Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

January 2014

1 participants
62 discussions

ffmpeg-devel.log.20140125
by burek 26 Jan '14

26 Jan '14

[00:52] <BBB> ubitux: fixed the valgrind stuff [01:23] <BBB> ubitux: and that other patch fixes fuzzed2.ivf - I believe it fixes the others as well but please double check, I'm having issues reproducing reliably for some reason [01:31] <BBB> michaelni: and for you, a mergeable branch: https://github.com/rbultje/ffmpeg/commits/vp9-simd [01:32] <cone-484> ffmpeg.git 03Wim Vander Schelden 07master:af09be4f4b2f: Fixed a memory leak in dvbsubenc.c: sub->num_rects was reduced without freeing the associated rects. [01:32] <cone-484> ffmpeg.git 03Michael Niedermayer 07master:cf812d812967: avcodec/dvbsubdec: Remove unused display_list_size [01:39] <cone-484> ffmpeg.git 03Kostya Shishkov 07master:0e1ad2f591b8: dxtory: add more compressed and uncompressed modes [01:39] <cone-484> ffmpeg.git 03Michael Niedermayer 07master:2d0d1f7eb3f7: Merge commit '0e1ad2f591b87e944550c15b54e54f8189743289' [01:43] <cone-484> ffmpeg.git 03Kostya Shishkov 07master:28e1eed3c2e7: dxtory: compressed RGB555/RGB565 decoding support [01:44] <cone-484> ffmpeg.git 03Michael Niedermayer 07master:4b84a69ebb35: Merge remote-tracking branch 'qatar/master' [01:51] <cone-484> ffmpeg.git 03Ronald S. Bultje 07master:baf47020cd23: vp9/x86: 8x8 iadst SIMD (ssse3/avx) variants. [01:51] <cone-484> ffmpeg.git 03Ronald S. Bultje 07master:d43efa68bd53: vp9/x86: 4x4 iadst SIMD (ssse3) variants. [01:51] <cone-484> ffmpeg.git 03Ronald S. Bultje 07master:97474d527f9a: vp9/x86: iwht4x4 (lossless) mmx. [01:51] <cone-484> ffmpeg.git 03Ronald S. Bultje 07master:c9e6325ed984: vp9/x86: use explicit register for relative stack references. [01:51] <cone-484> ffmpeg.git 03Ronald S. Bultje 07master:4147b337c105: vp9: fix memory corruption if header decoding fails after size change. [01:51] <cone-484> ffmpeg.git 03Michael Niedermayer 07master:5554c6dd455c: Merge remote-tracking branch 'rbultje/vp9-simd' [02:50] <BBB> ubitux: https://github.com/rbultje/ffmpeg/commits/vp9-coef-opts for review [02:50] <BBB> ubitux: basically some coef reading and zero writing optimizations [02:57] <BBB> (renamed to vp9-context-opts) [05:00] <cone-484> ffmpeg.git 03Ramiro Polla 07master:222fb8276dc4: lavfi/drawtext: get bitmap from glyph in a separate step [05:00] <cone-484> ffmpeg.git 03Ramiro Polla 07master:78a9f185eb17: lavfi/drawtext: add option for drawing border around text [11:58] <ubitux> BBB: https://github.com/rbultje/ffmpeg/compare/vp9-context-opts ? [11:58] <ubitux> ok [11:58] <ubitux> didn't see the rename [11:58] <ubitux> will test [11:58] <ubitux> (the valgrind stuff) [12:15] <ubitux> BBB: all fuzzed files working here :) [12:15] <ubitux> thanks [12:20] <ubitux> BBB: want another one? :) [12:37] <BBB> hum, that bad? [12:37] <BBB> sure [13:29] <saste> michaelni, any insight about the codec time_base thing? [14:30] <ubitux> michaelni: https://github.com/ubitux/FFmpeg/compare/master...lossless opinion? [15:02] <michaelni> saste, i suspect some code doesnt expect that the timebase has been overridden [15:02] <michaelni> also ticks_per_frame probably would need to be set [15:22] <michaelni> ubitux, nice [15:35] <ubitux> heh, just won about 10 cycles in lpf 16x16 vert [15:37] <ubitux> 3931 decicycles in ff_vp9_loop_filter_v_16_16_ssse3, 4193221 runs, 1083 skips [15:37] <ubitux> 3827 decicycles in ff_vp9_loop_filter_v_16_16_ssse3, 4193594 runs, 710 skips [15:37] <ubitux> \o/ [15:41] <wm4> what's a decicycle anyway? [15:41] <ubitux> a tens of cycle? [15:42] <wm4> how can there be 0.1 cycles? [15:45] <kierank> wm4: it's an average [15:45] <kierank> http://git.videolan.org/?p=ffmpeg.git;a=blob;f=doc/examples/filtering_video… [15:45] <kierank> why is the input a sink? [15:49] <wm4> hm [15:49] <BBB> wm4: average of many runs [15:49] <BBB> ubitux: cool [15:50] <wm4> BBB: I always thought the minimum of all runs is the most useful value [15:50] <ubitux> BBB: and i found a way to make the asm simpler :p [15:50] <BBB> cool [15:50] <wm4> kierank: maybe it's misnamed? the buffer src is the input, buffer sink the output... [15:52] <wm4> oh I see [15:52] <wm4> the output is basically a filter pad which outputs something, and which is thus linked to an input pad of the newly created filters [15:53] <wm4> so the buffer src is an output, because it provides an output [15:58] <ubitux> BBB: https://github.com/ubitux/FFmpeg/compare/vp9-simd [15:59] <ubitux> and btw, this leads to a < 620 lines file [16:00] <kierank> wm4: any idea what filter_descr is? [16:01] <kierank> and how it is different to the parameters you pass to the filters directly [16:01] <wm4> looks like it is the graph [16:01] <wm4> it's the string passed to the -vf option on the command line [16:01] <wm4> oh wait this is the example [16:01] <kierank> yeah [16:01] <wm4> but the same code is in ffplay [16:01] <wm4> copy pasted, in fact [16:02] <ubitux> kierank: saste once told me why it was that way [16:02] <ubitux> but i forgot since it's completely unintuitive [16:02] <wm4> so the buffer src, sink are for putting in/receiving AVFrames [16:02] <ubitux> :( [16:02] <wm4> and the format parameter is to force output configuration (i.e. what buffer sink will receive) [16:02] <wm4> oh [16:02] <wm4> the other way around [16:03] <wm4> it's for forcing input [16:03] <wm4> whatever [16:15] <ubitux> michaelni: fate is covering that lossless code? [16:15] <ubitux> (the asm) [16:18] <michaelni> ubitux, it should [16:18] <ubitux> ok, so i can push? :P [16:18] <michaelni> yes [16:27] <cone-652> ffmpeg.git 03Clément BSsch 07master:cddbfd2a9554: x86/lossless_videodsp: simplify and explicit aligned/unaligned flags [16:27] <cone-652> ffmpeg.git 03Clément BSsch 07master:5267e850563d: x86/lossless_videodsp: use common macro for add and diff int16 loop. [16:27] <cone-652> ffmpeg.git 03Clément BSsch 07master:5f4d04d08470: x86/lossless_videodsp: silly one-line cosmetic. [16:33] <cone-652> ffmpeg.git 03Michael Niedermayer 07master:018e2b57ca83: avcodec/libx264: also consider ticks per frame for fps/timebase setup [17:00] <jnvsor> I have a fate test failing that's giving an MD5 diff but has the same CRC - I changed timecode.c to use fmod instead of % to support fractional FPS [17:00] <kierank> ??? [17:00] <kierank> fractional fps [17:01] <jnvsor> As in 29.97 [17:01] <kierank> that's a terrible idea [17:01] <kierank> use drop frame for that [17:01] <jnvsor> timecode.c uses int instead of AVRational, and it's hardcoded to drop frames so if you actually do want 30fps you'll get desync [17:02] <jnvsor> I would have thought AVRational would be the "right" way to do it [17:02] <kierank> it only uses drop frame if you ask it to [17:03] <jnvsor> So, timecode is supposed to be limited to specific framerates and I should look for this bug in the filter instead? [17:04] <kierank> yes [17:05] <jnvsor> Right, thanks for the info [17:05] <cone-652> ffmpeg.git 03Lars Kiesow 07master:7fc4c1846300: Factors for scale filter [17:06] <cone-652> ffmpeg.git 03Lars Kiesow 07master:e395f8de5ac6: Fixed factor for scale filter [17:06] <cone-652> ffmpeg.git 03Lars Kiesow 07master:69b1d1d99bc5: Documentation for scale filter factor [17:06] <cone-652> ffmpeg.git 03Lars Kiesow 07master:c49b0360966d: Documentation for scale filter factor [17:06] <cone-652> ffmpeg.git 03Michael Niedermayer 07master:682ddb89cf15: Merge branch 'scale-filter-factor' of https://github.com/lkiesow/FFmpeg [17:06] <cone-652> ffmpeg.git 03Michael Niedermayer 07master:1e48c39ece3e: avfilter/vf_scale: do aspect ratio and scale factor compensation together [17:06] <cone-652> ffmpeg.git 03Michael Niedermayer 07master:214a3b8bf939: avfilter/vf_scale: simplify alignment code [17:09] <ubitux> oh that's pretty cool [17:11] <ubitux> > It is a DOS program from 1998 [17:11] <ubitux> we definitely need to support that app :D [17:11] <Daemon404> dos in 1998? [17:11] <Daemon404> half a decade late [17:27] <ubitux> BBB: i have both 88_16 working, will improve and share soon [17:47] <ubitux> 9680 decicycles in loop_filter_v_88_16_c, 4193765 runs, 539 skips [17:47] <ubitux> 9233 decicycles in loop_filter_h_88_16_c, 4193751 runs, 553 skips [17:47] <ubitux> to: [17:47] <ubitux> 1932 decicycles in ff_vp9_loop_filter_v_88_16_ssse3, 4194162 runs, 142 skips [17:47] <ubitux> 3058 decicycles in ff_vp9_loop_filter_h_88_16_ssse3, 4193877 runs, 427 skips [17:47] <nevcairiel> its no longer over 9000 :( [17:48] <ubitux> :( [17:49] <ubitux> BBB: https://github.com/ubitux/FFmpeg/compare/vp9-simd [17:53] <ubitux> 5.994 5.478 overall decode time on ped1080p.webm [17:53] <ubitux> (-threads 1) [17:54] <JEEB> nice [18:01] <kierank> av_frame_free and avcodec_free_frame are not confusing at all [18:01] <wm4> with git master, you can just always use av_frame_free AFAIK [18:02] <wm4> with earlier releases, it depends on the situation, and doing the wrong thing may lead to memory leaks or corruption [18:02] <ubitux> kierank: see http://git.videolan.org/?p=ffmpeg.git;a=blob;f=doc/examples/demuxing_decodi… [18:02] <ubitux> somehow, the 3 methods are depicted here [18:02] <wm4> ubitux: that's even more confusing [18:02] <ubitux> none should leak/corrupt anything currently, but maybe something broke again [18:02] <wm4> it's ridiculous and everyone will hate ypou for it [18:03] <ubitux> wm4: wtf? [18:03] <ubitux> it shows how the old code was supposed to look like [18:03] <wm4> yes that will be the reaction of anyone reading this example [18:03] <ubitux> and how it's supposed to be done currently [18:03] <nevcairiel> the new and old api are synonyms now, so you could even mix-match them if you want [18:03] <ubitux> no [18:03] <wm4> ffmpeg is insanely hard to use right, and these verbose, long winded examples demonstrating decprecated features don't help [18:04] <ubitux> seriously [18:04] <ubitux> well whatever [18:04] <wm4> that's just how it is [18:04] <ubitux> i remember people complaining about how that was supposed to be done because anton didn't document the migration from one method to another [18:04] <nevcairiel> ubitux: the old api functions just call the new ones now, they have no single line of specific code anymore [18:04] <ubitux> this example shows how that's done, and allow testing it.. [18:05] <ubitux> nevcairiel: that's what you meant, ok [18:06] <wm4> sounds like anton is to blame, but personally I'm just glad the difference between these was removed [18:06] <wm4> as soon as libav 10 is released I can dump all my old code and stop thinking about it [18:06] <wm4> (until the next api transition) [18:06] <nevcairiel> released? dont you mean adopted by distros? :d [18:07] <ubitux> http://pastie.org/pastes/8666840/text 'seems lpf is no longer that much performance critical [18:10] <jnvsor> How do I get the output of a specific fate test if there are no errors so it doesn't generate a .err file? [18:11] <ubitux> make fate-thetest V=1 [18:11] <ubitux> re-run the command manually [18:11] <jnvsor> That works I guess [18:22] <ubitux> decode_coeffs_b is certainly slow.. [19:05] <BBB> ubitux: oh whoa good work [19:28] <BBB> ubitux: so for the 88_h, you don't need a 16x16 transpose right? you should just need a 16x8 [19:29] <ubitux> ah? [19:30] <BBB> ubitux: and filter4 in line 488, that's filter14 right? [19:30] <BBB> oh no it's filter4 nm [19:31] <BBB> yeah so my biggest comment would be that I think it should be possible to get away with a 16x8 transpose instead of a 16x16 for the 88 [19:33] <BBB> what is mask_mix? [19:34] <ubitux> BBB: for splatting the 2 values of E, I and H [19:34] <BBB> oh right I remember now [19:35] <BBB> I don't think they have to be dq right? it's really just times 8 db 0, times 8 db 1 [19:35] <BBB> (not that it matters) [19:35] <ubitux> yeah sure [19:35] <BBB> I'd add a one-line comment with what you just said [19:35] <BBB> I had to scratch my head there several times because I'm forgetful [19:35] <ubitux> i wasn't sure if i could mix multiple times like that when doing the first time [19:37] <BBB> my context-opts gains a few %, like 2-3% or so [19:37] <BBB> decoding time is now halfway 6.0sec and 6.1sec [19:38] <BBB> (so with your changes that'd be like 5.low) [19:38] <BBB> I'll add a few more changes that I had planned at some point soon [19:56] <ubitux> BBB: any idea how we could make decode_coeffs_b() faster? [21:35] <BBB> ubitux: you can't, it just takes long [21:35] <ubitux> sadness. [21:35] <BBB> ubitux: you can make it 10-20% faster by writing it in assembly, but I assure you it's no fun [21:35] <ubitux> :D [21:35] <BBB> mru did that for arm, where every cycle counts [21:35] <BBB> I never bothered to write a x86 version [21:36] <ubitux> for what codec? [21:36] <BBB> vp8 [21:36] <nevcairiel> disassemble gccs result and start optimizing there? :d [21:36] <ubitux> ok :) [21:36] <BBB> I was planning to move the tx32x32 division out of the loop [21:36] <BBB> that would make a small difference [21:36] <BBB> but other than that I have no ideas on how to optimize it by a big amount, just small 1-2 cycles differences [21:36] <BBB> I can give you my list if you want [21:37] <ubitux> BBB: btw, about the 16x8, i'll need to read 16 "lines" (actually 16 half-lines), i'm not sure the i can simplify the transpose itself much [21:37] <BBB> - coef parsing opts: [21:37] <BBB> o make cache i instead of rc ordered (means having to change nb to be i indexed) [21:37] <BBB> o move division to decode_coeffs outside txblk loop [21:37] <BBB> o merge eob and cnt [21:37] <BBB> o remove tx argument [21:37] <BBB> o do DC outside loop and remove indexing of qmul (?) [21:37] <BBB> o move c on stack [21:37] <BBB> 16 half lines = 16 movhs right? [21:37] <BBB> then the first stage of the transpose is much simpler [21:38] <BBB> it's just punpcklbw [21:38] <BBB> instad of SBUTTERFLY bw [21:38] <BBB> (which is punpckl+hbw [21:38] <BBB> ) [21:38] <ubitux> ah, i see [21:38] <BBB> then after that it's a 8-register transpose [21:38] <ubitux> mmh [21:38] <BBB> same for the reverse [23:12] <ubitux> beastd: so, when are we putting back this awesome logo? http://web.archive.org/web/20050227054954im_/http://ffmpeg.sourceforge.net/… [23:14] <beastd> ubitux: yay \o/ that one is classic :D [23:15] <ubitux> :) [23:15] <beastd> Maybe we should put is up at the next ffmpeg anniversary [23:29] <cbsrobot> ubitux: I thought I send the true peaks before march 2014 [23:53] <ubitux> cbsrobot: oh :) [23:54] <cbsrobot> ubitux: I remember you liked the soundtrack I sent you for testing [23:54] <ubitux> yes [23:54] <cbsrobot> here's the movie that goes with it: https://vimeo.com/43986997 [00:00] --- Sun Jan 26 2014

1 0

ffmpeg.log.20140125
by burek 26 Jan '14

26 Jan '14

[04:39] <fundies> Error: Track 2 with the CodecID 'S_VOBSUB' is missing the "codec private" element and cannot be extracted. How to fix? [05:10] <fundies> relaxed, mkvextract cant extract vobsub from mkv created w/ ffmpeg [05:11] <fundies> mkvextract tracks dvd.mkv 2:vobsub [05:11] <fundies> ffmpeg -fflags genpts -i dvd.mpg -c copy -map 0:1 -map 0:2 -map 0:3 -map 0:4 -map 0:5 dvd.mkv [05:15] <relaxed> fundies: does it work if you remux with mkvmerge instead of ffmpeg? [05:17] <fundies> mmg doesnt copy over the vobsubs at all [05:19] <relaxed> maybe someone in #matroska has an idea [05:52] <smit> where can I find the windows binary for ffserver? it's not in the packages on http://ffmpeg.zeranoe.com/builds/ [06:11] <dantxo> Hello. [06:11] <dantxo> I'm using ffmpeg to stream audio and video form a webcam. [06:12] <dantxo> whenever I reboot my computer the /dev/video* devices get reordered [06:12] <dantxo> so I can't trust that video1 is always the webcam [06:12] <klaxa> can't you create udev rules for that? [06:13] <dantxo> Is there a way to identify the video source by hw id or something? like what I doo with the audio part [06:15] <dantxo> klaxa: thank you [06:25] <bmuk> Hey everyone I'm trying to record my desktop and I'm getting terrible performance (1-4 fps). I'm not really sure what I am doing wrong. ffmpeg -video_size 1024x768 -framerate 25 -f x11grab -i :0.0+100,200 output.flv this is the command I am using [06:26] <bmuk> the x server is using insane amounts of cpu while this runs, but this itself only uses 3% [06:27] <relaxed> dantxo: do you have /proc/videodev ? [06:30] <bmuk> http://pastebin.com/D7kTXaeL [06:30] <bmuk> that is the command followed by the output [06:32] <relaxed> bmuk: add -c:v libx264 -preset:v veryfast after the input [06:34] <bmuk> I'm still getting 3-4 fps [06:34] <relaxed> OS? cpu? [06:35] <bmuk> arch linux, 3.12.8-1. amd fx 6300 [06:35] <dantxo> relaxed, sorry I was away from the computer, there is no videodev under /proc/ [06:38] <relaxed> bmuk: you're using the command directly from the guide, yes? [06:39] <relaxed> which resolution are you actually using on your desktop? [06:39] <relaxed> dantxo: how many /dev/video* device do you have? [06:39] <bmuk> yes I am. My monitor is 1920x1080 [06:40] <relaxed> devices* [06:40] <dantxo> 3, video0, video1 and video2 [06:40] <relaxed> what are the other two? [06:40] <dantxo> video 1 and 2 are two webcams, and I want them to be always the same name under /dev/ [06:40] <relaxed> bmuk: ffmpeg -video_size 1920x1080 -framerate 25 -f x11grab -i :0.0 -c:v libx264 -preset:v veryfast output.mp4 [06:41] <dantxo> I've added a udev rule as suggested but it does not seem to work [06:42] <bmuk> this is giving me 26 fps. Is processor just really slow? [06:43] <dantxo> the rule that I've created for the first webcam is SUBSYSTEM=="video4linux", BUS=="usb", SYSFS{idVendor}=="045e", SYSFS{idProduct}=="0761", NAME="video5" [06:43] <relaxed> bmuk: lowering the "-framerate" should help [06:43] <dantxo> for the second webcam SUBSYSTEM=="video4linux", BUS=="usb", SYSFS{idVendor}=="046d", SYSFS{idProduct}=="082b", NAME="video6" [06:43] <dantxo> they are still named video1 and video2 under /dev/ after reboot though [06:43] <bmuk> thank you relaxed. [06:44] <relaxed> bmuk: You can also scale down the output. Try adding -filter:v scale=w=1280:h=-1 after the input. [06:45] <bmuk> actually, I increased the framerate to 60 and it was much more usabel [06:45] <bmuk> useable* [06:53] <relaxed> dantxo: yeah, I don't know. maybe ask in your OS's channel [07:43] <fundies> so no way to extract vobsubs from a mkv? [08:30] <dantxo> how can I lower the picture quality of a webm stream? I'm over local network and the video is pretty slow, choppend, pretty bad overall [09:25] <__raven_> klaxa: how to detect for (amount of) edges with result for scripts? [09:31] <llogan> dantxo: http://trac.ffmpeg.org/wiki/vpxEncodingGuide [09:40] <skinkitten_> hi everybody [09:55] <goek> Hey I cannot get FFMPEG to connect to live stream could anyone help [09:55] <goek> kk [10:00] <goek> http://pastebin.com/adUrZCQZ [10:00] <goek> fflogger [10:00] <goek> llogan* [10:00] <goek> It hangs at that spot [10:02] <goek> it streams on justin.tv but not livestream oddly enough [10:02] <goek> I think its because the documentation only allowed sometging like rtmp://server/live/stream [10:03] <goek> and FFMPEG doesn't appear to support in URL RTMP authentication [10:03] <goek> or I don't know how rather i'm sure its possible [10:04] <goek> Has anyone here had success in streaming to livestream before? [10:05] <llogan> goek: apparently http://ffmpeg.zeranoe.com/forum/viewtopic.php?f=7&t=657&start=50#p3063 [10:05] Action: llogan blames livestream [10:06] <llogan> goek: i have to go now. try the ffmpeg-user mailing list if you don't figure it out. mayber there are other livestream users [10:09] <goek> ok thanks [10:31] <goek> exit [14:00] <theTroy> Hi, I am trying to simply crop a video with ffmpeg, but it seems to also re-encode the whole video, is there a way to not do that? [14:00] <theTroy> I am using crop filter* [14:01] <saste> theTroy, no [14:01] <JEEB> nope, cropping changes the video [14:02] <JEEB> and thus you have to re-encode [14:02] <JEEB> some formats have limited cropping flags in the header, but the cropping filter most definitely doesn't touch that, and you can't do major cropping with it anyways [14:03] <theTroy> I only want to crop the video to 2/3 of the width, it uses default h264 [14:03] <theTroy> wouldn't want to re-encode it as cubie board will take ages doing that :( [14:03] <JEEB> yeah, not gonna happen without re-encoding [14:04] <theTroy> thats unfortunate, thanks for the help! :) [14:14] <bencc> I'm getting "width not divisible by 2 (547x430)" when trying to convert a video from avi to mp4 [14:15] <bencc> can I ask ffmpeg to choose the closest even number for width and height? [14:16] <saste> bencc, yes, use the scale filter, check examples [14:16] <bencc> saste: thanks. searching [14:20] <bencc> saste: this seems to work: -vf scale='trunc(in_w/2)*2:trunc(in_h/2)*2' [14:20] <bencc> hopfully I'm doing this right [14:21] <saste> bencc, even better you can use the HSUB and VSUB variables, they will work also for other pix fmts than yuv420p [14:22] <bencc> saste: I need mp4 to play on browsers [14:22] <bencc> do HSUB and VSUB work on IE, Chrome, FF... ? [15:07] <lkiesow> saste: About https://trac.ffmpeg.org/ticket/309 [15:07] <lkiesow> At that time your approved of the idea and said that patches are welcome. Noting happened since then. [15:07] <lkiesow> Now I ran into the same problem and wrote a patch for that [15:08] <lkiesow> Do you have any suggestions about what is the best way to contribute that patch to ffmpeg? [15:08] <lkiesow> I found https://github.com/FFmpeg/FFmpeg [15:08] <lkiesow> Is a pull request against that repo a good idea? [15:12] <relaxed> lkiesow: http://www.ffmpeg.org/developer.html [15:14] <lkiesow> Thanks, I will have a look at that [16:17] <DelphiWorld> hi guys [16:18] <DelphiWorld> please would someone give me a CPU aware ffmpeg command for H.264/AAC live streaming? [16:18] <DelphiWorld> i'm using this: [16:18] <DelphiWorld> ffmpeg -re -f mpegts -i udp://@239.100.1.3:1234 -vcodec h264 -b:v 600k -acodec libfdk_aac -b:a 64k -f flv -threads 8 -preset veryfast rtmp://localhost:1935/myapp/rtdoc [16:41] <DelphiWorld> what's the best CPU friendly x264 presets? [17:13] <lkiesow> DeadSix27: There should also be an ultrafast preset for x264 which should need even less CPU power [17:26] <dantxo> Hello again [17:27] <dantxo> I don't really understand what I'm doing wrong [17:27] <dantxo> I'm streaming webm format video from my webcam, with a pretty low quality [17:27] <dantxo> but the video gets stucked, and there is a huge delay [17:27] <dantxo> audio & video are not in sync [17:31] <dantxo> ffserver config: http://pastie.org/8666773 [17:31] <dantxo> ffmpeg streams: http://pastie.org/8666779 [17:32] <dantxo> The mjpeg stream is really smooth without delay [17:32] <dantxo> but the webm stream is terrible (not in video quality, that I dont mind) but with a huge delay and audio out of sync [17:55] <DelphiWorld> ffmpeg -re -f mpegts -i udp://@239.100.1.4:1234 -vcodec h264 -b:v 600k -pix_fmt yuv420p -acodec libfdk_aac -b:a 64k -f flv -r 25 -s 1920x1080 -threads 8 -preset veryfast rtmp://localhost:1935/myapp/yemen [17:55] <DelphiWorld> :P [18:03] <dantxo> DelphiWorld: is that for me? [18:03] <DelphiWorld> dantxo: what was your question? :P [18:04] <dantxo> how to speed up webm stream [18:05] <DelphiWorld> ah i'm not sue [18:05] <DelphiWorld> sure [18:14] <relaxed> dantxo1: look at "ffmpeg -h encoder=libvpx" [18:16] <DelphiWorld> llogan: alive? [18:21] <shevy> got a question - the video with x11grab is kinda blurred. is there a way to have it much sharper? [18:40] <dantxo1> By doing this ffmpeg -f v4l2 -s 320x240 -i /dev/videovx2000 http://localhost:8000/ext.ffm -f v4l2 -s 320x240 -i /dev/videoc170 http://localhost:8000/int.ffm [18:40] <dantxo1> Shouldn't each webcam feed ext.mjpg and int.mjpg streams? I see the image from the same webcam in both streams [19:04] <lkiesow> shevy: Which command did you use? [19:05] <shevy> hmm [19:05] <shevy> ffmpeg -f x11grab -video_size 1024x800 -r 25 -i :0.0 /Depot/Temp/GRABBED.mpg [19:05] <shevy> may be related to .mpg default? [19:12] <lkiesow> I guess so [19:12] <lkiesow> Could you try: ffmpeg -f x11grab -video_size 1024x800 -r 25 -i :0.0 -c:v libx264 -preset ultrafast -qp 0 out.mp4 [19:13] <lkiesow> Though you probably don't want to use that for actual recordings (lossless h264) [19:14] <shevy> yeah a moment [19:15] <shevy> hmm Unrecognized option 'preset'. [19:15] <shevy> Error splitting the argument list: Option not found [19:18] <lkiesow> then just ignore preset for now [19:30] <DelphiWorld> what's -r option do? [19:34] <shevy> DelphiWorld, should be -video_frame_rate [19:34] <shevy> probably sets how many frames there will be in a second [19:35] <DelphiWorld> could be [19:42] <DelphiWorld> shevy: if i give you a live stream, can you ffprobe it and tel me how to produce similar stream? [19:43] <shevy> DelphiWorld but I have no experience with streams [19:44] <DelphiWorld> yeah but is h264 regardless [20:44] <shevy> llogan hmm? [20:44] <llogan> it will show why preset is not found [20:45] <shevy> I see [20:45] <shevy> http://pastie.org/pastes/8667156/text [20:46] <llogan> shevy: you have no libx264 support [20:46] <llogan> no --enable-libx264 [21:03] <shevy> ok, recompiling right now with --enable-libx264 [21:12] <llogan> or you could use rawvideo (huge files), or a lossless encoder other than libx264 (huffyuv, utvideo, ffv1, etc) [big files], and then re-encode that when you are done capturing the screen [21:15] <shevy> it worked now [21:16] <xlinkz0> i'm trying to connect with ffmpeg to a broken rtsp server [21:17] <xlinkz0> however it does not timeout or output anything, it just hangs [21:17] <xlinkz0> this was the whole output : http://codepad.org/i7Q6Qg3P [21:17] <llogan> xlinkz0: your command is missing [21:17] <xlinkz0> sorry [21:18] <xlinkz0> http://codepad.org/m2njeW3g [21:18] <xlinkz0> i'd like to use ffmpeg to check when that server is up [21:18] <xlinkz0> can i make it exit after some time if it doesn't receive anything from the server? [21:19] <llogan> -timeout [21:19] <llogan> http://ffmpeg.org/ffmpeg-protocols.html#rtsp [21:20] <xlinkz0> thank you ! [22:47] <lkiesow> michaelni: Thanks for merging my patch in &and cleaning it up :) [00:00] --- Sun Jan 26 2014

1 0

ffmpeg-devel.log.20140124
by burek 25 Jan '14

25 Jan '14

[00:15] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:78530240715c: avcodec/h264_cabac: Fix use with the checked bitstream-reader [03:08] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:52d2bcc78632: avcodec/libopenjpegenc: Check the return code of av_frame_ref() [03:08] <BBB> michaelni: is this the latest version of valgrind? [03:09] <michaelni> BBB dunno, its ubitux box [03:09] <BBB> hm... [03:09] <BBB> ubitux: ^^ [03:12] <BBB> michaelni: because I can't reproduce locally, fate-valgrind is clean for me [03:14] <michaelni> i can reproduce with 3.8.1 and 3.7.0 (these are the 2 i have installed locally it seems) [03:17] <BBB> just valgrind ffmpeg -threads 1 -i bla -f framemd5 -? [03:17] <BBB> valgrind --quiet --dsymutil=yes /Users/ronaldbultje/Projects/ffmpeg/x86-64-gpl/ffmpeg -nostats -cpuflags all -threads 1 -thread_type frame+slice -i /Users/ronaldbultje/Movies/fate-suite-ff/vp9-test-vectors/vp90-2-00-quantizer-46.webm -flags +bitexact -f framemd5 - [03:17] <BBB> that's what it executes for me [03:18] <BBB> oh it stumps it to stderr, duh [03:22] <michaelni> i was testing with whatever ./configure --valgrind=valgrind used and also with just valgrind ./ffmpeg ... [03:31] <BBB> fuck fuck [03:32] <BBB> was the emu_edge patch merged? [03:50] <michaelni> you mean 91e00c4a785d3f4cd44f1044d3e7b1b34fee8b5c ? [03:51] <michaelni> the commit message says its not merged [03:51] <michaelni> similar and conflicting changes where already in IIRC and IIRC someone said to skip vp9 stuff [03:52] <michaelni> but maybe i misremember [03:57] <michaelni> hmm actually this one doesnt look very much vp9 related [04:09] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:a26e9c1040af: avcodec/mjpegenc: Use av_frame_clone() instead of av_frame_ref() [04:09] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:9b68538fddf6: avcodec/libopenjpegenc: Replace av_frame_alloc() and av_frame_ref() by av_frame_clone() [04:09] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:40c218c60dce: avcodec/vc1: fix type of tmp [04:22] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:41003da94a59: avfilter/avfilter: fix use of uninitialized pointer [04:22] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:4b8c599e8490: avformat/nutenc: fix missing break in switch [04:41] <BBB> hm it's itxfm-related as ubitux said [04:41] <BBB> I'm still suspecting a bad interaction between valgrind and asm [04:41] <BBB> but I'll debug which part exactly [04:41] <BBB> I can (for now) reproduce with 00-q-64 block 0 (tx32x32 y) where the input is defined but the output isn't with asm, but not c [04:41] <BBB> it uses a sub16x16 tx [04:42] <BBB> it also reproduces with a full 32x32 tx (asm again) [04:42] <BBB> but sub8x8 (corrupt output) doesn't reproduce it [04:42] <BBB> very odd [04:42] <BBB> I'll look further, but for now, sleep [07:22] <ubitux> BBB: valgrind 3.9.0 [11:56] <saste> michaelni, may be possible to reject mails sent to @mplayerhq.hu? [12:03] <michaelni> saste, you mean ffmpeg-dev(a)mplayerhq.hu or something else ? the mplayerhq domain is used by mplayer [12:04] <michaelni> and its probably possible but i dunno how [12:07] <saste> michaelni, some users are still using the old address (after three years after it has been changed), and I don't want to add custom rules to my .procmail file [12:07] <nevcairiel> scream at them, loudly [12:07] <saste> michaelni, yes ffmpeg-devel(a)mplayerhq.hu, ffmpeg-user and libav-user [12:08] <saste> nevcairiel, that doesn't work, better to make them aware of the change in the hard way [12:10] <nevcairiel> well there is two ways the setup went down originally; either someone was lazy and just aliased the two domains so they get the same mboxes, or they are intentionally available on both ... the second case is easy to fix, the first needs someone to become not-so-lazy :p [12:12] <michaelni> the mail setup was done by arpi, i dont even know where that domain stuff is configured without investigating [12:12] <michaelni> saste, also you dont need a custom rule just to fix your rules [12:13] <michaelni> the List-Id points to ffmpeg.org even for a To: mplayerhq.hu mail [12:22] <cone-883> ffmpeg.git 03Stefano Sabatini 07master:433b153b681d: doc/ffmpeg: reference time syntax sections in ffmpeg-utils for itsoffset and timestamp options [12:22] <cone-883> ffmpeg.git 03Stefano Sabatini 07master:ca57659440fc: examples/filtering_audio,video: do not call avcodec_register_all() [12:38] <cone-883> ffmpeg.git 03Ronald S. Bultje 07master:bd01412313c7: vp9: fix mvref finding to adhere to bug in libvpx. [12:38] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:367245857399: Merge commit 'bd01412313c728400f1fc5448ede0ad8b51da0d1' [13:03] <cone-883> ffmpeg.git 03Guillaume Martres 07master:50866c8d95bf: vp9: fix bugs in updating coef probabilities with parallelmode=1 [13:03] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:a0e9dfb5ae5e: Merge remote-tracking branch 'qatar/master' [13:38] <saste> why the packet PTS can be different from the decoder PTS? [13:38] <saste> but then I see best_effort_ts is equal to the packet PTS [13:39] <saste> or in other words, what's the best_effort thing about? [13:39] <av500> random guessing I guess [13:39] <saste> or in other words, why the decoder changes the input packet PTS? [13:42] <kierank> 12:39 PM <"av500> random guessing I guess [13:44] <saste> kierank, not useful answer [13:44] <kierank> but it's the actual answer [13:44] <saste> the problem is that there is any guarantee that input PTS == output PTS, not even with copyts [13:44] <saste> muxers, decoders, and ffmpeg code cooperate to make it completely impredictable [13:53] <michaelni> saste, if i guess that with packet pts you mean demuxer pts, then your question "why packet and decoder pts can differ" has the awnser "because both the container and the codec bitstream can store pts and they can differ" [13:55] <saste> michaelni, ok [13:55] <saste> why is that happening? [13:55] <michaelni> and if they differ then thats probably violating some spec [13:55] <saste> it is happening with a sample I have [14:00] <av500> what kind of file? [14:01] <saste> av500, H.264 in MPEG-TS [14:32] <Keestu> dear all, it is possible to send custom command to rtsp server through ffmpeg ? [14:37] <j-b> https://trac.videolan.org/vlc/ticket/10451#comment:5 any idea about this backtrace? [16:59] <saste> michaelni, problem was due to rounding errors from converting timestamps [16:59] <saste> otoh I cannot set the timebase in the encoder [16:59] <saste> (the option has no enabled flags, so it is documented but can't be set in the commandline) [17:00] <saste> OTOH if I enable it then libx264 is confused, since it sets fsp = 1 /tb [17:00] <saste> fps [17:16] <kierank> well...I moved to ffmpeg [17:16] <kierank> was quite painless apart from --enable-avresample [17:28] <Compn> kierank : welcome, i guess ? :) [17:28] <Compn> whyd you switch over ? [17:28] <kierank> Compn: filters, though I have to fix bugs in those [17:29] <saste> kierank, i'minterested [17:29] <saste> (about bugs in filters) [17:29] <kierank> not really a bug but vf_overlay only supports 420 and 444 but it looks like a simple fix [17:30] <ubitux> how naive [17:30] <Compn> how many filters have to add colorspace to? :) [17:30] <kierank> ubitux: naive to be a simple fix? [17:30] Action: Compn guesses 10bit support [17:30] <ubitux> kierank: i remember some little crazyness about such thing in overlay [17:31] <kierank> even if i'm overlaying *everything* in 422 mode? [17:31] <ubitux> ask saste about the details i don't know :) [17:36] <saste> overlay is using drawutils, so we need to extend that and all user filters will be enabled to support 10 bits support (including pad) [17:36] <kierank> I don't need 10-bit [17:36] <kierank> I need 422 in overlay [17:36] <kierank> but you already support 420 and 444 so i don't see why 422 is special [17:38] <saste> kierank, so you need to extend format mode [17:38] <kierank> yeah [17:49] <cbsrobot> subtitling directly in 10-bit would be a nice enhancement [17:49] <cbsrobot> or generally in > 8-bit [17:49] <nevcairiel> blending in higher bitdepths isnt really hard [17:49] <nevcairiel> but is it worth it? :) [17:53] <cbsrobot> I have a lot of 10-bit clips and it would be awesome to add hardsubs without degrading to 8-bit [17:54] <saste> kierank, untested patch sent, please report [18:03] <ubitux> saste: we should make more tests with odd sizes :) [18:03] <saste> ubitux, why? [18:04] <ubitux> because it's easy to break without noticing [18:04] <saste> if you break and you notice is no fun anymore [18:05] <ubitux> :) [18:38] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:746350ea0f7b: avcodec/mpeg12dec: Make mpeg2_fast_decode_block_intra() more robust by breaking out on invalid vlcs [18:38] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:0a59055167ee: avcodec/mpeg12dec: check for overread in mpeg1_fast_decode_block_inter() [18:38] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:0c8e5fb21183: avcodec/mpeg12dec: Optimize mpeg1_decode_block_intra() [18:38] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:d82eccea2bf9: avcodec/mpeg12dec: check block index in mpeg2_fast_decode_block_non_intra() [18:39] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:76b5e99ce9c1: avcodec/mpeg12dec: Check for overread in mpeg_decode_slice() [18:39] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:5f54756f7e4f: avcodec/mpeg12dec: Disable the checked bitstream reader [18:39] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:934bb11ad79a: avcodec/mpeg12dec: fix mis-indented line [18:47] <michaelni> j-b, crash should be fixed, a more complete fix is on the ML [18:48] <j-b> michaelni: amazing [18:49] <j-b> but no more --fast for mpeg12? [18:52] <kierank> hmm maybe i didn't need saste's patfch after all [18:54] <Compn> nevcairiel : its always worth it [18:54] <Compn> more bits! [18:58] <michaelni> j-b, its still there and can still with some "luck" crash, with the patch from the ML its renamed to fast_crash which should ensure that noone can misunderstand what the option does [18:59] <j-b> michaelni: but, but, but, it makes it faster!! :) [19:00] <michaelni> well thats why i tried to make it more robust [19:00] <michaelni> but ive no idea how successfull i was [19:01] <michaelni> also mpeg1 / 2 should be faster now with and without "fast" than before [19:02] <Compn> awesome [19:02] <j-b> nice [19:02] <Compn> michaelni sped up mpeg1/2 :) [19:03] <j-b> Compn: don't joke. It's 70% of our libavcodec related crashes on iOS [19:04] <Compn> i'm happy its faster [19:04] <Compn> but you have that many crashes on ios ? :\ [19:04] <Compn> i thought ffmpeg was ROBUST! :P [19:04] <j-b> haha [19:04] <j-b> just fun [20:35] <jnvsor> Quick question about AVRationals, what's the preferred way to reduce it to a framerate? [20:36] <jnvsor> av_q2d()? [20:44] <michaelni> av_q2d() is the obvious way yes [20:45] <michaelni> if instead you want a simpler rational as output theres av_reduce() [20:48] <jnvsor> I'm trying to port timecode.c from int to rational, quite interesting [20:49] <jnvsor> Looks like there's a LOT of stuff relying on that int. Ew [21:14] <cone-883> ffmpeg.git 03João Bernardo 07master:290326711b1a: avutil/opt: Better print representation of number limits [21:17] <jnvsor> Right, so I've got it compiling using doubles from the drawtext filter timecode instead of ints - is there some sort of test suite I should run to make sure I haven't broken anything else? [22:18] <Compn> jnvsor : fate test ? [22:21] <jnvsor> Compn: Sure. Any idea how to set that up? [22:21] <jnvsor> Compn: nvm, found the web page [22:22] Action: BBB pokes ubitux [22:24] <Compn> make fate [22:24] <Compn> after pulling the fate samples [22:24] <Compn> or something, i am not sure :) [22:29] <cone-883> ffmpeg.git 03Serhii Marchuk 07master:f8051bd31a23: mpegts muxer: Change the default subtitle language to "und" [22:29] <cone-883> ffmpeg.git 03Serhii Marchuk 07master:2ebee19e1ed2: mpegts muxer: restore PMT table of DVB teletext from extradata [22:29] <cone-883> ffmpeg.git 03Serhii Marchuk 07master:1d0738505386: mpegts demuxer: store PMT values of DVB teletext to extradata [22:31] <jnvsor> So if I have a fate error, how do I debug it? I'm just seeing a bunch of FFmpeg runs in the .err file and I don't know which is causing the issue [22:34] <BBB> make fate will stop if there's an error [22:34] <ubitux> BBB: hey. [22:35] <ubitux> lpf this week end [22:35] <ubitux> i'm half dead from my week + lack of previous week end; will get some sleep tonight and will be ready again tomorrow :) [22:37] <BBB> no, just wanted to ask for review of the other patches [22:37] <BBB> lpf just whenever you have time and energy, didn't mean to push you on that [22:38] <ubitux> BBB: i saw the SWAPs update, the 3 patchs look ok [22:38] <ubitux> unless you meant another patch(es) i have missed? [22:40] <BBB> no, just that one [22:40] <BBB> did you ok them? maybe I missed it [22:41] <ubitux> no i didn't reply, sorry :) [23:09] <BBB> aha [23:09] <BBB> ok [23:09] <BBB> so [23:09] <BBB> valgrind error [23:09] <BBB> I have a pretty solid hypothesis [23:10] <BBB> so you guys see how we do stack arithmetic to get the correct location on the stack to place the idct? [23:10] <BBB> so we do sub rsp, 2048 [23:11] <BBB> for (i=0;i<4;i++) { do idct32x8_1d and place in rsp[0-512]; add rsp, 512 } sub rsp, 2048 [23:11] <BBB> and then the second idct32x8_1d x4 [23:11] <BBB> that's the error [23:11] <BBB> algrind thinks it's undefined after add rsp, x [23:13] <BBB> (it being the memory under rsp), so after add rsp, 512, valgrind thinks the first 512 bytes become undefined agian [23:13] <jnvsor> Is the last run in a fate .err file the one that caused the error or could it be any of them? [23:14] <BBB> I'm not sure that's defined in the C standard or anywhere [23:14] <BBB> jnvsor: last [23:15] <BBB> ubitux: any opinions on the above? or someone wanna read the c spec or the platform specs on what any of this means? [23:15] <BBB> I don't think I've ever come across something like that [23:17] <jnvsor> So I've found the error - a nice nondescript "Input/output error". Any way to track it down a bit further? [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:86bee7984e30: AVFormatContext: add metadata_header_padding field [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:72eeb18468b0: lavf/flacenc: use metadata_header_padding [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:c14b011a97f3: lavf/flacenc: fix comment after previous change [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:0de03fd6a1f1: lavf/id3v2enc: use metadata_header_padding [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:67270ccd3af9: lavf/id3v2enc: update comment about minimum padding [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:fa20babb4661: lavf/avienc: use metadata_header_padding [23:19] <cone-883> ffmpeg.git 03James Darnley 07master:2efdccac87cb: lavf/avienc: cosmetic indent [23:36] <jnvsor> Can I get fate to run the failing test in gdb or something? [23:37] <mark4o> BBB: if a signal handler is run then it will overwrite anything at the stack pointer [23:37] <nevcairiel> you can use "make V=1 fate-test-name", it'll show you the command used, and you can use that command line in gdb [23:38] <BBB> jnvsor: make V=1 fate-name-of-test [23:38] <BBB> oh late sorry [23:39] <BBB> hm I'll add a stackpointer to resolve this [23:39] <BBB> kind of annoying [23:39] <BBB> but ok [00:00] --- Sat Jan 25 2014

1 0

ffmpeg.log.20140124
by burek 25 Jan '14

25 Jan '14

[00:05] <fundies> can I make ffmpeg to copy all the tracks? [00:12] <fundies> lahttp://pastie.org/8661919 [00:13] <sacarasc> You might have to remove stream 0. [00:14] <fundies> how? [00:15] <sacarasc> Use -map, I guess... [00:15] <fundies> i tried that n it errored [01:25] <Aquaguy> Hello [01:26] <Aquaguy> I'm using ffmpeg to stream video from a webcam but when I launch it I get a notice saying that the program is deprecated. [02:06] <NoZero> Help me please? [02:07] <sacarasc> NoZero: What with? [02:07] <NoZero> I have tried at least 10 times to write c++/cli program with FFMPEG includes [02:07] <NoZero> get all kinds of external link errors at best [02:08] <NoZero> there is no libswresample/swresample.h [02:08] <NoZero> folder is empty [02:09] <NoZero> it has empty subfolders ARM, X86 [02:09] <NoZero> wtf? [02:09] <NoZero> I have to download 38,000 header files piece meal? [02:10] <NoZero> there is no help docs anywhere [02:10] <NoZero> turorials are 10 years old [02:10] <NoZero> Is there any hope? [02:10] <sacarasc> Did you install the libraries yet? [02:10] <NoZero> yes as far as I know [02:11] <NoZero> you mean DLLs? [02:18] <NoZero> Directory of C:\Program Files (x86)\Microsoft Visual Studio 9.0\VC\include [02:18] <NoZero> 01/23/2014 07:33 PM <DIR> libavcodec [02:18] <NoZero> 01/16/2014 08:27 AM <DIR> libavdevice [02:18] <NoZero> 01/16/2014 08:28 AM <DIR> libavfilter [02:18] <NoZero> 01/23/2014 07:35 PM <DIR> libavformat [02:18] <NoZero> 01/16/2014 08:29 AM <DIR> libavresample [02:18] <NoZero> 01/23/2014 07:34 PM <DIR> libavutil [02:18] <NoZero> 01/23/2014 07:35 PM <DIR> libswscale [02:19] <Aquaguy> Hello again, I'm trying to stream audio & video from my webcam using ffmpeg, but i get an error saying "Unknown input format: 'alsa'" this is the command I'm using: ffmepg -f v4l2 -s 640x480 -i /dev/video1 -f alsa -ac 1 -i default:CARD=VX2000 http://localhost:8000/exterior.ffm [02:19] <Aquaguy> What's wrong? Thank you for your help in advanced [02:32] <Aquaguy> anyone knows why I get that alsa error? [02:33] <sacarasc> Aquaguy: What version of ffmpeg are you using? [02:35] <Aquaguy> sacarasc: N-60111-g094516a [02:36] <sacarasc> You're using a static build? [02:36] <Aquaguy> I built it from source, cloned the git repository [02:37] <sacarasc> Did you have alsa libs/dev files and stuff installed before compiling? [02:38] <Aquaguy> I don't know, what I know is that alsamixer is installed and working. [02:38] <Aquaguy> So I guess it was installed before building ffmpeg [02:38] <Aquaguy> surew [02:41] <Aquaguy> klaxa: http://pastie.org/8662197 [02:42] <klaxa> you compiled it from source? can you pastebin your config.log? [02:44] <Aquaguy> http://pastebin.com/neA79SKy [02:45] <klaxa> >/tmp/ffconf.zZOQTLNG.c:1:28: fatal error: alsa/asoundlib.h: No such file or directory [02:45] <klaxa> install the asoundlib development packages, configure, build [02:45] <klaxa> Aquaguy: ^ [02:46] <klaxa> should be libasound2-dev [02:47] <Aquaguy> no package with that name [02:47] <Aquaguy> sorry, my bad [02:47] <Aquaguy> I was missing the a [03:04] <klaxa> Aquaguy, any news? [03:05] <Aquaguy> klaxa: just finished building and installing [03:05] <klaxa> kk [03:05] <Aquaguy> I'm getting stats so I guess it's working!! thanks! [03:06] <klaxa> nice :) [04:27] <Aquaguy> Hello once again. [04:27] <Aquaguy> The stream seems to be working as I can access it from another computer [04:27] <Aquaguy> but the video is all black [04:27] <Aquaguy> I'm streaming both in mjpg and mpg [04:28] <Aquaguy> and both look black. [04:29] <NoZero> I am trying to get FFMPEG support for VC++/cli any suggestions? [04:31] <NoZero> Is it only for 'nix? [04:32] <NoZero> Trying to make a simple Media player in c++/cli [04:33] <NoZero> :\ [04:42] <NoZero> Any WinDoze FFMPEG helpers here? [04:43] <NoZero> Know of any tutorials that really work? [05:06] <NoZero> hi [05:09] <NoZero> C:. [05:09] <NoZero> ____libavcodec [05:09] <NoZero> ³ ____alpha [05:09] <NoZero> ³ ____arm [05:09] <NoZero> ³ ____avr32 [05:09] <NoZero> ³ ____bfin [05:09] <NoZero> ³ ____mips [05:12] <NoZero> anyone here? [05:12] <NoZero> Hello... ello... llo... lo... o... [05:13] Action: NoZero pours himself a MOLSON ICE [05:13] <NoZero> My Windows hasn't crashed for 1day 13hrs 11mins 8secs [05:16] <NoZero> Anyone else use Windows? [05:22] <thebombzen> NoZero: if nobody answers it's because they're busy or afk. people aren't ignoring you such that asking again will get an answer this time [05:23] <thebombzen> AFAIK FFmpeg does not compile in MSVC++. I would recommend building the headers and the DLLs/LIBs with MinGW/MSYS. [05:23] <thebombzen> Then you can add it as an external library. [05:24] <thebombzen> Or you could download prebuilt Shared builds from http://ffmpeg.zeranoe.com/builds/ [05:28] <NoZero> I already downloaded the builds [05:28] <NoZero> MinGW is imperative? [05:28] <NoZero> dadgumm [05:29] <NoZero> Isn't there a way to use MinGW compiler in linker? [05:31] <NoZero> OK you already said it doesn't compile. [05:31] <NoZero> Thanks, heheh I kinda already knew that but I didn't want to believe it. [05:32] <NoZero> figure after several attempts and failing. Thought I was doing something wrong. [05:33] <NoZero> MinGW is such a clumsy interface. [05:43] <NoZero> What about QT creator? [10:22] <slavko> hello, i need to package a proprietary audio/video feed to feed ffmpeg with, can you reccomend a simple muxer? [10:22] <slavko> I'm feeding ffmpeg through a pipe with video and it works, now i need to add sound [10:22] <slavko> Thanks! [10:29] <Mavrik> slavko, can't you just pipe both through named pipes_ [10:29] <Mavrik> ? [10:40] <slavko> I dont seem to know how to use them [10:40] <slavko> thanks for the answer though [10:40] <slavko> they fill up? I'm afraid they will malfunction, the stream is live [10:40] <slavko> also, i suspect that ffmpeg sets their size to 4k [10:40] <slavko> do you have any advice about this? [10:41] <slavko> linux btw [10:41] <Mavrik> about what exactly? [10:41] <Mavrik> if your ffmpeg can't keep up with the live stream that's not a piping problem [11:01] <slavko> I've tried piping through a named pipe, but the pipe is limited to 4k, i have 32k frames [11:53] <__raven> how to detect wrong focus, darkness and dropouts? [14:11] <dh1jc> Hey guys, hope you can help me. [14:11] <dh1jc> I'm very experienced on ffmpeg, but currently a bit stuck on OPUS encoding. I can't change the current sampling rate. [14:12] <dh1jc> I would like to have a opus sampling of 44.100 or even less 22.050 but ffmpeg only allows 48k [14:12] <dh1jc> any idea? [14:13] <dh1jc> [my current ffmepg version is N-59852-g785dc14 on windows] [14:14] <dh1jc> the source is in 44100Hz [14:21] <dh1jc> sri ... got a disc [14:23] <dh1jc> http://pastebin.com/F77F7RCf [here the pastbin ..] [14:27] <saste> dh1jc, so what's the problem with -ar? [14:28] <dh1jc> ffmpeg just ignores it... or better gives me an error (second) [14:28] <dh1jc> Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width or height [14:29] <dh1jc> So for opus I can't set it to anything else than 48k ... :/ for all other formats it does work [14:29] <dh1jc> I pasted on pastbin the running one. [14:30] <saste> so are you ashamed of posting the one which has issues instead? [14:30] <dh1jc> http://pastebin.com/r37u5cVn [14:31] <dh1jc> no... just didn't ;) done now. [14:31] <saste> Specified sample rate 44100 is not supported [14:31] <dh1jc> afk for a minute (meeting... maybe you have an idea...) [14:31] <dh1jc> no. only 48k but libopus states a lot more than only that. [14:31] <dh1jc> so not even 11025, 22050, 44100... or 8000 [14:32] <saste> dh1jc, 48000, 24000, 16000, 12000, 8000, 0, [14:33] <saste> these are the supported rates, error message could be improved to list the supported ones [14:34] <saste> also try: ffmpeg -help encoder=libopus [15:08] <dh1jc> ok, will have a look there thank you. [15:09] <dh1jc> @saste ah ok... only 48k 24k 16k... any reason why not hte usual ones?! [15:10] <saste> dh1jc, complain with the opus designers [15:11] <dh1jc> hehe... from thier documentation: Sampling rates from 8 to 48 kHz [15:11] <relaxed> dh1jc: you can compile libopus with "custom modes" that will enable 44100 [15:11] <relaxed> but I'm not sure how wise that would be. [15:11] <dh1jc> ah ok, maybe we can ask the ffmpeg-dev crew to do sow. [15:11] <dh1jc> thanks for all your help [15:53] <NoZero> Hello [15:54] <NoZero> I saw this: "4.2.1 Linking to FFmpeg with Microsoft Visual C++" at http://www.ffmpeg.org/platform.html#Linking-to-FFmpeg-with-Microsoft-Visual… [15:55] <NoZero> Does this mean I can compile code using FFMPEG includes developed in the VC++ environment? [15:55] <NoZero> or am I wasting time trying? [16:01] <JEEB> NoZero, yes -- you can link FFmpeg libraries built with both mingw-w64 and MSVC with MSVC [16:03] <JEEB> the way of doing it with mingw-w64 has been around for ages [16:14] <juke_> what is trailing option on ffmpeg -i "rtsp://mafreebox.freebox.fr/fbxtv_pub/stream?namespace=1&service=372" -vf select='gt(scene\,0.2)' -vf scale=320:-1 -vsync 0 -f image2 image2%3d.png -qscale 1 [16:14] <juke_> ? [19:27] <Stepps> Hi there [19:27] <Stepps> I am trying to use ffserver or ffmpeg to rebroadcast a local http stream to rtsp [19:28] <Stepps> I am not getting very far with this [19:30] <Stepps> The input stream comes from zoneminder, I haven't been able to get ffserver to accept this stream [19:34] <Stepps> I can open the stream in a browser or vlc, so I know it works [19:34] <Stepps> this is the url zoneminder is serving [19:34] <Stepps> http://10.111.1.3/cgi-bin/nph-zms?mode=jpeg&monitor=1&scale=75&maxfps=5&buf… [19:39] <klaxa> Stepps, i won't be a great help with ffserver, but what have you tried so far? [19:49] <Stepps> wel,, I have just modified the default conf file to take the stream [19:49] <Stepps> but it can't open it [19:49] <Stepps> I'll nopaste that part of the config file. [19:49] <Stepps> If I could see the config file of a working rebroadcast with ffserver, I am sure I could figure it out [19:52] <Mavrik> um, you do know you need ffmpeg to read your stream [19:52] <Stepps> doh [19:52] <Stepps> I have read something in the config file, but I wasn't sure it was my case [19:54] <Stepps> so, I see a line saying to use ffmpeg to read the stream, does it have to be a different instance? [19:54] <Stepps> Or will it be enough to edit the line I have in the config [19:54] <Stepps> ? [20:08] <klaxa> it will be a separate instance using the output http://<ip>:<port>/some/path/ffm1.ffm or something similar [21:19] <__raven> how to detect wrong focus, darkness and dropouts? [21:55] <llogan> __raven: you'd have to code those features. [21:55] <llogan> AFAIK [21:56] <llogan> what i mean is there is no "detectfocus", etc type of existing filter (other than blackdetect/blackframe but those probably won't fit your needs) [21:57] <__raven> llogan: i did not program anything yet how to start that? and any chance to detect such things by kind of side channel effects such as bitrate changes/file size? [21:58] <llogan> i don't know. you've asked about this at least ~5 times. [21:59] <__raven> yes and i still hope to find anyone who knows every settings to increase the bitrate significantly for example for edges compared to no edges in a fuzzy image [22:02] <llogan> the problem is that nobody really knows what you are asking or what your are trying to do, how it relates to ffmpeg, and also how do you know if something is too dark or whtever in the first place? it's subjective. [22:07] <__raven> llogan: i for example need an overview about any vbr setting to use this to give the "side channel method" a chance [22:08] <__raven> for example if an image is not fuzzy there will be more edges and theese edges should be encoded as most as possible to significantly increase the bitrate and the final filesize - thats one kind of suggestion [22:09] <klaxa> can't you do some edge detect and weight it against a black frame? [22:11] <klaxa> or just check for edges in general [22:11] <klaxa> the problem is defined vague enough as it is [22:11] <klaxa> detecting those things by bitrates is not really feasible i think [22:47] <niso_> hi,newbie question please:) - if i have two movies (one is compressed with xvid and the second with avc which i think its x264?) , that qualitty of the movie is Determined by the resolution and not by the video kbps right? [00:00] --- Sat Jan 25 2014

1 0

ffmpeg-devel.log.20140123
by burek 24 Jan '14

24 Jan '14

[00:30] <gizmore> i am on gentoo and i get: "Option pixel_format not found" [00:30] <gizmore> i am of course asking for a pointer or hint :) thank you [00:57] <cone-800> ffmpeg.git 03Stefano Sabatini 07master:c92d2f98db68: examples/muxing: flush encoders at the end [01:08] <cone-800> ffmpeg.git 03Stefano Sabatini 07master:35fe88bb5169: examples/muxing: reindent after previous commit [01:48] <michaelni> gizmore, what are you trying to do and with what ffmpeg version ? [01:48] <gizmore> michaelni: [01:49] <gizmore> recordmydesktop --no-sound [01:49] <gizmore> ffmpeg -pix_fmt rgb24 -i out.ogv out.gif [01:50] <michaelni> gizmore, ffmpeg -i out.ogv -pix_fmt rgb24 out.gif [01:51] <michaelni> the pix_fmt sets the output codecs format not the input ... [01:51] <gizmore> http://pastebin.com/8QGxLWf4 [01:53] <michaelni> gizmore, the pix_fmt ... is before the "-i ..." it should be after it [01:53] <gizmore> geeeez [01:53] <gizmore> thank you so very very much [01:53] <gizmore> somehow i managed to mess that up twice :) [02:03] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:622d46300037: avcodec/bitstream: assert that no integer overflow happened when writing codes in build_table() [02:04] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:eaacfc7dd1ea: avcodec/lossless_videodsp: Pass AVCodecContext to init [02:04] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:e6d1c66d742d: avcodec/x86/lossless_videodsp: disable median optimizations for 16bps [03:20] <Nicias> Hello. I have an idea for a filter, and I'm looking to get started on writing it. Any suggestions for where I should look for references? [03:27] <Compn> what kind of info do you want? [03:27] <Compn> and what kind of filter ? [03:27] <Compn> let me see if i can dig up docs [03:27] <Nicias> The filter will just modify the PTS's and timebase of a video stream. [03:28] <Nicias> to smooth out judder. [03:28] <Compn> http://wiki.multimedia.cx/index.php?title=FFmpeg_filter_howto [03:29] <Compn> http://trac.ffmpeg.org/wiki/FilteringGuide [03:29] <Compn> theres -vf setpts which might do what you want already [03:29] <Compn> but i dunno [03:29] <Nicias> no, it doesn't. thanks. [03:30] <Nicias> I'll look into those doc's. [03:30] <Nicias> I think this might be a significant step up from my previous coding experience. [03:32] <Nicias> If I have specific questions, or a first draft, where would be an appropriate place to show it to get feedback? [03:37] <Compn> mailing list is best, but you can also ask here [03:37] <Nicias> ok. thanks. [03:37] <Nicias> I'll read those docs now. [03:40] <Nicias> I do have one question. [03:41] <Nicias> if I am messing with the timebase and PTS of the video stream do I have to worry about the timebase and PTS's of the other streams? [03:50] <Compn> Nicias : probably , depending on what you are doing [03:50] <Compn> i mean if you make the video 2x slower, the audio will play regular speed :P [03:50] <Nicias> for example. [03:55] <Nicias> if I had PTS's of 1, 2, 3, 5.. and a timebase of 1/30, I would output 5, 10, 15, 20 with a timebase of 120. [03:55] <Nicias> so same speed overall, but new timebase. [03:56] <Nicias> um 1/120. [03:56] <Compn> possibly then. i swear theres already a way to do it but who knows :D [03:57] <Nicias> Well, I can't find one that works for mixed frame rate sources. [03:58] <Nicias> pullup produces the judder above, and if you just raise the fps, you get frames like AAAABBBBCCCCDDDDDDDDEEEEFFFF...... [03:59] <Nicias> and decimate runs on the whole file. [04:10] <Nicias> @Compn : I'll see what I can put together and then send it to the list. [04:11] <Nicias> should I get it all worked up as a patch if I'm not quite sure it's right or should I post it just in a message? [04:14] <Compn> devs like looking at patches [04:15] <Compn> it doesnt have to be a git send email, just a simple diff -u [04:15] <Compn> even if its incorrect, then they can correct it :)\ [04:15] <Nicias> ok. Thanks. You've been very helpfull. [04:15] <Compn> no problem [04:15] <Nicias> have a good day. [04:15] <Nicias> bye [10:38] <durandal_1707> what you think about merging libavdevice into libavformat? [11:10] Action: ubitux has no opinion [12:31] <BBB> durandal_1707: it very much tells me libav isn't about features, speed or fixing bugs - it's just about moving code around in some way, either through reindenting, splitting or merging files and modules </troll> [12:41] <saste> there is a weird interaction between -copyts and -itsoffset [12:42] <saste> if -copyts is not specified, -itsoffset is ignored altogether, since the timestamp are reset to 0 [12:42] <saste> michaelni, is that intended? [12:43] <saste> also, there is any way to apply an offset to the *output* timestamps? [13:07] <relaxed> saste: there was a guy in #ffmpeg saying having issues with -itsoffset not working unless he paired it with -async 1. Could it be related? [13:08] <relaxed> he was concatenating streams, while trying to offset the audio on the second half. [13:11] <saste> relaxed, good luck at understanding how the async thing works [14:34] <cone-883> ffmpeg.git 03Matt Oliver 07master:306e1ac012d0: avformat/tls: Fix windows build with openSSL enabled. [14:48] <cone-883> ffmpeg.git 03Eli Kara 07master:da25a6573f9c: avformat/http: Properly handle cookies that specify sub-domain where the URL. [15:43] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:8456bd2c0f3b: avformat/matroskaenc: wrap V_QUICKTIME codec private in something that looks like its part of quicktime stsd [15:46] <Keestu> hi all, i have built ffmpeg for android, and i see rtsp protocol is not available. Not sure how i resolve this issue?. i checked like --enable-protocols. A question, In the libavformat/allformats.c, i done see any line related to REGISTER_PROTOCOL(...) for rtsp, but for rtp?. kindly forgive me if i am asking very stupid question. [15:49] <nevcairiel> rtsp is a demuxer, not a protocol in ffmpeg terms [15:50] <Daemon404> 'demuxer' [16:03] <Keestu> it is interesting that to explore the ffmpeg code. if someone helps me would be gr8. What i understood is av_register_all() prepares the handlers for all the modules that are asked from the configuration, and which is defined in allformats.c, if(CONFIG_##X##_MUXER) av_register_output_format(&ff_##x##_muxer); where this CONFIG__##x##_MUXER is checked from ? ./configure ? [16:08] <nevcairiel> config.h [16:08] <nevcairiel> which is written by configure [16:15] <Keestu> nevcairiel, ok. here is my question. in config.h -> CONFIG_RTSP_MUXER, AND CONFIG_RTSP_PROTOCOL is available, consider this, and getting back to av_register_all (), CONFIG_RTSP_MUXER is used, but not CONFIG_RTSP_PROTOCOL. [16:16] <Keestu> nevcairiel, what i am trying is how avformat_open_input () works for rtsp:// protocol. [16:18] <nevcairiel> there is no rtsp protocol [16:18] <nevcairiel> i have never seen CONFIG_RTSP_PROTOCOL [16:20] <Keestu> nevcairiel, ok. can u hint me for avformat_open_input(&context, "rtsp://<>:<>", NULL, NULL) fails me saying that no protocol available. not sure where i have to look at. spending whole day.:( [16:20] <saste> michaelni, should i push the cosmetical ffmpeg patch as well? [16:21] <nevcairiel> build it without disabling anything, and it will work [16:21] <nevcairiel> works for me. [16:21] <Keestu> nevcairiel, basically i m porting with android. [16:21] <nevcairiel> i dont care :) [16:21] <Keestu> nevcairiel, :) [16:21] <nevcairiel> and it doesnt matter [16:21] <Keestu> true. [16:21] <nevcairiel> build it with a simple ./configure without disabling any components [16:24] <Keestu> nevcairiel, interesting.. not much disable, except --disable-shared. :( [16:25] <Keestu> :) all muxers, demuxers, parsers,fileters.. enabled. [16:25] <michaelni> saste, if you want [16:30] <michaelni> BBB, ubitux if the vp9 valgrind failures are valgrind bugs, please add them to tests/fate-valgrind.supp so valgrind fate gets green again and we can see new failures as color change [16:31] <cone-883> ffmpeg.git 03Stefano Sabatini 07master:9f956d9d8d98: ffmpeg: apply misc cosmetics to process_input() code [16:31] <cone-883> ffmpeg.git 03Stefano Sabatini 07master:094516a5d0f8: ffmpeg: use av_err2str() macro [16:31] <saste> michaelni, done [16:36] <ubitux> michaelni: i don't know what BBB planed for this [16:38] <nevcairiel> michaelni: btw, some ffvhuff tests fail on mingw 64-bit since some changes yesterday, looks like a crash, but I didn't have time to check yet whats really going on; http://fate.ffmpeg.org/report.cgi?time=20140123040141&slot=x86_64-mingw-w64… [16:41] <michaelni> nevcairiel, i saw but i couldnt reproduce yesterday in mingw + wine [16:41] <nevcairiel> yeah the wine fate stations seem to still be happy, too [16:41] <nevcairiel> just the native one is broken [16:44] <saste> what happened of anton's work on bitstream filters? [16:49] <saste> michaelni, also, what about the -itsoffset/-copyts conflict? [16:50] <saste> itsoffset is not effective unless -copyts is applied, which is a bit counterintuitive [16:50] <saste> also it is not possible to apply an output offset (with a single remuxing) [16:51] <saste> to simulate that i need to do a remux (to remove the initial offset), and then do -copyts -itsoffset OFFSET [16:51] <michaelni> saste, do we have tickets for these ? (with reproduceale testcases) [16:51] <saste> michaelni, no but i'm not sure they are "bugs" [16:51] <michaelni> then they would be feature requests [16:51] <saste> indeed i couldn't say which is the proper behavior [16:52] <saste> at least we should clarify docs (is what i'm trying to do right now) [16:52] <saste> my current way to apply a custom output offset is to use a custom tool, which is not optimal [16:53] <saste> (or alternatively a 2-steps remuxing) [16:54] <saste> also what about adding a new -otsoffset? [16:55] <saste> ideally we should use a bsf for that, but the current bsf design is quite limited (doesn't allow to change pkt timestamps) [17:00] <michaelni> -otsoffset is an option [17:02] <saste> michaelni, uh? where? [17:03] <michaelni> its an option to add -otsoffset :) [17:03] <michaelni> its an option to add such option [17:03] <saste> confused... ok [17:22] <saste> aaah vsync, all the magik [17:29] <saste> michaelni, can you explain this: d7ebeba80c609e160a171168b3434c342a652237 [17:29] <saste> what is "JM"?? [17:30] <kierank> Reference encoder [17:30] <saste> reference encoder of what? [17:31] <michaelni> h264 [17:32] <JEEB> ref dec/enc for AVC/H.264, just like HM for HEVC [17:33] <saste> and being compliant with $random_stuff is good for ffmpeg? [17:33] <JEEB> unless it's against the spec of course [17:33] <JEEB> and/or breaks spec-compliant stuff [17:34] <nevcairiel> compliant with the reference is not random. :D [17:58] <mark4o> saste: with multiple inputs that you want to mux together, -itsoffset is used to specify a separate relative offset for each input; probably not as useful if you have only one input [17:58] <mark4o> saste: for output offset, setpts/asetpts should be able to do that [18:06] <saste> mark4o: only when transcoding [20:27] <cone-883> ffmpeg.git 03Martin Storsjö 07master:2aec9e228cb3: rtpdec_asf: Copy the need_parsing field from the chained demuxer [20:27] <cone-883> ffmpeg.git 03Michael Niedermayer 07master:c1333a762c0e: Merge remote-tracking branch 'qatar/master' [20:42] <cone-883> ffmpeg.git 03addr-see-the-website(a)aetey.se 07master:bf23642dcc86: Cinepak encoder: adding option handling for flexibility [21:41] <cone-883> ffmpeg.git 03Reynaldo H. Verdejo Pinochet 07master:30f4c58f6f92: libavformat/mtv: Check for min header size first [21:41] <cone-883> ffmpeg.git 03Reynaldo H. Verdejo Pinochet 07master:d467e7df1a74: libavformat/mtv: donwgrade probe score for < 512 [21:41] <cone-883> ffmpeg.git 03Reynaldo H. Verdejo Pinochet 07master:ba15aab4a4a2: libavformat/mtv: make clear we assume bpp is always 16 [21:41] <cone-883> ffmpeg.git 03Reynaldo H. Verdejo Pinochet 07master:4d888a0ad823: libavformat/mtv: add missing MTV_ prefix to macro [22:10] <JEEB> what very simple thing have I forgotten? include\libavutil\error.h(100): error C2054: expected '(' to follow 'inline' [22:11] <JEEB> I just moved some C++ code that used a single define from libavformat/avio.h to a C file (and fixed the little things that had to be fixed for that move) [22:11] <JEEB> the C++ one compiles just fine, but this file doesn't [22:12] <JEEB> so I must be missing something very simple :s [22:41] <JEEB> ok, I bet something is defining inline to __inline [22:44] <JEEB> ugh, it was the other way [22:45] <JEEB> seemed to work for C++ mode, but C mode fails if I don't #define inline __inline [22:48] <wm4> inline is C99 [22:48] <wm4> which means... the ffmpeg headers are broken [00:00] --- Fri Jan 24 2014

1 0

ffmpeg.log.20140123
by burek 24 Jan '14

24 Jan '14

[00:45] <cuba> hey [00:45] <cuba> when encoding images with huffyuv or ffv1 and very low framerates like 1fps i get a video [00:46] <cuba> but its just the same image [00:47] <cuba> http://pastie.org/8658649 [00:47] <cuba> -r can also be 1 or 2 without any change [00:47] <klaxa> >to show your EXACT ffmpeg command and the COMPLETE console output. [00:47] <klaxa> i don't see any output [00:47] <cuba> oh sorry [00:48] <klaxa> if it's a windows terminal and you can't copy paste, a screenshot is fine too [00:49] <cuba> http://pastie.org/8658654 [00:49] <cuba> na its fine i use cygwin [00:50] <klaxa> >frame= 200 fps=2.6 q=-1.0 Lsize= 18729kB time=00:04:45.71 bitrate= 537.0kbits/s dup=0 drop=6855 [00:50] <klaxa> your frames are being dropped, i don't see why exactly though [00:52] <cuba> without setting -r it works fine [00:52] <cuba> maybe such low framerates are not supported [00:53] <klaxa> i don't think so, but i can be wrong, can you try with -r 30 and see if that works? [00:53] <cuba> yes that works [00:54] <klaxa> that's weird [00:55] <cuba> seems like everything under 25 wont work [00:55] <klaxa> is there a reason you use ffv1 and avi? [00:55] <cuba> >=25 does not drop [00:55] <cuba> no not really which container shall i use? [00:55] <klaxa> i personally use x264 and mkv for almost anything [00:56] <cuba> same issue [00:58] <klaxa> let me extract some frames from a video and try what you are doing [00:58] <cuba> ah its the argument handling i think [00:58] <cuba> but ive to download the movie [00:58] <cuba> maybe it must stand in front [00:59] <cuba> yep [01:00] <cuba> as second argument it works very nice [01:00] <cuba> thank you very much klaxa [01:03] <klaxa> so did it work? [01:03] <klaxa> sorry my system just crashed because >lol 2 gb ram [01:03] <cuba> yes [01:03] <cuba> seems like it needs to stand upfront [01:03] <cuba> then it works perfectly [01:04] <cuba> even with -r 0.73 [01:04] <klaxa> ah err... yeah that makes sense [01:04] <cuba> not really, does it? [01:05] <klaxa> no it does [01:05] <klaxa> you put it before -i right? [01:05] <cuba> yes [01:05] <klaxa> that makes sense because that sets the input framerate of your images to 0.7 fps [01:06] <klaxa> instead of taking whatever is default (maybe that's even 0) and then set the framerate for the output to 0.7 [01:06] <cuba> ah okay [05:00] <GregDR> I have question about fragmented MP4 files& Is it legal to have a single sample group description box ('sgpd') in the 'moov' while each 'traf' has only a sample to group box ('sbgp') that references it? Or must the 'sgpd' and 'sbgp' always exist together in the same container box? [08:33] <x_> If I use ffmpeg like this: ffmpeg -f flv test.flv I am able to record my desktop with 0 xruns, however as soon as I try to do -f flv 'rtmp://someserver.com/feed' I instantly get a few xruns. The xruns last for only a second or two, as thing seem to 'stabilize' if you will, and never happen again. The problem is the audio is not synced with the video when the xrun occur. [09:07] <xlinkz0> is there any image format that supports YUV? [10:37] <ussher_> Im trying to merge 2 stereo mp3 files. When they are of different lengths the length of the merged file is the length of the shorter mp3. Is there a way to loop one file so it returns to the beginning while until the main file completes? http://pastebin.com/TPV9FwvN [10:51] <ussher_> figured it out. make sure the overlay-track is always longer than the target file. thanks. [11:06] <Winston__Smith> ez [11:39] <x_> How come one xrun desyncs the audio permanently is there a way to make it detect this and rsync the audio as its going? [11:40] <x_> Trying to record my desktop for live view. [11:42] <x_> I see -async but I have not had luck with it. [11:42] <x_> aresample [11:50] <x_> "Stretch/squeeze samples to the given timestamps, with a maximum of 1000 samples per second compensation:" what given timestamps? [11:55] <ussher_> x_: i've been having good luck recording 720p with this command http://pastebin.com/2gBvXMQt [11:57] <Winston__Smith> i played a bit with windows overlay ... here my result http://lpc.fioresardo.net/up/%C2%A9%20Lino%27s%20Screensaver%202.0.exe <<< play any video in overlay modus (dvbviewer or mediaplayer classic for example) and when screensaver switches on continous watching [11:57] <Winston__Smith> windows 7 compatible [12:01] <x_> ussher_, in the event of an xrun I don't want it to permanently desync [12:02] <x_> ussher_, I can record fine but rather then stopping xruns all together rather try to handle them. [12:02] <ussher_> x_: your probably better at ffmpeg thatn me. i'd need to look up both of those terms to know what your talking about. I've been doing a lot of screencasts with the above set of params and its been working. [12:05] <x_> ussher_, ah I see, thanks for your input! ;) [12:30] <x_> https://trac.ffmpeg.org/ticket/2420 wondering the same thing, but this is not very helpful. [12:43] <x_> It is very really unclear that you need to have --enable-avresample set for that filter to be enabled. [12:43] <x_> very really ha oops [14:01] <x_> Is there a way to make ffmpeg print out the timestamps it refers to? [14:23] <average> x_: erm..vague question, rephrase [14:24] <relaxed> x_: version? [14:30] <x_> so my video pts_time:0.190476 and my audio pts_time:1.19473 are not the same how do I make sure they are always the same or close to the same. like on the fly audio video sync. [14:30] <x_> or is that even what I am looking for? I am not even sure if this would actual keep audio and video in sync in the event of an xrun or something [15:04] <x_> relaxed, What I am asking is, is there a way of keeping audio and video in sync on the fly? [15:07] <x_> async vsync and filters like asyncts are not what I am looking for I think. [15:10] <x_> as they just make sure the rates for audio and video are consistent as far as I understand. [15:11] <x_> looks like concat may do what I want. [15:13] <x_> scratch that. [15:28] <Mavrik> x_, you're looking for solution at the wrong place [15:28] <Mavrik> if you have a discrepancy between audio and video [15:28] <Mavrik> then there's something wrong with the thing that sets those PTS values [15:28] <Mavrik> ffmpeg can't magically know what's synced if timestamps are wrong :) [15:36] <x_> [Parsed_ashowinfo_0 @ 0x1f64120] pts_time:1.42942 [Parsed_showinfo_0 @ 0x1fa0860] pts_time:0.0952381 So it does not really know? [15:36] <x_> when the video and audio are in sync the pts_times do match up and they don't when the video is not in sync. [15:38] <x_> Maverick, [15:38] <x_> Mavrik, [15:38] <x_> huh [15:40] <Mavrik> <x_> when the video and audio are in sync the pts_times do match up and they don't when the video is not in sync. [15:40] <Mavrik> yea no kidding [15:40] <Mavrik> that is kinda the point :P [15:41] <x_> yeah? then why do they become desynced then? [15:41] <x_> the timestamps are correct [15:41] <x_> it correlates to what is on screen the audio is behind by about one second. [15:41] <relaxed> x_: are you using a recent version of ffmpeg? [15:41] <Mavrik> PTS TELL the player on when to show a video frame [15:41] <Mavrik> or audio frame [15:42] <x_> I am using a git version [15:42] <Mavrik> if your PTS are not correct, your A/V will not be in sync [15:42] <relaxed> Just checking [15:42] <Mavrik> so you're somehow changing (or improperly generating) your PTS in input vidoes [15:42] <Mavrik> *videos [15:42] <Mavrik> since I didn't see where your source comes from and what you're doing exactly... I can't help more than that. [15:43] <x_> So the fact that the pts will be in sync with the output but when there is an xrun the pts no longer matcxh [15:47] <x_> Mavrik, ffmpeg is setting the pts as far as I understand, I am capturing with x11grab and jack. [15:49] <x_> Mavrik, Also ffmpeg will report one xrun at the beginning of the output, but never again. However the xrun will only occur if I output the file via -f flv 'someserver' other wise -f flv test.flv won't have the same behavior. [15:49] <x_> Jack does not report an xrun. [15:50] <x_> getting a pastebin [15:54] <x_> http://pastebin.com/eUw9SBuW to server, http://pastebin.com/1zNdXuBT to local file [15:55] <Mavrik> well, it seems either x11grab or jack don't properly signal their timestamps [15:56] <x_> Mavrik, so the fact that it only xrun when I send it to a server is jacks or xorgs fault? [15:56] <relaxed> x_: try -framerate 21 [18:52] <LucaS05> hi. do you know where can i find an analysis of the xvid format? i don't know where to ask [18:54] <Zeranoe> LucaS05: What sort of analysis are you looking for? [18:54] <LucaS05> how it is composed and how the compression happens (the steps) [18:55] <LucaS05> maybe you have a pdf or something [19:06] <JEEB> 14496-2 [19:06] <JEEB> is the format's specification [19:06] <JEEB> MPEG-4 Visual [19:06] <JEEB> divx/xvid and avcodec's 'mpeg4' implement this format [19:09] <LucaS05> do you have a link? [19:11] <JEEB> http://lmgtfy.com/?q=14496-2+.pdf [19:13] <LucaS05> thanks :D [19:17] <JEEB> not sure why you want to know about that format, though [19:32] <x_> JEEB, I am experiencing an Audio packet xrun as I start to broadcast via -f flv <server_address> by not when I do as such -f flv test.flv any ideas? It actually offsets the audio by about one second. [19:33] <x_> but not when I do as such* [19:33] <x_> the audio does not get offset when I output by the ladder. [20:33] <simmerz> Hi, is it possible to composite a bunch of videos on top of each other, and then scale the whole composition down, in one command? I'm using complex_filter to do the first bit, but not sure if I can do the second at the same time [21:02] <simmerz> answered my own question :) [21:03] <juke_> hi [21:04] <juke_> hi take capture of a stream with ffmpeg -i "rtsp://mafreebox.freebox.fr/fbxtv_pub/stream?namespace=1&service=372" -vf select='gt(scene\,0.2)' -vsync 0 -f image2 image%3d.png [21:05] <juke_> how can i obtain better quality than http://digitalashram.org/explode2/image046.png ? [21:30] <maujhsn> I Would like to clone an older version of ffmpeg that I still like How is this done using the git repository? [21:34] <clever> git log, find the hash of the version you want, git checkout hash [21:41] <maujhsn> clever...Is it just that simple? [21:41] <JEEB> yes [21:42] <maujhsn> JEEB where is the git log location in '10.04"? [21:42] <maujhsn> Or is it? [21:43] <JEEB> uhh [21:43] <JEEB> easiest way would be to run the ffmpeg binary and see the git hash noted :P Unless that really is from 2010 and was before the svn->git move [21:43] <JEEB> if it is the svn, then it would note the svn revision there, and you should be able to grep the git log for that [21:43] <JEEB> to get the hash [21:44] <maujhsn> JEEB Yes this would be 2010, 2011! [21:44] <JEEB> just run the bloody binary to see the revision or hash :P [21:47] <maujhsn> JEEB Thanks for the tip! I'll check into it! [22:24] <slavko> hello [00:00] --- Fri Jan 24 2014

1 0

ffmpeg-devel.log.20140122
by burek 23 Jan '14

23 Jan '14

[00:36] <BBB> nevcairiel: to answer on your doom9 comment, ffvp9 -cpuflags 0 (i.e. c): user 0m18.951s, default (ssse3 mostly): user 0m6.271s [00:40] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:28bc1d94b739: avformat/mov: redesign the fps calculation [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:7e8f3048464f: lavf/libssh: add private_key option [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:8d3f14e11b34: lavf/libssh: factorize open_file function [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:bf5d73b04d68: lavf/libssh: factorize file_stat function [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:77fadab1beb6: lavf/libssh: factorize create_ssh_session function [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:7fb5f55e1e5a: lavf/libssh: factorize create_sftp_session function [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:522814455cb9: lavf/libssh: add av_cold attributes [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:b271aac53602: lavf/libssh: rename context variable from s into libssh [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:e0d124a9209f: lavf/libssh: fix seek with whence==SEEK_CUR [01:35] <cone-745> ffmpeg.git 03Lukasz Marek 07master:9ef441c52553: lavf/ftp: remove double space [01:35] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:94a52417cb23: Merge remote-tracking branch 'lukaszmluki/master' [02:02] <cone-745> ffmpeg.git 03Lukasz Marek 07release/2.1:c5a2a65e1e92: lavf/libssh: fix seek with whence==SEEK_CUR [13:02] <nevcairiel> can someone enlighten me how the encoder/stream timebases interact? I encode using libx264, it produces a AVPacket with timestamps in codec timebase, which is then handed off to the muxer, which expects stream timebases .... am i supposed to rescale it? are the two timebases supposed to be the same? [13:03] <nevcairiel> lookiong at the example, i'm supposed to rescale [13:03] <nevcairiel> glad its not documented anywhere =P [13:04] <saste> nevcairiel, yes [13:04] <saste> you had to fight with it as well [13:05] <saste> so when you get the packet from the encoder the timestamps are expressed in the *encoder timebase* [13:05] <saste> then you need to rescale before to pass the packet to the output stream [13:05] <saste> this is a bit silly, but that's it [13:05] <saste> probably the rescaling should be automatically handled by the library, that's a constant source of troubles [13:06] <nevcairiel> at least that seems more logical now [13:07] <nevcairiel> even if ugly [13:09] <BBB> ubitux: ty for review, how's lf going? :-p [13:11] <ubitux> BBB: sorry didn't have time last days, stuff happened [13:11] <ubitux> OTOH i'll probably have a (cheap) avx2 machine "soon"© [13:13] <saste> can someone explain what parsers (avcodec.h) are good for? [13:13] <saste> why should an API user use them? [13:24] <sspiff> can someone explain to me why avformat_open_input tries to read id3v2 metadata? I'm looking at libavformat/utils.c:546 and below, and it looks like there's no way to prevent this? [13:26] <av500> because one must always read id3v2 [13:29] <sspiff> av500: I'm trying to force-feed an MPEG2 TS to AVFormat, and I'm providing this through an AVIOContext, yet it never returns from avformat_open_input. I made it break on the first read operation, and this id3v2 function showed up. [13:29] <BBB> saste: I think typically you don't use them, lavf uses them [13:30] <BBB> saste: but if you don't use lavf, you need to use them before you input your packetized demuxer data into the decoders [13:30] <sspiff> av500: I'm asuming it'll look at the first pieces of data, determine there's no ID3v2 tag there, and quit? [13:30] <sspiff> I mean, return [13:36] <saste> BBB, yes, the fact that the API is public is a bit confusing [13:43] <kierank> sspiff: there is some stupid apple id3 in mpeg-ts spec [13:58] <BBB> saste: like I said, outside demuxers but lavc decoders [13:58] <BBB> saste: so for mplayer or gstreamer it makes sense [14:00] <sspiff> kierank: oh, great -_- [15:25] <cone-800> ffmpeg.git 03Tomas Härdin 07master:59dbc36f49db: Adding Cinepak encoder [15:25] <cone-800> ffmpeg.git 03addr-see-the-website(a)aetey.se 07master:762c4dc08276: avcodec/cinepakenc: fixes and improvements [15:54] <cone-800> ffmpeg.git 03Diego Biurrun 07master:8267f9363532: configure: Set default HOSTCFLAGS/HOSTCPPFLAGS after compiler detection [15:54] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:383d2339e043: Merge commit '8267f9363532b23b3b8b4e27491ee2d412a8aec3' [16:08] <cone-800> ffmpeg.git 03Janne Grunau 07master:ea49f60523c4: h264: skip chroma edges at the picture boundary while deblocking 4:4:4 [16:08] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:0e59e30b141e: Merge commit 'ea49f60523c4f4c43cb06a35166b44478d473e51' [16:20] <cone-800> ffmpeg.git 03Diego Biurrun 07master:19987fa18cea: bfin: vp3: Mark all binary object sections appropriately [16:20] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:7b89e24151ea: Merge remote-tracking branch 'qatar/master' [17:29] <ubitux> saste: i'm not so sure about metadata [17:29] <ubitux> (for the cabac thing) [17:30] <ubitux> one thing i don't like with these meta is that they get randomly populated to output files for no reason [17:32] <ubitux> :( [17:36] <wm4> there should be separate metadata for file tags and "random other stuff the demuxer might want to export" [17:36] <wm4> in this case, it seems to be about codec metadata though [17:36] <wm4> does that really end up in output files? [17:37] <ubitux> i don't know if that's even possible [17:38] <ubitux> format metadata definitely ends up being copied randomly depending on the format [18:04] <xlinkz0> is there a feature request to remove \r characters from reports? [18:07] <kierank> dos2unix? [18:10] <xlinkz0> doesn't work [18:23] <xlinkz0> kierank: does it work for you? try running dos2unix and then vim/less on the reported file and go to the end of it [18:23] <kierank> what report [18:26] <xlinkz0> kierank: run ffmpeg with -report [18:26] <xlinkz0> i get this after dos2unix : http://i.imgur.com/RaLt32p.png [19:47] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:631939bde6e2: avcodec/x86/lossless_videodsp: add diff_int16_mmx/sse2 [19:56] <burek> "Guys, we are testing a new technology - an FFmpeg-based DirectShow decoder (one filter for all file formats) and are looking for feedback." :) [19:56] <burek> is this a spam? [19:56] <burek> should I delete this from the forum? [19:57] <nevcairiel> from the name of the poster, its this guy: http://blog.medialooks.com/0LyGa6/ [20:22] <burek> The filter will be available in 2 versions: a) a Pro version with support for MXF, LXF and closed captions available for 1450 euro; b) a Standard version without these broadcast-specific features, available for 950 euro. [20:22] <burek> I'll delete it [20:23] <nevcairiel> its pro software [20:34] <Compn> burek : ask him where the source code is [20:34] <Compn> ask him how distributing ffmpeg without source is ok under gpl terms [20:34] <Compn> :) [20:34] <burek> :) [20:34] <burek> +1 [20:38] <nevcairiel> its actually lgpl, and strictly speaking you only have to provide the source code on request, not to mention that he at least claims to use zeranoe's ffmpeg binaries [20:38] <Compn> well i'm requestin [20:39] <nevcairiel> unless he lied on his page, you'll get a vanilla git tree :P [20:39] <Compn> its the tought that counts [20:39] <Compn> +h [20:40] <burek> do we have any legal representative/person so that we could tell him to contact that person via email and resolve all the issues about licensing? [20:40] <nevcairiel> what is there to resolve? :) [20:43] <Compn> burek : no, we dont have that [20:43] <burek> well, if you guys are ok with this, i'm fine too :) [20:46] <Compn> 1) Redistribute the FFmpeg folder found in our product's setup folder. This folder contains the following: a) FFmpeg's source code (exactly the one that was used to compile the DLLs that are being used); [20:46] <Compn> there it is [20:46] <Compn> the source code is in their setup dir [20:46] <Compn> http://blog.medialooks.com/g8mF25/ [20:46] <Compn> so nevermind burek [20:47] <burek> ok, fair enough [21:00] <wm4> nevcairiel: is it right that we both want to replace our custom mkv demuxers with lavf's? and that we both need extended features like ordered chapters for it? [21:01] <wm4> well, replacing it in the far future [21:01] <wm4> not now [21:03] <nevcairiel> i have no big desire to replace anything [21:03] <nevcairiel> mine works fine [21:17] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:fee97f25fa12: avcodec/x86/lossless_videodsp: port add_hfyu_median_prediction_mmxext to 16bit [21:33] <Compn> do we have vdpau fate tests ? [21:34] <gilligan_> Evening [21:42] <gilligan_> i've got a rather basic question about AVFrame::data - i'm a bit confused about how the data is stored here. The documentation says it's a pointer to picture/channel planes. How is the raw video data stored by the decoding functions ? just in sequence of r,g,b,r,g,b... ? or data[0] = r,r,r,r, data[1] = g,g,g,g .. [21:43] <smarter> depends on the format used [21:43] <gilligan_> I'm sorry. Probably a stupid question. I'm trying to help someone out who is fiddling with ffmpeg on android via Java JNI interface who has absolutely 0 idea about C - I on the other hand have no idea about ffmpeg/video [21:44] <Compn> the filters can use different colorspaces so it depends on colorspace i think ? [21:44] <Compn> erp i mixed up avframe and filters , ignore me [21:45] <smarter> gilligan_: see https://libav.org/doxygen/master/pixfmt_8h.html#a9a8e335cf3be472042bc9f0cf8… for the different formats [21:48] <smarter> gilligan_: see also http://www.fourcc.org/yuv.php which explains what packed and planar formats are [21:48] <gilligan-> smarter, ah..thanks a lot for that [21:48] <ubitux> smarter: incomplete, http://ffmpeg.org/doxygen/trunk/pixfmt_8h.html#a9a8e335cf3be472042bc9f0cf80… [21:48] <ubitux> ;) [21:48] <smarter> ubitux: ah, I forgot on what channel I was :) [21:49] <gilligan-> smarter, are there any convenience functions to convert between packed and planar formats ? [21:49] <smarter> I don't know that much, sorry :) [21:49] <smarter> probably [21:50] <gilligan-> smarter, well you helped me plenty already. Thanks [21:51] <smarter> gilligan-: https://www.ffmpeg.org/libswscale.html might help [21:51] <gilligan-> smarter, It's often hard find the right place to look/start if someone confronts you with something like "I don't really know what I am doing here can you help me with it" ;] [21:52] <gilligan-> aaah.. indeed! [21:52] <smarter> the actual documentation is here: https://www.ffmpeg.org/doxygen/trunk/group__libsws.html [21:54] <smarter> there are some API usage examples here which might help: http://git.videolan.org/?p=ffmpeg.git;a=tree;f=doc/examples;h=b78bac14f25a1… [21:57] <smarter> gilligan-: and if you've never heard of it you might want to know what https://en.wikipedia.org/wiki/Chroma_subsampling is too [22:01] <gilligan-> https://github.com/chelyaev/ffmpeg-tutorial/blob/master/tutorial01.c [22:01] <gilligan-> that's perfect for me ;) [22:02] <smarter> note that this tutorial is already a bit old and might refer to deprecated/removed functions [22:03] <smarter> depends what version of ffmpeg you're using [22:05] <gilligan-> ah, well it is not using img_convert anymore which was mentioned in other places [22:05] <gilligan-> but yeah, last update was 5 month ago [22:06] <gilligan-> but in combination with the other links i should be good to go anyhow - will see tomorrow when i'm back at work [22:06] <gilligan-> smarter, thanks for your help and suggestions again [22:08] <smarter> gilligan-: see http://git.videolan.org/?p=ffmpeg.git;a=blob;f=doc/APIchanges for the API changes [22:08] <smarter> you're welcome :) [22:08] <gilligan-> also there is a pull request that fixes the use of deprecated functions: https://github.com/chelyaev/ffmpeg-tutorial/pull/19 [22:08] <gilligan-> sounds good [23:34] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:fad49aae28cf: avcodec/x86/lossless_videodsp: Port sub_hfyu_median_prediction_mmxext to int16 [23:34] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:ef00ef7553fc: avcodec/x86/lossless_videodsp: port sub_hfyu_median_prediction_int16 to yasm [23:43] <cone-800> ffmpeg.git 03Martin Storsjö 07master:cff480e49d73: mpegvideo_enc: Don't call ff_h263dsp_init unconditionally [23:43] <cone-800> ffmpeg.git 03Michael Niedermayer 07master:0269b30fefba: Merge remote-tracking branch 'qatar/master' [00:00] --- Thu Jan 23 2014

1 0

ffmpeg.log.20140122
by burek 23 Jan '14

23 Jan '14

[00:12] <sherafff> hi [00:13] <sherafff> i need to join video and audio file together, then video is shorter than the audio, i want stop joining when video is done [00:14] <sherafff> for now, i get something but the video stay at the end and audio keep playing until its done [00:16] <sherafff> ok i found [00:16] <sherafff> thx [00:16] <sherafff> http://superuser.com/questions/628996/adding-soundtrack-to-a-video-using-ff… [00:16] <sherafff> here [00:18] <brontosaurusrex> -shortest < or something [00:22] <_fil_> -itsoffset was the solution, llogan [00:22] <_fil_> thanks again [00:22] <_fil_> final cmd is: ./ffmpeg-2.1.3 -i incomplet.mp4 -itsoffset 00:01:58.5580 -i insert.mp4 -filter_complex "overlay=enable='between(t,118,220)'" -preset ultrafast output.mp4 [00:22] <_fil_> s/5580/580/ [00:25] Action: _fil_ happy [01:06] <SpecialEd> Hey guys, by any chance do any of you know if it is possible to use ffmpeg to transcode x264/aac videos using the GPU instead of the CPU? [01:06] <klaxa> it's not feasible [01:07] <SpecialEd> I thought that was the case [01:07] <SpecialEd> I just got these two 780ti video cards in SLI [01:07] <SpecialEd> and was hoping maybe I could put them to work lol [01:07] <sacarasc> If you send one of them to me, I'll show you how it is done. [01:08] <SpecialEd> I can do that just as long as you first send me $750 or whatever it is that I spent on one, lol [01:08] <sacarasc> Heh. [01:08] <SpecialEd> one card uses 3 pci slots [01:08] <SpecialEd> its so fat with the fans [01:09] <SpecialEd> i had to give up my PCIe raid controller to make space for the second one [01:11] <SpecialEd> Are there other programs that can do transcoding using GPU? Or is that still a fantasy of the future? [01:12] <klaxa> it's a fantasy for the never i think [01:12] <SpecialEd> ok [01:12] <klaxa> JEEB will probably be able to elaborate why it won't really work [01:13] <SpecialEd> i have a 4770k and a 4771 slaving away so im probably in much better shape than most others [01:13] <sacarasc> Something about it not being linear data makes GPU encoding not as good as CPU... [01:13] Action: sacarasc shrugs. [01:14] <SpecialEd> ok [01:14] <SpecialEd> i also have a 2700k in my other room, is there any tools for ffmpeg to do transcoding over the network? [01:14] <SpecialEd> so that I can use that cpu on files locally? [01:14] <SpecialEd> I was thinking of spinning up a headless VM on that desktop and doing a network share [01:15] <klaxa> not afaik [01:15] <klaxa> i started coding something like that in python [01:15] <klaxa> https://github.com/klaxa/Distributed-encoding [01:15] <klaxa> i will rewrite that at some point maybe [01:20] <SpecialEd> looks very interesting but unfortunately my 2700k is a windoze desktop so i'd have to go back to the virtualbox vm method [01:20] <SpecialEd> and at that rate i might as well just use a network share to handle it [01:31] <Irrelium> I have been using ffmpeg to record screencasts with two audio inputs using this command http://pastebin.com/YhvFRbyG . This seems to mix the two audio inputs together into one track. Is there a way I could record the tracks separately so I could go back later and remove one of them if I want to? [02:03] <Irrelium> llogan: I didn't think the console output was neccesary, but I can get it I guess [02:03] <llogan> it's always necessary [02:06] <Irrelium> Here is the console output http://pastebin.com/hHtwCzxB [02:06] <Irrelium> I'm glad Irssi alerts you when you try to paste more than a few lines, I accidentally tried to paste in the console output instead of the pastebin link :P [02:07] <llogan> Irrelium: remove the whole filter_complex line [02:08] <Irrelium> When I didn't have that line in there it didn't seem to record any audio at all [02:08] <llogan> and add the appropriate -map options [02:08] <llogan> -map 0 -map 1 -map 2 [02:08] <Irrelium> Oh [02:08] <Irrelium> I will have to look up how that works [02:08] <llogan> http://ffmpeg.org/ffmpeg.html#Stream-selection [02:08] <llogan> http://ffmpeg.org/ffmpeg.html#Advanced-options [02:09] <llogan> http://ffmpeg.org/ffmpeg.html#Stream-selection [02:09] <llogan> those should help [02:09] <Irrelium> Thanks [02:09] <llogan> oops...i meant http://ffmpeg.org/ffmpeg.html#Stream-specifiers-1 as well [02:10] <llogan> as for the "Advanced-options" link see the -map option specifically [02:10] <Irrelium> Ok [02:10] <llogan> it should have recorded at least one audio stream [02:11] <llogan> -threads 0 is default so you can omit that [02:12] <Irrelium> Oh ok [02:12] <llogan> why -b:v -minrate -maxrate instead of -crf? [02:12] <Irrelium> I originally based this off a command I got from the Internet [02:12] <Irrelium> So there might be some unneeded stuff in there [02:13] <llogan> also see https://trac.ffmpeg.org/ticket/3045#comment:7 [02:13] <Irrelium> I got the -b:v -minrate -maxrate from here https://trac.ffmpeg.org/wiki/x264EncodingGuide [02:14] <Irrelium> At the part where it says "There is no native CBR mode, but you can "simulate" a constant bit rate setting by tuning the parameters of ABR:" [02:14] <llogan> that's one of roger's additions [02:14] <Irrelium> Does -crf do the same thing? [02:14] <Hello71> you don't want constant bitrate [02:15] <Irrelium> I do when I am streaming and want to keep the bitrate below my upload speed so I don't slow down my Internet [02:15] <Hello71> so you don't actually want constant bitrate [02:15] <Hello71> you want qos [02:16] <Irrelium> The constant bitrate works fine :P [02:16] <storrgie> I'm using ffmpeg to convert ALAC to FLAC via "for f in *.m4a; do ffmpeg -i "$f" -c:a flac "${f%.m4a}.flac"; done" Is there any way for me to pass in an argument to flac such as --best? [02:16] <llogan> Irrelium: remove -minrate [02:17] <Irrelium> Yeah, I guess I don't really need a minrate [02:17] <Irrelium> Although I would like to keep the quality as good as possible [02:18] <relaxed> storrgie: you want the best compression? [02:18] <storrgie> relaxed, with flac yes [02:19] <storrgie> the cli flac package has a --best arugment [02:19] <llogan> storrgie: probably -compression_level <int> [02:19] <Irrelium> How does -crf differ from -b:v -minrate -maxrate? [02:19] <Hello71> it's a constant rate factor. [02:19] <Hello71> not a constant rate. [02:20] <relaxed> storrgie: put this after "$f", -c:a flac -lpc_type 3 -lpc_passes 8 -lpc_coeff_precision 0 -prediction_order_method 5 -min_partition_order 1 -max_partition_order 8 [02:20] <storrgie> woah [02:20] <storrgie> watttttt [02:20] <storrgie> relaxed, I have to ask, silly question maybe... is this going to be a lossless conversion from the ALAC to FLAC? [02:21] <relaxed> yes, flac is always lossless [02:21] <storrgie> relaxed, its kinda weird, my input files are like 5-7MB .m4a (ALAC) files and my flac files (with your options added) at 30-40MBs [02:22] <llogan> Irrelium: think of crf as the closest as you can get for an intelligent "constant quality level" [02:22] <relaxed> storrgie: hmm, can you upload one of your m4a somewhere so I can test? [02:23] <Irrelium> llogan: ah [02:23] <Irrelium> I would rather have a max bitrate [02:23] <storrgie> relaxed, one tic! [02:23] <Irrelium> Since I am trying to keep it just below my upload speed [02:23] <storrgie> relaxed, thanks man. I apprecaite it [02:28] <storrgie> relaxed, do you know of a service I could use to upload the files? [02:28] <storrgie> I tried mega but they now require you to make a user account [02:29] <relaxed> storrgie: http://db.tt/cDWPLzP [02:30] <storrgie> i dont have an account there either [02:30] <storrgie> im the worst [02:30] <relaxed> according to `du` on a sample I ran `flac --best` on: 36148 sample.flac [02:31] <relaxed> ffmpeg with the options I gave you: 36128 ffmpeg.flac [02:31] <relaxed> FLAWLESS VICTORY [02:31] <llogan> A winner is you. [02:31] <llogan> (guess the game) [02:31] <relaxed> We get signal. [02:32] <sacarasc> storrgie: Are you sure they're ALAC and not just AAC? [02:32] <relaxed> llogan: I had to google it :( [02:33] <relaxed> but I played it quite a bit as a kid [02:33] <llogan> Now the song is stuck in my head. [02:33] <storrgie> I believe they are (https://gist.github.com/storrgie/72a23c470eca92c4bb74) [02:34] <storrgie> when I try to play the resulting flac files in mpd I see errors [02:34] <sacarasc> Stream #0:0(eng): Audio: aac (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 265 kb/s (default) [02:34] <storrgie> sacarasc, oooo [02:34] <relaxed> they are aac, so there's no point in using flac [02:35] <storrgie> dese fakin uploaders [02:35] <storrgie> I do have some ALAC here [02:35] <storrgie> one tic [02:36] <storrgie> helps to test with the right data, sorry, thanks for pointing that out [02:37] <storrgie> so does ffmpeg even use flac under the hood? [02:38] <llogan> ffmpeg has a native flac encoder [02:38] <storrgie> llogan, that is quite cute [02:39] <storrgie> relaxed, is there any way I could also delete the input m4a files in that bash loop (pardon my bash newbism) [02:40] <relaxed> ffmpeg command; rm "$f";done [02:40] <bparker> oh hai [02:41] <storrgie> bparker, ohaifriend [02:42] <Irrelium> I got it working, thanks llogan [02:42] <relaxed> storrgie: it's wise to use `echo` to see what will be done when running loops and commands like that, just in case. [02:45] <storrgie> relaxed, it might not be working quite right... mpd seems to be not reading the files. lemme get the mpd log [02:45] <Irrelium> Seems a good idea to me to run a separate script to rm them after you are sure the outputs turned out ok [02:45] <storrgie> https://gist.github.com/storrgie/be4e30826f6b2ed68795 [02:45] <storrgie> not sure why, but mpd doesn't seem to like these flac files produced by ffmpeg [02:46] <crawln> maybe someone in here can tell me why the compiled version of ffmpeg messes up on this script. [02:46] <crawln> http://pastie.org/private/ffmtsnkmne1ldw0nmegn4g [02:47] <bparker> crawln: define 'messes up' [02:47] <bparker> llogan: wow, you must get a lot of that. [02:47] <crawln> I can't show the complete console output of the loop [02:47] <bparker> that there's a command for it [02:47] <crawln> TYVM [02:47] <bparker> crawln: why not? [02:47] <crawln> because I have yet to get it to print to anything [02:48] <llogan> crawln: see "/msg fflogger help" [02:48] <bparker> crawln: you mean running it doesn't print anything at all? [02:48] <bparker> what if you just run the ffmpeg line itself? [02:48] <crawln> no i mean when it runs it does 1st file fine then starts sending out unintelegible data [02:49] <crawln> works fine by itself [02:49] <bparker> so paste that [02:49] <crawln> but starts cutting characters off the filenames if i put it through the loop [02:49] <bparker> everything it outputs. [02:49] <bparker> crawln: do your files have spaces in them? [02:49] <crawln> no [02:49] <bparker> try not using while. [02:49] <crawln> this script works fine on the non-complied version of ffmpeg [02:49] <bparker> for file in `find ...`; do ffmpeg ... $file; done [02:53] <crawln> hmm let's see if i can put this in terms you understand better. Precompiled ffmpeg -- works with script. Compiled from git -- does not work from script because it's cutting filename characters out of the variable. [02:53] <crawln> but echoing the command shows that the filenames where never cut from the command [02:54] <crawln> so since the only things that have changed. is and are the fact that ffmpeg was compiled. is there a reason that i would start getting filename that are not what was sent in the command issued. [02:57] <crawln> Now since this has been tested on 2 different systems running 2 different OS's 1 ubuntu 12.04 server the other being Arch Linux x64. Both of which work fine with the ffmpeg from repository and this script. But if I compile ffmpeg on each system. FFmpeg starts cutting the filenames till there's nothing left. [02:58] <relaxed> crawln: http://mywiki.wooledge.org/BashFAQ/089 [02:58] <relaxed> or ask in #zsh [02:59] <crawln> this happens regardless of zsh or bash [02:59] <crawln> i've done both [02:59] <crawln> the zsh was to see if it was something I wasn't thinking of [03:01] <crawln> again your thinking it's a problem with my script. that does work. I have a problem with ffmpeg cuting the fucking filenames [03:01] <crawln> "ffmpeg -i ./Star.Trek.Voyager.s01e03.Parallax.mkv -c:v copy -c:a libfdk_aac ./Star.Trek.Voyager.s01e03.Parallax.mp4" does not fking equal "ffmpeg -i ./tar.Trek.Voyager.s01e03.Parallax.mkv -c:v copy -c:a libfdk_aac ./tar.Trek.Voyager.s01e03.Parallax.mp4" [03:01] <crawln> or how about by the time it get's to file 30 [03:02] <crawln> when it's "ffmpeg -i ./ax.mkv -c:v copy -c:a libfdk_aac ./ax.mp4" [03:02] <crawln> those are not the commands being issued to ffmpeg [03:02] <llogan> That's a feature. You don't want to watch Voyager anyway. [03:02] <llogan> It should re-encode them to TOS. [03:04] <storrgie> ooo, bash question... I'm about to try to process a ton of ALAC files... can I scan folders recursively for them? [03:04] <storrgie> this is what I've got so far: for f in *.m4a; do ffmpeg -i "$f" -c:a flac "${f%.m4a}.flac"; rm "$f"; done [03:04] <storrgie> I think there is a bash 'find' [03:05] <crawln> fking useless [03:06] <crawln> for the record. the find is because some things are more than a folder deep. hints for fking blah won't go more than the current folder. [03:07] <crawln> idiots [03:21] <relaxed> contrary to what that troll said, for i in download/*/*/*flac;do echo "$i";done [03:21] <relaxed> works fine [03:24] <relaxed> but using find is probably more effective [04:09] <thebombzen> storrgie: the find command is quite useful :D try: find -name \*.m4a [04:11] <thebombzen> or if you want to convert, try find -name \*.m4a | parallel ffmpeg -i {} -c:a flac {.}.flac [04:11] <thebombzen> GNU Parallel reads a list of files from stdin, and {} means the filename, and {.} means the filename - the extension [04:12] <thebombzen> or you could use {/} to mean the filename without the directory, or {/.} for the name without the directory or the extension [04:13] <thebombzen> that is to place myalac/*.m4a to myflac/*.flac do find myalac -name \*.m4a | parallel ffmpeg -i {} -c:a flac flac/{/.}.flac [04:14] <thebombzen> (note that \* prevents the shell from expanding the * but "*.m4a" works too [04:25] <ac_slater> hey guys, I'm working with some embedded devices and opencv. I'm curious if anyone in the ffmpeg world has anything bad to say about opencv's ffmpeg layer [04:25] <ac_slater> debating creating a replacement for my own purposes [08:06] <mohit> Hi folks [08:06] <mohit> i am encoding camera frames using FFMPEG in android [08:07] <mohit> But frames pixel format is YV12 [08:08] <mohit> i swapped U and V like this http://stackoverflow.com/questions/16392521/how-to-convert-from-yv12-to-yuv… [08:08] <mohit> But its not working and i am getting green colored frames [08:08] <mohit> anyone help please [08:09] <anshul> how are you trying, are yiu using ffmpeg command or library [08:10] <mohit> library code [08:10] <anshul> have you defined the pix_fmt [08:11] <anshul> I think ffmpefg support yv12 if you set the pix_fmt [08:11] <mohit> where> [08:11] <mohit> http://stackoverflow.com/questions/16392521/how-to-convert-from-yv12-to-yuv… [08:12] <mohit> i am swapping after filling frame from data [08:27] <SirCmpwn> I have a microphone on computer 1 [08:27] <SirCmpwn> I have ffmpeg on computer 1 and computer 2 [08:27] <SirCmpwn> I can ssh into computer 2 [08:27] <SirCmpwn> can I talk and have sound come out of computer 2 [08:28] <relaxed> stream the mic audio to computer 2 [08:29] <SirCmpwn> thanks, that was super helpful [08:29] <relaxed> https://trac.ffmpeg.org/wiki/StreamingGuide [09:34] <anshul> mohit:you can look at the link http://code.google.com/p/video-manager-and-generator/source/browse/overlay.c [11:51] <__raven> i need to build a quality test chain for some params - anyone who did such already? [15:13] Last message repeated 1 time(s). [16:07] <x_> I am getting [jack @ 0x2f5f240] Audio packet xrun while using x11grab but the jack server itself does not indicate any xrun. Secondly the audio coming threw the speakers is just fine. Is there a way around this? [16:08] <x_> working on a pastebin of my command. [16:12] <x_> Going to tinker with it a bit first. [16:35] <__raven> i need to build a quality test chain for some params - anyone who did such already? [17:25] <Maverick|MSG> what's the typical framerate of an animated gif? [17:25] <Maverick|MSG> (even though a gif doesn't really have a framerate) [17:37] <x_> Probably has more to do with the over all size of the animated gif [18:30] <Maverick|MSG> what pixel formats are available when encoding to a gif? [19:45] <DeadSix27> Maverick|MSG: rgb8 is what i know of [19:45] <Maverick|MSG> yeah, rgb8 and pal8 are the only ones I've found so far, but an 8 bit palette is pretty weak [19:46] <DeadSix27> rgb24? [19:46] <DeadSix27> not sure if that works [19:46] <Maverick|MSG> tried that as well; ffmpeg resets back to using pal8 [19:46] <Maverick|MSG> Incompatible pixel format 'rgb24' for codec 'gif', auto-selecting format 'pal8' [19:47] <Maverick|MSG> tried increasing the bitrate as well but no luck [19:49] <Maverick|MSG> oh, bgr8 also works apparently [19:50] <Maverick|MSG> same scenario though [19:51] <klaxa> gif is limited to 256 colors by format definition [19:52] <klaxa> there's nothing you can do about that [19:52] <klaxa> you can, however, defined an optimized color pallet, not with ffmpeg though [19:53] <Maverick|MSG> ah, right, sorry. I was more complaining about the fixed 8 bit color palette [20:02] <Maverick|MSG> klaxa do you know if ffmpeg has plans to integrate a dynamic palette? [20:02] <klaxa> i don't think so [20:02] <Maverick|MSG> dang, ok [20:02] <Maverick|MSG> thank you for the help [20:02] <klaxa> you can export it to individual frames though [20:02] <Maverick|MSG> how? [20:03] <klaxa> and use something like gimp to generate an optimized color pallette [20:03] <Maverick|MSG> oh, yeah. I was hoping for ffmpeg to do everything in terms of turning a video into an animated gif [20:03] <Maverick|MSG> rather than use a secondary program [20:03] <klaxa> it can do it, but it will look like crap :P [20:03] <klaxa> http://linuxers.org/tutorial/how-extract-images-video-using-ffmpeg [20:04] <klaxa> you might want to look at imagemagick instead of gimp [20:04] <klaxa> just remembered [20:04] <Maverick|MSG> yeah, googled around a bit and imagemagick's convert application does it [20:04] <klaxa> i haven't used the gif feature of imagemagick extensively but i'm quite positive they support optimized color pallets [20:05] <Maverick|MSG> http://superuser.com/questions/556029/how-do-i-convert-a-video-to-gif-using… [20:06] <Maverick|MSG> it looks kinda ugly, but I can live with it [20:06] <Maverick|MSG> was just hoping for some magical switch that would fix the palette [20:07] <klaxa> i bet one could port that from gimp or imagemagick, nobody has felt the need to do so though, or it's not feasible and one gets told to use those tools instead [20:27] <Freakshow> folks, what is has_b_frames=2 telling me from ffprobe? [20:44] <burek> Freakshow, does this help? https://www.google.com/search?q=has_b_frames+site%3Affmpeg.org [20:44] <burek> http://ffmpeg.org/doxygen/trunk/structAVCodecContext.html#a686a77363668795c… [20:47] <Freakshow> burek: that was one step away from lmgtfy [20:48] <Freakshow> but effectively the same [20:49] <Freakshow> I'm not clear on how this would be getting set to 2 then [20:49] <burek> sadly yeah [20:49] <burek> docs say nothing about that option, so.. :/ [20:50] <burek> has_b_frames integer [20:50] <burek> that's all :/ [20:50] <burek> http://www.ffmpeg.org/ffmpeg-all.html [21:04] <Freakshow> it's curious that I see this [21:04] <Freakshow> 3115 if (h->avctx->has_b_frames < 2) [21:05] <Freakshow> in h264.c [21:06] <Freakshow> it's clearly expected to be 2 at some point [21:20] <iive> ever heard of b-frame pyramid? [21:21] <Freakshow> I have but only in reading through other docs, would you mind explaining a little? [21:21] <Freakshow> googling as well [21:22] <iive> well, with b-frame pyramid, you can have b-frames using other b-frames for prediction. aka something like second level of decoding. [00:00] --- Thu Jan 23 2014

1 0

ffmpeg-devel.log.20140121
by burek 22 Jan '14

22 Jan '14

[01:05] <lukaszmluki> Hi, can I ask somebody to review http://ffmpeg.org/pipermail/ffmpeg-devel/2014-January/153441.html - at least doc update part, I want to send pull request with some other commits before going sleep. [01:11] <Zeranoe> Does ffplay support SDL 2.x? [01:13] <lukaszmluki> i think ffmpeg in general support only 1.2 unless you change configure [01:28] <kierank> lukaszmluki: why does it need inlining [01:30] <lukaszmluki> it does not, but i wont be called anything else [01:35] <BBB> ? [01:47] <mark4o> lukaszmluki: The option is the private key itself, or a file name, or full path name? The doc should be more clear. [01:47] <mark4o> lukaszmluki: s/search for keys in/searches for keys in the/ [01:47] <mark4o> lukaszmluki: s/invalid or not existing key/an invalid or nonexistent key/ [01:48] <mark4o> lukaszmluki: s/successed/successful/ [01:49] <lukaszmluki> ok, thanks [01:52] <BBB> lukaszmluki: if not performance-critical, we typically prefer to not mark with inline attribute and leave it to the compiler [01:53] <BBB> inline etc. should only be used in performance-critical code [01:53] <lukaszmluki> got it, i'm removing it [01:53] <BBB> ty [01:58] <lukaszmluki> just as question, does it matte when you move some code for better readability to a function and make it inline? [01:59] <lukaszmluki> i guess compiler figure it out, but it wont hurt to make a hint [02:06] <BBB> lukaszmluki: again, all this depends on how performance critical it is [02:06] <BBB> lukaszmluki: typically it won't matter either way [02:07] <BBB> lukaszmluki: but in inner loop function in audio and video decoders, there's special needs, so to say, so then it sometimes makes a measurable difference [02:12] <lukaszmluki> I see . Sometimes it is hard to know what is better :) [02:13] <lukaszmluki> I hope av_cold is ok anyway because it added it :) [02:13] <BBB> av_cold means "is only called during initialization" [02:13] <BBB> so it's probably ok [02:28] <Timothy_Gu> FFmpeg is now used to endorse Coverity: http://blog.coverity.com/2013/06/18/scanning-ffmpeg-for-your-audio-video-en… [02:37] <Compn> huh [02:37] <Compn> they dedicate 13 pages to ffmpeg in their 2012 report too [03:27] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:63d2be7533b7: avcodec/x86/lossless_videodsp: use SPLATW in add_int16 [03:27] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:83b67ca05609: avcodec/x86/lossless_videodsp: Port lorens add_hfyu_left_prediction_ssse3/sse4 to 16bit [03:27] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:3715f9e2c68b: avcodec/lossless_videodsp: fix diff_int16_c on MIPS [03:27] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:13e0109a5f2c: fate: add a few tests for >8bpc ffvhuff [04:11] <cone-469> ffmpeg.git 03Justin Ruggles 07master:d01e684186bc: mov: do not set avg_frame_rate in the demuxer [04:11] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:e9b836cd87c2: Merge commit 'd01e684186bc1631bc176f06b89d33c27ec0d24d' [04:11] <michaelni> Daemon404, do you have a sample / can you share it that shows wrong fps value ? [04:13] <michaelni> also did someone compare the wrong code in mov against the default code in terms of which is more accurate for more (real world) files ? [04:20] <cone-469> ffmpeg.git 03Martin Storsjö 07master:24eb3c791606: rtmpproto: Avoid using uninitialized memory [04:20] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:ca9d7c57f4b0: Merge commit '24eb3c791606fe98a1591c13a8b2ba6c342bb3b5' [04:26] <cone-469> ffmpeg.git 03Martin Storsjö 07master:89564be444d2: rtmpproto: Send a full, absolute timestamp if it isn't monotonically growing [04:26] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:1c7d2870440f: Merge commit '89564be444d24f75ea5add8b6987e414cf7aa7d5' [04:36] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:dd2d0039b640: vc1: Always reset numref when parsing a new frame header. [04:36] <cone-469> ffmpeg.git 03Michael Niedermayer 07master:a459891e624c: Merge remote-tracking branch 'qatar/master' [06:11] <Guest66336> while compiling ffmpeg i'm getting error /usr/bin/ld.bfd.real : libavutils/aes.o : reloaction R_ARM_THM_MOVW_ABS_NC against a 'local symbol' cannot be used when making a shared object, recompile with -fPIC libavutils/aes.o : could not read symbols : Bad value [08:17] <anshul> I was using mjpeg codec from libavcodec, I am getting [mjpeg @ 0xaaf14d20] overread 8 message in the new git code, while I am not getting any error in the fourier release of ffmpeg [10:01] <nguydavi> Hello ! Can anyone tell me if there is a date for ffmpeg release 2.2 ? As there are many securities patches recently, I would like to know if it's better to wait for the next release or apply all the patches now. [10:28] <anshul> nguyadavi: if you are not distributor why not use git [11:56] <cone-745> ffmpeg.git 03Stefano Sabatini 07master:e34ad128a363: examples/muxing: reduce duration, remove wrong and misleading comment [12:31] <saste> ubitux: https://trac.ffmpeg.org/ticket/3093 [12:32] <saste> you wrote: Filtering audio still needs to some adjustments. [12:32] <saste> what's wrong with it? [12:32] <saste> should it use the refcounted obscure API? [12:40] <nguydavi> I am asking if there are any dates for the next release because I am interested in certain patches such as 459db51271807ba26162db7b67ac1ff444cc0fa9, e630ca5111077fa8adc972fe8a3d7e2b3e8dc91f, 165f96cd2d687122748f862a0bc6e9908fe3d5d2 and more. From the shortlog of 2.1.3 I couldn't find those commits. [13:03] <Daemon404> michaelni, see pm [14:12] <sspiff> ubitux: the leak is gone for me [14:13] <ubitux> send a patch :p [14:13] <sspiff> the indentation is wrong in that function, can I fix that in the same patch? [14:17] <ubitux> better not [14:23] <michaelni> anshul, how can that jpeg issue be reproduced ? [14:31] <michaelni> nguydavi, these 3 dont look security relevant, 2 are memleaks the 3rd is a assert failure with non default build flags [14:34] <nguydavi> michaelni, I am interested to know what release will include these commits http://secunia.com/advisories/56352/ [14:36] <nguydavi> michaelni, and if there is an approximate date, I would like to know if it's better to manually apply the patches or simply wait for the next release [14:47] <wm4> Gynvael Coldwind really is a cool name [14:49] <michaelni> nguydavi, these commits seem to be a random list of commits, certainly not a list you blindly want to backport ,some of which like 75647dea6f7db79b409bad66a119f5c73da730f3 are backports of ffmpeg fixes in a fork, this one for exxample was fixed in 9ed388f5985992a0a6a43fdc0b1732962b6b5619 in ffmpeg [14:51] <nguydavi> michaelni, oh so all of these fixes should be included in 2.1.3 ? [14:52] <michaelni> no, some are missing though the ones i saw did look like normal everyday bugs not security issues, but iam not through with the list yet [14:53] <michaelni> ill make sure they get backported to the next 2.1 release [14:53] <nguydavi> great ! thanks ! do you know when is the next release ? [14:54] <nguydavi> (thank you very much for your time btw, michaelni) [14:54] <michaelni> dunno about release dates [14:55] <nguydavi> ok [14:57] <nguydavi> thanks again for your time [15:10] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:a7a07cc98ac5: hevc: check that VPS referenced from SPS exists [15:10] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:42a29015e1bd: Merge commit 'a7a07cc98ac548297b5b0628cb81280e11952e3f' [15:15] <michaelni> Daemon404, thx, will look at it later [15:16] <Daemon404> k [15:16] <Daemon404> we get anough of those samples that we actually remporarily patched ffmpeg in production with that patch [15:16] <Daemon404> enough* [15:16] <Daemon404> (that is currently the only modification to ffmpeg) [15:18] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:d5c15ebeaf19: hevc: Fix modulo operations [15:18] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:7c4b96f55eb6: Merge commit 'd5c15ebeaf1914ea5e3e0599d4316d7c4cf74434' [15:24] <cone-745> ffmpeg.git 03Luca Barbato 07master:b37e796082b2: hevc: Use uint64 to check for tile dimensions [15:24] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:29ba1cff2b79: Merge commit 'b37e796082b2d787aff3cd5631bb89c4fd374708' [15:27] <cone-745> ffmpeg.git 03Guillaume Martres 07master:a246d06fe0dc: hevc: clip pixels when transquant bypass is used [15:28] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:ee3e1951af5c: Merge commit 'a246d06fe0dc6c2ea65e95327624b4537ff9bd0d' [15:30] <cone-745> ffmpeg.git 03Guillaume Martres 07master:faf03ecba031: hevc: Remove useless clip [15:30] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:c7b454f7a42e: Merge commit 'faf03ecba03155bb1f5416713bd01da043863b43' [15:36] <cone-745> ffmpeg.git 03Luca Barbato 07master:838740e64205: hevc: Prevent some integer overflows [15:36] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:5b93b2722dea: Merge commit '838740e6420538ad45982da6b1d3aa3ae91307f5' [15:39] <cone-745> ffmpeg.git 03Luca Barbato 07master:0d999333f96a: hevc: Bound check slice_qp [15:40] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:58f437c3f667: Merge commit '0d999333f96a34903448579bf13a3209deaee9da' [15:53] <cone-745> ffmpeg.git 03Luca Barbato 07master:e22ebd04bcab: hevc: Bound check cu_qp_delta [15:54] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:a69dd1163b1a: Merge commit 'e22ebd04bcab7f86548794556c28ecca46d9c2ac' [16:01] <cone-745> ffmpeg.git 03Sam Lantinga 07master:5b2b23f2d69e: dxva2: Retry IDirectXVideoDecoder_BeginFrame() [16:01] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:9056d0c94af5: Merge commit '5b2b23f2d69e05c5fcd1c933e383fe60e185574d' [16:09] <cone-745> ffmpeg.git 03Sam Lantinga 07master:9d80b1ae9590: dxva2: Log errors verbosely [16:09] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:91f4394ed31e: Merge remote-tracking branch 'qatar/master' [16:16] <Daemon404> ... 90 comments [16:16] <Daemon404> wow [16:16] <nevcairiel> the first 80 or so are clueless people poking at random things [16:16] <nevcairiel> took me 30 minutes to fix the actual issue [16:17] <nevcairiel> first 70 is more accurate [16:17] <Daemon404> sounds about right [16:17] <nevcairiel> the last couple is just some xbmc dudes worrying that one allocation per slice is going to hurt their performance [16:18] <Daemon404> i am not a big fan of the xbmc devs' practies. [16:18] <Daemon404> at all. [16:57] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:e6299a4cf9ef: cmdutils: update year [16:57] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:c9a8dfa5ae16: Merge commit '6892d145a0c80249bd61ee7dd31ec851c5076bcd' [16:57] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:9e8464e81bfe: Merge commit '9eef9eb3014b2ed9c3ff4aac510a9f04edb555cf' [16:57] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:23ae7bfb4e94: dnxhdenc: fix mb_rc size [16:57] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:94e2673f4e45: avformat/ape: free packet on avio_read() failure [16:58] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:d35916f6ea49: avformat/mpegts: check sl.timestamp_len [16:58] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:fedbba5ea051: avformat/rmdec: move packet allocation down [16:58] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:5f56e495aea4: avcodec/apedec: more checks for k [17:46] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:32262ca7d781: avcodec/vmnc: Check that rectangles are within the picture [17:46] <cone-745> ffmpeg.git 03Michael Niedermayer 07release/2.1:a8ed3685e193: avcodec/jpeg2000dec: fix error detection in pix_fmt_match() [19:20] <saste> ffmpeg -f lavfi -i testsrc -c:v h264 -map 0 -f segment x-%d.ts [19:20] <saste> michaelni, this leads to a crash ^^ [19:28] <ubitux> saste: for the audio filtering, as i said, there was the memleak/error, and the splitted packets [19:32] <saste> ubitux, there was / there is? [19:32] <saste> i tested with valgring, found no leaks [19:43] <saste> michaelni, was a segment problem... [19:44] <michaelni> ahh ok, i just wanted to take a look :) [19:47] <cone-745> ffmpeg.git 03Reimar Döffinger 07master:76421982d042: lossless_videodsp.asm: fix compilation. [19:53] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:8e5e84c2a2a2: avformat/mov: Ignore the last frame for duration and fps calculation if it looks suspect [20:10] <cone-745> ffmpeg.git 03Stefano Sabatini 07master:b539a72bba30: examples/filtering_audio,video: drop call to avcodec_get_frame_defaults() [20:11] <cone-745> ffmpeg.git 03Stefano Sabatini 07master:a2e78161ce77: lavf/mpegtsenc: fix weird indent [20:11] <cone-745> ffmpeg.git 03Stefano Sabatini 07master:169065fbfb3d: lavf/segment: remove duplicated and inconsistent cleanup code in seg_write_packet() [20:11] <cone-745> ffmpeg.git 03Stefano Sabatini 07master:f57baf743f0e: lavf/segment: drop pointless variable oc from seg_write_packet() [20:43] <ubitux> saste: i guess that's ok [20:45] <cone-745> ffmpeg.git 03Diego Biurrun 07master:766df7ca89a2: dxva2: Add missing #includes [20:45] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:075989c33c84: Merge commit '766df7ca89a2398e71182f5f2b46053e3aa9bd69' [20:46] <saste> ubitux, so, shall we close the ticket? [20:46] <ubitux> go ahead [21:01] <cone-745> ffmpeg.git 03Diego Biurrun 07master:ade4ecb42d2d: dxva2: Use correct printf format strings [21:01] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:57a0c5fa9ca2: Merge commit 'ade4ecb42d2dacd18d04eb8df2afa8131e5ad653' [21:08] <cone-745> ffmpeg.git 03Kostya Shishkov 07master:e91a3f1bdba9: dxtory: correctly handle YUV slices with average odd height [21:08] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:9eb954b91d36: Merge commit 'e91a3f1bdba9b4945e42c191d2e35e9844625fb4' [21:25] <cone-745> ffmpeg.git 03Kostya Shishkov 07master:025fd76e1a26: dxtory: change error code for unexpected slice configuration [21:25] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:a298f934ec9b: Merge commit '025fd76e1a2623c858d8c686a73cc30980a314b0' [21:44] <cone-745> ffmpeg.git 03Anton Khirnov 07master:e0ab5078a7d8: lavc: do not force the emu edge flag [21:44] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:8a77baae6e79: Merge commit 'e0ab5078a7d865f8f6fd6a6d3cbe0f380ead4a3d' [21:54] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:eb01a25fe145: swscale: fix stride used in planarToNv12Wrapper() [21:54] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:797d807869f0: Merge commit 'eb01a25fe1452740a7f3ae2cbaff356a5c6e7806' [22:00] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:9047491f8bcd: swscale: add nv12/nv21->yuv420 converter [22:01] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:53feab7a4e99: Merge commit '9047491f8bcd87673eed55fb310647a03b0981e9' [22:09] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:7597e6efe492: swscale/x86/rgb2rgb: add support for AVX [22:09] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:ef25595b7181: Merge commit '7597e6efe492cb2449bb771054d64cc7fdf62ff5' [22:13] <saste> merging lavd into lavf, any thought? [22:14] <kierank> see my response [22:14] <saste> kierank, "oh my god no" that's not really an argument :) [22:15] <nevcairiel> i agree with kierank's arguments [22:24] <JEEB> is this still about that one person trying to make lavd take in lavf packets and do VO with it or whatever? [22:24] <nevcairiel> nah this is about a certain anton wanting to merge them [22:33] <wm4> that shit makes me semi-mad... if someone really wants to add video and audio output to ffmpeg, why not create a new API, instead of coercing it into the muxing API? [22:34] <nevcairiel> isnt most of the stuff in there input anyway [22:34] <wm4> yes, I guess it used to be for capturing video from various sources [22:35] <wm4> and then someone had the idea to use it for output [22:37] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:91c981857bc6: rgb2rgb_template: add MMX/SSE2/AVX-optimized deinterleaveBytes [22:37] <cone-745> ffmpeg.git 03Michael Niedermayer 07master:977abf9aedec: Merge remote-tracking branch 'qatar/master' [23:02] <Daemon404> michaelni, did you get a chance to look at that sample? [23:02] <Daemon404> since i noticed tou didnt merge justins patch [23:02] <Daemon404> s/tou/you// [23:05] <michaelni> Daemon404, see 8e5e84c2a2a21a979b48e80c5a8dd44754ab3f21 [23:06] <Daemon404> im not sure i think thats a great idea [23:06] <Daemon404> a lot of the problem files we get are only slightly longr or shorter than the track says [23:07] <Daemon404> but enough to make it noticable [23:07] <Daemon404> e.g. 8:20 vs 8:30 [23:08] <nevcairiel> didnt someone comment on the patch, saying the duration used there is derived from the wrong data, which is why it produces wrong fps results [23:08] <Daemon404> well [23:09] <Daemon404> more like track duration is container defined [23:09] <Daemon404> and not related to the real turation [23:09] <Daemon404> container lies. [23:09] <Daemon404> which is why it should be removed [23:10] <wm4> is there a way yet to update track duration mid-playback [23:11] <Daemon404> im going to have to keep patching ffmpeg in production if thats the solution... [23:14] <michaelni> Daemon404, can you share a file where the my change fails = [23:14] <michaelni> s/=/?/ [23:15] <Daemon404> i dont have a ginat collection ive saved [23:15] <Daemon404> but in general i dont understand why we choose to believe the container [23:16] <nevcairiel> the container also has the timestamps [23:16] <nevcairiel> you gotta believe it somehow [23:16] <Daemon404> well yes [23:16] <nevcairiel> isnt the duration just computed by adding up all timestamps or something [23:16] <Daemon404> you can [23:16] <Daemon404> but duration is a single coded value in a mov [23:16] <nevcairiel> mp4 is one of these formats where the info is actually neatly available [23:17] <Daemon404> and its wrong [23:17] <Daemon404> a lot. [23:17] <Daemon404> you can also get teh duration by parsing timestmaps, or doing stuff with stts [23:17] <nevcairiel> the code michaelni fixed looked like it was doing something with stts, iirc [23:18] <nevcairiel> not reading the container duration header [23:18] <Daemon404> its making up a sample_duration as far as ican tell [23:18] <Daemon404> based on heuristics [23:19] <nevcairiel> i guess the last stts can also be easily wrong, since it can contain a much larger sample count then it has actual samples [23:19] <nevcairiel> who would ever notice [23:19] <Daemon404> i rely on avg_frame_rate ;) [23:19] <michaelni> stts data in fps.mov contains a lost sample thats very long [23:19] <michaelni> lAst [23:20] Action: nevcairiel still uses r_frame_rate for some reason [23:20] <Daemon404> dont worry [23:20] Action: nevcairiel forgot if there was a difference [23:20] <Daemon404> we're dumping avg_frame_rate soon [23:20] <Daemon404> using our own code [23:20] <Daemon404> nevcairiel, yes [23:20] <Daemon404> theyre entirely different things [23:20] <nevcairiel> i'm not too worried about the framerate being wrong in any case [23:21] <nevcairiel> its really just informational [23:21] <nevcairiel> cant assume its cfr anyway, since you never know [23:21] <Daemon404> i dont [23:21] <Daemon404> this is for picking an output framerate [23:22] <Daemon404> we normalize to cfr [23:22] <Daemon404> not somethign lav would have to o [23:22] <Daemon404> do [23:22] <nevcairiel> thats really the only thing it might be used for in directshow too, determining the frame rate to switch your display to [23:22] <nevcairiel> s/frame/refresh/ [23:23] <Daemon404> that never works well for me unless i use d3d rendering [23:23] <Daemon404> which is super annoying since mpc-hc doesnt render a seekba then [23:23] <JEEB> exclusive mode that is [23:23] <Daemon404> yes [23:23] <nevcairiel> mpc-hcs built in switcher is stupid [23:24] <Daemon404> and yet nothing better exists [23:24] <nevcairiel> can only hope the new vesa vfr standard makes it into consumer hardware [23:25] <nevcairiel> (and not only gaming displays to battle vsync issues) [23:25] <JEEB> yeah [23:25] <JEEB> user-based refresh would be nice [23:25] <Daemon404> ... consumer hardware doesnt even implement vesa stuff rght as it is [23:25] <nevcairiel> no need to change refresh, just pain the image when it comes in [23:25] <nevcairiel> paint* [23:25] <Daemon404> and you expect it to do that? [23:25] <Daemon404> lol. [23:25] <nevcairiel> AMD is pushing it for gaming [23:26] <nevcairiel> maybe TVs will also pick it up [23:26] <nevcairiel> because of consoles [23:26] <nevcairiel> who knows [23:31] <michaelni> Daemon404, about e.g. 8:20 vs 8:30 the heursitic will drop the last frame from the average if its bigger than 10 times the average, a 10sec long last frame will always be droped [23:31] <Daemon404> hmm ok [23:32] <michaelni> also either way i want this to be as accurate and fast as possible i dont care if its done with that heuristic or something else [23:34] <Daemon404> we will see how it works in practice [23:34] <Daemon404> whenever we roll out a new build [00:00] --- Wed Jan 22 2014

1 0

ffmpeg.log.20140121
by burek 22 Jan '14

22 Jan '14

[00:30] <l_r> why do i get [alsa @ 0x14e2d80] sample format 0x15002 is not supported from ffmpeg -f alsa -c:a aac -i hw:0,0 -f flv rtmp://myserver [00:53] <pyBlob> Hello there [00:54] <pyBlob> I'm trying to do some audio capturing and I'm curious about the different raw sample formats ... what's the difference between the formats with "p" and without p? [01:25] <Gambi_> hi is there a way to get remaining time in place of time ? (when it's reading a videofile with -re ) [01:25] <jkli> does ffmpeg support image filters for thumbnails? [01:26] <jkli> i saw some effects being applied on thumbail images like contrast optimization and wondered if they can be done with ffmpeg [01:26] <jkli> or if it has to be done externally through php [01:27] <Gambi_> you need to use filters or complex filters [01:29] <jkli> is there a link where all ffmpeg filters are shown with their effects? [01:30] <Gambi_> http://trac.ffmpeg.org/wiki/FilteringGuide [01:31] <jkli> no images [01:32] <Gambi_> http://trac.ffmpeg.org/wiki/FancyFilteringExamples [01:33] <jkli> ah ok [01:33] <jkli> thx [02:17] <l_r> my god the builtin aac codec consumes 100% cpu [02:17] <l_r> in my raspberry [02:18] <l_r> i wonder if libfaac or others make any difference [02:40] <Hello71> builtin aac is bad [03:02] <albator> does creating multiple out like -f flv rtmp url -f mp4 output.mp4 raise cpu ? [03:11] <klaxa> albator, most likely? [03:11] <klaxa> it's doing more stuff, right? [03:11] <albator> I never felt this but on this particular encoding, the difference is obvious [03:11] <albator> so much i can't broadcast basically :) [03:12] <albator> broadcast and keep a local file [03:12] <klaxa> what about putting the copy into flv too? [03:13] <albator> how is that? [03:14] <albator> i tried to do -f flv rtmp//stream -f flv output.flv [03:14] <albator> but it s a no go [03:15] <albator> i need to make a local output at least , then i ll just send using -c:v copy etc [03:15] <klaxa> why is the -f output.flv thing a no go? [03:18] <albator> for local file with .flv extension, it says [03:18] <albator> Could not write header for output file #1 (incorrect codec parameters ?): Invalid data found when processing input [03:18] <jkli> is there a way to add a line into the user data of an mp4 file? [03:19] <jkli> like a url? [03:20] <klaxa> hmm [03:20] <klaxa> i'm not too sure how ffmpeg handles multiple output files, but i'm guessing then that it will not use the same output options for the second output as for the first output [03:21] <klaxa> meaning that even if you used the same settings, it will encode everything twice [03:21] <klaxa> with piping you can encode it once and use -c copy in another ffmpeg instance to get it to work without encoding twice [03:21] <albator> still pretty curious it doesn't work when requesting same output extension [03:22] <albator> while it works with -f flv & -f mp4 [03:22] <klaxa> like i said [03:22] <albator> hmmm [03:22] <klaxa> ffmpeg probably applies different encoding options to different outputs [03:22] <albator> u must be right [03:23] <klaxa> anyway, maybe try something like this: ffmpeg -i <input> [encoding options] -f flv pipe:1 | ffmpeg -i - -c copy rtmp:// -c copy output.mp4 [03:23] <albator> i have been using piping and never felt this before [03:23] <klaxa> that will probably convolute your terminal though [03:23] <albator> but with the concat muxer, i went to something else, and using multiple output without piping [03:23] <albator> so that explain this [03:23] <albator> very good, i m sure u fix my problem [03:23] <albator> thanks [03:24] <albator> i ll first use piping to make both stream+local file , then concat the resulting local file [05:30] <liam-kelly> How do I compile ffmpeg with support for h265/vp9? [05:34] <liam-kelly> Nevermind, I reread the ubuntu compilation guide and it is included in libvpx. [06:10] <Guest66336> while compiling ffmpeg i'm getting error /usr/bin/ld.bfd.real : libavutils/aes.o : reloaction R_ARM_THM_MOVW_ABS_NC against a 'local symbol' cannot be used when making a shared object, recompile with -fPIC [06:10] <Guest66336> libavutils/aes.o : could not read symbols : Bad value [07:59] <anshul> I was using mjpeg codec from libavcodec, I am getting [mjpeg @ 0xaaf14d20] overread 8 message in the new git code, while I am not getting any error in the fourier release of ffmpeg [08:02] <anshul> can any one suggest what should I do to remove [mjpeg @ 0xaaf14d20] overread 8 message. should I look to my input or ffmpeg code, or should I comapare forrier release code and git code. [08:07] <gnos> Hey, is there a way to use ffmpeg to cut off, say, the leftmost 2/5 of a video stream? [08:16] <anshul> @gnos you want to cut the video frame? [08:18] <gnos> Yeah, I want to crop out the left 2/5 of the frame [08:24] <gnos> I'm lookint at http://libav.org/libavfilter.html#crop, but I don't really understand it. [08:36] <kvz> I'm having trouble with this ffmpeg command + output: http://pastebin.com/yMt7P1fx - does anyone know what the "real" error in this output is? [08:38] <CommaCrazy> hi all, question, how can I restream a mpts stream and just add buffer to the stream using ffmpeg? [09:05] <Keestu> is it possible to capture rtsp streaming data using ffmpeg in android?. i am new to ffmpeg, sorry for basic question. [09:17] <sspiff> I'm having a weird memory leak, which valgrind says is caused by me calling avcodec_decode_subtitle2() - yet I always call avsubtitle_free() when *got_sub_ptr is non-zero. [09:18] <sspiff> Does anyone have a clue what I might be missing? Is there something else I need to free after calling avcodec_decode_subtitle2()? [09:19] <ubitux> anshul: if it's a regression, please open a ticket [09:19] <ubitux> gnos: cut the width? [09:20] <ubitux> kvz: [mp4 @ 0x24c6a00] Malformed AAC bitstream detected: use audio bitstream filter 'aac_adtstoasc' to fix it ('-bsf:a aac_adtstoasc' option with ffmpeg) [09:20] <ubitux> sspiff: not that i know of [09:21] <sspiff> ubitux: argh! I've been breaking my head over this. [09:21] <sspiff> I'm no doubt doing something retarded [09:21] <ubitux> sspiff: call free even if no *got_sub_ptr [09:21] <ubitux> maybe [09:21] <kvz> ubitux: The bsf:a aac_adtstoasc is already passed though.. So, broken input file? It happens on a few input files. [09:23] <ubitux> kvz: you need to use bitstream filter inside the segmenter [09:23] <sspiff> ubitux: didn't change anything :( I'll continue looking [09:23] <ubitux> sspiff: what's the leak? [09:24] <ubitux> kvz: probably :) [09:25] <sspiff> I'm not sure, I can't get line numbers for ffmpeg to appear in valgrind [09:26] <sspiff> I'm rebuilding ffmpeg from source and going to link against that [09:26] <sspiff> ubitux: http://pastie.org/8652952 [09:26] <sspiff> but that probably doesn't mean much to you :) [09:27] <ubitux> does it leak with ffmpeg itself? [09:27] <sspiff> when using the ffmpeg command? [09:27] <sspiff> haven't tried [09:27] <sspiff> I need to go to a meeting, I'll try when I get back [09:28] <ubitux> kvz: mmh, maybe start simplifying your cmd line [09:28] <ubitux> your crazy x264 settings are not needed anymore [09:28] <ubitux> then can you reproduce with a simple transcode? [09:35] <anshul> there must be some @gnos you are using scale command, its function is not to crop [09:35] <anshul> @gnos you are using scale command, its function is not to crop [09:36] <anshul> I was using mjpeg codec from libavcodec, I am getting [mjpeg @ 0xaaf14d20] overread 8 message in the new git code, while I am not getting any error in the fourier release of ffmpeg [09:36] <anshul> an any one suggest what should I do to remove [mjpeg @ 0xaaf14d20] overread 8 message. should I look to my input or ffmpeg code, or should I comapare forrier release code and git code. [09:36] <anshul> sorry I didn't got what is regression [09:37] <anshul> sorry I also mixed up with gnos question and kvz output [09:44] <gnos> anshul, can you please tell me how to use the crop function? The docs don't explain it very well [09:52] <anshul> gnos: can you tell me, what you tried yet. so that I may understand that the document need to be more elaborated or its only problem with you [09:53] <ubitux> http://ffmpeg.org/ffmpeg-filters.html#crop the doc is here btw. [09:54] <gnos> I haven't tried anything yet, I don't know what to try. I'm using livestreamer to watch my friends twitch.tv streams, but he has a placeholder for his webcam that he doesn't have yet, and I just want to see the part of the stream that actually matters [09:54] <gnos> ubitux, I looked at that and I just don't understand it [09:54] <ubitux> ffplay randomvideo.avi -vf crop=<try-something> [09:55] <ubitux> gnos: and that's not the doc you pasted [09:55] <ubitux> there are examples below [09:55] <gnos> Oh. Is it using the libav syntax? [09:56] <ubitux> you probably want crop=2/5*in_w [09:56] <ubitux> "the libav syntax"? [09:56] <ubitux> libav is a fork, their doc is outdated [09:56] <gnos> Ok, I see [09:56] <anshul> gnos:http://trac.ffmpeg.org/wiki/FilteringGuide try this if you are using ffmpeg [09:56] <gnos> I was reading on the forum that it uses the libav syntax. I guess they're just full of it [09:57] <anshul> gnos:basic of applying filters [09:57] <ubitux> what you say makes no sense gnos [09:57] <gnos> On "a forum" I meant [09:57] <ubitux> not that part [09:57] <gnos> I don't know why they would say that [09:57] <gnos> anshul, that link doesn't work [09:58] <renihs> works for me [09:58] <renihs> link url is valid :) [09:58] <gnos> I messed up [09:59] <gnos> Ok, I guess I'll just play around with it then [09:59] <gnos> Was hoping I' [09:59] <anshul> ubitux:should i discuss about regression on developer forum, i dont know how to decide regression [10:01] <ubitux> anshul: open a ticket on the trac [10:02] <ubitux> saying "it works with version N, doesn't with version N, here is a sample, command line and full output in both versions: [...]" [10:02] <ubitux> (doesn't with version M*, you get it) [10:04] <anshul> ubitux thanks [10:13] <sspiff> ubitux: problem is present but less severe with ffmpeg (I'm using a different version of the same sample, my tool uses MPEG PES streams for testing, not transport streams [10:13] <sspiff> ubitux: http://pastie.org/8653090 [10:14] <ubitux> use ./ffmpeg_g [10:17] <sspiff> ubitux: sorry, re-running [10:18] <DeadSix27> woah [10:18] <DeadSix27> i just foudn out how useful ffmpeg will be for what i planned [10:18] <sspiff> ubitux: http://pastie.org/8653106 :) [10:19] <DeadSix27> -> test.txt -> "file '..' | file '..'" -> "ffmpeg -f concat -i test.txt -c copy out.mkv" [10:19] <DeadSix27> that was the most satisfying thing i found out ever. [10:19] <sspiff> I assume this means there's a leak in the dvbsub decoder? Or am I jumping to conclusions? [10:20] <sspiff> DeadSix27: good for you :) [10:20] <renihs> hmm is there a way to record the audio output (speakers) with ffmpeg/alsa only? [10:20] <renihs> or do i need jack/alsaloop/pulse/something for that [10:20] <ubitux> sspiff: please share a sample :) [10:21] <sspiff> ubitux: you mean the input file? Sure thing! Do I upload to my own storage or is there a place for these things? [10:22] <gnos> can ffmpeg read from stdin? [10:22] <sspiff> it will appear here in a short while: https://dl.dropboxusercontent.com/u/197940/SubTitleHD.ts [10:23] <sspiff> gnos: yep [10:23] <gnos> Do I have to pass it any options to make that happen? [10:23] <sspiff> cat ~/Documents/extract_subtitles/samples/SubTitleHD.ts | ./ffmpeg_g -i - -map 0:#0x1580 -scodec dvdsub foo.mkv worked for me [10:24] <sspiff> so just point to "-" as input [10:25] <sspiff> ubitux: upload is done [10:26] <anshul> does anyone know what does the Ah and Al parameter means in function mjpeg_decode_scan file libavcodec/mjpegdec.c [10:26] <gnos> How do I pipe it's output to another program after that? [10:26] <anshul> write pipe:1 instead of outputfilename [10:27] <gnos> Ok, thanks [10:30] <CommaCrazy> how can I restream a mpts stream and just add buffer to the stream using ffmpeg? [10:43] <gnos> Is it possible to use the filters in ffplay? [10:46] <gnos> The man page mentions getting a list of all filters, and detailed information about a filter, but not whether it's possible to use a filter [10:57] <strigga> hi, quick quesiton: is there a way to tell avconv not to use 100% of my memory? [11:01] <sspiff> ubitux: should I file a bug? [11:01] <saste> strigga, !fork [11:02] <ubitux> sspiff: probably [11:08] <sspiff> I'll look into the leak this afternoon [11:08] <sspiff> perhaps I can fix it easily. If not, I'll file a bug. [11:53] <sspiff> weird, the leak originates from an av_mallocz for creating subtitle rects, but they seem to be free'd cleanly in avsubtitle_free [11:58] <sspiff> the indentation of dvbsub_display_end_segment seems broken [12:10] <sky2> average: hello [12:10] <sky2> microchip_: hi [12:11] <average> sky2: :) so you've returned [12:11] <average> I thought you gave up [12:12] <average> nice to see someone's ambitious [12:12] <sky2> average: no. im not a guy to give up so easily [12:13] <sky2> im always a learner & really wish to learn linux stuufs [12:13] <sky2> stuffs* [12:14] <sky2> until i get my mission fulfilled of converting .flv files to .3gps or vice -versa [12:14] <average> interesting [12:14] <average> so what have you brought to the table today ? [12:15] <sky2> right now im heel bent on getting .flv converted into mobile compatible .3gp & wish to enjoy them on my mobile :) [12:16] <sky2> hell* [12:16] <sky2> come what may..i must get it done :p [12:16] <average> sure, have you read the links I gave you yesterday ? [12:17] <average> show me some new commands and some output/errors [12:18] <sky2> average: it was too late in the night here (midnite) & was dog tired & a bit off-mood for not having my desired job done.. so could not afford myself watching the full video that you provided me with.. [12:19] <sky2> i was speaking of yesterday.. [12:19] <average> the video was like 3 minutes long, but nevermind that, you can watch it today [12:19] <sky2> yea [12:19] <sky2> sure [12:19] <sky2> i bookmarked it [12:20] <sky2> will watch it today for sure [12:21] <sky2> average: may I PM you if you dont mind pls [12:23] <average> sky2: pm me [12:24] <sky2> thanks a ton [12:29] <clu3> hi all, im relatively novice with ffmpeg, i would like to split an mp3 files into multiple smaller continous mp3 files, is this possible with ffmpeg? [12:30] <saste> clu3, segment muxer [12:33] <average> clu3: continous ? [12:33] <clu3> saste, thanks. i did a quick google and still hopeless [12:33] <clu3> average, i mean if those split files are concatenated together, they will produce the original mp3 file [12:34] <saste> clu3, why don't you read the ffmpeg documentation instead? [12:34] <average> ffmpeg -i file.mp3 -ss <start> -t <duration> -vn piece.mp3 [12:34] <average> clu3: ^^ [12:35] <average> clu3: saste knows what he's talking about. you will think that "wow, that was so easy, let's see what else I can do" . but to find out, you have to read the documentation and try stuff out.. [12:36] <average> <start> is in the format HH:MM:SS and <duration> is a number of seconds [12:36] <average> I think <start> is even HH:MM:SS.mm [12:36] <average> where mm is milliseconds [12:36] <clu3> average, thanks alot. @saste, I did, but it's really not easy to find out, as I'm quite dumb with media and stuff [12:37] <clu3> average, any chance if it is possible to split into multiple files at once? like the input would be an array [12:37] <average> clu3: you see what I mean ? [12:37] <average> you just did it [12:37] <average> you did exactly what I predicted you would do [12:37] <average> before you would even do it [12:37] <clu3> average, ^^ [12:37] <average> clu3: now move your ass and go read some docs for your own sake [12:38] <clu3> wow you ffmpeg guys are so hardcore [12:38] <average> saste: am I hardcore ? is he right ? [12:39] <average> I don't see myself as hardcore [12:39] Action: relaxed is softcore [12:39] <clu3> seriously, for a utility where I just need a command line to split it, this is pretty basic command and i've googled, i didn't find it [12:40] <average> clu3: listen, ffmpeg is not for *everything* , you might need some bash here and there to achieve what you want... [12:40] <clu3> to read the whole documentation to use it is really too much, hence i'm here [12:40] <saste> clu3, segment muxer, ffmpeg-formats(1) [12:40] <saste> you can use it to split the input into multiple files, containing a given media duration [12:41] <saste> also splitting and joining is not as simple as people think [12:41] <average> clu3: nobody forced you to read the whole documentation. there is stackoverflow, there is avp.stackexchange.com ... [12:41] <average> clu3: there are mailing lists where people talk about this [12:41] <clu3> plus, look at this http://www.ffmpeg.org/documentation.html , where is a novice like me supposed to go [12:41] <average> clu3: it's not like "omg, poor clu3 , he doesn't have any place to read about this stuff.." [12:41] <clu3> average, as i told u i already googled [12:41] <clu3> and SO always comes first in the result [12:42] <average> clu3: You did not google, do not contradict me otherwise I will google and I'll prove you wrong from the very first page of google search results ! [12:42] <saste> clu3, http://ffmpeg.org/ffmpeg-formats.html#segment_002c-stream_005fsegment_002c-… [12:42] <saste> that implies some basic understanding of how multimedia works [12:48] <clu3> average, really i do not wanna argue with u but just so you know. This is what i got http://stackoverflow.com/questions/43890/crop-mp3-to-first-30-seconds [12:49] <clu3> now if u look at the answer and with a media-dumb guy like me, that's even more confusing [12:49] <clu3> that's why i wanna talk directly to you guys in here [12:49] <clu3> i was like "wtf is transcoding" when i read that stackoverflow answer [12:53] <clu3> nvm, best wishes guys [12:53] <clu3> i gotta go [12:53] <clu3> peace [13:21] <sspiff> average: he just doesn't have enough background knowledge to understand the documentation/SO/the google results. [13:21] <sspiff> which begs the question if he can pull off what he's trying, but still [13:54] <sspiff> ubitux: found it! [13:56] <sspiff> there's a mismatch between DVBSubContext.display_list_size and the actual number of entries in DVBSubContext.display_list [13:57] <sspiff> allocation is based on the display_list_size, but the number of rects can subsequently be lowered if there are fewer entries in the list, and there's no free there [13:57] <sspiff> I'm going to change allocation to happen based on the actual number of entries, if that's alright with you? [14:03] <ubitux> sspiff: #ffmpeg-devel :) [14:03] <ubitux> sspiff: i'm not the maintainer of that code [14:32] <xlinkz0> i'm trying to record from a webcamera using ffmpeg, i set -r 30 but i get a message saying 'The driver changed the time per frame from 1/30 to 1/10 [14:33] <xlinkz0> i can watch video at 30fps with vlc [14:57] <Hello71> !pastebin [14:57] <Hello71> !triggers [15:07] <xlinkz0> Hello71: http://codepad.org/3l2q8vJV [15:11] <sky2> average: i have been able to convert the said .flv into a .3gp using ffmpeg [15:11] <sky2> its playing successfully in my mobile [15:11] <sky2> :) [15:12] <sky2> \o [15:31] <average> sky2: nice work :) [15:31] <average> sky2: \o/ [15:32] <sky2> average: ty [15:33] <average> np [15:33] <sky2> now concentrating on how to convert an.mp4 to .3gp [15:33] <sky2> :p [15:35] <NiksenX_> Hey guys! Could you please suggest me channel about p2p? [15:40] <__raven_> anyone there [15:44] <spaam> nope. everyone is dead. [15:47] <__raven_> i need to build a quality test chain for some params - anyone who did such already? [16:08] <Keshl> __raven_: As far as I know, no, and it's not really possible. Each video is different. One level of -crf or bitrate or whathaveyou can alter your footage differently than another. [16:09] <Keshl> In general, though, if you're using x264 and -crf, 23 is "sane" and 18 is "visually lossless". If you're worried about people pausing your video and being all OCD, I've found that 11 makes that appear lossless. [16:10] <Keshl> Of course your mileage may vary. You may be able to use higher values (Thus smaller sizes and bitrates) without knocking off quality if your video moves more slowly or only has action in one area. [16:12] <__raven_> Keshl: i am talking about technical valuas as like as fokus, dropouts, blackframes, light and so on [16:41] <sspiff> seems like I can't figure out how to disable format probing for my AVIOContext [16:52] <sspiff> Can anyone tell me why avformat_open_input is calling ff_id3v2_read for an MPEG2 TS? [17:33] <x_> What is the best way to capture your desktop at a high frame rate like 60 fps. Right now if use ffmpeg -f x11grab -i :0.0 -r 60 -s 640x480 I can barely get 30 fps. FFMPEG is only using at around 20-30% cpu as well. Now 640x480 is not what I am looking for I would like to do 1080 or at the very least 720 resolution. [17:40] <klaxa> x_ maybe your storage is too slow [17:41] <x_> klaxa, Really? IOtop was reporting it using under 1MB/s [17:42] <JEEB> you should first of all try how fast capture by itself is [17:42] <JEEB> -f null and set output to - (slash) [17:42] <JEEB> and if it doesn't do rawvideo then, add rawvideo to be the output "encoder" [17:44] <x_> -f null? [17:44] <JEEB> yes [17:44] <JEEB> as in "don't output it anywhere" [17:44] <JEEB> it should default to raw video and audio "encoders" in that case [17:44] <JEEB> but I'm not sure :P [17:44] <spaam> but null is somewhere! :D [17:44] <JEEB> and you need the slash [17:45] <JEEB> because ffmpeg wants an output descriptor [17:45] <JEEB> even if it's just stdout (which will get nothing because of -f null) [17:50] <sky2> average: could you pls suggest how the command should be to convert an .mp4 to a .3gp ? I used this which says that there is " No such file directory " whereas the file is on the Desktop & I changed the current directory to cd Desktop [17:50] <sky2> here goes the command: [17:51] <sky2> ffmpeg -i source.mp4 -s qcif -vcodec h263 -acodec aac -ac 1 -ar 8000 -r 25 -ab 32 -y output.3gp [17:52] <x_> so what you're saying something like ffmpeg -f x11grab -s 1920x1080 -i :0.0 -r 60 -f null [17:52] <sky2> x_: u r asking me ? [17:53] <x_> JEEB, my apologies [17:53] <sky2> x_: no prob [17:53] <JEEB> x_, and the slash for the output "file name" [17:53] <JEEB> otherwise ffmpeg will error out [17:54] <sky2> JEEB: the flv i was trying with to convert to .3gp was sucessfully got converted & worked fine on my phone [17:54] <x_> JEEB, of course, it reports 60 fps so encoding would be a bottleneck? [17:55] <JEEB> x_, so it keeps the 60fps when going through the capture? [17:56] <x_> JEEB, yes it keeps 60 fps [17:56] <JEEB> x_, ok, now add in the output settings -c:v libx264rgb -preset ultrafast -crf 0 [17:56] <JEEB> and test that [17:56] <JEEB> still keep -f null [17:56] <JEEB> it will encode the video with libx264 into lossless RGB [17:57] <JEEB> (and then proceed to throw out the result, but yeah) [17:57] <x_> JEEB, ah it appears to work at 60 fps with libx264 but how do we get something we can keep? [17:58] <JEEB> now remove -f null and the slash, and make it out.mkv or so [17:58] <JEEB> it will thus mux it into matroska [18:00] <x_> as soon as I do that it reports ~30 fps the file is pretty small [18:00] <JEEB> uhh [18:00] <JEEB> post full command line and the output [18:00] <JEEB> in a pastebin [18:00] <JEEB> and link here [18:00] <x_> okie doke [18:02] <x_> http://pastebin.com/Fqq8uEvp [18:03] <JEEB> you forgot the -crf 0 :P [18:03] <JEEB> also your input seems to be read as 30fps [18:03] <JEEB> looking by that [18:03] <JEEB> Stream #0:0: Video: rawvideo (BGR[0] / 0x524742), bgr0, 1920x1080, 1988667 kb/s, 29.97 tbr, 1000k tbn, 29.97 tbc [18:04] <x_> rame= 705 fps= 26 q=13.0 Lsize= 4350kB time=00:00:29.11 bitrate=1223.9kbits/s? [18:04] <x_> is that not to be trusted? [18:04] <JEEB> it just says the runtime encoding speed [18:04] <x_> huh [18:04] <x_> wait [18:05] <x_> So what is capping it at 29.97 I wonder. [18:05] <JEEB> I thought about saying this before, but try moving the -r to before -i [18:05] <JEEB> because as far as I can see this is doing exactly what I thought it'd do [18:05] <JEEB> as in, converting to 60fps [18:06] <x_> ah [18:06] <x_> that makes sense. [18:06] <x_> So how do we improve encoding speed then? [18:07] <x_> of course faster hardware but perhaps there i something I am missing. [18:07] <JEEB> well to be honest this doesn't really look too promising looking at the x264 enabled asm line "using cpu capabilities: MMX2 SSE2Fast LZCNT" [18:08] <JEEB> you could try removing -c:v libx264 -preset ultrafast -crf 0 and switching to -c:v ffvhuff [18:09] <x_> JEEB what other capabilities will it "use"? [18:09] <JEEB> x_, what? [18:10] <x_> JEEB, Perhaps I did not ubild libx264 with flags that would be useful? that my cpu supports of course :P [18:10] <JEEB> nah [18:10] <JEEB> it doesn't get limited like that [18:11] <JEEB> only your cpu limits it, everything gets built [18:11] <JEEB> anyways, also don't jump around [18:11] <JEEB> 1) test just capture and -f null with raw video "coding" 2) add the video encoder 3) add file output [18:12] <JEEB> (just noting because you suddenly switched to 1080p and all that) [18:12] <JEEB> also, I have to note that the ffmpeg version you're using is pretty old. [18:13] <x_> JEEB, Yeah I should compile 2.x somewhere and take alook at that. [18:13] <JEEB> anyways, libx264rgb with preset ultrafast and crf 0 will be lossless and RGB [18:13] <JEEB> so will ffvhuff [18:14] <JEEB> those are pretty much the fastest stuff you can get around [18:14] <JEEB> so if that goes too slow [18:14] <JEEB> then you're out of luck :P [18:14] <x_> JEEB :) indeed! thanks a bunch [18:15] <JEEB> (crf set to 0 with a 8bit x264 enables lossless coding, with >8 bit depth you will have to set qp to zero as crf values get dragged lower because of normalization of lossy values with various bit depth encoding) [18:16] <JEEB> (and preset ultrafast is the fastest x264 preset, its usage for video capture is just fine, otherwise it's not really useful for anything else but encoding speed competitions) [18:16] <JEEB> (I mean, _lossless_ video capture) [18:17] <x_> JEEB, hmm thanks for the info I'll continue reading up on what I can perhaps I'll stumble on something else to ask ^^ [18:17] <JEEB> you can gain some extra speed on the encoder side if you add a colorspace conversion to 4:2:0 YCbCr, but this will cause degradation :) [18:18] <JEEB> -pix_fmt yuv420p , with x264 you have to switch to the normal encoder name (libx264), and with ffvhuff it should work as-is? [18:19] <x_> JEEB I'm interested in speed atm I could drop some quality :P [18:20] <JEEB> yeah, I just really recommend using RGB for the capture in most cases :) Because swscale can do funky things with the conversion, esp. in relation to colormatrix etc. [18:20] <Mista-D> HLS segment adds a lot of TS overhead compared to a source TS file, used "-f hls -c copy". Started with 5 MB file, ended up with 8.2 MB (its a 1 fps slide show at 64kbps). Whenever its TS fat is trimmed (ffmepg -i segment_%03d.ts -c copy -copyts out_segment_%03d.ts), the WebVTT stops working... something related to the TS/CC timing is removed or changed. Is this a ticket, or i'm doing something [18:20] <Mista-D> wrong? [18:20] <x_> Really cool! to know interested in learning some more :D [18:21] <sky2> converted .mp4 to .3gp \o [18:21] <sky2> wow [18:21] <x_> cool beans ;-) [18:42] <x_> JEEB works really well now. thanks for your knowledge I appreciate it :-) [18:43] <JEEB> np [18:44] <x_> am able to squeeze out 60 encoding fps on a phenom 2 1090t [18:45] <x_> with latest git. [18:46] <l_r> is there a way to read both and video and audio via pipe with dd? [18:47] <JEEB> dunno about dd, but if you use a container you could squeeze both :P [18:47] <l_r> dd if=/dev/vdie0 + dd if=/dev/audio | ffmpeg -i - etc..? [18:47] <l_r> the fact is that [18:48] <l_r> this consumes all my cpu ffmpeg -f alsa -i hw:0,0 -c:a aac -f flv out.flv [18:48] <l_r> the aac codec is damn slow [18:48] <JEEB> uhh [18:48] <l_r> i am on a raspberry [18:48] <JEEB> oh [18:48] <JEEB> yeah [18:48] <JEEB> you will barely have CPU time to do anything on that thing encoding-wise [18:49] <l_r> but why so slow? [18:49] <JEEB> you would actually have to use some profiling tool to find that out :P [18:49] <l_r> woould libfaac make any difference? [18:49] <JEEB> no idea, but I wouldn't use faac but fdk-aac [18:50] <l_r> is it a matter of builtin aac vs eterenal? [18:50] <JEEB> no [18:50] <l_r> ok [18:50] <JEEB> as in, it doesn't matter if the thing is internal or external, both can be slow or fast [18:50] <JEEB> you really don't want to throw out overly vague statements :P [18:50] <JEEB> anyways, you might want to do the same thing as I recommended to that other screen capture guy first tho [18:51] <JEEB> test how fast you can capture only [18:51] <JEEB> -f null [18:51] <l_r> and then? [18:51] <JEEB> and the slash symbol as the "output file name" [18:51] <l_r> ok..but what is the goal? [18:51] <JEEB> to see which part is making it slow [18:51] <l_r> it's aac codec [18:51] <l_r> since streaming in wav only takes 50% [18:52] <JEEB> instead of "I have this thing that captures, encodes and muxes and writes it to a file" [18:52] <l_r> "only" [18:52] <JEEB> ok [18:52] <JEEB> yeah, the raspi isn't exactly fast :P [18:52] <l_r> i see [18:52] <l_r> it's a pity [18:52] <l_r> actually h264 of the cam is good for streaming video [18:52] <l_r> but audio needs some encoding [18:52] <l_r> in software [18:52] <JEEB> yes, a hardware ASIC makes things simpler :P [18:53] <JEEB> for such devices [18:53] <JEEB> in any case, if you want to learn what exactly is slow in there [18:53] <JEEB> you profile [18:53] <l_r> does not usb device exist for this? [18:53] <l_r> with an internal aac encoder? [18:53] <l_r> usb-sound cards i mean [18:53] <l_r> cheapy [18:54] <JEEB> noidea [20:33] <llogan> x_: with x11grab try -framerate instead of -r (as an input option) [21:28] <l_r> what is the best aac encoder? [21:30] <klaxa> fdk-aac [21:30] <klaxa> i think that's the one that is referred to by "the best" [21:37] <Irrelium> I am trying to use ffmpeg to record a screencast at 30 FPS, but whenever I have a bit of lag and the framerate drops below 30 FPS, it makes the video speed way up at that point when I play it back, which also causes the video to get ahead of the audio [21:37] <Irrelium> Is there any way I can prevent this? [21:38] <Irrelium> This is the command I am using right now http://pastebin.com/6fy4shQi [21:40] <re-G> try ultrafast preset at least [21:40] <Irrelium> I know that would make the video lag less, but it is still bound to lag at some point even if I use ultrafast [21:41] <Irrelium> And I would like to prevent the audio getting out of sync if it does [21:41] <l_r> klaxa, but is it also faster? [21:42] <klaxa> fast, good, cheap [21:42] <klaxa> pick any two [21:42] <klaxa> Irrelium: move the -r 30 flag behind the input [21:42] <klaxa> otherwise ffmpeg will accumulate 30 frames and put them in one second of the video [21:43] <Irrelium> By behind do you mean before the '-f x11grab'? [21:45] <Irrelium> Well, I am going to assume that is what you mean and give it a try [21:45] <Irrelium> Thanks [21:45] <re-G> -i is input [21:46] <Irrelium> Oh, so it should go at the end of the first line [21:47] <klaxa> yes [21:47] <Irrelium> (I always get confused when people say 'behind something' or 'in front of something' :P) [21:47] <Irrelium> I can never remember which way is which [21:47] <klaxa> :P [21:47] <klaxa> <before -i> -i <behind -i> [21:49] <l_r> ERROR: libfdk_aac not found [21:50] <l_r> where can i find it [21:50] <klaxa> probably in your distro's repos [21:50] <JEEB> you build it and then add its prefix to the search path [21:50] <JEEB> and no, it generally isn't in repos binary-wise [21:50] <JEEB> because the license says that binaries can only be distributed with a license [21:50] <klaxa> not even the library development binaries? [21:50] <klaxa> oh... [21:50] <JEEB> which is why it's nonfree [21:51] <JEEB> the source code can be distributed of course [21:51] <l_r> my only hope is fdk-aac [21:51] <l_r> for the cpu issue [21:51] <l_r> the internal codec is consuming too much [21:52] <JEEB> your only hope is to either have someone write optimized code for your crappy ARM board, or see if any other encoder is better at it :P [21:53] <klaxa> does it have to be aac? [21:53] <l_r> whatever can be in flv [21:53] <klaxa> can't mp3 be in flv? [21:53] <l_r> aac and mp3 re the only one i am aware of [21:53] <l_r> but [21:53] <klaxa> but? [21:53] <l_r> mp3 cannot be read on ipads [21:53] <l_r> by default [21:53] <l_r> you know... [21:53] <l_r> apple... [21:54] <llogan> Irrelium: why use -f flv and then give output name .mp4? [21:54] <llogan> also you should always include the complete ffmpeg console output with yout command [21:55] <llogan> l_r: can you stream copy the audio and then re-encode it later? [21:55] <klaxa> wikipedia states: >MP3, Nellymoser, ADPCM, Linear PCM, AAC, Speex [21:56] <l_r> llogan, hmm...there are thousands of clients...a short answer would be no [21:56] <llogan> you did not provide much information so we are making guesses [21:56] <klaxa> thousands of clients and you're running a server on an ARM board? [21:56] <llogan> libshine may encoder faster than libmp3lame on ARM [21:57] <klaxa> mp3 has the ipad incompatibility thing though [21:57] <l_r> yes [21:57] <Irrelium> llogan: I guess because I didn't notice that :P [21:58] <l_r> only mp3 and aac can be in a flv? [21:58] <Irrelium> I was using this command to stream and modified it to record to a file [21:58] <Irrelium> So I guess I just left the flv in there [21:58] <llogan> l_r: also, if you're using raspberry you may be using Rasbian, and if you're using Rasbian you're probably using Libav, not FFmpeg [21:58] <klaxa> l_r: <klaxa> wikipedia states: >MP3, Nellymoser, ADPCM, Linear PCM, AAC, Speex [21:59] <l_r> llogan, but i am recompiling everything [21:59] <l_r> the version in raspian is years old [21:59] <l_r> speex? [21:59] <l_r> wow [21:59] <l_r> nellymoser <- never heard of it [21:59] <klaxa> this is wikipedia, enjoy with care [22:00] <l_r> what's this nellymoser? [22:00] <klaxa> probably unsupported on ios [22:00] <llogan> it's nothing. [22:00] <llogan> carry on [22:00] <klaxa> http://en.wikipedia.org/wiki/Nellymoser_Asao_Codec [22:01] <JEEB> well, iOS doesn't support FLV to begin with :P [22:01] <JEEB> so his explanations are already all over the place [22:14] <_fil_> hello, I'm stuck with a video in which I want to replace a portion of the video track by another track [22:15] <_fil_> I've tried to cut/merge and so on but it always breaks [22:33] <llogan> _fil_: sounds like you want to concatenate. or maybe overlay. [22:36] <_fil_> yes [22:40] <_fil_> llogan: precisely, I need to overlay video2 over video1, starting from time t (video2 is much shorter than video1) [22:45] <_fil_> how do I specify the timestamp to begin the overlay? http://pastebin.com/kEjFNBDW [22:58] <llogan> _fil_: use timeline. https://ffmpeg.org/ffmpeg-filters.html#Timeline-editing [23:02] <_fil_> hmm and how do I specify that the overlay should start at 00:01:58.680 with the content of video2? [23:05] <_fil_> ffmpeg -i complet2.mp4 -i insert.mp4 -filter_complex "overlay enable='between(t,118,219)'" output.mp4 -t 00:01:58.680 [23:05] <_fil_> " No such filter: 'overlay enable'" [23:05] <_fil_> something with the syntax probably :DD [23:06] <llogan> _fil_: you may need the trim filter as well, but i can't remember and i should be editing videos... [23:06] <llogan> also your ffmpeg may be old. does it have timeline support? ffmpeg -filters [23:07] <llogan> https://ffmpeg.org/download.html#MacOSXBuilds [23:08] <_fil_> ffmpeg version 1.2.4 [23:08] <llogan> note that the online docs are synced nightly to the most recent code, so if you must use an old version then refer to the local docs [23:09] <_fil_> i just reinstalled it with homebrew :) [23:09] <llogan> ffmpeg -filters has a key that will indicate which filters have timeline support [23:09] <_fil_> apparently it's the 1.2 branch [23:09] <_fil_> should I switch to branch 2.x ? [23:10] <llogan> brew install ffmpeg --devel [23:10] <llogan> http://trac.ffmpeg.org/wiki/MacOSXCompilationGuide#Shortcut:CompileFFmpegth… [23:10] Action: llogan gets back to editing [23:10] <_fil_> thank you! [23:17] <_fil_> almost there -- I just need to fix the timestamp on video2 so that I overlay it in sync [23:23] <llogan> _fil_: see setpts/overlay examples in filter docs [23:28] <_fil_> I'm trying hard! [23:28] <_fil_> does this look liek it's going to work? => [23:28] <_fil_> ./ffmpeg-2.1.3 -i incomplet.mp4 -i insert.mp4 -filter_complex "overlay=enable='between(t,118,219)';[0:v] setpts=PTS-STARTPTS;[1:v] setpts=PTS-STARTPTS+118*25" output.mp4 [23:40] <llogan> _fil_: i recommended setpts because: "Be aware that frames are taken from each input video in timestamp order, hence, if their initial timestamps differ, it is a good idea to pass the two inputs through a setpts=PTS-STARTPTS filter to have them begin in the same zero timestamp" [23:41] <llogan> so you would use setpts first: "[0:v]setpts=PTS-STARTPTS[out1];[1:v]setpts=PTS-STARTPTS[out1];[out1][out2]overlay stuff[filtered]" -map "[filtered]" [23:42] <_fil_> I'm lost [23:44] <_fil_> maybe I can just change the pts in my insert.mp4 file so that it begins with 118s of a blank image [23:47] <_fil_> I'm trying this now: [23:47] <_fil_> ./ffmpeg-2.1.3 -i incomplet.mp4 -i insert.mp4 -filter_complex "[1:v] setpts=PTS-STARTPTS-118*25-17 [decal];[0:v][decal] overlay=enable='between(t,118,220)'" output.mp4 [23:47] <_fil_> (each try costs 10 minutes) [23:48] <llogan> add "-preset ultrafast" when testing [23:49] <llogan> otherwise by default it will use -preset medium (assuming your build supports libx264, then it will default to the encoder mpeg4) [23:52] <_fil_> HA! thanks! :) [23:57] <llogan> _fil_: you can use the trim filter to specify which section of the input you want [00:00] --- Wed Jan 22 2014

1 0