1.5.28 Skip settings for new recordings

[johnsonx42]

It's every recording though, not just one that's a little off. Still, nothing is certain until I put the original 1.5.31 version of Unmanaged.dll back in and test, which I will do tonight some time.

[johnsonx42]

false alarm I guess, I don't know any more. A recording that I was watching tonight had a way-off auto-skip, a good 5+ seconds early, so I put the original 1.5.31 Unmanaged.dll back in. The auto-skip came in the same spot. Then I watched one of my wife's shows that I thought she was having a problem with earlier, and it's auto-skips were dead accurate to the point where a couple almost looked like professional edits with perfect fades out an back in with nary an audio glitch. Now I can't remember which shows I was watching last night when I first complained about the comskip points being off again.

At this point I can't put my finger on exactly what is happening, but my HTPC just doesn't seem to work right any more. The high point seemed to be about a month ago with 1.5.28 + some of the patches. Since then it's been downhill. I don't claim it's changes in NPVR causing it, but the whole thing just seems 'off' now. I wsh I could be a little more specific. Just venting for now I guess....

[jksmurf]

Is this still an issue for you? I'm having comskip problems, but wanted to see if I could try a patch first?
k.

[Reddwarf]

Is this still an issue for you? I'm having comskip problems, but wanted to see if I could try a patch first?
k.

I'm having som e issues with comskip too, but it is not consistent. On some channels the skip is a bit too early and resumes a little early too, but on other channels the skip is sevral seconds late.

[johnsonx42]

discussion of current skipping and comskip related issues is here: http://forums.nextpvr.com/showthread.php...d-up-again

in this thread here, my problems in my last posts from 9/13 and 9/14 ultimately turned out to be due to high cpu utilization in nrecord which was actually causing the clock to drift so severely it could drift during the recording by several seconds - therefore the timing.info written into the recording would drift. the high cpu-use turned out to be due to the real-time signal strength info being communicated to the NPVR Tray icon, thus sub removed that via a patch in 1.5.33 (and of course is now getting complaints about why the signal strength info is gone).

I still think there is some issue with skipping in 1.5.36, as something is definitely different than it was with prior versions. However the new <comskip offset> patch offered there has helped me some. I find that skipping is a bit variable from day to day... I set the comskip offset to 3 seconds, and that made skips spot-in that day... but a couple of days later I'm catching 1-2 seconds of commercial, even on the exact same recordings I used to tune the comskip offset a few days ago. Still, I'd rather catch a couple seconds of commercial than miss the last half-line of dialog on the way out, so the <comskip offset> is golden regardless of what may or may not be wrong with skipping.

[johnsonx42]

also, let's continue any other discussion on this in the 1.5.36 "skipping fouled up again thread"... don't MAKE ME wield my moderator hammer and close this one!

after that I start to go power-mad, issuing bans, laying down edicts... it gets ugly, no one wants to go down this road