March 2014 - Ffmpeg-devel-irc

ffmpeg-devel.log.20140320
by burek 21 Mar '14

21 Mar '14

[07:15] <wm4> vdpau is broken in git master [07:26] <wm4> and it works in libav git [08:17] <wm4> ah, big WTF [08:17] <wm4> struct Picture is supposed to alias struct H264Picture, but the field offsets don't match [08:17] <wm4> so hwaccel_picture_private is "off" [08:29] <wm4> found the issue, it was caused by libav refactoring [08:55] <ubitux> wm4: but doesn't affect libav? [08:55] <wm4> probably by coincidence [08:55] <wm4> I sent a patch to libav-devel [09:02] <nevcairiel> didnt michael fix such aliasings in various places [11:07] <anshul_> wm4,the code should be cleaned up not to call av_log from a signal. What would you suggest in How to tell the user that he has got seg-fault similar signal [11:18] <wm4> anshul_: don't handle segfaults, that's just broken [11:18] <wm4> for terminal, you could start a cleanup thread [11:38] <anshul_> wm4,thanks [12:47] <cone-171> ffmpeg.git 03James Almer 07master:32291ba6eaac: swresample: add swri_resample_float_sse [13:03] <cone-171> ffmpeg.git 03Janne Grunau 07master:6a74ebc34dd3: float-dsp-test: do not use C99's predefined __func__ [13:03] <cone-171> ffmpeg.git 03Michael Niedermayer 07master:3d04b1aab140: Merge remote-tracking branch 'qatar/master' [13:24] <cone-171> ffmpeg.git 03Michael Niedermayer 07master:9d60527a1315: configure: fix direct symbol ref support detection [16:34] <cone-706> ffmpeg.git 03Michael Niedermayer 07master:d3ed8a4d2b10: ffmpeg: try to avoid calling av_log() from sigterm_handler() [17:19] <Case> I have one mp4 file that causes some trouble for ffmpeg.exe. Even a simple remux attempt gives "Assertion next_dts >= 0 failed at libavformat/movenc.c:620" message and Windows tells the exe crashed [17:28] <ubitux> Case: open a ticket with a sample [17:28] <ubitux> or you're here to fix it yourself? [17:29] <Case> no. I don't understand about mp4 format. But I'm a bit hesitant to go make a ticket as the sample is over 1 GB and copyrighted video file. The cut versions work [17:31] <JEEB> I would recommend grabbing L-SMASH and seeing what you can poke at it with the boxdumper tool [17:31] <JEEB> it will output a human-readable dump of whatever's in the file [17:40] <Case> --timestamp dump that outputs a list of DTS at least doesn't seem to have zeroes anywhere but for the first frame [17:45] <Case> foobar2000's File Integrity Verifier says the file has garbage at the end [17:52] <JEEB> well that assert just means that there's a DTS that's going backwards suddenly [17:52] <JEEB> not sure what next_dts is [17:52] <JEEB> but if it is the next dts value from the previous, that would mean it's not more or the same as the previous one [17:52] <JEEB> thus making the time difference negative [17:52] <wm4> but isn't this the encoder? [17:52] <wm4> also av_assert is AFAIK for internal errors [17:53] <JEEB> yeah, it should... [17:53] <JEEB> can't say much without a sample :P [17:53] <Case> it happens without encoding. ffmpeg -i file.mp4 -c:a copy -c:v copy test.mp4 [17:53] <JEEB> well, it is pushing the data through a demuxer and muxer at the very least [17:53] <Case> it happens very close to the end so it may go to that garbage data [17:53] <JEEB> and muxers are also sometimes called "encoders" [17:53] <wm4> yeah, I meant muxing [17:54] <JEEB> (as with lavf/movenc.c f.ex.) [17:54] <wm4> that might mean ffmpeg.c is feeding the demuxer with invalid data [17:54] <wm4> *muxer [17:58] <Case> I modified that boxdumper tool and made it check for timecode errors. None found [20:50] <cone-796> ffmpeg.git 03Diego Biurrun 07master:e8eace32765d: hpel_template: K&R formatting cosmetics [20:50] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:da760a64d113: Merge commit 'e8eace32765da4b42ac3c575fdb2c70cce8aa224' [21:08] <cone-796> ffmpeg.git 03Diego Biurrun 07master:2972087e3984: hpel_template: Drop some unnecessary parentheses [21:08] <cone-796> ffmpeg.git 03Diego Biurrun 07master:2539cf1c8141: hpeldsp_template: K&R formatting cosmetics [21:08] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:8b66b43128ea: Merge commit '2972087e398454122a82a970ed07924185175c6d' [21:08] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:76d7676fccee: Merge commit '2539cf1c814190a017cb8986b41aa2f949e37fd9' [21:29] <cone-796> ffmpeg.git 03Diego Biurrun 07master:cce791b17bec: hpeldsp_template: Merge some declarations and initializations [21:30] <cone-796> ffmpeg.git 03Diego Biurrun 07master:82ee14d2cedd: ppc: dsputil: comment formatting and wording/grammar improvements [21:30] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:75cc29a8a8e6: Merge commit 'cce791b17becc99f47e097adb93f4b4bbd382e7e' [21:30] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:047cf46e9754: Merge commit '82ee14d2cedd7867920529b408ed6c7ec2f13ff1' [21:43] <cone-796> ffmpeg.git 03Diego Biurrun 07master:30f3f959879e: ppc: dsputil: K&R formatting cosmetics [21:43] <cone-796> ffmpeg.git 03Diego Biurrun 07master:022184a64689: ppc: dsputil: more K&R formatting cosmetics [21:43] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:13405ecdfe7d: Merge commit '30f3f959879eee7890973e8cc9ce076450ced111' [21:43] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:adf82fe4f71c: Merge commit '022184a646896c6b27c6cca387039b030685266e' [21:50] <cone-796> ffmpeg.git 03Diego Biurrun 07master:8bd6f8826674: ppc: dsputil: Drop some unnecessary parentheses [21:50] <cone-796> ffmpeg.git 03Diego Biurrun 07master:b045283f2126: ppc: dsputil: Simplify some ifdeffed function definitions [21:51] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:d648ffe446bc: Merge commit '8bd6f88266740068415f7da0628117f1b802efb9' [21:51] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:eb680f2ab007: Merge commit 'b045283f2126637477a597f184638ebef09bf898' [21:57] <cone-796> ffmpeg.git 03Diego Biurrun 07master:b7d24fd4b221: ppc: dsputil: Merge some declarations and initializations [21:57] <cone-796> ffmpeg.git 03Diego Biurrun 07master:1675975216ec: ppc: dsputil: Drop trailing semicolon from macros [21:57] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:ba91ac5c0e52: Merge commit 'b7d24fd4b2213104c001ed504074495568600b9c' [21:57] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:b20b7724744b: Merge commit '1675975216ecbea8e4826235f204dc0f4234383f' [22:12] <cone-796> ffmpeg.git 03Diego Biurrun 07master:cf7a2167570e: arm: dsputil: K&R formatting cosmetics [22:12] <cone-796> ffmpeg.git 03Diego Biurrun 07master:5169e688956b: dsputil: Propagate bit depth information to all (sub)init functions [22:12] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:a87188ebdb83: Merge commit 'cf7a2167570e6ccb9dfbd62e9d8ba8f4f065b17e' [22:12] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:ba85bfabf3a1: Merge commit '5169e688956be3378adb3b16a93962fe0048f1c9' [22:18] <cone-796> ffmpeg.git 03Diego Biurrun 07master:54a6e08a650a: dsputil: Conditionally compile dsputil code on all architectures [22:18] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:3775224b2f8a: Merge commit '54a6e08a650a71fd375551585b0ea40a3a06d228' [22:27] <cone-796> ffmpeg.git 03Diego Biurrun 07master:0e083d7e4380: build: Group general components separate from de/encoders in arch Makefiles [22:27] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:011d83de488d: Merge commit '0e083d7e43805db1a978cb57bfa25fda62e8ff18' [22:37] <cone-796> ffmpeg.git 03Diego Biurrun 07master:82bb30480132: dsputil: Use correct type in me_cmp_func function pointer [22:37] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:e98bac82e515: Merge commit '82bb3048013201c0095d2853d4623633d912252f' [22:46] <cone-796> ffmpeg.git 03wm4 07master:7948a51b5c3d: vdpau: don't assume Picture and H264Picture are the same [22:46] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:4d216420bed9: Merge commit '7948a51b5c3d08e1a1173442a7ff72b220def303' [22:56] <cone-796> ffmpeg.git 03wm4 07master:1b1094a19d9e: vdpau: switch ff_vdpau_get_surface_id from Picture to AVFrame [22:56] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:1ec295a5c52b: Merge commit '1b1094a19d9e41baf3253c83841f9e5343cecbd0' [00:00] --- Fri Mar 21 2014

1 0

ffmpeg.log.20140320
by burek 21 Mar '14

21 Mar '14

[00:32] <llogan> Kolizer: "ffplay -af volume=0.5 input.foo" worked for me [03:52] <aaa> How do i get my microphone and "what i hear" to work with avconv? http://pastebin.com/mdPyfUZd [03:52] <sacarasc> aaa: #libav is the avconv channel. [03:53] <aaa> ah thanks! [08:23] <Kolizer> Ask in my stream "twitch.rv/kolizer/" Welcome [08:25] <Kolizer> " twitch.tv/kolizer/" * sorry [11:11] <abique> Hello! [11:11] <caral> hi, I'am trying to compile ffmpeg with x264 support on ubuntu according to this guide https://trac.ffmpeg.org/wiki/UbuntuCompilationGuide But the ./configure of x264 tells me http://pastebin.com/2afLgKZj so x264 is looking at the system wide ffmpeg install. Which option I have to set, that x264 is looking in ~/ffmpeg_build instead of the system wide installation? [11:13] <abique> Hi, I need a software library to do high performance scaling and cropping of images, would you recommand ffmpeg for that? [11:14] <Kolizer> "twitch.tv/kolizer/" Welcome [11:20] <relaxed> caral: x264's ./configure has --disable-lavf [11:20] <relaxed> caral: always look at "./configure --help" [11:25] <Max__> I have some dvvideo files that I'm trying to convert to something manageable. It can be lossy, I don't need super high quality. I've tried many (many many) variations of commands I found online, but nothing works "just right". I don't have the video savvy to concoct my own params. Can someone please help me out? This is the file I'm trying to convert: http://pastebin.com/JvQ2SFnd [11:26] <caral> ah thx, I missed ./configure --help , but I do not want to disable the libavformat support. Now I update the system wide installation, seems atm easier to me ;-) [11:26] <Max__> I am not particular about container or codec, only that it be viewable and a reasonable size. [11:28] <relaxed> Max__: ffmpeg -i input.avi -map 0 -vf yadif -c:v libx264 -threads 0 -crf 16 -c:a flac output.mkv [11:28] <Max__> Thanks. [11:29] <relaxed> add -preset veryslow for better compression [11:29] <Max__> What does the -threads flag do? [11:29] <relaxed> it means use multithreaded encoding [11:30] <relaxed> also, add -t 20 for a 20 second sample [11:30] <Max__> What will that do? [11:30] <Max__> Ah, never mind. [11:30] <Max__> Thaks. [11:31] <relaxed> just to test it looks good, etc [11:31] <Max__> Yeah... [11:31] <relaxed> Max__: -crf controls the quality size. Raise its value for smaller/lower quality files. [11:32] <relaxed> er, quality/size [11:32] <relaxed> -crf 16 should look visually lossless [11:33] <Max__> This is the new -qp ? [11:33] <Max__> I found an old command that had -qp 0, and the comments said it went up to 51, being the lowest quality. [11:33] <relaxed> think of it as a smarter -qp [11:34] <relaxed> -qp 0 and -crf zero enable lossless mode for libx264 [11:35] <Max__> I'm getting occasional "AC EOB marker is absent" errors. I understand that this is a bug and I should pay it no mind? [11:36] <relaxed> Don't worry unless it stops [11:37] <Max__> Video came out good, the two audio streams are corrupt. VLC says that it cannot play the "undf" construct. [11:37] <Max__> ffmpeg reports them as flac... [11:39] <relaxed> try -c:a libvorbis -q:a 0 [11:40] <Max__> Hmm... ogg files... [11:40] Action: Max__ broke several MP3 players trying to get them to play Ogg Vorbis. [11:41] <relaxed> 0 is low quality, I was thinking libmp3lame [11:41] <relaxed> use -q:a 5 [11:42] <relaxed> 10 is the highest [11:42] <Max__> Why isn't there a consistent quality scale? [11:42] <relaxed> you can use -b:a 160k instead [11:43] <Max__> That's higher than the input... [12:02] <Max__> relaxed, thanks for all the help. [13:43] <hub_> I downloaded the latest ffmpeg and tried 'ffmpeg -f alsa -i hw:0 out.wav' and it give me this 'Unknown input format: 'alsa' [13:43] <hub_> It's the same with pulse [13:55] <anshul_> is there any way to see specific frame of mp4 in image veiwer [13:59] <JEEB> you can use your favourite thing that uses lavf/lavc in a frame-exact way I guess :P one simple way would be by using aegisub for example (which uses ffms2 in the background to load the file in a frame-exact way). If you are interested in making sure yourself that you're actually looking at the exact same picture then a player like mpv and the framestep buttons probably could be helpful? [14:00] <JEEB> but do remember that if you start comparing pictures you have to make sure they get converted to RGB the exact same way [14:00] <JEEB> which pretty much means that you have to compare them all in the same thing [14:19] <anshul_> jebb, i will try, thanks [14:30] <Aiena> How do I tell ffmpeg to encode all data passing through my soundcard as audio for a video capture from screen ? [14:31] <Aiena> and JEEB thanks for the upscaling tips [15:26] <salamanderrake> where did all the presets go for ffmpeg 2.1.4? [15:29] <JEEB> if you are talking about libx264-related stuff you should switch to x264's internal presets that are available with the -preset option [15:29] <JEEB> http://mewiki.project357.com/wiki/X264_Settings#preset [15:35] <salamanderrake> ok [19:22] <Voicu> hello, is this the right place to ask something related to avformat_open_input ? [19:30] <llogan> Voicu: yes, but libav-user mailing list may be better for API questions [19:32] <Voicu> I see [19:32] <Voicu> well I searched a lot and I'm kinda stuck [19:32] <Voicu> I might've found something but I might as well ask now [19:33] <Voicu> I need to read some video data that doesn't come from a file (so avformat_open_input doesn't work) [19:33] <Voicu> I am now looking into using avio_open_dyn_buf - am I in the right direction ? [19:34] <Voicu> I'm thinking I can somehow create a context, write the data to it and then to avformat_read_packet on it [19:36] <synth_> hello [19:38] <synth_> i'm having issues getting ffmpeg to convert an exr sequence to mov. the command and error i'm getting is at http://pastebin.com/0UdX4fU1 [19:44] <r3c9n> sup [20:09] <gers4302> I have an an mp4 file with an aac stream in it, ffprobe is showing the following: Audio object type 39 is not supported. What does this mean? [20:36] <synth_> i'm having issues getting ffmpeg to convert an exr sequence to mov. the command and error i'm getting is at http://pastebin.com/0UdX4fU1 does anyone know what could be wrong? [20:41] <JEEB> resolution not specified for the exr file in one way or another [20:44] <synth_> oh interesting. let me try specifying that. [20:46] <rsdrsdrsd> I am re streaming an rtmp stream to libx264 and libfdk_aac but I get the following error: missing picture in access unit with size [20:51] <rsdrsdrsd> I am using the following command ffmpeg -re -i rtmp://localhost:80/live/m1 -c:v copy -c:a libfdk_aac rtmp://localhost:80/hls/m2 [20:57] <synth_> JEEB : I tried specifying the input file size and output and am getting a little bit further, however i'm still encountering errors: http://pastebin.com/tSqkYbXG at the bottom i ran a mediainfo on one of the exr files and it shows the size i specified in the command [21:17] <rsdrsdrsd> Please can anyone help? I have been busy with it all day and can't fix it [21:17] <rsdrsdrsd> I also have a bunch of these messages AVC: nal size 21102612 [21:18] <rsdrsdrsd> the rtmp is comming from a flashplayer stream (webcam) [21:33] <rsdrsdrsd> missing picture in access unit with size 40 , what does that mean? [21:38] <smo_> hi, i develop an application wich transcode http stream then pipe it to client... i have a problem my client videoplayer reports duration as Nan:Nan any idear ? [21:39] <smo_> and i tried many formats for ouput only matroska works [22:13] <rsdrsdrsd> I have version ffmpeg version git-2014-03-20 [22:13] <rsdrsdrsd> it must be a bug or somewhat [22:13] <maujhsn> Compiling issues? [22:14] <rsdrsdrsd> no [22:14] <rsdrsdrsd> i want to transocde an rtmp stream [22:14] <rsdrsdrsd> to h264 and aac [22:15] <rsdrsdrsd> I use this command: ffmpeg -re -i rtmp://localhost:80/live/m1 -c:v copy -c:a libfdk_aac rtmp://localhost:80/hls/m2 [22:15] <rsdrsdrsd> I get errors of: AVC: nal size 21102612 [22:16] <rsdrsdrsd> missing picture in access unit with size 40 [22:17] <maujhsn> Can you transcode the stream using gstremer instead? [22:17] <rsdrsdrsd> Really can't find anything to solve it [22:17] <rsdrsdrsd> haven't compiled it [22:18] <maujhsn> Haven't compiled what? [22:18] <rsdrsdrsd> gstreamer [22:18] <maujhsn> Goto #gstreamer [22:20] <rsdrsdrsd> does ffmpeg use gstreamer? [22:20] <maujhsn> Yes! [22:21] <maujhsn> The problem with ffmpeg is that you have to have the right branch installed! If you don't then your screwed! [22:22] <maujhsn> Too many factions...No what I mean! [22:23] <rsdrsdrsd> what is a goodbranch then? [22:24] <maujhsn> A needle in a hay stack! :P [22:25] <maujhsn> Seriously there are good branches out there you just have to ask around! [22:26] <rsdrsdrsd> which one do you advice? [22:27] <maujhsn> Unfortunately The one I would recommend I wouldn't part with it if my life depended upon it! [22:28] <maujhsn> But if you do a search "ffmpeg + fakeoutdoorsman" He had great branches years ago! [22:29] <maujhsn> The latest stuff I wouldn't touch with a ten foot pole! [22:30] <rsdrsdrsd> why not? [22:30] <rsdrsdrsd> are they so bad? [22:32] <synth_> i'm having issues getting ffmpeg to convert a exr sequence to h264 and retain the same colorspace used in the exr sequence. when looking at the generated h264 in linear space, it appears that it's been lifted slightly, however if i use mjpeg, i get an almost identical result to the exr sequence [22:33] <maujhsn> rsdrsdrsd You go with a brand or branch that you can trust! [22:35] <maujhsn> rsdrsdrsd Tell you what, I am considering trying this guys branches! https://ffmpeg.org/download.html#release_2.0 [22:37] <rsdrsdrsd> you mean the static builds? [22:39] <maujhsn> Quality Control? [22:40] <rsdrsdrsd> well let's try to recompile [22:42] <maujhsn> Hint: If anyone is willing to speak up about this guys work...I'm all ears! https://ffmpeg.org/download.html#release_2.0 [22:47] <maujhsn> rsdrsdrsd Are you using libav or ffmpeg? [22:54] <rsdrsdrsd> ffmpeg [00:00] --- Fri Mar 21 2014

1 0

ffmpeg-devel.log.20140319
by burek 20 Mar '14

20 Mar '14

[09:40] <ubitux> i don't find anything to do a ABCDEFGH to ACEGBDGH permutation with aarch64, wtf oO [12:23] <ubitux> ahah @ http://ffmpeg.org/pipermail/ffmpeg-devel/2014-March/155697.html [12:42] <wm4> how witty [12:46] <thardin> maybe I should start actually reading the ML again [12:46] <thardin> but that seems needlessly hostile [12:51] <thardin> why not make explicit c++ examples instead of hackishly making the c examples c++-compatible? [12:56] <nevcairiel> from a maintenance perspective I can understand. The c++ examples have differences in rather small places only, and are otherwise exactly the same, so you would have to update twice the number of examples now, and they run easily out of date as it is [12:58] <ubitux> i like the way we make use of C features [12:59] <thardin> you could librarize the C example [12:59] <thardin> or rather, make a exampletools.c which both the C and C++ examples make use of [13:00] <BBB> it's certainly true that most people just copy the code without even reading it, let alone understanding it [13:01] <thardin> add it to the FAQ (assuming there is one) [13:01] <thardin> considering it is in fact a frequently asked question [13:02] <thardin> then you can just point to it when users ask in libav-user [14:35] <ubitux> wbs: for gas-pp; it seems xcode doesn't like stuff like movi v0.16B, #255, lsl #0 [14:36] <ubitux> but v0.8H "works" (doesn't do the same though) [14:36] <ubitux> that mandatory extra ", lsl #0" is pretty "funny" btw [15:56] <Kovensky> http://seclists.org/fulldisclosure/2014/Mar/332 [16:52] <merbanan> Kovensky: https://twitter.com/nicholemonias/status/446287258682155009 [16:54] <Kovensky> merbanan: he should post that to full-disclosure ;) [16:54] <merbanan> rumor is that he is the reason they closed the list [16:59] <ubitux> oh cool, about 2k$ income this month on spi :) [16:59] <ubitux> (http://lists.spi-inc.org/pipermail/spi-general/2014-March/003315.html) [17:04] <ubitux> the 2 tweets sound like a joke about the whole google drama thread [17:04] <ubitux> (on fd) [17:05] <ubitux> btw, seems debian finally dropped mplayer [17:05] <ubitux> it's funny that they replace it with an even more unmaintained one (mplayer2) [17:05] <wm4> hey, there were two commits in the past year [17:05] <ubitux> :) [17:24] <Plorkyeran_> wasn't the problem that mplayer doesn't even compile? [17:24] <Plorkyeran_> a dead project that does compile is better than a dead project which doesn't [17:24] <nevcairiel> it doesn't compile because they refuse to have ffmpeg [17:39] <ubitux> Plorkyeran_: what nevcairiel said + problem is that it's used as an argument to drop mplayer [17:41] <ubitux> having mplayer2 is not relevant anymore as an argument; mpv might be, but it's a real fork, so not compatible [17:42] <wm4> and who will provide glorious mencoder? [17:42] <ubitux> :) [17:42] <ubitux> wm4: can we close https://trac.ffmpeg.org/ticket/3287 ? [17:42] <wm4> ubitux: yes [17:43] <wm4> I forgot that this existed [17:43] Action: wm4 will try to forget again [18:23] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:1083c479eb6e: avcodec/mjpegdec: factorize parse_avid out [18:23] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:98f9aa389b2d: avcodec/mjpegdec: switch interlaced_polarity to 0 for PAL AVID [18:23] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:046a75eea091: avcodec/mjpeg: print info from parse_avid() if requested [18:23] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:493296800c4c: avcodec/mjpegdec: parse avid data from extradata [18:23] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:dde16f5aaed3: avcodec/mjpegdec: Switch default interlaced polarity for MJPG to 1 [20:46] <cone-969> ffmpeg.git 03Luca Barbato 07master:92ebb3321571: mxf: Add additional H264 PictureEssenceCoding [20:46] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:bd9e0127c9fb: Merge commit '92ebb332157195b7c1212e734c5668afbf19cf2b' [20:53] <cone-969> ffmpeg.git 03Luca Barbato 07master:842ea36c747e: mxf: Override faulty PreviousPartition entries [20:53] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:8263c1469164: Merge commit '842ea36c747eb6537c4236114346cb7a3e9a8df1' [20:54] <michaelni> j-b, how can the change in 842ea36c747eb6537c4236114346cb7a3e9a8df1 be tested ? (it says Reported-By: Jean Baptiste Kempf ...) [21:04] <cone-969> ffmpeg.git 03Luca Barbato 07master:6920fb2b33f8: mxf: Verbose debug logging [21:04] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:3e59db606ed4: Merge commit '6920fb2b33f84b114ea4561ae1c623f4a26562e5' [21:14] <cone-969> ffmpeg.git 03Luca Barbato 07master:af9481d7019d: mxf: Lowercase UL strings [21:14] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:2ef6c0fdf569: Merge commit 'af9481d7019d46c9f18c4ccac77a5d0c54cf6004' [21:20] <cone-969> ffmpeg.git 03Diego Biurrun 07master:606cc8afa1cb: configure: ppc: Disable ldbrx instruction by default [21:20] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:9da88d5e56a8: Merge commit '606cc8afa1cb782311f68560c8f9bad978cdcc32' [21:31] <lindenle> Hi all. Building ffmpeg libraries and have enabled hwaccels but nothing is being listed in the hwaccels section of my configure output. Does this mean i dont have any? [21:32] <J_Darnley> More likely you don't have the required packages/libraries [21:32] <lindenle> J_Darnley: like? [21:32] <lindenle> J_Darnley: I see NEON ... yes but the hwaccels listing is empty [21:33] <J_Darnley> linux's thingy, microsoft's thingy, things like that [21:34] <J_Darnley> dxva, dxva2, vaapi, vdpau, whatever [21:35] <J_Darnley> But since you're compiling for ARM you may not have any of these anyway [21:36] <lindenle> J_Darnley: right, i thought it used to print NEON there...but im not 100% sure. [21:37] <J_Darnley> Probably not, that's an instrcution set [21:37] <lindenle> J_Darnley: ah ok, so it seems there are no hwaccels for amrv7a ? [21:38] <J_Darnley> No idea [21:38] <cone-969> ffmpeg.git 03Martin Storsjö 07master:ae23809184c9: float_dsp: Use LOCAL_ALIGNED for instead of DECLARE_ALIGNED within functions [21:38] <cone-969> ffmpeg.git 03Michael Niedermayer 07master:f502234f9341: Merge remote-tracking branch 'qatar/master' [21:38] <lindenle> J_Darnley: what would I search for [21:38] <lindenle> ... [21:39] <J_Darnley> Find out where ffmpeg has its dxva code then look for an arm equivalent [21:40] <J_Darnley> Alternatively look in configure [21:42] <J_Darnley> Oh wait here you go: ./configure --list-hwaccels [21:47] <lindenle> J_Darnley: yeah looking at that, i was more asking how can i figure out what is available from the chipset... [21:49] <Compn> lindenle : what hw do you have ? [21:49] <lindenle> Compn: well we are building for armv7/android [21:49] <lindenle> Compn: just looking for ways to get more performance out of our JNI code. [21:51] <lindenle> Is libstagefright faster than the internal h264 decoder? [21:51] <Compn> yes [21:51] <Compn> because it uses hardware decoder [21:51] <Compn> i think [21:52] <lindenle> Compn: Maybe that is the best thing to try next... [21:53] <Compn> yes [21:59] <lindenle> Compn: Is there some instructions to build libstagefright? [22:01] <lindenle> nm i found the script in tools [22:41] <cone-969> ffmpeg.git 03Ben Avison 07master:e555e1bc39ea: truehd: tune VLC decoding for ARM. [00:00] --- Thu Mar 20 2014

1 0

ffmpeg.log.20140319
by burek 20 Mar '14

20 Mar '14

[00:02] <Jack64> n-st: you're just trying to get a resized snapshot at a specific time? [00:03] <Jack64> n-st: if so, http://blog.roberthallam.org/2010/06/extract-a-single-image-from-a-video-us… [00:03] <n-st> yes, but i want to resize it to fit given dimensions and preferably center it on a black background [00:03] <Jack64> n-st: check out the link, see if it helps [00:04] <n-st> Jack64: nope, they're all resising the frame without preserving its aspect ratio [00:04] <n-st> i got that working a while ago, but it's less than ideal... [00:05] <llogan> if you want it on a black background you can use pad [00:05] <Jack64> how about using -s 800x-1 ? [00:05] <Jack64> not sure if that works [00:11] <n-st> Jack64: i don't know which dimension is larger with respect to the aspect ratio, i.e. which one i need to specify to get the result to fit the required dimensions... [00:12] <n-st> so i don't know in advance if i need 800:-1 or -1:600 [00:14] <llogan> n-st: is this what you're trying to do? http://superuser.com/questions/547296/resizing-videos-with-ffmpeg-avconv-to… [00:18] <n-st> llogan: wow, that actually works! [00:18] <n-st> the "black" padding isn't exactly black, though [00:18] <n-st> how can i set that to #000? [00:19] <llogan> the default is black [00:19] <llogan> http://ffmpeg.org/ffmpeg-filters.html#pad [00:21] <n-st> llogan: strangely, even if i specify 0x000000, the padding ends up being 0x101010 [00:23] <n-st> llogan: http://pastebin.com/Jip9ibX6 [00:24] <n-st> llogan: 0xFF0000 ends up producing 0xee0e0e [00:24] <n-st> color profile stuff, possibly? [00:24] <n-st> could "Incompatible pixel format 'yuv420p' for codec 'mjpeg', auto-selecting format 'yuvj420p'" have something to do with it? [00:25] <llogan> you didn't include the complete ffmpeg console output [00:25] <n-st> oh, the output... sorry! [00:26] <n-st> llogan: here you go: http://pastebin.com/FzRbUt1T [00:26] <llogan> you're not using ffmpeg from FFmpeg [00:27] <llogan> third party tools, forks, and scripts are not supported here [00:27] <llogan> or go to #libav [00:29] <n-st> llogan: i see. unfortunately, this has to work on a debian server where installing anything is not an option. :( [00:29] <llogan> static builds don't require installation. you just download, extract, and execute. [00:29] <llogan> but if you must use Libav stuff then you'll have to get help at their channel [00:30] <n-st> llogan: i think i'll just live with the dark grey for now ;) [00:31] <n-st> thanks to you and Jack64, you have helped me a lot! [00:36] <llogan> it's a shame that our time is sometimes spent on helping fork users [00:38] <n-st> llogan: i didn't even know that avconv was a fork until today [00:38] <llogan> i'm not blaming you. [00:38] <n-st> i always thought it was a new version created by the original programmers to eventually replace ffmpeg [00:39] <n-st> i've even been slightly worried about basing scripts on ffmpeg, because it seemed to be about to disappear in favour of avconv... [00:39] <llogan> yes. that's what they wanted you to think. [00:43] <caral> Is it possible to have multiple lines with the filter 'drawtext' like \n? [00:43] <caral> like "text \n next line of text" [00:51] <relaxed> caral: try "$(printf %s "hi\nhi")" [00:52] <relaxed> if not, libass would give you much more control. [00:53] <llogan> or you could possibly use textfile option [00:55] <roger21> hello, [00:56] <roger21> when i downmix a dts 5.1 with -ac 2 into a wav, it doesn't do it right [00:56] <roger21> it's like it just put rear or front chanel or something [00:56] <roger21> if i put it into a mp3 or ogg/vorbis it's ok [00:57] <roger21> and if i do the wav from the said mp3 or ogg it' ok too [00:57] <roger21> argh [00:58] <roger21> it just that you know : ffmpeg.exe -i "source.mkv" -map_metadata -1 -vn -c:a pcm_s16le -ac 2 -sn -t 9 "audio.wav" [00:58] <roger21> ...ok [01:02] <roger21> http://pastie.org/8949065 [01:06] <caral> The printf try doesn't work. Output is "hinhi" But textfile works like a charm. Extra \n gives me also an extra Linebreak for the eg. timecode. thx :-) [01:07] <roger21> the mp3 or ogg command just use the -ac 2 too [01:07] <roger21> i could try to remove it though [01:08] <llogan> i can't duplicate the issue [01:09] <roger21> nop same, w/o it [01:09] <roger21> well sucks for me [01:10] <roger21> well no it create a 6 chanel wav in this case [01:10] <roger21> but i just hear two chanel when playing it -> so same sound but not same file [01:11] <roger21> could my ffmpeg version be buggy? [01:13] <roger21> can i manually down mix it, to force using the 6 channels [01:13] <roger21> ? [01:15] <llogan> roger21: you can choose whaever channels you want with pan audio filter [01:20] <roger21> how does it do it normaly, it just pick 2 channels or it mix all together? [01:21] <roger21> because the pan things just let me pick channels [01:21] <roger21> when is do the mp3 i clearly have all the channels or something, the sound is "big" [01:21] <roger21> when i do the wav it just suck [01:23] <Hello71> ... [01:25] <roger21> can you see what i mean, the files : http://roger21.free.fr/tests/med/ [01:38] <roger21> i have to remove the files [02:07] <cpen> hello, I have a partial flv that I fetched from a server that supports flv "pseudo-streaming". this partial flv can be played in flash (i.e., when seeking to a point beyond where you've buffered). however, when I try to use ffmpeg to copy the video and audio from that partial flv into an mp4, it's able to grab the audio, but not the video. what gives? how can I help ffmpeg be able to copy both streams? here is a pastebin with several of t [02:07] <cpen> commands I've run on my test file: http://pastebin.com/NNRidTrz [02:17] <decci> Hi [02:18] <decci> My ffmpeg is not working. I am running ffmpeg on CentOS 6.4 [02:18] <decci> http://ffmpeg.upeoapp.com/upload_processor.php [02:19] <decci> If you try to upload any video file, it is not working full fledge [02:21] <wallbroken> hi guys, can I please ask a question about avidemux? [02:21] <decci> How to install ffmpeg under /usr/local/bin [02:22] <sacarasc> That's not really in the scope of this channel, wallbroken. [02:22] <wallbroken> yes, i know, but i don't know wherever to ask [02:22] <sacarasc> decci: ./configure --prefix=/usr/local [02:22] <wallbroken> it's a simple little think [02:23] <wallbroken> i only want to ask why in avidemux xvid level goes from 1 to 4, and 5 it is not [02:23] <decci> sacarasc: http://paste.ubuntu.com/7117339/ [02:23] <decci> sacarasc: Does it mean its working [02:23] <sacarasc> decci: I have no idea what script that is you are using. If you use just ffmpeg, maybe someone here could help. [02:23] <decci> sacarasc: It says File Uploaded but not stored [02:24] <sacarasc> That sounds like a third party script problem, which is not supported here. [02:24] <decci> sacarasc: I just followed https://trac.ffmpeg.org/wiki/CentosCompilationGuide [02:26] <decci> sacarasc: I installed ffmpeg under /root/ffmpeg_sources dfirectory. Will simply a symbolic link help as ln -s /root/ffmpeg_sources/ffmpeg /usr/local/bin/ffmpeg [02:44] <relaxed> decci: wouldn't that be /root/ffmpeg_sources/ffmpeg/bin/ffmpeg ? [02:49] <decci> relaxed: I want it to be under /usr/local/bin/ffmpeg [02:49] <decci> relaxed: Can you suggest [02:49] <decci> relaxed: Can simply symbolic link help here [02:50] <decci> relaxed: I followed this https://trac.ffmpeg.org/wiki/CentosCompilationGuide [02:52] <relaxed> did you follow that guide exactly? [02:54] <relaxed> decci: ^^ ? [02:55] <decci> yes [02:55] <decci> relaxed: anything I missed out [02:55] <relaxed> ok, run this: echo $'#!/bin/sh\nLD_LIBRARY_PATH=/root/ffmpeg_build/lib /root/bin/ffmpeg "$@"' > /usr/local/bin/ffmpeg [02:55] <relaxed> chmod +x /usr/local/bin/ffmpeg [02:56] <relaxed> now test: /usr/local/bin/ffmpeg [02:57] <decci> relaxed: echo $'#!/bin/sh\nLD_LIBRARY_PATH=/root/ffmpeg_build/lib /root/bin/ffmpeg "$@"' > /usr/local/bin/ffmpeg -bash: /usr/local/bin/ffmpeg: Is a directory [02:57] <decci> relaxed: I opened up a new putty console and the ln -s wont persist it seems [02:57] <decci> How to fix it? [02:57] <relaxed> rm -rf /usr/local/bin/ffmpeg [02:58] <relaxed> then run the commands I gave you again [02:58] <decci> relaxed: doe [02:58] <decci> done [02:58] <decci> relaxed: it went well [02:58] <relaxed> I will fix that guide sometime soon to be more clear. [02:59] <decci> relaxed: can you try uploading and checking this [02:59] <decci> relaxed: http://ffmpeg.upeoapp.com/upload_processor.php [03:00] <decci> relaxed: http://paste.ubuntu.com/7117467/ [03:00] <relaxed> I do not have time. [10:44] <Kasper^> how do you specify a keyframe interval when encoding to vp9 ? [10:46] <JEEB> I think ffmpeg cli has the general option -g [10:46] <JEEB> not format-specific [10:49] <Kasper^> "I" frames ? [10:49] <Kasper^> option -g defines space between I frames; are these keyframes ? [10:50] <JEEB> no, it defines the GOP length. Depending on the format (and settings) it's either an I picture, an IDR picture, an I picture with refresh signaling or whatever, or an IRAP picture [10:50] <JEEB> personally I prefer calling them RAPs, Random Access Points [10:51] <JEEB> how they're called specifically is format-specific and if something calls them "I" pictures that's due to MPEG-1 and MPEG-2 where I pictures were RAPs [10:52] <Kasper^> so basically using -g would help seeking through the video at random points [10:52] <JEEB> but yeah, GOP is a group of pictures that has a RAP in the beginning of it, so -g effectively sets the maximum distance between RAPs [10:53] <Kasper^> that is what I am after actually [10:53] <JEEB> yes, a shorter GOP length lets you seek quicker to random points [10:54] <Kasper^> thanks [10:54] <JEEB> also I wish thee luck to encode in vp9 :P [10:54] <JEEB> the encoder is damn slow [10:54] <Kasper^> now will just keep fingers crossed for vp9 multithreaded encoding [10:55] <JEEB> not gonna happen any time soon :P [10:55] <Kasper^> hard to parallelize the code ? [10:55] <JEEB> not harder than other similar formats I would guess, no idea about the code [10:55] <JEEB> it's just that it definitely doesn't seem to be a thing Google wants to put effort into [10:56] <Kasper^> hmm, that's weird, they own youtube [10:56] <JEEB> well, almost all major video sites nowadays encode in parallel, so the speed of a single instance doesn't matter too much [10:56] <Kasper^> and why put effort in an format if nobody can use it [10:57] <JEEB> so you either encode multiple videos at once, or you run a quick first pass with something to create the GOPs (I guess x264's pretty good for this?), and then you encode those separately [10:58] <Kasper^> I wonder if 2-pass encoding improves compression with vp9 [10:58] <JEEB> dunno, but it makes it even slower [10:58] <JEEB> personally I would just use whatever their version of CRF is [10:58] <JEEB> but yeah, I'm not gonna see VP9 being much useful any time soon [10:58] <JEEB> unless you want to code that parallel encoding setup and then stitch the bitstream together [10:59] <Kasper^> what's weird is that during compression the fps keeps dropping [10:59] <JEEB> have fun with libvpx :) [10:59] <Kasper^> smells like some inefficient search/loop to an ever increasing data structure [11:00] <JEEB> "Pass 2/2 frame 57/32 182737B 15553831 ms 0.22 fpm [ETA 77:15: 44]" [11:00] <JEEB> quote from a friend who wanted to test libvpx's vp9 [11:03] <Kasper^> and I need to convert about 50gb of old clips [11:05] <JEEB> then you don't want to use vp9 :P [11:19] <exed_> are there common pitfalls for decoding an unicast/multicast mpegts stream to plain s16le audio? [11:20] <exed_> we always having jitter inside the recorded audio as well as PES pkg mismatches and we don't know how we could get rid of them? [11:20] <exed_> or at least how we can debug this problem better? [12:10] <alket> is it possible to update file every 1 minute while playing video ? [14:01] <nyc3j> Hello! I ran into an odd problem: I launch ffmpeg-git compilation on Archlinux and I got "ERROR: gnutls not found", then I have run it again and it is (almost) working! Is it a known issue? [14:01] <nyc3j> *launched [14:27] <DrSlony> Hi, my screencast doesn't play in Chromium. I use ffmpeg-1.2.6 and chromium-33.0.1750.149. Exact code here: http://paste2.org/1GGxO8kh What do I do to make my h.264 videos more compatible? [14:28] <DrSlony> oh, sample video here http://rawtherapee.com/bugs/2301/2301pipettesegfault.mp4 [14:31] <klaxa|work> use webm instead maybe? [14:32] <klaxa|work> that will require you to encode to vp8 though [14:32] <klaxa|work> (or vp9 if you are using that) [15:48] <hayuto> hey [15:49] <hayuto> is there a way to capture imgae from rtmp stream to file.png and then exit and then capture again (in cron) [15:49] <hayuto> not to capture all the time [15:49] <hayuto> i want ffmpeg to exit after capturing [15:50] <klaxa|work> ffmpeg -i rtmp://wherever -vframes 1 -c:v png picture.png [15:51] <hayuto> oh vframes [15:51] <hayuto> ill try that [15:55] <hayuto> but if there is no stream it hangs [15:55] <hayuto> and stays hanging [15:55] <hayuto> can it return error instead so i can check $? [15:56] <hayuto> or whatever is last return value [15:57] <hayuto> i want to have something like in dailymotion [15:57] <hayuto> an actual thumbnail on stream [15:57] <hayuto> but if theres no stream and cron starts ffmpeg to capture it it will just stay hanging [15:58] <klaxa|work> maybe use something beforehand to check if the stream is up? [15:58] <klaxa|work> it also shouldn't hang [16:00] <hayuto> just stays like this http://i.imgur.com/5BAtxSU.png [16:01] <klaxa|work> well the port is open... so it will send a request and if nothing comes in return it will block, that makes sense [16:01] <klaxa|work> maybe the server is not working correctly [16:02] <hayuto> so if i put it in cron it will make so many process with it [16:02] <hayuto> i could just kill it after 10 seconds or so anyways [16:14] <hayuto> ok i did it like this /usr/local/bin/screen -UdmS lalalalalala /usr/local/bin/ffmpeg -i rtmp://37.59.202.234:13335/live/hayufdfdfto -vframes 1 -c:v png img3.png && sleep 10 && screen -X -S lalalalalala quit [16:14] <hayuto> tought probably there's better way [17:26] <roger21> hello, i'm confuse about a 6 channel audio track [17:26] <roger21> media info says Front: L C R, Side: L R, LFE [17:27] <roger21> well .. i don't know how they are called in ffmpeg [17:27] <roger21> i guess FL FC FR SL SR ... and er [17:28] <roger21> also that makes 2 central chanel the FC and the LFE [17:28] <tsjiller> ffmpeg --tracks or something lists everything [17:28] <roger21> how does that map into 5.1 ? [17:28] <roger21> oh i'll try that [17:28] <tsjiller> LFE is the .1 [17:29] <roger21> but what is FC then [17:29] <tsjiller> Front Center [17:29] <roger21> that make 4.1.1 [17:29] <roger21> i suck at home cinema [17:29] <tsjiller> no, the center channel is full frequency [17:30] <roger21> ok i guess [17:31] <roger21> so the or something will come useful here [17:31] <roger21> i don't find the tracks option [17:32] <DrSlony> klaxa|work i dont want to use webm or vp* ;] [17:32] <klaxa|work> then your videos won't get more compatible [17:33] <klaxa|work> chromium and firefox on linux don't support h264 within the browser afaik [17:36] <spaam> Ohh.. they used "Lavf55.19.104" to make the Free To Play movie http://store.steampowered.com/app/245550/ :) [17:37] <roger21> nobody on my tracks? option or something? [17:37] <DrSlony> firefox too?? [17:37] <DrSlony> huh [17:38] <DrSlony> how widely compatible is webm or vp*? [17:38] <roger21> wide enougth [17:38] <DrSlony> ok [17:39] <DrSlony> best two-word answer i could hope for :] [17:39] <DrSlony> what si the recommended ffmpeg syntax for encoding to vp9? [17:39] <klaxa|work> vp8 and vp9 are royalty free, anyone may implement encoders and decoders free of charge [17:39] <roger21> https://developer.mozilla.org/en-US/docs/HTML/Supported_media_formats [17:42] <roger21> ffmpeg -layouts [17:47] <roger21> holly shit i did it [17:48] <DrSlony> do presets work with ffmpeg/vp9? [17:48] <DrSlony> -c:v libvpx-vp9 -preset slower ? [17:48] <roger21> -af "pan=stereo:FL=FL+FC+SL+LFE:FR=FR+FC+SR+LFE" to have a proper -ac 2 [17:49] <roger21> yeah me [17:50] <JEEB> DrSlony, -preset is libx264-specific as far as I know [17:50] <JEEB> it's a "private option" [17:50] <JEEB> so you basically will have to look at libvpx encoder wrapper to see if it has a mapping for the -preset option [17:50] <JEEB> libavcodec/libvpx whatever dot c [17:51] <DrSlony> ok JEEB [17:51] <JEEB> but I'm pretty sure it doesn't have a mapping for it [17:51] <JEEB> because as far as I know, libvpx doesn't have "presets" [17:51] <DrSlony> so then how is it controlled? [17:51] <DrSlony> i read it just has two profiles [17:51] <DrSlony> but thats more a compatibility thing [17:51] <JEEB> I'd say looking at the wrapper most probably would be the quickest way to check what private options there are :) [18:00] <roger21> ok that is better -af "pan=stereo|FL=FL+0.5*FC+SL+0.5*LFE|FR=FR+0.5FC+SR+0.5*LFE" /nobody care [18:01] <roger21> can i now what is supported for a said format ? [18:02] <roger21> for exemple for wav, isee i can put 16le 32le and other won't work, is ther a way to list what goes [18:17] <Aiena> Hi I need some help [18:18] <Aiena> I made an ffmpeg script it works fine on my system. IN my freinds machine it used to work fine. But of late he gets these errors http://pastebin.com/E3p7sn1W and the file output is corrupt. Can someone elnlighten me as to the cause as I cannot understand the error. [18:26] <jmul> Hello [18:26] <jmul> I'm doing mp4 / h264 muxing / encoding [18:27] <jmul> Using libavformat, libavcodec, etc in C [18:27] <jmul> Everything works, but I'd like to minimize the ffmpeg build for mp4 muxing and h264 encoding [18:27] <Mavrik> aaand? :) [18:27] <jmul> so I'm wodering which configure flags I need to pass while building ffmpeg to allow for mp4 muxing and h264 encoding [18:28] <jmul> I've tried the following but it doesn't work (give me a minute to make a pastebin): [18:28] <Mavrik> --disable-everything --enable-muxers=<what you need> --enable-encoders=<what you need> ? [18:28] <Mavrik> its pretty well documented in ./configure --help [18:29] <jmul> I've looked through ./configure --help, but there are some gotchyas sometimes [18:29] <jmul> for example, --enable-encoder=h264 is meaningless while --enable-libx264 is the actual flag needed for enabling the h264 encoder [18:31] <Mavrik> because you're using an external library, yea. [18:31] <Mavrik> of course, you need --enable-encoder=libx264 too when using --disable-everything ;) [18:31] <Mavrik> ffmpeg -formats and ffmpeg -codecs will give you info what to pass there [18:31] <Mavrik> also remember to enable any filters if you're using them. [18:33] <jmul> Mavrik, I've actually built without --enable-encoder=libx264 and it seems to work. [18:33] <jmul> Here's my current config: [18:33] <jmul> http://pastebin.com/CkG0uK9k [18:35] <jmul> I'm fairly sure I'm missing a muxer. But, I'm not sure what muxer to enable. [18:36] <jmul> When I compile with that, then avformat_alloc_output_context2(&outCtx, NULL, "mp4", filename); fails. [18:37] <jmul> Mavrik: Thanks for your help, btw. [18:37] <kxra> does fate only do regression testing, or benchmarking as well? [18:37] <Mavrik> hmm, --enable-protocol=file might be missing :) [18:39] <jmul> Aha! Good catch. Unfortunately, it's only missing in my pastebin and not in my script. [18:41] <Mavrik> lemme check other build scripts [18:42] <jmul> Mavrik: Thank you so much. [18:43] <jmul> http://www.ffmpeg.org/ffmpeg-formats.html#mov_002c-mp4_002c-ismv lists two other muxers: mov and ismv [18:43] <Mavrik> oh yeah [18:43] <jmul> Do you think those are relevant? [18:43] <Mavrik> mp4 muxer is actually mov muxer [18:43] <Mavrik> so yeah :) [18:43] <Mavrik> hrrm, I'm not using mp4 anywhere, gah [18:44] <jmul> bingo. I'll try that. fingers crossed. [18:44] <Mavrik> this part of build process is kinda broken :/ [18:44] <jmul> What do you mean? [18:44] <kxra> from the website it looks like FATE just does regression testing? but i wanted to double check [18:55] <kxra> but there are these two posts which make me wonder if benchmarking was added: http://multimedia.cx/eggs/?s=benchmarking&submit=Search [18:56] <llogan> roger21: LFE is usually recommended to be excluded when downmixing 6 to 2 channels [18:56] <roger21> oh [18:57] <llogan> what's wrong with -ac 2? [18:57] <roger21> results sucks [18:57] <roger21> it's all low [18:57] <llogan> oh, the wav vs mp3 difference? [18:57] <roger21> yes... [18:58] <llogan> you can ask about that on ffmpeg-user mailing list. might get an answer. i don't have one. [18:59] <roger21> well.. i won't, i'm in a hurry, i just need a set of test files i'll jump to something else [18:59] <roger21> so has it works quite ok with -af pan i'll go with that [18:59] <llogan> should only take a few minutes to send a message to ffmpeg-user(a)ffmpeg.org (and then up to 12 hours for me to approve it depending on when you send it) [19:00] <llogan> also seaarch for LFE in http://lists.ffmpeg.org/pipermail/ffmpeg-devel-irc/2014-March/001958.html [19:00] <llogan> *depending on when you send it and if you are a subscribed user or not [19:04] <jmul> Mavrik: I think that was a solution to the muxer problem. [19:05] <jmul> Mavrik: I also think you were right about --enable-encoder=libx264, because now that I've disabled everything and am getting past the avformat alloc I'm getting an error that it can't find the h264 encoder. :P [19:15] <jmul> Mavrik: I've just made a minimal build that works in my project. Thank you for all your help. [19:16] <jmul> It's not exactly minimal, I've built with "--enable-muxer=h264,mp4,mov,ismv --enable-encoder=libx264", but I can rule out the unecessary muxers later. [19:22] <llogan> kxra: not that i know of, but you could user the fate-suite as a corpus to perform your own benchmarks [19:22] <llogan> *use [19:23] <llogan> Aiena: your paste is missing the command. [19:25] <Aiena> llogan oops sorry [19:25] <Aiena> one moment [19:27] <Aiena> llogan http://pastebin.com/aBfZP5sq [19:28] <Aiena> ^^ I included the script and the output I cannot reproduce it [19:28] <Aiena> for him sometimes its succeeds sometimes it fails for me it hasn;t failed but I admit he uses it much more [19:28] <Aiena> first i thought it was a window out of bounds issue [19:28] <Aiena> but it does not seem like it [19:29] <Aiena> if it was window out of bounds it would record the wrong area of the screen that is all [19:30] <cpen> hello [19:31] <cpen> I have a question about using ffmpeg to copy streams from a partial flv [19:31] <cpen> I fetched it from a server that supports flv pseudo-streaming [19:31] <cpen> when I try to copy the video and audio, ffmpeg can get the audio, but not the video [19:32] <cpen> here is my attempt: http://pastebin.com/NNRidTrz [19:37] <llogan> Aiena: the console output indicates that the input video size is larger than the max size of the display, and/or that the offsets cause it to go out of bounds [19:38] <Aiena> llogan but the strange thing is that the video is uscaled later the offsets are from xwininfo I guess that the wondow is not positioned correctly by him [19:38] <Aiena> upscaled [19:40] <llogan> the output upscaling doesn't matter (and why upscale)? [19:45] <diroots> hi there, looking for some help on building ffmpeg with decklink support on mac os x ... [19:55] <Kolizer> hello people [19:56] <tjablin> Is it possible for ffmpeg to load a filter from a shared object? [19:58] <tjablin> Is it just a matter of calling avfilter_register in the shared object's __attribute__((constructor)) method? [19:58] <tjablin> And then doing an LD_PRELOAD? [19:59] <Kolizer> As can turn off the sound in ffplay? [20:01] <Kolizer> There is such a button? [20:03] <jmul> Kolizer: From the man page (man ffplay) it seems that passing -an when opening ffplay will turn off the sound [20:04] <jmul> Kolizer: not sure if that helps or if you're looking to disable the sound after the content has started playing. [20:04] <Kolizer> But in time viewing the video, you can turn off the sound and after you enable? [20:04] <jmul> I do not know [20:04] <Kolizer> ok [20:05] <Kolizer> There is a Russian? [20:05] <Kolizer> Curious [20:09] <Aiena> llogan so that text is sharper when downscaled during playback [20:09] <Aiena> as video players downscale in RGB I think [20:10] <Kolizer> I just Russian, byloby nice to talk [20:11] <Kolizer> ahaha [20:11] <maister> I'm trying to stream a TS over RTP. Is this supposed to work? ffmpeg -f mpegtsraw -i foo.ts -codec:d copy -f rtp rtp://127.0.0.1:8000. I'm getting an error: "Output file #0 does not contain any stream". [20:12] <maister> -f mpegtsraw creates a stream with "fake data codec" mpeg2ts. [20:35] <maister> llogan, http://pastebin.com/5pm3BhRu [20:41] <llogan> maister: did you try: ffmpeg -i /tmp/test.ts -codec copy -f rtp rtp://127.0.0.1:8000 [20:42] <maister> llogan, ye, but the TS has multiple streams, so it fails with "Only one stream supported" in the RTP muxer. [20:43] <tjablin> I found a thread from 2010 in which the ffmpeg developers declined a patch to implement dynamically loaded plugins. Does ffmpeg still discourage dynamically loaded plugins? [20:43] <maister> llogan, http://pastebin.com/GniW3fcj [20:50] <llogan> what's liblinelet? [20:54] <maister> codec I'm working on, but that shouldn't be relevant I think. [20:58] <llogan> maister: does it have to be RTP? also you may want -re as input option but that is unrelated to one stream issue. [20:58] <llogan> possibly relevant: http://stackoverflow.com/a/16468600/1109017 [20:59] <maister> llogan, yes. Problem here is that somehow the "raw mpegts" codec isn't forwarded properly to the muxer. I'll try to debug a bit. [21:00] <llogan> i don't know. i'm not much of a streaming guy. [21:02] <llogan> you could ask ffmpeg-user mailing list if you want more useful input (possibly) [21:05] <maister> ok [21:29] <Aiena> llogan found the issue [21:29] <Aiena> nothing wrong with the script [21:30] <Aiena> I asked my freidn several times if the window was out of bounds he did not understand because of a language barrier [21:30] <Aiena> but then he finally got it :) [22:04] <maister> llogan, I think I figured out what happens. in ffmpeg_opt.c: 1781 it checks for default maps for video/audio/subtitle codecs, but *not* data. [22:04] <maister> Is it possible to use -map or something here to force it? [22:08] <maister> ye, -map 0 did the trick :D [22:29] <Kolizer> It is possible to adjust the sound in ffplay? [22:31] <DeadSix27> volume? [22:31] <DeadSix27> -> http://trac.ffmpeg.org/wiki/How%20to%20change%20audio%20volume%20up-down%20… [22:31] <DeadSix27> same filter should work in ffplay. [22:31] <Kolizer> yes [22:31] <DeadSix27> -af 'volume=0.5' [22:32] <Kolizer> oh no [22:32] <Kolizer> ffplay [22:32] <DeadSix27> ? [22:32] <Kolizer> volume control [22:33] <Kolizer> need [22:33] <Kolizer> in ffplay [22:33] <Kolizer> or off volume [22:34] <DeadSix27> ye just realised volume filter does not work [22:37] <pyBlob> why can't I stream from one pc to another using these commands? [22:37] <pyBlob> http://pastebin.com/vwWj8utq [22:40] <Kolizer> I want to use ffplay in the form of player [22:41] <Kolizer> It turns out he is not able to control the sound [22:42] <Kolizer> Maybe there are some solutions? [22:44] <pyBlob> Kolizer: you could adjust the global audio volume [22:44] <Kolizer> yes))))) [22:45] <Kolizer> i know [22:47] <Kolizer> Such a cool functionality and such important things no( [22:47] <pyBlob> well ffplay is just a viewer, not a fully fledged player [22:48] <Kolizer> ok [22:48] <Kolizer> There are Russians? [22:49] <DeadSix27> Kolizer you could also use the volume mixer, if in windows. [22:49] <Kolizer> yes yes i know [22:50] <Kolizer> i no use windows bue [22:52] <Kolizer> thanks people [22:53] <pyBlob> any ideas, why this isn't working? http://pastebin.com/vwWj8utq [22:56] <brontosaurusrex> pyBlob, try to use -i file.input [22:57] <brontosaurusrex> for the file that is supposed to be the input file [22:57] <brontosaurusrex> also, you are supposed to post the full-log [22:57] <pyBlob> whoops ... my fault, but still doesn't work [23:04] <pyBlob> full logs: [23:04] <pyBlob> client http://pastebin.com/uGQviP0f [23:04] <pyBlob> server http://pastebin.com/RVUXmpDS [23:14] <pyBlob> woo, it works =) [23:15] <pyBlob> damn firewall, I've written 192.168.2.0 instead of 192.168.2.0/24 [23:15] <Kolizer> nice [00:00] --- Thu Mar 20 2014

1 0

ffmpeg-devel.log.20140318
by burek 19 Mar '14

19 Mar '14

[01:22] <BBB> oh classic [01:22] <BBB> plepere: you removed SECTION .text at the top [01:22] <BBB> plepere: so the whole section (file) is .rodate, which (on mac) is non-executable [01:23] <BBB> plepere: so it crashes on first execution of any dsp function in that file [01:23] <BBB> plepere: so I'd re-add SECTION .text [01:23] <BBB> mraulet: for you also ^^ [01:24] <mraulet> cool thanks [01:45] <mraulet> BBB: it works back, thanks [02:05] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:8ab80707841a: avcodec/utils: fix sizeof(AVFrame) dependence in avcodec_encode_audio2() [02:13] <cone-714> ffmpeg.git 03Carl Eugen Hoyos 07master:bb97bcec6b2b: Fix TARGET_SAMPLES and TARGET_PATH for some fate tests. [02:13] <cone-714> ffmpeg.git 03Marton Balint 07master:2daf6be01162: MAINTAINERS: add myself az mpegts demuxer maintainer [02:13] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:aa86cccd1145: Merge remote-tracking branch 'cehoyos/master' [03:12] <BBB> mraulet: yw [09:17] <plepere> thanks BBB [11:16] <ubitux> there are some really cool things in aarch64 [11:17] <ubitux> but really sometimes it's ugly as hell [11:17] <ubitux> it's worse than the previous aliasing [12:47] <BBB> sometimes I think they should rename codec-devel to codec-newbie or codec-fanboy or so... [12:55] <anshul_> I was adding examples of c++, I heard that its lack of intigration with build system, what are the things that I have to take care to intigrate with build system [13:05] <ubitux> anshul_: look for "example" in the configure [13:12] <anshul_> ubitux, example_list is same for c and cpp [13:27] <ubitux> anshul_: and so it builds both when you make examples? [13:27] <anshul_> how and where example target is defined, i dont find it in Makefile, and make example works [13:27] <ubitux> anshul_: it's in doc/Makefile [13:28] <ubitux> you have all kind of rules there [13:28] <anshul_> ok, I will look o that [13:50] <anshul_> I did added some target but its searching for c files rather then cpp, while the Makefile inside examples_cpp is working [13:50] <anshul_> its giving me error No rule to make target `doc/examples_cpp/avio_reading.c', needed by `doc/examples_cpp/avio_reading.o' [14:15] <anshul_> I made a test make file http://pastebin.com/Ff10sxCz [14:16] <anshul_> which looks for [14:16] <anshul_> [anshul@daku_daddy C]$ make examples [14:16] <anshul_> g++ johnny.cpp -o johnny [14:17] <anshul_> but in ffmpeg for same rule it does not consider cpp file only ask for c file [14:26] <anshul_> ok I got something from make -d command, our makefile have: Found an implicit rule for `doc/examples_cpp/avio_reading.o' [14:50] <anshul_> Ohh In ffmpeg we have disabled default suffix rule [17:08] <shubhamjain1> I hope this channel is active. [17:08] <shubhamjain1> My question is not specifically related to FFMPEG. I am building a simple flv to mp3 convertor. I succeeded at extracting audio frames from flv but can't build proper mp3 frames. When converting with ffmpeg I observed that it does resampling even if we convert at the same bitrate. I was curious if thats a necessary part of conversion. Is it not possible to just copy samples as is? [17:30] <ubitux> is there something to get ACEGBDGH from ABCDEFGH with aarch64/simd/16bits*8lanes ? [17:30] <ubitux> i might be able to get something with uzp+uzp2 but that looks overkill [17:30] <ubitux> vuzp was doing the correct thing in armv7 neon [17:30] <ubitux> not sure that's the good place to ask though [17:41] <ubitux> does anyone knows if gdb can be used to exec instruction on the fly? [18:03] <mopp> stumbled on some weird close caption on video.adultswim.com. did a bit of research and found this: https://trac.ffmpeg.org/ticket/3172 any news on this? [18:30] <lindenle> Hi all. How can i pass tune parameters to libx264? [18:31] <lindenle> or preset for that matter? [18:31] <Case> sounds like you want to ask that in #ffmpeg, lindenle [18:32] <lindenle> Case: im talking about programatically in libavcodec. [18:32] <Case> ok [18:39] <lindenle> Case: looks like i can use av_opt_set ... [20:10] <michaelni> what happened to cone ? [21:38] <llogan> a broken game format. where's mike m? [22:04] <Compn> llogan : hes not on irc, have to email him [00:00] --- Wed Mar 19 2014

1 0

ffmpeg.log.20140318
by burek 19 Mar '14

19 Mar '14

[02:33] <Compn> anyone here with ffmpeg want to test rtmp://cp49989.live.edgefcs.net/live/streamRM1@2564 and see if it works ? i'm curious if its native rtmp or librtmp bug [02:36] <relaxed> Compn: http://pastie.org/8942222 [02:37] <relaxed> ffplay plays it [02:41] <Compn> relaxed : yes, i got it working with "rtmp://cp49989.live.edgefcs.net/live/streamRM1@2564 live=yes" [02:41] <Compn> thanks for testing [02:42] <Compn> my version was last month, strange :) [04:26] <Gregor> Anybody know of a filter (or other system) in ffmpeg that could generate some reasonable notion of the motion between frames? I looked at using deshake's logs, but it gives up so easily that most frames are marked as having no motion. [04:27] <Gregor> I just want some unitless value that vaguely represents how much motion there was from one frame to the next. I'm currently just using difference in pixel luminance, but with some motions, particularly sweeping pans, it artificially inflates the amount of change. [05:17] <sinclair|work_> Gregor: not sure mate, but i am intrigued by the things you just said [09:50] <newbie2> Is this the correct place to ask for FFMPEG help? [09:53] <relaxed> newbie2: yes [09:56] <Eventh> Is there a way to select two streams (audio and video) in the tee pseudo-muxer? select allows me to specify one or all streams? http://pastie.org/8946733 I would like each ouput file to contain a video stream and the audio stream. [09:59] <relaxed> which streams do you want? [10:00] <relaxed> is -flags +global_header needed? [10:00] <Eventh> each output file should have 1 video and 1 audio [10:00] <Eventh> If I remove it I get the following error "[mp4 @ 0x3f5b4a0] Codec for stream 0 does not use global headers but container format requires global headers" [10:01] <relaxed> I don't think you want the tee muxer then. [10:01] <relaxed> ffmpeg -i input -map 0 [10:02] <relaxed> er... [10:02] <newbie2> Hope someone can help: Currently I have Centos 6.3 with FFMPEG 0.6.5 from repository and recording all works well (RME Hammerfall HDSP 9632 card). I wanted to use the latest FFMPEG and all appears to compile fine, but when I record with the same command I just get silence. Also had the same problem with compiling ecasound. Any ideas? [10:02] <relaxed> ffmpeg -i input -map 0:v:0 -map 0:a output1 -map 0:v:1 -map 0:a output2 .. [10:02] <newbie2> command I am running on both version of ffmpeg: 'ffmpeg -f alsa -i default ~/ffmpeg.wav' [10:04] <Eventh> relaxed: I could do that, but then I would demux etc twice? And encode audio twice? [10:04] <relaxed> newbie2: http://trac.ffmpeg.org/wiki/Capturing%20audio%20with%20FFmpeg%20and%20ALSA [10:05] <relaxed> Eventh: yes [10:05] <Eventh> I'm trying hard to avoid that, as I actually need to create 10-15 different outputs, for rtmp/hls and more [10:06] <newbie2> relaxed: Thanks for the link, the problem I am trying to get more info on is why my compile of the latest ffmpeg isn't working with the same command as I'm running on ffmpeg 0.6.5 [10:07] <relaxed> newbie2: 0.6.5? please download the most recent release and use that. [10:07] <relaxed> which is 2.1.something [10:08] <relaxed> Eventh: tee is for copying output [10:09] <Eventh> True [10:17] <newbie2> relaxed: I am trying to use the latest version (Downloaded and compiled) the issue is that this version does not record any sound and I think it may be a possible compile issue. The old version 0.6.5 that comes with Centos 6.3 repository all works. [10:22] <newbie2> relaxed: sorry. [10:22] <newbie2> fflogger: please see: http://pastebin.com/Rix5p1jW [10:22] <relaxed> fflogger is a bot :) [10:23] <relaxed> did you read http://trac.ffmpeg.org/wiki/Capturing%20audio%20with%20FFmpeg%20and%20ALSA ? [10:24] <relaxed> as well as you read fflogger's message :/ [10:28] <newbie2> relaxed: I have read the link and still no luck. [10:29] <relaxed> have you tried anything other than "-i default" ? [10:29] <newbie2> relaxed: Yep tried hw:0 this gives me error: [alsa @ 0xb1c0d80] cannot set access type (Invalid argument) [10:29] <newbie2> hw:0: Input/output error [10:30] <newbie2> relaxed: Same for hw:0,0 [10:30] <relaxed> did you run "arecord -l" to get the name you need? [10:30] <relaxed> or are you just trying random crap? [10:31] <newbie2> relaxed: This was the aplay -l output: [10:31] <newbie2> **** List of CAPTURE Hardware Devices **** [10:31] <newbie2> card 0: DSP [Hammerfall DSP], device 0: RME Hammerfall HDSP 9632 [RME Hammerfall HDSP 9632] [10:31] <newbie2> Subdevices: 1/1 [10:31] <newbie2> Subdevice #0: subdevice #0 [10:31] <newbie2> As I said previosly I can record with ffmpeg 0.6.5 that was taken using yum from repository. But when I use the compiled latest version, I get no errors, just empty wav file :( [10:32] <newbie2> relaxed: Might just try completely rebuilding the box :( [10:33] <anshul_> hey, I am trying demuxing_decoding.c with c++ and i am getting error: expected ) before PRId64. any one has idea what to do? [10:34] <relaxed> newbie2: pastebin the command and all console output [10:41] <newbie2> relaxed: http://pastebin.com/fbJ8cQdz [10:45] <relaxed> newbie2: try, ~/ffmpeg/ffmpeg -f alsa -ac 1 -ar 44100 -i default ~/ffmpegNew.wav [10:47] <newbie2> relaxed: Thankyou it is working!!! what do -ac and -ar do? [10:48] <relaxed> it sets the audio rate and channel number [10:49] <newbie2> relaxed: Thankyou so much, I've been trying this for days. Glad I came back to IRC, hadn't been on here since my teen days :) Now I'm going to try and split stereo channels etc... time to play. Thanks again! [11:03] <newbie2> relaxed: How can I record both stereo chanels I thought it would be -ac 2 but I just get silence... :( [11:03] <relaxed> well, the old ffmpeg used one channel. That's how I knew to try it. [11:04] <newbie2> relaxed: sorry, you've lost me [11:04] <relaxed> Is this a mic? Maybe it only records in mono. [11:05] <newbie2> Could this be my alsa config? [11:48] <GT1> Hi, I'm trying to compile ffmpeg for android and before that I would need to compile x264, but I have a problem it says no wokring c compiler found http://pastebin.com/FHiULysC [11:48] <GT1> Can anyone help me with this? [11:49] <Mavrik> well it's pretty obvious what the error is [11:49] <Mavrik> arm-linux-androideabi-gcc: command not found [11:49] <GT1> yeah, but how can I find it? [11:49] <GT1> what I did so far [11:49] <GT1> I added my ndk path to path [11:50] <GT1> and I can find it, if I type arm tab tab, it will come up [11:50] <GT1> arm-linux-androideabi-gcc [11:50] <GT1> so what's wrong? [11:51] <GT1> hmm, may it be that I don't have permission to execute? would that cause the same error? what permission should I give it? [11:54] <GT1> Hm? [11:55] <GT1> I tried 667, still nothing [11:55] <GT1> same error [11:55] <GT1> and as I said, I added it to the path, so I can find it [12:08] <relaxed> GT1: maybe --cc=/path/to [12:11] <GT1> rofl, I believe I Have an interesting situation... I was trying to build a toolchain to help me out [12:11] <GT1> and it said [12:11] <GT1> host system linux-x86 is not supported by the source sdk [12:11] <GT1> good to know I have x86 linux installed... [12:12] <GT1> nvm.. [12:13] <GT1> it's not [12:13] <GT1> nvm, I believe I can figure this out,b ut I may need your help in a bit [12:17] <GT1> k, I have no idea what am I doing wrong, can somebody help me trough to compile x264 then ffmpeg? [12:24] <GT1> Anyone have the slitest clue why do I get No wokring C compiler found? [12:29] <GT1> Is there anybody here who build x264 for ffmpeg before? [12:31] <sacarasc> Do you have a C compiler installed? [12:36] <GT1> isn't the compiler in the ndk what I need? [12:37] <GT1> I'm pointing him the compiler the androidabli what so ever, no? [12:42] <GT1> if I don't use the cross-prefix I can build it [12:42] <GT1> but, I believe to use it on android, I would need to use the cross-prefix no? [12:45] <GT1> So, my problem is that I won't find the ndk cross prefix compiler, if I leave that ocmmand out it compiles, but I need it to be able to run it on android, any idea how to solve this? [12:51] <GT1> really there is no one where who could help me build x264 for ffmpeg on android? [12:54] <GT1> How can I point x264 configure to use ndk-s gcc for android? [12:54] <sacarasc> [11:08:49] <relaxed> GT1: maybe --cc=/path/to [12:54] <sacarasc> I believe that's a ./configure option. [12:59] <GT1> no cc ocnfig [12:59] <GT1> ifor cross compilation in the help it says --cross-prefix [12:59] <GT1> and I added that, but it won't find it [13:01] <GT1> rofl [13:01] <GT1> wtf [13:01] <GT1> if I typed the full name, it found it [13:01] <GT1> so, my question is, if I added it to the path [13:01] <GT1> why couldn't it find it? [13:06] <GT1> but the build of ffmpeg still fails [13:07] <GT1> nvm typo [13:08] <relaxed> GT1: http://fate.ffmpeg.org/report.cgi?time=20140317222807&slot=armv7l-panda-gcc… [13:09] <relaxed> ^^ might be helpful to look at [13:11] <relaxed> GT1: this might be better http://fate.ffmpeg.org/report.cgi?time=20140317214754&slot=armv7a-android-g… [13:27] <GT1> but guys, if I add something to PATH, should it try to find it there too? [13:27] <GT1> why did I had to add the full path to my gcc? [15:23] <anshul_> how to disable silent rule of makefile [15:23] <anshul_> in ffmpeg [16:11] <semslie> I've been playing around with converting from a webm video to frames and I've successfully used ffmpeg to create an webm video with working transparency. What I would like to do now is use ffmpeg to split that video into png frames, maintaining the transparency. This works correctly when I convert from a .mov, but when I convert from .wemb then the resulting png images don't have any transparency. I'm a bit stuck - any pointers would be ap [16:11] <semslie> preciated! [16:13] <semslie> The command I'm running is in the form: ffmpeg -i foo.webm -pix_fmt rgba -r 25 -frames 2000 -s 854x480 /bar/%04d.png [16:14] <relaxed> try -pix_fmt bgra [16:16] <Hello71> uh. [16:18] <semslie> relaxed: ffmpeg doesn't seem to like that. I get this "Incompatible pixel format 'bgra' for codec 'png', auto-selecting format 'rgba'" [16:20] <semslie> This works perfectly when the input is a png-encoded mov. Unfortunately this results in a massive file and I'm trying to cut down the size, hence the attempted move to webm. [16:20] <relaxed> what about -pix_fmt rgb32 ? [16:21] <semslie> relaxed: strangely, ffmpeg gives me the same messages about 'bgra' being incompatible for 'png', and autoselecting 'rgba' instead [16:22] <semslie> I've tried it anyway, but I can confirm that any transparent areas are black instead [16:28] <semslie> Just a random guess, but perhaps ffmpeg isn't compiled with the right png support? Any ideas on how to check? [16:30] <semslie> running "ffmpeg -encoders" produces the following line for png: [16:30] <semslie> VF.... png PNG (Portable Network Graphics) image [16:35] <sacarasc> semslie: What does ffmpeg -codecs | grep -i png give? [16:36] <semslie> sacarasc: DEV..S png PNG (Portable Network Graphics) image [16:36] <sacarasc> So, you can decode, encode, it's video and lossless. \o/ [16:37] <semslie> fwiw I tried "ffmpeg -h encoder=png" to see what pixel formats are supported and got "rgb24 rgba rgb48be rgba64be pal8 gray gray8a gray16be monob" [16:38] <semslie> I did try rgb24, but that produced the same results as rgba, so that's not it [16:48] <shubhamjain1> My question is not specifically related to FFMPEG. I am building a simple flv to mp3 convertor. I succeeded at extracting audio frames from flv but can't build proper mp3 frames. When converting with ffmpeg I observed that it does resampling even if we convert at the same bitrate. I was curious if thats a necesarry part of conversion. [17:09] <anshul_> shubhamjain1, personal opinion why are you reinventing wheel, if you want something low weight you can always keep flv decoder and mp3 encoder and use light version of ffmpeg apis. [17:10] <shubhamjain1> anshul_, Just a fun project. Thats all. Nothing fancy I wish to achieve. [17:13] <anshul_> Resample changes the raw format of data, like planar to nonplanar stuff, it depend how you have specified in your mp3 header about raw format of audio [17:26] <semslie> if anyone's interested, I found out a bit more about that webm weirdness: It looks like ffmpeg doesn't currently decode transparency from webm well, though encoding is fine: http://ffmpeg-users.933282.n4.nabble.com/Encode-VP9-or-VP8-WITH-alpha-chann… [19:09] <brontosaurusrex> semslie, vp9? [19:10] <semslie> brontosaurusrex: vp8 as far as I can tell [19:48] <RAZ0REDGE> hey i'm having some troble with libx264, i have got the lastest git source and install it.. i can even do x264 -V and it shows on my system, when i try to configure ffmpeg dev (git) source --enable-libx264 and --enable-gpl, i get libx264 not found error [19:48] <sacarasc> What does the config.log say? [19:55] <RAZ0REDGE> http://pastebin.com/BePG3Q9t [19:56] <RAZ0REDGE> sorry took awhile i went to get coffee [19:56] <RAZ0REDGE> i'm using OpenBSD os [19:56] <sacarasc> Where is your x264.h file? [19:57] <RAZ0REDGE> hmm should be in /usr/inclub [19:57] <RAZ0REDGE> include* [19:57] <sacarasc> But is it? [19:58] <RAZ0REDGE> aaa its in /usr/local/include [19:58] <RAZ0REDGE> weird [19:58] <RAZ0REDGE> i have --prefix=/usr [20:01] <RAZ0REDGE> hmm its funny how it puts in in /usr/local/include [20:03] <RAZ0REDGE> hmm [20:56] <roger21> hello [20:57] <roger21> can we put opus with vp8 and vp9 into webm already? [20:58] <roger21> s/and/or [21:03] <llogan> roger21: you can with "-strict experimental" (or the neckbeard alias "-strict -2") [21:04] <llogan> but note that it is considered "experimental" [21:05] <roger21> oh ok ... noted, thanks [21:05] <JEEB> I have no idea with ffmpeg cli still notes the -2 variable instead of the name [21:05] <llogan> i just mentioned why [21:05] <JEEB> the number in theory could easily change, but the textual version generally wouldn't as easily :P [21:06] <JEEB> no you did not note why ffmpeg shows the number instead of the textual version. you noted what it does [21:06] <JEEB> (or so) [21:06] <llogan> "neckbeard" [21:06] <JEEB> no, I'm pretty sure that's not it [21:07] <JEEB> it's just because it's from before the variable was available as text [21:07] <JEEB> and/or the one who wrote that av_log line was lazy [21:07] <llogan> IIRC, console outputs would mention "experimental" in the past instead of -2 [21:07] <JEEB> if so, the change makes absolutely no sense :P [21:08] <JEEB> it's less readable and notes a number that could in theory change [21:08] <llogan> i'm not sure when (of if in the case i'm making it up unknowingly) it was changed [21:08] <JEEB> let's first git grep for it [21:08] <JEEB> and then git blame [21:09] <JEEB> oh yes [21:10] <JEEB> I have this feeling that it was originally only available as a number [21:10] <JEEB> and then it was copypasted [21:10] <Aralucc10> hi, can anyone give any hint on this error (im trying to stream a https online stream so I recompiled with openssl / https support...but still error) [qtp736595190-506] WARNING - 02:46:45.269 - j : ffprobe error exit code=143 [21:10] <JEEB> the files noting vstrict=-2 | -strict -2 are huffyuvenc, snowenc, lavf/matroskaenc [21:11] <JEEB> although I remember seeing that message somewhere else too... [21:11] <JEEB> it's probably somewhere generic and my git grep for "\-strict \-2" isn't finding it [21:12] <Aralucc10> ok...thanks...a sec [21:12] <llogan> JEEB: i prefer the words to the numbers too [21:13] <JEEB> ah [21:13] <JEEB> I guess it's the thing in lavc/utils.c [21:13] <Aralucc10> here? http://pastebin.com/kkL1wLxU [21:14] <JEEB> http://up-cat.net/p/eb7be34a [21:14] <JEEB> and it just outputs the number of FF_COMPLIANCE_EXPERIMENTAL [21:15] <Aralucc10> I can stream the same stream with windows version [21:16] <llogan> i don't see any ffmpeg or ffprobe commands nor any complete console outputs from these cli tools [21:17] <Aralucc10> oh... I put a bigger log? [21:26] <Aralucc10> http://pastebin.com/5yTqDcb5 still protocol not found... I hoped I added https support [21:26] <Aralucc10> is there a prebuilt binary with https support to download for testing purpose? [21:26] <Aralucc10> or..the right configure string? [21:27] <Aralucc10> btw if I check the -protocols options https is there :/ [21:27] <llogan> you need to show your actual ffmpeg command and the complete ffmpeg console output [21:28] <Aralucc10> I see... I pasted the client log... you need the ffmpeg log [21:28] <Aralucc10> sorry for ignorance... does ffmpeg store log to var/log? [21:30] <llogan> no. in your console, where you type your ffmpeg command, ffmpeg will show the console output [21:30] <Aralucc10> I dont do that... I use a client soft which does that... ill try to mimic he same command... [21:31] <llogan> scripts and third party tools are not supported here [21:31] <Aralucc10> well..its a front end...it just choose a preset and apply to a url...ill try to execute the fuill command... [21:48] <Aralucc10> I think the error is this...with the command executed: http://pastebin.com/3A8NU4w8 [21:56] <caral> hi, I try to use the 'drawtext' filter and found some examples like http://pastebin.com/9reHaNwN . How do I find the meaning of e.g. "%H\:\%M\:\%S | \%a \%d/\%b/\%Y" I also tried "ffmpeg -h filter=drawtext" but there is no explanation for this kind of parameters. [22:06] <llogan> caral: see "man strftime" [22:08] <caral> ah :-) thx [23:07] <Aralucc10> im trying to recompile ffmpeg to add https support..but I receave a openssl not found error... i read around but not been able to fix... this is the log http://pastebin.com/Q6c34aSN [23:07] <Aralucc10> if I remove --enable-openssl it configure is ok [23:08] <caral> what about installing openssl? [23:08] <llogan> Aralucc10: also view the tail of config.log [23:09] <Aralucc10> I have it installed [23:09] <Aralucc10> ok [23:09] <Aralucc10> http://pastebin.com/M8SuxY3p ebough? [23:09] <Aralucc10> enough [23:10] <Aralucc10> im so newbie...and its so frustrating :) [23:11] <caral> you know how to install a package? [23:12] <Aralucc10> yes [23:12] <Aralucc10> I think it look for openssl bunary in the wrong path [23:12] <llogan> where did you install openssl/ssl.h? [23:12] <llogan> what's your distro? [23:13] <Aralucc10> ./usr/include/openssl/ssl.h [23:13] <Aralucc10> Linux XBMC-I5 3.11.0-18-generic #32-Ubuntu SMP Tue Feb 18 21:11:14 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux [23:14] <Aralucc10> this is a longer config.log if needed http://pastebin.com/sKrEFwbs [23:22] <caral> maybe you missed the dev package libssl-dev ? Don't know if you really need it, but its worth a try [23:23] <Aralucc10> ok I try [23:24] <Aralucc10> I have it [23:25] <Aralucc10> im pretty sure I tried before and that option worked... maybe I should try with less configure options and see if it works [23:26] <caral> and you can also reading this https://trac.ffmpeg.org/wiki/UbuntuCompilationGuide [23:27] <Aralucc10> sure..ok [23:29] <Aralucc10> !! I removed this --extra-libs=-static [23:29] <Aralucc10> and now openssl works [23:29] <Aralucc10> is that option needed? [23:30] <llogan> i can't answer that for you, but in your case I will guess "no". [23:30] <llogan> also you list several options more than once [23:30] <caral> that's a diplomatic answer ;-) [23:31] <Aralucc10> yes... the double are there..im sorry..its because i merges 2 config examples :) [23:31] <Aralucc10> merged [23:31] <Aralucc10> now I try to cleanup [23:31] <Aralucc10> and see if it works [23:52] <n-st> hi, i'm trying to capture a resized frame from a video using ffmpeg -i in.mp4 -ss 2 -vframes 1 -vf scale='if(gt(a,800/600),800,-1)':'if(gt(a,800/600),-1,600)' out.jpg [23:53] <n-st> ... which results in the error "No such filter: '800/600)'" [23:53] <n-st> what am i doing wrong? [23:56] <Jack64> n-st: "800/600" maybe? [23:57] <Jack64> n-st: just throwing it out there, I never tried to do what you're doing [23:59] <n-st> Jack64: doesn't work either, unfortunately... [00:00] --- Wed Mar 19 2014

1 0

ffmpeg-devel.log.20140317
by burek 18 Mar '14

18 Mar '14

[00:11] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:2f955d572b08: swscale/x86/swscale: remove unused constants [00:19] <cone-714> ffmpeg.git 03Matt Oliver 07master:590805b7c3e4: Fixed 64bit conformance with mvzbl. [00:54] <cone-714> ffmpeg.git 03Matt Oliver 07master:9eb3f11c55d6: Add missing external declarations. [02:11] <BBB_> Compnn: I don't want their review - I really don't care about spaces and stuff. I don't mind someone else reformatting my code, but not as a requirement of me committing it - or more functionally, it _should not_ slow me down [02:11] <BBB_> Compnn: other than that no objections [02:38] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:8e92ff25469f: avcodec/h264: be more tolerant on what pixel format changes trigger reinits [02:49] <Rodeo> michaelni: I'm not sure the commit hash you posted in that issue on trac is correct [02:49] <Rodeo> I believe you meant http://git.videolan.org/?p=ffmpeg.git;a=commit;h=8e92ff25469f75f5c1fcbb9ba5… [02:51] <michaelni> oops, fixed [02:52] <michaelni> i had edited the commit message and used the hash from the previous one [02:52] <michaelni> thx for pointing it out [02:52] <Rodeo> np [02:52] Action: Rodeo sleeps [03:12] <Compn> BBB_ : i think many people would like to work together. but i am not sure how the review/commit policies could combine [03:12] <Compn> but yes i prefer your method, other people can feel free to fix whitespace [03:23] <llogan> Compn: why do you ask? [03:24] <Compn> llogan : libav person asking me about it [03:24] <Compn> probably a few downstream projects and distros would rather only have one libavcodec library as well... [03:24] <llogan> i assumed they would never want anything to do with us as long as michael is involved. [03:28] <Compn> llogan : its difficult to get communications with them, so i really dont know what their opinion is [03:33] <iive> there are few libav developers who have never worked with michael [03:36] <iive> imho, mans and diego should never be let around ffmpeg. they were root admins and they abused their power. [03:38] <iive> i've heard that mans doesn't work on libav anymore, so this leaves only diego. ;) [03:38] <Compn> iive : cant you forgive people? :P [03:38] <Compn> ehe [03:39] <Compn> forget this libav stuff [03:39] <Compn> where can we get more CC608 samples [03:42] <iive> Compn: it is not about revenge... it is very important what kind of people work in the project and how they interact. [03:43] <iive> at the moment ffmpeg is mostly trouble free place, where everybody are happy to work on improving code. [03:48] <Compn> yes we've had a lot less bikesheds and fights imo [03:49] <iive> I have no reason to think that Diego have changed for good. The fact that Ronald left libav because of Diego's cosmetics is good indication why it would be best to keep him away, no matter how hardworking he might be. [03:50] <iive> (Diego, that is). [03:54] <iive> the old ffmpeg had problems. libav tried to improve the things, by fixing the most obvious issues. However the new ffmpeg is radically different and better than both old_ffmpeg and libav. This is as community dynamics, not codebase. [03:56] <iive> So in the long run, ffmpeg managed to create healthier environment and attract development, while libav started to slowly stagnate. It is even more amazing, given that the biggest distributions package libav instead of ffmpeg. [03:56] <iive> well, [03:58] <iive> all this comes to say one thing. regular libav developers are welcome to work on ffmpeg. [04:07] <Compn> probably j-b wants us to be friends again [04:07] <Compn> ehe [04:12] <iive> I can't recall j-b ever been offensive toward somebody. [04:12] <iive> imho he is a nice person. isn't he a primary vlc developer? [04:22] <iive> n8 ppl, have fun. [04:28] <cone-714> ffmpeg.git 03Matt Oliver 07master:b73aae6fe9b7: avcodec/x86/idct_sse2_xvid: move offsets out of MANGLE() [04:41] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:75af13a0e4c5: mpegvideo: directly use frames in ff_draw_horiz_band() [04:41] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:3ddf9b5b7750: Merge commit '75af13a' [04:46] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:1c79b1625d4d: h264: directly use frames in ff_h264_draw_horiz_band() [04:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:3dab9e804afe: Merge commit '1c79b1625d4d257bfd01eccb84cc0ab355fb9a9e' [05:42] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:d66e305bd1b4: er: move relevant fields from Picture to ERPicture [05:42] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:533bc4c0a30d: Merge commit 'd66e305bd1b4f3e91ae4e7e549148509d0811672' [05:42] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:8ef9dcf1d74a: avcodec/mpegvideo: ff_mpeg_set_erpic() clear destination [05:42] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:4f22e39e65e9: avcodec/error_resilience: fix the case when MVs are not available [05:58] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:9b749c8274f6: h264: move relevant fields from Picture to H264Picture [05:58] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:c237e88d51b9: Merge commit '9b749c8274f6b6f35dde2cf29b99fa4f719abf87' [05:58] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:6102dda1d754: avcodec/h264: h264_set_erpic() clear destination [06:11] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:5d1c2e53ab3c: h264: Replace mpegvideo-specific MAX_PICTURE_COUNT by private define [06:11] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:a81a2b514eec: Merge commit '5d1c2e53ab3ce27b48c138d22bb01ff8e8304f27' [06:23] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:e3c2d0f3d41f: h264: Replace mpegvideo-specific MAX_THREADS by private define [06:23] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:bb3c0571d31a: Merge commit 'e3c2d0f3d41f79f7be7ba944aaca2e287c7d5c7c' [11:40] <BBB_> Compn: it's important to listen to people like j-b, he's an important distributor plus he's a good dude [11:40] <BBB> ok time to get the kids out of bed... [12:31] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:136034d86b5c: h264: Remove MotionEstContext and move the relevant fields to H264Context [12:31] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:3e5833802ec7: Merge commit '136034d86b5cb1819a2c3e6ecdfeb05dcba7140d' [12:39] <plepere> hey BBB, did you find the time to look a bit at my code ? [12:45] <BBB> plepere: a little, not halfway done yet [12:45] <BBB> it's so big dude [12:45] <plepere> BBB : okay okay. :) [12:46] <BBB> as for the crash, I can reproduce it, I'm trying to install valgrind to look further but it seems it's incompatible with mavericks [12:46] <BBB> so it'll take me a bit [12:46] <BBB> had to update all my ports :( [12:46] <BBB> every os update is such a mess [12:46] <BBB> anyway, I'll look at the crash first (if mraulet hasn't fixed it yet by the time I have updated my tools), and then more review [12:47] <plepere> BBB : no pressure, I'm looking at the ASM idct. (I'll take a look back at my 10bit qpel_hv_unw problem another time) [12:47] <BBB> oh yeah that question [12:47] <BBB> does fate trigger the 12px function? [12:47] <plepere> BBB : mraulet is looking at the merging of all the MC and weighting IIRC. [12:47] <BBB> it might just not, or the rounding might be off [12:47] <BBB> and maybe that doesn't trigger a defect in width=12 but does in w=8/4 [12:48] <plepere> BB : if I break the function, there is an error on 12, 16, 24, 32, 48, 64 [12:48] <plepere> I can activate all of them, but not the 4 or 8. [12:53] <BBB> hm... interesting [12:53] <BBB> ok I'll look [12:54] <BBB> I'm not entirely sure [12:55] <BBB> darn it valgrind doesn't install from ports, how annoying [12:55] <plepere> thanks, you're awesome. :) [12:55] <BBB> now I have to compile it by hand [12:55] <BBB> grrr [13:09] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:e0c16e4e3259: mpegvideo: move mpegvideo formats-related defines to mpegutils.h [13:09] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:9517900bef52: Merge commit 'e0c16e4e3259cf50b5bac4c23bb6e517f397c74b' [13:09] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:377cfc28a226: avcodec/vaapi: fix pointer types after H264Picture changes [13:09] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:2a37e560dccb: avcodec/vdpau: fix ff_vdpau_get_surface_id() argument after H264Picture [13:30] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:be039278b5eb: mpegvideo: move ff_draw_horiz_band() to mpegutils.c [13:30] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:29be9b530121: avcodec/dxva2: fix pointers after H264Picture [13:30] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:329a3286d648: Merge commit 'be039278b5ebd8075d90a3508db2aed5adf59e02' [13:41] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:7245a0ae872d: mpegvideo: remove h264-only fields [13:41] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:dc226c023d2b: Merge commit '7245a0ae872d4f65396a37d13f5d1d2c2efe11c2' [13:41] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:3e3b0b3a40ee: avcodec/vdpau: update ff_vdpau_h264_set_reference_frames() to H264Picture [13:51] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:d24e9a99a401: h264 does not depend on mpegvideo any more [13:51] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:ad9a6e19e1a8: Merge commit 'd24e9a99a40166bf881ccd2e3ae5688af4726658' [14:43] <cone-714> ffmpeg.git 03Jean First 07master:1481d24c3a0a: RGBA64 pixel formats [14:43] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:6b1ca1709fe5: Merge commit '1481d24c3a0abf81e1d7a514547bd5305232be30' [14:51] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:6d7b22fb8f24: isom: fix C99-style declaration [14:51] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:85d28a865681: Merge commit '6d7b22fb8f247a51e668c83b2f2460a9b3e948a8' [15:02] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:f7518f1a0e0a: codec_desc: update dvaudio tag [15:02] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:b361a0bf1fbb: Merge commit 'f7518f1a0e0a1788c303de3c6198da07e575710c' [15:30] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:c598b569fb3d: png: K&R formatting cosmetics [15:30] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:a00d4c5956e1: Merge commit 'c598b569fb3d1f4b6c4868fe64f6989254df5186' [15:38] <cone-714> ffmpeg.git 03Carl Eugen Hoyos 07master:cccac7654f3f: png: Support rgb48 and rgba64 encoding [15:39] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:3731c409890c: Merge commit 'cccac7654f3f6b82967bc6ae08b35ae8b4fbe1a9' [15:50] <cone-714> ffmpeg.git 03Carl Eugen Hoyos 07master:58c215961a90: raw, nut: Support rgba64 encoding [15:50] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:c738ffcf2c07: Merge commit '58c215961a9067bb670387bbd72286c30de04b93' [15:52] <cone-714> ffmpeg.git 03Jean First 07master:6d78e852fbf3: img2: add j2c file extension [15:53] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:ba7cdb970f1c: Merge commit '6d78e852fbf3cab438a2085bcb64b90457c4a851' [15:56] <cone-714> ffmpeg.git 03Carl Eugen Hoyos 07master:5b1b91cfb82d: tiff: Support rgba encoding [15:56] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:1c79ca272de8: Merge commit '5b1b91cfb82d56c09846c68b09d2b8b7c8faccc4' [16:06] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:6612a03d7470: fraps: set color_range [16:06] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:ca0e6fa7317c: Merge commit '6612a03d7470af4aa6b8aa313b1eff013691d181' [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/1.2:e3ce3d373caf: avcodec/h264: be more tolerant on what pixel format changes trigger reinits [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/1.2:620d80d5720e: avcodec/adpcm: ADPCM_IMA_DK3 packets are padded to 16-bit packet boundary [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/1.2:b1a960cc8367: avcodec/adpcm: squelch 'mismatch in coded sample count' warning for AV_CODEC_ID_ADPCM_EA_R2/3 [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.1:db53b2b2c0f7: ffmpeg_opt: check that a subtitle encoder is available before auto mapping streams [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.1:9098da032975: avcodec/h264: be more tolerant on what pixel format changes trigger reinits [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/2.1:2e99fab15c23: avcodec/adpcm: ADPCM_IMA_DK3 packets are padded to 16-bit packet boundary [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/2.1:02d46127f5f5: avcodec/adpcm: squelch 'mismatch in coded sample count' warning for AV_CODEC_ID_ADPCM_EA_R2/3 [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.2:d8fe6957796a: avcodec/h264: be more tolerant on what pixel format changes trigger reinits [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.2:48609236daf8: ffmpeg_opt: check that a subtitle encoder is available before auto mapping streams [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/2.2:4aab3f868f35: avcodec/adpcm: ADPCM_IMA_DK3 packets are padded to 16-bit packet boundary [16:08] <cone-714> ffmpeg.git 03Peter Ross 07release/2.2:dce2f820e952: avcodec/adpcm: squelch 'mismatch in coded sample count' warning for AV_CODEC_ID_ADPCM_EA_R2/3 [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.2:90d6b563fe81: avcodec: Move STRIDE_ALIGN to internal.h [16:08] <cone-714> ffmpeg.git 03Michael Niedermayer 07release/2.2:7b7d8b879428: avcodec/mpegvideo_enc: dont use direct mode for unaligned input [16:14] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:1c1fbc70cae4: mjpeg: set color_range [16:14] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:020baa0be64b: Merge commit '1c1fbc70cae44c10af8ff865826e31c17bc9f347' [16:22] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:2183432e6dc8: mdec: set color_range [16:22] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:987b873049a8: Merge commit '2183432e6dc8aedf1ef3db63006a2a8195479abd' [16:27] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:bf0d7da7cbbf: ljpeg: check color_range [16:27] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:34e325efa5d0: Merge commit 'bf0d7da7cbbf869605086c2a47cdf87f0a533e24' [18:09] <Keestu> Why is that setting UPD read timeout does not impact ? [18:09] <Keestu> AVDictionary *stream_opts = 0; [18:09] <Keestu> av_dict_set(&stream_opts, "timeout", 3000000) // in microseconds. [18:10] <Keestu> avformat_open_input ( &fmtCtx, "ip", NULL, &stream_opts); [18:10] <Keestu> av_read_frame takes 20 seconds timeout always. [18:12] <Keestu> as in mentioned socket.h:#define SO_RCVTIMEO 20 [18:13] <Keestu> extremely sorry for disturbance if it is not the right channel to ask this question!! [18:46] <cone-714> ffmpeg.git 03Vittorio Giovara 07master:75177b2f5e18: libx264: check color_range [18:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:b1eb92a6bcda: Merge remote-tracking branch 'qatar/master' [18:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:f6f3c85c0eff: avformat/isom: Simplify mov_rewrite_dvd_sub_extradata() [18:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:b6ad5893ce58: avformat/isom: only write the size if its non zero in mov_rewrite_dvd_sub_extradata() [18:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:8ba432bc56f2: avformat/mov: fill in subtitle dimensions after parsing tkhd [18:46] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:e21235f9b9e2: avformat/mov: call mov_rewrite_dvd_sub_extradata() after parsing dimensions from tkhd [21:03] <cone-714> ffmpeg.git 03Andrey Utkin 07master:e10ab43614c1: doc/examples: Add example transcoding.c [21:34] <cone-714> ffmpeg.git 03James Almer 07master:aa1f38015cb0: x86/synth_filter: improve FMA version [00:00] --- Tue Mar 18 2014

1 0

ffmpeg.log.20140317
by burek 18 Mar '14

18 Mar '14

[00:59] <thangalin> Hey folks. I'm getting Xext not found when building ffmpeg. [00:59] <thangalin> http://pastebin.com/raw.php?i=axhNMhJ9 [01:00] <thangalin> x11proto-xext-dev is installed. [01:00] <thangalin> Any ideas how to fix this to get a static build? [10:10] <Aiena> HOw do I tell ffmpeg to exit cleany in a shell script ? ctrl+c seems to not be the best method as it can tend to corrupt videos if trigged at the wrong time. [10:11] <Aiena> it corrupts videos on occasion [10:15] <relaxed> Aiena: killall -INT ffmpeg [10:15] <Aiena> relaxed but will ffmpeg encode properly and exit like that [10:15] <Aiena> that will just kill ffmpeg on the spot [10:16] <Aiena> *or that will just kill ffmpeg on the spot [10:16] <Aiena> I had read online that this video corruption could be due to a memory leak but those reports are so old that they are no longer relevant [10:19] <relaxed> It will make ffmpeg exit cleanly. Test it. [10:22] <Aiena> relaxed and I need to specify it as -INT or an integer number ? [10:24] <relaxed> I gave you that exact command. [10:25] <relaxed> s/that/the/ [10:26] <Aiena> relaxed thanks [10:26] <Aiena> I did not know what -INT was [10:26] <Aiena> now I know [10:26] <Aiena> i thought it was integer but it is a type of signal [10:26] <Aiena> relaxed but now I am stuck again [10:27] <Aiena> once ffmpeg is recording it will only stop recording after I tell it to but at the moment I can tell it to stop recording only with ctrl+c [10:27] <Aiena> I do not know how I would tell the script to use your command instead [10:28] <Aiena> i wonder if there is a way to overload ctrl + c to do something else [10:30] <pron> Aiena -t [10:30] <Aiena> pron an example [10:30] <pron> t duration' [10:30] <pron> Restrict the transcoded/captured video sequence to the duration specified in seconds. hh:mm:ss[.xxx] syntax is also supported. [10:30] <pron> i dont have one [10:31] <pron> but basicly add -t 30 for 30 seconds long video [10:31] <Aiena> ok [10:32] <Aiena> pron is it possible to tell ffmpeg to encode for 10 more seconds after ctrl+c is pressed ? [10:32] <pron> idk [10:32] <pron> dont think so [10:36] <Aiena> ok [10:55] <relaxed> Aiena: you can run killall -INT ffmpeg from another terminal [10:55] <Aiena> Ah ok [12:22] <q_bit> I need to make an .exe for windows and it seems that this cross compile process is the easiest. Are there other cross compile processes I should look at? https://github.com/rdp/ffmpeg-windows-build-helpers [12:25] <JEEB> have a sane enough mingw-w64 compiler or build with MSVC 2013 [12:26] <JEEB> and how simple or not simple it is to compile depends wholly on what third party libraries you want to link to FFmpeg [12:33] <q_bit> thanks JEEB that makes sense with what i'm seeing starting with this helper script. [12:35] <JEEB> basically third party libraries are almost always encoders, so note what you are going to be doing with your binary and set things up accordingly [12:35] <JEEB> one exception is libopus which has a decoder as well as an encoder, but if you've never heard of opus or never seen a file with it, you're OK [12:40] <cody_> hi [12:41] <cody_> i want to decode and display a h264 bitstream in realtime [12:41] <cody_> how could i do that [12:41] <cody_> no buffering, no audio [12:42] <cody_> http://ffmpeg.org/doxygen/trunk/h264_8h.html [12:42] <cody_> is that the right place to look at? [13:34] <qxt> Got a so called *.ts file here that is 1080i. The interlacing looks bad. Any recommendation on how to de-interlace and not loose to much quality? [13:34] <qxt> btw dropped the mess into a *.mkv container already. [13:37] <qxt> Using Debian GNU/Linux stable. [13:37] <qxt> Only main repos [13:41] <iive> -vf yadif should do reasonable job at good speed. [14:30] <jarainf> qxt, -vf "fieldmatch, decimate" does the job [14:31] <jarainf> It will deinterlace the transport stream and decimate it to 23,976 fps [14:31] <jarainf> If you still have interlaced frames after fieldmatch you can add yadif between fieldmatch and decimate [14:34] <jarainf> horizontally scrolling text might still look bad, but as far as I know, there is no filter to fix this in ffmpeg yet (don't quote me on that) [15:21] <pron> is there a way to tell ffmpeg not to exit when input isnt available [15:22] <pron> like when input is http:// [15:28] <klaxa> what else should ffmpeg do? [15:29] <pron> idk , cook me a dinner ~.~ [16:53] <namccarty> Alright, so yesterday I was experiencing an issue where ffmpeg was outputting slightly corrupt mp3s when down sampleing correct files and collecting the output over a pipe [16:53] <namccarty> wound up having to leave before we got anywhere [16:54] <namccarty> i am able to reproduce the problem on both linux and windows across multiple versions of ffmpeg [16:56] <namccarty> and the files are only corrupt when the output is to the standard out [16:56] <namccarty> where do I go from here? [16:56] <namccarty> also ffmpeg is consistent on which files trigger the issue [17:22] <canci> is there a way to start ffplay in fullscreen mode? [17:24] <jarainf> namccarty, uhm... Do you need the pipe? [17:25] <namccarty> jarainf for what i am using ffmpeg for, yes [17:26] <jarainf> I've just tested what you did, and indeed the piped file seems to be a bit broken, but that's nothing which would stop it from playing back (atleast in mplayer/mpv that is) [17:26] <namccarty> yeah, its only specific files that it breaks on for me, and its slight enough that vlc can play it [17:26] <namccarty> but the media player on my phone cant [17:26] <jarainf> namccarty, well, I don't really get what you need the pipe is for... [17:26] <jarainf> -us [17:27] <jarainf> -is [17:27] <namccarty> im using a piece of software called madsonic to automically downsample and stream to my phone [17:27] <namccarty> it requries that its transcoders output to standard out [17:30] <namccarty> it works perfectly for maybe 99% of the songs in my library [17:30] <namccarty> but there are a few albums that can reliably trigger this problem [17:32] <jarainf> Soooo... [17:33] <jarainf> You are not doing a reencode at all? [17:34] <namccarty> normally i stream to my phone at a set 128kbps to save data, and pretty much all of my music is encoded above that, so usually i end up downsampleing [17:37] <namccarty> manual testing has confirmed that it is only specific files that this issue shows up for when reencoding [17:39] <jarainf> isn't madsonic supposed to do the downsampling? [17:39] <namccarty> it passes the downsampling off to the command you have set to do so [17:39] <namccarty> which is by default, ffmpeg [17:39] <namccarty> and collects the output from standard out and streams it off [17:42] <namccarty> This is what transcoding settings look like [17:42] <namccarty> http://puu.sh/7yVM0.png [17:43] <jarainf> you don't happen to have an example file which is cc? [17:44] <namccarty> not that i am aware of [17:45] <jarainf> Is that a default setting? [17:45] <namccarty> mostly default, yes [17:45] <namccarty> the defualt for downsampleing had the pipe:1 replaced with - [17:47] <namccarty> this is the subsonic page on reccomended settings, subsonic being the software madsonic was forked off of [17:47] <namccarty> http://www.subsonic.org/pages/transcoding.jsp [19:25] <namccarty> so is this anything we can think of a sloution for or devise a hack to get around, or is it time for me to go file a bug report [19:56] <sisco> Hello all, is there any way to do timeshifting with ffmpeg? like 6 hours [19:57] <llogan> canci: -fs [19:59] <llogan> namccarty: you should provide your command and the complete ffmpeg console output and any required samples so we can attempt to duplicate the issue [20:00] <sisco> well no idea how to do timeshift thats why im asking here how to do so [20:00] <llogan> sisco: what do you mean by "timeshift"? [20:00] <sisco> like iptv, stream live and stream timeshift back 6 hours [20:01] <namccarty> llogan alright, let me go fetch those [20:01] <namccarty> made those yesterday, just need to go find where i put them [20:03] <sisco> or to be more sample ffmpeg will record an rtmp stream but keeps always only the last 10 mins [20:05] <llogan> sisco: you can try the segment muxer http://ffmpeg.org/ffmpeg-formats.html#segment_002c-stream_005fsegment_002c-… [20:06] <llogan> -segment_time and -segment_wrap being options you will want to investigate [20:08] <sisco> yes i saw it before , its gonna create segment file with specific time or size [20:08] <sisco> but im gonna be stuck to restream them to create one rtmp [20:11] <namccarty> llogan here is the console output [20:11] <namccarty> http://pastebin.com/Zq6Mz9U6 [20:12] <namccarty> the only samples i have that trigger the issue are copyrighted [20:13] <namccarty> not sure how to procede, I honstly don't care about the copyright issues here since no profit is being made, but i don't know how you feel or what the procedures are for that [20:14] <llogan> namccarty: if the input is needed to duplicate the issue then please provide it [20:16] <namccarty> llogan: here is a sample that triggers the output accompanied with the outputs i get when piping and not piping [20:16] <namccarty> https://dl.dropboxusercontent.com/u/42936303/samples.tar [20:17] <llogan> namccarty: thanks. which players does the output not play correctly? [20:17] <namccarty> the apollo and ultrasonic media players on my phone, it plays correctly on vlc on my desktop [20:19] <namccarty> i've used a couple of tools that claim to check the validty of mp3 files, and they all claim that the pipe version has some sort of mpeg stream error [20:27] <hi117> does x265 have a irc channel? [20:28] <hi117> my package manager is warning that the libs have writable and executable sections [20:28] <llogan> namccarty: WMP 12 doesnt like your piped.mp3 either [20:30] <namccarty> llogan: any idea what is casuing me to get a bad output when its via standard out? [20:33] <llogan> not yet. but i did encounter a seg fault that i neglected to investigate last week [20:34] <namccarty> this issue has been plauging me for several months and I have only just now gotten around to realizing that the problem was with ffmpeg and not the program consuming its output [20:34] <llogan> if you exclude the video stream (album art) it may work. [20:34] <namccarty> how would i go about doing that? [20:34] <llogan> -vn or -map 0:a [20:35] <namccarty> let me give it a shot [20:36] <BlackBishop> any ideas on how I could get the subtitle from a file then make it hardcoded in the output ? [20:36] <pyBlob> I've managed to capture audio from my microphone on windows using ffmpeg and dshow, the only problem is that the audio is delayed by ~1s, but I want something about 0-50ms [20:36] <klaxa> BlackBishop: http://trac.ffmpeg.org/wiki/How%20to%20burn%20subtitles%20into%20the%20video [20:37] <BlackBishop> the subtitle is a stream in the initial input file ( mkv container ) [20:37] <BlackBishop> :/ [20:37] <klaxa> just specify the .mkv in -vf subtitle=somefile.mkv [20:38] <klaxa> .mkv can be handled as a file containing subtitles [20:38] <klaxa> because it is [20:38] <klaxa> font attachments could become an issue [20:38] <namccarty> llogan: looks like it did it windows media player is accepting the piped file now and it looks like i can now stream to my phone without anything breaking [20:38] <namccarty> thank you [20:41] <llogan> namccarty: i guess that narrows it down. you should report it as a bug. make sure to include the command, complete console output, and the input file. you can omit "-loglevel verbose". ...but first test with a more recent build if you can [20:42] <llogan> also, why re-encode instead of stream copy? [20:42] <klaxa> BlackBishop: here is a shellscript that extracts fonts from a matorska file, it is pretty bad, but it works: https://gist.github.com/klaxa/5651164 [20:42] <namccarty> llogan: the reencoding is to save bandwidth streaming to my phone, its done by the madsonic media server using ffmpeg [20:42] <namccarty> ive had the same issue with the git version i pulled down yesterday testing [20:43] <namccarty> the logs i just put together were using the version that ships with the windows version of madsonic because i forgot where i put the git verion [20:44] <namccarty> i guess im off to go report a bug then [20:44] <llogan> you should use the new version for your report. also test to see if stream copying instead of re-encoding can reproduce the issue because then you can possibly omit libavcodec as a possible culprit [20:44] <BlackBishop> klaxa: thanks .. but if there's a special font, would there be any problems if I'm not using them ? [20:44] <klaxa> it would render the subtitles incorrectly [20:44] <klaxa> gotta go afk, back whenever [20:45] <BlackBishop> [AVFilterGraph @ 0x20f6c70] No such filter: 'subtitles'' [20:46] <BlackBishop> :-( [20:46] <BlackBishop> altough I have it compiled with libass [20:53] <Kasper^> can you specify keyframes when you encode to vp9 ? [21:18] <pzich> I'm using `ffmpeg -i <in> -vcodec h264 -preset slow -pix_fmt yuv420p -profile:v main -level 3.1 -movflags +faststart <out>` to encode some videos. Does faststart have any quality implications for the file, or does it just help getting the video to start playing faster? [21:26] <BlackBishop> damn, ffmpeg is finishing faster than the client can view via the rtmpstream :| [21:27] <sacarasc> There's a setting to make ffmpeg encode at real time. [21:33] <BlackBishop> re ? [21:33] <BlackBishop> -re * [21:33] <BlackBishop> I think I just discovered it :) [21:33] <sacarasc> Yeah. [22:43] <RAZ0REDGE> hey i have install libmp3lame 3.9.5 or something [22:43] <RAZ0REDGE> but ffmpeg git source is asking for something different [22:43] <RAZ0REDGE> ERROR: libmp3lame >= 3.98.3 not found [22:43] <RAZ0REDGE> why do i need a older lame for this ffmpeg [22:43] <llogan> pzich: no quality implications [22:44] <llogan> although note that -level does not change refs [22:45] <jarainf> RAZ0REDGE, '>=' means greater than or equal to 3.98.3 [22:45] <llogan> RAZ0REDGE: it says greater than or equal to 3.98.3 [22:45] <jarainf> heh, llogan [22:45] <llogan> stereo [22:50] <RAZ0REDGE> LAME 64bits version 3.99.5 (http://lame.sf.net) [22:50] <sacarasc> Did you also install the dev files? [22:51] <RAZ0REDGE> no [22:52] <RAZ0REDGE> hmm dev files ? [22:52] <jangle> greetings all, attempting to use avcodec_video_decode2. Its my understanding that the AVPacket struct needs to be loaded with complete frames each call [22:52] <sacarasc> I am assuming that you just installed Lame through your package manager. Most of them split the binaries and stuff from what is needed to compile using them. [22:53] <jangle> so I maintain a buffer that has an h264 bitstream consisting of the sps, pps, idr nals [22:53] <RAZ0REDGE> yeah openbsd installs these lame and ffmpeg that i dont want to use [22:53] <RAZ0REDGE> perhaps i will delete them [22:53] <sacarasc> You need the dev files. [22:53] <sacarasc> Of the current version only. [22:54] <jangle> when i call avcodec_decode_video on that buffer, I get a successful decode, but when I append a non-idr, type 1, nal, and call decode, I get Missing reference picture, default is 0 [22:54] <RAZ0REDGE> where are the dev files ? [22:55] <jangle> as I continue to append p frames, I eventually get a successful decode call, but it is my understanding that the addition of the type 1 nal should still constitute a complete frame [22:55] <sacarasc> If you're going to uninstall things, use your package manager whenever possible, RAZ0REDGE. [22:56] <RAZ0REDGE> ok, as i know that, but what do you mean by developer files [22:56] <RAZ0REDGE> is this on the ffmpeg website ? [22:56] <llogan> RAZ0REDGE: "dev" files being debian/RHEL talk for whatever package that installs lame.h [22:57] <llogan> or you can compile lame yourself [22:57] <RAZ0REDGE> i have install lame [22:57] <RAZ0REDGE> by source yes [22:57] <RAZ0REDGE> oh i'm not using linux [22:57] <RAZ0REDGE> as i mention that i'm using openbsd [22:57] <sacarasc> You still need them on BSD. [22:58] <llogan> so the question then becomes why ffmpeg does not see your lame [22:58] <RAZ0REDGE> not sure really [22:58] <RAZ0REDGE> i type in lame [22:58] <sacarasc> RAZ0REDGE: If you do `which lame` what does it output? [22:58] <RAZ0REDGE> and it seems to use the 3.99.5 [22:59] <RAZ0REDGE> perhaps i have to link it [23:00] <pzich> llogan: thanks, any idea how I can up the quality of the video in general? from what I've read online the h264 encoder is pretty automatic in quality settings, but I'm trying to go with "very best quality" and am still getting more artifacting than I'd like [23:04] <RAZ0REDGE> aaa [23:04] <RAZ0REDGE> this work [23:04] <RAZ0REDGE> i have done lame as ./configure --prefix=/usr [23:04] <RAZ0REDGE> i suppose its linking in /usr instead of /usr/local [23:05] <jangle> so in summary, i'm presenting buffers that look like this [7 8 5] [7 8 5 1] [7 8 5 1 1] [7 8 5 1 1 1] and so on, is this correct? The library seems to tell me, no. so, what is correct? [23:07] <RAZ0REDGE> now re-building ffmpeg with lame support [23:07] <RAZ0REDGE> so far so good :) [23:13] <llogan> pzich: right now you're using the default setting for -crf (since you did not include this option), which is -crf 23. you can use a lower value. [23:14] <llogan> see https://trac.ffmpeg.org/wiki/x264EncodingGuide [23:16] <pzich> llogan: thanks, I'll take a look [23:19] <llogan> jangle: you might get more help for library usage at libav-user mailing list [23:19] <jangle> 1logan: thanks [23:21] <RAZ0REDGE> hmm odd [23:22] <RAZ0REDGE> thanks for everything [23:22] <RAZ0REDGE> cheers [23:31] <pzich> llogan: this is exactly what I needed, thanks, much appreciated [23:35] <pzich> since you've known all the answers thus far: do you know (or have an article that states) the compatibility of yuv444? the settings I got were optimized for iPad compatability a few gens back, so is using 3.1 and yuv420 [23:36] <pzich> looks like the last answer in the FAQ is saying to go yuv420 for quicktime [23:37] <llogan> pzich: yes, use yuv420 for non-FFmpeg based players [23:37] <pzich> thanks [23:38] <llogan> which is the oldest iPad you're supporting? [23:38] <pzich> it's actually possible that this won't need to support iPad or any iOS device, but I'm precautinarily supporting back to iPad 2/iOS 6 [23:39] <pzich> I recall there being a table listing iOS devices and the h264 level settings, so I'll take a look at that again [23:40] <pzich> oh yes, further down on that very page [23:41] <llogan> i should update that by mentioning refs [23:56] <pzich> is there a way I can specify frame numbers using -ss instead of HH:MM:SS.xxx? or at a higher level, is there a good way to export a single still of a particular frame from a video? [23:58] <pzich> looking like the 'select' video filter might be able to do it [00:00] --- Tue Mar 18 2014

1 0

ffmpeg-devel.log.20140316
by burek 17 Mar '14

17 Mar '14

[01:39] <wm4> michaelni: is FFmpeg 2.2 the release that's supposed to be equivalent to Libav 10? [01:39] <wm4> aka the next release [01:39] <michaelni> yes [01:40] <wm4> thanks [01:48] <wm4> there are some strange things [01:49] <wm4> ffmpeg 2.2 will have libavresample 1.2.0, while libav 10 has 1.1.0 (???) [01:50] <wm4> while ffmpeg 2.2 has libavutil 52.66.100, and libav 10 has 53.3.0 (??????) [01:53] <wm4> version table retrieved by hand (so there can be errors): http://privatepaste.com/84c5030dc0 [02:04] <michaelni> wm4, about avresample, i guess we branched a moment later and that included that minor bump, about libavutil, libav bumped its major version without real need, we didnt, we can bump but we must then bump ALL other libs major versions then. This was discussed already twice IIRC [02:04] <michaelni> if you have a magic solution that avoids this, iam all ears [02:04] <michaelni> but one cant be compatible to something that isnt compatible to itself [02:05] <wm4> well, it's probably not a problem, it's just strange [02:05] <michaelni> yes, iam certainly not happy about it either [02:05] <michaelni> avresample could be fixed, i can revert that one commit if people prefer it [02:06] <michaelni> but not sure what advantage that would have ? [02:07] <wm4> not much [03:11] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:b920c1d5ad5c: configure: Support older version of openjpeg1 [03:11] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:00ecce5c8bc0: http: Return meaningful error codes [03:11] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:738d68de8571: http: Drop doxy comments [03:11] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:eac281b06c45: Merge commit 'b920c1d' into release/2.2 [03:11] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:7430f3064fef: Merge commit '00ecce5' into release/2.2 [03:11] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:8796c3b7d3e2: Merge commit '738d68d' into release/2.2 [03:38] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:f1de93dec302: http: K&R formatting cosmetics [03:38] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:25d14b716aef: http: Refactor process_line [03:38] <cone-678> ffmpeg.git 03Luca Barbato 07release/2.2:e2811c2ede42: http: Add support reading ICY metadata [03:38] <cone-678> ffmpeg.git 03wm4 07release/2.2:facd3dbc6e07: http: handle ICY in presence of chunked transfer encoding [03:38] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:6776c2c04fe6: Merge commit 'f1de93dec302cc860c718e89656f184510b2d239' into release/2.2 [03:38] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:a7338ae8acf3: Merge commit '25d14b716aef43bf9e82f1d67a364e5922332f47' into release/2.2 [03:38] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:bdb219435e0b: Merge commit 'e2811c2ede428a20aec3630bef4378f2927fb306' into release/2.2 [03:56] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:40de74d0eb8c: http: Export Content-Type information [03:56] <cone-678> ffmpeg.git 03Clément BSsch 07release/2.2:09dca5106698: http: Support setting custom User-Agent [03:56] <cone-678> ffmpeg.git 03Anssi Hannula 07release/2.2:991e6fa35b4a: http: Add support for selecting a request range [03:56] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:bcc6429c01a8: Merge commit '40de74d0eb8c1898e2184a1484fe246aed5b295d' into release/2.2 [03:56] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:6a10263f163a: Merge commit '09dca5106698871920328ae8041d1437ce3cff43' into release/2.2 [03:56] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:bcc25353cf25: Merge commit '991e6fa35b4ad2df389ba7e1c2c58d064e295c76' into release/2.2 [04:12] <cone-678> ffmpeg.git 03Anssi Hannula 07release/2.2:f859fed03d5d: http: Always allow no-op seek [04:12] <cone-678> ffmpeg.git 03Reinhard Tartler 07release/2.2:82a3e469c658: http: Improve options descriptions [04:12] <cone-678> ffmpeg.git 03Clément BSsch 07release/2.2:cd874cf8e69f: http: Allow setting a Content-Type for POST requests [04:12] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:e727cbf0be20: Merge commit 'f859fed03d5d039430baea03a0ef35cfd6ac4abd' into release/2.2 [04:12] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:251331491299: Merge commit '82a3e469c6584e6d9e56a3fc38eb001d0650defa' into release/2.2 [04:12] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:9ae2aaea501d: Merge commit 'cd874cf8e69f1b31986fd978577994b45efa3d5c' into release/2.2 [04:24] <cone-678> ffmpeg.git 03Martin Storsjö 07release/2.2:10379d50be18: http: Declare more parameters as const where possible [04:24] <cone-678> ffmpeg.git 03Alessandro Ghedini 07release/2.2:b5210f4eae30: http: Properly initialize icy headers string [04:24] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:0e5d9fe2a7f4: Merge commit '10379d50be18325a07ef297bd2120d85a58ec78b' into release/2.2 [04:24] <cone-678> ffmpeg.git 03Michael Niedermayer 07release/2.2:f4489c9558c9: Merge commit 'b5210f4eae30df587815b5f005f219aa723591d3' into release/2.2 [06:37] <cone-678> ffmpeg.git 03Michael Niedermayer 07master:4d17179b370b: ffmpeg: fix printing INT64_MIN pts at the end [14:05] <cone-139> ffmpeg.git 03wm4 07master:8f629a986c46: isom: convert mp4 dvdsub extradata to vobsub format [14:05] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:09680951dfdf: Merge commit '8f629a986c46f227abda1811a6ad1f449871ca35' [14:15] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:cd0982e1899d: avformat/isom: fix mix of declaration and statement [14:31] <cone-139> ffmpeg.git 03Anton Khirnov 07master:e84361269500: af_channelmap: fix ONE_STR mapping mode [14:31] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:c198af648a2a: Merge commit 'e843612695007cc623813073754c651ab43021f7' [14:37] <cone-139> ffmpeg.git 03Anton Khirnov 07master:a84c8af042f3: FATE: add a test for the ONE_STR mapping mode of the channelmap filter [14:37] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:267c5723e08e: Merge remote-tracking branch 'qatar/master' [15:23] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:25bcf24d4d0f: ffmpeg_opt: check that a subtitle encoder is available before auto mapping streams [15:55] <burek> there might be a bug in the ffmpeg's configure script for the libopus library.. even though I've set the --prefix, ./configure runs pkg-config without PKG_CONFIG_PATH=$PREFIX or something like that [15:56] <burek> im not sure if i should report this as a bug, because im not sure if im not doing something wrong here [15:56] <burek> anyway, ffmpeg doesn't see libopus installed in a custom dir, specified with --prefix param [15:58] <nevcairiel> isnt --prefix meant for the output prefix, not where it looks for libraries [16:01] <burek> you're trying to say that --prefix doesn't mean that ffmpeg should look in there for its dependencies? hmh.. i'm pretty sure that all the other libraries work except those that are being checked with pkg-config.. [16:02] <burek> that's why i assumed --prefix also isolates an area where the relevant stuff should be searched for and put into [16:07] <burek> thanks nevcairiel, it works now [17:16] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:a17ee4117dcb: avformat/movenc: only ommit encoder tag of the metadata for bitexact mode [17:16] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:10c14264db73: avformat/movenc: omit LIBAVCODEC_IDENT for bitexact mode from uuidusmt_tag [17:16] <cone-139> ffmpeg.git 03Michael Niedermayer 07master:9b1d264e8744: avformat/movenc: omit avformat ident from isml_manifest for bitexact mode [17:42] <cone-139> ffmpeg.git 03Jasper Taylor 07master:0d5ae023b2b7: avfilter/vf_tinterlace: Fix vf_tinterlace mode 6 (interlacex2) [19:12] <cone-139> ffmpeg.git 03Peter Ross 07master:97bb0076c5bb: avcodec/rawdec: decode 16-bit aligned and packed 'raw' pixel formats where bits_per_coded_sample < 16 [19:12] <cone-139> ffmpeg.git 03Peter Ross 07master:40f88796c69a: Phantom Cine demuxer (iteration 2014.3) [22:25] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:b74792a02721: Changelog: move post 2.2 branching entries out of the 2.2 changelog [22:38] <cone-714> ffmpeg.git 03Martin Storsjö 07master:0412cb67cda0: aacdec: Don't count probed ADTS frames if there are false positives [22:38] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:6dd007ad99f9: Merge commit '0412cb67cda05b08dfca6bfc4ff664ea917fa932' [22:48] <cone-714> ffmpeg.git 03Martin Storsjö 07master:3b1c9eb0ffb6: aacdec: Lower the number of frames required to detect ADTS [22:48] <cone-714> ffmpeg.git 03Michael Niedermayer 07master:0efacc7f9a36: Merge remote-tracking branch 'qatar/master' [23:18] <Compn> just an informal poll, how many developers would want to work with libav team ? just say yes or no. thank you. [23:18] <Compn> if we could find some common place to work together [00:00] --- Mon Mar 17 2014

1 0

ffmpeg.log.20140316
by burek 17 Mar '14

17 Mar '14

[01:57] <everdred> Hello all. [01:57] <everdred> Does anyone have experience compiling ffmpeg on Slackware? [01:57] <iive> i'm on slackware [01:59] <everdred> Good to hear! I'm having trouble using the Slackbuild [01:59] <everdred> I install the packages it asks for one by one as it asks for them... [01:59] <everdred> but then it asks for X11. [02:00] <everdred> I'm doing this on a server without X, but I'll install whatever I need to to get ffmpeg built. [02:00] <everdred> Except I can't tell which actual package it needs. [02:01] <relaxed> X is only required if you want ffmpeg's x11grab [02:01] <everdred> Ah. I probably don't need that. [02:04] <everdred> I don't see that among the "optional features" listed on the Slackbuild page for ffmpeg: http://slackbuilds.org/repository/14.1/multimedia/ffmpeg/ [02:05] <everdred> Does x11grab support fall under one of the options they have listed there? [02:07] <everdred> Oh wait, I see there's a line in the Slackbuild script for x11grab, even though it isn't listed on the page. [02:28] <everdred> Removing that line appears to have worked; it's actually building now. [02:28] <everdred> Thanks, relaxed! [03:36] <t4nk622> hola buena noches [03:37] <t4nk622> soy usuario de Blag 140000 he instalado Gnash y me pide q instalee los codecs Gstreamer-ffmpeg alguien me puede dar un empujon por ahi [03:38] Last message repeated 1 time(s). [03:38] <t4nk622> alguien quien me pueda ayudar [03:40] <t4nk622> soy usuario de Blag 140000 he instalado Gnash y me pide q instalee los codecs Gstreamer-ffmpeg alguien me puede dar un empujon por ahi [03:41] <mmint> gstreamer-ffmpeg is not part of ffmpeg. It's just a plugin that interfaces the ffmpeg library with gstreamer. http://gstreamer.freedesktop.org/modules/gst-ffmpeg.html [03:45] <t4nk622> por favor los comandos para instalar y desempaquetar con blag140000, las disculpas del caso soy algo novato en esto [03:48] <mmint> I don't know. Try asking in #blag on irc.indymedia.org [03:51] <t4nk622> mil gracias <mmint> tratare de resolver por el otro medio [03:52] <mmint> buena suerte [14:08] <maister> Does the RTP reader in libavformat reorder packets? If so, is there a way to increase the buffer size? [14:43] <roger21> hello, i get "Error getting first pass stats" while trying to do a 2 pass theora encode [14:44] <roger21> here is the command : ffmpeg.exe -i "video.mkv" -map_metadata -1 -c:v libtheora -s 256x144 -aspect 16:9 -r:v 30000/1001 -q:v 10 -pass 1 -passlogfile lolilol -an -sn -f rawvideo -t 9 -y NUL & [14:44] <roger21> -& [14:46] <roger21> i also have that : "Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height" [14:48] <roger21> oh it seems to be the -q:v 10 [14:48] <klaxa> i'd be surprised at that [14:48] <klaxa> i don't think you can specify -f rawvideo when using -c:v libtheora [14:49] <klaxa> also given the resolution, are you really aiming for high quality? i would think a two-pass encode is unnecessary [14:49] <klaxa> unless you have to match a certain filesize [14:51] <feld> does anyone know how to get ffmpeg to encode alac files that are "optimized for streaming" ? [14:52] <feld> there are a few tools out there that can do this. dbPowerAmp describes it as this: [14:52] <feld> "m4a Layout can be 'Optimized for Streaming' (where the audio details and tags are placed before audio data)" [14:52] <klaxa> is a lossless codec ever optimized for streaming? [14:52] <klaxa> ah [14:52] <klaxa> that's more of a container thing is it not? [14:53] <feld> yeah, sort of, but I was hoping I could fix this as I convert things from FLAC to ALAC with ffmpeg [14:54] <feld> the resulting files that ffmpeg produces do not work well with streaming to iOS devices, etc and any tool that doesn't require me to use a Mac or Windows machine to touch all these files would be ideal [14:54] <feld> I'd like to be able to do it on the server itself [14:54] <klaxa> i don't know a lot about alac and streaming lossless audio, sorry [14:54] <klaxa> i usually encode to a lossy codec to stream [14:54] <klaxa> i think it also depends on the streaming software you are using [14:55] <feld> wait, i may have found this finally [14:55] <feld> i should have looked for AAC first -- -movflags +faststart [14:56] <feld> i think if I add that to my ffmpeg flags it will produce what I want. I'll give it a shot. [14:56] <klaxa> sounds good [14:57] <roger21> no of corse, it's test, what should i put to -f then ? [14:57] <klaxa> ogg maybe? [14:58] <klaxa> matroska should work too [14:58] <roger21> no but it's the first pass [14:58] <roger21> i put ogg at the second one [14:59] <klaxa> you still need to specify a valid container i think [15:12] <roger21> if i use -b instead of -q it seems ok [15:12] <roger21> why no -q in two pass [15:45] <Riso> hello [15:45] <Riso> I am trying to "burn" in srt subtitles to a .mkv; I used "-i in.mkv -vf subtitles=in.srt out.mkv" [15:46] <Riso> it created a "out.mkv" which is 1/2 of "in.mkv" and no subtitles in the video [15:46] <Riso> can someone advise what is the correct way to achieve this? [15:52] <Riso> fflogger: http://pastie.org/8935716 [15:53] <Riso> or klaxa ;) [15:55] <klaxa> there are fontconfig errors [15:55] <klaxa> i haven't seen them before [15:55] <klaxa> i don't know if that makes a difference [15:55] <Riso> all i downloaded was the ffmpeg binary - i am on MACOS [15:55] <Riso> i followed http://trac.ffmpeg.org/wiki/How%20to%20burn%20subtitles%20into%20the%20video [15:56] <klaxa> as to why the file is so much smaller, your encoding settings most likely don't match the input's quality [15:56] <Riso> i want the video to be untouched [15:56] <Riso> what should i use so it is the same? [15:57] <klaxa> that's impossible [15:57] <klaxa> at least with lossy compression, and you don't want lossless compression for production because of the resulting size [15:58] <Riso> what does this mean? [15:58] <Riso> Input stream #0:1 frame changed from rate:48000 fmt:fltp ch:2 chl:stereo to rate:48000 fmt:fltp ch:6 chl:5.1(side) Input stream #0:1 frame changed from rate:48000 fmt:fltp ch:6 chl:5.1(side) to rate:48000 fmt:fltp ch:2 chl:stereo [15:58] <klaxa> anyway, back to the issue at hand, could you try using ass instead of srt? the encoding guide describes how to convert srt subs to ass subs [15:58] <Riso> yes :) with ass its encoding [15:59] <klaxa> is it rendering the subtitles correctly? [16:00] <Riso> in vlc, mplayer always played them perfectly [16:01] <klaxa> are you sure the subtitles are not rendered by the player, but actually burned into the video? [16:01] <dvnl> Hi Everyone! Please, help me with this issue: how can I cross-compile ffmpeg to an ARM bare-metal (so no linux). I tried the following: ./configure --disable-everything (#--selecting only the needed decoders) --disable-asm --disable-yasm --enable-cross-compile --arch=arm --target-os=none --disable-optimizations. I'm trying to do this from Windows, MinGW. What other options should I add? [16:02] <Riso> here is with "ass" http://pastie.org/8935733 [16:03] <klaxa> well is it burning in the subtitles correctly? [16:03] <Riso> when i Play the original .mkv it fetches the original corresponding .srt [16:03] <klaxa> can you disable that? [16:03] <Riso> when i played out.mkv there were no subtitles in the video [16:03] <klaxa> that is not good... [16:04] <Riso> the buring with "ass" it says [Parsed_ass_0 @ 0x101b06260] Added subtitle file: 'in.ass' (2 styles, 770 events) [16:05] <Riso> it did not say that when i used srt [16:05] <klaxa> you mean the styles and event counts? that's .ass specific [16:05] <klaxa> although the event counts should be the same [16:05] <Riso> ah hmm [16:06] <Riso> if i play out.mk which is being written to, the video plays, no subtitles in the video :( [16:06] <klaxa> :/ [16:07] <klaxa> without having the actual material, i'm out of ideas, sorry [16:23] <c_14> Riso: about the output file being smaller than the input file, that's because you're reincoding the source video and audio. In order to burn the subtitles into the video you need to reencode the video, but you might want to try messing with the quality settings, for the audio you can use -c:a copy. [16:23] <c_14> And about the subtitles not burning, I found this bug report: https://trac.ffmpeg.org/ticket/2100 [16:24] <c_14> Apparently the issue is that fontconfig can't find a config file and therefore doesn't have font paths. [16:24] <c_14> Not sure if this is your issue, but it might help. [16:28] <Riso> c_14: thanks for the tip, how do i get the fonts to be fixed? [16:29] <c_14> You could try what the person with the trac ticket did. Second to last comment on the link I posted. [16:29] <Riso> c_14: how do i leave audio untouched [16:29] <Riso> ? [16:29] <c_14> Just add -c:a copy to the commandline. [16:37] <Riso> ok.. trying again. no warning with fonts seen this time.. [16:38] <Riso> Stream mapping: Stream #0:0 -> #0:0 (h264 -> libx264) Stream #0:1 -> #0:1 (vorbis -> libvorbis) Press [q] to stop, [?] for help [Parsed_subtitles_0 @ 0x101b03b00] Neither PlayResX nor PlayResY defined. Assuming 384x288 frame= 717 fps= 54 q=28.0 size= 3440kB time=00:00:30.13 bitrate= 935.2kbits/s [16:38] <Riso> works!!! [16:39] <Riso> c_14: thanks a lot!! [16:39] <c_14> No problem. [17:06] <namccarty> Alright, I am having a rather confusing issue, running the command ".\ffmpeg.exe -i .\song.mp3 -ab 96k -v 0 -f mp3 - > test-a.mp3" results in a 6MBish unplayable file, but ".\ffmpeg.exe -i song.mp3 -ab 96k -v 0 -f mp3 test-b.mp3" results in a 3MBish playable file [17:06] <namccarty> what am I doing wrong here? [17:06] <namccarty> I should also note that this issue is only affecting specific files [17:11] <namccarty> hold on, i lied a little, apparently output to stdout isnt working at all for me now [17:19] <namccarty> here is the output with loglevel set to verbose with the working version: http://pastebin.com/H2Ks5eRH , and the non-working version: http://pastebin.com/u0TnfmTX [17:20] <namccarty> any help would be appreciated, as this problem has been plaguing me for a while [17:23] <klaxa> namccarty: use pipe:1 instead of - for output [17:24] <namccarty> mp3 is still unplayable and twice as big as the playable one [17:25] <namccarty> command and output here: [17:25] <namccarty> http://pastebin.com/dbpfaWhH [17:26] <namccarty> also the file is a few KB larger than the one that i got when using - for the output [17:27] <klaxa> that is weird [17:27] <klaxa> also according to the output ffmpeg provides all files should be the same size [17:27] <namccarty> yeah, that is not the result I am getting [17:28] <klaxa> maybe windows handles pipes weirdly? [17:28] <namccarty> the result is persisting and identical on another build i just swaped out [17:28] <klaxa> what i can imagine is that some text-output is mixed with the audio output [17:29] <klaxa> can you upload the resulting file? [17:29] <namccarty> yes [17:29] <klaxa> the larger one [17:29] <namccarty> one moment [17:30] <namccarty> https://dl.dropboxusercontent.com/u/42936303/help3.mp3 [17:31] <namccarty> I've been having the same problem on linux, but I've only done serious troubleshooting on windows so far [17:31] <namccarty> let me get out the laptop and see if I can reproduce these results on linux [17:33] <klaxa> it seems that almost every other byte is 0x00 [17:33] <namccarty> that is disturbing [17:38] <namccarty> alright, testing on the laptop [17:41] <namccarty> output appears to be correct for both - and pipe:1 on that particular file, let me locate one i was having trouble with in the past [17:45] <namccarty> appernently the issue isnt happening on my laptop at all [17:45] <namccarty> given that i was having this issue on linux before, and across two machines, i am now questioning my own sanity [17:46] <namccarty> i dont have sound on my laptop though, so the files are the same size, i dont know if they are both valid though [17:46] <Riso> if i run "~/ffmpeg -i in.mkv -c:a -vf subtitles=in.srt out2.mkv" it says [17:46] <Riso> File 'subtitles=in.srt' already exists. Overwrite ? [y/N] ^C [17:46] <Riso> if i ommit "-c:a" it runs fine [17:47] <c_14> -c:a needs an option/value after it. [17:47] <klaxa> use -c:a copy to copy the audio stream [17:47] <klaxa> as in: ~/ffmpeg -i in.mkv -c:a copy -vf subtitles=in.srt out2.mkv [17:47] <Riso> i want the audio to be untouched [17:47] <Riso> "-c:a copy" ? [17:47] <klaxa> yes, that will do that [17:48] <namccarty> welp, laptop is doing everything fine, must have managed to get a janky build of ffmpeg multiple times [17:48] <Riso> ah thanx [17:49] <namccarty> oh wait, now i remember, i managed to get it working on my desktop right before i started encountering cripiling graphics driver issues [17:51] <klaxa> maybe windows is just bad at piping [17:52] <namccarty> possibly [17:54] <namccarty> okay, turns out it is just powershell that is bad at piping [17:54] <namccarty> now to figure out why madsonic is trying to ship my phone the version powershell is making [17:58] <namccarty> okay, its shipping out a valid mp3 after the switch to pipe:1 from - [17:59] <namccarty> but its not one my phone wants to play [18:03] <namccarty> annoying, but i dont think it's ffmpegs problem at this point [18:03] <klaxa> the 0x00 bytes in the file won't come from ffmpeg [18:04] <klaxa> maybe switch to *nix as your default os? :> [18:05] <namccarty> problem with that on my desktop is graphics drivers [18:05] <namccarty> im waiting for the stable amd drivers to support the latest version of xorg [18:06] <namccarty> but it appears that even though i now have ffmpeg outputting something that VLC can play [18:06] <namccarty> for some songs it is still outputting an invalid mp3 on both windows and linux [18:07] <namccarty> but [18:07] <namccarty> only when it is being piped [18:07] <namccarty> huh [18:08] <namccarty> piped version contains one less mpeg frame [18:08] <namccarty> both the linux and the windows produced versions [18:08] <klaxa> huh [18:08] <klaxa> what happens if you use lame without ffmpeg inbetween? [18:11] <namccarty> hold on, let me see what happens when i try that [18:16] <namccarty> same thing with the one less mpeg frame, let me see what happens when i try to play it on my phone [18:16] <namccarty> works on my phone and the mp3 is totally valid [18:17] <klaxa> well yeah that's the way mp3 works [18:17] <klaxa> you take frames and concatenate them [18:17] <klaxa> frame size is specified in the 4-byte header [18:18] <namccarty> well, thing is though, the only diffrence is one file is being piped and another is being output as a named file, i know you could get the same data with a diffrent number of frames [18:18] <namccarty> just strikes me as odd that we are getting a diffrent number here [18:18] <klaxa> not really [18:18] <klaxa> if the number of frames is different, the file is different [18:18] <klaxa> and yeah it's pretty odd [18:19] <klaxa> let me see if i can reproduce [18:19] <namccarty> i meant data here as in what the person eventually hears [18:20] <namccarty> but yeah, something about using ffmpeg with this particular song (and all the others i have by the same artist) appears to be corrupting the file when it is piped [18:21] <namccarty> but in such a way that VLC is still able to play it, but the media player on my phone is not [18:22] <namccarty> here is one of the songs that triggers issue for me [18:22] <namccarty> https://dl.dropboxusercontent.com/u/42936303/all.mp3 [21:18] <hannes3> i encode images to a video with "-i %04d.png -vcodec libx264 -pix_fmt yuv420p out.mkv". now i would like to add an audio track. i have a mp3 but in the video the audio would need to be at 19.55s already. can i do that on the commandline or do i need to cut the mp3 beforehand? [22:33] <llogan> hannes3: ffmpeg -i video.mkv -i audio.mp3 -c copy -shortest output.mkv [22:33] <llogan> if i understand yuor question correctly [00:00] --- Mon Mar 17 2014

1 0