@mcraenz: If you set the DivX codec for TV then it will be used for playback of TS files as well. This is noted at the bottom of the TV and video codec tabs, and I've confirmed it with Graphstudio. Other than that exception (the TS format), the codecs used for video playback will be the ones you select on the video codec tab. The reason for the exception is that the TS format is used as the timeshifting format. MediaPortal can't tell the difference between a video that you might want to use other codecs for, timeshifting files, and recorded TV files.
Did you have a problem with the DivX codec and recordings?
I've had to go back to Monogram probably due to a MediaPortal 1.1.2 issue. Every 2nd or 3rd advance in a TV recording will cause the audio to go silent. I have to go back or forward again for the audio to come back. I've had this occur with Monogram as well but very rarely, perhaps once a week, not 6 times an hour while skipping ads. I don't watch C4 very often.
@rlevis: Yeah, I noticed that too. That happens for people using DVB-T2 and the DivX codec over in the UK. I don't think it is an MP issue - rather, I'd say it is a DivX codec issue. @mcraenz: I agree that it would be nice to have Monogram fixed, but Igor (RadScorpion) doesn't seem to be very active on Doom 9 or on his website. The South Americans seem to have been asking for a fix for an issue for some time with no response. I still wonder whether the stream really is compliant. VLC (which uses libfaad2 internally) can't play the audio at all. Note that Monogram is based on libfaad2, so that might explain a few things. I wish I had the knowledge to look at the code myself, however this is a bit too far outside what I feel capable of.
Yeah, I can code in c# but this is over my head. I guess if we can get a libfaad dev to look at it then if should just be a matter of re-compiling monogram aac aginst the fixed faad? But the question is whether it is compliant. If it's not then it will be harder to get someone to look at it.
Help me build a new way of doing politics in Aotearoa New Zealand
I've also reverted to Monogram again. I skip ads lots and losing audio altogether on any channel was a "deal breaker" for me. Regarding the actual issue: I've discovered that DVBViewer + Monogram + C4 leads to bursts of "electronic" noise with silence in between - don't really know how to describe it any better. That probably rules out a MediaPortal issue, but I don't know how to explain the difference in behaviour. I've encountered differences before when trying to play back VideoReDo edited TV One/TV 2 recordings (720p stuff, so relatively old) in various applications. VLC and DVBViewer (with Monogram) would sometimes be silent and MediaPortal (with Monogram) would always play fine. That issue seemed to depend on where I made the first cut in the recording. Shifting by a few frames would make the audio play. This seems to have been fixed in the latest VRD betas, although I haven't done much checking as I'm still waiting for the glitches at joins to be resolved/improved. One key thing I learnt from the VRD issue was that it was caused by the AAC stream being incorrectly tagged as low complexity when it was actually high efficiency.
What does that mean? Well I don't really know, except that I think that the same issue could be a factor here. Connecting to the MediaPortal graph shows that the Monogram input pin connection is configured as low complexity. The same information can be seen in the Monogram property page when viewing the channel in DVBViewer. From the PMT I think that the stream should be treated as high efficiency - TransEdit says the profile and level is 0x58, which according to the ITU h.222 standard (here, page 92 of the PDF, labelled page 78) is high efficiency level 2. In short, the problem might not be an exclusive Monogram issue...
Interesting. I reverted to monogram as I also had seizing video and audio after advancing when playing TS files. Now the stuttering during viewing C4 is gone and the playback of files is good. Fixed for me at this stage. Anyone got an idea why this has corrected itself?