Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

May 2014

1 participants
62 discussions

ffmpeg-devel.log.20140525
by burek 26 May '14

26 May '14

[00:40] <michaelni> probably [00:51] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:46e3883519b7: avcodec/opusdec: check alignment, misalignment could lead to crashes with avx [02:24] <cone-248> ffmpeg.git 03James Almer 07master:d94e255dd114: x86/dsputilenc: make the SUM_ABS_DCTELEM macro more readable [03:27] <cone-248> ffmpeg.git 03Aman Gupta 07master:4372fb7a5722: avcodec/webvttenc: add webvtt encoder [03:42] <clever> any good docs on how to use av_opt with the rtmp options? [03:43] <clever> i dont see any clear way to get the void* for the 1st argument to av_opt_find [03:59] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:fa915d4193e1: avcodec/aac: fix () in IS_CODEBOOK_UNSIGNED macro [03:59] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:97e6b5ee3a16: avcodec/golomb-test: fix () in EXTEND() macro [03:59] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:af62b42736c0: avcodec/h264: fix () in macros [03:59] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:0f45e643cc09: avcodec/h264_slice: fix macro () [03:59] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:70f671c39e53: avcodec/hevc: fix () in macros [04:00] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:f276bf303cbb: avcodec/ivi_dsp: add some missing () to macros [04:02] <michaelni> ubitux, iam not sure EOVERFLOW is supported on all platforms, i remember we removed it elsewhere in the past [04:02] <michaelni> i think [06:20] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:fa160af08b6f: avcodec/mlpdec: fix () in MSB_MASK() macro [06:20] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:6e720c5c815e: avcodec/mss34dsp: fix () in SOP* macros [06:21] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:cf7ff0146c76: avcodec/mss4: Fix () in MKVAL() macro [10:47] <cone-128> ffmpeg.git 03Clément BSsch 07master:657c603263fb: avcodec/webvttenc: do not use EOVERFLOW. [10:47] <ubitux> michaelni: fixed [12:49] <michaelni> ubitux, thx [13:15] <cone-128> ffmpeg.git 03Nidhi Makhijani 07master:e0d01dc7d7fc: smoothstream: check malloc calls [13:15] <cone-128> ffmpeg.git 03Michael Niedermayer 07master:ee9b0199c843: Merge commit 'e0d01dc7d7fc3ce4c23f286a10870e9599c8b8b9' [14:32] <BBB> I'm looking at that swresample avx inline... [14:32] <BBB> really michaelni [14:32] <BBB> you have to learn to write more maintainable assembly [14:54] <compn> you mean non-michael maintainable asm [14:54] <compn> michael can maintain it :P [14:55] <compn> hehe [14:55] <Daemon404> 1/g 35 [14:55] <Daemon404> fffs. [14:55] <compn> thats the old trick, code so no one but you can fix it. "job security" [14:56] <compn> Daemon404 : expensive pot ? [14:56] <Daemon404> eh? [14:57] <Daemon404> i only typo'd by one character.... this time [14:57] <compn> thought you were buying 1gram of something for 35 bux, nevermind [14:58] <nevcairiel> dont mind compn, he is the resident off-topic troll [14:58] <Daemon404> maybe compn is right [14:58] <Daemon404> that old numb3rs clip says irc is like boats in the ocean [14:59] <nevcairiel> if you could ignore some of those crazy inaccuracys, it was a decent show tho [15:00] <compn> i need to setup an x86 android build environment i think. got some old googletv hardware 'logitech revue' [15:00] <Daemon404> nevcairiel, but the entire show is based around it [15:00] <Daemon404> i cringed myself into a coma when they were using uncertainty theory to predict where a bomb will go off [15:01] <nevcairiel> luckily i only cringed at the tech references as I know shit about such math theories :p [15:02] Action: Daemon404 enhances nevcairiel [15:02] <nevcairiel> and really, who gets tech references right [15:03] <nevcairiel> no money for one geek in the multi-million-dollar budgets [15:03] <Daemon404> the girl with the dragon tattoo did [15:03] <Daemon404> i mean the proper swe version [15:03] <nevcairiel> the swe version was way better than the hollywood version anyway [15:03] <Daemon404> why yes, it was [15:04] <Daemon404> i have a vague memory of watching the dude run dpkg or apt-get in the bg of that movie [15:05] <nevcairiel> according to hollywood every mad hacker needs 6 screens, with at least 5 of those running large text or hex dumps of stuff over and over, anyway [15:06] <Daemon404> yes: http://www.smbc-comics.com/index.php?db=comics&id=2526 [16:03] <cone-128> ffmpeg.git 03Michael Niedermayer 07master:6e725f19f88c: configure: try to fix the disappearing "r" bug in fate [18:29] <cone-128> ffmpeg.git 03Timothy Gu 07master:ad733089b024: x86: dsputilenc: convert ff_sse{8, 16}_mmx() to yasm [19:14] <kurosu> michaelni, iive, hi [19:15] <kurosu> I noticed simple_idct has a bias, and I'm wondering if that's expected [19:15] <kurosu> for only a dc value, output is (dc+3-(dc>>15))>>8 (empirical formula) [19:16] <kurosu> so the rounding differs between positive and negative values [19:35] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:c163f5e601ca: avcodec/pngdec: fix () in OP_* macros [19:35] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:d62a4707f92e: avcodec/opus_celt: Fix () in CELT_PVQ_V macro [19:35] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:eb718f4c532c: avcodec/opus: Fix () in ROUND_MULL() macro [19:36] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:9babf13764dd: avcodec/msvideo1enc: Fix () in MKRGB555() macro [19:40] <michaelni> kurosu, the idct is tuned to minimize rounding errors relative to what encoders used or rather relative to the material that was posted to us that showed artifacts [19:43] <kurosu> yeah I know about the specs requesting max/mean errors [19:43] <kurosu> but that was kind of weird [19:43] <kurosu> *requiring [19:44] <kurosu> if you think this is ok, then so be it [19:45] <kurosu> but this may have underlined a potential issue [20:07] <jamrial> kurosu: re 3dnow, i have a k10 so i can test [20:07] <jamrial> back when i made that comment i hadn't noticed you sent a 3dnow patch yourself [20:07] <jamrial> if you think it's worth it, i can send a patch adding a 3dnow version [20:07] <jamrial> iive at least seemed somewhat interested :P [20:08] <kurosu> jamrial, nice - I think only veryfying that it works is good enough [20:08] <kurosu> eg testing with cpuflags=mmx+3dnow [20:09] <jamrial> yeah [20:09] <jamrial> alright, i'll send a patch later [20:09] <kurosu> as for a patch, I tend to think: current patch/code doesn't change things for your setup? send patch [20:10] <kurosu> at least that's what the discussion about avx inline made me think [20:11] <kurosu> refusing code because of this is kind of baffling - I tend to thing something > nothing [20:11] <kurosu> *think [20:26] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:a3950a90f643: Revert "x86: dsputilenc: convert ff_sse{8, 16}_mmx() to yasm" [20:35] <kurosu> wasn't maillan adding the tag [FFmpeg-devel] to messages before ? [20:36] <kurosu> I never think of adding it manually [20:37] <nevcairiel> you shouldn [20:38] <nevcairiel> 't [20:38] <nevcairiel> its added automatically [20:38] <kurosu> I'm using gmail maybe it hides from me - but then that's gmail that has changed without me noticing [20:38] <kurosu> *it hides it [20:39] <kurosu> thanks [20:50] <nevcairiel> if i send mails via git send-email, it never showed the prefix in gmail for me [21:50] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:e5dc7439e33c: avcodec/proresenc_anatoliy: Fix () in macros [21:50] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:a45e6b7bc5ad: avcodec/psymodel: Fix () in AAC_CUTOFF() macro [21:50] <cone-679> ffmpeg.git 03Michael Niedermayer 07master:9f50d3b944a9: avcodec/rv34: Fix () in GET_PTS_DIFF() macro [21:52] <ubitux> michaelni: did you get a tool or compiler warning about these macro? [22:55] <michaelni> ubitux, no just grep define and my eyes so iam surely missing 50% or so of these issues [22:55] <ubitux> did you hit a bug or something because of this recently? [22:56] <michaelni> no [22:57] <ubitux> ok :) [22:57] <michaelni> do you think its not worth fixing them ? [22:57] <ubitux> no no, no problem with that [22:57] <ubitux> just curious [22:58] <michaelni> i thought its better to fix as that could lead to rather hard to debug issues [23:53] <kurosu> is msvc2013 express + yasm sufficient/able to compile ffmpeg? [23:54] <Daemon404> yep [23:54] <Daemon404> well you need msys as well obviously [23:54] <Daemon404> to run configure [23:55] <kurosu> yeah, the help page made me guess so [23:55] <Daemon404> youre looking at platform.{texi,html,1} ? [23:55] <kurosu> I was wondering how restricted the express version was [23:55] <Daemon404> express versins lacks uh [23:56] <Daemon404> MFC i think [23:56] <Daemon404> nothing meaningful [23:56] <kurosu> Daemon404, basically yes, through I just googled it :) [23:56] <kurosu> *though [23:56] <Daemon404> jab me if anythin is unclear [23:56] Action: Daemon404 wrote it [23:56] <nevcairiel> MFC and ATL is missing, so really only stuff older apps would be using [23:56] <nevcairiel> dont understand the limitation anymore really [23:56] <Daemon404> the IDE is missing some stuff too [23:56] <Daemon404> team stuff [23:56] <nevcairiel> yeah sure [23:56] <Daemon404> not like i use that though [23:57] <kurosu> well my issue is that it's getting hideous debuging some issues with gdb [23:57] <nevcairiel> the ide is actually quite a bit limited from some features [23:57] <Daemon404> debugging is pretty much the reason msvc support was added [23:59] <kurosu> last time I did debug asm with msvc it was less friendly, but that was another decade [00:00] --- Mon May 26 2014

1 0

ffmpeg.log.20140525
by burek 26 May '14

26 May '14

[02:37] <Logicgate> is there a way I can resize a video and add black padding at the same time [02:37] <Logicgate> example a video that's 856x480, I want it to keep its aspect ratio and become 480x480 [02:37] <Logicgate> and add black padding [02:41] <klaxa> use scale and pad, see: http://www.ffmpeg.org/ffmpeg-filters.html#pad [02:41] <Logicgate> http://puu.sh/8ZR07.png [02:42] <Logicgate> this is what I get though [02:42] <klaxa> your command string would clear up a lot [02:43] <Logicgate> ffmpeg -i tmp2/1.mp4 -ss 00:00:00 -t 00:00:05.972639 -vf scale=480:-1 -vf 'pad=480:480:0:106:black' -c:v libx264 -c:v libfaac -ac 2 -deinterlace -b:a 96k -c:v libx264 -preset:v slow -profile:v baseline -level 3 -maxrate 10000000 -bufsize 10000000 -b:v 1200k -f mp4 -threads 0 -strict experimental -movflags faststart -vsync 2 -pix_fmt yuv420p tmp2/test.mp4 [02:43] <sacarasc> Try -vf scale=480:-1,pad=480:480:0:106:black [02:44] <Logicgate> :D [02:44] <Logicgate> it's twerking! [02:45] <Logicgate> why does that work? [02:45] <Logicgate> because it's in one statement [02:46] <sacarasc> It works because it is one statement. [02:46] <sacarasc> It only takes one -vf string, so you have to chain them. [02:49] <Logicgate> awesome [02:50] <Logicgate> -pix_fmt, will it fail if video is 4:4:4? [08:34] <decci> Hi [08:34] <decci> http://paste.ubuntu.com/7514089/ [08:34] <decci> I am try running the ffmpeg command but get converter with 0 byte [08:35] <decci> How to fix it [08:58] <wizbit> i want clients to be able to stream webcam using ffmpeg to a remote server running ffserver, then the output of ffserver embedded into a webpage using <video> or something like jplayer, is that possible? [09:29] <decci> I encountered broken ffmpeg default settings detected issue while running ffmpeg [09:29] <decci> Adding -vpre hq or vpre max solved this issue. May I know what it means [09:30] <decci> I followed https://code.google.com/p/winff/issues/detail?id=89 [09:30] <relaxed> decci: read https://trac.ffmpeg.org/wiki/x264EncodingGuide [09:30] <JEEB> if you have a new enough ffmpeg you shouldn't get that unless you force a lot of retarded options [09:30] <JEEB> if you have an old old ffmpeg then you're not gonna get much help :P [09:31] <JEEB> because that stuff has gotten a /lot/ less retarded within the last three years [09:32] <decci> relaxed: If I run without vpre it throws the error http://paste.ubuntu.com/7514294/ [09:32] <decci> relaxed: If I run putting vpre hq or max it runs well [09:32] <decci> relaxed: Can you brief what was the issue [09:33] <decci> JEEB: ffmpeg FFmpeg version 0.6.5, Copyright (c) 2000-2010 the FFmpeg developers built on Jan 29 2012 [09:33] <relaxed> you're version is way out of date. [09:33] <relaxed> your* [09:33] <decci> JEEB: is that old..i just installed ffmpeg through yum [09:33] <decci> relaxed: How to install newer version of ffmpeg [09:33] <JEEB> yes, that version is from 2010 :) [09:34] <decci> JEEB: May I know the link which talks about latest version [09:34] <relaxed> decci: uname -a [09:34] <decci> JEEB: I would prefer yum based installation [09:34] <decci> relaxed: uname -a Linux serv01.m2170.sgded.com 2.6.32-279.22.1.el6.x86_64 #1 SMP Wed Feb 6 03:10:46 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux root@serv01 [~]# [09:35] <relaxed> http://johnvansickle.com/ffmpeg/ [09:35] <JEEB> decci, you'd have to find a repository then, which is something I think you wouldn't be able to find easily [09:35] <relaxed> then follow the wiki guide I gave you earlier [09:36] <decci> JEEB: let me check if newer repository is available [09:36] <JEEB> static compiles and/or compiling yourself just happens to be much simpler than finding someone packaging it for you [09:37] <decci> JEEB: May I know what version is latest [09:37] <decci> JEEB: I will search with that version [09:38] <JEEB> 2.2.2 [09:39] <decci> JEEB: http://dl.atrpms.net/el6-x86_64/atrpms/stable/ffmpeg-2.2.1-65.el6.x86_64.rpm [09:40] <wizbit> im coming to the conclusion that its impossible to stream mp4 so iphones and android devices can play it [09:40] <wizbit> mp4 is not a streamable format :( [09:47] <JEEB> decci, well if that random repository is fine enough for you, then it's fine. Although it seems like it's already behind :P [09:47] <JEEB> people often do one-offs or so [09:47] <JEEB> wizbit, both true and untrue [09:48] <JEEB> untrue because you can use movie fragments that will let a parser that supports them start playing right away [09:48] <JEEB> true because I have no idea if android or iOS supports fragments :P [11:38] <wizbit> JEEB: can ffserver stream mp4 files? i managed to get my webcam to stream in .swf format but not .mp4 [11:38] <JEEB> no idea about ffserver :P [11:52] <klaxa> mp4 is not made for live-streaming anyway, is it? [11:53] <JEEB> <JEEB> wizbit, both true and untrue [11:53] <JEEB> <JEEB> untrue because you can use movie fragments that will let a parser that supports them start playing right away [11:53] <JEEB> <JEEB> true because I have no idea if android or iOS supports fragments :P [11:54] <JEEB> it's not the most straightforward thing you could be using for streaming [11:54] <JEEB> but it isn't impossible [12:12] <klaxa> ah [12:18] <DelphiWorld> hi ffmpegsters [12:18] <DelphiWorld> please, please. i'm stuck for weeks on an issue with hls [12:18] <DelphiWorld> i cant include multiple audio track [12:18] <DelphiWorld> if i do -map 0:1 -map 0:2 i get them when probing but i dont see the language [12:19] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -map 0:0 -map 0:8 -map 0:2 -c:v libx264 -crf 18 -profile:v baseline -maxrate 400k -bufsize 1835k -pix_fmt yuv420p -flags -global_header -bsf:v h264_mp4toannexb -r 30 -ab 32k -f segment -segment_time 10 -segment_list /tmp/hls/eu.m3u8 -initial_offset 10 segment_%05d.ts [12:19] <decci> I am facing [12:19] <decci> I am facing an issue http://198.20.89.58/~upeoapp2/ffmpegtest/upload.php [12:20] <decci> When I upload any image I got error:-1 [12:30] <decci> Hello Guys [12:32] <decci> While I try to upload a file through http://198.20.89.58/~upeoapp2/ffmpegtest/upload.php [12:32] <DelphiWorld> guys... [12:32] <decci> It doesnt work [12:32] <decci> It reports error:1 [12:38] <DelphiWorld> relaxed ? [12:39] <DelphiWorld> anyone do hls here? [12:59] <DelphiWorld> hi iive [13:00] <DelphiWorld> iimay you be able to help me with a small hls problem? [13:00] <DelphiWorld> iive: maybe you be able to help me with a small hls problem? [13:01] <iive> i know nothing of http live stream [13:01] <DelphiWorld> iive: its not a hls issue but i belieuv its a ts issue, may i explain? [13:01] <iive> ask and be patient. somebody may know [13:03] <DelphiWorld> ffmpeg command: [13:03] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -map 0:0 -map 0:2 -map 0:8 -c:v libx264 -vb 512k -c:a libfdk_aac -profile:a aac_he_v2 -b:a 32k -ac 2 -ar 44100 -f mp4 eu.mp4 [13:03] <DelphiWorld> this one give multi audio track on a mp4 [13:03] <DelphiWorld> work perfectly [13:03] <DelphiWorld> but [13:03] <DelphiWorld> if i segmant it into hls ansd not mp4... [13:04] <DelphiWorld> when i probe i see the tracks but dont see the audio language describtion [13:04] <DelphiWorld> like fr or en or ara [13:04] <DelphiWorld> that's my problem [13:07] <iive> just a wild guess. could this be related to not processing metadata? there was something about it... [13:07] <DelphiWorld> iive: yeah, but why if i use .mp4 i see the tracks? [13:07] <DelphiWorld> iive: you want a video? [13:07] <iive> no, i can't do anything with it. [13:08] <DelphiWorld> iive: you may just probe it online:P [13:08] <iive> does mp4 have proper track description, like language? [13:08] <DelphiWorld> yeah iive but ts is not doing that [13:08] <decci> DelphiWorld: Can you please take care of your spellings [13:09] <DelphiWorld> decci: yeah, if you give me your eyes :D [13:09] <DelphiWorld> decci: i am blind but using tts:) [13:10] <iive> decci: you are not helping. [13:10] <decci> iive: I am aware [13:10] <decci> iive: Sorry [13:11] <DelphiWorld> iive: to be sure, i made a mp4 and now i'm hlsing it [13:12] <DelphiWorld> ok, confirmed [13:13] <DelphiWorld> -f mp4 ...test.mp4 give the track describtion [13:13] <DelphiWorld> but -f hls or -f mpegts wont [13:16] <iive> DelphiWorld: could you put the text output of ffmpeg in some pastebin site. first with output to mp4 and second time with output to hls. [13:16] <DelphiWorld> iive: the entire log? [13:17] <iive> it shouldn't be more than one page. and yes, that is why i'm asking you to paste it to a pastebin website. [13:17] <iive> if it is too much trouble for you, don't bother. [13:18] <DelphiWorld> iive: no, i must [13:18] <DelphiWorld> iive: you want it during encoding? [13:18] <iive> but I wonder if it would say something different about metadata [13:18] <DelphiWorld> iive: i'lle provide now [13:20] <iive> I don't know about hls, but mpeg-ts dvb standard supports language codes. No idea about the libavformat muxer. [13:21] <iive> e.g. from 'man ffmpeg' I found this example: [13:21] <iive> To set the language of the first audio stream: ffmpeg -i INPUT -metadata:s:a:1 language=eng OUTPUT [13:22] <DelphiWorld> iive: mp4 output: http://paste.debian.net/101670/ [13:28] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -map 0:0 -map 0:8 -map 0:2 -c:v libx264 -vb 512k -c:a libfdk_aac -profile:a aac_he_v2 -b:a 32k -ac 2 -ar 44100 -f hls -metadata:s:a:1 language=ara -metadata:s:a:1 language=eng /tmp/hls/eu.m3u8 [13:28] <DelphiWorld> this one provide also a ts, but... still metadata is not comming [13:33] <iive> :s specifies it is about subtitles :a specifies it is (also) about audio and :1 specifies that it is about second stream. [13:34] <iive> but it should have specified at least one of the streams. [13:34] <DelphiWorld> iive: i'lle provide you ts output now [13:35] <DelphiWorld> iive, http://paste.debian.net/101671/ [13:39] <iive> both text outputs are almost identical. I'm out of ideas. [13:39] <DelphiWorld> yeah... [13:39] <iive> try moving the metadata command after the output [13:39] <DelphiWorld> evean with the idea you gave me there's still no metadata [13:40] <DelphiWorld> after the filename? [13:40] <iive> yes [13:40] <DelphiWorld> hold on [13:40] <iive> the position of the options does matter. [13:40] <DelphiWorld> i see [13:42] <DelphiWorld> lol, same problem... [13:46] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -map 0:0 -map 0:8 -map 0:2 -c:v libx264 -vb 512k -c:a libfdk_aac -profile:a aac_he_v2 -b:a 32k -ac 2 -ar 44100 -f hls ./test.m3u8 -metadata:s:a:1 language=ara -metadata:s:a:2 language=eng [13:52] <DelphiWorld> back [13:52] <DelphiWorld> iive, fyi. the issue is not only a problem of metadata [13:52] <DelphiWorld> but i dont evean see the tracks listed in client SIDE! [13:53] <iive> as I said, I know nothing of hls. or libavformat muxers. [13:53] <DelphiWorld> any way of getting it live except of hls? [13:53] <iive> i hope somebody here would be able to help you. [13:53] <iive> c_14 seems to know a lot, but he is away atm. probably asleep. [13:53] <DelphiWorld> ;) [13:54] <decci> Hi [13:54] Action: DelphiWorld would be probably c_13 :P [13:54] <decci> I need help with http://198.20.89.58/~upeoapp2/ffmpegtest/upload.php [13:54] <DelphiWorld> Hi decci [13:54] <dloo> hello [13:55] <dloo> i'm trying to record v4l2 input, so i try : ffmpeg -f alsa -i hw:0,1 -f v4l2 -channel 1 -i /dev/video0 -vcodec libx264 -b 2500k -acodec copy test.mkv [13:55] <dloo> i get this : http://i.imgur.com/2el9tub.png [13:55] <dloo> what's wrong? :( [13:56] <decci> DelphiWorld: While I run http://paste.ubuntu.com/7515147/ it runs well as upeoapp2 user [13:56] <DelphiWorld> decci: i realy know nothing about this upload [13:56] <decci> DelphiWorld: But whenever I run from web it throws error : -1 [13:58] <decci> DelphiWorld: Ok [14:35] <dloo> nobody can help me? [14:42] <iive> dloo: are you able to view normal image with any other program. [14:42] <iive> dloo: is the channel 1 an composite input, not RF modulated one? [14:43] <iive> sorry gtg [14:47] <dloo> http://pastie.org/private/ml0o7tu2c6jdq2ryljhsbw [14:47] <dloo> igitoor: yes, with vlc [14:47] <dloo> damn [14:54] <sacarasc> You missed the second part of what fflogger asked. :( [14:55] <dloo> wait.. i think i got it :) [16:25] <DelphiWorld> c_14: alive? [19:43] <pqatsi> About ffserver, its possible to ffserver act just as a streaming relay, w/o re-encoding anything and being just a proxy? [21:12] <satanist> wo krig ich jetzt ne maus her [21:13] <satanist> ich will doch nur sauerbraten spielen [21:38] <Tylertwo> Hey, anyone with experience streaming to twitch? [21:40] Action: DelphiWorld back [22:01] <Tylertwo> I'm trying to do so. Using the command "avconv -f x11grab -s 200x200 -r 15 -i :0.0 -c:v libx264 -pre fast -pix_fmt yuv420p -s 640x480 -threads 0 -f flv "rtmp://live.twitch.tv/app/live_<myPrivateKey>"" I get the error from twitch that "Audio codec must be set to MP3 or AAC (it is currently "unknown")". Any idea about fixing this? [22:01] <c_14> -c:a $codec [22:02] <c_14> This channel, however, does not provide support for avconv. [22:02] <c_14> That would be #libav [22:56] <Tylertwo> ok, thanks [23:06] <dloo> hi [23:06] <dloo> why this [harold@855a-0d8d-2-53-k8vsed ~]$ ffmpeg -t 00:00:01 -f alsa -i hw:0,1 -f v4l2 -channel 1 -i /dev/video0 -standard 3 -vcodec libx264 -b 2500k -acodec copy /run/media/harold/52F82754F827361F/k7/le_college_c_est_pas_sorcier.mkv does not stop the record after 1 sec? [23:07] <rsdrsdrsdrsd> Hi, I have a small question about filtering [23:07] <c_14> dloo: -t isn't an input option, it's an output option [23:07] <c_14> rsdrsdrsdrsd: ask, if someone can help you they will [23:07] <dloo> c_14: i('ve tried to put it at the end of the command, it do the same [23:07] <rsdrsdrsdrsd> ok that would be great [23:07] <rsdrsdrsdrsd> thank you [23:07] <rsdrsdrsdrsd> Here it comes [23:08] <c_14> dloo: it should be right before output.mkv [23:09] <rsdrsdrsdrsd> If I have the following filter -vf "crop=720:384:0:48,scale=trunc(480*dar/2)*2:480,setsar=1" the black border are chopped of but it looks like the dar in the scale filter isn't changed properly after the crop [23:10] <dloo> c_14: thanks you, it is working :) [23:11] <rsdrsdrsdrsd> if I remove the crop part and compare the videos the one with the crop part seems a bit more stretched than without the crop [23:11] <rsdrsdrsdrsd> I would assume that with or without black borders the video should have the same aspect ratio [23:12] <rsdrsdrsdrsd> but due to the cropping something goes wrong [23:12] <c_14> try crop=720:384:0:48:1 [23:20] <rsdrsdrsdrsd> Hmm, that option removes the complete sar info [23:26] <c_14> what exactly are you trying to do? [23:27] <rsdrsdrsdrsd> I have a bunch of videos some with none square pixels I want to crop the bar and encode to fixed height [23:27] <rsdrsdrsdrsd> and to square pixels [23:28] <rsdrsdrsdrsd> I first run a crop detect command to check for the black bars [23:29] <rsdrsdrsdrsd> and insert the outcome in the filter [23:37] <rsdrsdrsdrsd> I am using ffmpeg version N-39361-g1524b0f [23:45] <c_14> Why are you assuming that the video should have the same aspect ratio after cropping? Unless the bars are equidistant on all sides. [23:47] <rsdrsdrsdrsd> I found out that without the 1 in the crop the frame size is changes as expected and the dar is also changed, the sar is kept the same [23:47] <rsdrsdrsdrsd> that is ok [23:47] <rsdrsdrsdrsd> buit after that someting goes wrong [23:54] <rsdrsdrsdrsd> hmm, while comparing, cropping removes some 6x extra on both sides, that makes the difference [00:00] --- Mon May 26 2014

1 0

ffmpeg-devel.log.20140524
by burek 25 May '14

25 May '14

[02:10] <cone-205> ffmpeg.git 03James Almer 07master:7538ad224835: x86/hevc_deblock: improve chroma functions register allocation [02:15] <cone-205> ffmpeg.git 03Lou Logan 07master:9eaa8c22bc40: MAINTAINERS: remove myself as website maintainer [03:43] <cone-205> ffmpeg.git 03Christophe Gisquet 07master:f0aca50e0b21: x86: hpeldsp: implement SSE2 versions [03:52] <cone-205> ffmpeg.git 03Christophe Gisquet 07master:9722a6a3f35c: x86: hpeldsp: implement SSE2 put_pixels16_xy2 [05:09] <cone-205> ffmpeg.git 03Billy Shambrook 07master:308188be3412: Add metadata injection to blackdetect [11:10] <cone-248> ffmpeg.git 03Clément BSsch 07master:cba92a222615: avformat/vobsub: do not create empty streams. [13:33] <cone-248> ffmpeg.git 03Nidhi Makhijani 07master:8692e6284f51: rdt: check malloc calls [13:33] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:726316240bcc: Merge commit '8692e6284f5169257a537c8fc25addf32fc67c87' [13:59] <ubitux> why is there a call to emms_c() at the beginning of ff_faandct()? [14:00] <ubitux> ah mmh well it makes sense somehow [15:52] <ubitux> why do we have reference/simple (i)dct only for 8x8 blocks? [15:55] <ubitux> we have dozens of dct implementation ± accurate for 8x8 but i can't see any for 4x4, 16x16, etc, except in codec specific code [15:56] <kierank> probably because those are defined to be bitexact [16:02] <ubitux> AFAICS, we have: jfdctint which defines some fdct not that much accurate based on aan, jfdctfst.c is the aan one, so not yet bitexact (but more accurate than the previous one), then we have a float version of the aan (faandct.c), then we have the idct in simple_idct*, then we have the reference in dct* [16:02] <ubitux> and then, there all the codec specific ones [16:02] <ubitux> did i miss any? [16:03] <ubitux> ah, and all of those are 8x8 [16:04] <ubitux> i don't know which of those have an asm version btw [16:45] <ubitux> ah there is jrevdct.c for some more matching idct [16:46] <Daemon404> there should be matching asm [16:46] <Daemon404> theyre from ijg, so you could swap out jpeg-turbo asm [16:46] <Daemon404> no? [16:50] <ubitux> no idea. [16:53] <ubitux> btw, does anyone knows some algorithm names for making a sliding window 2d dct faster? [16:53] <ubitux> like i mean, there is just one column of the input changing after each iteration (of course shifted in one direction) [16:53] <ubitux> nothing can be re-used between these dct? [17:00] <cone-248> ffmpeg.git 03Christophe Gisquet 07master:81aa0f4604f9: x86: hpeldsp: implement SSSE3 version of _xy2 [18:59] <kurosu_> michaelni: I have some questions about mpegvideo, in particular decoding and mpeg2 [18:59] <kurosu_> my intent is to do like all cool decoders, take into account the last coeff index [19:00] <kurosu_> (so as to apply a DC only, but one may want to implement a pure vertical or horizontal version) [19:00] <kurosu_> in the idct [19:01] <kurosu_> in profiling idct is now 25% of the decoding time and I guess most of the time is spent on vlc decoding then [19:01] <kurosu_> however, that last coefficient count is wrong [19:02] <iive> vlc is in the dequant functions. [19:02] <nevcairiel> as if you could turn mpeg2 into a cool decoder [19:02] <nevcairiel> psh [19:03] <kurosu_> yeah, there's dxva/vdpau/whatever for that [19:03] <kurosu_> but I specialize in useless stuff [19:04] <michaelni> kurosu_, with mpeg2 be carefull as th dequantization can flip the LSB of coeff 63 even when coeff 63 is 0 [19:04] <kurosu_> michaelni: already noticed that [19:04] <kurosu_> the dequant function are not taking this into account to update the index [19:04] <kurosu_> although the alternate scan forces 63 but at least the index is higher than the real one [19:05] <iive> if I remember correctly, the last coefficient is actually in the linear sense, so after the zig-zag when it is put in 2d matrix it might be somewhere else completely. [19:05] <kurosu_> so my question is, do you know if there could be other issue [19:06] <kurosu_> on the other hand, one would expect idx=1 to mean "dc" but the index is often completely different [19:07] <michaelni> know? no [19:07] <iive> idx=0 is dc [19:07] <michaelni> iive, yes [19:07] <kurosu_> err, yeah off-by-one error [19:07] <kurosu_> anyway, my comment still applies [19:08] <iive> afaik, idct routines check if all coefficient in a row/col are zero, and skip processing it if that is the case [19:09] <kurosu_> maybe the c implementation, I haven't checked the inline asm [19:09] <iive> also, the matter is a little bit more complicated, as to help paralization, some mmx variants use strange permutations. [19:10] <kurosu_> ok, important to know [19:10] <kurosu_> but I don't think it has an impact for dc-only [19:11] <kurosu_> in the add/put dct functions, I recomputed the last index, and if 0, did a add offset and clamp [19:11] <kurosu_> which makes me thing that every codec is reimplementing that dc-only version, which is kind of inefficient [19:12] <kurosu_> maybe I could start there instead: convert all those to use a common function, and then codec-specific versions only compute the average pixel offset from the dc coeff [19:13] <kurosu_> and pass it to that function [19:14] <iive> well, it depends... [19:15] <iive> doning a check in the codec before calling function by a pointer might be faster than doing the check in the given function. [19:18] <iive> some cpu can speculatively execute or fetch instructions in advance, even when there is function jump. however I'm not sure that would be done if the function address is not fixed, should be read from memory or is held in register. [19:18] <iive> so, better benchmark it. [19:18] <iive> if you want ideas.... [19:19] <iive> i've thought is donig dequantization in SIMD could be beneficial. [19:20] <iive> the idea is that you need 3 linear arrays. in the first you write the coefficient. in the second you write the runlength, or coefficient position. in the third you write the reciprocial quantization matrix coefficients. [19:21] <iive> then you can do SIMD on 1'st and 3'd arrays using SIMD. [19:21] <iive> ideally, the first stage of the idct could be done without having to unpack the coefficients. [19:24] <iive> guh, why i'm writing donig instead of doing.... [19:25] <clever> and doning [19:25] <iive> yeh... [19:27] <iive> another idea. [19:27] <kurosu_> actually, dequantization is ideally done when either parsing vlcs or doing idct, but the former is simpler [19:29] <iive> in h264 cow and row operations are the same, so they just do a permutation and repeat the function. [19:30] <iive> older mpeg idct doesn't do that. [19:31] <iive> so, different route is to do the horizontal stage in parallel for a number of different blocks. [19:31] <kurosu_> my biggest issue now is more that all these add clamped functions / add_dc across vp3/vp8/rv40/vc1 x86 asm are the same concept [19:31] <kurosu_> let's go for the easiest and pratical path maybe ? [19:32] <iive> i'm kind of afraid that the easiest paths might have already been tried :} [19:39] <iive> but it doesn't heart to try [19:39] <iive> hurt [19:39] <iive> gah... [19:39] <kurosu_> the biggest roadblock being the other archs where I'd need to add equivalent assembly [19:42] <iive> you can always do C wrapper that does the dc check and fallbacks to the old existing asm. [19:43] <iive> should be better that falling back to pure C, but there is always the danger of been too good and nobody bothering writing the asm ;) [21:43] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:9a7d332b923b: avcodec/aacenc: dont use global quality if its negative [21:43] <cone-248> ffmpeg.git 03Michael Niedermayer 07master:ddeb58b90c41: avcodec/asvenc: dont use a negative global_quality [22:17] <cone-248> ffmpeg.git 03James Almer 07master:61eea421b23f: x86/dsputilenc: port sum_abs_dctelem functions to yasm [23:41] <BBB> saves 11 cycles? [23:41] <BBB> he probably means 1.1 (i.e. 1) right? [00:00] --- Sun May 25 2014

1 0

ffmpeg.log.20140524
by burek 25 May '14

25 May '14

[00:26] <jgh-> is there a way I can grab the fragments of an HLS stream without doing any sort of transmuxing or changing the container format at all? [00:33] <klaxa> can you explain that a little further? [00:33] <klaxa> HLS is just streams muxed in an mpeg-ts container [00:33] <klaxa> then segmented and downloaded over http [00:33] <jgh-> yeah, I'm basically just trying to download those segments without having to rig up a script [00:33] <jgh-> I figured FFmpeg might be able to accomplish such a thing [00:34] <klaxa> not sure if you can copy it segmented [00:34] <klaxa> copying the stream as a combined stream is pretty easy though [00:34] <klaxa> otherwise, just write a script, HLS is not that complex [00:35] <jgh-> yeah I guess that's what I'll have to do - I'm streaming to Wowza using RTMP but encountering some issues with the .ts files it's generating, so combining them into some other container would not be very helpful [00:36] <jgh-> in figuring out what's wrong with the files [01:11] <temhawk> where can i get the man-page from? i have pre-built ffmpeg binary [01:14] <c_14> Either from the source tree or from a distribution package. [01:14] <temhawk> c_14: ok, thanks [01:14] <c_14> Or check the online documentation: https://ffmpeg.org/ffmpeg.html [01:23] <temhawk> c_14: so I gotta compile it myself to get the man-pages? [01:25] <c_14> Well, technically you only need to compile the manpages themselves, but I don't really know how doxygen works. [01:27] <c_14> You could try ./configure --disable-all --enable-manpages and then see if make gets you the manpages without building everything else. [01:27] <temhawk> Yeah, I don't know either. I guess I can just compile it any way, take the man-pages and delete the rest? [01:29] <c_14> You can always just compile ffmpeg and then copy the manpages into your manpath or just man $file [01:30] <temhawk> Yeah. --disable-all doesn't seem to work. [01:31] <c_14> --disable-everything maybe? [01:33] <temhawk> Shouldn't there be a "build" directory after I make? [01:35] <klaxa> nope [01:38] <temhawk> where do I find the files then? Do I have to make install? [01:39] <c_14> In ffmpegdir/doc it should be all the .1 and .3 files [01:39] <c_14> Where you probably only want the .1 files since those are for the binaries. [01:39] <temhawk> hm, not there. maybe it failed [01:40] <c_14> I just tried it with ./configure --disable-everything --enable-manpages; make and they're there. [01:41] <temhawk> When I do that it says: "yasm/nasm not found or too old. Use --disable-yasm for a crippled build." [01:42] <c_14> just do it, you don't need assembler for the docfiles [01:42] <temhawk> but when I add --disable-yasm it fails. ok, nevermind it seems to be working now [01:43] <temhawk> c_14: success! [01:44] <temhawk> I had --disable-all before as well which looks like was the problem [01:44] <temhawk> everything != all [01:44] <temhawk> confusing [01:46] <llogan> you can probably just do "./configure && make doc" [01:47] <c_14> As a matter of fact, yes, yes you can. [01:47] <c_14> My mind. It must be going. [01:52] <voip> do have ffmpeg daemon mode ? [01:52] <voip> without & [01:52] <llogan> what you say??? [01:53] <voip> i need to run ffmpeg in daemon mode [01:53] <llogan> ffmpeg -d [01:53] <voip> one way use with "&" [01:53] <voip> ok thanks [01:54] <llogan> that is undocumented, AFAIK [02:01] <JodaZ> jgh-, the problem with segments coming from simple segmenting solutions is that they aren't seamless when not generated one after another [02:03] <JodaZ> jgh-, oh, you are streaming to wowza? increase keyframes [02:05] <jgh-> JodaZ: I have keyframes set to 2 seconds right now, does it need to be more frequent than that? [02:05] <JodaZ> no, that sounds about often enough [02:05] <JodaZ> what issues are you encountering? [02:06] <jgh-> Seems like the issue is more closely related to audio. Disabling audio seems to fix the issue (except for the fact that now there's no audio). [02:07] <jgh-> I am getting ERROR: (-12703) failed to parse segment as either an MPEG-2 TS or an ES from Apple's mediastreamvalidator [02:07] <jgh-> occasionally (not on every segment) [02:07] <JodaZ> well, as long as it plays, i wouldn't bother [02:07] <jgh-> VLC plays through it, but QuickTime and quicktime-based players (i.e. iOS) don't like it too much [02:07] <voip> llogan, -d dont works :( [02:07] <JodaZ> hmm [02:08] <JodaZ> jgh-, you sure there is no packet loss streaming to the server? [02:08] <jgh-> I'm streaming over TCP, but besides that I'm just streaming from an iPhone to my laptop that is on the same network [02:09] <JodaZ> jgh-, did you try downloading the segments and looking at them manually? [02:09] <jgh-> That's what I'm going to do next [02:10] <jgh-> I came in here wondering if I could do that quickly with FFmpeg, but it looks like I'll jsut have to rip up a script to download the segments [02:17] <llogan> voip: what's wrong with it? [02:19] <voip> whan runing with -d . ffmpget like freese not starting no messages, no carret retrn [02:19] <llogan> i don't understand you. [02:30] <voip> llogan, i am typing command "ffmpeg -d -i input ...output" then hitting "enter" button. cursor goes too new line and blinking, i didnt see any messages regarding ffmpeg starting. In servers process i see ffmpeg running instances. When i am typing ctrl-c all ffmpeg process are desaper from ps [02:33] <llogan> that's daemon mode [02:36] <voip> llogan, after ctrl-C or terminal disconnection ffmpeg shuld resume works, in my case its stoppong [02:40] <llogan> voip: i don't know what the correct or expected behavior is supposed to be, but if you think it is a bug then please report it. [02:40] <voip> ok found it works when i using -d and "&" togetther [02:40] <llogan> ok [03:13] <haithere> hi. is it possible to use ffmpeg on a remote Linux machine without installing it (I don't have sudo rights)? coming from Windows world, I'd expect something like a portable version... [03:13] <haithere> on Windows all you need is a "ffmpeg.exe" file [03:14] <haithere> but on Linux I need to do "apt-get install ffmpeg" which needs sudo rights [03:14] <c_14> Just get one of the static builds I posted. [03:22] <haithere> hmm [03:22] <haithere> it says: FATAL: kernel too old Segmentation fault [03:22] <haithere> so I guess I should try older version? [03:23] <c_14> probably... [03:23] <c_14> What does uname -a say? [03:23] <haithere> 2.6.9-103.ELsmp [03:24] <haithere> so I'll take my ffmpeg from the second link you gave? [03:24] <c_14> ye [03:26] <haithere> dang... he has 64bit version there, I need the 32 one [03:28] <c_14> What distribution is that server running? [03:28] <c_14> You might want to compile from source. [03:28] <c_14> You won't need root and you can compile for 32bit [03:29] <c_14> https://trac.ffmpeg.org/wiki/CompilationGuide/Generic [03:30] <haithere> well, that's a disappointment. the server is RHEL. will ffmpeg even work on it? [03:31] <c_14> You should be able to compile from source as long as the basic build tools are installed: https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [03:31] <c_14> That one is more specific to rhel than the generic. [03:34] <haithere> guess I should ignore the first "get the dependencies" command, since I don't have root permissions. I'll just start from "Compilation and installation"? [03:35] <c_14> probably a good place to start, if those programs aren't there you might need to cross-compile [03:38] <haithere> "./configure --prefix="$HOME/ffmpeg_build" --bindir="$HOME/bin" gave me an error, though it did not stop the process. [03:38] <haithere> configure: WARNING: Python not found [03:38] <haithere> configure: WARNING: Could not find a suitable version of Python [03:39] <haithere> is Python necessary? [03:39] <c_14> Don't think so. [03:39] <c_14> Then again, never had a computer without python... [03:39] <sacarasc> I didn't know ffmpeg needed python for anything. [03:39] <c_14> If it's not listed on the wiki it shouldn't be required. [03:39] <c_14> sacarasc: Me neither... [03:42] <haithere> "git clone --depth 1 git://git.videolan.org/x264" gives me: [03:42] <haithere> "-bash: git: command not found" [03:42] <haithere> so if I need x264 I need git on the machine? [03:42] <haithere> can't I get it with wget or curl? [03:43] <sacarasc> Go to the x264 site and see if there's a daily snapshot. [03:43] <sacarasc> That should do it. [03:43] <c_14> ftp://ftp.videolan.org/pub/videolan/x264/snapshots/ [03:45] <haithere> "wget ftp://ftp.videolan.org/pub/videolan/x264/snapshots/last_stable_x264.tar.bz2" should do it then. [03:46] <sacarasc> More like ftp://ftp.videolan.org/pub/videolan/x264/snapshots/x264-snapshot-20050824-2… [03:50] <haithere> 2005 is a bit old I guess [03:50] <sacarasc> Oh, I clicked the wrong one/. :D [03:51] <sacarasc> The one you picked is probably okay though. SOrry. [03:51] Action: sacarasc goes to bed as it's almost 03:00. [03:51] <haithere> np np :) [03:51] <haithere> haha [03:51] <haithere> it's 4:51 here [03:51] <haithere> am [03:51] <haithere> should go to bed too :D [03:51] <c_14> Night Owls unite. [03:52] <haithere> ok, I downloaded this tar.bz2 x264 file. googled how to extract it, did this: "bzip2 -d last_stable_x264.tar.bz2" [03:53] <c_14> tar -xjvf $file [03:53] <haithere> this changed the file to "last_stable_x264.tar" [03:53] <haithere> aa [03:53] <c_14> If you already bzip2 -d 'd it you can probably just tar -xvf $file [04:03] <haithere> got libfdk_aac from sourceforge: [04:03] <haithere> "wget http://downloads.sourceforge.net/project/opencore-amr/fdk-aac/fdk-aac-0.1.3…" [04:03] <haithere> worked perfectly [04:03] <haithere> but now I got an error [04:03] <haithere> "autoreconf -fiv" gives me: Makefile.am: C objects in subdir but `AM_PROG_CC_C_O' not in `configure.ac' [04:04] <haithere> autoreconf: automake failed with exit status: 1 [04:06] <haithere> well I'm too lazy to fix this.. not going to use aac at the moment any way [04:13] <haithere> I think I don't need libogg, libvorbis etc as well [04:13] <haithere> here's what I'm trying to achieve: http://pastebin.com/ncDUe1jM [04:14] <haithere> what encoders do I need? [04:15] <haithere> I did install libmp3lame from the tutorial, so audio is done? [04:15] <haithere> c_14 ? :) [04:17] <c_14> I don't think ffmpeg uses external libs for mpeg2video or mp2 [04:18] <haithere> ohh.. all this hassle :( [04:18] <haithere> could've just done ffmpeg straight away [04:19] <c_14> Just use the configure flags in the tutorial with exception of everything after nonfree [04:22] <haithere> used "wget http://ffmpeg.gusari.org/static/32bit/ffmpeg.static.32bit.latest.tar.gz" instead of "git clone --depth 1 git://source.ffmpeg.org/ffmpeg" [04:22] <haithere> I extracted it [04:22] <sacarasc> The former is not source, but a compiled version. [04:22] <haithere> tutorial tells me to: cd ffmpeg [04:22] <haithere> aa [04:23] <c_14> https://ffmpeg.org/releases/ffmpeg-snapshot.tar.bz2 [04:27] <haithere> did this: ./configure --prefix="$HOME/ffmpeg_build" --extra-cflags="-I$HOME/ffmpeg_build/include" --extra-ldflags="-L$HOME/ffmpeg_build/lib" --bindir="$HOME/bin" --extra-libs=-ldl --enable-gpl [04:28] <haithere> did ok [04:28] <haithere> and then this: make [04:28] <haithere> and this gives me: common.mak:18: *** unterminated call to function `foreach': missing `)'. Stop. [04:30] <haithere> googled it. might be that my version of "make" is too old [04:31] <haithere> whatever that "make" is :D [04:31] <c_14> It works for me, so I'm guessing that's it. [04:33] <haithere> trying this workaround: "./configure --prefix="$HOME/ffmpeg_build" --extra-cflags="-I$HOME/ffmpeg_build/include" --extra-ldflags="-L$HOME/ffmpeg_build/lib" --bindir="$HOME/bin" --extra-libs=-ldl --enable-gpl --disable-indevs --disable-outdevs" [04:34] <haithere> dang... still the same error [04:40] <c_14> hmm, you might have to cross-compile/build a static version yourself on a different machine [04:44] <haithere> what would be the requirements? I can use my Windows and Debian machine. or do I need RHEL? [04:45] <c_14> I'd probably use the debian. Is it 32 bit? [04:45] <haithere> nope, but I can install it as virtual machine in Virtualbox [04:53] <c_14> Then try compiling ffmpeg in a 32bit environment with this configure http://sprunge.us/TBig [04:53] <c_14> should hopefully work [08:02] <ParkerR> c_14, kinda unrelated but: is a staticly linked ffmpeg all compiled into the main binary? May be a dumb question but Ive always kinda wondered that for something like toolkit where you can compile one binary with all the featrues and codecs [08:02] <ParkerR> *statically [08:02] <ParkerR> "Static builds provide one self-contained .exe file for each program " oh neat [08:05] <ParkerR> Obviously the downside beng having to keep it up to date yourself [08:05] <ParkerR> If Im reading correctly [10:18] <sruz25> so I have a video from camera which is 1440x1080 16:9, I couldn't find any info about what to do with it. Can you point me at something, or do I just recode it with chosen resolution? [10:20] <pzich> by default it uses the resolution of the input file/stream/device, did you want to scale it? [10:21] <sruz25> I want it to use square pixels and it would be probably best to downscale it to 720p, since the quality isn't great anyway [10:22] <pzich> You're saying the footage is 1440x1080, but scaled to 16:9? [10:22] <pzich> instead of 1920x1080, which is true 16:9? [10:22] <sruz25> the pixels are apparently somehow not square [10:23] <pzich> ok, well if it should be 16:9, you can just add -vf scale=1920:1080, or 1280:720 if you want 720p [10:23] <pzich> it should stretch it [10:23] <sruz25> ok, thanks [10:24] <pzich> np, I'm going to sleep now, but if you still need help, post it here and maybe someone else can help, or I can later [11:47] <Mavrik> sruz25, that's just anamorphic video :) [11:48] <sruz25> hmm, might help with googleing, if it's official term [11:48] <Mavrik> anyway, what do you want to do with it? [11:48] <Mavrik> I mean, do you have problem playing it? :) [11:48] <Mavrik> (sorry, just came) [11:50] <sruz25> just wanted to recode into something more usable. As it is now, the PC isn't able to play it smoothly and the bitrate is rather big [11:50] <Mavrik> ah, can you run ffprobe on it and paste it into a pastebin? [11:51] <Mavrik> anamorphic video is video that's stored (encoded) in a different resolution than what it should display as, it usually has a flag in it to tell player how to stretch it :) [11:51] <Mavrik> ffmpeg will keep anamorphic video like that unless you explicitly rescale it [11:51] <sruz25> Duration: 00:23:16.80, start: 2236.323122, bitrate: 7338 kb/s Stream #0.0[0x1011]: Video: h264 (High), yuv420p, 1440x1080 [PAR 4:3 DAR 16:9], 50 fps, 50 tbr, 90k tbn, 50 tbc [11:52] <sruz25> using scale filter to classic 720p should work then... [11:53] <Mavrik> sruz25, see, it has a "PAR" (pixel aspect ratio) of 4:3 [11:53] <Mavrik> 1:1 would mean square pixels [11:53] <Mavrik> 4/3 * 1440 = 1920 :) [11:55] <Mavrik> so if you want to fix that you'd do something like [11:55] <Mavrik> scale=1920:720,setsar=1:1 [11:55] <Mavrik> for video filter :) [11:55] <Mavrik> sorry, scale=1920:1080 of course :) [11:55] <Mavrik> that should give you a video that's 1920x1080 [PAR 1:1 DAR 16:9] [11:56] <JEEB> also funny how setsar uses the more official SAR naming, while the info output says it's PAR (which is AFAIK only used in MPEG-4 Part 2 and Part 12) [11:56] <JEEB> sample versus pixel [11:56] <JEEB> meaning is the same though, the aspect ratio of a single entity [11:56] <sruz25> oh, so if I did jsut scale=1920x1080 (or other 16:9 ratio) it wouldn't work properly? [11:56] <JEEB> sample is just more generic :) [11:56] <JEEB> nah, it would generally become 1:1 there too [11:57] <Mavrik> hmm [11:57] <JEEB> it's just that in some cases when you resize the scaler doesn't notice that it's exactly 1:1 [11:57] <Mavrik> sruz25, I don't really know by heart what scale filter does by default :) [11:57] <Mavrik> should look at doc [11:57] <sruz25> I'll use it, just to be safe [11:57] <JEEB> it becomes a thing esp. when downscaling and the resolution because of mod2 etc becomes not exactly what the DAR was [11:58] <Mavrik> yeah, you get those wierd SAR values like 4866:4865 and stuff :D [11:58] <Mavrik> also, the fact that scale/ffmpeg still can't handle rounding to what codec expects annoys me to no end [11:59] <Mavrik> about the PAR... I think PAR is just generally more "clear" term to people not really into encoding :) [12:00] <JEEB> still doesn't change the fact that it hasn't been used since MPEG-4's first specs :) [12:00] <Mavrik> true [12:01] <Mavrik> it's especially more confusing when documentation/filters have SAR everywhere except the probe output :D [15:14] <sruz25> is it possible to use encoded part, if the encoding process was interrupted with "Write failed: Broken pipe0 size= 214830kB time=00:08:52.89 bitrate=3302.5kbits/s"? [15:16] <klaxa> just try to play it back [15:16] <klaxa> depending on video codec and container format it may be valid [15:34] <haithere> hi c_14, you there? [15:34] <c_14> Hmm, ye. What's up? [15:38] <haithere> well, I was the one who tried to get ffmpeg on RHEL. we stopped with the conclusion, that I should cross compile/[another option here], since I had problems with the "make" command with ffmpeg. so you told me to get debian 32 bit, which I got and installed. [15:39] <haithere> so where should I start to get ffmpeg on RHEL up and running? :) [15:39] <haithere> remote RHEL [15:41] <JEEB> uhh [15:41] <JEEB> cross-compiling from a completely different distro sounds like trouble [15:41] <JEEB> can you share some info on your issue? [15:42] <JEEB> preferably in a pastebin or something [15:42] <haithere> uhhh... I'm wondering if this channel has history or anything [15:42] <haithere> well, I have a remote RHEL, on which I'd like to use ffmpeg. I don't have root permissions on it [15:43] <haithere> I tried this tutorial: https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [15:43] <haithere> the RHEL is 32 bit btw [15:43] <JEEB> yeah, unless your system is somehow locked down, it should be possible [15:43] <JEEB> some people do lock down their servers, of course [15:44] <c_14> http://sprunge.us/jdHJ <- here's a paste of our conversation up to this point in case you're interested [15:45] <JEEB> I'm quickly scrolling through, but did you try and get a basic ffmpeg built? [15:45] <JEEB> as in, just configured without any options [15:45] <JEEB> and then it should give you a basic ffmpeg binary [15:46] <JEEB> after that you are supposed to move further [15:46] <haithere> guess that what I should need, since all I need is mpegvideo and mp2 audio [15:46] <JEEB> oh [15:46] <JEEB> yeah [15:46] <haithere> and as I understand, they don't need external libraries [15:46] <JEEB> yup [15:46] <c_14> The only flags we used were prefix cflags, ldflags, bindir, extra-libs and enable-gpl [15:46] <JEEB> and I think both of those don't even need you to enable GPL [15:47] <JEEB> haithere, anyways, go grab a tarball of current ffmpeg sources [15:47] <JEEB> then extract it and run ./configure there without any parameters [15:47] <JEEB> let's see what it says [15:47] <JEEB> pastebin the output, basically [15:47] <JEEB> and link here [15:48] <JEEB> (since you seem to lack git :P ) [15:48] <haithere> ok. "wget http://ffmpeg.org/releases/ffmpeg-snapshot.tar.bz2" [15:48] <JEEB> or you could just get a tarball from the latest revision of the git repo [15:48] <JEEB> http://git.videolan.org/?p=ffmpeg.git [15:48] <JEEB> see the URLs for the "snapshot" for every revision [15:48] <JEEB> but yeah, both should be fine [15:50] <JEEB> (also you'll probably need a recent enough yasm, but we'll see about that after we see how far the configure script goes) [15:50] <haithere> those two are necessary too? PKG_CONFIG_PATH="$HOME/ffmpeg_build/lib/pkgconfig" [15:50] <haithere> export PKG_CONFIG_PATH [15:50] <JEEB> uhh [15:50] <JEEB> you don't need ANYTHING [15:50] <haithere> aa [15:50] <haithere> ok [15:50] <JEEB> anything else but just the source code [15:50] <JEEB> and run ./configure in the source dir without any parameters [15:50] <haithere> so download, extract and this ./configure [15:51] <JEEB> dot-slash pretty much is a shorthand for "in this current directory, execute something named "configure"" [15:51] <haithere> haaaa [15:51] <haithere> yasm/nasm not found or too old. Use --disable-yasm for a crippled build. [15:51] <JEEB> okies [15:51] <haithere> well well [15:52] <JEEB> http://www.tortall.net/projects/yasm/releases/yasm-1.2.0.tar.gz [15:52] <JEEB> extract, configure similarly, but add a --prefix to point at your home dir's somewhere, I usually use /home/username/ownapps/installed_things_name (f.ex. /home/jeeb/ownapps/yasm ) [15:53] <haithere> this link has instructions for yasm. " Yasm is an assembler used by x264 and FFmpeg." https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [15:53] <JEEB> so ./configure --prefix=/your/prefix [15:53] <JEEB> then you make and then you make install, which copies things where they are supposed to go into your set prefix [15:54] <JEEB> after that you can add the bin directory of your prefix to your PATH. PATH is a list of directories that your shell looks for things to run for (which is why you need to actually say "here" for the configure script) [15:55] <JEEB> which would then be something alike to "export PATH=${PATH}:/your/set/prefix/bin" [15:55] <JEEB> which then adds that /your/set/prefix/bin into the list of directories [15:55] <JEEB> (and without the "", btw) [15:56] <JEEB> after that you should have "yasm" as a command available [15:56] <JEEB> since binaries of things get installed to bin/ of the prefix you set when you configured [15:57] <JEEB> (you can later add this to a file executed every time you log in so that you will always have that directory in PATH for that user) [16:00] <haithere> ok, yasm is installed [16:00] <JEEB> how far did you get? [16:00] <JEEB> as in, did you just make install or did you test to run it after adding the directory to your PATH? [16:02] <haithere> how could I test it? if I type "yasm", it tells me: "yasm: No input files specified" [16:03] <haithere> but I did "./configure" in ffmpeg folder [16:03] <haithere> and did not get any errors [16:03] <haithere> it gave me a list of what's enabled [16:03] <haithere> then in the last line: "Creating config.mak, config.h, and doc/config.texi..." [16:04] <haithere> so when we began it said it did not find yasm, I guess not it does [16:04] <haithere> now it does* [16:05] <JEEB> "yasm: No input files specified" <- that already tells you that it found it ;) [16:06] <JEEB> since it's yasm that's outputting that message [16:06] <JEEB> I should have probably said yasm --version [16:06] <JEEB> which would have output 1.2.0 or so [16:06] <JEEB> :) [16:06] <haithere> jop [16:06] <JEEB> and yeah, sounds like you got ffmpeg configured [16:06] <JEEB> now you can try running 'make' [16:06] <JEEB> and see if anything fails [16:07] <haithere> oh ffs. "common.mak:18: *** unterminated call to function `foreach': missing `)'. Stop." [16:08] <JEEB> if you actually got configure to finish nicely, then I guess you've hit a bug in an old version of make [16:08] <JEEB> or well, lack of support for something [16:09] <haithere> if it really is an issue of old version of make, could I update it? [16:09] <JEEB> yes [16:09] <haithere> without root of course [16:09] <JEEB> as you can see, unless specific care has been given to setting up the system, as long as you can make a binary, you can also call it [16:09] <JEEB> as you just did with yasm [16:10] <JEEB> http://mirrors.kernel.org/gnu/make/ [16:10] <JEEB> grab a tarball, configure with a prefix and off you go ;) [16:10] <JEEB> also holy crap [16:10] <JEEB> make 4.0 O_O [16:10] <haithere> I'm wondering what my version of make is [16:10] <haithere> what's yours? [16:10] <JEEB> make --version [16:11] <haithere> 3.80. is it old? [16:11] <haithere> 2002 [16:11] <haithere> guess it is :D [16:11] <JEEB> yes :D [16:12] <JEEB> funny, my new ubuntu box still has 3.81, I wonder if it's just patched to hel [16:12] <JEEB> *hell [16:12] <JEEB> (debian and ubuntu love to keep old versions just with a lot of patches applied) [16:13] <JEEB> haithere, also btw, since you already have something called 'make' in your PATH, you will either have to add the prefix before your currently already-in-PATH things, or you can just call it with the full path to the binary :) [16:14] <JEEB> former can be done with "export PATH=/your/prefix/bin:${PATH}", and latter can just be achieved by doing /your/prefix/bin/make instead of just make in the ffmpeg source code directory [16:19] <haithere> ok, I'm so confused at the moment [16:19] <haithere> I downloaded and extracted make [16:20] <JEEB> ok [16:20] <haithere> but this whole path stuff is confusing [16:20] <haithere> what do I do after extracting ma [16:20] <haithere> make [16:20] <haithere> go to the make folder and "./configure"? [16:21] <JEEB> do note that I mean different things when I use the word PATH and the word path, PATH is the system variable that your shell uses (a list of directories) to find executables. a path is just the path to a directory (f.ex. /hurr/durr/meh) [16:21] <JEEB> yes, with a prefix preferably [16:22] <haithere> from the ffmpeg tutorial, an example: "./configure --prefix="$HOME/ffmpeg_build" --bindir="$HOME/bin" [16:23] <JEEB> you only need to set --prefix, bindir is basically when you want to override the place the executables go into [16:24] <JEEB> in general you have all the directories under your prefix (bin, lib, include f.ex. [the binaries (executables), the libraries and the headers]), but in that case it overrides the place where the executables get copied [16:24] <haithere> so after prefix, I'll just put make-4.0 folder path? something like this? "./configure --prefix="$HOME/ffmpeg_sources/make-4.0" [16:25] <JEEB> uhh, whatever you use as a prefix :P [16:25] <JEEB> some people like to separate things, some people put all into one prefix [16:25] <JEEB> mine was just one example [16:26] <haithere> the tutorial, starting from "Yasm is an assembler used by x264 and FFmpeg." was easy to follow: https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [16:26] <haithere> just copy-pasted commands and yasm worked [16:27] <haithere> but now I'm sooo confused [16:27] <JEEB> yes, you actually have to think [16:27] <JEEB> how terrible [16:27] <JEEB> </sarcasm> [16:27] <JEEB> anyways, you're making it be harder than it is [16:28] <JEEB> prefix is the place where things get copied into, the "root" directory, into which things get "installed" (copied) into. [16:28] <JEEB> by default it tends to be something along the lines of /usr/local [16:28] <JEEB> which is generally root-only :P [16:28] <JEEB> (for writing) [16:28] <JEEB> so you have to override that [16:29] <JEEB> anyways, you configure make in a similar vein as you seemingly have done with yasm [16:29] <JEEB> make [16:29] <JEEB> and then make install [16:29] <JEEB> make sure everything went fine [16:29] <JEEB> after that you can just call the make with its full path [16:30] <haithere> aaaa, prefix is for specifying the installation directory? [16:30] <JEEB> basically yes [16:30] <haithere> lol [16:30] <JEEB> <JEEB> prefix is the place where things get copied into, the "root" directory, into which things get "installed" (copied) into. [16:30] <JEEB> <JEEB> in general you have all the directories under your prefix (bin, lib, include f.ex. [the binaries (executables), the libraries and the headers]), but in that case it overrides the place where the executables get copied [16:30] <another> make: sure everything went fine is not a target [16:31] <JEEB> also as I explained, the other setting that gets set overrides the place where the executables get placed [16:31] <JEEB> usually a directory called "bin" is made under the prefix [16:31] <JEEB> (or used if already existing) [16:31] <haithere> so that'd be logical: "./configure --prefix="$HOME/make_build" [16:31] <JEEB> and --bindir overrides that [16:32] <JEEB> well, there's no convention, so there's nothing logical or not logical :P [16:32] <JEEB> there are just conventions that some people use [16:32] <haithere> ok, configured [16:32] <haithere> now what? :) [16:32] <JEEB> you compile it with make? [16:32] <JEEB> and then if that went fine [16:32] <haithere> compile make with make? :D [16:32] <JEEB> well make just calls the compiler and other things [16:33] <haithere> ok [16:33] <JEEB> and as long as the Makefile that gives the orders is supported [16:33] <JEEB> it should go just fine :) [16:33] <JEEB> s/gives/contains/ [16:33] <JEEB> and then if the compilation went fine, you copy (install) things into the prefix you set with make install [16:33] <JEEB> after that you can call /home/your_username/make_build/bin/make instead of make [16:34] <JEEB> in ffmpeg's directory [16:34] <JEEB> and compile yourself a ffmpeg [16:35] <haithere> ok, "make" did ok [16:35] <haithere> ummm [16:35] <haithere> should I "make install" too? [16:35] <JEEB> <JEEB> and then if the compilation went fine, you copy (install) things into the prefix you set with make install [16:35] <JEEB> actually read what I write, if only possible ;) [16:40] Action: sacarasc waves at Rathann. [16:41] <haithere> great... I don't have "make_build" folder in my username folder. thought this command would create the folder: "./configure --prefix="$HOME/make_build", because ffmpeg_build folder got there itself, I did not create it [17:02] <JEEB> haithere, pastebin the output of what you've done lately? [17:03] <JEEB> also ffmpeg_build was indeed created by another thing, because you set that directory as the prefix :P [17:03] <JEEB> in other words, you did indeed create it, but not directly [17:03] <haithere> well, then this command: "./configure --prefix="$HOME/make_build" should've made the "make_build" folder as well [17:03] <JEEB> yes [17:04] <JEEB> thus I asked for your pastebin [17:04] <haithere> well, it didn't [17:04] <JEEB> of the terminal logs [17:05] <haithere> would, if I could, but putty (windows client for ssh connections) only displays last 100 (?) lines? [17:06] <JEEB> you should be able to scroll those unless you set your backlog to be retardedly small :P [17:07] <haithere> haven't changed any settings regarding this, so this should be default [17:07] <JEEB> I have no idea about putty's defaults but it should have a backlog [17:07] <JEEB> unless you're in a screen [17:08] <JEEB> screen is a bit harder to deal with regarding scrolling (in other ways, simpler) [17:10] <another> haithere: putty has an option for backlog [17:10] <another> not sure where. maybe somewhere under window? [17:29] <haithere> how should I Google it, if I want to update "make"? [17:29] <haithere> "update gnu make" I guess [17:30] <JEEB> you just did that :P [17:30] <JEEB> and please post as much as you can [17:30] <JEEB> in a pastebin [17:30] <JEEB> among useful things is the config.log file under make's source directory [17:31] <JEEB> which will tell me among other things what was your full configuration in case you failed at something there [17:32] <haithere> well, I couldn't do it correctly [17:33] <JEEB> and I'm trying to goddamn help you in finding out what went wrong [17:35] <haithere> jep, config.log might help you [17:51] <haithere> http://pastebin.com/GwTsHAZh [17:56] <JEEB> this actually seems fine'ish [17:56] <JEEB> so you ran make and that finished successfully? [17:56] <JEEB> and then make install [17:56] <JEEB> and that finished successfully as well? [17:58] <haithere> in the "make-4.0" source folder? [17:58] <JEEB> in its source directory, yes [17:58] <JEEB> where the configure script is [18:03] <haithere> seems like make_build folder appeared in the home folder [18:04] <haithere> now I should go to ffmpeg's source folder and then do this: "/home2/r/robert/make_build/bin/make" [18:05] <haithere> coz this is where the new 4.0 version is [18:06] <JEEB> yes [18:06] <JEEB> without the "" of course [18:06] <JEEB> it should call the new make there [18:07] <haithere> well, it's doing something [18:07] <haithere> did not get the error I did before [18:07] <JEEB> well there could be two problems still [18:07] <JEEB> 1) make ended up being too new (quite improbable I guess) [18:07] <JEEB> 2) the compiler you have is too old (quite possible) [18:08] <haithere> noooooo, I don't want any problems :D [18:08] <JEEB> and to be honest if it's 2) , I don't want to go through installing a whole toolchain so my free help ends there [18:11] <haithere> hmm, it "make"-s quite a long [18:11] <haithere> how long does the process usually take? [18:12] <JEEB> depends on your machine, but there's plenty to compile [18:12] <JEEB> so it'll take a while [18:21] <haithere> ok, "make" is done [18:21] <haithere> now I should do "/home2/r/robert/make_build/bin/make install"? [18:27] <vklimkov_> hi folks. where i can found source example of good video player based on ffmpeg? [18:28] <JEEB> haithere, no - you didn't set a prefix when configuring [18:28] <JEEB> you can use the ffmpeg binary that is now in the ffmpeg source directory [18:29] <JEEB> dot-slash ffmpeg should show you the version info [18:29] <haithere> ffmpeg version 2.2.git Copyright (c) 2000-2014 the FFmpeg developers [18:30] <haithere> built on May 24 2014 19:17:22 with gcc 3.4.6 (GCC) 20060404 (Red Hat 3.4.6-11) [18:30] <haithere> yaaay [18:30] <JEEB> congrats [18:30] <JEEB> and yes, lol old [18:30] <JEEB> at least it compiled [18:30] <haithere> not sure if it was that difficult. but I hadn't done this before [18:35] <JEEB> haithere, now you can either call it by its full path, or put it somewhere and make that somewhere be in your PATH so you can call it with just 'ffmpeg' [18:52] <haithere> jep, seems like everything's working as I wanted. haven't even said thanks for wasting your time, hehe. so, thanks JEEB and c_14 ! [19:34] <Logicgate> i need help from the pros [19:34] <Logicgate> -c:v libfaac -ac 2 -deinterlace -c:v libx264 -preset:v slow -profile:v baseline -level 3 -maxrate 10000000 -bufsize 10000000 -b:v 1200k -f mp4 -threads 0 -strict experimental -movflags faststart -vsync 2 [19:35] <Logicgate> that's what I'm using at the moment for my video uploading function on my website [19:35] <Logicgate> the videos uploaded are 7 seconds [19:35] <Logicgate> I need to lower the quality of the uploaded video to what would be the quality of an iPhone 5's recorded video [22:04] <bencc2> is it possible to sharpen text in video? [22:07] <klaxa> http://www.ffmpeg.org/ffmpeg-filters.html#unsharp [22:07] <klaxa> apparently also does sharpening [22:08] <Logicgate> anyone around [22:08] <Logicgate> http://puu.sh/8ZwQk.png [22:08] <Logicgate> http://puu.sh/8ZwRO.png [22:08] <Logicgate> 2 videos, recorded with the same phone [22:08] <Logicgate> Why is the video bitrate different? [22:09] <Logicgate> How can I re-encode incoming videos to mimmick the properties of those 2 videos? [22:10] <klaxa> as to why one video has a higher bitrate, try recording pitch black with the same phone and watch the bitrate drop even further [22:10] <klaxa> depending on the content, it can be compressed better or worse [22:11] <Logicgate> I see [22:11] <Logicgate> klaxa, are you familiar with Vine? [22:11] <bencc2> klaxa: I'll try it. thanks [22:12] <klaxa> Logicgate: no [22:12] <Logicgate> I made a website which basically enables users to upload 6.8 second videos to Vine [22:12] <Logicgate> The biggest problem is that videos come in all different containers, video bitrates, audio bitrates, width, height, codecs, etc. [22:13] <Logicgate> Right now what I'm doing is taking the video uploaded, extracting the video stream, the audio stream, and glueing them back together with FFMPEG for a final result in an mp4 [22:13] <Logicgate> -c:v libfaac -ac 2 -deinterlace -c:v libx264 -preset:v slow -profile:v baseline -level 3 -maxrate 10000000 -bufsize 10000000 -b:v 1200k -f mp4 -threads 0 -strict experimental -movflags faststart -vsync 2 [22:13] <Logicgate> those are my current settings [22:14] <Logicgate> I would like to basically re-encode the videos to match what would normally be recorded with an iPhone 5 within the app [22:15] <klaxa> you might want to have a look at this: https://trac.ffmpeg.org/wiki/x264EncodingGuide#Compatibility [22:16] <Logicgate> what seems to be the output of the Vine application is 480x480 (yes, 1:1 ratio, weird), 64kpbs 44.1hz 2 Channels AAC Audio with 29fps AVC Level 3 video [22:16] <Logicgate> Baseline 3 seems to be what I need, since we need reverse compatibility for all devices [22:17] <Logicgate> I just want to know if I'm on the right path here or not. [22:17] <Logicgate> How would you change my existing command line [22:18] <klaxa> hmm... maybe use -crf instead of -b:v [22:18] <Logicgate> crf 0? [22:19] <Logicgate> is the vsync important? [22:20] <Logicgate> what about maxrate, that's the max bitrate? [22:21] <Logicgate> Will downsampling the audio cause major problems? [22:37] <klaxa> Logicgate: crf 0 would be lossless, you don't want that, maxrate is max bitrate afaik, i don't even know what vsync does, downsampling shouldn't cause problems [22:54] <Logicgate> right [22:54] <Logicgate> why wouldn't I want lossless [22:55] <Logicgate> i do want to preserve the quality of the uploaded video [22:55] <Logicgate> it's only 6 seconds [22:55] <Logicgate> not gonna take hours [22:55] <c_14> -crf 18 is usually considered visually lossless. Going full lossless usually just bloats filesize. [23:05] <Logicgate> thanks c_14 [23:19] <bencc2> I'm trying to understand what parameters to use with unsharp http://www.ffmpeg.org/ffmpeg-filters.html#unsharp [23:20] <bencc2> I have a video export from a powerpoint presentation and I want to improve the text quality [23:20] <bencc2> this gives nice result but I want to make it a bit better: ffmpeg.exe -i before.mp4 -filter:v "unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5" after.mp4 [23:21] <bencc2> is there an explnation on the params? [23:34] <eris0xff> hi [23:37] <eris0xff> quick question on the proper way to generate pts for a raw x264 stream at 23.976 FPS while muxing with audio. i've tried doing the copy video codec, but that really ignores every attempt at modification (no matter how minor). So I guess using a setpts filter would do the trick. Do you simply increment the current PTS value by the number of PTS ticks in your frame rate (ie: 1/23.976 * 1/90000)? [23:40] <eris0xff> oh well guess no one out there. i'll try the mailing list [00:00] --- Sun May 25 2014

1 0

ffmpeg-devel.log.20140523
by burek 24 May '14

24 May '14

[00:00] <jamrial> haha, i see [01:06] <jamrial> kurosu: i can send a 3dnow version of vector_clipf to make sure there are no yawns since everybody will be busy laughing :P [01:07] <iive> I won't. [01:40] <cone-833> ffmpeg.git 03James Almer 07master:1d36defe94c7: x86/dsputil: port ff_vector_clipf_sse to yasm [01:41] <Daemon404> michaelni, ping [01:46] <michaelni> Daemon404, pong [01:47] <Daemon404> michaelni, http://git.videolan.org/?p=ffmpeg.git;a=commit;h=20fa3fb93d0f3d3eab2b1f63a0… [01:47] <Daemon404> this is causing sigabort on some files [01:47] <Daemon404> apparently [01:48] <michaelni> where can i find these files ? [01:48] <Daemon404> let me see... [01:51] <Daemon404> PM. [01:58] <michaelni> and how do i reproduce it with that file ? (the sig would be from the muxer id assume) [01:59] <Daemon404> its something to do with copying chapters [01:59] Action: Daemon404 checks [01:59] <Daemon404> (-map_chapters -1 'fixes' it for me) [02:02] Action: Daemon404 greps logs [02:06] <Daemon404> michaelni, [02:06] <Daemon404> ffmpeg -i a.mov -vn -strict -2 a.mp4 [02:06] <Daemon404> ^ [02:06] <Daemon404> doesnt need to be aac obviously [04:15] <BBB> kurosu: jamrial: can I please say how happy I am with your yams stuff? it's awesome, thank you [04:16] <BBB> in general, any line of inline non-inlined asm converted to yasm brings a smile on my face, but when you add sse2/ssse3 and make it 2x as fast, I'm just happy [04:17] <jamrial> you're welcome :P [05:10] <cone-402> ffmpeg.git 03Michael Niedermayer 07master:744df0a0319c: avformat/movenc: Check that packet duration is valid in ff_mov_write_packet() [05:10] <cone-402> ffmpeg.git 03Michael Niedermayer 07master:220df143b962: avformat/utils: Check end time in avpriv_new_chapter() [05:10] <cone-402> ffmpeg.git 03Michael Niedermayer 07master:64bf8455442b: avformat/mov: Dont blindly trust the stream duration in seting chapter times [05:11] <michaelni> Daemon404, should be fixed [06:51] <jamrial> nevcairiel: can you test this patch http://pastie.org/9201161 and see if it fixes the fate failures on msvc 2013 x86_64? [08:25] <kurosu> jamrial, but can you test that 3dnow patch ? ;) [08:25] <kurosu> bye [08:44] <ubitux> why is Don Moir arguying on such a meaningless issue? :D [08:44] <ubitux> (re ffplay rotated video) [15:43] <Daemon404> michaelni, cool [15:45] <cone-556> ffmpeg.git 03James Almer 07master:6a4832caaede: x86/diracdsp: mark all functions as yasm [17:55] <nevcairiel> jamrial: your patch didnt work [17:55] <jamrial> wonder why it's failing only with msvc and not mingw64 then... [17:59] <nevcairiel> how big is len usually? it seemed to have a rather astronomically large value [18:00] <nevcairiel> i can debug in a bit [18:01] <jamrial> ok, thanks [18:14] <nevcairiel> jamrial: the "len" it reads into a register is 0x0000000900000040 ... somehow I think this should just be 0x40 without the 9 in the middle? needs register expansion from int to 64-bit? [18:16] <nevcairiel> (also, only with your patch applied, otherwise it assigns the wrong arg to len in general) [18:20] <jamrial> a disass of the old inline version with mingw64 shows a cltq, so i guess it's what you said [18:21] <nevcairiel> or load it manually from the stack only reading 32-bit, instead of letting the macros generate it [18:30] <nevcairiel> adding a movsxd lenq, lend fixes the issue in any case [18:39] <nevcairiel> is that a reliable rule that using any operation on the 32-bit part of a register clears out its upper half? [18:43] <nevcairiel> (practically, if i change the "shl" in that function to use lend instead of lenq, is it guaranteed to empty the upper half?) [18:55] <jamrial> no idea [18:55] <jamrial> does using "shl lend, 2" without the movsxd fix the failure there? [18:55] <nevcairiel> yea, thats why I ask [18:55] <jamrial> it at least doesn't break anything here [18:55] <nevcairiel> on top of your earlier patch, that is [18:56] <nevcairiel> but yes, it works just fine with that change [18:56] <jamrial> ok, guess then i'll go with that. better than adding an extra mov [18:56] <nevcairiel> too bad the cglobal macro doesn't make manual loading of args easier [18:56] <nevcairiel> then you could just load 32-bit [18:58] <jamrial> yeah, improving PROLOGUE is marked as TODO in x86inc [20:45] <cone-205> ffmpeg.git 03Niels Möller 07master:6b7b8585d735: dca: Convert dca_dmixtable to integers [20:45] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:c867be03f61e: Merge commit '6b7b8585d73548f08b98959df39d80b97677662f' [20:45] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:4e4dbb99831d: dcadec: fix xxch_dmix_coeff and xxch_dmix_sf after merge [21:13] <cone-205> ffmpeg.git 03Vittorio Giovara 07master:21f68c2489cb: avcodec: bump version after rotation api [21:13] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:d062db788012: Merge commit '21f68c2489cba2a1a4a41d0c5c828266e6162800' [21:21] <cone-205> ffmpeg.git 03Carl Eugen Hoyos 07master:27d460fe9821: Refuse to mux H.264 with fourcc H264 into avi without startcode. [21:21] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:9340ced47b5b: Merge remote-tracking branch 'cehoyos/master' [22:09] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:6da06ef6bbc0: avcodec/dcadec: make dca_dmix_code() 650% faster [22:12] <nevcairiel> are compilers smart enough to replace pow(2, -15) with a constant? that seems kinda too smart for them! [22:13] <thardin> check [22:13] <nevcairiel> looks like gcc does it [22:15] <wm4> why not use a fp constant [22:16] <wm4> 0x1p-15 or so? (not sure if the - goes there) [22:19] <nevcairiel> if it would be a power of ten, that might be easier, but powers of two dont have an easy syntax, so it would be something like 3.0517578125e-5 [22:20] <nevcairiel> (or in hex, if you prefer) [22:20] <wm4> nevcairiel: that is a power of 2 [22:21] <nevcairiel> the hex syntax is power of 2, not 10 like the decimal syntax? [22:22] <wm4> yes, it's power of 2 [22:22] <wm4> the advantage of hex floats are that they can represent floats exactly [22:23] <nevcairiel> its a C99 feature, i wonder if msvc supports that :D [22:24] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:9ccb5455ca8c: avcodec/dcadec: use a constant instead of assuming every compiler can optimize pow(2,-15) [22:24] <nevcairiel> guess we'll find out! [22:25] <nevcairiel> at the very least i expect older msvc to break [22:25] <michaelni> if something breaks ill replace it by a dezimal constant [22:26] <michaelni> that is please ping me if it breaks somewhere [22:27] <nevcairiel> libavcodec/dcadec.c(2110) : error C2059: syntax error : 'bad suffix on number' [22:27] <nevcairiel> well, no luck! [22:28] <wm4> it might actually be in C++11, but yeah, even if MSVC adds it to C, it probably will break with older versions [22:31] <nevcairiel> msvc also doesn't optimize it out, the pow call remains if i put it back [22:33] <nevcairiel> anyhow michaelni, obligatory ping, it does indeed break msvc, even the latest version [22:34] <ubitux> michaelni: don't forget to add a com along the constant though [22:45] <cone-205> ffmpeg.git 03Michael Niedermayer 07master:6b88f22e8953: avcodec/dcadec: dont use hex float, msvc doesnt support it [22:52] <ubitux> so libav is going into the mountain... to reindent code as fast as possible? [23:03] <ubitux> btw, should i apply the ffplay rotation patchset? [23:06] <cone-205> ffmpeg.git 03James Almer 07master:518cbf9b4a0a: x86/dsputil: fix VECTOR_CLIP_INT32 macro [23:07] <michaelni> ubitux, either that or send cuz a mail [23:08] <michaelni> cuS I meant [23:08] <ubitux> too bad he isn't on irc [23:08] <ubitux> i guess i'll ping [23:08] <michaelni> yes, he probably missed the patch or is busy [23:08] <Plorkyeran_> +wm4> it might actually be in C++11 <-- it isn't [23:09] <Plorkyeran_> c++11 only imported the c99 standard library [23:09] <Plorkyeran_> and not the language changes [23:11] <wm4> I didn't look, but g++ printed a wanring in c++98 mode, but not c++11 mode [23:30] <cone-205> ffmpeg.git 03James Almer 07master:584327f22fbf: x86/dsputil: fix argument declaration in vector_clipf [23:30] <cone-205> ffmpeg.git 03Clément BSsch 07master:fa972dcf1a36: avutil/display: use hypot(). [23:30] <cone-205> ffmpeg.git 03Clément BSsch 07master:e63a17bfdc99: avformat/mov: use av_display_rotation_get() for rotate metadata. [23:30] <cone-205> ffmpeg.git 03Clément BSsch 07master:08c51e12b1c3: ffplay: support rotated video. [23:34] <Timothy_Gu> michaelni: did you see Diego's patch to Libav about huffyuvdsp [23:35] <Timothy_Gu> Is that the same thing as losslessdsp? [23:46] <michaelni> i think its not exactly the same [00:00] --- Sat May 24 2014

1 0

ffmpeg.log.20140523
by burek 24 May '14

24 May '14

[00:53] <temhawk> how do i prevent the black screen when i cut a video with -codec copy? i need to cut at a keyframe, right? how do i find out what frames are keyframes if my framerate is 29.97002? [01:14] <c_14> It should cut by the nearest keyframe automatically. [01:18] <temhawk> c_14: i think it does but there is no video for a moment when i then play it in VLC [01:22] <temhawk> http://pastie.org/9200663 [01:24] <c_14> How long is the black/does the timestamp advance while the screen is black? [01:26] <temhawk> between 3-4 seconds [01:26] Action: debianuser "solved" the problem with file not detected as 3D - that's a player bug, after RENAMING files around several times all the same files were detected as 3d video. [01:27] <c_14> And the timestamp advances? cc temhawk [01:29] <temhawk> c_14: i don't know what you mean [01:31] <c_14> vlc should tell you at which point in the video you are while playing, are the first three seconds of video literally black or is vlc just showing a black screen for three seconds and then playing [01:32] <temhawk> c_14: ahh. it's not black in vlc. only in quicktime. in vlc it just shows the playlist until it starts after 3-4 seconds. the thingy moves during that time in both VLC and quicktime [01:33] <c_14> thingy? progress bar? [01:34] <temhawk> ye [01:36] <temhawk> i have a gif maker app and when i open the video in it i can jump by frames and when i click "jump to next frame" it goes immediately to where the video starts between 3-4 seconds [01:36] <temhawk> there is audio the whole time btw [01:37] <c_14> Is ffmpeg seeking correctly? ie does the sound start where it's supposed to or does the video start where it's supposed to [01:38] <temhawk> the video starts where it's supposed after 3-4 seconds (i.e. 4 seconds in compared to the original footage) and the audio is sync'd [01:39] <temhawk> it almost seems like the audio was just not cut but the video appears sooner than 4 seconds... [01:40] <c_14> Can you try it again using the trim and atrim filters instead of -ss and -t ? [01:40] <c_14> -vf 'trim=4:19,atrim=4:19' should do it [01:42] <temhawk> i'm using streamcopy [01:42] <c_14> right, i forgot ehhh [01:42] <temhawk> it says i can't use that with streamcopy [01:43] <c_14> can you try using -ss as an input option? [01:46] <temhawk> c_14: yeah, that works! except now the video is 19 seconds and the last 4 seconds is nothing! [01:47] <c_14> adjust -t down by 4 seconds [01:48] <temhawk> now it's 15 seconds but the last 4 seconds are still gone! [01:48] <c_14> whaaaa [01:49] <c_14> The last 4 seconds of audio are missing? [01:49] <temhawk> both video and audio [01:49] <temhawk> there's just a still frame [01:49] <c_14> and with -t 19 the last 4 seconds were empty? [01:50] <temhawk> no, -t 15 was 19 seconds, then i tried -t 11 after you said to reduce it by 4 [01:50] <temhawk> and both have the same problem at the end [01:51] <c_14> you want everything from second 4 to second 19 in the original video? [01:51] <temhawk> yes [01:51] <temhawk> the framerate is 29.97002 [01:52] <c_14> The video stays at a frame but the audio progresses? [01:52] <temhawk> no, no audio [01:52] <c_14> How long is the original video? [01:53] <temhawk> 38s [01:54] <c_14> try using -to 15 and if that doesn't work -to 14 and then -to 13, I'm guessing it might be cutting at vframes somewhere and then just lengthening the video [01:58] <temhawk> c_14: the same problem in all cases. last 4 seconds is no video or audio. -to 15 gives me a 19s video, -to 14 gives me an 18s video, etc [02:02] <c_14> So -to 15 gives you everything you want plus 4 seconds of nothing? [02:04] <temhawk> c_14: yeah! [02:04] <temhawk> c_14: it gives me a little bit extra, actually [02:04] <c_14> I'd probably blame that on the stream copying, the blank seconds are still bothering me a bit though. [02:10] <temhawk> c_14: so i tried trimming the last 4 seconds off with another ffmpeg command and& mhmmmm& now the video starts where the ORIGINAL video starts! [02:10] <c_14> wait, what? sure you grabbed the right file? [02:11] <temhawk> yeah, i'm positive... [02:12] <c_14> I'm relatively certain deleted video shouldn't be coming back from the dead. [02:12] <c_14> Your computer is scaring me. :P [03:36] <BtbN> What's the current situation with libavresample vs. libswresample? Are those compatible? [03:54] <fuff> Has anybody used JPEG2000 with FFMPEG? [03:54] <fuff> I would like to try encoding lossy intra-frame video. Anything but classic JPEG would be good. [04:09] <fuff> wow, libopenjpeg is slow as hell [06:28] <help_me> Is there a lossy intra-frame codec I can use that is actually playable? [06:33] <help_me> I tried libopenjpeg (jpeg2000) on HD video and it could not play back in real-time. I am having trouble finding documentation on this. [07:03] <BtbN> help_me, what about h264? [07:06] <help_me> I did not know that h264 did intra-frame, or that it did it well [07:07] <help_me> I'm leaning towards JPEG2000 because digital cinema packages use it [07:07] <help_me> trouble is I can't get it to play (or encode) very fast [07:07] <help_me> I think I may have it on losslessmode or something but I don't know how to change it [07:08] <help_me> I want to experiment, like I've been doing with lossless codecs, but there doesn't seem to be much information on lossy intra-frame compression with FFMPEG [08:21] <help_me> OK so it's j2k that is lossy, nt libopenjpeg [08:21] <help_me> http://sinclairmediatech.com/use-ffmpeg-to-encode-jpeg2000-270mbit/ [08:31] <help_me> wrong again, it's jpeg2000 (I wish they'd stop renaming things) [09:12] <ericbutters> hello, how do i configure/build libasound? [09:22] <help_me> ericbutters: I can't help you (never compiled an ffmpeg component before) and pretty much nobody is online right now so I'm just hangin out and waiting [09:23] <help_me> I've been trying to get JPEG2000 t work but it looks like the encoders and decoders for it seriously suck [09:23] <help_me> http://dcinemaforum.com/forum/index.php?topic=44.0 [09:24] <help_me> The FOSS people haven't cracked the JPEG2000 thing yet for real-time decoding [09:24] <help_me> pretty sure you need special hardware for it [09:25] <help_me> Motion JPEG seems to run pretty fast, but macroblocking is a nightmare [09:26] <help_me> I'm going to try AVC-intra next, and maybe se if I can get JPEG XR to work [09:26] <help_me> that's about all of the losy intra-frame codecs there are though, so if none of those are satisfactory then I will be very sad [09:32] <help_me> there's also webp I suppose [09:40] <help_me> fuck, webp is too new and was aded less than a week ago [09:40] <help_me> unbelievable [09:53] <relaxed> help_me: my builds have had webp support for a while, http://johnvansickle.com/ffmpeg/ [09:54] <tm512> does -rtbufsize need to be located anywhere special in the command line? [09:54] <tm512> because I'm still getting stuttering in my video [09:55] <tm512> trying to do x11grab [09:58] <help_me> thanks, relaxed [09:58] <relaxed> tm512: did you try before the input? [09:58] <relaxed> tm512: what's your command? [09:58] <help_me> I prefer to use the official builds, but I might give yours a try for testing purposes [10:00] <help_me> I am on Windows though [10:00] <tm512> http://hastebin.com/cozasodezu [10:01] <tm512> it's always had this weird issue of having the video freeze for a bit and then resume [10:02] <relaxed> tm512: does it happen when you encode lossy? [10:02] <tm512> I'm not recording lossy [10:02] <relaxed> no shit, that's why I asked. [10:03] <tm512> sorry, I'm tired and didn't read correctly [10:03] <relaxed> try -crf 20 just to test [10:07] <tm512> doesn't look like it happens [10:07] <tm512> I thought the initial problem was an I/O one [10:07] <tm512> but figured that rtbufsize would allow it to catch up [10:09] <relaxed> see if it happens with -c:v rawvideo -pix_fmt yuv420p [10:09] <tm512> was I mistaken? [10:09] <relaxed> all is fine with -crf 20? [10:10] <tm512> I'm double checking [10:13] <relaxed> -c:v ffvhuff is lossless and might be faster [10:13] <tm512> looks like all is fine but that isn't completely conclusive [10:13] <tm512> why did you have me try -crf 20? [10:14] <relaxed> I figured it would be less intensive than lossless mode [10:14] <tm512> for i/o? [10:15] <relaxed> yes [10:15] <tm512> is -rtbufsize useless then? [10:15] <relaxed> I've never used it [10:15] <tm512> doesn't it buffer the input to make a smoother output? [10:15] <Keestu> hi, i am reading packets from rtsp server through av_read_frame (context, finished, packet). How can i copy this particular packet ? [10:18] <tm512> relaxed: rawvideo lags a lot [10:18] <relaxed> and ffvhuff? [10:21] <tm512> lags [10:21] <help_me> you can give utvideo a try except it's intra-only (I have no idea what you guys are talking about BTW) [10:21] <tm512> I'm going to see if rtbufsize is an input specific option [10:22] <tm512> and put it after -f x11grab [10:22] <tm512> then I'm going to sleep [10:22] <rsdrsdrsdrsd> I have a bunch of videos on which the dar is ok, but the sar is wrong. The sar should be set to 1, is there a tool how I can reset the sar of the video to 1 without transcoding? [10:23] <rsdrsdrsdrsd> If i play the video, it is really stretched because of the sar. I need it to be 1 [10:24] <DelphiWorld> hey FFmpegsters [10:25] <relaxed> rsdrsdrsdrsd: there should be a way using certain players. I believe you'll have to rencode to fix it. [10:26] <DelphiWorld> hey relaxed :D [10:27] <rsdrsdrsdrsd> isn't there an option to just change the metadata info of sar [10:28] <tm512> relaxed: also it might be worth noting that I thought these problems were fixed but they came back after getting a mouse (well, trackball) that polls a minimum of 4 times as much as my old mouse [10:28] <tm512> not sure if that has anything to do with it [10:28] <DelphiWorld> anyone know how to add main profile? [10:28] <tm512> but it makes X11 take up more CPU [10:28] <DelphiWorld> this is my ffmpeg line: /usr/local/bin/ffmpeg -re -i udp://@239.100.1.13:1234 -c:v libx264 -vb 512k -c:a libfdk_aac -ar 44100 -ac 2 -ab 64k -f flv rtmp://192.168.0.13:1935/live/cp; [10:29] <relaxed> DelphiWorld: -profile main [10:29] <DelphiWorld> relaxed: vprofile or profile? [10:29] <relaxed> er, -profile:v main [10:30] <DelphiWorld> thank relaxed [10:30] <DelphiWorld> relaxed, do you recomand adding vb to video? [10:30] <relaxed> setting the video bitrate? [10:31] <DelphiWorld> yes [10:31] <DelphiWorld> otherwise how is bitrate defined if i dont usevb? [10:31] <relaxed> I think using -crf is best [10:31] <DelphiWorld> relaxed: crf is specifying bitrate right? [10:32] <relaxed> https://trac.ffmpeg.org/wiki/x264EncodingGuide [10:33] <rsdrsdrsdrsd> crf is specifying a quality [10:33] <rsdrsdrsdrsd> try a few and choose your quality you can also set a max rate [10:34] <DelphiWorld> rsdrsdrsdrsd: thank, i want to be bandwidth aware [10:35] <tm512> also I specifically ran X and ffmpeg on different CPU cores to see if that had any effect but the stutturing still happens [10:35] <rsdrsdrsdrsd> anyone know how to set the reset sar info [10:36] <DelphiWorld> relaxed: its normal to specify profile if using crf? [10:36] <relaxed> there's the setsar filter, but it recuires encoding [10:36] <relaxed> requires* [10:37] <DelphiWorld> also i dont know what preset must i use for live encoding [10:38] <DelphiWorld> this is not file but stream [10:38] <rsdrsdrsdrsd> yeah but from what i read it also changes the frame size [10:38] <tm512> minimal choppiness in this lossless x264 recording with rtbufsize set after -f x11grab [10:38] <rsdrsdrsdrsd> ok renecoding isn't a problem [10:38] <tm512> dunno if it's just a fluke [10:38] <rsdrsdrsdrsd> I can do copy codec [10:39] <DelphiWorld> guys? [10:42] <help_me> hi DelphiWorld [10:42] <DelphiWorld> hi help_me [10:42] <help_me> I'm tired [10:42] <DelphiWorld> help_me: lol [10:42] <help_me> and somewhat horny [10:43] <help_me> I've been trying out lossy intra-frame codecs [10:43] <help_me> they all suck [10:43] <DelphiWorld> i dont realy know what quality must i chouse for IPTV [10:43] <help_me> well [10:44] <help_me> are you broadcasting? details please [10:44] <DelphiWorld> help_me: i am getting streams from satellite and pushing to rtmp & hls [10:44] <help_me> oh cool [10:44] <help_me> well, it's easy to squeeze that into net-streamable bandwidth with good quality [10:44] <help_me> provided you have a fast enough encoder [10:45] <help_me> depends on the codec [10:45] <tm512> and it doesn't look like rtbufsize is used at all in x11grab :/ [10:45] <DelphiWorld> help_me: i have a 16cors Server [10:45] <DelphiWorld> and want H.264/AAC [10:45] <tm512> guess I have a new programming project for tomorrow [10:45] <help_me> then no problemo [10:45] <help_me> DelphiWorld: your quality is limited only by your bandwidth [10:46] <DelphiWorld> help_me: bandwidth... 1GB :) [10:46] <help_me> then your quality is unlimited [10:46] <DelphiWorld> help_me: lol [10:46] <help_me> what exactly is the problem? [10:46] <DelphiWorld> help_me: but what encoding settings you recomand [10:46] <DelphiWorld> help_me: problem is confusion that's all [10:47] <help_me> depends on the resolution of the streams [10:47] <help_me> what's the width and height? [10:47] <DelphiWorld> help_me: lol, pm [10:47] <help_me> pm? [10:47] <help_me> oh I see ok [10:47] <DelphiWorld> help_me: private message:P [11:27] <help_me1> DelphiWorld: my IRC client glitched out [11:30] <help_me> I am having a really rough night [11:40] <DelphiWorld> back [11:43] <DelphiWorld> how to add a subtitle to a mp4 stream? [12:44] <shoikan> hi folks.. I've got a question about the -f tee option, to output to two separate things... [12:45] <shoikan> I've found the doc about it on the web, but I haven't been able to get it to work properly [12:46] <shoikan> I am trying to use a single multicast input, encode this to 5 different bitrates plus aac audio, then output it in ismv format to 2 separate ingests (http servers).. [12:47] <shoikan> if I specify -f ismv and indicate a single http server, it works like a charm... [12:48] <shoikan> but when I do -f tee "[f=ismv]http://server1/url1|[f=ismv]http://server2/url2" it doesn't seem to produce any output on neither mountpoint) [12:58] <\\matthias> Hey there im running version 1.2.1 on my server and im trying to convert an avi2mp4 but im getting "x264 [error]: high profile doesn't support 4:4:4" - what did i miss? :o [12:58] <\\matthias> and i think that setting is kinda essential for the mp4 to work in web [13:02] <\\matthias> i just tried the same with the latest available version for Win. [13:02] <\\matthias> [libx264 @ 0000000002b9fca0] Error setting profile high. [13:02] <\\matthias> [libx264 @ 0000000002b9fca0] Possible profiles: baseline main high high10 high422 high444 [13:02] <\\matthias> still getting those errors :( [13:02] <\\matthias> im trying ffmpeg -i input.avi -codec:v libx264 -profile:v high -preset slow -b:v 500k -maxrate 500k -bufsize 1000k -vf scale=-1:480 -threads 0 -codec:a libfdk_aac -b:a 128k output.mp4 [13:37] <\\matthias> noone alive eh ? :/ [13:46] <spaam> \\matthias: or people have no idea on how to fix it :( [13:47] <spaam> \\matthias: have you tried adding -level 4.1 or 4.0 after -profile:v high ? [13:53] <\\matthias> -level ... ill try that :) and will pastebin the output.. just a moment. [13:57] <\\matthias> http://pastebin.de/125918 [13:58] <\\matthias> now that was with version 1.2.1 [14:00] <\\matthias> no luck with -level [14:03] <\\matthias> cant be too hard to convert avi 2 mp4 .. in a way that it is still good quality and playin in a browser [14:05] <\\matthias> i converted the video earlier .. but i must say that all those who got converted cant be launched in a brwoser. meh :( ^^ [14:11] <klaxa> >793x480 [14:12] <klaxa> h264 can only encode videos with width and height divisible by 2 [14:12] <klaxa> cc: \\matthias [14:17] <\\matthias> oh [14:18] <\\matthias> so i would have to turn that avi into something else / diff resolution and then like rerun it [14:23] <klaxa> \\matthias, you can use a filter, like -vf "scale=792x480" [14:27] Action: DelphiWorld love -map [14:36] <DelphiWorld> can rtmp have multi audio stream? [14:37] <\\matthias> yeay [14:37] <\\matthias> its working :d [14:37] <\\matthias> :D [14:37] <DelphiWorld> how? [14:37] <\\matthias> thanks guys :D [14:37] <DelphiWorld> lol [14:38] <\\matthias> ias klaxa said. ffmpeg was not able to divide [14:38] <\\matthias> because [klaxa] >793x480 [14:39] <\\matthias> so i used the scale setting first to rescale it (avi 2 avi) xD pretty "senselesse" but anyway [14:39] <\\matthias> so the video now is 792x480 [14:39] <\\matthias> and then i was able to run the avi 2 mp4 command successfully :D [14:40] <\\matthias> sry for my crap english but i think u get what i mean xD [14:43] <DelphiWorld> lol [14:48] <\\matthias> klaxa yea i did. but only when im doing "avi 2 avi" first it doesnt work inline directly when u want to go from avi 2 mp4 [14:48] <\\matthias> what so funny over there :D [14:48] <\\matthias> xD [14:48] <klaxa> it should work [14:49] <\\matthias> i have some :D [14:49] <\\matthias> u want some? xD [14:49] <\\matthias> lol [14:49] <klaxa> it's read-input -> decode -> filter -> encode -> write-output [14:49] <klaxa> so there should be no problem filtering right away [14:50] <DelphiWorld> guys. will rtmp support multi language video? [15:21] <DelphiWorld> how to include subtitles if using mp4? [15:21] <DelphiWorld> i am encoding mpeg2ts into mp4 [15:54] <DelphiWorld> hi guys [15:54] <DelphiWorld> i'm doing hls like this [15:55] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -c:v libx264 -c:a libfdk_aac -ac 1 -ar 44100 -f segment -segment_time 4 \ -segment_list euronews.m3u8 -segment_format mpegts stream%05d.ts [15:55] <DelphiWorld> but its saying output file #0 dont have any stream [16:06] <DelphiWorld> hls... guys [16:06] <Keestu> dear all, could someone help me on this issue ? http://ffmpeg.org/pipermail/libav-user/2014-May/006841.html [16:14] <DelphiWorld> relaxed: any clue? [16:15] <DelphiWorld> hls is not working for me guys [16:15] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -pix_fmt yuv420p -vcodec libx264 -acodec libfdk_aac -r 25 -profile:v baseline -b:v 512k -maxrate 1500k -force_key_frames 50 -s 640×360 -flags -global_header -f segment -segment_list /tmp/hls/euronews.m3u8 -segment_time 10 -segment_format mpeg_ts -segment_list_type m3u8 /tmp/hls/segment%05d.ts [16:15] <DelphiWorld> Output file #0 does not contain any stream [16:26] <DelphiWorld> saste: are you available to help with hls? [16:35] <DelphiWorld> ffmpeg -re -i udp://@239.100.1.8:1234 -map 0:0 -map 0:2 -map 0:8 -c:v libx264 -vb 512k -c:a libfdk_aac -profile:a aac_he_v2 -b:a 24k -ac 2 -ar 44100 -f flv rtmp://localhost:1935/live/eu/join #videolan [16:35] <DelphiWorld> oh [17:00] <DelphiWorld> anyone using crtmpserver? [18:13] <SpecialEd> Hey guys, having troubles getting my Haswell based dedicated server to use all CPU cores with ffmpeg to transcode with the libx264 codec. [18:13] <SpecialEd> I configured (and compiled) ffmpeg manually with the flags: --enable-pthreads --enable-zlib --disable-stripping --enable-runtime-cpudetect --enable-postproc --enable-swscale [18:14] <SpecialEd> Also when I use ffmpeg I use the flag -threads 8. However [18:14] <SpecialEd> I'm only seeing about 40% usage across all 8 cores [18:14] <c_14> try getting rid of -threads 8 or increasing it [18:15] <c_14> libx264 should thread automatically [18:15] <SpecialEd> ok [18:15] <SpecialEd> I have a small sampling of files set aside [18:15] <SpecialEd> and im timing how long it takes to run [18:15] <SpecialEd> which is always 96 seconds [18:16] <SpecialEd> so Ill try that right now without --threads # and then --threads 0 [18:16] <SpecialEd> so ill be back in about 192 seconds or so :) [18:21] <SpecialEd> Almost no difference with --threads 0 or without even using the --threads flag... :( [18:21] <SpecialEd> however when I dropped --threads flag it did complete the job 2 seconds faster @ 94 seconds [18:21] <SpecialEd> sure thing, brb [18:25] <SpecialEd> http://pastebin.com/5Uk9Zchw [18:26] <SpecialEd> crap, I guess I posted my script, but its just the ffmpeg command wraped in a for loop for each file [18:26] <SpecialEd> so its not like you'll need to decipher whats doin [18:28] <SpecialEd> Also, I'm not sure if its related but I am also seeing the terminal getting spammed by ffmpeg with the following: [18:28] <SpecialEd> [h264 @ 0x3b21620] missing picture in access unit with size 40 [18:28] <SpecialEd> [h264 @ 0x3b21620] AVC: nal size 21102622 [18:29] <SpecialEd> The videos still play fine so I just normally have --loglevel quiet set [18:29] <c_14> hmm, I can't see anything about the command that would limit ffmpeg from using full cpu... [18:30] <SpecialEd> yea, and I thought maybe it could be iowait but I checked the disk I/O and it was low [18:30] <SpecialEd> this particular doesn't have SSDs, rather a RAID0 configuration of two 2TB disks [18:31] <SpecialEd> well let me take a step back here then [18:31] <c_14> I'm guessing the factor limiting the encoding speed isn't the x264 encoding, but rather the filter or something. [18:31] <SpecialEd> I am running this on many files each day [18:31] <SpecialEd> I could parallelize the script itself [18:31] <SpecialEd> and that way I do see the CPU usage go up [18:32] <SpecialEd> So overall, would it be faster if I were to parallelize the script 3x or would it be faster to get just 1 instance of it to use more CPU by itself? [18:32] <SpecialEd> Does my question make sense? [18:33] <c_14> In this case, I think it would be better to make the script run in parallel. [18:33] <SpecialEd> ok [18:34] <SpecialEd> What would be faster overall if lets say I had 10,000 videos to transcode to x264: [18:34] <SpecialEd> 1. 3 parallel processes of ffmpeg running down a list of the files [18:34] <SpecialEd> 2. 30 parallel processes of ffmpeg running down a list of the files [18:35] <l0rd_hex> if I've already encoded a video with "ffmpeg -i input -t 20 -c:v libx264 -tune grain -preset veryslow -crf 16 -movflags +faststart output.mp4" and I want to adjust the crf value (higher) do I need to pass -tune and -preset again? [18:35] <SpecialEd> Like, at any point does over parallelizing slow ffmpeg down? Obviously eventually you'll saturate the disk I/O which would be a bottleneck, but lets ignore that for the scope of the question. [18:36] <c_14> SpecialEd: I would never increase the amount of cpu-dependent tasks over 1+numcpu, in this case I would probably go for 3 since that should fully saturate the cpu. Basically, use the smallest number that fully saturate the cpu. [18:37] <c_14> l0rd_hex: you want to encode the output video again? [18:37] <SpecialEd> OK, last question. Do you have any tips for any flags that I should try to help increase the transcode speed? Like should I try to specify -crf and -preset ? [18:37] <c_14> l0rd_hex: it would probably be better to encode the input video again with a higher crf or else you'll introduce more loss [18:38] <c_14> SpecialEd: if filesize doesn't matter, use -preset {faster,veryfast,ultrafast} [18:38] <SpecialEd> ok ty [18:38] <c_14> SpecialEd: -crf is mainly quality based and doesn't really influence encode speed [18:38] <SpecialEd> gotcha, thanks [18:39] <c_14> l0rd_hex: but, if you want to encode the output video again, yes I would pass tune and preset again [19:06] <l0rd_hex> c_14: oops, stepped away [19:06] <l0rd_hex> thanks, it's security cam footage so it actually looks the same crf 16 -> crf 20 [19:06] <l0rd_hex> I let it run in parallel and it didn't take very long [19:11] <tm512> does -rtbufsize even do anything? [19:12] <DelphiWorld> guys i want to serv a live stream over hls and i'm stuck [19:12] <DelphiWorld> can someone help please [19:14] <tm512> in the source code it looks like rtbufsize sets unsigned int max_picture_buffer; which is not used for anything other than printfs [20:06] <alteregoa> someone has a static linked ffmpeg handy for mipsel? ty [20:50] <rsdrsdrsdrsd> I want to convert to square pixels with a fixed height. Should I do setsar=1 before or after scale=trunc(dar*720/2)*2:720 ? [20:50] <rsdrsdrsdrsd> This keeps me busy for a few days now [20:50] <rsdrsdrsdrsd> and get come to a solution [20:50] <rsdrsdrsdrsd> because all files are different [20:50] <rsdrsdrsdrsd> The docs aren't clear enough for me [20:51] <rsdrsdrsdrsd> does setsar change the frame size or not? [21:07] <klaxa> it shouldn't [21:07] <klaxa> sample aspect ratio is different from display aspect ratio [21:08] <rsdrsdrsdrsd> yeah, but they are related to each other aren't they? [21:09] <rsdrsdrsdrsd> if I do it before the frame size of the source is the same as the dar of the output [21:10] <rsdrsdrsdrsd> if i do it after the dar of the output is the same as the frame size of the input times the sar of the input [22:11] <sdfgdf> I forgot how to list the pixel formats supported by a codec [22:11] <sdfgdf> internet searches are turning up nothing and -pix_fmt list is failing [22:18] <c_14> -pix_fmts [22:18] <sdfgdf> thank you, my savior [22:19] <sdfgdf> wait no [22:19] <c_14> -h encoder=$encoder [22:19] <sdfgdf> I mean all of the pix_fmts supported by a secific codec [22:19] <sdfgdf> there is a command for that but I forgot it [22:19] <c_14> -h encoder=h264 should list them for h264 etc etc [22:19] <sdfgdf> -h etc. tring... [22:20] <sdfgdf> ah thank you [22:21] <sdfgdf> hey, it looks like mjpeg does not suport more than 8 bits per pixel [22:21] <sdfgdf> well, that's really all I need to know [22:21] <sdfgdf> many thanks [00:00] --- Sat May 24 2014

1 0

ffmpeg-devel.log.20140522
by burek 23 May '14

23 May '14

[01:59] <cone-362> ffmpeg.git 03Dale Curtis 07master:d5d6b8e25979: Fix compilation error around struct visibility when VDPAU disabled. [01:59] <cone-362> ffmpeg.git 03Michael Niedermayer 07master:f932e5f71692: avcodec/vdpau_internal: move "struct vdpau_picture_context" up to avoid "#else" [03:14] <kierank> http://embed.cs.utah.edu/creduce/ [03:16] <alpa_astero> kierank: http://alpaastero.wordpress.com/2014/05/19/tracking-down-a-bug-in-clang-fsa… :D [03:17] <kierank> ah [03:17] <kierank> :) [03:17] <alpa_astero> (if you found it via hacker news, the post is by a guy who is a fellow Hacker School alum, who learned about it from my blog post) ^-^ [03:18] <alpa_astero> it's a good tool :) [03:20] <alpa_astero> (here's Michael's patch to work around the clang bug - http://git.videolan.org/?p=ffmpeg.git;a=commit;h=bebce653e5601ceafa004db0eb… - he did it within a few hours of my opening the bug report on the Libav bugtracker) :-) [14:20] <nevcairiel> Daemon404: gcc 4.8.3 is out [14:21] <Daemon404> \o/ [14:40] <Daemon404> nevcairiel, now gcc 4.9.1 where [14:41] <Daemon404> eta N months [14:48] <Case> nevcairiel: have you made a mingw set with that? [14:49] <nevcairiel> i'm working on it [14:50] <Daemon404> im making mine right now :P [14:50] <nevcairiel> i finished compiling, just need to pack it up with zlib and upload it [14:50] <Daemon404> nevcairiel, mingw is rather annoying for c++ right now... [14:51] <Daemon404> there is a hard dep on winpthreads (bleck) if you want c++11 threads [14:51] <Daemon404> kind of annoying [14:51] <Daemon404> mind you, you may as well just use msvc for c++ [14:51] <Daemon404> ;) [14:51] <nevcairiel> yeah i dont enable threads in gcc because of that nonsense [14:52] <Daemon404> winpthreads is really bad still [14:52] <Daemon404> lots of failed compiles, deadlocks, terrible perf [14:52] <Daemon404> i wonder why they didntj ust import pthreads-win32 [14:53] <nevcairiel> NIH? [14:53] <Daemon404> i guess so [14:53] <nevcairiel> i switched back to building ffmpeg with win32threads anyway [14:53] <Daemon404> why would you ever use pthreads on windows [14:53] <nevcairiel> pthreads-w32 has a handle leak if you mix it with native windows threads [14:53] <Daemon404> for ffmpeg anyway [14:54] <nevcairiel> mostly because of some thing in the udp receiving code that caused packet loss without the worker thread, which only worked with pthreads [14:54] <nevcairiel> but i hacked it together to work with win32threads now [14:54] <Daemon404> is it submit-able? [14:54] <Daemon404> also, gross. [14:54] <nevcairiel> probably not [14:55] <Daemon404> :/ [14:55] <nevcairiel> i mean, it works fine, since closing a socket on windows makes all blocking calls in other threads return immediately [14:55] <nevcairiel> (which is what pthread_cancel does on *nix) [14:55] <nevcairiel> http://git.1f0.de/gitweb?p=ffmpeg.git;a=commitdiff;h=b7c9992477d9d48dd211c7… [14:55] <nevcairiel> i just hacked it in :P [14:56] <nevcairiel> (also, the API for pthread_cond_timedwait is just horrible) [14:56] <Daemon404> thats not *that* bad [14:56] <nevcairiel> why not just an api that says "wait 100 ms", noo, it has to be an API that takes a silly timespec with an absolute time [14:57] <Daemon404> i was going to say "holy shit thats a lot of hacks" to your repo [14:57] <Daemon404> but like 50% of them are haali [14:57] <nevcairiel> well there is two other patches in the w32threads header to add the new function and add a few return values [14:57] <nevcairiel> but yeah [14:57] <nevcairiel> it felt kinda ugly [14:57] <nevcairiel> but i tested it to work reliably at least [14:58] <nevcairiel> (important to note, pthread_cancel in pthreads-w32 is pretty much a NOP) [14:58] <nevcairiel> (i checked) [14:58] <Daemon404> there are still a bunch of things you could submit i see [14:58] <Daemon404> http://git.1f0.de/gitweb?p=ffmpeg.git;a=commit;h=f819cef397b46212bc8b2fbcc3… [14:58] <Daemon404> this looks useful actually [14:58] <Daemon404> i hit that a lot with user content [14:59] <iive> nevcairiel: are you talking about classic mingw, or mingw64 ? [14:59] <Daemon404> iive, mingw64 [14:59] <nevcairiel> classic mingw is dead [14:59] <Daemon404> yea ^ [14:59] <iive> ok. so one problem less. [15:00] <ubitux> could we get the github pull requests automatically forwarded to ffmpeg-devel? [15:00] <Daemon404> should be easy enough with the github API [15:00] <Daemon404> but i really think we ought to just autoreply with "send to the ML" [15:00] <Daemon404> people who dont even take the time to figure out how to submit it to us... i question how goood their patches are [15:00] <Daemon404> :P [15:01] <ubitux> well for occasionnal patches, they might not want to mess with mailing-list [15:01] <ubitux> typically this: https://github.com/FFmpeg/FFmpeg/pull/70 [15:02] <Daemon404> right [15:02] <Daemon404> well such a script would be only a few lines of code + cron job [15:03] <J_Darnley> Is it hard to say in your opening email "please CC me"? [15:03] <Daemon404> J_Darnley, then you run into ML moderation [15:03] <Daemon404> mind you, you can also use gmane [15:03] <Daemon404> if you dont want to subscribe [15:04] <J_Darnley> I think that's even worse that github's features [15:04] <Daemon404> gmane is horrible [15:04] <Daemon404> but also the best thing for what it does available [15:04] <Daemon404> life sucks, etc [15:07] <J_Darnley> I wonder if that's why some of my emails to other places have just disappeared. [15:08] <Daemon404> nevcairiel, wtf is with gcc's mirrors [15:08] <Daemon404> 4.8.3 isnt on any of them [15:08] <nevcairiel> Daemon404: dunno, i got i t [15:08] <Daemon404> which mirror [15:08] <Daemon404> i tried 3 so far [15:08] <nevcairiel> http://ftp.gnu.org/gnu/gcc/ [15:08] <Daemon404> ohlol [15:08] <Daemon404> i was using the local mirror [15:09] <Daemon404> as in closest [15:09] <Daemon404> i guess nothing rsync'd it yet [15:15] <nevcairiel> Case: http://files.1f0.de/mingw/mingw-w64-gcc-4.8.3-stable-r11.7z [15:16] <Case> nice, thanks [15:17] <Case> I wish they hurried up with fixed 4.9 release. It seems to have some nice performance improvements for compiled binaries [15:19] <cone-833> ffmpeg.git 03Michael Niedermayer 07master:e8e9186a70ae: avcodec/pgssubdec: Use av_malloc(z)_array() [15:19] <cone-833> ffmpeg.git 03Michael Niedermayer 07master:aca48725afe8: avcodec/proresdec2: Use av_malloc_array() [15:19] <cone-833> ffmpeg.git 03Michael Niedermayer 07master:124c94a1c8de: avdevice/v4l2: Use av_malloc_array() [15:23] <nevcairiel> updated my fate box as well, hopefully nothing breaks ;) [15:32] <Daemon404> ;p [15:32] <cone-833> ffmpeg.git 03Carl Eugen Hoyos 07master:eff913c9725b: Autodetect Asterisk raw pcm with extension "sln". [15:32] <cone-833> ffmpeg.git 03Michael Niedermayer 07master:a996eacce269: Merge remote-tracking branch 'cehoyos/master' [17:15] <Daemon404> oh nevcairiel [17:16] <nevcairiel> ? [17:16] <Daemon404> https://github.com/FFMS/ffms2/commit/ead726b7964edaf7dc725c68c41ce00a82eb3d… [17:16] <Daemon404> relevant to your haali code? [17:16] <nevcairiel> I have a similar hack somewhere [17:17] <nevcairiel> After a ass script with 1mb privates crashed [17:17] <Daemon404> yea [17:17] <Daemon404> same [17:17] <Daemon404> it was for a .ass [17:17] <nevcairiel> Silly alloca everywhere [17:17] <Daemon404> yep [17:18] <Plorkyeran_> probably should lower the limit for alloc to like 1KB [17:19] <Plorkyeran_> since codecprivate sections seem to usually be more like 10 bytes [17:19] <nevcairiel> http://git.1f0.de/gitweb?p=ffmpeg.git;a=commitdiff;h=f570ac731b90e12664b961… [17:19] <nevcairiel> Mine seems longer [17:20] <nevcairiel> More error free [17:20] <nevcairiel> I don't use alloca at all anymore for CP [17:21] <j-b> (ASS files with > 1MB CodecPrivate) o_O [17:22] <Plorkyeran_> fansubbers do dumb things [17:22] <Plorkyeran_> the only reason I fixed it was because it was stopping you from opening the video track [17:23] <nevcairiel> I should've looked into what was actually in that codec private block [17:23] <Plorkyeran_> commented lines [17:23] <nevcairiel> Can you have 1mb format rules? [17:24] <Daemon404> if you hit such a stupid file, still better to fail than crash [17:24] <Daemon404> as was the previous behavior [17:24] <nevcairiel> It just crashed because the 2mb limit in the code was already too large [17:25] <nevcairiel> At least on windows default stack is 1mb [19:04] <cone-833> ffmpeg.git 03Vignesh Venkatasubramanian 07master:37cef8c77449: libavformat/matroskaenc: Add WebM DASH support [20:16] <cone-833> ffmpeg.git 03Vignesh Venkatasubramanian 07master:bda8ceb9f8d9: maintainers: adding myself as a maintainer to webm dash [21:39] <cone-833> ffmpeg.git 03Christophe Gisquet 07master:f8de35ebc4f2: x86: hpeldsp: kill hpeldsp_mmx.c [21:39] <cone-833> ffmpeg.git 03Christophe Gisquet 07master:17ac9980555a: x86: hpeldsp: mark _xy2 versions as approximate [21:39] <cone-833> ffmpeg.git 03Christophe Gisquet 07master:c081ca851c9b: x86: hpeldsp: avg_pixels_xy2 for mmx2&3dnow [21:44] <kurosu> woah that went smoother than expected [23:40] <cone-833> ffmpeg.git 03Florent Le Coz 07master:fdcb2873e1c8: Fix the mime types for MPEG and MPEG-TS formats [23:59] <kurosu> jamrial, I have SSSE3 versions of _xy2 that should cause less yawns [23:59] <kurosu> currently finetuning the loading of constants [00:00] --- Fri May 23 2014

1 0

ffmpeg.log.20140522
by burek 23 May '14

23 May '14

[00:45] <sgfdsf> Is it possible to use a differet codec for each GOP? [00:47] <sgfdsf> sometimes x264 gives better compression and sometimes utvideo gives better compression so I was wondering if there was some sort of smart algorithm that tried encoding segments in each codec and selected the smaller pieces [00:48] <voip_> hello guys, i need to convert avi to mp4 [00:49] <c_14> sgfdsf: pretty sure you can't do that [00:49] <llogan> voip_: mp4 is just a container that can contain various formats of various types [00:49] <Plorkyeran_> matroska segment linking could theoretically support it [00:49] <Plorkyeran_> but nothing does [00:50] <blippyp> voip_: short answer, try this: ffmpeg -i video.avi -c copy video.mp4 [00:50] <voip_> llogan, i uloaded avi file to wowza and cant play [00:50] <Plorkyeran_> comedy option would be to generate an avisynth script that stitches together each individual segment [00:51] <voip_> blippyp, tryed no success [00:51] <blippyp> then you probably need to re-encode using a different codec [00:52] <blippyp> something like: ffmpeg -i video.avi -c:v libx264 -crf 17 -c:a ac3 video.mp4 [00:53] <voip_> blippyp, whats mean -crf 17 ? [00:53] <llogan> read this https://trac.ffmpeg.org/wiki/x264EncodingGuide [00:54] <blippyp> refer to llogan's link... ;) [00:54] <voip_> :) [00:56] <voip_> blippyp, thanks [00:56] <blippyp> np [01:04] <voip_> [NULL @ 0x26c9a00] Unable to find a suitable output format for 'libx264' [01:04] <voip_> version 2.2.1 static buil [01:08] <voip_> ok found [02:45] <vklimkov_> using ffmpeg on android to play mp4. i works but plays video corruped like this: http://picpaste.com/exeample-7z5JC5hS.png. What might be the reason?? [02:45] <vklimkov_> it* [02:49] <klaxa> wild guess is you selected the wrong resolution for playback? it seems like the lines are offset by a few pixels every row [02:49] <klaxa> but i don't know how you could set it up that way even [02:49] <klaxa> maybe android is doing something weird [02:51] <vklimkov_> klaxa: i'm using tutorial without modifications: http://www.roman10.net/how-to-build-ffmpeg-with-ndk-r9/. So it looks like implementation staff, not compilation issue? [02:52] <klaxa> no idea [02:53] <klaxa> most likely your android application messing up [02:53] <klaxa> not ffmpeg itself [02:57] <vklimkov_> well. i was able to start one of the videos properly. no idea what i've done though. thanks anyway. [06:04] <sgfdsf> so I'm encoding a 1080p video file in lossless x264 4:2:0 with -preset placebo. It is encoding at less than 1 FPS. Is there any way to make it faster while still using the same settings? I'm using 64-bit ffmpeg BTW. [06:10] <JEEB> lol, 1) make sure your x264 has asm opts enabled (you should have a line of SIMD [SSE2 etc] enabled in libx264's log that it outputs to terminal) 2) get a faster computer [06:10] <JEEB> preset placebo is slow [06:10] <JEEB> I mean, very very slow [06:10] <JEEB> should only be used when doing quality benchmarks where you aim for maximum quality on both/all encoders [06:10] <JEEB> unless you of course get a fast enough speed from it :P [06:11] <JEEB> (for example I get ~4fps with 720p24 and that's kind of OK) [06:11] <JEEB> but by choosing a faster preset like even veryslow will only lose you some compression, while giving a nice boost to speed [06:12] <JEEB> if you need to go faster, then just pick a faster preset :) (or get a faster machine, but that's a whole separate area of discussion) [06:12] <sgfdsf> ok so a couple questions [06:13] <sgfdsf> do the speed presets work with all video codecs? [06:13] <JEEB> no [06:13] <sgfdsf> utvideo? [06:13] <JEEB> no [06:13] <sgfdsf> ok [06:13] <sgfdsf> and regarding the ASM thing [06:13] <JEEB> ut video just has the prediction mode, although I recommend just keeping it in the median mode [06:14] <sgfdsf> why do you recommend that? [06:14] <sgfdsf> I did not know it had multiple encoding modes [06:14] <JEEB> not encoding modes [06:14] <JEEB> it's just a manually hand-written SIMD code path [06:14] <JEEB> which, surprise surprise, is faster than what a compiler can produce :P [06:14] <sgfdsf> I mean the utvideo thing, sorry [06:15] <JEEB> yes, ut video has two (actually three) encoding modes [06:15] <JEEB> left and median [06:15] <sgfdsf> and right? [06:15] <JEEB> left is pretty much not worth it [06:15] <JEEB> no [06:15] <JEEB> the third is "none" [06:15] <JEEB> which means there is no prediction [06:15] <sgfdsf> so just RLE or something? [06:16] <JEEB> well Ut Video is just RLE in the end :D it's just what gets done to the picture before that [06:16] <sgfdsf> so none is no inter-frame encoding? [06:16] <JEEB> there's no inter-frame prediction in Ut Video [06:16] <JEEB> it's all intra [06:16] <sgfdsf> fuck [06:16] <sgfdsf> no [06:17] <sgfdsf> a PNG sequence is way huger [06:17] <sgfdsf> how can that be? [06:17] <JEEB> the fuck I know :P There's plenty to PNG encoders, among other things [06:17] <sgfdsf> so it's basically a still image encoder then [06:17] <JEEB> yes [06:18] <sgfdsf> awesome [06:18] <JEEB> and if you have multiple slices it encodes multiple pictures [06:18] <sgfdsf> slices? [06:18] <JEEB> basically parts of the picture encoded separately [06:18] <JEEB> that's how the official Ut Video decoder and encoder multithread [06:18] <sgfdsf> but it's all intra [06:18] <JEEB> and I've lately enabled slices by default in the lavc encoder [06:18] <JEEB> yes, it's all intra [06:19] <JEEB> basically you cut the picture after X height :P [06:19] <sgfdsf> interesting [06:19] <JEEB> and encode X pictures separately [06:19] <JEEB> that's what slices are [06:19] <sgfdsf> to use multiple cores [06:20] <sgfdsf> it's a speed boost I presume [06:20] <sgfdsf> so anyhow I assume lossless x264 is inter-frame then [06:20] <sgfdsf> since lossy is [06:20] <sgfdsf> and there's "correction frames" unless I'm mistaken [06:21] <sgfdsf> but sometimes the inter-frame makes bigger files than the intra-frame [06:22] <sgfdsf> so basically utvideo's image encoder uses fairy magic [06:22] <sgfdsf> I bet it would make a great new lossless still image file format [06:22] <JEEB> no, it most definitely does not use any kind of fairy magic :P [06:22] <JEEB> you can check libavcodec/utvideoenc.c [06:23] <JEEB> and H.264 can be all-intra or intra+inter [06:23] <JEEB> depends on your setting [06:23] <sgfdsf> oh boy [06:23] <JEEB> default is a long'ish GOP so inter is used (since it is quite useful) [06:23] <sgfdsf> yes indeed [06:23] <sgfdsf> it looks weird at medium-bitrate lossy though [06:24] <sgfdsf> like let's say it's at 20 mbps with a GOP size of 80 [06:24] <sgfdsf> and a moving camera [06:24] <sgfdsf> there's image shifting all over [06:24] <sgfdsf> I prefer just having more blur per frame [06:25] <sgfdsf> but regarding lossless mode [06:25] <sgfdsf> I assume the GOP size defaults to something pretty big [06:25] <sgfdsf> the max was 100 according to somebody, don't know why though [06:26] <sgfdsf> I would think automatic keyframe detection would be a bit better for detecting cuts and flashes and such [06:26] <sgfdsf> then again I'm no expert [06:38] <sgfdsf> sorry if I'm beng adummy [06:38] <sgfdsf> how do I check if libx264 is compiled to use ASM? [06:38] <sgfdsf> or am I misunderstanding [06:42] <sgfdsf> ok I see now [06:42] <sgfdsf> yes, my ffmpeg has "SSE2Fast" [06:43] <sgfdsf> MMX2 and other SSE stuff [06:43] <Plorkyeran_> pretty sure placebo 1080p would be way slower than 1fps without asm [06:43] <sgfdsf> oh yeah? [06:43] <sgfdsf> I hate to imagine "way slower" [06:44] <Plorkyeran_> the non-asm stuff isn't particularly optimized [06:44] <Plorkyeran_> while the asm is excessively optimized [06:44] <sgfdsf> it's hard to optimize [06:44] <sgfdsf> I used to make computer games [06:44] <sgfdsf> the redering time was a killer [06:45] <sgfdsf> I can save a PNG pretty fast [08:08] <bencc2> can I place mp4 video ontop of other mp4 video without loosing quality? [08:09] <relaxok> hello.. does cutting clips with ffmpeg -ss and -t seek to/cut to timecode if there's a timecode track or is it only from a 00:00:00 start? [08:09] <relaxok> and if its from 0 is there a way to cut on timecode points? [08:10] <relaxok> or is there a way to output the timecode start time? [08:10] <relaxok> (i want to script cutting up a video on a set of timecode points, without human interaction) [08:10] <relaxok> if i can parse some text output like the timecode start and then subtract it from the timecode cut times, i could cut with -ss -t starting at 0 [10:01] <brainopia> hi, I'm encoding a big mp4 file for http streaming, I'm using faststart, but moov atom is pretty big around 1mb, so there is a very noticeable delay at the beginning, how to circumvent it? [10:06] <Paranoialmaniac> brainopia: movie fragment with empty initial movie. [10:09] <brainopia> Paranoialmaniac: sorry, I'm quite a noob in this matter, I would be grateful for any additional hint or a link to work through to understand your advice [10:13] <Paranoialmaniac> the feature movie fragment is splitting media and index into pairs of media and index. empty initial movide is has no media data and no index of media data, and has only the initialization of decoder. thus you can start playback with minimal delay. [10:15] <Paranoialmaniac> a weakpoint of fragmentation is random access. you shall create own index and cache or get the index somewhere if present. [11:03] <b_jonas> strange. I'm processing multiple videos with basically the same command line, only the filenames replaced, but for one of them, I get a strange error message. the video could be corrupt, but the message doesn't seem to indicate that. [11:04] <b_jonas> [buffer @ 0000000003599560] Unable to parse option value "-1" as pixel format [11:04] <b_jonas> [buffer @ 0000000003599560] Error setting option pix_fmt to value -1. [11:04] <b_jonas> [graph 0 input from stream 0:0 @ 00000000002c9f00] Error applying options to the filter. [11:04] <b_jonas> Error opening filters! [11:04] <b_jonas> my command line is: ffmpeg -i "rawinput\2014-05-17-ceghetvege\208-3.mp4" -vf crop=w=160:h=16:x=0:y=0,fps=1/60 -pix_fmt gray "tmp\idok\cegh-ido-208-3-%05d.png" [11:07] <b_jonas> I wonder what that means [11:07] <blippyp> I don't think it's liking your pixel format - try using gray16le instead [11:19] <b_jonas> fflogger: I will, yes, I'll just wait till all my commands run on the last video too [11:19] <b_jonas> um [11:19] <b_jonas> ubitux: I will paste the complete output, with -v 99 [11:25] <b_jonas> here's the full log: http://dpaste.com/2CSDTMF/ [11:27] <b_jonas> argh, it complains about that even if I don't give the -pix_fmt option [11:28] <sacarasc> It the input where it's having the problem. [11:29] <OrbWeaver> I have a Blu Ray with an AC-3 audio stream that does not play correctly with mplayer or ffplay. Where is the correct place to report this? [11:31] <OrbWeaver> ok [11:36] <OrbWeaver> Pasted here: http://pastebin.com/y24PTdFD [11:36] <b_jonas> I don't understand. no matter how I try to convert this particular video with ffmpeg, I get the same error. I can still play it with ffplay. [11:37] <sacarasc> OrbWeaver: Did you get the latest git to try with this file? [11:37] <sacarasc> If so, then it's just an unsupported feature of AC3. [11:37] <OrbWeaver> Yes, mplayer automatically pulls in the latest ffmpeg git, I will check the commit [11:38] <OrbWeaver> commit f932e5f7169282daaa06b08500d2c632cf70e14e [11:41] <ubitux> b_jonas: maybe you can open a bug report? [11:42] <b_jonas> ubitux: I might, but I want to investigate a bit more first. it's likely that the video really is corrupt, [11:42] <b_jonas> and it's just the particular error message that's wrong [11:42] <b_jonas> but it's strange that I can play this with ffplay. [11:42] <ubitux> b_jonas: well, if ffplay plays it, there is a bug in ffmpeg [11:42] <ubitux> ffmpeg -i bla.mp4 -frames:v 1 test.png works? [11:43] <b_jonas> also, the error won't be easy to reproduce, because it occurs on only this video, and I can't give this particular video to you [11:43] <b_jonas> a moment, I've tried something like that, let me try that exact statement [11:44] <b_jonas> ubitux: no, that still gives the same error [11:44] <OrbWeaver> sacarasc: so am I out of luck with this disk? Would it help to upload the sample file, or is this an well understood unimplemented feature? [11:45] <b_jonas> I can try upgrading to a latest ffmpeg (not release) first, though this one is not too old [11:45] <ubitux> b_jonas: wait, you have 2 streams [11:45] <sacarasc> OrbWeaver: Ask ubitux, I am, just a common user, he is a dev. :D [11:45] <b_jonas> oh damn... you're right [11:45] <b_jonas> I didn't even notice that [11:45] <b_jonas> why does this have two video streams? [11:45] <ubitux> 2 *videos* streams [11:45] <ubitux> i don't know [11:45] <b_jonas> yeah, that's the problem [11:45] <ubitux> try to play both [11:45] <b_jonas> right [11:45] <ubitux> it MIGHT be a cover or something [11:46] <b_jonas> let me look up the options, it's been a while since I handled multiple streams, and it was complicated back then too [11:46] <b_jonas> (two video streams, argh. I wouldn't be surprised if it was just one video and one audio.) [11:50] <b_jonas> I presume the png writer wouldn't accept two streams [11:50] <ubitux> well [11:51] <ubitux> does both -map v:0 and -map v:1 works individually? [11:51] <b_jonas> and, wait for it, ffprobe says stream 0 has unknown pixel format [11:51] <ubitux> OrbWeaver: check http://trac.ffmpeg.org/query?status=new&status=open&status=reopened&keyword… [11:52] <ubitux> if you see an issue similar to yours, comment & upload your sample, otherwise, just open a new one [11:52] <ubitux> b_jonas: one of the stream is probably badly probed, that's the reason it fails [11:53] <OrbWeaver> ubitux: ok, thanks [11:53] <ubitux> b_jonas: try ffplay -vst 0 and ffplay -vst 1 [11:54] <b_jonas> -map v:1 works, can extract image that way [11:54] <b_jonas> will try original command with -map v:1 then [11:54] <b_jonas> let me try ffplay -vst 0 [11:54] <OrbWeaver> ubitux: is it possible that the Blu Ray has failed to decrypt correctly and this is manifesting as codec errors? Or would that result in a more obvious error? The video seems OK, as does the first few seconds of the audio. [11:55] <b_jonas> indeed, ffplay -vst 0 doesn't work (no wonder, unknown pixel format), -vst 1 works [11:55] <b_jonas> that explains a lot: [11:55] <b_jonas> ffplay could play the video because it chose video stream #1 automatically, [11:56] <b_jonas> whereas ffmpeg doesn't have this freedom, it chose video stream #0 and couldn't decode it [11:56] <ubitux> OrbWeaver: well the file might have multiple audio streams, and you can only read one, i guess [11:56] <b_jonas> ubitux: thank you for the pointer [11:56] <OrbWeaver> ubitux: all of the other audio streams seem to work fine. Unfortunately they are all foreign language or commentary tracks. [11:57] <ubitux> b_jonas: please make sure you report it [11:57] <b_jonas> I might still try to file a bug report, because the error message is unclear, though like I said it will be hard to reproduce [11:57] <ubitux> OrbWeaver: i don't know :) [11:57] <ubitux> b_jonas: hard to reproduce? why? [11:58] <ubitux> b_jonas: can you generate a sample with the same issue with ffmpeg -i bla.mp4 -map 0 -c copy -t 10 sample.mp4? [11:58] <b_jonas> ubitux: hard to reproduce because I can't give you the video file, but indeed, stream copying the first stream is a good idea, I'll try that [11:59] <b_jonas> if that fails, I'll try to truncate the video file bytewise, hoping I can reproduce the message that way [11:59] <ubitux> if the header is not at the beginning it won't work [11:59] <ubitux> and you'll need to run a qt-faststart pass or something [12:00] <ubitux> remuxing should be fine though [12:00] <b_jonas> yes, and this is a crazy mov container :( [12:12] <b_jonas> hmm... ffmpeg wouldn't even stream copy that video stream #0 for me [12:25] <Keestu> why ffmpeg programming does not have detailed documentation ? [12:27] <blippyp> actually, it has quite good documentation: http://www.ffmpeg.org/documentation.html [12:29] <Keestu> blippyp, do u think the information provided in the link you provided is suffice to develop with out the help of community folks ? [12:30] <blippyp> yes, look at the doxygen link near the bottom [12:30] <blippyp> under api documentation [12:30] <ItsMeLenny> when converting a list of images to a video, is there a way to have it start at 30, or rather, my images themselves start at 30, but of course it's looking for the file named 0001 [12:31] <blippyp> ItsMeLenny, yes, let me find the option unless someone else beats me to it... [12:32] <blippyp> This might take me a minute though, not sure what script I did that in [12:32] <ItsMeLenny> i'm usinig %04d.png at the moment [12:32] <ItsMeLenny> oh ok [12:32] <sacarasc> ItsMeLenny: https://trac.ffmpeg.org/wiki/Create%20a%20video%20slideshow%20from%20images See the glob images section. [12:33] <ItsMeLenny> ah [12:33] <ItsMeLenny> -start_number [12:33] <ItsMeLenny> oh, or using glob [12:34] <blippyp> that might of been it - there's a couple of different options for it [12:34] <ItsMeLenny> i'll give them all a go :P [12:34] <ItsMeLenny> tahnks very much [12:34] <brainopia> segmented mp4 can't be played by chrome until it's fully downloaded, is there a way around? [12:36] <Mavrik> use HLS? [12:36] <Mavrik> hmm, that doesn't work on desktop tho [12:38] <brainopia> Mavrik: is there a cross-platform solution? [12:38] <Mavrik> well [12:38] <Mavrik> using non-segmented MP4s with MOOV at the start usually works everywhere [12:38] <Mavrik> MP4s with segments are an iffy tech when it comes to wide support anyway [12:40] <brainopia> yeah, but how to prevent a big moov atom? if it's bigger than 1mb then the startup time will be noticeible [12:41] <brainopia> does ffmpeg allow to control chunk size to use in metadata index? [12:43] <Mavrik> 1MB of moov atom? just how big is your file? [12:44] <minodudd> I have a compilation issue. I'm on a CentOS 6.5 box and I'm following this guide: http://trac.ffmpeg.org/wiki/CompilationGuide/Centos [12:44] <brainopia> Mavrik: ~ 300MB [12:44] <minodudd> I've installed amrnb-devel with yum, but i keep getting this: http://pastebin.com/Gw0vbFa0 [12:45] <sacarasc> minodudd: What does config.log say? [12:47] <Keestu> do we have any API to get details explanation of the error ? like perror() ? [12:49] <minodudd> cnofig.log: http://pastebin.com/eaG0mrF9 [12:50] <minodudd> should I post the entire file? [12:50] <sacarasc> Does /usr/include/opencore-amrnb/interf_dec.h exist? [12:51] <blippyp> Keestu - looks like there's error information in the header files [12:51] <minodudd> yes and no, it's in a very similar location: http://pastebin.com/DpfWbxjp [12:52] <Keestu> for example , avcodec_decode_video2 retValue of avcodec_decode_video2 -1094995529 [12:55] <blippyp> Keestu - I think the libav library deals with that - you might have to look into their documentation for more details on their errors? (I've never tried playing with the libraries myself) but there's a lot more to ffmpeg than just ffmpeg - You will likely have to go to different sources depending on your situation.... [13:00] <minodudd> sacarasc: I noticed there's some more info around those lines in config.log (last 20 lines or so), here's the full log: http://pastebin.com/fCSY406X [13:01] <sacarasc> Maybe it's because you're not using opencore amr? :\ [13:01] <minodudd> huh, very possibly, just blindly installed the first thing that came up in yum search. [13:02] <minodudd> I'll follow that lead. Thanks a bunch! [13:02] <blippyp> Keestu: http://www.ffmpeg.org/doxygen/trunk/libavcodec_2utils_8c_source.html look at line 2225 [13:08] <ItsMeLenny> blippyp, worked nicely for me https://www.youtube.com/watch?v=DmZmesbJekI [13:08] <Keestu> blippyp, thanks for the link. but i m unable to map the error no. :) [13:08] <Keestu> :( [13:12] <blippyp> ItsMeLenny: nice :) [13:13] <blippyp> Keestu:how about http://www.ffmpeg.org/doxygen/trunk/group__lavu__error.html [14:04] <b_jonas> great, I think I might be able to reduce the bug to a broken video I can give you [14:17] <b_jonas> ok, I reduced the file. now I'll review the bug report instructions and file a bug report. [14:38] <b_jonas> there, report filed [17:25] <Logicgate> I need the highest quality iphone preset for converting [17:25] <Logicgate> anybody? [18:08] <esperegu> [ffplay_abuffer @ 0x7f9eb400f780] Setting 'sample_rate' to value '16000' [18:08] <esperegu> [ffplay_abuffer @ 0x7f9eb400f780] Setting 'sample_fmt' to value '(null)' [18:08] <esperegu> [abuffer @ 0x7f9eb400f8a0] Unable to parse option value "(null)" as sample format [18:08] <esperegu> anybody an idea why ffmpeg suddenly does this? it played it before without issues. [18:17] <b_jonas> esperegu: that sounds somewhat similar to what I just bug reported [18:18] <esperegu> b_jonas: the strange thing is that it worked before [18:18] <b_jonas> esperegu: can you ffprobe -show_streams the video and see if the values look right? [18:18] <b_jonas> esperegu: in particular, "options" in that probably does _not_ refer to your command-line options [18:18] <esperegu> b_jonas: I tried without any options. [18:18] <b_jonas> esperegu: exactly, the error is not in your options [18:19] <b_jonas> here, this is the bug ticket for what I just met: https://trac.ffmpeg.org/ticket/3662 [18:20] <esperegu> b_jonas: thx. [18:22] <esperegu> b_jonas: I just try to play it with ffplay. without any options. and I was able to play it before but now not anymore [18:22] <esperegu> b_jonas: plays fine with mplayer [18:23] <esperegu> the fact that it played before drives me insane :-) [18:24] <b_jonas> esperegu: so does ffplay play it right now? [18:24] <esperegu> b_jonas: no [18:24] <l0rd_hex> hi friends, I'm trying to convert some security camera footage for the trial of the CENTURY!! - the footage is not great quality (looks a little fuzzy) and the frame rate is low (robot walking) - right now it's recorded at 4600kb/s (video), yuv444p, 640x480 I'd like to make it a bit smaller in a lossless way (as to not affect the quality but reduce the size). Is there a decent set of guidelines I should be [18:24] <l0rd_hex> looking at or is this best left up to trial (ha!) and error [18:26] <Logicgate> trial of the century? [18:26] <Logicgate> which one may that be [18:27] <l0rd_hex> the year 1900 is suing the year 2000 [18:28] <Logicgate> lol [18:28] <relaxed> l0rd_hex: https://trac.ffmpeg.org/wiki/x264EncodingGuide [18:29] <l0rd_hex> relaxed: thanks, I've been using that as far [18:29] <relaxed> l0rd_hex: use "fmpeg -i input -t 20 ..." to create some 20 second samples [18:29] <relaxed> +f [18:30] <relaxed> l0rd_hex: then what's the problem? [18:31] <l0rd_hex> relaxed: I've been working through it slowly, but I'm impatient :) [18:31] <l0rd_hex> thanks for the -t 20 tip [18:33] <relaxed> try, ffmpeg -i input -t 20 -c:v libx264 -tune grain -preset veryslow -crf 16 -movflags +faststart output.mp4 [18:34] <l0rd_hex> ahh nice, it's noticing duplicate frames [18:35] <relaxed> is it interlaced? [18:35] <l0rd_hex> I don't think so [18:36] <l0rd_hex> wow that looks great [18:36] <l0rd_hex> but holy crap it's slow :) [18:38] <relaxed> that's the cost for great compression [18:38] <l0rd_hex> and great justice [18:38] <l0rd_hex> thank you relaxed :) [18:39] <relaxed> you're welcome [19:10] <debianuser> Hello, I have a 3D player and two mp4 video samples 1280x720 ~30MB each, they both are side-by-side 3D, but first file [https://sendspace.com/file/pdkfr0] is shown as a 3D video, while second one [https://sendspace.com/file/059123] is not. [19:10] <debianuser> How can I find the difference between them and convert second file to be 3D too? [19:11] <Mavrik> debianuser, it's probably missing SBS 3d flag [19:11] <Mavrik> use stereo3d filer to inject it [19:49] <debianuser> Mavrik: Tried `ffmpeg -i smaug.mp4 -vcodec libx264 -acodec copy -crf 20 -vf stereo3d=sbs2l:sbs2l smaug3d.mp4` - file is 18% larger, but is still shown as 2D video, not 3D. Have I used wrong command? [19:57] <jimmy2015> hi there [19:58] <jimmy2015> I've got a little quenstion, anyone who might be able to help me? [19:58] <c_14> just ask and if someone can help, they will [19:59] <jimmy2015> I'm sending out a stream with rtmp://ip:port/file , which i was able to view with red5 [19:59] <jimmy2015> thing is, i would like to read that stream and write it into a directory + streaming it [19:59] <jimmy2015> i thought ffserver 'd be good for that [20:00] <jimmy2015> but i don't get it working [20:01] <jimmy2015> thx, gonna do that [20:02] <jimmy2015> have been on that the whole day, trying to get everything together again .... . [20:13] <jimmy2015> Ok, here we go: [20:14] <jimmy2015> First of all, I'm not directly using ffmpeg, but JavaCV , https://github.com/bytedeco/javacv to get ffmpeg funtionality within an android phone. [20:14] <jimmy2015> which 'sort-of' uses ffmpeg with the very same syntax [20:14] <jimmy2015> The stream gets generated within the java files, [20:14] <jimmy2015> format is set to: flv [20:14] <jimmy2015> the 'so-called' ffmpeg_link is: rtmp://192.168.1.53:1935/streamTest [20:14] <jimmy2015> Using red5 as a media server, i can view the stream wihout a problem [20:14] <jimmy2015> (audio and video). [20:14] <jimmy2015> ATM i can't tell u more about the very special settings here, caus i don't know them .... . [20:14] <jimmy2015> But i hope these information are enough for you to help me :) [20:15] <jimmy2015> This is my ffserver.conf [20:15] <jimmy2015> http://pastie.org/9199842 [20:16] <jimmy2015> Beacsue of not beeing able to change the java stuff right away, I thought someone might have a clue of what to do :) [20:16] <c_14> So the phone is streaming media to ffserver which is $somewhere and then you're accessing the stream from ffserver using red5? [20:18] <c_14> Can you give me a basic rundown of who is generating the media content, and where it's going jump by jump? [20:18] <c_14> And then what isn't working/what you want to work differently. [20:18] <jimmy2015> Nope, sry for that. I'm not using red5 anymore, made some other problems. that's what i wanted to use ffserver for. Receiving the rtmp:// stream, write it into a temp file: File /home/jimmy/ffserver_streaming_test/streamTest1 [20:18] <jimmy2015> so I'd be able to set a max size for that file + streaming it via cam1.asf [20:19] <jimmy2015> to any other player [20:19] <jimmy2015> like vlc [20:20] <jimmy2015> The media content is from my phone's camera, so this is a live stream from my phone [20:21] <c_14> You'll have to tell your phone to stream the content to http://$ffserverip:$port/streamTest [20:22] <jimmy2015> yeha, i thought something like that + tried it, didn't work (the way i did it .... :P ) [20:22] <c_14> or you can have your phone rtmp stream $somewhere and have an ffmpeg process there capture the rtmp stream and send it to http://[..] [20:22] <jimmy2015> so there's no way for ffserver to use that rtmp stream? [20:22] <c_14> since you probably don't want your phone handling the encoding [20:23] <c_14> ffserver only accepts input via http (afaik), but you can run an ffmpeg process on the same device where ffserver is on and just capture the rtmp stream there and send it to ffserver [20:24] <c_14> ffmpeg -i rtmp://[..] http://127.0.0.1:$port/feed [20:25] <jimmy2015> ok, that'd be possible, thanks for the idea :) [20:26] <jimmy2015> what about RTSP? [20:26] <jimmy2015> i'm not realy sure how this is realy diffrent from RTMP, but ffserver can (?) handle this, right? [20:26] <c_14> ye [20:27] <c_14> well, ffmpeg can anyway. [20:28] <jimmy2015> yeha, but i'd like not do encoding again [20:28] <jimmy2015> should not use too much ressources [20:29] <c_14> according to the manual page, ffserver can output to rtsp [20:30] <jimmy2015> output, ok, what about input :P [20:30] <jimmy2015> ? [20:31] <jimmy2015> if there was a generated live video stream via RTMP, which i delivered to the server via rtmp://IP:PORT/FILE what whould happen to that stream, if i change from rtmp:// to rtsp:// ? [20:31] <l0rd_hex> I wish ffmpeg had "enhance" voice recongnition [20:31] <l0rd_hex> http://www.youtube.com/watch?v=Vxq9yj2pVWk [20:31] <c_14> if rtmp works, rtsp should work as well [20:32] <jimmy2015> yeha, rtmp did definitly work with red5 [20:33] <jimmy2015> what about the ffserver.conf http://pastie.org/9199842 shouldn't it work when giving a RTSPPORT 1935 ? [20:34] <jimmy2015> + changing the ffmpeg_link within my java files to rtsp:// ..... [20:40] <jimmy2015> so setting RTSPPORT within fsserver.conf should make ffserver beeing able to accept RTSP streams, just like http? [20:41] <ac_slater> hey all. Sorting throught th ffmpeg code, I don't see that ffmpeg has an H.264 RTP payloader. Am I missing it? [20:45] <c_14> jimmy2015: I just tested, and I think ffserver only accepts http post as input. [20:49] <jimmy2015> @c_14 mh, if so, why is there the possibility to set RTSPPORT ? [20:49] <c_14> ffserver can output to rtsp [20:49] <c_14> ie, it can act as an rtsp server, but not as an rtsp sink [20:50] <jimmy2015> well then, ok :) [20:51] <jimmy2015> what a pitty, i thought this could wokr :) [20:51] <jimmy2015> do you know a small, easy to handle rtmp server, which can act like i thought ffserver could? [20:52] <c_14> you want to accept an rtmp stream and then do what with it again? [20:52] <c_14> restream it? [20:56] <azk> Maybe https://github.com/arut/nginx-rtmp-module ? [20:56] <ldiamond> Anyone knows if FFMPEG is able to rewrite the header of a corrupted .mp4 file? [20:56] <ldiamond> or recover the file or part of the file somehow [20:58] <c_14> Try ffmpeg -i file.mp4 -codec copy out.mp4 [20:59] <ldiamond> [mov,mp4,m4a,3gp,3g2,mj2 @ 0x241b9c0] moov atom not found [20:59] <pzich> do you know if it's corrupted beyond the header? [21:00] <ldiamond> It's a video that was recording and the phone crashed [21:00] <ldiamond> the data is correct, the format isn't [21:07] <ldiamond> using untrunc https://github.com/ponchio/untrunc I managed to get the audio [21:07] <ldiamond> the video is unfortunately all gray now [21:17] <benlieb> First time in the channel. Hope people are here to help... [21:17] <benlieb> I'm trying to use the drawtext filter but I'm not succeeding. [21:17] <benlieb> https://gist.github.com/pixelterra/3b8abe995a2ab5f5be7a [21:18] <benlieb> I keep getting "No such filter: 'drawtext'" even after I uninstalled and recompiled with --enable-libfreetype [21:18] <benlieb> please take a look at the gist to see what I'm doing wrong.... [21:19] <llogan> benlieb: but your ffmpeg configure line lacks --enable-libfreetype [21:19] <benlieb> llogan: it's brew install ffmpeg --with-theora --with-libogg --with-libvorbis --enable-libfreetype [21:19] <benlieb> the line that shows up in brew lacks it.... [21:20] <pzich> the "==> ./configure..." [21:20] <benlieb> right [21:20] <benlieb> but that is coming from brew, even though I passed the options on the command line. [21:21] <benlieb> should I do a manual install apart from brew? [21:21] <pzich> maybe you need to pass them differently? manual install would work better, possibly [21:21] <llogan> if you're lazy just use a build http://www.evermeet.cx/ffmpeg/snapshots/ [21:22] <benlieb> this is my first project with ffmpeg. This is now the 4th time that something I want to use isn't available and ffmpeg has to be recompiled. Is there just a way to install it once with everything available? [21:24] <benlieb> llogan: wouldn't I still have to supply the options to build that? [21:24] <benlieb> or no, those are binaries with everything working? [21:24] <llogan> that link provides a binary [21:25] <llogan> but i'm not sure why brew isn't doing what you want. i've never used it and dont have a OSX machine. [21:26] <llogan> why are you using "--enable-hardcoded-tables"? [21:26] <pzich> it looks like brew is adding that for him [21:26] <benlieb> llogan: that's not an option I supplied [21:28] <llogan> i wonder what "brew info ffmpeg" shows [21:29] <jimmy2015> @c_14 still there? sry, was trying diffrent things. [21:29] <c_14> ye [21:29] <jimmy2015> @c_14 i managed to get the stream via http to the point i can see it a the status.html file [21:29] <benlieb> llogan: https://gist.github.com/pixelterra/bc4b115e310ec484e0c5 [21:30] <jimmy2015> now i get this within my cmdl where i started ffserver [21:30] <jimmy2015> " Codec for stream 1 does not use global headers but container format requires global headers" [21:30] <jimmy2015> i guess (not sure about all that new stuff ) that *.flv is the container? [21:30] <jimmy2015> that's what i set as 'format' [21:30] <llogan> benlieb: i guess you have to give brew "--with-freetype" instead of "--enable-libfreetype" [21:31] <c_14> jimmy2015: didn't you have asf earlier? [21:31] <c_14> anyway, try adding this: AVOptionVideo flags +global_header [21:31] <benlieb> llogan: will try... [21:31] <llogan> hopefully brew will install the required dependency, otherwise you may have to do "brew install freetype" first. [21:31] <c_14> and possibly: AVOptionAudio flags +global_header [21:32] <llogan> benlieb: and fdk-aac is better than faac [21:33] <jimmy2015> |c_14 ope, using the Java ffmpeg stuff, i was able to switch format from my phhones stream, flv, avi, 3gp, mp4 . my ffserver.conf hast has: >stream cam2.asf> in it [21:33] <jimmy2015> where to? [21:33] <jimmy2015> within the stream section of ffserver.conf? [21:33] <c_14> those lines should go into the stream section of your ffserver conf [21:33] <c_14> ye [21:34] <llogan> benlieb: then you'll probably have to change the fontfile path in your drawtext example. the one you're using is from an Arch linux machine [21:36] <bencc2> does it matter what video height I use as long as I use the same aspect ration? [21:36] <bencc2> will height = 720 give me better quality than height = 750? [21:36] <jimmy2015> still the same error [21:36] <jimmy2015> an nothing happens [21:37] <llogan> bencc2: are you downscaling? [21:38] <bencc2> llogan: I'm using the createVideo feature of powerpoint and than transcoding with ffmpeg [21:38] <c_14> jimmy2015: Can you save a short section of what you're recording to file on a pc with ffmpeg installed and then ffmpeg -i file http://ffserver:port/feed ? [21:39] <bencc2> I'm trying to improve the quality before transcoding and wonder if it matters what size I use [21:39] <c_14> Just to make sure it isn't the java thingy. [21:39] <llogan> or at least "ffmpeg -i inputfrompp" [21:40] <bencc2> llogan: my question is more basic. are there preferable sizes? [21:40] <jimmy2015> @c_14 best way to do so? [21:40] <llogan> bencc2: it depends on your input, the encoder, the encoder options, and how you're intending on using the output [21:40] <bencc2> is there a difference between 960X720 and 1000X750? [21:41] <c_14> jimmy2015: I'd just use the phone's built in recorder, then copy the file to a computer somehow. usb cable or something [21:41] <llogan> bencc2: other than the obvious, one may requre upscaling which is usually not recommended or desired [21:41] <bencc2> llogan: even before that. is there a fondumental difference? like multiple or 4 or something similar? [21:41] <bencc2> powerpoint is vector so I guess it doesn't need to downscale or upscale [21:42] <jimmy2015> @c_14 :D [21:42] <jimmy2015> c_14 ok, that'd be an easy way [21:43] <llogan> bencc2: width and height must be divisible by 2 if using libx264 to encode (assuming YUV 4:2:0 planar) [21:44] <jimmy2015> c_14 was able to get ridd of those errors [21:44] <c_14> jimmy2015: what'd you do? [21:44] <jimmy2015> c_14 removed the no audio part [21:44] <jimmy2015> c_14 so i had to state the flags for audio too ... [21:45] <jimmy2015> c_14, any way, there's no error, but screen remains black .... [21:45] <c_14> No process is reporting any errors? [21:45] <jimmy2015> right [21:45] <llogan> bencc2: also you must consider and limitations or restricitons for devices you will be using for playback [21:46] <jimmy2015> this is from my commandline:http://pastie.org/9200168 [21:46] <llogan> jimmy2015: you're using a fork [21:47] <llogan> counterfeits, forks, and other third-party tools are offtopic here [21:47] <llogan> you'll have to use the real thing or get help from the fork [21:51] <bencc2> llogan: thanks [21:52] <benlieb> thank for the help earlier folks [21:52] <benlieb> Here's another question [21:52] <benlieb> The following works to create an empty mpg file. How Would I create an empty mp4 file? [21:52] <benlieb> ffmpeg -t 3 -s 640x480 -f rawvideo -pix_fmt rgb24 -r 25 -i /dev/zero empty.mpg [21:53] <llogan> ffmpeg -f lavfi -i color=c=black:s=640x480 -t 10 output.mp4 [21:58] <llogan> "ffmpeg -f lavfi -i color=c=black:s=640x480 -f lavfi -i aevalsrc=0:0 -t 10 output.mp4" if you want silent stereo audio too [22:03] <benlieb> llogan: tnx! [22:04] <benlieb> I love ffmpeg, but I don't really know how to go about learning it well. It's like magic and smoke and mirrors [22:04] <benlieb> What should I do to learn it well? Just read the entire manual? [22:05] <benlieb> I'm not opposed to that. Or is there a book? [22:10] <c_14> benlieb: start encoding lots of stuff [22:11] <c_14> listening to the problems people who come here have helps a lot as well [22:11] <benlieb> c_14: I'll start with the manual. [22:11] <benlieb> :) [22:15] <thebombzen> benlieb: I learned by toying with it, just transcoding various things and using google when things didn't work. If you need help on some filter then the filtering guide on the wiki is very helpful [22:32] <debianuser> Does anybody knows how to detect container options from existing file? [22:33] <debianuser> I have a 3D player and two ~30MB 1280x720 avc1 mp4 side-by-side 3d video samples. But first file [https://sendspace.com/file/pdkfr0] is detected as a 3D video, while second one [https://sendspace.com/file/059123] is not. How to change second file so that it was also detected as a 3D video? [22:33] <debianuser> I guess there's something magical in that .mp4 container, but what? [22:48] <llogan> benlieb: the wiki is good place to look too https://trac.ffmpeg.org/wiki#CommunityContributedDocumentation [23:19] <benlieb> google isn't helping much on this one. how do I apply a filter that fade to black for the last 1 second? [23:20] <c_14> benlieb: tried the fade filter? [23:21] <benlieb> c_14: looking that up now... [23:21] <c_14> man ffmpeg-filters [23:22] <c_14> search for fade [23:22] <benlieb> c_14: what if I want the last 1 second to fade to black, but I have many videos of different lengths? [23:25] <c_14> You could try giving a negative start_time, but I don't know what that does. just guessing [23:26] <c_14> nvmd, found a bugreport: open, enhancement [23:27] <c_14> Only way would be to get the framecount/total time programatically and use that. [23:30] <c_14> that I know of anyway [23:34] <benlieb> my fade filter is turning the whole thing black. Tried everything I can think to. [23:34] <benlieb> https://gist.github.com/pixelterra/b7418b99af96900665d5 [23:35] <benlieb> can anyone take a look? [23:35] <c_14> You can't have multiple -vf lines [23:35] <benlieb> c_14: oh [23:35] <c_14> The last one overrides all the previous ones. [23:35] <c_14> separete them with commas [23:35] <c_14> -vf filter1,filter2 [23:35] <benlieb> ah [23:36] <c_14> *separate [23:36] <benlieb> with no spaces? [23:36] <temhawk> how can I use an image sequence if the numbering starts at like 300? [23:37] <llogan> benlieb: you can use spaces if you like if you place the filtergraph in quotes: -vf "filter, filter" [23:37] <temhawk> i tried -pattern_type glob but that gets it messed up because the zero-padding isn't consistent [23:38] <llogan> temhawk: can you show some examples of your input file names? [23:38] <c_14> temhawk: you can probably use -start_number 300 [23:38] <benlieb> llogan: what if one of the filters is using double quotes already? [23:38] <c_14> benlieb: escaping is fun [23:39] <c_14> try putting \ before them [23:39] <benlieb> c_14: the whole point of spaces would be to make it more readable :) [23:39] <c_14> The more quotes you have, the more you get to escape. [23:40] <c_14> Like I said, escaping is fun :) [23:40] <benlieb> can I quote you on that? [23:40] <c_14> sure? [23:40] <benlieb> I wonder if there is a ruby library that makes a pretty OO api wrapper for ffmpeg [23:40] <benlieb> \c_14 [23:40] <c_14> This is going to come back and bite me, I'm sure of it. [23:41] <temhawk> c_14: thanks, that did it! llogan: e.g. clip-300.png, clip-301.png, etc, then clip-1000.png, clip-1001.png, etc [23:41] <temhawk> llogan: no zero-padding at all, actually [23:42] <llogan> i would expect glob to be fine with that. -pattern_type glob -i '*.png' [23:43] <temhawk> llogan: nah, it's not. it added all the frames 300-999 at the end of the output video [23:44] <llogan> then i'll be damned. again. [23:48] <benlieb> is there a way to "wrap" text with draw text? If not how do you handle text that may not fit... [23:50] <llogan> easiest way is with textfile option [23:50] <benlieb> llogan: how would that solve the issue exactly? [23:51] <llogan> line breaks [23:52] <benlieb> llogan: I can add that in my script I guess I just have to eyeball it. [23:52] <benlieb> basically I want a biggish title, and underneath, the author's name smaller [23:53] <llogan> you can use two drawtext instances [23:53] <benlieb> If I have multiple lines, how would I know where to put the small line of author's name? [23:55] <benlieb> llogan: by instances you mean comma-separated filters? [23:55] <llogan> yes. you can take into account the fontsize of the first drawtext + some padding to get your placement [23:56] <llogan> or something like that [00:00] --- Fri May 23 2014

1 0

ffmpeg-devel.log.20140521
by burek 22 May '14

22 May '14

[00:11] <kurosu> michaelni, while on the topic of those dsp functions, I can't find symbols for sse2 variants (they doesn't seem to be implemented anyway) [00:11] <kurosu> in particular, I would expect the 16-wide functions to benefit from sse2 [00:11] <kurosu> but maybe they are simply named otherwise [00:16] <michaelni> ff_put_pixels16_sse2 / ff_avg_pixels16_sse2 [00:19] <kurosu> yes, but that's the only ones [00:20] <kurosu> michaelni, I'm thinking of the 2d ones like put_pixels16_x2 [00:21] <kurosu> most of the other ones are defined by called twice the 8-wide version (mmxext/3dnow) [00:27] <michaelni> maybe they where forgotten somewhere between the early cpus that where not faster with sse2 than 2 x mmx and today [00:30] <kurosu> I had the same issue with 3dnow and sse [00:33] <kurosu> do you have any idea where the _x2_ and _y2_ would be used? for the others, it's mostly mpeg2, and it looks like this too [00:51] <kurosu> some mpeg2 decoding going 3.71s -> 3.53s - I guess they are useful [01:11] <jamrial> kurosu: 8388241 runs at 55 cycles each vs 49 cycles each is a ~12% difference, whereas 131042 runs at 449 cycles each vs 430 cycles each is ~4% [01:11] <jamrial> so technically, and unless I'm failing hard at math, avg16 is the negligible one :P [01:13] <kurosu> yes, strictly speaking [01:13] <kurosu> but the bigger cycle difference may actually result in a bigger difference at the end [01:13] <kurosu> that and 4*8x8 << 16x16 so the 16x16 might prevail [01:15] <kurosu> anyway, we are checing hard on <1%, while sse2 cpus have a 10% slowdown because there are no sse2 version of other mpeg2 dsp functions :D [01:54] <BBB> kurosu: pure laziness [01:54] <BBB> kurosu: so no, not intentional [01:55] <BBB> kurosu: I believe the big thing was (at that time) to get some codecs to compile w/o dsputil [02:20] <michaelni> kurosu, _x2 _y2 _xy2 would be used in mpeg1, mpeg2, h263, mpeg4-asp, msmpeg4, some realvideo variants [03:23] <cone-670> ffmpeg.git 03Michael Niedermayer 07master:19e66c7232d9: avcodec/libvorbisenc: dont add the duration to AV_NOPTS_VALUE [04:47] <cone-670> ffmpeg.git 03Christophe Gisquet 07master:d1a32c3f4912: x86: kill fpel_mmx.c [04:48] <jamrial> great commit message there :P [07:27] <kurosu> michaelni, ok, wasn't sure half pel was changed away from bilinear in mpeg4 [07:27] <kurosu> (or h263) [07:28] <michaelni> mpeg4 asp supports qpel but its optional and not every video uses it [11:05] <nevcairiel> you guys broke msvc builds again [11:06] <nevcairiel> if you're messing with inline asm, test a --disable-inline-asm build after plz :P [13:47] <cone-225> ffmpeg.git 03Hendrik Leppkes 07master:cd9e08e110b9: hpeldsp: fix build without inline asm [13:47] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:3d4e365073c0: avcodec/x86/hpeldsp_init: remove redundant if() [14:13] <BBB> in a way I'm so happy with all the msvc support nowadays [14:13] <BBB> even if I don't actually use it :-p [14:14] <BBB> that was probably the most political issue ever [14:14] <BBB> progress yay [15:08] <compn> BBB : was it? i guess msvc had a lot of roadblocks , c89 > c99 stuff [15:08] <compn> inline asm > external yasm [16:07] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:f0211f4189a5: ffprobe: Use av_malloc_array() [16:07] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:606a49d2e6f1: avcodec/atrac3: Use av_mallocz_array() [16:07] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:e61055fd8ba3: avcodec/atrac3plusdec: Use av_mallocz_array() [17:40] <cone-225> ffmpeg.git 03Dale Curtis 07master:ef0c466a0fef: Replace assert with AVERROR when recode_subtitle called w/o ICONV. [17:40] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:91a8262b9dbc: avcodec/utils: add error message for the recode_subtitle() without iconv case [19:02] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:7f7080dc73df: avformat: add av_stream_get_end_pts() [19:02] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:1f359c85599b: ffmpeg: use av_stream_get_end_pts() [19:12] <ubitux> michaelni: can't we replace most of the "last" stuff (pts & friends) in AVStream by a pointer to the last packet (since now it's ref counted)? [19:14] <michaelni> i guess that should be possibe, does this simplify something ? [19:15] <ubitux> i don't know, i was just wondering about the need to expose a new av_stream_get_end_pts() [19:16] <ubitux> since api users can already keep the last packet with ref counting [19:16] <ubitux> and iirc they're is not only the "end pts" [19:16] <ubitux> i remember seeing other "last" stuff [19:16] <ubitux> but i'm not familiar with that code [19:18] <michaelni> well, api users cant get the last pts by just keeping the last packet before muxing when that packet had a AV_NOPTS_VALUE pts [19:18] <michaelni> it would require some api from libavformat to keep a modified packet [19:18] <ubitux> ok [22:21] <cone-225> ffmpeg.git 03Michael Niedermayer 07master:ce4d91ba2a97: sws: document color range >8bit oddity [23:10] <cone-225> ffmpeg.git 03Carl Eugen Hoyos 07master:afaa39b46dbc: Don't overwrite codec_id with 0 in ff_mp4_read_dec_config_descr(). [23:10] <cone-225> ffmpeg.git 03Anshul 07release/2.1:83f58778cb6a: ffprobe: fix crash happening because of new streams occuring [23:10] <cone-225> ffmpeg.git 03Anshul 07release/2.2:acafd1814e68: ffprobe: fix crash happening because of new streams occuring [23:57] <cone-225> ffmpeg.git 03James Almer 07master:7b05267239ed: x86/dsputil: port clear_block functions to yasm [23:57] <cone-225> ffmpeg.git 03James Almer 07master:80ee2dfcf674: x86/dsputil: port ff_put_signed_pixels_clamped_mmx to yasm [00:00] --- Thu May 22 2014

1 0

ffmpeg.log.20140521
by burek 22 May '14

22 May '14

[01:19] <kingsob> does encoding H.264/AVC using ffmpeg benefit from GPU's? I'm trying to determine if it would be beneficial to use Amazon's GPU optimized instances... [01:20] <relaxed> no [01:48] <iive> kingsob: ffmpeg uses libx264 library, so you can also ask there. afaik, you'll get the same answer. [02:02] <Zeranoe_> I'm trying to get better playback out of libvpx, but it still seems to be steaming at like 1fps. I'm not even seeing full CPU usage [02:03] <llogan> VP8 or 9? [02:03] <GEEGEEGEE> I have a live video stream that has x264 video and aac audio, and I want to keep the same video data, but change the protocol to rtmp and -f flv. Is there a way to do this, or would I need to transcode it? I want x264 and aac as the output, but in flv and being sent to my rtmp server. [02:03] <Zeranoe_> llogan, What does -c:v libvpx default to [02:04] <llogan> vp8 [02:04] <kingsob> iive: yeah thanks, google seems to say no as well :( [02:05] <Zeranoe_> llogan, vp9 is no better, except now I get realtime buffer overflow messages [02:06] <Zeranoe_> Command is: ffmpeg -f dshow -pix_fmt yuyv422 -s 640x480 -r 30 -i video="USB 2.0 UVC HD Webcam" -r 30 -c:v vp9 -an -f webm - | ffplay - [02:06] <Zeranoe_> [02:06] <c_14> GEEGEEGEE: ffmpeg -i input -codec copy -f flv rtmp://url should do the trick [02:07] <llogan> Zeranoe_: vp9 is too slow to be usable last time i checked more than a month ago [02:07] <Zeranoe_> vp8? Is theora any better? [02:07] <llogan> no [02:08] <llogan> how do you know it's not the input device that is the issue? [02:08] <Zeranoe_> I've seem some examples of people streaming live through html, I have it working but the codec speeds are killing me [02:08] <Zeranoe_> llogan, ffplay handles it fine without lag [02:08] <Zeranoe_> ? [02:09] <Zeranoe_> No need for pb [02:09] <llogan> you must be new here [02:11] <llogan> i'm referring to your libvpx encoding speed issue. not the "ffplay can play input device just fine". [02:12] <Zeranoe_> lol im not new here [02:12] <llogan> i know. [02:12] <llogan> but without your actual command i don't know what you're doing [02:14] <Zeranoe_> Well i already said the command, just not the output. I'll add [02:14] <llogan> your command had vp9. which is uselessly slow. [02:15] <Zeranoe_> ffmpeg -f dshow -pix_fmt yuyv422 -s 640x480 -r 30 -i video="USB 2.0 UVC HD Webcam" -r 30 -c:v vp8 -an -f webm - | ffplay - [02:17] <Zeranoe_> im really just looking for guidelines to improve vp8 encoding speed [02:18] <llogan> i guess if you don't give a shit about quality you can add the private option "-quality realtime", but I'm fairly ignorant of this encoder. [02:20] <llogan> there is http://trac.ffmpeg.org/wiki/vpxEncodingGuide but it does not go into details about encoding speed [02:21] <GEEGEEGEE> c_14, thanks [02:21] <GEEGEEGEE> would mp4 give better view quality at all? its looking a little blurred :/ [02:21] <llogan> Zeranoe_: also -speed option. i guess i should learn how to use this encoder. [02:22] <llogan> GEEGEEGEE: are you stream copying with -codec copy? [02:22] <GEEGEEGEE> yes [02:22] <GEEGEEGEE> it should look the smae right [02:22] <llogan> yes [02:23] <GEEGEEGEE> oh i was on a SD channel instead of HD lol [02:23] <GEEGEEGEE> nvm [02:26] <llogan> Zeranoe_: http://www.webmproject.org/docs/encoder-parameters/#2-encode-quality-vs-spe… [02:32] <Zeranoe_> llogan, Thats helpful, can I pass some of those options direct to libvpx though FFmpeg, or will I need to look into pipes [02:36] <llogan> Zeranoe_: you can use it via ffmpeg, but the option named may be different. see "ffmpeg -h encoder=libvpx" [08:25] <esperegu> I reinstalled my pc and when I run an old script&binary I now get the error: Specified probe size value 32 cannot be < 2048. I thought 32 is the minimum. anybody an idea what might cause this? [11:21] <Katharsis> i have two notebooks with the same Debian version system (updated / upgraded) [11:21] <Katharsis> one of them has ffmpeg 0.8.10, another ffmpeg 1.0.9 [11:22] <Katharsis> (i don't know why) [11:22] <sacarasc> Are you using different repos? [11:22] <Katharsis> nope [11:22] <Katharsis> I check it [11:22] <Katharsis> but this is not a proble [11:22] <Katharsis> not a problem* [11:22] <Katharsis> the problem is with recording screen with voice [11:22] <sacarasc> Different mirrors of the same repo? [11:22] <Katharsis> i used the same command on two of them [11:23] <Katharsis> and one notebook (with ffmpeg 1.0.9) gives me mp4 with voice delay [11:23] <Katharsis> i show you my command [11:24] <Katharsis> ffmpeg -f alsa -i pulse -f x11grab -r 30 -s 1366x768 -i :0.0 -vcodec libx264 -r 25 -preset ultrafast -threads 4 -y -sameq $1 [11:26] <Katharsis> delay is not much when i used: [11:26] <Katharsis> ffmpeg -f alsa -i pulse -f x11grab -s 1366x768 -i :0.0 -vcodec libx264 -preset ultrafast -threads 0 -r:v 30 -vsync 2 -async 1 $1 [11:26] <Katharsis> but it is still not the same result like in first notebook [14:40] <sdfgsdfh> Does anybody know the difference between rgb24 and rgb0? [14:40] <sdfgsdfh> Is rgb0 really 0 bpp? [14:44] <klaxa|work> that sounds only useful for blank videos [14:44] <klaxa|work> is it listed in ffmpeg -pix_fmts? [14:45] <sdfgsdfh> well, FFV1 wants to default to it [14:46] <sdfgsdfh> I was joking about the 0 bpp thing [14:46] <sdfgsdfh> there is no way it's hat [14:46] <sdfgsdfh> *that [14:47] <klaxa|work> is it listed in ffmpeg -pix_fmts though? [14:47] <klaxa|work> that should give you some clue maybe [14:47] <sdfgsdfh> so we've got... 0rgb, rgb0, 0bgr, and bgr0, all listed [14:48] <sdfgsdfh> they are listed as 24 bpp [14:49] <sdfgsdfh> and 3 channels [14:49] <sdfgsdfh> WTF, that's like the same as RGB24, BGR24, etc. [14:49] <sdfgsdfh> same flag too [14:50] <sdfgsdfh> but I don't know what kind of subsamplin they might have [14:50] <sdfgsdfh> I don't get the "0" part [14:50] <sdfgsdfh> FFV1 does not supportthe "24" formats for some reason [14:51] <sdfgsdfh> I am actually avoiding FFV1 right now for that reason [14:56] <sdfgsdfh> this pixel format limitation has been bugging me for days [14:56] <sdfgsdfh> I would like to get a definitive answer on FFV1's pixel format limitations [14:56] <sdfgsdfh> I do not know who to ask [14:57] <sdfgsdfh> I need to sleep but the chat will fill up so fast... [14:59] <relaxed> sdfgsdfh: ffvhuff does [15:00] <sdfgsdfh> ffvhuff? [15:00] <sdfgsdfh> not huffyuv? [15:01] <JEEB> sdfgsdfh, if you really want to see what pix_formats a video format supports, you can check the encoder definition in the matching libavcodec/***.c file [15:01] <relaxed> that too [15:01] <JEEB> it has a list of the supported pix_fmts [15:01] <sdfgsdfh> I can't understand code [15:01] <JEEB> it's a simple list :P [15:01] <sdfgsdfh> ahg [15:01] <relaxed> ffmpeg -h encoder=ffv1 [15:02] <sdfgsdfh> I'll try that, but that doesn't tell me wat the formatsactually are [15:03] <sdfgsdfh> ok so there's bgr0 and bgra [15:03] <JEEB> https://github.com/FFmpeg/FFmpeg/blob/master/libavcodec/ffv1enc.c#L1336 [15:03] <JEEB> is this really that hard to see as a list :P [15:03] <JEEB> the AV_PIX_FMT_* listing [15:03] <JEEB> NONE being the "end" of the list [15:04] <JEEB> but yes, if -h encoder=xxx outputs the same [15:04] <JEEB> that's probably simpler :P [15:04] <sdfgsdfh> well almost [15:04] <sdfgsdfh> the constants are more descriptive [15:05] <sdfgsdfh> very telling actually [15:05] <sdfgsdfh> it looks like FFV1 has to pad out 32 bits per pixel, period [15:05] <JEEB> anyways, the difference between huffyuv and ffvhuff is that the former IIRC tries to be compatible with "Ye Olde Huffman" [15:05] <JEEB> while ffvhuff adds features etc [15:06] <sdfgsdfh> of course huffyuv is old [15:06] <sdfgsdfh> I don't use it because of bad compression ratios, but it makes a good fallback [15:06] <relaxed> ffv1 is much slower to encoder/decode [15:06] <sdfgsdfh> I care more about decoding speed [15:06] <sdfgsdfh> for playability [15:06] <sdfgsdfh> so utvideo is awesome [15:07] <sdfgsdfh> not triedffvhuff yet [15:07] <sdfgsdfh> didn't know it existed [15:07] <sdfgsdfh> I assume it supports various pixel formats [15:08] <sdfgsdfh> not as many, but the important ones are there [15:08] <sdfgsdfh> including rgb24 [15:08] <relaxed> yes, see ffmpeg -h encoder=ffvhuff [15:08] <JEEB> libx264's RGB encoding might be useful, too (as lossless). Try out whatever can do well enough on your hardware [15:08] <JEEB> ffvhuff/ut video are going to be pretty damn fast since they're so simple [15:09] <sdfgsdfh> I heard libx264 with RGB is a minefield [15:09] <JEEB> depends in what way [15:09] <sdfgsdfh> well, taking a RGB24 source [15:09] <JEEB> yes? [15:09] <sdfgsdfh> putting it straight into a libx264 mkv [15:09] <sdfgsdfh> trying to be 100% lossless [15:09] <sdfgsdfh> jut something I read on stackoverflow [15:10] <JEEB> yes, by default it will use pix_fmt yuv420p unless you use the RGB-specific encoder name [15:10] <JEEB> which is libx264_rgb or so [15:10] <sdfgsdfh> ORLY [15:10] <JEEB> for playback compatibility with hardware devices and such :P [15:10] <JEEB> yarly [15:10] <sdfgsdfh> I am learing so much [15:10] <sdfgsdfh> so that's like 2 or 3 other encoders I can use [15:10] <JEEB> and then with 8bit libx264 just use -crf 0 (or quantizer zero) to set lossless [15:11] <sdfgsdfh> I know [15:11] <JEEB> quantizer zero also works with >8bit libx264 [15:11] <JEEB> okies [15:11] <sdfgsdfh> well, there is no libx264_rgb [15:12] <relaxed> libx264rgb [15:12] <sdfgsdfh> why yes [15:12] <sdfgsdfh> and what a long description of functionality it gives me [15:13] <sdfgsdfh> is it strictly lossless or both lossy and lossless? Does it default to RGB24? [15:13] <relaxed> -crf 0 -pixfmt rgb24 [15:13] <JEEB> there's only one type of RGB in libx264 [15:13] <relaxed> -crf 0 -pix_fmt rgb24 [15:14] <sdfgsdfh> it looks like there's bgr24 as well [15:14] <JEEB> and as always, there's speed/compression presets and you can encode both lossy and lossless [15:14] <sdfgsdfh> I do prefer faster decoding than encoding [15:14] <sdfgsdfh> but I tend to shy away from those speed presets [15:15] <JEEB> well, decoding will anyways be pretty fast [15:15] <JEEB> the presets are just if you need to either make the encoding faster (f.ex. in realtime scenarios) or if you need to compress more [15:15] <sdfgsdfh> I see [15:16] <sdfgsdfh> but I assume the compression ratio is pretty much the best ffmpeg has to offer [15:16] <sdfgsdfh> libx264 with 4:2:0 gave me some amazing lossless compression [15:16] <sdfgsdfh> -crf 0 actually makes a file smaller than -crf 3 [15:17] <sdfgsdfh> for some werd reason [15:17] <sdfgsdfh> *weird [15:17] <JEEB> with specific kinds of sources it indeed is better to code something lossless than with high bit rate lossy [15:17] <sdfgsdfh> I'm not going to ask why :P [15:22] <sdfgsdfh> Encoding now... [15:22] <sdfgsdfh> BTW, I am operating under the assumption that wavpack defaults to lossless [15:23] <sdfgsdfh> WB telex [15:30] <sdfgsdfh> wow, that seemed to do away with 1/3 of the file size VS utvideo [15:36] <sdfgsdfh> OK so that libx264rgb (rgb24) video is unplayable in VLC and glitches out in smplayer [15:36] <sdfgsdfh> in VLC it's all green [15:37] <sdfgsdfh> I can't believe I pulled an all-nighter for codecs and pixel formats [15:39] <klaxa|work> happens all the time [15:39] <c_14> True words, true words. [15:40] <sdfgsdfh> especially considering those who frequent this channel :) [15:43] <JEEB> sdfgsdfh, probably old versions of things. [15:43] <JEEB> everything current should be fine with it [15:43] <JEEB> as in, current since... a year or two ago [15:43] <JEEB> I think 4:4:4 and RGB first got used around ~2011 or so [15:44] <sdfgsdfh> these are he latest stabe builds of VLC and smplayer [15:44] <sdfgsdfh> I think VLC is the more up-to-date version [15:45] <sdfgsdfh> as of like 2 months ago [15:45] <sdfgsdfh> smplaer may be a year old [15:45] <sdfgsdfh> and I think RGB has been around for some time in the form of uncompressed AVIs [15:46] <sdfgsdfh> I used that format all the time when exporting from Blender back in high school [15:46] <JEEB> ugh, there's a commit from april in VLC that fixes planar RGB :V [15:46] <JEEB> yeah, it's mostly a case of players thinking that H.264 can only be either YCbCr or 4:2:0 YCbCr [15:46] <JEEB> and mplayer can just be old in any case [15:47] <sdfgsdfh> all true [15:47] <sdfgsdfh> regarding that commit [15:47] <JEEB> if you try a recent build of mpv or in DirectShow's LAV Filters it should just work though [15:47] <sdfgsdfh> is there a specific day [15:47] <JEEB> http://git.videolan.org/gitweb.cgi?p=vlc.git;a=commit;h=0e10c9d6bf06785eb39… [15:47] <JEEB> anyways, I use mpv as a "reference" in many cases [15:48] <sdfgsdfh> Mar 20 [15:48] <sdfgsdfh> well [15:48] <sdfgsdfh> I don't know what mpv is [15:48] <JEEB> mplayer -> mplayer2 -> mpv [15:48] <JEEB> http://mpv.io/ [15:50] <spaam> mpv is the new black? [15:50] <JEEB> well, mplayer2 pretty much died :P [15:51] <JEEB> since everyone but uau moved to mpv [15:51] <JEEB> and mplayer, while still being kept on life support, is pretty much a lost cause in my opinion [15:51] <spaam> why? so people dislikes uau ? [15:51] <JEEB> mostly because uau wasn't really co-operative at the time [15:51] <JEEB> it took like two years of things to people to fork [15:52] <JEEB> so chances were given and various people did ragequit before a fork was made [15:55] <sdfgsdfh> Well, ffvhuff is a disaster when it comes to file size so I won't be using that again [15:58] <sdfgsdfh> Wow, MPV has a .com file [15:58] <sdfgsdfh> what is this, 1985? [15:59] <JEEB> no [15:59] <JEEB> it's so that you don't get a command prompt [15:59] <JEEB> in certain ways you launch it [15:59] <sdfgsdfh> I did not know that [16:00] <JEEB> well, the COM binary basically just is a "GUI" binary so it doesn't spawn a command prompt [16:00] <JEEB> the exe one is a "cli" binary that spawns a command prompt [16:00] <JEEB> and I think one of them, com or exe, comes before the other in the listing of executable things in PATH [16:00] <sdfgsdfh> the COM is 13 KB, the EXE is 26 MB and has an icon [16:00] <JEEB> yes [16:00] <JEEB> the COM just calls the exe [16:01] <sdfgsdfh> I see [16:01] <JEEB> it's a hack to not get a command prompt under certain circuimstances, basically [16:01] <sdfgsdfh> so since there is no built-in GUI (right?) I open my videos wth said program? [16:02] <JEEB> there's the on-screen controls, but that's it [16:02] <sdfgsdfh> I'm fine with that [16:04] <sdfgsdfh> well, it plays everything perfectly [16:04] <sdfgsdfh> amazing [16:05] <sdfgsdfh> kind of laggy on a 720p libx264rgb file though [16:05] <sdfgsdfh> I think it may be the format [16:05] <JEEB> or the video renderer [16:05] <JEEB> there can be plenty of reasons [16:06] <sdfgsdfh> the video renderer that is built into MPV? [16:07] <sdfgsdfh> oh yeah looks like t [16:07] <JEEB> yeah, the one that gets used by default. depending on your hardware [16:07] <sdfgsdfh> it's slower to play utvideo! [16:07] <sdfgsdfh> about smplayer speeds [16:07] <sdfgsdfh> unsurprisingly [16:07] <JEEB> well naturally :P ut video is simple huffman [16:07] <JEEB> anyways, you can make H.264 faster to play back in various ways [16:08] <JEEB> -tune fastdecode being probably the thing that works in all presets [16:08] <sdfgsdfh> what I mean is that VLC handles utvideo better [16:08] <JEEB> well, of course :P it didn't have a bug with handling it :P [16:09] <sdfgsdfh> OK so basically MPV is faster with x264 for some reason [16:10] <sdfgsdfh> does fastdecode do things like disable deblocking and drop B frames or whatever? [16:10] <sdfgsdfh> it has to be "fast" somehow [16:10] <JEEB> it disables deblocking and CABAC [16:10] <JEEB> b-frames aren't used in lossless to begin with [16:10] <sdfgsdfh> CABAC? [16:11] <sdfgsdfh> there still is lossy x264 [16:11] <sdfgsdfh> I don't know what CABAC is [16:11] <JEEB> yes, but we were talking about lossy, no? :P [16:11] <JEEB> CAVLC and CABAC are the two modes of compression you have in H.264 [16:11] <sdfgsdfh> I'm just trying to cover my ignorance :P [16:11] <JEEB> in HEVC they removed CAVLC [16:11] <JEEB> and left only CABAC [16:11] <JEEB> CABAC is basically slower to encode/decode, but compresses better [16:11] <JEEB> CAVLC is more like huffman in a way [16:12] <sdfgsdfh> so you can encode or decode in either one arbitrarily without problems? [16:13] <sdfgsdfh> I heard HEVC/x265 is indeed slower for some reason [16:13] <JEEB> well, it is designed 10 years later :P [16:14] <JEEB> also you can still make a fast (and crappy) HEVC encoder, just like you could with H.264 [16:14] <JEEB> it just has more features that can be harder on the encoder [16:14] <JEEB> decoder-wise it shouldn't be much slower [16:14] <klaxa|work> i did some encodes with libx265 on my server [16:15] <klaxa|work> 90 seconds 1080p anime source took like 30 minutes with 8 threads, filesize was like ~150 mb with x264 to ~37 mb with x265 [16:15] <sdfgsdfh> I assume they turned out well [16:15] <klaxa|work> i don't think i remember the settings [16:15] <klaxa|work> but decoding in real-time on my crappy laptop was impossible [16:15] <JEEB> klaxa, that sounds like the incorrect type of testing :P I mean, the CRF values are not and probably never will be the same [16:16] <JEEB> but currently x265 only wins in very very low rate scenarios [16:16] <klaxa|work> heh yeah pretty much pointless waste of cpu time [16:16] <JEEB> I mean, in cases where both files look crappy [16:16] <klaxa|work> but the server is in idle most of the time anyway [16:16] <JEEB> but the HEVC encode looks a bit better [16:17] <klaxa|work> i couldn't see a visual difference between the encodes though [16:17] <klaxa|work> http://dedi.klaxa.eu/public/op_x264.mkv vs. http://dedi.klaxa.eu/public/op_hevc.mkv [16:17] <JEEB> then you were testing in a bad way [16:18] <klaxa|work> most likely [16:18] <JEEB> anyways, match up file sizes (x265 is slower so encode with CRF with it first) and basically max out both encoders (preset placebo, ref/bframes 16) [16:18] <JEEB> I think around crf 32 was pretty good for this for x265 in february or so [16:19] <klaxa|work> it was more a test to satisfy my curiosity as to how low the filesize can go with x265 with no noticeable difference [16:19] <klaxa|work> which is absolutely subjective in the first place [16:20] <JEEB> that's OK too [16:24] <sdfgsdfh> so what if you wanted to compress a 2hr long movie at 640x360 to 20 MB with x265? [16:25] <sdfgsdfh> I've done some tests with x264 and I can say that I am OK with crf 35 in many cases [16:25] <sdfgsdfh> I have seen worse bitrate abuse [16:26] <sdfgsdfh> and I assume some day there will be x266 which understands the context of video and encodes subjective information based on human consciousness [16:28] <JEEB> the CRF values between x264 don't match [16:28] <JEEB> *between x264 and x265 [16:28] <JEEB> so you can't really compare them [16:28] <JEEB> but with some random 720p animoo I tested in february CRF 32 was ~415kbps [16:28] <JEEB> which was quite nice to see good/bad points of both encoders [16:31] <sdfgsdfh> I am not realy concerned with the CRF values of x265, but the artifacts that crop up with x264 do bother me, especially at super low bitrates [16:32] <sdfgsdfh> there is a jump at each keyframe transition because there is no blending, for instance [16:33] <sdfgsdfh> x265 is still under development so it's probablytoo early to judge anyways [16:44] <sdfgsdfh> it looks like mpv does no have a -tune option [16:44] <klaxa|work> why should it have one? [16:45] <sdfgsdfh> because apparetly mplayer has it or something [16:45] <sdfgsdfh> JEEB suggested using it [16:45] <klaxa|work> -tune is used for encoding, mpv does decoding [16:45] <JEEB> uhh [16:45] <JEEB> it's an encoder option, yes [16:45] <JEEB> for ffmpeg [16:45] <sdfgsdfh> h [16:45] <sdfgsdfh> oh [16:46] <JEEB> mpv does also encode, but it uses lavf/lavc on the background, and using it will make you read some documentation :P [16:46] <sdfgsdfh> that would make moe sense [16:46] <sdfgsdfh> I wasn't really planing on using mpv as an encoder at this point [16:46] <klaxa|work> oh? they put the mencoder parts back into mpv? [16:46] <JEEB> no [16:46] <JEEB> it's new code [16:46] <klaxa|work> ah [16:47] <sdfgsdfh> so how does -tune fastdecodework with lossless? [16:47] <JEEB> it switches from CABAC to CAVLC [16:48] <JEEB> and some other things that make compresson efficiency somewhat worse [16:48] <JEEB> but that speed up decoding [16:48] <sdfgsdfh> how much worse? [16:48] <JEEB> depends on the stuff, and depends on a whole lot of things [16:48] <sdfgsdfh> well, in a typical use case [16:48] <JEEB> I can't just come up with a random number :P [16:48] <sdfgsdfh> ok [16:48] <JEEB> in any case, it should still be better than ut video for example [16:49] <JEEB> for various reasons [16:49] <sdfgsdfh> oh wow ok [16:49] <sdfgsdfh> well, ut video is still way better than huffman [16:49] <JEEB> because ut video is in the end so simple :P [16:49] <sdfgsdfh> it was made by one Japanese guy, so it would have to be: :) [16:49] <JEEB> well ut video is just huffman with a simple median prediction in most cases :P [16:49] Action: JEEB wrote the avcodec ut video encoder [16:50] <sdfgsdfh> you actually re-implemented utvideo all yourself for ffmpeg? [16:50] <JEEB> yes [16:50] <sdfgsdfh> you are a hero! [16:51] <sdfgsdfh> you did a wonderful thing [16:51] <sdfgsdfh> it's such a reliable codec compared to others [16:51] <sdfgsdfh> it's got lots of pixel formats, a good compression ratio, and fast decoding [16:52] <sdfgsdfh> and you didn't even invent it, so I can imagine it was lots of work [16:52] <Venti> cabac produced like 20% better bitrate for hevc back when cavlc was still being proposed [16:53] <JEEB> yeah, it was and still is (?) in HM, too [16:53] <JEEB> at least leftovers of the code were all around HM at one point [16:53] <sdfgsdfh> HM? [16:54] <JEEB> ther reference implementation of HEVC [16:55] <Venti> I think the decoder side still has them.. no idea if it works though [16:55] <JEEB> wouldn't be surprised if it didn't [16:55] <JEEB> also at this point removing CAVLC just makes sense [17:06] <MrNaviPacho> RPM Fusion is listed on the downloads page for CentOS, however, it looks like they are on 0.x, is there somewhere I can get RPMs for 2.x? [17:08] <relaxed> MrNaviPacho: if it's 64bit then http://johnvansickle.com/ffmpeg/ [17:08] <relaxed> which is a static build, not an rpm [17:12] <MrNaviPacho> relaxed: that's fine, not sure how I missed that [17:12] <MrNaviPacho> thanks [17:23] <MrNaviPacho> relaxed: does this not have aac support? (Unknown encoder 'libfaac') [17:24] <sdfgsdfh> you can use the built-in experimental aac encoder [17:28] <MrNaviPacho> ok, thanks [17:28] <sdfgsdfh> np, sorry I couldn't come up with the command line [17:32] <sdfgsdfh> x264 with fastdecode played smoother than anything, to bad VLC is a little behind right now [17:33] <sdfgsdfh> no color loss either so that's good [17:58] <svm_invictvs> Can ffmpeg do chroma key postprocessing? [18:04] <sdfgsdfh> maybe through a series of steps with libavfilter or something, but it's complicated [18:05] <sdfgsdfh> I'm not sure if there's any pre-mae keying filter [18:06] <sdfgsdfh> if you want to do keying or compositing manually, I would recommend Blender. If you need it in real-time, I'm afraid I can't be of much help. [18:30] <iive> sdfgsdfh: ffmpeg should support alpha channel [18:44] <sdfgsdfh> I'm feeling shell shocked from al these angry neighbors slamming doors [18:45] <sdfgsdfh> some kind of domstic dispute going on for sure [18:45] <sdfgsdfh> time for a nap [18:46] <sdfgsdfh> iive: ffmpeg does support encoding videos with alpha channels, but I don't know anything other than that [18:46] <iive> that's almost as much as I know too. [18:46] <iive> some filters might support it :) [18:46] <sdfgsdfh> yes I suppose so [18:47] <sdfgsdfh> I've never used filters with ffmpeg [18:47] <sdfgsdfh> there's probably an overlay filter [18:47] <sdfgsdfh> if real-time keying is an issue though, an alpha channel is the least of your concerns [18:48] <sdfgsdfh> rather, you'll need per-pixel math to blend one video frame into another [18:48] <iive> there is definitely and overlay filter. [18:48] <iive> and/an [18:48] <sdfgsdfh> cool [18:48] <iive> that's why I am talking about alpha channel [18:49] <iive> it is a whole plane with an alpha value for each and every pixel. [18:49] <sdfgsdfh> yep [18:49] <blippyp> that's basically what I did once - I had to make an algorithm to get the keying to work - I think frei0r has a filter for it though (haven't tried it yet) [18:52] <sdfgsdfh> a good start would be something like A=(R+B)/2-G [18:53] <sdfgsdfh> I don't know if it would be more efficient to use the alpha channel or to peek and poke the pixels directly [18:54] <sdfgsdfh> one trick you could do is using a lower-resolution image to generate your alpha map if you're using HD [18:54] <blippyp> probably an alpha channel - I think in the end, that's how I went about it - I first started with an algorithm and then migrated to alpha channel approach - it's been a while since I did it, so I honestly don't remember exactly how I did it. But it worked reasonably well. An actual filter designed to do the job is probably better tbh.... [18:55] <blippyp> Yeah - I also generated a low res image as well - and decreased the colors to simplify to generate the mask i think.... [18:55] <sdfgsdfh> decreased the color bitrate? [18:56] <sdfgsdfh> the edges mightget weird [18:57] <blippyp> it worked for me... not sure why - I'm no expert on this kind of thing - all I know is that I was pretty happy with the result - and it was much faster than analyzing each pixel... [18:57] <sdfgsdfh> I'm sure [18:58] <sdfgsdfh> an the filtering probably helped to blur the edges a bit, which one might do anyways [18:58] <blippyp> also, my wants were simple - I was converting the image into a 'cartoon' in the end anyways... for a real-time image the approach I took probably isn't very useable... [18:58] <svm_invictvs> sdfgsdfh: I'm not trying to do it in real time [18:59] <blippyp> I'd look at the green screen technique used in the frei0r filter - it will likely give you the details you want for a much better green screening [18:59] <svm_invictvs> sdfgsdfh: What about just decomposing the video into frames, then i can write code to dot he chromakey myself? [18:59] <blippyp> oops - didn't mean real-time - more of a 'real life' image [18:59] <blippyp> with many colors and high res etc.... [19:00] <blippyp> that could work, but is also probably quite slow - if you're willing to code - I would seriously take a look a frei0r - it will probably help you tons [19:07] <sdfgsdfh> I can't stay here any more [19:07] <sdfgsdfh> I'm sorry [19:36] <mjuszczak> Can anyone recall a bug fixed in ffmpeg where you would pass a filename on the command line, particularly for srt files, and it would say "file not found", yet if you copied the same exact path and ran stat/ls on it, it's listed? This would have been a very old bug. [19:38] <pzich> http://stackoverflow.com/questions/8672809/use-ffmpeg-to-add-text-subtitles ? [22:01] <midzer> hi nypenick [22:01] <nypenick> hi midzer [22:05] <nypenick> hey ffmpeg guys.. i've got a simple question. I'm trying to seek while decoding audio. So i'm calling av_seek_frame and after that I go on decoding with av_read_frame and avcodec_decode_audio4. [22:06] <nypenick> this code works but the seeking is not very precise [22:07] <nypenick> it always ~1 sec befor or behind the actual seek position [22:11] <pzich> nypenick: can you use -ss? [22:12] <nypenick> what does -ss mean? [22:12] <pzich> it's an ffmpeg flag to seek, and depending on whether you place it before or after your input files (with '-i') it seeks quickly or precisely [22:12] <pzich> https://trac.ffmpeg.org/wiki/Seeking%20with%20FFmpeg [22:12] <c_14> pzich: He's using the library not the binary. [22:13] <nypenick> correct I#m using the library [22:13] <pzich> ah, right [23:02] <sad> hi guys! is there a way to recompile just the libavcodec in ffmpeg? [23:11] <sad> if i modify source code from mjpegdec.c, is there a way to compile/build only this part of libavcodec? [23:25] <iive> if you run make second time, it would only compile the changed .c file and then link again all that depend on it. [23:25] <iive> that is, libavcodec, ffmpeg/play/server etc. [00:00] --- Thu May 22 2014

1 0