March 2012 - Ffmpeg-devel-irc

ffmpeg-devel.log.20120330
by burek 31 Mar '12

31 Mar '12

[00:23] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r30ced7e69f 10ffmpeg/libavformat/utils.c: [00:23] <CIA-17> ffmpeg: lavf: print ts values in pts < dts error message [00:23] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [00:23] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rd61ef05bc5 10ffmpeg/ffmpeg.c: [00:23] <CIA-17> ffmpeg: ffmpeg: check pts like dts for being in the past. [00:23] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [00:23] <CIA-17> ffmpeg: 03Brian Olson 07master * r0be130e37b 10ffmpeg/ (configure doc/ffmpeg.texi libavformat/img2dec.c): [00:23] <CIA-17> ffmpeg: img2: glob matching for image series [00:23] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [00:24] <CIA-17> ffmpeg: 03Alexander Strasser 07master * r3e1ff8eb27 10ffmpeg/libavformat/img2dec.c: (log message trimmed) [00:24] <CIA-17> ffmpeg: img2dec: Recognize glob meta chars only if prefixed by % [00:24] <CIA-17> ffmpeg: This changes globbing support to only be used if the character [00:24] <CIA-17> ffmpeg: contains at least one glob meta character that is preceded by [00:24] <CIA-17> ffmpeg: an unescaped %. To escape a literal % one would use %% which is [00:24] <CIA-17> ffmpeg: identical to the way to match a % with image2 sequence generation [00:24] <CIA-17> ffmpeg: feature. [00:36] <michaelni> ubitux, read or write ? [00:37] <michaelni> for read theres an example in mov.c [05:21] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rd3d5e84f33 10ffmpeg/libavformat/img2dec.c: [05:21] <CIA-17> ffmpeg: img2dec: remove GLOB_TILDE support. [05:21] <CIA-17> ffmpeg: tilde expansion should/can be done by the shell [05:21] <CIA-17> ffmpeg: Reviewed-by: Alexander Strasser <eclipse7(a)gmx.net> [05:21] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [06:23] <CIA-17> ffmpeg: 03Kostya Shishkov 07master * rf704eb612b 10ffmpeg/libavformat/id3v2.c: id3v2: add another mimetype for JPEG image [06:23] <CIA-17> ffmpeg: 03Mashiat Sarker Shakkhar 07master * r28e8c4d59a 10ffmpeg/libavcodec/wmalosslessdec.c: [06:23] <CIA-17> ffmpeg: WMAL: Remove inaccurate and unnecessary doxy [06:23] <CIA-17> ffmpeg: A call to decode_packet() does not always decode a complete WMA packet. [06:23] <CIA-17> ffmpeg: Moreover, this is not the correct place to document calls that are part [06:23] <CIA-17> ffmpeg: of the public API. [06:23] <CIA-17> ffmpeg: Signed-off-by: Diego Biurrun <diego(a)biurrun.de> [06:23] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rbd508d435b 10ffmpeg/libavcodec/truemotion2.c: [06:23] <CIA-17> ffmpeg: truemotion2: convert packet header reading to bytestream2. [06:23] <CIA-17> ffmpeg: Also use correct buffer sizes in calls to tm2_read_stream(). Together, [06:23] <CIA-17> ffmpeg: this prevents overreads. [06:23] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [06:23] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [06:23] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r2b7c0c9fe1 10ffmpeg/: (log message trimmed) [06:24] <CIA-17> ffmpeg: id3v2: add another mimetype for JPEG image [06:24] <CIA-17> ffmpeg: lzw: prevent buffer overreads. [06:24] <CIA-17> ffmpeg: WMAL: Remove inaccurate and unnecessary doxy [06:24] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rddcf67c8a5 10ffmpeg/libavcodec/lzw.c: [06:24] <CIA-17> ffmpeg: lzw: prevent buffer overreads. [06:24] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [06:24] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [06:24] <CIA-17> ffmpeg: 03Kostya Shishkov 07master * ra05c41acd1 10ffmpeg/libavcodec/mpegaudiodec.c: [06:24] <CIA-17> ffmpeg: mp3dec: perform I/S and M/S only when frame mode is joint stereo. [06:24] <CIA-17> ffmpeg: Looks like some LAME versions produce dual stereo mode MP3s with [06:24] <CIA-17> ffmpeg: flags for intensity and middle stereo set. In this mode those flags [06:24] <CIA-17> ffmpeg: should be ignored like the reference decoder and derived ones do. [06:24] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r63a1b481f6 10ffmpeg/libavcodec/h264_cabac.c: h264: fix cabac-on-stack after safe cabac reader. [07:23] <ubitux> michaelni: for read; i missed the one in mov.c, thanks :) [07:52] <ubitux> ok well, it does the uncompress before, not transparently in the avio, too bad [11:58] <ubitux> michaelni: heh, nice :) [11:59] <ubitux> michaelni: any html output somewhere? [12:16] <michaelni> ubitux, ill generate html/pdf/... once iam closer to done [12:17] <michaelni> makes not so much sense in the middle while iam working on it [12:36] <TimNich> michaelni: meantime what is a good reader for that format? [12:38] <michaelni> TimNich, lyx [12:39] <michaelni> for actual reading you probably want to convert it to pdf or something probably, it should look better then [12:40] <michaelni> lyx File->Export->pdf will do that [12:44] <TimNich> Ahh, I hadn't seen the export options, as my main machine died while trying to view it... [12:45] <TimNich> Mind you lxy is no lightweight. the download, plus deps was 900M! [12:45] <michaelni> hmm [12:45] <michaelni> thats a bit bigger than what i had expected [12:46] <TimNich> lxy itself was 9M but texive etc were enormous [12:47] <av500> or just use plain LaTex and dvi2pdf [14:47] <j-b> [mpeg4 @ 0x7f9580cbac20] looks like this file was encoded with (divx4/(old)xvid/opendivx) -> forcing low_delay flag [14:47] <j-b> [mpeg4 @ 0x7f9580cbac20] Width/height/bit depth/chroma idc changing with threads is not implemented. Update your FFmpeg version to the newest one from Git. If the problem still occurs, it means that your file has a feature which has not been implemented. [14:47] <j-b> what does this mean? [14:48] <av500> you cannot read? [14:48] <j-b> I can. [14:49] <j-b> never had that in the past, though [14:49] <j-b> so I am WTF? [14:49] <av500> anyway, thats a question for #ffmpeg-users [14:49] <av500> :) [14:49] <av500> happy birthday too [14:50] <j-b> well, #ffmpeg-devel is not really high traffic though... [14:50] <j-b> I could get naked here and noone would see it [14:51] <iive> j-b: imho, it is something with the multithreaded decoding. [14:52] <j-b> seems so. [14:52] <iive> probably it thinks 2 bitstreams are concatenated.. if this is even possible with mpeg4. [14:57] <ubitux> is it me or the zlib api is braindead? [15:02] <ubitux> it seems to randomly fail if you don't allocate enough for the output for a given input [15:34] <iive> hum? [15:35] <iive> it is supposed to work, even if you output one byte at a call. [15:36] <ubitux> well i have a stream, i read 1024 for input, prepare 1024 for output, call inflate DATA_ERR [15:36] <ubitux> it works only if output is sth like 65536 [15:37] <ubitux> i just don't get it [15:39] <ubitux> the input stream is not "segmented", so not in a final state, maybe it's one of the reason [15:39] <ubitux> (1024 is arbitrary) [15:40] <iive> shouldn't matter. [15:45] <j-b> who understands simple questions about openjpeg in libavcodec code? [15:47] <kierank> durandal when he is around [15:47] <j-b> ok [16:09] <michaelni> j-b, if you have a mpeg4 file that worked before and doesnt work anymore, then iam interrested (that is both old and new with same number of threads) [16:11] <j-b> michaelni: seems the issue is only with threads > 1 [16:11] <j-b> we did not use threads > 1 before. [16:26] <iive> j-b: i think that if the file is not broken, then we need to have a sample :) [16:29] <j-b> I cannot, you need to sign a VideoLAN NDA... [16:29] <j-b> j/k, here it goes: [16:29] <j-b> ftp://ftp.axon.tv/TracsSampleClips/MPEG4%20small%20512kbps.mpg [16:31] <iive> bbml [16:34] <Compn> mpeg4 in mpeg [16:34] <Compn> yuckkkkkkkkkkkk [16:41] <michaelni> j-b, i dont get a "Width/height/bit depth/chroma idc changing with thread" here with that file [16:43] <j-b> with threads > 1 ? [16:44] <michaelni> -threads 8 [16:48] <j-b> weird++ [16:50] <michaelni> can you reproduce it with ffmpeg (instead of vlc) ? [16:53] <j-b> nope [16:54] <j-b> well, tested with ffplay [16:54] <michaelni> ubitux, ive put a pdf of the latest ffv1 spec up (http://ffmpeg.org/~michael/ffv1-draft/) if you still want to see it [16:56] <ubitux> nice, thanks :) [16:56] <michaelni> ubitux, html up too [16:56] <j-b> EMU_EDGE + threads issue? weird [16:59] <michaelni> cant see any issue with ffplay either [19:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rec0965be36 10ffmpeg/libavcodec/cavsdec.c: [19:39] <CIA-17> ffmpeg: cavsdec: ensure the tables have been allocated before using them [19:39] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [19:39] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rcb9fe2745b 10ffmpeg/libavcodec/j2kdec.c: [20:39] <CIA-17> ffmpeg: j2kdec: fix format strings [20:39] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r90f9b06386 10ffmpeg/libavcodec/dvbsubdec.c: [20:39] <CIA-17> ffmpeg: dvbsubdec: Fix "set but not used" warning [20:39] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra76a2ffe9d 10ffmpeg/libavformat/aviobuf.c: [20:39] <CIA-17> ffmpeg: libavformat: Fix several "incompatible pointer type" warnings. [20:39] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r03757d488f 10ffmpeg/libavcodec/j2kenc.c: [20:39] <CIA-17> ffmpeg: j2kenc: Fix "function declaration isnt a prototype" warning [20:39] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:39] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8307025467 10ffmpeg/libavcodec/mjpegdec.c: [22:06] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r280beebd39 10ffmpeg/libavcodec/cavs.c: [22:06] <CIA-17> ffmpeg: cavsdec: initialize all tables to zeros. [22:06] <CIA-17> ffmpeg: This ensures that they dont contain invalid values. [22:06] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [22:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:06] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r96df29c318 10ffmpeg/libavcodec/svq3.c: [22:06] <CIA-17> ffmpeg: svq3: Fix pointer type mismatch warning. [22:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:33] <CIA-17> ffmpeg: 03Martin Storsjö 07master * r9bd41bb191 10ffmpeg/configure: (log message trimmed) [22:33] <CIA-17> ffmpeg: configure: Check for a different SDL function [22:33] <CIA-17> ffmpeg: This one is available both in SDL 1.2 and in 1.3 (which is the current [22:33] <CIA-17> ffmpeg: version available e.g. in macports), while 1.3 doesn't contain [22:33] <CIA-17> ffmpeg: SDL_Linked_Version(). [22:33] <CIA-17> ffmpeg: The current check for SDL_Linked_Version() (available since SDL 1.2.13) [22:33] <CIA-17> ffmpeg: was added in 8f1b06c8, because including the headers for SDL_Init() [22:33] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rbf39d3b59d 10ffmpeg/libavcodec/truemotion2.c: [22:33] <CIA-17> ffmpeg: truemotion2: handle out-of-frame motion vectors through edge extension. [22:33] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [22:33] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [22:33] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r4f7c7624c0 10ffmpeg/libavformat/mov.c: [22:33] <CIA-17> ffmpeg: mov: don't overwrite existing indexes. [22:33] <CIA-17> ffmpeg: h264: Make ff_h264_decode_end() static, it is not used externally. [22:33] <CIA-17> ffmpeg: Also drop the now unnecessary ff_ prefix from its name. [22:33] <CIA-17> ffmpeg: 03Diego Biurrun 07master * rb691fd7a4d 10ffmpeg/libavcodec/h264.c: h264: drop ff_h264_ prefix from static function ff_h264_decode_rbsp_trailing() [22:33] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r2b6325cacb 10ffmpeg/: (log message trimmed) [22:33] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master' [22:33] <CIA-17> ffmpeg: * qatar/master: [22:33] <CIA-17> ffmpeg: h264: drop ff_h264_ prefix from static function ff_h264_decode_rbsp_trailing() [22:33] <CIA-17> ffmpeg: h264: Make ff_h264_decode_end() static, it is not used externally. [22:33] <CIA-17> ffmpeg: output-example: K&R formatting cosmetics, comment spelling fixes [22:33] <CIA-17> ffmpeg: avf: make the example output the proper message [22:34] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rb7b1509d06 10ffmpeg/libavcodec/truemotion2.c: (log message trimmed) [22:34] <CIA-17> (6 lines omitted) [00:00] --- Sat Mar 31 2012

1 0

ffmpeg.log.20120330
by burek 31 Mar '12

31 Mar '12

[00:15] <rschmidt> hi guys, I've got a bit of a strange situation while trying to read from a capture device with a freshly-built ffmpeg [00:16] <rschmidt> I'm trying to read from a Hauppauge HD PVR in linux, and convert that to an FLV(h264+aac or mp3 audio) for Flash [00:17] <rschmidt> when I pass the video source (/dev/video0) to ffmpeg with -i, it encodes a few frames (typically between 100 and 200) and it hangs there... doesn't respond to keyboard input [00:17] <rschmidt> doesn't crash either, though [00:18] <rschmidt> and if I cat the contents of /dev/video0 and read them in ffmpeg through stdin, it keeps on chugging with no hang [00:18] <rschmidt> so it's not a big problem per se, but it is a bit odd that doing things the 'right' way hangs and the roundabout way works. [03:31] <Kittens> oh dear where do I start. [03:33] <Kittens> I own a copy of wirecast and have been using justin/twitch.tv as my streaming service for about a year or two. yesterday they suspended my account because I took my medication on camera and that constituted "illegal drug use." I would like to run a stream from my own server (which has plenty of bandwidth to spare) and someone told me that ffserver would be the best choice. (server runs archlinux) [03:39] <Kittens> this is the error I keep getting [tcp @ 0x246e180] TCP connection to localhost:8090 failed: Connection refused [08:29] <burek> Kittens, [08:29] <burek> can you please use pastebin.com, to show your command line and its output? [08:30] <burek> moses1, you too? [08:32] <moses1> hi kittens, surely, but i have to leave the office right now [08:32] <moses1> so you later [08:33] <burek> ok :) [09:08] <rsv> if i want to convert yuv422 to rgb888 which function i have to use to do it [10:29] <lahwran> hi. I'd like to resample a 11025hz sound file to 44100hz, because I want to play it on a device that doesn't like 11025hz [10:30] <Tjoppen> ffmpeg -i foo.wav -ar 44100 bar.wav [10:31] <lahwran> ! [10:31] <lahwran> how nice and easy :D [10:31] <lahwran> Tjoppen: thanks! [10:34] <Tjoppen> :) [12:33] <brontosaurusrex> how can i overlay an alpha-rich png still over first 8 seconds of my video? [13:38] <filo1234> hi all, for extract audio from an flv video is it a good choise? ffmpeg -i input.flv -acodec mp3 -ab 192k out.mp3 [13:45] <filo1234> or there is some other options to add? [13:46] <moses> @burek [13:55] <zimnyx> I've encoded a movie using xvidcap (codec=FFMpeg MPEG-4), its size is 45MB. When I bzip2 it, its size falls down to 25MB. Is it fine? I mean shouldn't MPEG be maximaly compressed? [14:17] <brontosaurusrex> filo1234: flv audio could be mp3 allready, so you just need to demux [14:19] <brontosaurusrex> filo1234: your example is transcoding from lossy to lossy [14:19] <filo1234> brontosaurusrex: so what you could to be use for? [14:19] <filo1234> which string [14:20] <filo1234> brontosaurusrex: simply ffmpeg -i input.flv out.mp3? [14:20] <brontosaurusrex> ffmpeg -i in.flv -acodec copy out.mp3 maybe [14:21] <filo1234> brontosaurusrex: but if I want to set best quality audio like ab 190k? [14:21] <filo1234> -ab* [14:23] <brontosaurusrex> filo1234: you cant get any better than it allready is, so the goal is to keep it equaly small [14:23] <filo1234> brontosaurusrex: ok thanks a lot [14:23] <filo1234> I'm not very expert about audio/video :| like you can see lol [14:24] <brontosaurusrex> how can i overlay an alpha-rich png still over first 8 seconds of my video? [14:56] <globus> Does any one know how to use the rtpenc.c file ? [17:23] <madwill> Anybody feels all mighty on this friday and would like to throw a compiled x264 lib in 64 bit this way for a poor mans that can't get it to work [17:23] <madwill> I can only build 32 bit for some reason [17:25] <burek> why? [17:26] <madwill> Currently : DIED: endian test failed [17:26] <madwill> I have no idea what i means [17:26] <burek> did you try google? [17:28] <burek> I tried and this is what I've found [17:28] <burek> http://www.linuxquestions.org/questions/linux-newbie-8/config-error-729725/ [17:28] <burek> "Then I guess, the problems are solved ... the "endian test failed", error origins from a missing package 'glibc-headers' ." [17:28] <burek> do you have build tools installed? (gcc, make, ...) [17:29] <burek> if not: apt-get install build-essentials [17:29] <madwill> Oh! good idea, i'M building it on my windows machine but i should build it on a linux one [17:29] <burek> build-essential* [17:30] <burek> well you can build it on a Windows machine too, but you need build tools too :) [17:30] <madwill> I do have them, [17:30] <madwill> gcc and minGW the 32 bit build work flawlessly [17:30] <madwill> but i need the 64 as well [17:31] <madwill> can't link dynamic shared lib with a 64 bit process [17:31] <burek> hm, I'm weak at cross-compiling, so you might wait for someone more experienced to come on irc [17:32] <madwill> i'm simply going to try on linux [17:33] <burek> ok :) [17:35] <madwill> Thanks :P [17:36] <burek> :beer: :) [17:41] <madwill> Haha same error i must be simply getting the command wrong [17:42] <madwill> Used this : ./configure --cross-prefix=x86_64-w64-mingw32- --host=x86_64-pc-mingw32 [17:43] <burek> try configure --help | less [17:43] <burek> :) [18:34] <Diogo> hi one question.. [18:34] <Diogo> Unrecognized option 'directpred' anyone know what this it? [18:35] <burek> can you please use pastebin.com, to show your command line and its output? [18:40] <Diogo> http://pastebin.com/qxSqFhNz [18:41] <burek> oh [18:41] <burek> vpre is deprecated [18:41] <burek> ffmpeg now uses native x264's -preset -tune and -profile [18:41] <burek> try typing x264 --help [18:42] <burek> and see presets/tunes/profiles [18:43] <burek> after that use: ffmpeg ... -preset .. -profile ... -tune ... (which ever you need or all of them) [19:41] <Diogo> where i can find the last ffmpeg preset? [19:49] <burek> Diogo, there are no [19:49] <burek> its deprecated [19:49] <burek> ffmpeg now uses x264's native presets/profiles/tunes [19:49] <burek> type x264 --help [19:50] <burek> meaning, after you specify -vcodec libx264 you can use all the x264's options: http://mewiki.project357.com/wiki/X264_Settings [19:51] <burek> for example: ffmpeg -i input.avi -vcodec libx264 -preset fast -tune film -crf 20 ... [21:50] <Kakadu> hallo! [21:51] <Kakadu> I have a video a.flv (h264/aac) and I want to copy part from this video to another file with the same codecs. [21:52] <Kakadu> I write `ffmpeg -ss 01:02:08 -t 00:52:20 -acodec copy -vcodec copy -i a.flv 1.flv` but ffmpeg doesn't understand that I want copy codecs [21:52] <Kakadu> (I've found example with `copy` here: http://blog.yimingliu.com/2008/10/07/ffmpeg-encoding-gotchas/ ) [21:53] <burek> can you please use pastebin.com, to show your command line and its output? [21:54] <Kakadu> burek: http://paste.in.ua/4089/ [21:55] <burek> built on Feb 14 2012 [21:55] <burek> are you serious? :) [21:55] <Kakadu> debian unstable [21:55] <burek> oh my bad :) [21:55] <burek> sorry :) [21:55] <Kakadu> :) [21:56] <burek> -i [21:56] <burek> goes before [21:56] <burek> -acodec and -vcodec [21:56] <burek> parameters order do matter [21:56] <Kakadu> I should swap parameters? [21:56] <Kakadu> ok... [21:56] <burek> try something like this [21:56] <burek> ffmpeg -ss 01:02:08 -i live_user_sc2tv_ru_1331297204.flv -acodec copy -vcodec copy -t 00:52:20 1.flv [21:57] <JEEB> generally stuff before -i is related to decoding, while after -i is encoding [21:57] <burek> if you want to seek after the video is decoded (more accurate) then use this [21:57] <Kakadu> burek: WoW. Thanks a lot! [21:57] <burek> ffmpeg -i live_user_sc2tv_ru_1331297204.flv -ss 01:02:08 -acodec copy -vcodec copy -t 00:52:20 1.flv [21:58] <Kakadu> burek: -ss and -t before -i also works :) [21:58] <burek> yes I know, but it's a good practice to form a kind of parameters flow [21:58] <burek> like from input to output [21:58] <burek> in order what needs to be done [21:59] <burek> "take the input, seek some time, skip encodings, set the final duration, write to file" [21:59] <Kakadu> burek: OK, I understand [23:17] <globus> I am trying a couple of weeks now to use libav* for streaming purposes only. I want to stream over rtp already encoded frames. I encode the frames using the x264-api. Thanks [23:18] <globus> How can I use the libav* to stream only not to encode ? I have made it to encode and stream over rtp frames using the libav* but now I want to use it only to stream [23:26] <burek> did you try asking in #libav ? [23:27] <sacarasc> libav* are the libraries ffmpeg is made up of, burek. :p [23:27] <burek> just try.. [23:30] <globus> oops sorry [23:30] <globus> sorry guys [23:30] <burek> :beer: :) [23:31] <globus> hahahaha [23:48] <Lorra> Hi everyone! I am trying to convert a wmv video to an mp4 one and actually the video comes out good, but sound in the output file gets lost. Do you have a quick command line to perform this conversion? [23:48] <JEEB> -c:a aac -strict experimental -ab 192k [23:49] <Lorra> (maintaining quality, that is, I think, bitrates, size and audio sampling frequency) [23:49] <JEEB> this is, in the case ffmpeg can read the audio off that wmv file [23:49] <Lorra> JEEB, with ffmpeg -i input.wmv in front of that [23:50] <Lorra> and the name of the output file at the end of it? [23:50] <JEEB> uhh [23:50] <Lorra> JEEB, what would the whole command be? [23:51] <Lorra> JEEB, with input and output files [23:51] <JEEB> ffmpeg -i input.derp -c:v libx264 -crf 20 -preset:v fast -c:a aac -strict experimental -ab 192k out.mp4 [23:51] <Lorra> what is the -crf? [23:51] <JEEB> constant rate factor [23:51] <JEEB> closest we have to "constant quality" in x264 [23:51] <JEEB> if it looks good you can try raising it, if it looks bad you can lower it [23:52] <Lorra> isn't the number of frames per second mentioned in this command? [23:52] <JEEB> the idea is to use the highest number that still looks good [23:52] <JEEB> why would it? [23:52] <JEEB> in most cases you want the timestamps from input [23:52] <JEEB> as in, not changing a thing [23:52] <JEEB> -preset:v sets the libx264 preset [23:52] <JEEB> http://mewiki.project357.com/wiki/X264_Settings#preset [23:52] <JEEB> ^ a listing of them [23:53] <JEEB> slower presets will use techniquest that will help you compress the clip better, faster are meant for cases when compression plays a smaller role for you than speed [23:54] <JEEB> (everything except for ultrafast and placebo should be usable, placebo is "everything up to eleven" and "ultrafast" is to show that libx264 can be quite fast if needed -- for speed benchmarks) [23:54] <Lorra> JEEB, it says this: ffmpeg: failed to set value 'libx264' for option 'c:v' [23:54] <Lorra> I do have the package with that library installed though [23:55] <JEEB> how old is your ffmpeg? [23:55] <JEEB> I gave you commands primed for the current ffmpeg HEAD [23:55] <Lorra> It's 0.7.3 [23:56] <JEEB> then -c:v and -c:a would be -vcodec and -acodec [23:56] <JEEB> and yes, that's quite an old'ish version [23:57] <Lorra> and what would preset:v be? [23:57] <Lorra> ffmpeg: failed to set value 'fast' for option 'preset:v' [23:57] <Lorra> it's the latest version shipped with Ubuntu 11.10 [23:57] <JEEB> which usually means that it's from around 4-5 months before that :) [23:58] <JEEB> preset:v probably is -preset [00:00] --- Sat Mar 31 2012

1 0

ffmpeg-devel.log.20120329
by burek 30 Mar '12

30 Mar '12

[04:32] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * re74d6daa29 10ffmpeg/tests/ (fate/video.mak ref/fate/kgv1): [04:32] <CIA-17> ffmpeg: fate: add kgv1 fate test. [04:32] <CIA-17> ffmpeg: Tested to be bit-exact across x86-64, x86-32 and ppc. [04:32] <CIA-17> ffmpeg: 03Anton Khirnov 07master * re2e165c00f 10ffmpeg/tests/ (fate/vpx.mak ref/fate/vp8-size-change): FATE: add a test for vp8 with changing frame size. [04:32] <CIA-17> ffmpeg: 03Anton Khirnov 07master * ra839dbb94e 10ffmpeg/libavcodec/ (dv.c dvdata.c dvdata.h): dvenc: print allowed profiles if the video doesn't conform to any of them. [04:32] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r856c8e0a04 10ffmpeg/libavcodec/ (dct-test.c faandct.c faandct.h mpegvideo_enc.c): [04:32] <CIA-17> ffmpeg: faandct: Remove FAAN_POSTSCALE define and related code. [04:32] <CIA-17> ffmpeg: It is not a user-accessible option and unlikely to ever be changed. [04:32] <CIA-17> ffmpeg: 03Diego Biurrun 07master * ra92be9b856 10ffmpeg/ (39 files in 5 dirs): [04:32] <CIA-17> ffmpeg: Replace memset(0) by zero initializations. [04:32] <CIA-17> ffmpeg: Also remove one pointless zero initialization in rangecoder.c. [04:32] <CIA-17> ffmpeg: 03Mans Rullgard 07master * r2efb710a34 10ffmpeg/doc/Makefile: (log message trimmed) [04:32] <CIA-17> ffmpeg: build: fix doc generation errors in parallel builds [04:32] <CIA-17> ffmpeg: The $(dir) function used to construct OBJDIRS includes a trailing slash [04:32] <CIA-17> ffmpeg: in the names returned, which GNU make 3.82 does not match to the [04:32] <CIA-17> ffmpeg: slash-less 'doc' in the documentation dependencies, causing parallel [04:32] <CIA-17> ffmpeg: build to fail. Adding a slash fixes this and still works with make [04:32] <CIA-17> ffmpeg: 3.81. [04:32] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r0a82f5275f 10ffmpeg/libavcodec/ (lagarith.c lagarithrac.c): [04:32] <CIA-17> ffmpeg: lagarith: fix buffer overreads. [04:32] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [04:32] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [04:32] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r7374fac804 10ffmpeg/libavcodec/cabac_functions.h: [04:32] <CIA-17> ffmpeg: h264: fix overreads in cabac reader. [04:32] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [04:32] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [04:32] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rdb025929f2 10ffmpeg/libavcodec/x86/cabac.h: h264: use struct offsets in get_cabac_bypass_sign_x86(). [04:33] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * ra0bdcb019e 10ffmpeg/libavcodec/x86/cabac.h: h264: add overread protection to get_cabac_bypass_sign_x86(). [04:33] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rd40ff29cac 10ffmpeg/: (log message trimmed) [04:33] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master' [04:33] <CIA-17> ffmpeg: * qatar/master: [04:33] <CIA-17> ffmpeg: asf: only set index_read if the index contained entries. [04:33] <CIA-17> ffmpeg: cabac: add overread protection to BRANCHLESS_GET_CABAC(). [04:33] <CIA-17> ffmpeg: cabac: increment jump locations by one in callers of BRANCHLESS_GET_CABAC(). [04:33] <CIA-17> ffmpeg: cabac: remove unused argument from BRANCHLESS_GET_CABAC_UPDATE(). [04:33] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r44257ef426 10ffmpeg/libavformat/asfdec.c: [04:33] <CIA-17> (34 lines omitted) [04:57] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r9aeacc9546 10ffmpeg/ffmpeg.c: [04:57] <CIA-17> ffmpeg: ffmpeg: dont disable dr1 when changing dimensions are encountered in codec_get_buffer(). [04:57] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:37] <CIA-17> ffmpeg: 03Michael Bradshaw 07master * rb678dd9690 10ffmpeg/libavcodec/libopenjpegdec.c: [05:37] <CIA-17> ffmpeg: Fixes ticket 1127. [05:37] <CIA-17> ffmpeg: I'm still looking into why bpp is getting set to 0. [05:37] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * raebce0b0fa 10ffmpeg/: [05:37] <CIA-17> ffmpeg: Merge git://github.com/mjbshaw/FFmpeg-OpenJPEG-J2K-Encoder [05:37] <CIA-17> ffmpeg: * git://github.com/mjbshaw/FFmpeg-OpenJPEG-J2K-Encoder: [05:37] <CIA-17> ffmpeg: Fixes ticket 1127. I'm still looking into why bpp is getting set to 0. [05:37] <CIA-17> ffmpeg: Merged-by: Michael Niedermayer <michaelni(a)gmx.at> [05:37] <CIA-17> ffmpeg: 03Piotr Bandurski 07master * rddd8974008 10ffmpeg/libavcodec/iff.c: [05:37] <CIA-17> ffmpeg: iff: prevent a possible crash with broken/prepared IFF PBM [05:37] <CIA-17> ffmpeg: Based on fix by Peter Ross for ticket #1054. [05:37] <CIA-17> ffmpeg: Reviewed-by: Paul B Mahol <onemda(a)gmail.com> [05:37] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [06:21] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra37a281778 10ffmpeg/ffmpeg.c: [06:21] <CIA-17> ffmpeg: ffmpeg: pass bitexact flag to vsrc_buffer [06:21] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [06:21] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r1dab5efa01 10ffmpeg/tests/ (fate/vpx.mak ref/fate/vp8-size-change): [06:21] <CIA-17> ffmpeg: fate/vp8-size-change: set bitexact flag [06:21] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [08:03] <CIA-17> ffmpeg: 03Carl Eugen Hoyos 07master * r9a0f2b7b3a 10ffmpeg/libavcodec/Makefile: [08:03] <CIA-17> ffmpeg: Revert "Fix compilation with hardcoded tables." [08:03] <CIA-17> ffmpeg: This reverts commit b17b2963692f5f7c19a2188e61922e00680c5fb4. [13:10] <CIA-17> ffmpeg: 03Carl Eugen Hoyos 07master * r282ec7289d 10ffmpeg/libavcodec/mjpegdec.c: [13:10] <CIA-17> ffmpeg: Do not try to use lowres for unusual jpg subsampling. [13:10] <CIA-17> ffmpeg: Fixes ticket #1144. [19:27] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r4204364678 10ffmpeg/ffmpeg.c: [19:27] <CIA-17> ffmpeg: ffmpeg: switch duration calculation code from avg to r frame_rate. [19:27] <CIA-17> ffmpeg: This heuristic is less likely to overshoot and cause encoder failure. [19:27] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [19:39] <ubitux> michaelni: i'm getting the DTS invalid droping warning (introduced in 29034e65) while transcoding a rtsp stream in real time [19:39] <ubitux> not sure it is justified [19:39] <ubitux> ./ffmpeg -i rtsp://son.radio-france.fr/locales/iledefrance/info/elysee120329.rm -aq 0 -y out.mp3 [19:40] <ubitux> also, it seems i can't run an audio copy because of the pts<dts [19:52] <michaelni> ubitux, works if the input is marked as containing ts discontinuities [19:52] <michaelni> iam not saying this is the correct solution though .. [19:55] <michaelni> hi beastd [19:55] <beastd> hi [19:55] <tborg> hi beastd, hi michaelni [19:57] <michaelni> ubitux, it seems start_time/first_dts are 0, this probably confuses the code when TS discontinuities are not supposed to be present but iave not verified this hypothesis [20:12] <ubitux> ok [20:19] <MP4_maniac> ubitux: i have never seen isom/mov files with shadow sync sample atom, and i can't understand its behaviour... [20:21] <ubitux> according to the annex describing a seeking method, it seems appropriate for "streaming seeking" iirc [20:21] <ubitux> but i can't understand why :p [20:21] <michaelni> ubitux, the timestamps from the rtsp demuxer dont look clean at all [23:35] <ubitux> is it possible to define a "zlib" IO? [23:35] <ubitux> (aviocontext etc) [00:00] --- Fri Mar 30 2012

1 0

ffmpeg.log.20120329
by burek 30 Mar '12

30 Mar '12

[00:03] <frostschutz> burek: http://pastebin.com/NsLUFM1z [00:04] <burek> well, there are no errors in that paste [00:04] <burek> sprzybilla, [00:04] <burek> can you please use pastebin.com, to show your command line and its output? [00:05] <frostschutz> burek: I thought you were referring to the audio issue (wrong channel mapping) I mentioned earlier. I made no other inquiries [00:05] <burek> oh.. my bad.. [00:05] <burek> sorry, i mean to ask markatto ... :( i need some sleep obviously.. :/ [00:05] <burek> let me see your issue [00:05] <burek> :) [00:05] <burek> just a sec [00:06] <sprzybilla> burek: no problemo! [00:09] <burek> frostschutz, you are just reencoding 7 channels audio right? [00:09] <burek> and you get a bad audio quality at the output? [00:10] <burek> also, can you try to use mp4 or flv as an output format/container and see if that helps? [00:10] <frostschutz> burek: yup. the dts sounds fine, the encoded ones poor. (the dts is from samples.mplayerhq if you want to try yourself) [00:11] <frostschutz> burek: container doesnt matter I also tried mkv [00:11] <burek> did you try just using uncompressed audio encoder? [00:12] <burek> -uncompressed +lossless [00:12] <burek> like wav or s16le [00:13] <markatto> burek: the ` was just for clarity, I wasn't typing that in [00:13] <frostschutz> burek: Unknown encoder 's161e' - what would I need to put in the cmdline? [00:15] <burek> frostschutz, ffmpeg -codecs [00:15] <burek> markatto, ok [00:19] <markatto> I am having another issue now, something involving sample formats for aac [00:19] <burek> can you please use pastebin.com, to show your command line and its output? [00:19] <markatto> http://pastie.org/private/ijolsmotbf3wbi9xfcqxcg [00:20] <burek> try reading the output a little bit more careful :) [00:20] <burek> and add -strict -2 [00:20] <burek> or -strict experimental [00:20] <burek> something like that [00:22] <markatto> burek: okay I figured out that the -strict stuff needs to be after the input file [00:23] <markatto> but isn't s16 standard for aac? [01:11] <songderman> Hi, I have a question about makin multiple thumbnail for video example: every 10 minute for video with 1 hour duration [01:11] <songderman> with 1 hour duration [01:16] <burek> songderman, you want to create thumnails for every 10 minute of the video? [01:17] <burek> it would be the best to use the batch file with for loop [01:17] <burek> but [01:18] <burek> one non-standard way, we could try here, would be to "convert" your video to png/jpg images [01:18] <burek> and use some very odd fps for the output, to match that 10 minutes interval [01:18] <burek> let me test it first, and I'll let you know if it can work in 1 cmd line [01:20] <burek> until that, try playing with: ffmpeg -i input -ss start_time_in_seconds -f image2 out%02.png [01:20] <burek> until that, try playing with: ffmpeg -i input -ss start_time_in_seconds -vframes 1 -f image2 out%02.png [01:20] <burek> typo :) [01:23] <burek> make it out%d.png [01:24] <burek> yup it works in 1 cmd [01:25] <burek> ffmpeg -i input.flv -f image2 -r 1 out%d.png [01:25] <burek> that would give you 1 image each second [01:25] <burek> ffmpeg -i input.flv -f image2 -r 1/60 out%d.png [01:25] <burek> each minute [01:25] <burek> and since you want each 10 minutes: ffmpeg -i input.flv -f image2 -r 1/600 out%d.png [01:43] <icat> hey, is it possible to change the fourcc code of the audio stream w/out reencoding? [01:44] <icat> i got a file from tvitch.tv, and it contains speex audio: Stream #0.1: Audio: 0x0000, 16000 Hz, mono, s16 [01:44] <icat> but the tag is messed up so vlc etc won't play it [01:46] <burek> can you use a hex editor? [01:47] <icat> oh :) [01:47] <icat> if you can recommend a shell-based one, sure [01:48] <burek> oh [01:49] <burek> well, my bad :) but hey, there is -vtag option :D [01:50] <burek> so, ffmpeg -i input.flv -vcodec copy -acodec copy -vtag BLAH output.flv [01:50] <burek> :) [01:54] <icat> thx [02:01] <burek> :beer: [02:01] <icat> there's an additional problem [02:01] <burek> yes [02:01] <burek> ? [02:01] <icat> [NULL @ 0x9de6db0]flv does not support that sample rate, choose from (44100, 22050, 11025). [02:01] <burek> can you please use pastebin.com, to show your command line and its output? [02:02] <icat> sure: http://pastebin.com/mizLbd0s [02:02] <icat> looks like the problem is the 16k sample rate [02:13] <burek> hm [02:13] <burek> Input #0, flv, from 'finale.flv': [02:13] <burek> Stream #0.1: Audio: 0x0000, 16000 Hz, mono, s16 [02:13] <burek> ffmpeg did not recognize your audio stream [02:13] <burek> I don't know why [02:13] <icat> speex audio [02:13] <burek> but I would suggest updating your ffmpeg to the latest git [02:16] <icat> ok [02:18] <icat> thanks, burek [02:18] <icat> helps me out a lot [02:36] <burek> :beer: :) [02:40] <Durmush> hi, how do i remux mp4 files so they are streaming compatible? [02:54] <burek> Durmush, try using -f mpegts [02:54] <burek> also, try describing your goal a little bit more in detail [03:02] <Durmush> burek,i'm trying to stream an mp4 file via Flash Media Server's RTMP. the stream played nice in mplayer, but in embedded flash player, the video doesnt show. just audio. but i guess this is beyond file format problems right? it is gotta be an issue with flash player... [03:12] <burek> which embedded flash player [03:12] <burek> jw? [03:12] <burek> Durmush, you can always paste the url to the stream on their support forum, so they can check and tell you what's wrong [03:59] <bencahill> hmm, I've got a Radeon HD3450 on XP 32-bit...any chance ffmpeg could use it to speed up en/decoding? [03:59] <bencahill> (h264->libx264, to be exact) [04:00] <bencahill> hmm, -preset for x264 I am not using :-/ [05:02] <sacarasc> bencahill: No, ffmpeg can't do that. [05:23] <bencahill> sacarasc: kthx [06:26] <LiNoRgirl> i noticed vp6 is awful, is vp3 just as bad [08:35] <ruby_on_tails> i have recorded my screen for 20 mins in a mpeg format screencast which is 120mb in size [08:36] <ruby_on_tails> i did : ffmpeg -i test-0000.mpeg -r 24 -b 64k movie.mpeg tocompress it, the quality degrades superly, but the size is 130mb, why ? [08:39] <jmbc> about quality: i suppose a nice medium video should have like -b 700k [08:40] <ruby_on_tails> the original -b is 500 [08:40] <jmbc> good then [08:40] <ruby_on_tails> i shifted it to 64k [08:40] <ruby_on_tails> the quality degraded [08:40] <ruby_on_tails> but the size inflated [08:40] <ruby_on_tails> why ? [08:40] <jmbc> also i suppose using a codec like xvid or divx will compress it alot more than a plain mpeg [08:41] <jmbc> not compressed/encoded? [08:41] <ruby_on_tails> no its simple mpeg [08:41] <ruby_on_tails> i want to do without a codec [08:42] <jmbc> what codec does it had? [08:45] <ruby_on_tails> no codec for now [08:45] <ruby_on_tails> it doesnt have any codec [08:46] <mystica555> if its an mpeg, its got a codec. by default, mpeg2video [08:46] <ruby_on_tails> oh [08:46] <ruby_on_tails> hmm [08:47] <mystica555> you need a much more efficient codec to try and get a reasonable quality at low bitrate, and 64K is indeed much too low for much quality with any codecs [08:47] <ruby_on_tails> i am new to this [08:47] <ruby_on_tails> i just need to bring the file size < 20mb [08:47] <mystica555> h264 for 640x480 tends to be about 1 megabit for good quality [08:47] <mystica555> 1 megabyte per minute? [08:47] <jmbc> ffmpeg -i test-0000.mpeg will tell you all [08:47] <mystica555> of screen capture? [08:47] <ruby_on_tails> mystica555: if possible yes [08:48] <mystica555> what is the activity on-screen [08:48] <ruby_on_tails> its a simple tutorial [08:48] <ruby_on_tails> browsing a web app [08:48] <ruby_on_tails> nothing complex, low quality will work [08:49] <ruby_on_tails> and just that it can run on a windows machine without irritating the user for some weird things to be installed [08:50] <jmbc> video codecs allways do :D [08:50] <jmbc> just get a mega pack [08:50] <mystica555> http://camstudio.org/ seems to offer what you want, screencasts, with deployment in flash... i can't vouch for the quality.. but, it is windows software... [08:51] <mystica555> as in, it runs on windows.. so if you are wanting to capture a linux interaction with the shell, perhaps use it on windows with putty..or with X11 perhaps use on windows VNCing into your x11 session [08:52] <mystica555> i really can't think of anything specifically to make a screen capture with even highest compression/quality h264 work at that low of a bitrate.. [08:52] <ruby_on_tails> i dont care about the bitrate [08:52] <ruby_on_tails> i care only about the size [08:52] <mystica555> you said under 20 megabytes [08:52] <mystica555> that dictates the bitrate [08:52] <ruby_on_tails> oh [08:52] <ruby_on_tails> ok [08:52] <mystica555> and for 1 megabyte per minute, you are at 128 kilobits max per second. [08:52] <mystica555> or perhaps less [08:53] <mystica555> 128 kilobits, h264, will work for...320x240 or perhaps 640x480 at very low framerate [08:53] <jmbc> you might be a noob, but xvid or h264 are not rare to be on systems for playback [08:53] <mystica555> try perhaps capturing with 5fps [08:54] <mystica555> jmbc: depends on your environment; in a corporate environment, there is flash, but not always a way to play back raw h264 avis/mkvs/mp4s [08:54] <mystica555> so i consider this.. [08:54] <mystica555> jwplayer would make it work however [08:54] <mystica555> as a bridge between flash and the h264 input [08:54] <ruby_on_tails> mystica555: i tried to bring down the fps to 6 but it throws an error [08:55] <mystica555> ruby_on_tails: you were doing a specific container format (mpeg) which has set limits. you also were potentially violating the miniumum framerate of the codec, but im not certain [08:55] <jmbc> then a .flv still would allow h264 compression [08:55] <ruby_on_tails> oh [08:55] <ruby_on_tails> hmm [08:55] <ruby_on_tails> i am trying .avi now [08:55] <ruby_on_tails> i think avi runs fine on windows [08:55] <jmbc> so does .flv on flash ;D [08:55] <mystica555> id suggest for test and deployment purposes to focus on mp4; then you can later process that with flash or other utilities to make a web playable presentation [08:57] <mystica555> there is also something to consider with mp4: the 'fast-start' ability (the MOOV atom at the beginning of the stream) with 'qt-faststart' [08:59] <ruby_on_tails> i did a -i i cant find the screen size of the video [08:59] <ruby_on_tails> i think i will reduce the dimensions a bit [08:59] <ruby_on_tails> but i dont know the original :) [09:00] <ruby_on_tails> oh i see it [16:41] <smvv> hi. I'm writing a tool to interleave external data with a mkv file. I use the new libav* api (0.10.2), but the bitrate of the output file is incorrect. Did I miss something when copying the video stream? My code: http://sprunge.us/KfEf?c and av_dump_format output: http://sprunge.us/jgEH [16:42] <smvv> (the code is a stripped down version, e.g. without error checking, to reduce the length of the code) [16:58] <sprzybilla> anyone mind helping me troubleshoot my crappy looking MPEG2-TS output? I usually work with x264, so quality mpeg2 is a new world: http://pastebin.com/NRP9SBs5 [18:17] <stan_theman> If I've got a wmv file that uses g2m4 codec. Can I convert this to *anything* else and keep the audio+video? [18:18] <stan_theman> been reading and am having problems [18:18] <jmbc> ffmpeg -codecs | grep g2m4 [18:19] <stan_theman> i think it sent it to stdout, but it looks like i dont have it installed. i've got a debian box and a mac on hand at the moment [18:19] Action: jmbc wonders what is convert apart but then keep the audio+video [18:20] <stan_theman> i can play the audio but can't see the video. sorry, I'm not super good with the terminology and didn't want to offend if i was wrong [18:22] <jmbc> ffmpeg -codec |grep g2m4 gives nothing, what's that codec? more info on it? [18:23] <stan_theman> it's the GoToMeeting codec, it looks like there was some chatter on the mailing list about it [18:23] <stan_theman> but it's kind of sparse [18:25] <jmbc> nope, nothing. seems unsuported, very specific. [18:25] <stan_theman> yeah, it sucks. the marketing guy got a video in this format and VLC won't even go for it [18:27] <stan_theman> looks like there are some windows solutions, just need to find an iso and spin up virtual box [21:10] <madwill> if i compile x264 using : ./configure --disable-cli --enable-shared --extra-ldflags=-Wl,--output-def=libx264-120.def its going to be 32 bit right ? [21:11] <madwill> i need both 32 and 64 bit [21:12] <high-rez> I'm attempting to transcode a video using ffmpeg (ffmpeg -i input.dvr-ms -deinterlace -target ntsc-dvd output.mpg). The result is very choppy, and it looks like almost 30% of the frames are being dropped. [21:12] <high-rez> E.g. I"m seeing this: frame= 697 fps= 52 q=2.0 Lsize= 16138kB time=22.40 bitrate=5901.9kbits/s dup=348 drop=231 -- what am I doing wrong ? [21:16] <specie> hi i just read the legal compliance information on the website and i was wondering .. [21:17] <specie> what if i build a product not for commercial purpose, but giving it away for free. Can i keep the -non-free and -gpl libs in ? [21:19] <Plorkyeran> no [21:22] Action: specie sighs deeply. [21:23] <JEEB> nonfree stuff can't be distributed in any case, (L)GPL stuff is distributable if you follow the license. [21:23] <JEEB> (and no, there isn't a single difference from ffmpeg's side whether or not it's payware or not) [21:24] <specie> ok thanks. [21:26] <JEEB> LGPL = with dynamic linking you release sources for the LGPL part, with static linking sources for the LGPL part + object files for everything else (IIRC), GPL = sources of the whole thing. [21:29] <specie> ic.. well it wasn't exactly the answers i was looking for.. but it will have to do. Getting sued was not the aim of the project tbh. [21:29] <JEEB> well, it wasn't exactly clear what you were asking :P [21:29] <specie> i'll see if i can work arround it somehow.. leaving out the nonfree and gpl bits. [21:30] <JEEB> no real need to work around anything if you only use LGPL parts of ffmpeg [21:30] <specie> well i got my anwsers.. but i just hoped they came out different being a free product ;) [21:30] <JEEB> since you only need to release the ffmpeg sources if you link dynamically [21:31] <JEEB> nonfree libraries pretty much means that the library was later or from the beginning found to be something that couldn't be licensed under LGPL or GPL [21:31] Action: specie nods solemnly. [21:31] <JEEB> and thus you still have the chance to compile with such, but you can't give those binaries out [21:32] <specie> well that currently takes out a chunk of the basis functions :) [21:32] <JEEB> does it really? [21:32] <JEEB> I only remember faac and libaacplus being nonfree [21:40] <JEEB> of course GPL'd libs such as libx264 make you a bigger problem if you were thinking about using them :) [21:41] <specie> uhm. yes. [21:41] <Plorkyeran> openssl too [21:41] <JEEB> (although x264 now does have an alternative licensing scheme for corps) [21:41] <JEEB> Plorkyeran, tru dat [21:41] <Plorkyeran> since it's 4-clause bsd [21:41] <specie> was just checking if it was actually was using libaacplus but it isnt. [21:42] <specie> i'm transcoding to webm aswell [21:45] <specie> it should transcode to .webm and .mp4. But it seems mp4 is using libfaac and libx264 so i guess i'm loosing a chunk indeed. [21:46] <JEEB> you can replace libfaac with something like Quicktime's API, or ffmpeg's aac encoder -- libx264 becoming the main problem if you want to do a closed source app :P [21:48] <specie> what if i don't want to do a closed source app? [21:49] <JEEB> and use libx264? Get a commercial license, lol [21:49] <JEEB> oh [21:49] <JEEB> _don't_ want to? [21:49] <specie> heh [21:49] <JEEB> if you use libx264, then you just have to make it GPL [21:49] <JEEB> and that's it [21:51] <feelcal> Hi,I am a new user of libavcoed and trying to decode a h264 encoded video from a rtsp server with libavcodec but I cannot figure out how to initialize the AVCodecContext from the sdp information. It seems I need to extradata in it, but not sure in what order& Anyone could have an idea on where to look for help here? [21:52] <feelcal> (and it was libavcodec ;) ) [21:58] <specie> JEEB: thank you for your anwsers so far. I guess i need to dig a little deeper in all the legal matters before giving some combined technologies; each with their own licenses, away. [21:59] <specie> this is actually a lot harder then i expected it to be of 'free software' [21:59] <JEEB> uhh, the software license side should be rather straightforward unless you are not telling me something I should know [22:01] <specie> i honestly don't know what you referring at. It's just a bunch of tech bound together in a 'appliance' (VM). But with all those license term i doubt i can just put it online. [22:01] <specie> ffmpeg being just of the technologies i'm using. [22:01] <specie> s/ffmpeg/FFmpeg [22:06] <madwill> if we use x264 we have to be GPL ? [22:06] <madwill> even by dynamic linking [22:06] <madwill> Anyway [22:06] <JEEB> x264 is GPL [22:07] <JEEB> (or payware license) [22:07] <madwill> So basically Smaxe demo (java rtmp publisher) using shared libx264-79.dll is a fraud [22:07] <madwill> Good to know [22:08] <JEEB> if they don't release sources, yeah -- because I don't think that old of an x264 had payware licensing yet [22:08] <madwill> still need this demo to work on 64 bit jvm which refuse to take on 32bit compil. this only works on 32bit jvm [22:09] <madwill> back to google [22:09] <JEEB> I think that's because it uses 32bit libraries and you can't load them into a 64bit process :P [22:09] <madwill> yep it is [22:34] <madwill> it appears i'm doing it wrong [22:34] <madwill> Damn why can,t you take copy and paste stupid MINGW32 [22:35] <JEEB> that depends on the shell you're using for msys/mingw :P [22:35] <JEEB> I use mintty [22:36] <madwill> i don't think its me : ./configure --cross-prefix=x86_64-w64-mingw32- --host=x86_64-pc-mingw32 just gives : No working c compiler found [22:36] <madwill> MSYS-1.0.11.exe [22:37] <JEEB> and that's not the name of the thing you're using for your tty/shell :P [22:37] <JEEB> and see config.log why it fails [22:37] <madwill> Oh thanks [22:42] <madwill> checking whether x86_64-w64-mingw32gcc works... no lol [22:43] <madwill> i guess it should be looking for : i686-w64-mingw32-c++ [22:43] <moses1> hallo zusammen [22:44] <moses1> ok, somebody in her to help to ompiling and customizing ffmpek with librtmp on debian squeeze [22:49] <moses1> is there somebody out who can talk about librtmp errors? [00:00] --- Fri Mar 30 2012

1 0

ffmpeg-devel.log.20120328
by burek 29 Mar '12

29 Mar '12

[02:00] <Daemon404> j-b, i traced that crash to a windows directshow filter [02:00] <Daemon404> as in one that comes with it [02:00] <Daemon404> no codec packs here. [02:01] <Daemon404> just a thought @ your reply. [02:02] <j-b> ok [02:02] <j-b> still, notvlc [02:02] <Daemon404> yeah [02:02] <Daemon404> i gave the bug report to a freind at MS though [02:02] <Daemon404> he opened a internal defect [02:34] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * re6093e3640 10ffmpeg/ffplay.c: [02:34] <CIA-17> ffmpeg: ffplay: pause rdft column when ffplay is paused [02:34] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [02:34] <CIA-17> ffmpeg: Signed-off-by: Marton Balint <cus(a)passwd.hu> [02:34] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * re206b51fd3 10ffmpeg/: [02:34] <CIA-17> ffmpeg: Merge remote-tracking branch 'cus/stable' [02:34] <CIA-17> ffmpeg: * cus/stable: [02:34] <CIA-17> ffmpeg: ffplay: pause rdft column when ffplay is paused [02:34] <CIA-17> ffmpeg: Merged-by: Michael Niedermayer <michaelni(a)gmx.at> [03:02] <durandal11707> how to handle format where data in video and audio packet is splitted in bunch of small chunks and than mixed in random order [04:07] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r4a619fcae9 10ffmpeg/libavcodec/h263dec.c: [04:07] <CIA-17> ffmpeg: h263dec: Restore w/h values to a consistent state if a change is rejected. [04:07] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [04:07] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:07] <CIA-17> ffmpeg: 03Paul B Mahol 07master * re4cc9f309e 10ffmpeg/ (libavcodec/flicvideo.c libavformat/riff.c): [04:07] <CIA-17> ffmpeg: Support FLIC in AVI. [04:07] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [04:07] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:07] <CIA-17> ffmpeg: 03Piotr Bandurski 07master * r3c6397404b 10ffmpeg/libavformat/id3v2.c: [04:07] <CIA-17> ffmpeg: id3v2: support JPEG APIC with incorrect (image/jpg) MIME type [04:07] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:26] <CIA-17> ffmpeg: 03Diego Biurrun 07master * rd7ce1a48a5 10ffmpeg/configure: configure: Drop stray duplicate entry for --disable-fft from help output. [04:26] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r8e0b7e2f14 10ffmpeg/doc/Makefile: [04:26] <CIA-17> ffmpeg: build: Do not explicitly add the doc directory to the OBJDIRS list. [04:26] <CIA-17> ffmpeg: Now that a documentation generator is built in the doc directory, [04:26] <CIA-17> ffmpeg: this is no longer necessary. Fixes the Make warning: [04:26] <CIA-17> ffmpeg: Makefile:188: target `doc' given more than once in the same rule. [04:26] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r98df2e2414 10ffmpeg/libavcodec/rawdec.c: [04:26] <CIA-17> ffmpeg: raw: forward avpicture_fill() error code in raw_decode(). [04:26] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [04:26] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [04:26] <CIA-17> ffmpeg: 03Diego Biurrun 07master * rd724fe665b 10ffmpeg/libavcodec/ (Makefile dv.c dv_tablegen.h dvdata.h dvdec.c): dv: Split off DV video decoder into its own file. [04:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r7b453d1c60 10ffmpeg/: (log message trimmed) [04:26] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master' [04:26] <CIA-17> ffmpeg: * qatar/master: [04:26] <CIA-17> ffmpeg: westwood_vqa: fix SND0 chunk handling [04:29] Action: Compn wonders what dv decoder was split [04:30] <durandal11707> so that people who do not care about encoders are happy [04:32] <Compn> ah [04:33] <durandal11707> name Heart of Darkness fits paf format [04:34] <Compn> Daemon404 : how many decoders/ demuxers are you going to work on? :) [04:35] <Compn> hehe [06:23] <Daemon404> Compn, ? [06:25] <oneal_gao> hello , I'm a developer for video processing , image procesing . Now I want to the hareware acceleration in ffmpeg , but it seems the hardware is not compliled when I use --enable-dxva2 [06:26] <oneal_gao> ./configure --enable-dxva2 --enable-hwaccel=mpeg2_dxva2 --prefix=/home/ffmpeg/ffmpeg-v0.7-1696-gcae4f4b/install , I want to eanble the mpeg2 hardware accelartion in ffmpeg. [06:27] <oneal_gao> the mpeg2_dxva2 is not compiled [06:29] <oneal_gao> Anyone can tell me how to enable the dxva2 hardware acceleration [07:08] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rc30d04cedd 10ffmpeg/libavcodec/lagarithrac.c: [07:08] <CIA-17> ffmpeg: lagarithrac: fix length used in ff_lag_rac_init() [07:08] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [07:08] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r44bd50f529 10ffmpeg/libavcodec/lagarith.c: [07:08] <CIA-17> ffmpeg: lagarith: fix flipped return value [07:08] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [07:08] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r96d0494123 10ffmpeg/libavcodec/lagarith.c: [07:08] <CIA-17> ffmpeg: lagarith: Fix various issues that lead to out of array reads. [07:08] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [07:08] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [07:53] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r11cc209226 10ffmpeg/libavcodec/lzw.c: [07:53] <CIA-17> ffmpeg: lzw(gif): Fix overread [07:53] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [07:53] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [07:53] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rd0dafebb75 10ffmpeg/libavcodec/truemotion2.c: [07:53] <CIA-17> ffmpeg: tm2: Fix overread of token array. [07:53] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [07:53] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [08:07] <CIA-17> ffmpeg: 03Carl Eugen Hoyos 07master * rb17b296369 10ffmpeg/libavcodec/Makefile: Fix compilation with hardcoded tables. [08:51] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r936951ca5c 10ffmpeg/libavcodec/tta.c: [08:51] <CIA-17> ffmpeg: tm2dec: check total_frames and extradata_size. [08:51] <CIA-17> ffmpeg: Fixes overread [08:51] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [08:51] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [08:51] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r7a5e587249 10ffmpeg/libavcodec/apedec.c: [08:51] <CIA-17> ffmpeg: apedec: check bits <= 32 [08:51] <CIA-17> ffmpeg: Fixes FPE [08:51] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [08:51] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [09:42] <cbsrobot> Tjoppen, if you are interested on a dcp mxf with subtitle and font embeded: http://www.michaelcinquin.com/get/TEST-MCINQUIN.zip [10:18] <Tjoppen> might be useful. yoink! [15:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8db2935db0 10ffmpeg/libavcodec/apedec.c: [15:22] <CIA-17> ffmpeg: apedec: fix global array overread [15:22] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:22] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra60a4d7041 10ffmpeg/libavcodec/vc1dec.c: [15:22] <CIA-17> ffmpeg: vc1dec: Fix global array overread. [15:22] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:22] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r92f7f1db42 10ffmpeg/libavcodec/indeo4.c: [15:22] <CIA-17> ffmpeg: indeo4: Check for mismatching scan tables [15:22] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:22] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:22] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rc44417e15a 10ffmpeg/libavcodec/indeo4.c: [15:22] <CIA-17> ffmpeg: indeo4: Dont leave tables in random state on errors. [16:34] <ubitux> oh nice patches from nicolas [16:54] <ubitux> in mp4, there are shadow sync sample atoms; just like sync samples but more mysterious [16:54] <ubitux> they look as useful as a fibonacci sequence [16:54] <ubitux> does anyone know if any app is making use of them? [16:54] <CIA-17> ffmpeg: 03Nicolas George 07master * r247fbf071b 10ffmpeg/ (doc/filters.texi libavfilter/version.h libavfilter/vf_ass.c): ass: fix aspect ratio computation. [16:55] <ubitux> MP4_maniac: any idea? :) [17:48] <ramiro> hi [17:48] <j-b> ramiro: hello [17:49] <ramiro> is there an ffmpeg option to use the same bit_rate for an output stream as its input stream? [17:49] <av500> how does that work for vbr? [17:50] <ramiro> in my case it's cbr [17:50] <av500> well, but not all containers have bitrate [17:50] <av500> reliably [17:51] <av500> so in case the input format does not provide it, what would you fallback to? [17:51] <ramiro> it's a specific use-case. i just need to set codec->bit_rate = icodec->bit_rate; [20:12] <jkdev> Hi, i would like to terminate cleanly (releasing all resources) the encoding process. Just like pressing 'q' in the "Press [q] to stop". What function should i call to do this properly? [23:44] <CIA-17> ffmpeg: 03Stefano Sabatini 07master * r99bb88c588 10ffmpeg/ (doc/ffmpeg.texi ffmpeg.c): [23:44] <CIA-17> ffmpeg: doc/ffmpeg: update and extend documentation for -copytb [23:44] <CIA-17> ffmpeg: Address trac ticket #1120. [00:00] --- Thu Mar 29 2012

1 0

ffmpeg.log.20120328
by burek 29 Mar '12

29 Mar '12

[00:28] <sh_t> hey are there any good references for encoding options to use for various formats ie iphone/ipad with the latest ffmpeg? ive found a lot to be outdated syntax wise [00:42] <Wntrvnm> sh_t: https://github.com/Wintervenom/Scripts/blob/master/video/ffmpeg/ffmpeg-andr… [00:42] <Wntrvnm> sh_t: (But take out the 'r 23.976'.) [00:42] <Wntrvnm> * -r [00:43] <Wntrvnm> And adjust the resolution height to 640. [00:54] <sh_t> nice, thanks. I also found this http://www.virag.si/2012/01/web-video-encoding-tutorial-with-ffmpeg-0-9/ [00:54] <sh_t> I am curious though, what's the difference between -s and -vf scale:w:h [00:54] <sh_t> seems that scale is more flexible since you can use -1 to maintain the aspect ratio [00:55] <sh_t> -vf scale=w:h* sorry [00:58] <Wntrvnm> sh_t: «-s» doesn't maintain ratio. [00:59] <sh_t> alright just confirming thanks :) [01:00] <Wntrvnm> You might remove the «unsharp» filter from that, too. [01:00] <Wntrvnm> The only reason I put that there is because it helps make things a little less blurry on small screens. [01:02] <sh_t> i notice you use qt-faststart there [01:02] <sh_t> im used to using MP4Box to fix the streaming issues [01:02] <sh_t> do you have any input on that? [01:03] <Wntrvnm> «qt-faststart» was the first one I learned about and already comes with FFmpeg. [01:33] <sh_t> Wntrvnm: I'm finding that the -vf scale parameter is being ignored.. any idea why? http://pastebin.com/sSq5UXPz [01:34] <sh_t> ah, a correction in the pastebin. my $DIMENSIONS is "-vf scale=-1:240" [01:34] <sh_t> outputs the same as the source [04:25] <realazthat> hi [04:26] <realazthat> I got two large x264 files (1GB) [04:26] <realazthat> I need to upload them to the net, was wondering what codec is best for lowering them to reasonable size w/o reducing quality too much [04:26] <realazthat> (advice) [10:20] <someone-noone> hello, i'm writing own video-player and get next problem: [10:20] <someone-noone> my program time from time crashes on av_read_frame (some heap memory error). I'm running on windows with VS 2010 [10:21] <someone-noone> Here is how i do in ReadThread: [10:21] <someone-noone> 1. av_read_frame [10:21] <someone-noone> 2. av_dup_packet [10:21] <someone-noone> 3. if video or audio packet -> push to queue , else av_free_packet [10:21] <someone-noone> in DecodingThread (s): [10:21] <someone-noone> 1. queue->GetPacket [10:22] <someone-noone> 2. Decode packet [10:22] <someone-noone> 3. av_free_packet [10:22] <someone-noone> why can it happening? [11:38] <sarathy> \help [11:43] <madlag> Hi there ! [11:43] <madlag> I described here a problem with ffmpeg truncating videos yesterday. [11:44] <madlag> It looks like I get "Enter command: <target> <time> <command>[ <argument>]" at some random time when I run ffmpeg. [11:44] <madlag> Something like after 15 minutes of encoding. [11:44] <madlag> What is this stuff ? [15:17] <godfath3r_> hello, I'm facing the following issue: I playback a swf file with mplayer and it's duration is wrong. swf originally is 10 seconds but in mplayer it is 4... Any suggestions? This is my mplayers msg http://pastebin.com/XEZDrFxX I also get the same (wrong) duration when I'm trying to encode swf file to avi/mpg with ffmpeg... [15:29] Last message repeated 1 time(s). [17:30] <madlag> Hi. [17:30] <madlag> I used the -d option to get over random freeze of ffmpeg, as it was maybe related to log file freeze ... [17:31] <madlag> It works a bit better, but I have still some issues [17:31] <madlag> For very long videos, from time to time, ffmpeg continue to accept data from input (named pipe) [17:31] <madlag> But it stops writing anything on the output [17:32] <madlag> (output file is a mp4 h264 file) [17:32] <madlag> So in the end, the video is readable, but truncated. [17:33] <madlag> Anybody ? [17:33] <madlag> An older version of ffmpeg just worked fine for now 3 years ... [17:34] <Mavrik> hmm [17:34] <Mavrik> that's probably something you should report to bug tracker [17:34] <madlag> OK. [17:34] <madlag> Do you just have any idea of what I could try ? [17:35] <madlag> I have disabled threading (-threads 1) [17:35] <madlag> I have set loglevel to quiet [17:36] <madlag> The most strange thing is that it continues accepting data, but it just throw it. [18:44] <dv_> christ, FAAD is slow compared to the ffmpeg aac decoder [18:44] <dv_> i expected a difference, but faad is about 7 times slower! [18:47] <mystica555> im not certain, and i think one of the f's stands for 'fabrice', but i consider ffmpeg to be f'n fast mpeg. [18:49] <dv_> also, faad seems to be dog slow [18:50] <jmbc> fast forward mpeg [19:00] <frostschutz> Is ffmpeg by any chance not good with dts? Mplayer works fine for me but ffmpeg/ffplay gives me everything on the left channel. There's a file on samples.mplayerhq.hu that gives me the same problem http://samples.mplayerhq.hu/A-codecs/DTS/lotr_5.1_768.dts - I'm using 0.10.2, btw [19:06] <JEEB> frostschutz, I would guess that's a mixing problem rather than a decoding problem [19:06] <JEEB> if it's a decoding problem, then a bug report is appropriate [19:07] <JEEB> (also trying the current git HEAD is a good'ish idea) [19:10] <ubitux> downmixing to two channels (-ac 2) seems to work with ffmpeg [19:10] <ubitux> i'm not sure how ffplay is supposed to deal with this [19:12] <ubitux> seems to depend on sdl config [19:12] <frostschutz> mplayer lotr_5.1_768.dts sounds much better to me, than ffplay -ac 2 lotr_5.1_768.dts. There's water noise in the background and centered bird chirping with mplayer, ffplay only gives birds on the left. [19:13] <ubitux> it might need a volume boost [19:13] <frostschutz> and when I encode it to flac with ffmpeg, it sounds just as bad with mplayer [19:13] <ubitux> oh? [19:13] <ubitux> ok [19:14] <ubitux> feel free to open an issue then [19:14] <frostschutz> yes, just wondering if anyone besides me could reproduce the issue. I'm on Gentoo so it may always be my system [19:17] <frostschutz> if it's just a matter of channel mapping I should be able to do get it right with -map_channel? [19:18] <frostschutz> on the other hand mplayer says Selected audio codec: [ffdca] afm: ffmpeg (FFmpeg DTS) - so if it's using ffmpeg to decode the audio how can it be wrong when using ffmpeg directly *scratches head* [19:19] <JEEB> mplayer IIRC still has its own copy of ffmpeg in its code tree [19:19] <JEEB> could be a regression [19:19] <JEEB> ...or something else [19:21] <jermy> So, I haven't been near head in a while. Is it possible to use libavfilter to take video with - for example - a stereo and a mono audio stream, and mix it down to a single stereo pair whilst keeping sync with the video? [19:28] <MStoykov> question: is there a way to embed (into the container) subtitles (from srt file for example) into webm video ? [19:34] <JEEB> MStoykov, as far as I know webm still hasn't set itself a subtitle format [19:35] <JEEB> it's basically matroska, so in all theory you could just mux srt or whatever there [19:35] <JEEB> but that would then be against webm specs :D [19:36] <MStoykov> hm QQ so .... html5 <video> is the way ? [20:24] <sprzybilla> is there a simple way to set a PAL resolution adherent to the non-square pixel aspect ratio? [20:45] <Hfuy> Hello [20:45] <Hfuy> I have some videos that I want to rotate 270 degrees, or, equivalently, flip both horizontally and vertically. [20:45] <Hfuy> Google sources differ on the best way to do this - any pointers? [21:03] <Mista_D> I have mov with 8 mono channel tracks, how can I mix them into 1 eight channel track (PCM WAV)? [21:15] <silverrocker> hi, can I ask my question about ffplay here aswell? [21:17] <Hfuy> is "-vf vflip" the appropriate way to mirror something vertically? can I combine it with a horizontal flip? [21:22] <Hfuy> Why would it complain "Error opening filters"? [21:23] <Hfuy> Also, why would it complain "no such filter hflip" when many documents insist there is a horizontal flip filter? [21:26] <Hfuy> It also complains "missing parameter such as bit_rate", but when I attempt to insert a -bit_rate option it claims there is no such option. [21:28] <Hfuy> it doesn't like "width" either, despite asking for "width" [21:30] <sprzybilla> hfuy -vf "filter=setting" [21:30] <Hfuy> what settings do I need [21:31] <Hfuy> it's a vertical flip (there seems to be no horizontal flip available, for some reason) [21:31] <sprzybilla> Hfuy: ohoh sorry -vf "vflip" [21:31] <Hfuy> I don't think that's the problem. [21:31] <Hfuy> It's an "error opening encoder" issue [21:32] <Hfuy> I'm after mp4 output at quality similar to the 3gp input, but I don't know what it needs. [21:32] <Hfuy> It's expressly asking for some options that it then expressly refuses to use. [21:33] <sprzybilla> try setting a constant bitrate like this "-b:v 4000k -minrate 4000k -maxrate 4000k -bufsize 1500k" [21:33] <sprzybilla> there is no bit_rae parameter, bitrate is set differently depending on your desired results [21:34] <Hfuy> Unrecognized option 'b:v' [21:34] <sprzybilla> just use -b [21:34] <Hfuy> Error while opening encoder for output stream #0.1 - maybe incorrect parameters [21:34] <Hfuy> such as bit_rate, rate, width or height [21:34] <sprzybilla> set your size with -s [21:35] <sprzybilla> i.e. "-s 640x360" [21:35] <Hfuy> no change [21:35] <sprzybilla> also, look at the top of the output, is sometimes there will be another error towards the top [21:35] <Hfuy> I love the way it's asking for options it refuses to use [21:36] <Hfuy> http://pastebin.com/9exYN9YC [21:37] <sprzybilla> your output needs to be at the end [21:37] <Hfuy> Unable to find a suitable output format for 'pipe:' [21:38] <sprzybilla> hey a new error we are making progress! [21:38] <sprzybilla> ;) [21:38] <Hfuy> This has to be the single most user-hostile piece of software I've ever come across. [21:39] <Hfuy> OK, I typo'd [21:39] <Hfuy> but if I correct the typo, we're back to "missing parameter" stuff [21:39] <sprzybilla> poop [21:40] <sprzybilla> oh [21:40] <sprzybilla> take the audio out once [21:40] <sprzybilla> -an [21:40] <sprzybilla> [aac @ 02ef3f70]Too many bits per frame requested [21:40] <Hfuy> I do need the audio.l [21:41] <sprzybilla> hehe okay, then do -acodec faac -ab 96k [21:41] <sprzybilla> if you have libfaac [21:41] <Hfuy> I suspect I don't. [21:42] <sprzybilla> taking the audio out would at least tell us if that was the original issue [21:42] <Hfuy> Unknown encoder 'faac' [21:42] <sprzybilla> yeah [21:42] <Hfuy> It will encode with -an [21:42] <sprzybilla> jsut do -an [21:42] <sprzybilla> okay [21:42] <sprzybilla> so that was the problem [21:42] <Hfuy> Is there any way to feature-request slightly less foggy error messages? [21:42] <sprzybilla> lol, well, there was an error in there, it was just hidden [21:43] <sprzybilla> i think it's because the error at the bottom is ffmpeg, the errors for the libraries appear as they happen [21:43] <Hfuy> I don't know what the 3gp file uses for audio. [21:43] Action: Hfuy tries -acodec copy [21:43] <sprzybilla> i.e. just use -acodec aac [21:43] <sprzybilla> oh yeah [21:43] <sprzybilla> that might work too [21:43] <Hfuy> Could not write header for output file #0 (incorrect codec parameters ?) [21:44] <sprzybilla> haha [21:44] <sprzybilla> try aac [21:44] <Hfuy> encoder 'aac' is experimental and might produce bad results. [21:44] <Hfuy> Add '-strict experimental' if you want to use it. [21:44] Action: Hfuy pounds his head against the monitor screen [21:44] <sprzybilla> yeah, its a shitty aac encoder i think, that's why i use faac [21:44] <sprzybilla> here [21:44] <sprzybilla> do ffmpeg -codecs [21:45] <sprzybilla> that will tell you which decoders/encoders you have [21:45] <Hfuy> there are...many [21:45] <Hfuy> Christ on a souped-up harley, this thing supports CINEPAK? [21:45] <sprzybilla> (and all the acceptable parameters for -vcodec -acodec) [21:45] <sprzybilla> be careful [21:46] <sprzybilla> DEVSD [21:46] <sprzybilla> those letters by the name will tell you what your currently compiled ffmpeg can do [21:46] <Hfuy> It supports cinepak, but it doesn't support mp3? [21:46] <sprzybilla> ah - it explains what the letters mean at the top of the output if you're curious [21:46] <Hfuy> this is hopeless! [21:47] <sprzybilla> lol NOOOOOO, your doing an mp4 output? [21:47] <Hfuy> Well it claims it supports mp3 for at least some opertaions. [21:47] <sprzybilla> the most important are the first two letters [21:48] <sprzybilla> D = decodec E = encoder [21:48] <sprzybilla> decoder* [21:48] <Hfuy> Why does ffmpeg not have an mp3 encoder? Isn't that a massive oversight? [21:48] <JEEB> there's libmp3lame [21:48] <Hfuy> there isn't, I tried that [21:48] <JEEB> which is one of the best mp3 encoders around [21:49] <JEEB> it means you haven't built ffmpeg with the library [21:49] <Hfuy> I haven't built ffmpeg at all, I downloaded a binary [21:49] <JEEB> well durr [21:50] <JEEB> the reason why ffmpeg doesn't have encoders for anything you can see in the ogg vorbis encoder that's inside ffmpeg :P [21:50] <JEEB> (not libvorbis) [21:51] <Hfuy> oh, screw this noise [21:51] <Hfuy> I'll do it in Premiere [21:52] <Hfuy> it'll take 3 hours to go through every video file in this folder, doing it all manually, but at least it can produce output in a format invented after 1990. [21:52] <Hfuy> DIdn't it occur to anyone that mp3 might be useful!? [21:52] <JEEB> yes? which is why there's a decoder [21:52] <JEEB> the encoder's not there because there's a library which you can link to ffmpeg [21:52] Action: sprzybilla giggles [21:53] <Hfuy> I'm not a software engineer. [21:53] <JEEB> then you find a better compile, that's all >_> [21:53] <sprzybilla> Hfuy: I know your pain, ffmpeg made me learn all about compiling software [21:53] <sprzybilla> but it was worth it! [21:53] <Hfuy> I suspect that libmp3lame doesn't work because it isn't happy with the codec parameters [21:53] <JEEB> anyways, -acodec aac's current state should be ok at around 192k+ [21:53] <JEEB> Hfuy, is it in the codec listing? [21:53] <Hfuy> but there's no documentation on how to give it the proper parameters. [21:53] <JEEB> also how old is this build of yours? [21:54] <Hfuy> let's just do LAME. [21:55] <Hfuy> it's saying "mp4 @ 01640400]track 1: muxing mp3 at 8000hz is not supported" [21:55] <Hfuy> but I've specified -ab 192k [21:55] <JEEB> ah [21:55] <JEEB> that's not bitrate [21:55] <Hfuy> Oh, sample rate [21:55] <JEEB> yup [21:55] <JEEB> 441000 or 48000 should be ok, or even 22050 [21:55] <Hfuy> is it trying to match the sample rate in the input file, or something? [21:55] <JEEB> yeah [21:55] <JEEB> which is usually what you want [21:55] <Hfuy> Aha. [21:56] <Hfuy> Can I make it not? [21:56] <JEEB> yes [21:56] <JEEB> unfortunately I don't remember the setting for that :P [21:56] <Hfuy> Aurrgh, the pain [21:56] <JEEB> at least ffmpeg has a web site with most settings now [21:56] <sprzybilla> -ar [21:56] <sprzybilla> = samplerate [21:56] <Hfuy> -ar [21:57] <Hfuy> By the wibbly bits of my squeedilyspooch, it works! [21:57] <sprzybilla> hooray! [21:57] <sprzybilla> grats! [21:57] <sprzybilla> feels good don't it? [21:57] <Hfuy> No, it makes me revile and despise opensource software with a new and cutting vitriol. [21:57] <JEEB> just make sure that whatever you're wishing to play that on likes mp3 as audio [21:57] <JEEB> lol [21:57] <Hfuy> How do I make it do all the files in a folder? [21:57] <sprzybilla> now I need to figure out how to make this 720x390 video get into a standard PAL resolution/PAR with padding.. [21:58] <JEEB> Hfuy, make a batch file with a for loop? [21:58] <JEEB> (I will take the approximation that you're on windows) [21:58] <Hfuy> powershell I think [21:58] <Hfuy> OK thanks folks [21:58] <JEEB> powershell is probably less roasting for the brain [22:34] <sprzybilla> is SAR the same thing as Pixel aspect ratio in FFMPEG lingo? [22:37] <Tjoppen> yes. SAR = sample aspect ratio [22:37] <Tjoppen> SAR is the proper term I think [22:39] <sprzybilla> hrm, I have a source with square pixels, but FFMPEG reads a SAR of 1:1 and a DAR of 24:13, which shouldn't be the case according to http://en.wikipedia.org/wiki/Storage_Aspect_Ratio#Storage_Aspect_Ratio [22:39] <sprzybilla> according to that, they should "agree" [22:39] <sprzybilla> by agree I would assume be both 24:13 [22:40] <sprzybilla> since that would mean 24:13/24:13 = 1 [22:40] <sprzybilla> god I hate broadcast standard and their unaquare pixels [22:40] <sprzybilla> and my typing abilities.. [22:43] <Tjoppen> erm, DAR = width/height * SAR [22:43] <Tjoppen> some files have separate width/height + SAR in the container though. this is common with IMX50 MOVs for instance [22:46] <sprzybilla> Tjoppen: where does pixel apect ratio fit into your equation. Again I am referencing that wikipedia article [22:47] <sprzybilla> I need to make a video with a PAR of 16:15 for PAL [22:47] <sprzybilla> and my source is reading differntly than I would think it should [22:47] <sprzybilla> this is some deep stuff, and can get super confusing hehe [22:54] <Tjoppen> 16:15 is typically 720x576. you're getting the resolution right but not the SAR? [22:57] <sprzybilla> Tjoppen: pm'd this might get hairy ;) [23:07] <Diogo> hi this is possible to encode video and generate thumbnails in the same command? [23:08] <Diogo> and another question...what is the best way to detect video 3d? [23:08] <Diogo> i'm using exiftool and mediainfo..but this is possible to detect a video 3d? [23:08] <Diogo> thanks.. [23:27] <markatto> when I use the option `-vf scale='min(500, iw*3/2):-1'` (copy-paste out of the man page) I am getting the error `No such filter: 'iw*3/2):-1'` [23:27] <markatto> any help? [23:28] <frostschutz> tried without space? [23:28] <markatto> yup [23:28] <emerica> passed over a line this morning that said sometimes the commas may need to be escaped [23:29] <markatto> emerica: hmm, escaping the comma seems to work [23:29] <emerica> yay [23:30] <markatto> except now it complains about the height not being a multiple of 2 :( [23:33] <markatto> if the -1 height has a chance of calculating heights that are invalid for my codec it is pretty useless [23:52] <burek> frostschutz, why ` [23:52] <frostschutz> why what? [23:53] <burek> backtick [23:53] <burek> can you please use pastebin.com, to show your command line and its output? [23:58] <sprzybilla> why, with a high bitrate, would my q level be riding at 31? [23:59] <sprzybilla> isn't that supposed to reflect the quality for availabel resources? Or am I misinterpreting? [00:00] --- Thu Mar 29 2012

1 0

ffmpeg-devel.log.20120327
by burek 28 Mar '12

28 Mar '12

[00:29] <durandal_1707> bytestream2_get_byte should return uint8_t [01:14] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r43fd321252 10ffmpeg/libavcodec/targa.c: [01:14] <CIA-17> ffmpeg: targa: Fix input buffer size check. [01:14] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [01:14] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:14] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rd6eef545c1 10ffmpeg/libavcodec/vc1dec.c: [01:14] <CIA-17> ffmpeg: vc1dec: move mquant zero check down. [01:14] <CIA-17> ffmpeg: This way it catches all cases, and prevents later segfaults. [01:14] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [01:14] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:14] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r7f9b42ff9d 10ffmpeg/libavcodec/j2kenc.c: [01:14] <CIA-17> ffmpeg: j2kenc: cosmetics: vertically align AVCodec members. [01:14] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [01:14] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:14] <CIA-17> ffmpeg: 03Paul B Mahol 07master * ra65016205f 10ffmpeg/libavcodec/libopenjpegdec.c: [01:14] <CIA-17> ffmpeg: libopenjpegdec: hack to fix GRAY16 decoding [02:42] <ikrabbe> Hi, I'm trying to get an overview over av_close_input* deprecation. Are there any docs about the libav* API in the ffmpeg tree? [02:43] <funman> docs/APIchanges [02:46] <ikrabbe> funman: yes there the deprecation is mentioned [02:46] <ikrabbe> funman: but the api itself isn't docuemented within ffmpeg tree? [02:47] <funman> possibly not, you have to read avcodec.h and/or git log and figure it on your own [02:51] <ikrabbe> hmm in the libav package there seems to be some documentation. Does ffmpeg provide its own state of libav or does it simply merge in the libav changes? [03:04] <funman> what is libav? [03:05] <Compn> short for libavcodec [03:05] <Compn> :) [03:05] <durandal_1707> michaelni: libopenjpegdec segv on file created with j2kenc [03:05] <Compn> and libavformat, libavfilter, libavutils... [03:05] <funman> yeah that's what i thought too it's not very clear [03:05] <durandal_1707> lav is enough [03:06] <ikrabbe> www.libav.org? [03:06] <funman> it's the same thing wrt doc [03:06] <Compn> ikrabbe : ffmpeg merges most libav changes [03:07] <funman> i.e. api changes are wrongly documented if at all [03:07] <Compn> but libav doesnt merge much of ffmpeg :\ [03:07] <Compn> api may be broken forever :( [03:09] <durandal_1707> @deprecated use avformat_close_input() [03:09] <durandal_1707> it just takes ** instead of * [03:09] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rb62b9df798 10ffmpeg/ffmpeg.c: [03:09] <CIA-17> ffmpeg: ffmpeg: more verbose error messages in case of audio encoding failure. [03:09] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:09] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra0820eaa8a 10ffmpeg/ffmpeg.c: [03:09] <CIA-17> ffmpeg: ffmpeg: fix handling or empty audio packets [03:09] <CIA-17> ffmpeg: Fixes Ticket1131 [03:09] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:09] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8b58f6b5c2 10ffmpeg/libavcodec/rawdec.c: [03:09] <CIA-17> ffmpeg: Revert "raw: move buffer size check up." [03:09] <CIA-17> ffmpeg: This reverts commit cc5dd632cecc5114717d0b90f8c2be162b1c6ee8. [03:09] <CIA-17> ffmpeg: The change was redundant, it has been fixed long ago (422e3a7) [03:09] <CIA-17> ffmpeg: Conflicts: [03:09] <CIA-17> ffmpeg: libavcodec/rawdec.c [03:09] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:10] <durandal_1707> and this one clears looks to clear everything [03:18] <durandal_1707> libopenjpegdec segv an it is not wrapper fault but bug in openjpeg and IMHO this decoder should be marked experimental [03:19] <ohsix> what sort of reply are you looking for, and from whom [03:32] <ramiro> hi [03:34] <ramiro> michaelni: i'm transcoding from wmav2 to wmav2 (don't ask why =) and I got a sample that has 20kbps bitrate. I need to transcode to the same bitrate, but the wma encoder says at least 24000 is needed. where did you get that limit from? [03:36] <adem1> hello [03:38] <michaelni> ramiro, i guess our rate control code failed with lower bitrates [03:39] <michaelni> that is the packts ended up too big [03:39] <michaelni> more aggressive quantization, lower sample rate only mono could lower the limit [03:39] <michaelni> with the existing rate/quant choosing code [03:40] <adem1> this os my first time on gsoc, is there anyone who can help me a bit [03:40] <adem1> ? [03:41] <michaelni> hi adem1 [03:41] <michaelni> FFmpeg sadly has not been accpted by google this year for their gsoc [03:42] <michaelni> they had 400 or so organizations and could only accept 180 and they droped some long time participating onces [03:42] <adem1> ok then sorry [03:42] <michaelni> wait a moment :) [03:43] <michaelni> theres a small chance that ffmpeg might find some donator(s) and could make its own FF-summer of code [03:43] <michaelni> and of course you are welcome to work on anything you have fun to work on as long as you dont mind to do it without getting money [03:44] <michaelni> either way FFMpeg will apply next year again to gsoc so maybe we will be in again then [03:46] <adem1> thank you [03:46] <CIA-17> ffmpeg: 03Paul B Mahol 07master * rddfa3751c0 10ffmpeg/libavcodec/j2kdec.c: [03:46] <CIA-17> ffmpeg: j2kdec: convert to bytestream2 API [03:46] <CIA-17> ffmpeg: Protects against overreads. [03:46] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [03:46] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:46] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r75625c6138 10ffmpeg/libavcodec/j2kdec.c: [03:46] <CIA-17> ffmpeg: j2kdec: Remove useless .pix_fmts. [03:46] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [03:46] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:46] <CIA-17> ffmpeg: 03Paul B Mahol 07master * rebbf10efb0 10ffmpeg/libavcodec/j2kdec.c: [03:46] <CIA-17> ffmpeg: j2kdec: vertically align AVCodec members [03:46] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [03:46] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [03:46] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r0f3886fa7d 10ffmpeg/libavcodec/j2kdec.c: [03:49] <ramiro> michaelni: do you have any suggestions to improve the rate control code so it may go lower than 24kbps? [03:51] <durandal_1707> wtf why -lavdopts lowres=5 doesnt work for mplayer... [03:54] <michaelni> ramiro, use less bits :) [03:54] <michaelni> i think there was a second variant to encode scale factors [03:54] <michaelni> it might be more compact [03:54] <michaelni> a bigger block size might help too in lowering the lowest achiveable point [03:55] <michaelni> and 24000 is maybe not as aggressivly low as it could be, i dont remember [03:57] <michaelni> also make sure the rate/quant control code tests down to the lowest possible quality [03:57] <michaelni> and that it would always reach all coefficients = 0 no matter what input at that lowest quality [03:58] <ramiro> the second variant does scoreL and scoreR, I don't know what difference that makes [03:58] <michaelni> they should match [03:59] <michaelni> these are the quant/rate control code [03:59] <michaelni> one is just a simplified variant of the other IIRC [03:59] <michaelni> its simple bisection [03:59] <michaelni> both actually [04:09] <Compn> shit [04:09] <Compn> forgot to tell adam if he wanted to work for ffmpeg, we could probably sponsor him [04:09] <Compn> :) [04:09] <Compn> oh you told him [04:09] <Compn> good work michaelni :) [04:10] <funman> ramiro: http://download.videolan.org/pub/videolan/vlc/2.0.1/win64/ [04:11] <funman> 2.0 version is built with FFmpeg btw [04:14] <CIA-17> ffmpeg: 03Compn 07master * rab31db0610 10ffmpeg/doc/indevs.texi: doc: clarify dshow input requirements [04:17] <Compn> ramiro : the dshow input doesnt work with regular mingw ? [04:21] <Daemon404> it doesnt? [04:21] <Daemon404> worksforme(tm) [04:24] <funman> what is 'regular' mingw? [04:24] <ramiro> that would be mingw32 [04:24] <Compn> mingw.org mingw is regular [04:24] <darkstarbyte> How would one start making an interface for ffmpeg in C? [04:24] <ramiro> and no, it doesn't. there are no headers [04:24] <Compn> well Daemon404 is being a jerk then [04:24] <Compn> ;P [04:24] <ramiro> there are the headers written by videolan, but they're not part of mingw32 [04:25] <Compn> thanks [04:25] <funman> written by, or stolen by? [04:25] <ramiro> darkstarbyte: it already exists. it's the only one actually [04:25] <ramiro> oh, you mean ffmpeg.c? [04:25] <Compn> darkstarbyte : you start by copying and pasting a bunch of code from another C frontend from the projects page :) [04:25] <ramiro> funman: that's a very thin line [04:25] <funman> ramiro: dshow headers come from wine i think (header says (C) codeweavers) [04:26] <darkstarbyte> I was wondering if there was a header I needed? [04:26] <funman> and Alexandre Julliard [04:26] <Compn> hellokitty.h [04:27] Action: darkstarbyte murders Compn silently like a ninja [04:28] <darkstarbyte> I need it to do one specific task [04:28] <darkstarbyte> and almost the exact same everytime [04:30] <darkstarbyte> So if I were to steal some source code, where would I start? [04:30] <darkstarbyte> (Which project) [04:31] Action: darkstarbyte ninja suicide? [04:32] <Compn> sorry, i uhhh really havent a clue [04:32] <Compn> i'm just here to troll people [04:32] <Compn> someone here will know [04:32] <Compn> they are useful people [04:32] <darkstarbyte> Your good at trolling. [04:33] <darkstarbyte> The only reason I want to do this, is to improve an app someone wrote, except it is in python so I can't understand it. The other program uses [04:33] <darkstarbyte> ffmpeg to convert stuff, but does not have a command line tool which I want to write [04:34] <Compn> i mean uh [04:34] <Compn> what error are you running into when including ffmpeg ? [04:34] <Compn> you said you think you were missing a header [04:34] <darkstarbyte> ??? [04:34] <Compn> which means you get some errors when compiling your program ? [04:34] <darkstarbyte> I just wanted to write a command line tool for a program [04:35] <darkstarbyte> I need some place to start [04:35] <Compn> + [04:36] <darkstarbyte> Nothing [04:36] <Compn> a command line frontend for a command line program [04:36] <Compn> i guess it can be done :) [04:36] <darkstarbyte> no a command line front end for another program that uses ffmpeg [04:36] <Compn> ahhh [04:36] <Compn> sounds very complicated [04:36] <darkstarbyte> yeah [04:37] <darkstarbyte> It converts file to compliant mpeg2 for dvd players [04:37] <darkstarbyte> it authors them, turns them into an iso [04:37] <darkstarbyte> devede is what I am talking about [04:38] <Daemon404> i dont use mingw from mingw.org [04:38] Action: Compn hits head against wall [04:39] <Daemon404> i use mingw-w64, from komisar [04:39] <Daemon404> i didnt think anyoen didnt used mingw-w64 anymore. [04:39] <Compn> ... [04:39] <darkstarbyte> Daemon404, I am sorry I can't help, but how did you bold that word? [04:39] <Daemon404> that varies between clients? >_> [04:40] <Compn> what do you mean bold? [04:40] <funman> 1b2o3l4d ? [04:40] <Compn> ;P [04:40] <darkstarbyte> xchat [04:40] <Daemon404> google it [04:40] <Compn> try ctrl+b [04:40] Action: Daemon404 slaps Compn [04:40] <funman> /bin/sh: 1: toilet: not found [04:40] <funman> oops [04:40] <Daemon404> im trying to promote RTFM [04:40] <Daemon404> ;) [04:41] <Compn> i'm trying to see how darkstarbyte is going to write a c program if he cant read a manual [04:41] <Daemon404> fun fact: i wrote a good chunk of the xchat manual [04:41] <funman> he'll boldly ask other people to write it for him :P [04:41] <Daemon404> and stopped using it right after [04:41] <Compn> Daemon404 : did anyone read it ? [04:42] <Compn> lol burn [04:42] <Daemon404> Compn, of course [04:42] <Daemon404> i am the only one whoever documented all its /set vars [04:42] <Daemon404> someone else is maintaining it now though [04:43] <darkstarbyte> sorry I just didn't know about the bold thing, I had no intention of actually using it. [04:43] <darkstarbyte> I only asked because I didn't know about it. [04:43] <Compn> darkstarbyte : its good idea to ask questions, fast way to learn :) [04:43] <Compn> dont listen to these jerks [04:43] <Daemon404> hey! im not old enough to be jaded yet! [05:09] <durandal_1707> lol jp2 use ftyp from mp4/qt [05:09] <Compn> yes, they were infected [05:34] <ramiro> michaelni: the wma encoder seems to expect a fixed frame_size. but I have some samples that decode using different frame_sizes. what would need to be changed in the encoder to support this? or is it better to just buffer up the samples prior to calling the encode function? [05:36] <ramiro> Zeranoe: hi [05:51] <Compn> wbs : rtsp://208.86.19.32:8000/crtn [05:51] <Compn> wbs : you ever seen h264 in asf over rtsp ? [06:07] <michaelni> ramiro, dunno, i think the code should mostly be able to handle variable bytes per frame if something makes it variable [06:09] <michaelni> i need to sleep a bit now though (send me a mail if you have more questions ill reply tomorrow) [06:24] <Zeranoe> ramiro: Hi Ramiro [12:37] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r5484170ac7 10ffmpeg/libavcodec/rv34.c: (log message trimmed) [12:37] <CIA-17> ffmpeg: rv34: set mb_num_left to 0 after finishing a frame [12:37] <CIA-17> ffmpeg: Prevents running error resilience on a previous frame which will write [12:37] <CIA-17> ffmpeg: to the pic->mb_type[] array of the previous image. The array might [12:37] <CIA-17> ffmpeg: already be re-used for a new image in a subsequent thread, thus cause [12:37] <CIA-17> ffmpeg: two threads to write to the same pic->mb_type[] array, causing a race [12:37] <CIA-17> ffmpeg: condition which can crash in rv34_decode_cbp(), called by [12:37] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r2ad405548b 10ffmpeg/libavcodec/targa.c: [12:37] <CIA-17> ffmpeg: targa: convert to bytestream2. [12:37] <CIA-17> ffmpeg: Also remove some write-only variables or write-only variable [12:37] <CIA-17> ffmpeg: assignments, remove internal colorspace conversion to native [12:37] <CIA-17> ffmpeg: endianness (that can be done by swscale much more efficiently), [12:37] <CIA-17> ffmpeg: and some cosmetics. [12:37] <CIA-17> ffmpeg: 03Mashiat Sarker Shakkhar 07master * r95b192de5d 10ffmpeg/libavcodec/ (vc1.c vc1data.c vc1data.h vc1dec.c): [12:37] <CIA-17> ffmpeg: vc1: Do not read from array if index is invalid. [12:37] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [12:37] <CIA-17> ffmpeg: CC: libav-stable(a)libav.org [12:38] <j-b> michaelni: are there many remaining issue from j00ru ? [12:38] <CIA-17> ffmpeg: This patch allows the user to force flushing of all queued packets [12:38] <CIA-17> ffmpeg: by calling av_interleaved_write_frame() with pkt set to NULL. [12:38] <CIA-17> ffmpeg: Signed-off-by: Jindrich Makovicka <jindrich.makovicka(a)nangu.tv> [12:38] <CIA-17> ffmpeg: Signed-off-by: Martin Storsjö <martin(a)martin.st> [12:38] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r6999f8bcf5 10ffmpeg/: (log message trimmed) [12:38] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master' [12:38] <CIA-17> ffmpeg: * qatar/master: [12:38] <CIA-17> ffmpeg: make av_interleaved_write_frame() flush packets when pkt is NULL [12:38] <CIA-17> ffmpeg: mpegts: Fix dead error checks [12:38] <CIA-17> ffmpeg: vc1: Do not read from array if index is invalid. [12:38] <CIA-17> (1 lines omitted) [13:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rdcd013a535 10ffmpeg/libavformat/omadec.c: [13:36] <CIA-17> ffmpeg: oma: dont over-read buffer [13:36] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [13:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [13:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * re3bd61248f 10ffmpeg/libavcodec/vc1dec.c: [13:36] <CIA-17> ffmpeg: vc1dec: avoid == -1 checks for errors. [13:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [13:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r1acc553e54 10ffmpeg/libavcodec/vc1dec.c: [13:36] <CIA-17> ffmpeg: vc1dec: dont ignore ff_vc1_parse_frame_header_advs return value [13:36] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [13:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [13:36] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rb378a23395 10ffmpeg/libavcodec/vc1dec.c: [13:36] <CIA-17> ffmpeg: vc1dec: sanity check state used in get_mvdata_interlaced() [13:36] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [13:37] <j-b> more vc1 fixes [13:37] <j-b> interesting [15:54] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rdfacef9e73 10ffmpeg/libavcodec/alsdec.c: [15:54] <CIA-17> ffmpeg: alsdec: make sure no invalid opt_order stays in the context. [15:54] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:54] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:54] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r28a613b09b 10ffmpeg/libavcodec/ (h264.c h264.h h264_loopfilter.c svq3.c): [15:54] <CIA-17> ffmpeg: h264: dont mess with chroma planes for grayscale h264. [15:54] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:54] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:31] <CIA-17> ffmpeg: 03Thilo Borgmann 07master * r7ea948bac3 10ffmpeg/libavcodec/alsdec.c: alsdec: check return values. [20:31] <CIA-17> ffmpeg: 03Tim Nicholson 07master * r6fa998c491 10ffmpeg/doc/filters.texi: [20:31] <CIA-17> ffmpeg: Clarify the use of the interl=1|-1 option of the scale filter. [20:31] <CIA-17> ffmpeg: It is not entirely clear that whilst for width and height only an [20:31] <CIA-17> ffmpeg: expression needs to be provided, for interlace the option must [20:31] <CIA-17> ffmpeg: also be given. [20:31] <CIA-17> ffmpeg: It is also unclear that the default is non interlaced aware scaling. [20:31] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:31] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r1cd8510912 10ffmpeg/libavcodec/j2kdec.c: [20:31] <CIA-17> ffmpeg: j2kdec: move ff_j2k_init_tier1_luts() to j2kdec_init() [20:31] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [20:31] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:31] <CIA-17> ffmpeg: 03Paul B Mahol 07master * ra0340dfbd7 10ffmpeg/ (5 files in 3 dirs): [20:31] <CIA-17> ffmpeg: XBM decoder [20:44] <Daemon404> why is searching disabled on git's web ui? [20:44] <funman> thresh disabled a few stuff in gitweb recently [20:44] <Daemon404> any idea why? [20:45] <durandal11707> so that libav folks have hard time porting stuff [20:45] <funman> Daemon404: avoid DoS - not sure if search was part of them though you should ask him [20:45] <Daemon404> durandal11707, if thats true [20:45] <Daemon404> it's pretty sad [20:45] <Daemon404> and hypocritical. [20:46] <Daemon404> funman, oic [20:47] <funman> no it's not true :P [20:49] <ubitux> < durandal11707> so that libav folks have hard time porting stuff // like they care& [20:49] <ubitux> :) [20:59] <iive> they can follow ffmpeg-cvs and get all the patches they want. [21:00] <iive> and they like to rewrite the fixes, to avoid giving credit :) [21:01] <Daemon404> i forward them fixes sometimes (some projects i use are built with libav) [21:01] <Daemon404> theyre usually accepted pretty easily [21:02] <bcoudurier> :) [21:06] <ubitux> they just don't give a shit to most things really [21:06] <Daemon404> thats not really true [21:06] <ubitux> their NIH syndrom is particulary marked [21:06] <ubitux> well, did you follow what was said about lavfi? [21:07] <ubitux> there are at least 3 of them who want to "rewrite everything" [21:07] <Daemon404> i just think they need to bikeshed less. [21:08] <ubitux> i recently realized libav has even less than what i thought about audio processing [21:08] <ubitux> of course they're still stick with the old api which doesn't allow much (and no third app can really use it) [21:08] <ubitux> but they also have nothing in lavfi for audio [21:09] <Daemon404> i dont use either project for much besides encoding/decoding [21:09] <Daemon404> so i cant really say [21:09] <Daemon404> (both are rather inadequate for my usual uses) [21:10] <ubitux> anyway, it's too late now to think libav will rise imo [21:11] <ubitux> they should have removed their blinkers [21:11] <ubitux> and it seems they don't even plan to soon [21:11] <ubitux> so much manpower lost is a shame imo, but well. [21:12] <durandal11707> manpower? [21:13] <iive> yes. The rewrite of prores encoder was fundamental waste of time. [21:13] <ubitux> well, they should just work with us instead of rewriting everything [21:14] <iive> well, they forked because they don't want to work with, you know who :) [21:15] <gnafu> Elvis Presley? [21:15] <gnafu> ;D [21:15] <iive> Presisely [21:15] <gnafu> Hehe, nice. [21:16] <durandal11707> Elvis Presley is still alive? [21:16] <iive> yes, I guess it is relativity time delay. You know, aliens. [00:00] --- Wed Mar 28 2012

1 0

ffmpeg.log.20120327
by burek 28 Mar '12

28 Mar '12

[01:40] <mfwitten> I don't know what this means: Incompatible sample format 's16' for codec 'aac', auto-selecting format 'flt' [01:40] <mfwitten> Isn't ffmpeg supposed to, you know, CONVERT? :-P [01:42] <saste> yes, but different codecs support different sample formats, so ffmpeg is converting to a format which can be dealt by aac [01:42] <saste> or maybe is just a limitation in the implementation, can't say [01:49] <mfwitten> Well, anyway, I get `Error while opening encoder for output stream #0:1 - maybe incorrect parameters such as bit_rate, rate, width or height'; there is a problem with the audio stream, but I don't know what what; I get this error even when I don't specify parameters. [01:49] <mfwitten> Am I to assume that there are no defaults and it's complaining about that? [01:49] <saste> uhm, you usually get some contextual error which explain the error in details [01:50] <saste> try to post a pastebin [04:20] <darkstarbyte> How would one start making an interface for ffmpeg in C? [04:24] <darkstarbyte> maybe I should ask the development channel [04:45] <_klk_> darkstarbyte: ffmpeg is in C... [04:45] <_klk_> do you mean a UI? [04:48] <darkstarbyte> yeah [04:50] <darkstarbyte> I would like to make one of my own [04:53] <_klk_> darkstarbyte: hmm, sorry don't know much about UI programming in C. There are existing UI wrarppers around ffmpeg though - mplayer/mencoder and Handbrake i believe [04:54] <darkstarbyte> oh [06:45] <grepper> darkstarbyte: #ffmpeg-devel is for ffmpeg development only, not what you want [07:29] <darkstarbyte> yeah [07:29] <darkstarbyte> I just thought they would know [08:23] <Wntrvnm> Hello. Is it possible to specify a maximum frame rate when batch-converting videos (but not touch the framerate of ones that have a lower source framerate)? [08:42] <ikrabbe> if, in the ffplay output is written SDL_OpenAudio, I assume ffplay tries to use sdl as outdev (which I disabled during configure) [08:42] <ikrabbe> how can I stop that [08:43] <ikrabbe> or as a first work-around, how can I set alsa as audio output when playing a video with audio [08:45] <ikrabbe> also: using ./configure ... --disable-outdev=sdl, still shows sdl as part of ./configure --list-outdevs, while the ./configure output itself just shows alsa now. [10:03] <ikrabbe> http://paste.pocoo.org/show/571923/ [10:14] <hellop> I need to stream a live event over two cellphone modems. Can anyone think of a way to split an incoming Video/Audio stream (RTP, UDP, RTSP, RTMP, etc..), send each half over a different network connection, then reconstruct and play the stream? [10:25] <ikrabbe> hellop: this will be very tricky to synchronize [10:26] <ikrabbe> I don't even know if there is a transport format that is able to synchronize with another transport stream [10:27] <hellop> Have to do something custom, I guess. [10:31] <hellop> Wonder what some possible strategies are.... [10:32] <hellop> You could record the source stream to a file. Then upload the file later, each half over a different connection. [10:33] <hellop> But, then we have to wait until the end of the event.. [10:35] <hellop> Can ffmpeg can read/write to named pipe. What format should I use? Netcat/udp, a file format? If I can get that working then I can put something in the middle to split the data and recombine it. [10:58] <grepper> hellop: should be able to read from a fifo [11:00] <grepper> you may need to do something like (for example): ffmpeg -pix_fmt yuv420p -f yuv4mpegpipe -i mypipe ... [11:03] <malinens> *** glibc detected *** /usr/bin/ffmpeg: invalid fastbin entry (free): 0x0000000000d5c4c0 *** ffmpeg 0.10 I get this while trying to start transcoding process- sometimes I get this error and sometimes I don't. what should I do about it? [13:43] <hellop> grepper: thanks, I'm looking into that, it looks promising. What about somehow splitting and/or combining a stream? [13:44] <hellop> Could I split into top half and bottom half, or only re-stream every other frame? [14:07] <hellop> Maybe use crop and overlay? [16:19] Action: insider is away: auto-away [16:34] <hellop> Is there a way to save two different output files using avfilter split? From the docs: Pass on the input video to two outputs. Both outputs are identical to the input video. [16:35] <hellop> I've been hacking for some time and can't figure out the syntax to save... It seems like any split must have an overlay. [16:50] Action: insider is back (gone 00:30:45) [17:46] Action: insider is away: auto-away [17:49] Action: insider is back (gone 00:02:44) [17:59] <globus> Does anyone knows how to use libav* to do rtp-streaming of x264 encoded NAL units, I want to use it only for streaming the encoding is done using x264-api. Thanks [18:09] Action: insider is away: auto-away [18:10] Action: sacarasc stabs insider. [18:10] Action: insider is back (gone 00:00:34) [18:10] <ckb> hey guys, I'm encoding a 400k/s 360px width video from a source of 6.2k/s and an aspect ratio of 16:9. When I encode the video, ffmpeg is reporting a 180:10 DAR, why is this? [18:11] <ckb> 180:10 in the encoded video** [18:12] <hellop> I think I found a bug. I can't seem to crop then pad unless I pad greater than the video before cropping. So I can't do this: "crop=iw:ih/2:0:0, pad=iw:ih*2:0:0" But I can do this: "crop=iw:ih/2:0:0, pad=iw:ih*2+2:0:0" [18:14] <iive> hellop: put a scale between crop and pad [18:15] <iive> screenshot may also do the same. [18:15] <iive> oops, sorry, wrong channel. [18:18] <ckb> would if help if someone saw ffmpeg's output? [18:18] <ckb> s/if/it [18:20] <iive> btw, does it crash, or it just undoes the cropped image? [18:20] <hellop> iive, thanks, I have this now: "crop=iw:ih/2:0:0, scale=iw:ih, pad=iw:ih*2:0:0" [18:20] <hellop> So, I guess that resets the height/width for the input on the next filter, eh? [18:27] <ckb> no one? :( [18:29] <iive> hellop: crop could be done via stride(linesize) arithmetic, using the same image, and no full frame copy. The problem is that pad does the opposite, it allocates bigger image and puts the "input" into the center, so it does the exact opposite of crop. [18:30] <iive> there must be option to explicitly clear the added bands. [18:31] <iive> ckb: the 18:1 ratio looks wrong and could be bug, put the text log in some pastebin site. [18:31] <ckb> iive: ok [18:31] <iive> i won't hunt the bug, but somebody else might. I want to see the options you used too. [18:34] <ckb> http://pastie.org/3679100 [18:34] <ckb> iive: the options are the first 2 lines.. then re-executed [18:39] <iive> 180/101 not 180/10 [18:39] <iive> your pixel aspect ratio is 1:1 [18:39] <ckb> hm? [18:39] <iive> the display aspect ratio depends on resolution [18:40] <iive> and you change it [18:40] <ckb> So it's not supposed to stay consistant? [18:40] <iive> your par is 1:1 [18:40] <iive> it is ok [18:42] <ckb> thank you [18:42] <ckb> so 16:9 is dependent on the resolution... [18:49] <hellop> iive, since you're here, can you comment on using -vf split? Is there a way write to disk the two "outputs"? Or do I have to always use overlay to combine them back into one output, before I can write to disk? [18:54] <hellop> If you look in the docs for split it says: "Pass on the input video to two outputs." But, using the example code always gives error about "the filter out not connected to any source". I tried a bunch of different guesses. [18:58] <iive> hellop: it should be possible, but i'm not familiar with the syntax [19:01] <iive> in the past ffmpeg was able to encode multiple different codecs taking the same image [19:02] <hellop> yes, I can do that just by specifying new options after the first output file in the command line. Just can't get out the two streams from split. Can only merge them, then output. [19:03] <hellop> Also can't run two different -vf strings in one command line. [19:14] <iive> hellop: i'm sure saste would know, hope he answers :) [19:14] <iive> btw, you can fill the crop+pad as bug in the bugtracker, it is still bug. [19:17] <saste> hellop: what's the bug? [19:23] <globus> Does anyone knows how to use libav* to do rtp-streaming of x264 encoded NAL units, I want to use it only for streaming the encoding is done using x264-api. Thanks [20:46] <madlag> Hi ! [20:47] <madlag> I am getting a strange hang using ffmpeg + x264 on very long videos created using a pipe for audio and for video inputs. [20:47] <madlag> Before trying to reproduce it, I wanted to know if somebody had the same issue. [20:48] <madlag> (it's fully deterministic) [20:49] <madlag> The version is 0.10 [20:50] <madlag> The process just don't exit. But the file itself is ok. [21:04] <hellop> I get "Error initializing filter 'movie' with args 'udp://127.0.0.1:10000'", is it possible to specify a live stream for the movie paramater for a filter chain? [21:04] <hellop> the command: ffmpeg -re -f mpegts -i udp://127.0.0.1:10000 -vf "movie=udp://127.0.0.1:10000 [logo];[in][logo] overlay=0:main_h/2 [out]" -vcodec libx264 -vprofile default -vprofile baseline -f rtsp rtsp://127.0.0.1:1935/live/myStream [21:10] <hellop> Docs say "can be a stream accessed by some protocol" [21:24] <sudo_dirk> hi. can anybody help me wit a command line issue? i want to use ffmpeg for an open source project... [21:25] <sudo_dirk> i want to start ffmpeg from inside python and want to extract a single jpeg from a video file. [21:25] <sudo_dirk> i found: ffmpeg -vframes 1 -ss 0.5 -i cimg0868.avi pipe:1 [21:26] <sudo_dirk> and get: "nable to find a suitable output format for 'pipe:1'" [21:26] <sudo_dirk> +U(nable) [21:26] <sudo_dirk> i want to catch the output from stdout then [21:30] <hellop> Try naming the pipe: cimg0868.pipe.avi so ffmpeg will recognize the format. [21:30] <sacarasc> Or use -f [21:34] <sudo_dirk> sacarasc: what format do i need after -f. jpg or jpeg is not valid. [21:35] <sacarasc> Anything that's listed in ffmpeg -formats as being encodable should work. [21:38] <sudo_dirk> thanks -f null worked fine ;-) [22:11] <hellop> Ok guys. Lets brainstorm here. How can I combine two live streams into one? I can merge 1 live and one file like this: ffmpeg -re -f mpegts -i udp://127.0.0.1:10000 -vf "movie=samplebottom.mp4 [logo];[in][logo] overlay=0:main_h/2 [out]" -vcodec libx264 -vprofile default -vprofile baseline -f rtsp rtsp://127.0.0.1:1935/live/myStream [22:11] <hellop> What are my options? Try to read the second udp stream to a named pipe, then open that in the filter chain? [22:13] <hellop> I tried ffmpeg -i samplebottom.mpg namedpipe.mp4.. But ffmpeg just hangs trying to read it.. any ideas? [22:33] <hellop> I can do all sorts of stuff now. I can send a UDP stream to another computer. I can receive it and send it to a named pipe. I can read the named pipe and send via RTSP and play it. But how do I get it into the filter chain? [23:20] <hellop> I find it weird that the filter chain an read in a live stream from a hardware source like a camera, but can't read a stream. Can I simulate a hard ware source using a live stream? [00:00] --- Wed Mar 28 2012

1 0

ffmpeg-devel.log.20120326
by burek 27 Mar '12

27 Mar '12

[00:17] <durandal_1707> nice you can make openjpeg encoding/decoding faster at expense of inrease in file size [00:59] <durandal_1707> michaelni: gray16 -> anything is broken in swscale [01:02] <durandal_1707> or not, looks like strange issue in openjpeg decoder [01:04] <durandal_1707> yup, it is definetly bug in libopenjpeg decoder [01:06] <CIA-17> ffmpeg: 03Aaron Colwell 07master * raa9c2e41b6 10ffmpeg/libavcodec/pthread.c: [01:06] <CIA-17> ffmpeg: pthread : Remove lock/unlock pairs in worker loop to avoid unexpected state changes. [01:06] <CIA-17> ffmpeg: Reviewed-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [01:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:06] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r25b67dc765 10ffmpeg/libavcodec/pngdec.c: [01:06] <CIA-17> ffmpeg: png: fix warning about discarding qualifiers [-Wincompatible-pointer-types] [01:06] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [01:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:06] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r40821980ae 10ffmpeg/ (libavfilter/vf_hflip.c tests/ref/lavfi/pixfmts_hflip): [01:06] <CIA-17> ffmpeg: vf_hflip: accept PIX_FMT_RGB444 and PIX_FMT_BGR444 pixel formats [01:06] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [01:06] <CIA-17> ffmpeg: added fate checksum update [01:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [02:26] <durandal_1707> i get segv for this sample: http://samples.ffmpeg.org/jpeg2000/jpeg2000_mxf_first_10mb.mxf [02:34] <durandal_1707> michaelni: av_pix_fmt_descriptors have holes and they are not filled with 0 [02:42] <durandal_1707> michaelni: shouldn't hflip work for any pix_fmt like vflip? [02:45] <michaelni> YUYV shoudl fail [02:50] <durandal_1707> so jpeg2000 stream in mxf is not supported? [02:57] <michaelni> hflip should fail with YUYV, i dunno about j2k-mxf [02:57] <durandal_1707> got hflip part already ;) [02:58] <durandal_1707> j2k in mxf is not supported, and sample i mentioned is causing segv [02:59] <michaelni> i can confirm that [02:59] <durandal_1707> libav does not segv [03:01] <michaelni> ill fix the segv [03:05] <michaelni> segv fixed, will be in my next push [03:07] <michaelni> and libav does crash on the file [03:07] <durandal_1707> not for me [03:07] <michaelni> ==21862== ERROR: AddressSanitizer crashed on unknown address 0x000000000010 (pc 0x000000927be8 sp 0x7ff7fc2a3160 bp 0x7ff7fc2a32d0 ax 0x100000000000 T3) [03:08] <durandal_1707> ahh, it is lucky [03:08] <michaelni> does it actually display anything for you with any tool ? [03:10] <durandal_1707> tools here are at minimum [03:38] <durandal_1707> libopenjpeg decoder is broken and it likes to segfault [04:05] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r33f39c02aa 10ffmpeg/libavformat/mxfdec.c: [04:05] <CIA-17> ffmpeg: mxfdec: pass correct context to av_timecode_init() [04:05] <CIA-17> ffmpeg: fixes null ptr dereference [04:05] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:05] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rc855ece101 10ffmpeg/libavcodec/ (indeo5.c ivi_common.c): [04:05] <CIA-17> ffmpeg: indeo5: check motion vectors. [04:05] <CIA-17> ffmpeg: fixes out of frame reading [04:05] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [04:05] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:05] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r140c54ea0b 10ffmpeg/libavcodec/indeo4.c: [04:05] <CIA-17> ffmpeg: indeo4: better mv check [04:06] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [04:31] <durandal_1707> michaelni: actually for that sample which causes segv libopenjpeg managed to show something [05:45] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r8e45005a84 10ffmpeg/libavcodec/libopenjpegenc.c: [05:45] <CIA-17> ffmpeg: libopenjpegenc: use pix_fmt descriptor to get bpp [05:45] <CIA-17> ffmpeg: This will also simplify adding support for RGB565 and similar. [05:45] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:45] <CIA-17> ffmpeg: Reviewed-by: Michael Bradshaw <mbradshaw(a)sorensonmedia.com> [05:45] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:45] <CIA-17> ffmpeg: 03Paul B Mahol 07master * rdec8eb6b29 10ffmpeg/libavcodec/libopenjpegenc.c: [05:45] <CIA-17> ffmpeg: libopenjpegenc: merge common code [05:45] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:45] <CIA-17> ffmpeg: Reviewed-by: Michael Bradshaw <mbradshaw(a)sorensonmedia.com> [05:45] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:45] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r67d5fcc989 10ffmpeg/libavcodec/libopenjpegenc.c: [05:45] <CIA-17> ffmpeg: libopenjpegenc: allow changing encoding parameters [05:45] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:45] <CIA-17> ffmpeg: Reviewed-by: Michael Bradshaw <mbradshaw(a)sorensonmedia.com> [05:45] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:45] <CIA-17> ffmpeg: libopenjpegenc: use pix_fmt descriptor to get numcomps [05:45] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:45] <CIA-17> ffmpeg: Reviewed-by: Michael Bradshaw <mbradshaw(a)sorensonmedia.com> [05:45] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:45] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r3dc0b9d6da 10ffmpeg/libavcodec/libopenjpegdec.c: [05:45] <CIA-17> ffmpeg: libopenjpegdec: refactor some code and fix memory leaks [05:45] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:45] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:45] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r02fb320ada 10ffmpeg/libavcodec/libopenjpegdec.c: [05:45] <CIA-17> ffmpeg: libopenjpegdec: always check image because decoding may still fail [05:45] <CIA-17> ffmpeg: Fixes bunch of segfaults. [05:46] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:46] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [10:18] <CIA-17> ffmpeg: 03Martin Storsjö 07release/0.8 * rceeaf42451 10ffmpeg/libavcodec/avcodec.h: (log message trimmed) [10:18] <CIA-17> ffmpeg: avcodec: Remove a misplaced and useless attribute_deprecated [10:18] <CIA-17> ffmpeg: If attribute_deprecated is used in an enum declaration, it [10:18] <CIA-17> ffmpeg: should follow the 'enum' keyword, otherwise it's ignored [10:18] <CIA-17> ffmpeg: silently. This is the only case of attribute_deprecated for [10:18] <CIA-17> ffmpeg: enum declarations currently. [10:18] <CIA-17> ffmpeg: Currently, this attribute_deprecated doesn't have any effect. [15:47] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r3b370abf16 10ffmpeg/libavformat/mxfdec.c: [15:47] <CIA-17> ffmpeg: mxfdec: fix off by one error. [15:47] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:47] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:47] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r7d74aaf698 10ffmpeg/libavcodec/qdm2.c: [15:47] <CIA-17> ffmpeg: qdm2dec: fix out of array read [15:47] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [15:47] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:47] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rcece491daa 10ffmpeg/libavcodec/qdm2.c: [15:47] <CIA-17> ffmpeg: qdm2: Fix out of array writes [15:47] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:47] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rfe91becc2d 10ffmpeg/libavcodec/qdm2.c: [15:47] <CIA-17> ffmpeg: qdm2: fix out of array read [15:47] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [15:54] <durandal_1707> michaelni: looks like we need implementing icc profile [15:55] <j-b> for? [15:56] <durandal_1707> jp2 use this [15:59] <j-b> brrr [15:59] <j-b> in XYZ? [15:59] <durandal_1707> jp2 can store anything [15:59] <durandal_1707> and icc describe what colorpsace is stored [15:59] <durandal_1707> like 12bit gray [16:01] <durandal_1707> does anyone know link to some mj2 samples? [16:01] <j-b> I do [16:01] <j-b> and I will feed you soon with DCP files [16:01] <j-b> from actual theatres [16:06] <durandal_1707> there is hardware which can display XYZ? [16:07] <j-b> theatre projectors [16:08] <Tjoppen> probably only a subset of XYZ, but makes more sense than feeding them sRGB [16:14] <cbsrobot> I don't think the projectors can show xyz colors [16:15] <j-b> I don't think so either. but they take that as input,IIRC [16:15] <cbsrobot> the dci specification use xyz colorspace in order to not use a limited colorspace - for futrue usage [16:16] <cbsrobot> I think the hardware converts it to sRGB for the projection - but yes, the files need to be in XYZ [16:17] <j-b> I am just wondering about that, since we got so many requests about DCP playback since a few weeks [16:18] <cbsrobot> well DCP is the new 35mm - so .... [16:18] <durandal_1707> j-b: perhaps they need XYZ -> rgb/yuv conversion [16:19] <j-b> durandal_1707: well, so far, the latest VLC displays a nice green screen on 2.0.1 on Mac [16:19] <durandal_1707> how many bits XYZ use? [16:19] <j-b> 12 [16:19] <durandal_1707> not 16? [16:19] <j-b> using lavf MXF demuxer [16:19] <cbsrobot> ffplay shows the DCP nicely [16:19] <j-b> and lavc openjpeg wrapper decoder [16:20] <cbsrobot> no only 12bit [16:20] <j-b> cbsrobot: it could very well be a VLC integration issue (that is what I assumed, first) [16:20] <j-b> but it could be the machine is too slow. [16:20] <durandal_1707> hmm, libopenjpeg wrapper doesn't do colorspace conversion [16:20] <cbsrobot> j-b: I think I sent you the screenshot ... http://bayimg.com/dakIkAADK [16:21] <cbsrobot> although ist mplayer - but ffplay shows it aswell [16:21] <j-b> I do not remember this screenshot [16:21] <cbsrobot> it's from a few month ago [16:21] <j-b> and trying to compile against openjpeg 1.5 did not work. [16:22] <cbsrobot> j-b I have a patch [16:22] <j-b> configure rejected it, or something [16:22] <cbsrobot> http://pastebin.com/sgffmm84 [16:22] <j-b> openjpeg 1.5 was supposed to be a bit less slow [16:22] <cbsrobot> for openjpeg 1.5 [16:22] <j-b> ohhh [16:22] <durandal_1707> headers for both versions are same [16:23] <durandal_1707> and my ffmpeg build use 1.5 [16:23] <cbsrobot> j-b yeah - one or two milliseconds maybe [16:23] <cbsrobot> or even half [16:23] <j-b> If we have a way to display that with powerful machines, I might find the money to sponsor more ASM/development around that. [16:24] <cbsrobot> j-b: that would be great [16:25] <j-b> well, depends on what is needed :_ [16:25] <durandal_1707> it looks to me conversion would need poverfull machine anyway [16:25] <j-b> conversion? [16:26] <durandal_1707> xyz-> ... [16:26] <j-b> should be doable in GLSL [16:27] <durandal_1707> you mean outside of sws? [16:27] <j-b> yes [16:28] <j-b> I need decoding to YUV 12bits or something that I can convert to RGB in GLSL shaders [16:28] <durandal_1707> the only remaining problem i see is XYZ detection for which ICC is needed [16:28] <durandal_1707> if j2k stream actually have ICC [16:28] <cbsrobot> durandal_1707: could be a command line switch - no ? [16:29] <durandal_1707> or where in the file is stored informationn about used colorspace? [16:29] <cbsrobot> I think it's not stored [16:30] <cbsrobot> but all files using cinema_*k profile are in xyz [16:31] <j-b> cbsrobot: no idea what kind of money is needed [16:35] <CIA-17> ffmpeg: 03Thilo Borgmann 07master * r04dc7e7e5e 10ffmpeg/libavcodec/alsdec.c: [16:35] <CIA-17> ffmpeg: ALS error message pretty print [16:35] <CIA-17> ffmpeg: One prettier message. [16:35] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [16:35] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r3f26a87488 10ffmpeg/libavcodec/libopenjpegenc.c: [16:35] <CIA-17> ffmpeg: libopenjpegenc: remove leftover because older version of patch being applied [16:35] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [16:35] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [16:35] <CIA-17> ffmpeg: 03Thilo Borgmann 07master * r5afd9a80db 10ffmpeg/libavcodec/alsdec.c: [16:35] <CIA-17> ffmpeg: als: consistent messages [16:35] <CIA-17> ffmpeg: "." for warnings, "!" for errors. [16:35] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [16:37] <cbsrobot> converting a 5 min 1920x2080 prores file to dcp takes about 40 min ... lol [16:37] <cbsrobot> *1920x1080 [16:37] <durandal_1707> you mean using libopenjpeg encoder? [16:39] <cbsrobot> yeah its more like prores --(ffmpeg)-->16bit tiff --(opendcp using openjpeg)--> dcp [16:41] <durandal_1707> hmm what needs doing to make something like this: prores --(ffmpeg)--> dcp [16:41] <cbsrobot> yes of course [16:42] <cbsrobot> dcp is more like a package - there is mxf inside - so [16:56] Action: durandal_1707 gonna fix j2kdec color_space detection for jp2 [17:02] Action: Compn guesses all of this jp2k stuff will break something in mplayer codecs.conf [17:03] <Compn> er nm, i'm thinking ffj2k :) [17:03] <gnafu> s/mplayer codecs.conf/something that's already broken/ [17:03] Action: gnafu ducks. [17:21] <durandal_1707> sucks, it sucks that you need to completly rewrite j2kdec before you start fixing [17:21] <durandal_1707> it implements own get_bits() braindead [17:31] <durandal_1707> michaelni: when get_bits stuff where introduced in FFmpeg? [17:38] <michaelni> durandal_1707, the code in j2kdec probably with the first j2kdec version, the code in libavcodec get bist probably with the first ffmpeg checkin [20:04] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * rfca62f0500 10ffmpeg/ (libavcodec/flicvideo.c libavformat/isom.c): [20:04] <CIA-17> ffmpeg: Support FLIC in MOV. [20:04] <CIA-17> ffmpeg: Fixes trac issue #626. [20:04] <CIA-17> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [20:04] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * r699c61d2f0 10ffmpeg/libavformat/utils.c: (log message trimmed) [20:04] <CIA-17> ffmpeg: Handle 0-size sync packets when only parsing headers. [20:04] <CIA-17> ffmpeg: Currently, the duration of those packets is just discarded [20:04] <CIA-17> ffmpeg: when enabling parsing, thus the output of the Metal Gear Solid [20:04] <CIA-17> ffmpeg: demuxer breaks completely when just setting AVSTREAM_PARSE_HEADERS. [20:04] <CIA-17> ffmpeg: The result will not be correct if a parser creates a delay even [20:04] <CIA-17> ffmpeg: with PARSER_FLAG_COMPLETE_FRAMES and there might be other cases [20:04] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * r60497cb984 10ffmpeg/libavformat/mgsts.c: [20:04] <CIA-17> ffmpeg: Enable generic index/seeking for mgsts demuxer. [20:04] <CIA-17> ffmpeg: This depends on the proposed parser change for 0-size packets [20:39] <juanmabc> Hi!, i got a flv vbr video going slowly issue (audio ok), so i got pts and i do: if(pts * av_q2d(media.video.codec->time_base) * 1000000.0 <= av_gettime() - media->masterclock) update_gltexture [20:43] <juanmabc> if pts is in time_base units and this is in seconds this should be ok, but doesn't, it's like time_base is wrong (?) [20:57] <michaelni> juanmabc, there are multiple timebases :) [20:58] <juanmabc> could you explain yourself? [20:58] <michaelni> AVFormatContext.streams[].time_base and AVCodecContext.time_base [20:58] <michaelni> one is from the demuxer layer and one codec layer [20:59] <juanmabc> so i use codec layer should i use demuxer layer? [20:59] <michaelni> if your pts are from the demuxer layer (AVPackets) yes [20:59] <juanmabc> nice, i'll give it a try [20:59] <juanmabc> (hope it solves seeking issue with delta frames ;P) [20:59] <juanmabc> thanks [21:00] <michaelni> np [21:02] <juanmabc> yep, seems to do it, yay :D [21:06] <juanmabc> mmm, nop, despite this solving sync, my seeking unsync issue persists [21:08] <juanmabc> i use avformat_seek_file, but the pts of the frame is weird, seems a lot later of audio, and seems to come from the first non delta video frame [21:10] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rb2af057a36 10ffmpeg/libavcodec/smacker.c: smacker: convert palette and header reading to bytestream2. [21:10] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rc3bbd0b53b 10ffmpeg/libavcodec/alac.c: alac: convert extradata reading to bytestream2. [21:10] <juanmabc> mmm, i use avcodec_flush_buffers [21:10] <CIA-17> ffmpeg: 03Jindrich Makovicka 07master * r3fadb29baf 10ffmpeg/libavformat/mpegtsenc.c: [21:10] <CIA-17> ffmpeg: mpegtsenc: allow user triggered PES packet flushing [21:10] <CIA-17> ffmpeg: Signed-off-by: Jindrich Makovicka <jindrich.makovicka(a)nangu.tv> [21:10] <CIA-17> ffmpeg: Signed-off-by: Martin Storsjö <martin(a)martin.st> [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * r9c47f2b294 10ffmpeg/doc/APIchanges: APIchanges: mark the place where 0.7 was cut. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * rdca9c81d82 10ffmpeg/libavformat/avformat.h: lavf doxy: clarify that an AVPacket contains encoded data. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * rf58b8cc3e3 10ffmpeg/libavformat/avformat.h: lavf doxy: add some basic documentation about reading from the demuxer. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * re44ada129c 10ffmpeg/libavformat/avformat.h: [21:10] <CIA-17> ffmpeg: lavf doxy: improve AVStream.time_base doxy. [21:10] <CIA-17> ffmpeg: Remove confusing sentence that implied the user should set the timebase. [21:10] <CIA-17> ffmpeg: Elaborate on how the timebase is set for muxing. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * r10fa4ff7bc 10ffmpeg/libavformat/avformat.h: lavf doxy: document passing options to demuxers. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * r967923abd1 10ffmpeg/libavformat/avformat.h: lavf doxy: expand AVStream.codec doxy. [21:10] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r6a7c5312d8 10ffmpeg/configure: build: drop some unnecessary dependencies from the H.264 parser [21:10] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r193d7eea95 10ffmpeg/ (configure libavutil/Makefile libavutil/lzo.c): libavutil: Remove broken and pointless lzo test program. [21:10] <CIA-17> ffmpeg: arch subdirectory Makefiles, just putting them in place is enough. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * r5d20e7b7ea 10ffmpeg/doc/APIchanges: APIchanges: fill in missing dates and hashes. [21:10] <CIA-17> ffmpeg: 03Anton Khirnov 07master * r75bdd55e77 10ffmpeg/doc/APIchanges: APIchanges: mark the place where 0.8 was cut. [21:11] <juanmabc> michaelni: do you know what could mess the pts on seek ? [21:18] <michaelni> juanmabc, check if it works with ffplay and if it does check what it does different [21:31] <funman> nefrir: did you work on mpeg4 with dxva2 already? [21:31] <juanmabc> ok, i'm allways with it as help code, still this issue persist, gonna recheck [21:33] <juanmabc> it's like there is some trash on the first read packets for video [21:41] <durandal_1707> -win1 [21:47] <durandal_1707> is there way to ban bug reporters? [22:27] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * ra84851bef8 10ffmpeg/libavcodec/indeo3.c: [22:27] <CIA-17> ffmpeg: indeo3dec: check mv bitstream pointer [22:27] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [22:27] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:27] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8a521d57b3 10ffmpeg/libavcodec/indeo3.c: [22:27] <CIA-17> ffmpeg: indeo3dec: Fix end pointer. [22:27] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [22:27] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:45] <ubitux> michaelni: ok to update RELEASE to 0.10.2.git? [22:46] <michaelni> sure [22:48] <CIA-17> ffmpeg: 03Clément BSsch 07master * r21d650cb69 10ffmpeg/ffmpeg.c: [22:48] <CIA-17> ffmpeg: ffmpeg: more expressive sample rate automatic selection. [22:48] <CIA-17> ffmpeg: Output now lists the available sample rates with commands like [22:48] <CIA-17> ffmpeg: ffmpeg -f lavfi -i aevalsrc=0 -ar 20000 -y /tmp/out.mp3 [22:48] <CIA-17> ffmpeg: 03Clément BSsch 07master * ra67d9cfa58 10ffmpeg/ (4 files in 2 dirs): lavfi: add avfilter_fill_frame_from_{audio_,}buffer_ref(). [22:48] <CIA-17> ffmpeg: 03Clément BSsch 07master * r8f0d9b43be 10ffmpeg/RELEASE: RELEASE: bump to current version (0.10.2). [22:48] <CIA-17> ffmpeg: 03Clément BSsch 07master * r29679f4a37 10ffmpeg/libavcodec/ (aacenc.c libvo-aacenc.c): aac: reference supported sample rates in encoders. [22:48] <CIA-17> ffmpeg: 03Clément BSsch 07master * rfb8469dd7e 10ffmpeg/doc/APIchanges: APIchanges: fill in missing dates and hashes. [22:51] <ubitux> i wonder if there is no "auto hash fill" [23:15] <teratorn> ffs AVDictionary is just the worst thing ever [23:15] <durandal_1707> why? [23:15] <teratorn> is it possible to use the new tree.c/h code to set codec options [23:16] <teratorn> durandal_1707: it's the most brain-dead key/value datastructure imaginable :) [23:16] <ubitux> why? [23:16] <ubitux> you mean it's not optimized? [23:16] <teratorn> it's an array, yeah [23:16] <teratorn> the documentation says its deprecated and slow [23:17] <teratorn> *shrug* [23:17] <ubitux> do you have performance issues at option parsing? [23:17] <teratorn> I just felt like ranting [23:17] <teratorn> nope, it's fine [23:17] <teratorn> but I think a hash-table would even be easier to implement.. but the tree is probably faster to iterate [23:18] <ubitux> iirc michaelni tried to convince the author about this [23:18] <ubitux> not sure about the outcome [23:18] <ubitux> certainly some kind of blinkers issue :) [23:18] <teratorn> with a hastable you have to iterate all buckets to iterate the key/values, even though most buckets are empty [23:19] <teratorn> so the binary tree might be nice for a basic key/value store [23:19] <teratorn> anyway, just ranting... [23:19] <ubitux> feel like sending a patch to make AVDictionary use AVTree? [23:19] <teratorn> and wonder if the tree can actually be used for setting codec options [23:19] <teratorn> ubitux: hrmph. possibly [23:20] <teratorn> ubitux: I need to get some code working first, but it might not be too hard to do assuming AVTree works [00:00] --- Tue Mar 27 2012

1 0

ffmpeg.log.20120326
by burek 27 Mar '12

27 Mar '12

[03:54] <grumpy_> Hello I hope someone can help. I get the following:"ERROR: libmp3lame >= 3.98.3 not found" then :"If you think configure made a mistake, make sure you are using the latest version from Git. If the latest version fails, report the problem to the [03:54] <grumpy_> ffmpeg-user(a)ffmpeg.org mailing list or IRC #ffmpeg on irc.freenode.net. Include the log file "config.log" produced by configure as this will help solving the problem." I have no idea how to get the "config.log" can anyone help? [03:56] <grumpy_> I'm cutting and pasting from "ubuntuforums.org/showpost.php?p=11483157&postcount=1945" "HOWTO: Install and use the latest FFmpeg and x264 " [03:58] <grumpy_> I'm on no. 7: cd git clone --depth 1 git://source.ffmpeg.org/ffmpeg cd ffmpeg ./configure --enable-gpl --enable-libfaac --enable-libmp3lame --enable-libopencore-amrnb \ --enable-libopencore-amrwb --enable-libtheora --enable-libvorbis --enable-libvpx \ --enable-libx264 --enable-nonfree --enable-version3 --enable-x11grab make sudo checkinstall --pkgname=ffmpeg --pkgversion="5:$(./version.sh)" --backup=no \ --deldoc=yes --defa [03:58] <grumpy_> ult hash x264 ffmpeg ffplay ffprobe" and I get this when I enter .config... This is my second reinstall of ubuntu I thought I messed it up the first time. [04:24] <grumpy_> can anyone help? [08:15] <linuxdev> Hi, i'm getting "[libx264 @ 0xa6ffc00] Error setting profile baseline" by using "-profile baseline", this used to work in a previous ffmpeg version. Does anyone know the new syntax? [10:34] <ikrabbe> what does "DTS -1, next:-200000 st:0 invalid droping" on encoding of a stream means? Actually the audio stream is dropped? (pastbin follows) [10:35] <ikrabbe> http://paste.pocoo.org/show/571376/ [10:36] <ikrabbe> the effect of this problem seems to be, that the audio stream #0.1 is dropped on output [10:44] <ikrabbe> I can convert the mic.wav to mic.aac seperately [10:44] <ikrabbe> but the error still happens on -i mic.aac -acodec copy [10:57] <ikrabbe> hmm the problem seems the time measurement, as without shortest it works [10:59] <ikrabbe> so: [10:59] <ikrabbe> which time rules, without shortest? a- the first stream, b- the longest stream? [11:01] <ikrabbe> it seems to be the longest then [13:25] <andrewebdev> Does anyone know what the prefix is for specifying a custom libmp3lame path is? I'm currently combiling with --enable-libmp3lame ... but I dont have it installed globally. [13:53] <ikrabbe> I would try with --extra-cflags [13:53] <ikrabbe> I don't see any extra-ldflags in configure [13:54] <ikrabbe> ah wait, there is an --extra-ldflags actually [13:57] <ikrabbe> andrewebdev: so simply add --extra-cflags and --extra-ldflags should do the job and as far as I understand the configure script, these flags can be used multiple times without negative side effects [14:21] <andrewebdev> ikrabbe, thanks, I actually found that out just a couple of mins ago :) [15:29] <ikrabbe> luckily the ffmpeg configure is no autoconf script, so it's readable :) [15:45] <orangey1> hello all! [15:46] <orangey1> I'm trying to use ffmpeg (?avconv??) to resync a file. I would like to avoid recompressing the video / audio, though. How do I get it to just fix the audio / video location, and do nothing else? [15:47] <ikrabbe> orangey1: I would first try with -acodec copy, -vcodec copy [15:47] <orangey1> Currently, this is the command I'm trying, but it compresses the video and audio to something else: avconv -i in.avi -itsoffset -0.3 -i in.avi -map 1:0 -map 0:1 out.avi [15:47] <orangey1> ikrabbe: thank you. [15:47] <ikrabbe> no guarantees, I'm just a noob user too [15:48] <orangey1> ikrabbe: Worked!!!! [15:48] <ikrabbe> hmm, maybe I'm not that noobish anymore :D [15:48] Action: orangey1 bows [15:49] <orangey1> ooh. and wickedly fast.. 500 fps over my NAS [15:49] <orangey1> for 1080p [15:59] <orangey1> aaah [15:59] <orangey1> this problem is going to have me pull out my hair. [15:59] <orangey1> initially I thought it was 0.3 seconds off no matter what [15:59] <orangey1> but it looks like the sync problem is variable [16:10] <orangey1> ah. OK, so, the sync issue in the original file *is* 0.3 seconds *no matter what* [16:10] <orangey1> however, once I render it with: avconv -i in.avi -itsoffset -0.3 -i in.avi -map 1:0 -map 0:1 -acodec copy -vcodec copy out.avi [16:11] <orangey1> it becomes a *variable* sync problem [16:11] <ikrabbe> orangey1: maybe avi is not the best format to solve sync problems [16:11] <orangey1> ikrabbe: I thought avi was just a container? [16:11] <orangey1> ikrabbe: what do you suggest? [16:11] <orangey1> mp4? [16:11] <ikrabbe> though that leads you to reencoding again [16:12] <orangey1> ikrabbe: At this point, no problem if it'll work. [16:12] <orangey1> I can also re-render [16:12] <orangey1> which isn't a big deal. [16:13] <ikrabbe> for sync problem it might be best to demux the audio and video into mp2 video and some audio format (mp3 or aac, or even wav), and then recompose it [16:14] <orangey1> should I render as mpeg2 to begin with? [16:15] <ikrabbe> As a result I would choose a h264/aac .mp4 [16:15] <orangey1> ah. that's what i'm rendering to right now.. [16:16] <ikrabbe> yes I would do mpeg2 [16:16] <orangey1> ikrabbe: OK, thank you [16:16] <ikrabbe> (because that's what I generally get anyway from my tv card ;) [16:17] <ikrabbe> you might need to set a high bitrate, to not loose quality [16:25] <gues0x00> Hi there! :-) [16:26] <gues0x00> How can I convert an .ogg stereo file to mono? [16:26] <gues0x00> is ffmpeg the tool I should use for doing this? [16:28] <teratorn> gues0x00: Audacity might have an easy thing for that [16:28] <teratorn> gues0x00: but ffmpeg probably has a thing [16:29] <ikrabbe> gues0x00: you can use -ac 1 for the output to set the number of channels in the output [16:29] <ikrabbe> but I don't know what it does if the stereo differs much [16:31] <ikrabbe> there are some downmixing options from 7.1/5.1 or else to stereo [16:33] <ikrabbe> see ffmpeg manpage Mixing examples [16:33] <gues0x00> -ac 1 seems to work, thanks ikrabbe :-) [16:33] <ikrabbe> For example, if you want to down-mix from stereo to mono, but with [16:33] <ikrabbe> a bigger factor for the left channel: [16:33] <ikrabbe> for more advanced fun [17:01] <OneFix_Work> Is ffmpeg no longer being developed? When I run a newer copy of ffmpeg, I am getting "This program is not developed anymore and is only provided for compatibility. Use avconv instead (see Changelog for the list of incompatible changes). [17:01] <OneFix_Work> " [17:02] <OneFix_Work> I'm assuming that this is caused by one of the libraries that it has been built with... [17:03] <TimNich> OneFix_Work: It is vey actively developed. However those responsible for the avconv fork would wish you to believe otherwise... [17:03] <OneFix_Work> TimNich: So, how did that line get included in the Ubuntu version??? [17:03] <TimNich> because Ubuntu switched to the fork by default... [17:15] <Gepmff> Hey guys, I've written some code that decodes video's, all the h264 files I throw at it are decoded fine, but I can't seem to decode mkv's or mp4's I get "damaged header" during the decoding process, anyone have an idea as to why? [17:30] <vivienschilis> does aacplus supports sample rates < 44100 ? [17:39] <Aegwyn11> Hi there, I have some feedback regarding Prores decoding and not sure where to send it. I have a Prores 422 file that has some periodic issues. The libavcodec decoder manifests the issues as grey or grey/green macroblocks, while the quicktime decoder does a VERY good job of concealing them to the point that most of them aren't visable at all. The errors are all shown as "[prores @ 02cb9b80] ac [17:39] <Aegwyn11> tex damaged -3, 1024". [19:27] <teratorn> I'm encoding a VP8 stream using libvpx - using codec default values - but it looks pretty bad. what should I do to (presets?) to set up a higher-quality encode? [19:28] <Mavrik> teratorn, http://www.virag.si/2012/01/webm-web-video-encoding-tutorial-with-ffmpeg-0-… [19:28] <teratorn> Mavrik: thx [19:30] <teratorn> Mavrik: I'm doing this in C code btw - do you know anything about setting up these parameters using the API? [19:32] <ubitux> check doc/examples/*.c [19:32] <Mavrik> teratorn, you'll have to checkout libvpx.c in avcodec folder [19:32] <Mavrik> to see how those vpx-specific parameters are parsed [19:34] <teratorn> OK, thanks [19:37] <teratorn> is there not like a preset I can use? [19:44] <_klk_> Hi all - I have a bunch of images that I would like to merge into a video file using ffmpeg. I know how to do this, but is there a way to tell ffmpeg to repeat each image for N frames. (e.g. 30 frames showing image1, 30 frames showing image2, etc.) [19:46] <teratorn> _klk_: as a hack you could create your direction of images with symlikes for 29 of 30 images [19:46] <teratorn> *directory [19:53] <_klk_> teratorn: i suppose i can try that, thanks [19:53] <teratorn> well [19:53] <teratorn> I supposed you can slow down playback by 30x [19:53] <teratorn> and achieve the same effect [19:56] <GreatEmerald> Is it possible to record audio and video with one instance of ffmpeg, but have it output two files (audio and video, separately)? The program segfaults every time I try to do that... [19:58] <GreatEmerald> This is what I'm attempting to use: http://pastebin.com/raw.php?i=cERbgBda [20:14] <juanmabc> hi, i get a weird ~1fps on a flv file on my own opengl/openal player based on ffmpeg, i use the pts correctly but i have a time_base issue, perhaps this could help: http://stackoverflow.com/questions/9854293/encoding-video-only-flv [20:16] <juanmabc> also other issue is when i seek, it haves to wait for a non delta frame (but audio doesn't) so audio starts, and video keeps going with pts non drawable until scene gets moved to non delta frame [20:38] <juanmabc> so i got pts and i do: if(pts * av_q2d(media.video.codec->time_base) * 1000000.0 <= av_gettime() - media->masterclock) update_texture [20:46] <_klk_> teratorn: i did the symlink approach. ffmpeg barfs when trying to process the symlinks. i give it an input wildcard like this: ffmpeg -i image%03d.jpeg , but it says no files found. [20:47] <_klk_> when i try an equivalent wildcard for non-symlinked files it works. [21:16] <GreatEmerald> Aha, so my order of arguments was wrong, I have to set all input options first, and then all output options [21:17] <GreatEmerald> Does ffmpeg make sure the audio and video that is being recorded is in sync? [21:20] <GreatEmerald> Although either way, I guess that due to the fact that you can't specify individual input for each output file, there is no way to use ffmpeg to sync more than two streams and output them to different files to begin with... [21:26] <GreatEmerald> Hmm, unless I can map the streams manually... [21:28] <GreatEmerald> Yes I can map them! Awesome! [22:03] <Kieran> Hi, I'm having a very strange problem. Apparently commas can't appear in the output filename unless a space has appeared somewhere earlier in the filename. E.g. "test,test.m4a" triggers a "test,test.m4a: No such file or directory", where as "te st,test.m4a" or "test ,test.m4a" works. "test, test.m4a" or "test,te st.m4a" don't work either. Using latest "64-bit (Static)" Windows build found at http://ffmpeg.zeranoe.com/builds/ [22:04] <Kieran> Is this a bug, or is there something I'm missing? [22:04] <teratorn> _klk_: huh well that sux - I wonder why it doesn't like symlinks [22:05] <teratorn> sounds like a bug [22:05] <rav0__> do you have the quotes in your commandline? (i'm thinking about windows' handling of commas for multiple files) [22:05] <Kieran> yes [22:06] <Kieran> ffmpeg -i 218.m4v -i metadata.txt -y -c copy -map 0:0 -map_metadata 1 -map_metadata:s 1 -t 00:03:23.400 -ss 00:03:00.220 "test,test.m4a" [22:06] <Kieran> triggers "test,test.m4a: No such file or directory" [22:06] <Kieran> putting a space anywhere before the comma works fine [22:11] <GreatEmerald> Oh, so -vsync and -async can sync the audio and video, that's even more awesome. Too bad both the arguments have very different meanings out of context :\ [22:11] <GreatEmerald> both names, rather (vertical sync and asynchronous) [22:12] <GreatEmerald> So searching for "ffmpeg async" or "ffmpeg vsync" won't give any good results at all... [22:16] <GreatEmerald> It says that the -async argument takes samples per second. Does that relate to the sampling rate? [23:43] <teratorn> could anyone help me setting codec options? I'm using av_dict_set() to set the "rc_target_bitrate" option to "2048" before passing the options in to avcodec_open2(...) - and I've got DEBUG level logging on - when the codec options are dumped out rc_target_bitrate is still 256? [23:49] <teratorn> OK, well the "b" option does seem to have an effect... [00:00] --- Tue Mar 27 2012

1 0