March 2018 - ffmpeg-user - lists.ffmpeg.org

Effects of -bufsize
by Dan Norton 25 Mar '18

25 Mar '18

During stream copy, if inputs and output are on the same disk, I notice that although speed starts high, it eventually settles to a much lower value. Does the value specified for -bufsize have a copying speed benefit if input and output are on the same disk? - Dan

1 0

concat Missing part2.mov
by Dan Norton 25 Mar '18

25 Mar '18

I have two .mov files which were made on Debian 9 by running the following: $ ffmpeg -y -i "Norton 8mm Reel 04.mov" -ss 00:00:00 -to 00:20:00 -c copy -bufsize 2048K part1.mov $ ffmpeg -y -ss 00:21:12 -i "Norton 8mm Reel 04.mov" -c copy -bufsize 2048K part2.mov Viewed individually part1 and part2 are OK. When I tried to concatenate them, the output is missing part2. Here's what I did: $ ffmpeg -i "concat:part1.mov|part2.mov" -c copy out.mov ffmpeg version 3.2.10-1~deb9u1 Copyright (c) 2000-2018 the FFmpeg developers built with gcc 6.3.0 (Debian 6.3.0-18) 20170516 configuration: --prefix=/usr --extra-version='1~deb9u1' --toolchain=hardened --libdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --disable-stripping --enable-avresample --enable-avisynth --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libebur128 --enable-libflite --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmp3lame --enable-libopenjpeg --enable-libopenmpt --enable-libopus --enable-libpulse --enable-librubberband --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-libzmq --enable-libzvbi --enable-omx --enable-openal --enable-opengl --enable-sdl2 --enable-libdc1394 --enable-libiec61883 --enable-chromaprint --enable-frei0r --enable-libopencv --enable-libx264 --enable-shared libavutil 55. 34.101 / 55. 34.101 libavcodec 57. 64.101 / 57. 64.101 libavformat 57. 56.101 / 57. 56.101 libavdevice 57. 1.100 / 57. 1.100 libavfilter 6. 65.100 / 6. 65.100 libavresample 3. 1. 0 / 3. 1. 0 libswscale 4. 2.100 / 4. 2.100 libswresample 2. 3.100 / 2. 3.100 libpostproc 54. 1.100 / 54. 1.100 [mov,mp4,m4a,3gp,3g2,mj2 @ 0x55b324506d80] Found duplicated MOOV Atom. Skipped it Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'concat:part1.mov|part2.mov': Metadata: encoder : Lavf57.56.101 major_brand : qt minor_version : 512 compatible_brands: qt Duration: 00:20:00.00, start: 0.000000, bitrate: 313526 kb/s Stream #0:0(eng): Video: prores (apch / 0x68637061), yuv422p10le(top coded first (swapped)), 1920x1080, 205645 kb/s, SAR 1:1 DAR 16:9, 29.97 fps, 29.97 tbr, 11988 tbn, 11988 tbc (default) Metadata: handler_name : DataHandler encoder : Apple ProRes 422 (HQ) timecode : 00:00:00;00 Stream #0:1(eng): Data: none (tmcd / 0x64636D74) Metadata: handler_name : DataHandler timecode : 00:00:00;00 File 'out.mov' already exists. Overwrite ? [y/N] y Output #0, mov, to 'out.mov': Metadata: compatible_brands: qt major_brand : qt minor_version : 512 encoder : Lavf57.56.101 Stream #0:0(eng): Video: prores (apch / 0x68637061), yuv422p10le(top coded first (swapped)), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 205645 kb/s, 29.97 fps, 29.97 tbr, 11988 tbn, 11988 tbc (default) Metadata: handler_name : DataHandler encoder : Apple ProRes 422 (HQ) timecode : 00:00:00;00 Stream mapping: Stream #0:0 -> #0:0 (copy) Press [q] to stop, [?] for help frame=35964 fps= 57 q=-1.0 Lsize=30124305kB time=00:19:59.96 bitrate=205654.3kbits/s speed=1.91x video:30123882kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.001403%

2 2

Re: [FFmpeg-user] How to extract DVD subtitle from this VOB file?
by Farhad Mohammadi Majd 25 Mar '18

25 Mar '18

at 2018 Mar 15 15:42:32, Adi Marvillo (adi5 at gmx.at) wrote: > This is your command: > *ffmpeg -i ./dump.stream -g 25 -keyint_min 3 -vf -aspect "16:9" -map > v > -c:v libx264 -map a -c:a copy ./output.mkv > *ffprobe doesn't show a subtitle stream, don't know where it is... as > much as I know subtitles are streams as well. Let us know when you > find > out.CU Thanks for trying to help me but this is the second time your command is wrong, some options are wrong, I don't known what options, anyway the important issue is that I don't want to transcode it. at 2018 Mar 8 22:49:17, Nicolas George (george at nsup.org) wrote: > FFmpeg can read them, but it cannot put them into anything useful at > this time. MEncoder can extract them. thanks, I searched the WWW and it seems there are many documentation about it, I will try to do that later. good luck and bye

1 0

unable to view this video as one screen
by JD 25 Mar '18

25 Mar '18

https://www.youtube.com/watch?v=eqboujtUk4Y My browser shows 3 parallel views instead of one, of the same "picture". can ffmpeg do something so I can use ffplay to view it as one "picture" ?

2 1

nvenc on GT1030
by Piergiorgio Sartor 24 Mar '18

24 Mar '18

Hi all, I've recently got a Nvidia GT1030 graphic card, which should be Pascal generation, supporting HEVC encoding. The system is Fedora 27, up to date, with official Nvidia driver and CUDA libraries form rpmfusion repository. Also ffmpeg is from there. Now, I'm trying to encode an h264 video into HEVC, but ffmpeg complains the device is not nvenc capable. Command line: ffmpeg -loglevel 40 -i input.mp4 -c:a copy -c:v hevc_nvenc -profile:v main -preset slow -y test.mp4 Result: --- --- --- ffmpeg version 3.3.6 Copyright (c) 2000-2017 the FFmpeg developers built with gcc 7 (GCC) configuration: --prefix=/usr --bindir=/usr/bin --datadir=/usr/share/ffmpeg --docdir=/usr/share/doc/ffmpeg --incdir=/usr/include/ffmpeg --libdir=/usr/lib64 --mandir=/usr/share/man --arch=x86_64 --optflags='-O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic' --extra-ldflags='-Wl,-z,relro -specs=/usr/lib/rpm/redhat/redhat-hardened-ld ' --extra-cflags='-I/usr/include/nvenc ' --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libvo-amrwbenc --enable-version3 --enable-bzlib --disable-crystalhd --enable-fontconfig --enable-frei0r --enable-gcrypt --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libcdio --enable-indev=jack --enable-libfreetype --enable-libfribidi --enable-libgsm --enable-libmp3lame --enable-nvenc --enable-openal --enable-opencl --enable-opengl --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-libschroedinger --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libvorbis --enable-libv4l2 --enable-libvidstab --enable-libvpx --enable-libx264 --enable-libx265 --enable-libxvid --enable-avfilter --enable-avresample --enable-postproc --enable-pthreads --disable-static --enable-shared --enable-gpl --disable-debug --disable-stripping --shlibdir=/usr/lib64 --enable-libmfx --enable-runtime-cpudetect 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 libavresample 3. 5. 0 / 3. 5. 0 libswscale 4. 6.100 / 4. 6.100 libswresample 2. 7.100 / 2. 7.100 libpostproc 54. 5.100 / 54. 5.100 [h264 @ 0x556c2f685a80] Reinit context to 720x400, pix_fmt: yuv420p Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input.mp4': Metadata: major_brand : isom minor_version : 512 compatible_brands: isomiso2avc1mp41 creation_time : 2018-03-10T08:16:57.000000Z encoder : Lavf57.83.100 Duration: 00:28:06.40, start: 0.000000, bitrate: 1131 kb/s Stream #0:0(und): Video: h264 (High), 1 reference frame (avc1 / 0x31637661), yuv420p(left), 720x400 [SAR 80:81 DAR 16:9], 995 kb/s, 30 fps, 30 tbr, 15360 tbn, 60 tbc (default) Metadata: creation_time : 2018-03-10T08:16:57.000000Z handler_name : VideoHandler Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 127 kb/s (default) Metadata: creation_time : 2018-03-10T08:16:57.000000Z handler_name : SoundHandler Stream mapping: Stream #0:0 -> #0:0 (h264 (native) -> hevc (hevc_nvenc)) Stream #0:1 -> #0:1 (copy) Press [q] to stop, [?] for help [h264 @ 0x556c2f7eee20] Reinit context to 720x400, pix_fmt: yuv420p [graph 0 input from stream 0:0 @ 0x556c2f7f3500] w:720 h:400 pixfmt:yuv420p tb:1/15360 fr:30/1 sar:80/81 sws_param:flags=2 [hevc_nvenc @ 0x556c2f967560] Loaded Nvenc version 8.1 [hevc_nvenc @ 0x556c2f967560] Nvenc initialized successfully [hevc_nvenc @ 0x556c2f967560] 1 CUDA capable devices found [hevc_nvenc @ 0x556c2f967560] [ GPU #0 - < GeForce GT 1030 > has Compute SM 6.1 ] [hevc_nvenc @ 0x556c2f967560] OpenEncodeSessionEx failed: unsupported device (2) [hevc_nvenc @ 0x556c2f967560] No NVENC capable devices found [hevc_nvenc @ 0x556c2f967560] Nvenc unloaded Error initializing output stream 0:0 -- Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height Conversion failed! --- --- --- Now, using "h264_nvenc" returns the same... Somewhere someone mentioned an Intel CPU with QSV is required to run the nvenc codec, which seems strange to me The CPU here is an AMD Opteron 3350 HE. Any idea? Suggestions? Thanks a lot in advance, bye, -- piergiorgio

3 2

Questions about h264_amf encoder
by Chao Liu 24 Mar '18

24 Mar '18

Hi, I am trying to see how good h264_amf encoder is. I use ffmpeg to encode a 1080 video on Windows 10 running Ryzen 5 2500U. Two observations: - The encoded video quality is pretty poor, much worse than x264 with a smaller bitrate. - I got about 70fps with ~70% GPU usage. I was expecting to get better numbers though. As a comparison, on a much slower intel CPU Celeron n3450, using h264_vaapi to encode the same file, I got 30 fps with less than 20% GPU usage. The command I use is: ffmpeg -i transcode-test.mkv -pix_fmt nv12 -c:v h264_amf -profile:v constrained_baseline -rc cqp -qp_i 0 -qp_p 0 -qp_b 0 h264_amf.mkv (looks like the qp parameters don't change anything..) Did I do sth. wrong?

1 0

ffserver in prod env
by Good Developer 24 Mar '18

24 Mar '18

Is anybody using ffserver successfully in production environment?

2 1

Creating filter calling external library
by Jonathan Viney 22 Mar '18

22 Mar '18

Hi everyone, We have a filter that sends video frames to another C library for processing, then collects the result and applies some transformations to the frames. This filter runs as part of a longer filter chain that may include things like rotate and scale, as well as then producing several h.264 outputs. Currently the library runs synchronously. We send it 1 frame, and it gives us one back. Would there be any potential performance benefit from making the calls from the ffmpeg filter into the library non-blocking? We would most likely run the library in another thread. Eg, instead of: library.send_frame(frame); result = library.get_result(); ... process result and send frame to output Something like: if (ff_outlink_frame_wanted(outlink) && library.has_results()) { ... process results, sending frame(s) to output } if (ff_inlink_check_available_frame(inlink) && library.ready_for_next_frame()) { library.send_frame(frame); } return FFERROR_NOT_READY; I would expect that for a reasonable number of calls to activate(), there will be no forward steps that can be taken, so it will return FFERROR_NOT_READY. Could this be a problem for the filter system? Would the redesign potentially allow ffmpeg to do some useful work instead of being blocked on calls into the external library? Cheers, -Jonathan.

1 0

Skipping RTSP for RTP packets
by Dave Shapiro 22 Mar '18

22 Mar '18

Hi - I have a scenario where an IP camera is pushing up raw RTP data to the server. Currently I am spawning an ffmpeg process and telling it to connect to my server via RTSP. I then handle the handshake to setup the stream session. Once I receive the PLAY request I start piping data from the camera socket to the ffmpeg RTSP socket. I would like to remove the RTSP part of this and just send the RTP packets directly into the ffmpeg process. I have not found any documentation for this, is this possible? If so, what settings should I pass to the ffmpeg process so that it knows the data is interleaved? Thank you ahead of time for the help!

1 0

./configure finish with fail
by milan.konf＠seznam.cz 22 Mar '18

22 Mar '18

Hi all first, ffmpeg is good job. Thank you. When I try compile older version (from about 2018-01-14), all is OK but when I try the newest version from git with the some params, configure is finished with error. I have git-clone from 2018-03-20 (about 15 hours) and I try compile it with this params. ../ffmpeg/configure --enable-gpl --enable-version3 --enable-static --enable- nonfree --enable-nvenc --enable-cuda --enable-cuvid --extra-cflags=-I../ cudautils --extra-ldflags=-L../cudautils but configure is finished with error "ERROR: cuda requested, but not all dependencies are satisfied: ffnvcodec" Regards Milan

2 1