May 2025 - ffmpeg-user - lists.ffmpeg.org

Would it be possible to load xattr-related media-files like a container?
by Peter B. 02 Aug '25

02 Aug '25

Hi everyone :) I'm working a lot with extended attributes on filesystems, and I'm amazed! I was wondering if it would be possible for ffmpeg to load "related media-streams" as related objects: Having a 0-Byte "videofile.aha": * video = /path/to/my_movie.h264 * audio = ./my_movie-DE_AT.flac * audio.1 = my_movie-EN_GB.mp3 * video.default = video * audio.default = audio.1 * dc.title = "My Movie" * dc.creator = "..." * ... Like an input demuxer which resolves the related objects and treats this like loading a container? I imagine this to be quite fun (and useful actually). One could then drag-and-drop to remux tracks, and right-click-edit metadata, I imagine. Feedback greatly appreciated :) Have a great day! Peter

1 1

Re: [FFmpeg-user] [Cin] FranceBB's LUT collection
by Andrew Randrianasulu 01 Jun '25

01 Jun '25

пт, 16 июл. 2021 г., 09:36 Andrea paz via Cin <cin(a)lists.cinelerra-gg.org>: > I made a forum post to keep track of the repository. So if a request > is made we can direct to that post. > > > https://www.cinelerra-gg.org/forum/help-video/francebb-luts-for-space-color… It was updated yesterday with interesting message: https://forum.doom9.org/showpost.php?p=2018552&postcount=310 ==== quotation start === This feature was requested by ossnorry. As a bit of history, in 1953 the NTSC version of BT601 was born, however the tube phosphors originally used by the TVs of the time had issues displaying those values and the standard was later changed (reduced) in 1987 to make life easier for the hardware of the time, however one country never stopped using the original NTSC version and that country is Japan, which is why the original NTSC version has become known as "NTSC-J" which stands for "Japan". Anyway, the old BT601 NTSC Japan (1953) is based on the following values: x white = 0.3100 y white = 0.3160 x red = 0.670 y red = 0.330 x green = 0.210 y green = 0.710 x blue = 0.140 y blue = 0.080 and is indicated with --colormatrix smpte170m --transfer bt470m --colorprim bt470m while the normal BT601 NTSC SMPTE C (1987) that all other NTSC countries adopted has the following values: x white = 0.3127 y white = 0.3290 x red = 0.630 y red = 0.340 x green = 0.310 y green = 0.595 x blue = 0.155 y blue = 0.070 and is indicated with: --colormatrix smpte170m --transfer smpte170m --colorprim smpte170m As you can see, all values are different, including the white point. === end of quotation ==== I surely recall Japan was different, but it completely evaded my mind that old NTSC 1953 was still in use there up to relatively modern times! cc ffmpeg-users because it tend to be more popular list, so visibility will be higher. > -- > Cin mailing list > Cin(a)lists.cinelerra-gg.org > https://lists.cinelerra-gg.org/mailman/listinfo/cin >

6 38

old version of ffmpeg
by Harald Müller-Dempf 30 May '25

30 May '25

How and where can I download an FFMPEG app which works with Mac OS 10.11.6 ? (I had it before but lost it.) Thank you very much ! —————————————— Dr. Harald Müller-Dempf Fraenkelufer 50 10999 Berlin harald.mueller-dempf(a)gmx.de www.harald-mueller-dempf.de

4 8

Seeking a more detail explanation of err_detect options
by Grilled Tanuki 30 May '25

30 May '25

Hi all, I am writing some software to parse through a repository of videos to detect encoding errors that would result in a visual or audible glitch. I'm not concerned about minor issues that a decoder can handle gracefully. I have been using the following command. FFMPEG_PATH -xerror -err_detect ERROR_DETECT -loglevel error -hwaccel cuda -hwaccel_output_format cuda -i FILE_PATH -f null - The documentation for err_detect lists the following options. crccheck - verify embedded CRCs bitstream - detect bitstream specification deviations buffer - detect improper bitstream length explode - abort decoding on minor error detection ignore_err - ignore decoding errors, and continue decoding. careful - consider things that violate the spec and have not been seen in the wild as errors compliant - consider all spec non compliances as errors aggressive - consider things that a sane encoder should not do as an error I have been using explode, however this fails on inconsequential errors where the video still plays through fine. I want to switch to an option that should only flag noticeable glitches, but the documentation on the options is brief and vague to me. Does anyone have any insight on what these options do and could give me a more detailed description? Thanks

2 1

Detelecine: I got the right result from the wrong command?
by Alex Xu 30 May '25

30 May '25

Hi everyone, I'm working to detelecine a video, attached here (12 MB, first 32 seconds truncated): https://www.mediafire.com/file/x4ly6moqb8ayfyy/detelecine_input.mkv/file I wasn't sure what was the right command to get the result I wanted so I looped over all possible combos like so: for i in 0 1 2 3 4; do for j in 23 2332; do ffmpeg -i detelecine_input.mkv -c:v libx264 -crf 18 -c:a aac -b:a 192k -pix_fmt yuv420p -vf "detelecine=start_frame=$i:pattern=$j" $i$j.mkv; done done It was very clear which option was correct, since the correct option encoded the best and had a significantly smaller file size. I visually verified it too. In the source, every set of 5 frames had 3 clear frames and 2 interleaved frames. In the final output, it preserved those 3 clear frames and combines the 2 interleaved frames into 1, going from 30i to 24p. But I'm just not sure why the correct result was start_frame=3:pattern=2332. Examining the video, it seems pretty clear to me the pattern should be 23. Since it has two interleaved frames. I learned about 23 from https://en.wikipedia.org/wiki/Telecine#2:3_pulldown Could someone explain the gap in my knowledge? Alex

7 31

Why does the 'detelecine' filter exist?
by Mark Filipak 29 May '25

29 May '25

Why does the 'detelecine' filter exist? It is weak and clumsy. The 'fieldmatch' filter followed by 'decimate' do a better job and are robust. By continuing to carry obsolete filters, FFmpeg emasculates itself. That makes FFmpeg appear to be chaotic and ties up names that could be, should be reused to better effect. The same could be written about many FFmpeg filters. Dump them, or create a "Legacy" category in the documentation and put them there. Also, include documentation in "Legacy" that shows how to replace the obsolete filters in existing scripts. Promote the best filters, not the worst.

6 27

Conversion and Down sampling of ".Wav" to ".mp3"
by Connor Grooms 28 May '25

28 May '25

Hello, I am currently working on converting a large number of files from wav (Max freq 64000 Hz) to mp3 (max Freq 48000 Hz). My current call is: ffmpeg -i "INPUTFILENAME.WAV" -af "aresample=resampler=soxr:precision=33" -q:a 2 -ar 48000 -first_pts 0 "OUTPUTFILENAME.mp3" Also without the "first_pts 0" call I was noticing that there was a taper of the waveform amplitude at the start and end of the output files. I want to use the full dynamic range of the file and each file has a varying maximum sound levels. I have investigated ffmpeg-normalize however it seem to cause artifacts in the mp3 files over the last 3 seconds where the file is repetitively clipped in 0.1 second intervals. Normalizing a file to the maximum recorded level seems like a trivial thing to do with ffmpeg... am I missing something obvious. Any ideas or assistance are welcome.

2 1

[FFmpeg] about conversion of movies
by p db 26 May '25

26 May '25

Dear FFmpeg, Imagine that you have provided to convert a movie in *.mkv; *.ogv; *.webm; there is not a a log after conversion that says, how was converted the movie, if there are errors in the movie, why it stopped during the conversion... and the possibility to type commands on the shell to work on the movie and to know if the movie was fine converted. Awaiting your reply, Best Regards, Paolo Del Bene

3 2

Benefits from Chroma Upsampling v4l2 8bit 422 to 10bit 422 FFV1 backup?
by Terje J. Hanssen 25 May '25

25 May '25

My first backup generation of my PAL Analog S-video footages on tapes was digitized to SD 576i50 DV25 with 8-bit 4:2:0 color depth and 25 Mb/s video bit rate. IMO playback from DV25 backup files visually lack some brightness and colors compared with native playback from S-video tapes. Now I am testing another backup method with purpose of best practical video quality using ffmpeg to encode lossless compressed FFV1.mkv files: Analog video camera/player --> S-video/HDMI converter --> HDMI/USB3 capture card --> v4l2 --> ffmpeg/ FFV1.mkv 10-Bit 4:2:2 Video is recommended for Analog Tape Transfer due to color shades and branding https://www.archivalworks.com/blog/10-bit-video-transfer Well, my ms2130 HDMI/USB3 video capture card achieves YUYV2 or 8-bit 4:2:2, which is not 10-bit but yet twice as much color info compared with DV25 4:2:0. Q1 Therefore my question here if it yet may be Benefits from Chroma "Upsampling" v4l2 8bit 422 to 10bit 422 FFV1 backup? While some say this is fake upsampling like filling 8-bit data in a 10-bit container, other say this may be beneficial with temporal noise filtering https://www.eoshd.com/comments/topic/26710-8bit-%E2%86%92-10bit-video-with-… Q2 So let me hear if this may be applied with some ffmpeg denoising filters like time base correcting, tape grain or snow? ms2130 seemingly doesn't support interlaced video (25fps), so I have used 50fps for PAL. v4l2-ctl -V Format Video Capture: Width/Height : 1920/1080 Pixel Format : 'YUYV' (YUYV 4:2:2) Field : None Bytes per Line : 3840 Size Image : 4147200 Colorspace : sRGB Transfer Function : Rec. 709 YCbCr/HSV Encoding: ITU-R 601 Quantization : Default (maps to Limited Range) Flags : ffmpeg -hide_banner -f v4l2 -list_formats all -i /dev/video0 [video4linux2,v4l2 @ 0x556f439fc7c0] Raw : yuyv422 : YUYV 4:2:2 : 1920x1080 1600x1200 1360x768 1280x1024 1280x960 1280x720 1024x768 800x600 720x576 720x480 640x480 [in#0 @ 0x556f439fc500] Error opening input: Immediate exit requested Error opening input file /dev/video0. As seen I have tested four cases with different input and output formats in the table below. Q3 I also wonder why the Video bitrate differs so much for the 10-bit output files? Case Input format Video size Output format Bit depth/YUV Video bitrate ---- ------------ ---------- ------------- ------------- ------------- 1) - 720x576 - 8 bits 422p 55.6 Mb/s 2) - 720x576 yuv422p10 10 bits 422 68.0 Mb/s 3) yuv422p10 720x576 yuv422p10 10 bits 422 32.0 Mb/s 4) y210 720x576 y210 10 bits 422 35.8 Mb/s Attach ffmpeg output for Case 2) below: ffmpeg -hide_banner -f v4l2 -video_size 720x576 -framerate 50 -i /dev/video0 -c:v ffv1 -level 3 -vf format=pix_fmts='yuv422p10' -t 10 output.mkv Input #0, video4linux2,v4l2, from '/dev/video0': Duration: N/A, start: 239.890699, bitrate: 331776 kb/s Stream #0:0: Video: rawvideo (YUY2 / 0x32595559), yuyv422, 720x576, 331776 kb/s, 50 fps, 50 tbr, 1000k tbn Stream mapping: Stream #0:0 -> #0:0 (rawvideo (native) -> ffv1 (native)) Press [q] to stop, [?] for help [ffv1 @ 0x5571b7d55280] bits_per_raw_sample > 8, forcing range coder Output #0, matroska, to 'output.mkv': Metadata: encoder : Lavf61.7.100 Stream #0:0: Video: ffv1 (FFV1 / 0x31564646), yuv422p10le(tv, progressive), 720x576, q=2-31, 200 kb/s, 50 fps, 1k tbn Metadata: encoder : Lavc61.19.101 ffv1 [out#0/matroska @ 0x5571b7d56d00] video:84659KiB audio:0KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: 0.007280% frame= 500 fps= 50 q=-0.0 Lsize= 84665KiB time=00:00:10.00 bitrate=69357.5kbits/s speed=0.998x

4 7

Eclipsa
by Martin Broerse | Martinic 25 May '25

25 May '25

Hi, This is my first post here. We are working on an audio plugin that can be used with the new YouTube Eclipsa. If you want to create a 7.1.4 Eclipsa file with FFMPEG there are instructions here: https://github.com/AOMediaCodec/iamf-tools/blob/main/docs/external/encoding… that use standard SMPTE channel order ‘7.1.4’ FL+FR+FC+LFE+SL+SR+BL+BR+TFL+TFR+TBL+TBR for the audio input. FFMPEG uses a channel order I have not seen before. See: https://ffmpeg.org/ffmpeg-utils.html‘ 7.1.4’ FL+FR+FC+LFE+BL+BR+SL+SR+TFL+TFR+TBL+TBR I know FILM and SMTPE are standards but on what standard is the FFMPEG channel order based? Regards, Martin Broerse Martinic

1 0