Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

November 2014

1 participants
60 discussions

ffmpeg-devel.log.20141129
by burek 30 Nov '14

30 Nov '14

[00:23] <cone-437> ffmpeg.git 03Neil Birkbeck 07master:dd5d61795690: avfilter/vf_idet: Fixing idet for single-frame inputs. [01:08] <cone-437> ffmpeg.git 03Lukasz Marek 07master:0cb994dfe32e: lavu/opt: add escaping to av_opt_serialize [04:27] <cone-784> ffmpeg.git 03Vadim Kalinsky 07master:2db29482f1ae: avutil/bprint: C++ compatible AVBPrint definition. [05:35] <jamrial> http://fate.ffmpeg.org/report.cgi?time=20141128085005&slot=x86.os2.446 something broke mp2 decoding on os/2 a month or so ago [05:35] <jamrial> encoding, rather [10:23] <ubitux> pross: yeah so actually, there is this: c70a6a41ddb759a5c6e8e810ebd603e39c36a315 [12:03] <cone-428> ffmpeg.git 03Dave Yeo 07master:090a7801a881: libavutil/thread.h: Support OS/2 threads [12:03] <cone-428> ffmpeg.git 03Dave Yeo 07master:32eadfe453c3: libavutil/threads.h: correct an include to be local [13:20] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:c299b6fd0807: avutil/buffer: Move USE_ATOMICS to thread.h to avoid it becoming out of sync with it [13:49] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:5bf4cd8e5b77: avutil/ppc/util_altivec: add () to VEC_SPLAT16 macro [14:25] <cehoyos> Daemon404: Hi Derek, what's wrong with pastebin (or Compn)? [14:25] <cehoyos> Anyway, I will leave the winodw open, thank you! [15:03] <cone-428> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n0.5.15': unknown revision or path not in the working tree. [15:03] <cone-428> Use '--' to separate paths from revisions [15:03] <cone-428> refs/tags/n0.5.15:HEAD: avutil/ppc/util_altivec: add () to VEC_SPLAT16 macro [15:43] <kurosu> Is there a non-convoluted way to export options for a bitstream filter ? [15:44] <kurosu> There's AV_OPT_FLAG_FILTERING_PARAM but no class/... way to then export said options except for libavcodec/options_table.h [15:46] <kurosu> passing a const char *args as a parameter to the bsf seem a dead giveaway it predates that kind of things [19:04] <BtbN> wow, linking ffmpeg.exe takes longer than the entire compiling, in cygwin [19:06] <Compn> mingw configure script takes foreeeeeeeever [19:06] <BtbN> Configure runs as fast as on my linux box [19:07] <BtbN> But it takes like 10 minutes to link ffmpeg.exe [19:07] <Compn> mingw and cygwin are diff , of course :p [19:07] <Compn> static ? [19:07] <BtbN> yes [19:08] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:cae851c78921: avcodec/ac3enc: Use avpriv_float_dsp_alloc() [19:08] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:14285c3331bc: avcodec/aacenc: Use avpriv_float_dsp_alloc() [19:08] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:b0464212bd81: avcodec/on2avc: Use avpriv_float_dsp_alloc() [19:09] <jamrial> linking on windows also trashes my hdd like mad [19:10] <jamrial> 10 min is excessive, though [19:14] <Compn> whenever i install mingw and try compiling ffmpeg or mplayer, seems like i quickly pick up a bunch of errors on the disk [19:14] <Compn> >windows problems [19:17] <jamrial> in general, running a cross-compiler on a linux VM is much faster than using a cygwin/msys environment [19:19] <jamrial> i rarely do it because i'm too lazy to boot up one. especially now that i started using msys2 which has git send-email working out of the box without a hassle :p [19:20] <Compn> ooo [19:20] <kurosu> does msys2 allow building windows binary without linking to the equivalent of cygwin's DLLs ? (ie a bit like -mno-cygwin) [19:20] <Compn> but yes cross compile is faster in all cases. [19:21] <Compn> if msys is using mingw, then yes. because mingw uses native binary not cygwin [19:21] <kurosu> iirc, msys2 is based off cygwin from I had read from afar [19:22] <jamrial> mingw is independent from the msys/msys2 environemnt [19:22] <Compn> iirc msys is just the shell, so [19:23] <Compn> i dont know what msys2 is doing yet [19:23] <kurosu> ok but cygwin usually impacts both the environment and the toolchain, hence my question [19:23] <Compn> cehoyos should force me to use msys2 , because i dont use git-send-email on windows [19:23] <Compn> send patches attached to emails ehe [19:24] <jamrial> yeah. the same happens if you use msys gcc [19:24] <kurosu> depends on the number of patches [19:24] <kurosu> if I'm sending only one as an update to a previous post, then I do it manually [19:24] Action: Compn prefers direct commit anyway [19:24] <jamrial> but mingw gcc is independent regardless of what environment you use [19:24] <BtbN> I'm using cygwin for everything on Windows. I basicaly cross compile in it, if i want to build for normal windows. [19:24] <BtbN> mingw is available as cygwin package, so you use it like you would with a cross compiler on linux. [19:24] <kurosu> probably I could learn how to use message ID to do just that, but... [19:25] <jamrial> you mean "git send-email --in-reply-to=<message ID> file.patch"? [19:29] <kurosu> yeah [19:38] <BtbN> Is there some example on how to use the dynarray from avutil/dynarray.h? [21:26] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:2336c76d5a6e: avcodec/ra288: Use avpriv_float_dsp_alloc() [21:26] <cone-428> ffmpeg.git 03Michael Niedermayer 07master:fc9ced41e48f: avcodec/twinvq: Use avpriv_float_dsp_alloc() [21:44] <BtbN> https://bpaste.net/show/6755cbec2e4d Is this me, doing some weird memory corruption in my nvenc encoder, or is this an actual bug in the mkv muxer? [23:01] <michaelni> BtbN, no idea from this, try valgrind or asan [23:20] <BtbN_> michaelni, already found it. Just forgot to set a variable, so i called av_malloc with a size of 0 [23:21] <BtbN_> It's interesting that this ringbuffer actualy does squeeze out a few more fps over the old linked list [23:21] <BtbN_> is malloc/free realy that expensive? [23:22] <nevcairiel> it can be [23:23] <BtbN_> yeah, but i didn't expect it to actualy take some time, even if the CPU isn't busy [00:00] --- Sun Nov 30 2014

1 0

ffmpeg.log.20141129
by burek 30 Nov '14

30 Nov '14

[00:17] <mopp> hm, I guess this has something to do with x264. I tried to input the http stream to a native x264.exe and it didn't work either. [04:37] <eristisk> How does gstreamer-ffmpeg work? It implements transcoding functionality via the ffmpeg libraries (libavcodec and libavformat, etc.) separate from the ffmpeg binary? [04:38] <eristisk> I have a Python software that is transcoding videos happily without even the ffmpeg binary installed on my system.... [04:48] <hipsterdufus> hi, I'm trying to use ffmpeg. I have a .avi video and want to convert it to multiple resolutions and encode with H.264, is that possible? [04:53] <c_14> hipsterdufus: yes [04:53] <c_14> https://trac.ffmpeg.org/wiki/Encode/H.264 [04:53] <c_14> https://trac.ffmpeg.org/wiki/FilteringGuide [04:53] <c_14> https://trac.ffmpeg.org/wiki/Creating%20multiple%20outputs [05:10] <hipsterdufus> ok, lets start smaller. How do I convert a file from .avi to mp4 with h.264 encoding? [05:16] <c_14> ffmpeg -i .avi -c:v libx264 out.mp4 [05:16] <c_14> That's all described very well in the first link I posted. [07:40] <KoolAidPitcher> Hello. I've been trying to debug a SIGSEGV from ffmpeg for a few days now... [07:40] <KoolAidPitcher> I've searched online to no avail, and while I can get it to crash with gdb attached to retrieve a backtrace, it does not crash with valgrind. [07:41] <KoolAidPitcher> The crash occurs when I call avcodec_flush_buffers, but only about 50% of the time... [07:41] <KoolAidPitcher> I'm typically someone who does not ask for help, but I am completely at a loss for what is causing it. I was hoping I could find some advice [07:42] <KoolAidPitcher> oh, right [07:42] <KoolAidPitcher> my version of ffmpeg is 2.4.3; however, it was also happening in 2.2.10, I tried both. [07:43] <KoolAidPitcher> The exact crash happens inside of malloc.c. The final few lines of the crash are: [07:43] <KoolAidPitcher> #0 malloc_consolidate (av=av@entry=0x7ffff1476760 <main_arena>) at malloc.c:4099 [07:43] <KoolAidPitcher> #1 0x00007ffff113bc66 in _int_free (av=0x7ffff1476760 <main_arena>, p=0x1f43230, have_lock=0) at malloc.c:3999 [07:43] <KoolAidPitcher> #2 0x00007ffff6c7cc5c in av_freep (arg=arg@entry=0x7fffb9459858) at /home/me/Workspace/avidemux2/buildCore_debug/ffmpeg/source/libavutil/mem.c:239 [07:43] <KoolAidPitcher> #3 0x00007ffff6f9fa93 in ff_h264_free_tables (h=h@entry=0x7fffb9458040, free_rbsp=free_rbsp@entry=1) at /home/me/Workspace/avidemux2/buildCore_debug/ffmpeg/source/libavcodec/h264.c:378 [07:43] <KoolAidPitcher> #4 0x00007ffff6fa2ee1 in flush_dpb (avctx=<optimized out>) at /home/me/Workspace/avidemux2/buildCore_debug/ffmpeg/source/libavcodec/h264.c:1119 [07:43] <KoolAidPitcher> #5 0x00007ffff71c017f in avcodec_flush_buffers (avctx=0x199a400) at /home/me/Workspace/avidemux2/buildCore_debug/ffmpeg/source/libavcodec/utils.c:3075 [07:43] <KoolAidPitcher> #6 0x00007ffff7bd47d9 in decoderFF::flush (this=0x1a33c50) at /home/me/Workspace/avidemux2/avidemux_core/ADM_coreVideoCodec/src/ADM_ffmp43.cpp:349 [07:47] <KoolAidPitcher> What is peculiar to me, is the segmentation fault is occuring when trying to free a pointer &h->non_zero_count [07:51] <relaxed> KoolAidPitcher: try again with git master and if it still happens open a bug report. [07:52] <fahadash> I have downloaded a PBS video which I would like to burn into DVD, but my DVD burner says Not-Supported-Format on most videos... and AVI files it says that the audio is not supported. Is there any way ffmpeg can convert my videos into formats that could easily be burned into DVDs ? [07:52] <KoolAidPitcher> hmm... okay [07:52] <KoolAidPitcher> so you have never seen this before? [07:52] <relaxed> fahadash: does your dvd player support mpeg4 video? [07:53] <fahadash> no [07:53] <fahadash> My DVD player supports old .VOB files [07:53] <fahadash> I guess thats mpeg2 , i dont know [07:54] <relaxed> correct, there's "-target ntsc-dvd" (and pal-dvd) [07:55] <fahadash> my dvd player supports ntsc [07:57] <KoolAidPitcher> Hi relaxed [07:57] <KoolAidPitcher> before I submit a bug report, I was curious if I could get any advice on debugging ffmpeg [07:58] <KoolAidPitcher> our version of ffmpeg is forked-- we apply patches to it prior to using it. [07:58] <relaxed> KoolAidPitcher: https://www.ffmpeg.org/bugreports.html [08:02] <KoolAidPitcher> wow [08:03] <KoolAidPitcher> => 0x00007ffff113b0e3 <malloc_consolidate+307>: mov 0x10(%rbx),%rax [08:03] Last message repeated 1 time(s). [08:03] <KoolAidPitcher> that's quite a moverax 0x34706d2e74736574 3778640133653489012 [08:03] <KoolAidPitcher> rbx 0xcb8f92d18d6e233c -3778640133621931204 [08:03] <KoolAidPitcher> rax 0x34706d2e74736574 3778640133653489012 [08:04] <KoolAidPitcher> I'm going to look into this a little further. This might be our patches. Thanks! [11:55] <sant527> I am using ffmpeg -i CAM00440.mp4; it shows "Stream #0:0(eng): Video: mpeg4 (Main Profile)". Is this mpeg4 referring to mpeg4 Part 2 [11:57] <relaxed> sant527: yes [11:59] <sant527> relaxed: Ok. [12:00] <sant527> relaxed: What are the various profiles of mpeg4 [13:32] <mopp> Hi! I'm capturing and on-the-fly transcoding a dvb-s2 stream via sat-ip. Everything works fine, this is the ffprobe output https://paste.ee/p/DKNYu As you can see, the mpegts frame rate is a 50fps progressive h264 stream. 50 FPS is due to PAL standards, so the TV station just doubles every frame from 25 to 50fps. That means every other frame is redundant. Can I drop every other frame? [13:36] <mopp> -x264opts video-filter=select_every:2,0,1 throws this error (complete output): https://paste.ee/p/dl44r [13:43] <sacarasc> Is that an actual x264 option? Or is it for a different part of ffmpeg? [13:45] <JEEB> probably an x264cli option he is trying to use via x264opts :P [13:45] <JEEB> mopp, go look at the video filter documentation on ffmpeg.org [13:45] <JEEB> there are various filters with documentation there [13:45] <JEEB> you'll probably find one that matches your use case :P [14:12] <mopp> "select_every" seems to be an actual x264 option. I'm using x264opts because select_every is not mapped to x264 directly. Well, I'll keep looking. Is it possible at all to drop frames from a h264 stream in order to reencode? I'm thinking of I-frames etc. [14:13] <JEEB> select_every is a x264CLI option of the video filter feature it has [14:13] <JEEB> it is not a LIBx264 feature [14:14] <JEEB> as I said, look at this page http://ffmpeg.org/ffmpeg-all.html [14:14] <JEEB> and see the documentation for the video filters [14:19] <Borys> hello [14:20] <Borys> anyone could tell me syntax for set icecast server as output ? [14:24] <mopp> JEEB: Huh. Using x264opts was overkill. A simple -vf fps=fps=25 was enough. thx! [14:28] <fahadash> can I convert mp4 to NTSC DVD ? [14:34] <sacarasc> ffmpeg -i blah.mp4 -target ntsc-dvd output.mpeg then author it and burn. [14:34] <Barvinok> hi. I use ffmpeg to play video on -f xv :0. how do I force playback to be at the same framerate as the source video? currently ffmpeg forces fps to be that of display refresh rate. [14:35] <Barvinok> because of this, the playback is sped up more than 2 times [14:43] <smo__> hello can i use -re with -ss? [14:48] <fahadash> thanks sacarasc [15:17] <fahadash> Which CPU architecture build is recommended 32bit or 64bit ? [15:20] <Barvinok> better match OS arch [15:33] <pagios> hi all what is the difference between smil and m3u8 [15:36] <fahadash> Hello [15:37] <fahadash> I ran the command sacarasc told me to, it output-ed the audio-only file [15:38] <fahadash> I can't open that with windows media player [15:39] <fahadash> Nevermind, Windows-Media-Player sucks [16:05] <Soft> Is there any documentation for libavformat aside from the doxygen docs? [16:11] <Borys_> how to set output of encoded mp3 from line-in to icecast? [16:12] <Moppel> In regards to transcoding a h264 mpegts with libx264 and -vf fps and cropping: would you advise to convert to rawvideo first or use the h264 stream directly? [16:39] <Alina-malina> can i use ffmpeg as virtualwebcam, so windows recognize its output as some webcamera and various programs capture that stream as webcamera(like skype, yahoo messenger etc) [16:47] <Alina-malina> facebook video [17:59] <debianuser> Alina-malina: Check something like manycam or VLC2VCam, it should support streaming sources, so you'll be able to stream ffmpeg to it [18:00] <Alina-malina> eh i need some pythonish stuff, that manycam doesnt designed as i expect [18:41] <droid909> thats what i try to do http://pastebin.com/8aT9uKcU [18:41] <droid909> why do i have this error [18:42] <droid909> i'me trying to stream so that my iphone could watch it via safari browser [18:42] <droid909> stream my webcam [19:00] <luc4> Hello! Any idea why I cant transcode from stdin? http://paste.kde.org/pcqefkmzm [19:08] <klaxa> broken file maybe, or it's because mp4 is bad at streaming [19:08] <klaxa> moov atom at the end or something [19:11] <luc4> klaxa: it works perfectly when not taking it from stdin. [19:12] <klaxa> probably the moov atom thing then [19:12] <luc4> You mean that it cant seek so being that at the end it wont work? [19:13] <klaxa> yes, you can move it to the front though [19:13] <klaxa> see: http://stackoverflow.com/questions/8061798/post-processing-in-ffmpeg-to-mov… [19:16] <luc4> This wont probably solve my original problem. I wanted to have a way to show a progress bar when transcoding with ffmpeg. This is not a good option. The only way is to probably modify ffmpeg itself. [19:21] <c_14> luc4: just take frame=$num; $num/total_frames [19:25] <luc4> c_14: sorry? [19:26] <droid909> what should i do with:This feed is already being received. [19:26] <droid909> ? [19:26] <c_14> When ffmpeg is encoding it outputs a little informational message with frame= fps= etc. Just take the current output of frame= and divide by the total number of frames [19:26] <droid909> c_14: me? [19:27] <c_14> nah, luc4 [19:27] <c_14> droid909: kill ffserver and restart it [19:28] <luc4> c_14: yes, I see that. But is there the toal amount somewhere? Also, that means modifying the ffmpeg sources. Is there a particular reason why it was never done? [19:28] <droid909> c_14: tried, the same [19:29] <droid909> c_14: can i post my config file, will you take a look? [19:30] <droid909> c_14: config http://pastebin.com/jiSgrLZM [19:31] <droid909> c_14: and i run it like ffmpeg -r 25 -f video4linux2 -i /dev/video0 http://localhost:8090/feed1.ffm [19:32] <c_14> luc4: because ffmpeg doesn't know how many frames there will be until it's done [19:33] <luc4> c_14: then I cant do the division... [19:33] <c_14> luc4: you can find out manually with ffprobe -count_frames -show_streams iirc [19:33] <c_14> though if it's cbr you can just do fps * total_time [19:34] <c_14> droid909: I can't see anything wrong with that. What FFmpeg version are you running? [19:34] <luc4> c_14: ah ok, so is there a particular reason why ffmpeg does not do that? I may add another param to add a progress bar& [19:34] <droid909> c_14: ffmpeg version 2.4.3 Copyright (c) 2000-2014 the FFmpeg developers [19:36] <c_14> luc4: because ffmpeg doesn't know if the file is constant framerate or variable framerate, and it doesn't always have the total length of the video [19:36] <c_14> droid909: what's the complete console output of both ffmpeg and ffserver [19:37] <luc4> c_14: mmh& concept is& if ffmpeg cannot do it automatically, I cant do that either& [19:38] <c_14> luc4: if you have the whole file, you can use ffprobe to count each individual frame, grab the output from that and use that for the progress [19:38] <droid909> c_14: ffmpeg http://pastebin.com/ciiRwHGQ and ffserver http://pastebin.com/rTAyNjCq [19:39] <luc4> c_14: which means I can patch ffmpeg to count the frames if this is a file and create the progress bar. Correct? [19:39] <c_14> luc4: yes [19:39] <luc4> c_14: thanks! [19:40] <droid909> c_14: maybe i should try another format? [19:41] <droid909> c_14: i need any that will work in iphone's browser [19:42] <c_14> droid909: what's the problem? It looks like it works. [19:44] <droid909> c_14: when i access it in my google chrome it writes: This feed is already being received. [19:44] <droid909> c_14: and starts downloading the feed [19:47] <c_14> You're accessing http://ip/test.swf ? [19:49] <droid909> c_14: ohh... [19:49] <droid909> c_14: i tried feed1.ffm [19:49] <droid909> c_14: it works now .. [19:50] <droid909> still i need to choose something that work in safari, and swf isn't for safari, i was trying just to make it work [19:50] <droid909> c_14: what format should i chose? [19:55] <droid909> mjpeg? [19:55] <c_14> that might ork [19:57] <c_14> *work [20:07] <droid909> c_14: that is my input i guess : Stream #0:0: Video: rawvideo (YUY2 / 0x32595559), yuyv422, 320x240, 36864 kb/s, 30 fps, 30 tbr, 1000k tbn, 1000k tbc [20:56] <hexafluoride> hey [20:56] <hexafluoride> this command: ffmpeg -i in.mp3 -f s16le -ar 48000 -ac 1 out.wav [20:57] <hexafluoride> should produce a valid wave file with a RIFF header, right? [20:57] <hexafluoride> or am I missing something? [21:14] <MadTBone_> In general, it seems like container formats have audio and video packets interleaved. Are there any common formats where they aren't, but rather all video at the beginning followed by audio, or something similar? [21:34] <Mavrik> MadTBone_, I haven't heard about any: that greatly increses costs of video player hardware :) [21:34] <Mavrik> MadTBone_, perhaps RTP which transmits audio and video on separate bands [21:38] <MadTBone_> Mavrik, that's what I figured. It would introduce lots of seeking/random i/o into what should be a mostly linear data read /write [21:41] <Mavrik> mhm, also most formats are defined for hardware decoders where memory was premium [21:41] <Mavrik> hence most formats have hard limits on how far audio, video and other references can be between them [22:28] <mopp> I'd like to do -vf cropdetect and -vf crop in one step, in other words I'd like to parse the cropdetect output into -vf crop, in one line preferable. Is that possible from CLI without a bash script? I've only found this so far: http://pastebin.com/n8Jk3DjS but I'm on a windows box and I have very little scripting experience. [22:29] <mopp> *preferably [22:50] <wyatt8740> Hi, having a problem here. How do I separate the chroma and luma of a video in ffmpeg? I tried '-vf pp=noluma' but it spat out an error. [22:50] <wyatt8740> 1 errors in postprocess string "noluma" / [AVFilterGraph @ 0x2ed7e00] Error initializing filter 'pp' with args 'noluma' / Error opening filters! [22:50] <wyatt8740> ffmpeg 2.3.3 [23:21] <wyatt8740> okay I got around it, but it looks like the postproc (pp) filter is still not working properly. [23:21] <wyatt8740> What worked was '-vf lutyuv=y=128' [23:44] <wyatt8740> how does the mergeplanes filter work? [23:45] <wyatt8740> I'm trying to merge a video with only chrominance to a video with only luminance [23:55] <wyatt8740> now I'm getting more problems [23:55] <wyatt8740> ffmpeg -i 03luminance.mkv -i 03chrominance.mkv -filter_complex [0:0][1:0]mergeplanes=0x00010210:yuv422p out.mkv [23:55] <wyatt8740> [Parsed_mergeplanes_0 @ 0x3973de0] output plane 0 width 640 does not match input 0 plane 1 width 320 [23:55] <wyatt8740> but all the inputs are 640 pixels wide. [23:57] <wyatt8740> nvm might have gotten it [00:00] --- Sun Nov 30 2014

1 0

ffmpeg-devel.log.20141128
by burek 29 Nov '14

29 Nov '14

[00:00] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:449bd3c0c33c: Merge commit '96fda42a8f9bf84beaaf7f5991d17f2a057de86c' [00:12] <ramiro> Zeranoe: new patch sent in your forum [00:28] <cone-146> ffmpeg.git 03Anton Khirnov 07master:1973079417e8: opusdec: make sure all substreams have the same number of coded samples [00:28] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:fae545ca2a34: Merge commit '1973079417e8701b52ba810a72cb6c7c6f7f9a56' [00:57] <cone-146> ffmpeg.git 03Anton Khirnov 07release/2.4:12700b021952: mp3enc: fix a triggerable assert [00:57] <cone-146> ffmpeg.git 03Anton Khirnov 07release/2.4:46a17d886b85: lavu: add wrappers for the pthreads mutex API [00:57] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:86b532898d8a: Merge commit '12700b0219521a5f20c8ba47b3ad7857ea9e0554' into release/2.4 [00:57] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:15fd62110fd8: Merge commit '46a17d886b8559723c40b9f5cdf0e0c6b1c95180' into release/2.4 [01:15] <cone-146> ffmpeg.git 03wm4 07release/2.4:517ce1d09b5e: lavu: fix memory leaks by using a mutex instead of atomics [01:15] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:b6691fba77af: Merge commit '517ce1d09b5e6b72afc2ef9490b5f8ca42fa6a65' into release/2.4 [01:15] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:f783259fdb37: avutil/buffer: use the old atomics based code for the release branch [01:26] <cone-146> ffmpeg.git 03Anton Khirnov 07release/2.4:ca78ee73db9e: opusdec: make sure all substreams have the same number of coded samples [01:26] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:0e216ed40789: avutil/buffer_internal: leave the buffer pool entries volatile [01:26] <cone-146> ffmpeg.git 03Michael Niedermayer 07release/2.4:56b84b023d70: Merge commit 'ca78ee73db9e059f501706ba6108e23902e84933' into release/2.4 [02:09] <cone-146> ffmpeg.git 03Timo Rothenpieler 07master:95fc80672ffe: Move extralibs variables using ldl after ldl definition [02:09] <cone-146> ffmpeg.git 03Benoit Fouet 07master:7dfee8d69793: avcodec/pngdec: split P frames handling to a separate function. [02:09] <cone-146> ffmpeg.git 03Benoit Fouet 07master:cfd83a8af6bf: avcodec/pngdec: add support for 'over' blend operation for 'none' dispose operation. [04:17] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:ea38e5a6b757: avcodec/hevc_ps: Check num_long_term_ref_pics_sps [11:56] <cone-437> ffmpeg.git 03Martin Storsjö 07master:857e6667f906: rtmpproto: Clarify a comment [11:57] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:a105c1f204bd: Merge commit '857e6667f9061ae261c0b951113e4efc4329b05e' [12:13] <cone-437> ffmpeg.git 03Martin Storsjö 07master:3c3b8003a13d: rtmpproto: Simplify code for copying data into the output packet [12:14] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:42a095d095b5: Merge commit '3c3b8003a13d9c3668c0bb6d79d2376da3b2b352' [12:25] <cone-437> ffmpeg.git 03Martin Storsjö 07master:44127b157e9f: rtmppkt: Make pkt->data reallocable [12:25] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:900fff89c9ed: Merge commit '44127b157e9f8acb837d4bb3a094f56b40da3ef5' [13:04] <cone-437> ffmpeg.git 03Bryan Huh 07master:a9d8d35e4833: dashenc: Add options to make segment names configurable [13:04] <cone-437> ffmpeg.git 03Martin Storsjö 07master:fcae9f212a60: dashenc: Avoid a VLA-like construct [13:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:5ce070c16c72: Merge commit 'a9d8d35e4833fc4dfbf557ce73c84e9ca6224427' [13:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:f001a2bd344e: Merge commit 'fcae9f212a6001d966c52dc22cd4b22e9851b428' [15:13] Action: Compn forgets timothy gu's irc nick [15:14] <Daemon404> Timothy_Gu [15:14] <Daemon404> shocking, i know [15:14] <Compn> ah [15:14] <Compn> so not online [15:14] <Compn> thus, he does not exist! [15:15] <Daemon404> i cant remember: did we move to his nodejs fate? [15:15] <Compn> dunno, fate server would probably look different if so ? [15:15] <Daemon404> no [15:15] <Daemon404> html/css would be the same probably [15:15] <Daemon404> it's a new backaend [15:15] <Compn> fate looks diff [15:15] <Compn> ah [15:15] <Daemon404> backend* [15:25] <Daemon404> oh carl isnt online when i finally get him is dang prores thing [15:25] <Daemon404> OH WELL [15:29] Action: koda still is unsure what he needs it for [15:34] <Compn> Daemon404 : you got the sample or what ? [15:34] <Compn> where is it ? [15:34] <Compn> i'll relay it to him [15:34] <Daemon404> he wanted output from a shitty prores-in-avi sample [15:34] <Daemon404> ill just wait [15:34] <Compn> whatever prores thing is :P [15:34] <Compn> ok [15:35] <koda> Compn: i got the sample, but i cannot share it [15:36] Action: koda also wonders why people are asking Daemon404 rather than the patchs author [15:43] <Compn> koda : i have no idea what is going on, i was just offering my services since it seems like Daemon404 is tired of non-instant communication with some people :P [15:43] Action: koda hugs Compn [15:44] <cone-437> ffmpeg.git 03Rong Yan 07master:57c89c50bd2f: avcodec/ppc/h264dsp: POWER LE support for h264_idct8_add_altivec() h264_idct_dc_add_internal() h264_loop_filter_luma_altivec() write16x4() VEC_1D_DCT() weight_h264_W_altivec() biweight_h264_W_altivec() VEC_LOAD_U8_ADD_S16_STORE_U8() ALTIVEC_STORE_SUM_CLIP() [15:44] <cone-437> ffmpeg.git 03Rong Yan 07master:89f3043c7fa6: avcodec/ppc/h264chroma_template: POWER LE support for PREFIX_h264_chroma_mc8_altivec() PREFIX_no_rnd_vc1_chroma_mc8_altivec() CHROMA_MC8_ALTIVEC_CORE_SIMPLE() CHROMA_MC8_ALTIVEC_CORE() [15:44] <cone-437> ffmpeg.git 03Rong Yan 07master:8cc5a78e45d9: avcodec/ppc/h264qpel_template: POWER LE support for PREFIX_h264_qpel16_h_lowpass_altivec() PREFIX_h264_qpel16_v_lowpass_altivec() PREFIX_h264_qpel16_hv_lowpass_altivec() [15:44] <cone-437> ffmpeg.git 03Rong Yan 07master:bd67d0ead145: avcodec/ppc/h264qpel: POWER LE support for put_pixels16_l2_altivec() and avg_pixels16_l2_altivec() [15:44] <Compn> in my capacity as a 'dev' , i pretty much have no idea what is going on 100% of the time. [15:44] <Compn> ehehe [15:55] <Daemon404> nevcairiel, looked into schannel.... my god [15:57] <kierank> koda: :) [15:57] <koda> kierank: o/ [16:16] <kierank> Welcome to the dark side [16:27] <cone-437> ffmpeg.git 03Benoit Fouet 07master:aff50ae1d176: avcodec/pngdec: do not blend on transparent black [16:27] <cone-437> ffmpeg.git 03Benoit Fouet 07master:af8804ac8638: avcodec/pngdec: allow for some code path optimizations. [16:27] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:885a763cacf5: avcodec/ppc/h264qpel_template: protect unistd.h by #if HAVE_UNISTD_H [16:59] <aleskandro> Hi all, I'm trying to learn to code with ffmpeg and vdpau API. I'm looking for some examples and found the ffmpeg_vdpau.c file in ffmpeg.org... What is it? how can I use that? [17:02] <aleskandro_> Hi all, I'm trying to learn to code with ffmpeg and vdpau API. I'm looking for some examples and found the ffmpeg_vdpau.c file in ffmpeg.org... What is it? how can I use that? [17:02] <BtbN> That's the vdpau code for the ffmpeg cli util as far as i know. [17:05] <aleskandro_> BtbN: do you know any basic player as ffplay.c that use vdpau? [17:05] <BtbN> no [17:05] <BtbN> using vdpau with ffmpeg is complex and not well documented [17:06] <nevcairiel> ffmpeg_vdpau.c is not very long, and should serve as an example on how to implement it more than actual useful function for ffmpeg itself [17:10] <aleskandro_> nevcairiel: yeah but there aren't comments about the use case of that functions so how can I understand it? [17:11] <BtbN> aleskandro_, https://gist.github.com/BtbN/517f7d11aad228f2eb9d that's from one of my projects, it uses libavcodec with vdpau. Look for the HAVE_VDPAU blocks. [18:32] <cone-437> ffmpeg.git 03Benoit Fouet 07master:ebf2052a6177: avcodec/pngdec: apng: fix output buffer filling when no blending is needed. [19:13] Action: ubitux wonders if there is a relationship between AV_AUDIO_SERVICE_TYPE_* and AV_DISPOSITION_* [19:13] <Daemon404> i doubt it [19:14] <ubitux> AV_DISPOSITION_HEARING_IMPAIRED [19:14] <ubitux> AV_DISPOSITION_VISUAL_IMPAIRED [19:14] <ubitux> AV_AUDIO_SERVICE_TYPE_VISUALLY_IMPAIRED [19:14] <ubitux> AV_AUDIO_SERVICE_TYPE_HEARING_IMPAIRED: [19:14] <ubitux> also AV_AUDIO_SERVICE_TYPE_KARAOKE vs AV_DISPOSITION_KARAOKE [19:14] <ubitux> etc.. [19:15] <Daemon404> lol awesome [19:15] <ubitux> i just don't understand why Anton seem to be targetting random pkt side data with this information [19:16] <ubitux> instead of just flagging the stream [19:16] <ubitux> there is probably a good reason but well. [19:16] <ubitux> i probably miss something [19:17] Action: Daemon404 stabs schannel more [19:17] <Daemon404> no wonder nevcairiel never did this [19:18] <JEEB> I'm happy I have nfi about wtf that is [19:18] <ubitux> schannel? [19:18] <koda> ubitux: ask him? :) [19:18] <ubitux> koda: i'm on his ignore list afaik [19:19] <ubitux> because i'm "demotivating" him last i heard [19:19] <ubitux> he's not interested in talking to me and even less helping me, so i'm not going to ask him anything [19:19] <koda> i meant on the mailing list [19:20] <ubitux> i'm not going to force myself in here [19:21] <koda> dunno, try helping him so hell help you =) [19:21] <koda> afaik it has something to deal with the avcodecparameters [19:22] <ubitux> it's not important, i'll eventually have a look another time [20:02] <cone-437> ffmpeg.git 03Anshul Maheswhwari 07release/2.1:712b8e712828: v4l2enc: adding AVClass [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:d5eca1651f4a: avcodec/dvdsub_parser: never return 0 when the input isnt 0 [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:f99675627d1c: avcodec/dvdsub_parser: Check buf_size before reading 32bit packet size [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:384be84bbe8a: avcodec/dvdsub_parser: print message if packet is smaller than the packet size field [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ded44bda27ec: ffmpeg_opt: Use av_guess_codec() instead of AVOutputFormat->*codec [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:72149fcb1b79: avformat/tee: flip assigment direction [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:7d69132775e6: avcodec/wavpackenc: Fix log2sample() result value [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:150ae7692efc: ffserver: initialize pbuffer in prepare_sdp_description() [20:02] <cone-437> ffmpeg.git 03Anton Khirnov 07release/2.1:f7ed48938a33: cdgraphics: do not return 0 from the decode function [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:f405267493c8: avcodec/iff: check pixfmt for rgb8 / rgbn [20:02] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:c9da441eb460: avcodec/snow: fix null pointer dereference in cleanup after allocation failure [20:03] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:e1ed566c1cda: proresenc_kostya: remove unneeded parameters [20:03] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:92096acc0a04: proresenc_kostya: report buffer overflow [20:03] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:407982b8f945: proresenc_kostya: properly account for alpha [20:03] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:656bf0ca79ba: wavpack: report if there is no bits left [20:03] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:80b6632b360a: wavpackenc: proper buffer allocation [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:6ab793c2b9bb: avcodec: fix aac/ac3 parser bitstream buffer size [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:cb8f645fafec: avcodec/utils: add GBRP16 to avcodec_align_dimensions2() [20:03] <cone-437> ffmpeg.git 03Timothy Gu 07release/2.1:0f8863df8640: bktr: Fix Fabrice's name [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:8c8950f982c5: avcodec/snow: check coeffs for validity [20:03] <cone-437> ffmpeg.git 03wm4 07release/2.1:17f30ab6a154: oggdec: fix invalid free on error [20:03] <cone-437> ffmpeg.git 03Mika Raento 07release/2.1:cb10e05ff06e: segment: don't access outside seg->frames array [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:255ebf3aff52: avformat/swfdec: Use side data to communicate w/h changes to the decoder [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:72f0d13802ba: avformat/swfdec: Do not change the pixel format [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:f7086be79a59: avcodec/h264: Allow partial escaping [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:7644b292222c: avcodec/mpegvideo: Use "goto fail" for all error paths in ff_mpv_common_frame_size_change() [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:31f5d5f6903c: avcodec/mpegvideo: check that the context is initialized in ff_mpv_common_frame_size_change() [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:cd190f0c3fd3: avcodec/mpegvideo: Set err on failure in ff_mpv_common_frame_size_change() [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:5bd45a1b27dd: avformat/m4vdec: Check for non startcode 00 00 00 sequences in probe [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:5e4a821b8a41: tools/crypto_bench: fix build when AV_READ_TIME is unavailable [20:03] <cone-437> ffmpeg.git 03Katerina Barone-Adesi 07release/2.1:36ec1c2c5519: apetag: Fix APE tag size check [20:03] <cone-437> ffmpeg.git 03Pascal Massimino 07release/2.1:5ecc4a644a1f: libavcodec/webp: treat out-of-bound palette index as translucent black [20:03] <cone-437> ffmpeg.git 03James Almer 07release/2.1:84487650e23b: x86/dsputil: add emms to ff_scalarproduct_int16_mmxext() [20:03] <cone-437> ffmpeg.git 03Gianluigi Tiesi 07release/2.1:34ef754854ee: avcodec/libilbc: support for latest git of libilbc [20:03] <cone-437> ffmpeg.git 03Pascal Massimino 07release/2.1:77367f27280c: avcodec/webp: fix default palette color 0xff000000 -> 0x00000000 [20:03] <cone-437> ffmpeg.git 03Benoit Fouet 07release/2.1:dc319a52f920: avformat/riffenc: Filter out "BottomUp" in ff_put_bmp_header() [20:03] <cone-437> ffmpeg.git 03Philip DeCamp 07release/2.1:9c61b4494113: libavutil/opt: fix av_opt_set_channel_layout() to access correct memory address [20:03] <cone-437> ffmpeg.git 03lvqcl 07release/2.1:8dd6075a7f28: avutil/x86/cpu: fix cpuid sub-leaf selection [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:c658f6c34d0f: avcodec/ac3enc_template: fix out of array read [20:03] <cone-437> ffmpeg.git 03Reimar Döffinger 07release/2.1:d61a325a68aa: configure: add noexecstack to linker options if supported. [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:23a22b0da90d: avcodec/jpeglsdec: Check run value more completely in ls_decode_line() [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:9ec550c36483: avcodec/mjpegdec: check bits per pixel for changes similar to dimensions [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:b5298c464f05: avcodec/utils: Add case for jv to avcodec_align_dimensions2() [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ce9d497755fd: avcodec/mmvideo: Bounds check 2nd line of HHV Intra blocks [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ceb9d67a0ad1: avcodec/tiff: more completely check bpp/bppcount [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:5d852f338d5b: avcodec/pngdec: Check bits per pixel before setting monoblack pixel format [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:4841b2759f11: avcodec/pngdec: Calculate MPNG bytewidth more defensively [20:03] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:96357894ff9a: avcodec/cinepak: fix integer underflow [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:05d7e92e4fdd: avcodec/gifdec: factorize interleave end handling out [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:c10c71452fff: avcodec/qpeg: fix off by 1 error in MV bounds check [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:bf2605c35610: avcodec/smc: fix off by 1 error [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:21808e218180: avcodec/vorbisdec: Fix off by 1 error in ptns_to_read [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ea61dfe0abf4: avformat/mpegts: Check desc_len / get8() return code [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:f335275c50b8: avcodec/h264: Check mode before considering mixed mode intra prediction [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:11313263ab37: swresample/swresample: fix sample drop loop end condition [20:04] <cone-437> ffmpeg.git 03Christophe Gisquet 07release/2.1:3fd2ff1b4f2e: utvideoenc: properly set slice height/last line [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:a90ed80d05b7: postproc/postprocess: fix quant store for fq mode [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:95fa91d97525: postproc: fix qp count [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:917946e1c96e: avcodec/diracdec: Use 64bit in calculation of codeblock coordinates [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:824f29e4fbb7: avcodec/diracdec: Tighter checks on CODEBLOCKS_X/Y [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:dc2f4b72673a: avcodec/dirac_arith: fix integer overflow [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:8c5c45b9e1b6: avcodec/dxa: check dimensions [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:a2619a0a36db: avcodec/dnxhddec: treat pix_fmt like width/height [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:a4b6d5574f05: avcodec/utils: Align dimensions by at least their chroma sub-sampling factors. [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ffd5ccee5df5: avcodec/g2meet: check tile dimensions to avoid integer overflow [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:3467dfed6bf3: avcodec/cook: check that the subpacket sizes fit in block_align [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:bc0a6add0aea: avcodec/svq1dec: zero terminate embedded message before printing [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:4352a971afeb: avcodec/h264_slice: Clear table pointers to avoid stale pointers [20:04] <cone-437> ffmpeg.git 03Carl Eugen Hoyos 07release/2.1:1987afe5a0c5: lavc/utils: Make pix_fmt desc pointer const. [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:de259f32ac12: avcodec/options_table fix min of audio channels and sample rate [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:58e673b9a345: avcodec/utvideodec: fix assumtation that slice_height >= 1 [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:d3e19509cf43: avcodec/wmaprodec: Fix integer overflow in sfb_offsets initialization [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:c4034e4e192e: avformat/hlsenc: Free context after hls_append_segment [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:51c32a535bb4: swscale/x86/rgb2rgb_template: handle the first 2 lines with C in rgb24toyv12_*() [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:b4ce4f94e781: avcodec/mjpegdec: Fix context fields becoming inconsistent [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:126ee72e47c9: avcodec/utils: Check that the data is complete in avpriv_bprint_to_extradata() [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:431d66b1d560: avcodec/flacdec: Call ff_flacdsp_init() unconditionally [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:ef8fbb4878dc: avcodec/pngdec: Check IHDR/IDAT order [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:6212758c0757: avcodec/rawdec: Check the return code of avpicture_get_size() [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:35cfb99c15dd: avcodec/hevc_ps: Check num_long_term_ref_pics_sps [20:04] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:e4f0f854c476: avcodec/svq3: Dont memcpy AVFrame [20:17] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/2.1:27172a5ca360: Update for 2.1.6 [20:28] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:371403601121: avcodec/jpeglsdec: Check run value more completely in ls_decode_line() [20:28] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:f7170c48328d: avcodec/mjpegdec: check bits per pixel for changes similar to dimensions [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:9b4507e423e5: avcodec/mmvideo: Bounds check 2nd line of HHV Intra blocks [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:b2f2cbdb1caf: avcodec/gifdec: factorize interleave end handling out [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:eac21ee7ba0f: avcodec/qpeg: fix off by 1 error in MV bounds check [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:7128f67c3f20: avcodec/smc: fix off by 1 error [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:b8129b1a7ab9: avformat/mpegts: Check desc_len / get8() return code [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:831416692b81: avcodec/dxa: check dimensions [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:473b01609627: avcodec/pngdec: Check IHDR/IDAT order [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:e74795e54108: huffyuvdec: check width more completely, avoid out of array accesses [20:29] <cone-437> ffmpeg.git 03Michael Niedermayer 07release/0.5:15d6b44ddcce: update for 0.5.15 [20:59] <cone-437> ffmpeg.git 03Benoit Fouet 07fatal: ambiguous argument 'refs/tags/n2.1.6': unknown revision or path not in the working tree. [20:59] <cone-437> Use '--' to separate paths from revisions [20:59] <cone-437> refs/tags/n2.1.6:HEAD: avcodec/pngdec: apng: fix output buffer filling when no blending is needed. [22:01] <cone-437> ffmpeg.git 03Michael Stypa 07master:cb58c771ade6: fix Makefile objects for pulseaudio support [22:21] <cone-437> ffmpeg.git 03Martin Storsjö 07master:491805636cef: rtmpproto: Fix a typo in a comment [22:21] <cone-437> ffmpeg.git 03Michael Niedermayer 07master:27897d2ef6d3: Merge commit '491805636cef50d3f582bd345e1460eeb739ea48' [22:57] <pross> ubitux: AV_AUDIO_SERVICE_TYPE_ lol [22:58] <pross> disposition was first! [00:00] --- Sat Nov 29 2014

1 0

ffmpeg.log.20141128
by burek 29 Nov '14

29 Nov '14

[00:06] <Demon_Fox> Jack64, You mean mpeg2 files from a dvd? [00:08] <kepstin-laptop> Jack64: well, if you have the encryption keys, that would be a good start... [00:18] <Jack64> I mean from a stream [00:18] <Jack64> I was just doing a small test and it worked [00:19] <Jack64> so I'm trying to interface with an app that delivers tv streaming [00:19] <Jack64> i login to the service and get the m3u8 file [00:19] <Jack64> and the key [00:19] <Jack64> and the ts streams on the m3u8 file [00:20] <Jack64> then I put them on my local webserver [00:20] <Jack64> and I create a new m3u8 file pointing to the files + key file [00:20] <Jack64> and then i load the m3u8 file on vlc and it worked [00:21] <Jack64> well the point is to make a small relay on my server so i can watch the stream on my tv with XBMC [00:22] <Jack64> because I can't do the authentication there [00:30] <DanielSa_> Hi [00:34] <DanielSa_> have you heard about PJPG and 0511? I have two cameras with these "Pixel Format", according to "v4l2-ctl --get-fmt-video" command. And I'm trying to use both at "-input_format" and I'm always getting this error "No such input format: PJPG". Any idea? Thank you in advance. [00:36] <jack_thompson> what is the point of ffserver when i can only stream 1 file per <stream>? [00:36] <jack_thompson> i keep failing to stream more than one file using ffserver over the same stream... [00:36] <jack_thompson> i thought it transcodes everything into the same stream... [00:37] <jack_thompson> but ffserver doesn't do that... it doesn't even keep clients connected when the stream stops... [00:37] <jack_thompson> or push new frames to the clients when the broadcast is up again... [00:38] <jack_thompson> it does absolutely nothing than just existing and /not/ showing the least bit of functionality.. [00:39] <jack_thompson> ffserver has no functioniality other than being a glorified proxy... why that has even to be configured.. i seriously dont know... [00:40] <jack_thompson> at least a way to configure a timeout before it breaks over ffmpeg reconnecting would make it 100% more useful than it is now... [00:40] <jack_thompson> imagine you'd using it to livestream something... [00:41] <jack_thompson> how fucking often that would go offline... people would /hope/ to see at least a commercial... [00:42] <jack_thompson> someone should actually try using ffserver... like i have tried the last days... you'd see what a disfunctional piece of software it is.. [00:42] <jack_thompson> sometimes it just crashes(!).. and goes offline(!)... what server does that? :O [00:45] <jack_thompson> all i want is ffmpeg to feed completely different sources to ffserver... and connect to ffserver with a client that does play everything (transcoded to the same format and size(s))... [00:45] <jack_thompson> *it* *does* *not* *happen* [00:46] <sacarasc> I heard ffserver is a candidate for removal because it's not good and no one cares about it. [00:47] <jack_thompson> ffserver, at least, would have to send a blank screen to the playback client until ffmpeg connects with another source... but it doesn't do that... [00:47] <jack_thompson> sacarasc, if that is so... that would explain *everything* i have gone through... [00:47] <jack_thompson> i wonder what people use instead... [01:14] <MrSavage> I'm getting a blackscreen when I'm trying to re-encode an rtmp stream with libx264 [01:14] <MrSavage> any help? [01:29] <MrSavage> anyone? [01:31] <MrSavage> c_14: http://pastebin.com/iFpTaWD8 [01:32] <c_14> And the output? [01:32] <MrSavage> just audio [01:32] <c_14> The console output [01:33] <MrSavage> c_14: is it because I"m trying to reencode x264 and it says in ffmpeg -codecs that only encoding is available with x264? [01:34] <c_14> wait, what [01:34] <c_14> It doesn't show decoding for h.264? [01:34] <MrSavage> c_14: I was unable to see the console output due to the command being run by nginx [01:34] <c_14> can you exec with a &>file ? [01:34] <MrSavage> for h264 yes [01:34] <MrSavage> but not for libx264 [01:34] <c_14> libx264 doesn't decode [01:34] <c_14> libx264 is only the encoder [01:35] <c_14> the decoder is ffmpeg internal [01:35] <MrSavage> ok [01:37] <MrSavage> c_14: I'm not sure how I can see the output when the command is being run by another program [01:37] <c_14> worst case: exec sh -c 'ffmpeg command &>/path/to/file' [01:39] <gabriel_> Hi again everyone. Can anyone tell me why I might only be seeing the first frame of an image sequence using image2pip as the input format, mpeg4 as the vcodec, and m4v as the output file? I ame using -framerate 25 and -r 25 [01:40] <gabriel_> If I do the same exact parameters but set the output format to mov, I can see both frames [01:40] <c_14> The only thing I can think of is a bug in the muxer. [01:43] <MrSavage> hold on [01:43] <MrSavage> almost done with printing output [01:45] <gabriel_> Ok, if it helps Im using this exact command: ['/usr/local/bin/ffmpeg', '-framerate', '25', '-i', 'test_rand/11-small.png', '-i', 'test_rand/25-small.png', '-f', 'image2pipe', '-filter_complex', 'concat=n=2:v=1:a=0', '-vcodec', 'mpeg4', '-b:v', '800k', '-r', '25', '-f', 'm4v', '-y', 'pipe:1'] [01:45] <MrSavage> can I just state -vcodec with no other options for video? [01:46] <c_14> MrSavage: normally yes [01:47] <c_14> gabriel_: does the same problem occur when writing to a file [01:47] <c_14> Have you tried setting -framerate for the second input as well? [01:47] <gabriel_> Yeah, if the output format is the same [01:47] <gabriel_> ohhhhh [01:47] <gabriel_> interesting [01:47] <gabriel_> let me try that [01:49] <MrSavage> c_14: this is not writing a log exec sh -c 'ffmpeg -i rtmp://localhost:1935/$app/$name -vcodec libx264 -acodec copy -f flv rtmp://localhost:1935/small/test &> /usr/local/nginx/logs/ffmpeg.log'; [01:50] <c_14> hmm, is the file being created? [01:51] <c_14> Does the nginx process have write permissions there? [01:51] <gabriel_> c_14: like this? ['/usr/local/bin/ffmpeg', '-f', 'image2pipe', '-framerate', '25', '-i', 'test_rand/11-small.png', '-framerate', '25', '-i', 'test_rand/25-small.png', '-filter_complex', 'concat=n=2:v=1:a=0', '-vcodec', 'mpeg4', '-b:v', '800k', '-r', '25', '-f', 'm4v', '-y', 'pipe:1'] [01:52] <c_14> What's that first -f image2pipe doing there? [01:52] <c_14> unless 11-small.png is a fifo [01:53] <c_14> Also, I don't think you can't write mp4 to a pipe [01:53] <MrSavage> c_14: it should have permission. I'm executing as root [01:53] <c_14> MrSavage: is the file created? [01:53] <MrSavage> c_14: no [01:54] <c_14> try replacing the command with exec sh -c 'touch /usr/local/nginx/logs/ffmpeg.log' [01:54] <gabriel_> c_14: basically Im only interested in the resulting filesize, so I dont want to write the resulting file to disk. I thought the -f image2pipe was neccesary to get the resulting bytes into memory [01:55] <c_14> You're using the image2pipe as an input format, that should only be necessary when you're actually receiving images from a pipe [01:55] <MrSavage> c_14: still no file [01:55] <gabriel_> Im able to pipe the results to output.m4v and play it with vlc but thats all I know [01:55] <gabriel_> Ok so I just want -f image [01:55] <gabriel_> 2 [01:56] <c_14> MrSavage: so either nginx is jailing the exec or it's failing outright, what about exec touch /usr/local/nginx/logs/ffmpeg.log [01:56] <c_14> gabriel_: ye, but it should normally autodetect that in most cases [01:57] <gabriel_> c_14: All I care about is using mpeg compression, the output file doesnt matter because I just want the resulting filesize. Do you have any tips? [01:57] <c_14> just pipe it out raw... [01:57] <c_14> -f raw [01:58] <gabriel_> Ill try that. [01:58] <c_14> should work [01:58] <c_14> if not, use mkv or mpegts [01:59] <gabriel_> http://pastebin.com/2FNmHXrB [01:59] <gabriel_> Thats interesting. Why is input#1 a png_pipe? [02:00] <MrSavage> c_14: exec touch /usr/local/nginx/logs/ffmpeg.log; does not work [02:01] <c_14> gabriel_: ehhh, no clue, but use -f mpegvideo [02:01] <gabriel_> mpegts seemed to work [02:02] <c_14> MrSavage: can you try /tmp/ffmpeg.log ? in case nginx is dropping permissions [02:02] <c_14> (assuming /tmp is 777, which it usually is) [02:10] <MrSavage> c_14: I see it in /tmp/ [02:11] <gabriel_> This is driving me bonkers. I fixed the input error, but Im still only seeing the first frame! :/ [02:12] <MrSavage> c_14: this doesn't give me anything in the file though exec sh -c 'ffmpeg -i rtmp://localhost:1935/$app/$name -vcodec libx264 -acodec copy -f flv rtmp://localhost:1935/small/test &> /tmp/ffmpe.log'; [02:14] <c_14> MrSavage: try exec FFREPORT=/tmp/ffmpeg.log ffmpeg -i rtmp://[..] -c:v libx264 -c:a copy -f flv rtmp://[..] ; [02:16] <MrSavage> c_14: the file isn't being created now [02:16] <c_14> eh, FFREPORT=file=/tmp/ffmpeg.log [02:18] <MrSavage> c_14: still not creating the file [02:18] <c_14> gabriel_: maybe https://trac.ffmpeg.org/wiki/Create%20a%20video%20slideshow%20from%20images… ? [02:18] <c_14> MrSavage: geh, does nginx have an errorlog? [02:18] <c_14> Maybe it's throwing errors there [02:18] Action: c_14 hopes [02:20] <MrSavage> c_14: it does but it doesn't show anything related to ffmpeg [02:22] <c_14> I'm out of ideas. If I could get the ffmpeg console output I might be able to help. Without it, I can only guess. [02:23] <c_14> And my guess is pretty limited to "try updating ffmpeg". [02:24] <gabriel_> -vf/-af/-filter and -filter_complex cannot be used together for the same stream. Haha nice. I will try using filterchain [02:28] <gabriel_> Now Ive confused myself [02:31] <gabriel_> c_14: Do you know how do get round that error? [02:32] <c_14> -filter_complex 'concat=n=2:v=1:a=0,fps=25' [02:32] <gabriel_> Oh, doesnt that set the framerate on the input stream, not the output stream? [02:33] <c_14> hmm, nah [02:51] <gabriel_> c_14: You still there? Im not sure thats doing what I want it to do. The resulting video has no frames now. :P Hey thanks for you help [02:52] <c_14> eh [02:52] <c_14> that shouldn't be happening... [02:54] <gabriel_> Yeah, let me fetch my command for you [02:56] <gabriel_> ['/usr/local/bin/ffmpeg', '-f', 'image2', '-framerate', '25', '-i', 'test_rand/11-small.png', '-f', 'image2', '-framerate', '25', '-i', 'test_rand/12-small.png', '-filter_complex', 'concat=n=2:v=1:a=0,fps=25', '-vcodec', 'mpeg4', '-b:v', '800k', '-f', 'm4v', '-y', 'pipe:1'] [02:56] <c_14> Try changing the format to mpegvideo or mpegts? [02:56] <gabriel_> isnt mpegts for mpeg2? [02:57] <gabriel_> Requested output format 'mpegvideo' is not a suitable output format [02:57] <gabriel_> pipe:1: Invalid argument [02:57] <c_14> geh [02:57] <c_14> Use matroska then [02:58] <gabriel_> This is madness XD [02:59] <gabriel_> Hey were back to just the first frame with matroska [02:59] <c_14> Just want to see if it's the format. [02:59] <gabriel_> nothing else changed but the format [02:59] <c_14> try adding the 2nd png as another input and change concat to n=3 [02:59] <c_14> don't mind the dirty hack [02:59] <gabriel_> :O [03:01] <gabriel_> ['/usr/local/bin/ffmpeg', '-f', 'image2', '-framerate', '25', '-i', 'test_rand/11-small.png', '-f', 'image2', '-framerate', '25', '-i', 'test_rand/12-small.png', '-f', 'image2', '-framerate', '25', '-i', 'test_rand/12-small.png', '-filter_complex', 'concat=n=3:v=1:a=0,fps=25', '-vcodec', 'mpeg4', '-b:v', '800k', '-f', 'matroska', '-y', 'pipe:1'] [03:02] <gabriel_> Still no luck [03:02] <c_14> Still only one frame? [03:02] <gabriel_> Yeah! so crazy [03:03] <c_14> Can you try changing the -framerate 25 s to -framerate 1 s? [03:03] <gabriel_> yeah [03:03] <gabriel_> still a 0 second output with just one frame [03:04] <gabriel_> I must be doing something wrong... [03:05] <c_14> What version you running? [03:05] <c_14> Cause my version throws an error with the command like that... [03:06] <c_14> unless I messed something else up. [03:06] <c_14> nvmd [03:37] <c_14> Right, if gabriel_ comes back and I'm not here, could someone tell him to use the concat demuxer like so: `ffmpeg -f concat -i concat -r 25 -c:v mpeg4 -b:v 800k -f matroska pipe:1' ? (or in the hope he reads the logs...) [03:54] <croco> hello [03:55] <croco> i have problem record audio with ffmpeg [03:55] <croco> the line is 'ffmpeg -f alsa -ac 2 -i pulse -f x11grab -s 800x600 -r 30 -i :0.0 -vcodec libx264 -strict -2 ./out.mp4', have alsa and pulseaudio, all sound function correctly and i listen all (also mplayer), only have problem in ffmpeg for record [03:57] <croco> if without -strict -2 , the ffmpeg requesting obligatory [03:57] <croco> any ideas? [04:05] <gabriel> c_14: Did you get it to work? [04:06] <gabriel> (on your machine) [04:08] <gabriel> ffmpeg version 2.4.2 Copyright (c) 2000-2014 the FFmpeg developers [04:20] <gabriel> Ive upgraded to 2.4.3, ill see if that changes anything [04:23] <gabriel> Hmmm nothing. [04:28] <gabriel> Can anyone help me understand why this is happening? I would expect the output stream to have 3 frames, but it says it only has 1. http://pastebin.com/S9zEdSRH [04:29] <croco> i have a problem with record sound with ffmpeg... said most upper chat [04:30] <klaxa> croco: pick a different audio codec [04:30] <klaxa> to avoid the -strict -2 [04:31] <klaxa> for the audio problem, open pavucontrol and see what stream ffmpeg is recording [04:31] <klaxa> change it to "Analog Input" or something similar [04:31] <croco> what audio? if avoid the -strict, ffmpeg try request [04:32] <croco> i see /proc/asound/cards and arecord -l/-L [04:33] <klaxa> you are using pulse though [04:33] <croco> have alsa and pulse audio, i trying use -f alsa and -i pulse [04:33] <klaxa> so you have to open pavucontrol to control what ffmpeg records [04:33] <croco> i not have pavucontrol [04:34] <klaxa> can you install it with your package manager? [04:34] <klaxa> if you don't want to install it, you will have to use pactl/pacmd and read the manpages [04:36] <croco> klaxa, i have install the pavucontrol and see in the config, not have none of ffmpeg [04:36] <klaxa> start recording with your command [04:36] <klaxa> then open the "Recording" tab in pavucontrol [04:39] <croco> klaxa, ok, and i what checking? [04:39] <klaxa> on the right side there is a dropdown menu [04:40] <klaxa> set that to "Analog Stereo" [04:40] <croco> the record say ALSA plug-in [ffmpeg]:Alsa capture from [04:40] <klaxa> yes, on the right there should be a menu [04:40] <klaxa> just select... something i guess [04:40] <klaxa> try changing it until you are recording what you want to record [04:41] <croco> i have two: "monitor of audio intern" (listen) and "audio intern" (not listen) [04:41] <klaxa> oh, do you want to record your computer's sound? or your microphone's sound? [04:41] <croco> the record from computer sound [04:42] <klaxa> then select "monitor of audio intern" [04:43] <croco> klaxa, change option, and equal not record sound [04:44] <klaxa> huh? that is weird... can you pastebin your command and the output? (use pastebin.com or something similar) [04:44] <croco> ok [04:45] <gabriel> klaxa: do you have a quick second to look at the pastebin above? Im totally at a loss [04:46] <croco> http://pastebin.com/PNvmHN8m [04:48] <klaxa> croco, can you try using -f pulse instead of -f alsa? [04:48] <klaxa> gabriel, sorry, i don't know either [04:49] <gabriel> So weird. I dont understand it at all [04:49] <klaxa> maybe something with mp4 not supporting low framerates? [04:49] <klaxa> have you tried using mkv instead? [04:49] <gabriel> It doesn the same thing with PAL or NTSC framerates [04:49] <gabriel> and with mkv [04:49] <croco> klaxa, the pavucontrol none output in the record [04:50] <klaxa> ffmpeg has to be running to show up in pavucontrol [04:51] <croco> klaxa, http://pastebin.com/xjFec9jN [04:52] <gabriel> klaxa: do you mean m4v or do you mean mkv? [04:52] <croco> klaxa, not show the ffmpeg in pavucontrol [04:52] <klaxa> croco: are you making sure there is sound to record? are you playing music or something? [04:52] <croco> klaxa, yes [04:52] <klaxa> gabriel: mkv, i'm currently trying with some pictures myself [04:52] <gabriel> I dont see mkv in ffmpeg -formats, could you explain what it is? [04:53] <klaxa> croco: can you take a screenshot of pavucontrol's record tab while ffmpeg is running? also change the menu at the bottom right to "All Streams" before doing so [04:53] <klaxa> gabriel: mkv is matroska [04:53] <gabriel> oh, thanks [04:53] <gabriel> Ive tried with matroska as well. D: [04:54] <croco> klaxa, with '-f pulse' only the ffmpeg not record sound (only have #0 for Video and none for sound) [04:55] <klaxa> oh right, i didn't see that... i wonder why that is [04:55] <klaxa> ah, try -f pulse -i default [04:55] <klaxa> instead of -f pulse -i pulse [04:55] <croco> ok [04:58] <gabriel> klaxa: any luck? [04:58] <klaxa> none so far [04:59] <croco> klaxa, with '-f pulse -i default' function sound! :D now only have problem with phase sincronization with Audio and Video [05:02] <klaxa> gabriel, i also only get one frame... weird... [05:03] <gabriel> Do you know who might have some clue as to why this is happening? [05:03] <klaxa> err... not me. [05:04] <klaxa> however, what you could do instead would be: cat file1.jpg file2.jpg file3.jpg > ffmpeg -f image2pipe -i - out.mp4 [05:04] <gabriel> Oh really? [05:04] <gabriel> whats the -i? [05:05] <klaxa> wait... [05:05] <klaxa> i'm sure this used to work [05:05] <klaxa> the - is standard input [05:07] <croco> the sincro between audio and video not import so. something is that function the sound :) Very Thanks klaxa [05:07] <klaxa> np :) [05:08] <klaxa> gabriel, i'm starting to think it's filter_complex's fault [05:09] <klaxa> if i use the concat protocol instead of the filter, i get 3 frames [05:09] <gabriel> ooooo [05:09] <gabriel> Could you explain how to use that? [05:09] <klaxa> if i then add filter_complex to change the framerate, i get one frame [05:09] <gabriel> that sounds promising [05:10] <klaxa> http://pastebin.com/s9Rfa4MM [05:10] <klaxa> that's what my concat.txt looks [05:10] <klaxa> ffmpeg -f concat -i concat.txt -c:v libx264 test.mkv [05:10] <klaxa> that was the command used [05:11] <gabriel> let me try it with the mpeg4 encoder [05:14] <gabriel> Odd& I cant get that to work. [05:15] <gabriel> It runs but I get 0 frames using mkv, 1 frame using m4v [05:17] <gabriel> Oh my gosh. It might be a vlc bug [05:17] <gabriel> klaxa: what are you using to play the resulting mkv file? [05:18] <klaxa> mpv [05:28] <gabriel> is it possible to use the concat protocal from stdin? [05:28] <gabriel> rather than a text file [05:29] <gabriel> yes. [05:37] <gabriel> Haha yes! it feels a little hacky, but Ive managed to work around it [05:37] <gabriel> klaxa: do you think this is worth raising an issue over? [05:37] <gabriel> thanks for you help, by the way. [05:38] <klaxa> maybe try the user-mailinglist first [05:38] <klaxa> i think i would probably file a bugreport [05:38] <klaxa> if it's not a bug, then maybe it should be documented somewhere [05:42] <gabriel> Yeah, its definitely been confusing. I still dont really understand why one works over the other [05:48] <gabriel> Ill put something in the user-mailing list [07:53] <Matador> Hello :) [08:18] <Matador> man cross-compiling is hard :( [08:45] <Matador> How do you set the -- if [[ "$non_free" = "y" ]]; then ; in a script ? [10:20] <aleskandro> Hi all, I would to learn how to code with vdpau and ffmpeg api... Can you link me some examples? I didn't found anything, only documentation but it is very small [10:48] <Matador> What is the recommended container to publish a live stream with libx265 ? (since FLV is not compat with this) [10:55] <davidw> is there a way to make ffserver not take an input feed and just redirect to an ffmpeg that's outputing to rtp itself? [11:24] <DelphiWorld> yo [12:28] <braunr> hello [12:29] <braunr> when ffmpeg is built with an instruction-set specific option such as --enable-mmx, does it mean mmx instructions are used whenever possible, or just that such code is compiled in and can be run if runtime cpu detection allows it ? [12:57] <Mavrik> braunr, afaik --enable-mmx will enable compilation options in a way where MMX is always used [12:58] <Mavrik> I think only libswscale and libx264 can do dynamic detection [13:03] <braunr> Mavrik: hm, not good [13:05] <Mavrik> braunr, what is your usecase and which encoders are you trying to use? [13:08] <DanielSa> hi [14:03] <DanielSa> I want to create a live event, transmit audio and video. I'm planning to use a Zoom h4n audio recorder as usb mic and a GoPro 3 Wi-Fi (I'd like to move the camera around the room), but before buying this equipments, I'm trying with the desktop mic and my D-Link DCS-930L IP Camera. Until now I just managed to write it on a file http://pastebin.com/vryFNUYP . Can anyone show me the way? Thank you in advance. [14:11] <davidw> DanielSa, there are a bundle of streaming guides on the internet.... maybe try one and see where you get stuck? [14:18] <DanielSa> davidw: Yes I did it yesterday, until late at night :D so was when I start doubt if this 'architecture' works, retransmit a video from a IP Camera to the avserver/ffserver. [14:22] <braunr> Mavrik: i maintain a home-made linux distribution at a company, where userspace must be the same binary for a range of different boards [14:23] <braunr> microchip_: for now it's only about decoding and playing videos [14:23] <braunr> woops [14:23] <braunr> Mavrik: ^ [14:24] <braunr> Mavrik: my current idea is to build ffmpeg libraries with all optimizations along with runtime cpu detection, so that the same binaries work everywhere with optimized instructions [14:26] <Mavrik> As I said, I think detection only works for swscale, but don't hold my word for it, I've checked it in version 1.2 last time :P [14:28] <braunr> Mavrik: i understand :) [14:30] <davidw> DanielSa, somehow or another, these things do work [14:30] <davidw> you just have to pound your head against the tablle a lot [14:41] <waNNaBe> anyone got time to help out an unexperienced user with an explanation? [14:41] <braunr> waNNaBe: just ask [14:41] <braunr> (and be patient) [14:44] <waNNaBe> ok, thx. i'm struggling with a simple task. converting an .avi containing a single cvid video stream with rgb24 color model to a .mp4 with h264 encoding [14:44] <waNNaBe> for some reason ffmpeg won't let me use any h264 profile other than high444 [14:45] <waNNaBe> seems like with every profile i use, the encoder tries to use yuv444p color model (which is then a wrong parameter to e.g. baseline profile) [14:46] <waNNaBe> my question to this is, whether it might be a bug or I am just lacking some insights into this stuff [14:53] <braunr> waNNaBe: it's quite specific, you should also try the mailing list i guess [15:01] <waNNaBe> I'll try, thx. [15:18] <pagios> question, if i am capturing audio from a composite in card and video from an ipcam (network) will there be a out of sync between the audio/video? assuming the network is a local network with good latency? i mean i dunno if data trasnfered over the audio cable would be faster than the ipcam which is encoding and sending video over the network cable [15:19] <BtbN> depends on the camera and microphone. There is no general answer to that. [15:20] <pagios> BtbN: what do you mean [15:20] <pagios> how can i make sure they are in sync [15:22] <BtbN> measure the difference and add an apropiate offset. [15:44] <pagios> let me make it a general question, how can i make capturing an audio stream over analog and a videostream over digitial be in harmoney and synch'd [15:46] <BtbN> <BtbN> measure the difference and add an apropiate offset. [15:57] <pagios> BtbN: the difference would be constant? [16:00] <BtbN> maybe [16:00] <BtbN> like i said, there is no general answer to that. [16:11] <MrSavage> Hi everyone, I'm having issues trying to encode a stream's video into libx264 [16:12] <MrSavage> as it's giving me a black screen, but I can receive the stream fine and hear the audio but only see black [16:12] <MrSavage> I'm putting in this ffmpeg -i rtmp://localhost:1935/$app/$name -c:v libx264 -c:a copy -f flv rtmp://localhost:1935/small/test [16:13] <MrSavage> it's giving me, Failed to set value 'libx264' for option 'c:v' [16:14] <MrSavage> i guess i'll try avonv [16:14] <MrSavage> avconv* [16:18] <MrSavage> well i'll try to get it to work with ffmpeg since there's no #avconv channel [16:18] <MrSavage> So far my command is this and still a blackscreen: exec ffmpeg -i rtmp://localhost:1935/$app/$name -vcodec libx264 -acodec copy -f flv rtmp://localhost:1935/small/test [16:19] <BtbN> check ffmpeg -encoders [16:19] <BtbN> if you actualy have the x264 encoder [16:20] <MrSavage> BtbN: I have it [16:20] <MrSavage> also here's my log output so far http://sprunge.us/WZIc [16:20] <BtbN> That's libav, not ffmpeg [16:21] <MrSavage> what do you mean? It says ffmpeg at the top [16:21] <MrSavage> and the command is ffmpeg [16:21] <BtbN> it says Libav in the very fist line [16:23] <MrSavage> but isn't it still ffmpeg? [16:23] <BtbN> no [16:23] <BtbN> just read the first few lines it print [16:23] <BtbN> libav is a fork, that's similar but not the same [16:23] <MrSavage> OH [16:24] <MrSavage> Thanks [16:26] <BtbN> Apart from that, that log looks successfull [16:26] <BtbN> only problem might be that you specified no bitrate, and 0.4Mbit/s is _very_ low for 720p [16:27] <MrSavage> BtbN: would the maxbitrate command be sufficient? [17:03] <pcherry> Can I ask regarding the audio/sync issue in MP4 (x264)? [17:12] <zybi1> hi [17:13] <zybi1> I would like to convert lossless mkv 67 minutes video to 5000kbit/s, what would the command line be for that? [17:14] <zybi1> right now I use something like http://www.pasteall.org/55392 to convert videos to a smaller file, but VIMEO can use 5000kbit/s at it's best [17:16] <Mavrik> zybi1, add -maxrate 5000k -bufsize 5000k [17:16] <Mavrik> to limit maximum bitrate to 5000k [17:16] <Mavrik> maybe take 4800 just in case [17:17] <Mavrik> other stuff looks ok [17:42] <MrSavage> Anyone know where there's a webpage explaining advopts of libx264? [17:42] <MrSavage> I'm trying to put CBR on [18:09] <zybi1> Mavrik: thanks! so maxrate AND bufsize 4800 or only maxrate? [18:20] <kaotiko> hi [18:54] <Toropisco> hi, how can I mux an h.264+AC3 streams from mkv to avchd (.mts or .m2ts)? [19:16] <zybi1> Mavrik: it resulted in a file with overall bitrate of only 3072 kbps and very poor quality [19:19] <zybi1> I would like to convert lossless mkv video (67 minutes, 29giB) to a video file of 5000-10000kbit/s, what would the command line be for that? [19:19] <zybi1> see http://www.pasteall.org/55395 what I tried so far [20:05] <c_14> zybi1: you're missing a zero in your maxrate [20:13] <c_14> Toropisco: ffmpeg -i file -c copy -f mpegts out.m2ts [20:16] <mopp> hey there. I'm trying to live-transcode a transport stream from my sat-ip server (Digibit R1). This is the stream address I use as input: http://pastebin.com/24cQeVbU Unfortunately, ffmpeg doesn't seem to read the input stream. Any hints? [20:20] <c_14> If you curl the stream to a file, can ffmpeg decode said file? [20:21] <mopp> Yes [20:21] <c_14> but ffmpeg -i http://url doesn't work? [20:22] <mopp> no, it just lists all the enabled libs but then nothing happens [20:22] <c_14> !ub mopp [20:28] <mopp> here is the complete console output: http://pastebin.com/fMkP2Bpu [20:30] <c_14> Surround the url in '' [20:30] <c_14> I'm pretty sure the & is causing your shell to fork. [20:33] <mopp> Yup, working now :) only a LITTLE embarassing. Thank you very much! [20:39] <Toropisco> c_14, thanks [20:43] <croco> hello [20:44] <croco> i am the same of day after. [20:44] <croco> have a problem with sincronism between video and sound [20:45] <croco> use -f pulse -i default, and probe -async with non result. [20:47] <croco> not sincro between video and sound, i get '[swscaler @ 0x8369060] Warning: data is not aligned! This can lead to a speedloss' in output text [20:48] <croco> http://pastebin.com/V9B2Rj8P [20:48] <croco> ? [21:08] <croco> what is the problem with sincronism between audio and video?? [21:19] <zybi1> thanks c_14, but now I tried again resulting in only 4051 kbps again - http://www.pasteall.org/55397 [21:23] <c_14> Try lowering the crf? [21:25] <c_14> Also what version of FFmpeg are you using?^ [21:27] <c_14> croco: can your pc encode h264 in realtime? [21:27] <c_14> maybe try adding -preset fast, veryfast or superfast? [21:29] <croco> c_14, yes [21:30] <croco> use this command: 'ffmpeg -f pulse -i default -ac 2 -f x11grab -s 800x600 -r 30 -i :0.0 -vcodec libx264 -strict -2 ./out.mp4' probes with -preset with same result [21:31] <croco> the encode realtime not affect the efficiency of computer [21:31] <c_14> croco: have you tried outputting to mkv? [21:31] <c_14> But to me it looks like pulse is giving you weird pcm. [21:32] <croco> no, use mp4 and detect automaticaly h264 (or libx264) [21:34] <croco> the pcm_s16le is (i see) a default sound input source. [21:35] <croco> also probes change output codec in the same pcm_s16le and getting error of pcm_s16le not exist [21:36] <croco> [aac @ 0x8590ae0] Queue input is backward in time [21:36] <croco> [mp4 @ 0x858ecc0] Non-monotonous DTS in output stream 0:1; previous: 199391, current: 198759; changing to 199392. This may result in incorrect timestamps in the output file. [21:37] <croco> this various lines outputs [21:37] <croco> and get init forever this line: '[swscaler @ 0x84ef060] Warning: data is not aligned! This can lead to a speedloss' [21:38] <croco> retrace sound 0,5 second with video [21:42] <croco> and not most retrace in the time, not needed -preset, the computer is very fast and none lowest for ffmpeg, intentional retrace sound 0,5 second with the video [21:47] <c_14> hmm, no clue. sorry [23:40] <mopp> hey, still trying to live transcode from my Sat-IP server. However, now I can't get "x264opts videofilter" to work. Here the complete output: http://pastebin.com/56WpT8W0 [23:47] <c_14> I think you need video-filter=select_every:2,1 [23:53] <mopp> hm, no, now I get "Error while opening encoder for output stream #0:0 - maybe incorrect parameters [23:53] <mopp> such as bit_rate, rate, width or height" [23:54] <mopp> which is strange, because the stream is recognized correctly with all parameters [00:00] --- Sat Nov 29 2014

1 0

ffmpeg-devel.log.20141127
by burek 28 Nov '14

28 Nov '14

[00:01] <kierank> there is no interlaced coding mode [00:15] <iive> hevc doesn't have interlaced coding mode?! Now i'm liking it even more. [00:17] <Compn> it means non standard interlacing [00:17] <Compn> vendor specific [00:17] <Compn> bwahahaha [00:25] <gnafu> Compn: http://www.reactiongifs.com/r/jZ2J5Yv.gif [01:05] <michaelni> kierank, i didnt mean hevc, but in general that i think as soon as interlaced was invented it was hated ... [01:06] <michaelni> but i could be wrong, iam no archeologist ... [01:07] <kierank> michaelni: yes exactly [01:07] <kierank> that's why the sample is just 2x 1920x544 [01:07] <kierank> there is no mbaff [01:07] <kierank> nor paff (mv chroma compensation part) [03:02] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:9f9440bd8122: avcodec/hevc_ps: Check return code from pps_range_extensions() [10:57] <ubitux> http://lists.libav.org/pipermail/libav-devel/2014-November/065317.html any idea from where this was taken? oO [10:57] <ubitux> git log -p --author=dericed libavformat/mov.c raises nothing [10:59] <nevcairiel> the change isnt in ffmpeg [10:59] <ubitux> yes [11:00] <ubitux> and knowing Dave a bit, he probably didn't do that on libav tree [11:16] <anshul_mahe> Do we have some parity check function that use processor power to do it rather then using bitwise operator [11:27] <BtbN> Which C standard does ffmpeg use? Can i use variable-length automatic arrays? As in C99 [11:28] <kierank> No you can't use vla [11:29] <anshul_mahe> for eg in x86 it use pushf for parity flag [11:29] <BtbN> hm, that makes the code quite a bit more inefficient [11:30] <cone-146> ffmpeg.git 03Martin Storsjö 07master:9326d64ed1ba: Share the utf8 to wchar conversion routine between lavf and lavu [11:30] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:097de4d1d68f: Merge commit '9326d64ed1baadd7af60df6bbcc59cf1fefede48' [11:43] <cone-146> ffmpeg.git 03Martin Storsjö 07master:79fd186a5035: lavf: Use MoveFileEx instead of rename/_wrename on windows [11:44] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:cc663bd13a97: Merge commit '79fd186a5035cf16fc0ab288d8f59da8b1ba2c0e' [12:35] <cone-146> ffmpeg.git 03Martin Storsjö 07master:675ac56b7ee0: Revert "lavf: Don't try to update files atomically with renames on windows" [12:35] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:4ae1d6021be9: Merge commit '675ac56b7ee0f204963fde55295197c5df80aa91' [12:35] <cone-146> ffmpeg.git 03Rong Yan 07master:22e557917d08: libavutil/ppc/util_altivec.h : fix load_with_perm_vec() add marcos vcswapi2s() vcswapc() VEC_SPLAT16() VEC_SLD16() for POWER LE [12:50] <cone-146> ffmpeg.git 03Matthew Oliver 07master:0167fa006044: msvc: Fix compilation errors due to header include order. [14:06] <BtbN> Does ffmpeg realy run the init_static_data function just when enumerating the encoders? I'm not quite sure if i should use it for my nvenc encoder, or just make assumptions about which format(s) are supported. [14:15] <Compn> good question, but i think many devels are asleep right now [14:15] <Compn> BtbN : and i dunno the answer [14:15] <Compn> :) [14:16] <BtbN> The problem is, finding out the supported pixel format involves initializing CUDA and NVENC, and is quite costly and can cause troubles. [14:17] <BtbN> Just assuming it's NV12 works for all hardware that currently exists, but it's far from ideal. [14:17] <BtbN> And i can't see any other option [14:22] <j-b> why far from ideal? [14:22] <j-b> All hw decoders and encoders are using it. [14:22] <j-b> x264 uses it internally. [14:22] <BtbN> Because NVENC does not guarantee that it's supported [14:23] <j-b> oh lol [14:23] <BtbN> It's the assuming that's not ideal. NV12 itself is fine [14:23] <BtbN> It's propably fine to just hardcode it to NV12, as they won't ever drop it. [14:30] <j-b> WTF is this license [14:30] <j-b> are they stupid or what? [14:31] <BtbN> Which license exactly? [14:32] <j-b> nvenc one [14:33] <BtbN> There are 2 which apply to the header [14:33] <j-b> https://developer.nvidia.com/nvidia-video-codec-sdk-license-agreement [14:34] <BtbN> Yeah, that one's pretty annoying [14:34] <BtbN> The one in the headers also is [14:38] <j-b> wow, indeed [14:38] <BtbN> Now the question is.. does it make the ffmpeg binaries nonfree or not? [14:41] <j-b> BtbN: yes, it does [14:43] <cone-146> ffmpeg.git 03Stefano Sabatini 07master:29208e6dcf94: lavu/imgutils: remove redundant and wrong check in av_image_fill_arrays [14:51] <BtbN> Is it ok to call ff_lock_avcodec? Can't see it beeing used anywhere [14:53] <wm4> what are you trying to do? [14:54] <BtbN> I have to do non-threadsafe initialization [14:54] <wm4> <j-b> are they stupid or what? <- it's nvidia [14:54] <BtbN> Like loading the cuda and nvenc library [14:54] <wm4> BtbN: what for? usually this is only needed for stupid global tables [14:54] <BtbN> Yes, i load the nvenc library globaly [14:54] <wm4> is the library always the same? or can the user specify which one to use? [14:54] <BtbN> With a reference counter, to avoid loading it multiple time [14:55] <BtbN> It's allways the very same library [14:55] <wm4> the OS should be doing the refcounting for you [14:55] <BtbN> The OS doesn't know how many times the avcodec codec is in use [14:56] <wm4> BtbN: seems ff_lock_avcodec is already held during init [14:57] <wm4> so codec->init is always restricted to 1 thread in the process [14:57] <Compn> BtbN : where is the api header license ? [14:57] <j-b> wm4: I hate hardware vendors when they do software. It makes no sense, their headers are crap, the API very bad, and their licenses are always "you're not allowed to use this". [14:57] <BtbN> Compn, in the nvEncodeAPI.h [14:57] <BtbN> and at https://developer.nvidia.com/nvidia-video-codec-sdk-license-agreement [14:57] <BtbN> but they are not the same [14:57] <Compn> thx [14:57] <wm4> j-b: isn't that normal? they got to protect their IP!!!!!!11111111111111111111 [14:58] <BtbN> j-b, the NVENC api is actualy quite nice [15:00] <Compn> probably we should just ask nvidia to relicense this header for us :P [15:00] <Compn> since we used to have contacts with them... [15:06] <Daemon404> i wonder whats the use case of nvenc? [15:06] <Daemon404> isnt it quite bad [15:06] <BtbN> It's encoding with nearly no cpu usage. [15:07] <Daemon404> of course [15:22] <nevcairiel> the quality is OKish on recent GPUs [15:22] <nevcairiel> for realtime streaming things or so [15:24] <nevcairiel> its been a todo for me to offer nvenc and quicksync for real-time transcoding uses [15:25] <nevcairiel> but the timetable doesnt have that in it until next year, soo [15:25] <BtbN> The quality is the exact same on all GPUs [15:25] <BtbN> Maxwell just got faster, quality is unchanged [15:26] <BtbN> QSV on linux is horrible [15:26] <nevcairiel> 1hm oh well, then it was just intel that increased quality as well [15:26] <BtbN> There is only libva, and libva has the most horrible api i have ever seen [15:26] <BtbN> You have to manualy generate the h264 nal [15:26] <nevcairiel> i only need windows for now [15:26] <BtbN> but only parts of it [16:01] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:57e5812198aa: avcodec/hevc_ps: More complete window reset [16:01] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:98e8a9e2f238: ffmpeg: Print a debug message if the frame parameters mismatch the context [16:20] <BtbN> Is it ok to av_assert0 after av_malloc, or does the code have to continue eben when oom? [16:22] <j-b> don't assert, just properly quit [16:22] <j-b> assert should be for impossible code-path. [16:23] <BtbN> Hm [16:23] <BtbN> so i need to refactor half my code, just because of that [16:52] <b_jonas> BtbN: beware of http://xkcd.com/292/ [16:52] <BtbN> Well, i'm using goto anyway. Best way to do error handling. [16:53] <BtbN> Better than a growing list of deinitializations in every error case [16:53] <Daemon404> it is standard in most c codebases to use goto error [16:54] <Daemon404> including ours [16:54] <BtbN> It's basicaly the C version of exception handling [16:55] <BtbN> At least i got rid of any non-constant static global variable while i was on it [16:55] <rcombs> yeah, goto is only really considered harmful in languages that _do_ have higher-level exception handling [16:55] <Daemon404> i wouldnt call it the same as exception handling [16:56] <b_jonas> I'm not saying you can't use goto, just be careful with it [16:56] <nevcairiel> Just don't goto upwards [16:56] <Daemon404> i.e. you dont accidentally let a goto slip through 3 calls up [16:56] <Daemon404> and be uncaught [16:56] <Daemon404> ;) [16:56] <nevcairiel> goto fail is common practice in C [16:56] <BtbN> Could be avoided if you realy want to, but that would involve ugly wrapping of every function [16:56] <BtbN> so you return from the inner function instead of goto error [16:57] <BtbN> and the outer function does the cleanup [16:57] <rcombs> nevcairiel: which makes it quite funny to hear people talk about "the goto fail bug" (and mean that one Apple TLS one) [16:57] <Daemon404> rcombs, yeah but goto didnt cause that one [16:57] <Daemon404> allowance of lack of { did [16:57] <nevcairiel> Well if you forget the actual error handler... [17:05] <wm4> rcombs: "goto considered harmful" is often mistaken as a complete damnation of goto [17:05] <wm4> rcombs: but actually this was just to contrast goto spaghetti programming to structured programming, when structured programming was still new [17:06] <kepstin-laptop> my understanding was that it was referring to how people learned to program in older basic variants which had very limited flow control, and mostly relied on goto [17:06] <wm4> there's nothing wrong with simple gotos, they're probably less error prone than "break" or "continue" [17:07] <Daemon404> yes it refers to loops using goto ala BASIC [17:07] <Daemon404> and such [17:07] <Daemon404> anyone who takes any essay as Pure Truth should probably not be programming anyway. [17:09] <rcombs> Daemon404: you should write an essay on that :3 [17:09] Action: wm4 goes writing a function that uses 10 gotos [17:09] <rcombs> it'd be delightfully meta [17:09] <b_jonas> wm4: no, that's a different strip, http://www.xkcd.com/148/ [17:09] <b_jonas> um [17:09] <b_jonas> sorry, I mena [17:09] <b_jonas> rcombs: that's a different strip [17:43] <koda> hey ubitux [17:43] <koda> i got your question on the other chan [17:43] <koda> i took the additional metadata info from https://github.com/l-smash/l-smash/blob/master/lsmash.h#L3861 [18:07] <ubitux> koda: what do you mean "additional"? i don't even see what was the original commit [18:08] <koda> the original commit is the one you pointed me? or i don understand the question [18:08] <ubitux> the author is dericed in your patch [18:08] <ubitux> so i'm wondering where this comes from [18:10] <koda> oh the other one, right [18:10] <koda> i think i saw something on a github pull request that got ignored for a few years [18:12] <ubitux> ok [18:20] <cone-146> ffmpeg.git 03Clément BSsch 07master:5ab882d7283f: avformat/mov: strengthen some table allocations [18:20] <cone-146> ffmpeg.git 03Clément BSsch 07master:92fa1d9231b8: avformat/mov: change conjugation for "Duplicate" [19:21] <cehoyos> Hi! Has anybody here ever compiled x265 on msys? How is it supposed to work? [19:23] <rcombs> ow [19:24] <Daemon404> i think you mean mingw via msys, and its on the wiki [19:24] <Daemon404> https://bitbucket.org/multicoreware/x265/wiki/CrossCompile [19:25] <cehoyos> I don't want to crosscompile, I want to build on msys [19:25] <Daemon404> its the same whether you compile natively with mingw or cross-compile from linux [19:25] <Daemon404> ^^^^^^^6 [19:25] <cehoyos> Ok, so I have cmake in my path and gcc.exe [19:25] <cehoyos> the gcc is able to compile ffmpeg (and passes fate) [19:26] <Daemon404> you might not need a toolchain file for windows, iirc there is an option in cmake-gui if you compile it natively, to use mingw [19:26] <cehoyos> If I run ./make-Makefiles.sh it finds gcc.exe but claims "is not able to compile a simple test program." [19:26] <Daemon404> eh [19:26] <jamrial> x265 has a script to run cmake with the correct arguments depending on you using x86 or x86_64 [19:26] <Daemon404> to quote myself from earlier today, elsewhere [19:26] <jamrial> check build/msys [19:26] <Daemon404> [15:36] < Daemon404> dont use their shell script [19:26] <Daemon404> [15:36] < Daemon404> ever. [19:26] <Daemon404> [15:36] < Daemon404> just use cmake [19:26] <Daemon404> [15:36] < Daemon404> their shell script is for the benefit of their outsourced developers. [19:27] <cehoyos> I originally wanted to build x86_64 (I have a compiler that succeds FFmpeg fate) but atm, I can neither build x86 nor x86_64 [19:27] <cehoyos> What means "just use cmake"? (Sorry, I never used it) [19:27] <Daemon404> cmake-gui.exe is pretty straight forward [19:27] <rcombs> this all looks way more complicated than it should have to be [19:27] <Daemon404> you point it at a build and source dir [19:27] <Daemon404> and select your toolchain [19:28] <Daemon404> rcombs, the gist is: people dont like learning new tools (cmake) [19:28] <Daemon404> (also cmake is bad, but thats orthogonal) [19:28] <cehoyos> Who invented this... (sorry): What is the "source code" is it /x265? or /x265/build? or /x265/source? [19:28] <jamrial> the latter [19:28] <Daemon404> x265/source [19:29] <rcombs> `cd build; cmake "-GUnix Makefiles" ../source; make; make install` [19:29] <rcombs> alternately, -GNinja if you've got that [19:29] Action: Daemon404 barfs [19:29] <jamrial> well no, not for msys [19:29] <jamrial> cmake -G "MSYS Makefiles" [19:29] <rcombs> oh, is that a thing [19:29] <rcombs> why is it different :| [19:30] Action: jamrial shrugs [19:30] <Daemon404> i dont know [19:30] <Daemon404> i hate cmake. [19:30] <Daemon404> a lot. [19:30] <rcombs> well, either MSYS or MinGW or UNIX depending on fuckall, try until one works [19:31] <rcombs> or just cross-compile and maintain some semblance of sanity [19:31] <Daemon404> i run nightly builds of x265, but with msvc and icl [19:31] <jamrial> cross-compiling with cmake sucks [19:31] <cehoyos> cmake-gui now asks for a generator: Which one should I choose: mingw or msys? [19:31] <Daemon404> jamrial, i used to work on openembedded as my job. [19:31] <Daemon404> guess why i hate cmake? [19:32] <Daemon404> cehoyos, good question. it's... unclear [19:32] <jamrial> haha [19:32] <rcombs> I hate MinGW/MSYS [19:32] <rcombs> "it's like cygwin, but half-assed!" [19:32] <cehoyos> Error in configuration [19:33] <jamrial> rcombs: try msys2. it's like cygwin, but not half assed [19:33] <cehoyos> Ok, I chose "mingw" which was wrong, how can I change what I chose? [19:33] <jamrial> clean the build folder [19:33] <rcombs> jamrial: maybe one day the differences will be documented! [19:34] <jamrial> well, for starters, it's in active development :p [19:35] <Daemon404> cehoyos, it worked here by using msys makefiles and choosing to specify native compiler location [19:35] <Daemon404> and pointing it at gcc and g++ [19:35] <Daemon404> and specifying yasm iin the config of course [19:36] <cehoyos> I chose "Msys makefiles" now and I get the identical error as with the shell scripts. [19:36] <cehoyos> gcc.exe not able to compile a simple test program. [19:36] <rcombs> "paths are POSIX-style except when they're not!" [19:36] <jamrial> maybe your msys environment is broken [19:36] <rcombs> "drive letters may or may not be required!" [19:37] <Daemon404> rcombs, thats untrue [19:37] <cehoyos> http://pastebin.com/vJ7MKyU9 [19:37] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:970a8f1c256f: avcodec/mjpegdec: Fix integer overflow in shift [19:38] <Daemon404> cehoyos, works here with msys+mingw for me... [19:39] <cehoyos> What did you do? [19:39] <cehoyos> I mean, did you run cmake-gui from msys? [19:39] <cehoyos> Or from cmd.exe? [19:39] <Daemon404> i double clicked it. [19:40] <Daemon404> x265 [info]: build info [Windows][GCC 4.9.1][64 bit] 8bpp [19:40] <jamrial> does it not work if you run 'cmake -G "MSYS Makefiles" <path to source>' first, then 'cmake-gui <path to source>'? both on msys and in the build folder [19:40] <cehoyos> How did you pass the path to the toolchain to cmake-gui [19:40] <cehoyos> ? [19:41] <cehoyos> Assuming you are using Hendriks compiler or aren't you? [19:41] <Daemon404> shouldnt matter [19:41] <Daemon404> you select 'specify native toolchain' in the wizard thing [19:41] <cehoyos> The toolchain doesn't matter? [19:41] <Daemon404> and you get file open dialogs to specify teh paths to gcc and g++ [19:42] <Daemon404> jamrial, that should work too [19:43] <cehoyos> jamrial: I get the identical output as above: gcc is not able to compile a simple test program. [19:43] <cehoyos> See http://pastebin.com/vJ7MKyU9 [19:44] <jamrial> then there's something wrong with your gcc toolchain or msys environment [19:44] <cehoyos> The same toolchain passes fate for FFmpeg... [19:44] <cehoyos> Of course, you are right, I just wonder what I should do. [19:45] <cehoyos> My msys is antique, I just wonder how this affects cmake-gui... [19:45] <Daemon404> it works without gui too [19:45] <Daemon404> running cmake -G "MSYS Makefiles" in msys works fine [19:47] <cehoyos> The following shows if I run cmake-gui with a double-click: [19:47] <cehoyos> CMake Error: CMake was unable to find a build program corresponding to "MSYS Makefiles". CMAKE_MAKE_PROGRAM is not set. You probably need to select a different build tool. [19:47] <Daemon404> ... is gcc even in your path [19:47] <Daemon404> actually it doesnt need to be [19:48] <Daemon404> because you specify the toolchain path [19:48] <cehoyos> It is in my msys path, if I double-click cmake-gui, how can gcc be in my path? [19:48] <jamrial> cmake doesn't recognize "MSYS Makefiles"? [19:48] <cehoyos> Or do I just misunderstand? [19:48] <cehoyos> Where can I specify the toolchain path (I believe I asked that above)? [19:49] <Daemon404> there is an option to specify it when you set the generator [19:49] <Daemon404> but as i said, you can just 'cmake -G "MSYS Makefiles"' from the same msys prompt you bult ffmpeg in [19:52] <cehoyos> Daemon404: http://pastebin.com/cXvyDcyC [19:53] <Daemon404> your environment seems broken then... [19:53] <cehoyos> Ok (of course), but how can I proceed? [19:53] <cehoyos> And why does fate succeed? [19:54] <Daemon404> i have no idea how your stuff is set up [19:54] <cehoyos> A very old msys installer [19:55] <jamrial> did you successfully compile ffmpeg on this environment? [19:56] <Daemon404> he said he did [19:56] <cehoyos> No, I mean yes, as said above, fate passes both for the native (old) compiler on my msys environment and with Hendrik's current compiler. [19:56] <cehoyos> I have a second Windows box, which version of msys should I install there? [19:56] <Daemon404> idea: make sure your build directory is empty [19:57] <Daemon404> especially no *.cache files [19:57] <cehoyos> I removed it after every try. [19:57] <jamrial> latest msys using their installer, or just use msys2 [19:57] <cehoyos> That is my question. [19:57] <jamrial> both should work. but if you want something in active development, go with msys2 [20:01] <cehoyos> Allow me to try again: What did you use when you did a successful install last time, and would you recommend it? [20:01] <cehoyos> Daemon404: Would you mind uploading a 64bit x265.a with 10bit support? [20:03] <jamrial> i used msys until a month or so ago, then switched to msys2. i compiled x265 successfully with both [20:04] <jamrial> and i'd say use msys2 [20:05] <cehoyos> Ok, thank you. If you can upload a x265.a (64bit, 10bit), I would be grateful. [20:08] <cone-146> ffmpeg.git 03Vincent Bernat 07master:5269cef4087f: avformat/udp: Allow to specify DSCP class [20:08] <cone-146> ffmpeg.git 03Vincent Bernat 07master:d0f8b94b432e: avformat/rtpproto: Allow to specify DSCP class [20:16] <Daemon404> cehoyos, why 10bit? [20:17] <Daemon404> x265 RD currently has issues in 10bit mode [20:17] <cehoyos> TherAk [20:17] <Daemon404> ? [20:18] <cehoyos> wrong window [20:18] <cehoyos> what is rd? [20:18] <Daemon404> rate distortion [20:18] <Daemon404> basically, 10bit is not prod-ready. [20:18] <cehoyos> So there is no reason to test it? [20:19] <Daemon404> i dont think so, yet [20:19] <cehoyos> Ok, Thank you. [20:19] <Daemon404> its on their list to work on [20:19] <cehoyos> Any news from the prores file? Where you able to run ffmpeg -i on it? [20:19] <cehoyos> Were... [20:20] <Daemon404> koda has it, i never remembered to poke him [20:20] <cehoyos> Perhaps you could do so now? [20:20] <Daemon404> he's not around, it's dinner time here [20:20] <Daemon404> ill just leave a message. [20:20] <cehoyos> I will ask again, thank you! [20:20] <cehoyos> gtg [20:35] <Daemon404> maybe in 2025 the aac patch will be pused [20:51] <rcombs> is there any need for more samples that have trouble on the current AAC encoder but work with the patch (and don't give assertion failures)? [20:51] <rcombs> 'cause I've got one [21:12] <nevcairiel> We don't need any proof that the old one sucks, we know! :P [21:12] <rcombs> nevcairiel: do you need more easily-noticeable proof that it's better now? [22:22] <cone-146> ffmpeg.git 03Anton Khirnov 07master:2443e522f005: lavu: add wrappers for the pthreads mutex API [22:22] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:476027800721: Merge commit '2443e522f0059176ff8717c9c753eb6fe7e7bbf1' [23:00] <Zeranoe> ramiro: Any progress on that DeckLink issue? [23:21] <cone-146> ffmpeg.git 03Lukasz Marek 07master:568853b8f533: lavf/ffmdec: add common options to recommended encoder configuration [23:21] <cone-146> ffmpeg.git 03Lukasz Marek 07master:500d6088618a: lavf/ffmenc: store recommended encoder configuration [23:22] <cone-146> ffmpeg.git 03Lukasz Marek 07master:3ff39901049f: ffmpeg_opt: make use of recommended encoder configuration [23:30] <cone-146> ffmpeg.git 03Lukasz Marek 07master:f00e9c4b10f5: lavu/opt: add consts where possible [23:42] <cone-146> ffmpeg.git 03wm4 07master:fbd6c97f9ca8: lavu: fix memory leaks by using a mutex instead of atomics [23:43] <cone-146> ffmpeg.git 03Michael Niedermayer 07master:6db8cd8f37b7: Merge commit 'fbd6c97f9ca858140df16dd07200ea0d4bdc1a83' [23:59] <cone-146> ffmpeg.git 03Kieran Kunhya 07master:96fda42a8f9b: vf_interlace: get rid of useless loads [00:00] --- Fri Nov 28 2014

1 0

ffmpeg.log.20141127
by burek 28 Nov '14

28 Nov '14

[00:02] <quixotic> i'll try that. thanks everybody for the help. you're great [00:02] <llogan> anyway, if you just do "ffmpeg -i input -f md5 -" for each input you'll see that the hash is the same [00:03] <llogan> each input being your new ogg files [00:03] <quixotic> i see, i'll use this info in my new automation process. thanks [00:03] <llogan> and you could probably specify the desired stream with -map if you want to just check a particular stream [00:06] <llogan> quixotic: and make check that flac isn't the default encoder. i think it used to be default for ogg even with libvorbis supported. (next time provide the complete console output) [00:06] <llogan> *make sure [00:17] <bakers> quixotic: Did you solve your problem? I'm curious now [00:40] <quixotic> bakers: i'll work around the issue by comparing the ffmpeg hashes to see if the file is actually the same. haven't found out how to specify the serial number [01:43] <samthewildone> How do I use your application to compress a 1GB video to 500mb ? [01:46] <MDTech-us_MAN> samthewildone: http://superuser.com/questions/4244/how-do-i-reduce-the-size-of-a-huge-mp4-… [01:51] <llogan> samthewildone: https://trac.ffmpeg.org/wiki/Encode/H.264#twopass [02:56] <Snaggle> I have a movie whose source is 133 frames, but which apparently got exported as 60 fps BUT instead of staying 133 frames played over 2 seconds, it became 8025 frames played over 133 seconds. I do not have access to the source. Is it possible to use ffmpeg to keep every 133rd frame ? [03:02] <Snaggle> ahhh. framestep filter... now to learn how to use it. [09:54] <indieross> so i'm trying to compile ffmpeg on an f21 machine with an nvidia gpu and i'm having trouble with --enable-opencl [09:54] <indieross> i'm getting a not found error, i have files in /usr/include/CL and /usr/include/opencl-utils [09:55] <indieross> am i missing any deps or something? i;m pretty sure i have every i need installed [09:55] <indieross> everything i* [09:55] <sacarasc> What does the configure log say? [09:56] <indieross> error: OpenCL/cl.h: No such file or directory [09:57] <indieross> should i link one of the cl.h's in the dirs i previously mentioned to that dir? [09:58] <sacarasc> So it's looking in /usr/include/OpenCL/, not in either of the places you have the files. [09:59] <indieross> i got that [09:59] <sacarasc> I can' help well wih a broken keboard, sorr. :D [10:00] <Hfuy> Hello. [10:00] <Hfuy> I'm in the process of converting an AVI to various quicktime-wrapped output formats. [10:01] <Hfuy> In each case, black becomes grey; it looks very much like a luma range encoding glitch. [10:01] <Hfuy> Are there any commandline switches that control this behaviour? [10:03] <sacarasc> indieross: I believe it is possible to set where configure looks for things. [10:04] <indieross> yeah, i think ill try changing it to /usr/include/opencl-utilc/CL/ [10:05] <Akagi201> Could ffmpeg modify the timecode of a flv stream? [10:07] <Hfuy> I've just tried -color_range but it seems to have no effect on anything. [10:10] <indieross> yuss, found the missing dep [10:11] <indieross> lets see how this nvenc works out [10:14] <K4T> how I can send ONE stream with FFMPEG to multiple udp multicast addresses? [10:15] <K4T> http://pastebin.com/dKjV8zQS here is my command line [10:16] <K4T> -f mpegts -flush_packets 0 "udp://239.239.4.100:1234?pkt_size=1316" [10:16] <K4T> I would like to use something like this: -f mpegts -flush_packets 0 "udp://239.239.4.100:1234?pkt_size=1316, udp://239.239.4.101:1234?pkt_size=1316, udp://239.239.4.102:1234?pkt_size=1316" [10:16] <K4T> is it possible? [10:18] <viric> hi! [10:19] <viric> I'm having troubles again... [10:19] <sfan5> K4T: https://trac.ffmpeg.org/wiki/Creating%20multiple%20outputs#Duplicateoutputs [10:20] <viric> I have a mts as input, and an image2 with loop, with overlay. ffmpeg does not end... starts dropping [10:20] <viric> at the end of the mts, all frames go to "drop", and never ends. '-shortest' does not help. [10:20] <viric> How can I make it finish once the mts finishes? [10:20] <viric> (I'm just putting an image cover to the first 4 seconds of the mts) [10:25] <K4T> sfan5, thx [10:31] <viric> shortest=1 in overlay does not help ... [10:31] <viric> weird [10:33] <viric> ah, updating ffmpeg fixed it. [10:33] <viric> \o/ [10:44] <bro45> hey [10:44] <bro45> I need an help compiling ffmpeg [10:44] <bro45> I have configured with --enable-shared [10:45] <bro45> now when I make it says: /usr/bin/ld: libavutil/display.o: relocation R_X86_64_PC32 against undefined symbol `hypot@@GLIBC_2.2.5' can not be used when making a shared object; recompile with -fPIC [10:45] <bro45> what can I do? [10:46] <c_14> add -fPIC to your CFLAGS [10:46] <bro45> how? [10:48] <c_14> either with --extra-cflags="-fPIC" in your configure line or with export CFLAGS="-fPIC" [10:48] <c_14> making sure to keep whatever's already in your cflags [11:05] <bro45> usr/bin/ld: /usr/local/lib/libavformat.a(rmsipr.o): relocation R_X86_64_32 against `.rodata' can not be used when making a shared object; recompile with -fPIC [11:06] <bro45> anyone knows what is this problem when I compile vlc? [11:10] <KeyboardNotFound> Hi [11:11] <Bugson> guys where can I found quality scale for codecs? some man page or so? [11:13] <davidw> Hi, I'm using the Ubuntu PPA, and am trying to stream to UDP: ffmpeg -f x11grab -s 1600x900 -r 50 -vcodec mpeg2video -f mpegts udp://10.0.0.55:9002 - as per an example I found on the web. [11:13] <davidw> However, I get: Output #0, mpegts, to 'udp://10.0.0.55:9002': Output file #0 does not contain any stream [11:15] <c_14> add -map 0 before -f mpegts [11:16] <Hfuy> Are any questions ever actually answered in here? I keep seeing people ask, but very few responses. [11:16] <c_14> depends on the time/question [11:17] <davidw> c_14, it gives me: Invalid input file index: 0. [11:18] <davidw> c_14, fair enough: http://pastebin.com/eTDE3zYs [11:19] <davidw> I'm looking at this web page for my examples, BTW: http://www.waitwut.info/blog/2013/06/09/desktop-streaming-with-ffmpeg-for-l… [11:23] <davidw> More googling reveals commentary about an audio stream. Don't have/want that.... could it be a problem? [11:47] <davidw> do I have to start the client at the same time I start the 'server' ? [12:49] <c_14> davidw: you don't have a -i input option [12:49] <c_14> https://trac.ffmpeg.org/wiki/Capture/Desktop [13:02] <DanielSa> hi [13:16] <davidw> damn, that's still way laggy [13:20] <davidw> how does webrtc manage to be so fast -( [14:19] <davidw> Ok, I wrote it up here: http://superuser.com/questions/845699/get-webrtc-like-latency-with-ffmpeg - any ideas are welcome [14:26] <samfisher> hi, i have ffserver running, ffmpeg doing ffmpeg -f video4linux2 -i /dev/video0 http://localhost:8090/webcam1.ffm [14:26] <samfisher> but I don't know how to access video feed [14:27] <samfisher> the remote machine connects with mplayer and sais No stream found to handle url http://192.168.255.103:8090/webcam1.ffm [14:29] <pcherry> Hi, I have audio/video sync problem (MJPEG -> H264) [14:29] <pcherry> ffmpeg -i mjpeg.mov -c:v libx264 -preset fast -crf 13 -pix_fmt yuvj420p -c:a libfdk_aac -b:a 192k output.mp4 [14:30] <pcherry> audio is ahead of video, there is about 0.2sec cut [14:31] <pcherry> if I encode just audio (MJPEG -> AAC) it is fine [15:15] <pcherry> Can anybody help me with a/v sync problems? [15:19] <kevincheri> guys, new here and I need help converting rtsp to rtmp and stream via web [15:20] <kevincheri> ffmpeg -i "rtsp://url/" -vcodec copy -f flv -r 25 -s 1920x1080 -an "rtmp://127.0.0.1:1935/live/livestream11" [15:20] <kevincheri> I am using that in the web script [15:20] <kevincheri> the error is Press [q] to stop encoding [15:20] <kevincheri> [flv @ 0x15c2560]st:0 error, non monotone timestamps 1 >= 1 [15:20] <kevincheri> av_interleaved_write_frame(): Operation not permitted [15:20] <kevincheri> can somebody help please? [17:44] <davidw> is there a way to make ffserver simply reroute to an ffmpeg that is outputting rtp already? I want to reduce latency [18:47] <smo__> hi ! i m building my ffmpeg now but it won t build ffplay ! (i added the --enable-ffplay) [18:47] <smo__> an idear? [18:51] <BtbN> You need sdl. [19:20] <Borys> Hi there :) [19:21] <Borys> It's possible to stream line-in input to icecast server? I use now darkice to do it, but i looking for method to decrease delay [19:22] <Borys> So i thought I can try encode audio input via ffmpeg and stream it icecast on my vps [19:22] <c_14> https://trac.ffmpeg.org/wiki/Capture/ALSA [19:22] <c_14> https://ffmpeg.org/ffmpeg-protocols.html#Icecast [19:26] <Borys> I already reading it, but I;m not sure about syntax, ffmpeg -f alsa -i hw:0,2 icecast://[username[:password]@]server:port/mountpoint will works? [19:31] <c_14> should [19:41] <DanielSa> c_14: very nice tip about ALSA [19:45] <Popara> hello i have an avi movie that i convert it to mp4 using -codec copy -f mp4 test.mp4 [19:45] <Popara> the new movie has dropped frames and its a bit laggy and id ont understand why,i will post the command output to pastebin though i didnt see something strange there [19:47] <Popara> http://pastebin.com/QmdnxHLR [19:48] <kepstin-laptop> well, that 'pts has no value' message looks strange, and seems like it would cause stuff like dropped frames or laggy playback [19:48] Action: kepstin-laptop has no idea why that would happen or how to fix it [19:48] <Popara> the whole movie is laggy though [19:57] <jack_thompson> do i see that right that ffserver only supports flv? [19:58] <jack_thompson> is webm even suitable for expanding live streams? because i have no problem when i stream it from a file with ffserver.. [19:58] <jack_thompson> but when i stream something that is live encoded it doesn't work... [19:59] <jack_thompson> the only thing that always works is flv [20:04] <azk> You can stream with webm [20:04] <azk> Oh sorry, I misunderstood [20:05] <jack_thompson> no... webm is complete crap... it is too slow... and even when i use a server that takes way too much electricity (so that webm encodes in realtime)... the encoded stream still stutters [20:05] <jack_thompson> the only case where i have seen webm working was when it was streamed directly from a file... without any touching.. [20:06] <azk> Well, I have it working fine on my end [20:06] <jack_thompson> can you pm me the [20:06] <jack_thompson> <stream> part of your config? [20:06] <azk> I wrote it up some time ago [20:06] <azk> sec [22:07] <smo__> hello, i try to open some rtsp links served by my adsl box exemple rtsp://mafreebox.freebox.fr/....... when opening it i have tcp error can t resolve hostname mafreebox.freebox.fr [22:07] <smo__> what do i miss in my ./configure build options to fix this ? [22:08] <smo__> (same link works well in vlc) [22:09] <jack_thompson> dude told me i'd had to adjust preroll and leys [22:09] <jack_thompson> wrong channel [22:11] <c_14> smo__: try installing nscd and see if that works [22:12] <c_14> As in, see if it works once it's installed. [22:12] <smo__> ok i try [22:13] <smo__> yup it works [22:13] <smo__> hummm soo it s os relevant ? [22:15] <c_14> I don't know the details, I just know that that might help... [22:15] <smo__> ok thanks but strange :p [23:34] <Demon_Fox> The site really cleaned up [23:35] <Demon_Fox> Is the h.264 decoder closed source? [23:35] <c_14> No? [23:36] <Demon_Fox> I asked because [23:36] <Demon_Fox> I was unsure [23:57] <Jack64> hey guys, does anyone have any experience decrypting hls ts files? [00:00] --- Fri Nov 28 2014

1 0

ffmpeg-devel.log.20141126
by burek 27 Nov '14

27 Nov '14

[01:11] <cone-477> ffmpeg.git 03Carl Eugen Hoyos 07master:5badcdf20d98: Rename sync() functions in libavformat. [01:11] <cone-477> ffmpeg.git 03Benoit Fouet 07master:4acefd25215a: avformat/apngdec: account for blend and dispose operations. [01:55] <llogan> michaelni: we could always ask users on twitter to come up with a name. they'll probably like that, and we'll get a free name such as "Ur mom". [02:18] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:2ad38c6e02cc: avformat/mxfdec: dont ask for samples with field dominance 0 anymore [02:25] <kierank> what's the difference between yuv2yuvX and yuv2planeX [02:32] <kierank> ah the latter has accurate rounding [04:30] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:e5c01ccdf5a9: avcodec/flacdec: Call ff_flacdsp_init() unconditionally [04:30] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:5f30522894d1: avcodec/flacdec: fix off by 1 error [06:05] <Zeranoe> Perhaps another DeckLink issue: http://ffmpeg.zeranoe.com/forum/viewtopic.php?f=10&t=1823#p7554 [12:00] <cone-393> ffmpeg.git 03Martin Storsjö 07master:ee37620b6ae4: movenc: Add a flag for indicating a discontinuous fragment [12:00] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:b78074fd13ad: Merge commit 'ee37620b6ae4783cda637408422044b2d14a688c' [13:34] <cone-393> ffmpeg.git 03Martin Storsjö 07master:234fb81e3145: movenc: Expose the fragment index as an avoption [13:34] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:0df95fa327c8: Merge commit '234fb81e3145e9c9aec4ec16266676fab7dc21fa' [13:34] <cone-393> ffmpeg.git 03Benoit Fouet 07master:8b8cb30d1188: avformat/apngdec: use packet pts and duration instead of altering stream framerate. [14:59] <kierank> michaelni: can you apply my v210 stuff [15:18] <michaelni> kierank, is there a patch that took care of James comments ? [15:19] <michaelni> 1122 23:10 James Almer (5.9K) [15:30] <kierank> michaelni: ok I will squash the two patches and resubmit [16:14] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:79ceaf827be0: avcodec/pngdec: Check IHDR/IDAT order [17:00] <kierank> michaelni: patch sent with changes [17:04] <kierank> hah and now libav has a patch [17:39] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:9a53707e86eb: avcodec/pngdec: Fix paeth prediction with small images [17:46] <michaelni> kierank, reply sent, seems it breaks fate [17:47] <kierank> how do I run that single fate test? [17:47] <Daemon404> make fate-vsynth1-v210 [17:48] <kierank> wtf is that test doing [17:49] <ubitux> use V=1 [17:49] <kierank> it compares a 422 file by converting it to 420 [17:49] <kierank> i did use V=1 [18:08] <jamrial> it fails even without running the asm [18:11] <saste> arwa: http://trac.ffmpeg.org/wiki/SponsoringPrograms/OPW/SponsoringPrograms/OPW/2… [18:11] <saste> fast&furious edit, I have to dig it more myself [18:12] <saste> about eq/eq2, James Darnley was working on it, you should contact him and ask for more [18:12] <saste> arwa: ^ [18:13] <kierank> is an upsampling to 422 and a downsampling back to 420 meant to be bitexact? [18:14] <nevcairiel> i would be surprised [18:15] <nevcairiel> maybe if you use a point scaler for downsampling or something :p [18:25] <kierank> a lot of pixels are off by one [18:26] <Compn> who said we had bit exact colorspace conversion? :P [18:27] <Compn> it would be interesting to see if the c version was bitexact vs the asm [18:27] <wm4> basically libswscale is being liquid shit _again_ [18:28] <Compn> or rather , no one is making a proper bugreport so it can be fixed [18:28] <Compn> instead wasting time on complaining :) [18:55] <kierank> i'm pretty sure there's an off by one somewhere [18:55] <kierank> or a clipping ug [18:59] <michaelni> kierank, in your code or swscale ? [19:00] <kierank> don't know [19:02] <kierank> ./ffmpeg -pix_fmt yuv422p -s 1280x720 -i in.yuv -vcodec v210 -y out.mov && ./ffmpeg -i out.mov -pix_fmt yuv422p10 -y out.yuv && ./ffmpeg -pix_fmt yuv422p -s 1280x720 -i in.yuv -f rawvideo -pix_fmt yuv422p10 -y in10.yuv && md5sum in10.yuv out.yuv [19:02] <kierank> in10.yuv and out.yuv match [19:02] <kierank> but in.yuv and out.yuv don't [19:02] <kierank> so there's something going on somewhere [19:03] <kierank> some broken colourspace conversion [19:05] <kierank> michaelni: ^ [19:07] <kierank> the c seems to be ok [19:07] <kierank> the asm doesn't [19:24] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:1b5d11240692: avformat/mov: Fix memleaks for duplicate STCO/CO64/STSC atoms [19:24] <cone-393> ffmpeg.git 03Michael Niedermayer 07master:1d3a3b9f8907: avcodec/rawdec: Check the return code of avpicture_get_size() [20:12] <kierank> Gramner: ping [20:12] <Gramner> pong [20:12] <kierank> Gramner: do you mind being a second pair of eyes and looking at my v210 8-bit chroma shuffles? [20:12] <kierank> there's something subtly wrong with the code [20:13] <Gramner> yeah sure, in a few minutes [20:13] <kierank> thanks [20:22] <kierank> ah I think it's because I don't merge the luma and chroma correctly [20:23] <kierank> i mean the u and v planes actually [20:24] <michaelni> kierank, i found the issue [20:25] <kierank> the shuffle is wrong in my asm for one thing [20:26] <michaelni> yes, posted correction to ml a sec ago [21:14] <cone-393> ffmpeg.git 03Kieran Kunhya 07master:36091742d182: v210enc: Add SIMD optimised 8-bit and 10-bit encoders [21:57] <ramiro> a user got in touch with me asking to add strftime to the segment muxer [21:57] <ramiro> the segment muxer uses av_get_frame_filename() to get the filename [21:58] <ramiro> would it be A Good Thing to make av_get_frame_filename() use strftime, or better to just add strftime to the segment muxer? [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:b424e67abf0d: avfilter/signalstats: fix different buffers for out frame if burn is enabled [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:c7e8f610f281: avfilter/signalstats: remove pointless sub filter init system [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:56b98dfc4f09: avfilter/signalstats: integrate height loop into subfilters [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:cc5c667eb13b: avfilter/signalstats: reindent after previous commit [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:fad6865748c7: avfilter/signalstats: fix repitition/repetition typo [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:9db78a296c37: avfilter/signalstats: add slice threading for subfilters [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:7acbd56a8a99: avfilter/signalstats: isolate sat hue computation metrics in a function [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:82dda8e4eaa6: avfilter/signalstats: add threading in compute_sat_hue_metrics [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:9cb1d81a60bd: avfilter/signalstats: localize a few variables [21:59] <cone-393> ffmpeg.git 03Clément BSsch 07master:c2ea7069c489: avfilter/signalstats: re-use yuv/yuvu/yuvv vars in diff [22:35] <Zeranoe> Perhaps another DeckLink issue: http://ffmpeg.zeranoe.com/forum/viewtopic.php?f=10&t=1823#p7554 [22:42] <nevcairiel> the free of the name in that line should use SysFreeString, not free() [22:42] <nevcairiel> on win32 anyway [22:42] <nevcairiel> no idea about the other operating systems [22:43] <nevcairiel> (obviously not SysFreeString on other operating systems) [22:43] <ramiro> nevcairiel: thanks [22:44] <ramiro> I wonder why I chose free... [22:44] <nevcairiel> on osx its a CFString, whatever is required to free that, i don't know! [22:45] <ramiro> Zeranoe: care to make a patch? [22:47] <Zeranoe> ramiro: Pretty busy right now [22:49] <ramiro> nevcairiel: how about you? =) [22:50] <nevcairiel> i dont write patches i cant even compile [23:44] <cone-393> ffmpeg.git 03Lukasz Marek 07master:e98aced69955: ffserver_config: cosmetic: simplify functions calls. [23:44] <cone-393> ffmpeg.git 03Lukasz Marek 07master:d57a6d20875f: ffserver_config: cosmetic: move line_num into FFServerConfig [23:44] <cone-393> ffmpeg.git 03Lukasz Marek 07master:f61cb6453d22: ffserver_config: map ffserver options to AVOptions [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:6c2ed67c2f6f: ffserver_config: remove ffserver_apply_stream_config function [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:aaf6cc925f7f: ffserver: allow skip setting defaults [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:ec6e035b8b1f: ffserver: export recommented encoder configuration [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:3cb0bec6870c: ffserver: dont leak child arguments [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:3d0867917faa: ffserver: dont leak pb_buffer [23:45] <cone-393> ffmpeg.git 03Lukasz Marek 07master:345cfd04d093: lavc/options: fix leaks in avcodec_free_context [23:54] <rcombs> >HEVC >interlaced [23:54] <rcombs> weren't there plans to not even support that [23:57] <gnafu> >broadcast [23:57] <gnafu> >industry [23:57] <gnafu> :-P [23:57] <cone-393> ffmpeg.git 03Benoit Fouet 07master:e2b8b4caf6c0: avformat/apngdec: validate frame dimensions. [23:58] Action: michaelni has the suspicion that the plans to kill interlaced are as old as interlaced formats [00:00] --- Thu Nov 27 2014

1 0

ffmpeg.log.20141126
by burek 27 Nov '14

27 Nov '14

[00:04] <llogan> shitty-ass proprietary security camera videos... "let's take a format and shit on it so it can't be played" [00:05] <ScottMichaud> Yeah... [00:23] <ubuntuaddicted> if anyone else cares, i did end up getting ffmpeg compiled with libnvenc enabled (in Arch) using this github fork https://github.com/agathah/ffmpeg_libnvenc along with the Nvidia NVENC SDK for windows because a single header file was neeeded. [01:22] <ubuntuaddicted> how would I find out what codec are available in my ffmpeg I have installed. when I run ffmpeg --version I get ffmpeg: error while loading shared libraries: libx265.so.35: cannot open shared object file: No such file or directory [01:23] <c_14> Is libx265.so.35 in your LD_LIBRARY_PATH ? [01:24] <ubuntuaddicted> c_14, no idea how to find that out. sorry, i'm pretty much a noob that's been getting help from others [01:25] <ubuntuaddicted> c_14, i know that when I compiled ffmpeg I had x265 as a depends and had --enable-libx265 within the configure line [01:25] <c_14> echo $LD_LIBRARY_PATH; locate libx265.so; find / -iname 'libx265.so.*' [01:26] <ubuntuaddicted> c_14, can i narrow it down to /usr/ instead of /? [01:26] <c_14> On arch, probably. [01:27] <c_14> They moved everything to /usr a while ago anyway. [01:27] <c_14> Though, check /home as well [01:27] <c_14> if it's not in /usr [01:27] <c_14> In case you compiled but didn't install libx265 [01:27] <c_14> or installed in a non-standard path [01:28] <ubuntuaddicted> c_14, /usr/lib/libx265.so [01:28] <ubuntuaddicted> /usr/lib/libx265.so.35 [01:28] <llogan> ldd /path/to/ffmpeg [01:28] <c_14> And pastebin it. [01:28] <ubuntuaddicted> so just ldd /usr/bin/ffmpeg? [01:29] <c_14> And put that on a pastebin, yes. [01:29] <ubuntuaddicted> https://clbin.com/eftgn [01:31] <c_14> Is /usr/lib/libx265.so.35 a regular file? [01:32] <ubuntuaddicted> c_14, that file doesn't exist [01:32] <c_14> Didn't you just tell me it did? [01:32] <ubuntuaddicted> ls -la /usr/lib/ | grep libx265 returned only 2 files. libx265.so > libx265.so.37 and libx265.so.37 [01:33] <c_14> recompile ffmpeg against the newer version of libx265 [01:33] <c_14> Or (and I don't recommend this) link libx265.so.37 to libx265.so.35 [01:33] <c_14> Probably just ignore I said that. [01:34] <llogan> current x265 package in Extra will provide libx265.so.35. not sure where you got 37 [01:34] <ubuntuaddicted> c_14, my locate db was old, redoing the search reveals no /usr/lib/libx265.so.35 [01:35] <c_14> Or install that package from Extra [01:35] <ubuntuaddicted> oh, apparently that's from x265-hg [01:36] <llogan> either use the package from Extra, or re-compile ffmpeg [01:36] <ubuntuaddicted> ok, all fixed. thank you [01:36] <ubuntuaddicted> llogan, installing x265 from extra solved it [01:37] <llogan> most users should need to use any CVS type package in AUR if there is an equivalent package in the official repos [01:37] <llogan> should *not* [01:42] <ubuntuaddicted> llogan, understood. i had misread a comment on the obs-studio-git package in the AUR. so it was totally my own fault. thanks for helping me guys, i can now utilize my gtx 760 for encoding with ffmpeg. :) [01:46] <sacarasc> I thought that video cards sucked for encoding video still? Has this changed? [02:32] <foonix> anyone got idea if this works and what are the proper tags names? ffmpeg -i INPUT.mkv -metadata:s:v:0 "Matrix coefficients=BT.709" -metadata:s:v:0 "Color primaries=BT.709" -metadata:s:v:0 "Transfer characteristics=BT.709" -c:v copy OUTPUT.mkv [02:37] <c_14> I'm no expert, but if the colormatrix is already BT.709, then all relevant tags should be set and if it isn't setting them will probably just break things. [02:37] <c_14> Why exactly do you want to do this? [02:44] <foonix> c_14: when i play such files (without tags) with mpc-hc+madvr the decoder uses BT.601 when the horizontal resolution is below 1280 i guess [02:45] <foonix> so i did quick test with -x264-params colorprim=bt709:transfer=bt709:colormatrix=bt709 and the behavior is OK for hd480 resolution.. [02:49] <c_14> Can you pastebin an ffprobe of the file it's playing incorrectly? [03:13] <foonix> c_14: diff is yuv420p10le vs yuv420p10le(tv, bt709), i just dl zeranoe's build to check it because i dont have ffprobe on my box...thats why such delay, if you want both 1 frame files i can give you url for look... [03:14] <c_14> kk, then the source file doesn't have the colorspace metadata set so the program is just guessing [03:14] <c_14> s/space/matrix [03:14] <foonix> yes [03:16] <foonix> and it seems for 1280 and higher its ok but for lower resolutions nope (which makes sense, but not in my case) [03:18] <c_14> I'm not sure you can change the colormatrix without reencoding since I'm pretty sure that's stored in the video stream and not the container. [03:18] <c_14> s/change/"set" [03:19] <c_14> The easiest option, especially if it's just a few videos would be to set the colormatrix in the player. [03:22] <foonix> well, yeah i think i could script that in madvr easily.. but wondering about the remux way.. [03:23] <foonix> did another test with -map_metadata -1 and colormatrix is still set [03:23] <c_14> Ye, I don't think it's container metadata. [03:23] <c_14> It's stream metadata. [03:23] <foonix> so i guess i cant manipulate it so easily [03:23] <c_14> you can set it using the colormatrix filter, but that'll require reencoding [03:26] <c_14> You could probably set it with a hex editor/a specially crafted program, but you'd have to dig into the spec[s]. [03:26] <c_14> You can always try asking on the user ml. [03:38] <foonix> interresting h264_changesps bitstream filter - http://forum.doom9.org/showthread.php?t=152419 [03:42] <foonix> but im off now, thx for help [04:15] <molikto> hi, i disabled almost all things i do not need to achieve a smaller file size, it works for compressiong most mp4 files, but one with amrnb audio encoder failed me. i compiled ffmpeg with amrnb decoder and aac encoder, but it says i need aresample filter, but i configured it with this, but it doesn't shows up in the configure result [04:19] <molikto> anyone can help me? [06:34] <JDuke256> hi , is ffmpeg supporting cedarx vs tms320 omap optimized codes ? or davinci ? [07:20] <ubuntuaddicted> here's some interesting reading http://on-demand.gputechconf.com/gtc/2014/poster/pdf/P4188_real-time_panora… [07:42] <gabriel> ffmpeg [07:42] <gabriel> -f image2pipe [07:42] <gabriel> -filter_complex concat=n=2:v=1:a=0 [07:42] <gabriel> -vcodec mpeg4 [07:42] <gabriel> -b:v 800k [07:42] <gabriel> -r 2, [07:42] <gabriel> -y, pipe:1 [07:43] <gabriel> Can anyone tell me why the above command gives me the error: Unable to find a suitable output format for 'pipe:' [07:44] <gabriel> there are inputs in the actual command, I just forgot to put them in there [07:45] <sacarasc> Add -f <someformat> to the output options. [07:49] <gabriel> Oh, I thought it would pickup on that from the -vcodec option [07:50] <relaxed> that applies to the codec, not the container [07:52] <gabriel> Got it, thanks. That worked. You guys are like mountain sages [07:52] <gabriel> I always think Ive got an obscure problem and someone always seems to get it straight away [10:14] <eristisk> Hi all, I compiled ffmpeg for a web CMS inside a non-system directory, but a dir specifically for the CMS installation (Plumi) [10:15] <eristisk> It compiled successfully, but when I try to run it, I get the error that it cannot find the libraries it needs to run [10:15] <eristisk> bin/ffmpeg: error while loading shared libraries: libavdevice.so.54: cannot open shared object file: No such file or directory [10:17] <eristisk> I see that the library compiled successfully, but it is in the path ffmpeg/parts/ffmpeg-build/lib [11:42] <alexvf> hi all [11:43] <alexvf> when encoding to an H264 file, is there a way (set of parameters) to ensure that the frames in the resulting file will have decoding_order = display_order? [12:12] <alexvf> anyone knows hot to encode an H264 without frame reordering using ffmpeg? [13:27] <erp2themaxx> I'm trying to stream a video from ffmpeg to ffserver. Why does ffmpeg exit without streaming and without saying what's wrong? http://pastebin.com/cJhvZp2C [13:58] <DanielSa> hi [14:10] <Mavrik> g'day [14:32] <pagios> do windows phone support HLS? [14:42] <Alina-malina> hello all, guys do you know what software can i use to make stuff like this: http://www.youtube.com/watch?v=-yM8mO63yE0 i know he is professional dj, and he use sound software and video separately, but anyways, would love very much to learn how he does this [15:35] <dougquaid> I've been searching on google but I haven't found the answer yet. Can ffmpeg record an h264 video stream from an IP camera? [15:38] <dougquaid> Also, can ffmpeg add metadata to video files? I'd like to record the time when the video starts recording [15:38] <__jack___> dougquaid: which is the output format or your cam ? a 'raw' h264 video stream ? [15:39] <__jack___> eg with or without a container [15:39] <dougquaid> __jack___ let me check [15:40] <dougquaid> It says h.264 (mpeg-4) part 10/avc, as well as Motion JPEG [15:42] <dougquaid> __jack___: Does that spec describe the output format? Based on my limited knowledge of video files I'm guessing the format is mpeg-4? [15:42] <__jack___> hum, it's not very explicit to me [15:43] <__jack___> ffmpeg should be able to handle such stream [15:59] <hefest> anyt ideas why resulting mp4 video has audio out of sync? ffmpeg -y -i 0CAGS5AIZMQ.flv -vf scale=320:240 -q:v 0 -profile:v baseline -r 25 -vcodec libx264 -crf 21 -ac 2 test.mp4 [16:34] <BtbN> hefest, try adding -c:a copy [16:36] <hefest> BtbN: Could not find tag for codec nellymoser in stream #1, codec not currently supported in container [16:37] <BtbN> What even is nellymoser oO [16:37] <hefest> BtbN: Stream #0:1: Audio: nellymoser, 8000 Hz, mono, flt [16:37] <hefest> BtbN: in source FLV file [16:38] <BtbN> yes, i have never heard that before. [16:38] <hefest> source flv file has variable frame rate, i think that's causing issues but not sure how to solve it [16:44] <hefest> man this is killing me, anyone else? any hints how to solve this? [17:20] <FunkyELF> hey guys... using the latest static build and I'm getting this... [17:20] <FunkyELF> Unknown encoder 'libfdk_aac' [17:20] <FunkyELF> trying to follow https://trac.ffmpeg.org/wiki/Encode/AAC [17:21] <kepstin-laptop> the static build doesn't include the fdk aac encoder, because (due to license incompatibilities) distributing a build with that enabled would not be legal. [17:21] <kepstin-laptop> if you want to use it, you have to build your own ffmpeg [17:21] <FunkyELF> kepstin-laptop: how can I get aac?... it doesn't seem that libfaac works either [17:22] <kepstin-laptop> the static build has the native aac encoder enabled; if you don't want to build your own ffmpeg, you could use that. [17:22] <FunkyELF> I will try... I have a bunch of .mkv files that don't play on XBMC on my Nexus Player, but after I re-encode them with Handbrake they do [17:23] <FunkyELF> I used ffmpeg to change containers without re-encoding... put it in a .mp4 and it still wouldn't play [17:23] <FunkyELF> so now I want to use ffmpeg to re-encocde the audio but leave the video... I'm trying to track down what the problem is [17:25] <kepstin-laptop> it's probably more likely to be an issue with the video encoding; for example, if the mkv used 10bit h264, that won't work on any hardware players. [17:26] <FunkyELF> title : Shaun The Sheep S01E01 540p HDTV x264-DEADPOOL [17:26] <FunkyELF> encoder : HandBrake 0.10.0 2014112300 [17:26] <FunkyELF> whoops... wrong lines [17:26] <FunkyELF> Stream #0:0: Video: h264 (High 10), yuv420p10le, 960x540 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 1k tbn, 50 tbc (default) [17:27] <FunkyELF> is that (High 10) mean 10bit? [17:27] <kepstin-laptop> there you go, that's exactly the problem. [17:27] <kepstin-laptop> yep. [17:27] <FunkyELF> kepstin-laptop: okay... so I will have to re-encode no matter what [17:27] <FunkyELF> kepstin-laptop: I can actually play these on my Android phone / tablet with mxplayer or vlc. [17:28] <kepstin-laptop> If you use software decoding, it would play. but the hardware decoder wouldn't work [17:28] <FunkyELF> kepstin-laptop: okay... that must be what is going on [17:59] <excalibr> Does ffmpeg have switch to drop all user metadata from original input vid? [18:03] <rflmota> Hi everyone [18:05] <rflmota> I'm trying to setup a live continuous (24/7) WebM streaming w/ ffmpeg feeding to icecast server. All appears to run just fine but after some days of running, ffmpeg stops and I managed to capture the final 'debug log' which is here : http://pastebin.com/SNDqbc5s [18:07] <rflmota> i'm issuing the following command for ffmpeg : 'screen /bin/bash -c 'sudo ffmpeg -y -loglevel debug -f video4linux2 -re -i /dev/video0 -i usora_logo.png -i artica_small_bw.png -filter_complex "[0:v][1:v] overlay=0:H-h [V1]; [V1][2:v] overlay=W-w:H-h" -pre libvpx-720p-copy -vf scale=-1:720 -strict experimental -f WebM -content_type video/webm icecast://artica:artica9900@usora-artica.noip.me:8000/stream.webm 2> log2.txt'' [18:07] <rflmota> can somebody help me figuring out what's the problem? [18:09] <DanielSa> rflmota: you just send your user and password to us. [18:09] <DanielSa> rflmota: artica:artica9900 [18:15] <DanielSa> rflmota: By the way, I came here today just to ask this command that you just send. I'll study your parameters. [18:17] <ocx> very quick question, is it possible to play mp4 files with FLASH RTMP? [18:18] <kepstin-laptop> flash can play mp4 files over http. But when you're using rtmp it's streaming the video over rtmp; no mp4 is involved anywhere :) [18:18] <kepstin-laptop> you should be able to use ffmpeg in codec copy mode to send the contents of an mp4 file over rtmp if you want [18:19] <rflmota> thanks DanielSa [18:38] <DanielSa> Is it possible, in a live broadcast using one mic and two or more video sources, switch between the videos without stop the transmission? To be more specific, using two webcams, without any analogic tool. [18:39] <c_14> If you code it yourself/can find a program, yes. [18:41] <relaxed> DanielSa: you could pull off side by side, or picture in picture, video with ffmpeg. [19:16] <DanielSa> Thank you guys. [19:39] <llogan> excalibr: "-map_metadata -1" might do what you want [19:40] <llogan> and don't forget "-map 0 -c copy" if you want to map all streams and stream copy everything if you dn't want to re-encode [19:44] <bakers> What is the aac encoder marked experimental? [19:44] <llogan> the native FFmpeg AAC encoder, "aac" [19:45] <llogan> there is talk about removing the experimental status, but that may have to wait for bug #2686, aka the "epic thread". [19:52] <bakers> That would be great, I'd love to get rid of libvo [19:53] <kepstin-laptop> even now with the 'experimental' status, the aac encode is better than libvo-aacenc [19:54] <bakers> kepstin-laptop: I've been using it, but it's a pain to have to type "-strict -2 -a:c aac" everytime [19:54] <bakers> it'd be nice if that were the default [19:55] <JEEB> the experimental flag is a leftover from the times when it actually produced broken audio at relatively high bit rates [19:55] <JEEB> which was true up until early 2011 or so? [19:55] <JEEB> but no-one decided to remove the flag [19:56] <bakers> I'm converting a bunch of OLD avis to mp4 so my chromecast can play them [19:56] <bakers> they have .mp3 audio that's pretty crappy [20:04] <c_14> Converting mp3 audio to aac audio will never make it better. [20:04] <kepstin-laptop> unless you want it smaller and lower quality, just leave it. [20:05] <kepstin-laptop> chromecast should be able to play mp3 audio in an mp4 container [20:24] <bakers> kepstin-laptop: Good point... [20:25] <excalibr> llogan, thanks. By the way, any idea why mkv file muxed with ffmpeg is not fast-forwardable on my tv media player? It is fast-fwd able though, when I muxed it with mkvmerge [20:26] <kepstin-laptop> excalibr: player might need the seek index at the start of the file, I think ffmpeg writes it to the end by default (since it doesn't know how big the seek index is until the encode is done) [20:27] <kepstin-laptop> there's an option to reserve some space at the start in ffmpeg, you could give that a try [20:28] <excalibr> kepstin-laptop, what's the option/switch? [20:28] <kepstin-laptop> excalibr: https://www.ffmpeg.org/ffmpeg-formats.html#Options-4 [20:28] <kepstin-laptop> (make sure your ffmpeg is writing to a file rather than a pipe, too) [20:33] <excalibr> kepstin-laptop, does the option take size prefix for its arg? or it must be in bytes only [20:33] <bakers> On this page: https://trac.ffmpeg.org/wiki/Encode/AAC#NativeFFmpegAACencoder [20:33] <bakers> Is this still accurate? "Effective range for -q:a is around 0.1-10. This VBR is experimental and likely to get even worse results than the CBR. If ffmpeg ignores -q:a then get a newer build (see ticket #1346). " [20:33] <kepstin-laptop> excalibr: dunno, i'd have to look at the source coded. [20:39] <llogan> bakers: i don't know. that was written a year ago, but by kamedo2 who knows more about it than i do [20:55] <excalibr> kepstin-laptop, just gave it a go and no dice :(. No idea what's wrong [20:56] <excalibr> here's the command I used: ffmpeg -i input.mp4 -i en.srt -map 0:v -map 0:a -map 1 -map_chapters -1 -map_metadata -1 -c copy -reserve_index_space 100000 ffmpeg-fastfwd.mkv [20:59] <hay> hi, I would like to add looped audio input (mp3 file) to the video and can't find any useful options.. can it be done and if so, how? thanks a lot! [21:13] <llogan> hay: maybe the loop option in the amovie filter? or maybe use the concat demuxer or concat filter [21:15] <llogan> there's a loop example using concat demuxer you can adapt here: https://trac.ffmpeg.org/wiki/How%20to%20concatenate%20%28join,%20merge%29%2… [21:16] <hay> llogan, thanks, will look at it [21:30] <dert78> How can I do multiple files at once? * doesnt seem to work [21:30] <c_14> What do you mean? [21:31] <dert78> "ffmpeg -f image2 -loop 1 -i picture.png -i *.mp3 -c:v libx264 -tune stillimage -c:a copy -shortest movie.mp4" for example [21:32] <c_14> That's not how it works. [21:32] <c_14> What are you trying to do? [21:32] <c_14> Merge or concat or do the same thing for multiple files? [21:32] <dert78> I am trying to merge audio and a picture into video file format. (but not one by one) [21:33] <c_14> You have 3 choices, merge the audio tracks, concat the audio tracks, create one output per audio track. [21:33] <c_14> Pick one. [21:34] <dert78> I have no idea what those things mean? Im not very tech savy [21:34] <c_14> merge ie, imagine having 1 boombox per audio track and playing a different audio track on every boombox at the same time [21:34] <c_14> concat: playing each audio track in succession on one boombox [21:35] <c_14> create one output per audio track: have numerous boomboxes, play one song on one boombox then the next song on the next boombox, etc [21:35] <dert78> the last one. [21:35] <kepstin-laptop> have multiple audio tracks in one output: have songs playing on multiple radio stations, and pick between them with a dial on the boombox [21:36] <c_14> Right, that was option 4. Forgot about that one. [21:36] <c_14> Do you have one picture or one picture per mp3? [21:36] <dert78> one pic for them all [21:37] <kepstin-laptop> so you have a bunch of mp3s, one picture, and you want to make a bunch of video files where each video file plays one song and shows the picture? [21:37] <dert78> Exactly [21:37] <c_14> for mp3 in *.mp3; do ffmpeg -f image2 -loop 1 -i picture.png -i $mp3 -c:v libx264 -tune stillimage -c:a copy ${mp3%.*}.mp4; done [21:38] <kepstin-laptop> probably add some quotes in case your mp3 filenames have spaces [21:38] <kepstin-laptop> but that's basically it. :) [21:39] <llogan> you may need -pix_fmt yuv420p as an output option depending on your target player/device [21:39] <dert78> Quotes around "${mp3%.*}.mp4" ? [21:39] <kepstin-laptop> and around "$mp3" as well [21:41] <llogan> and "-movflags +faststart" as an output option may be desired if this is for progressive download playback (such as via a browser) [21:41] <dert78> Well, it is for youtube. [21:42] <llogan> you won't need either of the options i mentioned then [21:42] <llogan> but add -crf 18 [21:42] <llogan> since youtube will re-encode it anyway [21:44] <llogan> you can even add lyrics with a subtitles file (hardsub it) or other text with drawtext filter [21:44] <dert78> $mp3: no such file or dir [21:44] <dert78> hmm [21:46] <c_14> Did you use single quotes anywhere? [21:46] <dert78> ffmpeg -f image2 -loop 1 -i publicdomain.large.png -i "$mp3" -c:v libx264 -tune stillimage -c:a copy "${mp3%.*}.mp4" [21:47] <kepstin-laptop> dert78: you're supposed to have that wrapped inside of a for loop... [21:48] <dert78> How do I do that? [21:48] <c_14> exactly like I wrote up there [21:49] <llogan> you could add -framerate 2 as in input option. less waiting. [21:49] <llogan> dert78: what is your OS? [21:50] <dert78> Oh I didnt know you mean to add for in the code. /facepalm. [21:50] <dert78> Windows 7 [21:50] <c_14> eh [21:50] <c_14> that won't work in windows [21:50] <c_14> probably [21:50] <kepstin-laptop> oh, then that won't work, the command c_14 gave was a linux shell command [21:50] <kepstin-laptop> unless you use cygwin or something :/ [21:50] <c_14> You'll need a batch script. [21:53] <dert78> I will install cygwin real quick and see if that works [21:53] <llogan> might be easier to just use a normal batch script [21:53] <ramiro> "cygwin" and "real quick" don't go together in the same sentence... [21:54] <dert78> How does the batch script look like then? [21:54] <llogan> there are many examples on the 'net [21:54] <llogan> or ask in #windows if such a channel exists [21:55] <llogan> but we gave you all of the ffmpeg stuff for it to work nicely, you just need to figure out the batch part [21:56] <dert78> Okay, I will do that. Thanks for all the help :) [23:09] <quixotic> hello. i'm using ffmpeg to convert from mp3 to ogg. My problem is that running the exact same command results in binary different ogg files. I would like the conversion to be deterministic, so that I can rely on the fact that if the input file did not change, I get the exact same binary file as output. I was wondering if somebody would have a tip for me? (http://pastebin.com/RStsM2va) [23:14] <bakers> quixotic: You run that same command, on the same machine, and get different outputs? [23:14] <quixotic> yes [23:16] <quixotic> bakers: they are both legit files and sound the same, but the binary data is slightly different which causes problems in my automation process/revision control system. [23:17] <bakers> how are you comparing the binary data? [23:17] <bakers> Filesize? [23:17] <bakers> Checksum? [23:18] <quixotic> git is noticing the file discrepancy, i verified via winmerge [23:19] <bakers> quixotic: Wild... I just did the same thing you did 5 times [23:19] <bakers> http://www.fpaste.org/154471/41704037/ [23:19] <quixotic> bakers: thanks for taking the time. are you getting the same file? [23:19] <bakers> I don't know what to tell you, that doesn't make any sense [23:20] <quixotic> bakers: at least i'm not crazy [23:20] <bakers> quixotic: All exactly the same size [23:20] <bakers> http://www.fpaste.org/154472/14170404/ [23:20] <bakers> I wonder if the ogg is the same, but there is some metadata about "encode time" or something that's unique to each one [23:21] <klaxa> i would think metadata is messing you up [23:21] <klaxa> yeah [23:21] <klaxa> try: -map_metadata -0 to unmap metadata from the first input [23:22] <bakers> If I use ogginfo to look at the metadata [23:22] <bakers> http://www.fpaste.org/154474/41704056/ [23:23] <bakers> Line #5 [23:23] <bakers> Each file has a different serial number [23:23] <quixotic> ah yes, i would need a way to force the serial number [23:24] <bakers> I didn't know ogg files had serial numbesr [23:25] <bakers> quixotic: https://xiph.org/ogg/doc/oggstream.html [23:25] <bakers> That page mentions unique serial number several times [23:25] <bakers> not sure how to set that [23:27] <quixotic> klaxa: not sure if i'm doing it right, but i think the -map_metadata does not help [23:28] <bakers> It's gotta be that serial number, I'm not sure how you would force that [23:29] <c_14> -metadata:[foo]:[bar] baz=thing [23:29] <c_14> check the manpage [23:38] <somebody_useless> Hi Guys, I work for a small time service provider and we have run into an issue while transcoding live video. We tweak the settings per channel and SD channels work great, but we have run into an issue when a channel changes audio from stereo to dolby or vice versa (from what we can tell). [23:38] <somebody_useless> This causes an issue and we have to restart the ffmpeg process manually. [23:38] <somebody_useless> While checking the output stream using ffmpeg it appears to be encoding the Descriptive Video channel perfectly fine (constant stereo?), however it doesn't appear to allow the primary audio channel (stream/pid) to work properly, we suspect that it is because this channel changes. [23:38] <somebody_useless> The crazy thing, volumedetect SHOWS that volume is working, yet it isn't. I would LOVE to hear someones incite into this matter! [23:40] <quixotic> tried setting serial or "s" as metadata, but does not seem to work.. or at least i have no idea what the correct param name should be [23:51] <llogan> quixotic: use the md5 muxer to test https://ffmpeg.org/ffmpeg-formats.html#md5-1 [23:52] <klaxa> that will require decoding to compare hashes, right? [23:53] <llogan> yes, but it could also be used to test the encoded output [23:53] <llogan> by adding -c:v [23:53] <llogan> to avoid any differences caused by metadata [00:00] --- Thu Nov 27 2014

1 0

ffmpeg-devel.log.20141125
by burek 26 Nov '14

26 Nov '14

[01:12] <Theywentviral> im having trouble using a new software called stream-m [01:13] <Theywentviral> has anyone heard of this? [01:15] <Theywentviral> im using a method called concat and its causing me to use 90% cpu usage when its looping [01:18] <Theywentviral> ok [01:18] <Theywentviral> well i can't see anyones replies [01:25] <TheyWentViral> ok [01:26] <J_Darnley> Why would you come here for help with other software? [01:28] <TheyWentViral> _Darnley it says for the development of ffmpeg , and the development of other software that uses ffmpeg clearly in the documentation [01:29] <TheyWentViral> This software stream-m uses ffmpeg [01:30] <llogan> This is the channel for discussion of FFmpeg development only. user support is at ffmpeg-user IRC and mailing list, but we only support FFmpeg stuff, and not 3rd party stuff (including stream-m) [01:30] <TheyWentViral> im having trouble with the concat of a stream i beleive its causing me to use 90% CPU usage i was just wondering if anyone ran into a similar problem when it loops using this method... [01:31] <J_Darnley> So give us some ffmpeg command lines and terminal output, then maybe we can help you. [01:31] <llogan> ...so you'll have to get help from stream-m, or if you are using ffmpeg directly then use the mailing list or IRC channel. [01:31] <TheyWentViral> ok sorry llogan maybe i inturpreted the doucmentation wrong im just looking for a little help i'll search there [01:31] <llogan> (ffmpeg-user that is, and not here or ffmpeg-devel mailing list) [01:35] <TheyWentViral> Thanks again [01:55] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:03a17f2bbf36: avcodec/mjpegdec: Print the number of bits in the unsupported pixel format error [01:55] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:172d22a071b5: avcodec/mjpegdec: Add YUVA420 formats to *scale asserts [02:12] <cone-890> ffmpeg.git 03Uwe L. Korn 07master:40665d27e38e: flvdec: Document how the duration is retrieved at the end of the file [02:12] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:b83beb131f0d: Merge commit '40665d27e38e6a2f65037878202bd1a398c7683e' [02:48] <cone-890> ffmpeg.git 03Martin Storsjö 07master:b9d08c77a443: lavf: Don't try to update files atomically with renames on windows [02:48] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:71ecfcf2d31e: Merge commit 'b9d08c77a44390b0848c06f20bc0e9e951ba6a3c' [04:04] <cone-890> ffmpeg.git 03Martin Storsjö 07master:960aff379da4: lavf: Use wchar functions for filenames on windows for mkdir/rmdir/rename/unlink [04:05] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:ecfafc5f2b63: Merge commit '960aff379da46dcaff61504a57714d4d4e758e41' [04:05] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:60420fa3dc88: avdevice/dshow: fix build, ensure that feature enable #defines are set before includes [04:05] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:c724c82a7828: doc/print_options: Fix build on mingw after 960aff379da46dcaff61504a57714d4d4e758e41 [04:19] <cone-890> ffmpeg.git 03Carl Eugen Hoyos 07master:0e3ea5b28a32: Include stddef from snow_dwt.h. [04:19] <cone-890> ffmpeg.git 03Carl Eugen Hoyos 07master:20f3cdf0c894: Include config.h from huffyuvdsp.h. [04:19] <cone-890> ffmpeg.git 03Michael Niedermayer 07master:67dbf8bec977: Merge remote-tracking branch 'cehoyos/master' [04:38] <Zeranoe> michaelni: Any chance you could commit the attached patch for this bug https://trac.ffmpeg.org/ticket/4130 [04:56] <cone-890> ffmpeg.git 03Carl Eugen Hoyos 07master:1f3d4788983d: avdevice/decklink_common: Fix "Cross-compiling FFmpeg on Debian for Windows with MinGW-w64" [04:56] <michaelni> Zeranoe, applied [05:01] <Zeranoe> michaelni: Thanks [11:17] <cone-477> ffmpeg.git 03Vittorio Giovara 07master:bc75b64cff37: vc1pred: remove logically dead code [11:17] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:0aa208837e24: Merge commit 'bc75b64cff37d58f3944e2da3da45c37f35f019a' [11:55] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:c117da9d3e0d: lpc: remove unneeded {} [11:55] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:62e52b94e684: vorbis_parser: Move vp check to avoid a null pointer dereference [11:55] <cone-477> ffmpeg.git 03Luca Barbato 07master:b67138598ce1: vc1: Simplify a little setting the intra variables [11:55] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:a63ec9d5737a: Merge commit 'c117da9d3e0db7dc311d817054988364b3ef4587' [11:55] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:bcccb2c2986b: Merge commit '62e52b94e684491dfc5a6b7ca688bb86f7cd0f3f' [11:55] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:ed25ca166171: Merge commit 'b67138598ce158e3083f6295a27b63e2065d5ecb' [12:06] <cone-477> ffmpeg.git 03Luca Barbato 07master:d25afb579fac: vc1: Set the is_intra bitfield to all 1 when needed [12:06] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:79f77ce31c6c: Merge commit 'd25afb579facc83fd3a839f21411124d0b09f0ba' [12:26] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:51946d2de8bd: vc1: Use the correct shift amount [12:26] <cone-477> ffmpeg.git 03Luca Barbato 07master:16158da9607f: hnm4: Use av_image_check_size [12:26] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:0e85a28fa434: Merge commit '51946d2de8bd4a4aada43b6ab41340b0f5eb4ecb' [12:26] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:ba1a19bc35e4: Merge commit '16158da9607f2f84232d3dd381406b2f2449ec74' [12:44] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:57ed5a64feec: hnm4: change width/height to int to fix hypothetical integer overflows [12:44] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:ca5c3ff90972: vf_interlace: x86: improve asm performance [12:44] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:8c9945285e30: Merge commit '57ed5a64feec4af1f16f9a74c63cfa9aa8147242' [12:44] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:3fe3c8abb15a: Merge commit 'ca5c3ff90972a5c97aabda2ace57ba72dcd7d83b' [12:44] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:ca59b5b6eceb: avfilter/x86/vf_interlace: remove redundant instructions [13:35] <wm4> ubitux: went through utils.c; I wonder why everything checks whether pb is NULL (except some code doesn't) [13:36] <wm4> ubitux: I don't think the API allows the demuxer to close the pb at all, if that's what you planned [13:37] <wm4> even setting to NULL might not be fine (due to the API) [13:43] <__gb__> hi kierank, you can use vaapi without rendering, yes -- even without a display server actually (using the VA/DRM backend) [13:45] <__gb__> and, if you have a kernel version recent enough (>= 3.15 to have that feature enabled by default), then you can use render nodes too for using vaapi remotely through ssh, even if another user has locally started a display server [13:46] <__gb__> if you don't want to go over elaborated libudev stuff, you could just open /dev/dri/renderD128 -- that's the render nodes companion to /dev/dri/card0 [13:47] <__gb__> and provide the resulting fd to vaGetDisplayDRM() and use the resulting VADisplay wherever needed [14:10] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:0eecf40935b2: avcodec/mjpegdec: Fix context fields becoming inconsistent [14:56] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:3d5d95db3f5d: avcodec/utils: Check that the data is complete in avpriv_bprint_to_extradata() [15:28] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:8cd80b5fcbfa: avformat/jacosubdec: Cleanup when avpriv_bprint_to_extradata() fails [15:48] <wm4> "Getting board?" [15:48] Action: wm4 facepalms [16:11] <Zeranoe> michaelni: I think ecfafc5f2b632780dd310def1d6b3a2c10565465 or 960aff379da46dcaff61504a57714d4d4e758e41 introduced a new compiling error http://paste.ubuntu.com/9233871/ [16:18] <Daemon404> Zeranoe, that looks like a C++ error... weird [16:22] <Zeranoe> Daemon404: More context: http://paste.ubuntu.com/9234011/ looks like decklink_enc.cpp is being included [16:25] <michaelni> Zeranoe, does a cast to wchar_t* fix that ? [16:26] <Daemon404> michaelni, a cast is the wrong fix [16:26] <Daemon404> this is not a valid error in C [16:27] <wm4> well it's compiled as C++ apparently? [16:27] <Daemon404> that seems majorly wrong [16:27] <wm4> these are just stubs for win32 [16:27] <wm4> so nobody cares whether they're inline [16:27] <wm4> same with win32 pthread wrappers [16:28] <Daemon404> doesnt matter, it can still be done right in C++ [16:28] <wm4> with the new operator? [16:28] <Daemon404> let me sort it [16:28] <Daemon404> also [16:28] <Daemon404> #include "libavformat/internal.h" [16:28] <Daemon404> in libavdevice [16:28] <Daemon404> yum yum [16:29] <wm4> apparently this is "ok" [16:29] <wm4> because it's a compile time dep only [16:30] <Daemon404> i think vfwcap.c is misleading [16:30] <Daemon404> Zeranoe, were you building multithreaded? [16:30] <Daemon404> i think ti *actually* failed in the .cpp [16:32] <Daemon404> ok so... [16:32] <Daemon404> https://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavdevice/decklink_enc.cp… [16:32] <Daemon404> what the fuck g++ [16:32] <Daemon404> you should not be doing that when the func is under extern "C" [16:32] <wm4> uh [16:32] <wm4> extern C doesn't mean it's C code [16:32] <nevcairiel> extern C only controls linkage [16:32] <nevcairiel> nothing else [16:32] <wm4> just that C mangling is used [16:33] <nevcairiel> common misconception [16:33] <Daemon404> i see [16:33] <Daemon404> that's kinda lulz [16:33] <wm4> also you could have tortured the contributor into writing that file in C [16:33] <Daemon404> i figured it would be smarted than to try and inline C into C++ [16:34] <Daemon404> because inline is a hint to the compiler [16:34] <Daemon404> not a guarantee [16:34] <nevcairiel> the standard can probably not assume that every C++ compiler cna also speak C [16:34] <nevcairiel> while turning off name mangling is easy [16:34] <Zeranoe> Daemon404: I am, pthread [16:34] <Daemon404> Zeranoe, not what i meant, but its not relevant now [16:35] <Daemon404> i guess a cast it is [16:35] <Daemon404> c++/c interop sure is fugly [16:35] <Zeranoe> Daemon404: Yes to what you meant, make -j4 [16:37] <BtbN> https://github.com/BtbN/FFmpeg/commit/13b5cceea6300e36c2d8d7d6b8b2cfacbcdc5… Is this the correct way to add a dependency on an external header file? [16:37] <BtbN> This configure script is extremely confusing and way too much stuff happens automagicaly based on naming [16:38] <BtbN> I'm still not sure if it would propperly disable my code if nvenc is disabled [16:41] <BtbN> Seems like it does, but i have no idea why [16:41] <BtbN> It also doesn't seem to matter that in allcodecs there is no compile-time removal of anything [16:42] <Zeranoe> Daemon404, michaelni: Do you need anything else from me on this? [16:42] <Daemon404> Zeranoe, no, the fix is clear [16:43] <BtbN> Hm, what qualifies a library to be flagged as non-free? As far as i know, linking ffmpeg with nvenc doesn't imply any redistribution restrictions. You just can't include its header in ffmpeg. [17:01] <michaelni> Zeranoe, a tested patch would be best but i can try to blindly fix it and maybe after a few tries we will have found all cases th [17:01] <michaelni> that need changing [17:04] <Zeranoe> michaelni: I would be happy to test any patchs [17:17] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:d0879a93eac6: avformat/os_support: try to fix build when included from a c++ file like libavdevice/decklink*cpp [17:18] <michaelni> Zeranoe, pushed a fix for that one line, does it build now or are there more issues ? [17:21] <Zeranoe> michaelni: running with a single make job now [17:43] <Zeranoe> michaelni: Everything seems fine now, thank you for the prompt fix [17:44] <michaelni> np [18:02] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:2f1de5ca139f: avcodec/huffyuvdec: apply vertical filter in steps of 1 line for interlaced BGRA [18:27] <cone-477> ffmpeg.git 03Yu Xiaolei 07master:bc3d02fa88c4: lavf/avio: clarify the buffer parameter of avio_alloc_context [18:30] <cone-477> ffmpeg.git 03Benoit Fouet 07master:155f4dd668b8: doc: add entry for APNG demuxer where needed. [18:48] <Daemon404> so uh.... [18:48] <Daemon404> would a patch be accepted to disable xmp printing in ffprobe? [18:48] <Daemon404> because some files print shit for 5+ minutes [18:48] <Daemon404> its ridiculous [19:22] <ubitux> wm4: alright, so what do you propose (re:closing pb) [19:22] <wm4> let the API user handle it [19:23] <wm4> just export the necessary information [19:26] <ubitux> wm4: but, i don't see what the user is supposed to do here [19:27] <BtbN> ffmpeg has pull requests now? [19:27] <wm4> ubitux: the user can do things with the AVIOContext, or even unset it (if lavf allows that) [19:28] <ubitux> so the patch as is looks ok to you? [19:28] <wm4> yes [19:29] <wm4> although granted I'm not perfectly sure how I'll use it [19:29] <ubitux> i'll push it if you're able to make use of it then [19:29] <wm4> fair enough [19:29] <ubitux> maybe we forgot something so i don't want to change the api yet [19:54] <BtbN> Are github PullRequests accepted nowdays? [19:54] <kierank> no [19:54] <BtbN> That's highly annoying... [19:55] <Daemon404> BtbN, some of us would say github PRs are highly annoying [19:55] <Daemon404> they have a number of disadvantages. [19:55] <BtbN> Setting up mails, managing changes, and getting tons of ML mail is annoying [19:56] <Daemon404> im saying its personal preference, and the overwelming majority of people here prefer the ML. [19:56] <BtbN> I have that stuff flying around since ages now, and didn't send it in just because of having to do that git mail stuff [19:56] <Daemon404> ok? [19:57] <Daemon404> and you can attach patches too. [19:57] <BtbN> Makes it kinda hard to discuss some of the changes [19:57] <BtbN> I could send a mail pointing to my github branch, as request for comments [19:59] <Daemon404> someone might review it, but i cant guaranatee it [20:00] <BtbN> I'm still not sure if it qualifies as nonfree or not [20:00] <llogan> BtbN: some github pull requests have been accepted, IIRC. we prefer patches sent to the mailing list, but you can try a PR. not nearly as many people see them. [20:00] <BtbN> It's a new h264 encoder, needs some feedback [20:00] <BtbN> nvidia nvenc stuff [20:01] <Daemon404> non-free with dlopen/loadlibrary is a massive grey area [20:01] <Daemon404> i dont think youll ever know conclusively [20:01] <BtbN> There is no non-dlopen way though. That's the official way nvidia wants it to be used [20:01] <wm4> ML + github PRs are the worst combination [20:01] <Daemon404> BtbN, i know [20:02] <wm4> pick either one, and ffmpeg picked ML-based workflow [22:15] <cone-477> ffmpeg.git 03Martin Storsjö 07master:f20141d73f08: vorbis_parser: Include stdint.h in the header, to make it work standalone [22:15] <cone-477> ffmpeg.git 03Michael Niedermayer 07master:23e91a1bfd7f: Merge commit 'f20141d73f08ed0c8e875bd993a7143e19b266e3' [22:18] <cone-477> ffmpeg.git 03Lukasz Marek 07master:e29153f414f5: lavc/avuienc: fix mem leak in case of init failure [22:18] <cone-477> ffmpeg.git 03Lukasz Marek 07master:2a89afb376ae: lavc/libxvid: return meaningful error codes [22:42] <kierank> kurosu_ / jamrial : any more asm comments? [22:44] <kurosu_> kierank, nope [22:45] <kurosu_> was the unrolling actually helpful? you said it was but it looked more than I would have assumed [22:45] <kurosu_> and I'm not sure if it breaks the C part [22:46] <kurosu_> I mean, handling batches of 12 instead of 6 as I thought it assumed [22:47] <kurosu_> anyway, I don't really care: if it breaks, I bet you'll be eager to fix it [22:48] <kierank> for 8-bit yuv422p -> v210 you have to unroll [22:48] <kierank> whereas in 10-bit mode you only have 8 pixels in your register [23:10] <cone-477> ffmpeg.git 03Lukasz Marek 07master:ea0d8938173e: lavu/opt: handle NULL obj in av_opt_next [23:10] <cone-477> ffmpeg.git 03Lukasz Marek 07master:1907ff0a67ed: lavc/utils: free private options on avcodec_open2 fail [23:19] <jamrial> kierank: nothing aside from what i mentioned for the 10bits patch [23:19] <kierank> ok [00:00] --- Wed Nov 26 2014

1 0

ffmpeg.log.20141125
by burek 26 Nov '14

26 Nov '14

[00:00] <TSM> the odd thing is how it plays on the BB10, it starts, knows the length of the video 27s then stops after 7 seconds, if i take qtfaststart off then BB10 complains of a codec error [00:01] <TSM> another odd thing, if i do not send a Content-Length header to the device over html then the video plays correctly [00:01] <TSM> all this time it works fine on desktop chrome/firefox [00:02] <TSM> I have done webinspection on the BB10 and it always downloads the whole file before playing it so very odd [00:04] <llogan> TSM: are you playing it "directly" on the device (uploaded to device and play with native player) or via a browser, or some other way? [00:06] <TSM> via browser, if i save it then play it all is fine [00:07] <TSM> the browser on the BB10 offloads the playing to the video app [00:07] <TSM> it is odd [00:13] <llogan> TSM: i don't know what to suggest here other than trying a simpler command using defaults and seeing if any of your options cause the issue [00:13] <llogan> omit audio, etc [00:15] <llogan> unrelated to the issue, but you have support for libfdk_aac, so why are you using the native AAC encoder? [00:20] <TSM> dont know about the aac, i read about libfdk so based on your question it seems like it should be the better encoder [00:21] <TSM> i wish the x264 encoder was faster but i suspect that is because of my lack of CPU extensions that some of the newer procs have [00:24] <llogan> yes, libfdkaac is better. you could use a faster preset [00:26] <TSM> superfast is a awfull on the output at 1M, fast seems to be the best tradeoff, im also using baseline to make sure that it will play on andrioid devices [00:26] <TSM> i still average about 30-35fps encode speed with watermark,drawtext,scale [08:57] <pagios> when we say MPEG-DASH server does it mean a plain HTTP Server that servers the MPD file? [10:04] <sim590> I'm trying to configure urxvt. I've xrdb ~/.Xresources and now urxvt doesn't show with an interactive terminal but only a blank screen. [10:04] <sim590> oops [10:05] <The_NetZ> any one care to assist me in assisting another user in setting up libnvenc with ffmpeg? [10:07] <The_NetZ> https://clbin.com/GFc2C << what we're getting. [14:01] <Cas_> Hi, I try to stream audio from a decklink card (SDI) to an audio loopback but I keep getting silence. If I stream to a wav it works fine, anyone knows what I could be doing wrong? [14:02] <Cas_> My commands are: `ffmpeg -f decklink -i 'DeckLink SDI (1)@8' -f alsa hw:0` and `ffmpeg -f alsa -i hw:0 foo.wav` [14:02] <Cas_> `ffmpeg -f decklink -i 'DeckLink SDI (1)@8' foo.wav` works fine though [15:44] <ubuntuaddicted> morning/afternoon/evening having a real difficult time compiling ffmpeg with libnvenc enabled in Arch using a PKGBUILD. anyone else got ffmpeg to compile with libnvenc enabled? [15:46] <BtbN> ffmpeg does not have any nvenc support. Ask the guy who made that package what he did there. [15:48] <ubuntuaddicted> BtbN, oh, so apparently he patched ffmpeg to be able to use nvenc? [15:49] <ubuntuaddicted> curious why ffmpeg wouldn't want to utilize the onboard encoder chip of kepler and up GPU's [15:49] <BtbN> Quite surprising though, when i made working nvenc support for ffmpeg months ago, nobody cared. Now it's big news. [15:50] <ubuntuaddicted> nvidia just released a driver that supports it like a couple fridays ago as far as i'm aware [15:50] <BtbN> I am using it since months [15:50] <BtbN> it worked since forever, they just wrote it in their changelog for some reason [15:51] <ubuntuaddicted> ok, as i said for us normal users who don't compile our own nvidia kernel modules it just came available to us [15:51] <ubuntuaddicted> oh ok [15:51] <BtbN> https://github.com/BtbN/FFmpeg/tree/nvenc could use an update to current master though [15:51] <ubuntuaddicted> in your experience does nvenc provide any better quality than libx264? [15:52] <BtbN> no [15:52] <BtbN> just like any other hw encoder [15:52] <BtbN> x264 is just unbeatable in terms of quality per bitrate [15:52] <ubuntuaddicted> hmmm [15:52] <BtbN> But hw encoders use no cpu power and nearly no hardware resources [15:53] <ubuntuaddicted> the advantage i see is taking all the encoding load off the cpu for those that don't have the best cpu but have a kepler or higher gpu right? [15:53] <ribasushi> BtbN: doesn't HVEC/x265 outperform x264 by double-digit-percentages in that department? [15:54] <BtbN> what? [15:54] <ribasushi> <BtbN> x264 is just unbeatable in terms of quality per bitrate <--- responding to that [15:54] <BtbN> Nobody was talking h265, let alone h265 hw encoders. [15:54] <ubuntuaddicted> ribasushi, i don't think rtmp services like twitch support x265 yet though [15:54] <ubuntuaddicted> does youtube even support it yet? [15:55] <BtbN> twitch has trouble because users struggle to decode 720p with 60 fps because of too much cpu load. They don't even dare to think about h265. [15:55] <ubuntuaddicted> lol [15:55] <ribasushi> they don't, again I was just responding to BtbN's line in isolation [15:56] <ubuntuaddicted> BtbN, so ffmpeg isn't going to be adding support for libnvenc? [15:56] <BtbN> I have no idea what libnvenc even is [15:56] <BtbN> It's not the nvidia nvenc library. [15:57] <BtbN> That's nvEncodeApi [15:58] <ubuntuaddicted> well now i'm thoroghly lost. lol the people who have ffmpeg working with libnvenc enabled say they got ffmpeg to work by using --enable-libnvenc per here https://github.com/Brainiarc7/ffmpeg_libnvenc [15:58] <BtbN> Using that fork propably [15:58] <BtbN> which isn't even a fork, because only 4 commits [15:59] <ubuntuaddicted> BtbN, ok, well he got it from this guy's github which actually adds the support for libnvenc https://github.com/agathah/ffmpeg_libnvenc/commit/3b5a7bdccd5ed6b4189f59654… [16:00] <BtbN> Again, i have no idea what libnvenc even is [16:01] <ubuntuaddicted> BtbN, looking at that commit it's utilizing the nvEncodeAPI [16:01] <iive> most likely nvidia gpu accelerated encoder [16:01] <BtbN> Nope [16:01] <BtbN> That library has a diffrentname [16:01] <BtbN> -" " [16:02] <BtbN> libnvidia-encode.so [16:02] <ubuntuaddicted> libavcodec/nvencoder.c [16:02] <ubuntuaddicted> libavcodec/nvenc.h [16:03] <BtbN> That's just the avcodec part. [16:03] <ubuntuaddicted> BtbN, maybe i should start over. :) i would like to be able to utilize my GTX 760 hardware encoder to encode video/audio with ffmpeg, how would I do that? [16:04] <ubuntuaddicted> doc/examples/libnvenc.c [16:04] <BtbN> There is no supported way for that currently [16:05] <ubuntuaddicted> it appears that he used the windows nvenc sdk, reviewed the Sample example and then implmented his own libnvenc encoder type for ffmpeg. is that a fair statement? [16:05] <ubuntuaddicted> BtbN, apparently there is a way since it's been done by 2 people using ffmpeg [16:06] <BtbN> Not using the official version of ffmpeg, but some fork. [16:06] <ubuntuaddicted> BtbN, right, he started with RELEASE NOTES for FFmpeg 2.4 "Fresnel" [16:07] <ubuntuaddicted> BtbN, he did it back on Oct 24. https://github.com/agathah/ffmpeg_libnvenc/commits/master [16:10] <ubuntuaddicted> BtbN, i'm a little confused on naming of these things. I don't fully understand C+ so i'm clearly in way over my head. you call it libnvidia-encode.so where he calls it libnvenc. the name is irrelavant isn't it? both use a dedicated SIP block for accelerated video processing, and as such, is independent of CUDA cores. [16:10] <BtbN> He doesn't call it like that, he _links_ against that [16:11] <BtbN> so it's not the same thing [16:12] <ubuntuaddicted> ok but again i'm just trying to understand why he was able to get NVENC encoding enabled with his ffmpeg but i'm having troubles [16:13] <ubuntuaddicted> my config.log is here https://clbin.com/8i0O7 [16:13] <ubuntuaddicted> specifically common.mak:49: recipe for target 'libavutil/opencl.o' failed [16:31] <ubuntuaddicted> looks like it's getting hung right here libavutil/atomic_gcc.h:31:5: warning: "HAVE_ATOMIC_COMPARE_EXCHANGE" is not defined [-Wundef] [16:48] <ubuntuaddicted> is there currently plans to enable nvEncodeAPI into ffmpeg? [17:11] <ubuntuaddicted> BtbN, i see you were the one who worked on the nvenc implmentation in windows. [17:11] <ubuntuaddicted> BtbN, for OBS [17:40] <BtbN> yes [17:47] <Dominique_Michel> Hi, with the following webcam "uvcvideo" 3.17.0 capabilities : 0x84000001 [VIDEO_CAPTURE,STREAMING,(null)] [17:48] <Dominique_Michel> v4l2 or guvcview didn't find any standard [17:49] <Dominique_Michel> According to guvcview, the frame rate of the picture is changing on the fly when the webcam is streaming. [17:50] <Dominique_Michel> My goal is to record the video from the webcam together with the audio from JACK (jackdbus). [17:50] <Dominique_Michel> I don't know if it is possible with ffmpeg. The nearest I get is with the command: [17:51] <Dominique_Michel> ffmpeg -f jack -ac 2 -i jack -f video4linux2 -s 640x480 -r 8.5 -i /dev/video0 -qscale 3 -r 30 out.mpg [17:52] <kepstin-laptop> framerate changing on the fly is pretty common with webcams, particularly in the dark; they increase the exposure time pretty dramatically to make the image visible [17:52] <Dominique_Michel> but I get 2 issues: 1) the aaudio is not in sync with the video, and 2) the frame rate of the video is not 100% correct. [17:53] <kepstin-laptop> do not use -r as an input option, that's what's breaking the video framerate and sync [17:53] <kepstin-laptop> you can use the "-framerate" option to request the webcam to give you some framerate [17:54] <kepstin-laptop> but you'll probably get variable framerate output, that's usually ok. [17:54] <kepstin-laptop> note that with most webcams, you can get higher framerates by using mjpeg (or h264 if supported) input format; the limit when using raw video is sometimes the usb bandwidth. [18:03] <Dominique_Michel> With "ffmpeg -f jack -ac 2 -i jack -f video4linux2 -input_format h264 -framerate 3 -i /dev/video0 -qscale 3 -r 30 out.mpg", ffmpeg quit with " Cannot find a proper format for codec 'h264' (id 28), pixel format 'none' (id -1) [18:03] <Dominique_Michel> Assertion *codec_id != AV_CODEC_ID_NONE failed at /var/tmp/portage/media-video/ffmpeg-2.2.9/work/ffmpeg-2.2.9/libavdevice/v4l2.c:805" [18:06] <kepstin-laptop> well, then your webcam probably doesn't support it (you can check with the -list_formats all input option) [18:09] <kepstin-laptop> (as an additional note, the webcam will usually round up the "-framerate" option to its nearest supported framerate; e.g. the one in my laptop only supports 15 and 30fps) [18:12] <Dominique_Michel> Thanks, mine use a very low framerate, around 3 fps, 10 fps as very short peaks [18:12] <Dominique_Michel> Also, with -list_formats, ffmpeg bail with "/dev/video0: Invalid argument" [18:13] <Dominique_Michel> It look to be consistent with v4l-info output which find no standard at all [18:15] <Dominique_Michel> v4l-info find Video_capture "YUV 4:2:2 (YUYV)", 640x480 [19:27] <excalibr> How do you embed idx/sub file into mkv? I know how to do it with .srt but this idx/sub Im not sure which file I should use as the input [19:32] <ubuntuaddicted> trying to complile ffmpeg with libvnenc enabled using a ffmpeg fork located on github but it's getting hung up on opencl, here's the config.log if anyone can help please. https://clbin.com/flKPy [19:35] <llogan> ubuntuaddicted: we do not support forks here [19:36] <ubuntuaddicted> llogan, ah that sucks. so if I try to compile regular ffmpeg and it fails at the same opencl error then you'll help me? [19:37] <llogan> if i know an answer, but i probably won't since i know nothing of opencl [19:40] <ubuntuaddicted> would opencl be required for software like kdenlive (a video editor) or simplescreenrecorder (a video capture program) or obs-studio (a livestreaming software) [19:41] <ubuntuaddicted> or better yet, what is opencl used for? i guess i don't even know if I want it enabled. lol [19:42] <llogan> for ffmpeg specificially, nothing worthwhile [19:44] <ubuntuaddicted> i see that x264 can use it for lookahead per this http://git.videolan.org/?p=x264.git;a=commit;h=3a5f6c0aeacfcb21e7853ab4879f… [19:46] <llogan> you would have to test to see if your encodings would benefit from it or not [19:46] <ubuntuaddicted> llogan, right. it's just that i can't compile ffmpeg because it appears like during ffmpeg compilation it can't find some opencl headers. [19:48] <llogan> IIRC, opencl in ffmpeg is optionally used only by a few filters [19:49] <ubuntuaddicted> llogan, i see it's used by x264, motionsearch, downscale etc etc. according to this anyway http://git.videolan.org/?p=x264.git;a=commit;h=3a5f6c0aeacfcb21e7853ab4879f… [19:50] <llogan> yes, that's x264 [19:57] <ubuntuaddicted> ok, so x264 encoder uses opencl, or should I say "can" use opencl for some things like lookahead etc etc [20:25] <ubuntuaddicted> i'm just hung up on my opencl, i see some header files located in /opt/cuda/include/CL/opencl.h AND /opt/cuda/include/CL/cl.h [20:25] <ubuntuaddicted> but also /usr/include/gegl-0.2/opencl/cl.h [20:37] <brontosaurusrex> interlacing in 2014 > http://paste.debian.net/plain/133499 < do "we" agree? [20:39] Action: kepstin-laptop agrees [20:40] <kepstin-laptop> obviously european-centric, americans have to switch all the 25 to 24 or 30, and the 50 to 60 :) [20:40] <brontosaurusrex> kepstin-laptop: any problems that you see, or would you agree 100% ? [20:40] <brontosaurusrex> yeah, its a short doc for a specific tv station in eu [20:40] <kepstin-laptop> main issue is that getting cameras and production equipment that can do 50/60p is more expensive. [20:41] <brontosaurusrex> well, then shoot 25p, isnt that obvious? [20:41] <ubuntuaddicted> for some reason ffmpeg isn't seeing my libOpenCL headers. [20:41] <kepstin-laptop> ubuntuaddicted: do you actually mean ffmpeg? or are you having issues with x264? [20:41] <ubuntuaddicted> kepstin-laptop, ffmpeg but it's a fork so not sure you want to help me. :( [20:42] <kepstin-laptop> ubuntuaddicted: you don't want x264 built with opencl anyways, it apparently makes it slower and lower quality. [20:42] <ubuntuaddicted> kepstin-laptop, im trying to get ffmpeg built with nvEncodeAPI capabilities. [20:43] <kepstin-laptop> hmm. there are very limited use cases for wanting to use the hardware encode blocks on graphics cards, pretty much limited to "i want to do realtime streaming, and my cpu is already being used for other stuff" [20:43] <ubuntuaddicted> kepstin-laptop, yeap, that's it [20:43] <kepstin-laptop> but if that fits your use case have fun, but I don't know anything about it :) [20:45] <kepstin-laptop> but either way, opencl shouldn't be involved, since opencl means using the graphics card compute resources, so it would e.g. compete with gaming performance. [20:45] <ubuntuaddicted> kepstin-laptop, trying to utilize this https://github.com/Brainiarc7/ffmpeg_libnvenc but for some reason when I try to configure ffmpeg it's not finding my nvidia.icd [20:46] <ubuntuaddicted> kepstin-laptop, my last config.log https://clbin.com/9ykup [20:47] <kepstin-laptop> yeah, you should really try to contact the people working on that repo directly for that. Unless you get lucky and they're hanging around here... [20:48] <mam_> Hi [20:49] <ubuntuaddicted> kepstin-laptop, well i have the opencl-headers20 installed on my machine. i also have cuda installed. so i'm not sure why it's failing [20:49] <mam_> How to merge 2 avi file [20:50] <kepstin-laptop> mam_: please describe what you mean by merge. do you want a single file that plays the two videos in order? do you want one overlayed over the other but both playing at the same time? [20:50] <ubuntuaddicted> mam_, this may help http://unix.stackexchange.com/questions/43896/is-it-possible-to-merge-video… [20:51] <kepstin-laptop> mam_: ffmpeg has an actual faq entry for this: http://www.ffmpeg.org/faq.html#How-can-I-join-video-files_003f [20:52] <mam_> Thanks I will Chack [21:15] <mam_> Hi I checked only first video show [21:23] <mam_> ubuntuaddicted: I checked only first video showed [21:24] <ubuntuaddicted> mam_, not sure sorry. personally i use a linear video editor for joining video files but thats overkill for what you want to accomplish and i'm not sure how to accomplish it using ffmpeg and the command line. sorry [21:56] <MamuDingo> Hi there! Instead of writing a stream to a file, I'd like it to output to stdout. I see in a lot of online documentation that I can use "-" as the output parameter to accomplish this, but whenever I try I get the following error: [NULL @ 0x1be9240] Unable to find a suitable output format for 'pipe:' [21:58] <MamuDingo> http://pastebin.com/39sTt9E4 [21:58] <kepstin-laptop> MamuDingo: when outputting to a pipe, you have to specify the -f output option, since it can't guess from the filename [21:59] <MamuDingo> Ah ha... Okay, I'll try that. Thanks! [21:59] <kepstin-laptop> note that not all output formats can be sent to a pipe; for example, mp4 doesn't support it. [22:00] <MamuDingo> Is there any way to tell which formats support piping? [22:00] <kepstin-laptop> hmm, I don't actually know of any good lists. You can always try and see, it'll give an error if it's not supported. [22:01] <MamuDingo> Well, trial and error is definitely one way. :) [22:01] <MamuDingo> Thanks, again! [22:04] <MamuDingo> rawhide seemed to work for what I needed to do. [22:04] <MamuDingo> *rawvideo ... autocorrect. :( [22:05] <llogan> Cut 'em out, ride 'em in Rawhide! [22:08] <gokuh> hello i need parameter for ffmpeg: add srt in video but want choose font : dejevu.tiff , but how? [22:10] <gokuh> "ffmpeg -i video.avi -vf subtitles=subtitle.srt out.avi" this is OK but i want chose font.tiff how...??? [22:20] <Raccoon> best to just wait until someone in the channel can answer your question. [22:29] <klaxa> gokuh, i think your best bet is to convert the .srt subs to .ass and set a font there (for example with aegisub if you have a GUI) [22:44] <gokuh> klaxa: THANK YOU i try to do convert ok? thank you [23:18] <Vimeo> Hello Everyone i need some help with video filter configuration in ffmpeg. I wanne show a picture if on the udp input of ffmpeg is no signal [23:19] <Vimeo> the output of ffmpeg is published to a rtmp server [23:51] <ScottMichaud> Got to say, there's one thing that I don't understand about FFMPEG: It sometimes brings my PC to ~100% utilization, but it remains completely responsive. o.O [23:51] <ScottMichaud> That's some magic. :p [23:52] <c_14> You can blame your Operating System's load management policies. [23:52] <ScottMichaud> Or, thank? [23:52] <c_14> What, you actually want a responsive system? [23:52] <ScottMichaud> Ideally, yes. :p [23:54] <ScottMichaud> I often need to babysit processes otherwise my machine stutters like mad. Not so with FFMPEG. Not even aware it's still running if I don't look at PowerShell or my CPU charts. [23:57] <klaxa> most of the time i can tell from my fans it's running [23:57] <c_14> You mean the room heater? [23:58] <klaxa> i only have laptops, they don't heat up rooms very well [23:58] <c_14> It's all about placement. [23:58] <ScottMichaud> Pfft. Who wants more than a 35W block heater? :p [00:00] --- Wed Nov 26 2014

1 0