The bmx project has moved to GitHub: - https://github.com/bbc/bmx - https://github.com/bbc/libMXF - https://github.com/bbc/libMXFpp No more development will be done in the Sourceforge git repos. The Sourceforge "General Discussion" forum will remain open for a couple months before it is made read-only. The discussions can now happen in the bmx discussions on GitHub . The existing file releases will be kept on Sourceforge.. The aim is to make future file releases on GitHub. Philip
Home
Fix assignment of UL - error: expected expression
Fix forcing HD colorimetry for MPEG2 SD essences
Looks good, thanks.
Hi Philip, Nicolas fixed it and prepared a new commit. :) https://sourceforge.net/u/ngaullier/bmxlib/ci/Fix_MXFSDColorDesc/ You can consider this ticket closed. ;)
Fix forcing HD colorimetry for MPEG2 SD essences
readme: markdown linter updates
readme: remove reference to libmxf debian package
readme: markdown linter updates
readme: remove reference to libmxf/libmxf++ debian packages
Hi Philip, Hope you are well ! I'm still alive in BC S/W after 2 take-overs... Hi Sergino, I was the one back then emitting the patch for Avid growing files. I recall that the patch enabled growing OPAtom files for fixed-frame-size essence and was tested back then with AVC-Intra in the then available Media Composer application versions and worked growing Ok. I also had a patch for growing Long GOP/non-fixed-frame-size essence (like XDCAM) that needs regular updating of the Index Tables, but that...
I don't remember the details but one thought is that you may need something like an AAF file to register the clips in Avid. Creating an AAF file for the growing MXF OPAtom files is not something bmx provides. This thread may be of use, https://sourceforge.net/p/bmxlib/discussion/general/thread/23d0e9d3/ Philip
Hi I'm attempting to generate from an MXF OP-1A DNxHD MXF file (8 audio tracks) an OP ATOM clip for AVID media composer in growing mode (that is with enabled the EditWhile Capture) option. Command line is : bmxtranswrap.exe -t avid --avid-gf -o "Avid MediaFiles\MXF\1\testGrow" "Source_20220830_160126.mxf" I found that op atom are correctly created, but media Composer doesn't see them while writing. There is something I missed ?
imf_audio_track_files: fix md link
README: fix md link
imf_track_files: fix md link
test
test
test
SF markdown requires 4 spaces for list indent
Remove additional whitespace
bmxtranswrap: break if/else to avoid msvc limit
Various updates to imf docs
Fix markdown lint issues
Various updates to imf docs
Shift emphasis to IMF for JPEG 2000 doc
Shift emphasis to IMF for JPEG 2000 doc
Link to BT 2100
Improve and lint IMF-related documentation
Reference new docs in README
Doc: add relative paths to file links
Doc: correct language tag for code blocks
Doc: correct language tag for code blocks
Split IMF Audio Track Files into separate doc
docker: add bmxtimecode to list of tools
Rename label to avoid confusion with a define with the same name
docs: Rename IMF track files doc
docs: improve IMF doc sentences
docs: add reference to docs directory in imf doc
Improve IMF doc intro and refer to Track Files
Set DisplayF2Offset 0 for progressive RDD36 in IMF
Add IMF essence component README
apps: Add imf clip type to usage messages
timed_text.md: change mono-essence timed text example to be IMF
bmxtranswrap: check file descriptor not null before updating
as02: add PrepareHeaderMetadata stage
Add IMF regression tests
test: change '_' to '-' in create_samples/create_data
Output IMF for jpeg2000 tests and readme examples
Add mxflib props & types for IMF
Add options to set IMF sound & pict props
apps: add 'imf' channel assignment label to --audio-layout
apps: Default to singlemca for clip wrap OP1a
apps: use edit unit in usage if can be video frames or audio samples
apps: use edit unit in usage if can be video frames or audio samples
Only use leading zero in duration string for rates <= 99Hz
apps: don't require audio frames if audio only
Replace fps with Hz in messages for audio rates
op1a: Don't write CBE index if duration == 0
op1a: add option to force index follows essence
tests: fix tests to exclude new non-static info log messages
Fix doc typo
Slightly improve presentation of MCA label readme
mxf2raw: log message when writing non-info files
Various fixes to MCA label file readme
Clarify/expand slightly MCA label text file doc
Add description of properties to MCA label text file doc
Add regression test for MCA label file parsing and string properties
Add new MCA properties to mxflib dict
Support additional label properties in text file
Allow redundant GOSG labels to be omitted in text file
Make minor clarifications
Combine MCA labels into single index
Add markdown description of audio labels file format
Lint Markdown
Error messages differentiate between Options and Input Options
State that mono is default for --track-map
mxf2raw: fix usage typo
mxf2raw: add primary package property to file and clip section
tests: update checksums after primary package in as02 exposed in mxf2raw
Add new properties to MCALabelSubDescriptor
Add IMF generic picture & sound properties
examples: fix uninitialised var reported by gcc
Add IMF generic picture & sound properties
Add new properties to MCALabelSubDescriptor
Add ability to build MXFDump on/for Apple Silicon
Yes I meant the macro name. Your changes look good as-is and so I've pushed. Thanks. Philip
Fix macro, type definitions for ARM MacOS
Ah yes, I had missed that the AAF SDK is publically available and did not double-check, that's my bad. Thank you for letting me know! I'm not quite sure what you are referring to when saying " the name is different", are you referring to the MXF_COMPILER_GCC_ARM_MACOSX macro? As for the type definitions, you are correct; I've pushed a change to fix this error. If you'd like, I could rename the macro and switch __aarch64__ for __arm64__ to align with the SDK as well, though these changes are already...
I had a look at the AAF SDK, which is where MXFDump originates from, and it does something different: https://sourceforge.net/p/aaf/code2/ci/a4a578fa0dca53126d0f6925ee11039ec402c9e7/#diff-6 The name is different but more importantly the typedef for mxfUInt32 is "unsigned int" in the AAF SDK change, whereas your change uses "unsigned long int". Based on the other 64-bit definitions I would guess that "unsigned int" is the correct one. Could you confirm whether the definitions in the AAF SDK are correct?...
Add ability to build MXFDump on/for Apple Silicon
I contacted him on Linkedin two days ago, I hope he's gonna reply. :)
Support for MPEG-2 576i was only recently contributed to bmx https://sourceforge.net/p/bmxlib/bmx/ci/a96406a18bc936b0a6d711dae134ebdd808195ba/ . Before that bmx only supported D-10 MPEG-2 SD. The support is only for a narrow range of files. The aspect ratio is hard coded to 16:9 and the transfer characteristic is hardcoded to BT.709 (Color Primaries and Matrix coefficients are not set). You can try contacting the contributor to see if they can help you to extend support for MPEG-2 SD. Philip
Hi there, it's me again; sorry to bother you with an SD File issue in 2022, but looks like they're still in use, sadly... :( Essentially, I'm encoding SD files with the following command line as per our Omneon based playout specs: ffmpeg.exe -i "A:\MEDIA\temp\Test.avs" -an -pix_fmt yuv420p -vcodec mpeg2video -s 720:576 -aspect 4:3 -vf setfield=tff -flags +ildct+ilme+cgop -b_strategy 0 -mpv_flags +strict_gop -r 25 -sc_threshold 1000000000 -b:v 12000k -minrate 12000k -maxrate 12000k -g 15 -bf 2 -profile:v...
Hi there, it's me again; sorry to bother you with an SD File issue in 2022, but looks like they're still in use, sadly... :( Essentially, I'm encoding SD files with the following command line as per our Omneon based playout specs: ffmpeg.exe -i "A:\MEDIA\temp\Test.avs" -an -pix_fmt yuv420p -vcodec mpeg2video -s 720:576 -aspect 4:3 -vf setfield=tff -flags +ildct+ilme+cgop -b_strategy 0 -mpv_flags +strict_gop -r 25 -sc_threshold 1000000000 -b:v 12000k -minrate 12000k -maxrate 12000k -g 15 -bf 2 -profile:v...
Hi there, it's me again; sorry to bother you with an SD File issue in 2022, but looks like they're still in use, sadly... :( Essentially, I'm encoding SD files with the following command line as per our Omneon based playout specs: ffmpeg.exe -i "A:\MEDIA\temp\Test.avs" -an -pix_fmt yuv420p -vcodec mpeg2video -s 720:576 -aspect 4:3 -vf setfield=tff -flags +ildct+ilme+cgop -b_strategy 0 -mpv_flags +strict_gop -r 25 -sc_threshold 1000000000 -b:v 12000k -minrate 12000k -maxrate 12000k -g 15 -bf 2 -profile:v...