User Activity

  • Posted a comment on discussion General Discussion on bmx

    The 'L'ength in KLV can be encoded using 1 to 9 bytes depending on how big the 'V'alue is and other constraints such as a requirement to use a fixed number of bytes (e.g. many essence mappings specify 4 bytes be used). The first byte tells you how many bytes are used to encode the length, e.g. if the first byte is < 128 then it is only 1 byte. The 'llen' parameter is the number of bytes that were used to encode the 'L'ength and 'len' is (decoded) 'L'ength. The mxf_read_l function therefore returns...

  • Posted a comment on discussion General Discussion on bmx

    You could rewrap all the .w64 audio to an MXF file using ffmpeg as that should be relatively quick. Then use bmxtranswrap to produce your output file from the 2 MXF files. Use --disable-audio in front of Test.mxf to exclude that audio. Use the --track-map option to reorder the audio channels if required. Use --pass-vbi and --pass-anc all to passthrough anc data. Philip

  • Posted a comment on discussion General Discussion on bmx

    You would have to copy the UMIDs manually across using the --mp-uid and --ps-uid options. The wrap utilities are very basic and don't copy across descriptive metadata (other than some copying functionality that was required for AS-11/DPP). Philip

  • Posted a comment on discussion General Discussion on bmx

    The video has a height of 508 which doesn't match the required height - the code is fairly restrictive as it requires the display dimensions to be 720x486. The audio is uncompressed AIFF big-endian, not little-endian. These essence types are unfortunately not supported in bmx. Philip

  • Posted a comment on discussion General Discussion on bmx

    Thanks. I've pushed a fix to the git repository. Philip

  • Posted a comment on discussion General Discussion on bmx

    bmx is an open source project. The source code is available in a git source code repository; see the "Git" menu for links to the repositories. The fix I made was added ("pushed") to the git repository. You can re-compile the source code (see the "Home" menu) to produce a new set of executables that will have the fix. The last release I posted was compiled with the earliest still available Microsoft compiler, Visual C++ 2010, rather than version 2005 as stated in the "Home" page. Philip

  • Posted a comment on discussion General Discussion on bmx

    I'm sorry bmx is not providing what you need. bmx was not intended to be a fully fledged MXF processing tool, and so I'm not surprised that changes to some formats have meant that it can no longer support them. I'd suggest trying some commercial offerings which are more likely to track changes to the Avid format for example. Hopefully they would also provide the AAF functionality you are after. The AAF SDK provides a library that would allow you to create the .aaf file, but I don't think the AAF...

  • Posted a comment on discussion General Discussion on bmx

    I've pushed a fix to the git repo for the sample file you provided. Your sample file had an index table that was different from the ones I've seen previously from Avid. The difference resulted in an invalid read which could result in a segmentation fault (e.g, the error you saw) which is not good. It looks like bmx doesn't recognise the essence types in your files. I suspect Avid have made changes to their format (such as the index table change) which means bmx no longer supports the format. I don't...

View All

Personal Data

Username:
philipn
Joined:
2003-04-29 10:38:41

Projects

Skills

  • No skills entered.

Personal Tools

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

JavaScript is required for this form.





No, thanks