Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

February 2012

2 participants
60 discussions

ffmpeg-devel.log.20120204
by burek 05 Feb '12

05 Feb '12

[00:48] <swone> is there an 8-bit luma maximum in the ffmpeg code? [00:49] <swone> bad question. i suppose i really meant, is there an 8-bit depth limit in the intermediate phases of exporting 12-bit 4:0:0 data? [00:50] <swone> my ultimate goal is 12-bit monochrome output [00:50] <durandal_1707> what? [00:50] <Daemon404> um [00:51] <swone> is there any way to encode/decode 12-bit monochrome data at full fidelity? [00:51] <Daemon404> what part of "12-bit monochrome" makes any sense? [00:51] <durandal_1707> monochrome is 1 or 0 - 1 bit [00:51] <swone> sure [00:51] <swone> grayscale [00:51] <Daemon404> thats a different thing. [00:51] <swone> right [00:52] <swone> i misused the term, but my question is basically the same [00:52] <durandal_1707> there is 8bit and 16bit single plane gray pixel format [00:53] <durandal_1707> actually it is not planar but packed [00:54] <Daemon404> how can something that is one plane be planar or packed [00:54] <durandal_1707> Daemon404: yes, I feel stupid right now [00:54] <Daemon404> :P [00:55] <swone> so would any of the compression codecs (e.g., x264) be able to use at least 12 bits for luma values? [00:55] <Daemon404> no. [00:55] <Daemon404> x264 is 10-bit max [00:55] <swone> or any other codec, really [00:55] <Daemon404> prores might be able to do 12-bit [00:55] <Daemon404> but i think that icnludes alpha [00:56] <swone> i'm not familiar with that one -- i'll have to read up on it [00:56] <swone> my ultimate goal is 12-bit grayscale interframe compression [00:56] <Daemon404> i dont think ffmpeg's implementation supports 12-bit prores anyway [00:56] <Daemon404> prores is intra only [00:56] <durandal_1707> 4:0:0 does not mean much to me [00:56] <Daemon404> iirc [00:57] <Daemon404> durandal_1707, why not? [00:57] <gnafu> Dirac can do at least 10- and 16-bit: http://www.schleef.org/blog/2012/01/23/new-schrodinger-release/ [00:57] <durandal_1707> it can be packed or planar [00:57] <Daemon404> no it cant [00:57] <Daemon404> 4:0:0 only has one plane. [00:57] <swone> daemon404, yep, looks like ProRes is intra-frame only [00:57] <gnafu> Although, I think that is intra-only right now. There may be 12-bit inter someday, but I don't know. [00:57] <durandal_1707> and swone need 12bit gray, 4:0:0 is not good format [00:57] <Daemon404> well no [00:58] <Daemon404> well 4:0:0 is gray [00:58] <Daemon404> so uh [00:58] <Daemon404> <_< [00:58] <Daemon404> same diff [00:58] <durandal_1707> Daemon404: define 4:0:0 [00:59] <Daemon404> do you understand chroma subsampling notation? [00:59] <swone> isn't it just all luma? [00:59] <Daemon404> yes [00:59] <Daemon404> luma with no chroma [00:59] <swone> ok, well that's what i had meant at least [00:59] <durandal_1707> so how many bits 4:0:0 have? [01:00] <Daemon404> durandal_1707, depends on the bit-depth of course [01:00] <Daemon404> in swone's case, 12 per pixel. [01:01] <durandal_1707> swone: only mjpeg supports 16 bit GRAY [01:01] <swone> excuse my ignorance, but isn't mjpeg also just intra-frame? [01:01] <durandal_1707> ignoring image formats and obviously raw [01:01] <durandal_1707> yes it is [01:01] <Daemon404> durandal_1707, ffv1 can essentially [01:01] <Daemon404> interframe even too [01:02] <Daemon404> but lossless, not lossy. [01:02] <durandal_1707> Daemon404: for pix fmt for ffv1? [01:02] <Daemon404> durandal_1707, pass it a null chroma [01:02] <Daemon404> same deal. [01:02] <durandal_1707> ah ane with 16 bits, rest of planes will be empty... [01:02] <Daemon404> so? [01:02] <Daemon404> itll comrpess to nothing [01:02] <Daemon404> if theyre empty [01:03] <Daemon404> and iirc, ffv1 works on a plane-basis [01:03] <Daemon404> so adding gray support is insanely trivial [01:04] <durandal_1707> would need to add new colorspace [01:04] <Daemon404> no? [01:04] <Daemon404> why would you? [01:04] <Daemon404> (assuming you mean in swscale) [01:04] <durandal_1707> how one could differ between yuv and rgb? [01:04] <Daemon404> [19:03] <+Daemon404> and iirc, ffv1 works on a plane-basis [01:04] <Daemon404> you dont. [01:05] <Daemon404> or maybe for rgb you subtract G fro mR and B [01:05] <Daemon404> and then proceed as normal. [01:05] <durandal_1707> ffv1 already can store 16bit gray lossless just fine and sws will handle lossless conversion. [01:06] <Daemon404> then tehre you go. ffv1 works. [01:06] <durandal_1707> Daemon404: i was more talking about adding direct support [01:06] <Daemon404> define "Direct support: [01:06] <Daemon404> " [01:07] <durandal_1707> adding GRAY16 into encoder input formats [01:07] <Daemon404> like i said [01:07] <Daemon404> thats trivial [01:07] <Daemon404> unless ffv1 has a number of planes hardcoded everywhere liek snow [01:07] <Daemon404> -_- [01:07] <durandal_1707> ffv1 colorspace=0 is YUV, 1, is RGB [01:07] <Daemon404> (snow codes "3" everywhere) [01:08] <Daemon404> it should be pretty simple [01:08] <durandal_1707> and there is transparency symbol for extra 4 plane [01:08] <durandal_1707> Daemon404: but would break compatibility with older decoders [01:08] <Daemon404> well yeah [01:09] <durandal_1707> because ffv1 is lazy and have if else check.... [01:09] <Daemon404> i dont see a problem with just passing it GRAY16 as luma [01:09] <Daemon404> youre not going to lose compression [01:09] <durandal_1707> yes that is alternative and "hacky" [01:10] <Daemon404> it's practical enough for me [01:10] <Daemon404> as long as you can output/input with ease [01:10] <Daemon404> and it works [01:10] <Daemon404> sometimes using what already works is a better idea than what is 'technically correct' [01:10] <durandal_1707> it should work because 16 bit YUV planar formats are already supported [01:10] <Daemon404> in terms of effort and time [01:11] <Daemon404> yes [01:16] <ixxvil> hi [01:18] <durandal_1707> so ffv1 is only codec which supports bunch of formats, comparing with ut video and vble? [01:18] <Daemon404> vble is only 4:2:0 8-bit [01:18] <Daemon404> ut video supports a bunch of 8-bit formats [01:18] <Daemon404> ffv1 is slow as hell. [01:18] <Daemon404> so noone uses it. [01:18] <swone> gnafu: thanks for the Dirac link. unfortunately, it doesn't look like any implementations use anything beyond 8-bit luma [01:20] <durandal_1707> hmm, ffv1 support only le? [01:20] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * rf57ab7ed37 10ffmpeg/libavcodec/zmbv.c: [01:20] <CIA-40> ffmpeg: zmbv: remove unused variable [01:20] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:20] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r670229ef43 10ffmpeg/libavcodec/apedec.c: [01:20] <CIA-40> ffmpeg: apedec: remove unused variable [01:20] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:22] <pengvado> ffv1 supports only native-endian. like any sane codec. [01:22] <Daemon404> a wild pengvado appears [01:22] <pengvado> translating le<=>be is for swscale to do, not for the codec [01:22] <Daemon404> which reminds me i should go finish that ffv2-replacement with marc fd... [01:23] <Daemon404> btw pengvado, i had a bet going about you -- have you ever heard of teh tunstall transform? [01:25] <durandal_1707> but how decoder will know if pix fmt is le or be? [01:26] <Daemon404> durandal_1707, it outputs whatever teh system has [01:26] <Daemon404> simple. [01:27] <durandal_1707> so file created on le will looks wrong played on be? [01:27] <Daemon404> the bitstream is likely endianless [01:27] <Daemon404> i.e. doesnt matter [01:27] <durandal_1707> whatever [01:28] <durandal_1707> dont blame me if it does not work on big endian [01:28] <Daemon404> :P [01:28] <Daemon404> for all that video encoding one does on mips and ppc of course [01:28] <Daemon404> but it should in any case i thing [01:28] <Daemon404> s/g$/k/ [01:32] <pengvado> Daemon404: no, but now that I've looked it up I see that I have implemented something like that. [01:32] <Daemon404> lol ok [01:32] <Daemon404> we;re planning on using it [01:37] <durandal_1707> heh i would need to feed zeroes [01:38] <swone> thanks! [01:42] <pengvado> oh, but what I did was variable-length on both sides. build two huffman trees and a bijection between their leaves. this ends up with just as good compression as rangecoding even for rather tiny trees. otoh, it wasn't any faster than rangecoding. [01:43] Action: Daemon404 notes speaks and thinks eerily similar to marc... [01:43] <Daemon404> pengvado* [01:44] <ohsix> not [01:51] <durandal_1707> ffv1 encodes at 21fps [01:51] <durandal_1707> on crappy celeron M [01:51] <Daemon404> now try its AC :P [01:52] <durandal_1707> Daemon404: what you need higher that 30fps? [01:52] Action: Daemon404 is an anime pirate [01:53] <Daemon404> (fansubber) [01:53] <Daemon404> 20 fps is slow. [01:53] <Daemon404> its AC is even slower iirc [01:53] <durandal_1707> but this is withn single thread on CPU from previous century [01:54] <Daemon404> the worst bit is teh decoding speed btw [01:54] <Daemon404> with ffv1, decoding is always teh bottleneck for me. [01:56] <durandal_1707> i wrote gray16 hack and it decodes with green tint and blue artifacts [01:56] <Daemon404> sounds like 00 chroma [01:56] <durandal_1707> probably pecause i do not write extra zero stuff [01:56] <Daemon404> (as in all the chroma vals are 00) [01:57] <durandal_1707> there are no chroma vals, i did not encoded chroma samples at all [01:57] <Daemon404> but is it trying to write them? [01:57] <Daemon404> >_>< [01:57] <Daemon404> or read or w/e [01:57] <durandal_1707> I just encoded luma plane [01:58] <Daemon404> is it trying to decode chroma? [01:58] <Daemon404> and getting nothing [01:58] <durandal_1707> yes [01:58] <Daemon404> well there u go [01:58] <durandal_1707> but I cant touch decoder - there is no way for decoder to know that encode is only single plane [01:59] <Daemon404> well yea... [02:00] <Daemon404> anyway... [02:00] Action: Daemon404 boots up steam [02:06] <durandal_1707> tunstall transform? [02:14] Action: durandal_1707 feels stupid [02:19] <Daemon404> its not well known. [02:20] Action: Daemon404 vanishes back to steam [02:20] <durandal_1707> Daemon404: i was wrong, ffv1 decoder checks for colorspace, making gray implementation easy [03:05] <durandal_1707> latest schroedinger, 1.0.11 have 4:4:4 16bit support [03:31] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r22c0babbb2 10ffmpeg/libavcodec/apedec.c: apedec: remove unneeded #include of get_bits.h and associated macro [03:31] <CIA-40> ffmpeg: 03Justin Ruggles 07master * re4169612a1 10ffmpeg/libavcodec/apedec.c: [03:31] <CIA-40> ffmpeg: apedec: av_fast_malloc() instead of av_realloc() [03:31] <CIA-40> ffmpeg: av_realloc() does not guarantee alignment, which is required for [03:31] <CIA-40> ffmpeg: DSPContext.bswap_buf(). [03:31] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r0759c8eb10 10ffmpeg/libavcodec/apedec.c: apedec: fix handling of packet sizes that are not a multiple of 4 bytes [03:31] <CIA-40> ffmpeg: 03Anton Khirnov 07master * r61a099680c 10ffmpeg/avconv.c: avconv: factorize setting stream_index for the output packet. [03:31] <CIA-40> ffmpeg: 03Anton Khirnov 07master * rd2afbd9a56 10ffmpeg/ (398 files in 2 dirs): (log message trimmed) [03:31] <CIA-40> ffmpeg: frame{crc/md5}: set the stream timebase from codec timebase. [03:31] <CIA-40> ffmpeg: Right now those muxers use the default timebase in all cases(1/90000). [03:31] <CIA-40> ffmpeg: This patch avoid unnecessary rescaling and makes the printed timestamps [03:31] <CIA-40> ffmpeg: more readable. [03:31] <CIA-40> ffmpeg: Also, extend the printed information to include the timebases and packet [03:31] <CIA-40> ffmpeg: pts/duration and align the columns. [03:31] <CIA-40> ffmpeg: 03Anton Khirnov 07master * r324fbadff9 10ffmpeg/avconv.c: avconv: remove debugging cruft from do_video_out(). [03:31] <CIA-40> ffmpeg: 03Alex Converse 07master * r29a20ac4a1 10ffmpeg/libavformat/mov.c: [03:32] <CIA-40> ffmpeg: movdec: Avoid av_malloc(0) in stss [03:32] <CIA-40> ffmpeg: * qatar/master: [03:32] <CIA-40> ffmpeg: dsputil: remove debug message in dsputil_init(). [03:32] <CIA-40> ffmpeg: movdec: Avoid av_malloc(0) in stss [03:32] <CIA-40> ffmpeg: build: Drop YASM-OBJS-FFT from SUBDIR_VARS. [03:32] <CIA-40> ffmpeg: build: Drop unused X86-OBJS variable. [03:32] <CIA-40> ffmpeg: 03Diego Biurrun 07master * r7f2885bbb7 10ffmpeg/Makefile: build: Drop unused X86-OBJS variable. [03:32] <CIA-40> ffmpeg: 03Diego Biurrun 07master * r58fc740059 10ffmpeg/Makefile: [03:32] <CIA-40> ffmpeg: build: Drop YASM-OBJS-FFT from SUBDIR_VARS. [03:32] <CIA-40> ffmpeg: FFT code only appears in libavcodec, so there is no need to reset the variable [03:32] <CIA-40> ffmpeg: when the build templates are instantiated for other libraries. [05:31] <durandal_1707> lossless schroedinger is much slower that ffv1 [05:33] <durandal_1707> ^encoding [06:07] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r25f35df115 10ffmpeg/libavcodec/libxvid_rc.c: [06:07] <CIA-40> ffmpeg: xvid_rc: check that write() succeeded. [06:07] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [06:08] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r7f8027b76f 10ffmpeg/libavutil/adler32.c: [06:08] <CIA-40> ffmpeg: adler32: avoid "too big" check in the inner loop [06:08] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [09:06] <NapHtaKeRoSene> hi [09:06] <NapHtaKeRoSene> if i do fabs(doublevar) then is it possible doublevar be < 0 ? [09:19] <ohsix> thats basic math [09:27] <NapHtaKeRoSene> sure but i'm getting faulty values at some pixels [09:27] <NapHtaKeRoSene> whiteish, where it should be black [09:27] <NapHtaKeRoSene> even tried clamping 0,1 [19:14] <CIA-40> ffmpeg: 03Reimar Döffinger 07master * rd7b542ae29 10ffmpeg/libavformat/oggdec.c: [19:14] <CIA-40> ffmpeg: Fix potential infinite discard loop. [19:14] <CIA-40> ffmpeg: Fixes trac issue #438. [19:14] <CIA-40> ffmpeg: Seeking in that sample would cause ogg_read_timestamp to fail [19:14] <CIA-40> ffmpeg: because ogg_packet would go into a state where all packets [19:14] <CIA-40> ffmpeg: of stream 1 would be discarded until the end of the stream. [19:14] <CIA-40> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [20:01] <CIA-40> ffmpeg: 03Clément BSsch 07master * r785d8a3868 10ffmpeg/tests/ (3 files in 3 dirs): [20:01] <CIA-40> ffmpeg: fate: use compression level 0 for flashsv2 tests. [20:01] <CIA-40> ffmpeg: Output sizes mismatch (and thus MD5) between zlib version 1.2.5 and [20:01] <CIA-40> ffmpeg: 1.2.6 if compression level > 0. The issue doesn't affect the intra-only [20:01] <CIA-40> ffmpeg: tests. [20:19] <CIA-40> ffmpeg: 03Reimar Döffinger 07master * r5b0103590c 10ffmpeg/libavformat/oggparseskeleton.c: [20:19] <CIA-40> ffmpeg: Remove warning about Ogg skeleton v4. [20:19] <CIA-40> ffmpeg: We can handle v4 just fine, the parts we currently use [20:19] <CIA-40> ffmpeg: are the same for v3 and v4. [20:19] <CIA-40> ffmpeg: v4 can in addition contain an index which we so far do [20:19] <CIA-40> ffmpeg: not use though. [20:19] <CIA-40> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [20:26] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r26585d2a7f 10ffmpeg/libavutil/adler32.c: [20:26] <CIA-40> ffmpeg: adler32: rewrite using integer SIMD. [20:26] <CIA-40> ffmpeg: about twice as fast as before. [20:26] <CIA-40> ffmpeg: the not CONFIG_SMALL case is also droped as it is not faster than the [20:26] <CIA-40> ffmpeg: CONFIG_SMALL case. [20:26] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:30] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * rd3f912c1f0 10ffmpeg/libavutil/adler32.c: [20:30] <CIA-40> ffmpeg: adler32: add comment for the SIMD code [20:30] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [21:36] <pills> i hope this is the right chan to ask...I notice there is a separate decode path when using video filters (libavfilter/src_movie.c) this path doesn't seem to honor the decode options given on the CLI to the "main" video stream. [21:37] <pills> so I guess my question is there a plan to allow more advanced decode options to "movie" sources when they are not all encoded the same or need special options to decode (despite encoding errors) [21:49] <Compn> yes there may be video filter plan [21:49] <Compn> ask stefano and or michaelni [21:49] <Compn> well, probably stefano [21:53] <Compn> his nick is saste, but hes not here right now [00:00] --- Sun Feb 5 2012

1 0

ffmpeg.log.20120204
by burek 05 Feb '12

05 Feb '12

[00:10] <madsage> if i were to keyframes value to 4 seconds using flash media encoder. what is the equivalent with ffmpeg. I see that ffmpeg refers to keyframes as GOP 'groups of pictures' would it be -g 4 ? [00:10] <madsage> or thats a keyframe every 4 frames? [00:11] <madsage> at 23fps i would need -g 92 ? [00:11] <madsage> for a keyframe every 4seconds? [00:13] <madsage> my wowza is complaining; Fragment duration greater than suggested range of 1-4 seconds. Adjust keyframe interval accordingly [00:14] <madsage> i guess i can just try some values. i've seen others using -g 20 in some examples [00:15] <teratorn> madsage: I *think* that is the GOP size yes [00:16] <madsage> sweet thanks, i kind of figured that out. but i appreciate the confirmation. i'm fairly new at this stuff. [00:17] <madsage> i will try a value of -g 20 and see if the warnings go away [00:17] <burek> -g is the size of group of pictures [00:18] <burek> (in frames) [00:18] <madsage> the translation of keyframes to groups of pictures is a little unclear to me still but i can try by trial and error [00:18] <burek> -g 25 gives you 1 keyframe/sec at 25fps display rate [00:18] <madsage> keyframes in Flash media encoder are per second. [00:19] <burek> it means that distance between 2 keyframes is 25 frames [00:19] <madsage> ahh ok cool, so i was close [00:19] <madsage> right on, i appreciate the clarification [00:20] <madsage> i was off quite a bit [00:20] <burek> :beer: [00:20] <madsage> heh [00:20] <burek> :) [00:20] <madsage> thanks, its 4:20 here [00:20] <burek> "GOP size is 300 which means one intra frame every 10 seconds for 29.97fps input video" [00:21] <madsage> =] [00:21] <burek> http://ffmpeg.org/ffmpeg.html [00:21] <burek> :) [00:21] <madsage> aye [00:21] <madsage> almost beer:30 [00:21] <burek> :) [00:23] <madsage> wowza seems to want a keyframe every 2-4 seconds. -g 50 should be close most of my media is NTSC standard 23fps or close [00:23] <madsage> we'll give it a shot. and thanks again [01:18] <pills> how do i get the "movie" video source when using video filters to obey the vsync option on decode? [01:38] <praedo> hello [01:38] <praedo> anyone knows how to duplicate audio channels? [01:39] <praedo> i have a football match recording where one channel is the environment sound and one channel is an XLR voice speaker input [02:20] <teratorn> "duplicate" ? [02:21] <teratorn> praedo: ^ [04:19] <pills> how do i get the "movie" video source when using video filters to obey the vsync option on decode? Trying to combine two video streams using video filters and they are out of sync. [04:56] <ayaka> does ffmpeg support multi thread? [07:54] <pills> I have 4 videos I am trying to combine into a single video. This is easy enough to do using the video filter options. My problem is the input videos don't seem to decode properly (first frame is dropped). This seems to be due to how they were encoded wrt timecodes. I can fix this problem for a single video by adding the -vsync 0 option however when combined with the video filters only the first video seems to inherit the vsync option. Is [07:54] <pills> there anyway around this (besides reencoding the video)? Thanks. [09:09] <NapHtaKeRoSene> hi [09:09] <NapHtaKeRoSene> if i do fabs(doublevar) then is it possible doublevar be < 0 ? i believe i am getting negative float value after clamping [09:12] <Tjoppen> uhm, of course? [09:13] <Tjoppen> why else would you use fabs() if doublevar was guaranteed >= 0? [12:21] <tim> hi, i am using kdenlive to capture a full-hd video with a lossless codec. however i have some issues when using huffyuv ... so i wonder, can anyone recommend a lossless video codec that has a high encoding performance? [12:34] <fr0ggie> can i make ffmpeg put MOOV atoms at the start of a mpeg4/h.264 file? [12:34] <gfto> fr0ggie: qt-faststart [12:35] <gfto> moov atoms can be generated only after the file is ready [12:35] <fr0ggie> hmm [12:35] <gfto> run qt-faststart on the encoded mp4 [12:35] <fr0ggie> basically what i have happen is motion (video motion detection/recording) program uses ffmpeg to record videos of events.. Hmm i suppose i could hook on_* to run it [12:36] <gfto> fr0ggie: unfortunately mp4 is not ment to be used as streaming container (thats what we have mpegts for), so movng the moov is a hack but it works [12:36] <gfto> the problem is that it can't be put in front during generation [12:37] <fr0ggie> can you put h.264 into a mpeg-[12] container and expect android devices to support it? [12:37] <gfto> 3.0 support mpegts [12:37] <gfto> in addition to http live streaming (the apple prptocol) [12:38] <gfto> for older devices you can use a player for this [12:38] <gfto> there is a free one drood player or something that supports hls [12:38] <fr0ggie> hmm [12:39] <fr0ggie> See, i have a directory full of recordings from the security cameras which i view from either a nice webpage (on a real browser) or from android browser not so nicely ;) [12:39] <fr0ggie> I guess qt-faststart as an on_* will work, thanks [12:40] <fr0ggie> on_movie_end ;) [12:42] <gfto> i've used this to play hls sreams on android 2.4 devices https://market.android.com/details?id=com.daroonsoft.player [12:43] <gfto> yeah just run qt_faststart once the file is available and then you can use flash player to view the mp4 [12:43] <gfto> also this file will be playable on iXXX devices (assuming the video is coded correctly) [12:44] <gfto> using flash you can view them on adroid too [12:45] <fr0ggie> flash? [12:45] <fr0ggie> i dont have flash [12:46] <fr0ggie> I personally dont even use android to view things, but my friend who i set this up for does and he's stuck on like 2.1 [12:46] <fr0ggie> (htc hero on crappy carrier) [12:49] <gfto> well try daroon then, it may work [12:53] <fr0ggie> will try later today when i see him, thank you [12:53] <fr0ggie> ffmpeg is pretty awesome ;) [16:51] <NapHtaKeRoSene> VIDEO: [WMV3] 480x360 24bpp 1000.000 fps 832.0 kbps (101.6 kbyte/s) [16:52] <NapHtaKeRoSene> :) [16:52] <NapHtaKeRoSene> mplayer output [16:52] <NapHtaKeRoSene> though it plays ok [16:54] <phed> hi, is there a way to "test" if a file would be decodable by ffmpeg? [16:55] <Tjoppen> ffprobe? [16:59] <phed> hmm i found this example "ffmpeg -v 5 -i episode01.avi -f null -" but i am not sure if this fully decodes or not :) [17:01] <phed> hmm never used ffprobe, i will try that next :) [17:01] <Tjoppen> it also depends on what you mean by "decodable" [17:01] <Tjoppen> you can just run ffmpeg and check its exit code ($?) [17:01] <Tjoppen> like, just do the transcode. if $? != 0 then something went wrong [17:01] <phed> i believe i am doing that there. it's diagnostics [17:02] <phed> but i worry if that becomes "optimized" since tere's no output :) [17:03] <phed> but that doesn't seem to be the case [17:57] <SIFTU> I have a bunch oof small mpeg2 files I want to join. I ran the following command "cat $(eval echo {$first..$last}.mpg) | ffmpeg -i - -target ntsc-dvd -f dvd -q:v 3 dvd$i.mpeg" [17:57] <SIFTU> and all worked by the audio gets out of sync [17:58] <SIFTU> any tips? [18:01] <pills> how do i get the "movie" video source when using video filters on the CLI to obey the vsync option on decode? [20:54] <avamk> Hello, I've been converting several H.264 .mov videos to WebM via this command: ffmpeg -i [input].mov -sameq [output].webm But for some of the converted videos (why not all?) they would freeze after playing for a few seconds, while the audio would continue. How can I troubleshoot this? [20:56] <Mavrik> avamk, don't use sameq, it doesn't do what you think it does [20:57] <Mavrik> avamk, http://www.virag.si/2012/01/webm-web-video-encoding-tutorial-with-ffmpeg-0-… [20:57] <avamk> Mavrik: Oh I see, checking out the link now... [20:58] <Mavrik> avamk, basically there's no easy way of doing it "the same quality" [20:58] <Mavrik> so you'll have to set approximately the same bitrate manually [20:58] <Mavrik> for more troubleshooting you should really paste the commandline and output to some pastebin and show it [20:59] <avamk> Mavrik: OK, I'll try the steps outlined on that page. If it still doesn't work I'll use pastebin. Thanks for your quick response. [21:03] <avamk> Basic question from new user: my ffmpeg says it doesn't recognise the 'codec:v' option? [21:06] <drv> some of the options were renamed recently; -vcodec should work instead [21:06] <drv> (or you can update to a newer ffmpeg release) [21:07] <pills> I have 4 videos I am trying to combine into a single video. This is easy enough to do using the video filter options. My problem is the input videos don't seem to decode properly (first frame is dropped). This seems to be due to how they were encoded wrt timecodes. I can fix this problem for a single video by adding the -vsync 0 option however when combined with the video filters only the first video seems to inherit the vsync option. Is [21:07] <pills> there anyway around this (reencoding is fine)? Thanks. [21:10] <avamk> drv: Thanks, is there a list of the renamed options? Looks like my ffmpeg also doesn't like others such as '-cpu-used'. [21:13] <drv> hmm, i'm not sure what -cpu-used does [21:14] <drv> you might be able to just leave it out [21:14] <avamk> drv: I am trying to replicate what's here: http://www.virag.si/2012/01/webm-web-video-encoding-tutorial-with-ffmpeg-0-… [21:15] <Mavrik> avamk, update your ffmpeg [21:15] <Mavrik> since it's so old it doesn't support libvpx-specific flags [21:15] <Mavrik> don't encode WebM with old ffmpegs ;) [21:15] <avamk> Mavrik: Oh OK. I'll try to update it. But I got it from MacPorts on my Mac, looks like they didn't update it in a while... [21:16] <Mavrik> hmm [21:16] <avamk> Might have to compile from scratch... [21:16] <Mavrik> basically you want 0.9+ [21:16] <avamk> OK [21:56] <brontosaurusrex> funny problem, if i add -map 0:0 -map 1:0 to command line, then x264 pass loging gets broken, any clues how this two could be connected? [21:59] <brontosaurusrex> a log http://pastebin.com/J6RgHQF7 [22:31] <jackNDbox> sry for the noob question but what does it mean when I do ./configure and I get a message saying yasm not found? [22:32] <Monkeypaws> the configure script calls on an app named yasm, which you dont have installed [22:32] <jackNDbox> k thx i'll go get that [22:33] <Monkeypaws> YASM is a modular assembler, it is required by the x264 package. [22:33] <Monkeypaws> http://wiki.razuna.com/display/ecp/FFMpeg+Installation+on+CentOS+and+RedHat [22:33] <Monkeypaws> ok gl [22:34] <jackNDbox> oh wow thanks for that link [23:19] <jackNDbox> anyone have any idea? I am running this command: [23:19] <jackNDbox> ./configure --enable-gpl --enable-nonfree --enable-postproc --enable-libfaad --enable-avfilter --enable-pthreads \ [23:19] <jackNDbox> --enable-libxvid --enable-libx264 --enable-libmp3lame --enable-libfaac --disable-ffserver --disable-ffplay \ [23:19] <jackNDbox> --enable-libtheora --enable-libvorbis --disable-ffplay --enable-shared --enable-avfilter-lavf --arch=x86_64 [23:20] <jackNDbox> and I get libmp3lame not found [23:21] <jackNDbox> should I just delete --enable-libmp3lame and move on? or will I have serious problems later? [23:23] <iive> jackNDbox: do you have the headers needed for using the mp3lame library? usually they are in -dev package. [23:26] <jackNDbox> honestly i'm not really sure I followed this step by step and didn't have a problem until now [23:26] <jackNDbox> http://wiki.razuna.com/display/ecp/FFMpeg+installation+on+CentOS+and+RedHat [23:30] <jackNDbox> I am running cent OS 5.7 [23:45] <iive> jackNDbox: can you find the mp3lame check in the configurel.log file? [23:45] <iive> it should say what is failing. [23:52] <jackNDbox> it's saying something like: [23:53] <jackNDbox> ./configure: line 2845: libswscale/libswscale.pc: no such file or directory... [23:53] <jackNDbox> ./configure line 2861: libswscale/libswscale-uninstalled.pc no such file or directory [23:58] <jackNDbox> i dunno i think i'm just gonna take a break [23:58] <jackNDbox> i've been configuring this server for like 12 or 13 hours now [23:59] <jackNDbox> i'm probably just brain dead at this point [00:00] --- Sun Feb 5 2012

1 0

ffmpeg-devel.log.20120203
by burek 04 Feb '12

04 Feb '12

[00:43] <ubitux> durandal_1707: you are Paul right? [00:44] <durandal_1707> for you yes [00:46] <ubitux> just a little suggestion about the apedec patch you sent a few minutes ago; it's generally a good idea to have more expressive message about the error: for instance you could specify what is the unsupported bps [00:56] <CIA-40> ffmpeg: 03Clément BSsch 07master * r9477fa094b 10ffmpeg/libavutil/timecode.c: lavu/timecode: use @see for URL in doxygen. [00:57] <CIA-40> ffmpeg: 03Clément BSsch 07master * r174880f580 10ffmpeg/libavfilter/vf_thumbnail.c: [00:57] <CIA-40> ffmpeg: lavfi/thumbnail: rename @url to @see. [00:57] <CIA-40> ffmpeg: @url isn't a valid Doxygen directive. [01:04] <durandal_1707> some thing is strange when seeking in ape [01:32] <ubitux> nevcairiel: it seems at least the AAC decoder is changing the allocated extradata [01:32] <ubitux> sorry i'm still on that ;) [01:41] <dalecurtis> Hi all, I merging upstream ffmpeg into chromium and noticed an issue with videos w/o sound. It's reproducible with tot ffplay as well. Essentially the last 0.5 secs of video is being dropped. Has anyone else ran into this issue? [01:41] <dalecurtis> ffplay cli: ffplay -autoexit -i sync0.mp4 , test case: http://dl.dropbox.com/u/242784/sync0.mp4 [01:42] <dalecurtis> Repros with mp4, ogv, webm. [01:42] <ubitux> is it a regression? [01:42] <dalecurtis> Yes, ffmpeg from 6 mo ago doesn't show this issue. [01:42] <ubitux> could you git bisect and find the problematic commit? [01:43] <dalecurtis> Sure, let me look. [01:59] <durandal_1707> michaelni: are you sure your ape changes are correct? I get strange problem when decoding 24bit samples [01:59] <durandal_1707> by changes I mean from first ape commit to today [02:01] <durandal_1707> i can give you sample and patch to support 8 and 24 bit [02:07] <michaelni> durandal_1707, does reverting the commit fix it ? [02:07] <michaelni> and all patches are always welcome [02:08] <durandal_1707> will start reverting commits, because seeking is broken too [02:20] <dalecurtis> ubitux: Looks like the bad commit is 4478e9d8db65ca827f2b3ef3ef6ee806bffdba45 , which is sadly a merge commit from qatar/master. [02:20] <dalecurtis> Came in on Dec 31. [02:20] <durandal_1707> dalecurtis: that is threads [02:20] <durandal_1707> force threads to 1, does issue still happens? [02:23] <dalecurtis> Yes, -threads=1 solves the problem. [02:30] <durandal_1707> michaelni: i cant reproduce crashes with png [02:31] <dalecurtis> durandal_1707: Should I file an issue for this? Or is this known? [02:31] <durandal_1707> dalecurtis: it is known issue [02:31] <michaelni> png is a misalignment issue with linesize % 8 != 0 ithink but i didnt debug further [02:31] <michaelni> or better said that is the reason for the segfault i think [02:31] <michaelni> i dont know what causes the vis artifacts [02:32] <dalecurtis> durandal_1707, ubitux: Thanks. [02:45] <durandal_1707> michaelni: i think your changes actually removed >16bit stuff [02:45] <michaelni> durandal_1707, which changes ? [02:46] <durandal_1707> tables [02:46] <michaelni> which commit are we talking baout ? [02:46] <durandal_1707> 8d4bef64b4ab7e66cf6ca1fb2bc762103390950a [02:47] <michaelni> march 2008 :) [02:48] <michaelni> now if i could remember ... [02:48] <michaelni> but if that entries are usefull i can put them back ? [02:48] <durandal_1707> thats for are logs [02:48] <durandal_1707> you should write in commit message all stuff [02:48] <michaelni> i know [02:49] <michaelni> log messages are something i always was bad at [02:51] <durandal_1707> i hope that I would not be forced to remove bunch of commits, exp ones from Loren [02:54] <ubitux> michaelni: consider your commits a blog post entry instead of a tweet :) [02:54] <ubitux> i'm sure you'll try to be a bit more descriptive [02:59] <durandal_1707> some filese caused apedec to crash, now I would need to search for them [03:06] <durandal_1707> this is going to be harder [03:28] <sassan> hi all. I'm writing a protocol plugin and due to the nature of the system it connects to, the protocol receives the metadata seperately, rather than as part of the encoded data. Is there a way for the protocol plugin to make ffmpeg aware of this metadata? [03:33] <michaelni> sassan, easy way is to throw it into the data stream if thats easy possible [03:33] <sassan> it's an mp3 stream. not sure how easy that would be [03:39] <michaelni> what metadata is it ? [03:39] <sassan> artist, title, etc [03:44] <michaelni> you could put it in a id3 at the start of the stream [03:45] <sassan> hmmm, I'll give that a try [03:45] <sassan> thanks :) [03:51] <Compn> sassan : like icyhttp / icecast ? [03:51] <Compn> i think it sends metadata seperately [03:51] <Compn> somehow [03:52] <ohsix> in http headers :] [03:53] <sassan> Compn, last.fm [04:03] <Compn> you are working on last.fm ? [04:08] <sassan> Compn, yep [04:08] <sassan> Compn, well, I'm ripping apart an existing player to turn it into an ffmpeg protocol plugin, yes [04:11] <Compn> so... working against last.fm's player then :) [04:11] <Compn> what was that other site that merged with last.fm before last.fm got bought [04:11] <Compn> something audioscrobbler [04:12] <durandal_1707> someone with bunch of 24bit ape files? [04:14] <durandal_1707> or wav files, MAC.exe does not want to encode wavs generated with ffmpeg [04:14] <Compn> possibly kostya or some ape site, or whoever did the 24bit [04:14] <Compn> try using mplayer -ao pcm :P [04:16] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r83ce51cc7d 10ffmpeg/libavcodec/mpc7.c: [04:16] <CIA-40> ffmpeg: mpc7: use av_fast_padded_malloc() [04:16] <CIA-40> ffmpeg: Avoids doing malloc/free for each frame. [04:16] <CIA-40> ffmpeg: Also fixes valgrind errors due to use of uninitialized padding bytes. [04:16] <CIA-40> ffmpeg: Based on a patch by Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [04:16] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r3c432631e9 10ffmpeg/libavcodec/pcm.c: [04:16] <CIA-40> ffmpeg: pcmenc: Do not set avpkt->size. [04:16] <CIA-40> ffmpeg: It is already the correct size as set by ff_alloc_packet(). [04:16] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r236a550c3f 10ffmpeg/libavcodec/x86/dsputil_yasm.asm: [04:16] <CIA-40> ffmpeg: Fix a typo in the x86 asm version of ff_vector_clip_int32() [04:16] <CIA-40> ffmpeg: Specifies the correct number of xmm registers used so that they can be saved [04:16] <CIA-40> ffmpeg: and restored on Win64 if necessary. [04:16] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r1a670973a7 10ffmpeg/libavcodec/ (internal.h utils.c): [04:16] <CIA-40> ffmpeg: ff_alloc_packet: modify the size of the packet to match the requested size [04:16] <CIA-40> ffmpeg: This will simplify encoders which use this function to request the exact [04:16] <CIA-40> ffmpeg: packet size rather than the maximum size. [04:16] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r41dd77bdbc 10ffmpeg/doc/general.texi: [04:16] <CIA-40> ffmpeg: doc: decoding Forward Uncompressed is supported [04:16] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [04:16] <CIA-40> ffmpeg: Signed-off-by: Justin Ruggles <justin.ruggles(a)gmail.com> [04:16] <CIA-40> ffmpeg: 03Mans Rullgard 07master * r2c98f407c8 10ffmpeg/ (5 files in 4 dirs): [04:16] <CIA-40> ffmpeg: fate: make acodec-ac3_fixed test output raw AC3 [04:16] <CIA-40> ffmpeg: There is no point in this test using the RM format. [04:16] <CIA-40> ffmpeg: Signed-off-by: Mans Rullgard <mans(a)mansr.com> [04:16] <CIA-40> ffmpeg: 03Janne Grunau 07master * rfcc518a024 10ffmpeg/doc/APIchanges: APIchanges: add missing commit hashes [04:16] <CIA-40> ffmpeg: 03Ronald S. Bultje 07master * rde53b9068a 10ffmpeg/libswscale/x86/ (input.asm swscale_mmx.c): swscale: implement MMX, SSE2 and AVX functions for RGB32 input. [04:17] <CIA-40> ffmpeg: Do not use AVStream's duration for dts generation since it contains in [04:17] <CIA-40> (33 lines omitted) [05:00] <durandal_1707> looks like issue is in ape demuxer and not in decoder, ... sigh* [05:01] <bcoudurier> durandal_1707, where is your patch for ape ? [05:01] <bcoudurier> I don't see it [05:01] <durandal_1707> on libav ml [05:02] <bcoudurier> oh right [05:03] <Compn> durandal_1707 enjoys the pain of sending patches to two lists :D [05:04] <bcoudurier> he didn't send this one so I was wondering [05:04] <Compn> hes not sending same patches to both lists [05:04] <Compn> just to add in confuuuusion [05:04] <Compn> and less merges i guess [05:04] <bcoudurier> :) [05:05] <durandal_1707> libav bikesheed more - i like that [05:06] <bcoudurier> let's troll ! [05:06] <ohsix> kind of makes me wonder if people know they can reply off-list [05:28] <sassan> michaelni, adding an id3 tag worked, and easier than I thought it would be :) Thanks :) [13:59] <ubitux> flashsv2 fate is broken or it's me? [16:49] <michaelni> ubitux, fate.ffmpeg.org is green [17:12] <durandal_1707> lets make party [18:01] <ubitux> michaelni: i was wondering if fate.ffmpeg.org wasn't dead [18:03] <ubitux> i may be just smoking weed, forget this [18:10] <michaelni> anyone has an idea how to make darwin llvm green too ? [18:19] <ubitux> mmh i still have vsynth1-flashsv2 failing locally& [18:20] <ubitux> i'm not sure fate.ffmpeg.org is taking into account the last reports [18:23] <ubitux> i have no idea what's going on actually [18:24] <ubitux> my last commit is indeed the d77294c merge, just like what fate reports [18:28] <ubitux> http://pastie.org/3310367 [18:28] <ubitux> this is the failure i have [18:29] <ubitux> maybe a zlib update? [18:59] <CIA-40> ffmpeg: 03Ramiro Polla 07master * rc9e5acad61 10ffmpeg/libavdevice/dshow.c: [18:59] <CIA-40> ffmpeg: dshow: fix AVInputFormat declaration after ABI breakage [18:59] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [18:59] <CIA-40> ffmpeg: 03Ramiro Polla 07master * r1f0fa80bfa 10ffmpeg/libavdevice/dshow.c: [18:59] <CIA-40> ffmpeg: dshow: use DirectShow information to set time_base [18:59] <CIA-40> ffmpeg: Fixes build after removal of AVFormatParameters. [18:59] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [19:53] <ubitux> michaelni: ok i think a zlib update or something broke it [19:53] <ubitux> http://fate.ffmpeg.org/report.cgi?time=20120203183521&slot=x86_64-archlinux… [19:53] <ubitux> i just did the updates on the box [19:58] <ubitux> ok, confirmed, downgrading to zlib 1.2.5 fixes the issue [19:59] <ubitux> something changed in 1.2.6 [20:06] <CIA-40> ffmpeg: 03Vincent Torri 07master * red3c0fe85d 10ffmpeg/libavformat/ (network.h os_support.h): [20:06] <CIA-40> ffmpeg: Remove some warnings when compiling with mingw (mingw-w64, 32 bits) [20:06] <CIA-40> ffmpeg: Reviewed-by: Ramiro Polla <ramiro.polla(a)gmail.com> [20:06] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:42] <CIA-40> ffmpeg: 03Reimar Döffinger 07master * rda1ba4e88b 10ffmpeg/libavcodec/x86/rv40dsp.asm: [20:42] <CIA-40> ffmpeg: Fix NASM compilation. [20:42] <CIA-40> ffmpeg: movd needs explicit register size prefix for NASM. [20:42] <CIA-40> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> [23:13] <CIA-40> ffmpeg: 03Reimar Döffinger 07master * r58dabf7bf2 10ffmpeg/libavcodec/ (pngdsp.h x86/pngdsp.asm): (log message trimmed) [23:13] <CIA-40> ffmpeg: Fix png decoding on x86. [23:13] <CIA-40> ffmpeg: Line sizes are only 8-byte aligned, so use unaliged loads [23:13] <CIA-40> ffmpeg: for add_bytes_l2 pointers. [23:13] <CIA-40> ffmpeg: Increasing the alignment requirement to 16 seemed a bit extreme [23:13] <CIA-40> ffmpeg: (png may be used for rather small sizes). [23:13] <CIA-40> ffmpeg: Also fix a mov that had its arguments swapped, leading [23:34] <danverde> I am writing a program that is using libavfilters. I currently setup the filter chain with avfilter_graph_parse. [23:35] <danverde> I now need dynamic arguments for my filters [23:35] <danverde> is there an example of setting up the graph and parameters a different way [23:35] <danverde> for example I want to dynamically change a text overlow and scale value of output video frames [23:49] <michaelni> danverde, theres some code in ffmpeg.c that can dynamically change a limited set of things [23:50] <michaelni> see avfilter_graph_send_command() [00:00] --- Sat Feb 4 2012

1 0

ffmpeg.log.20120203
by burek 04 Feb '12

04 Feb '12

[00:25] <tiglionabbit> how do I get ffprobe to tell me what codecs a file uses? I am using this command right now but it isn't telling me what I want to know: ffprobe chair.avi -show_format -print_format json -loglevel quiet [00:26] <Tjoppen> -show_streams | grep codec [00:26] <tiglionabbit> ah, -show_streams, that's what I need [01:24] <Freakshow> hmm... is --enable-network still a valid configure option? [01:37] <Freakshow> particularly interested in this snippet in stackoverflow [01:37] <Freakshow> http://stackoverflow.com/questions/7906944/ffmpeg-api-how-to-connect-to-rts… [01:40] <drv> it's on by default [01:40] <Freakshow> thought so... [01:40] <Freakshow> but I'm not sure that the rtsp authentication is working as advertised [01:40] <drv> if you find that it isn't enabled, you can check config.log to see why [01:40] <Freakshow> well, if it weren't, nothing would stream [01:40] <Freakshow> so I think it's safe to say it's on [01:40] <Freakshow> ;-) [01:41] <Freakshow> I'm seeing auth failures when giving it the credentials in the rtsp uri [02:42] <Freakshow> yup, I'm perplexed [05:54] <xjkx> My dvd player runs AVI files. When I get the file online, sometimes, it comes in the proper way to run in my dvd player, so it works. But when I convert it myself, I can't make it work...here is an example of file that runs fine, "RIFF (little-endian) data, AVI, 688 x 288, 23.98 fps, video: XviD, audio: MPEG-1 Layer 3 (stereo, 44100 Hz)", but when I convert with ffmpeg -i simples.mp4 -sameq avi/simples.avi I get a file like this "RIFF (little-endian) data, [06:35] <xjkx> Its a freely distrubuted video by the way, just in .mp4 yet and I need to pass to avi [06:35] <grepper> your output line for the new file was not complete [06:37] <xjkx> what is it missing ? [06:38] <grepper> it ends with: "but when I convert with ffmpeg -i simples.mp4 -sameq avi/simples.avi I get a file like this "RIFF (little-endian) data," [06:39] <xjkx> Alright let me explain on another way anyway, I want to watch this video on my dvd player www.khanacademy.org/video/simple-equations?topic=core-algebra [06:40] <xjkx> If you click on download there, you'll see they distribute it in .mp4 only, not avi [06:40] <xjkx> My dvd player runs avi, so I converted like this ffmpeg -i simples.mp4 -sameq avi/simples.avi [06:40] <xjkx> The file does not run [06:40] <xjkx> If I run the linux command "file simples.avi" I get this "RIFF (little-endian) data, AVI, 1280 x 720, video: FFMpeg MPEG-4, audio: MPEG-1 Layer 1 or 2 (mono, 44100 Hz)" [06:41] <grepper> now I see the complete output :) [06:41] <drv> if you just add "-vtag xvid" it will probably play fine [06:41] <xjkx> I have an avi file that I downloaded some time ago, and it works on my dvd, if I run the linux command "file thisfile.avi", I get this "RIFF (little-endian) data, AVI, 688 x 288, 23.98 fps, video: XviD, audio: MPEG-1 Layer 3 (stereo, 44100 Hz)" [06:41] <xjkx> Hmm [06:41] <grepper> probably need the xvid tag [06:42] <xjkx> I'll try that [06:42] <grepper> default is to use ffmpeg's mpeg4 [06:42] <grepper> dvd players are dumb [06:42] <xjkx> Unfortunately [06:42] <grepper> I've experienced that before [06:43] <grepper> some even need the divx tag (not yours obviously) [06:45] <grepper> http://www.fourcc.org/codecs.php [06:45] <xjkx> thanks [06:46] <xjkx> Trying to convert with xvid at first [07:07] <arbin> isn't 720p xvid a bit much for a dvd player? [07:07] <xjkx> I guess so [07:08] <xjkx> Not sure how to reduce it though, -s ? [07:09] <arbin> -s would work yea [07:09] <arbin> well, that's how i do it [07:10] <arbin> what with all the -vf stuff nowaday that might be the "new" way of resizing [07:10] <xjkx> Feel free to share how you do and what number I should choose for resolution [07:10] <arbin> but for now just try doing -s 640x272 [07:10] <xjkx> Hmm [07:10] <arbin> that should do for 16:9 [07:10] <arbin> er 360 [07:13] <arbin> 640x360 [07:13] <arbin> 272 gets you 2.35:1 or something [07:13] <xjkx> Alright, so ffmpeg -i simples.mp4 -s 640x360 -sameq -vtag xvid avi/simples2.avi [07:13] <xjkx> Going to try now, thanks [07:17] <grepper> xjkx: I don't think you need to covert it if you are just changing the tag. Encoding again will lower the quality. [07:18] <xjkx> But I gotta change resolution and such I guess, can do it without converting ? [07:19] <grepper> doesn't seem worth it to me, but no, you would have to convert in that case [07:20] <grepper> not sure why you are changing the resolution [07:20] <grepper> your dvd player can't handle it ? [07:20] <xjkx> Not sure, just trying everything [07:21] <grepper> I would say its definately the fourcc tag [07:21] <grepper> but you'll have to try :) [07:21] <grepper> though its possible your player can't handle the res, I have no clue [07:22] <xjkx> We'll see in a minute, else we try some fourc tags, I'm not familiar with that though [07:22] <grepper> come to think of it my old player might not have been able to handle 720p [07:23] <grepper> in which case ignore what I said about just copying and changing the tag :P [07:23] <grepper> or maybe it was 1080i it couldn't handle, can't remember [07:49] <xjkx> ffmpeg -i simples.mp4 -s 640x360 -sameq -vtag xvid avi/simples2.avi worked, but the screen is not perfectly fit, I can't see a 7 on the screen, so I converted it to 200x160.avi and woot, it seems ok, but maybe I exagerated on resolution, would you recommend something else ? [08:09] <oasisfleeting> hello. I'm trying to install ffmpeg-php on fedora 16 [08:09] <oasisfleeting> I've followed several tutorials and I can get as far as make install [08:10] <oasisfleeting> but then i get an error cannot stat "modules/*" [08:10] <oasisfleeting> where is the so file at once the make command has finished? [08:27] <cbreak-work> wonder if there's a mem leak in ffmpeg [10:24] <dusf> is there a way with ffmpeg to take multi screencaps with says 16 small screens in one image in storyboard format? [10:26] <relaxed> dusf: yes, use the pad filter to create a canvas for the other videos and the overlay filter to place them. [10:29] <dusf> relaxed: can that be done through the gui? i have never encoutered the pad filter [10:30] <Mavrik> gui? [10:30] <relaxed> probably not [10:33] <dusf> i wish there was a decent video editor for linux with a gui [10:33] <Mavrik> what's wrong with kdenlive and other such software? [10:33] <dusf> it's so much easier just to boot into windows for 2 minutes and open media player classic then have to search for a manual, and find the specific command just to take a storyboard screencap [10:33] <dusf> Mavrik: i had never heard of kdenlive, will it take storyboard screencaps? [10:34] <Mavrik> no idea, it's something my video friends use :) [10:34] <Mavrik> that's why I asked :) [10:34] <dusf> i'll check it out, thanks :) [11:17] <mage_> hello [11:17] <mage_> any idea how I can fix a [mov,mp4,m4a,3gp,3g2,mj2 @ 0x29a19010] moov atom not found ? [11:17] <mage_> I'm trying to convert a .mp4 -> .webm [11:18] <JEEB> let me guess, the input file's broken (didn't download to the end or whatever) [11:18] <JEEB> moov atom is basically the "index" of the file [11:19] <JEEB> if it has no index, it doesn't know where things are in the container [11:20] <mage_> ok so there is now way to fix it ? [11:21] <mage_> it's strange because the file plays good in mplayer [11:21] <JEEB> oh [11:21] <JEEB> that's really strange [11:21] <JEEB> because mplayer should be using libavformat as well for parsing the file [11:21] <JEEB> unless you have different versions of ffmpeg doing the job .-. [11:21] <mage_> nop :( [11:21] <JEEB> or well, different libraries [11:22] <JEEB> pastebin mplayer's terminal output [11:23] <mage_> http://www.pastie.org/3308200 [11:24] <mage_> and this is the ffmpeg one http://www.pastie.org/3308212 [11:24] <Monkeypaws> Hello... I want to use ffmpeg or ffserver to input a stream from my ipcamera and output to a stream apache can use in an iframe. I have the input working fine, but, do i need to output a file or do i have to use ffserver to stream the output? (thanks) [11:25] <JEEB> you can see that libavformat seems to fail in a similar way with mplayer as well... and then it takes it in as MPEG-ES? [11:25] <mage_> yeah.. [11:26] <mage_> but.. I can read it :) [11:29] <Monkeypaws> hi? [11:30] <JEEB> hmm [11:30] <JEEB> I wonder if you can make ffmpeg try and read it as a mpeg-4 part 2 elementary stream [11:30] <JEEB> -f mpeg4 or something [11:30] <JEEB> because if it succeeds at reading it as an elementary stream, it could be it isn't an ".mp4" file in reality [11:31] <mage_> let's try .. :) [11:32] Action: Monkeypaws waves his paws [11:32] <JEEB> Monkeypaws, you're pretty much on your own with ffserver. There's no-one keeping it up really, and very few know anything about it. [11:32] <JEEB> thus you're not being ignored, no-one around just knows anything you've asked about :P [11:33] <Monkeypaws> oh ok.. i didnt know if what i was doing was commonly done. sounds like not... [11:33] <mage_> JEEB: same error [11:33] <Monkeypaws> thanks [11:33] <JEEB> you could just stay around and wait if anyone knows (without spamming the channel with the same question all the time) [11:34] <JEEB> mage_, you did set it before -i, right? [11:35] <mage_> works better with -f mpeg [11:35] <mage_> Unknown input format: 'mpeg4' [11:35] <mage_> let's try to convert it now [11:36] <mage_> it stalls [11:36] <mage_> http://www.pastie.org/3308264 [11:37] <mage_> WS Public Health.mp4: could not find codec parameters [11:37] <JEEB> -f mpeg is probably something else :P [11:38] <JEEB> also, I do recommend building a newer ffmpeg if possible. 0.7.x is getting old. Not directly related, but happened to see the version number :P [11:45] <mage_> JEEB: anyway, thanks for your help [12:28] <frogs> other can compile? no rule to make target libavcodec/x86/rv40dsp.c needed by libavcodec/x86/rv40dsp.o ;;;;; did compile 4 days ago, looking for breakage since [12:32] <frogs> think this: http://thread.gmane.org/gmane.comp.video.ffmpeg.cvs/46891 [12:42] <frogs> hello iive [12:42] <iive> yo [12:47] <frogs> ok, it compiled now that i did clean before build this time [14:06] <Schneewittchen> Hello! I use ffmpeg for merging a video track (flv) and a audio track (mp3) together without new encoding. It runs on a debian linux server. For building a new header I use flvtool2. That works perfect with a flv file encoded with h.264 but does not work with flv files encoded with VP6. I need this to use alpha channel. Is this a bug or is VP6 not provided by ffmpeg? Here is the command... [14:06] <Schneewittchen> ...line: ffmpeg -i "video.flv" -i "audio.mp3" -vcodec copy -acodec copy "video_and_audio.flv" [15:10] <Snaggle> I'm trying to build ffmpeg-0.10 w/ --disable-swresample, but when ffplay gets linked, I get the following error about undefined symbols: http://pastebin.com/G5akQfFQ [17:22] <qxt_> Got a weird compile error that ffmpeg can't create and execute file in /tmp. Debian using noexec for tmp in fstab. Is this what is causing the error? Compile tons of stuff and never been hit with this error before. [17:25] <qxt_> Debian recommends fstab to look look like this when using an SSD (tmpfs). The noexec is used (Debian recommendation) for used even with an hdd. fstab -> tmpfs /tmp tmpfs rw,noexec,nodev,nosuid,noatime,mode=1777 0 0 [17:34] <qxt_> Yup it was the noexec. [17:39] <grepper> good idea doing it in /tmp (tmpfs) I do it this way some times but haven't benchmarked it, so I'm not don't have a figure for the speedup due to caching. [18:20] <kenyabob> Whats the simplest way to re encode a divx (mpg4) video to be a much smaller size (orig video encoded without target bitrate) [19:01] <naeluh> Hi all ! [19:02] <naeluh> I got his error when I ran ./configure [19:02] <naeluh> WARNING: pkg-config not found, library detection may fail. [19:02] <naeluh> This is ffmpeg 0.9.1 [19:02] <naeluh> thanks [19:07] <hi117> did make work? [19:10] <slackyman> hi [19:10] <slackyman> how to cross-compile static xvid lib?!? [19:11] <naeluh> its still running [19:11] <naeluh> hi117:it is still running [19:12] <hi117> oh, what i should ask is does ./compile report everything works besides that warning? iirc it prints a list of what the compiled ffmpeg will support [19:15] <naeluh> hi117:make just finished and seemed to work yeah that was the only error [19:15] <hi117> meh i fail [19:15] <hi117> did ./configure report all the codecs/muxers/other stuff you want it to support? [19:16] <hi117> if it didnt, install pkg-config [19:16] <hi117> and recompile [19:17] <naeluh> this was the ouput http://pastebin.com/UdFfGF4c [19:17] <naeluh> does that look right?? [19:18] <naeluh> it seems that nothing is enabled [19:18] <hi117> ya, you need pkg-config [19:18] <hi117> http://www.freedesktop.org/wiki/Software/pkg-config [19:20] <naeluh> cool I used apt-get install pkg-config [19:20] <naeluh> it installed [19:20] <naeluh> what exactly doe this do [19:20] <praedo> hello [19:20] <praedo> i use this binary for windows: [19:20] <praedo> http://sourceforge.net/projects/mplayer-win32/files/FFmpeg/git-N-28849-ga82… [19:21] <praedo> but it doesn't recognize the -preset option [19:21] <praedo> why? [19:21] <hi117> naeluh it helps programs detect where libraries are [19:21] <hi117> now run ./configure and make again [19:22] <naeluh> cool [19:23] <naeluh> it keeps installing /usr/local/bin/ffmpeg [19:38] <naeluh> how do I do a test conversion? [19:49] <madsage> morning [19:51] <madsage> i have a question on the -loop command. i cant seem to get it to work. Where in the command syntax exactly should it reside. [19:52] <madsage> i want to loop this infidelity: ffmpeg -re -i live1.mpg -threads 0 -c:v libx264 -crf 30 -c:a libaacplus -ab 48k -ac 2 -vbsf h264_mp4toannexb -f mpegts udp://127.0.0.1:10000?pkt_size=1316pegts udp://127.0.0.1:10 [19:52] <madsage> er infinitely, damn spell check [19:53] <Monkeypaws> im tryin to input a mjpeg http stream from an ip camera and output to something apache can stream. Is ffmpeg what i need or is there a better way? [19:54] <madsage> oh just found something. -loop_input [19:57] <madsage> hrmm now to figure out how to use it. Unrecognized option 'loop_input' [19:58] <madsage> oh and i am using ffmpeg version N-37346-g371946b [20:01] <madsage> Monketpaws, i would say yes. but i push media to a media broadcast server. [20:02] <Monkeypaws> do you know what would be a good output format i could use on a web page? [20:02] <naeluh> Hey how can I move my install of ffmpeg from /usr/local/bin/ffmpeg to /usr/bin/ffmpeg for some reason it installed in the wrong directory [20:04] <Monkeypaws> you might be able to just mv the binary [20:04] <Monkeypaws> thats probably not koscher though [20:04] <naeluh> yeah [20:04] <naeluh> I have tried a sym link but I not able to access the codes [20:05] <naeluh> codecs [20:06] <naeluh> is there any good doc installing ffmpeg cause I think I am doing something wrong [20:06] <naeluh> ' [20:10] <slackyman> naeluh, have you just tried to move the ffmpeg directory? [20:11] <slackyman> have you installed ffmpeg from a repository or... what? [20:11] <naeluh> no I tried to compile it myself [20:11] <naeluh> is there a good repository? [20:11] <slackyman> ok, then [20:11] <slackyman> enter the src directory and [20:11] <slackyman> make distclean [20:11] <slackyman> so you remove your installed ffmpeg [20:12] <slackyman> then [20:12] <slackyman> make install prefix=/Usr [20:12] <slackyman> sorry [20:12] <slackyman> make install prefix=/usr [20:12] <slackyman> or maybe you can get e precompiled ffmpeg [20:13] <slackyman> which distro are u using? [20:13] <slackyman> I'm compiling my ffmpeg with some static libs so that I can have a single executable binary file without dependences [20:15] <naeluh> I am ubuntu and 0.9.1 ffmpeg [20:16] <slackyman> ok [20:16] <slackyman> I guess Ubuntu repos don't have the latest ffmpeg :) [20:16] <slackyman> so that you chose to compile it by yourself, right? [20:17] <naeluh> yeah [20:17] <madsage> you want to follow the url in the topic. use the dig clone and get the latest trunk built. [20:17] <madsage> built/build [20:18] <naeluh> slackyman: I want to compile myself but when I used the install docs it just never wrks properly [20:19] <naeluh> is it ./configure first [20:19] <slackyman> have you followed the link in the topic? [20:19] <slackyman> have you read the guide? [20:20] <madsage> sure it does. you arnt following the docs then [20:20] <naeluh> I used this guide http://bit.ly/3xSE5 [20:20] <naeluh> yeah I am its pretty strait forward [20:20] <madsage> nor did you remove your old libriaries or the old ffmpeg, it sounds [20:20] <naeluh> I used that first command [20:20] <madsage> sorry for interrupting slacky [20:22] <naeluh> this one sudo apt-get remove ffmpeg x264 libx264-dev [20:22] <slackyman> I'm not comfortable with Ubuntu [20:22] <naeluh> right on [20:22] <madsage> your configure line should be full of --enable's and if you dont want your libs and bins in /usr/local like default. prefix like slacky said [20:22] <naeluh> cool okay thanks! [20:23] <madsage> usr/local is fine, if your ldconfig cache and PATH= is setup like the docs [20:24] <madsage> your prolly dont have /usr/local/bin in your $PATH [20:24] <Snaggle> I'm trying to build ffmpeg-0.10 w/ --disable-swresample, but when ffplay gets linked, I get the following error about undefined symbols: http://pastebin.com/G5akQfFQ [20:25] <madsage> and /usr/local/lib needs to be in /etc/ld.so.conf [20:25] <madsage> refering to nueluh [20:26] <naeluh> I think I have to have in the usr/bin cause I have a drupal install and I think it requires it [20:26] <naeluh> it for this video module [20:27] <slackyman> nealuh, do you need ffmpeg-php? [20:27] <naeluh> he has offered these options http://video.heidisoft.com/documentation/ffmpeg-installtion-scripts [20:28] <naeluh> maybe I am thinkin its most likely I need both maybe? [20:32] <slackyman> I have another kind of problem, cross-compiling rtmp [20:35] <naeluh> damn [20:41] <madsage> monkeypaws was a chick, we missed out. should have helped her out more. =] [20:44] <naeluh> madsage: hey when I install ffmpeg should I install from the root folder or does it matter [20:44] <madsage> it doesnt matter, as long as you are installing as root, via sudo or root [20:45] <naeluh> cool thanks your awesome! [20:45] <madsage> well i dont know about awesome. but you are welcome. i'm just another ffmpeg user [20:46] <madsage> the developers and the ffmpeg team are awesome [20:46] <naeluh> haha right on [20:46] <madsage> some serious skillsets they must have [20:47] <naeluh> yeah for sure [21:59] <kriegerod> is ffmpeg util still not capable of concatenating input files into one output? FAQs say user should transcode to mpg, then just `cat`, then transcode back. [21:59] <durandal_1707> kriegerod: there is concat protocol [22:04] <durandal_1707> and FAQ probably have not been changed for ages..., like rest of documentation [22:07] <madsage> durandal, can you clue me in on -loop ? i'm trying to stream media into mpegts to wowza and want to loop it. any ideas? i cant seem to get -loop to work [22:09] <kriegerod> concat looked encouraging, but it gave me on output just first chunk ( [22:09] <kriegerod> i used it with flv files [22:10] <durandal_1707> kriegerod: what is your cmdline [22:10] <durandal_1707> madsage: -loop on works for some demuxers ,,, are you using ffserver? [22:12] <madsage> no i am using ffmpeg sorry i wasnt spacific [22:12] <kriegerod> durandal_1707: http://pastebin.com/h2LGMXCU [22:12] <madsage> latest build from trunk [22:13] <durandal_1707> "ffmpeg -help" list -loop only for some demuxers and GIF muxer [22:13] <madsage> i'm encoding media with ffmpeg and outputitng it mpegts to a wowza server [22:13] <madsage> ahh ok [22:13] <madsage> yeah it prolly wont work with what i am doing [22:14] <madsage> i seen some things googling suggesting only for jpeg/gif [22:16] <durandal_1707> kriegerod: perhaps you need to escape | with \: "\|" ? [22:16] <kriegerod> durandal_1707:no, i've embraced string into single quotes [22:17] <kriegerod> escaping | breaks things [22:17] <kriegerod> i sent you logs with libav, now tried with ffmpeg, the same behaviour [22:22] <durandal_1707> strange it worked with mp3 last time i tried [22:23] <durandal_1707> wtf empty flac file [22:24] <durandal_1707> kriegerod: could you try with something else, like mp3? [22:27] <kriegerod> ok [22:28] <durandal_1707> tested with flac -codec copy will cause ffprobe to report incorrect duration for created file [22:29] <durandal_1707> but file size is first one + second one [22:30] <durandal_1707> if i remmember correctly mp3 did not had this issue [22:31] <durandal_1707> so it works but it is not perfect, probably should report this issue anyway [22:32] <durandal_1707> is created flv file bigger than all 3 parts? [22:34] <kriegerod> tested with mp3, file size is like ok [22:34] <kriegerod> flv file size is almost equal to first chunk size [22:36] <kriegerod> can it be that util stops pulling frames from source when "duration" is reached? i hope no... [22:36] <durandal_1707> i never tried it with videos, perhaps if you remux flvs into somethink else like mkv,nut,avi,mov will work... [22:37] <durandal_1707> anyway i think issue with flv should be reported - it may point to bug in demuxer/muxer [22:38] <durandal_1707> just going to try with two avis I have... [22:46] <durandal_1707> produced file have audio out of sync and 2nd part of file is not seekable [22:48] <durandal_1707> so this trick works only for some formats [22:48] <kriegerod> with flv it seeks wrongly. i get [flv @ 0x97b6a80] skipping flv packet: type 70, size 5002753, flags 0 - wrong packet type, wrong size [22:50] <durandal_1707> i guess this happens because concat does not allow seeking so evertyhing get messed up for some containers [22:50] <kriegerod> no, concat implements seeking [22:51] <Synthead> I'm going to encode a video of a full-screen video demo here soon. I'd like to keep most of the quality but have the smallest size (but don't we all?). What settings and codec would you use if it had to have good compatibility? [22:51] <kriegerod> although this warning happens before any seek is done [22:52] <Synthead> oh, by "full-screen video demo", I mean a recording of my screen [22:52] <kriegerod> Synthead: first, choose compatibility more exactly. Overall, good is h264 codec, for very good compression use two-pass encoding [22:52] <cbreak> crf [22:53] <Synthead> kriegerod: compatibility being that it'll be shared with people online. how do I do two-pass encoding? [22:53] <cbreak> crf > n-pass [22:53] <Synthead> cbreak: what? [22:55] <cbreak> crf is better than n-pass encoding. It'll give you about what a multi pass encoding would converge to if you allow it to vary the bitrate as needed to get a constant "quality" [22:55] <cbreak> in a fraction of the time [22:56] <Synthead> cbreak: ahhh, I see. that's neat :) [22:56] <cbreak> multi pass encoding is used if you have a bandwidth/size target [23:01] <kriegerod> durandal_1707, i think this problem is not a bug. concat protocol concatenates at protocol level. [23:01] <kriegerod> No wonder it breaks flv processing [23:02] <kriegerod> because gives bytes stream, not avpackets stream [23:02] <kriegerod> in my case concatenation after demux is needed [23:03] <Synthead> kriegerod: with h264, if barely anything moves on the screen, the video size should be small, right? [23:03] <pills> I have 4 videos I am trying to combine into a single video. This is easy enough to do using the video filter options. My problem is the input videos don't seem to decode properly (first frame is dropped). This seems to be due to how they were encoded wrt timecodes. I can fix this problem for a single video by adding the -vsync 0 option however when combined with the video filters only the first video seems to inherit the vsync option. Is [23:03] <pills> there anyway around this (besides reencoding the video)? Thanks. [23:04] <kriegerod> Synthead, seems. I'm not an expert. If nothing moves and filesize is still huge, try increasing 'gop' (group of picture length) param [23:04] <Synthead> I'm going to do the same thing here, smashing a few high-res videos into a single, smaller video :) [00:00] --- Sat Feb 4 2012

1 0

ffmpeg-devel.log.20120202
by burek 03 Feb '12

03 Feb '12

[00:11] <Daemon404> michaelni, can you clarify ea535ed50d1b8d751e2d194a987295ab38daf1a2? [00:12] <michaelni> Daemon404, i dont remember ATM thats a year ago ... [00:12] <michaelni> but [00:12] <michaelni> as we are already speaking [00:12] <michaelni> what is with postproc ? [00:13] <Daemon404> libav people want it out of their repo and separate, and i thought writing a script to do so would be interesting [00:13] <michaelni> sofar ok [00:14] <Daemon404> not much more to it than that... [00:14] <michaelni> some chance it can be merged with ffmpegs ? [00:14] <michaelni> so there is just 1 postproc ? [00:14] <Daemon404> michaelni, thats why im asking.. im backporting all the ffmpeg-only commits to the standalone postproc [00:15] <Daemon404> the one i linked is the last one [00:15] <michaelni> well, i already merged everything in my repo a year ago :) [00:15] <Daemon404> well they should be identical now [00:16] <Daemon404> only without the merge meta-commits [00:16] <Daemon404> (i did a diff on the commit logs) [00:16] <michaelni> good [00:17] <nevcairiel> do a diff on the code, and you'll know if they are identical. ;) [00:17] <ubitux> Daemon404: i think that commit could be squashed with the one just after (84fb4e9df737c856cca7767016110fc74bee56cb) [00:17] <ubitux> which explains a bit what's wrong [00:17] <michaelni> ubitux, iam not a friend of history rewriting [00:17] <Daemon404> nor am i [00:18] <ubitux> i didn't meant to actually do the squash [00:18] <ubitux> but they are related [00:18] <Daemon404> i only do it if i havent made somethign public yet [00:18] <michaelni> such squashing would contradict in a confusing way any discussion that happened for example ... [00:19] <ubitux> i don't have much opinion on writing/rewriting the history, just pointing out some clues [00:20] <Daemon404> man i dont know how i survived withouy git before.. [00:21] <michaelni> Daemon404, placebos ... [00:21] <Daemon404> michaelni, ? [00:21] <michaelni> https://en.wikipedia.org/wiki/Placebo [00:22] <ubitux> Daemon404: the workflows were pretty different, so problems were not the same [00:22] <ubitux> and thus, most of the solutions git solve are related to the new way of working with vcs [00:22] <ubitux> s/solutions/problems/ [00:22] <ubitux> i guess © :) [00:22] <Daemon404> ubitux, most problems git solves are due to it beign distrobuted, yes [00:23] <Daemon404> and its lovely branching [00:23] Action: gnafu prefers things that are disco-buted. [00:23] Action: gnafu gets down and funky. [00:25] <michaelni> Daemon404, when you are done with postproc, can we move it to a place i have write access to? [00:26] <michaelni> because id like to maintain it [00:26] <Compn> did you see j-b's repo ? [00:26] <Daemon404> michaelni, http://git.videolan.org/?p=libpostproc.git;a=shortlog [00:26] <Daemon404> j-b clone my repo there [00:26] <Compn> [03:16] <j-b> Compn: http://git.videolan.org/?p=libpostproc.git;a=shortlog [00:26] <j-b> and michaelni already has write access to it... [00:27] <j-b> what do you think? That I am lazy? [00:27] <j-b> oh, alright, I am lazy, but I did it :) [00:27] <Daemon404> michaelni, all i have left to do is -properly- strip down configure, and also figure out how the hell to handle this version stuff [00:27] <Daemon404> i.e. the +100 [00:28] <michaelni> Daemon404, can you fix the comiter & commit dates too ? [00:29] <michaelni> its all 2 days ago ;) [00:29] <Daemon404> the commit dates are fine [00:30] <Daemon404> the push dates are obviously 2 days ago [00:30] <michaelni> http://git.videolan.org/?p=libpostproc.git;a=commit;h=29f1f61a4a74c08070eb1… [00:31] <michaelni> commiter and date looks unlikely relative to author and date [00:31] <michaelni> 2008 vs 2012 [00:31] <michaelni> also "From original Libav commit:" <-- this is unneeded [00:31] <Daemon404> authorMichael Niedermayer <michaelni(a)gmx.at> [00:31] <Daemon404> Thu, 10 Jan 2008 06:05:12 -0500 (11:05 +0000) [00:31] <Daemon404> ^ looks right to me [00:31] <michaelni> yes [00:32] <michaelni> but it was also commited in 2008 [00:32] <Daemon404> "comitter" refers to the push date [00:32] <Daemon404> and it WAS pushed in 2012 [00:32] <Daemon404> to the new repo [00:32] <Daemon404> git will likely throw a shitfit if i try and alter push dates [00:32] <michaelni> theres no push date [00:33] <Daemon404> commit date IS the push date. [00:33] <Daemon404> or well, when it was applied to my standaloen rpeo [00:33] <michaelni> "git commit" time is the commit time [00:33] <Daemon404> repo* [00:33] <Daemon404> the time that teh author comitted it is under the top date [00:33] <Daemon404> the second date is when i applied it to mine [00:33] <Daemon404> which is correct and shouldnt be changed [00:34] <michaelni> thats not correctly converted then [00:34] <Daemon404> why exactly should the date i put it in my repo say the date you put it in ffmpeg? [00:34] <Daemon404> that doesnt make any sense. [00:34] <michaelni> look at ffmpeg or libav [00:34] <michaelni> git log --pretty=fuller [00:34] <Daemon404> the info on when it was originally coded is there. [00:35] <Daemon404> there's a reason git has 2 dates. [00:35] <Compn> Daemon404 : so you can pull older versions of ffmpeg and have it work ? [00:35] <Compn> due to correct git dates ? [00:35] Action: Compn guessing [00:35] <Daemon404> Compn, that doesnt make any sense [00:35] <michaelni> In the real history these commits where commited long ago [00:36] <Daemon404> yes but this is in a new repo [00:36] <Daemon404> those dates MUST be relative to this repo [00:36] <Daemon404> the original coding date is in-tact too [00:36] <Daemon404> so i fail to see the problem [00:36] <michaelni> the problem is it looks unprofessional [00:37] <michaelni> like someone not knowing git making this convertion [00:37] <Daemon404> youre complainign about irrelavant data [00:37] <Daemon404> in a way that doesnt make any sense [00:37] <Daemon404> i dont think you really understand the point of that 2nd date yourself. [00:38] <Daemon404> the "commit date" is the date i applied that patch to my own repo. [00:38] <Daemon404> and it is accurate. [00:38] <michaelni> yes [00:38] <michaelni> but it was not your repo [00:38] <michaelni> you are rewriting history [00:38] <Daemon404> it IS my repo [00:38] <Daemon404> i did git init . [00:38] <Daemon404> and created a new repo [00:38] <Daemon404> and imported all the patches [00:39] <Daemon404> the only history being lost is the date you originally pushed it [00:39] <Daemon404> and thats not even useful [00:41] <michaelni> you should read: http://lwn.net/Articles/328438/ [00:41] <michaelni> its from linus [00:43] <Daemon404> michaelni, that contains nothing about push dates. in fact, it seems to solely concern author history. [00:43] <Daemon404> i.e. the first date listed [00:43] <michaelni> the commit messages are changed too [00:43] <michaelni> "From original Libav commit:" [00:43] <Daemon404> i took that idea from git-svn [00:43] <Daemon404> and how it didi t [00:44] <michaelni> see git filter-branch [00:44] <Daemon404> git filter-franch is a massive massive pain in the ass to use tbh [00:45] <Daemon404> i dotn really see a massive issue here, but if youre gonna bikeshed over it [00:45] <Daemon404> ill just change it [00:45] <Daemon404> and make a new repo [00:45] <Daemon404> rather than argue [00:45] <michaelni> please do, and i volunteer to help [00:45] <michaelni> if you want my help [00:45] <Daemon404> i do know how to use filter-branch yes [00:46] <Daemon404> it's just a ugly ugly command :P [00:46] <Daemon404> with nested commands [00:46] <michaelni> ive not used it yet, but i know it should result in something cleaner [00:46] <Daemon404> yeah [00:46] <michaelni> unchanged commit messages [00:46] <michaelni> unchanged commit dates [00:46] <Compn> or you could keep postproc in ffmpeg and skip this extra repo stuff :P [00:46] <Daemon404> the changed commit messages were solely my doing [00:47] <michaelni> its ok to add the original hash as reference but iam a bit alergic to the libav advertisement [00:48] <michaelni> libav hasnt written or maintained this code [00:48] <Daemon404> i just use their repo as the base for it because theyre the ones who wanted it [00:52] <michaelni> ok, so we agree, you redo it with unchanged commits, put it in videolan and we have one libpostproc and a happy world ? [00:53] <Daemon404> indeed [00:54] <gnafu> \o/ [00:54] <michaelni> \o/ [00:58] Action: Compn asks dumb question on list, expects dumb answer [00:59] <Daemon404> now to wait a bunch of time with filter-branch runs [00:59] <durandal_1707> Compn: why you do not ask here [00:59] <Daemon404> this thing is SLOW [00:59] <Compn> durandal_1707 : because i also crossposted to other list [00:59] <michaelni> we can surely give you a dumber awnser than any list [01:00] <Compn> michaelni : btw , a note about ffv1 having (whichever colorspace(s) were added) and being worked on in the changelog would be nice :) [01:00] <Compn> i know a few people were interested in ffv1 development [01:00] <durandal_1707> Compn: interested in what? [01:01] <Compn> in using ffv1 as an intermediate format [01:01] <Compn> users i mean [01:01] <Daemon404> Compn, nobody uses ffv1 [01:01] <Daemon404> it is FAR too slow [01:01] <Compn> isnt it faster than huff ? [01:01] <Daemon404> LOL [01:01] <Daemon404> no [01:01] <durandal_1707> Daemon404: than nobody use x264 lossless too [01:01] <Daemon404> x264 lossless is faster [01:01] <Daemon404> <_< [01:02] <Daemon404> for all intents and puroses [01:02] <Daemon404> purposes* [01:02] <durandal_1707> Daemon404: you are not making any sense [01:02] <Daemon404> how so? [01:03] <Daemon404> nobody uses ffv1 because it is incredibly slow; x264 is not as slow [01:04] <Compn> Daemon404 : we await your patches to include x264 into ffmpeg ;P [01:04] <Compn> ehe [01:04] Action: Compn trolling now [01:04] <Daemon404> ;p [01:08] <durandal_1707> Daemon404: there are two coders in ffv1 [01:08] <Daemon404> theyre both slow compared to e.g. huffy [01:08] <Daemon404> and the one that is less slow really doesnt compress better anywya [01:12] <Daemon404> michaelni, git filter-branch -f --index-filter 'git ls-files -s | grep $'\t'libpostproc\/.*$ | GIT_INDEX_FILE=$GIT_INDEX_FILE.new git update-index --index-info && mv $GIT_INDEX_FILE.new $GIT_INDEX_FILE || true' --prune-empty -- --all [01:12] <Daemon404> yup. nice and ugly. [01:16] <j-b> michaelni: you don't happen to have some Dirac 10bits sample? [01:17] <bcoudurier> I use ffv1 [01:17] <bcoudurier> it compresses better than x264 lossless [01:17] <Compn> compression.ru site does lossless video codec comparasions [01:18] <Compn> or did one in the past anyways [01:18] <Daemon404> bcoudurier, for what? archiving? [01:18] <Daemon404> not useful for much else [01:18] <bcoudurier> yes and mezz files sometimes [01:18] <durandal_1707> Daemon404: for intermediate x264 is not realtime anyway [01:18] <Daemon404> bcoudurier, Marc FD has drafted me to code his new lossless codec [01:18] <Daemon404> which supposed has huffy speeds + ffv1 comrpession [01:19] <Daemon404> (needs SSSE3 though) [01:19] <bcoudurier> that's nice [01:19] <Compn> erm, why not just work on another open source one ? [01:19] <Daemon404> Compn, it's being coded into libavcodec. [01:19] <durandal_1707> Compn: FFV2 - FFV9 [01:19] <Daemon404> as well as a qtx/vfw [01:22] <Compn> i mean like lagarith was leading in 2007 :P [01:22] <Compn> http://www.compression.ru/video/codec_comparison/lossless_codecs_2007_en.ht… [01:22] <Daemon404> lagarith was always crap [01:22] <Daemon404> slow and error-prone [01:22] <Daemon404> for practical use i mean [01:22] <durandal_1707> use floating point [01:22] <Daemon404> yes it does. leads to awesome random errors [01:22] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r9430c232e8 10ffmpeg/ffserver.c: [01:22] <CIA-40> ffmpeg: ffserver: fix program reference [01:22] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:23] <CIA-40> ffmpeg: 03Paul B Mahol 07master * re39487efe3 10ffmpeg/doc/general.texi: [01:23] <CIA-40> ffmpeg: doc: ffv1 is not experimental any more [01:23] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [01:23] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:23] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r5cd8afee99 10ffmpeg/libavcodec/diracdec.c: [01:23] <CIA-40> ffmpeg: diracdec: Check for negative quants which would cause out of array reads. [01:23] <CIA-40> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [01:23] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:23] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r0065080320 10ffmpeg/libavcodec/proresdec2.c: [01:23] <CIA-40> ffmpeg: proresdec: Fix read via negative index in a global array. [01:23] <CIA-40> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [01:23] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [01:23] <Compn> Daemon404 : but ok, :) [01:23] <Daemon404> also lagarith's code is terrible [01:23] <Daemon404> kb and kb or msvc intrinsics [01:23] <durandal_1707> I think I will stick with uncompressed formats for a while [01:23] <Daemon404> of* [01:24] <Compn> no on lossless wavelet? [01:24] <Compn> :P [01:24] <michaelni> j-b, i dont think i do, sorry [01:24] <Daemon404> lol wavelets [01:25] <michaelni> or at least iam not aware of having one ... [01:27] <Compn> j-b : we dont have many dirac samples iirc [01:27] <Compn> j-b : carl has been collecting a bunch of samples tho, he may . [01:28] <Daemon404> i must ask tho [01:28] <Daemon404> who actually uses dirac? [01:28] <Daemon404> and why? [01:29] <durandal_1707> who can not pay for H.264 [01:29] <Daemon404> vp8 > dirac [01:30] <durandal_1707> vp8 is full of patents [01:30] <Daemon404> not legally so, yet [01:30] <Daemon404> and so is dirac [01:30] <Daemon404> :) [01:30] <Daemon404> any non-trivial code is full of patents [01:30] <Daemon404> its near-impossible to write code that doesnt violate someone's patent [01:39] <Daemon404> michaelni, looks liek filter-branch is gonna take like an hr to run... [01:39] Action: Daemon404 goes to make food [01:39] <michaelni> j-b, ask anuradha, i guess she either has a sample or knows someone who does or can make one [01:41] <durandal_1707> can i get bigger mthp sample? [01:57] <Daemon404> i cant believe how slow this is... [01:57] <durandal_1707> git is slow [01:58] <Daemon404> yet faster than anything else [02:00] <durandal_1707> Daemon404: there are faster but use more space [02:00] <Daemon404> like? [02:00] <durandal_1707> i forget name [02:01] <Daemon404> i hope you dont mean bzr or hg [02:01] <Daemon404> cause theyre most certainly are not [02:46] <CIA-40> ffmpeg: 03Alex Converse 07master * rc0bc7bd1e7 10ffmpeg/libavformat/swfdec.c: swfdec: Simplify sample rate calculation. [02:46] <CIA-40> ffmpeg: 03Janne Grunau 07master * re67e3a3f4a 10ffmpeg/ (3 files in 3 dirs): [02:46] <CIA-40> ffmpeg: fate-golomb: extend golomb-test to get_ue_golomb_long() [02:46] <CIA-40> ffmpeg: get_ue_golomb_long() is only tested for values up to 2^15 - 2 since [02:46] <CIA-40> ffmpeg: we can not write larger values. [02:46] <CIA-40> ffmpeg: Silence the test on success and return a non-zero value on error. [02:46] <CIA-40> ffmpeg: Use an heap scratch buffer instead of large stack buffer. [02:46] <CIA-40> ffmpeg: Remove unneeded includes. [02:46] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r635bcfccd4 10ffmpeg/libavformat/dv.c: (log message trimmed) [02:46] <CIA-40> ffmpeg: dv: check stype [02:46] Last message repeated 1 time(s). [02:46] <CIA-40> ffmpeg: Fixes part1 of CVE-2011-3929 [02:46] <CIA-40> ffmpeg: Possibly fixes part of CVE-2011-3936 [02:46] <CIA-40> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [02:46] <CIA-40> ffmpeg: Reviewed-by: Roman Shaposhnik <roman(a)shaposhnik.org> [02:46] <CIA-40> ffmpeg: 03Alex Converse 07master * r2d1c0dea5f 10ffmpeg/libavformat/dv.c: [02:46] <CIA-40> ffmpeg: dv: Fix small stack overread related to CVE-2011-3929 and CVE-2011-3936. [02:46] <CIA-40> ffmpeg: Found with asan. [02:46] <Daemon404> blech, that wasnt teh right git filter-branch [02:46] <CIA-40> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [02:46] <CIA-40> ffmpeg: Signed-off-by: Alex Converse <alex.converse(a)gmail.com> [02:46] <CIA-40> ffmpeg: 03Mans Rullgard 07master * r034b03e7a0 10ffmpeg/libavcodec/ac3dsp.c: [02:46] <CIA-40> ffmpeg: ac3: Do not read past the end of ff_ac3_band_start_tab. [02:46] <CIA-40> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind [02:46] <CIA-40> ffmpeg: Signed-off-by: Alex Converse <alex.converse(a)gmail.com> [02:46] <CIA-40> ffmpeg: 03Janne Grunau 07master * rcb0b284381 10ffmpeg/libavcodec/zmbv.c: zmbv: check av_realloc() return values and avoid memleaks on ENOMEM [02:46] <CIA-40> ffmpeg: 03Paul B Mahol 07master * rd4eeadcbbf 10ffmpeg/libavcodec/truespeech.c: [02:46] <CIA-40> ffmpeg: truespeech: align buffer [02:46] <CIA-40> ffmpeg: DSPContext.bswap_buf() requires aligned output [03:07] <durandal_1707> rgb24 is failing for pixfmts_copy [03:09] <durandal_1707> michaelni: what that test actually do? [03:35] <michaelni> copy should do nothing, just pass the data through [03:35] <durandal_1707> why it have changed then [03:42] <michaelni> where did it change ? [03:43] <durandal_1707> http://fate.ffmpeg.org/report.cgi?time=20120202011422&slot=powerpc-linux-gn… [03:44] <michaelni> the ppc box overheats sometimes AFAIK [03:45] <michaelni> ill wager a bet it will be green on the next run [03:45] <durandal_1707> we will see :) [05:39] <CIA-40> ffmpeg: 03Paul B Mahol 07master * rb8b77abe92 10ffmpeg/libavcodec/allcodecs.c: [05:39] <CIA-40> ffmpeg: cosmetics: realign vertically [05:39] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [05:39] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:39] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r6462d28dcc 10ffmpeg/libavcodec/apedec.c: [05:39] <CIA-40> ffmpeg: apedec: Fix alignment and fate. [05:39] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:40] <durandal_1707> michaelni: that ape fix is going to be pretty slow [05:42] <durandal_1707> it should use av_fast_malloc(), possible fix is on Libav ml waiting review [05:48] <michaelni> i made no attempt to make it fast, just fix the code so it _works_ until someone writes a better fix ... [14:02] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * rde05e41bfc 10ffmpeg/libavdevice/x11grab.c: [14:02] <CIA-40> ffmpeg: Use the correct pix_fmt for 32bit x11grab. [14:02] <CIA-40> ffmpeg: Remove adding a constant value to each pixel to make it opaque. [14:06] <ubitux> i'm going to push the new timecode api today [14:07] <ubitux> and all the patches of the patchset [14:07] <ubitux> unless someone has something to say about it [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rb18ebcbe83 10ffmpeg/tests/ (6 files in 2 dirs): timecode: add write regressions tests. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r0eaa123b34 10ffmpeg/ (5 files in 2 dirs): lavu: add public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r11e5d3b9cf 10ffmpeg/libavformat/dv.c: dv: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r77971609de 10ffmpeg/libavformat/ (isom.h mov.c): mov: honor tmcd flags while extracting timecode meta. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rdc386a5e3c 10ffmpeg/ffprobe.c: ffprobe: use av_mpegtc_to_timecode_string(). [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rf65600d519 10ffmpeg/ (doc/filters.texi libavfilter/vf_drawtext.c): drawtext: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rd8804905eb 10ffmpeg/libavcodec/ (mpeg12enc.c mpegvideo.h): mpeg12enc: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * re2407556f1 10ffmpeg/libavformat/gxfenc.c: gxfenc: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rc79eddaff1 10ffmpeg/ (6 files in 3 dirs): [14:41] <CIA-40> ffmpeg: lavfi/aconvert: use libswresample. [14:41] <CIA-40> ffmpeg: This commit also drops the planar parameter; you now need to use the 'p' [14:41] <CIA-40> ffmpeg: suffix in order to request a planar sample format. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * re96be8409f 10ffmpeg/ (configure libavfilter/Makefile libavfilter/af_aresample.c): lavfi/aresample: use libswresample. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r9f0b0db0d3 10ffmpeg/libavfilter/af_aformat.c: [14:41] <CIA-40> ffmpeg: lavfi/aformat: use do..while(0) form for macro. [14:41] <CIA-40> ffmpeg: This avoids some empty statements. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rbf6e83a8e8 10ffmpeg/libavformat/mxfenc.c: mxfenc: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rd055c3286c 10ffmpeg/doc/ (ffmpeg.texi filters.texi): doc: document amerge filter as an alternative for the -map_channel limitation. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r85c66793d7 10ffmpeg/libavfilter/af_pan.c: [14:41] <CIA-40> ffmpeg: lavfi/pan: copy ref props after filtering samples. [14:41] <CIA-40> ffmpeg: At least PTS needs to be copied to avoid breaking options such as -t in [14:41] <CIA-40> ffmpeg: ffmpeg. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rd50a4c4a5b 10ffmpeg/libavfilter/af_amerge.c: [14:41] <CIA-40> ffmpeg: lavfi/amerge: copy ref props after filtering samples. [14:41] <CIA-40> ffmpeg: This fixes various issues with ffmpeg -ss and -t. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rbd10f01aa8 10ffmpeg/libavformat/mov.c: mov: use new public timecode API. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * red67dac093 10ffmpeg/libavcodec/mpeg12.c: mpeg12: use av_mpegtc_to_timecode_string(). [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * rb90d79ec1f 10ffmpeg/libavcodec/ (timecode.c timecode.h): timecode: drop lavc timecode on next bump. [14:41] <CIA-40> ffmpeg: 03Clément BSsch 07master * r6f55156234 10ffmpeg/libavformat/dvenc.c: dvenc: use new public timecode API. [18:46] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * raf0f8c09cc 10ffmpeg/libavformat/isom.c: [18:46] <CIA-40> ffmpeg: Cosmentics: Fix AVUI comment. [18:46] <CIA-40> ffmpeg: The codec contains no alpha information but deinterleaved interlaced video. [18:46] <CIA-40> ffmpeg: Found by Maksym Veremeyenko. [19:30] <durandal_1707> is ffv1-img supposed to be extension for image encoded with FFV1 ? [19:33] <Compn> where do you see ffv1-img ? [19:34] <durandal_1707> in img2 muxer [19:34] <durandal_1707> actually in img_tags [19:35] <durandal_1707> it could be only raw format, because ffv1 does not store dimensions [19:38] <Compn> could be, good question tho, i dunno [19:38] <Compn> michaelni knows [19:39] <durandal_1707> but nothing generate such files (automaticaly with mentioned tag), but support for decoding does exist [19:42] <durandal_1707> look at 03cfe134ca7b07edfdcfa269ac12da52aa363cc3 [19:42] <durandal_1707> with this its possible to encode&decode any video codec to individual (1 file per frame) files [20:06] <Daemon404> i see that OS/2 guy is still sending in fixes for long-dead things [20:06] <Daemon404> like aout [20:11] <durandal_1707> aout is not dead [20:11] <funman> KO is crazy [20:12] <Daemon404> durandal_1707, what uses it? [20:13] <durandal_1707> Daemon404: my sick relative [20:13] <Daemon404> durandal_1707, what? [20:13] <Daemon404> i was asking why it isn't dead [20:13] <Daemon404> i cant think of anythign even remotely prominent that uses it [20:13] <durandal_1707> because some apparently still use it [20:13] <Daemon404> for all intents and purposes it is [20:14] <Daemon404> except for that oen dude who is sitll using os/2 [20:14] <Daemon404> (aka kO) [20:14] <durandal_1707> Daemon404: ask that guy, he will have best answer [20:14] <Daemon404> i dont expect someone who thinks os/2 is still alive to be grounded i nreality [20:14] <Daemon404> but i see you wish to argue semantics [20:14] <Daemon404> for all practical use... it is dead. [20:15] <Daemon404> a few crazies using it does nto make it alive. [20:15] <durandal_1707> aout is dead, long live aout [20:15] <Daemon404> i cant tell if youre trolling or just being really anal about teh use of teh word [20:16] <durandal_1707> stop beating dead horse [20:16] <Daemon404> i have no idea what youre on about now [20:16] <durandal_1707> i mean, just ignore it [20:16] <Daemon404> you havent communicated a single thign clearly. [20:18] <durandal_1707> there are people who still use old aout legacy code [20:18] <durandal_1707> for whatever reason [20:18] <Daemon404> [14:14] <+Daemon404> for all practical use... it is dead. [20:18] <Daemon404> im saying in general it is dead except for a few crazies. [20:19] <Daemon404> im sure there's still someone, somewhere using a VAX system too. [20:37] <cbsrobot> somebody in possesion of 12 bit yuv files ? [20:38] <Daemon404> cbsrobot, no but they're trivial to make... [20:39] <Daemon404> tho if ffmpeg doesnt hav a 12-bit yuv colorspace then maybe a tad more annoying [20:39] <Daemon404> (given an e.g. 16-bit rgb input) [20:41] <cbsrobot> I was wondering if it is a useful addition, but I didn't find any codec that uses it ... [20:41] <cbsrobot> maybe prores4444 ? [20:44] <CIA-40> ffmpeg: 03Clément BSsch 07master * r8c48652ff0 10ffmpeg/libavcodec/dct-test.c: dct-test: remove odivx_idct_c dead prototype. [20:46] <cbsrobot> it seems the arri alexa uses 12 bit prores 4444 - anybody a sample ? [20:46] <cbsrobot> hmm - but is it rgb or yuv ... ? [20:46] <Daemon404> prores is always yuv afaik [20:53] <cbsrobot> http://documentation.apple.com/en/finalcutpro/professionalformatsandworkflo… [20:53] <cbsrobot> yuv and rgb - it seems [20:54] <nevcairiel> maybe that means input or something? All encodings i know of are yuv [20:55] <cbsrobot> http://documentation.apple.com/en/finalcutpro/professionalformatsandworkflo… [20:55] <cbsrobot> The R, G, and B channels are lightly compressed, with an emphasis on being perceptually indistinguishable from the original material. [20:55] <cbsrobot> lol [20:55] <nevcairiel> hm maybe they just encode the color planes as if its yuv, who knows [20:56] <Daemon404> well prores is kidna just a modified mjpeg... [20:56] <Daemon404> so... [20:58] <bcoudurier> I'm not sure prores supports rgb [20:58] <bcoudurier> it probably gets converted to yuv444 [21:01] <cbsrobot> bcoudurier: does it support 12 bit ? [21:01] <nevcairiel> I thought it always was 10bit [21:01] <cbsrobot> some say the alexa outputs 12bit (soon) ... [21:04] <nevcairiel> According to some white paper i found, the 4:4:4:4 mode can work with 12-bit image data and 16-bit alpha [21:08] <nevcairiel> Although its not really all that clear if it really encodes the 12-bit or just "supports image sources up to 12 bits" as the paper claims <.< [21:41] <gnafu> "< cbsrobot> The R, G, and B channels are lightly compressed, with an emphasis on being perceptually indistinguishable from the original material." <-- Sounds like a menu item in a fancy French restaurant. [21:57] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r9b9bf5ab73 10ffmpeg/libavcodec/ffv1.c: [21:57] <CIA-40> ffmpeg: ffv1: cosmetics: indentation [21:57] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [21:57] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [21:57] <CIA-40> ffmpeg: 03KO Myung-Hun 07master * rc853124fb0 10ffmpeg/libavcodec/x86/pngdsp.asm: [21:57] <CIA-40> ffmpeg: Use SECTION_TEXT instead of section .text for the compatibility [21:57] <CIA-40> ffmpeg: aout does not support 'align='. [21:57] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:57] <j-b> Daemon404: pingie [22:57] <Daemon404> hi [22:58] <j-b> Daemon404: update on the libpostproc repo? [22:59] <Daemon404> git was taking so long to run filter-branch i fell asleep last night [22:59] <Daemon404> and ive been at uni all day today... [23:00] <Daemon404> eta tomorrow for all intents and purposes [23:01] <j-b> Daemon404: I was just wondering, no pressure. [23:01] <j-b> and I can give you write access on git.v.o if you ever care. [23:03] <Daemon404> nah i can use github [00:00] --- Fri Feb 3 2012

1 0

ffmpeg.log.20120202
by burek 03 Feb '12

03 Feb '12

[00:02] <relaxed> praedo: try using the flv container [00:18] <Freakshow> ha! relaxed: I saw that comment but missed the notion if was for someone else and thought to myself, 'why?' [00:18] <Freakshow> :P [00:18] <Freakshow> I'm seeing this in the output... [00:18] <Freakshow> [rtsp @ 0x101849200] method ANNOUNCE failed: 401 Unauthorized [00:18] <Freakshow> Could not write header for output file #0 (incorrect codec parameters ?) [00:19] <praedo> relaxed, i do [00:19] <praedo> this is my commnad: [00:20] <praedo> ffmpeg -i file -re -acodec copy -vcodec copy -f flv rtmp://wowzaserver [00:20] <praedo> some of the mp4 files play and some don't [00:20] <praedo> and they are all encoded in the same way [00:20] <praedo> why? [00:23] <indigo> uhm... I'm still getting "libxyz.dll not found" [00:23] <Freakshow> praedo: what's the output on the wowza side look like? is it giving you any ideas of what's going on with those inputs? [00:24] <Freakshow> when they don't work; that is [00:24] <indigo> if I copy the libs (é.dll) the executable does... nothing! [00:25] <darkstarbyte> Where can I get the ffms header for x264? [00:35] <indigo> I think that ffmpeg enable automatically zlib and zlib2 if it finds those libs in libpath [00:35] <indigo> is it right? [00:35] <praedo> Freakshow, wowza logs show this: [00:36] <praedo> INFO server comment 2012-02-01 23:35:23 - - - - - 2427.688 - - - - - - - client connectionClosed [1232396645] pingtimeout [00:36] <praedo> for files that don't play [01:04] <Freakshow> which of the wowza logs is that? [01:04] <Freakshow> error or access? [01:07] <Freakshow> I'm judging by the output it's access [01:07] <Freakshow> but wanted to be sure [01:11] <Sorcey> praedo: I never saw 'pingtimeout' at wowza before? [01:11] <Sorcey> did you try serving the file through http to flowplayer? [01:11] <Sorcey> posting the link to a place where we can check out your .mp4 file might be useful as well [01:15] <praedo> in flowplayer as http both files work [01:15] <praedo> but as rtmp some work and some don't [01:15] <praedo> can i send you both files to compare? [01:15] <praedo> i used mediainfo and all seems equal for me [01:16] <Sorcey> try streaming to flowplayer without -re [01:16] <Sorcey> err, to wowza [01:16] <Sorcey> you can't dcc to me but you can pm me a link [01:16] <Sorcey> I'll be going to bed soon tho :) [01:17] <praedo> http://frilab.frigola.cat/working_in_rtmp.mp4 [01:17] <praedo> http://frilab.frigola.cat/not_working_in_rtmp.mp4 [01:17] <praedo> Sorcey, please have a look at them [01:19] <Sorcey> I see only one difference: the one that's working has a variable bitrate aac soundtrack [01:19] <Sorcey> the other one has constant bitrate [01:20] <praedo> strange because encoding was exactly the same [01:20] <Sorcey> I'll try streaming to wowza using my own test setup [01:20] <praedo> yes please, thanks [01:21] <praedo> ffmpeg -i file -re -acodec copy -vcodec copy -f flv rtmp://wowzaserver [01:21] <praedo> like this? [01:23] <Sorcey> yup [01:23] <Sorcey> both work using rtsp in vlc [01:23] <Sorcey> hmm, buffering [01:25] <Sorcey> by the way, these files are 48 kHz [01:25] <praedo> i tried flowplayer both 3.2.5 and the latest 3.2.7 [01:25] <praedo> with rtmp plugin [01:26] <Sorcey> I remember something about flv not supporting other samplerates than 11, 22 and 44 kHz [01:26] <praedo> but i have many working mp4 that can be streamed being 48 [01:26] <Sorcey> yeah, I know [01:26] <Sorcey> it works here too [01:26] <praedo> and the non_working too? [01:26] <Sorcey> ah, that's an mp3 restriction only I think [01:27] <Sorcey> people used to stream using h263 + mp3 audio back in the days :P [01:27] <praedo> the flash player can't play it because wowza logs timeout every time [01:27] <praedo> only for some files [01:27] <praedo> always for that non_working file [01:28] <Sorcey> I get this in my logs: [01:28] <Sorcey> Fragment duration greater than suggested range of 1-4 seconds. Adjust keyframe interval accordingly: Fragment durations: [10.0,10.0,10.0] [01:29] <Sorcey> 10 seconds is the default [01:29] <Sorcey> Fragment duration greater than suggested range of 1-4 seconds. Adjust keyframe interval accordingly: Fragment durations: [7.1,10.0,2.8] [01:29] <Sorcey> that's the working file [01:29] <Sorcey> might fail because flowplayer receives too much data without receiving a keyframe maybe? [01:29] <praedo> what version? [01:30] <Sorcey> the non_working = 10+10+10, the working = 7+10+3 [01:30] <praedo> what that means? [01:30] <Sorcey> add the following parameter in ffmpeg: [01:31] <Sorcey> -g 100 [01:31] <Sorcey> when encoding the .mp4 [01:31] <Sorcey> that means 1 keyframe every 4 seconds [01:31] <Sorcey> or better: [01:31] <Sorcey> -keyint_min 4 [01:31] <praedo> not both? [01:31] <Sorcey> -keyint_min 100 [01:31] <Sorcey> sorry [01:31] <Sorcey> I oughtta go to bed :x [01:31] <Sorcey> no, just use -keyint_min 100 [01:32] <Sorcey> it means ffmpeg should use at least 1 keyframe every 4 seconds [01:32] <praedo> can you wait for me to encode and upload the files again? [01:32] <Sorcey> yeah, sure [01:32] <praedo> 10min. please [01:32] <Sorcey> with a bit of luck they'll work immediately [01:33] <praedo> i hope so... [01:37] <praedo> ffmpeg -i working.avi -keyint_min 100 -deinterlace -vcodec libx264 -preset medium -crf 24 -threads 0 -acodec libfaac -aq 100 working.mp4 [01:37] <praedo> is it correctly placed there? [01:47] <Sorcey> I think so yeah [01:47] <Sorcey> you should use -vf yadif instead of -deinterlace [01:47] <praedo> uploaded again [01:48] <Sorcey> deinterlace doesn't work on your file, or at least not the mp4 you sent earlier [01:48] <praedo> you can download them again from the same place and compare [01:48] <praedo> the not_working still doesn't play [01:48] <praedo> http://frilab.frigola.cat/working_in_rtmp.mp4 [01:48] <praedo> not this link... [01:48] <praedo> http://frilab.frigola.cat/working.mp4 [01:48] <praedo> http://frilab.frigola.cat/not_working.mp4 [01:49] <praedo> deinterlace worked now and audio vbr should be the same in both [01:49] <praedo> but for some reason the not_working can't be streamed [01:51] <Sorcey> I can't help you, I'm sorry. Don't know what the problem is... but I know I can stream them without any problem to vlc [01:51] <Sorcey> rtsp://servername:554/application/stream_id [01:52] <Sorcey> so you might wanna test that to convince yourself that wowza is not the problem [01:52] <praedo> do i have 554 in my wowza too by default to test with vlc? [01:53] <praedo> i get that timeout in wowza logs for the not_working file [01:54] <Sorcey> my mistake [01:54] <Sorcey> 1935 [01:54] <Sorcey> rtsp://servername:1935/application/stream_id [01:55] <Sorcey> should work in vlc for the not_working file [01:55] <praedo> it asks for a pass and i don't have any set [01:56] <Sorcey> weird... it does that for both files I guess? [01:57] <praedo> yes [01:57] <praedo> is it possible to set a pass authentication in ffmpeg? [01:57] <praedo> applications should have a pass for security i guess [01:57] <Sorcey> I don't think it's possible that ffmpeg causes wowza to ask for authentication @ clients [01:58] <Sorcey> it's probably your wowza settings [01:58] <praedo> i use MediaSecurity addon [01:58] <praedo> but not for this application [01:59] <Sorcey> is rtsp enabled for this application? [01:59] <Sorcey> anyway, it's 2am [01:59] <Sorcey> and I have no clue what your poblem is [02:00] <Sorcey> hmm [02:00] <Sorcey> well, 1 thing though: you use a non 1:1 PAR [02:01] <praedo> what about audios this time? [02:01] <Sorcey> you have a 720:576 file (1.25:1) and you have a 768:576 output aspect (1.333:1) [02:02] <praedo> i didn't know... why? [02:02] <Sorcey> that's common, but flash doesn't support actively reading the aspect ratio so unless you force it to scale to 1.333:1 it will display as 1.25:1 [02:03] <Sorcey> solution is altering the aspect ratio in ffmpeg, scaling to 768:576 (@ vf scale) and add -aspect 1.25 (for some reason that's necessary to reduce the PAR to 1:1) [02:04] <Sorcey> then your output file is native 768x576 [02:04] <Sorcey> Reminds me doing all this stuff right costs a lot of time :P [02:05] <praedo> but source DV footage isn't 768:576 [02:05] <Sorcey> Anyway, good luck with fixing your broken file. Most solutions to problems like these come to me 5 minutes after I wake up in the morning :P [02:05] <praedo> okay, have a rest then [02:05] <Sorcey> 4:3 content is often stored as 720x576 and then expanded [02:05] <praedo> thanks! [02:06] <Sorcey> just like 16:9 is also stored as 720x576 and then expanded to 1024x576 while displaying it [02:06] <Sorcey> some DVB providers (like here in NL) abuse the pixel aspect ratio settings and broadcast tv in 520x576 and then expand to 768 of 1024 width [02:07] <Sorcey> if you play it in vlc and then check media info you'll see 768x576 there [02:07] <Sorcey> nn! [02:31] <praedo> INFO server comment 2012-02-02 02:19:15 - - - - - 927.844 - - - - - - - client connectionClosed [1505833773] watchdog [02:31] <praedo> i get this error each time i try to stream the not_working.mp4 file [02:36] <praedo> INFO server comment 2012-02-02 02:35:58 - - - - - 1930.478 - - - - - - - client connectionClosed [77176954] pingtimeout [02:37] <Freakshow> have you tried increasing the logging output of wowza to see what else is going on? [02:37] <Freakshow> and which version of wowza are you publishing to? [02:45] <Freakshow> praedo? [02:49] <praedo> yes [02:49] <praedo> 2.2.3 [02:49] <praedo> how is logging increased? [02:51] <Freakshow> hang on [02:52] <Freakshow> you'll modify log4j.properties [02:52] <Freakshow> top line should be something like [02:52] <Freakshow> log4j.rootCategory=INFO, stdout, serverAccess, serverError [02:52] <Freakshow> if I'm not mistaken, you'll need to add debug to this list [02:53] <Freakshow> ah, change INFO to DEBUG [02:53] <Freakshow> :D [02:53] <Freakshow> lmgtfy [02:53] <Freakshow> Debug Logging [02:53] <Freakshow> To turn on Wowza Media Server debug logging, edit the [install-dir]/conf/log4j.properties file and change the log level on the first line from INFO to DEBUG. [02:54] <Freakshow> *boom* [02:54] <Freakshow> but be prepared... it's A LOT logging... scroll blindness will ensue quickly [03:34] <tiglionabbit> can I use ffmpeg to change the container type of a video file without re-encoding it? [03:34] <tiglionabbit> oh wait I should have just googled that [03:37] <tiglionabbit> though I just tried it and ended up with something unplayable =[ [03:38] <relaxed> tiglionabbit: -vcodec copy -acodec copy [03:39] <tiglionabbit> yes that's exactly what I did, but the result is garbage. libmagic even just calls it "data" [03:41] <relaxed> command? [03:42] <tiglionabbit> ffmpeg -i chair.avi -vcodec copy -acodec copy chair2.mp4 [03:42] <tiglionabbit> with that one at least file knows what type it is, but it wont play in quicktime. It's all black. it's 1 minute 28 seconds of black. [03:43] <tiglionabbit> when I try and make it an mkv, file doesn't know what it is [03:43] <tiglionabbit> what is a good container format to use with web browsers btw? [07:02] <elkng> does ffplay supports playing one frame and then go into pause, same as using "." in mplayer ? [07:09] <elkng> I got *.flv video file from youtube didnt watch it, did converted it into *.avi, the video is 80 seconds long and there is 3 black frames on 70th second with text: "WARNING: NOTHING TO OUTPUT BFRAME DECODER LAG", is this ffmpeg's issue ? [10:14] <slackyman> Hi everybody [10:15] <slackyman> cross compiling ffmpeg results in an executable that requires zlib-1.dll [10:15] <slackyman> but I want to make a singe, huge, all-in-one executable [10:16] <slackyman> more, once i copy zlib-1.dll the ffmpeg.exe is not functional [10:16] <slackyman> running it on a windows machine DOES nothing [10:50] <nadavrub> Any1 with ffmpeg/andoid experiance ? [10:53] <nadavrub> Any1 with Android experiance ? [11:53] <slackyman> connection lost, again [11:53] <slackyman> sorry [12:17] <slackyman> is there any1? [12:39] <dddbmt> Hi guys, I am trying to convert an FLV into SWF. When I do that the sounds is converted correct, but I do not get any video in the SWF output file. [12:39] <dddbmt> http://pastie.org/private/pvrxvp6jlxqpsbgu9vd9w [12:40] <dddbmt> If I open the SWF i VLC I _do_ get some video, but it's very fast and only a few seconds. [13:06] <hroi> hi [13:06] <hroi> I have a video which I want to cut, start frame and end frame, but not change the encoding or anything else... [13:06] <hroi> I want to have as little image quality loss as possible... can ffmpeg do this? [13:07] <hroi> In theory it should be possible to avoid loss when cutting right? [13:13] <zap0> depends on keyframe positions [13:26] <hroi> zap0: seems now that cutting doesnt work well -vcodec copy on mp4 files. [13:26] <hroi> zap0: I guess because the video needs to be recompressed [13:26] <hroi> zap0: perhaps I should set -sameq instead. [13:48] <hroi> How do I set the codec manually using -vcodec? I want it to be the same as the input... but I dont want the copy option... I want it to recompress. [13:48] <hroi> I tried -vcodec h264... and it dont accept that codec... or any I see name in ffmpeg -formats [13:49] <hroi> how do I get a list of valid -vcodec options. [14:14] <Mavrik> hroi, ffmpeg -codecs [14:14] <Mavrik> for h264 you'll have to use -vcodec libx264 [14:14] <Mavrik> (and have the codec compiled with lib264) [14:14] <Mavrik> er... ffmpeg compiled with libx264 not codec, sorry :) [14:14] <hroi> Mavrik: hmm.. so even though my ffmpeg can decode it... it may not be able to encode it [14:15] <Mavrik> yes [14:16] <Mavrik> encoders and decoders are actually quite different things :) [14:17] <hroi> Mavrik: ffmpeg -codecs dont list anything... it says missing argument [14:18] <Mavrik> just how old is your ffmpeg? [14:18] <hroi> 2007 [14:19] <Mavrik> yeah... [14:19] <Mavrik> how about updating it :) [14:19] <hroi> :p I love old distros... nostalgia [14:20] <hroi> Mavrik: yeah I will... just stuck on this system right now [14:20] <Mavrik> http://www.virag.si/2012/01/web-video-encoding-tutorial-with-ffmpeg-0-9/ [14:20] <Mavrik> this can probably help you a little :) [14:21] <hroi> Mavrik: is there an alternative h264 codec sometimes used? [14:21] <Mavrik> that's... probably the wrong question to ask [14:21] <Mavrik> what are you trying to achieve? [14:22] <hroi> keep h264 type video [14:22] <hroi> mp4 or something [14:22] <Mavrik> ok [14:22] <hroi> the xvid codec seems to be wroking for me. [14:22] <Mavrik> H.264 is a video format [14:23] <Mavrik> MP4 is a container [14:23] <Mavrik> that's a difference, MP4 does not imply x264 ;) [14:23] <Mavrik> you're mixing up video formats, containers and encoders [14:23] <Mavrik> that's why I'm asking you what are you trying to achieve. [14:23] <hroi> Mavrik: ahh ok right... it has many options right, like quicktime [14:23] <hroi> and wmv [14:24] <Mavrik> those are all containers :) [14:24] <Mavrik> what you put in those containers is another thing :) [14:24] <hroi> Mavrik: hey... after cutting my vid... it loses the right aspect ratio... perhaps I can reset it manually [14:24] <hroi> -aspect 1:1.85 [14:24] <hroi> I think [14:25] <hroi> HDTV aspect [14:25] <hroi> again the ffmpeg help and man dont actually tell me what to write after aspect. [14:25] <hroi> just tell me there is -aspect option [14:57] <Mavrik> hroi, seems very clear in ffmpeg documentation here: http://ffmpeg.org/ffmpeg.html [14:57] <Barvinok> hi all. My ffmpeg 0.7.8 seems to ignore video bitrate settings when used with -vcodec libtheora. Is that a known bug? [14:58] <Barvinok> evidently, http://pastebin.com/LgwpQf8R . I set 500 kbps everywhere and get ~3800 kbps on output [15:07] <eebrah> Hi everyone. Trying to compile ffmpeg 0.10 with libx264 using "--enable-libx264" and I am getting the message "ERROR: libx264 not found" despit having compiled it( x264 ) a couple of minutes ago, can anyone help? [15:44] <relaxed> eebrah: install libx264-devel [15:45] <relaxed> oh, you compiled it. make sure you don't have libx264-devel installed :) [15:46] <darkstarbyte> You may also want to install ffmpeg source [16:07] <rbrt> hello! i just read "trac issue #92". anybody knows if somebody is working on a "packet-level filtering mechanism" ? [17:00] <TheBigD> I'm trying to stream the microphone output from an usb soundcard connected to an openwrt router, ideally creating something like a wireless IP baby monitor. [17:01] <TheBigD> When I first try to test the concept by streaming an mpr file I get the following output: Unknown input or output format: rtp [17:01] <TheBigD> or the complete output: http://pastebin.com/VYhSccTw [17:03] <TheBigD> Am I using it wrong? [17:16] <relaxed> TheBigD: ffmpeg versions from 2009 are hard to support. I recommend installing a recent version, which may fix your problem. [17:19] <TheBigD> hmm, I see. I have never compiled anything and the most recent version for openwrt I can find is 0.8.7-1, would that be recent enough? [17:24] <relaxed> that might work [18:13] <Rango3> anyone know where I can readup on the new syntax like -v:codec ? [18:24] <relaxed> you mean -c:v (codec:video)? [18:24] <relaxed> man ffmpeg [19:06] <Rango3> yea I did relaxed thanks :) [19:13] <grepper> relaxed: do you know if the old specifiers (as well as aliases) will be eventually removed ? Like -ab instead of -b:a and -b instead of -b:v and -vcodec/-acodec [19:45] <relaxed> grepper: I believe so. '-ab' would be removed but '-b' would apply to all streams, so it would stay. [19:49] <grepper> relaxed: okay, thanks for the info [19:50] <grepper> not sure if there is a better way of making a script (tovid) backwards compatible except by parsing ffmpeg --help output [19:57] <tiglionabbit> hi. I'm having a little trouble with ffmpeg. I have a 1 minute 28 second avi here that according to ffmpeg is mpeg4+mp3. I'm trying to change the container format like so: http://friendpaste.com/78AvO29GcRoJ7yGl2v7Jnu . However, the result is 1 minute and 28 seconds of black nothingness when I play it in quicktime. I've tried making it an mkv but that has the same result. Making it an mov wo [19:57] <tiglionabbit> rks though. [21:26] <slackyman> I resolved all my issues regarding cross-compiling a static ffmpeg executable [21:26] <slackyman> can be useful to let you know? [21:49] <ubitux> slackyman: yes [21:49] <ubitux> you might want to add such thing in wiki [21:49] <slackyman> ok, thans [21:50] <ubitux> https://ffmpeg.org/trac/ffmpeg/wiki/CompilationGuide [21:50] <ubitux> maybe add a page or something] [21:50] <slackyman> thanks [21:50] <slackyman> bye [21:59] <tiglionabbit> can I get ffmpeg to tell me the codecs a file is currently encoded with? [22:01] <jkramer> Ahoy [22:02] <jkramer> Is it possible to build ffmpeg without all the X11 dependencies? I'd like to use it on a server only for generating video thumbnails [22:02] <cbreak> I am quite sure you can [22:02] <cbreak> just --disable the stuff you don't want [23:15] <tiglionabbit> how do I find out what codecs a container has in it? [23:15] <tiglionabbit> like if I have an mp4, how can I check if it has h264 video or not? [23:16] <cbreak> ffprobe? mplayer -identify? [23:21] <tiglionabbit> cbreak: thanks. I was looking for avprobe but couldn't find it in any packages. I assume ffprobe is the same thing? [23:21] <cbreak> no idea [23:21] <cbreak> ffprobe is built when compiling ffmpeg [23:34] <Freakshow> hmm... how would one pass user/pass to ffmpeg for rtsp output? [23:35] <Freakshow> can that be done via command line? [00:00] --- Fri Feb 3 2012

1 0

ffmpeg-devel.log.20120201
by burek 02 Feb '12

02 Feb '12

[00:17] <durandal_1707> kierank: is there a way to pick between them? [00:17] <michaelni> durandal_1707, ffv1s uses a context adaptive coder which is not reset between frames by default [00:17] <kierank> if you care about latency then use slice threading, if you don't use frame threads :) [00:17] <kierank> not that simple but meh [00:59] <walker> Hi folks - I'm having an issue with h263 Mode B packets. From looking at ffmpeg-devel archives, it seems that ffmpeg only supports Mode A for h263 RTP? [01:00] <walker> Is there any work on support for Mode B? [01:01] <Compn> walker : good question. might want to ask whoever worked on h263 rtp [01:02] <durandal_1707> walker: do you have sample? [01:04] <walker> durandal_1707: I do. [01:05] <durandal_1707> then open bug on ffmpeg trac explaining your problem and upload it there [01:05] <walker> Okay. [01:05] <walker> I just want to know at this point, though, if Mode B is supported at all? [01:06] <Compn> i dont know :) [01:06] <Compn> maybe wbs does [01:06] <Compn> wbs : ping [01:06] <durandal_1707> if it is not supported how it will get support if you never report it? [01:07] <durandal_1707> only if someone else reports it or fix it [01:07] <walker> From https://lists.libav.org/pipermail/ffmpeg-devel/2009-April/067074.html - "It seems reasonable to only implement Mode A" [01:07] <Compn> walker : got a url that has problems ? [01:07] <walker> Compn: Lemme stash it somewhere public. [01:08] <Compn> walker : the guy who resent that patch is wbs , and michaelni is the one who reviewed it [01:09] <Compn> did that patch ever get applied ? [01:09] <walker> As far as I can see, it still supports only mode A. [01:09] <walker> The stupid thing is, this h263 stream contains both mode A and mode B. [01:09] <Compn> lol [01:09] <walker> (Commercial videophone.) [01:09] <Compn> i'm surprised there are still h263 streams [01:09] <Compn> ah [01:10] <Compn> h263 is last century maaaan [01:10] <walker> Tell that to my client! I developed the software for h264 and h263+. Submitted it. They came back with, "It no woooork!" [01:10] <Compn> haha [01:12] <walker> http://static.carbonatedblog.com/h263_test.h263 [01:13] <walker> That's the raw h263 rtp stream containing mode A and B packets. Wireshark confirms both A and B in there. [01:14] <walker> ffmpeg does its best: It takes the mode A packets and renders them to a rather bad video, making my ugly mug even uglier. [01:14] <walker> But playing back that stream straight to the videophone gets crystal clear video. [01:15] <Compn> alrighty, well stick around and wbs should come on sooner or later. or take it to the bug tracker [01:19] <walker> Thanks! [01:19] <walker> Poking around bugtracker. [02:27] <xxthink> which video format mp4 does NOT support? [03:06] <CIA-40> ffmpeg: 03Martin Storsjö 07master * r89f6e8a20c 10ffmpeg/libavformat/flvdec.c: (log message trimmed) [03:06] <CIA-40> ffmpeg: flvdec: Interpret a toplevel 'object' type metadata item as normal metadata, too [03:06] <CIA-40> ffmpeg: Previously, we've only passed the key string on to the recursive [03:06] <CIA-40> ffmpeg: amf_parse_object for the mixedarray type, not for 'object'. By [03:06] <CIA-40> ffmpeg: passing the key string on, the recursive amf_parse_object can [03:06] <CIA-40> ffmpeg: store the amf objects as metadata. [03:06] <CIA-40> ffmpeg: This kind of data was seen in data from XSplit Broadcaster, received [03:06] <CIA-40> ffmpeg: 03Diego Biurrun 07master * r91bafb52ae 10ffmpeg/libavcodec/x86/ (Makefile rv40dsp.c): x86: Give RV40 init file a more suitable name. [03:06] <CIA-40> ffmpeg: 03Martin Storsjö 07master * r75ab1e62d4 10ffmpeg/libavformat/mov.c: (log message trimmed) [03:06] <CIA-40> ffmpeg: movdec: Ignore sample_degradation_priority bits when checking first_sample_flags [03:06] <CIA-40> ffmpeg: This makes the first packet of a track fragment run to get [03:06] <CIA-40> ffmpeg: the keyframe flag set properly if sample_degradation_priority [03:06] <CIA-40> ffmpeg: is nonzero. [03:06] <CIA-40> ffmpeg: This makes the keyframes flag be set properly for ismv files [03:06] <CIA-40> ffmpeg: created by Microsoft. [03:06] <CIA-40> ffmpeg: 03Christophe Gisquet 07master * re5c9de2ab7 10ffmpeg/libavcodec/x86/ (Makefile rv40dsp.asm rv40dsp_init.c): (log message trimmed) [03:06] <CIA-40> ffmpeg: rv40: x86 SIMD for biweight [03:06] <CIA-40> ffmpeg: Provide MMX, SSE2 and SSSE3 versions, with a fast-path when the weights are [03:06] <CIA-40> ffmpeg: multiples of 512 (which is often the case when the values round up nicely). [03:06] <CIA-40> ffmpeg: *_TIMER report for the 16x16 and 8x8 cases: [03:06] <CIA-40> ffmpeg: C: [03:06] <CIA-40> ffmpeg: 9015 decicycles in 16, 524257 runs, 31 skips [03:06] <CIA-40> ffmpeg: 03Justin Ruggles 07master * reac31dd163 10ffmpeg/libavcodec/mpc7.c: [03:06] <CIA-40> ffmpeg: mpc7: align local temp buffer [03:06] <CIA-40> ffmpeg: DSPContext.bswap_buf() requires aligned output [03:06] <CIA-40> ffmpeg: 03Justin Ruggles 07master * rddf70db6d7 10ffmpeg/libavcodec/adpcmenc.c: [03:07] <CIA-40> ffmpeg: The new name is more appropriate as only golomb functions are tested. [03:07] <CIA-40> ffmpeg: 03Diego Biurrun 07master * r9e3a2736bd 10ffmpeg/libavcodec/golomb-test.c: golomb-test: K&R formatting cosmetics [03:07] <CIA-40> ffmpeg: 03Diego Biurrun 07master * r52afc97168 10ffmpeg/tests/ (fate/libavcodec.mak ref/fate/golomb): fate: add golomb-test [03:07] <CIA-40> ffmpeg: 03Diego Biurrun 07master * rc30b198381 10ffmpeg/libavcodec/x86/ (fmtconvert_mmx.c rv40dsp.c): [03:07] <CIA-40> ffmpeg: x86: Place mm_flags variable declaration below the appropriate #ifdef. [03:07] <CIA-40> (14 lines omitted) [05:08] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r4b4d3d7249 10ffmpeg/libavcodec/aacenc.c: [05:08] <CIA-40> ffmpeg: aacenc: fix library references [05:08] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:08] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * rdbb7009ea5 10ffmpeg/libavformat/westwood_aud.c: [05:08] <CIA-40> ffmpeg: westwood_aud: Fix project reference [05:08] <CIA-40> ffmpeg: "Copyright (c) 2003 The ffmpeg Project" and "part of libav" [05:08] <CIA-40> ffmpeg: is ROTFL [05:08] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:08] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * rada28b7f72 10ffmpeg/libavcodec/golomb-test.c: [05:08] <CIA-40> ffmpeg: h264 golomb: Fix copyright violation. [05:08] <CIA-40> ffmpeg: Copy and paste does not change authorship [05:08] <CIA-40> ffmpeg: See 0da71265 for the original checkin [05:08] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:08] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r92f9b26c1e 10ffmpeg/doc/muxers.texi: [05:08] <CIA-40> ffmpeg: doc: fix convertion tool references [05:08] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:09] Last message repeated 1 time(s). [05:09] <CIA-40> ffmpeg: 03Michael Niedermayer 07master * r9fdae05607 10ffmpeg/libavformat/avformat.h: [05:09] <CIA-40> ffmpeg: libavformat/avformat.h: fix stray libavformat reference [05:09] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [05:09] <durandal_1707> funny stuff [05:32] <Compn> lavf: add functions for accessing the fourcc<->CodecID mapping tables. [05:32] <Compn> that is an odd commit [05:33] <Compn> i dont even know what its doing [05:33] <Compn> well i know what its doing, just not why [09:12] <j-b> Compn: http://git.videolan.org/?p=libpostproc.git;a=shortlog [13:38] <Compn> j-b : whyd libav fork postproc? :P [13:38] Action: Compn not involved :P [13:41] <ubitux> :) [13:52] <JEEB> The CodecID <=> fourcc thing was IIRC done as it was useful for some people implementing their own parsers but wanting to use libavcodec to decode. [13:53] <nevcairiel> indeed, I use these tables in my mkv demuxer [13:53] <Compn> riff was put into libavformat.o [13:54] <Compn> or .a, whichever [13:54] <Compn> oh nm [13:54] <Compn> thats needed for new demuxer anyway [15:52] <CIA-40> ffmpeg: 03Michael Bradshaw 07master * rec0d02e441 10ffmpeg/libavcodec/libopenjpegdec.c: Support decoding gray8a with libopenjpeg. [15:52] <CIA-40> ffmpeg: 03Michael Bradshaw 07master * rd76864d9ce 10ffmpeg/libavcodec/libopenjpegdec.c: Support decoding rgba64 with libopenjpeg. [16:13] <ubitux> bcoudu, bcoudurier: i'll push the new timecode api (the fullpatch) pretty soon; if you want to comment, please do it soon [16:40] <ubitux> btw, i'm going to push the lavfi stuff too [18:05] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * r93a3ba9bcd 10ffmpeg/libavdevice/v4l2.c: Fix pix_fmt mapping for v4l2 rgb5x5. [18:05] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * r4632066713 10ffmpeg/libavdevice/v4l2.c: Fix v4l pix_fmt mapping for rgb32: There is no transparency information. [18:05] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * rb2327a8202 10ffmpeg/libavdevice/v4l2.c: Add some missing v4l2 pix_fmt mappings. [19:11] <ubitux> btw, i updated the string interpolation hash, and used for (i = 0; i < len; i++) hash = (hash<<5) ^ (hash>>27) ^ key[i]; (starting at hash=0) [19:12] <ubitux> seems it is more efficient with ascii keys [19:12] <ubitux> which will certainly be the main kind of input certainly [19:12] <ubitux> drop the duplicate. [19:15] <ubitux> i also changed the way of dealing with the formated buffer [19:35] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r917803babd 10ffmpeg/libavcodec/ffv1.c: [19:35] <CIA-40> ffmpeg: ffv1enc: PIX_FMT_YUV440P support [19:35] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [19:35] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [19:35] <CIA-40> ffmpeg: 03Paul B Mahol 07master * rec09230c9a 10ffmpeg/libavcodec/ffv1.c: [19:35] <CIA-40> ffmpeg: ffv1dec: PIX_FMT_YUV440P support [19:35] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [19:35] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [20:32] <Tjoppen> uh.. ffpresets need to be copied into ~/.ffmpeg for them to work? [20:33] <Tjoppen> that seems.. dumb? [20:42] <funman> or in /usr/share/ffmpeg ? [20:42] <Tjoppen> that's where they are, but they don't work [20:46] <michaelni> Tjoppen, there are 2 mxf patches on ffmpeg-dev [20:48] <michaelni> #define FFMPEG_DATADIR "/usr/local/share/ffmpeg" [20:48] <michaelni> maybe its missing to search in both /usr and /usr/local [20:49] <Tjoppen> ok, will check. have been sick and without internet for a while [20:49] <funman> michaelni: hmm [20:49] <funman> you should respect --prefix= option [20:50] <Tjoppen> my ffpresets are in /usr/share/ffmpeg/ (and ~/.ffmpeg/ of course) [20:50] <funman> for example i use debian package, and build/install my own with --prefix=$HOME/.local/ffmpeg [20:50] <michaelni> DATADIR=\$(DESTDIR)$datadir [20:51] <michaelni> #define FFMPEG_DATADIR "$(eval c_escape $datadir)" [20:51] <michaelni> i assume this does end up respecting --prefix but ive not checked [20:51] <funman> probably [20:51] <michaelni> the quote from before was from my config.h [23:49] <Daemon404> bcoudurier, https://github.com/dwbuiten/postproc/commit/5a5d853ed3aedcffe7eab235a43460e… [00:00] --- Thu Feb 2 2012

1 0

ffmpeg.log.20120201
by burek 02 Feb '12

02 Feb '12

[00:37] <darkstarbyte> Is there a way with ffmpeg to get the resolution of a video, and store it in a bash variable? [00:40] <ShadowJK> I heard of ffprobe [00:40] <darkstarbyte> Is there a way to make it give me just the resolution? [00:44] <plut0> hi [01:04] <indigo> hi [01:04] <indigo> I'm trying to cross-compile ffmpeg with libopenjpeg [01:05] <indigo> my config.log says "undefined reference to `_opj_version'" [01:05] <indigo> end configure ends up with libopenjpeg not found [01:08] <indigo> I sue --extra-libs, --extra-cflags etc [01:08] <indigo> I have compiled libopenjpeg for mingw32 [01:10] <indigo> I hope someone can explains me why I'm having this error [01:14] <indigo> anyone? [01:17] <indigo> thanks anyway [01:17] <indigo> bye [01:28] <kenyabob> is there a way I can compile a movie using sequential images, but simply refer to the folder that encloses them, since they are sequential by a date time stamp [01:52] <magn3ts> Hi all! Is there any reason that this command would be producing an output file with rather badly unsynced video/audio? Any tips on how I would go about debugging it? [01:52] <magn3ts> ffmpeg -ss 600 -i /media/Videos/Feature\ Length\ Films/Avatar.mkv -codec:v libvpx -quality realtime -cpu-used 4 -b:v 500k -qmin 10 -qmax 42 -maxrate 500k -bufsize 1000k -threads 6 -codec:a libvorbis -b:a 128k -f webm output.webm [01:53] <magn3ts> Happens even without multithreading/coring the output and it happens whether I spit out to a file or stdout. [01:54] <magn3ts> Seems like it would be an issue in the decoder (mkv/h264) [02:49] <desti> does ffmpeg 0.10 break multithreading for libvpx? just want to do some benchmarks with libvpx 0.9.6, 0.9.7 and 1.0, using ffmpeg 0.1, but there is no difference between threads 1 and threads 8 (it spawns more htreads, but only 1 is using cpu cycles) [02:52] <cynyr> so it seems all the libx264 presets are gone now? [02:55] <desti> ok, downgraded to 0.7 and smp is working again, makes libvpx encoding speeding up with more threads [03:01] <desti> looks like there is no bug report about this open [04:10] <Hyperi> 04:56:32 <@ dtd> House S08E10 julkaistu! [04:10] <Hyperi> 07:50:21 <@ dtd> House S08E10[720p] julkaistu! [04:10] <Hyperi> 17:54:53 <@ dtd> National Geographic S00E00 - Apocalypse The Rise of Hitler 1of2 Becoming Hitler julkaistu! [04:15] <Hyperi> doh [04:46] <\DSAFEW\> so I'm still trying to capture PCM with ffmpeg [04:47] <\DSAFEW\> I have alsa, and I'm trying -i hw:0,0 and that gets the mic [04:47] <\DSAFEW\> and -i hw:0 is the same thing [05:55] <arbin> \DSAFEW\, what about -i hw:0,1? [05:55] <arbin> or -i hw:1,0 [05:55] <arbin> you trying for line-in? [05:58] <\DSAFEW\> arbin, I have a normal microphone going into an onboard ALC892, not sure what you mean by line in, in alsamixer it's just "rear mic" or something like that [05:58] <arbin> well, fiddle with it [05:59] <arbin> only thing i could think of is changing the #s to 1 [05:59] <\DSAFEW\> for both those devices: cannot open audio device hw:0,1 (No such file or directory [06:01] <\DSAFEW\> when I do it for the microphone, the audio stream says: Stream #1.0: Audio: pcm_s16le, 48000 Hz, 2 channels, s16, 1536 kb/s [06:02] <\DSAFEW\> afaik pcm_s16le is supposed to be all the sounds coming from the sound card [06:02] <\DSAFEW\> something wrong with my alsa? [06:04] <\DSAFEW\> the only solution I think might work involves plugging physically a cord from line in to line out [06:05] <\DSAFEW\> but I don't have any extra cables, and the details on that are still fuzzy, after that step I don't know how I'd make line-out my default capture for the stream [06:14] <relaxed> \DSAFEW\: This would be pretty easy if you used pulse. [06:15] <\DSAFEW\> mmmmmkay, thanks relaxed I think I'll install that now. [08:26] <cbreak-work> I have a weird problem with ffmpeg, I have an application that pipes real time raw data into an ffmpeg instance, which only scales it, changes the framerate and then applies a filter. [08:26] <cbreak-work> the result gets written to a named pipe, from which a vlc instance reads. [08:27] <cbreak-work> the problem is that this is somehow slow, the memory usage of the ffmpeg instance continually increases, and the vlc instance gets much less image data than it should [08:27] <cbreak-work> the weird thing is that as soon as I stop the input program from sending new data, and make it close its pipe, the ffmpeg instance pushes out all the stored data and the encoding in vlc runs at full speed [08:28] <cbreak-work> why could that be? [08:28] <cbreak-work> is ffmpeg somehow overwhelmed with reading so it has no time to write? [08:34] <cbreak-work> if I change the output to be a file, or a cat instance instead of vlc, the memory consumption stays constant, so I assume that the writing is in real time [08:49] <desu> is there a way to make ffmpeg automatically use the calculated input bitrate for encoding the output? [08:50] <cbreak-work> that sounds kind of pointless [08:52] <desu> i'm trying to extract the audio stream from a video file [08:52] <desu> using higher bitrate would be waste, using a lower one would reduce quality [08:52] <desu> hence i want to preserve the bitrate [08:53] <cbreak-work> no [08:53] <cbreak-work> preserving bitrate would ALSO reduce quality [08:54] <desu> yeah, but not as much as, say, using 128 for a 196 input bitrate [08:55] <cbreak-work> how about just not reencoding it? [08:57] <cbreak-work> like with -acodec copy -vn [08:57] <desu> i need to do it because it's in a different format than i need.. [08:57] <desu> it's in AAC, i need it in ogg [08:59] <cbreak-work> then just use a custom bitrate [08:59] <cbreak-work> why do you think that quality in aac and ogg vorbis correspond with similar bit rates? :) [09:00] <desu> hmmm, good point.. but i don't want loss of too much information [09:00] <desu> :) [09:38] <cbreak-work> so, no one has an idea why ffmpeg is stalling? [09:57] <zap0> debug it [09:59] <cbreak-work> I attempted that [10:00] <cbreak-work> my current results are: [10:00] <cbreak-work> once my writer (the source for ffmpeg, with pipe:0 input) closes its pipe, ffmpeg writes out the data to vlc fast [10:01] <cbreak-work> if I make ffmpeg write to a file instead of a named pipe, it is also not increasing its memory consumption [10:01] <cbreak-work> if I make cat instead of vlc read from the pipe (and write into /dev/null), it's also fast [10:01] <cbreak-work> so, maybe ffmpeg can't handle having a pipe as input and output at the same time [10:12] <cbreak-work> the weird thing is: on a different computer the same scripts worked [10:13] <cbreak-work> (not the same ffmpeg/vlc though) [15:46] <Cervajz1> Hello guys [15:47] <Cervajz1> Where the options -newaudio and -newvideo disappeared, please? :c) [15:47] <Cervajz1> What should I use instead of it? [15:48] <JEEB> see the documentation on mapping options (-map) [15:49] <Cervajz1> JEEB: I see, it should be -map 0 [16:16] <praedo> hello [16:17] <praedo> i'm using ffmpeg to stream video files to rtmp [16:17] <praedo> but some of them have drop frames and i'm having problems [16:17] <praedo> while streaming i see fps passing by but video doesn't appear on web (flowplayer) [16:18] <praedo> other videos encoded exactly the same way, they appear correctly [16:18] <praedo> the only idea is that those bad videos may have at least one drop frame unnoticeable but ffmpeg may detect it and stop working correctly [16:19] <praedo> or maybe it's rtmp server being too sensitive [16:19] <praedo> or myabe flowplayer [16:19] <praedo> how could i debug this problem? [16:19] <praedo> or better: is there any way to tell ffmpeg to skip drop frames and continue? [16:22] <praedo> i use this option: [16:22] <praedo> -re Read input at native frame rate. Mainly used to simulate a grab device. [16:23] <praedo> files read are mp4 [16:23] <praedo> any idea why some files are streamed correctly and some not? [16:23] <praedo> being encoded the same way [17:32] <praedo> [15:17] <praedo> i'm using ffmpeg to stream video files to rtmp [17:32] <praedo> [15:17] <praedo> but some of them have drop frames and i'm having problems [17:32] <praedo> [15:17] <praedo> while streaming i see fps passing by but video doesn't appear on web (flowplayer) [17:32] <praedo> [15:18] <praedo> other videos encoded exactly the same way, they appear correctly [17:32] <praedo> [15:18] <praedo> the only idea is that those bad videos may have at least one drop frame unnoticeable but ffmpeg may detect it and stop working correctly [17:32] <praedo> any suggestion please? [17:42] <Sorcey> hi! Does anyone know if there could be any reason ffmpeg stops ingesting an axis rtsp stream after 90 seconds but ffplay doesn't, when it's tunneled over a http proxy server? (-rtsp_transport http) [17:43] <Sorcey> Is it possible this problem is re-introduced? [17:43] <Sorcey> http://libav-users.943685.n4.nabble.com/Re-IP-camera-amp-receiver-report-pr… [17:43] <Sorcey> Weird that the problem only occurs in ffmpeg [17:50] <Sorcey> wrong link, meant this one: http://ffmpeg.org/trac/ffmpeg/ticket/129 [19:51] <aster1sk> What's wrong with this : http://pastebin.com/eQwLWGyy [19:53] <relaxed> aster1sk: add -f flv [19:54] <aster1sk> Great works, ty relaxed -- I'm curious, can you point me to why ? [19:57] <relaxed> ffmpeg needs to know which container to use for audio/video and can't deduce it from "-", and flv is a streamable container [19:58] <aster1sk> Thank you, I believe instead of net cat I should be using ffserver& Just found it - duhh. [20:33] <indigo> Hi everyone [20:33] <indigo> I need an info [20:33] <indigo> I'm trying to compile (again) ffmpeg with some libraries (dirac, lame etc) [20:34] <indigo> and all is going well [20:35] <indigo> except that the executable (ffmpeg.exe) requires an external bz2-1.dll [20:35] <indigo> anyone here knoing WHY?!? [20:38] <iive> matroska demuxer seems to use it. [20:38] <indigo> I'm not enabling matroska [20:39] <indigo> I'll try to explain [20:39] <indigo> I set a benchmark just to make a proof [20:39] <indigo> first I compiled ffmpeg without ANY lib [20:39] <indigo> then I try ONLY with dirac [20:40] <indigo> (I only added --enable-dirac) [20:41] <indigo> in this second case I found that bzlib2 in selected but it doesn't seem to me that dirac uses bz2 [20:44] <indigo> now I've just tried again and on a windows machine a message appears! [20:44] <indigo> libdirac-decoder0.dll NOT FOUND [20:49] <praedo> [15:16] <praedo> hello [20:49] <praedo> [15:17] <praedo> i'm using ffmpeg to stream video files to rtmp [20:49] <praedo> [15:17] <praedo> but some of them have drop frames and i'm having problems [20:49] <praedo> [15:17] <praedo> while streaming i see fps passing by but video doesn't appear on web (flowplayer) [20:49] <praedo> [15:18] <praedo> other videos encoded exactly the same way, they appear correctly [20:49] <praedo> [15:18] <praedo> the only idea is that those bad videos may have at least one drop frame unnoticeable but ffmpeg may detect it and stop working correctly [20:49] <praedo> any suggestion please? [21:03] <indigo> I'm only try to get HOW to make a huge monolithic ffmpeg executable without any external dependence [21:13] <Mavrik> static linking of everything ftw :) [21:17] <Sorcey> praedo: using red5 or wowza? [21:20] <praedo> wowza [21:22] <indigo> maverik, yes, exactly [21:22] <Sorcey> so you stream ondemand content using -acodec copy -vcodec copy? [21:23] <praedo> exactly [21:23] <praedo> and using -re [21:23] <Sorcey> you should check the exact contents using mediainfo [21:24] <Mavrik> I came slightly late, praedo what are you trying to do? [21:31] <praedo> this is the command exactly: [21:32] <praedo> ffmpeg -i file -re -acodec copy -vcodec copy -f flv rtmp://wowzaserver [21:32] <praedo> some of the mp4 files play and some don't [21:33] <praedo> and they are all encoded in the same way [21:33] <Mavrik> hmm [21:33] <Mavrik> I've rather used MPEG2-TS to stream to wowza server, I've found RTMP support to be unreliable [21:34] <praedo> but i need rtmp because i play them using flowplayer rtmp plugin [21:34] <Mavrik> praedo, that doesn't matter, since wowza remuxes it anyway [21:34] <Mavrik> you stream MPEG2-TS to wowza and wowza remuxes it to whatever you connect to it with [21:35] <Mavrik> ffmpeg -i file -re -acodec copy -vcodec copy -f mpegts -bsf h264_mp4toannexb udp://<wowzaserver>:<port> [21:35] <Mavrik> check wowza docs for configuration, you just need a stream file somewhere to set the listening port [21:35] <praedo> and flash player will connect to it using rtmp too? [21:35] <Mavrik> yeah [21:36] <praedo> so the rtmp plugin is still useful if source comes from ts [21:36] <Mavrik> the container you use to stream TO wowza doesn't matter (as long as Wowza supports it) since wowza has to remux it for each output format anyway [21:36] <Mavrik> that's how it can provide RTMP, RTSP, HTTP and other output supports for same file [21:37] <praedo> so -f mpegts is better than -f flv? [21:37] <praedo> better quality [21:38] <Mavrik> it doesn't affect quality since it's a container [21:38] <Mavrik> maybe you should see basics of digital video before you continue [21:38] <praedo> Unrecognized option 'bsf' [21:38] <praedo> ffmpeg: failed to set value 'h264_mp4toannexb' for option 'bsf' [21:40] <Sorcey> Mavrik: rtmp was pretty stable in my tests (even over longer periods of time) [21:40] <Sorcey> I don't think this is praedo's problem tho [21:40] <Sorcey> praedo: did you check the wowza error log? [21:40] <Sorcey> by default it's pretty verbose [21:40] <Mavrik> praedo, -vbfs h264_mp4toannexb [21:41] <Mavrik> Sorcey, I agree [21:41] <Sorcey> and pretty useful too [21:41] <Mavrik> Sorcey, I had some negative experiences in old ffmpeg versions and I've used MPEG2-TS for extensive streaming so I trust it more [21:41] <Sorcey> so how do you deal with security when using mpeg2-ts input? [21:41] <Mavrik> Sorcey, IP-level firewall [21:42] <praedo> for files that aren't played, no error appears in ffmpeg... it just continues sending frames [21:42] <Mavrik> Sorcey, that is a problem though [21:42] <Sorcey> praedo: so did you check the wowza log? [21:42] <Mavrik> but that's not our primary problem so using time to swtich to something more secure isn't that much of a priority [21:43] <Sorcey> Mavrik: ok, dunno what you use it for now [21:43] <praedo> i check wowza log now.. [21:44] <Sorcey> the error log, not the access log [21:45] <Sorcey> hm, I think we found a bug in ffmpeg causing broken pipes when using http-tunneled rtsp input :P [21:51] <praedo> i checked the error log and nothing new appears when i start streaming using -vbfs [21:51] <praedo> vbsf [21:51] <praedo> and one file that could be played with rtmp, now they don't pplay with vbsf [21:51] <relaxed> what is your command? [21:52] <Sorcey> no, what happens if you stream it 'normally'? [21:52] <Sorcey> what does wowza say then? [21:52] <Sorcey> (in its error log) [21:52] <praedo> ffmpeg -i file -re -acodec copy -vcodec copy -f mpegts -vbsf h264_mp4toannexb udp://wowzaserver [21:52] <praedo> working files aren't working with this [21:52] <praedo> i check logs with non-working files with rtmp now... [21:53] <Sorcey> of course not, cause wowza isn't listening on udp if you don't config it :P [21:53] <relaxed> praedo: "file" is a mp4 container? [21:55] <praedo> yes [21:55] <praedo> no error log output in wowza using rtmp [21:56] <Sorcey> praedo: how did you encode those files? in what format? [21:57] <praedo> both working and non-working are encoded in the same way [21:57] <Sorcey> with what application/settings? [21:58] <praedo> ffmpeg -i file.avi -deinterlace -vcodec libx264 -preset medium -crf 24 -threads 0 -acodec libfaac -aq 100 file.mp4 [22:00] <Sorcey> what's the files' approximate duration? [22:00] <Sorcey> you might want to constrain the bitrate [22:00] <Sorcey> but that shouldn't be the problem, and if it were wowza would complain about it in its error log [22:09] <praedo> how can i check mediainfo? [22:10] <relaxed> drop -deinterlace and start using -vf yadif [22:13] <Sorcey> http://mediainfo.sourceforge.net/en/Download praedo [22:15] <Sorcey> you could also try and upload the .mp4 to a normal webserver and see if flowplayer can play it back from there [22:15] <praedo> i'll try... [22:15] <Sorcey> (see if the problem is with flowplayer+file or with flowplayer+wowza+file [22:15] <Sorcey> these kinds of problems always take up a lot of time, good luck! [23:38] <praedo> Sorcey, still here? [23:38] <indigo> I tried again compiling ffmpeg with --enable-gpl and --enable-version3 and keep getting "zlib1.dll not found" [23:38] <indigo> maybe some gpl feature requires zlib [23:38] <praedo> i finally got an error in wowza logs after streaming a mp4 file with rtmp [23:53] <praedo> INFO server comment 2012-02-01 23:35:23 - - - - - 2427.688 - - - - - - - client connectionClosed [1232396645] pingtimeout [23:53] <praedo> this is the error that appears in wowza logs when i stream an mp4 file using rtmp [23:53] <praedo> Sorcey, any clue? [23:54] <Freakshow> ugh... is there any magic that I need to perform in order to output to rtsp? [23:58] <Freakshow> ffmpeg -i $SOURCE -b:v 500k -bt 500k -r 30 -vcodec libx264 -b:a 64k -acodec libfaac -ar 44100 -s 720x480 -x264opts keyint=30:min-keyint=15 -f rtsp rtsp://server ip:1935/path/to/stream.sdp [00:00] --- Thu Feb 2 2012

1 0

ffmpeg-devel.log.20120131
by burek 01 Feb '12

01 Feb '12

[00:04] <CIA-108> ffmpeg: 03Carl Eugen Hoyos 07master * rea604de91e 10ffmpeg/libavcodec/pngdec.c: Simplify "deloco" and support decoding of 48bit loco png. [00:14] <CIA-108> ffmpeg: 03Carl Eugen Hoyos 07master * r97da38c99b 10ffmpeg/libavcodec/pngdec.c: [00:14] <CIA-108> ffmpeg: Allow decoding of 64bit png images. [00:14] <CIA-108> ffmpeg: Fixes a part of ticket #639. [00:23] <CIA-108> ffmpeg: 03Michael Niedermayer 07master * rb8c1655882 10ffmpeg/libavformat/avidec.c: [00:23] <CIA-108> ffmpeg: avidec: print informative error messages if seeking fails. [00:23] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [00:46] <ubitux> Daemon404: what's wrong with lpostproc? [00:46] <ubitux> why libav ppl want to get rid of it? [00:46] <Daemon404> it's deprecated and not maintained, and noone wants to touch its code [00:46] <kierank> unmaintainable [00:47] <ubitux> isn't just a 1k line .c? [00:47] <Daemon404> go look at its code [00:47] <Daemon404> :) [00:47] <kierank> 3500 lines of inline asm [00:47] <Daemon404> and macros [00:47] <bcoudurier> let's remove vsfilter [00:48] <Daemon404> ^ i wish [00:48] <ubitux> :/ [00:48] <bcoudurier> let's remove avisynth [00:48] <Daemon404> libass forever! [00:48] <Daemon404> avisynth is idneed balls but it has nothing to replac it [00:48] <ubitux> is there a replacement for postproc? [00:48] <Plorkyeran> vsfilter > fontconfig~ [00:49] <bcoudurier> what's replacing postproc ? [00:49] <Plorkyeran> just not using it? [00:49] <Plorkyeran> it's not very useful in the first place [00:49] <bcoudurier> :) [00:49] <ubitux> i'm not anyone is very well suited to decide of the needs of ppl like that [00:49] <ubitux> +sure [00:50] <Daemon404> ubitux, how is moving it to its own repo damaging btw [00:50] <Daemon404> it;s literally no different [00:50] <ubitux> the maintainer of postproc is the maintainer of the ffmpeg project afaik [00:50] <Daemon404> [18:48] < Plorkyeran> vsfilter > fontconfig~ <-- i have hopes! [00:50] <ubitux> why moving it outside? [00:51] <Daemon404> ubitux, this is libav [00:51] <ubitux> are you planing to nuke every code of the ffmpeg maintainer then? :) [00:51] <Daemon404> no [00:51] <ubitux> why? [00:51] <ubitux> it's not "maintained" anymore [00:52] <Daemon404> but it's pretty much consensus that postproc should die, among more than just libav developers [00:52] <Daemon404> i guess not here [00:52] <ubitux> yes, i understood that, but couldn't get the reasons [00:52] <bcoudurier> I like that, names please ? [00:52] <Daemon404> say hi to ffms2 devs [00:52] <Daemon404> ubitux, because it is not very useful and its code is unmaintainable [00:52] <Daemon404> i dont actually know ANYTHING using it [00:52] <ubitux> mplayer? [00:53] <bcoudurier> vlc [00:53] <ubitux> :) [00:53] <Daemon404> oh really? [00:53] <kierank> yeah [00:53] <ubitux> yes& [00:53] <Daemon404> anyway, we could bikeshed this all day and you can flame me all you want [00:53] <ubitux> not very useful you don't know, unmaintainable: just improve it; is it broken? [00:53] <Daemon404> but nothing will change [00:54] <ubitux> it's not bikeshed, and we're not flaming you [00:54] <bcoudurier> there is a recrudescence of nih in some part of the sphere [00:54] <Daemon404> i just did it cause others wanted it and it was an interesting script to write to import teh git history [00:54] <durandal_1707> the problem is it is not used by ffmpeg/libav at all [00:54] <ubitux> the reasons are just obscure [00:54] <Daemon404> so what is the problem with having libpostproc be external? [00:54] <ubitux> libav* are a/v libs [00:54] <ubitux> postproc is an a/v lib [00:55] <ubitux> no? [00:55] <Daemon404> thats some pretty bad reasoning... [00:55] <ubitux> :/ [00:55] <ubitux> ffmpeg/libav* are supposed to provide API for processing a/v [00:55] <durandal_1707> ubitux: make ffplay use libpostproc and I'm with you [00:55] <ubitux> why doesn't it belong there? [00:55] <Daemon404> so let me ask [00:55] <ubitux> durandal_1707: ffmpeg doesn't only provides programs, but also libs [00:55] <Daemon404> given we have libavfilter [00:56] <Daemon404> what is the poitn of postproc? [00:56] <ubitux> why not move it to lavfi ? [00:56] <ubitux> instead of *removing* features? :/ [00:56] <TheFluff> I thought you lunix people were supposed to be against bloat [00:56] <Daemon404> because i, liek everyone else am not fuckign touchign the wall of brainfuck that is postproc [00:56] <ubitux> it just looks like another try of purification without caring about users [00:56] <Daemon404> and porting it [00:56] <TheFluff> oh wait hahahahahaha gnome and kde [00:57] <Daemon404> for teh nth time [00:57] <Daemon404> NOTHING IS BEIGN REMOVED [00:57] <Daemon404> it is being MOVED [00:57] <ubitux> it's the same [00:57] <Daemon404> no it isn;t. [00:57] <ubitux> since no one will ever maintain the "working state" [00:57] <TheFluff> wait why would it be the same [00:57] <ubitux> it's just a "let it die" [00:58] <Daemon404> [18:57] <@ubitux> since no one will ever maintain the "working state" [00:58] <Daemon404> no [00:58] <TheFluff> nobody maintains it now either so I don't see your point? [00:58] <Daemon404> thats a fuckign assumption [00:58] <Daemon404> you make [00:58] <bcoudurier> TheFluff, scratch vsfilter ;) [00:58] <Daemon404> [18:58] < TheFluff> nobody maintains it now either so I don't see your point? <-- tell you what, ill backport any changes that get made to postproc [00:58] <ubitux> TheFluff: it's maintain at least to "keep builtin" [00:58] <Daemon404> all one every 2 years [00:58] <TheFluff> there are actually several maintained forks of vsfilters that add new functionality and fix old bugs [00:58] <TheFluff> just sayin' [00:59] <ubitux> Daemon404: you should have move it to lavfi instead of moving it outside the project [00:59] <bcoudurier> wow, so it is maintainable after all ? [00:59] <durandal_1707> what is vsfilter? [00:59] <TheFluff> ubitux: how is it builtin [00:59] <TheFluff> durandal_1707: a subtitle renderer [00:59] <Daemon404> ubitux, [18:56] <+Daemon404> because i, liek everyone else am not fuckign touchign the wall of brainfuck that is postproc [00:59] <Daemon404> feel free too [00:59] <ubitux> TheFluff: i meant "building" sorry [00:59] <Daemon404> but im nto touching a giant mess of inline asm and macros. [00:59] <TheFluff> it's not integrated with any other lav* libraries at all except for the fact that you can pass it a table of quantizers directly from a lavc codeccontext [01:00] <TheFluff> and god help you if you ever change the structure of that particular table [01:01] <bcoudurier> Daemon404, thankfully people didn't remove libswscale because of that reason [01:01] <TheFluff> in fact I do not see any arguments whatsoever in favor of keeping it in the libav repository other than historical reasons, because really it has absolutely nothing to do with the rest of the libav stuff except using its version defines [01:02] <Daemon404> bcoudurier, swscale actually does important and useful things [01:02] <Daemon404> that cant be replaced [01:02] <ubitux> TheFluff: we could port the vf pp from mplayer into lavfi pretty "easily" [01:02] <ubitux> that would create a dependency [01:02] <ubitux> would that be enough for you as a reason to keep it? :) [01:03] <Daemon404> ubitux, by that logic you should keep libass in the same repo because vf_ass needs it. [01:03] <TheFluff> hm yes a circular argument [01:03] <TheFluff> I am very convinced [01:03] <ubitux> the author isn't a ffmpeg developers [01:03] <ubitux> it's an independent project [01:04] <Daemon404> i fail to see how that matters [01:04] <TheFluff> does that mean everything fabrice bellard ever writes should be in ffmpeg too [01:04] <Daemon404> and nobody "develops" postproc [01:04] <ubitux> what's the point if it works? [01:04] <TheFluff> seriously where do you keep coming up with these arguments [01:05] <bcoudurier> at least fork it from ffmpeg [01:05] <bcoudurier> Daemon404, [01:05] <Daemon404> ? [01:05] <bcoudurier> nobody from libav did anything to it [01:06] <Daemon404> the code is the same in both repos [01:06] <ubitux> TheFluff: libass is somehow related to image processing afaik& i wouldn't mind having it distributed with ffmpeg if the author did it in ffmpeg tree at first, that was my point [01:06] <bcoudurier> headers and stuff [01:06] <Daemon404> i merely chose to to import the history from for its files [01:06] <Daemon404> oh the headers [01:06] <bcoudurier> seems only fair to me [01:06] <TheFluff> libpostproc has exactly one "killer feature" that isn't easily available elsewhere and that's the quantizer-based deblocking, just move that to libavfilter if you are so incredibly sensitive to "losing features" [01:07] <ubitux> i don't care very much how much the libav is being purified of features [01:07] <Daemon404> bcoudurier, id rather jusr s/Libav/FFmpeg/\Libav/g; [01:07] <ubitux> i was just quite surprised [01:07] <bcoudurier> from the great evil [01:07] <durandal_1707> dump libpostproc, that obfsucated code and rewrite new one in lavfi from scratch [01:07] <Daemon404> and keep everyone less butthurt [01:08] <TheFluff> (keep in mind though that said feature works in the killer way with maybe two or three of the lavc codecs; with the rest it's just a very dumb deblocking filter) [01:08] <durandal_1707> just forget about [01:08] <durandal_1707> it [01:08] <TheFluff> it's also really buggy and completely undocumented [01:08] <ubitux> durandal_1707: rewriting "from scratch" is rarely a good idea [01:08] <Daemon404> ubitux, for postproc it is. [01:09] <Compn> i must say, your reasoning for moving postproc is fine :P [01:09] Action: Compn not flaming, just strange logic going on in here [01:09] <Compn> and i have no opinion one way or the other. aside from not making it more difficult for mplayer to use the library... which moving it out of repo would do [01:09] <Daemon404> the whole idea is that is isnt really maintained, nothing in libav or ffmpeg uses is, and it may be useful outside of libav/ffmpeg [01:09] <Compn> if mplayer does use it [01:09] <Daemon404> so make it standalone. [01:10] <ubitux> it may be useful inside too& [01:10] <Compn> yeah but mplayer and vlc wont maintain it either, so making it standalone wont make it maintained [01:10] <ubitux> having it in the repo reminds us that "hey there is something to do there" :) [01:10] <Compn> what does making it standalone do to improve it ? [01:10] <Compn> just curious [01:10] <ubitux> the point is to purify the holy project Compn ;) [01:10] <Daemon404> it doesnt "lose" anything by moving [01:11] <Compn> PURITY OF ESSENCE [01:11] <Compn> ubitux : tap water. [01:11] <durandal_1707> whatever you do with it, it is going to be really slow death [01:11] <Compn> should just start a rewrite [01:11] <Daemon404> durandal_1707, i figured people would get butthurt when i sent that emai :/ [01:11] <TheFluff> I dunno about you guys but I don't really like having broken undocumented shit sitting around in my libraries that other people are expected to use [01:11] <durandal_1707> except if you write better one [01:11] <TheFluff> if you do it, do it right [01:11] <Compn> TheFluff : send patch.... [01:12] <Compn> ehhe [01:12] Action: Compn wins one for the cliche [01:12] <Daemon404> [19:11] < Compn> should just start a rewrite <-- implying anyone can actually comprehend postprocs code.... [01:12] <durandal_1707> that it will die very fast [01:12] <Compn> Daemon404 : i meant a rewrite of ffmpeg [01:12] <Daemon404> wat? [01:12] <ubitux> TheFluff: asking the users would have been wise instead of just avoid the problem [01:12] <TheFluff> I am a postproc user [01:12] <Compn> if you are going to whittle it down, one directory, one codec, one format at a time [01:12] <Daemon404> ubitux, TheFluff is ffms2 developer [01:12] <TheFluff> or well, "user", my library uses postproc and my library's users use it through mine [01:13] Action: Compn should probably stay out of this, since he knows nothing about the topic [01:13] <TheFluff> I think it's a terrible crock of shit and people ask me questions about why it doesn't work all the time [01:13] <ubitux> then why didn't you work on exporting the API properly and make a simple lavfi filter? [01:13] <TheFluff> that I can't answer because you don't maintain it and it's not documented [01:13] <Daemon404> and you cant even follow the code [01:13] <ubitux> where will you do bug reports if it's not in the project anymore? [01:13] <Daemon404> cause its a buttlaod of inline asm, [01:14] <TheFluff> nowhere, I'm going to deprecate it [01:14] <Daemon404> ubitux, same bugtracker different heading/topic? [01:14] <ubitux> with what replacement TheFluff? [01:14] <TheFluff> actually I already did deprecate it with the excuse "libav is going to remove it" [01:14] <Daemon404> it's merely a different repo [01:14] <TheFluff> none [01:14] <Daemon404> nobody misses it. [01:14] <TheFluff> as I said it's not very useful [01:14] <TheFluff> I accidentally posted a build with it disabled a while ago [01:14] <TheFluff> it took weeks before someone even mentioned it wasn't there [01:14] <ubitux> how come there is so much code if it isn't useful? [01:15] <Daemon404> that is a good question. [01:15] <TheFluff> are you seriously implying [01:15] <TheFluff> that all code is useful [01:15] <Compn> libass in ffmpeg would be useful. +1 [01:15] <TheFluff> inherently useful, even [01:15] <ubitux> Compn: :) [01:15] <Compn> but probably too many bloats [01:15] <Daemon404> Compn, remember how ffmpeg was talking about its own subttile filter? [01:15] <Daemon404> <_< [01:15] <bcoudurier> TheFluff, there is shitload of useless code here [01:16] <ubitux> Daemon404: we are still talking about it, and i'm working on it [01:16] <bcoudurier> that people rarely used [01:16] <durandal_1707> bcoudurier: what is shit? [01:16] <Daemon404> ubitux, is it libass based at least? [01:16] <Compn> Daemon404 : i remember a lot of talk and no coding. [01:16] <TheFluff> also, people really liked their quantizer-based deblocking back in 2003 when you sorta had to watch divx 3 movies all the time [01:16] <TheFluff> it was kinda useful in mplayer then, I guess [01:16] <ubitux> Daemon404: will be for overlay yes, just like it is now [01:16] <Daemon404> ok [01:16] <Daemon404> it would be foolish not to [01:16] <TheFluff> incidentally the last major functional changes in postproc were probably made sometime around that time [01:17] <ubitux> Daemon404: there is something problematic with this [01:17] <Compn> can i ask one more dumb question. which developer new to the project has looked at libpostproc and tried to work with it , without knowing it was depreciated ? [01:17] <ubitux> Daemon404: you can't really have a hard dep on it since it's an external project [01:17] <Compn> or 'deprecated' as some people would have you say [01:17] <ubitux> we need to just make the overlay depends on it [01:17] <Compn> freakin grammar nazis around here [01:17] <durandal_1707> Compn: i never entered that directory -> that says it all [01:17] <Daemon404> ubitux, it seems like youre redoign things libass has already accomplished [01:17] <Daemon404> for naught [01:17] <ubitux> ? [01:17] <ubitux> what am i redoing? [01:18] <bcoudurier> Compn, the nazis have left ;) [01:18] <Compn> bcoudurier : yes, i was in the other channel for a bit :P [01:18] <Daemon404> if you dont use libass, i mean, how do you expect to use it's functionality without redoing soem of its code? [01:18] <Daemon404> (i.e. dont make it a hard dep) [01:18] <Compn> ubitux : Daemon404 is wondering if it would be better if you just made a vf_ass and didnt work on subtitles, passed it all to libass, etc [01:18] <Daemon404> font handling etc [01:18] <ubitux> Daemon404: i was thinking of having it in the repo ;) [01:18] <Daemon404> is that a joke? [01:18] <ubitux> Daemon404: but of course that won't happen :) [01:19] <ubitux> Daemon404: the point is, it wouldn't have been an issue if libass was part of ffmpeg project [01:19] <Daemon404> i dont see a reason nto to require it for building the subtitle filter [01:19] <Compn> does ffmpeg have any external libs ? [01:19] <Daemon404> ubitux, because NIH? [01:19] <Daemon404> and external deps are satan? [01:19] <durandal_1707> Compn: libc [01:19] <TheFluff> what is the problem with having the subtitle filter being an optional component that depends on an external library [01:19] <ubitux> Daemon404: because it's harder to deal with the dep [01:19] <Daemon404> this soudns fucking dumb. [01:19] <TheFluff> I mean it's not like you don't have other such optional components [01:19] <Daemon404> what TheFluff said. [01:19] <bcoudurier> libass need not be in the repo [01:19] <TheFluff> oh wait, it was Not Invented Here, clearly you cannot use it [01:19] <bcoudurier> although libass dependencies are hell [01:20] <ubitux> as far as libass is limited to ass overlay it's fine [01:20] <bcoudurier> and I mean _hell_ [01:20] <TheFluff> yes [01:20] <TheFluff> yes they are [01:20] <Daemon404> you should render most things with libass really [01:20] <TheFluff> but that's Somebody Else's Problem [01:20] <Daemon404> s/most/all/ [01:20] <bcoudurier> mostly harfbuzz to be fair though [01:20] <ubitux> that's the intention Daemon404 [01:20] <Daemon404> harfbuzz fork you mean, bcoudurier [01:20] <Daemon404> ;) [01:20] <Compn> when did anime pirates dictate which subtitle renderers we must use ? [01:20] <bcoudurier> the fork has issues [01:20] <Compn> thats what i want to know. [01:20] <bcoudurier> I told greg already [01:20] <ubitux> Compn: :D [01:21] <Daemon404> Compn, since always [01:21] <durandal_1707> everything is going 3D lets deprecate ffmpeg/libav [01:21] <bcoudurier> ligatures are fucked in some situations [01:21] <Plorkyeran> shockingly enough people who do things with subtitles are the ones that care about subtitle renderers [01:21] <Compn> there was talk about switching to libass in mplayer [01:21] <bcoudurier> well x264 is the future anyway ;) [01:21] <Compn> but libass was so much slower than mplayer's optimized sub renderer [01:22] <Plorkyeran> uh [01:22] <Plorkyeran> what [01:22] <Compn> just different things for different people [01:22] <TheFluff> 00:47:41 <+kierank> 3500 lines of inline asm <-- wow, you sure weren't kidding, I just took a look at postprocess_template.c [01:22] <durandal_1707> bcoudurier: hevc [01:22] <Daemon404> bcoudurier, guess where lorren came from? :P anime pirate. [01:22] <Plorkyeran> libass is mplayer's sub render [01:22] <ubitux> isn't the internal osd just dealing with gray levels Compn? [01:22] <ubitux> Plorkyeran: not the default [01:22] <Plorkyeran> it wasn't even spun off into its own repo until a few years ago [01:22] <Compn> ubitux : its color in -vo gl, using gl extras [01:22] <Compn> but yes, otherwise gray [01:23] <Plorkyeran> yes, it also has its hilariously broken one that isn't actually usable [01:23] <Plorkyeran> that mostly just serves to confuse users [01:23] <ubitux> -noass isn't the default for various reasons [01:23] <Compn> mplayer likes to confuse, mplayer doesnt want dumb users ;P [01:23] <Compn> send them to vlc... :P [01:23] <bcoudurier> poeple reimplemting libass are stupid [01:23] <durandal_1707> mplayer uses own version of libass [01:23] <Compn> durandal_1707 : no, its synced to svn now [01:24] <Compn> manually, anyways [01:24] <TheFluff> vlc uses libass too so idgi [01:24] <Compn> its cool that libass went on to its own project. means more players can use it [01:25] <Daemon404> Compn, i hope it gets its windows backend working well soon enough [01:25] <Daemon404> so a dshow filter can be written, and vsfilter brough out behind the barn and shot [01:25] <Compn> Daemon404 : isnt that in ffdshow... [01:25] <Daemon404> no? [01:25] <Daemon404> and ffdshow is on its way out [01:25] <Daemon404> its code is so bad everyone is ditching it [01:25] <Compn> whats it being replaced with ? [01:25] <Daemon404> for lav filters [01:26] <Compn> heh [01:26] <Compn> ffdshow is a bunch of different things at once [01:26] <Compn> its lavf + mp vf system + lavc [01:26] <ubitux> erh with all the flames i started i didn't had time to finish my lavfi/libswr work :( [01:26] <ubitux> 'night ppl :) [01:26] <Daemon404> night. [01:26] <Compn> ubitux : dont flame, only code , gnight [01:27] <Compn> so what else can we argue about [01:27] <Compn> what tasks are we coming up with in GSOC this year ? [01:28] <ubitux> oh, forgot something [01:28] <Compn> the great libav conspiracy > spend hours fixing every grammar , spelling and cosmetic problem in the code. if come across a file without any of those (all inline asm) just rm it . [01:29] Action: durandal_1707 kicks Compn [01:29] <Compn> lol [01:29] <ubitux> Daemon404: sorry, that will make you certainly angry but& why no leave the libpostproc maintainance to ffmpeg (and ask user to build postproc from independant ffmpeg repo?) [01:30] <Compn> ubitux : because they want to move it to a diff repo [01:30] <ubitux> it is still available though ffmpeg [01:30] <Daemon404> not my call (tm) [01:30] <TheFluff> postprocess_template.c is hilarious, there's inline asm, "templating" of function using the c preprocessor, inline asm "functions" generated dynamically with the preprocessor, bitwise operation tango with magic numbers picked seemingly at random, reference in the comments to optimizing for the P4, and probably several hundred lines of commented-out debug code [01:30] <Compn> Daemon404 : anyways, did you ask michaelni ? [01:30] <ubitux> Compn: it is a different repo [01:30] <Daemon404> Compn, i did this at teh request of libav. [01:30] <ubitux> ffmpeg is a different repo as libav afaik [01:30] <Compn> Daemon404 : yeah, but you better ask michael, since hes maintainer iirc [01:30] <Compn> its just me and ubitux complaining in here, we arent maintainers :p [01:31] <ubitux> i'm not complaining [01:31] <ubitux> trying to understand, somehow [01:31] <Compn> discussing [01:31] <Daemon404> Compn, he is not listed as having touched it since 2008... [01:31] <Compn> michaelni : Daemon404 wants to know if libav can move postproc to another repo [01:31] <ubitux> i see little point in letting libpostproc rotting in an abandonned repository since it's still available in ffmpeg [01:31] <Daemon404> Compn, no [01:31] <Daemon404> it's getting moved regardless [01:32] <Compn> yes we know that [01:32] <TheFluff> it's rotting in ffmpeg too so what's the difference [01:32] <Daemon404> ^ [01:32] <ubitux> TheFluff: less work? [01:32] <Daemon404> wat? [01:32] <Compn> ubitux : stop biting and go code/sleep :P [01:32] <ubitux> TheFluff: i mean, just drop it and redirect libpostproc/ffmpeg repo [01:32] <durandal_1707> faster git co? [01:32] <Daemon404> it's not like this affects ffmpeg at all [01:32] <Daemon404> so i dont see why you give a crap? [01:32] <ubitux> TheFluff: what is a point of the burden of the export? [01:33] <TheFluff> having a broken feature is worse than not having it at all [01:33] <ubitux> don't be so aggressive Daemon404, i'm just trying to understand why you are trying to do that [01:33] <Daemon404> ubitux, look for the "nuke libpostproc" thread on libav-devel [01:33] <Daemon404> by adam [01:33] <Daemon404> perhaps you will find your answer there [01:33] <ubitux> i read it [01:33] <ubitux> i'm on the mailing list [01:34] <Compn> ubitux : are you looking for the reason why they want to move it ? [01:34] <Compn> the reason why it must be moved ? [01:34] <ubitux> i just see little point in the standalone libpostproc if you don't feel like maintaining it [01:34] <Daemon404> [19:32] < durandal_1707> faster git co? <-- no not really at all [01:34] <Daemon404> not the way git works. [01:34] <ubitux> and you KNOW no one except the ffmpeg maintainer will ever maintain it [01:34] <Daemon404> ubitux, i said i WOULD maintain it [01:34] <Daemon404> with any fixes needed [01:34] <ubitux> so why do you care about the export? [01:34] <Daemon404> jeez [01:34] <ubitux> Daemon404: why would you do that? [01:35] <Daemon404> it will get the exact amount of "maintaining" that it gets in ffmpeg/libav [01:35] <Daemon404> period. [01:35] <ubitux> what's the point of keeping it in sync with ffmpeg? [01:35] <ubitux> if you don't plan to add a plu-value [01:35] <Compn> ubitux is wondering why not just rm it [01:35] <Compn> ? [01:35] <ubitux> Compn: yes [01:35] <Compn> if no one uses it [01:35] <TheFluff> I can't even find the last time someone made an actual functional change to libpostproc [01:35] <ubitux> the point is, there is still a copy available [01:35] <Compn> let mplayer/vlc fork it [01:35] <ubitux> even if libav drop it [01:35] <Daemon404> Compn, hey i wanted to rm -rf it [01:35] <TheFluff> it was before 2006 at least [01:36] <TheFluff> i.e. 6 (six) years ago [01:36] <Compn> its an old bikeshed, i'm sure postproc has come up in the list before [01:36] <Daemon404> ANYWAY, i did what people wanted, and other peopel jumped down my throat [01:36] <Daemon404> so /care [01:36] <Daemon404> crew y'all [01:36] <Daemon404> im going to get drunk [01:36] <Daemon404> screw* [01:36] <ubitux> :/ [01:36] <TheFluff> all of the "maintenance" in the last six years has basically consisted of keeping up with changes in various libavutil macros [01:36] <Compn> i'm not jumping [01:36] <Compn> just like to talk [01:37] <ubitux> i wonder if my question was even understood [01:37] <Compn> TheFluff : sorry, i'm talking about the generic 'anything with no maintainer is maintained by michael' rule. not that that file is maintained or not, he is 'maintainer' of it [01:37] <ubitux> well whatever, seems it's hot enough in there [01:38] <TheFluff> I don't understand why you are talking about the importance of maintaining libpostproc in the first place [01:38] <Compn> i'm not [01:38] <TheFluff> you act like it is being maintained now [01:38] <Compn> i just said ask michael [01:38] <Compn> what he wants to do with it [01:38] <TheFluff> when in fact it has not been for over six years [01:39] <TheFluff> I think if he wanted to do something he has more than enough time to do it now [01:39] <Compn> lol [01:39] <Compn> then dont ask michael [01:39] <ubitux> :) [01:39] <Compn> you are being contradictory [01:39] <TheFluff> judging by the comments in the postproc source files the last time someone actually fixed or changed something in it was sometime in 2004 [01:39] <TheFluff> contradictory how [01:40] <Compn> someone comes here and asks if its ok to move, but its being moved already, so whats the point in asking, and you say its not important to ask? [01:40] <TheFluff> 01:39:06 < TheFluff> I think if he wanted to do something he has more than enough time to do it now <--- s/he has more/he has had more/ [01:40] <Compn> so i'm confuseeeeeeeeeeeeed [01:40] <TheFluff> I thought the entire debate was about how it was not going to be moved because you didn't want to lose features/make it unmaintained/whatever I don't even know [01:41] <Compn> i thought the debate was 1. does anyone use postproc and 2. does michael care about moving it [01:41] <TheFluff> that was mostly ubitux saying that and not you though [01:41] <Compn> heeh [01:42] <Compn> ubitux 's question was , whats the point in moving it [01:42] <ubitux> TheFluff: that was my first point, but i guess i have an answer for this [01:42] <ubitux> the second point is still unanswered though [01:42] <TheFluff> what was the second poin [01:42] <TheFluff> t [01:42] <ubitux> the second point is: why do you care about moving it outside libav if it is still available for third-party usage in ffmpeg? [01:43] <Compn> ubitux : 'because its ugly' apparently [01:43] <ubitux> libpostproc users can still build it independtly, and libav don't have to maintain it [01:44] <ubitux> and can just drop it [01:44] <TheFluff> I think I missed that part of the discussion entirely because I have no idea what you mean by "third-party usage in ffmpeg" [01:44] <ubitux> Compn: i meant, libav could just drop it and let ffmpeg maintains it, not worth forking a dead project [01:44] Action: Compn spent too much time on this bikeshed [01:44] <ubitux> TheFluff: mplayer/vlc users [01:44] <ubitux> for example [01:44] <Compn> ubitux : yes, but i think libav wanted some kind of cooperation [01:44] <ubitux> or anyone using it independently of ffmpeg [01:44] <Compn> i'm not sure what cooperation, Daemon404 was mum on that [01:44] <TheFluff> oh you meant libav as in the libav project, not as in libav* [01:44] <ubitux> yes [01:44] <ubitux> the name is confusing. [01:45] <TheFluff> I agree [01:45] <TheFluff> I don't know, I don't know what d404 was trying to do nor who he was (mis)representing, I just think you should get rid of that piece of shit [01:45] <TheFluff> hope this helps! [01:46] <ubitux> i can understand you just want to get rid of it (first point), the second was about "why so much trouble exporting it if you don't plan to do anything particulary constructive with it"? [01:46] <ubitux> (and since it's still available in another "maintained" repo [01:47] <TheFluff> because I happen to maintain a library that has the great misfortune of being able to use it and I need a good excuse for dropping it [01:47] <ubitux> you still have it [01:47] <Compn> lol [01:47] <Compn> that is a hilarious reason. [01:47] <ubitux> "libav is dropping libpostproc" [01:47] <ubitux> with or without proper export [01:47] <TheFluff> unfortunately we have gone to great lengths to be compatible with both ffmpeg and libav [01:48] <ubitux> you can still use libpostproc from ffmpeg and use libav for the rest [01:48] <Compn> TheFluff : so have lots of other projects [01:48] <Compn> :( [01:48] <ubitux> (also, ffmpeg is compatible with libav) [01:48] <TheFluff> only almost, but that's a discussion for another time [01:48] <ubitux> yes [01:48] <Daemon404> ubitux, i wanted to rm -rf postproc. not my decision [01:48] <TheFluff> personally I think you should just rm -rf libpostproc though [01:48] <Daemon404> now, booze yime. [01:49] <ubitux> Daemon404: if you're doing work i was naively thinking you knew why [01:49] <TheFluff> I don't think you should spend a second more of effort on it than absolutely necessary [01:49] <Daemon404> ubitux, because it was interesting to write teh import script [01:49] <Daemon404> basically. [01:49] <ubitux> hehe ok :) [01:49] <ubitux> but exporting it as no point by itself? [01:50] <ubitux> expect the learning point, which is a good one [01:50] <Daemon404> ask libav folks, i dont know. [01:50] <Daemon404> now srsly, booze. [01:50] Action: Daemon404 gone [01:50] <ubitux> 'night :) [01:50] <ubitux> i won't ask libav folks [01:50] <ubitux> i'm fed up being ignored ;) [01:50] <Compn> TheFluff : sorry for calling you contradictory, i was mixing your and Daemon404's discussion into one opinion [01:51] <Compn> two people with two opinions ;) [01:51] <TheFluff> no problems [01:51] <TheFluff> if you're going to be that way I guess I'll have to pretend to be sorry for being an confrontational asshole, but I'm not, really, it's just how I roll [01:53] <iive> TheFluff: I guess you can save a lot of headache by supporting just FFmpeg :P [01:54] <TheFluff> summary: I think libpostproc is a (buggy) misfeature at best and that alone should be reason enough for considering getting rid of it; the facts that it hasn't been touched in any significant way since 2004 and is extremely unlikely to be fixed due to its extremely hairy codebase do not help its case [01:54] <iive> TheFluff: what bugs? [01:56] <TheFluff> image corruption with certain resolutions, filter string parsing has a number of funny issues (you need to add a comma to the end of the filter string or it'll corrupt the heap iirc), other silly stuff [01:57] <Compn> bugs that TheFluff does not feel like reporting to bug tracker :P [01:57] Action: Compn trolls hard [01:57] <TheFluff> I can probably dig up a few more in excruciating detail if you really want to know [01:57] Action: Compn rolls his troll die [01:57] <TheFluff> Compn: hurpa durpa [01:57] Action: Compn should get a troll die [01:57] <TheFluff> >implying they'd get fixed [01:57] <TheFluff> etc [01:57] <Compn> >implying they will be fixed if you dont report them [01:57] <Compn> >implying someone will fix it [01:58] <TheFluff> they've been there for at the very least six years, if nobody else has reported them then clearly nobody actually uses postproc and that too is an argument speaking in favor of its deletion [01:58] <iive> I want you to dig the bugs up and unload them on the bugtracker [01:59] <TheFluff> :effort: [01:59] <TheFluff> actually hum maybe I should do it [01:59] <Plorkyeran> the heap corruption without a comma at the end bug had a patch sent to the mailing list at some point [01:59] <Plorkyeran> I don't think it ever got applied [01:59] <Compn> a forgotten patch! [01:59] <Compn> :P [02:04] <durandal_1707> there are myriad of them... [02:06] <iive> durandal_1707: do you have a list? [02:07] Action: durandal_1707 shame, i coded nothing today [02:07] <durandal_1707> iive: yes i have, gmane [02:07] <iive> durandal_1707: that's full of committed patches, too hard to find the good one that were left out. [02:08] <durandal_1707> iive: not really, just google for random feature missing in ffmpeg [02:09] Action: iive googles "random feature" [02:09] Action: durandal_1707 looool [02:11] <Compn> hehe [02:14] <durandal_1707> iive: cyuv encoder, rpza encoder, sanm decoder, mthp demuxer [02:14] <michaelni> the heap corruption was IIRC fixed in 84fb4e9df737c856cca7767016110fc74bee56cb [02:14] <michaelni> If someone knows of open bugs in libpostproc, please report them [02:15] <michaelni> iam happy to fix them [02:24] <Compn> michaelni : libav is going to move postproc out of tree [02:24] <Compn> you going to keep it in tree ? [02:24] <Compn> repo i mean [02:26] <funman> move it on videolan.org ? [02:27] <funman> it would stay not too far and on neutral lands [02:27] <Compn> i dont actually know what Daemon404 is asking, if libav already made up its mind [02:27] <Compn> they want michael to take it out of tree? [02:27] <Compn> Daemon404 was not clear [02:28] <Compn> i'll ask in #libav-devel [02:28] <michaelni> ill try to do what is least painfull for its users and myself. Dunno yet what that is [02:29] <Compn> alrighty [02:29] <michaelni> It would be nice if there was just one libpostproc btw [02:30] <Compn> yes, i'm guessing they just want you to move it , so its not forked [02:30] <Compn> they can ignore it, and ffmpeg can use it as external [02:30] <Compn> but this is only my guess [02:30] <Compn> and its quiet in libav channel [02:31] <Compn> probably should ask j-b and reimar before changing things [02:31] <iive> i really don't get, what's the problem with libpostprocess remaining in ffmpeg. [02:32] <Compn> did you read the thread on libav-devel ? [02:33] <iive> i guess i've missed it. [02:41] <funman> i've always wondered what it did [02:42] <funman> enabling it in vlc always resulted in absolutely no visible changes to my eyes [02:45] <j-b> funman: it is not like it was completly borken in 1.1 :) [02:45] <j-b> and of course, no issue to host it [02:47] <funman> still can't see the slightiest effect [02:47] <durandal_1707> funman: you are using too high bitrate [02:48] <funman> should i use shitty video to notice it? [02:49] <durandal_1707> postproc helps with poor youtoube clips last time i watched poor youtube clips (last year or two) [02:49] <durandal_1707> i think i also used 2xsai [02:49] <iive> funman: that's the idea. to smooth a little these huge macroblocks. [02:51] <funman> ah perhaps because i don't build postproc [02:52] <funman> j-b: i didn't eat your placebo pill with that dummy postproc menu! [02:53] <j-b> :) [03:00] <Compn> [20:56] <Kovensky> Compn: Daemon404 ported it to build outside of the libav tree in https://github.com/dwbuiten/postproc [03:00] <funman> hm i could ask which irc bot the mingw-w64 people use [03:01] <funman> if you talk to it on freenode it repeats what you said on oftc and vice versa [03:01] <Compn> lol [03:01] <Compn> i've seen those irc relays [03:01] <Compn> on a few diff channels/servers [03:20] <durandal_1707> add 96bpp and 128bpp pixel format? [03:21] <iive> durandal_1707: let's round it to 256 [03:22] <durandal_1707> iive: http://wiki.multimedia.cx/index.php?title=WDP [03:26] <durandal_1707> 96bpp would be needed to support rgbe [03:36] <CIA-108> ffmpeg: 03Ronald S. Bultje 07master * r59f474b49d 10ffmpeg/libavcodec/x86/ (Makefile pngdsp-init.c pngdsp.asm): png: convert DSP functions to yasm. [03:36] <CIA-108> ffmpeg: 03Mans Rullgard 07master * r3715d841a6 10ffmpeg/ (avconv.c libavcodec/aacenc.c): [03:36] <CIA-108> ffmpeg: Fix non-C89 declarations in for loops [03:36] <CIA-108> ffmpeg: Some compilers still do not support this syntax. [03:36] <CIA-108> ffmpeg: Signed-off-by: Mans Rullgard <mans(a)mansr.com> [03:36] <CIA-108> ffmpeg: 03Ronald S. Bultje 07master * rf91c4b7824 10ffmpeg/libavcodec/x86/ (pngdsp-init.c pngdsp.asm): png: add SSE2 version for add_bytes_l2. [03:36] <CIA-108> ffmpeg: 03Mike Melanson 07master * rcc09dc7863 10ffmpeg/libavformat/mpegtsenc.c: [03:36] <CIA-108> ffmpeg: s/vbsf/bsf/ [03:36] <CIA-108> ffmpeg: -vbsf doesn't exist anymore. It got renamed to -bsf somewhere along the [03:36] <CIA-108> ffmpeg: line. Update print statement accordingly. [03:36] <CIA-108> ffmpeg: Signed-off-by: Anton Khirnov <anton(a)khirnov.net> [03:36] <CIA-108> ffmpeg: 03Mike Melanson 07master * rb864b38397 10ffmpeg/libavformat/yuv4mpeg.c: (log message trimmed) [03:36] <CIA-108> ffmpeg: yuv4mpeg: allow YUV4MPEG2 demuxer to recognize 'C420' colorspace. [03:36] <CIA-108> ffmpeg: Current demuxer recognizes several colorspace formats that begin with 'C420' [03:36] <CIA-108> ffmpeg: but does not yet recognize plain 'C420'. GStreamer's y4menc component [03:36] <CIA-108> ffmpeg: generates .y4m files with a 'C420' colorspace. This new comparison is [03:36] <CIA-108> ffmpeg: placed after the other 'C420' checks so that it doesn't interfere with [03:36] <CIA-108> ffmpeg: them. [03:36] <CIA-108> ffmpeg: 03Rafaël Carré 07master * r420df8b7c4 10ffmpeg/libavformat/utils.c: [03:36] <CIA-108> ffmpeg: avformat_write_header(): detail error message [03:36] <CIA-108> ffmpeg: Give the exact aspect ratios when there is a mismatch between encoder [03:36] <CIA-108> ffmpeg: and muxer. [03:36] <CIA-108> ffmpeg: Signed-off-by: Anton Khirnov <anton(a)khirnov.net> [03:36] <CIA-108> ffmpeg: 03Diego Biurrun 07master * ra846202343 10ffmpeg/libavformat/rtsp.c: rtsp: Remove some unused variables from ff_rtsp_connect(). [03:36] <CIA-108> ffmpeg: 03Paul B Mahol 07master * r8b933129b9 10ffmpeg/ (doc/APIchanges libavutil/Makefile libavutil/avutil.h): [03:36] <CIA-108> ffmpeg: avutil: make intfloat api public [03:36] <CIA-108> ffmpeg: The functions are already av_ prefixed and intfloat header is already provided. [03:36] <CIA-108> ffmpeg: Install libavutil/intfloat.h [03:36] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [03:36] <CIA-108> ffmpeg: Signed-off-by: Anton Khirnov <anton(a)khirnov.net> [03:36] <CIA-108> (82 lines omitted) [04:17] <NapHtaKeRoSene> hii [04:17] <NapHtaKeRoSene> what do you use for BSR BSF in C ? [05:18] <michaelni> NapHtaKeRoSene, in ffmpeg you can use av_log2() for one direction [05:19] <michaelni> there are no portable standard functions to do this, gcc has buildins though [05:21] <funman> propose them for next C standard! oh wait you need to wait ten years now [05:22] <canbal> Hi, is there anyone who can answer a few questions of mine about CABAC? Thanks... [06:11] <bizzybody> Does this project still need the iFinish transcoder to implement support for that codec? [06:11] <Compn> bizzybody : we need samples of a lot of things [06:11] <Compn> which one was ifinish? [06:11] Action: Compn looks [06:11] <bizzybody> Media 100 [06:12] <Compn> yeah, i dont think we have samples for that [06:12] <Compn> sample video with that codec [06:12] <Compn> any weird codecs we dont support, we need samples for :) [06:13] <bizzybody> I wonder if there's something useful on the tutorial disc with the Media 100 2.6.1 software? [06:13] <Compn> possibly, but usually they have raw sample videos, not encoded sample videos [06:13] <bizzybody> I do have the transcoder for Windows, which does support realtime playback in spite of what Media 100 claimed on their website. [06:15] <Compn> i dont know what kind of codec it is really [06:15] <Compn> i just picked some random ones we didnt have samples for [06:16] <Compn> anyways, upload some samples if you can, to our bug tracker or any file host [06:16] Action: Compn goes afk [06:19] <bizzybody> OK, copying stuff off tutorial CD with TransMac, probably have to flatten the MOV files. [07:08] <bizzybody> I've a 2.2 meg rar file with a couple of small MOV videos, their resource forks, the iFinish transcoder for windows and a little flattener utility. Where do you need it uploaded? [07:17] <bizzybody> I've uploaded the Media 100 stuff to http://partsbyemc.com/pub/ Hopefully you can use that to add support to FFMPEG [11:23] <ubitux> michaelni: we have two presets directories: ffpresets/ and presets/ (i guess because of 8096fdf0b6886305ea1a8cb2c869ab2732cd8e11); should we move all the ffpresets to the new presets directory? [13:51] <ubitux> what's the gain of having do_avconv_crc in addition of do_avconv? [13:52] <ubitux> do_avconv seems to run a md5 while do_avconv_crc seems to be the same with crc; but is it really summing the same thing? [17:42] <durandal_1707> michaelni: how to test unscaled conversion stuff with ffplay/ffmpeg? [17:43] <durandal_1707> i'm asking because unscaled yuv2rgb in case of yuva444p is never going to be called [17:44] <durandal_1707> and i can't get message "using unscaled %s -> %s special converter" [17:45] <CIA-108> ffmpeg: 03Clément BSsch 07master * rc673671333 10ffmpeg/ffmpeg.c: ffmpeg: fix -map_channel being ignored when resampling is not needed. [18:18] <michaelni> durandal_1707, i thought it would be used by default when possible [18:18] <durandal11707> michaelni: what? [18:18] <michaelni> youd have to add a av_log or 2 to find out why not [18:22] <michaelni> durandal_1707, rereading what you wrote and thinking, i dont think 444 is supported in the yuv->rgb unscaled code [18:25] <durandal11707> michaelni: of course it is not (only yuva420p is checked in code), but cant get msg even with yuva420p too [18:43] <j-b> http://git.videolan.org/?p=ffmpeg.git;a=summary has switched HTTP server engine. [18:51] <durandal_1707> what "get_buffer() cannot be called after ff_thread_finish_setup()" means? [19:11] <beastd> hi [19:16] <ubitux> hi beastd :) [19:21] Action: beastd goes AFK now [19:21] <beastd> bbl [20:04] <ubitux> :) [22:00] <CIA-108> ffmpeg: 03Michael Niedermayer 07master * r146ef3f37c 10ffmpeg/libavcodec/pngdec.c: [22:00] <CIA-108> ffmpeg: pngdec: fix warning about pointer types [22:00] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:00] <CIA-108> ffmpeg: 03Michael Niedermayer 07master * re4a714f65a 10ffmpeg/libavcodec/h264.c: [22:00] <CIA-108> ffmpeg: h264: Return the correct number of bytes for mid strea, extradata. [22:00] <CIA-108> ffmpeg: Fixes the hang with Ticket952 [22:00] <CIA-108> ffmpeg: Tested-by: Nicolas George <nicolas.george(a)normalesup.org> [22:00] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:00] <CIA-108> ffmpeg: 03Michael Niedermayer 07master * r05ebe51e00 10ffmpeg/libavcodec/ (h264.c mpegvideo.c): [22:00] <CIA-108> ffmpeg: mpeg/h264: update thread context even if it is not initialized. [22:00] <CIA-108> ffmpeg: Fixes decoding of Ticket952 [22:00] <CIA-108> ffmpeg: Tested-by: Nicolas George <nicolas.george(a)normalesup.org> [22:00] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:01] <CIA-108> ffmpeg: 03Paul B Mahol 07master * rb0a30ea7c8 10ffmpeg/libavcodec/ffv1.c: [22:01] <CIA-108> ffmpeg: ffv1enc: PIX_FMT_YUVA444P support [22:01] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [22:01] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:01] <CIA-108> ffmpeg: 03Paul B Mahol 07master * r7054629f04 10ffmpeg/libavcodec/ffv1.c: [22:01] <CIA-108> ffmpeg: ffv1enc: PIX_FMT_YUVA420P support [22:01] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [22:01] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:01] <CIA-108> ffmpeg: 03Paul B Mahol 07master * rc8eba9f9d5 10ffmpeg/libavcodec/ffv1.c: [22:01] <CIA-108> ffmpeg: ffv1dec: PIX_FMT_YUVA444P support [22:01] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [22:01] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [22:01] <CIA-108> ffmpeg: 03Paul B Mahol 07master * rab7da16ea4 10ffmpeg/libavcodec/ffv1.c: [22:01] <CIA-108> ffmpeg: ffv1dec: PIX_FMT_YUVA420P support [22:01] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [22:01] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [23:21] <CIA-108> ffmpeg: 03Mashiat Sarker Shakkhar 07master * r6ca1016b3a 10ffmpeg/libavcodec/wmalosslessdec.c: Use correct variable type for 32-bit samples buffer [23:21] <CIA-108> ffmpeg: 03Mashiat Sarker Shakkhar 07master * rbfbd22f9f0 10ffmpeg/libavcodec/wmalosslessdec.c: Cosmetics: Fix some whitespace errors and indentation [23:22] <CIA-108> ffmpeg: 03Michael Niedermayer 07master * r0d17477e2c 10ffmpeg/: [23:22] <CIA-108> ffmpeg: Merge remote-tracking branch 'shariman/wmall' [23:22] <CIA-108> ffmpeg: * shariman/wmall: [23:22] <CIA-108> ffmpeg: Cosmetics: Fix some whitespace errors and indentation [23:22] <CIA-108> ffmpeg: Use correct variable type for 32-bit samples buffer [23:22] <CIA-108> ffmpeg: Merged-by: Michael Niedermayer <michaelni(a)gmx.at> [23:40] <CIA-108> ffmpeg: 03Paul B Mahol 07master * r0a3a69e8d7 10ffmpeg/libavcodec/ffv1.c: [23:40] <CIA-108> ffmpeg: ffv1dec: use correct linesize [23:40] <CIA-108> ffmpeg: Apparently this did not break anything. [23:40] <CIA-108> ffmpeg: Signed-off-by: Paul B Mahol <onemda(a)gmail.com> [23:40] <CIA-108> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> [23:57] <durandal_1707> michaelni: h264 decoder support both frame and slice threads - can they be used at same time? [23:57] <durandal_1707> michaelni: could ffv1 benefit from frame threads? [23:59] <kierank> no they can't be used at the same time afaik [00:00] --- Wed Feb 1 2012

1 0

ffmpeg.log.20120131
by burek 01 Feb '12

01 Feb '12

[01:39] <plut0> ugh ffmpeg crashed, out of memory [01:39] <plut0> why [01:44] <plut0> is there a way to limit the amount of memory used? [01:46] <undercash> how can i log ffmpeg output? [01:46] <undercash> i dont remenber .. [01:47] <relaxed> plut0: I've never seen that issue before. run `help ulimit` [01:48] <relaxed> undercash: ffmpeg ... 2>ffmpeg.log [01:48] <undercash> ok [01:48] <undercash> thx [01:49] <plut0> relaxed: i'm dealing with a 22G file [01:49] <undercash> my issue is i use a cmd like : ffmpeg -re -i "input.avi" -vcodec libx264 -preset fast -crf 28 -acodec copy -threads 8 -f flv rtmp://live.justin.tv/app/xxxx [01:49] <undercash> and it works fine, and suddenly ffmpeg stop [01:49] <plut0> relaxed: it wasn't writing any of the buffer to disk after almost 2 hours [01:49] <plut0> relaxed: then it crashed [01:49] <undercash> no idea what is happening [01:52] <relaxed> undercash: look at loglevel in the man page [01:52] <undercash> ok relaxed [01:52] <relaxed> plut0: ok, I've encoded files larger then that without problem. [01:53] <plut0> relaxed: so how should i troubleshoot? [01:53] <relaxed> first you should try a recent version to see if it still happens. [01:54] <plut0> relaxed: what else [02:17] <plut0> anyone got ideas on how to troubleshoot ffmpeg running out of memory? [02:26] <plut0> anyone? [03:02] <teratorn> plut0: ffmpeg used from command-line? [03:02] <plut0> teratorn: yes [03:03] <teratorn> plut0: well I'm sure that shouldn't be happening [03:03] <teratorn> plut0: what is your command line? [03:04] <plut0> time ffmpeg -i 00011.m2ts -vcodec libx264 -crf 22 -preset veryfast -acodec copy -scodec copy -threads 0 00011.mkv -scodec copy -newsubtitle -scodec copy -newsubtitle [03:06] <plut0> memory usage goes up and up until it's killed eventually [03:07] <teratorn> plut0: I would guess it's a broken encoder, but you aren [03:07] <teratorn> aren't encoding [03:07] <teratorn> this is latest ffmpeg HEAD? [03:07] <plut0> teratorn: libx264 is encoding right? [03:07] <plut0> teratorn: 0.7.8 [03:07] <teratorn> isn't that old? [03:07] <teratorn> oh, you don't have -vcodec copy [03:07] <teratorn> OK [03:08] <plut0> teratorn: i downgraded from 0.9.1 because of some bugs [03:08] <teratorn> well, try compiling git HEAD from source [03:08] <teratorn> and latest libx264 (and make sure you actually having ffmpeg /using/ latest x264) [03:09] <teratorn> memory errors are often fixed in newer versions of code [03:09] <plut0> i was looking for something stable [03:09] <teratorn> yes? [03:09] <plut0> you want me to try latest git though? [03:09] <teratorn> definitely [03:09] <teratorn> I know it's a pain, but there isn't any other way to debug it [03:10] <teratorn> the other thing I would try is to cut about 1 second of that input file, then transcode it as you are doing, but run ffmpeg under valgrind [03:10] <teratorn> with --leak-check=full [03:11] <plut0> will try latest git [03:11] <teratorn> (I am not 100% sure that ffmpeg actually cleans up after itself resulting in a clean valgrind run, but you can test it against other files and with different codecs, and see if there is something especially bad about x264) [03:12] <teratorn> there isn't any solving the problem without finding the fixing the bug, unless you encode to multiple files - small segments each - then -vcodec copy them all together [03:12] <teratorn> which may work without leaking [04:35] <kode54> huh, I take it there's no way to transmux a file without applying the "encoder" metadata string [04:50] <pasteeater> kode54: why do you want to get rid of that? [04:51] <kode54> being pedantic for some video files I'm remuxing for the sole purpose of stripping off the metadata [04:52] <pasteeater> you could re-mux it with MP4Box -add input.mp4 output.mp4 [06:42] <SIFTU> I'm trying to convert a 4:3 640x480 video to a video with 16:9 ratio (720x480). I have done this with help from the channel (pillarbox style) but the video looks stretched. Any ideas how I can get it too look normal again? [06:43] <SIFTU> command line - ffmpeg -y -i 9.AVI -vcodec mpeg2video -aspect 16x9 -vf "pad=720:480:40:0:black" -acodec ac3 -ab 448k -threads 0 -qscale 3 -f dvd 9.mpg [06:52] <relaxed> SIFTU: "pad=720:480:40:0:black,setdar=16:9" [06:52] <relaxed> remove -aspect 16x9 [06:52] <SIFTU> relaxed: thanks, I was looking at dar with no idea of what it means [06:52] <relaxed> display aspect ratio [06:55] <SIFTU> relaxed: still seems stretched [06:59] <relaxed> if the source is 4:3 then set the dar to that as well [07:01] <SIFTU> I'm not sure I follow.. is that setsar? [07:04] <relaxed> how does "-vf scale=720:480,setdar=4:3" look" [07:04] <SIFTU> squished [07:04] <SIFTU> oh hang on [07:04] <SIFTU> didnt see the scale [07:11] <SIFTU> relaxed: still not right.. hmm [07:24] <zro> so i gotta put some video on a website thats primaryily going to be viewed on mobile devices (ie. phones) ... should i be converting these to 3gp ? [07:42] <SIFTU> it seemed like -vf "pad=ih*16/9:ih:(ow-iw)/2:(oh-ih)/2" works [07:42] <SIFTU> from the man page [09:47] <lobstery> Hi, I'm running into this issue: https://ffmpeg.org/trac/ffmpeg/ticket/309 - libx264 complains width is not divisible by 2. Anyone know a way around this? [09:49] <JEEB> well, there's an example command there that rounds your picture to mod2 [09:49] <JEEB> libx264 by default encodes 420 YCbCr [09:49] <JEEB> which means that your picture has to be mod2 [09:49] <JEEB> in both width and height [09:49] <JEEB> you can also encode in 444 YCbCr, but that will most probably not be what you want :P [09:51] <lobstery> not sure I understood half of that - very new to ffmpeg and all this. [09:52] <lobstery> what is the example command? [09:52] <JEEB> https://ffmpeg.org/trac/ffmpeg/ticket/309#comment:3 [09:55] <lobstery> hm.. okay.. so what I want is to keep the height at 480 and then scale the width - so a 1270x720 becomes something like 854x480 [09:55] <lobstery> so changing that to scale="trunc(oh/a/2)*2:480" should work I believe [09:56] <JEEB> most probably [09:56] <JEEB> not pretty syntax, but derp [10:00] <lobstery> that seemed to scale it to 272x480 :/ [10:01] <JEEB> lol [10:01] <JEEB> prolly ow instead of oh [10:01] <JEEB> lol [10:02] <JEEB> oh wait [10:02] <JEEB> derp [10:02] <lobstery> yeah I think it should be oh [10:04] <lobstery> what does the trunc method do exactly? [10:04] <JEEB> that probably makes it an int, not sure tho [10:05] <JEEB> +Round the value of expression @var{expr} towards zero to the nearest [10:05] <JEEB> +integer. For example, "trunc(-1.5)" is "-1.0". [10:06] <lobstery> ah [10:18] <lobstery> scale="trunc(oh*a/2)*2:480" <- ftw [10:18] <lobstery> Thanks JEEB [10:21] <JEEB> np [12:08] <MikeW> Hey guys, if I have a byte array of mp3 audio stream data, is there a straightforward way to construct a new MP3 file using that data and then apply ID3/metadata to that file? [12:09] <MikeW> or would ffmpeg be a bad choice for this and would a different library be a better approach? [12:19] <vivienschilis> hi [12:19] <vivienschilis> Can some explain me this bufsize [12:19] <vivienschilis> someone* [12:20] <vivienschilis> does setting the bufsize has an impact on hardware decoding? I think about like phones [12:20] <JEEB> maxrate and bufsize come into play when there's a certain limit to how fast data can be pushed to the decoder [12:21] <JEEB> (and then there's the level limitations of course) [12:21] <JEEB> in most cases if you're making something to play from an SD card etc., unless the device really sucks you don't have to care much [12:22] <JEEB> some devices on the other hand suck really hard if you don't abide to some obscure limitations that depend on the device (I remember some blackberries having this, and maybe pre-3GS iDevices) [12:26] <vivienschilis> thanks [12:26] <vivienschilis> My initial issue was that [12:26] <JEEB> most obvious use cases where you need vbv maxrate and bufsize are disc media and network streaming [12:27] <vivienschilis> my client wants to predict his streaming billing [12:27] <vivienschilis> so he want's bounds on bitrates [12:28] <vivienschilis> to kind of predict the video file size [12:28] <vivienschilis> with a min/max bitrate [12:29] <vivienschilis> but I am not sure its the right mode, CRF would produce generally smaller files [12:29] <vivienschilis> don't you think? [12:29] <JEEB> CRF + maxrate and bufsize is what I'd use for general streaming [12:29] <vivienschilis> ok cool [12:30] <vivienschilis> how should i set the bufsize? [12:30] <JEEB> that depends on the client :P [12:30] <JEEB> how much does the player buffer [12:30] <vivienschilis> yeah that's my issue [12:33] <vivienschilis> thanks JEEB [12:34] <vivienschilis> if you though want to use the iPhone player for example [12:34] <vivienschilis> it's not a good choice right? cause it's not streaming anymore [12:34] <vivienschilis> it's http delivery [12:35] <JEEB> well, there's still network speeds to be thought of :P [12:35] <JEEB> so even if you pre-encode the file you usually want it to abide to some vbv limits [12:36] <JEEB> so that you know that "if the player buffers Xkb and has a speed of at least Ykbps it will play without further buffering" [12:37] <JEEB> with non-live http delivery of video you can of course let it be more lax, as the content isn't going anywhere, but it still gives a better experience for the user [12:39] <leandrosansilva> Hello to all. I get the lastest ffmpegt source code and it broken my app because I'm using AVFormatParamers. I'm seing it was replaced by AVDictionary. Where can I find a table which "maps" the old avformat options to the dictionary? [12:39] <leandrosansilva> for example, width, height? [12:40] <vivienschilis> brb [12:41] <leandrosansilva> The avio_set_interrupt_cb was also removed [12:41] <leandrosansilva> is it a replacement for it? [12:41] <leandrosansilva> ops is there* [12:54] <vivienschilis> JEEB, do you have some references I could read about this buffering? [12:54] <vivienschilis> to understand it better [13:05] <vivienschilis> is there a sense to put a buffer size < maxrate ? [13:07] <JEEB> maxrate is basically the average bitrate that is calculated over the bufsize amount of bits, so it's possible [13:08] <JEEB> (the average bitrate that is never exceeded) [13:08] <JEEB> s/bits/bytes/ [13:09] <JEEB> the smaller the bufsize is, the less freedom the encoder has to vary the bitrate [13:09] <JEEB> since it has to always keep within the "maximum average bitrate within the bufsize"'s limits [13:12] <vivienschilis> make sens [13:15] <vivienschilis> so in general if I double maxrate I get a fast buffering of about 2 sec? is that a good practice? [13:15] <JEEB> what [13:16] <vivienschilis> on the player side [13:16] <JEEB> what [13:16] <JEEB> re-phrase your question [13:16] <vivienschilis> ok [13:20] <vivienschilis> sorry I doesn't mean anything as it depends on the bandwdith [13:20] <vivienschilis> my question is what is the best practice to set the buffer size limit. my client as no requirement [13:23] <vivienschilis> a few people told me to just double the maxrate but I didn't get a valid reason from them so i don't trust those kind of formulas [13:26] <JEEB> oh, you mean bufsize = 2 * maxrate ? [13:27] <vivienschilis> YES [13:27] <vivienschilis> oups capital letters [13:29] <JEEB> the general idea vbv is that you'd either have a specification to which you abide on both sides, but yeah -- not sure if iPhone's player or whatever has those public [13:29] <JEEB> *-either [13:30] <JEEB> generally maxrate=bufsize or so is sane, but I don't think bufsize = 2 * maxrate is a bad idea (unless the player doesn't buffer enough and then you end up with more buffering) [13:30] <JEEB> if you have no specifications you'll just have to try and hope :P [13:31] <vivienschilis> better maybe it's better having a two pass + bitrate tolerance? [13:31] <JEEB> dunno how that is better :P [13:32] <JEEB> you still have the same problem [13:36] <vivienschilis> true [13:44] <JEEB> vivienschilis, well -- having some kind of vbv will at least let you have control over how much data goes over at max. :P [13:45] <vivienschilis> sure but I don't want to be constrained to set a buffer information on the player [13:46] <JEEB> what [13:46] <vivienschilis> you said I have to set the buffer size on the player as weel [13:46] <vivienschilis> well* [13:47] <JEEB> yes, or just abide to whatever default behavior it has [13:47] <JEEB> otherwise you can't stream really :P [13:48] <JEEB> vbv is needed for network streaming and so on, to be able to set a minimum network speed your stream can be watched with [13:48] <JEEB> (or well, to make it sure that a person with a speed of Xkbps or more can watch it without hick-ups) [13:57] <vivienschilis> They have their spec from their Sorenson software [13:57] <vivienschilis> they have 1 pass VBR [13:57] <vivienschilis> with a Data Rate... [13:57] <vivienschilis> how is that possible [14:00] <vivienschilis> with min and max in percent urggh!! [14:32] <ultramantis> hello is it possible to remove or change these lines in the logfile to something diffrent or custom frame=176115 fps= 49 q=27.0 [14:33] <Mavrik> they're hardcoded [14:34] <ultramantis> can they be changed? [14:34] <Mavrik> if you change them in the source [14:34] <ultramantis> which file in the source would do that [14:35] <Mavrik> grep over them and see? [14:35] <ultramantis> grep over them? [14:39] <iive> try to find the string "frame=" in the source files. [14:40] <ultramantis> right i under stand that part but which source file exactly is there a specific one and would i just # it out [15:16] <ultramantis> ok i think i figured it out [15:16] <ultramantis> whats the proper way to make a -vstats_file [15:16] <ultramantis> just add that line? [15:25] <ultramantis> figured that otu [15:25] <ultramantis> but im do i need to recompile n configure after i mess with the C file [15:35] <LexSfX> is there a build of ffmpeg that has a "libx26410bit" or something like that for encoding yuv444p10 h264 video with ffmpeg CLI? [15:35] <LexSfX> i thought it would work with libx264, but i was disappointedly surprised [15:35] <LexSfX> libx264 doesn't even do yuv444p [16:12] <evil_andy> I'm trying to run some code using the ffmpeg libs. Every time I try to decode some frames and convert them, I get the following error: http://pastebin.ca/2108154 [16:13] <evil_andy> http://pastebin.ca/2108155 is updated with the output of ldd pertaining to the ffmpeg libs [16:58] <Zeranoe> Could anyone tell me how FFmpeg uses GnuTLS? I see it's included now [17:54] <peace_> guys but wtf ; ffprobe is bad => I have worked on output with awk but ... it's a pain [17:54] <peace_> coudl someone take care of that ? [17:55] <peace_> look at this http://paste.ubuntu.com/824067 [18:01] <saste> peace: read the manual and you'll find plenty output options [18:07] <peace_> saste: uou mean this http://ffmpeg.org/ffprobe.html [18:08] <saste> peace_: same as man ffprobe [18:09] <relaxed> peace_: it supports JSON output which is nice [18:10] <peace_> i see [18:17] <peace_> relaxed: is this correct? fprobe -print_format JSON ~/output.mov [18:17] <peace_> it gives me an error [18:17] <peace_> mah [18:17] <peace_> ok -print_format json [18:18] <relaxed> or, ffprobe -print_format JSON -show_streams ~/output.mov [18:20] <peace_> ah damn... i hate fork i guess i have libav stuff [18:20] <peace_> -.- [18:20] <peace_> relaxed: [18:20] <peace_> Copyright (c) 2007-2011 the Libav developers [18:21] <peace_> built on Jan 24 2012 07:22:02 with gcc 4.6.2 [18:21] <peace_> i guess it's not the standard one? [18:21] <peace_> right ? [18:21] <saste> peace_: yes the output changes have not been ported to libav [18:21] <relaxed> correct, avprobe doesn't have JSON support [18:22] <peace_> wtf who the hell is has done the fork now everytime i have to check this shit [18:22] <\DSAFEW\> hey I figured out why ffmpeg was giving the file not found error for every rtmp address, networking wasn't compiled in [18:30] <relaxed> peace_: my avprobe doesn't output the trailing comma [18:30] <peace_> relaxed: i dunno but this fork is a pain i had my software broken for that [18:31] <relaxed> what broke? [18:31] <peace_> i did a service menu for dolphin to convert file with profiles [18:32] <peace_> now i have to analyze every files to get the proper bitrate for video and audio to use on ffmpeg [18:33] <relaxed> that tells me nothing [18:33] <peace_> mydistro provide some sort fo ffmpeg [18:33] <peace_> ffmpeg [18:33] <peace_> ffmpeg version 0.8-4:0.8-1ubuntu1, Copyright (c) 2000-2011 the Libav developers [18:34] <peace_> it's not ffmpeg i guess [18:34] <peace_> This program is not developed anymore and is only provided for compatibility. Use avconv instead (see Changelog for the list of incompatible changes). [18:35] <relaxed> again, what specific problem did it cause? [18:35] <peace_> :) it doesn't provide the json stuff [18:35] <peace_> that is the problem [18:36] <relaxed> uh, didn't you just learn about that 5 minutes ago? [18:37] <peace_> i have an old version of ffmpeg in another machine tht worked fine with my awk script [18:37] <peace_> and .... now i see 2 version of ffmpeg [18:37] <peace_> that doesn't respect the old style for output [18:38] <peace_> so my stuff is broken [18:39] <peace_> i have to check if the user uses ffmpeg standard of he use the avstuff [18:39] <peace_> bah [18:51] <relaxed> it should be trivial to add another awk gsub for trailing commas [18:51] <\DSAFEW\> I'm trying to capture the sound from my mic and my normal computer sounds with ffmpeg and alsa, any help would be appreciated [18:52] <\DSAFEW\> right now I can only capture my mic [19:09] <peace_> relaxed: :) [19:09] <peace_> awk ' [19:09] <peace_> /Video:/{$4 gsub(/\,/,"");gsub(/$Simple Profile$/,"");gsub( /\[[^\)]*]/, "" ) ;videocodec = $4 ; pixelformat = $5 ; resolution= $6 ;videobitrate = $7; fps = $9 } [19:09] <peace_> /Duration:/{$4 gsub(/\:/," ") ;realseconds= $4; seconds = int($4); minutes= $3 ;hours= $2;totalsoundtrack = (seconds+(minutes*60)+(hours*3600)); totalbitrate = $8 } [19:09] <peace_> /Audio:/{$4 gsub(/\,/,"");gsub(/$Simple Profile$/,"");gsub( /\[[^\)]*]/, "" ); audiocodec =$4; freq= $5 ; channel =$7 ; abitrate= $9} END {print "Video Codec: " videocodec ;print "Audio Codec: " audiocodec;print "Video Bitrate: " videobitrate ; print "Audio Bitrate: " abitrate;print "Total Bitrate: " totalbitrate ;print "Resolution: " resolution ; print "Fps: " fps ; print "Frequency: " freq ; print "Channel: " channel;print "Pixel [19:09] <peace_> Format: " pixelformat ;print "Total Duration: " hours ":" minutes ":" realseconds ;print "Total seconds " totalsoundtrack } ' [19:09] <peace_> ops [19:09] <peace_> :D [19:09] <peace_> sorry [19:12] <peace_> relaxed: [19:12] <peace_> awk ' [19:12] <peace_> /Video:/{$4 gsub(/\,/,"");gsub(/$Simple Profile$/,"");gsub( /\[[^\)]*]/, "" ) ;videocodec = $4 ; pixelformat = $5 ; resolution= $6 ;videobitrate = $7; fps = $9 } [19:12] <peace_> /Duration:/{$4 gsub(/\:/," ") ;realseconds= $4; seconds = int($4); minutes= $3 ;hours= $2;totalsoundtrack = (seconds+(minutes*60)+(hours*3600)); totalbitrate = $8 } [19:12] <peace_> /Audio:/{$4 gsub(/\,/,"");gsub(/$Simple Profile$/,"");gsub( /\[[^\)]*]/, "" ); audiocodec =$4; freq= $5 ; channel =$7 ; abitrate= $9} END {print "Video Codec: " videocodec ;print "Audio Codec: " audiocodec;print "Video Bitrate: " videobitrate ; print "Audio Bitrate: " abitrate;print "Total Bitrate: " totalbitrate ;print "Resolution: " resolution ; print "Fps: " fps ; print "Frequency: " freq ; print "Channel: " channel;print "Pixel [19:12] <peace_> Format: " pixelformat ;print "Total Duration: " hours ":" minutes ":" realseconds ;print "Total seconds " totalsoundtrack } ' [19:12] <peace_> -.- [19:12] <peace_> =>>>http://paste.ubuntu.com/824175/ [19:21] <relaxed> you just awked all over my channel [19:22] <peace_> relaxed: :D sorry pasted for mistake on channel [19:26] <relaxed> sub(/,$/,"") [20:33] <LiroXIV> Okay, I'm trying to build a little VB-based FFmpeg front end for making WebM videos. I downloaded some 64-bit binaries of FFmpeg, and hooked it into my program [20:34] <LiroXIV> But, when I try to convert an OGV video to WebM, it gives me an "ffmpeg.exe has stoped working" error. We're running Windows 7 SP1 x64 [22:42] <hachi> hello folks [22:44] <hachi> does anyone happen to know what raw video format I might want to use for output if my input file format is still image jpegs using image2? I'd like to start assembling them into containers so I can address each frame rather than having to have a directory with a few thousand files in it :) [22:47] <hachi> hmm, perhaps my best bet is to simply request a container and say -vcodec copy... I don't really know :) [22:48] <burek> use -f mjpeg [22:55] <hachi> right... so does mjpeg hold the same data as a jpeg image input for each frame? [23:02] <hachi> can I cycle an mjpeg encoded stream back out to jpeg files on disk using no recompression? [23:02] <hachi> with ffmpeg I mean [23:03] <relaxed> yes, use -vcodec ocpy [23:03] <relaxed> copy* [23:04] <hachi> I'm not getting an idempotent result [23:04] <hachi> if I input one file into an avi with copy, and then cycle it back out [23:04] <hachi> I get four images [23:05] <hachi> looking at what they contain now though [23:05] <hachi> first two are damaged... or not really images [23:06] <hachi> third is the exif thumbnail, I guess... it's a tiny copy of the image [23:06] <hachi> fourth one is possibly the original, but corrupted [23:07] <hachi> ffmpeg -i 104MEDIA/FILE0001.JPG -vcodec copy test2.avi [23:07] <hachi> ffmpeg -i test2.avi -f image2 -vcodec copy foo%04d.jpg [23:07] <hachi> should that be giving me an idempotent result... in theory at least? [23:07] <relaxed> ffmpeg -i 104MEDIA/FILE0001.JPG -vframes 1 -vcodec copy -f mjpeg test2.avi [23:09] <hachi> when I try to do the output from that, now I get an MP3 codec error :) [23:10] <hachi> http://nopaste.snit.ch/113720 [23:11] <hachi> well... hum [23:11] <hachi> debian appears to have some ancient ffmpeg [23:13] <burek> :) [23:13] <burek> just compile it [23:13] <hachi> I'm gonna assume that could be part of the problem... [23:13] <burek> git clone .. [23:13] <hachi> yeah :) [23:13] <burek> and compile [23:13] <pozican> not surprising tbh [23:13] <hachi> amazingly, this is a step up from some other version I was attempting this with [23:14] <burek> you can download static binary [23:14] <burek> just to try that [23:14] <burek> it will be lot faster [23:14] <hachi> is an OSX binary available by chance? [23:14] <hachi> I didn't see one [23:15] <burek> I'm not on that user account where I bookmarked that page, you can ask relaxed for a link [23:15] <hachi> meh, I'll just compile it... it's not exactly hard [23:15] <hachi> thank you both for the input.... on inputs [23:15] <burek> ok :) this way it would be download and run [23:15] <burek> but ok [23:16] <burek> https://sites.google.com/site/linuxencoding/builds [23:17] <burek> just to try something [23:17] <burek> oukey [23:24] <relaxed> hachi: omit '-f mjpeg', just use -vcodec copy for lossless conversion back and forth [00:00] --- Wed Feb 1 2012

1 0