[FFmpeg-devel] [PATCH] Limited timecode support for lavd/decklink

Thu Jun 7 19:12:43 EEST 2018

> On Jun 6, 2018, at 5:32 PM, Marton Balint <cus at passwd.hu> wrote:
> 
> On Wed, 6 Jun 2018, Dave Rice wrote:
> 
>>> On Jun 6, 2018, at 4:50 PM, Marton Balint <cus at passwd.hu> wrote:
>>> On Mon, 4 Jun 2018, Dave Rice wrote:
>>>>>> In my testing the timecode value set here has corrected been associated with the first video frame (maintaining the timecode-to-first-frame relationship as found on the source video stream). Although only having first timecode value known is limiting, I think this is still quite useful. This function also mirrors how BlackMagic Media Express and Adobe Premiere handle capturing video+timecode where only the first value is documented and all subsequent values are presumed.
>>>>> Could you give me an example? (e.g. ffmpeg command line?)
>>>> ./ffmpeg -timecode_format vitc2 -f decklink -draw_bars 0 -audio_input embedded -video_input sdi -format_code ntsc -channels 8 -raw_format yuv422p10 -i "UltraStudio 3D" -c:v v210 -c:a aac output.mov
>>>> This worked for me to embed a QuickTime timecode track based upon the timecode value of the first frame. If the input contained non-sequential timecode values then the timecode track would not be accurate from that point onward, but creating a timecode track based only upon the initial value is what BlackMagic Media Express and Adobe Premiere are doing anyhow.
>>> Hmm, either the decklink drivers became better in hinding the first few NoSignal frames, or maybe that issue only affected to old models? (e.g. DeckLink SDI or DeckLink Duo 1). I did some test with a Mini Recorder, and even the first frame was useful, in this case the timecode was indeed correct.
>>>>>>> I'd rather see a new AVPacketSideData type which will contain the timecode as a string, so you can set it frame-by-frame.
>>>>>> Using side data for timecode would be preferable, but the possibility that a patch for that may someday arrive shouldn’t completely block this more limited patch.
>>>>> I would like to make sure the code works reliably even for the limited use case and no race conditions are affectig the way it works.
>>>> Feel welcome to suggest any testing. I’ll have access for testing again tomorrow.
>>> I reworked the patch a bit (see attached), and added per-frame timcode support into the PKT_STRINGS_METADATA packet side data, this way the drawtext filter can also be used to blend the timecode into the frames, which seems like a useful feature.
>> 
>> 
>> That sounds helpful.
>> 
>> libavdevice/decklink_dec.cpp:734:21: error: unknown type name 'DECKLINK_STR'
>>                   DECKLINK_STR decklink_tc;
> 
> The patch I sent only replaces the second patch, the first one:
> 
> http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20180526/185eb219/attachment.obj <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20180526/185eb219/attachment.obj>

Thanks for the update. I continued testing and found this very useful, particularly with the side data.

Before I only tested with vitc but now have a serial cable connected as well and found a source tape that has distinct values for LTC and VITC timecodes. The LTC values are from 1:00:00 to 2:00:00 and the VITC values are from 07:00:00 - 08:00:00.

With the deckcontrol utility at https://github.com/bavc/deckcontrol <https://github.com/bavc/deckcontrol>, I can use the command gettimecode to grab the LTC value:

deckcontrol gettimecode
Issued command 'gettimecode'
TC=07:37:56:21
Command sucessfully issued
Error sending command (No error)

With these patches, I can only grab the vitc values:

for i in rp188vitc rp188vitc2 rp188ltc rp188any vitc vitc2 serial ; do echo -n "${i}: " ; ./ffprobe -v quiet -timecode_format "$i" -f decklink -draw_bars 0 -audio_input embedded -video_input sdi -format_code ntsc -channels 8 -raw_format yuv422p10 -i "UltraStudio Express" -select_streams v -show_entries stream_tags=timecode -of default=nw=1:nk=1 ; echo ; done
rp188vitc: 
rp188vitc2: 
rp188ltc: 
rp188any: 
vitc: 01:41:44;06
vitc2: 01:41:44;21
serial: 

Also it may be interesting in cases like this to support accepting multiple timecode inputs at once, such as "-timecode_format vitc+rp188ltc” though it would need to be contextualized more in metadata.

With a serial cable connected, I can access LTC via the deckcontrol utility but not with this patch.

./ffprobe -timecode_format rp188ltc -v debug -f decklink -draw_bars 0 -audio_input embedded -video_input sdi -format_code ntsc -channels 8 -raw_format yuv422p10 -i "UltraStudio Express"
ffprobe version N-91240-g3769aafb7c Copyright (c) 2007-2018 the FFmpeg developers
  built with Apple LLVM version 9.0.0 (clang-900.0.38)
  configuration: --enable-libfreetype --enable-nonfree --enable-decklink --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/include --disable-muxer=mxf --disable-demuxer=mxf
  libavutil      56. 18.102 / 56. 18.102
  libavcodec     58. 19.105 / 58. 19.105
  libavformat    58. 17.100 / 58. 17.100
  libavdevice    58.  4.100 / 58.  4.100
  libavfilter     7. 25.100 /  7. 25.100
  libswscale      5.  2.100 /  5.  2.100
  libswresample   3.  2.100 /  3.  2.100
[decklink @ 0x7fe2b8003000] Trying to find mode for frame size 0x0, frame timing 0/0, field order 0, direction 0, mode number 0, format code ntsc
[decklink @ 0x7fe2b8003000] Found Decklink mode 720 x 486 with rate 29.97(i)
[decklink @ 0x7fe2b8003000] Using 8 input audio channels
[decklink @ 0x7fe2b8003000] Unable to find timecode.
    Last message repeated 5 times
[decklink @ 0x7fe2b8003000] Probe buffer size limit of 5000000 bytes reached
Input #0, decklink, from 'UltraStudio Express':
  Duration: N/A, start: 0.000000, bitrate: 229869 kb/s
    Stream #0:0, 5, 1/1000000: Audio: pcm_s16le, 48000 Hz, 8 channels, s16, 6144 kb/s
    Stream #0:1, 6, 1/1000000: Video: v210, 1 reference frame (V210 / 0x30313256), yuv422p10le(bottom first), 720x486, 0/1, 223725 kb/s, 29.97 fps, 29.97 tbr, 1000k tbn, 1000k tbc

./ffprobe -timecode_format serial -v debug -f decklink -draw_bars 0 -audio_input embedded -video_input sdi -format_code ntsc -channels 8 -raw_format yuv422p10 -i "UltraStudio Express"
ffprobe version N-91240-g3769aafb7c Copyright (c) 2007-2018 the FFmpeg developers
  built with Apple LLVM version 9.0.0 (clang-900.0.38)
  configuration: --enable-libfreetype --enable-nonfree --enable-decklink --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/include --disable-muxer=mxf --disable-demuxer=mxf
  libavutil      56. 18.102 / 56. 18.102
  libavcodec     58. 19.105 / 58. 19.105
  libavformat    58. 17.100 / 58. 17.100
  libavdevice    58.  4.100 / 58.  4.100
  libavfilter     7. 25.100 /  7. 25.100
  libswscale      5.  2.100 /  5.  2.100
  libswresample   3.  2.100 /  3.  2.100
[decklink @ 0x7fb086003000] Trying to find mode for frame size 0x0, frame timing 0/0, field order 0, direction 0, mode number 0, format code ntsc
[decklink @ 0x7fb086003000] Found Decklink mode 720 x 486 with rate 29.97(i)
[decklink @ 0x7fb086003000] Using 8 input audio channels
[decklink @ 0x7fb086003000] Unable to find timecode.
    Last message repeated 5 times
[decklink @ 0x7fb086003000] Probe buffer size limit of 5000000 bytes reached
Input #0, decklink, from 'UltraStudio Express':
  Duration: N/A, start: 0.000000, bitrate: 229869 kb/s
    Stream #0:0, 5, 1/1000000: Audio: pcm_s16le, 48000 Hz, 8 channels, s16, 6144 kb/s
    Stream #0:1, 6, 1/1000000: Video: v210, 1 reference frame (V210 / 0x30313256), yuv422p10le(bottom first), 720x486, 0/1, 223725 kb/s, 29.97 fps, 29.97 tbr, 1000k tbn, 1000k tbc

Thanks,
Dave Rice