<div class="gmail_quote">On 11 December 2012 16:47, Karl Newman <span dir="ltr"><<a href="mailto:newmank1@asme.org" target="_blank">newmank1@asme.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Tue, Dec 11, 2012 at 2:54 AM, Brian J. Murrell <span dir="ltr"><<a href="mailto:brian@interlinx.bc.ca" target="_blank">brian@interlinx.bc.ca</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
On the subject of commskipping inaccuracies though, it seems really odd<br>
to me. There are lots of times that the commflagger flags beyond the<br>
end of the commercial break even when there is a very clear set of<br>
"black frames" at the end of the break. I suspect something fundamental<br>
has gone funky in the commskipping code. Maybe it's better on 0.26<br>
(although I'm not holding my breath). I'm still on 0.25-fixes.<br></blockquote></div><br></div>I doubt 0.26+ has fixed anything--I haven't heard about any progress in that regard. commflagging got completely broken in 0.25. It used to work nearly perfectly in 0.24, but something changed in 0.25 and now I'm lucky if it catches 1 out of 5 breaks. I also had to prevent it from running while the recording was occurring, otherwise it errored out with an end-of-file, apparently some old latent race condition that has only now manifested itself. There is reason for hope, though. Somebody (Stuart M?) was collecting samples of shows from mythtv users around the world to test the commflagger. But that's been 6 months or so ago. I don't know what's going on with it now.<span class="HOEnZb"><font color="#888888"><br>
<br>Karl</font></span><br></blockquote><div><br>It was Gavin Hurlbut, I'm just about to send my samples over, not sure if anyone else has done the same <br><br>Cheers<br><br>Martin<br></div></div>