TsReader - TV black screen and RTSP 'infinite loop' fixes (1 Viewer)

[Owlsroost]

As requested, supplied both unc and rtsp for comparison + FTR logs, would say there is a good 2 second difference, and I am using the loopback adapter with fixed ip.

View attachment 98812

View attachment 98813

View attachment 98814

New version 0_4_17a added - can you try this with 4TR + RTSP please ?

Thanks,

Tony

 

[glenn 1990]

As requested, supplied both unc and rtsp for comparison + FTR logs, would say there is a good 2 second difference, and I am using the loopback adapter with fixed ip.

View attachment 98812

View attachment 98813

View attachment 98814

New version 0_4_17a added - can you try this with 4TR + RTSP please ?

Thanks,

Tony

What's the problem with 4TR, slow zapping??

The 1.6.0.x release restarts the player on channel change to avoid the RTSP issues, so this will stay slow with every tsreader.
But this hack is removed in the 1.6.1 beta releases, because there is some progress in fixing the issue.

 

[Owlsroost]

The extra delay that Ray was getting was due to the TsReader mod - I could see that from the log. The 4TR client/server behaves differently to MP client/server.

Tony

 

[Paranoid Delusion]

Sorry for the delay, here are logs with latest tsreader, does seem slightly faster, but still a noticeable difference.

View attachment RTSP 0.4.1.7a.zip

 

[Owlsroost]

Hi Ray,

It looks like the latest version is working how it should

The zapping delay is a mixture of how 4TR is handling the zapping (as glenn1990 points out, it's completely rebuilding the graph on a channel change when using RTSP - which is the most reliable but slowest thing to do), and how long it takes TsReader.ax to find the first GOP boundary after the stream data start to arrive - which depends on how the stream is encoded and where it 'jumps in' relative to the GOP structure, amongst other things.

Thanks for the testing.

Tony

 

[Paranoid Delusion]

Thanks for explaining, sticking to unc anyway.

 

[elliottmc]

I have tested the 0.4.7.1a binary, and it is working fine with 1.2.2 + some other fixes. I didn't do any comparisons, but all seems fine so far.

Mark

 

[elliottmc]

These fixes are now approved and merged to the Release_1.2.X and master GIT branches. They will be in the next release (1.2.3).

Should we now close this thread?

Mark

 

[esas]

Just wanted to say that I've been testing 0.4.7.1a for a while, and it doesn't seem to break anything. Hoping that future updates will include work from your noStopMod-thread..

 

[Owlsroost]

These fixes are now approved and merged to the Release_1.2.X and master GIT branches. They will be in the next release (1.2.3).

Should we now close this thread?

Mark

I guess so

Tony