[mythtv-users] Problem making cut lists and transcoding recordings

[PJBrs]

Dear list,

I'm a first time poster here but I've used mythtv for several years now, 
first on a VIA EPIA SP8000 and now on an Intel D945GSEJT with Atom 
processor. I used to be able to make very precise cut points (by the 
frame) and the transcoded recordings would exactly honour the cutlists. 
However, recently I've noticed problems both in making a cut list and 
subsequently transcoding a recording. I was wondering whether anybody 
else has encountered problems as well. If not, I can open a bug report 
and help investigate.

The problem is as follows. First, I have trouble adding a cut point. 
When I'm searching _forward_ for the right frame, I sometimes see 
different frames searching forward than when I subsequently search 
_backward_, even though the frame numbers do not change. In other words, 
the same frame number appears to be associated with a different frame 
searching forward than searching backward. Searching backwards appears 
to display frames from just a little bit later in the recording that 
search forward, with a difference of no more than 1 second, as far as I 
can tell.

Second, when I do end up making a cut list, and when I next transcode 
the recording (usually mpeg->mpeg), it appears that the cut list is 
honoured exactly. I haven't experimented much, but it appears that 
several minutes are removed from the end of the recording (in addition 
to what I cut out), and some parts remain in the transcoded file that 
should have been cut out. I other words, the recording is basically 
corrupted and not very interesting to watch anymore.

Here's some context. I'm currently on 0.24.1. I did not have this 
problem earlier, but I cannot remember when I started to have it. It 
might be related to the Intel D945GSEJT that I'm using now, but it might 
also be a bug that has been introduced in a later version of mythtv 
(I've used several versions of mythtv, starting from 0.21, IIRC).

I'm using Slackware 13.37, kernel 2.6.38.8. So, does anyone recognise 
this problem? Does anyone have a solution?

Best regards,

PJBrs