May 2017 - ffmpeg-user - lists.ffmpeg.org

Unable to set ffmpeg frei0r facedetect filter parameter
by Frank Urbainsky 30 May '17

30 May '17

Hi all you friendly people! I am trying to set the classifier cascade used for the facedetect frei0r filter in ffmpeg: |ffmpeg -i people.mp4 -strict -2 -vf '[in] frei0r=facedetect:/usr/share/opencv/haarcascades/haarcascade_frontalface_alt2.xml [out]' blurred.mp4| I made sure that the .xml file is actually at that path. But i get the weirdest error messages: |ERROR: Could not load classifier cascade H�I�>H��D| I also posted that question at superuser.com: https://superuser.com/questions/1214314/unable-to-set-ffmpeg-frei0r-facedet… because i am THAT desperate. :) Please help me! Frank

2 1

Slow motion setpts equation
by Jonathan Girven 30 May '17

30 May '17

I know that it is possible to apply slow motion to a section of video using a combination of splitting the video into the relevant sections, using setpts to adjust the rate and then concat'ing them back together. E.g: ffmpeg -i input.mp4 -filter_complex "[0:v]trim=0:2,setpts=expr=PTS-STARTPTS[v0];[0:v]trim=2:3,setpts=expr=PTS-STARTPTS[v1];[0:v]trim=3:5,setpts=expr=PTS-STARTPTS[v2];[v1]setpts=PTS/0.5[v1_slowmo];[v0][v1_slowmo][v2]concat=n=3:v=1:a=0[out]" -map [out] output.mp4 I wondered if it is also possible to do this with an equation in setpts (and might be easier to read)? E.g: ffmpeg -i input.mp4 -filter_complex "[0:v]setpts=if(between(T\, 2\, 3)\, 0.5\, 1)*PTS[out]" -map [out] output.mp4 This command doesn't adjust the times correctly though, I guess because each segment of time doesn't account for the previous shifts. Therefore t=0--2 and t=2-3 (with slowmo) actually end up overlapping in PTS. Is the equation I am going to have to write going to end up more complicated that my first example? / Just impossible. Full output: $ ffmpeg -y -i input.mp4 -filter_complex "[0:v]setpts=if(between(T\, 2\, 3)\, 0.5\, 1)*PTS[out]" -map [out] output.mp4 ffmpeg version 3.3 Copyright (c) 2000-2017 the FFmpeg developers built with gcc 4.9.2 (Debian 4.9.2-10) configuration: --prefix=/usr/local --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/lib --bindir=/usr/local/bin --disable-doc --disable-static --enable-shared --disable-ffplay --extra-libs=-ldl --enable-version3 --enable-libfreetype --enable-libx264 --enable-gpl --enable-openssl --enable-nonfree --disable-debug libavutil 55. 58.100 / 55. 58.100 libavcodec 57. 89.100 / 57. 89.100 libavformat 57. 71.100 / 57. 71.100 libavdevice 57. 6.100 / 57. 6.100 libavfilter 6. 82.100 / 6. 82.100 libswscale 4. 6.100 / 4. 6.100 libswresample 2. 7.100 / 2. 7.100 libpostproc 54. 5.100 / 54. 5.100 Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input.mp4': Metadata: major_brand : isom minor_version : 512 compatible_brands: isomiso2avc1mp41 encoder : Lavf57.19.100 Duration: 00:00:11.05, start: 0.000000, bitrate: 5584 kb/s Stream #0:0(und): Video: h264 (Constrained Baseline) (avc1 / 0x31637661), yuv420p, 1280x720 [SAR 1:1 DAR 16:9], 5473 kb/s, 25 fps, 25 tbr, 12800 tbn, 50 tbc (default) Metadata: handler_name : VideoHandler Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 128 kb/s (default) Metadata: handler_name : SoundHandler Stream mapping: Stream #0:0 (h264) -> setpts setpts -> Stream #0:0 (libx264) Press [q] to stop, [?] for help [libx264 @ 0xd89820] using SAR=1/1 [libx264 @ 0xd89820] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 AVX2 LZCNT BMI2 [libx264 @ 0xd89820] profile High, level 3.1 [libx264 @ 0xd89820] 264 - core 148 - H.264/MPEG-4 AVC codec - Copyleft 2003-2016 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00 Output #0, mp4, to 'output.mp4': Metadata: major_brand : isom minor_version : 512 compatible_brands: isomiso2avc1mp41 encoder : Lavf57.71.100 Stream #0:0: Video: h264 (libx264) ([33][0][0][0] / 0x0021), yuv420p, 1280x720 [SAR 1:1 DAR 16:9], q=-1--1, 25 fps, 12800 tbn, 25 tbc (default) Metadata: encoder : Lavc57.89.100 libx264 Side data: cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1 frame= 78 fps=0.0 q=28.0 size= 403kB time=00:00:00.76 bitrate=4346.3kbits/s dup=25 drop=25 speed= 1frame= 122 fps=112 q=28.0 size= 1085kB time=00:00:02.52 bitrate=3525.7kbits/s dup=25 drop=25 speed=2.frame= 158 fps= 99 q=28.0 size= 1568kB time=00:00:03.96 bitrate=3243.6kbits/s dup=25 drop=25 speed=2.frame= 191 fps= 91 q=28.0 size= 2210kB time=00:00:05.28 bitrate=3428.1kbits/s dup=25 drop=25 speed=2.frame= 224 fps= 86 q=28.0 size= 2859kB time=00:00:06.60 bitrate=3549.2kbits/s dup=25 drop=25 speed=2.frame= 254 fps= 81 q=28.0 size= 3402kB time=00:00:07.80 bitrate=3572.8kbits/s dup=25 drop=25 speed= 2frame= 275 fps= 68 q=-1.0 Lsize= 4877kB time=00:00:10.88 bitrate=3672.0kbits/s dup=25 drop=25 speed=2.68x video:4875kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.046016% [libx264 @ 0xd89820] frame I:4 Avg QP:18.22 size: 89324 [libx264 @ 0xd89820] frame P:250 Avg QP:23.73 size: 18427 [libx264 @ 0xd89820] frame B:21 Avg QP:27.84 size: 1288 [libx264 @ 0xd89820] consecutive B-frames: 89.1% 2.2% 0.0% 8.7% [libx264 @ 0xd89820] mb I I16..4: 45.8% 20.2% 34.1% [libx264 @ 0xd89820] mb P I16..4: 6.4% 1.2% 1.2% P16..4: 53.7% 8.4% 4.7% 0.0% 0.0% skip:24.4% [libx264 @ 0xd89820] mb B I16..4: 0.5% 0.1% 0.0% B16..8: 5.0% 0.7% 0.1% direct: 0.3% skip:93.2% L0:53.8% L1:32.6% BI:13.5% [libx264 @ 0xd89820] 8x8 transform intra:14.6% inter:37.3% [libx264 @ 0xd89820] coded y,uvDC,uvAC intra: 53.2% 65.0% 12.8% inter: 27.7% 19.0% 0.1% [libx264 @ 0xd89820] i16 v,h,dc,p: 10% 46% 34% 9% [libx264 @ 0xd89820] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 8% 55% 24% 2% 1% 1% 3% 1% 6% [libx264 @ 0xd89820] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 13% 51% 22% 2% 2% 1% 3% 1% 5% [libx264 @ 0xd89820] i8c dc,h,v,p: 43% 40% 9% 8% [libx264 @ 0xd89820] Weighted P-Frames: Y:0.4% UV:0.4% [libx264 @ 0xd89820] ref P L0: 83.6% 14.8% 1.3% 0.2% 0.0% [libx264 @ 0xd89820] ref B L0: 84.2% 15.6% 0.2% [libx264 @ 0xd89820] ref B L1: 99.9% 0.1% [libx264 @ 0xd89820] kb/s:3629.85

2 1

forced keyframes and RTP source
by jonsmirl＠gmail.com 30 May '17

30 May '17

Does this: '-force_key_frames "expr:gte(t,n_forced*10)"' cause RTCP FIR commands to be generated when using a network RTSP source? -- Jon Smirl jonsmirl(a)gmail.com

1 0

Read EIA-608 captions and re-mux into Mpegts as stream
by Tim DeNike 30 May '17

30 May '17

Im successfully transcoding a udp input MPEG2 TS stream into h264 with NVENC and multicasting it onto the network.. The last piece of the puzzle is to be able to read in the EIA-608 embedded CCs in the original video stream and re-mux into the output TS as a separate stream. I can't seem to figure out how to do it.. I can pull out the EIA-608 captions but not simultaneously add them as a stream in the final TS file while transcoding video and copying audio. Im sure there is some way to do it. I just can't figure out how. Here is my command line for the transcode without CC: /usr/local/bin/ffmpeg -threads auto -i "udp:// 10.98.98.251:50029?fifo_size=262144&buffer_size=262144&overrun_nonfatal=1" -ss 00:00:10 -vf yadif=0:-1:0 -s hd1080 -c:v h264_nvenc -level:v 4 -profile:v high -maxrate 10000k -bf 4 -preset:v llhp -rc-lookahead 250 -rc:v ll_2pass_size -c:a copy -packetsize 1316 -avioflags direct -max_delay 0 -f mpegts udp://239.33.35.201:1234 Version info: ffmpeg version 3.3.1 Copyright (c) 2000-2017 the FFmpeg developers built with gcc 4.8.5 (GCC) 20150623 (Red Hat 4.8.5-4) configuration: --extra-libs=-ldl --extra-cflags='-I/root/sources/nvenc/Video_Codec_SDK_7.0.1/Samples/common/inc -I/root/sources/bmsdk1054/Linux/include' --enable-gpl --enable-nonfree --enable-nvenc --enable-libfdk_aac --enable-libmp3lame --enable-libvorbis --enable-libx264 --disable-libtheora --pkg-config=pkg-config --cpu=host --enable-libzvbi --enable-libass --enable-decklink --enable-libnut --enable-openssl --disable-ffserver --enable-cuda --enable-cuvid libavutil 55. 58.100 / 55. 58.100 libavcodec 57. 89.100 / 57. 89.100 libavformat 57. 71.100 / 57. 71.100 libavdevice 57. 6.100 / 57. 6.100 libavfilter 6. 82.100 / 6. 82.100 libswscale 4. 6.100 / 4. 6.100 libswresample 2. 7.100 / 2. 7.100 libpostproc 54. 5.100 / 54. 5.100 Any help would be appreciated. Thanks!

4 9

Float formatting in drawtext
by Wolfgang Hugemann 29 May '17

29 May '17

Hey, seems to be an easy question, but I have been searching for the answer quite a while without success: If I perform a float calculation via text=%{expr\:whatever} is there any way to format the output the printf way? I know, the result could be converted to integer and be formatted via text=%{eif\:whatever\:d} but this is restricted to integer results. Wolfgang Hugemann

2 4

memory leak with vaapi decode since ffmpeg_vaapi: Convert to use hw_frames_ctx only
by Andy Furniss 29 May '17

29 May '17

First reported on mpv github issues. Since below ffmpeg -hwaccel vaapi -i sommething-h264 -f null - will steadily eat memory. commit 542a65d0b33abf81e5087ec9142bb11a54f23cde Author: Mark Thompson <sw(a)jkqxz.net> Date: Sun Jul 31 22:51:45 2016 +0100 ffmpeg_vaapi: Convert to use hw_frames_ctx only Most of the functionality here has moved into lavc. (cherry picked from commit 3e8651a7ccd8e89cc2f162cf614a3c9f7f4d9fcf)

3 4

how to pass output of lavfi to xv?
by Anatoly 29 May '17

29 May '17

I need to monitor audio levels while capturing audio/video. To achieve this I tried simple thing first: # ffmpeg -i Like\ an\ e-bird.mp3 -lavfi showvolume -f xv audiolevel And I've got: ... Stream mapping: Stream #0:0 (mp3) -> showvolume showvolume -> Stream #0:0 (rawvideo) Could not write header for output file #0 (incorrect codec parameters ?): Not yet implemented in FFmpeg, patches welcome At the same time I can play any video on xv device without problem, like this: # ffmpeg -i video.avi -f xv video ... Stream mapping: Stream #0:0 -> #0:0 (mpeg4 (native) -> rawvideo (native)) Press [q] to stop, [?] for help Also, I can encode output of lavfi to file: # ffmpeg -i Like\ an\ e-bird.mp3 -lavfi showvolume -c:v h264 -an volmon.mp4

2 2

ffmpeg does plays only one essence in quicktime reference conversion
by Dan Ionescu 28 May '17

28 May '17

Hello! I am trying to convert a quicktime reference mov file exported from Avid MC 8.4.5. I am using the following conversion command: ffmpeg -threads 0 -enable_drefs 1 -use_absolute_path 1 -i testseq.mov -strict -2 -pix_fmt yuv420p test.mp4 -y After hardly managing to get ffmpeg to reach all the referenced files, here is the problem. In the converted mp4 the cuts (timecodes where source media is changing in the sequence) gets through allright, but only on source clip is used as source. The other source clips are just ignored. Is this behavior "normal". That's it, this is all that is implemented so fat in ffmpeg or: 1. I sould look for an updated version of the software 2. I am missing something out 3. I am doing something wrong Here is the command with the output during conversion. After conversion no error messages are displayed. root@mediasync-ubuntu:/share# ffmpeg -threads 0 -enable_drefs 1 -use_absolute_path 1 -i test\ seq.mov -strict -2 -pix_fmt yuv420p test.mp4 -y ffmpeg version 2.8.11-0ubuntu0.16.04.1 Copyright (c) 2000-2017 the FFmpeg developers built with gcc 5.4.0 (Ubuntu 5.4.0-6ubuntu1~16.04.4) 20160609 configuration: --prefix=/usr --extra-version=0ubuntu0.16.04.1 --build-suffix=-ffmpeg --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --cc=cc --cxx=g++ --enable-gpl --enable-shared --disable-stripping --disable-decoder=libopenjpeg --disable-decoder=libschroedinger --enable-avresample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmodplug --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librtmp --enable-libschroedinger --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx265 --enable-libxvid --enable-libzvbi --enable-openal --enable-opengl --enable-x11grab --enable-libdc1394 --enable-libiec61883 --enable-libzmq --enable-frei0r --enable-libx264 --enable-libopencv libavutil 54. 31.100 / 54. 31.100 libavcodec 56. 60.100 / 56. 60.100 libavformat 56. 40.101 / 56. 40.101 libavdevice 56. 4.100 / 56. 4.100 libavfilter 5. 40.101 / 5. 40.101 libavresample 2. 1. 0 / 2. 1. 0 libswscale 3. 1.101 / 3. 1.101 libswresample 1. 2.101 / 1. 2.101 libpostproc 53. 3.100 / 53. 3.100 [mov,mp4,m4a,3gp,3g2,mj2 @ 0x1c56520] Using absolute path on user request, this is a possible security issue [dnxhd @ 0x1c59160] frame size changed: 1920x1080 -> 1440x1080 Guessed Channel Layout for Input Stream #0.0 : stereo Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'test seq.mov': Metadata: major_brand : qt minor_version : 537199360 compatible_brands: qt creation_time : 2017-05-25 19:04:15 comment : Source ID = 060a2b340101010101010f00-13-00-00-00-{59271bed-5179-0018-060e2b347f7f2a80} comment-eng : Source ID = 060a2b340101010101010f00-13-00-00-00-{59271bed-5179-0018-060e2b347f7f2a80} timecode : 01:00:00:00 Duration: 00:00:44.48, start: 0.000000, bitrate: 1 kb/s Stream #0:0(eng): Audio: pcm_s16be (twos / 0x736F7774), 48000 Hz, 2 channels, s16, 1536 kb/s (default) Metadata: creation_time : 2017-05-25 19:04:16 handler_name : Apple Alias Data Handler Stream #0:1(eng): Video: dnxhd (AVdn / 0x6E645641), yuv422p(pc), 1440x1080 [SAR 4:3 DAR 16:9], 83558 kb/s, 25 fps, 25 tbr, 2500 tbn, 2500 tbc (default) Metadata: creation_time : 2017-05-25 19:04:16 handler_name : Apple Alias Data Handler encoder : Avid DNxHD Codec Stream #0:2(eng): Data: none (tmcd / 0x64636D74) (default) Metadata: creation_time : 2017-05-25 19:04:16 handler_name : Apple Alias Data Handler timecode : 01:00:00:00 [libx264 @ 0x1c8b780] using SAR=4/3 [libx264 @ 0x1c8b780] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 [libx264 @ 0x1c8b780] profile High, level 4.0 [libx264 @ 0x1c8b780] 264 - core 148 r2643 5c65704 - H.264/MPEG-4 AVC codec - Copyleft 2003-2015 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=9 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00 Output #0, mp4, to 'test.mp4': Metadata: major_brand : qt minor_version : 537199360 compatible_brands: qt timecode : 01:00:00:00 comment : Source ID = 060a2b340101010101010f00-13-00-00-00-{59271bed-5179-0018-060e2b347f7f2a80} comment-eng : Source ID = 060a2b340101010101010f00-13-00-00-00-{59271bed-5179-0018-060e2b347f7f2a80} encoder : Lavf56.40.101 Stream #0:0(eng): Video: h264 (libx264) ([33][0][0][0] / 0x0021), yuv420p, 1440x1080 [SAR 4:3 DAR 16:9], q=-1--1, 25 fps, 12800 tbn, 25 tbc (default) Metadata: creation_time : 2017-05-25 19:04:16 handler_name : Apple Alias Data Handler encoder : Lavc56.60.100 libx264 Stream #0:1(eng): Audio: aac ([64][0][0][0] / 0x0040), 48000 Hz, stereo, fltp, 128 kb/s (default) Metadata: creation_time : 2017-05-25 19:04:16 handler_name : Apple Alias Data Handler encoder : Lavc56.60.100 aac Stream mapping: Stream #0:1 -> #0:0 (dnxhd (native) -> h264 (libx264)) Stream #0:0 -> #0:1 (pcm_s16be (native) -> aac (native)) Press [q] to stop, [?] for help frame= 144 fps= 28 q=-1.0 Lsize= 1701kB time=00:00:05.95 bitrate=2340.9kbits/s video:1602kB audio:94kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.325520% [libx264 @ 0x1c8b780] frame I:3 Avg QP:20.55 size: 42916 [libx264 @ 0x1c8b780] frame P:75 Avg QP:21.30 size: 16263 [libx264 @ 0x1c8b780] frame B:66 Avg QP:23.50 size: 4410 [libx264 @ 0x1c8b780] consecutive B-frames: 23.6% 41.7% 12.5% 22.2% [libx264 @ 0x1c8b780] mb I I16..4: 41.7% 48.0% 10.2% [libx264 @ 0x1c8b780] mb P I16..4: 7.5% 12.9% 0.5% P16..4: 27.9% 6.1% 2.3% 0.0% 0.0% skip:42.7% [libx264 @ 0x1c8b780] mb B I16..4: 0.2% 0.5% 0.0% B16..8: 27.0% 2.1% 0.3% direct: 0.7% skip:69.1% L0:36.0% L1:59.6% BI: 4.4% [libx264 @ 0x1c8b780] 8x8 transform intra:59.6% inter:85.6% [libx264 @ 0x1c8b780] coded y,uvDC,uvAC intra: 37.0% 19.8% 0.6% inter: 9.3% 5.5% 0.0% [libx264 @ 0x1c8b780] i16 v,h,dc,p: 17% 58% 6% 19% [libx264 @ 0x1c8b780] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 24% 24% 34% 3% 3% 3% 3% 3% 4% [libx264 @ 0x1c8b780] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 11% 66% 9% 2% 2% 2% 2% 2% 3% [libx264 @ 0x1c8b780] i8c dc,h,v,p: 72% 16% 11% 1% [libx264 @ 0x1c8b780] Weighted P-Frames: Y:2.7% UV:1.3% [libx264 @ 0x1c8b780] ref P L0: 67.4% 15.4% 12.7% 4.5% 0.0% [libx264 @ 0x1c8b780] ref B L0: 91.0% 8.4% 0.6% [libx264 @ 0x1c8b780] ref B L1: 97.8% 2.2% [libx264 @ 0x1c8b780] kb/s:2277.16 Exiting normally, received signal 2. Thank you so much for reading. Any help will be greatly appreciated. BR++, Dan

2 1

Android compile with --enable-jni is using Objective C
by jonsmirl＠gmail.com 28 May '17

28 May '17

This configure test can't be right on an Android build..... I'm trying to turn on MediaCodec support. aarch64-linux-android-gcc -Werror=missing-prototypes --sysroot=/home/jonsmirl/arm64/sysroot -isysroot /home/jonsmirl/arm64/sysroot -D_ISOC99_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -Dstrtod=avpriv_strtod -DPIC -fpic -mcpu=cortex-a53 -std=c11 -fomit-frame-pointer -fPIC -pthread -c -o /tmp/ffconf.Nj0qd3Ec/test.o /tmp/ffconf.Nj0qd3Ec/test.m aarch64-linux-android-gcc: error: /tmp/ffconf.Nj0qd3Ec/test.m: Objective-C compiler not installed on this system ERROR: jni not found -- Jon Smirl jonsmirl(a)gmail.com

1 2

Determining how FFMPEG calculates start time for MPEG-2 files
by Harold Tessmann 26 May '17

26 May '17

Hi all, I'm trying to understand how ffmpeg calculates the start time for an MPEG-2 file containing one video track and one audio track. For example, when I run ffprobe, I get this output: Input #0, mpeg, from 'file.mpg': Duration: 00:00:11.06, start: 579.376000, bitrate: 10704 kb/s As a guess, I looked at the system clock reference in the first pack header and the PTS/DTS values in the first PES header. When I calculate those values by hand, or examine the same file in Wireshark, I get a value that differs from the one calculated by ffmpeg. The first pack header gives me a system clock reference of 578.4, while the first video PES header gives PTS = DTS = 579.4. If I look at the first audio PES header, I get 579.375655555556, which is so close that it may be just a matter of doing math in the wrong mode (integer vs. floating-point). From looking at those values, I would guess that ffmpeg takes the minimum value from all the PES headers, but I'm not confident about that. I see in av_dump_format() the lines that calculate & output seconds (ic->start_time / AV_TIME_BASE) and microseconds (% instead of /), and am working backwards from there. Any information or pointers on where to look would be appreciated. Thanks, Harold

2 1