[mythtv-users] How to transcode HD recording

John Pilkington J.Pilk at tesco.net
Sat Jan 11 13:07:27 UTC 2014


On 11/01/14 12:50, John Pilkington wrote:
> On 11/01/14 02:04, HP-mini wrote:
>> On Thu, 2014-01-09 at 20:14 +0000, John Pilkington wrote:
>>> On 07/01/14 23:29, John Pilkington wrote:
>>>> I made comments earlier in this thread about wide and variable spacing
>>>> of keyframes in FreeviewHD UK programmes.  I've been looking at this a
>>>> bit more, and my latest test recordings have been better behaved.  The
>>>> problem certainly exists, but it may be unusual.
>>>
>>> I have just found that for a recording from DVB-T2 FreeviewHD (h264), a
>>> listing of the markup data generated after mythcommflag --rebuild
>>> contains only a small subset of the lines in a similar listing made
>>> immediately after recording.
>>>
>>> In the 'original' table, keyframe separation is typically around 24
>>> frames.  In the 'rebuilt' table the spacing is variable, sometimes in
>>> the hundreds.  Points listed there are also in the original table.
>>>
>>> In my experiments I have often been working with recordings having a
>>> rebuilt seektable.  This explains why editing has sometimes been
>>> difficult.
>>>
>>> Should I open a Ticket?
>>>
>>
>> How about a 100MB sample cut from beginning of recording using "dd" ?
>>
>> The UK freeview H264 HD sample clips (I have here) behave well..my
>> seektables had to be generated by mythcommflag..
>>
>> Broadcast OTA H264 uses intra-refresh so the peak bitrate is
>> constrained. All the broadcast samples (I have seen) use this & have
>> about 2 sec keyframe intervals.
>>
>
> http://code.mythtv.org/trac/ticket/12010
>
> I will attach markup files and a link to a short clip asap.
>
The link to the clip was rejected as 65% probable spam.  Let's try here.

http://www.mediafire.com/watch/xeufvnx3etw9umi/1104_20140111121300.mpg




More information about the mythtv-users mailing list