Experimental TsReader development (3 Viewers)

[JackTramiel]

AW: Re: AW: Experimental TsReader development

EDIT: Ok, watching live TV some time suddenly audio keeps playing but picture freezed.

I'd be interested in logs for this situation.

Tony

Reverted back to stock dll and ax files - my wife was coming home.
There should be some logs anyway, so would those logs be enough?
Otherwise i need to start in debug mode and grab some full logs.
Not a problem, first way would just be a bit faster.

 

[Owlsroost]

Re: AW: Re: AW: Experimental TsReader development

EDIT: Ok, watching live TV some time suddenly audio keeps playing but picture freezed.

I'd be interested in logs for this situation.

Tony

Reverted back to stock dll and ax files - my wife was coming home.
There should be some logs anyway, so would those logs be enough?
Otherwise i need to start in debug mode and grab some full logs.
Not a problem, first way would just be a bit faster.

Any logs that show the problem would be fine (but it would be good if you could say what the time was when it happened - makes finding it in the logs much easier ).

Tony

 

[FreakyJ]

AW: Experimental TsReader development

seems there is a bug oO
There was a corruption in the stream, the picture freezes for 2 or 3 seconds, the sound stuttered 1 or 2 seconds and then I saw something what I watched ~45 minutes ago
I used the skipsteps to skip forward an I could jump over a half our forward.
Before the corruption I was on the live point oO

Logs attached Time: 13:47 maybe it was 13:46, but I think it was mire 13:47

 

[Owlsroost]

Re: AW: Experimental TsReader development

seems there is a bug oO
There was a corruption in the stream, the picture freezes for 2 or 3 seconds, the sound stuttered 1 or 2 seconds and then I saw something what I watched ~45 minutes ago
I used the skipsteps to skip forward an I could jump over a half our forward.
Before the corruption I was on the live point oO

Logs attached Time: 13:47 maybe it was 13:46, but I think it was mire 13:47

If you get big signal corruptions/dropouts the stream PCR and PTS data can get corrupted - this can cause all sorts of problems.....(your logs have some big dropouts and some continuity errors, plus 'seek' timeouts and big >10s PTS jumps). I suspect the (multi-segment) timeshift file handler might have got confused (or possibly the TV server end of things) - it's the only thing I can think of which could cause it to go back 45 minutes...

Tony

 

[mrmojo666]

HI Tony!

i've found this discussion in the last days, so it's quite hard to me find what is your goal rewriting the tsreader. can you please explain well your goal?

i've tried to follow many link found here and i'm surprised and confused cause the discussion is spanned wide over many threads.

even i haven't undestand what bugs you're trying to solve, i started to test those tsreaders. hoping could resolve a couple of small issue that i have

i'm in a multiseat environment

1) on my laptop using wifi, i have to switch in UNC mode to get tv working (rstp provide just a black screen)

2) i have three channel SD that work perfectly in UNC mode (over ethernet and wifi), but if i use rstp over ethernet when zap to one of those channels i get just the frozen picture of the previous working channel (sometimes the video and audio start then slow until freeze), skip fw or bw doesn't help. zap to other channel makes tv working again. (it happens on all my three mp clients)

well i didn't notice any change of beavoiur until i've tried the v41

it solved the point one, now i can use rstp also from WIFI. but it didn't solved the point 2.

i'm going to attach the tsreader log. i don't know if the point 2 is related to your goals or tsreader as well.

i hope my test could be useful.

thanks

bye

With the 'problem' channel TsReader doesn't receive any data after the first few seconds (as far as I can tell) - need the server logs as well to work out what's happening.

Tony

here the logs

there are log server/client using RTSP that show the problem and UNC logs that show on smae channel there is no problem


thank you

 

[Owlsroost]

With the 'problem' channel TsReader doesn't receive any data after the first few seconds (as far as I can tell) - need the server logs as well to work out what's happening.

Tony

here the logs

there are log server/client using RTSP that show the problem and UNC logs that show on smae channel there is no problem


thank you

You know the rules - you are using MD**I - no support unless you can reproduce the problem without it.....

Tony

 

[mrmojo666]

sorry i didn't thought this issue that happens just with rtsp is related to mdapi , so i haven't disabled it.
i could understand if we are talking about tune or crash or other issues......

but all is working ok by UNC so really i still think this "rule" should be weighted case by case


anyway i just want to help in make MP better and give help but this issue is happening on three crypted channels ( that i pay, so no piracy ).

at the moment i haven't yet zapped to all fta channel to find the issue.

if i'll find i'll post to you, but i hope you can understand and use the provided before

bye

 

[dyamant]

AW: Re: Experimental TsReader development

With the 'problem' channel TsReader doesn't receive any data after the first few seconds (as far as I can tell) - need the server logs as well to work out what's happening.

Tony

here the logs

there are log server/client using RTSP that show the problem and UNC logs that show on smae channel there is no problem


thank you

You know the rules - you are using MD**I - no support unless you can reproduce the problem without it.....

Tony

I understand mrmojo666.

and i undersand you that the rule is no support for MD****. But this short Dll is a very good solution to look the channels with my own payed PayTV-Card very stable and without a expensive Card Reader.

So maybe have a look on it. Please.Thnx.

 

[Owlsroost]

There's nothing I can see in the logs to explain what might be going wrong, but I'm not an expert on the TV server end of things.

RTSP runs different code at both ends of the system, so it's quite possible for something to work with singleseat and UNC but break with RTSP - sorry, but I can't help further.....

Tony

 

[FreakyJ]

AW: Experimental TsReader development

If you get big signal corruptions/dropouts the stream PCR and PTS data can get corrupted - this can cause all sorts of problems.....(your logs have some big dropouts and some continuity errors, plus 'seek' timeouts and big >10s PTS jumps). I suspect the (multi-segment) timeshift file handler might have got confused (or possibly the TV server end of things) - it's the only thing I can think of which could cause it to go back 45 minutes...

Tony

Thanks a lot for you info, I guess my bad signal quality is caused by a lot of wires behind the PC, because on a different place Signalquality is much better oO
But it is good for testing TSReader