ask a mod to move some of the posts to a new thread somewhere, although I'm not sure which is the correct sub-forum to post such ideas in, as there appears to be at least two which cover such discussions, which I posted about in https://forum.team-mediaportal.com/...est-one-place-improvement-suggestions-104281/

In reality I estimate we have more like 4 or 5 active TV Server devs and they're all occupied with other things at present. Your best bet might be to try and entice one of them with bribes and promises of beer and chocolate.

I'd have to find them first, as you're the only one I know of I can't imagine I'll be very popular if anyone finds out that I successfully bribed them to divert their efforts away from whatever important thing they're currently doing either

If you can find somebody who is willing and able to do that then all power to you. In the big scheme of things for TV Server I think there are bigger issues to deal with that will take precedence for me.

I don't even know if there's any suitable apps that could be used to scan for channels and then have the data imported into MP's database, but I imagine if there were it would take a lot less work to code something to do this import/translation rather than code a full scanning app. I'll have a look at what's available. Obviously as scanning is only done infrequently, issues with playback & recording are rather more important to fix, but after those I'd say scanning was pretty important to get working, as I imagine MP is bound to end up having problems if the channel list isn't kept up-to-date.

 

[infinite.loop]

For a commercial product, it is rather easy to deal with issues. You just assign a few developers and they will deal with it.
We have quite some troubles with solving issues that are only happening with specific hardware/datastreams.

We do not only need someone with access to that hardware/datastream, we also need someone who is interested (and skilled enought) to fix the issue.
I know that this is very frustrating for our users, but that is how it is. We can not just assign issues to our developers like big companies can (or could, Win7 MediaCenter still has major issues in europe).

So if an issue does not get fix as fast as the users would like to, it will definitly not help to push it, or demand a fix as if this would be a commercial product.

What will help is to collect all informations about the problem, get it confirmed and added to mantis. Maybe a user is even is lucky enought to find an developer who wants to fix the issue.
Also many of our current team members started to learn coding just because they wanted to contribute to MP.

So I want that this discussion, and the others, get back to a level, where the topic is discussed in a polite and respectful way. This also applies to new discussions/threads.
Otherwise I will be forced to step in which I really don't want to.

 

[doveman]

For a commercial product, it is rather easy to deal with issues. You just assign a few developers and they will deal with it.
We have quite some troubles with solving issues that are only happening with specific hardware/datastreams.

We do not only need someone with access to that hardware/datastream, we also need someone who is interested (and skilled enought) to fix the issue.
I know that this is very frustrating for our users, but that is how it is. We can not just assign issues to our developers like big companies can (or could, Win7 MediaCenter still has major issues in europe).

So if an issue does not get fix as fast as the users would like to, it will definitly not help to push it, or demand a fix as if this would be a commercial product.

What will help is to collect all informations about the problem, get it confirmed and added to mantis. Maybe a user is even is lucky enought to find an developer who wants to fix the issue.
Also many of our current team members started to learn coding just because they wanted to contribute to MP.

So I want that this discussion, and the others, get back to a level, where the topic is discussed in a polite and respectful way. This also applies to new discussions/threads.
Otherwise I will be forced to step in which I really don't want to.

I'm sorry if you think I've been out of order. Other than post #13 two days ago, where I expressed some frustration with jameson_uk for responding to my question to elliottmc with an unhelpful answer, and for seemingly telling me the same thing every time I post anything related to scanning/tuning, I think I've only politely asked for clarification on points made or made suggestions. I certainly don't think I've been pushing/nagging or demanding a fix and I quite understand that there's an issue with lack of developer time.

If you think something in my recent posts has been out of order or impolite or disrespectful, please do point it out to me as I certainly don't want to cause unnecessary offence.

If there's anything I can do to help "collect all informations about the problem, get it confirmed and added to mantis" do let me know, as I'd like to contribute in a positive way as much as I can.

Fairly sure deleting both these channels will remove these warnings from the logs

Thanks, that does seem to have fixed the problem

 

[doveman]

Just a heads-up, I've been thinking about a possible solution to the Freeview/DVB-T problem and have posted my ideas here https://forum.team-mediaportal.com/.../separate-app-freeview-dvb-t-scanning-104523/ if anyone would like to take a look.

Thanks.