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

ffmpeg-devel.log.20130720
by burek 21 Jul '13

21 Jul '13

[00:00] <bernie_> grrr... I'm a git patch newbie... I have 2 commits ahead of master, but I can only seem to grab each diff for each commit. I can do git format-patch -1 --stdout > patch, or I can do git format-patch -2 --stdout > patch, but I want to have a combined patch... [00:01] <wm4> AFAIK format-patch doesn't combine commits to a single patch [00:01] <wm4> you _could_ use git diff for that [00:01] <wm4> you could also just squash the two commits into one [00:01] <wm4> but I think if they are supposed to be separate commits - just send two patches [00:02] <bernie_> okay, tell me how to squash (I don't know how) [00:02] <bernie_> no, it's just fixing the bug I made earlier... no need to have 2 separate ones [00:02] <bernie_> it's just going to you, so I could just send you the patch for my patch ;) [00:03] <wm4> you could use "git reset origin/master" to go back to the upstream branch, and turn all commits into local changes [00:03] <wm4> then you can make a commit with these changes [00:03] <wm4> a patch on top of the previous patch would be ok [00:04] <bernie_> bombs away [00:05] <bernie_> I do have to say that I'm sorry to be chewing up your time with testing... [00:05] <wm4> no problem [00:05] <wm4> I (probably) want to implement this feature too [00:05] <wm4> for my own demuxer [00:05] <wm4> so some testing is good [00:05] <wm4> we already found a mkvmerge bug [00:06] <bernie_> hopefully I won't have to make a patch for a patch for a patch ;) [00:06] <bernie_> yes, you did! Thank you! [00:13] <wm4> bernie_: I think it's correct now [00:13] <bernie_> wm4: wooot! [00:13] <wm4> I can't guarantee it's correct though, it's just my wacky code which is riddled with debug-printfs [00:14] <wm4> and which seems to give the same result with ffmpeg-remuxed and mkvmerge-remuxed [00:14] <bernie_> Now alls we got to do is figure out how to change the fate test! [00:14] <bernie_> Well, I'm implementing code that relies on it right now [00:20] <bernie_> it's fate! [00:26] <llogan> durandal_1707: http://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavfilter/vf_separatefield… [00:27] <durandal_1707> llogan: ???? [00:27] <llogan> shouldn't the ;; be ;? [00:28] <durandal_1707> ahh, you stopped my p0rn watching because of ;; ? [00:28] <llogan> carry on. [00:29] Action: llogan is procrastinating [00:42] <bernie_> wm4: if I also wanted to implement the other way of fast seeking (ie: force Cluster start on keyframes, and controlling keyframe placement to where I need them), I'd need to augment ffmpeg. If I implemented this feature (it's quite trivial), would it be allowed into ffmpeg? It's obviously a pretty specialized use-case, and while I could always have a locally patched binary available, it would be better if I knew it was in the pu [00:43] <wm4> bernie_: your line was cut off at ..."it would be better if I knew it was in the pu" [00:43] <wm4> well, it might be accepted if it's an option, I don't know [00:43] <bernie_> that's terrible! [00:43] <bernie_> :) (that my line was cutt off) [00:43] <wm4> you should ask whoever maintains the ffmpeg matroska stuff (I don't know who) [00:44] <bernie_> "it would be better if I knew it was in the public version" [00:44] <bernie_> Hmm. David Conrad (according to the matroskaenc.c file) [00:47] <bernie_> aka yuvi [00:49] <bernie_> got his email. Nice. [00:50] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:74561680cd01: avfilter/vf_separatefields: fix ;; [01:12] <bernie_> wonder if I should email David directly, or post to ffmpeg-devel mailing list [01:21] <b_jonas> argh! I accidentally got a segmentation fault from ffmpeg when playing with parameters. Now I'll have to examine this with the lastest version and report it. [01:21] <b_jonas> Not now though. I'll write it up to myself. [01:21] <b_jonas> ffmpeg -filter_complex "testsrc=s=100x80,split[o1][o2]" -frames 200 -map "[o1]" -y a.mkv -map "[o2]" -y a2.mkv [01:22] <b_jonas> In fact, this already segfaults: ffmpeg -map "[o1]" a.mkv [01:31] <Compn> bernie_ : best to post to list with sample that causes problem hosted on other site, trac or file locker [01:42] <llogan> Compn: testsrc is the sample [01:43] Action: Compn wasnt paying attention [01:43] <Compn> llogan : did you email mike ? [01:44] <llogan> no. i wrote a message and then deleted it. i haven't seen him around in ages. [01:44] <bernie_> Compn: this is an optimization question, not so much something that does not work, etc. [01:46] <bernie_> I'll pose the question to the list and see what happens... ;) [03:03] <wm4> bernie_: btw. I think you could submit your CueRelativePosition patch too [03:17] <bernie_> wm4: yes -- that would be good. But I still must learn about how to teach fate about the change in file [03:18] <bernie_> wm4: change in the produced mkv file. I believe it is unhappy that the output changed. [03:19] <bernie_> wm4: but have not probed too deeply on that just yet. Need to figure it out. [03:29] <cone-261> ffmpeg.git 03Marton Balint 07master:782e06e292c5: ffplay: simplify audio decoding [03:29] <cone-261> ffmpeg.git 03Marton Balint 07master:b383498ea8b8: ffplay: estimate audio frame pts from the previous frame pts values [03:29] <cone-261> ffmpeg.git 03Marton Balint 07master:73b2043d7270: ffplay: use start_time as next audio pts on flush when seeking is not supported [10:48] <cone-577> ffmpeg.git 03Ed Torbett 07master:336353deaad6: rtpproto: Support IGMPv3 source specific multicast inclusion [10:48] <cone-577> ffmpeg.git 03Martin Storsjö 07master:4d97ca040b40: rtpproto: Check the source IP if one single source has been specified [10:48] <cone-577> ffmpeg.git 03Ed Torbett 07master:36fb0d02a1fa: rtsp: Support multicast source filters (RFC 4570) [10:48] <cone-577> ffmpeg.git 03Michael Niedermayer 07master:48353325372a: Merge commit '36fb0d02a1faa11eaee51de01fb4061ad6092af9' [12:12] <cone-166> ffmpeg.git 03Nicolas George 07master:c25d1ba55636: lavu/log: print prefix after \r. [15:11] <JEEB> btw, nevcairiel -- finally got some data off of the Panasonic crapdisc out, and it seems those bastards used some custom encryption scheme in there [15:11] <JEEB> so no simply available "12bit" data [15:11] <JEEB> lul [15:11] <Daemon404> wut [15:12] <Daemon404> how do you even play it back then [15:12] <Daemon404> on a regular bd player [15:12] <Daemon404> or is it BD+ javashit [15:12] <JEEB> the base layer is backwards compatible [15:12] <saste> what's the difference between -x264opts and x264-params? [15:12] <JEEB> only the second layer seems to be bullshit [15:12] <Daemon404> i see... so... how to do you play it at all [15:12] <Daemon404> saste, one is ffmpeg, one is libav iirc [15:12] <Daemon404> is all [15:12] <Daemon404> ive always used x264opts [15:12] <saste> Daemon404, semantically... are they the same? [15:12] <saste> so I can avoid some doc duplication [15:12] <Daemon404> no idea. ive never used x264-params [15:14] <saste> so no they are slightly different [15:14] <saste> stupidity is awesome [15:30] <Daemon404> saste, how so? [15:30] <Daemon404> i know x264opts just takes param=val:param2=val2 [15:30] <Daemon404> with x264's names [15:30] <Daemon404> anything else seems silly [15:31] <saste> Daemon404, actually, x264-params mat be even saner [15:31] <Daemon404> o? [15:31] <saste> x264-params performs escaping [15:31] <saste> x264opts doesn't, but it interprets opt: as opt=1 [15:32] <saste> in case the value is omitted [15:32] <saste> i'll send a patch to document the differences [15:32] <Daemon404> ah [15:32] <Daemon404> what x264 ops even need to be escaped? [15:33] <saste> Daemon404, in case you have a ":" in the value, for example [16:05] <Daemon404> i guess for e.g. deblock=-1:-2 [16:18] <wm4> awesome magic numbers [16:22] <Daemon404> its an example, why do i need to justify a random choice of numbers [16:23] <wm4> if the actual options are not full of magic numbers, that is good [16:30] <JEEB> wm4, or even better [16:30] <JEEB> a magical string of a GUID [16:30] <wm4> fortunately this is not windows programming [16:31] <nevcairiel> GUIDs are usually inside some header so you just use a define name for them. :p [16:41] <JEEB> nevcairiel, yet there are some within code without a name :D [16:41] <JEEB> I'm lovin' it [16:41] <wm4> "let's use random numbers to name things! no clashes anymore!" [16:42] <wm4> genius [16:42] <wm4> but extremely clunky [16:42] <nevcairiel> well you complain about everything, so i stopped listening anyhow [16:43] <wm4> I can't help that most computer technology is pure crap [16:43] <nevcairiel> complaining wont change that [16:43] <nevcairiel> changing that will change that [16:48] <Kovensky> <+nevcairiel> GUIDs are usually inside some header so you just use a define name for them. :p <-- *usually* [16:48] <Kovensky> and it's harder to map it the other wayaround :P [16:48] <Kovensky> +\ [16:49] <nevcairiel> well sure, but we can't complain about everything or we'll just get bitter and grow neckbeards, so instead focus on the important things :p [16:49] <Kovensky> dem neckbeards [16:49] <wm4> well, if I complained about _everything_, I'd spam this IRC channel 24/7 [16:50] <saste> the important question if why IT technology tends to be crap [16:50] <nevcairiel> because developers dont get enough time to do it properly [16:50] <Kovensky> "I'll just make this quick hack, I'll fix it later, I swear" [16:50] <wm4> often they don't care enough and just want get stuff done [16:50] <Kovensky> there's also the problem that most programmers in the world are just in for the paycheck and don't actually care about the quality of their output [16:50] <saste> nevcairiel, how is that different from architecture or mechanical engineering? [16:51] <nevcairiel> who says thats much better? [16:51] <wm4> collapsing bridges D: [16:51] <nevcairiel> many devices and buildings tend to be assembled rather flimsy :p [16:51] <Kovensky> flimsly? [16:52] <nevcairiel> no idea if that word can even be an adverb [16:53] <saste> Kovensky about the paycheck thing, well free software is mostly done by volunteers, and still it is mostly crap (ffmpeg is no difference) [16:54] <nevcairiel> usually too many people and their own ideas of "good" colliding giving you a common denominator that is below good [16:55] <nevcairiel> or like wm4 said, people that are after results instead of style [16:55] <saste> well but at the end saying that "everything is crap" is not useful, since everyone has a different definition of quality and of focus [16:55] <wm4> also, many don't even know the value of simplicity [16:55] <Kovensky> <@saste> Kovensky about the paycheck thing, well free software is mostly done by volunteers, and still it is mostly crap (ffmpeg is no difference) <-- that falls under the <Kovensky> "I'll just make this quick hack, I'll fix it later, I swear" [16:56] <nevcairiel> plenty code monkeys, not enough actual program designers [16:58] <Daemon404> [10:55] <+wm4> also, many don't even know the value of simplicity <-- git commit -a -m "fix a bug" [16:58] <Daemon404> cant get a simpler message [16:58] <Daemon404> ;) [16:59] <Kovensky> Daemon404: git commit -a -m '.*' [16:59] <Kovensky> Daemon404: the reader can fill in his own commit message; it'll match the regexp ;) [17:00] <Daemon404> ffmpeg has a one character commit message somewhere [17:00] <Kovensky> (I guess that's less simple in some ways, but simpler in others) [17:00] <wm4> wat [17:00] <wm4> must be from ffmpeg's dark cvs and svn past [17:01] <Daemon404> svn [17:01] <Daemon404> the commit message was a single semicolon [17:06] <saste> everyone has a different focus on "quality" [17:06] <saste> some people are obsessed by commit messages or code reindent [17:07] <saste> other focus on performance, on security, on usability, consistency etc etc [17:08] <saste> so the code will be always considered as "crap" in a way or another [17:09] <BBB> I think that era used "10l" a lot as commit message [17:14] <wm4> it has something to do with cola I heard? [17:16] <Daemon404> yes [17:16] <Daemon404> google yields intresting uses [17:16] <Daemon404> "DonDiego, you are honored by 10l cola per day for deliberately [17:16] <Daemon404> breaking MPlayer CVS. [17:16] <wm4> ah, mplayer cvs [17:16] <wm4> the ring of it [17:19] <JEEB> :D [17:19] <BBB> my preciousss [17:37] <bernie_> Hmmm kvminfo reports # of frames on a simple block, but I fail to see how the spec for SimpleBlock contains a frame count info [17:37] <bernie_> as in: | + SimpleBlock (key, track number 1, 1 frame(s), timecode 0.064s = 00:00:00.064) at 1423 size 19435 [17:38] <bernie_> oh never mind [17:38] <nevcairiel> simpleblock is always one frame, isnt it [17:38] <wm4> matroska blocks can be "laced", which means they contain multiple packets [17:38] <bernie_> If Lace is set [17:38] <bernie_> yep [17:38] <bernie_> :) [17:38] <wm4> but these are not organized as ebml elements [17:39] <nevcairiel> i dont think simpleblock allows lacing [17:39] <bernie_> it does allow it from the spec, I think [17:39] <bernie_> in the flags section [17:39] <bernie_> I was just not expecting to see reference to frame count... but all is well. [17:40] <wm4> yeah seems allowed [19:09] <saste> ffplay -i test.ogg -af "atrim=duration=2" -autoexit [19:09] <saste> doesn't exit and spam the log [19:09] <saste> is this known? [20:11] <saste> how do you convert stereo to mono? [20:11] <Daemon404> -ac 1 ? [20:12] <saste> I'm doing ffmpeg -i test.ogg -ac 1 test1.ogg [20:12] <saste> but then test1.ogg just contains random noise... [20:13] <Daemon404> ... does that use our WMD of a vorbis encoder? [20:14] <Daemon404> something's funky if you cant even dump pcm [20:20] <saste> uh cool... it's averaging the two channels, which are almost opposite [22:21] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:bbf6cb754cc8: avfilter/vf_scale: Add in/out yuv color matrix option [22:21] <cone-196> ffmpeg.git 03Michael Niedermayer 07master:835eee88ec4c: avfilter/vf_scale: add in/out color range option [00:00] --- Sun Jul 21 2013

1 0

ffmpeg.log.20130720
by burek 21 Jul '13

21 Jul '13

[00:02] <eth2wlan> I'll try to re-install tomorrow [00:03] <eth2wlan> Should i remove the current (wrong version of) ffmpeg with "apt-get --purge remove"? [00:06] <eth2wlan> I'm sleepy [00:06] <eth2wlan> (here i't midnight ) [00:07] <eth2wlan> Good night [00:07] <eth2wlan> Thx [00:11] <axorb> hey guys, I'm wondering if it's possible to copy codecs but fill in anything that is missing [00:12] <axorb> I want to start a video 10 seconds in but the first keyframe after that is at the 12 second mark [00:12] <axorb> so would it be possible to encode those first 2 seconds but reuse the rest? [00:27] <teratorn> axorb: it does not sound *impossible* [00:29] <axorb> I'm doing HLS so I'm experimenting with a few things [00:29] <axorb> overlapping the PTS doesn't seem to work unfortunately [00:32] <dx926> Hi! Can I ask a question about building compact version of ffmpeg on linux? [00:33] <llogan> dx926: just ask. no need to ask to ask. [00:33] <dx926> I have disabled everything by passing that option to configure script and enabled support for mp4, mov, flv. now I want to enable avi file support so that I can convert mp4, mov, flv file to avi file. Sorry if it's a stupid question. What option do I pass to configure script to enable avi? cause I can't convert to avi from one of the mentioned foromat. [00:36] <llogan> show your configure options [00:38] <dx926> "llogan" here it is. http://pastebin.com/yQV2kb41 [00:40] <llogan> why --enable-memalign-hack? [00:41] <saste> llogan, ^^ cargo cult [00:43] <llogan> in that case, where is --enable-small? [00:43] <dx926> Actually I did that for android and it's copied from a script found online. Added some of my own options. Sorry if I offended anyone by my stupidity. Before I deploy the program for android I wanted to test how many file format I can convert. I've no idea what that option does. [00:44] <llogan> no offense. i was just curious. [00:44] <llogan> nobody knows what it does since it's for windows, IIRC. [00:45] <dx926> is there any option I used there that's causing the problem? I thought android was armv6 or v7. [00:46] <llogan> i'm not sure what the problem is because you haven't shown any console outputs that show the problem. [00:46] <llogan> and --disable-everything --enable-foo is always tricky [00:47] <llogan> (not that i've really tried it myself) [00:50] <dx926> LOL here's execution of ffmpeg and error message. http://pastebin.com/z7dm2ZKX [00:53] <dx926> llogan Do you want any other output? [00:56] <llogan> i guess you forgot --enable-muxer=avi [00:58] <llogan> does all of this extra configure work make the result significantly smaller? [00:58] <dx926> No it [00:58] <dx926> it's 16MB [00:59] <dx926> I mean static binary ffmpeg [00:59] <dx926> that's why I'm trying to make it compact. [01:00] <llogan> heh. the static builds from burek are ~18 MB (but this particular build is old) [01:01] <dx926> but if i enable only mp4 and x264 thats 6MB. [01:04] <dx926> Thanks for your knowledgable help. I appreciate it and your community. Again thanks. Bye! [01:33] <axorb> hey guys, can you lie about the duration of a video file? [01:33] <axorb> like the duration reports 10 seconds but the actual content is 15 seconds? [01:37] <llogan> possibly a bug [01:43] <axorb> I want to cause it :P [01:43] <axorb> maybe I should explain more [01:44] <axorb> I'm generating a HLS video which means 10 seconds chunks, only I'm doing it in a time-sensitive environment [01:44] <axorb> so I'm faking the m3u8 and hard-coding a duration of exactly 10 seconds, instead of having the segmenter automatically split on keyframes [01:45] <axorb> this works for reencoding because I can seek and generate an exact amount of video [01:45] <axorb> but I'm trying to optimize things and pass-through the streams if the input settings match the output settings [01:46] <axorb> but since the original video does not have keyframes every 10 seconds exactly, each part has some unwatchable video at the start [01:46] <axorb> I either need to a) re-encode the part up until the first keyframe, then I can copy [01:46] <llogan> you lost me at HLS. i have no experience with that. [01:47] <axorb> it's pretty simple, .ts files of 10 second length [01:47] <llogan> http://ffmpeg.org/ffmpeg-formats.html#hls-1 [01:48] <llogan> -hls_time seconds [01:48] <axorb> the problem is that time is not exact, it uses the closest keyframe [01:48] <axorb> the usual work-around is to force a keyframe [01:48] <axorb> but I can't do that if I'm copying streams [01:50] <axorb> I can't download the entire file and scan for keyframes unfortunately [01:51] <axorb> because the file is stored on S3 and we want the user to be able to watch the first part without waiting for it all to be downloaded/scanned [01:52] <axorb> anyway, at this point I think the only solution is some way of encoding up until the first keyframe, then switching to copy stream mode [01:53] <axorb> teratorn: said it might be possible, but I imagine I'd have to add the feature myself [01:55] <axorb> alternatively, some quick way of getting the closest keyframes to 10 seconds without downloading the entire file [01:55] <axorb> (ie lots of seeking) [01:56] <axorb> something like [01:56] <axorb> ffprobe -ss 30 -i file.avi [01:56] <axorb> and just outputing the current time [08:58] <fling> When will vp9 encoder be supported? [09:14] <SirCmpwn> I'm experiencing unusually long encoding times for theora videos on my server [09:15] <SirCmpwn> ffmpeg there is minimal, only configured with what it needs [09:15] <SirCmpwn> is there something I might have missed that could speed up encoding? [09:15] <SirCmpwn> configuration: --enable-gpl --enable-libx264 --enable-libtheora [09:34] <R0SSI> which part of this code is wrong? [09:34] <R0SSI> ffmpeg -i "L01 - Realms of Physics".avi -map 0.1 -map 1.0 -vf nullsink\;color -acodec copy out.mkv [09:34] <R0SSI> I get this error: http://paste.ubuntu.com/5893278/ [09:36] <durandal_1707> R0SSI: you are not using ffmpeg [09:36] <R0SSI> alo? [09:37] <durandal_1707> that is "deprecated ffmpeg" from Libav, FFmpeg fork [09:41] <R0SSI> durandal_1707: So, how can I install ffmpeg; when I run `sudo apt-get install ffmpeg`; the package manager says: "ffmpeg is already the newest version." [09:44] <R0SSI> I want to decrease the video quality as much as possible, but I want to keep the original audio quality! [09:47] <durandal_1707> you want to replace video with black frame? [09:47] <durandal_1707> you could do this with several steps, if single one does not work [09:48] <durandal_1707> if you use reall ffmpeg, read: [09:48] <R0SSI> the black is the best! [09:49] <R0SSI> How can I use the reall ffmpeg! it seems mine is not the real. My operating system is Ubuntu 13.04 [09:50] <durandal_1707> i told you already, where you can download the realy thing [10:53] <fling> Mplayer2 is not playing encoded file :\ [10:53] <fling> http://dpaste.com/1311391/ [10:53] <fling> also tried mkv [11:10] <elkng> when I use option '-c:v png "%3d.png"' to convert video to many *.png images is there a way to set quality ? or png format has only one quality not like jpeg ? [11:11] <zap0> png should default to lossless quality [11:14] <durandal_1707> you can set compression [11:14] <durandal_1707> looks like its not documented at all [11:15] <durandal_1707> -compression_level 0-9 [11:58] <elkng> there is an option to set limit on time "-ss 00:10:30 -t 00:00:40" so for "-ss" you set "start" position but for "-t" you don't set "end" position but need to set amount of time you need from that video cuted, and its frustrating because evertime you should calculate that time, isn't it possible to set "end" position ? [12:00] <fling> elkng: no [12:01] <elkng> its really frustrating, isn't the logic of that case should be like: you have "start" position and "end" position of the piece of video you want to cut out ? [12:02] <elkng> but you can set "start" position with "-ss" but then you should calculate how much minutes,seconds its untill end position and use it for second option "-t" [12:02] <fling> elkng: no, I need to encode exactly 60s starting from 3s, so I use -t 60 -ss 3 [12:04] <fling> elkng: is not 00:10:30 equal to 630? [12:06] <elkng> yes, so what ? [12:09] <fling> no need to count minutes& [12:09] <fling> time ffmpeg -f concat -i svetik.txt -c:a libopus -c:v libvpx-vp9 -strict -2 -t 60 svetik.webm [12:09] <fling> took real 72m51.697s [12:10] <elkng> so should I make some script like: "a=$(date +%s 00:10:30) b=$(date +%s 00:00:40); ((c=b-a)); d=$(date -d @${c})" and then: "ffmpeg -ss 00:10:30 -t ${d}" ? [12:10] <fling> 0 fps is too slow& [12:10] <fling> elkng: you may script anything, yes. [12:11] <elkng> "may script anything" except girlfriends [12:11] <fling> right :D [12:13] <joeka> hello [12:15] <elkng> its strange I copied piece from video first using: "ffmpeg -ss 00:02:45 -t 00:06:54 -i file_in.avi -vcodec copy -acodec copy file_out.avi" then using command: "mencoder -ss 00:02:45 -endpos 00:06:54 -oac copy -ovc copy file_in_2.avi -o file_out_2.avi" and got two files 00:06:54 in length but they sizes are differ, first from "ffmpeg" is 82980KB s [12:16] <elkng> econd one from "mencoder" is 76126KB, about 6.5MB difference, why ? they should just raw copy from one file to another file ? [12:16] <fling> why is vp9 so slow? > http://dpaste.com/1311433/ [12:17] <fling> elkng: different muxing overhead? [12:17] <fling> elkng: try extracting streams and compare them [12:18] <fling> also use `-c copy' instead of `-vcodec copy -acodec copyŽ [12:18] <joeka> have some of you tried streaming to twitch? With my settings I only produce a black image while audio seems to work fine. If I output to a file with the same settings, it works. settings: http://pastebin.com/kPSsJbyQ [12:21] <fling> joeka: this works for me > http://bpaste.net/show/115833/ [12:21] <fling> joeka: also try piping with glc& [12:21] <fling> (if you are capturing opengl) [12:24] <joeka> only tune and preset different for video? [12:24] <fling> umm? [12:25] <joeka> oh wait, no [12:26] <joeka> you have one ffmpeg call nested in there [12:26] <joeka> why? [12:27] <fling> joeka: because it is not possible to double a single encoded stream (or not documented) [12:27] <joeka> ok [12:28] <fling> joeka: so if I want to write a file and rtmp with a single ffmpeg, I need to encode twice. [12:28] <fling> so I use one ffmpeg for encoding, and another ffmpeg for copying stream in a file and rtmp [12:34] <fling> joeka: just found this > http://ffmpeg.org/trac/ffmpeg/wiki/StreamingGuide [12:35] <joeka> fling, thanks [12:36] <joeka> I read that, but I still have no idea what could be wrong with mine [12:36] <joeka> maybe twitch just doesn't like me [12:37] <fling> joeka: use mine? [12:38] <joeka> noooooo I have to figure it out :D [13:16] <joeka> fling, I think I give up, even using all your video related options does not help. I can create an flv video without problems, only streaming to twitch does not work. [13:16] <fling> joeka: do you have a proper key? [13:16] <fling> joeka: have you tried my script? :D [13:17] <joeka> the key is not the problem, the audio stream works, only image is missing [13:18] <joeka> fling, ok ok I try yours [13:21] <joeka> fling it doesn't stream at all? [13:27] <joeka> fling I got something that works [13:29] <fling> give me a link :P [13:29] <joeka> oh I even know which specific option did the trick [13:29] <joeka> -pix_fmt yuv420p [13:32] <JEEB> yes, flash can only handle 4:2:0 and 4:2:2 YCbCr [13:33] <JEEB> and the streaming service can kick your ass for 4:2:2, possibly [13:33] <JEEB> the only thing that has hardware acceleration is 4:2:0 YCbCr [13:40] <joeka> if only the audio would be in sync with the video now :D [14:45] <zap0> just move closer/further away until it is. [14:45] <zap0> light/sound travel at different speeds. [17:16] <elkng> is it possible to convert *.gif file to a set of frames in one png or jpg file, like sprites for games they are all in one file with some grid and some background color that would be transparent ? [18:25] <iio7> I have a 3D mkv file side-by-side. Is it possible to use ffmpeg to convert it into 2D, and if yes, how do I do that? [18:25] <iio7> It's "Side by Side (left eye first)". [18:30] <mark4o> iio7: you can easily get left-only or right-only using crop or stereo3d filters [18:30] <mark4o> I don't think ffmpeg has a filter that can average them if that's what you are looking for [18:30] <iio7> How do I get left only? [18:31] <mark4o> just crop to left side, or use -vf stereo3d=sbsl:ml [18:31] <mark4o> http://ffmpeg.org/ffmpeg-filters.html#stereo3d [18:35] <iio7> I get "No such filter: 'stereo3d'" [18:36] <mark4o> are you using an old version? should be non problem, use crop filter instead [18:36] <mark4o> http://ffmpeg.org/ffmpeg-filters.html#crop [18:38] <mark4o> ffmpeg -i input -vf crop=in_w/2:in_h:0:0 output [18:41] <iio7> I'm going to compile the latest, do I need any kind of enable-filter-X? [18:42] <mark4o> stereo3d is enabled by default [18:42] <iio7> thanks! [18:43] <norbert_> hi, I'm trying to screen record a Flash game with ffmpeg; the output is a green rectangle; any ideas/suggestions? [18:46] <norbert_> if I try to make a screenshot of the game with GIMP then the result is a transparent area without content [18:46] <norbert_> it's probably related to how the video driver puts stuff on the screen, but I'm hoping I can keep the software settings as they are and still use ffmpeg to screen record [19:02] <norbert_> nobody have any suggestions? [19:02] <norbert_> *has [21:28] <Zerxer> I converted an AVI file that was created with Fraps, codec ID of FPS1, to an mp4 file. Plays fine on Windows and OS X. I also converted an AVI file that was created using a capture device, so it was a completely raw/uncompressed video, codec ID of UYVY. I did the same conversion (to an mp4 file) and it plays fine on Windows, but OS X has black video (audio is fine). I checked the media info of both files and each are nearly sim [21:28] <Zerxer> (using AVC format) with only minor differences. The first mp4 file has a profile of "High(a)L3.1" while the second has "High 4:2:2@L3.1". The Chroma subsampling is also different on each, 4:2:0 on the first and 4:2:2 on the second. Only other difference is the resolution and the bitrate. [21:29] <Zerxer> I didn't set any advanced settings when using ffmpeg at all, just setting output framerate to 30 [21:30] <Zerxer> I'm new to using ffmpeg so I'm hoping there's some command line arguments I can use to make the second video's conversion completely similar to the first, or something else that will make the video compatible with OS X playback. It's worth noting that I did install something on OS X for allowing QuickTime to play more codecs than what it can by default. [21:48] <Zerxer> Okay, so as it appears is true, OS X cannot play 4:2:2 YUV MPEGs. Any way to have ffmpeg force it to 4:2:0 ? [21:48] <klaxa> see ffmpeg -pix_fmts [21:51] <Zerxer> Thanks, would you know which one I should be using? sounds like yuv420p is what I'd want, just based off the name [21:51] <mark4o> -pix_fmt yuv420p [21:52] <Zerxer> then again, it says yuv420p would be 12 bits per pixel, while the media info on both of my files (the one that works and one that doesn't) has a bit depth of 8 [21:53] <Zerxer> unless that doesn't matter [21:54] <mark4o> yes it is 8 bits, but there is only 1 u and 1 v for every 4 pixels [21:54] <Zerxer> thanks, I'm going to try this right now. it's a 2 hour video so it'll take a bit to know if all is well [21:55] <klaxa> just encode the first few seconds and check [21:55] <klaxa> add -t 30 [21:55] <klaxa> that will encode the first 30 seconds [21:55] <klaxa> you don't have to encode the entire file to check if it would work [21:56] <Zerxer> ah great, will do. I figured a flag like that would exist and secretly hoped you'd mention it [21:59] <Zerxer> worked great, thanks! no problems on OS X now [21:59] <Zerxer> I'm knocking a 637GB file down to below 4GB with this, haha [21:59] <klaxa> haha wow [22:00] <Zerxer> using a blackmagic intensity shuttle to capture hdmi input from a game console, and it produces huge files at 720p [22:01] <klaxa> hardware encoder probably? [22:02] <Zerxer> not sure, possible since it is an external device connected via usb 3 [22:03] <Zerxer> sadly the video is out of sync, not sure if that's because of the hard drive speeds I'm writing to, or just the device/software isn't that great, but I figure I can fix that later by detaching the audio in my movie editor [22:04] <Zerxer> girlfriend is starting to do let's plays on youtube and I'm the lucky tech guy [22:05] <klaxa> heh [22:05] <klaxa> well you're the lucky guy for having a girlfriend [22:05] <Zerxer> no doubt [22:54] <braincracker> hey guys, there's me riding my new trike http://esteedv.files.wordpress.com/2011/11/death_rides_a_pale_trike_by_mjra… so what's up ? [00:00] --- Sun Jul 21 2013

1 0

ffmpeg-devel.log.20130719
by burek 20 Jul '13

20 Jul '13

[00:19] <bernie_> anyone have experience with fate? To figure out if a test is OK does it just do a check-sum of the output text? [00:21] <bernie_> because when I pull from git, and run one of the tests by hand, it shows an assertion failure, but somehow the test is accepted. [00:29] <BBB-work> make fate-list [00:29] <BBB-work> grep for your test of interest [00:29] <BBB-work> make fate-h264 groups all 264 tests [00:29] <BBB-work> etc. [00:50] <bernie_> cool -- thanks BB-work [00:52] <bernie_> could it be possible that a test that is marked as accepted has an assertion failure in it's output? I have not looked that closely at it, but does fate compare a hash of what is "accepted" vs. what is run, and if the "accepted" test has an assertion failure, then that fate test is still OK? (hope my question makes sense) [01:01] <bernie_> never mind -- looks like I messed something up real good with just a few lines of code. yay! [01:04] <BBB-work> bernie_, sorry, was off for a sec [01:05] <BBB-work> no, if a tests asserts, it will not be marked as passed [01:05] <BBB-work> the test needs to complete 100% succesfully and return a known-good hash before it's marked as completed and succesful [01:05] <BBB-work> but it looks like you figured that out already :) [01:17] <bernie_> BBB-work -- thanks. Yeah, investigating... [01:39] <bernie_> oh shit found my bug [01:39] <bernie_> I don't usually write code like this: [01:39] <bernie_> entries[cues->num_entries ].pts = ts; [01:39] <bernie_> entries[cues->num_entries ].tracknum = stream + 1; [01:39] <bernie_> entries[cues->num_entries++].cluster_pos = cluster_pos - cues->segment_offset; [01:40] <bernie_> because then I just went and added... [01:40] <bernie_> entries[cues->num_entries++].relative_pos = relative_pos - cluster_pos; [01:40] <bernie_> which is obviously stupid of me... [01:40] <bernie_> but I didn't look for the ++ right there, because I would never do something like that ;) [01:41] <bernie_> Not to start any wars or anything, but I personally like keeping statements with side-effects like that on a line by itself. [01:46] <bernie_> okay no more assertion failures, but the test is still failing. Should I be surprised that it's failing if I ended up adding a new EBML tag to the output of the video file? I guess not. But do I tell if it's failing because it's a different output file? [01:47] <bernie_> meant to say: How do I tell if it's failing because it's a different video output file? [01:50] <wm4> bernie_: I looked at it myself [01:51] <bernie_> wm4: cool [01:51] <wm4> bernie_: I wonder how this seeking is supposed to work, because it points directly to the block, instead of the block group [01:51] <wm4> but the block group is the containing element, and contains important information [01:51] <bernie_> Hrmmm [01:51] <wm4> so I guess it's somehow guaranteed that the Block is the first element of the BlockGroup? [01:54] <bernie_> I see your point. [01:54] <bernie_> Would also be nice to know the size of the block group... [01:56] <wm4> either way, if I understand this right, the Matroska devs fucked up yet again [01:56] <bernie_> So you'd need like a reference to the blockGroup, and the the offset into that blockGroup to seek to the block? [01:57] <wm4> a BlockGroup contains only one Block [01:57] <wm4> and some other elements to e.g. set the duration of the block [01:57] <bernie_> oh, gotcha [01:57] <wm4> so there'd be no downside to point to the BlockGroup instead (but mkvmerge doesn't do that) [01:59] <bernie_> I see, so your point being, if the Block is the first element, you can just continue to read the other tags, and pick those up, but it's unclean, because you are depending on the order of the EBML tags, which is kind of an implementation question. Not specified per spec. [01:59] <wm4> you also need to know when the BlockGroup ends [01:59] <wm4> which is hard [02:00] <bernie_> yep, which would be the reason to actually point to the blockgroup tag, which would have that info [02:00] <wm4> yes but mkvmerge doesn't do that [02:00] <bernie_> Poop [02:00] <wm4> indeed [02:01] <bernie_> do you ever talk to those guys? [02:01] <wm4> no [02:04] <bernie_> I wonder who in the world actually uses this currently? My guess is no one. So they should just fix their spec, and then fix mkvmerge and be done with. "what could possibly go wrong" ;) [02:04] <wm4> but there are broken files out there [02:04] <wm4> oh well, they still could introduce a separate, correct element [02:05] <bernie_> yeah [02:05] <bernie_> I'll see if they'll get back to me on that [02:06] <bernie_> Hmm they have an IRC #matroska [02:07] <wm4> joined there [02:09] <bernie_> you beat me to it :) [02:10] <bernie_> wm4: not too active in there... :/ [02:10] <cone-561> ffmpeg.git 03Ramiro Polla 07master:454c89dde3af: img2enc: add option to use strftime() for filename [02:10] <wm4> well, if we get any reply, it'll probably take half a day or something [02:11] <bernie_> groan [02:12] <bernie_> okay, so maybe for now, I'll point it to blockGroup for myself, but obviously won't put up a patch to ffmpeg for it, but I do need to keep going on my implementation [02:12] <bernie_> In the future, I'd hope they add a new EBML tag for this purpose [02:12] <wm4> if I were you I'd test with a mkvmerge produced file first [02:25] <bernie_> You know, I bet the only real other important information besides the block itself is the block duration, and the other elements do not get used typically -- of course, that's kind of a lame explanation, but given a cuepoint is an I-frame... maybe you only really need the block + the duration [02:27] <bernie_> and the block duration could be indicated by CueDuration [02:28] <wm4> that's even worse [02:28] <bernie_> hahaaa [02:28] <wm4> that'd you somehow need to get the duration value from the index [02:29] <bernie_> via CueDuration [02:29] <wm4> it's bad because it makes the code more complicated [02:30] <bernie_> I agree [03:26] <bernie_> wm4: cool, so all I have to do to force a new matroska cluster at the start of a key frame is to modify some logic in mkv_write_packet(), specifically right at the first logic block that has the comment about when it currently starts a new cluster. [03:26] <bernie_> seems quite easy [10:22] <bernie_> [10:53] <cone-261> ffmpeg.git 03Martin Storsjö 07master:d010e95f8608: avserver: Remove unused loadable module support [10:53] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:8740762c43dc: Merge commit 'd010e95f86089abe9a3d4d4a66ac8102312d28a4' [11:03] <cone-261> ffmpeg.git 03Diego Biurrun 07master:439902e0d68a: Employ consistent LIBAV_COMPAT_ multiple inclusion guards in compat/ [11:03] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:c88503e3f6bd: Merge commit '439902e0d68a0f0d800c21b5e6b598d5fa0c51da' [11:36] <cone-261> ffmpeg.git 03Diego Biurrun 07master:910042072a88: compat: Add missing license boilerplates [11:36] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:25ffaf527702: Merge remote-tracking branch 'qatar/master' [12:18] <cone-261> ffmpeg.git 03wm4 07master:06b269dacbe1: examples: demuxing: do partial audio packet decoding [12:18] <cone-261> ffmpeg.git 03wm4 07master:9f31c1608c74: examples: demuxing: simplify audio output [12:18] <cone-261> ffmpeg.git 03wm4 07master:72f5a6d0672a: examples: demuxing: print ffplay command even if sample format is planar [12:51] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:a4b55bbb6f8b: swscale/input: fix 16bit gbrp input [15:00] <cone-261> ffmpeg.git 03Michael Niedermayer 07release/1.2:d83ab33715cb: swscale/input: fix 16bit gbrp input [15:00] <cone-261> ffmpeg.git 03Michael Niedermayer 07release/2.0:8f9bc6f2ce49: swscale/input: fix 16bit gbrp input [15:07] <JEEB> hmm [15:07] <JEEB> I might have found the first semi-valid use for MVC [15:08] <JEEB> hacking high bit depth video onto blu-ray [15:08] <JEEB> I thought there was a guy here who was working on MVC decoding? :D [15:08] <JEEB> might have to read the logs back a few months [15:14] <saste> just got SOCIS mail [15:15] <saste> we were not accepted [15:16] <nevcairiel> JEEB: the only valid use is multi-view coding, using it as a hack is by definition completely invalid [15:17] <nevcairiel> also, since the second view is always a predicted frame against the first view, such a hack would also be very inefficient :p [15:17] <JEEB> I guess that's why they only went for 12bit and not for 16bit : [15:17] <JEEB> :D [15:17] <JEEB> also in such a way I completely agree, I wonder how much herp and derp panasonic got from Sony [15:17] <JEEB> for this hack [15:18] <nevcairiel> but i also heard of this hack before [15:18] <JEEB> yes [15:18] <JEEB> I brought it up on another channel earlier [15:19] <JEEB> we got the first blu-ray using that just a couple of days ago [15:19] <nevcairiel> if earlier is weeks ago, then sure [15:19] <JEEB> well, I remember linking it back in april when panasonic first announced it, too :D [15:19] <JEEB> http://dl.dropboxusercontent.com/u/17138887/ADC/Porco%20Rosso/Info.png "master grade video coding" [15:20] <nevcairiel> that they are still allowed to carry the Blu-ray Disc logo is mind boggling [15:21] <JEEB> well, the higher 8 bits are on the main view so it's backwards compatible in a way :D [15:21] <JEEB> until some player tries to play that as '3-D' :D [15:21] <saste> i get this email from an indian guy who calls me "mam", guess my name sounds like a female name in his (her?) language [15:21] <j-b> JEEB: WTF [15:21] <j-b> JEEB: you are joking, right? [15:21] <nevcairiel> even if backwards compatible, you usually cant bastardize a standard format like this [15:21] <JEEB> j-b, I'm not, although it will be some hours before my copy is here [15:21] <JEEB> so I can have a full look [15:22] <JEEB> but the article's out and that's a real thing [15:22] <JEEB> lol [15:22] <JEEB> nevcairiel, true [15:22] <JEEB> which is why [15:22] <JEEB> > I wonder how much herp and derp panasonic got from Sony [15:23] <nevcairiel> next up, half-4K discs with two 1080p frames coded as MVC and then stretchted vertically :p [15:23] <JEEB> :DDD [15:24] <j-b> nevcairiel: oh god... [15:24] <j-b> or just 4k interlaced with two 1080p frames? [15:24] <nevcairiel> not enough frame rate allowed for that to be viable [15:31] Action: Daemon404 just luls [15:39] <Compn> and we thought video stuff would get more normal [15:39] <Compn> we thought maybe interlaced was going away [15:39] <j-b> Compn: do you honestly think it will get more normal? [15:40] <Compn> i meant back in 2004 [15:40] <Compn> whenever h264 came out [15:41] <Compn> kinda thought we were done with format wars [15:41] <wm4> "more sins are committed in the name of optimization than something something" [15:41] <wm4> whatever was the correct quote [15:41] <Daemon404> grep dsputil [15:41] <Daemon404> im sure its there [15:42] <wm4> I guess cargo culting and micro optimizations are really big in multimedia, just like in game dev [15:42] <Daemon404> id wager its worse in game dev... from what ive seen anyway [15:42] <Daemon404> optimizing like it's 1994 [16:18] <gnafu> So what do you think it'd look like if you watched one of these 12-bit MVC videos decoded as 3D video? What would your right and left eyes see? [16:30] <nevcairiel> one eye sees an 8-bit image, the other garbage [16:35] <gnafu> nevcairiel: I'd love to try it :-). [16:35] <gnafu> Could be some real psychedelic stuff. [16:36] <nevcairiel> probably similar to this avisynth stacked high-bit hack [17:28] <Compn> i dont get why people are creating encodes that wont play with hwaccel decoders [17:28] <Compn> i guess reencoding for mpeg2 or mpeg4 without qpel never goes out of style [17:29] <nevcairiel> because they want to preserve more information than the default profile supports? :D [17:41] <wm4> saste: will the HAVE_INCOMPATIBLE_LIBAV_ABI issue be fixed at some point? [17:54] <cone-261> ffmpeg.git 03Piotr Bandurski 07master:5f39992ed925: bmp: fix decoding of flipped rle4 [18:42] <saste> wm4, what's the issue with have-incompatible-libav-abi? [18:43] <saste> if they will ever be completely compatible? only $deity knows [18:43] <wm4> include/libavfilter/avfilter.h:1314:5: warning: "HAVE_INCOMPATIBLE_LIBAV_ABI" is not defined [-Wundef] [18:44] <saste> uh? [18:44] <saste> wm4, how did you get it? [18:44] <wm4> by including avfilter.h [18:45] <saste> weird [18:45] <Compn> >warning [18:45] <Compn> >not error [18:46] <saste> wm4, what if you put AV_HAVE... [18:46] <saste> that should fix it, since it will use the public symbols [18:47] <saste> *symbol [18:49] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:db33010483a0: jpeg2000dec: silence unused variable warning [18:51] <wm4> saste: I think that works [18:54] <saste> wm4, please send a patch, I'm uncomfortable to commit patches which I didn't test [19:06] <wm4> saste: sent [19:07] <wm4> feel free to adjust the commit message [19:15] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:6d323ff57af2: swscale: remove unneeded include assert.h [19:15] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:8589d7a61c56: avfilter/vf_yadif:remove unneeded include assert.h [19:15] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:ca857c86bdd3: tests/tiny_psnr: remove unneeded include assert.h [19:15] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:56123fc721b1: avutil/softfloat: remove unneeded include assert.h [20:02] Action: llogan is looking forward to the resulting Kamedo2 tests [20:03] Action: nevcairiel is looking forward to the actual patches that give use these fixes [20:43] <bernie_> wm4: thanks for your help regarding the CueRelativePosition [20:49] <wm4> bernie_: so now that it turned out this was a mkvmerge bug, and CueRelativePosition is a bit more sane than we thought, which approach will you implement? [21:05] <bernie_> wm4: It's a good question. I really like the simplicity of starting a new cluster, and the code change is trivial in ffmpeg. I'd like to add this as an optional feature to ffmpeg, but would need guidance on if that is a reasonable thing to add, given the use-case is quite constrained to the specific project I'm working on. I like the idea of adding it, because it's an easy change [21:09] <bernie_> wm4: I do have a patch that I think implements CueRelativePosition correctly, but it blows up fate test -- I'm guessing fate looks at the output video file and checks if it's been unchanged from what is currently accepted? [21:10] <wm4> uh, I don't know [21:10] <nevcairiel> yeah it does a crc check on the file, so if you add new elements it would need to be updated, after validating with mkv validation tools that the file is valid, of course [21:10] <wm4> I guess it might be possible that there's a seek check [21:10] <wm4> so if the seek behavior changes, the test could blow up? [21:10] <wm4> much simpler [21:10] <nevcairiel> there might be seek tests too, i dont know about that [21:11] <bernie_> Well currently I have both things turned on: forcing a new cluster on a keyframe and generating CueRelativePosition. [21:11] <bernie_> So for the first path, I guess I should turn forcing a new cluser on a keyframe off. [21:12] <bernie_> ^path^patch [21:12] <wm4> bernie_: how much overhead do clustered keyframes cause? [21:12] <wm4> when remuxing [21:14] <bernie_> well -- for me it's not worth to worry about, because I basically constrain my keyframes to happen only at the percise points in the video file that I need fast-seeking to. In my case, that would be maybe 100 key frames for a video that's ~20 minutes long. [21:16] <bernie_> I'm using -force_key_frames and -g with a very high number. [21:16] <bernie_> afk for a bit. My wife wants to share a sandwich with me for lunch [21:37] <cone-261> ffmpeg.git 03Lukasz Marek 07master:a289f418e4fa: doc/filters: add sepia example for colorchannelmixer [21:56] <bernie_> wm4: seek behavior should not have changed -- code that doesn't make use of CueRelativePosition can safely ignore it, and i didn't make any modification to demux. Just adding an EBML element. Which mkv validation tools should I be using to verify, and how would I go about adding / changing the fate test expected output data? [21:57] <wm4> these are good questions, and I don't know how to answer any of them [21:58] <bernie_> d'oh! [21:58] <wm4> but it'll probably be hard to verify it's correct, because the only thing that is writing these elements (mkvmerge) did it incorrectly so far [21:58] <bernie_> yes [21:59] <bernie_> well, perhaps I can give you the patch, have you code review it, and then you could double check the EBML dump and verify a sample? :) (doesn't that sound like fun?) [21:59] <bernie_> after all, you are the one that caught the bug in mkvmerge [22:00] <wm4> I could try whether my code can read it correctly [22:00] <bernie_> yep [22:00] <bernie_> do you want me to email you the proposed patch real quick? [22:00] <wm4> sure [22:02] <wm4> you can also compile mkvmerge from git [22:02] <wm4> mosu said he fixed it [22:31] <bernie_> so many interruptions today [22:33] <bernie_> wm4: finally got around to emailing you the patch. I'll look at downloading mkvmerge and building that... though I'm totally new to that tool [22:38] <bernie_> aww man, mkvtoolnix is complaining about gcc, and my machine happens to be mac OS X, so it's really XCode underneath... configure: error: support for required C++11 features incomplete [22:38] <bernie_> grrr [22:39] <wm4> hm, try to get it to use clang [22:42] <bernie_> they suggest compiling the latest version of gcc [22:42] <bernie_> ouch [22:43] <bernie_> And with "they" I mean Mosu [22:44] <bernie_> http://forum.doom9.org/archive/index.php/t-155732-p-5.html ... "There's no workaround, I will not remove the code using these features again. See http://marcmutz.wordpress.com/2011/09/20/c98-support-costs-extra/ why. I like being an early adopter of good stuff :)" and "What you can do is compile gcc 4.6.1" [22:44] <wm4> yeah, isn't it a joy? [22:45] <nevcairiel> i think its great that a dev doesnt enslace himself to the stupid legacy systems in this world [22:45] <bernie_> and all I wanted to do was implement a video player. hahaha [22:45] <nevcairiel> enslave* [22:47] <durandal_1707> those are some c++ extensions? [22:47] <nevcairiel> no extensions, just a newer version of the standard [22:48] Action: wm4 is so glad that ffmpeg uses C [22:48] <wm4> also why does mkvmerge want boost::filesystem [22:49] <nevcairiel> its not like C doesnt have the opportunitys to use newer functionality, in C99 or even C11 [22:49] <nevcairiel> but its easier to decide for a project that you basically control alone [22:50] <wm4> now boost regex [22:50] <wm4> which depends on libicu [22:51] <nevcairiel> while i do like c++, i'm not a fan of boost =p [22:52] <wm4> aaand it compiles soo slow [23:07] <wm4> bernie_: I'm not sure yet (and mkvmerge is still compiling), but I think there might be a problem with your patch (or with my code) [23:07] <wm4> bernie_: we said the relative offset starts from the end of the cluster length field, right? [23:07] <wm4> while your patch looks like it takes the start of the cluster [23:08] <wm4> as in, the file position of the cluster ebml id [23:26] <wm4> bernie_: ok my code works with the newest mkvmerge [23:29] <bernie_> wm4: oh I screwed it up? sorry. [23:30] <bernie_> wm4: so you'll submit your patch? [23:31] <bernie_> wm4: I think you are right, I have a bug in my implementation [23:31] <wm4> I didn't write a patch against ffmpeg if you mean that [23:31] <bernie_> wm4: oh, okay -- then I shall go and fix my patch proposal for ffmpeg [23:34] <cone-261> ffmpeg.git 03wm4 07master:122536e01a0a: avfilter: fix preprocessor condition [23:34] <cone-261> ffmpeg.git 03Michael Niedermayer 07master:07704c61dd70: avformat/matroskaenc: Only change chapter ids if needed. [23:41] <durandal_1707> this mxf is nightmare [23:43] <bernie_> wm4: okay, just making sure I understand the structure correctly, ebml_master.pos points to the offset where the first child element goes. Given I'm interested in that point for relative cluster offset, I should be using mkv->cluster.pos, so specifically: (packet_start_pos - mkv->cluster.pos) [23:46] <wm4> cluster pos seems to point to the start of the cluster element itself [23:47] <wm4> but you need the position where the cluster element's header ends [23:47] <wm4> mkv->cluster = start_ebml_master(pb, MATROSKA_ID_CLUSTER, 0); [23:47] <wm4> after his, probably [23:47] <wm4> *this [23:47] <wm4> again I ask me why they did it this way... seems confusing [23:49] <bernie_> yeah, and start_ebml_master -- I think returns ebml_master struct that seems to populate the pos field with the position after writing the EBML tag and the size data. [23:50] <bernie_> so mkv->cluster.pos should point where we want it. [23:50] <bernie_> Just confirming... ;) [23:50] <wm4> I think it doesn't [23:51] <wm4> mkv->cluster_pos points to the start of the cluster element itself [23:51] <wm4> including the MATROSKA_ID_CLUSTER id and the cluster size field [23:51] <wm4> but you want the position after this [23:51] <bernie_> I agree [23:51] <bernie_> that is why I wrote, mkv->cluster.pos [23:52] <bernie_> and not mkv->cluster_pos [23:52] <bernie_> :) [23:53] <wm4> oh [23:53] <wm4> ok then it should be correct [23:54] <wm4> will you send an updated patch? [23:54] <bernie_> nice, okay -- I will do that [23:54] <bernie_> first have to undo my force-new-cluster-for-every-keyframe hack [00:00] --- Sat Jul 20 2013

1 0

ffmpeg.log.20130719
by burek 20 Jul '13

20 Jul '13

[02:45] <FunkyELF> hi guys... I have a bunch of .3g2 files from my mom's old phone. I want to change the container to mp4 or something that her new Android phone can play. I do ffmpeg -i video.3g2 -acodec copy -vcodec copy video.mp4 and what I get no longer plays in mplayer [02:47] <FunkyELF> Can I upload a file somewhere and get some help. The video is only 316K [02:49] <FunkyELF> I have a bunch of .3g2 files that I want to convert to .mp4 or .m4v or whatever so that my mom can play videos from her old phone on her new Android phone. I have tried using ffmpeg with -acodec copy and -vcodec copy but no luck. Any other tools I should try? [02:50] <LunixA380> why -acodec copy should work? [02:50] <LunixA380> What the .3g2 codec are using? [02:50] <LunixA380> mp4 often expect H.264 + MP3/ACC [02:50] <FunkyELF> LunixA380: I don't really care if it is mp4.... you tell me what the appropriate container is [02:51] <LunixA380> There's multiple appropriate container [02:51] <LunixA380> AFAIK, Android phones can read mp4 and WebM [02:51] <LunixA380> but, the unknown thing is what's the codec inside the .3g2, FunkyELF [02:52] <FunkyELF> LunixA380: I wouldn't mind even having her install some 3rd party video player like MX Player. I just need something to play. The videos are so terrible to begin with I don't want to re-encode them [02:52] <FunkyELF> [lavf] stream 0: video (mpeg4), -vid 0 [02:52] <FunkyELF> [lavf] stream 1: audio (qcelp), -aid 0, -alang eng [02:52] <FunkyELF> that is what mplayer says when I play the file [02:53] <LunixA380> hm [02:54] <LunixA380> Wait, why you don't want to reencode them, in fact? [02:55] <FunkyELF> LunixA380: because they're terrible quality from a terrible phone [02:56] <LunixA380> so reencoding them will make the quality worse... [02:56] <FunkyELF> yeah [02:56] <LunixA380> Hm, I see [02:56] <FunkyELF> well.... It looks like I can play the files as .3g2 on my Android phone but I can't hear the audio.... perhaps I need to re-encode audio [02:59] <FunkyELF> LunixA380: can I send you an example file... its only 316K [02:59] <LunixA380> it's a common bug on android forum [02:59] <LunixA380> I saw it reported while doing somme Google search [03:00] <FunkyELF> LunixA380: whats that? [03:00] <LunixA380> and sending me the file wouldn't be helful [03:00] <LunixA380> hm. I rephrase trhat. [03:00] <FunkyELF> LunixA380: I'd like to use ffmpeg to re-encode to something that my Android phone will play [03:00] <LunixA380> It's a common bug I saw reported on android communities* [03:00] <LunixA380> Hm yes. [03:00] <FunkyELF> LunixA380: I can't get this to work.... ffmpeg -y -i 0106021043.3g2 -acodec copy -vcodec copy test.3g2 [03:01] <FunkyELF> [3g2 @ 0x22b0740] Tag mp4a/0x6134706d incompatible with output codec id '86042' [03:01] <FunkyELF> er.... that would be dumb anyway. Copying streams to the same container. [03:01] <FunkyELF> What I wanted to show was the error I get trying to re-encode the audio [03:02] <FunkyELF> ffmpeg -y -i 0106021043.3g2 -acodec libmp3lame -vcodec copy test.mp4 [03:02] <FunkyELF> [mp4 @ 0x1971740] track 1: muxing mp3 at 8000hz is not supported [03:08] <LunixA380> you don't reencode it properly [03:08] <LunixA380> use ffprobe for see actual specs of the movie [03:09] <LunixA380> then you can know what you can/can't do [03:12] <FunkyELF> LunixA380: http://pastebin.com/F1qf9nRr [03:13] <FunkyELF> LunixA380: I was able to to an -acodec libmp3lame -vcodec copy video.avi [03:13] <FunkyELF> that converted the audio to something Android should be able to play but the players in Android only played like 2 seconds of the 20 second clip [03:15] <LunixA380> what's the encoding logs, FunkyELF ? [03:18] <FunkyELF> LunixA380: http://pastebin.com/Hhj8nthz [03:19] <FunkyELF> when I ffprobe test.avi I get this.... Duration: 00:00:00.00, [03:19] <klaxa> can you play it back with ffplay? [03:19] <FunkyELF> klaxa: I can with mplayer and it plays all 20 seconds just fine [03:19] <klaxa> also, you are not using ffmpeg but avconv [03:19] <FunkyELF> klaxa: lemmie try ffplay [03:19] <FunkyELF> klaxa: yeah test.avi plays with ffplay [03:20] <klaxa> wait, you already have a test.avi? [03:20] <klaxa> i thought that failed? [03:20] <FunkyELF> klaxa: I'm converting 0106021043.3g2 into test.avi [03:20] <klaxa> yes, the latest paste of you states that the encoding failed [03:20] <klaxa> >[avi @ 0xdcc740] Application provided invalid, non monotonically increasing dts to muxer in stream 0: 184 >= 184 [03:20] <klaxa> >av_interleaved_write_frame(): Invalid argument [03:21] <FunkyELF> klaxa: yet I have test.avi and it plays all 20 seconds [03:21] <FunkyELF> klaxa: even after deleting test.avi to make sure that I'm not crzy [03:22] <klaxa> the problem was the audio, right? [03:22] <FunkyELF> klaxa: I take that back... the avi didn't play all 20 seconds [03:22] <klaxa> can you hear the audio if you play back the .3g2 with ffplay? [03:23] <FunkyELF> klaxa: yeah [03:23] <klaxa> but if you encode it to mp3 it no longer works? can you test with: ffmpeg -i some_file.3g2 -c:a libmp3lame test.mp3 ? [03:24] <klaxa> (i also would like to encourage you to get ffmpeg instead of using avconv, you can grab a static build over here: http://ffmpeg.gusari.org/static/64bit/) [03:24] <FunkyELF> Failed to set value 'libmp3lame' for option 'c:a' [03:25] <FunkyELF> This program is only provided for compatibility and will be removed in a future release. Please use avconv instead. [03:25] <FunkyELF> am I not using ffmpeg? [03:25] <klaxa> no [03:25] <klaxa> it's badly worded [03:25] <FunkyELF> klaxa: is there a repo or should I just use that link you supplied? [03:26] <klaxa> if you want to, you can compile the latest git http://git.videolan.org/?p=ffmpeg.git [03:26] <klaxa> but using the static binary is fine too [03:26] <FunkyELF> klaxa: that link you supplied... seems to have daily builds until 06-28 then they stopped? [03:27] <klaxa> maybe so, but that's rather recent [03:27] <FunkyELF> downloading [03:28] <FunkyELF> klaxa: okay so now that command succeeded [03:29] <FunkyELF> ffmpeg -i 0106021043.3g2 -c:a libmp3lame test.mp3 [03:29] <klaxa> can you play back the mp3? [03:29] <FunkyELF> yeah [03:29] <FunkyELF> what does -c:a do? [03:29] <klaxa> it's the same as -acodec but a more sophisticated syntax [03:30] <klaxa> this way you can abstract the codec parameter further, if you use "-c copy" it will copy all codecs, earlier you had to do "-acodec copy -vcodec copy -scodec copy" [03:32] <FunkyELF> this command... ffmpeg -y -i 0106021043.3g2 -c:a libmp3lame -c:v copy test.avi ... with the static ffmpeg gives me an avi that plays all the way [03:32] <klaxa> sounds good to me then [03:32] <FunkyELF> klaxa: thanks. What would be a better audio codec to use... aac? Some default sane parameters? [03:33] <klaxa> um... well you are encoding from celt right? [03:33] <FunkyELF> qcelp [03:33] <klaxa> ah no [03:33] <klaxa> >Stream #0.1(eng): Audio: qcelp, 8000 Hz, 1 channels, flt, 14 kb/s [03:34] <FunkyELF> so what would be a good choice of audio codec (since that seems to be the problem playing it on Android). mp3 works but should I try another one? [03:34] <klaxa> yeah, well it's 14 kb/s mp3's lowest bitrate is 32 kb/s afaik, so you will have almost twice as much bandwidth even though it is mp3 i doubt it will be a big drawback on quality, otherwise i can recommend vorbis which has great quality for low bitrates [03:35] <klaxa> i once set up a vorbis stream at 48 kb/s it still sounded decent, 48 kb/s is about as low as you can go with vorbis [03:36] <klaxa> ah not sure avi supports vorbis audio though [03:36] <FunkyELF> This command works with avi but not mp4..... ffmpeg -y -i 0106021043.3g2 -c:a libmp3lame -c:v copy test.mp4 [03:36] <FunkyELF> what container should I use? [03:37] <klaxa> hmm... i would recommend mkv (matroska) since it's probably the most advanced container available, i'm not sure stock android has a matroska demuxer though, you could try to use vorbis with mp4 though [03:39] <LunixA380> vorbis, mp4? [03:40] <klaxa> dunno, does mp4 mux vorbis? [03:40] <klaxa> i rarely use anything besides matroska [03:40] <FunkyELF> libvo_aacenc is noticeably better than libmp3lame on this crappy audio [03:42] <LunixA380> Not really [03:42] <LunixA380> it's more about using MP3 with 8000Hz the problem [03:42] <LunixA380> I don't like MP3, it does has problems, but not this one. [03:42] <klaxa> well you can either use the internal aac encoder which is not recommended, or you can go with libfdk-aac [03:42] <klaxa> or libfaac [03:43] <klaxa> http://ffmpeg.org/trac/ffmpeg/wiki/AACEncodingGuide [03:43] <FunkyELF> I'm saying I noticed audio artifacts in the mp3 [03:43] <klaxa> > You can add metadata to any of the examples on this guide: [03:43] <klaxa> >ffmpeg -i input ... -metadata author="FFmpeg Bayou Jug Band" -metadata title="Decode my Heart (Let's Mux)" output.mp4 [03:44] <klaxa> haha [03:45] <FunkyELF> Unknown encoder 'libfdk-aac' [03:46] <klaxa> well, go with "-c:a aac -strict -2" for now, if you want to use libfdk-aac you will have to either get an (undistributable?) static build of ffmpeg with libfdk-aac or compile from souce with --enable-libfdk-aac [03:47] <FunkyELF> klaxa: with aac -strict -2 I get this .... [aac @ 0x34c9220] Too many bits per frame requested [03:47] <FunkyELF> ... and this Error while opening encoder for output stream #0:1 - maybe incorrect parameters such as bit_rate, rate, width or height [03:47] <klaxa> i uh... um... hmm... [03:48] <FunkyELF> but the experimental one works [03:48] <klaxa> the native encoder ("-c:a aac") is the experimental one [03:49] <FunkyELF> okay... libvo_aacenc works fine... aac -strict -2 doesn't work [03:49] <FunkyELF> klaxa: seriously thanks for the help. Got to go [03:49] <klaxa> ok then [03:49] <FunkyELF> thanks from my mom too [03:49] <FunkyELF> ;-) [03:50] <klaxa> heh :) [04:47] <gatorX2> Hi! Can I ask you a question about android-ffmpeg-x264-master obtained from git? [04:48] <gatorX2> It's about an error message. [04:51] <klaxa> don't ask to as-- [04:51] <klaxa> well... [05:15] <praveenmarkandu> hi guys. when i log an ffmpeg transcode [05:15] <praveenmarkandu> it provides a lot of ^M [05:15] <praveenmarkandu> how to i remove that [06:23] <gustavo_brazil> someone know how to use ffmpeg to record skype conversations ? [06:24] <klaxa> that's a bit tricky unless you are using pulseaudio [06:24] <klaxa> and using pulseaudio will then require you to use a loopback- and a null-module [06:24] <klaxa> actually two loopback-modules [06:25] <gustavo_brazil> Can I use alsa instead [06:25] <gustavo_brazil> ? [06:27] <gustavo_brazil> I want to use alsa, because I updated a pulse-audio package on my Chakra and its brakes my login [06:27] <klaxa> i don't know how to do it with alsa, sorry. [06:28] <gustavo_brazil> what is the pulse packages I have to install to get it ? [06:40] <gustavo_brazil> at least you can tell me the ffmpeg command using pulse, so i can test [10:46] <efyx> Hi everybody, is it possible with ffmpeg to encode a video with closed captioning (EIA-708) ? [10:53] <Mavrik> hmm [10:53] <Mavrik> no idea, I know it can encode DVB subs [11:43] <guru_shipu> is it possible to capture from webcam and stream to the network using http? [11:44] <guru_shipu> i know we can capture from webcam, but how to stream to http? [11:44] <Mavrik> depends on what do you mean by "stream to the network" [11:44] <guru_shipu> Mavrik, depends how [11:45] <guru_shipu> would you please elaborate [11:45] <Mavrik> nothing to elaborate [11:45] <Mavrik> what are you trying to do? which streaming technology do you need? [11:45] <Mavrik> since you already know you need HTTP, you should know what kind of playback are you attempting right? [11:45] <guru_shipu> just http [11:45] <Mavrik> what is "just http"? [11:46] <Mavrik> do you have a streaming server? [11:46] <guru_shipu> nope [11:46] <Mavrik> do you have another http endpoint you're trying to stream to_ [11:46] <guru_shipu> yes [11:46] <guru_shipu> all the other machines in my home network [11:46] <Mavrik> you do know http doesn't work like that right? :P [11:47] <Mavrik> so. [11:47] <Mavrik> you need a streaming server. [11:47] <Mavrik> the streaming server will accept http requests and serve them to your machines on the network [11:47] <Mavrik> you can use http to stream TO the streaming server with http, even though RTMP or UDP/TS would be better. [11:47] <guru_shipu> ok [11:48] <Mavrik> ffmpeg bundles "ffserver" but it's not that trivial to setup [11:48] <guru_shipu> ok then, suppose I want to capture the webcam and stream to just one machine using http [11:49] <Mavrik> see this: https://www.virag.si/2012/11/streaming-live-webm-video-with-ffmpeg/ [11:58] <guru_shipu> Mavrik, thank you [12:31] <CableNinja> mornin all. I've got a couple songs that were setup as 2849Kbps... these are much too large for my uses, how can I lower them down? [12:34] <Mavrik> reencode them into a compressed format [12:37] <CableNinja> .-. I know that much.. heh, sorry... I should rephrase... I'm horrible with ffmpeg and have no idea what I'm doing [12:44] <Mavrik> CableNinja, ok :) [12:44] <Mavrik> so, what format do you want the audio in? :) [12:45] <Mavrik> and what format is it currently in? :) [12:45] <CableNinja> flac -> flac [12:45] <Mavrik> um [12:45] <Mavrik> CableNinja, that won't be possible then [12:45] <CableNinja> I dont know whose idea it was to make it such a high bitrate, but I want to give them a firm pimphand... a 100mb for a single 5 minute piece of audio is insane [12:45] <Mavrik> you can lower number of channels and/or sample rate [12:45] <CableNinja> Mavrik: alright, flac->mp3? [12:46] <Mavrik> but FLAC is a losless format, so you can't change bitrate :) [12:46] <CableNinja> whatever I can do to drop the size of these... [12:46] <Mavrik> ffmpeg -i <file.flac> -codec:a libmp3lame -b:a 320 output.mp3 [12:46] <Mavrik> that should do what you want :) [12:47] <Mavrik> (note: you need ffmpeg with libmp3lame) [12:47] <CableNinja> should have those already. I'll give it ago, thanks [12:48] <Mavrik> -b:a sets bitrate [12:48] <Mavrik> 320 obviously would be 320kbps CBR :) [12:48] <CableNinja> yup [12:49] <CableNinja> waiting for it to copy somewhere I can mess with it... I forgot to do that first, heh [15:06] <gatorX2> Hi! Can I ask a question about converting mp4 file to 3gp conversion with ffmpeg? [15:07] <zap0> try [15:07] <gatorX2> I've built ffmpeg with libaacplus and libx264 support. How can I convert mp4 to 3gp file? [15:07] <Mavrik> ffmpeg -i file.mp4 -codec copy out.3gp _ [15:07] <Mavrik> ? [15:11] <gatorX2> `Mavrik` Thanks a lot and sorry for not knowing this before and bothering you. [15:11] <gatorX2> Bye! [15:12] <saste> that was fast [15:13] <Mavrik> yeah [16:12] <khali> oh, flv container supports AAC for audio now? sweet [16:37] <Fjorgynn> hm [16:57] <xlinkz0> why is the build script still trying to compile ffprobe.o if i configured with --disable-ffprobe ? [17:06] <saste> xlinkz0, because it's stubborn [17:22] <xlinkz0> saste: is there any way to make it obey? [17:27] <mkrufky> so, my transcoder application uses ffmpeg libraries... it runs fine against 1.2 but it segfaults in 2.0 ... after debugging in gdb it looks like a memory overrun, maybe an off-by-1 error somewhere. i spent some time doing a git bisection test to find the problem changeset and it ended at a merge changeset :-( 80e9e63c946660304fc65fa8141ccfdbe4d196d1 no idea where to begin now :-/ [17:28] <mkrufky> er, actually, the particular merge just merges in the patch "759001c534287a96dc96d1e274665feb7059145d" ... but this itself is a huge one [17:39] <shagoyjo> Transcoding interlaced content, does the +-alt flag still exist? [17:51] <shagoyjo> I ask because I get "unable to parse option value 'alt'" [19:55] <mkozjak> hi [19:56] <mkozjak> how to copy video and audio and hardburn srt? ffmpeg -i input.mp4 -vcodec copy -acodec copy -vf subtitles=subs.srt out.mp4 just copies everything and ignores subtitles [19:57] <llogan> mkozjak: -vcodec copy and -vf are mutually exclusive since you must re-encode to hardcord the subs [19:57] <mkozjak> aah, yeah.. [19:58] <mkozjak> llogan: is there a simple way to use the same picture properties (resolution, framerate) and just burn them? [19:58] <llogan> ffmpeg will use the same frame rate and frame size as the input [19:58] <llogan> unless you tell it not to [19:59] <llogan> depending on your ffmpeg configuration libx264 will be the default encoder for mp4 output container. you should mostly be concerned with -crf and -preset when re-encoding (unless the defaults look good enough) [20:01] <llogan> see http://trac.ffmpeg.org/wiki/x264EncodingGuide [20:02] <mkozjak> thanks [20:49] <_aeris_> hello everybody :) [20:49] <_aeris_> i have a question about ffmpeg [20:49] <_aeris_> i have a video, saying Input #0, avi, Video: msmpeg4v3 (DIV3 / 0x33564944) [20:50] <_aeris_> but if i try to change resolution with ffmpeg -i input.avi -s 1280x720 -c:v msmpeg4v3 output.avi [20:50] <_aeris_> i have on output Video: msmpeg4v3 (MP43 / 0x3334504D) [20:50] <_aeris_> input = DIV3, output = MP43 How i can have DIV3 on output too ? [20:54] <_aeris_> http://pastebin.com/qs33NG7S [20:59] <llogan> _aeris_: you can add "-vtag DIV3" if that is what you want [21:00] <_aeris_> :) [21:00] <_aeris_> so simple :D [21:01] <llogan> you should also add "-codec:a copy" so you do not re-encode the audio. [21:07] <mablae> Hi! [21:10] <mablae> I need libavcodec-dev <=55.0.0 on Ubuntu 13.04 Raring. Packages just give me libavcodec53-dev. What to do? [21:11] <llogan> mablae: you can compile it [21:11] <llogan> http://trac.ffmpeg.org/wiki/UbuntuCompilationGuide [21:14] <mablae> llogan, So current ffmpeg build we automatically install libavcodec-dev for me? Just make ; make install ? Or any other config needed? [21:14] <mablae> *will [21:15] <mablae> Okay, thats great then... :-) Thanks [21:16] <llogan> it won't make a "libavcodec-dev" package, but you can do that with checkinstall instead of "make install" if that's what you want [21:16] <llogan> the guide does not install to the system [21:18] <mablae> It would be nice to have some deb packages that can be uninstalled and dont mess up depedencies... [21:19] <llogan> dependencies of ffmpeg package are not interfered with in the guide since it does nto isntall to system [21:20] <llogan> *not install [21:23] <mablae> But I dont need new ffmpeg, I need new libs , dev libs to comile sth other against it [21:23] <mablae> Tvheadend to be precise [21:24] <mablae> It has some build in transcoding using libavcodec and its ./configure complains about: [21:25] <mablae> checking for pkg libavcodec <=55.0.0 ... fail [21:25] <mablae> BTW thanks for the help, llogan [21:26] <llogan> the source contains the libraries [21:29] <llogan> or you can try this if you get stuck: https://tvheadend.org/projects/tvheadend/wiki/AptRepository [22:02] <mablae> llogan, back again, when compiling x264 like described in that guide I got this error: undefined reference to `x264_encoder_open_130' [22:02] <klaxa> wrong x264 shared library installed [22:02] <mablae> /usr/local/lib/libavcodec.a(libx264.o): In function `X264_init' [22:03] <klaxa> do you have x264 installed? if so, can you post the version it states? (x264 --version) [22:03] <mablae> ah, fuck I messed around with avconv... [22:03] <mablae> sry for the f-word [22:03] <klaxa> also run: ldconfig -p | grep x264 [22:03] <klaxa> and see what version that states [22:04] <mablae> x264 --version: http://paste.ubuntu.com/5891913/ [22:05] <mablae> ldconfig -p | grep x264 : http://paste.ubuntu.com/5891914/ [22:05] <klaxa> hmm... your installed x264 is version 130, but the shared library is version 123 [22:06] <mablae> As I said I messed around with avconv, too [22:09] <mablae> klaxa, Thanks for the help! [22:09] <klaxa> oh it worked? did you re-install x264 or what fixed it? [22:09] <klaxa> or did you just go the dirty way and copy libx264.so.130 to /usr/local/lib =? [22:10] <mablae> No, I am tryin but you gave me a good hint [22:11] <mablae> I did a 'apt-get install x264' [22:13] <mablae> Okay : I had the x264 v130 on /usr/local/bin and 123 is now on /usr/bin [22:14] <mablae> remove the 130 [22:14] <mablae> *I removed [22:15] <mablae> But why do I need x264 when I actually want to compile it? Thats seems strange [22:16] <mablae> I think /usr/local/lib/libavcodec.a is the problem [22:16] <mablae> That is from my experiement with avconv I guess [22:18] <mablae> How would I fix my ldconfig to use ubuntu's system libavcodec again?! [22:23] <mablae> klaxa, haha I did some mixup with ffmpeg and avconv libs# [22:23] <mablae> Not good at all :-) [23:29] <eth2wlan> Hi [23:29] <eth2wlan> I need to capture X minute of videostream, save it to a file, then exit [23:29] <eth2wlan> I'm using this command line: [23:30] <eth2wlan> ffmpeg -i 'http://192.x.x.x/videostream.cgi?user=foo&pwd=bar&resolution=32&rate=2` -acodec copy -vcodec copy '/home/foobar/delme/foobar.mkv` [23:30] <eth2wlan> But, when I terminate process with CTRL [23:30] <eth2wlan> CTRL+C, foobar.mkv doesn't exists [23:30] <eth2wlan> Any idea? [23:30] <klaxa> see the -t parameter [23:30] <klaxa> -t is duration [23:31] <klaxa> -t 600 will record for 10 minutes [23:32] <eth2wlan> Ok, I'll try immediately [23:38] <eth2wlan> There is a wrong character in my previous command line [23:38] <eth2wlan> (`) istead of (') [23:39] <eth2wlan> With -t application break for error: [23:39] <eth2wlan> http://192.168.100.90/videostream.cgi?user=giadmin: Input/output error [23:39] <eth2wlan> Oh, sorry [23:44] <eth2wlan> ( "please hold the line" ) [23:44] <eth2wlan> :-) [23:46] <eth2wlan> Done [23:46] <eth2wlan> Look here: [23:46] <eth2wlan> http://pastebin.com/qyDLp1dT [23:47] <llogan> to get hel here you'll have to use ffmpeg, or you can go to #libav if you want to use fake ffmpeg [23:47] <llogan> *help here [23:50] <eth2wlan> So, I must compile for Ubuntu? [23:51] <eth2wlan> "apt-get install" can't work at all ? [23:56] Action: eth2wlan is reading UbuntuCompilationGuide [00:00] --- Sat Jul 20 2013

1 0

ffmpeg-devel.log.20130718
by burek 19 Jul '13

19 Jul '13

[00:46] <superware> I have an RTP/h264 video stream, when I ffplay -f h264 rtp://0.0.0.0:1234 it plays only half picture, the lower area seems corrupted. Do I need to specify more information about the stream? http://pastebin.com/MTxZw0PL [00:50] <superware> michaelni: hi [00:51] <kierank> dropping packets or something [00:54] <superware> kierank: it works with VLC, so I thought it's something I need to define in the command line [00:58] <superware> kierank: can I set RTP Packetization Mode from the command line? [00:58] <kierank> dunno [00:59] <superware> must I play an SDP file? [01:04] <cone-660> ffmpeg.git 03Fabian Neundorf 07master:353f302250fd: lavf/matroskaenc: using valid chapter ids [01:19] <superware> I have an RTP/h264 video stream, when I ffplay -f h264 rtp://0.0.0.0:1234 it plays only half picture, the lower area seems corrupted. Do I need to specify more information about the stream? http://pastebin.com/MTxZw0PL [01:27] <cone-660> ffmpeg.git 03Michael Niedermayer 07master:37ded5303719: vf_scale: use sws_init_context() [01:54] <cone-660> ffmpeg.git 03Lukasz Marek 07master:247e658784ea: ftp: fix interrupt callback misuse [01:54] <cone-660> ffmpeg.git 03Lukasz Marek 07master:db72b7742a94: ftp: remove unused headers [01:54] <cone-660> ffmpeg.git 03Lukasz Marek 07master:2217243e1234: ftp: comments [01:54] <cone-660> ffmpeg.git 03Lukasz Marek 07master:816c579cf3cb: ftp: warning about pure-ftp server used as and output [01:54] <cone-660> ffmpeg.git 03Michael Niedermayer 07master:1816f5509e4f: Merge https://github.com/lukaszmluki/ffmpeg [02:08] <cone-660> ffmpeg.git 03Lukasz Marek 07release/2.0:fcab45f39bda: ftp: fix interrupt callback misuse [08:00] <bernie_> the option -g, can take INT_MIN to INT_MAX. Not 0 to INT_MAX. What does a negative group of picture size signify? [09:13] <wm4> saste: forgot about it [09:13] <wm4> saste: so there was the linesize issue, right? [09:13] <wm4> saste: I don't understand why the linesize would include padding, that makes no sense to me [09:13] <wm4> planar audio doesn't use the linesize for access, or does it [10:02] <saste> wm4: I tested your patch, and noted the audio glitch [10:02] <saste> and yes, in general audio linesize is padded, for alignment reasons [10:03] <wm4> yes, I think I actually experienced the same problem with some files, I just didn't get what was wrong [10:03] <wm4> so why is the linesize padded? [10:03] <saste> for alignment reasons [10:03] <saste> linesize is the padded size of the plane [10:04] <wm4> I understand that for the allocation, but the linesize itself could be just the exact size, because unlike video frames, you don't need them for access [10:04] <saste> yes [10:04] <wm4> s/them/it/ [10:04] <saste> i don't remember the exact reasoning we did [10:04] <saste> also iirc that was merged from libav, so that's probably a better place where to ask [10:05] <wm4> I see [10:07] <wowdd1> hello [10:08] <wowdd1> is there some good book about multimedia ? [10:26] <cone-766> ffmpeg.git 03Diego Biurrun 07master:3ac7fa81b238: Consistently use "cpu_flags" as variable/parameter name for CPU flags [10:26] <cone-766> ffmpeg.git 03Michael Niedermayer 07master:9d01bf7d66df: Merge remote-tracking branch 'qatar/master' [11:21] <cone-766> ffmpeg.git 03James Almer 07master:562fb9c54099: lavf/riff: Add ITRK tag [11:30] <wm4> michaelni: just curious; how do you merge? do you manually merge until the first conflicting commit, or is there a magic git-merge option that does it for you? [11:53] <michaelni> i mostly use git merge and a text editor but occasionally i use/try different methods [11:55] <wm4> michaelni: I saw this, it sort of looks like it might be useful (though I haven't tried it myself, so I don't know if it works well): http://softwareswirl.blogspot.de/2013/05/git-imerge-practical-introduction.… [11:55] <wm4> it's a git extension for incremental merging [12:08] <cone-766> ffmpeg.git 03James Almer 07master:630fc7dcfc31: vorbiscomment: Add DESCRIPTION to ff_vorbiscomment_metadata_conv [13:51] <cone-766> ffmpeg.git 03Nicolas George 07master:ebaf20e94b99: lavfi/scale: allocate interlaced scalers only if needed. [14:32] <wm4> here's an interesting idea for libavfilter: make it possible for the graph to export variables [14:32] <wm4> which the application can set in a controlled manner [14:32] <wm4> even at runtime [14:32] <wm4> maybe that'd be better than the command stuff [16:54] <Daemon404> nevcairiel, fyi MSE still has issues with c00conv [16:54] <Daemon404> er c99conv [16:55] <JEEBsv> lol [16:55] <Daemon404> i just ran into it again [16:57] <nevcairiel> yeah i noticed it at home as well [16:57] <nevcairiel> apparently my laptop here just balances out between IO and CPU power [16:57] <nevcairiel> so you can get 100% [16:57] <Daemon404> lol [16:57] <Daemon404> aka magic [16:58] <nevcairiel> although at home its not one core anymore, its like 50% cpu now since i move my code onto a SSD [16:58] <nevcairiel> moved* [18:07] <cone-675> ffmpeg.git 03Ed Torbett 07master:7203dbde3910: avformat/rt*p: Joining a SSM multicast group using an SDP (Issue #2171) [18:51] <cone-675> ffmpeg.git 03Carl Eugen Hoyos 07master:42272e86fea5: lut3d: Fix reading 3dl files with leading comments. [18:51] <cone-675> ffmpeg.git 03Carl Eugen Hoyos 07master:36b21e17a23c: lavf/concat: Never fail for sample aspect ratio 0:1. [18:51] <cone-675> ffmpeg.git 03Carl Eugen Hoyos 07master:b39a6bbe7f43: Fix pix_fmt detection in the native jpeg2000 decoder. [18:51] <cone-675> ffmpeg.git 03Michael Niedermayer 07master:ed8c34a7666a: Merge remote-tracking branch 'cehoyos/master' [20:54] <mateo`> durandal_1707: do you know if some work (or brainstorming) has been started on how to handle stream with AV_DISPOSITION_ATTACHED_PIC ? [20:56] <wm4> what's there to handle? [21:04] <mateo`> wm4: try to find a better way to handle attached pics in lavf [21:06] <wm4> better in what way? [21:08] <mateo`> not treated as a normal stream [21:09] <mateo`> that's just an idea ... [21:10] <mateo`> the way the cover art are handled in the different muxer is quite ugly [21:10] <mateo`> imho [21:16] <mateo`> supporting such feature in the mov muxer for example results in intrusive code [21:21] <mateo`> storing the covert arts in some kind of metadata would be nice but we need to keep the ability to transcode the cover art itself [21:34] <durandal_1707> mateo`: nothing, and i [21:34] <durandal_1707> 'm not movitaved to do/fix it [21:35] <mateo`> i'm motivated to do it [21:37] <durandal_1707> imho, metadata should remain metadata [21:37] <durandal_1707> but some "smart" libav dev thought that it better to make it stream. so avplay can display it [21:38] <durandal_1707> and michaelni want abi compability with evil fork, so be it.... [21:38] <bernie_> I wonder if CueRelativePosition is implemented correctly in the matroska format. I'm getting byte offsets that easily fall outside the size of the cluster itself. [21:38] <mateo`> so we are basically doomed :D [21:38] <wm4> mateo`: well, you can access the AVPacket directly, instead of treating the AVStream as video stream [21:40] <bernie_> actually, possible just a decoding bug... (sorry) :) [21:42] <mateo`> wm4: but you'll still have to discard in some way this kind of stream in the muxer [21:44] <bernie_> yeah, okay -- now I see that CueRelativePosition is not being set in the encoding for matroska, though it is part of the spec (I guess it's pretty new). I'm looking for fast seek performance... would be good to set that EBML element for Cues. [21:44] <mateo`> wm4: assuming the muxer will automatically map all video streams [21:47] <bernie_> this would mean chaning mkv_add_cuepoint() in matroskaenc.c to include the relative byte offset in the cluster to get to the block. [22:03] <wm4> bernie_: or just use mkvmerge [22:04] <wm4> bernie_: how is seeking slow? are you sure it isn't just because of keyframes? [22:05] <bernie_> I'm writing my own video player for iOS for a laundry list of reasons... Seeking isn't really going to be slow I don't think, but using CueRelativePosition just makes sense, that way you do not need to parse through all preceding blocks before getting to the block you really want to go to. [22:05] <bernie_> I think I can make a patch that would just include this information [22:06] <bernie_> as per the matroska spec to-date. [22:06] <bernie_> It should be quite trivial to add... so I think... it just makes sense to add it. [22:07] <bernie_> The specific ID for CueRelativePosition is 0xf0 [22:07] <bernie_> as docuemnted here: http://matroska.org/technical/specs/index.html [22:08] <bernie_> In my situation, I'm in contorl over both the encoding and decoding process for the content I wish to present to the end-user, so this is a reasonable option. [22:08] <bernie_> I guess my question is... should I make this patch, if it would be of general interest to ffmpeg to include it. [22:10] <wm4> it probably would, though I don't really see how it's useful (cluster sizes are limited by convention anyway?) [22:10] <bernie_> it's useful like so: [22:11] <bernie_> http://permalink.gmane.org/gmane.comp.multimedia.matroska.devel/4454 [22:12] <bernie_> Specifically: At the moment seeking to a particular block requires [22:12] <bernie_> seeking to the cluster the block contains for two reasons: First, the [22:12] <bernie_> ClusterTimecode element has to be found. Second, the block's actual [22:12] <bernie_> position inside the cluster is unknown. A splitter usually finds the [22:12] <bernie_> ClusterTimecode as the first child in the cluster, but then it still [22:12] <bernie_> has to scan all children up to the required block. [22:12] <bernie_> With this proposal reading the ClusterTimecode would not change. [22:12] <bernie_> However, a splitter could then seek directly to the required block [22:12] <bernie_> skipping all the elements in between. This would speed up seeking [22:13] <wm4> ok so you can skip all blocks in the start of the cluster [22:13] <bernie_> correct [22:13] <bernie_> that would be faster [22:13] <wm4> but if the seek destination (including preroll) falls at the start of a cluster, you have to parse everything yet again? [22:13] <wm4> even if e.g. subtitles are sparse at best [22:13] <bernie_> cuepoints are typeically I-frames [22:14] <bernie_> in my situation they definately are [22:17] <bernie_> In fact, it only makes a cuepoint entry for keyframe. So there ought be no reason to have to re-prase. You should be able to go to the cluster, and then go to the correct offset to start reading the block that contains the I-frame you need to start. At least, this is how I understand it. [22:18] <wm4> I think there was something else that was actually useful [22:18] <wm4> oh right [22:18] <wm4> cue duration, I think [22:19] <wm4> so you can estimate whether you really have to seek before to get a subtitle [22:21] <bernie_> okay, so I'm working on a patch to implement this... not sure what the process is to submit it for consideration of inclusion. But first I will have lunch :) [22:22] <wm4> send it to the -dev mailing list [22:25] <durandal_1707> bernie_: read http://ffmpeg.org/developer.html [22:25] <bernie_> cool -- thanks! [22:26] <wm4> durandal_1707: when will cehoyos read these guidelines? [22:27] <wm4> durandal_1707: if you want to do something awesomely passive-aggressive, reject his next patch and point to the dev policy, and quote "When you submit your patch, please use git format-patch or git send-email. We cannot read other diffs :-)" [22:32] <cone-561> ffmpeg.git 03Michael Niedermayer 07master:99de97cabf35: jpeg2000dec: parse CDEF [22:32] <cone-561> ffmpeg.git 03Michael Niedermayer 07master:fc6de70c44be: jpeg2000dec: Support non subsampled 8bit planar pixel formats [23:04] <bernie_> wow make fate takes a while... :) [23:21] <bernie_> great -- now I have to also patch mkvinfo to support CueRelativePosition. [23:23] <bernie_> It's output is funny though, because it can ID the EBML tag by name: (Unknown element: CueRelativePosition; ID: 0xf0 size: 9) at 4993253 size 9 [23:24] <bernie_> One the one hand, it knows that 0xf0 is CueRelativePosition, but on the other it's unknown... grrr! [23:42] <bernie_> is there an easy way I can run just a single test? In my case, I need to re-run lavf-mkv, and a global make fate takes too long... [00:00] --- Fri Jul 19 2013

1 0

ffmpeg.log.20130718
by burek 19 Jul '13

19 Jul '13

[00:00] <relaxed> Maybe they removed that feature and it's now a default? [00:01] <LunixA380> Removed? [00:02] <LunixA380> Hm, wait, I remembered a log, I'll paste it [00:15] <LunixA380> So : http://pastebin.com/ja5LgFan [00:17] <LunixA380> I taken the static build from ffmpeg.zeranoe.com, and this uh...the 6 July [00:18] <LunixA380> ffmpeg-20130709-git-00b1401-win32-static.7z * [00:19] <LunixA380> line 202, we see rc_resize_allowed: 0 [00:19] <LunixA380> but I'm unable, for the time being, to change it. [00:24] <LunixA380> I stay connected ^^ [00:24] <LunixA380> I'll brb [00:24] <durandal_1707> LunixA380: when was it working? [00:25] <LunixA380> I never tested it [00:25] <durandal_1707> its supposed to be changeable from user? [00:25] <saste> LunixA380, was never a valid option [00:26] <LunixA380> oO [00:26] <LunixA380> So the website is lying? [00:26] <relaxed> If you need it you can encode with vpxenc directly. [00:26] <LunixA380> Okay. [00:26] <durandal_1707> this is trivial to add, if you reall need it [00:27] <relaxed> maybe a -vpx-params option is in order. [00:28] <saste> LunixA380, what website is lying? [00:30] <durandal_1707> that page even have failing link to libvpx in ffmpeg [00:32] <mark4o> yup, looks like http://wiki.webmproject.org/ffmpeg is wrong; maybe check on #webm or #vp8 ? [00:34] <saste> LunixA380, don't trust third parties wikis [00:34] <saste> otoh our internal documentation for livpx sucks [00:41] <superware> I have an RTP/h264 video stream, when I ffplay -f h264 rtp://0.0.0.0:1234 it plays only half picture, the lower area seems corrupted. Do I need to specify more information about the stream? [00:43] <superware> please see http://pastebin.com/MTxZw0PL [01:03] <superware> must I play an SDP file in order to play h264 over RTP? [01:05] <superware> vlad_starkov: hi [01:20] <superware> I have an RTP/h264 video stream, when I ffplay -f h264 rtp://0.0.0.0:1234 it plays only half picture, the lower area seems corrupted. Do I need to specify more information about the stream? http://pastebin.com/MTxZw0PL [01:25] <ilove11ven1> superware: check packet loss [09:15] <LunixA380> Welcome back ffmpeg support channel :) [09:15] <LunixA380> Nothing to ask for now ^^ [09:16] <relaxed> In soviet #ffmpeg, we welcome you. [09:17] <LunixA380> soviet #ffmpeg? Is there is regional ffmpeg channels :o? [09:20] <relaxed> No. [09:52] <braincracker> h [09:52] <braincracker> hack your 19" tft monitor CCFL to LED backlight http://pastebin.com/6QrP8ZbR neat? [18:13] <efyx> Hi everybody, is it possible with ffmpeg to encode a video with closed captioning (EIA-708) ? [18:14] <efyx> Or do you know any other open source tool that can do it? [18:52] <Sythic> Hey im trying to compile FFMpeg (need FFProbe as well). I configured it, hopefully correctly for what i need (screenshots from ffmpeg and some metadata from ffprobe). However it wont do the install. I get /root/deps/ffmpeg/libavcodec/libx264.c:552: undefined reference to `x264_encoder_open_135' [18:52] <Sythic> Any ideas what I might be doing wrong? [19:08] <teratorn> Sythic: you mean it won't compile? [19:09] <sacarasc> Sythic: Do you have multiple copies of x264 installed? [19:20] <Sythic> teratorn: sacarasc, um im not sure, make works, make install does not [19:20] <sacarasc> Yeah, that's what usually happens when you have it installed twice. [19:20] <Sythic> sacarasc: i dont think so.. i uninstalled the one i had, x264 command doesnt work, then i compile from the repo.. [19:21] <teratorn> "uninstalled" ? [19:21] <teratorn> that's not something that usually works on a computer [19:21] <Sythic> apt-get remove x264 [19:21] <teratorn> ok [19:21] <Sythic> idk if my versions are maybe off or what [19:22] <Sythic> i have the latest master of both x264 and ffmpeg [19:22] <teratorn> Sythic: it shouldn't be resolving references in .c files if it were already compiled [19:22] <Sythic> weird [19:22] <Sythic> maybe it failed but didnt tell me? [19:23] <teratorn> pastebin the full output of 'make' [19:23] <Sythic> well that will take about 20 minutes lol [19:23] <teratorn> it shouldn't, if it's mostly already compiled [19:23] <Sythic> oh you can do it again? [19:23] <teratorn> sure [19:24] <Sythic> would have been good to know 2 hours ago > [19:24] <Sythic> >< [19:25] <Sythic> running make again gives me the error that make install used to, is that normal? [19:26] <Sythic> sacarasc: any way I could check if i have 2 copies installed? [19:26] <teratorn> you didn't notice it the first time [19:26] <teratorn> pastebin the whole output [19:26] <Sythic> Gotcha [19:27] <Sythic> teratorn: http://codep.in/2r [19:32] <teratorn> Sythic: it sounds like you've linked against a different version of x264 than the headers you've compiled against [19:32] <Sythic> Must have, not sure what to do about that [19:32] <Sythic> i uninstalled x264 completely before hand [19:33] <teratorn> you can use objdump -t libavcodec/libavcodec.a to inspect what symbols have been compiled in [19:33] <teratorn> you'll have some symbol like x264_encoder_open_<BUILD_NUM> [19:34] <teratorn> so it's linking in a certain version of the x264 static library. which doesn't have the symbol the ffmpeg code expects (which it gets from the x264 headers, I think) [19:35] <teratorn> Sythic: where did you install x264 to (what --prefix ?) and what additional flags are you giving to ffmpeg ./configure so it knows where to find it? [19:36] <Sythic> I followed the guide on the ffmpeg wiki, i dont think any of them had me to prefixes [19:37] <Sythic> this one: https://ffmpeg.org/trac/ffmpeg/wiki/How%20to%20quickly%20compile%20FFmpeg%2… [19:37] <teratorn> so you sudo installed? to where? /usr/local? can you search your whole hard drive for x264.h and tell us where you find them? [19:37] <Sythic> im logged in as root [19:37] <Sythic> sure [19:42] <Sythic> teratorn: http://codep.in/2s [19:42] <Sythic> so yea, couple places [19:43] <teratorn> Sythic: the first is your git checkout? [19:43] <Sythic> yes [19:43] <teratorn> so you have two versions, one in /usr and one in /usr/local [19:43] <Sythic> which can i delete? [19:44] <teratorn> Sythic: the one in /usr is from system package, so you need to remove libx264-dev [19:44] <teratorn> Sythic: I assumed you already had done that as a result of apt-get removing x264 [19:44] <Sythic> i would have thought too [19:44] <teratorn> do you have that package? [19:45] <Sythic> ok so now its just the /usr/local [19:45] <Sythic> yea i did [19:45] <teratorn> weird [19:45] <teratorn> you must have libx264 then also [19:45] <teratorn> since the -dev depends on it [19:45] <teratorn> you need to remove that also, or just make sure you aren't linking against it [19:45] <Sythic> nope lol [19:45] <teratorn> what about libx264<something> [19:46] <teratorn> Sythic: here it is libx264-123 [19:46] <teratorn> Sythic: you have something im sure of it [19:46] <Sythic> I think its compiling [19:46] <Sythic> i got past that error [19:46] <Sythic> sweet it works [19:46] <teratorn> lol ok [19:48] <Sythic> I think its gonna do it, i so hope this works, tired of watching code compile hahah [19:48] <Sythic> thanks for the help teratorn sacarasc [19:49] <_janem> hi, why would ffmpeg -i input.mov -sameq output.mov not create an output.mov with the same fps? [19:49] <teratorn> Sythic: sure [20:33] <Sirisian|Work> I'm confused about something. If I want to convert a 1080p ogg file to webm and keep the same quality I can't use ffmpeg -i foo.ogg bar.webm. It seems to create a very low quality one. Is it trying to do the conversion at a low bitrate by default? I want it to convert using the best bitrate for the video to get an identical quality [20:36] <Sirisian|Work> hmm -sameq [20:39] Action: _janem guessing could it be to do that libx264 is not installed? "Unknown encoder 'libx264'" [20:47] <Sirisian|Work> I got it [20:47] <Sirisian|Work> had to use -b 1500k and other flags. Can't run it without flags [20:57] <LunixA380> hm Sirisian|Work ? [20:57] <LunixA380> It's weird. [21:07] <Sirisian|Work> LunixA380, oh I tried to use just ffmpeg -i foo.ogg bar.webm thinking it would convert with the same quality. Ended up having to use ffmpeg -i foo.ogg -b 1500k -vcodec libvpx -acodec libvorbis -ab 160000 -f webm -g 30 bar.webm [21:07] <Sirisian|Work> Simple mistake. First time using ffmpeg. [21:07] <LunixA380> I'll investigate a bit [21:07] <LunixA380> but I thought it was possible to repackage a file without reencoding [21:25] <_janem> LunixA380: well that what I thought -sameq is for (but I might missing libx264 to this might solve it) [21:40] <mark4o> do not use -sameq [21:40] <mark4o> http://superuser.com/questions/478549/what-is-the-sameq-or-same-quant-optio… [00:00] --- Fri Jul 19 2013

1 0

ffmpeg-devel.log.20130717
by burek 18 Jul '13

18 Jul '13

[00:12] <durandal_1707> .. compand A->A Compress or expand the dynamic range of the audio. [00:18] <wm4> michaelni: will you push the colorspace stuff, or is there still a need for discussion? [00:18] <michaelni> wm4, i intend to push it after double checking its ok [00:20] <wm4> michaelni: btw. you were saying that chroma sample location should not be a simple enum - doesn't the same apply to colorspaces? [00:20] <michaelni> that depends on your viewpoint [00:21] <michaelni> the chroma position changes with crop & field extract merge in current enum unfriendly ways [00:21] <michaelni> and these are normal operations [00:22] <michaelni> the colorspace shouldnt change in such ways with common operations at least i dnt see any ATM [00:22] <wm4> I see [00:42] <bernie2> howdy [00:47] <kml2> Hi michaelni . This http://pastebin.com/bj4whbbA . Build of `master` fails to play a video; build of `release` works. [00:47] <kml2> I'm not sure if you consider that 'usage' or 'dev' [00:47] <bernie2> I was wondering why ffmpeg is not allowing a VP8 video encoding and AAC for audio. I know that no browser out there will read that, but I'm developing something for a unique video player, and I'm confused by the assertion that webM doesn't allow it. I know that Matroska spec would allow any combination. Webm is really just matroska. [00:48] <Skyler_> webm is a subset of matroska, a subset that doesn't include AAC [00:48] <Skyler_> but writing to mkv should work just fine. [00:48] <Skyler_> vp8 + aac in mkv [00:48] <kierank> chrome might read it [00:48] <bernie2> oh, so instead of using a webm extension, using mkv will do the trick? [00:48] <bernie2> holy smoles my irc client sucks... [00:49] <kierank> maybe, try it and see [00:49] <kierank> chrome uses ffmpeg to demux and decode so it might work [00:49] <michaelni> kml2, if you submit a patch or intend to work on fixing the bug yoursef its dev [00:50] <kml2> michaelni: I'm not that good yet; just startin up. I am tracking down stuff in XBMC, e.g., and actually found a couple of bugs. My fixes are considered unacceptable hacks, though ;-) So if by "work on it" you include "help as much as possible", then yes :-) [00:53] <kml2> michaelni: Anyway, this ^^^ only happens on some machines I build the new ffmpeg-master on -- not all. So there's some sort of ENV dependency. But the built-from-src release version definitely works -- standalone AND as an external ffmpeg to XBMC. the -master branch does not. [00:53] <michaelni> kml2, probably best if you open a ticket on trac and upload the file that causes the problem unless these have already been done. next would be finding out the exact commit that introdeucd the problem [00:54] <kml2> michaelni: Ok. I usually find this stuff on my full movies. Like this time. Is there some handy way to 'cut' out a piece of a video source file that's more reasonable in size, and still useful to upload? [00:55] <michaelni> dd might work, it has to be tested if the problem is still reproduceable with a cut file [00:56] <cone-237> ffmpeg.git 03Stefano Sabatini 07master:0092bcdf06f2: lavc/avcodec.h: fix various typos in av_bitstream_filter_filter() docs [00:56] <michaelni> also -vcodec copy might work to cut pieces out [00:56] <Skyler_> ffmpeg -ss X -t Y -i input -vcodec copy -acodec copy output may work, where X is the starting time in seconds and Y is the duration in seconds , but like michaelni said, make sure the problem still occurs with the cut file [00:56] <Skyler_> as remuxing might not preserve the problem [00:57] <kml2> ok, thanks. I'll try to put together a test case. [00:58] <bernie2> omg -- weechat-curses isso slow it feels like I'm working over a 10 baud modem [01:00] <bernie2> Skyler: Thanks -- looks like -f Matroska is working [01:02] <bernie2> got to switch IRC client -- this weechat is entirely unsuable... [01:17] <cone-237> ffmpeg.git 03Michael Niedermayer 07master:a80e622924c8: avcodec/avutil: Add AVColorSpace and AVColorRange to AVFrames [01:27] <cone-237> ffmpeg.git 03Stefano Sabatini 07master:53c853e04916: lavc/h264_mp4toannexb: improve feedback in case of invalid bitstream [01:59] <kml2> michaelni: bug filed with attachment. thanks for the help. see 'whomever' over there. [01:59] <kml2> Skyler_: your 'copy' string works great. thx. [02:06] <mark4o> kml2: Any reason you used 0.11? ffmpeg 2.0 is the latest stable release, and 1.2.1 before that. [02:11] <kml2> mark4o: Apparently, because I didn't look far enough down the 'git tag -l [02:11] <kml2> list :-( [02:12] <kml2> mark4o: try 2, I suppose? [02:15] <mark4o> 2.0 was very recent so it is likely to be similar to git head, but it would help to narrow the problem, especially if you can bisect to find the specific commit [02:15] <mark4o> btw your file plays fine for me [02:15] <kml2> mark4o: It does for me, too, on other machines. Like I mentioned in the bug -- some, not all. [02:18] <kml2> atm, 2/6 have this problem. one of those 2 problem machines is exactly the same hardware as my desktop, with does NOT have the problem. [02:19] <mark4o> well if you are the only one that can reproduce it then you'll probably need to bisect it in order for anyone to be able to find the problem [02:19] <kml2> 'bisect' meaning -- find the release it appeared in? [02:21] <kml2> aha! there's a "git bisect"! (https://www.kernel.org/pub/software/scm/git/docs/git-bisect-lk2009.html) Who knew? [02:21] <kml2> reading [02:23] <mark4o> yes git bisect; there is also a bisect-create script in the ffmpeg/tools directory that might help but I haven't used it [02:24] <kml2> mark4o: Thx. Looking for the git release history ... so I can get GOOD & BAD commit#s [02:25] <kml2> got it: http://git.videolan.org/?p=ffmpeg.git;a=summary [02:27] <johnsmith_> Hi all [04:54] <cone-734> ffmpeg.git 03Michael Niedermayer 07master:b405f4e91624: swscale: Add support to specify chroma position [09:35] <cone-734> ffmpeg.git 03Stefano Sabatini 07master:34322ab27a50: doc/bitstream_filters: document dump_extra [09:35] <cone-734> ffmpeg.git 03Stefano Sabatini 07master:9836f47cf8b6: doc/filters: update/correct mp documentation [10:42] <cone-734> ffmpeg.git 03Sean McGovern 07master:50612484e058: pthread: Rename thread_init to avoid symbol collision [10:42] <cone-734> ffmpeg.git 03Michael Niedermayer 07master:10336ca6400a: Merge commit '50612484e058e8b241f0528584d64d9d2ccebf12' [10:50] <cone-734> ffmpeg.git 03Christophe Gisquet 07master:b6293e2798af: fmtconvert: Explicitly use int32_t instead of int [10:50] <cone-734> ffmpeg.git 03Michael Niedermayer 07master:b4fe41c981de: Merge remote-tracking branch 'qatar/master' [11:44] <j-b> https://www.ohloh.net/p/ffmpeg missing logo [12:35] <cone-734> ffmpeg.git 03James Almer 07master:088ed5314694: lavf/matroskaenc: Check for valid metadata before creating tags [13:20] <Compn> j-b : it was there the other day [13:21] <Compn> >Managers: [13:21] <Compn> Diego Biurrun [13:58] <cone-734> ffmpeg.git 03Michael Niedermayer 07master:066111bf1951: matroskaenc: simplify mkv_check_tag() [13:58] <cone-734> ffmpeg.git 03Vignesh Venkatasubramanian 07master:33968c201cc8: Fixing a stride issue in VP8 Alpha encode [13:59] <durandal_1707> michaelni: could you manually edit commit logs you commit so they make more sense? [14:12] <wm4> "make more sense"? [17:01] <cone-734> ffmpeg.git 03Michael Niedermayer 07master:a0b7e73f8bae: swscale/options: fix copy and paste typo [17:42] <wm4> michaelni: can we have at least a function that calculates libswscale chroma position from the libavcodec enum and a pixel format? [18:02] <kierank> that's non trivial [18:03] <kierank> especially with interlacing [18:14] <wm4> that's even more of a reason [18:15] <wm4> in what situations not involving interlacing is it not trivial, though? [19:03] <kierank> lots of pixel format conventions [19:04] <kierank> 4:1:0 etc [20:59] <Compn> michaelni : any response from rockbox re mips fixed point decoders ? [21:02] <Compn> imo just commit mips stuff. has rockbox sent any patches for their stuff in years ? [21:15] <saste> ubitux: you gonna review the lut3d patch? [21:53] <saste> how should we fix the movie loop thing? [21:53] <saste> should we reset the timestamps, or request to add a setpts? [21:53] <saste> or auto-insert a setpts filter? (which looks the more complicate solution at the moment)? [22:06] <saste> and now the socis website is down... [22:14] <Compn> saste : hows mencoder do it ? [22:14] <Compn> movie loop that is [22:15] <saste> Compn, why do you ask me about mencoder? [22:15] <Compn> dunno :P [22:16] <Compn> i figured comparing mencoder and ffmpeg might be ok [22:22] <durandal_1707> mencoder is dead [23:42] <saste> wm4, are you going to update your demuxing patches? [00:00] --- Thu Jul 18 2013

1 0

ffmpeg.log.20130717
by burek 18 Jul '13

18 Jul '13

[02:28] <johnsmith_> Hi all [02:29] <johnsmith_> does anyone here have any experiance encoding through x264 to get a successful call of avformat_write_header()..? [02:34] <bernie_> using -g 0 doesn't seem to supress I-frames [02:36] <axorb> hey guys, I fixed my original TS thing thanks to a patch merged in 6 days ago [02:37] <axorb> now I'm just missing some audio at the start of the clip for 0.2 seconds [02:38] <axorb> ffmpeg -ss 10 -i input.avi -ss 10 -f ssegment output%d.ts [02:38] <bernie_> I'm looking to force key frames with -force_key_frames, and not have any other key frames in the output file. I have not found the right way of doing that. I was experimenting with using -g as well, but that doesn't seem to help... [02:39] <bernie_> example: ffmpeg -y -re -i video_f06b84659c33731fdb34d1d43b0ddb14.mov -g 0 -dn -sn -map 0 -f Matroska -codec:v libvpx -quality good -cpu-used 0 -threads 4 -b:v 400K -maxrate 400K -force_key_frames 00:00:00.000,00:00:03.000,00:00:04.300,00:00:12.375,00:00:15.375,00:00:16.675,00:00:16.333,00:00:19.333,00:00:20.633,00:01:02.083,00:01:04.483,00:01:26.166,00:01:29.166,00:01:30.466 -pass 1 -an /dev/null && ffmpeg -y -re -i video_f06b84659c337 [02:40] <bernie_> sorry -- should have only shown a single pass of that... [02:45] <axorb> okay, let me rephrase the question [02:45] <axorb> ssegment is not encoding the first 0.05-0.2 seconds worth of audio for each part [02:53] <mark4o> bernie_: use a really big number after -g, and -scenecut 0 if you don't want keyframes on scene changes either [02:53] <mark4o> I mean -sc_threshold 0 [03:00] <klaxa> bernie_: -g 0 means GOP size 0 meaning you have only I-Frames [06:55] <ziadsa> hi guys, i was reading that we can now use ffmpeg with intel quick sync using libav? is that true? do i have to use h26 parameter instead if x264? [09:04] <natrixnatrix89> Would you guys recommend the ffmpeg integrated hls segmenter, or would it be better to search for something else? [09:05] <saste> natrixnatrix89, try it yourself, and see if it works for you [09:05] <saste> look in trac for the related open issues (keywords: hls, segment) [09:07] <natrixnatrix89> i tried it.. Opening the video on android caused the whole phone to crash [09:08] <natrixnatrix89> I'm just wondering if i've entered the wrong settings. or should I try a third party segmenter [09:19] <saste> natrixnatrix89, how are you opening the video? [09:19] <saste> does android support HLS? [09:19] <natrixnatrix89> yes it does.. [09:20] <saste> also what's the format of the video? [09:20] <natrixnatrix89> I was opening a html page with videojs player [09:20] <natrixnatrix89> video codec was h264.. and format obviously ts [09:20] <natrixnatrix89> format basically is hls. which creates the manifest file and segments the video [09:20] <natrixnatrix89> into ts files [09:21] <saste> natrixnatrix89, make sure you're streaming yuv420p [09:22] <natrixnatrix89> hmm. right.. i was streaming yuv422p [09:22] <saste> but you should check the player resources to find what it accepts [09:22] <saste> and a crash is always a (serious) bug [09:22] <natrixnatrix89> well yeah.. especially because the whole OS crashed.. and phone rebooted [09:22] <natrixnatrix89> when trying to play the video [09:23] <natrixnatrix89> what's the main difference between yuv420 and yuv422? [09:23] <saste> chroma sampling [09:23] <saste> *chroma subsampling [09:24] <saste> players are sometimes picky about that [09:24] <saste> and they will refuse to play a video for no apparent reason [09:24] <natrixnatrix89> ok [09:25] <natrixnatrix89> saste: do you have experience with ffserver? [09:25] <saste> no [09:25] <natrixnatrix89> ok.. thank's anyway [09:25] <natrixnatrix89> bye [09:25] <natrixnatrix89> hope changing pix fmt helps [10:14] <vincethechimp> Someone know any good article about ffmpeg filters? [10:17] <relaxed> did you read the man page? [10:20] <Mavrik> obviously not :P [11:03] <chris___> can anyone point me to an actual centos guide to install ffmpeg? i tried to follow http://ffmpeg.org/trac/ffmpeg/wiki/CentosCompilationGuide but end up in error with opus [11:21] <R0bin> ello [11:21] <R0bin> im on a amazon distro of linux [11:22] <R0bin> were can i get ffmpeg what kind of repos do they have? [11:22] <R0bin> (warning noob preseant) [11:23] <JEEB> "amazon distro of linux" [11:23] <JEEB> might want to check what distribution they use as the base if they have their own distro :P [11:23] <JEEB> also let me say this [11:23] <JEEB> if you're using Amazon's "cloud" [11:23] <JEEB> that will be costly as hell [11:23] <JEEB> if you will start encoding on it [11:24] <JEEB> low performance, high price, if you want the heavy stuff it will cost a fortune [11:40] <Gnintendo> Hello, I'm having trouble syncing up the video and audio in a stream I'm using ffmpeg for [11:40] <Gnintendo> the video is about 0.5 seconds before the audio [11:41] <Gnintendo> however, no matter what I do, I can't get -itsoffset or -ss to do anything [11:41] <Gnintendo> even when supplied longer times like 5s to verify, they appear to have no effect on the resultant video [11:41] <Gnintendo> I suspect this may be due to the fact that my output video is a stream, but this leaves me without an apparent way to delay the video [11:42] <Gnintendo> Here's what I'm currently doing: [11:42] <Gnintendo> ffmpeg -f v4l2 -input_format mjpeg -s "320x240" -r 10 -i /dev/video1 -f x11grab -s "1920x1080" -r 25 -itsoffset 3 -i :0.0+0,0 -filter_complex '[1]scale=1280:-1[vid];[0]scale=1280/6:-1[pip];[vid][pip]overlay=main_w-overlay_w:0' -f alsa -ac 2 -i hw:4,0 -vcodec libx264 -pix_fmt yuv420p -preset fast -vsync 1 -acodec libmp3lame -ab 196k -ar 44100 -threads 0 -re -f flv "rtmp://live.justin.tv/app/$(cat [11:42] <Gnintendo> ~/.twitch_key)" [11:42] <Gnintendo> however, the -itsoffset 3 (for 3 seconds), does not do anything [11:42] <Gnintendo> I need to delay the x11grab input by a period of time that is approximately 0.5 seconds, but that time will be refined as soon as I can find a way to delay it; can anybody here help me delay it? [14:59] <BoR0> any idea what might be causing this? I have this function seekpos() that does stream_seek, so if I do seekpos(50) it'll take me to the 50th second, but when I do seekpos(0) or seekpos(1) (up to 9-10) it takes me instantly to the 8th frame (audio is fine, i.e. starts from beginning, video is not) [15:00] <BoR0> but what's strange to me is that if I do seekpos(50) or any other number that is greater than 10 it works fine [15:01] <nerdshark> hey guys, can anyone clarify for me if the *priv* field in the avframe and avpacket structs are used by ffmpeg, or if they're meant to be used by end developers? [15:02] <nerdshark> (i'm unfortunately running on ffmpeg 1.1, for reference) [15:44] <mpfundstein> JEEB: thats not entirely true. you can create your own image, save it and than rent servers only if needed. I transcoded 30 000 vids this way for under 2k [15:44] <mpfundstein> JEEB: but you need an amazing big pipe there. [15:46] <JEEB> mpfundstein, I think it can still be costlier than dedicated servers from other places. Of course, if you need something you can take up/down quickly at will and as many instances as you want, then it can make sense -- even with the price tag [15:47] <JEEB> and no, I was not talking at all about their own encoding service. [15:47] <JEEB> Only VMs in their "cloud" [15:50] <mpfundstein> JEEB: no their own encoding service is crap [15:50] <mpfundstein> JEEB: they have a nice model where you rent instances for a certain price. lets say 0,19 cent / hour. if the demand rises the price rises [15:50] <mpfundstein> but as long as demand is low you pay nearly nothing [17:27] <Sythic> Hey, is FFProbe discussed here as well? [17:30] <stix_> hi. followed the CentosCompilationGuide but when compiling FFMPEG, I get "ERROR: opus not found" - I can locate opus in $HOME/ffmpeg_build/include, so not sure why it's complaining. [17:35] <mrAlmond> Hi everyone [17:52] <Sythic> I'm looking for an updated version of FFProbe for debian, or a way to compile it. My version was compiled in 2009 and is not compatible with a PHP library im using. Is there a newer version? [18:12] <teratorn> Sythic: it's not hard to compile [18:13] <Sythic> teratorn: Finding it is honestly the hard part. FFProbe doesnt seem to exist lol. Does it have its own repo or are there separate options to use when compiling or what? [18:13] <teratorn> Sythic: sudo apt-get build-dep ffmpeg; then ./configure and make the source [18:14] <teratorn> Sythic: it's part of the ffmpeg distribution [18:19] <johnsmith> hi all [18:19] <johnsmith> anyone know how to encode mpeg4 with av_write_header(..)? I get the "broken ffmpeg settings" error [18:39] <Sythic> teratorn: where exactly did apt put the source for me to build? or did i just install tools i need and still need to pull the code from git? [18:39] <teratorn> Sythic: you just installed the build dependencies of the ffmpeg package [18:39] <teratorn> Sythic: these will be more or less the same things that the source you need to get will want to link against [18:39] <Sythic> gotcha [18:40] <Sythic> teratorn: so do i just build as is or are there certain flags i need? I pretty much only need the meta-data and screenshot parts of this so transcoding stuff isnt really important [18:41] <teratorn> Sythic: defaults are probably fine [18:41] <teratorn> Sythic: thare are tons of --configure options [18:41] <Sythic> perfect, and branch releases are stable? [18:41] <Sythic> i noticed lol [18:41] <teratorn> Sythic: i'm not sure you might have to do --enable-gpl and such to get all codecs [18:42] <Sythic> teratorn: this is what i have now apparently http://codep.in/2q [18:42] <Sythic> this is the ffmpeg build from febuary i guess [18:43] <teratorn> do you need all those options? [18:44] <Carraway> question for you guys - is there a video or audio filter where I can put in a series of timecodes (00:00:05.000, 00:00:10.000, etc) and it will split the input at those markers [18:46] <Pet0r> I'm trying to compile ffmpeg with OpenCL support on Debian, I've compiled libx264 and configure for that outputs with opencl: yes, and ffmpeg compiles just fine with --enable-opencl, but when it comes to using it, I get "[libx264 @ 0x25b5920] OpenCL: not compiled with OpenCL support, disabling" [18:46] <Pet0r> any ideas? [18:52] <Mavrik> Pet0r, yeah. [18:53] <Mavrik> You compiled ffmpeg with OpenCL support. [18:53] <Mavrik> Not the x264 external library you're trying to use& x264 isn't part of ffmpeg so it doesn't give a darn about it's opencl support. :) [19:09] <Pet0r> Hey sorry, was in x264's chan also [19:09] <Pet0r> I feel like I'm missing something easy [19:09] <Pet0r> an option to pass to configure or something [19:10] <Pet0r> my ffmpeg check in configure is saying that opencl is disabled [19:10] <Pet0r> my understanding of the configure file is that it looks for OpenCL/cl.h or CL/cl.h [19:10] <Pet0r> this exists for me: /usr/include/CL/cl.h [19:10] <Pet0r> it's just obviously not seeing it? [20:16] <bernie_> hmm [20:36] <PasqualeIV> Hey fellas. I just preformed a fresh Ubuntu 12.04 server install, with instructions from UbuntuCompilationGuide. Verbatim with all options (minus those of xwin), ffmpeg ./configure responds ~"Unable to find opus". [20:37] <PasqualeIV> I confirmed that I had installed libopus. tried again. same issue. [20:38] <ezekiel> did you only confirm that you ran the opus installation _commands_? Or did you also confirm that the opus installation commands were _successful_ and that the opus libraries were _actually_ installed into a directory on your system? [20:39] <PasqualeIV> ezekiel, I confirm I ran the commands under a unprivileged user (as per the Comp guide). I'll be happy to rerun the opus install again and confirm each executed w/o error. standby. [20:44] <PasqualeIV> alright -- so same error upon ffmpeg ./configure "ERROR: opus not found" -- however I think I understand what you're saying -- as the opus install made the ld notices very clear. [20:46] <PasqualeIV> however, I would have expected the ffmpeg --extra-ldflags="-L$HOME/ffmpeg_build/lib" to compensate for this. [20:57] <PasqualeIV> ezekiel, the bottom line is; per the ubuntu comp guide; the install does not work under 12.04. [21:03] <PasqualeIV> adding more to the notes: opus-1.0.2.tar.gz does not include, what maybe an important file; autogen.sh [21:05] <PasqualeIV> installing opus per ?version=55 of the Comp guide, and using 'sudo make install' (opposed to checkinstall) gets opus running on the 12.04 system, and allows ffmpeg to be installed. [21:05] <mark4o> PasqualeIV: autogen.sh generates "configure". If you already have "configure" then you shouldn't need to use autogen.sh. [21:07] <mark4o> The tar.gz files should already have configure. You should only need autogen.sh if you are using git. [21:09] <PasqualeIV> mark4o, ok, so you think perhaps simply running "sudo make install" opposed to "make install" would have made the ffmpeg install successful? [21:09] <mark4o> The old version of the comp guide installed system-wide, which didn't work for people that did not have root access. The new guide installs in the home directory and does not require root/sudo. [21:10] <PasqualeIV> gotcha.. then without a doubt, the new guide did not work for a non root user in 12.04 [21:12] <PasqualeIV> the new guide does work when libopus is left out of the ffmpeg compile, but will fail if included. [21:13] <mark4o> I think it is for 13.04 but it is a wiki so if you figure out what is wrong with opus for 12.04 then please feel free to add a note for anyone else using 12.04 [21:14] <mark4o> btw there is now an opus 1.0.3 release, not sure if it would change anything for you [21:17] <PasqualeIV> I think the guide is attempting to support all "supported releases". I'll try 1.0.3 right now -- great suggestion. Once I find "the" answer, I'll be happy to update the wiki. [21:24] <Fjorgynn> when [21:36] <PasqualeIV> bahh: ERROR: opus not found -- even with 1.0.3. [21:47] <mark4o> PasqualeIV: what does pkg-config --libs opus say? [21:48] <PasqualeIV> No package opus found. [21:49] <mark4o> is PKG_CONFIG_PATH set? [21:50] <PasqualeIV> yes [21:50] <mark4o> It looks like the guide says to set it but it doesn't export it; can you try "export PKG_CONFIG_PATH" and then pkg-config --libs opus [21:51] <PasqualeIV> booyah! -L/home/ubuntu/ffmpeg_build/lib -lopus [21:51] <PasqualeIV> configure success! [21:51] <mark4o> cool [21:52] <PasqualeIV> you want to update the wiki? or should I? [21:52] <braincracker> ffmpeg! [21:53] <mark4o> was it just adding an "export PKG_CONFIG_PATH" that did it? [21:54] <PasqualeIV> once I did that -- the ffmpeg ./configure command worked. [21:54] <mark4o> ok, did the rest of the build work? [21:55] <PasqualeIV> I confirmed this by performing a fresh install of 12.04, following the guide from top to bottom, failing at the configure, did the export as you suggested, and make is running now. give it a few mins to confirm the build was a full success. [21:57] <mark4o> Ok, I can add that to the wiki. [21:57] <PasqualeIV> make success, make install success [21:57] <mark4o> Are you using opus-1.0.3 or 1.0.2? If 1.0.3 works for you then I'll update that also [21:59] <PasqualeIV> 1.0.3 [22:00] <PasqualeIV> it appears I have full success: built on Jul 17 2013 15:56:48 [22:00] <mark4o> very good; I've updated the wiki, thanks [22:01] <PasqualeIV> thanks for your help mark4o! [23:08] <johnsmith> Hi all, anyone have any experiance on how to convert mp3 files into mp4 (no video only audio) in aac? [23:08] <johnsmith> I am getting "aac bitstream error" [23:15] <ziadsa> guys can anyone guide me how to use intel quick sync. [23:15] <Fjorgynn> quick sync what? [23:19] <saste> johnsmith, command? [23:21] <johnsmith> I am working with libav files [23:21] <johnsmith> but if you know the command presets, information like that would be grateful [23:44] <LunixA380> Hello all [23:45] <LunixA380> I try to do spatial resampling with VP8/libvpx-vp8. How can I do it? [23:50] <LunixA380> according to https://sites.google.com/a/webmproject.org/wiki/ffmpeg, it's -rc_resize_allowed [23:50] <LunixA380> but -rc_resize_allowed doesn't work. [23:51] <LunixA380> it says that the option doesn't exists [23:53] <relaxed> LunixA380: look at $(ffmpeg -h encoder=libvpx | less) [23:54] <mark4o> johnsmith: ffmpeg -i in.mp3 -acodec libfdk_aac out.mp4 [23:54] <mark4o> johnsmith: https://ffmpeg.org/trac/ffmpeg/wiki/AACEncodingGuide [23:55] <LunixA380> The option isn't there, it's true relaxed [23:56] <saste> -metadata title="Decode my Heart (Let's Mux)" => fun [23:58] <relaxed> don't forget libass support [23:58] <johnsmith> mark4o, thanks [23:59] <LunixA380> libass ? Hm. I'll search about it [23:59] <relaxed> LunixA380: that wasn't directed at you [23:59] <LunixA380> Ops :( [23:59] <LunixA380> Thanks for your help though :) [00:00] --- Thu Jul 18 2013

1 0

ffmpeg-devel.log.20130716
by burek 17 Jul '13

17 Jul '13

[00:08] <cone-834> ffmpeg.git 03Martin Storsjö 07master:5b4eb243bce1: mov: Seek back if overreading an individual atom [00:08] <cone-834> ffmpeg.git 03Michael Niedermayer 07master:38b701a34924: Merge commit '5b4eb243bce10a3e8345401a353749e0414c54ca' [00:40] <cone-834> ffmpeg.git 03Martin Storsjö 07master:31931520df35: mov: Do not allow updating the time scale after it has been set [00:40] <cone-834> ffmpeg.git 03Michael Niedermayer 07master:13eed267f8c1: Merge commit '31931520df35a6f9606fe8293c8a39e2d1fabedf' [00:56] <cone-834> ffmpeg.git 03Martin Storsjö 07master:e740929a071a: lavf: Make sure avg_frame_rate can be calculated without integer overflow [00:56] <cone-834> ffmpeg.git 03Michael Niedermayer 07master:4f1d3e02122e: Merge remote-tracking branch 'qatar/master' [02:51] <cone-834> ffmpeg.git 03Michael Niedermayer 07master:37f4aa133d0a: swscale: move default colorspace setup to sws_init_filter() [03:20] <cone-834> ffmpeg.git 03Alexis Ballier 07master:8d9c1b325ed2: mastroka audio muxer: Set long_name to Matroska Audio so that it differs from the long_name of matroska video. [08:41] <noob384> hello, what are the params of ffv1? [08:46] <noob384> yesterday i converted avi uncompressed to avi ffv1 with ffmpeg and with some another gui converter (that use ffmpeg) [08:50] <noob384> i got the file ~6.5 GB (with gui program) instead ~8.5 GB (codec was ffv1 in both cases) [08:51] <noob384> is anyone here& [08:51] <noob384> ? [08:52] <noob384> or this thread is dead as #ffmpeg? [09:02] <noob384> only silent [09:03] <noob384> tshshsh..... [09:03] <noob384> hey, anyone is here? [09:37] <michaelni> noob384, this is a development channel, your question doesnt sound related to development [09:39] <noob384> @michaelni yes, but no anyone in #ffmpeg channel [11:49] <mateo`> michaelni: if you have some time, it would great if you can review my movenc patch :) [11:58] <cone-132> ffmpeg.git 03Michael Niedermayer 07release/1.0:03983c3650b5: avformat/matroskadec: Detect conflicting sample rate/default_duration [11:58] <cone-132> ffmpeg.git 03Michael Niedermayer 07release/1.1:ae72abf6521c: avformat/matroskadec: Detect conflicting sample rate/default_duration [11:58] <cone-132> ffmpeg.git 03Michael Niedermayer 07release/1.2:06190d75d09a: avformat/matroskadec: Detect conflicting sample rate/default_duration [11:58] <cone-132> ffmpeg.git 03Michael Niedermayer 07release/2.0:bc44d06c3de8: avformat/matroskadec: Detect conflicting sample rate/default_duration [13:29] <saste> what happens if i send empty packets to the encoder? [13:35] <TZer0> Quick question: what should I do if I want to modify cmdutils.c so that ffmpeg can accept additional (with information I'd like to use) files? [13:36] <BoR0> how good is AVSEEK_FLAG_FRAME, will it allow me to step to any frame? (something like QuickTime does) [13:37] <TZer0> if anyone wondered, yes, what I'm doing will be available and hopefully pushed back to the main ffmpeg-repo if successful. [13:56] <Compn> additional files ? [13:56] <Compn> what do you mean ? [13:56] <Compn> TZer0 [13:57] <Compn> ffmpeg has concat protocol for multi inputs ? [13:57] <Compn> or you mean metadata ? [14:00] <TZer0> Compn: well, it is sort of metadata [14:00] <TZer0> I'm looking to modify some stuff related to motion vector estimation [14:01] <TZer0> and I'd like to provide an additional file with information related to this [14:02] <TZer0> I must also admit that I am very fresh when it comes to ffmpeg. [14:02] <TZer0> just a heads up. [14:03] <TZer0> Oh, and thanks for any help I get. [14:34] <Compn> TZer0 : oh, well no idea :) [14:34] <Compn> TZer0 : try on libav-user list as well [14:40] <TZer0> Compn: well, if you know anything about command line parameter handling, that would always help [15:21] <TZer0> Compn: also, should this be in the libav-user list? I'm making modifications of ffmpeg, not using it thorugh other means. [15:40] <Compn> TZer0 : yeah i realized i was wrong, this sounds like the right place [15:41] <Compn> TZer0 : best thing is to make a trac issue, with a sample file that shows what you want to do, and what a finished file should look like [15:41] <Compn> that way other devs who can help and know where all the option parsing should go can figure out the right api path [15:41] <Compn> i dont program so i dont know any of this stuff [15:46] <saste> TZer0, what information? [19:25] <saste> michaelni, why don't we have a packet filtering layer? [19:26] <saste> bitstream filtering allows to only change data, but doesn't allow to modify/access the other packet properties, thus it is limited [19:56] <TZer0> saste: I've got some motion vectors which I want to feed FFmpeg [19:57] <saste> TZer0, what's the point? [19:57] <TZer0> saste: avoid having to calculate SEDs - speed up that part of the encoding. [19:58] <saste> TZer0, is that possible? are you tweaking the ME routines? [19:58] <TZer0> first of all, I'm kind of fresh to this. Just a warning - ME? [19:58] <saste> motion estimation [19:58] <TZer0> ah, yes. [19:59] <saste> motion vectors are usually computed by this component [19:59] <TZer0> I've got some pre-calculated ones. [19:59] <saste> this is internal API [19:59] <saste> used by encoders [19:59] <Skyler_> which encoder are you trying to feed with MVs? [19:59] <TZer0> to be honest: not sure yet. [19:59] <TZer0> I need to figure out this bit by bit. [20:00] <Skyler_> how things will work is going to depend a lot on which encoder it is, most likely [20:00] <TZer0> Hmm. [20:00] <TZer0> Most likely.. h264 [20:06] <TZer0> saste: where should I in your opinion start [20:07] <saste> TZer0, if i understand it correctly, you want to avoid the encoder to compute MVs, by providing them yourself [20:07] <TZer0> Yeah. [20:07] <saste> the problem is that this scenario is not currently supported, since the encoder is not able to "read" them, but always compute them [20:08] <saste> especially if you consider external encoders (e.g. libx264) for which such scenario was never contemplated [20:08] <TZer0> Well, I'm looking to add a seperate mode which simply reads them. [20:08] <Skyler_> for H.264, you want to look at libx264 [20:09] <Skyler_> however, keep in mind: [20:09] <saste> so in the best case you should modify libavcodec, but also the other external encoders you're interested in [20:09] <Skyler_> 1) at slower speeds, x264 will do a way better job than your motion search, so it's not worth it [20:09] <Skyler_> 2) at faster speeds, the motion search is not a very large portion of the encoding time [20:09] <Skyler_> so it is not as useful as people tend to thik it is [20:09] <Skyler_> doubly so if it turns out their motion vectors aren't very good, or aren't qpel [20:10] <saste> TZer0, what's the objective of such endeavor? optimizations? [20:10] <TZer0> Yeah [20:10] <Skyler_> furthermore, if you're using B-frames in x264, it's impossible to meaningfully use external motion vectors [20:10] <Skyler_> unless you force the B-frame pattern too which is particularly bad [20:18] <saste> why all this flaming in ffmpeg-user? people have nothing better to do than arguing about mail headers? [20:19] <TZer0> Aren't those super-important? :P [20:19] <Skyler_> so like if you're actually interested I can try to help with an implementation but there's a lot of icky gotchas [20:20] <TZer0> Skyler_: hmm, well, I'm not working on it right now. [21:26] <cone-237> ffmpeg.git 03Paul B Mahol 07master:5ceffb1bf648: matroskaenc: use ffio_fill() [22:10] <troker> Hi all, I'm looking to do some security research with ffmpeg as a testbed. Im interested in this vuln (http://people.canonical.com/~ubuntu-security/cve/2010/CVE-2010-3429.html) - can anyone help me find the git commit directly before that CVE was patched? Thanks much. [22:13] <wm4> you can find it by searching for the cve number in git log [22:14] <wm4> commit 6e73cef6 fixes it [22:15] <troker> wm4 - thanks so much for that [22:15] <troker> wm4 a bit new to git - so git checkout ba87f0801d77c21eb1e4891ca1f846500bbb0939^ (the commit before the fix) should get me what I want, yes? [22:17] <wm4> I don't really see how ba8... is related [22:46] <troker> wm4 Im looking for the commit directly before that CVE was fixed, is ba8.. the correct one? [22:46] <wm4> seems unrelated... [22:53] <durandal_1707> troker: how you came out with ba8? [22:54] <troker> I realize now that thats entirely incorrect ;P [22:54] <troker> Im new to git heh - so it looks like a carrot will give me the parent? So I want 6e73cef6^ [22:55] <durandal_1707> that is commit that fixes CVE [22:57] <troker> @durandal_1707 6e73cef6 fixes it or 6e73cef6^ does? [22:57] <troker> it looks like its 6e73cef6 [00:00] --- Wed Jul 17 2013

1 0

ffmpeg.log.20130716
by burek 17 Jul '13

17 Jul '13

[00:29] <Guest80252> I am facing issue during seeking for flv videos, i can see range request uses "ns" parameter and seeking fails any suggestions [00:33] <Guest80252> do we need special encoding for "ns" parameter? [00:40] <Guest80252> does anyone help about FLV Seeking issue [00:41] <klaxa> can you elaborate the problem further? are you using the libav* api or the command line ffmpeg tool? [00:41] <Guest80252> API [00:42] <klaxa> mh no idea, never used the api, sorry :( [00:42] <Guest80252> normally The query parameter "ns" is used to pass a byte offset which contain a byte offset from the start of the file that the video should seek to [00:43] <Guest80252> For FLV looks "from" and "to" offset s are encoded in the ns query parameter? [00:43] <Guest80252> i am really wondering how to decode that? [06:03] <Zeranoe> Is there any way to get a list of decoders and encoders that are LGPL only and cannot be included in a GPL build? I'm mainly wondering if MPNG or CSCD are LGPL only [06:03] <Zeranoe> GPL* only [06:10] <Plorkyeran> libcdio libutvideo libvidstab libx264 libxavs libxvid x11grab [06:10] <Plorkyeran> are the non-proprietary, non-avfilter things disabled by the configure script without --enable-gpl [08:15] <kamalbanga> My ffmpeg code succesfully compiles into object file without any warnings but when I link it, it gives following linker error: [08:17] <kamalbanga> http://pastebin.com/1hME8mdB [08:17] <kamalbanga> can i infer anything? [08:54] <noob384> hello, what are the params of ffv1? [08:54] <noob384> yesterday i converted avi uncompressed to avi ffv1 with ffmpeg and with some another gui converter (that use ffmpeg) [08:55] <noob384> i got the file ~6.5 GB (with gui program) instead ~8.5 GB (codec was ffv1 in both cases) [08:59] <noob384> so, gui had some -keys in command that i do not use (in console ffmpeg myself), what are they may be? [09:48] <michaelni> noob384, you can try different context sizes, vlc vs ac code, also -debug 1 with the ffv1 mght show some info about how its encoded [11:07] <misterno> hey i remember playing with video interpolation some time ago making videos run at 60 or even 120 fps smoothing them out [11:07] <misterno> i was wondering has this tech been improved over the years and what would be best tools these days to achieve it [11:08] <misterno> is it still avisynth + mvtools, or is it smooth video project? [11:36] <BoR0> how good is AVSEEK_FLAG_FRAME, will it allow me to step to any frame? (something like QuickTime does) [13:35] <Mercy404> Hello all, I was wondering if you could give me a bit of advice. I'm trying to stitch together a bunch of JPEGs into an mp4 using the following command: ffmpeg -r 10 -i "img (%d).jpg" -b:v 1500k "timelapse.mp4" [13:36] <Mercy404> This works for smaller sets of images, but with a 1.5gb set, malloc is failing. I have a machine with 16gb of ram, and I've tried rebooting to free up large contiguous chunks of memory, but I feel like I must be doing something rather simple wrong. [13:40] <Mercy404> switching to a 64bit build seems to have fixed the issue. It wasn't my first guess because malloc was failing to allocate just 716mb. [14:13] <misterno> is anyone into video interpolation making movies smoot at 60 fps? [14:13] <misterno> i was just wondering what would be the best tools these days to achieve that [16:00] <jdiez> hey guys, trying to compile ffmpeg after having compiled x264, I get this error: [16:00] <jdiez> ffmpeg/libavcodec/libx264.c:552: undefined reference to `x264_encoder_open_135' [16:00] <jdiez> x264 is the proper version, too: [16:00] <jdiez> # x264 --version [16:00] <jdiez> x264 0.135.2 f0c1c53 [16:00] <jdiez> help? [16:01] <saste> jdiez, make sure you're linking against the correct version [16:01] <BoR0> how good is AVSEEK_FLAG_FRAME, will it allow me to step to any frame? (something like QuickTime does) [16:02] <jdiez> saste: there's no other version in my system [16:02] <jdiez> # ldconfig -p | grep x264 [16:02] <jdiez> libx264.so.123 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libx264.so.123 [16:07] <Ph0bus> can i remove or add subs to mkv files with out reencoding? [16:07] <jdiez> saste: any other ideas? [16:11] <saste> jdiez, .so.123 => is it the right version? [16:12] <jdiez> saste: yep [16:12] <saste> shouldn't it be 135? [16:12] <jdiez> hmm.. [16:12] <saste> jdiez, it's the same here... [16:13] <saste> try to make a distclean and rebuild, it may help [16:13] <jdiez> you have 123 too? [16:13] <saste> yes [16:13] <jdiez> I did rebuild a couple of times... :/ [16:13] <saste> is it a configure or a linking problem? [16:14] <jdiez> oh, linking problem [16:14] <jdiez> it builds fine [16:14] <Ph0bus> can i remove or add subs to mkv files with out re-encoding? [16:22] <klaxa> Ph0bus: yes, see -c copy and -map options [16:23] <klaxa> to remove subs, you would do: ffmpeg -i input.mkv -map -0:s -c copy output.mkv [16:23] <klaxa> in general, you could have a look at mkvtoolnix especially the mkvmerge gui [16:23] <klaxa> the binary is called mmg, confused me for the longest time now :| [16:24] <Ph0bus> ok thanks i used mkvtoolnix bur it demuxes thats no reencoding? [16:24] <klaxa> jdiez: where is your libx264.so.135? [16:24] <klaxa> you wanted no reencoding [16:24] <Ph0bus> the yes [16:24] <Ph0bus> yes just delte the subs or add with our toutching the other files contained [16:25] <jdiez> klaxa: /usr/lib/x86_64_gnu_linux/libx264.so.135 [16:25] <klaxa> run ldconfig as root maybe? [16:25] <jdiez> did that [16:26] <klaxa> weird, it should put libx264.so.135 in the cache then [16:30] <jdiez> in the cache? [16:31] <klaxa> ldconfig cache [16:35] <klaxa> jdiez: so you said it compiled fine, right? in that case can you run ldd ffmpeg? [16:35] <jdiez> klaxa: it compiled fine, it didn't link [16:36] <klaxa> ah derp [16:37] <klaxa> well quick hack would be to just overwrite the libx264.so symlink [16:38] <klaxa> but meh [16:38] <jdiez> to what? [16:38] <jdiez> it's finding the file to link with, it's just not finding some symbols [16:39] <klaxa> it's finding the wrong version of the file [16:39] <jdiez> fun fact: [16:39] <jdiez> # nm /usr/local/lib/libx264.a |grep encoder_penen [16:39] <jdiez> 0000000000008020 T x264_encoder_open_135 [16:39] <klaxa> it's trying to link against libx264.so.123 apparently [16:39] <klaxa> no? [16:39] <jdiez> possibly [16:39] <jdiez> should I tell it to link statically? [16:40] <klaxa> if you have a static library that would make sense i guess? [16:40] <jdiez> I have both [16:40] <klaxa> and you have libx264.so.135 in your ldconfig path? [16:41] <jdiez> yep [16:41] <jdiez> # ldconfig -p | grep x264 [16:41] <jdiez> libx264.so.123 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libx264.so.123 [16:41] <klaxa> but that's the WRONG version! [16:41] <jdiez> is it [16:41] <klaxa> yes! [16:41] <klaxa> it's libx264.so.123, you need libx264.so.135! [16:42] <jdiez> klaxa: I'm afraid I don't have .135? [16:42] <Ph0bus> what are the most compatible mkv setings? [16:42] <klaxa> you compiled x264, right? did you configure it with "--enable-shared" ? [16:43] <jdiez> yes [16:43] <klaxa> did you run make install afterwards? [16:43] <jdiez> yes [16:43] <jdiez> and ldconfig [16:43] <klaxa> it should be in /usr/local/lib then [16:43] <Ph0bus> i have a bray player that supports mkv but it strugles with some files burned to dcd [16:43] <klaxa> see if it is there [16:43] <Ph0bus> dvd [16:43] <klaxa> unless you also configured it with "--prefix=/some/other/path" [16:43] <jdiez> klaxa: I have libx264.a in /usr/local/lib [16:43] <klaxa> that's a static library, not a shared one [16:43] <jdiez> correct [16:44] <jdiez> I don't have anything else x264 in there [16:44] <klaxa> yes, well then you have not built x264 version 135 with shared libraries [16:44] <Ph0bus> is there some standard seting for mkv encoding? [16:44] <jdiez> klaxa: will try to rebuild [16:44] <klaxa> Ph0bus: mkv is not a codec [16:44] <klaxa> jdiez, if you look into your x264 source directory, you should find libx264.so.135 [16:44] <jdiez> klaxa: damn right! [16:45] <Ph0bus> h264 [16:45] <jdiez> gee, I guess I need to copy that to /usr/lib and symlink libx264.so? [16:45] <klaxa> there's multiple ways to go about this now [16:46] <klaxa> either that, or you add /path/to/x264 to your /etc/ld.conf.d/my_additional_config.conf [16:46] <klaxa> but before that [16:46] <jdiez> meh, going to go the hacky route [16:46] <klaxa> see if you really ran make install and if it throws some errors [16:46] <jdiez> okay [16:46] <jdiez> I'm rebuilding it from scractch just to be sure [16:47] <jdiez> scratch* [16:47] <klaxa> i doubt that that's gonna help, but okay :I [16:47] <jdiez> can't hurt either [16:47] <jdiez> yeah, .135 is now in /usr/local/lib [16:48] <klaxa> try linking ffmpeg again [16:48] <jdiez> yep [16:49] <jdiez> worked [16:49] <jdiez> thanks for your help, klaxa [16:49] <klaxa> :) [16:50] <jdiez> excellent [16:51] <klaxa> Ph0bus: http://ffmpeg.org/trac/ffmpeg/wiki/x264EncodingGuide [16:51] <Ph0bus> ty [16:52] <Ph0bus> but are there something like a standar fit all setings? [16:52] <klaxa> not really [16:53] <Ph0bus> there should be a baseline compatibality seting [16:53] <klaxa> there are default values [16:53] <klaxa> you can just do: ffmpeg -i input.mkv -c:v libx264 output.mkv [16:53] <klaxa> that will encode with libx264 and apply default values set in ffmpeg, i don't know in how far that is compatible [16:53] <Ph0bus> but will it work on most hardware ? [16:53] <klaxa> you'll have to do some testing to find out what suits your player [16:54] <klaxa> probably [16:54] <Ph0bus> ok ty anyway [22:10] <troker> Hi all, I'm looking to do some security research with ffmpeg as a testbed. Im interested in this vuln (http://people.canonical.com/~ubuntu-security/cve/2010/CVE-2010-3429.html) - can anyone help me find the git commit directly before that CVE was patched? Thanks much. [22:10] <Fjorgynn> hi me [22:10] <Fjorgynn> dunno [22:17] <llogan> troker: git log 6e73cef [22:17] <troker> llogan thanks [22:18] <llogan> i got that hash from your link [22:18] <llogan> then look below the log entry for that hash for earlier ones [22:34] <Ericq> do we have any mechanism to get Byte savings during Adaptive streaming contents Optimization [22:36] <axorb> hey guys, I'm manually segmenting a video to TS while trying to keep the timestamps intact [22:36] <axorb> ffmpeg -ss 20 -i file.avi -t 10 ... output.ts [22:37] <axorb> it works but the video starts at 0, I want to shift the PTS to start at 20 so it works with HLS [22:37] <axorb> copyts and setpts work but they don't adjust the start time [22:38] <axorb> the segmenter works correctly (part2.ts starts at 20 seconds) and I need to emulate it but for just one part (can't afford to try and do all at once) [22:40] <Ericq> can we get any savings out of AS contents [22:51] <Ericq> does anyone has insight about optimiziing AS contents [22:59] <axorb> also, I noticed the ts muxer always sets the start time to 1.4 [23:29] <axorb> I can use the segmenter with only one segment and it correctly makes parts that start at 0 (instead of 1.4) [23:29] <axorb> now I just need to get the start PTS to be something like 20 instead of always 0 [00:00] --- Wed Jul 17 2013

1 0