The playback settings can be configured to be different based on the resolution of the recording being played. The difference in behavior could be a result of these rules. This gets discussed a bit in this
thread. In that thread I suggested the following settings:
Quote:
In UTILS/SETUP->SETUP->TV Settings->Playback->Playback Profiles (screen 3), you need to create a new Video Playback Profile with the following 3 entries. In theory they can be created in any order, but as a precaution, please create them in the following order:
1. Match Criteria1 >= 1280 720, Match Criteria2 < 1920 1080, decoder=standard, MAX CPUs=2, video renderer=xv-blit, osd-renderer=softblend, osd fade=NOT checked, primary deinterlacer=None, Fallback deinterlacer=None. FYI -- these settings will be used to play back 720p recordings.
2. Match Criteria1 >= 1920 1080, Match Criteria2 = blank, decoder=standard, MAX CPUs=2, video renderer=xv-blit, osd-renderer=softblend, osd fade=NOT checked, primary deinterlacer=One Field, Fallback deinterlacer=None. FYI -- these settings will be used to play back 1080i recordings.
3. Match Criteria1 < 1280 720, Match Criteria2 = blank, decoder=standard, MAX CPUs=2, video renderer=xv-blit, osd-renderer=softblend, osd fade=checked, primary deinterlacer=bob2x, Fallback deinterlacer=One Field. FYI - these settings will be used to play back 480i recordings.
Please let me know how this works out. Some additional observations:
1. Your CPU is borderline for hidef playback. You will not have success without HT or XVMC. In this configuration, we are trying to use HT instead of XVMC. If that does not work, we will try XVMC.
2. One field is not a great deinterlacer, but it should give you reasonable results and is the least CPU intensive of all the choices.
3. In your initial experiments, please watch an already recorded show while you are not recording anything else. Do not attempt to watch live TV to gauge if this helps. If playback works by itself, you can then move on to seeing how it does when recording while playing back a show; then while watching live TV.
4. It would be useful if you remote logged into your myth box, ran "top -d 1" and reported back the % idle time available while doing playback.
Before you change to those settings, I suggest you check to see what settings you currently are using. I would also check the myth frontend log file to see if there are any telling entries -- its in /var/log/mythtv. I don't recall the exact filename, but
Code:
ls /var/log/mythtv/*front*.log
should find it for you.
Marc