<div dir="auto"><div><br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">вт, 17 сент. 2024 г., 15:12 Stefan de Konink via Cin <<a href="mailto:cin@lists.cinelerra-gg.org" rel="noreferrer noreferrer noreferrer" target="_blank">cin@lists.cinelerra-gg.org</a>>:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">There are quite some topics online mentioning the when MTS files <br>
(exceeding 2GB) are naively added to a timeline. The problem that then <br>
appears is that the audio is typically "shorter" than the video, and it <br>
does not appear in the interface. Cinelerra will even shift the audio <br>
track under the previous track.<br></blockquote></div></div><div dir="auto"><br></div><div dir="auto">can you post example for the curious? :)</div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
It is mentioned that Adobe Premiere is able to parse an AVCHD structure, <br>
including the ability to use meta information such as timecode, which is <br>
not available from MTS files directly.<br>
<br>
Other workarounds that are mentioned to first join the MTS files. FFmpeg <br>
has the option concat demuxer and concat protocol. But it pulls another <br>
surprise...<br>
<br>
<br>
ffmpeg -i "concat:00152.MTS|00153.MTS|00154.MTS" -vcodec copy -acodec <br>
copy test.ts<br>
ffmpeg version n7.0.1 Copyright (c) 2000-2024 the FFmpeg developers<br>
built with gcc 14.1.1 (GCC) 20240522<br>
configuration: --prefix=/usr --disable-debug --disable-static <br>
--disable-stripping --enable-amf --enable-avisynth --enable-cuda-llvm <br>
--enable-lto --enable-fontconfig --enable-frei0r --enable-gmp <br>
--enable-gpl --enable-ladspa --enable-libaom --enable-libass <br>
--enable-libbluray --enable-libbs2b --enable-libdav1d --enable-libdrm <br>
--enable-libdvdnav --enable-libdvdread --enable-libfreetype <br>
--enable-libfribidi --enable-libgsm --enable-libharfbuzz <br>
--enable-libiec61883 --enable-libjack --enable-libjxl <br>
--enable-libmodplug --enable-libmp3lame --enable-libopencore_amrnb <br>
--enable-libopencore_amrwb --enable-libopenjpeg --enable-libopenmpt <br>
--enable-libopus --enable-libplacebo --enable-libpulse --enable-librav1e <br>
--enable-librsvg --enable-librubberband --enable-libsnappy <br>
--enable-libsoxr --enable-libspeex --enable-libsrt --enable-libssh <br>
--enable-libsvtav1 --enable-libtheora --enable-libv4l2 <br>
--enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpl <br>
--enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 <br>
--enable-libxcb --enable-libxml2 --enable-libxvid --enable-libzimg <br>
--enable-mbedtls --enable-nvdec --enable-nvenc --enable-opencl <br>
--enable-opengl --enable-shared --enable-vapoursynth --enable-version3 <br>
--enable-vulkan<br>
libavutil 59. 8.100 / 59. 8.100<br>
libavcodec 61. 3.100 / 61. 3.100<br>
libavformat 61. 1.100 / 61. 1.100<br>
libavdevice 61. 1.100 / 61. 1.100<br>
libavfilter 10. 1.100 / 10. 1.100<br>
libswscale 8. 1.100 / 8. 1.100<br>
libswresample 5. 1.100 / 5. 1.100<br>
libpostproc 58. 1.100 / 58. 1.100<br>
Input #0, mpegts, from 'concat:00152.MTS|00153.MTS|00154.MTS':<br>
Duration: 00:25:04.36, start: 1.040000, bitrate: 24022 kb/s<br>
Program 1<br>
Stream #0:0[0x1011]: Video: h264 (High) (HDMV / 0x564D4448), <br>
yuv420p(top first), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn<br>
Stream #0:1[0x1100]: Audio: pcm_bluray (HDMV / 0x564D4448), 48000 Hz, <br>
stereo, s16, 1536 kb/s<br>
Stream #0:2[0x1200]: Subtitle: hdmv_pgs_subtitle (pgssub) <br>
([144][0][0][0] / 0x0090), 1920x1080<br>
File 'test.ts' already exists. Overwrite? [y/N] y<br>
Stream mapping:<br>
Stream #0:0 -> #0:0 (copy)<br>
Stream #0:1 -> #0:1 (copy)<br>
Output #0, mpegts, to 'test.ts':<br>
Metadata:<br>
encoder : Lavf61.1.100<br>
Stream #0:0: Video: h264 (High) (HDMV / 0x564D4448), yuv420p(top <br>
first), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 25 fps, 25 tbr, 90k tbn<br>
Stream #0:1: Audio: pcm_bluray (HDMV / 0x564D4448), 48000 Hz, stereo, <br>
s16, 1536 kb/s<br>
Press [q] to stop, [?] for help<br>
[mpegts @ 0x600ed54cf540] Stream 1, codec pcm_bluray, is muxed as a <br>
private data stream and may not be recognized upon reading.<br>
<br>
This results in a video file cinelerra (and others) cannot read.<br>
<br>
Instead if we would say do it in pcm_sb16le we would be able to show <br>
that the MTS file does contain the audio and FFmpeg is able to read it. <br>
I have attached a screenshot showing what happens.<br>
<br>
I wonder: could we in Cinelerra implement something that would not open <br>
a file as single file or EDL, but instead a concat set of files?<br></blockquote></div></div><div dir="auto"><br></div><div dir="auto">hm, look for avformat_open_input in ffmpeg.C</div><div dir="auto"><br></div><div dir="auto">according to docs you can set 3rd argument there for specific demuxer:</div><div dir="auto"><br></div><div dir="auto">"@param fmt If non-NULL, this parameter forces a specific input format.</div><div dir="auto"><br></div><div dir="auto"><div dir="auto">int avformat_open_input(AVFormatContext **ps, const char *url,</div><div dir="auto"> const AVInputFormat *fmt, AVDictionary **options);</div></div><div dir="auto"><br></div><div dir="auto">libavformat/avformat.h lines 2240-2274<br></div><div dir="auto"><br></div><div dir="auto">but I am not sure if cingg can copy concat'd list of files from file picker? For testing you probably can just hardcode input format there and manually enter string of filenames? If this works we probably can think about how to GUI it</div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Something in the drop down: "Create a single new resource only". That <br>
single resource should be opened via the concat protocol, and then be <br>
able as single instance. An EDL, but then at protocol level.<br>
<br>
I think this would make sense for many recordings that are split due to <br>
filesystem limits, but even for image sequences, if they could be <br>
directly opened and edited as if they were single files.<br>
<br>
-- <br>
Stefan<br>
-- <br>
Cin mailing list<br>
<a href="mailto:Cin@lists.cinelerra-gg.org" rel="noreferrer noreferrer noreferrer noreferrer" target="_blank">Cin@lists.cinelerra-gg.org</a><br>
<a href="https://lists.cinelerra-gg.org/mailman/listinfo/cin" rel="noreferrer noreferrer noreferrer noreferrer noreferrer" target="_blank">https://lists.cinelerra-gg.org/mailman/listinfo/cin</a><br>
</blockquote></div></div></div>