On 19.05.2014 22:26, Karim wrote:

On 11.05.2014 15:13, Karim wrote:
On 08.05.2014 23:14, Martin wrote:
I'm having a small problem.

Some HD recordings are 1/2 in time of the SD recording.
Not every HD recording has this problem, but most of the time I
from RTL HD it has.

Recording is completed but only the length is wrong. So when I play
back and jump in time, I must jump 60 sec to skip 120 sec etc.

VDR 2.1.1 but had the same problem with other versions.

Is the problem still present in version 2.1.6?


FYI, I noticed this behaviour here with vdr-2.0.6 and french HD channels.

Can you make a small sample (like a minute or so) of such a recording

Of course.
Please download sample from here :


I have tested this sample with both VDR 2.1.1 and 2.1.6, and in both cases
'vdr --genindex..." creates an index file that is about twice the size of
your file and thus appears to be correct.

Was this sample taken from the very beginning of your recording or somewhere
in the middle? Maybe the error is only caused by a specific sequence in the
stream, which is not contained in the sample.
If you can produce a sample where vdr's --genindex option generates a faulty
index file, please let me know.


I think that sample is from very beginning, what I am not totally sure. I
will try to check again.
Could you please tell me how I can detect a "faulty index" ?

If you encounter a recording where the time shown in the progress display
is not correct, the index is most likely wrong. In such a case please do
as follows:

- change into the *.rec directory of that recording
- rename the file index to index.old (or whatever you like)
- run the command
  vdr --genindex=`pwd`
  (`pwd` will be replaced by the name of the current directory)
- do an
  ls -l index*
  and compare the size of the old and new index file. If they are the
  same, that means that --genindex produces the same index file as did
  the original recording, and thus I should be able to reproduce the problem
  here if you send me the first minute of that recording.


vdr mailing list

Reply via email to