skywalker-1 tuner scan all dish channel get unknown name (1 Viewer)

[wderbi]

No I am not Mondos...his guide came out sometime later in the same year as mine. My guide came out in Feb 2011, was better written (IMHO), and I actually did all of the groundwork for the conversion. Mondo's guide was written in October 2011, was based solely on my guide, but was: (1) slightly updated with updated drivers/software versions, (2) had a little more detail in how to build a HTPC from scratch and what parts to purchase for a nicely built HTPC. In comparison, my guide was written for the CW>HTPC conversion or for those who already knew how to build a PC.

Mondo's guide was just a repeat of my guide...Even though he states he was not trying to take credit for the contents of the guide, I didn't think he gave me the proper credit for all of the work that I had done detailing how to get MP to work as a replacement for the CW4000. FWIW, I was also the person who wrote the guides in how to get the CW4000 to work as a CS client and how to get it to work as a CS server. I spent a lot of time trying to get the CW4000 to last as long as possible by making it a viable box after the N2>N3 switch.

 

[wderbi]

Wow! I just did a Google search on "Converting Cw4000 to Mediaportal" and only Mondoman's guide showed up! This is a sad state of affairs as this is because all of the sites I posted my guide on have gone by the way side. I guess it is a good thing that Mondo did post his guide over at 'pros, otherwise there would be none on the net (since all record of my guides are buried with the shutdown sites.) If you do want to see my version, just send me a PM on this site (or over at pros) and I will send it to ya. I don't think it will be allowed to upload on the public forums here.

EDIT: I did find my version of the guide over at Satelllite Supreme. It was posted on that site by Priestmill (not me). Well, at least it still exist somewhere!

 

[mm1352000]

Sorry, no - it's been a busy week, and I'm now away on vacation until 26th Sept.

No problem - no rush. There was never going to be enough time to fix this properly for MP 1.5.

I don't really understand why a mux has only 'other' SDT data - it seems non-compliant ? i.e. it carries data for other muxes but not for itself ?

According to DVB spec it is required to carry an SDT actual table (see EN 300 468 page 18 - the coloured diagram with tables). So, yes Dish transmissions are not strictly DVB compliant... but the aim of TsWriter is not strict DVB compliance. Other "generic" SI and EPG standards (such as ATSC, SCTE, ISDB) as well as proprietary variants and schemes (eg. Freesat Huffman-compressed EPG) are already supported or would be useful to support. I have no problem with that. My opinion is that this is simply an example of the reason why great care needs to be taken when modifying our core scanning code.

I do not have knowledge to give the explicit details of what exactly are the areas that the EPG data is non compliant...

I've never had the opportunity to examine a Dish dump before, so this is interesting. As far as I know we've made no changes that would affect linking between EPG and services (channels). I'll see what I can see in the dump file...

 

[liewfr]

Wderbi, Thank for your offer.

Regarding this problem, mm1352000, you mean this will not be fixed on 1.5 ???

 

[mm1352000]

Regarding this problem, mm1352000, you mean this will not be fixed on 1.5 ???

Yes, that's right. It is simply too late. Our release process has rules, and at this stage we're locked and loaded for the final release - final internal testing is ongoing and no further changes are allowed.

 

[Owlsroost]

I've looked at the TS dump and I confirm that only SDT other info is present. We'd need to roll back that change to fix this problem.
@Owlsroost were you able to confirm that the "movement" option is able to prevent creation of dupes from Freeview?

I've just tried a TsWriter version with the '0x46' change rolled back.

I deleted all channels and then performed a scan on one tuner with 'movement detection' disabled - some of the channels had '10000' channel numbers.

I then deleted all channels again, enabled the 'movement detection' option and did a scan - some of the channels had '10000' channel numbers (so no change in behaviour with 'movement detection' enabled).

I repeated the tests a couple of times - same results each time.

Note that I just get some channels with '10000' channel numbers - no duplicates are created (and from memory the EPG gathering doesn't work for the '10000' channels, which is how I ended up investigating the issue originally)

 

[Owlsroost]

@mm1352000 - is there a way of fixing the 0x42/0x46 'Freeview' problem (since enabling movement detection doesn't help - see above) by using the 'otherMux' info field ?

 

[mm1352000]

I think I hadn't grasped the full scope of the issues involved here.
You're right - if TV Server doesn't end up assigning the correct TSIDs (ie. the TSID from the "actual" SDT) to channels then it does have potential to affect grabbing of LCNs and EPG. I wouldn't have noticed this because I manually assign my LCNs and use XMLTV for EPG.

So, simply rolling back the change is not going to be an adequate solution for the reasons you've described... but nor is simply using the OtherMux flag. Consider what happens if you receive the SDT other first and TsWriter scan complete conditions are met before SDT actual arrives to overwrite - you'd have the same problem with LCNs and EPG. I think the solution could involve setting the TSID from the TSID in the PAT (rather than using the TSID in the SDT) and then ignoring SDT records unless the TSID and SID match.

I'm almost completely focussed on getting a TS multiplexer written and working by the end of tomorrow so if you want to have a crack at this, by all means go ahead. I don't have the time until the muxer is done.

 

[FreakyJ]

I'm almost completely focussed on getting a TS multiplexer written and working by the end of tomorrow so if you want to have a crack at this, by all means go ahead. I don't have the time until the muxer is done.

If I should help you testing in some way let me know From Monday on I have quite some time again

 

[Owlsroost]

I'm almost completely focussed on getting a TS multiplexer written and working by the end of tomorrow so if you want to have a crack at this, by all means go ahead. I don't have the time until the muxer is done.

Hi mm,

I'm happy to have a go at fixing it (although I'll probably need to ask a few questions along the way, since I know very little about the internals of TsWriter/the scanning process).

In the meantime, would it be OK to attach a 'rolled-back' version of MP 1.5 TsWriter here for the benefit of the sat users with the scanning problems ?