Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

#390 Advanced MXF

More_attribute
closed-fixed
3
2014-03-04
2012-10-15
Anonymous
No

SMPTE 380M: DMS1 (a standard set of descriptive metadata to use with MXF files):
- production framework (title, locators)
- scene framework (title, locators)

SMPTE 436M: MXF Mappings for VBI Lines and Ancillary Data Packets:
- OP47 subtitles

Discussion

    • summary: Avanced MXF --> Advanced MXF
    • priority: 5 --> 3
    • labels: 553245 --> Libray - File recognition
    • status: open --> pending
     
  • SMPTE ST 380:
    I already support DMS1 for language of each track.
    For more DMS1 support, I need sample files in order to see how complex are the meta you want MediaInfo to support (currently, there is no place in the display for scene meta), and it would be great if you say how you, the end user, would imagine the output.

    SMPTE ST 436:
    I already support VBI and VANC detection in my code, currently I only detect line 21 presence (currently without VBI decoding, it is planned but not yet implemented) for VBI, VANC is already implemented for Bar data, AFD, CEA-708/CDP...
    I don't know a lot about OP47 subtitles, if I trust:
    http://www.tcube.tv/pdf/Teletext_Subtitling_HD_SD.pdf
    it is embedded in SMPTE ST 2031 (so OP47 inside SMPTE ST 2031 inside SMPTE 436, so I would need to implement it both SMPTE ST 2031 and OP47 (at least a minimal detection of OP47).
    I also need sample files.

    I can provide private FTP server access if you need one.

     
  • Printing out only the smpte 291 did, sdid, and data_count value from the smpte 436 anc packets would be a great start.

    I don't like the idea to display the presence of a stream without having detected real content, but if an OP47 stream without real content is rare, adding this DID/SDID to the list of format I know without parsing OP47 may be a first step.
    I don't see any reason to display the packet size (data_count), it is not part of the identification and may change in the stream.

    The main issue is not there. I will not add something in the official source code until I can do some automatic non-regression test on it.
    So this line is still valid: "I also need sample files"

     
  • I added all the DID/SDID I saw in SMPTE specs in this patch, but some of them are not tested (no file), so report issues if there are issues.

    Due to the lack of file for DMS1 (and it is too much risky to implement without a file), I close the ticket. If someone wants better support of DMS1, he must provide some sample files.

     
    • status: pending --> closed-fixed