July 2013 - Ffmpeg-devel-irc - lists.ffmpeg.org

ffmpeg-devel.log.20130725
by burek 26 Jul '13

26 Jul '13

[00:59] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:6db57fa8bc1a: avcodec/mjpegdec: Support RGBA lpegs [01:00] <durandal_1707> what is lpegs ? [01:08] <Compn> ljpegs [01:09] <durandal_1707> no its not, its typo [01:11] <durandal_1707> michaelni: what is J thing commented out? [01:13] <durandal_1707> also if i do conversion with scale filter from non YUVJ to YUVJ will it do right conversion now (without any additional options)? [01:17] <durandal_1707> color_range is not used by scale filter, but ideally it should be used with 'auto' .... [01:21] <michaelni> YUVJ->YUVJ should work [01:21] <durandal_1707> not YUVJ aka YUV -> YUVJ [01:21] <michaelni> that should work too [01:36] <durandal_1707> michaelni: i think subsampled yuv should not be hard to add to j2kdec [01:40] <durandal_1707> BuxiNess: does jpeg2000 performance matches libopenjpeg or its still worse? [01:41] <j-b> matches, IIRC [02:09] <axorb> hey guys, I'm modifying the mp3 encoder to include a guess of the total number of frames [02:09] <axorb> how would I go about getting the input duration? [02:10] <axorb> I' [02:10] <axorb> I'm dealing with streams here so it can't be 100% accurate, I just want to provide the user with something rather than default to 0 [02:15] <michaelni> axorb, see AVDurationEstimationMethod [02:15] <axorb> neat, thanks [02:25] <axorb> actually, I don't understand how the encoder and decoder interact [02:26] <axorb> I've got an AVFormatContext, but the duration is not set because it isn't used when encoding [02:26] <axorb> somebody I need to get the length from the decoder but I'm looking for examples but not finding anything yet [02:27] <axorb> somehow* and* [02:53] <axorb> yeah, all of the AVFormatContext is empty and there isn't enough in AVCodecContext [02:54] <axorb> I'm just trying to make it automatic, but I guess I can just have the user pass in the duration [02:55] <axorb> I don't understand the workflow enough at the moment to do it properly [03:01] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:a7e45cfa1185: swscale/sws_getColorspaceDetails: dont fail for non yuv [03:01] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:5156d482cd60: avfilter/vf_scale: use the inputs color range when its set and its not overridden [09:20] <BuxiNess> durandal_1707, j-b, Now jpeg2000 quicker [09:20] <BuxiNess> But not all pathces for speed up pushed yet [09:20] <BuxiNess> mainly SSE stuff [10:22] <cone-532> ffmpeg.git 03Martin Storsjö 07master:54ba52077c96: arm: Comment out unused labels in simple_idct_arm [10:23] <cone-532> ffmpeg.git 03Michael Niedermayer 07master:0815b230628e: Merge remote-tracking branch 'qatar/master' [16:19] <durandal_1707> michaelni: recent filter change introduced bunch of errors in valgring full fate box: http://fate.ffmpeg.org/report.cgi?time=20130725053855&slot=x86_64-archlinux… [16:26] <wm4> saste: how do I handle EOF correctly with libavfilter? just calling av_buffersink_get_frame() until no frame is returned doesn't seem to work [16:33] <durandal_1707> saste: will program thing for ffprobe be finally committed? [16:38] <xlinkz0> can ffmpeg be built using something other than gcc? [16:38] <durandal_1707> xlinkz0: yes [16:39] <xlinkz0> clang? [16:39] <Plorkyeran> clang, icc, msvc (with the help of c99conv) [16:39] <Plorkyeran> maybe the dumb sun compiler? [16:40] <cptspiff> that would surprise me ;) [16:42] <xlinkz0> damn apple llvm-gcc is old as shit doesn't support thread local [16:42] <nevcairiel> I think suncc even works [16:43] <nevcairiel> http://fate.ffmpeg.org/history.cgi?slot=x86-opensolaris-suncc5.10 found a fate box with it :D [17:15] <cone-532> ffmpeg.git 03Michael Niedermayer 07master:6bde1e9d14ff: avfilter/avfiltergraph: fix check using the wrong variables [17:15] <cone-532> ffmpeg.git 03Michael Niedermayer 07master:de0a1f63dff9: avfilter/can_merge_formats: fix memleak [17:21] <ramiro> saste: hi! [17:22] <ramiro> where's the best place to document that change I sent for img2enc? [17:22] <durandal_1707> doc/muxers.texi [18:17] <aballier> avformat_query_codec for ticket 2209 ? [18:51] <durandal_1707> cehoyos best of the best ^ [19:50] <llogan> durandal_1707: where did you find that sample (ticket 2810)? [20:02] <durandal_1707> llogan: iirc superuser something [20:02] <durandal_1707> by using google [20:02] <durandal_1707> searching for something unrelated.... [20:07] <durandal_1707> http://stackoverflow.com/questions/17809407/ffmpeg-cannot-open-a-simple-mic… [20:07] <durandal_1707> i did not checked is this because of file size or something else ... [21:57] <durandal_1707> stop the spam [22:14] <cone-400> ffmpeg.git 03Reuben Martin 07master:4a60df2d35e7: avformat/gxfenc: Added support for writing correct auxiliary data for DV streams. [22:23] <durandal_1707> do not use past tense in log messages [22:42] <cone-400> ffmpeg.git 03Paul B Mahol 07master:fd84b6fd3389: lavu/opt: show default values for AV_OPT_TYPE_PIXEL_FMT and AV_OPT_TYPE_SAMPLE_FMT too [23:16] <durandal_1707> beastd: no more logo contest? [23:16] <beastd> hi [23:16] <llogan> durandal_1707: you complained about it too much [23:17] <durandal_1707> ahh, you mean that airplane thing :).... [23:17] <cone-400> ffmpeg.git 03Carl Eugen Hoyos 07master:8ca3766f007d: Allow native compilation on the iPhone. [23:17] <beastd> it was a bit unfortunate this time. but i plan to restart in autumn [23:17] <cone-400> ffmpeg.git 03Carl Eugen Hoyos 07master:9375879d5702: Show max bitrate for mpeg2 video streams in avcodec_string(). [23:17] <cone-400> ffmpeg.git 03Michael Niedermayer 07master:1af921196fbc: Merge remote-tracking branch 'cehoyos/master' [23:19] <llogan> if we resume i'd like to see more submissions per contest, but i'm unsure of the best way to do that. [23:21] <beastd> durandal_1707: i was also thinking about your request about different sizes. and you mentioned you would also consider a new different logo. i don't want to have submissions that use a logo that will be changed. [23:22] <beastd> OTOH: if we would really decide to vote for a new logo the contest/search should be opened for month anyway. so i think autumn banner contest can be done anyway with the current logo. [23:22] <durandal_1707> well idea about new logo is just that, if others prefer curren 3d one.... i not going to insist [23:23] <beastd> agreed. a single person shouldn't decide about that anyway. [23:26] <beastd> we also should bring up logo licensing once again. i think e.g. one of the CC licenses should actually be suitable. i would like to present all logos on a wiki page so visitors can look at them and download them under a clear legal conditions (as clear as legal gets that is). [23:30] <beastd> maybe by-nc-sa 3.0 [23:33] <llogan> the 3d logo is troublesome for some printing [23:33] <beastd> ack [23:38] <llogan> also usage of a free font should be a requirement if there is a new logo. it's restrictive to only be able to work with text outlines because herve never said what typeface he used and he never responded to my questions. [23:38] <durandal_1707> i looked at photo of those stickers and they are ugly [23:38] <llogan> what's ugly about them? [23:38] <durandal_1707> or maybe it was low resolution.... [23:39] <llogan> i already mentioned in the RFC stickers that the printing quality could have been better if that is what you are referring to. [23:41] <llogan> beastd: you should have michaelni send you some stickers since you attend the conventions (thilo sent him too many) [23:41] <beastd> ok [00:00] --- Fri Jul 26 2013

1 0

ffmpeg.log.20130725
by burek 26 Jul '13

26 Jul '13

[00:00] <axorb> hey guys, I'm transcoding a flac file to mp3, writing to a pipe [00:00] <axorb> I need to pass along the total duration of the file [00:02] <axorb> if outputting to a file, the duration tag/header/? is properly set [00:07] <axorb> just a bit of explanation, I'm transcoding large flac files to mp3 and serving them with chunked-transfer (no content-length) [00:07] <axorb> so the player doesn't know the total duration because we didn't write a header of any sort [00:08] <axorb> I'd like to force ffmpeg to write the duration at the start, possibly calculated from ffprobe [00:09] <klaxa> mp3 is rather headerless afaik [00:10] <axorb> "For seekable output the muxer also writes a Xing frame at the beginning, which contains the number of frames in the file. It is useful for computing duration of VBR files." [00:10] <klaxa> it would be metadata [00:10] <klaxa> ah [00:10] <axorb> so I need to fake that frame [00:10] <klaxa> nvm then, all mp3 i messed with was just frames glued together [02:12] <Keshl> Anyone see this? http://www.tomshardware.com/reviews/x265-hevc-encoder,3565.html [02:13] <klaxa> yeah seen it [02:13] <klaxa> saw a thread on /a/ Daiz was being autistic again too [02:19] <llogan> mercurial and cmake. not interested. [02:19] <Samus_Aran> does anyone have any tips about encoding PNG frames from a 2D animation app (zero video noise) into a very high quality AAC/AVC? looking for almost lossless [02:19] <Samus_Aran> the video is 1080p/60 [02:20] <Keshl> Samus_Aran: Firstly, don't use VLC to play it. It introduces noise unless you go OCD on the encoder settings. [02:21] <Keshl> Hug http://mpc-hc.org/ instead. [02:21] <klaxa> Keshl: wtf, that's... not what he asked at all [02:21] <Keshl> I do 2D work sometimes too, sort of. [02:21] <Keshl> klaxa: Getting there. [02:21] <Samus_Aran> well, I don't get to choose the app people play it on [02:21] <Keshl> Now, assuming you use basically anything but VLC, you can get by using X264 with a crf of 23, pretty much. [02:21] <Keshl> If you think they might, you need to bring the CRF to like, 10 or less. x.x [02:22] <Samus_Aran> I'll be upping them to YouTube, but the original will be downloadable [02:22] <llogan> https://trac.ffmpeg.org/wiki/x264EncodingGuide [02:22] <llogan> https://trac.ffmpeg.org/wiki/AACEncodingGuide [02:23] <klaxa> ah also, re: x265, on my system it produces a floating point exception [02:23] <Keshl> o.O' How come everyone keeps dialing the wrong number with me? [02:23] <Keshl> Wrong window. [02:23] <klaxa> didn't bother to run it through gdb after that, though [02:23] <llogan> Samus_Aran: for youtube: ffmpeg -i input -codec:v libx264 -crf 18 -codec:a copy output.mkv [02:26] <llogan> for download: ffmpeg -i input -codec:v libx264 -crf <up to you> -preset <choose one> -tune animation -pix_fmt yuv420p -codec:a libfdk_aac -vbr 4 output.mp4 [02:29] <Samus_Aran> llogan: why would they be different for YouTube or download? doesn't YouTube re-encode everything received? [02:29] <Samus_Aran> I may just upload it to YouTube in 4K format, as YouTube now seems to support sizes above 1080p [02:36] <Samus_Aran> why does -vbr 5 only encode to 80-96 Kbps? [02:36] <Samus_Aran> that seems excessively low for full quality [02:37] <Samus_Aran> my audio is in 24bit/96KHz [02:39] <Samus_Aran> how do you disable the -cutoff? [02:49] <mAGiC69> Hi, when using 2 passes I see that options I pass are not used, for example: -me_method and -trellis. Am I doing something wrong that overrides the command line ? these are the command i'm running: ffmpeg -y -i input.mp4 -pass 1 -b:v 6000000 -ab 192000 -s 1920x1080 -passlogfile 1passlog -vcodec libx264 -g 90 -bf 3 -refs 1 -b_strategy 2 -coder 1 -qmin 10 -qmax 51 -sc_threshold 40 -flags +loop -me_range 16 [02:49] <mAGiC69> -me_method dia -subq 2 -i_qfactor 0.71 -qcomp 0.6 -qdiff 4 -direct-pred 3 -trellis 1 -partitions -parti8x8-parti4x4-partp8x8-partb8x8 -r 48 -keyint_min 25 -wpredp 2 -acodec libfaac -ar 44100 /output.mp4 [02:49] <mAGiC69> ffmpeg -y -i input.mp4 -pass 2 -b:v 6000000 -ab 192000 -s 1920x1080 -passlogfile 1passlog -vcodec libx264 -g 90 -bf 3 -refs 1 -b_strategy 2 -coder 1 -qmin 10 -qmax 51 -sc_threshold 40 -flags +loop -me_range 16 -me_method umh -subq 2 -i_qfactor 0.71 -qcomp 0.6 -qdiff 4 -direct-pred 3 -trellis 2 -partitions -parti8x8-parti4x4-partp8x8-partb8x8 -r 48 -keyint_min 25 -wpredp 2 -acodec libfaac -ar 44100 -movfl [02:49] <mAGiC69> ags faststart /output.mp4 [04:51] <mxmader> hello - just have a documentation update for you - WRT https://trac.ffmpeg.org/wiki/CompilationGuide [04:52] <mxmader> when compiling libx264 on fedora (18), i needed to add "perl-Digest-MD5" to the list of RPMs [04:52] <mxmader> apologies - the URL in question is https://trac.ffmpeg.org/wiki/CentosCompilationGuide [05:05] <mxmader> previously, the above would cause libx264 compilation to fail, and cascading that failure to ffmpeg itself. [05:05] <mxmader> it's compiling successfully now... [12:16] <smadamr> Anyone have some resources for learning the difference between -b:a and -ar flags? [12:23] <relaxed> smadamr: -b:a sets the bitrate and -a:r sets the sampling rate. [12:24] <relaxed> er, I think you meatn -ar [12:24] <relaxed> meant* [12:24] <relaxed> or maybe -r:a [12:25] <relaxed> oh, someone needs sleep [12:26] <smadamr> cool thanks....yea we all need sleep sometimes ^_^ [13:53] <t4nk915> hi [13:54] <t4nk915> does anyone know how to set timeout value in the arg list when I start ffmpeg? [13:54] <t4nk915> I mean http timeout [13:54] <t4nk915> for example [13:54] <t4nk915> my ffmpeg command is :ffmpeg -loglevel debug -f video4linux2 -r 30 -s 1280x720 -input_format h264 -i /dev/video1 -vcodec copy http://localhost:8090/feed2 [13:55] <t4nk915> then how to add http timeout value in agr list? [13:55] <t4nk915> any idea? [14:12] <smadamr> i'm converting speex to aac and have clicking and popping at the start of the stream, after the conversion. Anyone seen this before? [14:13] <durandal_1707> smadamr: what aac encoder you use? [14:14] <smadamr> originally plain, built-in aac, then I realized that could have been the problem and have changed to libfdk_aac [14:14] <smadamr> still, the popping/clicking occurs though [14:23] <durandal_1707> smadamr: what sample rate? [14:24] <smadamr> the original sample rate is 16000, i've converted to 16k and to 44100 - just for some experimenting to see if it helped [14:25] <smadamr> the strange thing is that I can convert speex -> vorbis without clicks, and then vorbix -> aac without clicks, but speex->aac always gives clicks [15:16] <luc4> Hi! When muxing, how do I tell ffmpeg what kind of muxer to use? I'm reading the muxing.c sample code, but it seems the container is set according to the extension of the output file. But if I tried to mux into a buffer? How can I specify the container to use? [15:16] <klaxa> luc4: -f <container> [15:17] <ItsMeLenny> is there a way to copy just one video stream? [15:17] <luc4> klaxa: sorry, I'm referring to libavformat [15:17] <klaxa> for example: ffmpeg -i <input> -c copy -f matroska pipe: | <matroskaplayer> [15:17] <luc4> klaxa: not the ffmpeg executable [15:17] <ItsMeLenny> and does specifying the output file name with extension automatically detect the container [15:17] <klaxa> ah... hmm... dunno [15:19] <klaxa> ItsMeLenny: you can copy only one video stream if you use -map [15:19] <klaxa> for example to copy the first video stream and the second audio stream use: ffmpeg -i <input> -map 0:v:0 -map 0:a:1 <output> [15:20] <klaxa> ah needs -c copy too [15:20] <klaxa> and yes ffmpeg will try to detect the correct container according to the output filename's extension [15:21] <ItsMeLenny> ah sweet [15:21] <ItsMeLenny> i'll try that now [15:22] <ItsMeLenny> i just found that a video i had contained 4 different quality versions within it :P [15:24] <ItsMeLenny> hmmm doesnt seem to be working for me [15:24] <klaxa> maybe your ffmpeg version is too old? [15:25] <Fjorgynn> probably [15:28] <ItsMeLenny> "Failed to set value 'copy' for option 'c'", also i'm actually using avconv [15:28] <ItsMeLenny> (ubuntu) [15:28] <ItsMeLenny> "Unrecognized option 'c'" [15:30] <ItsMeLenny> ah i think i got it [16:08] <luc4> Hi! Anyone with experience with avformat_alloc_output_context2? I'm using that function with a output file name and after muxing, the result is correct. When instead I pass a AVOutputContext created with av_guess_format, I get a result almost identical, apart from a couple of bytes, but the file does not play with vlc. [17:04] <luc4> Anyone able to write muxed frames using the callbacks instead of writing to file directly? It seems that slightly different data is passed when the callbacks are used. [18:16] <luc4> Is the write callback for avio supposed to do something particular before writing the resulting video file? [20:09] <rurtle> Hey guys [20:10] <rurtle> I'm trying to write a skeletal app for displaying rtsp streams from a cam connected over lan [20:10] <rurtle> *LAN [20:11] <rurtle> I've already compiled the ffmpeg for target platform and it gave out the ffmpeg so file just fine [20:11] <rurtle> can anyone suggest some pointers/links/documentation on how to write JNI wrapper for it [20:12] <rurtle> I'm absolutely new to android development [20:12] <rurtle> Will really appreciate your help on this matter [20:22] <rurtle> BTW, I missed the fact that the app is for Android (ICS) [20:35] <mkrufky> recent android has its own api for multimedia playback... i dont think u need to deal w ffmpeg [20:39] <rurtle> Yeah, right. However, the usage of native api's resulted in flicker in real-time streaming from IP cam [20:40] <rurtle> Also, I may have to make it to work for some earlier versions as well [20:40] <mkrufky> my guess is that you'll get better results debugging the media content to figure out why it plays weird on your device rather than switching out the player [20:41] <mkrufky> streaming on android before 4.2 is kinda not great [20:41] <rurtle> Hmm. I should have done some digging on that [20:41] <rurtle> meaning why the native player didn't perform well enough [20:41] <mkrufky> another helpful hint, try on some other devices and see if they have the same issues, or if they have similar but different issues [20:42] <mkrufky> what kind of flicker? is it a deinterlacing issue? [20:42] <mkrufky> mobile devices generally dont love interlaced content, and webcams love to stream interlaced content [20:42] <rurtle> like there is considerable latency and the playback freezes and comes back after some time and goes back to sleep again [20:42] <mkrufky> note im using the words "love to" ... im generalizing ... its really device specific [20:42] <rurtle> that sort of behavior [20:43] <mkrufky> ah, so its not a flicker [20:43] <mkrufky> it sounds like a buffering issue [20:43] <rurtle> my bad [20:43] <rurtle> and the device we tried it on, is an old one [20:44] <mkrufky> yea... debugging from that point is just spinning your wheels [20:44] <mkrufky> (in my opinion) [20:45] <rurtle> I did a local build of VLC and ported in on both 4.2.2 & 4.1.2 [20:45] <rurtle> hmm [20:45] <rurtle> and it works just fine on either of those platforms [20:45] <rurtle> however, when we have multiple streams to observe, VLC is not able to clean up cache [20:46] <rurtle> and leads back to old rtsp stream [20:46] <rurtle> so, thought of using the base ffmpeg engine and tweak it as per my need [20:47] <mkrufky> i thought u said you tried the native player [20:47] <mkrufky> vlc is not the native player :-P [20:48] <rurtle> :) [20:48] <rurtle> yeah, I tried several means [20:48] <rurtle> that includes building the native library and (trying to) write JNI wrapper [20:49] <rurtle> and doing a local build of VLC latest from git [20:51] <mkrufky> well, im not sure what the overall plan is [20:51] <mkrufky> but [20:51] <mkrufky> i was under the impression that the native playback api's of jellybean let you play rtsp streams [20:52] <mkrufky> i know for SURE it plays HLS, but thats not what u asked :-P [20:58] <rurtle> The overall plan is to stream several live cam feeds through an android app. The app should be flexible enough to run on multiple platforms [20:59] <rurtle> I am yet to have a clear visibility on the deliverable in terms of which all platform it should support among other things [20:59] <rurtle> Thanks for the tip on HLS player; I haven't checked it yet. Will definitely do. [21:18] <raj___> I am seeing some flv video contents which has range requests using "ns" and "nk" parameter associated with response..Can anyone help me to understand these params here is the example URL http://65.255.34.10/abc/676EFE25A034833FBAA744744/0300020100518AB037DBAD083… and for nk param here is the URL: /abc/6571DFBC8E132831F2D4DA3D31/03000203 [21:19] <raj___> and for nk param here is the URL: /abc/6571DFBC8E132831F2D4DA3D31/030002030051C99005966008565B268B39BDC7-D4DA-386A-772D-4B73E83A8B47.flv?nk=66933666410_22881099153&ns=0_216384&special=true HTTP/1.1\r\n [21:24] <raj___> <fflogger> i am using basic ffmpeg command for transcoding. this problem is when we do seeking.. i am not able to understand this "ns" and "nk" significanse [21:26] <llogan> seeking with what? [21:27] <raj___> seeking forward [21:29] <raj___> For FLV the "from" and "to" offset s are encoded in the ns query parameter? [21:29] <raj___> i am confused here [21:30] <raj___> http://65.255.34.10/abc/676EFE25A034833FBAA744744/0300020100518AB037DBAD083… [21:30] <raj___> &ns=0_22166660&special=true [21:34] <llogan> i don't see what this has to do with ffmpeg, but I don't quite understand what you're trying to do. [22:22] <Fjorgynn> so the internal mp3 encoder in ffmpeg [22:22] <Fjorgynn> lafv54 or whatever [22:26] <durandal_1707> Fjorgynn: there is not internal mp3 encoder in ffmpeg [22:28] <Fjorgynn> durandal_1707: encoder : Lavf54.50.104 [22:28] <Fjorgynn> what is this then? [22:30] <Fjorgynn> durandal_1707: [22:30] <durandal_1707> container != codec [22:31] <Fjorgynn> so I wonder what encoder that is [22:32] <durandal_1707> that is not encoder [22:32] <durandal_1707> libmp3lame is not internal mp3 encoder [22:33] <durandal_1707> it is external mp3 encoder [22:33] <Fjorgynn> yeah? [22:33] <Fjorgynn> but whatever [22:33] <durandal_1707> ... what was question? [22:35] <Fjorgynn> it just does -acodec copy hm [22:36] <Fjorgynn> I was using a script to download some files I know are mp3 [22:40] <durandal_1707> so everything works.... [22:54] <Fjorgynn> durandal_1707: yeah, I was just wondering what it did. Then I saw it copied it and I was asking what laf54 ment [22:58] <llogan> libavformat version 54.50.104 [22:59] <Fjorgynn> aha? [23:34] <salamanderrake> anyone know why the output from this screen is black on twitch.tv? http://bpaste.net/show/117396/ [23:43] <llogan> salamanderrake: i have an idea, but it is only a guess without the ffmpeg console output [00:00] --- Fri Jul 26 2013

1 0

ffmpeg-devel.log.20130724
by burek 25 Jul '13

25 Jul '13

[00:00] <durandal_1707> llogan: what? [00:00] <saste_> llogan, nice? doesn't sound like the correct term [00:01] <llogan> a worthless, but entertaining joke filter. [00:01] <saste_> i'd like a naive voice synth [00:02] <saste_> not sure if we can do timelined audio overlay at the moment [00:02] <durandal_1707> what is that? [00:03] <saste_> you overlay audio at the given interval [00:03] <saste_> like overlay=...:enable=expr [00:05] <llogan> does clement not code in the northern hemisphere summer? [00:05] <saste_> clement, not active in this period, he will be back in august he said [00:08] <durandal_1707> saste_: you have pan,amerge for this [01:14] <cone-528> ffmpeg.git 03Michael Niedermayer 07master:274a50ab5738: avcodec/utils: Warn the user about the lack of a lock manager if insufficient locks are detected [03:02] <kierank> that has got to be the greatest ticket known to man [03:05] <drv> it has cool pictures [03:05] <kierank> 5 years of no aac work and now this [03:05] <kierank> encoder work [04:49] <cone-815> ffmpeg.git 03Zhang Rui 07master:6bab3430a775: avformat/http: support reading compressed data [10:23] <cone-292> ffmpeg.git 03Luca Barbato 07master:70ecc175c7b5: alsdec: Fix the clipping range [10:23] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:865f34823c57: Merge commit '70ecc175c7b513a153ac87d1c5d219556ca55070' [10:39] <cone-292> ffmpeg.git 03Luca Barbato 07master:ca488ad48036: alsdec: Clean up error paths [10:39] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:a550926705fe: Merge commit 'ca488ad480360dfafcb5766f7bfbb567a0638979' [11:38] <cone-292> ffmpeg.git 03Luca Barbato 07master:2f034f255c49: 4xm: Reject not a multiple of 16 dimension [11:38] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:82654172ced5: Merge commit '2f034f255c49050e894ab9b88087c09ebe249f3f' [12:00] <cone-292> ffmpeg.git 03Luca Barbato 07master:bd7b4da0f462: 8bps: Bound-check the input buffer [12:00] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:da0f67b177ad: Merge commit 'bd7b4da0f4627bb6c4a7c2575da83fe6b261a21c' [12:16] <cone-292> ffmpeg.git 03Luca Barbato 07master:6fd221e5f8b0: 8bps: Make the bound-checks consistent [12:16] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:7dec8512be12: Merge remote-tracking branch 'qatar/master' [12:56] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:800637709c82: alsdec: minor simplification in read_block() [12:56] <cone-292> ffmpeg.git 03Anshul Maheshwari 07master:a553cb4683a2: documented that av_codec_close is required to call [13:06] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:101c7ea90673: avcodec/mpeg12dec: remove commented out assert [14:10] <cone-292> ffmpeg.git 03Bernie Habermeier 07master:c5f3cc40e33a: matroskaenc: implement CueRelativePosition [16:30] <cone-292> ffmpeg.git 03Timothy Gu 07master:11cb697501ed: doc/encoders: partially rewrite and reformat libx264 docs [16:31] <Daemon404> o shit it finally went it [16:31] <Daemon404> in* [16:37] <saste_> Daemon404, why, was you waiting for it? [16:37] <saste_> it took "only" a week, which is not so much for a 200+ lines documentation patch [16:38] <Daemon404> i was just following the very large email thread [16:57] <saste> ^^ epic ticket [17:17] <Daemon404> yes epic [17:57] <aballier> michaelni: (and others too) what is your opinion on an avframe based version of sws_scale/swr_convert that would allocate the output buffer as needed alike the avframe based avcodec_decode/encode_audio/video ? [17:59] <aballier> if the interface is generic enough it might even be possible to plug it in avcodeccontext so that input frames are converted before being sent to the encoder and output frames are converted before being returned [18:15] <michaelni> aballier, possible if someone wants to use it and someone volunteers to maintain it [18:17] <aballier> i can try to bake up a patch [18:17] <aballier> just wanted to be sure its not an idea that has been considered and discarded for some reason i dont know [18:23] <michaelni> i dont remember that being suggested previously (at least not in that form) [18:25] <wm4> michaelni: do I need to set any magical flags to make the libswscale chroma pos stuff for yuv->rgb take effect? [18:26] <michaelni> wm4, no but its very easily possible theres a bug, for example some optimized X->Y converter could end up being used that doesnt know about chrma pos [18:27] <nevcairiel> probably disable the inaccurate scalers, accurate_rnd and the full chroma flags [18:27] <nevcairiel> or something [18:27] <wm4> I was trying SWS_LANCZOS | SWS_FULL_CHR_H_INT | SWS_FULL_CHR_H_INP | SWS_ACCURATE_RND | SWS_BITEXACT [18:29] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:e31d20255d81: vf_scale: add output h/v_chr_pos parameterss [18:30] <wm4> oh, it only does something if I set the dst_ parameters [18:30] <wm4> but wouldn't refer dst_... for yuv->rgb position to RGB, which does not have chroma position stuff? [18:31] <wm4> are they just swapped? [18:31] <michaelni> i hope not [18:42] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:24a4367ae963: vf_scale: add input h/v_chr_pos parameters [18:49] <hotwings> according to the author of the softhddevice output plugin for vdr, theres a problem with the vdpau decoder not ffw/rew'ing h264 (h264 paff to be exact, iirc). would this be the right place to talk to someone about that? [18:50] <wm4> michaelni: so, is there something wrong with it? [18:56] <michaelni> hotwings, if theres someone around who knows vdpau well, not sure there is atm (carl is listed as maintainer) [18:56] <michaelni> reimar knows vdpau too [18:56] <michaelni> and probably others [18:57] <michaelni> hotwings, if its reproduceable with ffmpeg/ffplay then probably easiest to open a bug report [18:57] <michaelni> wm4, dunno, how can i reproduce it ? [19:07] <Compn> hotwings : upload a sample file that causes problems to the trac [19:08] <Compn> er, and report the bug on the trac i mean [19:19] <wm4> michaelni: I tried: ffmpeg -i chroma-pos-mpeg2.y4m -vf 'crop=32:32,scale=1000:1000:out_v_chr_pos=128:out_h_chr_pos=128:flags=accurate_rnd+full_chroma_int+full_chroma_inp+bilinear,format=rgba' avchroma.png -v debug [19:19] <wm4> michaelni: changing the in_v_chr... etc. params didn't change anything [19:19] <wm4> only out [19:19] <wm4> input is yuv420 [22:53] <cone-292> ffmpeg.git 03Michael Niedermayer 07master:436616fd4251: swscale: exchange src & dst chroma pos [00:00] --- Thu Jul 25 2013

1 0

ffmpeg.log.20130724
by burek 25 Jul '13

25 Jul '13

[00:23] <pinPoint> I have question, can ffmpeg be used to address issues with videos? [00:23] <pinPoint> corruption from let's say drive collapsing, etc? [00:23] <pinPoint> I have the file but it is a R3D file... :/ [00:37] <telly> im trying to compile ffmpeg for raspberry pi, but he says libx264 not found though there is libx264.a in the path i supply with -extra-ldflags [00:37] <telly> what can i try to fix this? [00:39] <Keshl> I don't know, but if you get it to work, can'ya tell me how the performence is? XD [00:41] <telly> ;D [00:41] <telly> well i mean he can take the whole day, hes got the time? :P [00:42] <pinPoint> heh [00:42] <pinPoint> is anyone aware of MP4Repair.org? [00:43] <Keshl> pinPoint: Both sites are unreated by WOT. [00:43] <Keshl> Chances are both the .com and .org site are malware distributors. [00:43] <pinPoint> unreated by WOT? [00:43] <Keshl> WOT, web of trust. It's a group that has user-made reveiws for basically every /known/ website. [00:44] <Keshl> And both the .com and .org sites are totally unrated. [00:44] <pinPoint> are you serious? [00:44] <Keshl> Yeah. [00:44] <Keshl> Generally it works. [00:44] <Keshl> http://www.mywot.com/en/scorecard/mp4repair.org?utm_source=addon&utm_conten… [00:44] <Keshl> http://www.mywot.com/en/scorecard/mp4repair.com?utm_source=addon&utm_conten… [00:44] <Keshl> Just search in the upper right corner, or go hug the addon for Chrome or Firefox. [00:45] <pinPoint> unrated is true [00:45] <pinPoint> but the point of the site is to repair damaged video files [00:45] <Keshl> And after some quick checks of my own... [00:45] <pinPoint> and they talk about ffmpeg [00:45] <Keshl> Yeah, this is just... Don't download stuff from them, you /will/ catch something. [00:45] <pinPoint> there is nothing to download [00:45] <Keshl> Exactly. [00:46] <pinPoint> http://mp4repair.org/blog/ [00:46] <pinPoint> look at their blog [00:46] <pinPoint> its a legitimate site [00:46] <Keshl> Link's there, it changes the page, nothing comes up. I'm sensing, failed exploit. [00:46] <pinPoint> how is that even possible? [00:46] <Keshl> Ah, I get what they're doiung. [00:46] <Keshl> They're not really repairing it, they're generating it. [00:47] <Keshl> Basically they're using motion estimation to track where pixels /might/ go, and then fill in the holes like JPEG decompression does. [00:47] <Keshl> It works for video, somewhat, but won't be an exact match, and the longer the gap the less accurate it'll be. [00:47] <Keshl> Audio's highly questionable. [00:47] <Keshl> Frankly you can do this for free. [00:48] <pinPoint> well I was wondering if anyone was aware of what they use to address issues with videos? [00:48] <Keshl> Either grab software that's designed for motion estimation (There's a lot around, but good luck getting it to run. It's /really/ picky stuff), or tweak ffmpeg's X264 compression algorithm a bit. [00:48] <pinPoint> because the site was able to preview R3D screenshots from a damaged .R3D file [00:48] <Keshl> I mean, all you really need is it to do motion estimation for every pixel at once no matter what. [00:48] <pinPoint> how are they pulling it off and could I do the same here at home? [00:48] <Keshl> pinPoint: Read my posts. [00:49] <pinPoint> i did [00:49] <Keshl> Normally X264 only estimates some pixels depending on the settings you (or ffmpeg) passes to it. [00:49] <pinPoint> compression alogrithm huh? [00:49] <pinPoint> it's not an x264 file [00:49] <Keshl> All you gotta do is make it do it to /every/ pixel, no matter the cirumstances, and it'll "repair" the file. [00:49] <Keshl> Doesn't need to be. [00:49] <pinPoint> it's redcode raw codec [00:50] <Keshl> It can be any input, you just need to run it through X264. [00:50] <Keshl> With some tweaks. [00:50] <Keshl> Dunno how to do those, I just know that's what you (and they) need (and do) do. [00:50] <pinPoint> yeah, i'm sorta new to this [00:50] <Keshl> pinPoint: Might wanna spoend a month or so reading up on the terms and current trends then. [00:50] <Keshl> *spend [00:50] <Keshl> Video (en/trans/re/etc)coding is a /lot/ more in-depth than you think. [00:50] <pinPoint> so they are doing about the same thing? [00:51] <Keshl> The exact same thing. [00:51] <Keshl> In fact, they probably use ffmpeg and X264 to do it. [00:51] <pinPoint> i gave them a damaged file and the generated a preview... [00:51] <pinPoint> I myself use a hex editor to get about 20 frames fixed out of 21 seconds of lost footage [00:51] <Keshl> And the fact their site is so poorly designed makes me think the patch's dead simple, or that there might be some obscure set of paramaters that can do it. [00:51] <pinPoint> my way was way toooo painful [00:52] <pinPoint> Keshl: they said this "REDCODE .r3d files can be parsed by identifying blocks of data with the atom structure: [00:52] <pinPoint> 32-bits length + token + payload [00:52] <pinPoint> " [00:53] <pinPoint> http://pastebin.com/mqiLwdEF [00:53] <Keshl> I'm assuming by "broken" you mean "the file has been corrupted in such a way that entire frames need to be discarded" [00:53] <pinPoint> probably [00:53] <pinPoint> it was a challenge from a forum on the internet [00:53] <Keshl> Then the way I wrote above will work. [00:54] <pinPoint> well [00:54] Action: pinPoint is confused [00:54] <Keshl> Like I said. Read stuff for a month. <É< [00:55] <Keshl> This is a very in-depth and complicated topic. A lot of people don't think it's that hard, but really, there's a LOT to this stuff. [00:55] <Keshl> Just go to Wikipedia, /read/ (don't skim!) the /entire/ article, and keep clicking "see also" links. For a month. Seriously. [00:55] <Keshl> And basically any other links if they use terms you don't fully understand. [00:56] <Keshl> (And as of now, assume that's all of them. A lot of people, and I mean a lot, take this stuff for granted and think "oh 720p is always better than 1080p" or "oh 480p is worse than 1080i" when in reality it's the opposite.) [00:57] <Keshl> (And "a bigger file means more quality" is another mis-conception.. "it's better to scale video up before playing it rather than in real-time" is another.. "higher compression means slower framerates" is a biggie..) [00:57] <Keshl> (None of those things are true, by the way. A lot of people think they are, but the exact opposite is.) [00:59] <pinPoint> what if the files i'm dealing with are 4096x2160? [00:59] <pinPoint> same routine? [00:59] <Keshl> No matter the input, same routene. [01:00] <Keshl> Frames are frames are frames are frames, oÉo. [01:01] <pinPoint> not to sound like broken record, in wikipedia, where or what do I start searching/reading about? [01:01] <Keshl> http://en.wikipedia.org/wiki/FFmpeg oÉo, and literally, if there is /any/ word in blue that you don't fully understand, ctrl click it and read that tab right away. [01:01] <Keshl> Do a depth-first search of what you don't know. [01:02] <Keshl> Do not skip things, if you're "iffy" on something, read the entire page. [01:02] <Keshl> There is no fast way to do this. oÉo. [01:02] <pinPoint> that will lead me to understanding a repair process down the road say 3months? [01:02] <Keshl> It might. It depends on how fast you read, how much you absorb and how much fluid intelligence you have. oÉo. [01:03] <pinPoint> i'm savvy with the computers [01:03] <pinPoint> i build my own I mean [01:03] <Keshl> That means nothing. XD [01:03] <pinPoint> trolol [01:03] <Keshl> Hardware != software, not by a mile. [01:03] <pinPoint> I did c++ 2yrs in HS, plus ~2 semesters in college. [01:03] <pinPoint> its been a while however [01:07] <pinPoint> Keshl: thanks [01:10] <Keshl> Welcomes, oÉo [01:10] <Keshl> And good luck -É- [01:11] <pinPoint> yeah, no kidding. :) [01:11] Action: pinPoint dead man walking. [02:09] <zN0z> hi, I would like to know if it is possible to create variables in a command line to store values of one file to be used to format another [02:21] <nfkd> Hello, #ffmpeg [02:21] <nfkd> We are working in a Skype replacement over #InsertProjectNameHere [02:21] <nfkd> And considering using ffmpeg [02:21] <nfkd> http://tox.im/ [02:22] <nfkd> I guess you guys would be interested [02:22] <klaxa> looks promising [02:23] <nfkd> Hopefully [02:23] <klaxa> oh it's p2p? [02:23] <nfkd> Yes, DHT [02:23] <nfkd> Decentralized, for better security [02:23] <nfkd> No central server [02:23] <klaxa> i like that idea [03:20] <sacredchao> I am using Kdenlive to edit videos (using ffmpeg for the actual A/V encoding) and I am trying to make a profile with 384k CBR AAC LC audio + x264 1080p video at 12Mb/s [03:22] <sacredchao> The command line switches I am using however seem to produce a file that VLC and mplayer cannot determine the bitrate of EITHER the audio or video stream via normal methods [03:22] <sacredchao> hq=1 acodec=aac ab=384k ar=48000 pix_fmt=yuv420p vcodec=libx264 minrate=12000k vb=12000k g=250 bf=3 b_strategy=1 subcmp=2 cmp=2 coder=1 flags=+loop flags2=dct8x8 qmax=51 subq=7 qmin=10 qcomp=0.6 qdiff=4 trellis=1 aspect=%dar pass=%passes [03:23] <sacredchao> Obviously I am doing something wrong. The destination is YouTube. [03:23] <Aprel> Any luck with ffprobe? [03:24] <sacredchao> Stream #0.0: Video: h264 (High), yuv420p, 1920x1080 [PAR 1:1 DAR 16:9], 30 fps, 30 tbr, 1k tbn, 60 tbc (default) [03:24] <sacredchao> Stream #0.1: Audio: aac, 48000 Hz, stereo, s16 (default) [03:24] <Aprel> I find (maybe moreso with VBR files) vlc and totem can't figure the bitrate, but play it fine otherwise. [03:24] <sacredchao> mediainfo cannot determine bitrate either [03:25] <Aprel> Does the file play fine otherwise? [03:25] <sacredchao> Yes. [03:25] <sacredchao> The problem is that I changed the video bitrate and the filesize was exactly the same as the previous render [03:26] <sacredchao> So I don't know if there is some limit configured elsewhere, but I need to make a render that is under 500MB for Vimeo and it keeps being 547MB [03:28] <Aprel> As a last resort, you could use kdenlive to output the file in some format close to raw video/audio, and then use ffmpeg to do the encoding on that file. It's possible kdenlive is mucking it up. I've have many problems with that software :/ [03:37] <sacredchao> Also, I read that there are better AAC encoding libraries to use besides the one built into ffmpeg [03:37] <sacredchao> Is this true, and does ffmpeg need to be specifically compiled to make use of them? [03:41] <Aprel> I don't know about that. I use ffmpeg for aac encoding and it's fine for me. I haven't heard of the encoding being particularly poorer to any other open source library codec [03:41] <relaxed> sacredchao: yes, compile ffmpeg with libfdk-aac support [03:43] <relaxed> if you need a specific size use two pass encoding [03:44] <relaxed> you can learn how to do both by following the ffmpeg compile guide and the x264 encoding guide, both found in the wiki. [03:53] <sacredchao> relaxed: OKay thanks, I will stick with simply using 'aac' for now and concentrate on 2-pass encoding [03:54] <sacredchao> On the wiki it simply reminds you of the bitrate = file size / duration formula before getting into 2-pass CLI switches [03:54] <sacredchao> I have a 394 second video I want at 500MB wth 384k audio [03:54] <sacredchao> So I am changing vb=10000k [03:55] <sacredchao> I wonder why this formula doesn't work just as well for 1-pass encoding. The wiki doesn't explain this. [04:09] <relaxed> the second pass is used to avaerage out the bits to achieve your size. [04:10] <relaxed> average* [05:04] <johnnny22> hey [05:04] <johnnny22> woah ffmpeg 2.0 [05:06] <johnnny22> I was wondering what's the best way to get ffmpeg to encode from a pipe into a file that only keeps the last X seconds of video [05:07] <johnnny22> and / or what keywords I could use to search for that technique [05:26] <johnnny22> i guess the best approach is to have a sort of buffer [09:06] <YeahBuddy> Hi everybody. Just built ffmpeg on a Raspberry Pi. Seems to have worked, but I have no idea where it was installed. So my question is A. Where is it? B. What's the easiest way to determine this from makefiles? [09:07] <YeahBuddy> BTW, this is a Debian derivative. [09:14] <YeahBuddy> This is a talkative bunch. [09:18] <dv-> use the which command [09:18] <dv-> which ffmpeg [09:19] <YeahBuddy> this returns nothing [09:22] <YeahBuddy> It could be that I'm an ass and forgot to run make install [09:22] <YeahBuddy> In fact, it's looking a lot like that [09:22] <YeahBuddy> I blame the bourbon [09:23] <YeahBuddy> Oh look, there it is right where one might expect: /usr/local/bin [09:23] <YeahBuddy> Well thanks for your help. [09:23] <YeahBuddy> I'm going to celebrate with chocolate chips. [10:24] <ThePuppetMaster> hi @all [10:26] <ThePuppetMaster> i have a shot question .. i try to create a atrle movie from png files. i need the alpha chan of the png inside the new movie. but if i encode it, ffmpeg shows me on the INPUT test: Stream #0:0: Video: png, rgb24, .... but, rgb24 is not argb ... what i need to do that ffmpeg read the png as argb? [10:27] <ThePuppetMaster> thats the output of ffmpeg: http://pastebin.com/D1aAYV0h [12:26] <anddam> slightly OT: is it actually possible to cut an MP3 file without recoding? [12:29] <JEEB> yes [12:30] <JEEB> compressed audio streams generally consist of independently decode'able chunks, so if you cut at those you get lossless cutting [12:30] <JEEB> the size of the chunks depends on the format/encoder/settings, but lossy encoders generally don't have too long ones, while f.ex. FLAC can have rather long ones [12:31] <anddam> is the chunks' length fixed in MP3 case? [12:31] <anddam> I see, thanks [12:31] <JEEB> I would say no :P [12:31] <anddam> so it can be done but at whole chunks granularity [12:31] <anddam> pretty much like minecraft [12:32] <JEEB> just like cutting video at random access points :) [12:33] <JEEB> (of course with video there are other things to consider as well, such as how the extradata is set and so forth) [12:33] <JEEB> anyways, ffmpeg with -ss and -t before -i (if it fails then move -t to after -i I guess?) and -c copy should work? :) [12:34] <JEEB> of course when the stream is in a container it's simpler to cut, dot-mp3 is generally a raw MPEG-1 Layer 3 audio stream, and those don't exactly make seeking simple :) [12:36] <anddam> :ls [12:51] <anddam> ok, that was the wrong vim buffer [12:52] <anddam> JEEB: thanks, bye [12:52] <Fjorgynn> :D [18:19] <Fjorgynn> 1,5 GHz Snapdragon S4 Pro [22:38] <exutux> hi [23:00] <exutux> guys I'll try to grab desktop video in a selected area but http://paste.ubuntu.com/5908953/ whats wrong in the command? [23:02] <exutux> is it possible to grab desktop in a selected region? [23:03] <klaxa> yes, what is your display resolution? [23:03] <klaxa> because my first guess is, the area you want to record is out of your display bounds [23:03] <exutux> current 1366 x 768 [23:04] <klaxa> okay so [23:04] <klaxa> >1024x768 -i :0.0+100,200 [23:04] <klaxa> that will be out of bounds [23:04] <klaxa> because you will be recording the rectangle: (100,200), (1124, 968) [23:04] <exutux> uhm how can I find right bound? [23:05] <klaxa> you can't record outside of your display [23:05] <klaxa> so your resolution kinda defines your bound [23:07] <llogan> exutux: you can use xwininfo if you want the coordinates for a certain window [23:08] <llogan> http://pastebin.com/xuYUmnGU [23:09] <llogan> also the defaults for flv output are going to be crappy. [23:11] <llogan> generally for capturing the screen a fast encoder is used, then this output is re-encoded later. [23:12] <llogan> ffmpeg -f x11grab -r 25 -s 1024x768 -i :0.0 -codec:v libx264 -preset ultrafast -crf 0 output.mp4 [23:12] <exutux> llogan: thank I'm reading [23:13] <exutux> llogan: yeah but in this case it will grab all screen [23:13] <llogan> yes, because i don't know the coordinates that you want [23:14] <llogan> that was just an example for you to modify [23:16] <llogan> then you can re-encode the first output to whatever you like: ffmpeg -i output.mp4 -codec:v libx264 -preset slow -crf 22 output2 [23:16] <llogan> of course if your computer is fast enough you can just do one command [23:17] <llogan> and you may also need to add -pix_fmt yuv420p as an output option for the second command [23:17] <exutux> http://paste.ubuntu.com/5908992/ [23:18] <exutux> so X 49 an Y 24? [23:19] <exutux> by clicking in differnets areas results is always the same [23:20] <llogan> -s 1317x744 -i :0.0+49,24 [23:20] <llogan> you click the window you want to record and it should give you the details of that particular window [23:22] <exutux> llogan: well I get understanting maybe :D [23:29] <exutux> llogan: [23:29] <exutux> ffmpeg -f alsa -ac 2 -i pulse -r 15 -s 1244x614 -f x11grab -i :0.0+63,121 -acodec libmp3lame -ab 96k -ar 48000 -ac 2 -vcodec libx264 -crf 30 -threads 2 test.mkv [23:29] <exutux> :D thanks [23:30] <exutux> klaxa: thanks to you too [23:30] <llogan> are you going to upload this to youtube or something? [23:30] <llogan> because -crf 30 seems high [23:32] <exutux> llogan: nope not now [23:32] <exutux> llogan: btw which is right crf ? [23:32] <llogan> screencasting can probably benefit from a higher frame rate [23:33] <llogan> https://trac.ffmpeg.org/wiki/x264EncodingGuide [23:33] <llogan> there is no "right" crf. it's subjective. [23:33] <llogan> unless you're uploading to a video sharing site, then use -crf 18 since they will re-encode it anyway. [23:34] <exutux> llogan: ok [23:34] <klaxa> youtube re-encodes are meh though [23:35] <klaxa> i have seen a guy that upscales his videos a little bit from 1080p to 1192p so he gets the "original source" quality [23:35] <klaxa> also applys a sharpening filter to make up for the weird scaling [23:35] <klaxa> don't know what i should think of that [23:36] <llogan> i think he has a lot of time on his hands [23:36] <klaxa> mmh [23:36] <llogan> "time on his hands". english is weird. [23:37] <llogan> i'm going to go chop down a tree and then chop it up. [23:39] <exutux> :D [23:53] <exutux> bye [23:53] <exutux> and thanks again [00:00] --- Thu Jul 25 2013

1 0

ffmpeg-devel.log.20130723
by burek 24 Jul '13

24 Jul '13

[00:45] <Compn> hmm [00:45] <Compn> how did this email make it to me [00:46] <Compn> Delivered-To: webmaster(a)ffmpeg.org [00:46] <Compn> oh ok [01:20] <cone-499> ffmpeg.git 03clook 07master:b9b1a2c3e470: libswscale: Adding RGB => XYZ support [01:20] <cone-499> ffmpeg.git 03clook 07master:4443213d2ea4: avcodec/libopenjpegen: XYZ 12 bit support for libopenjpeg encoder [03:21] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:b7397857e20d: avcodec: add avcodec_enum_to_chroma_pos() [03:21] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:abc8110f7e54: avcodec: add avcodec_chroma_pos_to_enum() [09:32] <bernie__> ok so I have a signed commit ready to go, but unsure if I should email the patch directly to ffmpeg-devel mail list or where... [09:36] <nevcairiel> usually you would use git send-email and send it to ffmpeg-devel list [09:37] <nevcairiel> or just send it as an attachment if you have issues with send-email [09:37] <bernie__> okay, I'll try that :) [09:37] <bernie__> thanks [09:38] <bernie__> when I use git send-email, will it show me what it's gonna send? (just want to make sure I don't spam something stupid to the whole list) [09:42] <bernie__> off it went... though I sent the patch as an attachment [09:42] <bernie__> :) [09:48] <bernie__> ... and I screwed it up... I'll just cut-n-paste it this time around [09:53] <nevcairiel> lets hope that your mailer didnt screw up the copy/paste then :d [10:39] <cone-528> ffmpeg.git 03Martin Storsjö 07master:69e6702c0132: arm: Mangle external symbols properly in new vfp assembly files [10:39] <cone-528> ffmpeg.git 03Michael Niedermayer 07master:7dc77613f653: Merge remote-tracking branch 'qatar/master' [14:16] <durandal_1707> lol [17:30] <cone-528> ffmpeg.git 03Hendrik Schreiber 07master:650355089cc4: avformat/aviobuf/ffio_init_context: set seekable automatically [18:11] <Compn> 99 comments, we have a winner [18:12] <Compn> wow look at all of those graphs [18:30] <hotwings> according to the author of the softhddevice output plugin for vdr, theres a problem with the vdpau decoder not ffw/rew'ing h264 (h264 paff to be exact, iirc). would this be the right place to talk to someone about that? [19:22] <cone-528> ffmpeg.git 03Michael Niedermayer 07master:fe328f080768: avfilter: Dont partially merge lists [21:25] <bernie_> wm4: hey I finally submittied my patch for the CueRelativeOffset... it's my first patch request to ffmpeg. yay! [21:29] <cone-528> ffmpeg.git 03Michael Niedermayer 07master:6a0b72f94fb7: avfilter/avfiltergraph: minor cosmetic [21:29] <cone-528> ffmpeg.git 03Michael Niedermayer 07master:55db06af64d1: avcodec/utils: use av_assert0() to check validity of input pointers for start code search [22:35] <johnsmith> Hi all, since most of you are developing in ffmpeg, I was wondering how much I need know about libavfilter's "graph"? [22:37] <durandal_1707> johnsmith: you want to use libavfilter or work on libavfilter? [22:38] <johnsmith> durandal_1707: I wish to use libavfilter, but I know it's offtopic within this channel, but you and the other guys are more likely to answer the question than on #ffmpeg [22:39] <johnsmith> To put the question in perspective to allow a simple answer: [22:40] <johnsmith> I am using libavfilter for encoding audio with one of ffmpeg's filter features, should I care at all about what this "graph" is from the filter_audio.c example provided from ffmpeg doxygen trunk? [23:07] <michaelni> bernie_, your patches are replies to random unrelated threads [23:18] <llogan> anyone have a rough idea of the size of the make fate suite? [23:18] <llogan> s/make// [23:20] <llogan> the samples I meant. [23:22] <durandal_1707> you mean size of fate directory? [23:23] <nevcairiel> the samples are ~670mb [23:23] <durandal_1707> 686 [23:26] <llogan> thanks [23:27] <nevcairiel> hm wonder if i havent synced for a while [23:30] <durandal_1707> 'du -m' gives 689 after 'gmake fate-rsync' [23:40] <durandal_1707> lol, libav gonna deprecate above mentioned target... [23:44] <durandal_1707> noone wants dynamic range filter? [23:55] <saste_> durandal_1707, dynamic range filter, what's that stuff? [23:56] <durandal_1707> compress or expand dynamic range of audio [23:59] <llogan> a wilhelm scream filter would be nice. -af wilhelm='eq(t,10)' [23:59] <saste_> durandal_1707, compand? [00:00] --- Wed Jul 24 2013

1 0

ffmpeg.log.20130723
by burek 24 Jul '13

24 Jul '13

[00:00] <brontosaurusrex> -crf = 0 = lossless, -crf = 23 = lossy [00:00] <Aprel> k, brb [00:00] <brontosaurusrex> basically -crf 23 will make smaller and uglier files [00:12] <w1jp> Is there an updated 'how-to' on building the ffmpeg-all on Mac OS X? For example, I understand that yasm isn't required anymore. [00:13] <w1jp> I also am noting that ffplay doesn't build and I didn't see any errors to that effect. [00:14] <brontosaurusrex> w1jp, what guide are you following? [00:14] <Aprel> K, I'm back. Still seeing desync very visible 3 mins into video from screen capture. 10min capture = 1GB file with -crf 23. [00:15] <w1jp> The one on the wiki (trac.ffmpeg.org) [00:15] <brontosaurusrex> Aprel, and the player that you are using is? [00:15] <w1jp> Want to build ffplay [00:15] <w1jp> I imagine sdl [00:16] <brontosaurusrex> yeah, i remember that ffplay (was never able to build it) [00:16] <Aprel> brontosaurusrex: totem. Also see the exact same desync in editing software pitivi and kdenlive, so it's not the paler [00:16] <brontosaurusrex> Aprel, how do you measure desync? [00:17] <Aprel> brontosaurusrex: I'm watching a video w people talking and I notice the lips don't match the words. Those framerates I gave earlier I got examine a video in editing software. [00:18] <Aprel> So I play a video, record the screen capture, and watch the screen capture (for this debugging) [00:18] <brontosaurusrex> Aprel, what would happen if you convert the clip with ffmpeg to something I-frameish? like prores maybe or that avid thingy [00:19] <Aprel> brontosaurusrex: ffmpeg command? [00:19] <brontosaurusrex> Aprel, also, what happens with mplayer ? [00:19] <brontosaurusrex> Aprel, hold on ... [00:21] <brontosaurusrex> Aprel, ffmpeg -i in.file -vcodec dnxhd -acodec pcm_s16le -b:v 120M -f mov out.mov [00:22] <brontosaurusrex> i'am pretty sure kdenlive should have some good support for dnxhd (from memory) [00:22] <Aprel> k, I will try that. [00:22] <brontosaurusrex> no promisses, i'am guessing [00:24] <Aprel> All right. It's encoding. [00:29] <Aprel> It's gonna take a little bit. I need to do like 10 mins of this video I have handy to really know if it's resyncing. [00:30] <brontosaurusrex> Aprel, another guess would be something to do with variable-frame-rate and mkv container, try mp4 [00:30] <brontosaurusrex> Aprel, or try to capture to say dnxhd [00:31] <Aprel> Also had another video up in a player just like 300x300 and screencapture that player with -crf 23. Very bad, quality, obviously. I can't really see people's mouths to judge the desync, but it doesn't seem to desync as much if at all with only 300x300 of the screen. [00:31] <brontosaurusrex> (But do some reading, i think dnxhd is resolution dependant, or it used to be) [00:31] <Aprel> Also I used to capture in the mp4 container before I knew what mkv was, and I got the same problem. [00:34] <Aprel> Just finished the encoding. -vcodec dnxhd didn't make a difference. [00:35] <brontosaurusrex> Aprel, try to capture to dnxhd directly [00:35] <brontosaurusrex> but i think you must use standard hd - 1920x1080 (not 1920x1200) [00:36] <Aprel> brontosaurusrex: like this?: ffmpeg -f x11grab -r 25 -s 1920x1080 -i :0.0 -f alsa -ac 2 -i pulse -vcodec dnxhd -crf 23 -preset ultrafast -acodec pcm_s16le output.mov [00:37] <brontosaurusrex> yeah [00:37] <brontosaurusrex> no [00:38] <brontosaurusrex> skip "-crf 23 -preset ultrafast -acodec pcm_s16le" [00:38] <Aprel> k. [00:39] <Aprel> brontosaurusrex: video parameters incompatible with DNxHD. [00:40] <Aprel> I think I need to set a bitrate... so add -b 175M? [00:40] <Aprel> i ffmpeg outputs Frame size: 1920x1080p; bitrate: 175Mbps; pixel format: yuv422p10; framerate: 24000/1001 as one of the valid profiles, but I don't know how to set that in the command line [00:41] <Aprel> nvm, just got it. [00:42] <brontosaurusrex> what command did you use? [00:43] <Aprel> just added -b 175M to the command line and it started without errors. [00:43] <brontosaurusrex> ok [00:45] <Aprel> btw, later I'll try capturing a long vdieo at just like 300x300 and see if the audio still is in sync at like 20 mins in. If it's fine at a small screen size, what could be the cause of the problem with only big sizes like 1920x1080? [00:46] <brontosaurusrex> Aprel, could be a lot of things, including drivers and whatnot [00:47] <Aprel> ugh, already had enough "fun" trying to fix a video tearing issue in ubuntu. :/ [00:47] <brontosaurusrex> also the problem is that kdenlive is something thats hard to trust [00:48] <brontosaurusrex> so the decoder side can be problematic somehow [00:48] <brontosaurusrex> what OS are you running? ubuntu? [00:48] <Aprel> haha, I'm just lucky if I can get kdenlive to not crash [00:48] <Aprel> yeah ubuntu 13.04 [00:49] <brontosaurusrex> Aprel, did you test lightworks beta? [00:50] <Aprel> brontosaurusrex, yeah, but it's a little above my learning curve. Plus I had some codec issue (what I encoded in ffmpeg didn't play right in lightworks), but I tried the beta like several months ago so could be better [00:51] <brontosaurusrex> yeah :| [00:51] <brontosaurusrex> video editing in nix is just sad [00:52] <Aprel> btw, doing it with dnxhd not only cause desyncing, but audio dropping (like people talking and their word drops as they say it). [00:52] <brontosaurusrex> I did some editing with lightwork like 10 years ago (they were using dos then) [00:52] <brontosaurusrex> so i had high hopes for a linux version [00:52] <brontosaurusrex> lightworks* [00:52] <Aprel> yeah, I want to use unix so bad to edit video, so many promising programs...so many problems with them. [00:53] <Aprel> damn, I have to run, but I'll be back in 30 mins, if you happen to be on. [00:53] <brontosaurusrex> have fun [00:54] <eggyknap> Can someone explain this "No channel layout" error for me? http://pastie.org/8165503 [00:55] <eggyknap> I originally figured it meant I needed -ac, to see that what ffmpeg thought were stereo inputs were really mono... but I added that, and it correctly sees they're mono, and it still complains [00:57] <eggyknap> The anull filters are just to allow me to refer to streams by nice names rather than [0:0], etc., which is helpful for the more general development I'm doing. [00:58] <eggyknap> ...but perhaps they're not working as I'd like, and I just don't know it. [01:34] <Aprel> Back. If there's a possibility that it's a driver issue causing the desync, it probably is, because I've got an nVidia GTX 550 Ti and as nice as the card is, it likes to cause a lot of problems. [01:37] <eggyknap> Is there any reason to believe multiple ffmpeg processes would perform better than one ffmpeg process with an intelligent -threads setting? [01:39] <Zerxer> I'm trying to use ffmpeg to record a dshow input but right now I'm having trouble with the audio. here's the command I'm using: ffmpeg -f dshow -s 1280x720 -r 59.94 -pix_fmt uyvy422 -i video="Blackmagic WDM Capture" -r 29.97 -pix_fmt yuv420p -t 15 out.mp4 [01:39] <Zerxer> that Blackmagic WDM Capture should include its own audio stream but that command itself makes my video not have any audio, is there something else I have to attach to the command? [01:39] <Aprel> Zerxer: does this help?: http://ffmpeg.org/trac/ffmpeg/wiki/How%20to%20grab%20the%20desktop%20%28scr… [01:41] <Zerxer> Aprel: I'm capturing from an external device but I did try looking into using :audio="" but there's no specific audio stream listed when I list devices. I tried using Blackmagic WDM Capture as both the video and audio settings but it threw an error that the audio stream wasn't found [01:45] <jankowiak> how can I check how long an encode lasted? [01:45] <Zerxer> I may have figured it out [01:46] <jankowiak> I'm looking at the out put but I don't see any time output [01:47] <Zerxer> my last question is this: if I'm capturing & encoding with a command like that, how do I safely tell it to stop capturing/outputting? do I simply close the command prompt, or will that cause it to be unclosed and corrupted in some way? [01:51] <Aprel> Zerxer, when I do screen capture, I use Ctrl-C to stop. If you press it only once, it tells ffmpeg to stop and clean up. 2+ times means abort. [01:52] <Aprel> Hardclosing the command prompt might corrupt the file. It will depend on the container. [01:52] <Zerxer> Aprel: awesome, thanks, Ctrl-C is what I wanted :) can't wait to give this all a spin and see how it does [01:53] <posciak> hi, I wanted to encode a stream that would have SEI recover points instead of IDRs when possible, I thought open_gop to x264 would be the option I need. I tried: ffmpeg -i in.mp4 -c:v libx264 -x264opts open_gop=1 -c:a copy out.mp4 and some variations of open_gop argument, but x264 keeps reporting "open_gop=0"... what am I doing wrong? [01:53] <Aprel> Zerxer, good luck :) [01:53] <Zerxer> the device I'm using outputs at 60fps so every other capture software I've used has wanted to write 60fps to disc and my hard drive is just not fast enough to do that, so I'm very excited that I discovered how to use ffmpeg to both capture it and write it to disc as 30fps [01:53] <Zerxer> hoping to finally be done with out of sync issues [01:55] <Aprel> Zerxer, well I hope it works out for you. I came here trying to work out out-of-sync issue captures myself :/ But I've got a graphics card, so I think that's overcomplicating things. [02:17] <Zerxer> hm still out of sync for me too& maybe I shouldn't try changing the pixel format while capturing? the cmd log did have a lot of dropped frames [02:19] <Zerxer> or if I can increase the buffer, says it hit 121% [02:23] <Zerxer> what's the best way to have it capture the raw/uncompressed video but still have it cut the framerate in half before outputting to a file? [02:28] <Aprel> Zerxer: check your CPU during the capture and see if it's close to 100%. [02:29] <Zerxer> good idea [02:30] <Aprel> Also feel free to report back with any progress you make. I'm having a similar issue with out-of-sync. Could be the ffmpeg code itself? [02:31] <Zerxer> I've been having nothing but out of sync problems with my Blackmagic Intensity Shuttle itself so I'm not sure I'd jump on ffmpeg being the cause for my issue :( [02:31] <Zerxer> the conclusion I came to is my hard drive is definitely not fast enough to handle writing 720p at 60fps [02:31] <Zerxer> which is why I'm resorting to ffmpeg now to cut the frames in half. my last option that I'm avoiding is to use motion jpeg [03:44] <Zerxer> I checked my CPU usage while recording and it stayed around 60% [03:44] <Zerxer> but the cmd log, frequently, would say in red text "real-time buffer 121% full! frame dropped" [03:49] <Aprel> hm, then you could be right about the hard drive being the bottleneck. btw, the blackmagick device, is that over usb (sorry I have no idea what that is) [03:49] <Zerxer> yup, usb 3.0 [03:50] <Zerxer> I know absolutely nothing about video-related items, completely new in this area, but should I also be capping the bitrate of the output file? [03:51] <Zerxer> I might as well try and not change the pixel format of it, see if that helps anything [03:55] <uersbob> hi, im writing a c# screen recording app, and i am using the following line to feed images taken to ffmpeg using image2pipe. It works, but the timestamps of the frames seem to be off, since when i sync it with audio, the frames are a bit to fast near the start, then slowly balance out near the end [03:55] <uersbob> here is the line i am using. -f image2pipe -i pipe:.bmp -c:v libx264 -an -y "video.mp4" [03:55] <Aprel> Yeah, I was thinking maybe the buffer overfill was on the usb side, but shouldn't be a problem with usb 3.0. Could be hard drive, could be ffmpeg...it's a frustrating issue, I know. You could try less bitrate=smaller file=less hd use and see if that helps. btw, what command line command are you using for the capture? [03:56] <Aprel> uersbob: funny you should ask. We have been having out-of-sync capture problems ourselves.... [03:57] <uersbob> this is after i run another ffmpeg command using PTS to make the framerate correct, as a new video file, since it plays to fast normally [03:59] <uersbob> i guess i am asking, how can i output the frames with the correct timestamps, when the recording is taking place [04:10] <Zerxer> I tried making the filename .avi since that's what my other capture programs made, for a raw/uncompressed file, and then I played 10 mins and monitored the output/cpu usage as I did. I was very impressed, no dropped frames, cpu below 15%, thought I figured it out [04:10] <Zerxer> then I went to play back the .avi file and it was very low quality, lol. but doing .mp4 causes 60% and dropped frames :( [04:13] <Aprel> Zerxer: what's your ffmpeg command line to creat teh avi file? [04:14] <Zerxer> ffmpeg -f dshow -s 1280x720 -r 59.94 -pix_fmt uyvy422 -i video="Decklink Video Capture":audio="Decklink Audio Capture" -r 29.97 out.avi [04:14] <Zerxer> (or out.mp4, which makes higher quality video, but dropped frames) [04:15] <relaxed> It drops frame to achieve 30pfs. [04:15] <relaxed> er, 30fps [04:15] <relaxed> Is the input interlaced? [04:16] <Zerxer> I mean it says the buffer is 121% full, frame dropped [04:16] <Zerxer> not that it's dropping frames to achieve the 30fps [04:16] <Zerxer> the input is 720p60 [04:17] <Zerxer> or well 720p59.94 to be completely true [04:17] <relaxed> 59.94 does not imply progressive content. [04:19] <Zerxer> oh, I thought the "p" did though. it's progressive [04:21] <relaxed> I thought fractional frame rates meant the content was interlaced [04:21] <relaxed> the correct way to specify the input frame rate is -r 60000/1001 [04:22] <Zerxer> hm do you think that's contributing to my issue then? [04:22] <relaxed> and the output should be -r 30000/1001 [04:23] <relaxed> You need to watch the source frame by frame and make sure it's not interlaced. [04:24] <Zerxer> I know it's not interlaced at all. it's coming out of an xbox 360 that's outputting at 720p and the blackmagic is set to progressive input as well [04:25] <Aprel> what if you tried -vcodec libx264 -crf 0 -preset ultrafast -acodec pcm_s16le ? This will use hardly any cpu and you can see if the difference is that 15%-60% cpu increase that's causing frame drop (i.e. overfilling the buffer into cpu processing). Maybe the codec can only use 2 cores so it's pushing them all the way and 65% is on a quad core cpu? [04:27] <Zerxer> that's possible. it is indeed a quad core [04:28] <Zerxer> would I put that before or after the -i input? does it apply to the input or the output? [04:29] <Aprel> After -i input. It just means don't do any on-the-fly cpu processing during capture. [04:30] <Aprel> So if the desync looks even worse with that command, it probably means a hdd bottleneck (because it write raw video and uses a lot of hdd space). If the desync goes away, then the cpu isn't keeping up with the encoding and you can capture in raw and encode after you're done capturing. [04:32] <Zerxer> since it's writing raw video, will I still be able to force the output to 30fps? cause I know my hdd is going to bottleneck on the raw 60fps [04:32] <Zerxer> and that'll ruin the test [04:32] <Zerxer> going to run it now [04:32] <Aprel> I think so. [04:33] <Zerxer> got some errors right away with that: track 1: could not find tag, codec not currently supported in container. and Could not write header for output file #0 (incorrect coedc parameters ?): Error number -1 occured [04:34] <Aprel> sorry, you need to use the .mkv container [04:34] <Zerxer> gotcha [04:34] <Aprel> can you play that? Otherwise, it might be compatible with other containers, just not sure which. [04:35] <Zerxer> got the buffer error right as it started, after I hit enter on the command, but it's going through now and cpu usage is below 20%. going to try actually doing things in the game now [04:35] <Zerxer> yeah I can [04:35] <Aprel> k, best of luck [04:42] <Zerxer> no issues at all while recording& 15% cpu, no dropped frames (except it giving the buffer error right when it started and right when it ended). playing back the MKV file now [04:45] <Zerxer> 6 minutes into the video and there's no sync issues! a great sign. you said that this should do no processing though, right? just a raw dump of quality? the quality just seems slightly less than what it was as the mp4 without those settings, but I could just be looking at it too much [04:48] <Aprel> Zerxer: I was under the impression it's raw. Got that command from http://ffmpeg.org/trac/ffmpeg/wiki/How%20to%20grab%20the%20desktop%20%28scr…. [04:49] <Aprel> It might do a very minimal processing. [04:50] <Zerxer> I think it's fine, I have my girlfriend watching the video now and she said the quality is good. I've just been staring at all this for too long and just got done playing it directly off the xbox [04:50] <Aprel> But it says this is the command to use for "uncompressed video/audio" which I read as raw. [04:50] <salamanderrake> hello having issues streaming to twitch.tv http://pastebin.com/SL8Df3jA [04:51] <salamanderrake> I used the command from here http://trac.ffmpeg.org/wiki/StreamingGuide [04:51] <Zerxer> and thanks, you've been extremely helpful :) I hope you're able to resolve the issue you're having [04:52] <salamanderrake> ffmpeg -f x11grab -s 1920x1200 -r 15 -i :0.0 -c:v libx264 -preset fast -pix_fmt yuv420p -s 1280x800 -threads 0 -f flv "rtmp://live.twitch.tv/app/live_********_******************************" [04:53] <Aprel> Zerxer, thanks, glad to help. You can continue to play around with it. Seems like your problem was a combo of hdd write speed and cpu limits. [04:53] <salamanderrake> oops thought you was talking to me Zerxer [05:01] <salamanderrake> shoud -pix_fmt yuv420p be for 16:10 resolution recording? [05:05] <relaxed> salamanderrake: yes, and it's the color format best supported for streaming. [05:05] <salamanderrake> ok [05:11] <salamanderrake> do I have to use libx264 for the video? [05:13] <relaxed> I guess you don't have to but it's a good codec and is well supported. [05:14] <salamanderrake> well for me its borked [05:20] <Zerxer> Aprel: just saw in the cmd output that it actually says to hit q to stop, so there's another way to do it [05:21] <Zerxer> then again, you do screen capture, so cmd may not be in focus for you (just caught myself there) [05:23] <Aprel> Zerxer: you're right. And I've seen that line every time I've used ffmpeg, lol. As soon as I read that, I remembered in my head the '[q]' in the output. Never actually read that whole output line, lol. Thanks for the tip :) [05:24] <Aprel> Zerxer: That might read sarcastic, but seriously, I never internalized the [q] shortcut despite seeing that output line all the time, lol :) [05:24] <Zerxer> I only caught a quick glance of it and was like "wait what was that" [05:25] <salamanderrake> here is the output from my terminal http://pastebin.com/mjGJvDeV [05:25] <relaxed> Zerxer: -t $time [05:26] <Aprel> Zerxer:Ha, yeah, ffmpeg output goes fast. Hope that capture command continues to work well for you. [05:26] <Zerxer> relaxed: ? I was using that flag earlier today but I wouldn't use it for actual capture runs since I don't know how long we'd be playing the game :) [05:27] <Zerxer> if you mean to see the output better, can at least scroll back up through it all once you end the capture [05:28] <Aprel> Zerxer: true. I should pipe it to file more often because it floods my terminal. [05:38] <eggyknap> Can someone explain this "No channel layout" error for me? http://pastie.org/8165503 [05:39] <eggyknap> I originally figured it meant I needed -ac, to see that what ffmpeg thought were stereo inputs were really mono... but I added that, and it correctly sees they're mono, and it still complains [05:39] <eggyknap> The anull filters are just to allow me to refer to streams by nice names rather than [0:0], etc., which is helpful for the more general development I'm doing. [06:23] <salamanderrake> is there a quick way to see if its libx264 thats causing my segmentation fault? [06:34] <Mr_Ewscoba> salamanderrake yes try mpeg2video as vcodec ? [06:49] <salamanderrake> how do I do that? [06:50] <salamanderrake> ffmpeg -f x11grab -s 1920x1200 -r 15 -i :0.0 -c:v libmpeg2video -preset fast -pix_fmt yuv420p -s 1280x800 -threads 0 -f flv rtmp://live.twitch.tv/app/`cat ~/.twitch_key` does not work [06:55] <salamanderrake> ok got the same error [07:10] <Mr_Ewscoba> salamanderrake no [07:10] <Mr_Ewscoba> -c:v mpeg2video [07:11] <Mr_Ewscoba> and preset wont work for mpeg2video [07:11] <salamanderrake> ahh not compatible with flv [07:12] <Mr_Ewscoba> try the mpeg4 codec then not libx264, it should be flv compatible :) [07:13] <salamanderrake> ok mpeg4 works [07:13] <salamanderrake> mpeg4 @ 0x7f0f45c66780] header damaged [07:13] <salamanderrake> saved it to mpg [07:13] <salamanderrake> thats what I got [07:15] <Mr_Ewscoba> try -vcodec msmpeg4 [07:16] <salamanderrake> instead of -c:v ? [07:16] <Mr_Ewscoba> Im not sure, but flv and x264 might not be a good mix, something tell me that you can only use mpeg4 for flv [07:16] <Mr_Ewscoba> yeah or just -c:v msmpeg4 [07:17] <salamanderrake> unknow format [07:17] <salamanderrake> missing an m [07:17] <salamanderrake> and no it did not work [07:18] <salamanderrake> [flv @ 0x6c54c0] Video codec 'msmpeg4v3' for stream 0 is not compatible with FLV [07:18] <salamanderrake> Could not write header for output file #0 (incorrect codec parameters ?): Invalid argument [07:21] <Mr_Ewscoba> What happends if you remove the -c:v blah completely from the line [07:21] <Mr_Ewscoba> ? [07:21] <Mr_Ewscoba> ffmpeg -f x11grab -s 1920x1200 -r 15 -i :0.0 -pix_fmt yuv420p -s 1280x800 -threads 0 -f flv [07:22] <salamanderrake> lol it works [07:22] <Mr_Ewscoba> It should choose the encoder best suited itself then :) [07:22] <Mr_Ewscoba> What encoder did it use then ? :) [07:22] <salamanderrake> looks like dirt [07:22] <salamanderrake> and libx264 worked before [07:22] <relaxed> probably flv as the codec [07:23] <salamanderrake> this script sorta worked I think before [07:24] <salamanderrake> http://pastebin.com/mmavyTQn [07:24] <salamanderrake> sorry [07:24] <salamanderrake> wrong one [07:24] <salamanderrake> thats to save to disk [07:25] <salamanderrake> why doesn't this one work from here for twitch.tv http://trac.ffmpeg.org/wiki/StreamingGuide [07:25] <salamanderrake> ahh its old [07:25] <salamanderrake> thats for 1.0 I have 1.2.1 [07:27] <Mr_Ewscoba> Yup its some old shit, and rtmp and flv is some quite nasty formats with ugly limits, they will always be bugged at best in ffmpeg :) [07:31] <Mr_Ewscoba> If you just want to stream to some vlc clients for example, you'd be better of using the apple segmenter and http. [07:34] <salamanderrake> apple? I am on linux [07:35] <Mr_Ewscoba> -f segment -segment_format mpegts -segment_list_type m3u8 -segment_list /var/www/live/mystream.m3u8 -segment_list_flags +live -segment_wrap 6 -segment_time 10 /var/www/live/mystream-%08d.ts [07:35] <salamanderrake> o ok [07:36] <salamanderrake> was trying to do this for a live session on twitch.tv [07:39] <salamanderrake> I think something is broke on my system [07:39] <salamanderrake> well I am off to bed [09:19] <pinPoint> hello [09:25] <pinPoint> I have question, can ffmpeg be used to address issues with videos? [09:25] <pinPoint> corruption from let's say drive collapsing, etc? [09:27] <pinPoint> I have the file but it is a R3D file... :/ [15:32] <mcr_> Hi, I have a problem. I'm using -t option to trim content. Here is full command I use: ffmpeg -ss 00:00:00 -t 00:10:10 -i big_buck_bunny_1080p_h264.mov -codec copy big_buck_bunny_1080p_h264_short.mov [15:32] <mcr_> I'm getting error: -t is not an input option [15:33] <mcr_> and my file isn't trimmed at all [15:33] <mcr_> when I do -t 00:00:30, ffmpeg trims file to 30 seconds, as expected [15:34] <mcr_> what am I doing wrong? [15:36] <relaxed> mcr_: -t goes after the input [15:39] <mcr_> ok, I have no error when i put -t after input, but output file size stays the same, same as reported length in VLC. [15:41] <mcr_> I'm sorry, everything's ok now :) [15:41] <mcr_> thanks [16:25] <Fjorgynn> :D [16:33] <salamanderrake> >D [16:53] <eggyknap> Can someone explain this "No channel layout" error for me? http://pastie.org/8165503 [16:53] <eggyknap> I originally figured it meant I needed -ac, to see that what ffmpeg thought were stereo inputs were really mono... but I added that, and it correctly sees they're mono, and it still complains [16:53] <eggyknap> The anull filters are just to allow me to refer to streams by nice names rather than [0:0], etc., which is helpful for the more general development I'm doing. [16:53] <eggyknap> (perhaps there's a better way to assign those names... that's the first way I came up with) [18:00] <bAndie9100> hi all [18:03] <bAndie9100> http://pastebin.com/QDpDt4vb [18:03] <bAndie9100> there is my question [18:03] <bAndie9100> i have raw video on stdin and audio on microphone. how to grab them togather? [18:56] <Bourbon> is it possible to do mp4-container output with ffmpeg? [18:56] <Bourbon> I mean, through pipes [19:10] <mkrufky> is there a guide anywhere for porting applications that depend on ffmpeg 1.x to ffmpeg 2.0 ?? i notice that some av_buffersink_foo & avfilter_foo functions have been deprecated [19:11] <Bourbon> there's 2.0? [19:14] <JEEBsv> yes, the current release is 2.0 [19:14] <mkrufky> heh [19:15] <mkrufky> ...then im guessin that you guys dont know of any porting guide or some kind of reference of which functions to use instead of the deprecated ones [19:15] <JEEBsv> the changelog should note some of the stuff [19:15] <JEEBsv> changes.txt or something [19:15] <JEEBsv> also going through the doxygen is useful as well [19:16] <mkrufky> Changelog just lists the new features [19:17] <mkrufky> i was hoping to port my code to use 2.0, in hopes that maybe the 2pass encoding might be better in 2.0 (even tho Changelog makes no mention of improvements to 2pass encoding, i still think its worth a try) [19:18] <mkrufky> ...also i stand to benefit from the 10% efficiency increase in aac [19:18] <JEEBsv> there was some API changes log as well, you should also try reading through the avfilter doxygen if you were using libavfilter before [19:18] <mkrufky> ah, perfect, that should help -- thanks [19:18] <mkrufky> awesome [19:18] <mkrufky> doc/APIchanges [19:18] <mkrufky> thanks :-) [19:23] <axorb> hey guys, is it safe to say that -ss x before -i will seek to the first keyframe after time x? [19:23] <axorb> will always* [19:33] <axorb> I know it will always seek to a keyframe, but is that keyframe always after the given time or the nearest match? [19:36] <xlinkz0> axorb: no [19:36] <xlinkz0> or actually nvm me [19:40] <axorb> the seeking guide example shows the fast seek picking the keyframe before the time, so I guess not [19:41] <axorb> makes sense anyway because you'd rather not seek backwards after going too far [21:04] <mkrufky> ok, so the doc/APIchanges file made the avfilter_graph_parse_ptr() port an easy one :-P [21:04] <mkrufky> but im still hunting for av_buffersink_get_buffer_ref(), av_buffersink_poll_frame(), avfilter_copy_buf_props(), avfilter_unref_bufferp() & avfilter_uninit() [21:04] <JEEBsv> p1 [21:05] <JEEBsv> ugh, damn small laptop keyboard [21:14] <matttt> Hi there I'm trying to take frames from an mts file. What I'm noticing are lines in certain parts of my images however when viewing the video in media player classic these lines are not there even when viewing frame by frame. I'm using the command: .\bin\ffmpeg.exe -i .\00001.MTS -dframes 100 -ss 00:00:00 image-%d.jpeg Any thoughts? [21:24] <wawrek> hello [21:24] <wawrek> [21:25] <wawrek> I have problem trying to run ffmpeg filtering examples [21:25] <wawrek> i get this error: Unknown input format: lavfi [21:44] <llogan> matttt: can you provide a link to one of the images that shows the lines? [21:46] <matttt> llogan: actually if it makes any sense to you I've narrowed it down to ffmpeg and all other programs seem to open my mts file as 1440x1080. Media player classic opens it at 1920x1080. It seems the scaling is what is causing the lines. I'll try get a link to you once I can! [21:54] <wawrek> fflogger: I will do that, i understand [21:56] <matttt> fflogger: will do [21:57] <llogan> fflogger is the channel bot [21:57] <matttt> thanks. always good to know hehe ;) [22:28] <johnsmith> Hi all, has anyone worked with the libav's before? I am just curious on what the "graph" is in the libavfilters, thanks [23:19] <Joske> Hello. av_find_input_format("video4linux2") returns NULL, does this mean FFMPEG no longer supports v4l2? [23:29] <llogan> Joske: ffmpeg still supports it. [23:31] <Joske> llogan: is it normal for it to return NULL? [23:32] <llogan> i'm not sure. i don't use this stuff programtically. [23:52] <Joske> llogan: oh, it's fixed now. i forgot to do avdevice_register_all(); [00:00] --- Wed Jul 24 2013

1 0

ffmpeg-devel.log.20130722
by burek 23 Jul '13

23 Jul '13

[00:39] <Compn> kierank : what airport are you flying into france on ? [00:39] <Compn> :P [00:44] <Compn> assuming flight [02:54] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:682e8a694f6c: avcodec/h264_sei: Try to make code more robust by jumping to the specified SEI ends [02:55] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:5169111145bc: avfilter/REDUCE_FORMATS: fix bug that ended reducing too early [03:11] <kierank> Compn: i am not flying [03:11] <kierank> but you will probably want to go to charles de gaulle [03:32] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:e2b718464e92: swresample: Make rematrix maxvalue user settable [03:32] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:7ac12599fe3a: swresample: fix negative rematrix volumns [04:37] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:5feaa451bb65: doc: add "rematrix_maxval" [10:32] <cone-499> ffmpeg.git 03Derek Buitenhuis 07master:4719040cd1b5: libx264: Define X264_API_IMPORTS on MSVC/ICL [10:32] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:b909428dabe3: Merge commit '4719040cd1b54381796b22056a9ae37bd2b4668d' [10:42] <cone-499> ffmpeg.git 03Hendrik Leppkes 07master:7ef760d95c73: configure: generate full msvc debug symbols when linking in debug mode [10:42] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:2ca6a13b5290: Merge commit '7ef760d95c73dfee021b8fdee04f65126eb7edd1' [10:57] <cone-499> ffmpeg.git 03Justin Ruggles 07master:8000206abcc8: libspeexdec: fix detection of final terminator code [10:57] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:9d4dece6b66c: Merge commit '8000206abcc82491c2a8c71e494c5477ede7264c' [11:48] <cone-499> ffmpeg.git 03Ben Avison 07master:41ef1d360bac: arm: Add VFP-accelerated version of synth_filter_float [11:48] <cone-499> ffmpeg.git 03Ben Avison 07master:ce9ed10ac27b: arm: Add VFP-accelerated version of int32_to_float_fmul_scalar [11:48] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:0129026c4e62: Merge commit '41ef1d360bac65032aa32f6b43ae137666507ae5' [11:48] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:877bbe05cef0: Merge commit 'ce9ed10ac27b9cf32a6257e083ea2f052692d971' [11:56] <cone-499> ffmpeg.git 03Ben Avison 07master:31c6f6f65c0e: fmtconvert: Add a new method, int32_to_float_fmul_array8 [11:56] <cone-499> ffmpeg.git 03Ben Avison 07master:26ffcc7de1d9: dcadec: Use int32_to_float_fmul_array8 [11:56] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:0d3400ec5329: Merge commit '31c6f6f65c0ed5a894e26ce44ab0c3e89c82b9a2' [11:56] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:1573274b914e: Merge commit '26ffcc7de1d98f77400a2ebe4e75055515c54587' [12:04] <cone-499> ffmpeg.git 03Ben Avison 07master:d6e4f5fef0d8: arm: Add VFP-accelerated version of int32_to_float_fmul_array8 [12:04] <cone-499> ffmpeg.git 03Martin Storsjö 07master:b63bb251ea6d: arm: Add VFP-accelerated version of imdct_half [12:04] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:3cab2280058c: Merge commit 'd6e4f5fef0d811e180fd7541941e07dca9e11dc0' [12:04] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:2305a6775dd6: Merge commit 'b63bb251ea6d6ba23295294e37a92625c0192206' [12:11] <cone-499> ffmpeg.git 03Martin Storsjö 07master:ba6836c966de: arm: Add VFP-accelerated version of dca_lfe_fir [12:11] <cone-499> ffmpeg.git 03Martin Storsjö 07master:8b9eba664eda: arm: Add VFP-accelerated version of fft16 [12:11] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:bacba8e5d1e9: Merge commit 'ba6836c966debc56314ce2ef133c7f0c1fdfdeac' [12:11] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:c30eb74d1820: Merge commit '8b9eba664edaddf9a304d3acbf0388b5c520781d' [12:17] <cone-499> ffmpeg.git 03Ben Avison 07master:800ffab48a78: dcadsp: Add a new method, qmf_32_subbands [12:17] <cone-499> ffmpeg.git 03Ben Avison 07master:ff30d121595e: arm: Add VFP-accelerated version of qmf_32_subbands [12:17] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:366e415836c5: Merge commit '800ffab48a7844dd5dc0a33b8f6b8e5ed718cf2e' [12:17] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:23f250d2bc53: Merge remote-tracking branch 'qatar/master' [13:18] <cone-499> ffmpeg.git 03Martin Storsjö 07master:47d57f24e35f: arm: Mangle external symbols properly in new vfp assembly files [17:39] <j-b> anyone has a raw thd samples? [18:03] <hotwings> according to the author of the softhddevice output plugin for vdr, theres a problem with the vdpau decoder not ffw/rew'ing h264 (h264 paff to be exact, iirc). would this be the right place to talk to someone about that? [20:15] <cone-499> ffmpeg.git 03Zhang Rui 07master:2a5891bb9dfe: avformat/hls: parse EXTINF duration as floating-point number [20:15] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:419a3d8a43d2: avformat/hls: avoid floating point arithmetic [20:38] Action: Daemon404 sighs at the continued copious av_restrict warnings [00:00] --- Tue Jul 23 2013

1 0

ffmpeg.log.20130722
by burek 23 Jul '13

23 Jul '13

[00:04] <t4nk574> nah [00:04] <t4nk574> doesnt work with the full command line [00:04] <klaxa> post it [00:05] <t4nk574> ffmpeg -i "$f" -i /home/logo.png -filter_complex overlay="10:10" -y -s 568x320 -c:v libx264 -vf ass="${f%.*}".ass -map 0:3 -b:v 300000 -movflags faststart -coder 1 -flags +loop -partitions +parti4x4+parti8x8+partp4x4+partp8x8+partb8x8 -refs 4 -trellis 2 -chromaoffset 0 -bf 3 -b-pyramid normal -weightb 1 -mixed-refs 1 -8x8dct 1 -fast-pskip 1 -b_strategy 1 -direct-pred 3 -wpredp 2 -me_method hex -subq 6 -me_range 16 -g 250 -b-bias 0 [00:05] <t4nk574> i tried without -map 0:3 before [00:05] <t4nk574> same problem [00:06] <t4nk574> maybe i just need to add ass=blablabal behind overlay [00:06] <t4nk574> where the other complex filter parameter is [00:06] <klaxa> yeah maybe [00:08] <t4nk574> Stream #0:0 (h264) -> overlay:main Stream #1:0 (png) -> overlay:overlay overlay -> Stream #0:0 (?) Stream #0:0 -> #1:0 (h264 -> libx264) Stream #0:1 -> #1:1 (ac3 -> libfaac) Encoder (codec none) not found for output stream #0:0 [00:11] <t4nk574> argh, this is hell [00:12] <t4nk574> just one tiny parameter [00:12] <t4nk574> cant they all wrap it into one simple command?? [00:16] <Mr_Ewscoba> I dont thik you can have filter complex [00:16] <Mr_Ewscoba> and -vf in the same line [00:16] <Mr_Ewscoba> one or the other [00:22] <t4nk574> it seems like it Mr_Ewscoba [00:22] <t4nk574> -vf and filter_complex seem to do the same thing [00:23] <t4nk574> but when I add complex filter only once [00:23] <t4nk574> it tries to output as the .ass file again [00:23] <t4nk574> or simply ignores .ass [00:27] <t4nk574> FINALLY [00:27] <t4nk574> -vf = simple filter and does not work with multiple filter streams [00:28] <t4nk574> so i had to use -filter_complex AND seperate overlay and ass with a comma [00:29] <Mr_Ewscoba> :) [00:29] <t4nk574> without the comma it doesnt load both filters, without the complex filter it does not encode two filters at the same time, without libass it does not parse ass correctly and without the ass conversion , the srt would not play [00:29] <t4nk574> so this was 4 level problem [00:30] <Mr_Ewscoba> subs is always a big problem, the dvbsub support in ffmpeg is so broken im just burning my subs instead as you do. [00:31] <t4nk574> in that case i hope that my problem will help other newbies [00:32] <t4nk574> in case somebody else encouters this problem [01:01] <t4nk574> hmm [01:01] <t4nk574> now there is a new problem sigh [01:02] <t4nk574> Mr_Ewscoba when you convert srt to ass, is there a way to specify screen size, font style? [01:06] <t4nk574> anyone? [01:06] <t4nk574> we cant be the only ones with the problem [01:50] <b6> is there some type of stream that can contain arbitrary text to serve as per-frame metadata for a video stream? [01:53] <Mr_Ewscoba> t4nk574 I would guess so but no idea, I only use ffmpeg for messing with dvb - iptv stuff, where srt/ass is useless :p [03:20] <norbert_> hi [03:20] <norbert_> what exactly is avconv? [03:21] <norbert_> I booted a Mint live CD and when I tried to run ffmpeg, the system said ffmpeg has been replaced by avconv [03:22] <norbert_> is it a fork of avconv [03:22] <norbert_> and, if so, why did the fork take place... what are the differences? [03:25] <klaxa> norbert_: ^ [03:27] <norbert_> thanks [05:14] <juanmabc> hi, my code is from ffmpeg-0.10/0.11+, but on 1.2 i hear sound only on left and low or glitches in right, sounds familiar? many api changes? [07:28] <posciak> hi, I wanted to encode a stream that would have SEI recover points instead of IDRs when possible, I thought open_gop to x264 would be the option I need. I tried: ffmpeg -i in.mp4 -c:v libx264 -x264opts open_gop=1 -c:a copy out.mp4 and some variations of open_gop argument, but x264 keeps reporting "open_gop=0"... what am I doing wrong? [10:04] <natrixnatrix89> I have a weird situation.. I'm creating a hls stream.. and it runs fine until media sequence 21 [10:04] <natrixnatrix89> ffmpeg -f hls [10:04] <natrixnatrix89> after that it stops and doesn't run further [10:05] <natrixnatrix89> I'm wondering why does it always stop on media sequence 21 [10:05] <natrixnatrix89> #EXTM3U [10:05] <natrixnatrix89> #EXT-X-VERSION:3 [10:05] <natrixnatrix89> #EXT-X-TARGETDURATION:10 [10:05] <natrixnatrix89> #EXT-X-MEDIA-SEQUENCE:21 [10:05] <natrixnatrix89> #EXTINF:10, [10:05] <natrixnatrix89> zalajs21.ts [10:05] <natrixnatrix89> #EXTINF:10, [10:05] <natrixnatrix89> zalajs22.ts [10:05] <natrixnatrix89> #EXTINF:10, [10:05] <natrixnatrix89> zalajs23.ts [10:11] <praveenmarkandu> when i output a transcode output to file, i see a lot of ^M in the output before frame= [10:11] <praveenmarkandu> how do i remove that? [10:15] <praveenmarkandu> if i do sed 's/^M//g' log.txt | less it still shows ^M [10:15] <Mr_Ewscoba> I think i had some semi luck at that using unix2dos convert scripts praveenmarkandu [10:15] <praveenmarkandu> i tried dos2unix [10:15] <Mr_Ewscoba> sed can't do it, that was my first try aswell [10:16] <praveenmarkandu> wrong way? [10:16] <Mr_Ewscoba> it can buy you need some special options [10:16] <Mr_Ewscoba> sec [10:16] <Mr_Ewscoba> but* [10:17] <praveenmarkandu> hmmm [10:17] <praveenmarkandu> tried tr , awk, [10:17] <praveenmarkandu> also didnt work [10:17] <Mr_Ewscoba> unix2dos -c 7bit faultyfile [10:17] <Mr_Ewscoba> try that [10:22] <Mr_Ewscoba> found the script where i knew i had used just that [10:22] <Mr_Ewscoba> dos2unix -l -f -c 7bit temp/$name.x264.log [10:26] <Mr_Ewscoba> praveenmarkandu [10:26] <Mr_Ewscoba> user@box:~$ cat log | tail -n 4 [10:26] <Mr_Ewscoba> Stream #0.0 -> #0.0 [10:26] <Mr_Ewscoba> Stream #0.1 -> #0.1 [10:26] <Mr_Ewscoba> Press ctrl-c to stop encoding [10:26] <Mr_Ewscoba> user@box:~$ 25 q=-1.0 size= 37926kB time=79.39 bitrate=3913.6kbits/s [10:26] <Mr_Ewscoba> user@box:~$ cat log | dos2unix -l -f -c 7bit | tail -n 4 [10:26] <Mr_Ewscoba> frame= 2095 fps= 25 q=-1.0 size= 40332kB time=83.56 bitrate=3953.8kbits/s [10:26] <Mr_Ewscoba> frame= 2110 fps= 25 q=-1.0 size= 40554kB time=84.00 bitrate=3955.1kbits/s [10:26] <Mr_Ewscoba> frame= 2134 fps= 25 q=-1.0 size= 41289kB time=85.15 bitrate=3972.3kbits/s [10:26] <Mr_Ewscoba> frame= 2158 fps= 25 q=-1.0 size= 41797kB time=86.01 bitrate=3980.8kbits/s [10:26] <Mr_Ewscoba> user@box:~$ [10:26] <Mr_Ewscoba> ;) [10:28] <praveenmarkandu> hmmm thanks [10:28] <praveenmarkandu> will try it out :D [10:31] <Mr_Ewscoba> np :) [12:01] <elkng> for options like "-ss 00:00:38" is it possible to set fraction of a second ? I would really like to have it start from 38.5 second instead of 38 [12:06] <elkng> also what if I convert audio from video to wav file, should I use "-ab 192" option ? [12:06] <elkng> or wav is lossless raw format ? [12:09] <Mr_Ewscoba> pcm is, wav is a container :) [12:11] <elkng> I use that command "ffmpeg -i file.avi -vn -ar 44100 -ac 2 -ab 192 -y -f wav file_out.wav" is it correct ? [12:11] <elkng> when I use "-f wav" option does ffmpeg understand it should use "pcm" encoding or it just uses wav container ? [12:12] <elkng> and I should also use option for pcm format ? [12:12] <wawrek> is this a channel for users? [12:12] <elkng> wawrek: are you from netherlands ? [12:13] <Mr_Ewscoba> -f rawaudio file.pcm [12:13] <wawrek> elkng: yes, why do you ask? [12:13] <wawrek> elkng: how did you know? [12:13] <elkng> wawrek: yes, why do you ask ? [12:13] <elkng> wawrek: your ip [12:13] <elkng> wawrek: "channel for users", yes, why do you ask ? [12:14] <wawrek> I want to know if it is possible to manipulate the content of video frames while compressing a clip with ffmpeg. we can apply filters.. [12:15] <elkng> wawrek: was that aquestion ? [12:15] <elkng> wawrek: there are filters for ffmpeg, try them out [12:16] <wawrek> yes it is a question, i want to create masks, overlay frames and so on. [12:16] <wawrek> I will do so, but i couldn't find examples so far. [12:17] <elkng> Mr_Ewscoba: if I use "-f rawaudio file.pcm" it says "[NULL @ 0x80896c0] Requested output format 'rawaudio' is not a suitable output format" [12:19] <Mr_Ewscoba> try just raw then [12:20] <elkng> I will use "-f wav" [12:21] <Mr_Ewscoba> Should be the same, just raw audio in a wav container [12:25] <Mr_Ewscoba> -acodec pcm_s16le -f rawvideo file.pcm [12:25] <Mr_Ewscoba> there you have it ^^ [17:26] <Azrael_-> hi [17:28] <Azrael_-> is it possible to supply ffmpeg data on windows from somewhere else beside files? it isn't possible to access the source-files directly, i want to supply the data as a stream to ffmpeg without the need to store it first into a file [17:28] <JEEBsv> stdin is always an alternative, the API also would let you deal with memory if you were using that [17:30] <Azrael_-> how could i use stdin on windows? [17:30] <JEEBsv> -i - ? there's also the pipe protocol [17:30] <JEEBsv> both should be documented [17:30] <Azrael_-> k, will have a look [17:33] <Fjorgynn> You where always on my minds... [17:34] <Fjorgynn> not files? [17:37] <Azrael_-> the big problem with stdin, ffmpeg keeps everything in memory if i supply non streaming-files [17:37] <Fjorgynn> ooooh [17:37] <Azrael_-> or am i wrong? [17:37] <JEEBsv> uhh, wat. No, it shouldn't [17:37] <JEEBsv> are you by chance using powershell or something? [17:38] <JEEBsv> that one actually does that if you try to pipe [17:38] <Azrael_-> using a c#-program directly [17:38] <JEEBsv> ffmpeg itself shouldn't cache more than absolutely needed [17:39] <JEEBsv> (unless you tell it to do something else of course) [17:39] <Azrael_-> k [19:07] <hotwings> according to the author of the softhddevice output plugin for vdr, theres a problem with the vdpau decoder not ffw/rew'ing h264 (h264 paff to be exact, iirc). would this be the right place to talk to someone about that? [19:51] <jankowiak> I've been looking at python wrappers for ffmpeg, and most of them seem to be dead [19:51] <jankowiak> Are there any that are recommended? [19:57] <JEEBsv> nope [20:32] <megaTherion> has the cli changed, I cannot do '-vcodec copy' - it says unknown decoder [20:33] <sacarasc> You've probably put it in the wrong place. [20:34] <megaTherion> sacarasc: ah, you're right [20:41] <natrixnatrix89> I have the following problem.. ffmpeg runs fine and stores an input stream if I run it as a user. [20:41] <natrixnatrix89> But if ffmpeg is ran programmatically by cron, then it freezes after first 2 minutes. [20:41] <natrixnatrix89> Is it because there is no stdout? Is there a way to suppress this behavior? [20:45] <natrixnatrix89> from what I read in stackoverflow, it's due to deadlock.. [20:45] <natrixnatrix89> how exactly do I fix that? [20:57] <Plorkyeran> > /dev/null 2> /dev/null, if that actually is the problem [20:57] <Plorkyeran> although dumping to a log file instead is likely to be useful at some point [22:44] <colinb> Hi all, I am using ffmpeg in the command line to encode my videos and I am running into a problem with concat. I record from the screen in FRAPS and then encode each piece, finally stitching them together with the concat demuxer. This time, around the 5th file, right at the stitch, things get weird. I don't know the technical terms but it seems like the dithering or encoding from frame to frame has a problem. Very blocky and distorted whe [22:44] <colinb> re both the source and the encoded files are clean. Any ideas? [22:45] <colinb> All the files are encoded with the same parameters and bitrate, by the way [22:46] <colinb> I even tried to re-concat them but no dice :| [22:51] <colinb> Howdy Carraway [22:51] <Carraway> hellooooo [22:52] <colinb> how goes it [22:52] <colinb> Are you familiar with the concat demuxer by any chance? :) [23:08] <colinb> Ugh, the new Gmail key shortcuts just made me prematurely send an e-mail to the ffmpeg users list :| [23:13] <brontosaurusrex> colinb, no big deal, just spammed few 1000 users. [23:13] <colinb> I know :( [23:13] <colinb> I blame it all on Gmail though! [23:13] <brontosaurusrex> just kidding. [23:16] <colinb> Agh this is such strange behavior. I am experiencing major distortion at the stitch point between two videos.. so I tried concatting the first section and then the second section at the problem point, both video are fine, but when I try to concat the two new sections together I get the same problem at the same point [23:17] <colinb> perhaps I need to re-encode the video at the problem point? even though the encoded section of video seems to be fine on its own? [23:24] <colinb> Nope, still the same issue. is it possibly because the clip preceding the clip which errors out is only 5 seconds long? [23:29] <brontosaurusrex> colinb, what codec they use? [23:30] <colinb> original is fraps to mpeg4 I believe [23:30] <brontosaurusrex> i mean the clips that you are concating? [23:30] <colinb> I have successfully used this method for several videos before this [23:31] <colinb> umm just to be certain and clear is there a ffmpeg command to analyze a file? [23:31] <brontosaurusrex> there is [23:32] <colinb> yes, mpeg4 [23:32] <colinb> in an avi container [23:33] <brontosaurusrex> and concat docs are clear that such codec/container is supported? [23:33] <colinb> FRAPS -> FMP4 [23:33] <colinb> I _think_ so.. but I am not entirely sure let me check again [23:34] <brontosaurusrex> i still don't know what you are concating, fraps or mpeg4, so yeah ... [23:34] <brontosaurusrex> the usual "paste the complete command and output to the paste bin." [23:34] <colinb> mpeg4 [23:34] <colinb> sure, lemme do that [23:38] <colinb> here's the paste for the encoding: http://pastebin.com/i1Ph4cdJ [23:39] <colinb> and the concat action http://pastebin.com/SBthgxbX [23:40] <colinb> sorry that concat paste got mangled a bit [23:40] <colinb> here's a cleaner version: http://pastebin.com/dvtiBGX4 [23:40] <brontosaurusrex> i would assume this has something to do with fraps and variable frame rate and variable frame rate flags and avi [23:41] <brontosaurusrex> so you are probably concating multiple streams with different framerates and ... [23:41] <colinb> could you perhaps elaborate on VBR flags and avi? are you saying that the vbr is causing issues? are you saying there is a vbr flag that might assist me? [23:42] <colinb> the strange thing is that all should be exactly the same.. bitrate, framerate.. [23:42] <colinb> I appreciate your attention I am just (obviously) a encoding newb [23:43] <brontosaurusrex> can you analize few of them with ffmpeg/ffprobe and see whats up with framerate? [23:44] <brontosaurusrex> basically what i would do is intermediate to some uncompressed constant-frame-rate format, then concat, then encode to final format [23:45] <colinb> do you have a suggestion for a constant rate format/codec? [23:46] <brontosaurusrex> perhaps ut-video or fv1 [23:47] <colinb> okay thanks for the assistance! I've run out of time to work on it for now so I gotta run! [23:47] <brontosaurusrex> have fun [23:50] <Aprel> hi compiled ffmpeg from source. Works great from cli, but video software like kdenlive can't find the binaries. How to I set up the proper links? On ubuntu 13.4 [23:53] <brontosaurusrex> Aprel, /j #kdenlive [23:53] <Aprel> One more q: When I do a screen capture, the audio and video desyncs, about a couple frames every 1-2 minutes--more frames if there's a lot of motion in the capture. Any way to avoid this, or programmatically repais desynced files? [23:54] <brontosaurusrex> Aprel, post the command line you use for capturing and all [23:55] <Aprel> I use this command: ffmpeg -f x11grab -r 25 -s 1920x1080 -i :0.0 -f alsa -ac 2 -i pulse -vcodec libx264 -crf 0 -preset ultrafast -acodec pcm_s16le output.mkv [23:55] <Aprel> I've got a i5 2500K cpu, and it's no wherre near %100, so it's not a cpu bottleneck. [23:56] <klaxa> might be harddrive bottleneck [23:57] <Aprel> klaxa: for a desync? I'd think hdd would cause frame and audio drops. [23:57] <klaxa> hmm true [23:57] <brontosaurusrex> Aprel, you can try -crf 23 or something like that [23:58] <brontosaurusrex> to see if it is a hdd issue [23:59] <Aprel> yeah, okay. Thing is I need to record like 8 mins video and watch it before the sync is easily noticeable to me, so hang tight :) [23:59] <Aprel> btw what does the crf switch do? [00:00] --- Tue Jul 23 2013

1 0

ffmpeg-devel.log.20130721
by burek 22 Jul '13

22 Jul '13

[01:08] <wm4> wow that ticket [01:08] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:3242747df392: avfilter/vf_scale; fix indent [01:08] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:0c480bcfd932: doc/filters: document vf_scales color range [01:08] <wm4> it even has graphs and stuff [01:10] <wm4> michaelni: are you not confident that avcodec_enum_to_chroma_pos() is the right thing? [01:13] <michaelni> just thought it cant hurt to wait a day or 2 extra to make sure noone has a better idea [01:20] <michaelni> (thats more about the chroma pos thing that the function depends on) [01:55] <Compn> michaelni : did rockbox ever reply about fixed point fft from mips ? [01:55] <kierank> yes [01:55] <kierank> see ml [01:55] <Compn> oh [01:57] <Compn> of course, :) [01:57] Action: Compn grumbles to himself [01:58] <kierank> Compn: are you going to come to vdd [01:58] <Compn> i was thinking about it [01:59] <Compn> i'd like to see france [02:00] <Compn> guess i'd have to register [02:00] <Compn> and ask those videolan guys :) [02:01] <Compn> sure why not [02:01] Action: Compn is shy person [03:44] <cone-196> ffmpeg.git 03James Almer 07release/1.0:ddfff763978d: oggparseskeleton: Fix fisbone header parsing [03:44] <cone-196> ffmpeg.git 03James Almer 07release/1.0:2d66486b8eab: oggparseskeleton: Replace avpriv_report_missing_feature() with a normal av_log() call [03:44] <cone-196> ffmpeg.git 03Michael Niedermayer 07release/1.0:be23cee317f5: oggparseskeleton: avoid header parsing failure [03:44] <cone-196> ffmpeg.git 03Dale Curtis 07release/1.0:858e06aabe12: avformat/utils: Keep internal and external av_read_frame() packets in sync. [03:44] <cone-196> ffmpeg.git 03Michael Niedermayer 07release/1.0:b896c1c98e06: update all trac links to use the trac subdomain [03:44] <cone-196> ffmpeg.git 03Michael Niedermayer 07release/1.0:9c7c2abad4a8: swfdec: check lens validity [03:44] <cone-196> ffmpeg.git 03Justin Ruggles 07release/1.0:d46713cc7e7d: swfdec: do better validation of tag length [03:45] <cone-196> ffmpeg.git 03Michael Niedermayer 07release/1.0:b8b73739141e: Merge remote-tracking branch 'jamrial/release/1.0' into release/1.0 [04:46] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:1434df3b93fd: jpeg2000dec: Support non subsampled 9-16bit planar pixel formats [04:46] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:7b79c84b40b7: avcodec/snowdec: remove assert.h include, its unused [04:46] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:953c7550f832: avcodec/snowenc: convert speed relevant asserts to av_assert2() [05:31] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:1f7acf3cff06: vfilter/vf_scale: avoid using "{}" [11:34] <bernie_> sigh [11:54] <cone-525> ffmpeg.git 03Luca Barbato 07master:fcae3ff124ee: dsicinav: K&R formatting cosmetics [11:54] <cone-525> ffmpeg.git 03Michael Niedermayer 07master:75fbe41fc61e: Merge commit 'fcae3ff124ee97c9265e3b93f3d41238b2aee9bd' [12:08] <cone-525> ffmpeg.git 03Luca Barbato 07master:dd0bfc3a6a31: dsicinav: Bound-check the source buffer when needed [12:08] <cone-525> ffmpeg.git 03Michael Niedermayer 07master:0a28a52d53e2: Merge commit 'dd0bfc3a6a310e3e3674ce7742672d689a9a0e93' [13:32] <cone-525> ffmpeg.git 03Luca Barbato 07master:fd8189932147: dsicinav: Clip the source size to the expected maximum [13:32] <cone-525> ffmpeg.git 03Michael Niedermayer 07master:db0384c9c962: Merge remote-tracking branch 'qatar/master' [15:54] <cone-499> ffmpeg.git 03Anshul Maheshwari 07master:0f229f9b91fe: avformat/utils: Close codec context since it is allocated by avformat_new_stream in refrence to ticket 2716 [16:11] <cone-499> ffmpeg.git 03Joakim Plate 07master:8710a634a5d4: h264: add frame packing as stereo_mode frame metadata [18:03] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:37ecd67b5e14: Revert "avformat/utils: Close codec context since it is allocated by avformat_new_stream in refrence to ticket 2716" [18:03] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:800ffa1fc089: avcodec/h264_sei/ff_h264_decode_sei: fix error codes for insufficient data [18:03] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:80c873a87ed0: avcodec/ff_h264_decode_sei: fix integer overflow with size. [18:03] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:45da7a0e8f27: avcodec/ff_h264_decode_sei: Check SEI size [18:03] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:118a1c2f4c23: avcodec/ff_h264_decode_sei: use skip_bits_long() [18:11] <cone-499> ffmpeg.git 03Marton Balint 07master:b198e33ad8b3: ffplay: fix build with disabled avfilter [18:11] <cone-499> ffmpeg.git 03Michael Niedermayer 07master:9042c8a4db9b: Merge remote-tracking branch 'cus/stable' [19:29] <cone-499> ffmpeg.git 03Stefano Sabatini 07master:f9f7f4c9bbd1: ffplay: improve error message in case of failure, mention filtergraph configuration [00:00] --- Mon Jul 22 2013

1 0

ffmpeg.log.20130721
by burek 22 Jul '13

22 Jul '13

[02:03] <Gnintendo> I recently upgraded to ffmpeg 1.2.1 from 1.0.(?), and my filter chain is no longer working [02:03] <Gnintendo> the error thrown is "Size and width/height expressions cannot be set at the same time." [02:03] <Gnintendo> and it is referring to the [0]scale=1280/6:-1[pip] filter in my chain [02:04] <Gnintendo> I am unsure exactly how to rectify this issue to still achieve the result I am aiming for here; any assistance would be appreciated [06:41] <elkng> I use few commands to convert same file into few others with different bitrate, I do it in two pass encoding, so first step command is: "ffmpeg -i file.mkv -c:v mpeg4 -vb 600k -s 650x365 -mbd rd -flags +mv4+aic -trellis 2 -cmp 2 -subcmp 2 -g 300 -pass 1 -passlogfile log_1 -an -f rawvideo -y /dev/null" then on that first pass it creates file [06:41] <elkng> "log_1" and then goes second pass that uses information from that "log_1" to createt new video file, then I change option "-vb 600k" to "-vb 800k" and do the same two pass encoding to get second video file with 800 bitrate and then same for 1000,1200,1500 bitrate files, but the strange thing is the first pass step should be similar for all of [06:41] <elkng> them and it should create "log_*" files that absolutely the same but they actually differ, why ? [07:22] <elkng> another question [07:23] <elkng> to cut out some piece of video I have two alternatives: "ffmpeg -ss 00:02:45 -t 00:06:54 -i file.avi -vcodec copy -acodec copy file_out.avi" or "mencoder -ss 00:02:45 -endpos 00:06:54 -oac copy -ovc copy file.avi -o file_out.avi" which way is more prefferable, "ffmpeg" or "mencoder" ? [07:24] <elkng> I have hyperthreaded CPU, and I noticed if I encode video with ffmpeg its about twice faster, so does ffmpeg use hyperthreading for every activity so even with cuting out from video will be faster with ffmpeg ? [07:49] <elkng> is it because of sunday no one here ? [11:25] <MrE_> What tool would you guys recommend to fetch a hls stream and broadcast it as udp/rtp? [11:26] <MrE_> Tried ffmpeg, ffmpeg -i url -vcodec copy -acodec copy - f mpegts "udp://239.0.0.1:5501?ttl=1&pkt_size=1316&buffer=262140" but that seems to dump every segment to udp every 10 seconds, and not properly stream it. [11:27] <MrE_> Adding -re fix that, but then ffmpeg will get out of sync with the segment list and skip segments. [16:05] <stefanha> Can ffmpeg concatenate ogg files and insert silence between some of the file segments without re-encoding? [16:06] <stefanha> From the documentation it seems the concat demuxer can concatenate without re-encoding. I don't see a way to set the starting time of each ogg segment though (I'd like to insert some silences). [16:06] <stefanha> For example: a.ogg, b.ogg, silence (10 seconds), c.ogg -> output.ogg [16:09] <saste> stefanha, what's the problem with generating a silence file, and then use concat? [16:09] <stefanha> saste: I'll try that. Not very familiar with ffmpeg so I wanted to check first. [16:10] <saste> you can use aevalsrc for example, or a simple read from /dev/zero [16:10] <stefanha> saste: Excellent, thanks! [16:42] <newbie|3> what does AV_DICT_IGNORE_SUFFIX do? [16:42] <newbie|3> can't seem to find any docs on it [16:42] <newbie|3> "metadata is read using the AV_DICT_IGNORE_SUFFIX flag" [16:42] <newbie|3> what does that mean? [16:44] <newbie|3> also, I can't really make sense of "filter_codec_opts", which seems to assume the type specificiers are prefixes, however, in all the docs I've found they are suffixes, e.g. "b:v" whatšs goign on there? [19:13] <1JTAADT7Z> Use -vf pad=width:height:x:y:color as in libav doc found here: [19:14] <1JTAADT7Z> i wanted to add 139pixels to top and bottom. What would be the pad parameters then? [19:35] <Yagger> I have a question concerning the --enable-sram configure flag. Will it make a difference when using the flag when compiling for a Core i7 machine? [19:43] <MrE_> What tool would you guys recommend to fetch a hls stream and broadcast it as udp/rtp? [19:43] <MrE_> Tried ffmpeg, ffmpeg -i url -vcodec copy -acodec copy - f mpegts "udp://239.0.0.1:5501?ttl=1&pkt_size=1316&buffer=262140" but that seems to dump every segment to udp every 10 seconds, and not properly stream it. [19:43] <MrE_> Adding -re fix that, but then ffmpeg will get out of sync with the segment list and skip segments. [19:45] <MrE_> Running ffmpeg twice one to do the downloading part and mux it as a mkv file, and then 10 seconds later, another ffmpeg that read that mkv file and broadcast it as udp, everything works, but should be posible with 1 command and no temp file... [19:58] <1JTAADT7Z> man, just noticed my weird nick. That's wrong [19:59] <defaultr0> i got the padding working [20:00] <defaultr0> I'm running this now and bitrate is only 10000kbit. I want it higher since source is huffyuv(lossless). I am happy if bitrate was showing 26000kbits [20:00] <defaultr0> ffmpeg -i g200mm.mkv -vf scale=1920x802,pad=1920:1080:0:139:black -vcodec libx264 -pix_fmt yuv420p -an -crf 15 -preset veryslow output2.mp4 [20:27] <relaxed_> defaultr0: use a lower crf value [20:32] <MrE_> or dont use crf, use -vb 26000k [20:36] <relaxed_> right, and if you want a specific size use two passes [21:38] <akovia> Hi, I am a moderator over on fanart.tv and have been using ffmpeg to make video tutorials for the site and for some users. The first ones I created seemed to not have any troubles but lately people have been having trouble playing them back. I asked for help and followed the "HOWTO Proper Sceencasting" guide on the ubuntu forums (which is of course down right now), but the help I received didn't fix the problem. My system and ffmpeg sp [21:38] <akovia> ecs to follow. [21:39] <akovia> Specs http://pastebin.com/DhpKE686 [21:39] <Fjorgynn> akovia: what are you using now? [21:39] <Fjorgynn> ffmpeg? [21:39] <Fjorgynn> so what's the output [21:39] <akovia> coming [21:39] <Fjorgynn> if you do ffmpeg -i latestscreencast.mp4 [21:39] <akovia> http://pastebin.com/tpwsFSXB [21:41] <akovia> and here's the script I use http://pastebin.com/2gf23nzm [21:42] <akovia> all that shows the info you needed or you need me to run that command? [21:42] <Fjorgynn> so you can watch it? [21:42] <akovia> it all plays back fine on my machine [21:43] <akovia> here's where i think the problem lies... [21:43] <akovia> if i open the file in smplayer, i'll get a time of around 1 minuts, in avidemux, it's like 20% shorter [21:44] <akovia> so i thought it may be a framerate thing? [21:44] <Fjorgynn> avidemux? [21:44] <Fjorgynn> why do you use that? [21:44] <akovia> well i noticed the discrepancy when doing some editing. [21:45] <akovia> i use avidemux to find good cutting points [21:46] <akovia> looking for keyframes, then i use mmg to do the cutting [21:46] <akovia> so this last screencast shows 57 seconds in smplayer, and only 36 seconds in avidemux [21:47] <akovia> it may not be related why some people can't watch them, but I didn't have any other ideas [21:50] <akovia> I've searched extensively to find a way to cut mkv's with frame accuracy, but never found a way. At least if i cut at the keyframes shown in avidemux, it always seems close to where i want the cut [22:02] <Mr_Ewscoba> akovia VideoRedo or TS Doctor [22:03] <Mr_Ewscoba> Both are made for cutting transport streams, videoredo should be able todo mkv files and frame accurate cutting :) [22:03] <akovia> thanks, but i think they are both for windows [22:04] <Mr_Ewscoba> They are, there is nothing that works for linux yet that is frame accurate and doesn't create artifacts [22:04] <akovia> yeah, that's what i found as well =/ [22:05] <Mr_Ewscoba> I need to run a virtual windows to be able to use videoredo with as near as host operation disk speeds [22:06] <akovia> i'm afraid my computer is quite up to the tak of video editing in a VM [22:06] <akovia> task [22:07] <Mr_Ewscoba> you could try wine, ive never had any luck with it thou [22:08] <akovia> i use wine for a few things, but it's more trouble than it's worth if it doesn't "just work" [22:08] <Mr_Ewscoba> tried mencoder and edl files ? [22:08] <akovia> edl files? [22:09] <Mr_Ewscoba> http://en.wikipedia.org/wiki/Edit_decision_list [22:09] <akovia> Fjorgynn, here is a link to one of the videos if you are still around [22:09] <Fjorgynn> hm [22:10] <akovia> will look into that, thanks! [22:10] <akovia> https://dl.dropboxusercontent.com/u/93550827/Toke2.mkv [22:11] <Mr_Ewscoba> There is a program called comskip that does commercial detection and creates such edl files, for 3rd party tools to use for cutting out video parts, you can find much more on that over at comskip's forum ;) [22:11] <akovia> Mr_Ewscoba, I'm familiar with mencoder, but never played with it. edl files look interesting though [22:11] <Fjorgynn> if you want to edit mp4-files you can use like kdenlive or OpenShot [22:11] <Fjorgynn> on linux [22:12] <Mr_Ewscoba> both sucks at cutting video parts out without artifacts Fjorgynn :) [22:12] <akovia> openshot and piviti both i've had bad luck with. they seem like dead projects. I'm assuming kdenlive requires all the KDE libraries to be installed which isn't an option =/ [22:13] <akovia> unfortunately, video editing in linux could use some more love [22:14] <akovia> Mr_Ewscoba, are you on windows? [22:14] <akovia> have you tried playing that video i linked? [22:15] <Mr_Ewscoba> Yes I am one of those that use windows on my main desktop and linux on everything else in the house :D [22:15] <akovia> hehe [22:15] <akovia> it took me a couple years to cut the strings but I'm glad I did [22:17] <Mr_Ewscoba> Hehe i can see the video fine open's up in vlc, not that i get anything out of it, never played with such video editing only cutting and reencoding :D [22:19] <akovia> i recommended those that couldn't play my videos to try VLC or MPC-HC, and most say they can watch them after that. Just curious why it can't be played without those. [22:22] <Mr_Ewscoba> Input #0, matroska,webm, from 'Toke2.mkv': [22:22] <Mr_Ewscoba> Metadata: [22:22] <Mr_Ewscoba> ENCODER : Lavf53.21.1 [22:22] <Mr_Ewscoba> Duration: 00:05:19.06, start: 0.000000, bitrate: N/A [22:22] <akovia> what would be a good command to re-encode to mp4 without taking forever but decent quality. [22:22] <Mr_Ewscoba> Stream #0.0: Video: h264 (High 4:4:4 Predictive), yuv420p, 1680x1026, PAR 1:1 DAR 280:171, 30 fps, 30 tbr, 1k tbn, 60 tbc (default) [22:22] <Mr_Ewscoba> No faults with that video, seems to work fine on anything based on ffmpeg/libav [22:23] <akovia> hmmm [22:24] <akovia> maybe i'm just getting people that don't play back many videos complaining then [22:24] <akovia> just thought it was curious to have such a big time discrepancy in avidemux. [22:25] <akovia> might not be the problem at all then [22:25] <akovia> that still screws up my editing flow though =/ [22:26] <akovia> it's like avidemux is basing the time on total frames of the clip to determine the time [22:26] <akovia> so when my framerate drops below 30, it cuts the time [22:31] <t4nk574> hi @ll [22:32] <t4nk574> Hope you are fine [22:32] <t4nk574> I understand it that this is THE way to go to, when it comes to ffmpeg settings [22:33] <t4nk574> I have been trying to hardcode external .srt files into a mp4 container for hours, but to no avail. [22:33] <t4nk574> After reading the documentary and the man ffmpeg, it pains me to say that I still do not understand why the srt are not included in the output mp4 file [22:34] <t4nk574> and thought maybe I could find some help from an experience encoder here? [22:35] <t4nk574> running ffmpeg 2.0 [22:35] <t4nk574> does somebody know something? [22:38] <Mr_Ewscoba> tryin to mux srt with mp4, or hardcode it onto the vide [22:38] <Mr_Ewscoba> video* [22:38] <Mr_Ewscoba> ? [22:38] <Mr_Ewscoba> akovia variable frame rate ? [22:40] <akovia> ffmpeg -f x11grab -s $SIZE -r $RATE -i :0.0 -vcodec $VCODEC -preset $PRESET -crf 0 -threads 0 -y "$OUTPUT"/"$input".mkv [22:40] <akovia> rate at 30 [22:40] <akovia> so i guess not [22:40] <t4nk574> mr_ewscoba to hardcode it into mp4 [22:41] <Mr_Ewscoba> convert it to ass and then, -vf aas=file.ass [22:42] <saste_> Mr_Ewscoba, simpler -vf subtitles=file.any [22:42] <akovia> if i re-encode to mp4 and open in avidemux the time is correct, but it crashes avidemux if I cancel safe mode [22:43] <akovia> "H.264 detected [22:43] <akovia> If the file is using B-frames as reference it can lead to a crash or stuttering. [22:43] <akovia> Avidemux can use another mode which is safe but YOU WILL LOSE FRAME ACCURACY. [22:43] <akovia> Do you want to use that mode?" [22:43] <Mr_Ewscoba> Sure, do you know how to use an internal track like if input is from a dvb adapter and stream has text based dvb subs saste_ ? :D [22:45] <JEEB> akovia, yes avidemux sucks balls handling b-frames even in containers where it's completely OK [22:45] <JEEB> that warning literally should be limited to AVI [22:45] <JEEB> anyways, that just shows how much of a trainwreck avidemux is ^^; [22:45] <Mr_Ewscoba> avidemux sucks balls at almost everything, stupid useless bug filled app :p [22:45] <akovia> lol, wish there were better options [22:46] <akovia> i use mmg for most editing, but just use avidemux for cut points [22:46] <JEEB> I generally use aegisub for the cut points [22:46] <JEEB> uses ffms2 for video loading [22:46] <klaxa> aegisub performes unexpectedly well on my computer [22:47] <akovia> never thought of that [22:47] <akovia> i have that installed [22:47] <akovia> hmmm, thought i did [22:47] <Mr_Ewscoba> ffmpeg -i inputfile.mkv -vcodec copy -acodec copy -f mpegts out.ts [22:48] <Mr_Ewscoba> you will have much more luck cutting in such ts file, than a mkv container [22:48] <JEEB> actually I wouldn't be that sure of it, since the tools for cutting on random access points in matroska are actually pretty usable :) [22:48] <Mr_Ewscoba> There are tons of programs made for that since thats what you get from a satelite cable tv etc feed [22:50] <akovia> well if i use my smplayer to just look for cut points where i want them in the timeline and then try using mmg to cut at those points i get really random results. If i use avidemux to search to keyframes and make my cuts there, i seem to get really good results [22:50] <Mr_Ewscoba> Well you can be sure of that, transport streams are most cutting friendly, VideoRedo for example first got mkv support after tvheadend came arround and in its first manny versions only could dvr mkv [22:51] <Plorkyeran> transport streams are no more cutting friendly than matroska, but what is in them often is [22:51] <Plorkyeran> TV broadcasts don't have the 10+ second GOPs that you often see in internet-distributed videos [22:52] <Plorkyeran> it'd kinda suck to have it take up to ten seconds for things to look right after changing channels [22:52] <t4nk574> Mr_Ewscoba , when I add that parameter it says Output file #0 does not contain any stream [22:52] <Mr_Ewscoba> A proper program like videoredo can do cutting on non gop boundary also.... it will just reencode that little part ;) [22:53] <akovia> i thought i had Aegisub installed and I did. Is there anything special you do to pick out cut cut points JEEB? [22:54] <t4nk574> im trying to convert .avi + .srt / .ass -> mp4 [22:54] <t4nk574> but it always says Output file #0 does not contain any stream [22:57] <JEEBsv> akovia: the time/framenum thing goes light blue with "random access points" [22:57] <JEEBsv> and they are visible on the timeline [22:58] <Mr_Ewscoba> t4nk574 can you put the complete command and output on paste ? [22:58] <t4nk574> on pastebin? [22:58] <Mr_Ewscoba> yup [23:00] <akovia> JEEBsv, I'm not seeing that on ver 3.0.1 [23:01] <JEEBsv> after loading a video file? [23:01] <JEEBsv> anyways, 3.x when I last tried still was doing that [23:02] <t4nk574> Mr_Ewscoba http://pastebin.com/PXJry2MY [23:02] <akovia> yeah, loaded that same video, tried sliding the seek video slider and also just let it play but the time/frame never highlights [23:03] <akovia> oh well, was worth a try [23:08] <t4nk574> Mr_Ewscoba have you seen it? [23:18] <t4nk574> hmm [23:18] <t4nk574> feels so unlucky [23:19] <t4nk574> why is it so difficult to hardcode subs with ffmpeg :( [23:20] <klaxa> but it's not [23:20] <klaxa> or is it? [23:20] <klaxa> -vf ass=subtitles [23:23] <t4nk574> doesnt work for me :( [23:23] <t4nk574> see the pastebin log i posted 23:02 [23:23] <Mr_Ewscoba> back [23:27] <klaxa> t4nk574, you didn't even compile ffmpeg with --enable-libass [23:29] <akovia> Just wanted to say thanks to all who helped. I'll try some of the suggestions and hopefully get a workflow going that works. Cheers! [23:30] <t4nk574> klaxa im doing now [23:33] <t4nk574> klaxa Output file #0 does not contain any stream [23:33] <t4nk574> that's what i still get ^^ [23:33] <klaxa> haha [23:33] <t4nk574> compiling ffmpeg with libass didnt seem to help [23:34] <klaxa> well try addming -map 0 [23:34] <klaxa> i doubt it'll work, but it's worth a shot [23:34] <t4nk574> does it matter where exactly? [23:34] <t4nk574> well i dont exaclty understand how map works [23:34] <klaxa> i think so, i'm not sure though, put it after the input [23:34] <t4nk574> ok [23:35] <klaxa> -map is pretty straightforward (and useful) once you understand the syntax [23:36] <t4nk574> klaxa http://pastebin.com/yytw0es3 [23:36] <t4nk574> welcome back Mr _Ewscoba [23:36] <klaxa> haha [23:36] <klaxa> >Output #0, ass, to 'ass=Man.Called.God.E01.450p.HAN.dramayou.com.ass': [23:36] <klaxa> that's uh... the problem [23:36] <t4nk574> oh i see [23:36] <t4nk574> so it thinks that im trying to output as as ass [23:37] <klaxa> ffmpeg wants to put stuff in the .ass container but only text goes into .ass so everything else is unmapped [23:37] <t4nk574> but you guys told me to wrap it as ass [23:37] <klaxa> so nothing is left [23:37] <klaxa> can you post the command line again? [23:37] <klaxa> oh [23:37] <klaxa> wait [23:37] <klaxa> >ffmpeg -i "$f" ass="${f%.*}".ass [23:37] <klaxa> make it: ffmpeg -i "$f" -vf ass="${f%.*}".ass [23:39] <t4nk574> at least its encoding now [23:39] <t4nk574> let see if there are subs [23:39] <t4nk574> i thought -vf was depreciated for -filter_complex [23:41] <klaxa> huh? maybe so... [23:43] <t4nk574> hm, ok so it works for the test file [23:43] <t4nk574> lets see if it works for the complete encoding line [23:43] <klaxa> hardsubbing for hardware compatibility? [23:48] <holgersson> Hi all [23:50] <Mr_Ewscoba> t4nk574 [23:50] <Mr_Ewscoba> Output #0, ass, to 'ass=Man.Called.God.E01.450p.HAN.dramayou.com.ass': [23:50] <klaxa> Mr_Ewscoba: we resolved that issue ;) [23:51] <holgersson> I want to downmix a soundstream and read several old posts at different places, that ffmpeg doesn't downmix. Instead, it was 'accused' to use just 2 channels and drop the rest. [23:51] <Mr_Ewscoba> kk Looks like somehow ffmpeg thinks it need to output aas not mp4 ^^ [23:51] <holgersson> Is this right? Were it right? [23:51] <klaxa> doesn't libavfilter have a mix audio filter? [23:52] <klaxa> http://ffmpeg.org/ffmpeg-filters.html#amix [23:52] <t4nk574> yes, klaxa pointed that out [23:53] <t4nk574> i fixed that part [23:53] <t4nk574> now it s a bit more complicated as i try to put it all into the encoding command line [00:00] --- Mon Jul 22 2013

1 0