Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

March 2014

1 participants
62 discussions

ffmpeg-devel.log.20140325
by burek 26 Mar '14

26 Mar '14

[01:25] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:c25d2cd20b76: avcodec/x86/mpegvideoenc_template: fix integer overflow [03:46] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:a81ee2651faf: avcodec/svq1enc: fix undefined pointer operation [10:31] <j-b> 'morning [10:33] <ubitux> BBB: i forgot to run the 20k encode for vp9 with the others at the beginning, so it's running since 2 days now and should be done tonight [10:33] <ubitux> you'll have full results soon, but here is a teaser: http://lucy.pkh.me/encodeshd/ssim-summary.txt [10:33] <ubitux> vs http://lucy.pkh.me/encodes/ssim-summary.txt [10:48] <ubitux> we need to do something about those :( [10:48] <JEEB> can anyone else test if they can replicate this? Going through `ffmpeg -i in.mp4 -vn -an -c:a aac -strict experimental out.mp4` multiple times seems to actually add the decoder delay into the encoded material without marking it correctly? [10:49] <JEEB> I re-encode a clip without seemingly any decoder delay (duration = 4561920 (00:01:35.040)) and get correct stuff (segment_duration = 95040). Then if I re-encode that, I get lolwut (segment_duration = 95061). [10:49] <pippin> ubitux: did you confirm that the gif encoder/muxer no longer does transparency diffing? [10:49] <JEEB> (values from tkhd and edit list checked by boxdumper) [10:50] <JEEB> 95061 being the tkhd duration of the first re-encode [10:51] <JEEB> tl;dr either I'm misparsing or ffmpeg cli is DoingItWrong [10:51] <ubitux> JEEB: -vn -an -c:a aac? huh? [10:51] <ubitux> pippin: how to reproduce? [10:51] <JEEB> ubitux, typoe [10:51] <JEEB> -sn [10:52] <JEEB> basically to make sure that it's only audio [10:52] <ubitux> wm4: not that simple, but feel free to do it [10:52] <wm4> ubitux: why not? [10:52] <ubitux> wm4: you probably want to change ff_subtitles_read_chunk() [10:52] <pippin> ubitux: bayer dithering and error diffusion ending up with roughly the same file size in a gif anim - that didn't happen in august when I was working on (other) dithering methods [10:53] <wm4> ubitux: the only hard part is not getting flamed and shitted on by nicolas [10:53] <ubitux> yeah nicolas had a complex WIP for that [10:53] <ubitux> but i havent heard any news since a while about that [10:53] <pippin> ubitux: the commandlines I used then, were of this form: $ ./ffmpeg -i bbb-original.mp4 -vf scale=320x180:sws_dither=bayer -r 10 -ss 40 -frames 100 bayer.gif [10:53] <ubitux> so please go ahead [10:54] <ubitux> pippin: did you try to revert to the commit you think cause problem, then drop it, and check if that was indeed the cause? [10:55] <pippin> opening up the gif anims in gimp, showed me that every frame in both versions are complete frames with no transparency [10:55] <ubitux> wm4: i support the utf16 "hack" in the chunk reader [10:56] <ubitux> wm4: but you need to change ff_get_line() and ff_subtitles_read_chunk() [10:56] <wm4> why is there both [10:56] <pippin> ubitux: I tried bisecting and building other versions; but at the moment I do not have sufficient battery/energy in my laptop to pursue further testing [10:56] <ubitux> wm4: because that's not the same thing [10:57] <ubitux> wm4: read_chunk() is "chunk" based (multi line read) [10:57] <ubitux> pippin: just try what i said [10:57] <ubitux> git checkout <the-hash-you-pointed-out> [10:57] <ubitux> do the test [10:58] <ubitux> git reset --hard HEAD^ [10:58] <ubitux> do the test again [11:00] <pippin> ubitux: I tried bisecting the problem, but came up empty handed - and at the moment I am unable to allocate my laptop battery to building ffmpeg [11:01] <ubitux> that's just one compilation [11:02] <pippin> the 1h40min of battery I have left, will be 10min left after that [11:03] <ubitux> pippin: note that transparency won't work with palette [11:04] <pippin> it used to work, and is the reason bayer / a_dither / x_dither are half the file size of error diffusion at http://pippin.gimp.org/a_dither/ [11:04] <ubitux> make sure it's not using pal8 [11:05] <ubitux> i could add a warning in case pal8 is used with gif encoder [11:06] <ubitux> it doesn't have any benefit [11:06] <pippin> why? [11:06] <ubitux> why why? [11:07] <pippin> if pal8 is 2 bit red, 4 bit green 2 bit blue; then it surely has benefit - as you can see on that page [11:07] <ubitux> pal8 can change palette, so it's not simple to handle properly [11:08] <ubitux> i think you're looking for rgb4 [11:08] <ubitux> or bgr4 [11:08] <ubitux> or simply, just use rgb8/bgr8 [11:09] <pippin> I think rgb8 is what ffmpeg-land calls what I refer to and use here [11:09] <ubitux> anyway, pal can not be used with transparency [11:09] <ubitux> or well, without larger heuristic [11:10] <ubitux> see the todo in lavc/gif.c [11:11] Action: pippin has no desire to dig himself much deeper into things; I provided patches half a year ago [11:12] <pippin> and when I updated the unapplied patches; there are unrelated regressions making the actual contribution pointess ;p [11:12] <ubitux> did you send them to ffmpeg-devel? were they forgotten? [11:12] <ubitux> if you want them not to break, make sure they go upstream [11:12] <ubitux> and add a test [11:12] <pippin> I nagged michaelni in here, and he reworked 2/3 of the patches but ignored the actual new dithering methods after my refactoring of making them fit in [11:13] <ubitux> pippin: where is the thread? [11:14] <pippin> my current re-iteration of it, was sent to ffmpeg-devel - and there has been no response on the list to it, but libav has contacted me off-list :] [11:14] <pippin> http://ffmpeg.org/pipermail/ffmpeg-devel/2014-March/155851.html [11:14] <ubitux> i saw that thread [11:14] <ubitux> i'm asking for the thread with the patches [11:15] <ubitux> where there maybe is review and requests to make it acceptable for upstream :p [11:15] <pippin> that was done ad-inifinitum in here in august [11:16] <ubitux> i wonder what you want then [11:16] <ubitux> irc is not a good way of keeping track of forgotten review [11:16] <ubitux> & patches [11:17] <pippin> which I guess is why I send it to the mailinglist instead after discovering the regressions? [11:18] <pippin> the actual features in there are pointless with the regression in the gif encoding [11:18] <ubitux> ah i see the patches are in the link.. you won't get review like that [11:18] <ubitux> can you workaround the regression with -pix_fmt rgb8 ? [11:19] <pippin> that is a slight rework of the 4th patch in a series; the start of the series were applied - which were the refactoring to make things able to deal with more dithering methods [11:25] <wm4> ubitux: looks I talked too early - there are somewhat deep dependencies on AVIOContext [11:26] <Compn> pippin : probably michaelni forgot about it. ping him :) [11:27] <Compn> michael is a busy guy, but also he doesnt want to miss patches [11:27] <ubitux> Compn: can't be applied, there is a regression he's trying to fix [11:28] <ubitux> i still don't know if it's because of the pix_fmt and if the suggested workaround works though [11:28] <Compn> ah [11:29] <pippin> ubitux: I am _not_ trying to fix the regression, I am trying to report the regression [11:30] <pippin> the patch with the new features is updates and is a gift that can be accepted or not [11:30] <pippin> s/updates/updated/ [11:32] <ubitux> pippin: i don't see the patch on the mailing list (just a link to follow with maybe not up-to-date patches, in a mail asking for information, not patch submit), for a regression not easily reproducible, which you don't seem to be able to bisect yourself [11:32] <ubitux> what do you expect from us? [11:34] <pippin> interest in the regression at least [11:35] <ubitux> well, i would if i could reproduce [11:36] <pippin> try to transcode a video to GIF and see if it treats all frames as I-frames fully encoded, or if it does transparency based delta encoding? [11:37] <pippin> or at least reply to the email, stating what is done wrong. if things should have been reported in a bug tracker, or if patches should be sent in a different way to the list [11:38] Action: pippin understands why such unresponded email yields off-list replies from the fork :] [11:38] <BBB> michaelni: do you remember why vp7 wasn't merged? I got that question from across the isle [11:39] <BBB> ubitux: yay! [11:39] <BBB> also, SLOW!!! (omg for real man) [11:39] <ubitux> :) [11:39] <ubitux> pippin: [11:39] <ubitux> -rw-r--r-- 1 ux ux 3.3M Mar 25 11:39 notrans.gif [11:39] <ubitux> -rw-r--r-- 1 ux ux 1.8M Mar 25 11:39 trans.gif [11:39] <ubitux> ./ffmpeg -ss 40 -i ~/samples/big_buck_bunny_1080p_h264.mov -vf scale=320:240 -r 10 -frames 100 -gifflags -transdiff -pix_fmt rgb8 -y notrans.gif [11:39] <ubitux> ./ffmpeg -ss 40 -i ~/samples/big_buck_bunny_1080p_h264.mov -vf scale=320:240 -r 10 -frames 100 -gifflags +transdiff -pix_fmt rgb8 -y trans.gif [11:39] <ubitux> works for me [11:40] <Compn> BBB : you have the last email in the vp7 thread, according to my email client [11:40] <BBB> Compn: and I said it was OK right? [11:40] <BBB> at least that's what I remember [11:40] <Compn> BBB : peter ross suggested splitting and copying the code into seperate decoders so as not to slow down vp8 by .25% [11:41] <Compn> > Sorry, lots of other stuff went around, I forgot. [11:41] <Compn> > Do you have specific ideas on what could cause the 0.25%? I am inclined to say that 0.25% is OK, given that it gives us a new decoder. I'm not happy to give up 0.25% all over the place but we sometimes do it (e.g. to add threading, or to support partition-threading in vp8, etc.) so it's not completely outrageous... [11:41] <Compn> BBB : ^^ thats your mail [11:41] <ubitux> pippin: your mail was too vague to get any answer [11:41] <BBB> hm... I still think it's OK to merge [11:41] <BBB> 0.25% isn't awful [11:41] <Compn> BBB : what about splitting the decoders up , having duplicated code ? [11:41] <pippin> ubitux: none of "-gifflags +transdiff -pix_fmt rgb8" was needed half a year ago, powering up generator and firing up a compile [11:42] <Compn> BBB : i'm not arguing, just curious [11:42] <ubitux> pippin: -gifflags +transdiff is not mandatory, it's the same without [11:42] <ubitux> pippin: -pix_fmt rgb8 might not have been mandatory because it was picking that pix format instead of pal8 for some reason [11:43] <ubitux> that is what i told you already [11:43] <Compn> BBB : i mean if you reply and say its OK , fine. but your last mail was confusing :) inclined not happy ?? :) [11:43] <BBB> Compn: nobody is happy when stuff is slower [11:43] <Compn> some people arent english majors [11:43] <BBB> but it's ok [11:43] <ubitux> pippin: now you didn't answer my simple question weither it was workarounding the problem or not [11:44] Action: wm4 is afraid that something slightly more advanced to support utf16 subs will be abused to add 8-bit codepage conversion support [11:44] <Compn> ok [11:44] <ubitux> pippin: if you can workaround the problem with -pix_fmt rgb8, then you can probably submit a patch [11:44] <BBB> Compn: duplicate code isn't a great idea, I think splitting decoders is fine but put common code in a shared file (header if you want to inline it)... but *effort* [11:44] <pippin> ubitux: I am checking that now,. and to do that I have started a gasoline powered generator :] [11:45] <Compn> coding off a generator :D [11:46] <BBB> too bad pross isn't here [11:47] <pippin> while coding/idle my laptop uses ~8 watts; big compiles are best done while charging [11:47] <BBB> michaelni: if it makes any difference, I'm still fine with merging vp7 as-is (I'd prefer pross was here, but ohwell) [11:50] <Compn> better to commit without waiting for pross [11:50] <Compn> hes busy [11:51] <Compn> i bugged him about bayer patch for years :) [11:51] <pippin> ubitux: with -pix_fmt rgb8 # it works [11:52] <pippin> ubitux: and then the question might be, if having to force the format for bayer (and my two methods) is a regression or not. [11:52] <ubitux> it's unreleated [12:13] <pippin> ubitux: thank you for your help - now the people that want to encode GIF anims with those methods can use master + the patch :) [12:19] <ubitux> would be nice if they could just have to use master [12:59] <michaelni> BBB, IIRC i didnt apply as i read the last mail from you as a question (about the 0.25%) and so waited ... [12:59] <michaelni> ill check if the patches still apply [13:15] <BBB> michaelni: hm, I see... no I think 0.25% is fine [13:21] <michaelni> ok, seems they still apply [13:31] <cone-700> ffmpeg.git 03Peter Ross 07master:7d4c0220f18a: avcodec/vp8: vp8_find_free_buffer [13:31] <cone-700> ffmpeg.git 03Peter Ross 07master:89f2f5dbd7a2: On2 VP7 decoder [13:31] <cone-700> ffmpeg.git 03Peter Ross 07master:4a031ce3c634: avformat/riff: add On2 VP7 fourccs [13:31] <cone-700> ffmpeg.git 03Peter Ross 07master:a490970af2f0: libavcodec/*/vp8dsp_init: indent [13:35] <michaelni> pippin, about the patch [13:35] <michaelni> theres "fprintf (stderr, "eeek1\n");" in it [13:36] <michaelni> and fprintf (stderr, "eeek2\n"); [13:58] <cone-700> ffmpeg.git 03Øyvind Kolås 07master:3e6016622e6f: swscale: add two spatially stable dithering methods [13:59] <ubitux> cool :) [14:06] <michaelni> any other patches that where "forgotten" and should be applied ? [14:22] <pippin> michaelni: \o/, I was about to fix up the patch by removing both the default and auto cases; other parts of the code should ensure that neither ever is the case. [14:43] <Daemon404> i see someone added GDI screen grabbing (cool) [14:43] <Daemon404> maybe ill go ahead and add the new (way, way better) screen grab API [14:43] <Daemon404> (win8+ only though) [14:45] <JEEBsv> yeah, the D3D 11 one seems rather naisu [14:46] <Daemon404> it is [14:50] <wm4> and who adds D3D output` [14:51] <wm4> ? [14:51] <Daemon404> output? [14:51] <Daemon404> it's input [14:52] <Daemon404> you get frames. [14:53] <wm4> Daemon404: I know [14:53] <wm4> but libavdevice can do output too now [14:53] <Daemon404> oh it was a trol [14:53] <Daemon404> nobody cares about output [14:53] <Daemon404> maybe one guy. [14:53] <wm4> hue [14:54] <wm4> that's almost mean [14:54] <Daemon404> but is it unture? [14:54] <Daemon404> untrue* [14:55] <Daemon404> www.virtualdub.org/blog/pivot/entry.php?id=356 [14:55] <Daemon404> what i was refering to btw ^ [14:55] <Daemon404> could is there to look at in vdub too [14:56] <Daemon404> it works really well [14:56] <kierank> Daemon404: wow gabest replies in that thread [14:57] <av500> oh, virtualdub [14:57] <Daemon404> kierank, gabest regularily replies to avery lee's posts [14:57] <JEEBsv> yeah [14:57] <Daemon404> also <3 avery lee [14:57] <JEEBsv> aye <3 [14:57] <av500> +1 [14:57] <av500> once he fixed vdub for me :) [14:57] <Daemon404> lol [14:58] Action: kierank used nandub [14:58] <av500> actually, I sent him a patch [15:02] <Daemon404> kierank, nandub has been irrelevant since the divx 3.11 days [15:02] <Daemon404> sorry [15:02] <Daemon404> divx .11 ;-) [15:02] <Daemon404> s/ [15:02] <kierank> I am aware of that [15:03] <Daemon404> ... ill stop typing now. [15:08] <ubitux> kierank: gpu encode? :( [15:08] <kierank> where [15:09] <ubitux> i don't know i'm waiting for your writeup [15:09] <ubitux> someone asked me [15:10] <kierank> I forgot about that [15:10] <kierank> I need time [15:10] <ubitux> asking me about stuff like https://developer.nvidia.com/nvidia-video-codec-sdk [15:10] <kierank> ubitux: tthat is different [15:10] <ubitux> and i must say i have not enough experience to know what to say [15:10] <kierank> GPU encoding and GPGPU encoding are not the same [15:11] <kierank> the former is just a hardware encoder on teh gpu [15:11] <kierank> the latter is using algorithms running on the main gpu [15:14] <ubitux> i'd really love an article anyway :° [15:14] <ubitux> i need to motivate myself to write something about the subtitles [15:14] <ubitux> wm4: you think you're going to be able to do sth about utf16? [15:14] <kierank> I need to start hacking on the dvbsub encoder [15:15] <ubitux> we need to move the subs to lavu and inject them into libavfilter :( [15:20] <wm4> ubitux: yes but tomorrow [16:56] <cone-700> ffmpeg.git 03Carl Eugen Hoyos 07master:e6fe804bdd62: Do not set swscale sizeFactor to -1. [16:56] <cone-700> ffmpeg.git 03Carl Eugen Hoyos 07master:0816abae1ed1: Add an .oga muxer. [17:52] <cone-517> ffmpeg.git 03rogerdpack 07master:10056bc964f7: dshow: log error and continue if unable to set audio buffer [18:02] <cone-517> ffmpeg.git 03rogerdpack 07master:4d9881e1cc20: udp: mention default in documentation [18:02] <cone-517> ffmpeg.git 03James Almer 07master:2d9821a2081a: x86/cpu: check for OS support before enabling AVX2 [18:30] <llogan> kierank: did you see if emotion fixed things as they claimed? i'll close the ticket if they did. https://trac.ffmpeg.org/ticket/3454 [18:32] <kierank> oh forgot about that [18:32] <kierank> i'll look when i am at home [19:25] <cone-517> ffmpeg.git 03Vittorio Giovara 07master:6ee55c7b111e: vf_transpose: K&R formatting cosmetics [19:25] <cone-517> ffmpeg.git 03Michael Niedermayer 07master:10e931d51477: Merge commit '6ee55c7b111eec6c51dfdc0fddbd46f5dd867fa3' [19:39] <cone-517> ffmpeg.git 03Vittorio Giovara 07master:aa499568afc0: avconv: More descriptive message about framedrop [19:39] <cone-517> ffmpeg.git 03Michael Niedermayer 07master:41e7e46cac50: Merge commit 'aa499568afc01d59215eef7e5b14b949a9671afc' [19:52] <cone-517> ffmpeg.git 03Vittorio Giovara 07master:dc0c70e018f6: avcodec: add missing includes [19:52] <cone-517> ffmpeg.git 03Michael Niedermayer 07master:8f20e3d4dfe5: Merge remote-tracking branch 'qatar/master' [22:04] <kierank> llogan: seems to be fixed [00:00] --- Wed Mar 26 2014

1 0

ffmpeg.log.20140325
by burek 26 Mar '14

26 Mar '14

[00:00] <RenatoCRON> because if the 'thing' is moving slowing in the second, may i don't get it. but if I check this only with 1 sec between thoses checks, less chace to that occour [00:06] <vl4kn0> Hi, I'd like to use scale video filter in my code, to rescale whatever format the codec uses to AV_PIX_FMT_GRAY8. I specify dst_format=AV_PIX_FMT_GRAY8, frames get filtered successfully (no error is produced) but I still get the original pixel format in the frames I load from buffersink. Any ideas why? [00:07] <llogan> RenatoCRON: i don't understand your question [00:08] <vl4kn0> this is the output I get http://fpaste.org/88222/57024951/ [00:12] <RenatoCRON> llogan, please a moment, i'm testing it and I will explain/output to you all [00:23] <RenatoCRON> llogan, here it is: http://pastebin.com/raw.php?i=ZNRxFfGu [00:23] <RenatoCRON> all commands and logs [00:24] <RenatoCRON> and comments! [00:26] <RenatoCRON> clearly use -vf fps=fps=1/8 is not the same as use a new video file. but my question is: is possible to, I don't know, use "gte(t,n_forced*1)*gt(scene\,0.01)" and scene so will only compare with 1 sec ago scene ? [00:34] <RenatoCRON> if I use select='not(mod(n\,4))*gt(scene\,0.009)' it's works, but scene is too low, isn't ? And I see the select filter saying -1 because of not(mod(n\,4)) ^ [00:34] <RenatoCRON> but it's fine. I will split then check! [01:35] <julienb> Hello, i'm looking for a tutorial to install the last version of ffmpeg on debian wheezy please [01:35] <sacarasc> Unless you're using it for rather specific things, try one of these links: [01:37] <julienb> sacarasc : Does the static version will be as powerfull as the compiled version? [01:37] <sacarasc> It won't have some of the non-free things (like fdk aac thingie) or X capturing, otherwise, yes. [01:38] <julienb> sacarasc i have 3.10 version [01:38] <julienb> of kernek [01:38] <julienb> kernel [01:39] <sacarasc> Then grab the 1) one, if you can live without the non-free/X capping. [01:39] <julienb> it's just to make some mp4 encoding in console [01:41] <julienb> scarasc with the static version i see the command will be ./ffmpeg [01:41] <julienb> how i can include it on my environment so i use it like a normal commande ffmpeg [01:42] <julienb> so i have not to write always the entire path [01:43] <sacarasc> If you remove your system ffmpeg (which is libav, I think) then you could put it in your PATH and it will work just as `ffmpeg` [01:45] <julienb> so i have to put the exec command in /sbin/ ? [01:45] <sacarasc> You could put it in /usr/local/bin or even ~/bin if it is in your PATH. [01:46] <julienb> ok, the same for qt-fastart? [01:47] <sacarasc> That is no longer needed, use -movflags faststart as output options. [01:49] <julienb> ok thanks a lot [01:51] <wodim> how can i loop a video to another video? i mean, i have a video and i want to make a new video which is the original video looped 100 times. can i do that? [01:55] <relaxed> wodim: which container? [01:56] <wodim> no specific one [01:56] <wodim> (you mean input or output?) [01:56] <rjp421> im currently trying to find how to loop my input so it replays the video over and over, transcoding to a live RTMP stream.. only thing i can find on google is use -loop 0 on image files only... [01:56] <wodim> yep, i read about that too [01:57] <relaxed> wodim: mkvmerge -o combined.mkv 1.mkv +2.mkv +3.mkv +4.mkv [01:59] <julienb> i download the last static version of ffmpeg but it show me ./ffmpeg -version [01:59] <julienb> ffmpeg version N-61445-gdded5ed [01:59] <julienb> i download ffmpeg.static.64bit.latest.tar.gz [01:59] <llogan> march 14. i already told burek about it. [02:00] <llogan> i mean that is from march 14 [02:00] <julienb> is this the last version 2.2 ? [02:00] <llogan> a could times a year that build machine shits out, so sometimes it stops until someone notices [02:01] <llogan> *couple [02:01] <sacarasc> julienb: That is the latest release, yes. [02:01] <julienb> ok thanks [02:02] <llogan> that is from git master, commit dded5ed from March 14. [02:02] <relaxed> wodim: mkvmerge -o combined input.mkv "$(for i in {1..99}; do printf %s "+input.mkv ";done)" [02:02] <sacarasc> Oh, I misread, sorry. [02:02] <julienb> sorry i have another question, i have to copy it to /usr/bin or /usr/sbin? [02:03] <sacarasc> You could put it in /usr/local/bin or even ~/bin if it is in your PATH. [02:03] <wodim> relaxed: what package contains mkvmerge? [02:03] <relaxed> mkvtoolnix [02:03] <julienb> but i just saw my /usr/local/bin and directory is emtpy [02:03] <llogan> relaxed: why not concat demuxer instead of mkvmerge? [02:03] <julienb> does it normal there are no command ? [02:04] <llogan> yes, for Debian and spawn, unless you compiled something [02:04] <wodim> relaxed: thanks [02:04] <julienb> ok so all my personal executable go there? [02:04] <julienb> usr/bin and usr/sbin are more for os commands right? [02:05] <sacarasc> sbin is for super user executables. [02:05] <relaxed> llogan: I've found mkvmerge to be more reliable [02:05] <julienb> yes but i have /usr/sbin and /usr/local/sbin [02:05] <julienb> both will working? [02:06] <sacarasc> julienb: And they should only be used for super user binaries. You could put the ffmpeg binary ANYWHERE in your path and it would work, but sbin/ directories are definitely not where you should. [02:07] <julienb> i don t understand, if sbin are the super user executables then it should go there [02:08] <sacarasc> Super user = root. [02:08] <julienb> yes [02:08] <sacarasc> If you are running it as root, you're doing it wrong. [02:08] <julienb> why? [02:08] <sacarasc> Because it's some random binary you're picking up from the Internet. [02:09] <julienb> you mean this static package is not 100% sure [02:09] <llogan> gusari is croatian for pirate [02:09] <sacarasc> Do you know who made it? [02:09] <julienb> no [02:09] <sacarasc> You should only run things as root that you *NEED* to. [02:09] <relaxed> it is known [02:10] <julienb> ok thanks [02:10] <julienb> so better to create an user to use it [04:22] <julienb> Hello, i contact you again because i tryed the static package of ffmpeg but when i launch a command i have an error message: [04:22] <julienb> Unknown encoder 'libfaac' [04:23] <sacarasc> That's because faac is non-free. [04:23] <julienb> but how i can use it? [04:23] <julienb> i need it for my encoding [04:23] <sacarasc> There are other AAC encoders. [04:24] <julienb> but i don't understand i was using this option on my other distribution with ffmpeg [04:24] <julienb> and it was working [04:25] <sacarasc> It is not legal to distribute libfaac statically compiled into a binary. [04:25] <relaxed> libfdk-aac is recommended for aac encoding, but you'll have to recompile ffmpeg with support for it. [04:25] <julienb> so i can't use the static package? [04:25] <sacarasc> I did warn you about the non-free things. [04:26] <julienb> so what i have to do? [04:26] <julienb> to make encoding of mp4 using libx264 and libfaac [04:26] <julienb> or libfdk-aac [04:26] <sacarasc> To get everything you want, you'll probably have to compile yourself, but you don't want libfaac anyway, it's poo. [04:27] <llogan> either 1) usethe native AAC encoder, or 2) compile ffmpeg to support the encoders you want 3) pipe from ffmpeg to your desired encoder and remux [04:28] <julienb> the aac native encoder is as good as libfaac? [04:29] <llogan> generally, no, but with enough bits it's probably good enough [04:29] <julienb> i don't know what to do [04:29] <julienb> i was looking for compile it myself at first but i didn't find a good tutorial [04:29] <sacarasc> https://trac.ffmpeg.org/wiki/UbuntuCompilationGuide should work. [04:29] <llogan> try -codec:a aac -strict experimental -b:a 192k. then listen to it. if it is acceptable use it. [04:30] <julienb> my original command was -acodec libfaac -ab 96k -ar 44100 [04:30] <julienb> so i put -acodec aac -strict experimental -ab96k -ar 44100 [04:31] <llogan> why -ar 44100? [04:31] <julienb> because sometimes its highger [04:33] <julienb> sometimes original audio is 48000 hz [04:33] <julienb> so i can optimize a bit [04:33] <julienb> without alterate a lot the quality [04:34] <julienb> the option -strict experimental is important? [04:42] <llogan> julienb: without it you can't use -acodec aac [04:43] <llogan> also see https://trac.ffmpeg.org/wiki/AACEncodingGuide [04:51] <julienb> ok thanks [11:23] <Bumble-Bee> Hello all [11:24] <Bumble-Bee> is there any way to pipe output a transcoded mp4 ? im looking at using the stream for live view on chromecast ? [12:30] <Guest34071> hello, anyone have ffmpeg binary for windows 32bit with libfdk_aac? [13:12] <relaxed> Guest34071: if they did, sharing it would violate the gpl. [13:16] <Guest34071> source code of ffmpeg and libfdk for the binary are available so should be no violation of gpl, but I read there is some problem with patent or something like that and then perhaps there is no problem in sharing outside USA [13:17] <JEEBsv> Guest34071: the licenses are not compatible [13:17] <JEEBsv> completely unrelated to any patent-related stuff [13:20] <Bumble-Bee> is there any way to use ISMV (smooth stream) back through pipe:1 ? [13:23] <Guest34071> JEEBsv: I am really stuck and enable to compile ffmpeg on windows and no tutorial or instructions works, maybe here are someone not caring about licenses violation to help noob me [13:25] <relaxed> Guest34071: http://ffmpeg.zeranoe.com/builds/ There's guides on compiling windows binaries. [13:25] <relaxed> also a forum where you can ask for help [13:33] <roger21> hello [13:33] <roger21> can't i create an .oga file [13:34] <roger21> same for ogx [13:34] <roger21> there arn't supported alredy ? [13:35] <roger21> they seems pretty hamless though (well for oga at least) [13:35] <Guest34071> relaxed: new script to test http://ffmpeg.zeranoe.com/blog/?p=269#more-269 , hope it works this time [13:36] <roger21> oh if i say -f ogg that will trick it :p [13:36] <roger21> that's sad though [13:36] <roger21> probably not the right way [13:36] <roger21> why no oga? [13:37] <relaxed> roger21: I'm assuming oga is ogg without a video stream, so -f ogg is right. [13:39] <roger21> well yes, but .oga is the official audio/ogg i would expect it to be native, not tricked into [13:39] <roger21> https://wiki.xiph.org/MIMETypesCodecs [13:40] <roger21> ogv is supported for instance [13:40] <relaxed> .oga is ogg, there are no tricks [13:40] <roger21> yes, but ffmpeg doesn't know about it [13:40] <roger21> that is sad [13:40] <roger21> it should [13:40] <relaxed> send a patch [13:41] <roger21> yes i will, thank you [14:22] <pyBlob> when playing an audio-file using ffplay, it displays a quite detailed fft, can ffmpeg render this to a video-file, so that I can display it in another player? [14:37] <relaxed> pyBlob: you could encode it with ffmpeg's x11grab [14:37] <relaxed> but that's probably not what you're after [14:38] <pyBlob> yep ... this will probably also cause synchronization issues [16:00] <share> hi [16:00] <share> "No pixel format specified, yuv422p for H.264 encoding chosen." [16:08] <Zeranoe__> share: I don't think you asked any question, but thanks for sharing [16:09] <share> I didnt [16:09] <share> lol [16:18] <ValdikSS> Please confirm or refute the following bug https://trac.ffmpeg.org/ticket/2975#comment:5 [16:25] <roger21> is there a static compilation available with libfdk_aac or libfaac or is it not possible? [16:26] <Zeranoe__> roger21: That would defeat the purpose of non-free, you cannot distribute with non-free [16:28] <roger21> you mean you cannot mix gpl with non-free? (because distributing non-free can be allowed depending on the liscence) [16:42] <roger21> i don't get it, reading the liscence of FDK AAC it seems pretty free to me [16:43] <roger21> they only talk about the patent but the patent would applly to any aac implementation as mp3 and so on [16:43] <Bumble-Bee> mixing licenses is messy [16:44] <Bumble-Bee> but you can include a non gpl module / lib and link into that from gpl code and not invaidate gpl [16:44] <JEEBsv> roger21: IIRC it has a note that you can't distro binaries if you haven't licensed the format, which is an extra limitation compared to (L)GPL, thus making the license not-(L)GPL [16:44] <Plorkyeran_> gpl doesn't allow imposing additional restrictions, and fdk-aac's license has restrictions the gpl doesn't [16:44] <Plorkyeran_> so you can't distribute the combined work under the gpl [16:44] <Plorkyeran_> and you can't distribute it under fdk-aac's license [16:45] <JEEBsv> also I remember there being something with the LGPL that makes it nonfree, I don't remember the exact reasonings [16:45] <Bumble-Bee> LGPL can be included in nonfree/nonglp packages as a prebuilt [16:46] <Bumble-Bee> but any changes to the library covered by LGPL have to be public [16:49] <roger21> the only "limitation" in fdk aac is to include the original liscence and say it is not the original if you modified the code, it is hard to call it a limitation [16:49] <JEEBsv> ugh, feel free to talk with the people who actually did the decisions [16:50] <JEEBsv> all I know is that the license is incompatible with both LGPL and GPL [16:50] <JEEBsv> partially because of the distribution limitation [16:50] <roger21> which there isn't either [16:50] <roger21> http://sourceforge.net/p/opencore-amr/fdk-aac/ci/master/tree/NOTICE [16:50] <Plorkyeran_> > You may not charge copyright license fees for anyone to use, copy or distribute the FDK AAC Codec software or your modifications thereto. [16:51] <Plorkyeran_> > Your modified versions of the FDK AAC Codec must carry prominent notices stating that you changed the software and the date of any change. [16:51] <Plorkyeran_> those are both gpl-incompatible [16:51] <roger21> hmmm [16:51] <roger21> ok then iguess [16:52] <Plorkyeran_> it's in the awkward range where it's a pretty permissive license that just happens to not line up well with other licenses [16:52] <JEEBsv> yes, because they went and wrote their own :D [16:52] <JEEBsv> instead of using a known one [16:54] <Bumble-Bee> anyone know if ISMV can be piped into stdout (pipe:1) ? [17:00] <roger21> man! the lib-faac is under gpl v2 you can't call it the limitation [17:03] <roger21> oh my bad, ther is another liscence [17:04] <roger21> which doesn't seems friendly (i won't read it) [17:04] <JEEBsv> roger21: faac's case is different [17:05] <JEEBsv> it was originally released as GPL [17:05] <JEEBsv> until someone noticed that it used reference code [17:05] <JEEBsv> that was under a license NOT compatible with GPL [17:05] <JEEBsv> so yodawg happened :P [17:05] <roger21> :) ok [17:12] <share> what codec do you recommend for analog stereo [17:12] <share> mp3 or ..? [17:13] <LithosLaptop> opus :) [17:14] <share> ffmpeg is using vorbis by default [17:14] <share> opus? [17:14] <share> lossy [17:15] <LithosLaptop> opus is lossy yes [17:15] <share> thats what i said :p [17:17] <share> -f alsa -i hw:1 -c:a opus ? [17:27] <share> maybe i need to install opus [17:49] <ValdikSS> share: libopus [18:55] <llogan> pyBlob: https://trac.ffmpeg.org/wiki/EncodeforYouTube#Usingfilters [19:16] <llogan> when will windows users learn to provide text instead of screenshots for console outputs? [19:47] <zap0> when they stop being retarded [19:56] <voip> Hello guys [19:57] <voip> Can you give me advice how to keep runing ffmpeg after crash (i have last version of ffmpeg). [19:58] <sacarasc> After what crashes? [19:58] <voip> my ffmpeg crashes every day 1 -2 times [19:58] <sacarasc> Write a script to restart it if the exit code isn't 0? [19:59] <voip> sacarasc, okt thanks [20:01] <voip> sacarasc, bdw do you have tested script example restarting ffmpeg after crash ? [20:07] <sacarasc> voip: http://996b7a3ffb1f23fe.paste.se/ That might work, it's untested and I've not really used bash for years. If you want it on Windows, I have NFI, though. [20:09] <voip> sacarasc, thank you i will try [20:11] <llogan> voip: why does it crash? [20:13] <voip> llogan, actualy i dont know, i didnt see any debug messages [20:13] <llogan> is it easy to reproduce? [20:14] <voip> llogan, yes its hapens 1-2 times in 24 hors [20:14] <llogan> can other people reproduce the issue? [20:15] <voip> i tink yes, i just coping stream [20:15] <llogan> provide the input file if it is required [20:16] <voip> ok [20:17] <pyBlob> llogan: that's what I've search for: "showspectrum" =) [20:24] <voip> llogan, fflogger, http://pastebin.com/C3MVA6M8 [20:28] <llogan> fflogger is a bot [20:29] <llogan> voip: that's not the latest ffmpeg and it doesn't look like it crashed [20:30] <voip> oh, let mi time paste new logs [20:32] <voip> with new version ffmpeg, and same parametrs i cant run ffmpeg http://pastebin.com/WftR0NQf [20:38] <llogan> you have conflicting options. you can't stream copy and use options such as -r and -s. [20:38] <llogan> your old ffmpeg is encoding. your new one is stream copying. [21:20] <voip> llogan, its 100% same command copy-paste, please chek http://pastebin.com/b4dKvxCh [21:54] <voip> ? [22:43] <voip> I have 2 versions of ffmpeg, 100% same command i can run on old version but have problem when i tryng to run on new version http://pastebin.com/VNjEvq1g [00:00] --- Wed Mar 26 2014

1 0

ffmpeg-devel.log.20140324
by burek 25 Mar '14

25 Mar '14

[00:14] <cone-365> ffmpeg.git 03Nicolas George 07master:26dea7731e00: lavf/concatdec: allow to match streams by id. [00:14] <cone-365> ffmpeg.git 03Nicolas George 07master:27cff4f6f8e0: tools: add dvd2concat. [00:14] <cone-365> ffmpeg.git 03Nicolas George 07master:521707d42e1e: ffmpeg: sub2video: send a last blank frame before closing. [00:14] <cone-365> ffmpeg.git 03Nicolas George 07master:c0c1fe797cdf: lavu: add myself as dynarray.h maintainer. [00:14] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:3788b8dbe698: Merge remote-tracking branch 'cigaes/master' [00:26] <cone-365> ffmpeg.git 03Luca Barbato 07master:fb7127b2c836: build: h264 parser depends on full h264 decoder [00:26] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:0dd4eff588c5: Merge commit 'fb7127b2c836b53ad5168887cd12871312c8708d' [00:41] <cone-365> ffmpeg.git 03Luca Barbato 07master:ede2b451ccb1: build: Sort h264 objects [00:41] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:3bf5c05925b5: Merge commit 'ede2b451ccb1b2317858c7a32784a9b739ba45f4' [00:58] <cone-365> ffmpeg.git 03Vittorio Giovara 07master:e9a77f4bc084: h264: move macroblock decoding into its own file [00:58] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:0626211b561f: Merge commit 'e9a77f4bc084e5174baba3a8bc869ce4d102c4f0' [01:43] <cone-365> ffmpeg.git 03Luca Barbato 07master:f51d0f39c0fb: h264: Split h264 slice decoding from nal decoding [01:43] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:ffd77f94a26b: Merge remote-tracking branch 'qatar/master' [02:34] <cone-365> ffmpeg.git 03James Almer 07master:fa25c4c40064: swresample/resample: mmx2/sse2 int16 linear interpolation [02:39] <cone-365> ffmpeg.git 03James Almer 07master:63dbba655e7b: swresample/resample: sse float linear interpolation [04:07] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:e03c3ac53900: tests/tiny_psnr: Print information about the supported sample types in the help and error texts [04:34] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:8683fa541bdf: fate: add tests for linearly interpolated resampling [05:14] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:c9c0451224fd: swscale/swscale: fix integer overflow [06:07] <cone-365> ffmpeg.git 03Tanja Batchelor 07master:5b8f0a5eac75: mpeg: K&R formatting cosmetics [06:07] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:fecebb7abab6: Merge commit '5b8f0a5eac75d16376af60a6f6bb79b3e9e6dab7' [06:16] <cone-365> ffmpeg.git 03Peter Krefting 07master:54f7881615a5: configure: Remove dcbzl check for e500v1 and e500v2 architectures [06:16] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:9456a8628b40: Merge remote-tracking branch 'qatar/master' [10:39] <j-b> 'morning [13:34] <ubitux> heh, libav advertises ffmpeg features :) [13:36] <ubitux> is there someone going to write a release news for 2.2? [13:38] <cone-898> ffmpeg.git 03Anton Khirnov 07master:0e2c3ee9a335: lavc: give a typedeffed name to the AVPacket.side_data struct [13:38] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:bd34e26b27e9: Merge commit '0e2c3ee9a335d8a0a5edf0509e222e804d7b2619' [13:49] <michaelni> ubitux, feel free to write release notes [13:49] <michaelni> s/notes/news/ [13:50] <ubitux> i'm not really motivated to [13:51] <michaelni> ok if noone wrtes any then ill copy and paste the previous and put the changelog in them + updated version [13:51] <Case> michaelni, I see not all Vorbis decoding fixes from git made it to 2.2. Is there a reason? [13:51] <michaelni> which are missing ? [13:51] <michaelni> that is which commits /hashes [13:52] <ubitux> michaelni: i think the hilight should be that this release is meant to be the one back in debian [13:52] <Case> 5171ae781a240cac3860c20f9aefc6d1b2c61cac [13:53] <cone-898> ffmpeg.git 03Anton Khirnov 07master:25b32586566f: lavf: add an AVStream field for exporting stream-global side data [13:53] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:595ba3b96258: Merge commit '25b32586566f285d797737863c97a1c5c9c84e2b' [13:56] <michaelni> ubitux, if someone wrtes news or provides me a text i can put that up, but i dont think iam the right person to be "editor" for this, you know iam not that much a english grammer/cosmetics guy [13:56] <michaelni> Case, locally backported [13:57] <michaelni> will be in 2.2.1 [13:57] <ubitux> michaelni: i know :) [13:57] <Case> ok. Thanks [13:57] <ubitux> michaelni: i think we should ask beastd, llogan, or compn :p [13:57] Action: pippin wondered why #ffmpeg was so quiet yesterday; not realising that it wasn't the dev channel :d [14:01] <Compn> what [14:01] Action: Compn runs and hides [14:02] <Compn> we should ask ubitux [14:02] <Compn> :) [14:06] <kierank> is JULIAN GARDNER on irc? [14:07] <ubitux> no [14:07] <ubitux> don't think so [14:13] <cone-898> ffmpeg.git 03Anton Khirnov 07master:5a7e35dd2351: Add replaygain side data type and code for parsing replaygain tags. [14:13] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:83e89787984d: Merge commit '5a7e35dd2351c30bab45177b9482cb8833a0ca78' [14:36] <cone-898> ffmpeg.git 03Anton Khirnov 07master:4a0f6651434c: libavcodec: when decoding, copy replaygain side data to decoded frames [14:36] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:c08e523586b6: Merge commit '4a0f6651434c6f213d830140f575b4ec7858519f' [14:50] <cone-898> ffmpeg.git 03Anton Khirnov 07master:59444c76e6d4: avconv: add stream-global side data to the first demuxed packet [14:50] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:82e6660ae237: Merge commit '59444c76e6d43529a12dbd80b6dd29c6ba4079a9' [15:00] <cone-898> ffmpeg.git 03Anton Khirnov 07master:d161ae0a3790: frame: add a function for removing side data from a frame [15:00] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:3e1f24131ac0: Merge commit 'd161ae0a37900cbd36c1390ca32a56b892c02ab5' [15:10] <cone-898> ffmpeg.git 03Anton Khirnov 07master:06c3cd3c0186: af_volume: support using replaygain frame side data [15:10] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:8de75f703ada: Merge commit '06c3cd3c0186803619bc6aad2d8f06c3e9015d15' [15:16] <cone-898> ffmpeg.git 03Anton Khirnov 07master:15c5a8d22d12: mp3dec: export replaygain tags from ID3v2 [15:16] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:1034fbdf5300: Merge commit '15c5a8d22d12d29a364ca2ab6438f1dee2fa08c7' [15:29] <Case> michaelni: the other Vorbis fix is missing too. I didn't notice as it was in a different source file. Its commit is 742d8601031fc69748f62fe7c5164a22a0751021 [15:43] <cone-898> ffmpeg.git 03Anton Khirnov 07master:b4679ecdbd80: {v,a}f_showinfo: print frame side data [15:43] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:15fe694e5ca1: Merge commit 'b4679ecdbd80ae050094733f137ccd7ebf1316ab' [15:44] <michaelni> Case, locally backported [15:45] <michaelni> if anyone finds other things that shuld be backported ping me or backport yourself and send pul req [15:47] <cone-898> ffmpeg.git 03Anton Khirnov 07master:781d97fd54b1: avconv: remove unused nb_frames_dup [15:47] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:c31b3e3878e6: Merge commit '781d97fd54b1f0d8f4f7536dbd932dca6445476e' [16:00] <cone-898> ffmpeg.git 03Anton Khirnov 07master:f759f66e367b: avconv: explicitly report when the muxing overhead is unknown [16:00] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:8a9878b4ff23: Merge commit 'f759f66e367bc62ed7738a51040f775b656a60d8' [16:22] <cone-898> ffmpeg.git 03Anton Khirnov 07master:baeb59d89bed: avconv: rewrite output data size tracking [16:22] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:13278b97aeb2: Merge commit 'baeb59d89beda499688e528c6a269d2b255b66f2' [16:28] <cone-898> ffmpeg.git 03Derek Buitenhuis 07master:206d5330f648: libx265: Simple cosmetic fix [16:28] <cone-898> ffmpeg.git 03Derek Buitenhuis 07master:ecb5b9c09800: libx265: Only use one memcpy for headers [16:36] <cone-898> ffmpeg.git 03Anton Khirnov 07master:a220b07b0f97: avconv: split printing the final statistics into a separate function [16:36] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:afa442c198fc: Merge commit 'a220b07b0f971f88213aef012aea51af2800be62' [16:45] <cone-898> ffmpeg.git 03Anton Khirnov 07master:4754345027eb: avconv: print verbose per-stream transcoding statistics [16:45] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:1046cd60d4b1: Merge commit '4754345027eb85cfa51aeb88beec68d7b036c11e' [16:51] <cone-898> ffmpeg.git 03Anton Khirnov 07master:d1780e01b8b7: avconv: don't warn on multiple frames per packet for codecs that expect it [16:51] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:33a7e288ab2e: Merge remote-tracking branch 'qatar/master' [19:55] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:cb640eb33542: ffmpeg: move packet deallocation into write_frame [20:17] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:bce2abc9362d: avcodec/h264: use ff_h264_set_erpic() instead of memset [21:07] <cone-898> ffmpeg.git 03Diego Biurrun 07master:8ef33497a6a3: build: Add config option for CABAC code [21:07] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:ab8d9fe9b84c: Merge commit '8ef33497a6a3ed05d2f931809c074ef3d1878834' [21:21] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:60728e8bab8d: libx265: Write global extradata when requested [21:21] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:fc2de9b7cf59: Merge commit '60728e8bab8d2a5f6bbb4baa7d53142dbc6047ed' [21:33] <cone-898> ffmpeg.git 03Derek Buitenhuis 07master:21a561176e5e: libx265: Only use one memcpy for headers [21:33] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:5d39ec2188e8: Merge remote-tracking branch 'qatar/master' [21:50] <michaelni> ubitux, "LTO (+ no inline asm) " says "lto1: fatal error: error writing to /tmp/ccpfPtEM.s: No space left on device" [21:50] <ubitux> heee [21:51] <ubitux> tmpfs 7.8G 2.0M 7.8G 1% /tmp [21:51] <ubitux> :/ [21:53] <michaelni> wild guess/suggestion "--disable-debug " might result in less space needs [21:54] <michaelni> but probably wont make a big diference [22:09] <cone-898> ffmpeg.git 03Michael Niedermayer 07master:575b95775867: dox/scaler:fix bicubiclin typo [22:29] Action: pippin tried bisecting the GIF anim regression; but it seems to hickup with make: *** No rule to make target `libavcodec/dct32_template.c', needed by `libavcodec/dct32_fixed.o'. Stop. [22:30] <pippin> time to stop wasting gasoline on FATE :] [22:31] <nevcairiel> sometimes it needs a full clean to fix those [22:33] <pippin> it isn't my regression; and since I currently use a gasoline generator as energy source ... maybe revisit it next time I'm charging but not on batter power :] [22:35] <ubitux> michaelni: i was using /tmp today, that might have affected the test, should be ok at the next run [23:47] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/0.11:12bf790b850b: avformat/mov: only force parsing for video tracks if stss is empty [23:47] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/0.11:7a56695ced39: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [23:48] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/1.0:519675c49c28: avformat/mov: only force parsing for video tracks if stss is empty [23:48] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/1.0:be66ae3b0151: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [23:48] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/1.1:cf4099161a33: avformat/mov: only force parsing for video tracks if stss is empty [23:48] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/1.1:c6f2ad9f944d: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [23:48] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/1.2:da6a8c9cfda5: avformat/mov: only force parsing for video tracks if stss is empty [23:48] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/1.2:5202621ac413: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [23:48] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/2.0:9ecd0f34adcc: avformat/mov: only force parsing for video tracks if stss is empty [23:48] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/2.0:5ee384c4eb6b: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [23:48] <cone-898> ffmpeg.git 03Hendrik Leppkes 07release/2.1:63dcf7c77c4a: avformat/mov: only force parsing for video tracks if stss is empty [23:48] <cone-898> ffmpeg.git 03Michael Niedermayer 07release/2.1:3c880d276513: avformat/mov: fix keyframe flags for sample from chromium Issue 340865 [00:00] --- Tue Mar 25 2014

1 0

ffmpeg.log.20140324
by burek 25 Mar '14

25 Mar '14

[00:04] <zumba_ad_> sacarasc: I want to stream it. How do I encode the mp4 file? [00:05] <sacarasc> zumba_ad_: Live stream? Or just able to be played without loading it all? [00:08] <zumba_ad_> it's a local file [00:08] <zumba_ad_> my aunt wanted to watch a video [00:09] <sacarasc> ffmpeg -i blah.mp4 -movflags faststart -c copy output.mp4 [00:09] <sacarasc> Then stream it however you want to. [00:13] <zumba_ad_> Thanks a lot [01:35] <kutemo> i convert an audio stream to wave file and then use another app to convert it to aac, is there a way to skip a two step and a one step command? if i try to pipe it, i get "[NULL @ 053f7720] Unable to find a suitable output format for 'pipe:' pipe:: Invalid argument" [01:37] <kutemo> this is my command: ffmpeg -i file.m2ts -c:a pcm_s16le -ac 2 -af volume=2:fixed -map 0:1 out.wav [01:38] <kutemo> actually this is my command: ffmpeg -i file.m2ts -c:a pcm_s16le -ac 2 -af volume=2:fixed -map 0:1 - | "my other app" [01:38] <klaxa> you have to specify a format if you are using a pipe output [01:38] <klaxa> i.e. -f wav [01:39] <klaxa> resulting in: ffmpeg -i file.m2ts -c:a pcm_s16le -ac 2 -af volume=2:fixed -map 0:1 -f wav - | "my other app" [01:40] <kutemo> ok, thanks! i swear, i could not found that format when i was looking. i'm glad you got it right away. [01:41] <justinfront> Hi facebook seems unhappy with this format ffmpeg -i 663_0174_01.MP4 -c:v libx264 -preset veryfast -vf scale=1280:-1 -crf 23 -c:a libmp3lame -b:a 192k -ar 44100 output.mp4 I chopped my video ffmpeg -i output.mp4 -ss 00:00:40 -to 00:03:22 -c copy outputCut.mp4 wondering what setting I should change, also VLC could play it but not quicktime [01:42] <justinfront> trying to put a salsa practice taken with an ex1 sony on facebook private group so we can discuss where we are going wrong. [01:45] <justinfront> maybe I can try vimeo or somewhere else but I can only upload it private as we have not done the student performance yet [01:45] <klaxa> have you tried youtube? [01:45] <klaxa> you can set it unlisted/private there too [01:47] <justinfront> is vimeo better or worse than youtube [01:47] <klaxa> dunno [01:48] <justinfront> any idea why fb and quick time don't like the video [01:48] <klaxa> not really, i don't really use either [01:50] <pippin> michaelni: seen my email about gif anim encoding regression? [02:03] <michaelni> pippin, yes, maybe ubitux has an idea, if not bisect might be a easy and quic way to find the lcluprint [02:04] <pippin> git revert didn't work, but git blame on the file made me think that 219b35f5d169d9ece408262ba2337dab9d19aafd is at fault [02:06] <pippin> though; that'd be only a hunch; it is one of few changes since then, and among them is mentioning of I-frame / p->key_frame = 1 [02:20] <vl4kn0> Hi, I've got AV_PIX_FMT_GRAY8 AVFrame and want to save it as an image, format doesn't matter. How do I proceed? [07:46] <ubitux> pippin: can you try to checkout 219b35f5d169d9ece408262ba2337dab9d19aafd test, and then reset hard to the previous commit, to confirm that's indeed where there is the problem? [07:47] <ubitux> it doesn't look like it's the cause to me [07:47] <ubitux> pippin: also, better join the devel channel to discuss this [09:27] <nyc3j> not compiling: [09:27] <nyc3j> In file included from libavformat/avio.c:29: In file included from libavformat/network.h:34: /usr/include/unistd.h:274:21: error: typedef redefinition with different types ('__socklen_t' (aka 'unsigned int') vs 'int') [10:11] <CalimeroTeknik> hello, I'm getting "library configuration mismatch" and ffmpeg crashes on "relocation error" http://sprunge.us/NYGS [10:11] <CalimeroTeknik> where is the configuration? [10:15] <spaam> CalimeroTeknik: read the message. you can find it there. [10:16] <spaam> CalimeroTeknik: read the first line with configuration and the second one . [10:26] <CalimeroTeknik> oh, the prefix is different [10:26] <CalimeroTeknik> I recompiled ffmpeg-compat but it doesn't suffice [10:27] <CalimeroTeknik> uninstalling it solves the problem though [11:10] <CalimeroTeknik> how do I identify whether a file is 10bit or 8bit h264? [11:10] <JEEB> the pix_fmt given by the decoder [11:11] <JEEB> 8bit will be yuv420p, and 10bit will be yuv420p10 [11:11] <CalimeroTeknik> oh! thanks [11:12] <JEEB> and the same for 422, 444 and RGB [12:11] <CalimeroTeknik> where is the list of options for <Stream x> Format for ffserver? [12:11] <CalimeroTeknik> I found one in an example file, but flv works for instance, and is not included [12:42] <sabgenton> is it possible to concatenate ffmpeg -i mount_point/video_ts/vts_01_3.vob -f mp3 file.mp3 with multiple vobs [12:43] <sabgenton> rather than doing each vob sepeartly and ending up with many mp3s [12:47] <CalimeroTeknik> do you only want the vob's audio? [13:33] <termos> are there any good examples of how to use tee muxing in the C interface? [13:53] <sabgenton> Ok I'm trying ffmpeg -f concat -i file.txt -map 0:2 -f mp3 file.mp3 . but it doesn't work [13:53] <sabgenton> I get [13:53] <sabgenton> [ac3 @ 027e6ca0] frame sync error [13:53] <sabgenton> Error while decoding stream #0:2: Invalid data found when processing input [13:54] <sabgenton> file.txt has file bla.vob on each line [13:54] <sabgenton> it does the first vob then fails [14:12] <justinfront> If I want to speed up some dance music from a video is that easy? [14:12] <justinfront> I say dance I am still on about the salsa [14:12] <justinfront> I want just to extract the music at something like 10% faster. [14:33] <xlinkz0> i'm saving one frame from a rtsp stream like so : ffmpeg -y -i $link -frames:v 1 -s 580x326 $path [14:33] <xlinkz0> the problem is sometimes it saves an incomplete frame, is it possible to only save a complete frame? [14:34] <xlinkz0> like the first i-frame that is fully decodable [16:38] <Euphoricz> Hi, I am having troubles with my live stream startup latency. I have read over the wiki documentation regarding this however it does not seem to apply due to the command line switches I'm using. Here is my command, do you have any suggestions for how I can decrease the startup latency? [16:38] <Euphoricz> ffmpeg.exe -rtsp_transport tcp -i "rtsp://admin:avgtest@Camera-test4.wxmet.com/defaultPrimary" -vcodec copy -f matroska -an -r 30 - [17:05] <TekniQue> does ffmpeg have a filter to pick a certain program from a multi program transport stream? [17:07] <hellhound> hey, anyone using ffplay on mac? [17:23] <VictorS> Hi! Can someone provide some help regarding an failed encoding from ASF(G2M4) to MP4? I'm getting an error like "Error setting option pix_fmt to value -1". Thanks! [17:35] <xlinkz0> is it possible to get the first i-frame that is fully decodable? [18:05] <share> hi [18:05] <share> this page https://trac.ffmpeg.org/wiki/x264EncodingGuide contains oudated comments (?) [18:06] <share> -preset is now -pre [18:06] <share> '-pre help' doesnt show anything [18:09] <klaxa> no -pre is deprecated, use -preset [18:12] <share> klaxa: no [18:13] <share> klaxa: check the manual -preset doesnt exist [18:14] <Euphoricz> Why are presets not available when using the -rtsp_transport command line switch? [18:16] <klaxa> maybe the manpage is outdated [18:17] <share> klaxa: do i need to specity the codec to list the presets? [18:17] <share> i tried ffmpeg -preset help [18:18] <share> doesnt show the presets [18:18] <klaxa> -preset is to be used with libx264 [18:18] <share> ok [18:18] <klaxa> same goes for -profile and -tune [18:18] <share> like x264 -preset [18:19] <klaxa> see ffmpeg -h encoder=libx264 [18:19] <share> doesnt show the presets [18:19] <share> oops [18:19] <share> ok thanks [18:19] <klaxa> presets are placebo, veryslow, slower, slow, medium, fast, faster, veryfast, ultrafast [18:20] <share> I know [18:20] <share> ill try medium [18:26] <llogan> hellhound: it is possible. http://www.evermeet.cx/ffplay/snapshots/ [18:29] <hellhound> llogan: hmm, I'm using the ffmpeg provided by macports. The funny thing is that I've installed ffmpeg latest svn revision. I'm assuming that macports patches it somehow before installing it [18:33] <klaxa> svn? wasn't that abandoned in 2008? [18:33] <klaxa> for ffmpeg i mean [18:36] <hellhound> klaxa: ok, I was assuming it was a svn repo, doesn't matter, I know for sure it's the current HEAD [18:45] <share> how to disable audio [18:46] <share> the tv is complaining about not supported audio [18:46] <share> -c:a null? [18:51] <share> -an [18:57] <share> OpenCL: Unable to query installed platforms [19:12] <VictorS> I'll come back tomorrow with pastebin commands. Thank you [19:13] <metaphysician> 32 bit static Linux builds have been discontinued? [19:30] <llogan> metaphysician: you mean from burek? [19:31] <llogan> the "gusari" builds [19:31] <llogan> probably just have to tell burek to poke the machine [19:41] <llogan> i let him know [19:41] <metaphysician> llogan: yup, http://ffmpeg.gusari.org/static/. Thanks. Looking forward to trying the new x265 encoder... [19:42] <JEEB> you're almost better off just compiling x265 yourself and using the command line encoder for it [19:42] <JEEB> the API gets broken all the time [19:42] <JEEB> and piping from ffmpeg for example [19:45] <metaphysician> JEEB: point taken [19:46] <JEEB> also let's just say that if you actually want to see better results under some scenarios compared to x264, you pretty much have to use presets like placebo with x265 :P [19:47] <JEEB> because that's pretty much the only preset that is relatively sure to give you results that you like [19:48] <JEEB> also last time I tested x265 doesn't max out bframes or refs with preset placebo, so you might want to do that, too [21:08] <ValdikSS> Please confirm the issue https://trac.ffmpeg.org/ticket/2975#comment:5 [21:16] <thedarkjedi> Hi Guyz [21:17] <thedarkjedi> One question [21:17] <thedarkjedi> I am running ffmpeg on my windows 64 bit and got the latest build from http://ffmpeg.zeranoe.com/ but when I use ffmpeg -filters I notice that a lot of filters are missing [21:18] <thedarkjedi> specially equalizer and bandreject [21:18] <thedarkjedi> Can anybody tell me how should I get missing filters? [21:39] <ValdikSS> thedarkjedi: you should probably recompile ffmpeg with needed filters enabled or find another build with enabled filters [21:41] <thedarkjedi> Thanks... it seems my problem is different though... My windows seem to be somehow dynamically linking to older libraries... [21:42] <thedarkjedi> I even deleted older files and replaced them with the new ones... [22:30] <thedarkjedi> Hi Guyz [22:30] <thedarkjedi> One more question.. I am using ffmpeg to extract audio for speech recognition [22:30] <thedarkjedi> the audio could be noisy...how do I reduce background noise to optimize speech [22:31] <thedarkjedi> I tried C:\Users\Sudh\Downloads\ffmpeg-20140324-git-63dbba6-win64-static\bin>ffmpeg -i i nput.wav -filter_complex "highpass=f=400,lowpass=f=1800" out2.wav [22:32] <thedarkjedi> and also C:\Users\Sudh\Downloads\ffmpeg-20140324-git-63dbba6-win64-static\bin>ffmpeg -i i nput.wav -af "equalizer=f=1000:width_type=h:width=900:g=-10" output.wav [22:32] <thedarkjedi> but couldnt filter out background noise properly... results were very disappointing [22:32] <thedarkjedi> :( [22:57] <TheCloudstrike> hey i have a question, how can i burn a mkv file with subtitle in it into a mp4 file. i dont want to include a external ass file, i want to have the included subtitle from the video [22:57] <TheCloudstrike> to be burned into mp4 [22:58] <TheCloudstrike> the thing is that some files have ass and some files not its a bash script what is used for many files [22:59] <TheCloudstrike> any idea? [23:46] <TheCloudstrike> hello? xD [23:48] <llogan> TheCloudstrike: i think subtitles and ass filters require a file name, so link lables and stream specifiers won't work, AFAIK [23:52] <llogan> you may have to output the subtitle from the input and use that [23:59] <RenatoCRON> hello guys. I have 20secs x264 videos. If I want know many changes was in the images each second (using gt(scene\,0.5) or something like that) do I need to first generate another video with (20 frames, time don't matter) to pipe to another ffmpeg with filter scene?? [00:00] --- Tue Mar 25 2014

1 0

ffmpeg-devel.log.20140323
by burek 24 Mar '14

24 Mar '14

[00:10] <cone-46> ffmpeg.git 03Marton Balint 07master:4acc86bcfbfc: mpegts: move last_pcr to MpegTSFilter level [00:10] <cone-46> ffmpeg.git 03Marton Balint 07master:6bab55b84c8b: mpegts: factorize mpegts filter creator code [00:10] <cone-46> ffmpeg.git 03Marton Balint 07master:d83a5b52184f: mpegts: add pcr filter for tracking standalone pcr pids [00:10] <cone-46> ffmpeg.git 03Marton Balint 07master:d2f60547c230: mpegts: fix teletext pts of streams with standalone pcr pid [00:10] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:2765c2ec7d5e: Merge remote-tracking branch 'cus/stable' [01:46] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:fc567ac49e17: avcodec: Add padding after the remaining AVFrames [03:20] <cone-46> ffmpeg.git 03Janne Grunau 07release/2.2:230c4c6ad9f7: armv6: vp8: use explicit labels in motion compensation asm [03:20] <cone-46> ffmpeg.git 03Martin Storsjö 07release/2.2:0f42e06651c9: doc: Point to the correct, actually maintained gas-preprocessor repo [03:20] <cone-46> ffmpeg.git 03Tim Walker 07release/2.2:daaef403d122: matroskaenc: write private data in hvcC format for HEVC. [03:20] <cone-46> ffmpeg.git 03Michael Niedermayer 07release/2.2:cad2958fd7ca: Merge commit '230c4c6ad9f74697e77351d32e679bf0c24478ea' into release/2.2 [03:20] <cone-46> ffmpeg.git 03Michael Niedermayer 07release/2.2:dde996bf999a: Merge commit '0f42e06651c9c4d7581b70b9ad2b160c2525094f' into release/2.2 [03:20] <cone-46> ffmpeg.git 03Michael Niedermayer 07release/2.2:d130fae51948: Merge commit 'daaef403d122b35d87b27d6b369d287b1ed06973' into release/2.2 [03:28] <cone-46> ffmpeg.git 03John Stebbins 07master:1e9db41e2a21: movenc: Allow override of major brand in ftyp atom [03:28] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:802e5fc8f811: Merge commit '1e9db41e2a2166be5671b088ef4ad06a40af459f' [03:43] <cone-46> ffmpeg.git 03Luca Barbato 07master:0f0d1d056e1a: configure: Always use -DOPJ_STATIC when checking for openjpeg [03:43] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:1abb03195930: Merge commit '0f0d1d056e1a3ad3db32b7d1a06286b2841e91b1' [03:54] <cone-46> ffmpeg.git 03Vittorio Giovara 07master:d5717bc4e9cc: changelog: switch libx265 and compand entries [03:54] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:acb8d655d750: Merge commit 'd5717bc4e9cc253a933ae8130eea3a9c8bc33c0b' [04:05] <cone-46> ffmpeg.git 03Vittorio Giovara 07master:6f6c0294646b: lavc: drop ff_pixfmt_list_420 [04:05] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:6ee2911996ed: Merge remote-tracking branch 'qatar/master' [04:28] <cone-46> ffmpeg.git 03Tim Walker 07release/2.2:9a6a7109985c: matroskaenc: enable Annex B to MP4 conversion for HEVC tracks. [04:28] <cone-46> ffmpeg.git 03John Stebbins 07release/2.2:29d61d73b1cb: movenc: Allow override of major brand in ftyp atom [04:28] <cone-46> ffmpeg.git 03Michael Niedermayer 07release/2.2:49f11e12d5af: Merge commit '9a6a7109985cbd3a20707b40cc31ed53a174195c' into release/2.2 [04:28] <cone-46> ffmpeg.git 03Michael Niedermayer 07release/2.2:efe259a27e56: Merge commit '29d61d73b1cbf0dcd1f79bc49c7df01b0a377c85' into release/2.2 [06:19] <cone-46> ffmpeg.git 03wm4 07master:686e66267628: http: restructure http_connect error handling path [13:37] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:82a90e7764ad: Move avpriv_find_pix_fmt() to utils.c [13:38] <michaelni> Case, build problem maybe fixed [13:46] <Case> michaelni: indeed it is. Thanks [15:41] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:c8246d37664a: avcodec/x86/h264_qpel: Fix typo introduced by 322a1dda973e802db7b57f2007fad3efcd5bab81 [17:52] <cone-365> ffmpeg.git 03Carl Eugen Hoyos 07master:942cded690bc: Autodetect VDA like all other hardware acccelerations. [19:01] <cone-365> ffmpeg.git 03Carl Eugen Hoyos 07master:cc6d549adbb8: Always pass the configure option sysroot to --sysroot and -isysroot. [19:06] <cone-365> ffmpeg.git 03Carl Eugen Hoyos 07release/2.2:13682b48e95c: Autodetect VDA like all other hardware acccelerations. [19:54] <cone-365> ffmpeg.git 03Vittorio Giovara 07master:2a38651033a4: RELEASE_NOTES: mention new filters [19:54] <cone-365> ffmpeg.git 03Michael Niedermayer 07master:5dfbecb3a374: Merge remote-tracking branch 'qatar/master' [19:56] <cone-365> ffmpeg.git 03Reinhard Tartler 07release/2.2:26bbc1c242eb: Prepare for 10 Release [19:57] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:9abe0bfb7f52: Merge commit '26bbc1c242ebfeec3c49585207efac7293632433' into release/2.2 [20:12] <cone-365> ffmpeg.git 03Anton Khirnov 07release/2.2:80239a8bb11f: af_channelmap: fix ONE_STR mapping mode [20:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:9a884b7b97c9: Merge commit '80239a8bb11f730d95f03dfbd0deab258b413b0e' into release/2.2 [20:13] <cone-365> ffmpeg.git 03Carl Eugen Hoyos 07release/2.2:d461e077a51c: Always pass the configure option sysroot to --sysroot and -isysroot. [20:31] <cone-365> ffmpeg.git 03Vittorio Giovara 07release/2.2:f25e6e0c255a: RELEASE_NOTES: mention new filters [20:31] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:5b0e2eb041db: Merge commit 'f25e6e0c255a6d81971fe66f6c32d29be0706f3d' into release/2.2 [21:12] <cone-365> ffmpeg.git 03Nicolas George 07release/2.2:dcf560204c5a: lavc/pthread: copy packet side data. (cherry picked from commit ea6825fd090ef3f8edbce506d2fbf6e187f585fa) [21:12] <cone-365> ffmpeg.git 03Nicolas George 07release/2.2:99905118a851: lavd/xv: take aspect ratio into account. (cherry picked from commit c37bbe54f400edc8cb0070138328048943e90b1a) [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:57a43142baed: avcodec/utils: fix sizeof(AVFrame) dependence in avcodec_encode_audio2() [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:c00beff5e075: avcodec/mjpegdec: factorize parse_avid out [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:af3d003658e5: avcodec/mjpegdec: switch interlaced_polarity to 0 for PAL AVID [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:7bdd348e5832: avcodec/mjpeg: print info from parse_avid() if requested [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:90c7bfb9be43: avcodec/mjpegdec: parse avid data from extradata [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:b052525f9b34: avcodec/mjpegdec: Switch default interlaced polarity for MJPG to 1 [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:7c6a8afa7ea7: ffmpeg: dont call exit_program() from a signal hander [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:25b462cab949: avcodec/mpegvideo: make mc_mb_var_sum_temp / mb_var_sum_temp 64bit [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:dca463b728c1: avcodec/mpegvideo: make mc_mb_var_sum / mb_var_sum 64bit [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:30ae080e9dec: avcodec/ratecontrol: make (mc_)mb_var_sum(_last) 64bit [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:72a12f61ef66: avcodec/snowenc: allow larger mb variances [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:e39a992bd155: avformat/mp3enc: use av_copy_packet() [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:0c88d539f828: avcodec: Add padding after the remaining AVFrames [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:938ff9371005: avformat/aviobuf: factorize buffer_size out [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:01507eb1f87d: avformat/aviobuf: optimize sequential backward seeking [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:b017785fa55c: avformat/mp3dec: optimize mp3_seek() for dir < 0 [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:e8919d6522c8: Revert "mpegts: do not set pts for missing dts in video streams" [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:4b7c149306cf: avcodec/g2meet: fix error returns [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:e1f51bbd1fbe: avcodec/g2meet: also reset local got_header when reseting the context got_header [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:8caaf260a66c: avformat/utils: detect MPEG streams with faulty DTS and discard affected DTS [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:4407b38b28d1: swscale/x86/swscale: fix missing xmm clobbers in yuv2yuvX_sse3() [21:12] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:9ff0467566e9: Move avpriv_find_pix_fmt() to utils.c [21:12] <cone-365> ffmpeg.git 03Andrey Myznikov 07release/2.2:4dc8b4d7d081: Fix pthread-related compile errors in iec61883.c [22:16] <cone-365> ffmpeg.git 03Michael Niedermayer 07release/2.2:d2c76782e0e8: update for 2.2 [23:28] <cone-365> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n2.2': unknown revision or path not in the working tree. [23:28] <cone-365> Use '--' to separate paths from revisions [23:28] <cone-365> refs/tags/n2.2:HEAD: Merge remote-tracking branch 'qatar/master' [00:00] --- Mon Mar 24 2014

1 0

ffmpeg.log.20140323
by burek 24 Mar '14

24 Mar '14

[00:23] <slorbast> Hey guys, is there any way to convert a .mp3 or any other typical audio or video file to .mid? I imagine there isn't a great way to do it systematically other than recomposing a midi myself but wanted to make sure [00:24] <klaxa> there are some programs that try to do that [00:24] <klaxa> there isn't anyone that does it reliably well though [00:25] <klaxa> *any one [00:26] <slorbast> Thanks, I'll poke around and try them out and see if they work "well enough" lol [01:06] <tabakd> I'm trying to record a stream of a x11 session seperate from the one I am in using ffmpeg. The second I leave my x11 sesion (:1) being recorded the stream freezes. [01:42] <BtbN> tabakd, that's most likely because it stops beeing rendered at all [01:46] <BlackFox> Greetings, quick question: I want to stream an existing video file, and I want to be able to play/pause/rewind it in realtime. Taking ffmpeg and giving it the file as input works, can I control it somehow though? [01:50] <sacarasc> I think you'd be better off with something like VLC for that. [01:50] <BlackFox> Hmm [01:51] <BlackFox> I suppose I can just have ffmpeg re-route the stream from VLC [01:52] <tabakd> Can ffmpeg record an x11 session I'm not actually connected to? [01:57] <clever> tabakd: if you can set DISPLAY right to connect to it, yes [01:58] <BtbN> The problem is not that ffmpeg can't connect to it, the problem is that it stops rendering when it's not active [01:58] <clever> yeah, thats a different problem [01:58] <clever> i would start an Xvnc4 server instead [01:58] <clever> then i can minimize the viewer or even DC without any impact [01:59] <BtbN> but also loose all performance [01:59] <clever> yeah, it depends on what he is trying to do as well [02:03] <tabakd> yea [02:04] <tabakd> the stream goes inactive when I leave the session [02:05] <BtbN> yes, that's not a bug and not a problem with ffmpeg, and most likely also nothing you can fix [02:08] <tabakd> does any other software do what I want? [02:46] <pippin> there has been a regression in the GIF animation encoder; it no longer makes pixels that are the same from the previous frame transparent (frame mode is still combine though) [02:49] <pippin> discovered when updating the ffmpeg patch at http://pippin.gimp.org/a_dither/ - now all of the methods; including bayer yields ~2.9M files [07:41] <pross-au> noob question: how do i play back a 352 kHz audio file using ffplay (sdl fails to open the audio device; e.g. how do i make ffplay resample it.) [10:26] <xhochy_> hey, I have a problem with rtmp streams where ffmpeg does not stop playing: in vlc the progressbar always stops 1/2s before the end and ffplay just outputs silence for an infinite time but the timecounter increases, the ffplay -loglevel debug log http://pastebin.com/WAANH0AD shows that a NetStream.Play.Stop is received so I'm wondering a bit why it does not [10:26] <xhochy_> stop [19:30] <ValdikSS> klaxa: https://trac.ffmpeg.org/ticket/3491 [20:08] <ValdikSS> klaxa: and it's solved! [22:24] <justinfront> Hi I would like to convert EX1 sony camera footage to something I can upload to the internet and something I can use with flash player and with quick time. I installed ffmpeg via macports and remember it was doing something but some of the links related seem to be broken, can anyone suggest how I check the current format and the best or simplest way to get it in a mac usable form using ffmpeg. [22:25] <klaxa> ValdikSS: oh nice [22:26] <zumba_addict> is h.265 in public? [22:29] <jbermudes> justinfront: if you just give ffmpeg an input without instructions on what to output, it'll tell you info about the file. ffmpeg -i filename.mov etc. [22:35] <klaxa> justinfront: if you want it to be compatible as possible, webm is probably what you want [22:36] <klaxa> to see if it satisfies your quality needs, just encode a test segment like so: ffmpeg -i somefile.mov -t 10 somewebm.webm [22:36] <klaxa> that will encode the first 10 seconds [22:36] <justinfront> jbermudes thanks I have pastebinned the output maybe someone can advise the best approach, been practicing a student dance show and really want to upload something to crappy facebook private group so quality can probably go out the window. [22:36] <justinfront> http://pastebin.com/dgavS9Qt [22:37] <justinfront> what does -t 10 do? [22:37] <justinfront> oh sorry did not read [22:37] <justinfront> 10 sec [22:37] <justinfront> Klaxa will try that [22:38] <justinfront> I think fb probably like h264 format ie flash, webm is likely heavier and not ideal for flash which is probably what the use best if they don't have to try "fixing" my quality further. [22:39] <klaxa> ah with flash player [22:39] <klaxa> i read without flash player [22:40] <klaxa> (nobody wants flash really...) [22:41] <klaxa> justinfront: to encode to h264 read: https://trac.ffmpeg.org/wiki/x264EncodingGuide [22:43] <justinfront> fb suggest [22:43] <justinfront> We recommend H.264 video with AAC audio in MOV or MP4 format [22:43] <justinfront> Make sure the video isn't more than 1280px wide [22:43] <justinfront> Try to keep your aspect ratio, or the length and width of the video, to multiples of 16px [22:43] <justinfront> Keep the frame rate of your video at, or below, 30fps [22:43] <justinfront> Use stereo audio with a sample rate of 44,100hz [22:44] <justinfront> my video is currently 1920x1080 [22:45] <sacarasc> -vf scale 1280:-3 [22:45] <justinfront> so would I change the video in stages or try to get all the settings in one go [22:45] <justinfront> what does the -3 mean? [22:46] <justinfront> thanks by the way guys this is really helpful. [22:50] <justinfront> that h264 link is not easy reading was hoping someone may have done something similar and I would not need to learn it all right now [22:51] <sacarasc> -3 means, IIRC, scale it properly, keeping aspect ratio and divisible by 16... [22:51] <sacarasc> I think. [22:52] <justinfront> cool maybe I need to read the wiki some [22:52] <klaxa> justinfront, for facebook you can probably just use -crf 21 and use whatever preset you find to have an acceptable speed [23:04] <justinfront> Unable to find a suitable output format for 'scale' scale: Invalid argument [23:05] <justinfront> = missing ? lets try [23:07] <justinfront> -1 ? [23:09] <sacarasc> -vf scale=1280:-3 [23:09] <sacarasc> I missed the =. [23:09] <justinfront> Currently I have [23:10] <justinfront> ffmpeg -i 663_0174_01.MP4 -c:v -vf scale=1280:-3 libx264 -crf 23 output.mp4 [23:10] <justinfront> is that order ok ? [23:10] <ValdikSS> yes [23:11] <ValdikSS> do you have interlaced video? [23:11] <justinfront> Stream #0:0(eng): Video: mpeg2video (Main) (mp4v / 0x7634706D), yuv420p(tv, bt709), 1920x1080 [SAR 1:1 DAR 16:9], 34959 kb/s, 25 fps, 25 tbr, 25k tbn, 50 tbc (default) [23:12] <justinfront> I don't know [23:12] <justinfront> does that tell me? [23:12] <Jack64> justinfront: libx264 should be after -c:v , no? [23:12] <klaxa> yeah make that: ffmpeg -i 663_0174_01.MP4 -c:v libx264 -vf scale=1280:-3 -crf 23 output.mp4 [23:13] <justinfront> [Parsed_scale_0 @ 0x7f8a10418500] Size values less than -1 are not acceptable. [23:13] <justinfront> [Parsed_scale_0 @ 0x7f8a10418500] Failed to configure output pad on Parsed_scale_0 [23:13] <justinfront> Error opening filters! [23:13] <Jack64> scale=1280:-1 ? [23:13] <ValdikSS> use -1 [23:14] <ValdikSS> or keep_aspect=1 [23:14] <klaxa> that *should* come out at 1280x720 anyway [23:14] <ValdikSS> so guys, is his video interlaced? [23:14] <justinfront> Is this ok to add The encoder 'aac' is experimental but experimental codecs are not enabled, add '-strict -2' if you want to use it. [23:14] <Jack64> use libfdk_aac [23:15] <klaxa> ValdikSS: probably not, otherwise it wouldn't be 25 fps [23:16] <justinfront> well I put the strict and it's doing something so guess wait and see [23:16] <justinfront> what it looks like [23:17] <justinfront> Do you guys often write wrappers to make it easier to select settings or use tools or is it best to just use a terminal? [23:19] <justinfront> Also wondering if you would shoot the video in high rez and then convert like I am or would you try to set the camera up to film at the output resolution? [23:20] <justinfront> by the way very grateful for the help thank you everyone. [23:21] <jbermudes> justinfront: It depends on the tools. On Linux, there is a control panel for adjusting some webcam settings but ffmpeg doesn't seem to honor some of the settings [23:21] <klaxa> if you have the storage, always record in the best quality possible [23:24] <justinfront> looks like it is using 48 not 44 for the audio. [23:26] <guest9184> Hello all, I am having trouble compiling ffmpeg for the Android NDK. Using these instructions http://www.roman10.net/how-to-build-ffmpeg-with-ndk-r9/comment-page-3/#comm… I am able to get right up to the point where I run the .sh build script. Unfortunatly, when I run it, I get the following error:"...arm-linux-androideabi-gcc is unable to create an executable file. C compiler test failed.". And it gives many "no such file or [23:28] <justinfront> do you need to do something like sudo and give the script admin permission, beyond that suggestion can't help [23:29] <justinfront> Duration: 00:04:40.00 er seems like it may take awhile its still on 00:00:31.14 :( [23:30] <justinfront> should I kill the shell or just go off and let it run and come back to see the output [23:33] <ValdikSS> justinfront: ffmpeg -i 663_0174_01.MP4 -c:v libx264 -preset veryfast -vf scale=1280:-1 -crf 23 output.mp4 [23:35] <justinfront> I need to add -strict -2 or something like -libfdk_aac [23:41] <Jack64> justinfront: replace -c:a aac with -c:a libfdk_aac [23:43] <justinfront> and I can set that to 44 instead of 48? [23:45] <Jack64> I assume you mean kHz [23:46] <Jack64> I'm not sure, I've never tried to change the audio frequency, I just change the bitrate with -c:b 320k for example [23:46] <justinfront> ok I can try that bitrate or is that a guess? [23:47] <justinfront> yes kHz [23:48] <justinfront> https://trac.macports.org/ticket/42190 [23:48] <zumba_addict> in an mp4 file, is there a atom data that I need to reposition? [23:50] <sacarasc> Only if you want to stream it. [23:50] <ValdikSS> justinfront: ffmpeg -i 663_0174_01.MP4 -c:v libx264 -preset veryfast -vf scale=1280:-1 -crf 23 -c:a libmp3lame -b:a 192k -ar 44100 output.mp4 [23:52] <justinfront> thanks will give that a go [00:00] --- Mon Mar 24 2014

1 0

ffmpeg-devel.log.20140322
by burek 23 Mar '14

23 Mar '14

[00:00] <J_Darnley> llogan: Sorry, I misread [00:01] <J_Darnley> I assume qscale will be ignored and qp will be used [00:05] <J_Darnley> Ah then that's a bug I jusst made [00:05] <J_Darnley> It does use qp but tell the use user its using crf [00:05] Action: J_Darnley grumbles [00:18] <J_Darnley> llogan: And what about if the user uses both -qscale and -x264opts? Do I need to warn in that case? [00:31] <llogan> J_Darnley: i forgot about that, but if they're using x264opts then i'd consider them to be on their own (or hopefully they know what they're doing). [00:32] <llogan> or x264-params [00:34] <J_Darnley> Is that a 3rd way to set options? [00:35] <J_Darnley> :( [00:37] <michaelni> maybe just say that qscale is ignored with libx264 [00:37] <michaelni> without any other checks or details [00:41] <J_Darnley> I can move my patch further down the init function to catch all options. [00:41] <J_Darnley> Just a simple statement that qscale is ignored may be better. [00:42] <J_Darnley> For that case, where should I print the message? From the top of the init function like my first patch? [00:43] <michaelni> iam happy with any location that works [00:44] <J_Darnley> llogan: Any more comments? [00:46] <J_Darnley> Should I ask the original poster for comments on trac? [00:48] <J_Darnley> How about "-qscale is ignored with libx264, -crf is preferred"? [00:55] <J_Darnley> Do I need to mention libx264 in the message because ffmpeg will print "[libx264 @ address] in the log? [01:05] <michaelni> i guess you can skip the libx264 in the message [01:08] <llogan> J_Darnley: up to you how complex you want it to be. im also fine with a simple "-qscale is ignored with this encoder" or whatever [01:13] <J_Darnley> In that case I will say that a simple message is least likely to become incorrect, wrong, misleading, or broken in the future so I will submit a patch to do just that. [01:19] <J_Darnley> Oh dammit [01:20] <J_Darnley> look at what you're actually commiting James! [01:21] <J_Darnley> I think it is time for bed when I send yet another email [01:39] <cone-406> ffmpeg.git 03Michael Niedermayer 07master:d003a0cd2e58: avformat/mp3enc: use av_copy_packet() [03:30] <cone-406> ffmpeg.git 03Michael Niedermayer 07master:87ecefdab009: elbg: change utility_inc to 64bit [03:54] <cone-406> ffmpeg.git 03rogerdpack 07master:6abc56e892c2: dshow: use distinct buffers per stream. [07:22] <Zeranoe> Looks like some updates on the -ac 2 channel downmix thread: http://ffmpeg.zeranoe.com/forum/viewtopic.php?f=17&t=1851 [13:18] <cone-46> ffmpeg.git 03James Darnley 07master:f02832904dbc: lavc/libx264: add a warning when using qscale [13:26] <BBB> re bug 3486, it's bicublin, not bicubiclin [13:27] <BBB> (I don't have a bug tracker account so can't do it myself) [15:42] <pawelrutka> are there any people to ask about using ffmpeg C Api ? [16:17] <cone-46> ffmpeg.git 03Matt Oliver 07master:cd5cf395f65a: Additional icl inline asm fix. [16:42] Action: michaelni wonders why people ask a question and then leave the channel 3 minutes later [16:43] <Compn> michaelni : we could setup an irc script/bot if someone joins and asks a question within 1 minute of joining it will say 'slow down! we're here but we check irc every 30 minutes, please leave the channel connected and we will respond shortly' [16:44] <michaelni> why am i thinking they will just leave after 30sec then, hmm [16:45] <cone-46> ffmpeg.git 03rogerdpack 07master:0b4c5c99bc6f: avformat/udp: cleanup comments, add a warning if receive buffer is unexpectedly unable to be set high enough [17:04] <cone-46> ffmpeg.git 03rogerdpack 07master:3eafe1d71c09: udp: try to clarify verbiage on buffer_size [17:04] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:ae4b11df0d9e: avcodec/snowenc: make snow non experimental [17:32] <Compn> http://askubuntu.com/questions/432893/libav-deinsterlace-option [17:32] <Compn> does that apply to ffmpeg ? cant test atm [17:37] <wm4> Compn: I would think so [17:37] <wm4> yadif is still the best deinterlacer in ffmpeg, so it should be used [17:37] <Compn> sure [17:37] <Compn> but i think we could make -deinterlace an alias for yadif [17:37] <Compn> so it wont break scripts [17:38] <Compn> michaelni : ^^ ? [17:38] <wm4> because apparently we don't care about backwards compatibility around here [17:38] <Compn> i care [17:39] <wm4> btw. does ffplay not use libavfilter for its audio display? [17:39] <wm4> seems like there's custom crap in ffplay.c [17:40] <Compn> ffplay's audio visualizer was made long before libavfilter [17:40] <Compn> dont think anyone ported it to libavfilter [17:40] <Compn> mr reading ffplay.c but not looking at git blame to see how old it is [17:40] <Compn> well bunch of useless hashes arent helpful anyhow [17:40] Action: Compn misses svn blame [17:41] <wm4> yeah that code is from 2003 [17:42] <wm4> written by fabrice himself [17:42] <Compn> lol [17:46] <michaelni> wm4, libav is not ffmpeg, we do care about compatibility [17:46] <michaelni> try ffmpeg -deinterace [17:46] <michaelni> try ffmpeg -deinterlace that is [17:47] <michaelni> it should insert yadif [17:48] <Compn> :) [17:48] Action: Compn afk bbl [17:51] <cone-46> ffmpeg.git 03Peter Ross 07master:54df8b4feaf4: fate: add cine demux test [17:59] <Actium> that's mine in case anyone would like to yell at me :) [18:19] <Case> latest sources give me some problems in mingw+msys with following configure command: [18:19] <Case> ./configure --disable-programs --disable-everything --disable-static --enable-shared --disable-swresample --disable-swscale --disable-network --disable-avdevice --disable-postproc --disable-avfilter --disable-debug --enable-runtime-cpudetect --enable-decoder=mp3float --enable-decoder=vorbis --enable-decoder=aac [18:20] <Case> make stops with error libavcodec/rawdec.o:rawdec.c:(.text.unlikely+0x34): undefined reference to `ff_dsputil_init' [18:20] <Case> it's my understanding this entire raw video decoder shouldn't be building with the above config command [18:30] <cone-46> ffmpeg.git 03Diego Biurrun 07master:7513234bdd93: avfilter: more correct printf format specifiers [18:30] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:7ef8d97b1aa9: Merge commit '7513234bdd93f80593bfee3eb2da8c8f0d3992da' [18:51] <cone-46> ffmpeg.git 03Diego Biurrun 07master:cc8163e1a360: avcodec: more correct printf specifiers [18:51] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:8d024c51071d: Merge commit 'cc8163e1a3601a56f722a4720516e860bf1c6198' [19:35] <cone-46> ffmpeg.git 03Diego Biurrun 07master:46caba4a6523: float_dsp: Replace arch optimization ifdefs by if cascade [19:35] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:cd1d4b51e3ed: Merge remote-tracking branch 'qatar/master' [19:38] <michaelni> Case, cant reproduce a build issue with that command [19:38] <michaelni> that is on linux havnt tries mingw [19:40] <michaelni> Case, try a make distclean and if that doesnt help look at config.log [21:22] <cone-46> ffmpeg.git 03Nicolas George 07master:7b9a310d0e27: lavu: add AV_DYNARRAY_ADD() macro. [21:22] <cone-46> ffmpeg.git 03Nicolas George 07master:b0dcf7653064: lavu/mem: reimplement the dynarray functions with the macro. [21:22] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:9ba119029796: avutil/mem: avoid using intptr_t to access void* in av_dynarray_add() [21:27] <wm4> this line is so wrong: https://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavutil/dynarray.h;h=3a3d… [21:27] <wm4> a log call just for resizing an array? [21:27] <wm4> probably even accidentally in here [21:53] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:82c742377a16: avutil/dynarray: remove debug av_log() from AV_DYNARRAY_ADD [21:53] <michaelni> wm4, removed [22:00] <wm4> thanks [22:11] <cone-46> ffmpeg.git 03Diego Biurrun 07master:1df0b061621b: nuv: Reuse the DSPContext from RTJpegContext [22:11] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:32da2fd42265: Merge commit '1df0b061621b10edde87e3ab7ea83aed381c574f' [22:16] <cone-46> ffmpeg.git 03Diego Biurrun 07master:1a8d0cf77ed2: x86: dsputil: Move inline assembly macros to a separate header [22:16] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:8fbc6e59118d: Merge commit '1a8d0cf77ed2611e542ae98f341d4c43a04467bd' [22:23] <cone-46> ffmpeg.git 03Diego Biurrun 07master:600b854ad817: imgconvert: Move ff_deinterlace_line_*_mmx declarations out of dsputil [22:23] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:d6d3cfb0aa23: Merge commit '600b854ad8173995518bd917e7f86120b5505088' [22:29] <cone-46> ffmpeg.git 03Diego Biurrun 07master:f9279ee74936: dsputil: Move ff_zigzag_direct and ff_crop_tab declarations to mathops.h [22:29] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:672c610d8439: Merge commit 'f9279ee74936b4990fc46287c27816ef828cc06c' [22:40] <Case> michaelni: I tried it in a linux VM too where it worked. make distclean doesn't help. I don't know what to look for in the huge config.log file, but I uploaded it to http://www.saunalahti.fi/~cse/temp/config.log [22:40] <Case> I can make it compile by copying the rawdec.c file from FFmpeg 2.1 sources [23:03] <cone-46> ffmpeg.git 03Diego Biurrun 07master:322a1dda973e: dsputil: Refactor duplicated CALL_2X_PIXELS / PIXELS16 macros [23:03] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:77bc34297596: Merge commit '322a1dda973e802db7b57f2007fad3efcd5bab81' [23:17] <cone-46> ffmpeg.git 03Diego Biurrun 07master:6655c933a887: x86: dsputil: Move fpel declarations to a separate header [23:17] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:9333bba6ed32: Merge commit '6655c933a887a2d20707fff657b614aa1d86a25b' [23:21] <cone-46> ffmpeg.git 03Diego Biurrun 07master:ae116cd3ed90: ppc: dsputil: Move hpeldsp-related declarations to a separate header [23:21] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:816d7936a714: Merge commit 'ae116cd3ed908d28b69d5198712217ec743d74f6' [23:26] <cone-46> ffmpeg.git 03Diego Biurrun 07master:82dd1026cfc1: x86: dsputil: Move hpeldsp-related declarations to a separate header [23:26] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:74fed968d1b6: Merge commit '82dd1026cfc1d72b04019185bea4c1c9621ace3f' [23:44] <cone-46> ffmpeg.git 03Diego Biurrun 07master:57f09608e160: dsputil: Move thirdpel-related bits into their own context [23:44] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:9d6a27d53365: Merge commit '57f09608e1600d1cf1679885a46f5004d522d68f' [23:50] <cone-46> ffmpeg.git 03Diego Biurrun 07master:4f4840377f80: lagarith: Fix typo in printf format string [23:50] <cone-46> ffmpeg.git 03Michael Niedermayer 07master:de182701ac9e: Merge remote-tracking branch 'qatar/master' [00:00] --- Sun Mar 23 2014

1 0

ffmpeg.log.20140322
by burek 23 Mar '14

23 Mar '14

[01:16] <Sander> Hi can someone help me with a mapping problem [01:22] <relaxed> Sander: I can try [01:23] <relaxed> pastebin "ffmpeg -i input" and tell me what you're after. [01:23] <Sander> i do it in combination with ffserver but i will past [01:23] <relaxed> you lost me at ffserver [01:24] <llogan> relaxed: let's go shopping. [01:24] <Sander> aw :( [01:24] <relaxed> llogan: shopping for what? is there a pub nearby? [01:25] <llogan> i could use a pub about now... [01:25] Action: llogan gets back to video editing, slowly, distractingly, and inefficently [01:26] <relaxed> are you working in OS X? [01:26] <relaxed> Sander: just pastebin it [01:27] <Sander> oke moment [01:27] <llogan> relaxed: no. a (excuse me as I sipt), a Wi-wi-wi...I can't say it. a work computer that I rarely use. [01:28] <llogan> *spit [01:29] <relaxed> premiere? [01:30] <llogan> yes [01:31] <relaxed> Sander: which streams do you want out of that? [01:32] <Sander> Stream #0:0 and Stream #0:7 [01:33] <relaxed> Sander: launch ffmpeg -i udp://@address -map 0:0 -map 0:7 ... [03:09] <Guest19387> hello [03:10] <Guest19387> in uotput i have about 800kb/s stream http://pastebin.com/khEtMXkq . how can i increase badwith up ti 1200kbps ? [03:10] <Guest19387> sorry 1200 kbps [03:11] <Jack64> /root/bin/ffmpeg -f flv -i rtmp://173.199.120.59:1935/CAREWORLDTVL/CAREWORLDTVL.stream-sd -acodec ac3 -vcodec -b:v 1200k [03:11] <Jack64> ? [03:12] <Jack64> after -vcodec libx264 sorry [03:13] <Guest19387> Jack64, thank you [03:14] <Jack64> you're welcome :) [03:14] <llogan> Guest19387: use -maxrate and -bufsize [03:15] <llogan> https://trac.ffmpeg.org/wiki/EncodingForStreamingSites [03:16] <Jack64> llogan: wow didn't realize that, thanks llogan [03:16] <llogan> that is for justin.tv, etc but it will describe VBV, which you should be using for cases with network speed is a limitation [03:17] <Jack64> yes, it's very clear and a propos. [03:39] <mjsmith> anyone used a windows named pipe as ffmpeg input? I keep getting pipe:pipename: Not enough space. [03:49] <Guest19387> guys what command i shuld use for setting jitter 100 ? [03:55] <Guest19387> ? [03:55] <Guest19387> what command i shuld use for setting jitter 100 ? [04:13] <mjsmith> relaxed, too bad he never told you the solution =\ http://mplayerhq.hu/pipermail/ffmpeg-devel-irc/2013-September/001636.html [04:19] <relaxed> mjsmith: what was the question again? [04:19] <mjsmith> anyone used a windows named pipe as ffmpeg input? I keep getting pipe:pipename: Not enough space. [04:19] <mjsmith> apparently the CreateNamedPipe() syntax is wrong or something [04:24] <relaxed> mjsmith: pastebin what you're trying. I'm at work and have access to some Windows 7 machines. [04:33] <mjsmith> its alright relaxed i think im going to call it a day on this [04:38] <Jaypan> Hello, I have just installed FFMPEG according to this page: https://trac.ffmpeg.org/wiki/CentosCompilationGuide [04:38] <Jaypan> I installed it as my root user, [04:39] <Jaypan> and FFMPEG is now available to my root [04:39] <Jaypan> But I need it to be available for all users, including cron [04:39] <Jaypan> The guide did not show how to do that [04:39] <Jaypan> can someone provide some assistance? [05:15] <Jaypan> I guess not. Thanks anyways! [08:18] <Keshl> Is there some magical set of parameters that'll cause ffmpeg to output a video with a steadily increasing bitrate that can be used to test when a device hits its limit, somehow? [08:25] <jbermudes> Keshl: using -t to set a max time you could loop through ever increasing bitrates [08:26] <jbermudes> Record for 5 seconds at rate X, then start recording again at a higher rate [08:26] <Keshl> I didn't follow half that. o_o [08:26] <Keshl> ...That post was not helpful to anybody. Uh, lemme reword that. [08:27] <Keshl> Are you saying that I should try recording on the device in question? [08:31] <jbermudes> Like, you could do it manually, recording for a few seconds at a certain bitrate, then try again at a higher one, or you could try to automate it with a script [08:31] <Keshl> Got'cha, oÉo. [08:31] Action: Keshl snugs on and such -É- [11:29] <voip> hello, guys [11:31] <voip> di you have any chart or someting like for aproximatly calculate how many streams can conver for example 1 xeon CPU ? [11:31] <voip> do [11:31] <relaxed> one per thread [11:32] <relaxed> what do you mean by streams? [11:33] <voip> relaxed, you mean if i need convert 1- streams a ineed 6 cores cpu with 12 threads? [11:33] <voip> for 10 streams [11:34] <relaxed> it's possible to encode 12 files at one time, each using their own thread. [11:35] <relaxed> "stream" is ambiguous when it comes to ffmpeg/encoding [11:36] <relaxed> did that answer your question? [11:37] <voip> i mean videostream converting [11:37] <voip> transcoding [11:37] <relaxed> yes, i answer that above ^^ [11:37] <relaxed> answered* [11:38] <voip> relaxed, intresting i have xeon cpu with 4 threats, but when i transcoding ip videostream all my 4 cpu "kernels" loads for 98% [11:39] <relaxed> You can tell ffmpeg to encode using only one thread, but by default it uses them all to transcode as quickly as possible. [11:39] <relaxed> ffmpeg -i input -threads 1 .... output [11:40] <voip> oooo, ok [11:40] <relaxed> libx264 uses all threads by default. I assume that's what you're seeing. [11:40] <voip> ok i see [11:40] <voip> one more question [11:41] <voip> can i use GPUs for acsleration [11:41] <relaxed> no [11:41] <voip> ok, tahnk you [11:42] <relaxed> you're welcome [13:04] <obiwahn> hi [13:04] <obiwahn> do you know why ffmpeg is not in debian sid? [13:05] <JEEB> obiwahn, because debian currently uses libav, which has the binary called avconv [13:06] <obiwahn> what a shame ... the world seems to use ffmpeg:) [13:10] <obiwahn> ok but avconv is quite useable as well:P [13:11] <sacarasc> The avconv channel, though, is #libav. [13:12] <JEEB> the main problem is that libav has had slow release cycles, and then debian and friends have had to migrate all their apps to the new APIs :P The latter of course counts for FFmpeg-using distros as well. [13:12] <JEEB> so the sad joke is that up to ubuntu 13.10 you've basically had a libav version from 2012 come with your distro :| [13:13] <obiwahn> Why did they switch? [13:13] <JEEB> because the maintainer was one of the people forking out [13:16] <JEEB> of course, old FFmpeg (up to 2010 or so? before the fork) was rather problematic, and I can see why some people would have seen libav as something better at the time. Of course current FFmpeg has been changed by the forking as well, and most of the problems of the old aren't really there :P (of course it has new problems, but so does Libav) [13:17] <obiwahn> http://lists.alioth.debian.org/pipermail/pkg-multimedia-maintainers/2011-De… [13:18] <JEEB> yes, the release schedule over at FFmpeg was rather quicker, although I'm not sure if it really manifested itself that much at that point, as the fork was done in early 2011 [13:20] <JEEB> ok, Libav wasn't /that/ bad at releasing at that point yet [13:20] <JEEB> because 0.7 was in June 2011 and 0.8 was in Jan 2012 [13:21] <JEEB> the real breeze then came with the <space> between 0.8 and 9 [13:22] <JEEB> which was a bit less than a year [13:25] <JEEB> and the joke kind of goes on since Libav 10 has been in preparing-for-release mode since december and still isn't out :P [13:26] <JEEB> (went to "beta" in february) [14:40] <ValdikSS> Hello everyone! Can somebody tell me why ffmpeg changes subtitle .ass file even if -c copy is set? [15:00] <Actium> hey there, i have a minor issue with the current ffmpeg git version. ./configure --enable-libopus results in "ERROR: opus not found"., although opus (including development files) is installed. i dug a little deeper and and found out that the problem can be solved by installing pkg-config. unfortunately the error message does not tell you just that. [15:37] <Actium> well ... i just took a look (or rather a couple of looks :) at the configure script and apparently it is supposed to warn about a missing pkg-config, but doesn't [15:38] <ValdikSS> Please test this: http://yadi.sk/d/NM69ZxZQKzNn7 and http://yadi.sk/d/qDFmsDPQKzNoQ [15:39] <ValdikSS> Is there any difference in subtitle effects for you? Please tell me the OS and the player on which you were testing [15:40] <Actium> no need to test anything as long as the ./configure fails [15:40] <ValdikSS> https://lists.ffmpeg.org/pipermail/ffmpeg-user/2014-March/020583.html [15:41] <Actium> sorry, thought you were refering to my issue ;) [15:41] <ValdikSS> no sorry [15:42] <klaxa> mpv 0.3.6 arch linux, ffmpeg remuxed subtitles seem broken [15:42] <klaxa> i wonder why though [15:43] <ValdikSS> cmplayer and vlc@archlinux seems broken too [15:43] <ValdikSS> but if i extract the subtitles to .ass file with ffmpeg and include it with -i subtitle.ass, it works as expected [15:43] <klaxa> maybe libass doesn't produce libass compatible subs :x [15:43] <ValdikSS> I though on mkv tags or something, but no [15:44] <klaxa> i think there was a similar issue like one and a half years ago [15:44] <klaxa> let me find the logs [15:44] <ValdikSS> Another strange thing that mplayer(1) plays fine [15:45] <ValdikSS> klaxa: that's not only this file or this subtitles issue, it happens to me on a lot of files [15:48] <klaxa> ValdikSS: https://gist.github.com/klaxa/197efb68e549d01193b8 [15:48] <klaxa> maybe there's a bugreport for that [15:48] <klaxa> what version of ffmpeg are you using? [15:48] <klaxa> note that this conversation dates back to november 2012 [15:49] <ValdikSS> I tried the latest git, it's the same [15:49] <ValdikSS> That file was muxed using 2.1.4 [15:51] <ValdikSS> klaxa: I'm downloading libav static build right now, would test with it [15:52] <ValdikSS> > [15:52] <ValdikSS> Nov 10 00:59:03 <klaxa> if you extract the subs from the encoded file and use the .ass file directly it gets rendered properly [15:52] <ValdikSS> Nov 10 00:59:25 <klaxa> if you use the embedded subtitles it doesn't work [15:52] <ValdikSS> The same for me [15:53] <klaxa> like i said, the same issue, just like a year and a half ago [15:53] <ValdikSS> Even with ffmpeg-muxed subtitles with changed lines [15:53] <klaxa> maybe CalimeroTeknik didn't open a ticket [15:54] <ValdikSS> klaxa: >Nov 10 01:17:29 <klaxa> if i use the .ass i get with mkvmerge and mux it with ffmpeg it doesn't work [15:54] <ValdikSS> it works for me btw [15:56] <klaxa> with codec copying or "re-converting" ass to ass? [15:56] <ValdikSS> both [15:56] <ValdikSS> tried with latest libav git, the same issue [16:17] <ValdikSS> klaxa: if there is no default-flag for subtitles written, subtitles won't appear at all [16:18] <klaxa> you will have to enable them in that case [16:18] <klaxa> because they are not enabled by default [16:18] <klaxa> afaik [16:18] <ValdikSS> klaxa: if i set default-flag to everything, it's set to "default" (not to yes or no), but subtitles would appear correctly [17:14] <teamen> I'm playing a playlist (m3u) file with vlc & ffmpeg (cvlc --demux ffmpeg --codec cedar --vout cedarfb --no-osd list.m3u). However, the format of the input file can't be detected ("avformat demux debug: couldn't guess format"). How can I make this work? [17:16] <teamen> The content of the M3U file is a list of video streams, when forcing the video format (--ffmpeg-format mpegts), I receive "avformat demux error: Could not open list.m3u: Operation not permitted" [17:27] <ValdikSS> teamen: you can't use playlists in vlc with forced demuxer afaik [17:27] <ValdikSS> OR you can try to "unblacklist" ffmpeg m3u demuxer [17:29] <teamen> ValdikSS: how can I do this? [17:34] <ValdikSS> teamen: i can't remember [17:34] <ValdikSS> why do you want to use demux globally anyway? [17:38] <teamen> I'm running this on particular hardware that require these arguments: http://linux-sunxi.org/VLC [17:40] <ValdikSS> teamen: well, since I don't remember unblacklisting parameter, you can try --demux playlist,ffmpeg [17:41] <ValdikSS> teamen: you'd better ask in vlc channel [17:42] <teamen> ValdikSS: that seems to work, very nice, thanks very much! [17:56] <ValdikSS> klaxa: mxplayer on android plays ffmpeg remux without issues [18:06] <Hfuy> Hello. [18:06] <Hfuy> Can anyone suggest some options I could use with the aim of creating an mp4 that's easy to replay? I need 1080p25 on a Turion 64 x2. [18:06] <Hfuy> Or, I suppose, another codec. Video in question is 30 seconds long and would need to play from an SD Card. [18:07] <Actium> have you tried x264 with fastdecode tune? [18:08] <Hfuy> I don't have x264 to hand, just ffmpeg. [18:08] <Hfuy> Actually to be accurate I need 1200p25 decode on a turion 64. These are 1920x1200 monitors :/ [18:09] <Actium> what codecs does your ffmpeg support? [18:09] <Actium> i assume/hope x264 is linked in :) [18:09] <Hfuy> It's a windows 64 bit build. It presumably does. [18:10] <Actium> when you took one of the pseudo official static builds it should [18:10] <Hfuy> Actually it looks a bit like -lavdopts threads=2 is enough to fix it. [18:10] <Hfuy> (when playing with mplayer) [18:10] <Hfuy> It's not -perfect-, but... [18:11] <Actium> have you tried mplayer -benchmark? it should tell you where the bottleneck is [18:11] <Hfuy> Didn't even know it existed. [18:11] Action: Hfuy tries it [18:12] <Hfuy> Hmm. Not sure what these results are trying to tell me. [18:13] <Hfuy> VC 13.6 VO 19.3 A 0.00 Sys 0.753s = 33.7s [18:13] <Hfuy> then another one [18:13] <Hfuy> VC 40.5 VO 57.6 A 0.00 Sys: 2.312 = 100% [18:13] <Hfuy> If my interpretation is right, it isn't bottlenecking at all. It's doing it right. No? [18:15] <Actium> how long is your video exactely? [18:16] <Actium> this is what a 10s clip's benchmark looks like when played back on my machine: [18:16] <Hfuy> 00:00:29:22 @ 25p [18:16] <Actium> BENCHMARKs: VC: 0.042s VO: 0.078s A: 0.035s Sys: 9.837s = 9.992s [18:16] <Actium> BENCHMARK%: VC: 0.4185% VO: 0.7765% A: 0.3474% Sys: 98.4576% = 100.0000% [18:17] <Actium> your first line tells you that mplayer spends a lot of time on video output [18:17] <Hfuy> On the basis that it totalled more than the theoretical duration, it's presumably bumming it up somewhere. [18:17] <Actium> yep [18:17] <Hfuy> This is an XP box. [18:17] <Hfuy> (well, the target machine is) [18:17] <Actium> in my case most time is spend on Sys (i.e. idling/sleeping) [18:17] <Hfuy> I assumed it would default to directshow output. [18:18] <Actium> try a different output first, if you found the optimal one you can try to tune the time spent on decoding [18:18] <Hfuy> Right now I'm simply trying to get this damn laptop to drive a 1920x1200 display correctly - at all. [18:18] <Hfuy> Bleagh. [18:19] <Hfuy> Ancient laptop lacks DVI. Monitor fails to understand concept of widescreen analogue. [18:20] <Actium> i didnt even know vga supported 1920x1200 :) [18:20] <Hfuy> I fear it doesn't. It's doing that strange sliding-around thing. [18:21] <Hfuy> I have an HTA (remember those?!) which draws alternating columns of red and blue to encourage monitors to phase detect widescreen VGA properly. It's got a nearly 100% success rate. Wish me luck... [18:22] <ValdikSS> klaxa: http://yadi.sk/d/5l7wnAECKzrnj another example [18:22] Action: Actium keeps his fingers crossed [18:22] <Hfuy> ...bugger [18:24] <Hfuy> Nnnnn...no. [18:24] Action: Hfuy rants [18:26] <cody_> i want to decode and display a h264 bitstream in realtime [18:26] <cody_> how could i do that [18:26] <cody_> no buffering, no audio [18:26] <cody_> http://ffmpeg.org/doxygen/trunk/h264_8h.html [18:26] <cody_> is that the right place to look at? [18:31] <Aiena> can ffmpeg be used to take a screenshot and encode it as png or jpg ? what would be the command for that ? [18:31] <Aiena> screenshot not video [18:31] Action: Hfuy pokes display settings w/big stick and succeeds in the end [18:32] <Hfuy> Is there a switch to make mplayer go fullscreen on an alternate display? It's insisting on using the laptop's built-in. [18:33] <Actium> --fs for fullscreen, but i dont know about the alternate display [18:34] <Hfuy> I'm just getting green overlay key. [18:41] <llogan> Aiena: linux or windows? [18:42] <Aiena> linux [18:42] <Aiena> llogan sorry I forgot to mention the platform [18:42] <llogan> Hfuy: ffmpeg has the -tune fastdecode option as mentioned by Actium [18:43] <llogan> when using the encoder libx264 [18:43] <Aiena> llogan I am using ffmpeg 2.1.4 on Suse 13.1 [18:43] <llogan> http://ffmpeg.org/ffmpeg-devices.html#x11grab [18:43] <llogan> ffmpeg -f x11grab -framerate 25 -video_size 1680x1050 -i :0.0 -vframes 1 output.png [18:44] <Hfuy> llogan: I'll get back onto that in a moment. [18:44] <llogan> ffmpeg -f x11grab -framerate 25 -video_size 1680x1050 -i :0.0 -vframes 1 -q:v 2 output.jpg [18:44] <Hfuy> Right now I'm distracted by trying to make mplayer work. [18:44] <Aiena> llogan that will get just one image ? [18:44] <llogan> yes [18:44] <Aiena> Great thank you :) [18:44] <llogan> -vframes 1 makes one image [18:45] <Aiena> ok llogan ok just a moment my script is a lttle different I make make an error in converting that line [18:45] <Aiena> let me pastebin it and you suggest How I should tailor the line [18:45] <llogan> we don't really support scripts here. just try it manually, and if it works then you can mess with the script. [18:54] <Aiena> thanks [18:55] <Aiena> llogan -s and -video_size is the same right ? [18:56] <llogan> Aiena: i guess so [18:56] <llogan> although -r and -framerate have different behaviors, IIRC [18:56] <llogan> not that frame rate really matters much for your case [18:56] <Aiena> Ok [18:58] <Aiena> llogan and the -i can accept an offset too right like "-i :0.0+5,5" ? [18:58] <Aiena> -i is same as in encoding for videos ? [18:58] <llogan> yes for the offset. the link i provided has examples. [18:58] <llogan> sorry, i don't understand the -i question [18:59] <Aiena> ok no problem I will test [18:59] <Aiena> if it does not work then I will elaborate and ask [18:59] <Actium> this might be of help, too: https://trac.ffmpeg.org/wiki/How%20to%20grab%20the%20desktop%20%28screen%29… [18:59] <Aiena> thank you so much llogan [18:59] <llogan> you can install xwininfo to get window coordinates [18:59] <llogan> then run it and click the window you want and it will barf out the infos [19:00] <Aiena> llogan that is what my script is already doing [19:00] <Aiena> then it calculates and offset to remove the window decorations [19:00] <Aiena> and that data is fed into ffmpeg -i option [19:00] <Aiena> I was just wonderig if -i works the same in this case (a still) as it does for encoding a video [19:01] <Aiena> as explained https://trac.ffmpeg.org/wiki/How%20to%20grab%20the%20desktop%20%28screen%29… [19:03] <Aiena> llogan yes it does [19:03] <Aiena> works great [19:03] <Aiena> thank you so much for your help llogan [19:03] <Aiena> I am very grateful. [19:03] <llogan> no problem. i have to go pack now for a 3 mile hike to a cabin in the woods by the ocean. [19:05] <Aiena> Alright. Travel safe and enjoy [19:05] <mjsmith> is there a way to make ffmpeg pad an audio input in order to make it in sync with a video captured separately? [19:06] <mjsmith> the input being from a pipe [19:06] <Aiena> yes mjsmith I had found a guide on google for it. [19:06] <Aiena> googling should help :) [19:06] <Aiena> let me look [19:06] <mjsmith> do you remember what you googled? :D [19:06] <mjsmith> i was trying to pad this manually by fwriting zeros but im having no luck [19:08] <Aiena> mjsmith but this may be more useful http://stackoverflow.com/questions/7426179/add-audio-with-an-offset-to-vide… [19:08] <Aiena> apad is there too but that is for padding at the end [19:08] <mjsmith> unfortunately its not a simple offset shift [19:09] <mjsmith> https://lists.ffmpeg.org/pipermail/ffmpeg-user/2013-December/018796.html more in line with what this guy was doing [19:27] Action: Hfuy punches the air [19:27] <Hfuy> Yyyyyyessss! [19:27] <Hfuy> Finally got it working! [19:38] <Hfuy> Right, back to the encoding issue. Can anyone recommend a reasonably up-to-date site with h.264 encoding examples? [19:38] <Hfuy> I've tried just ffmpeg -i foo -b:v whatever out.mp4 but the quality is awful [19:42] <Mavrik> well [19:42] <Mavrik> telling ffmpeg what quality you want actually helps [19:45] <Hfuy> any suggestions [19:46] <Hfuy> I've found "-crf" used variously and it seems to help but I have no idea what it means or whether 22 is a sane value. [19:47] <Mavrik> https://trac.ffmpeg.org/wiki/x264EncodingGuide [19:47] <Mavrik> this. :) [19:47] <Actium> dammit, mere 10 seconds too late :) [19:48] <Actium> but the wiki is great [19:48] <Hfuy> Yes, that's where I got it. [19:49] <Actium> 23 is the default value and is quite alright. depending on your demands you could decrease it a little. [19:49] <Actium> i tend to use 21 for my ... ehm ... bluray backups :) [19:50] <JEEB> basically there is no perfect value [19:50] <JEEB> use your eyes and ~5000 frames of your source, preferably from various points [19:50] <Actium> yep [19:50] <JEEB> and the highest value that still looks good to your eyes is the thing for that source [19:50] <JEEB> you probably will have to poke the value a bit depending on the source, but similar sources can be encoded with the same value [19:51] <Actium> this is a good description (link from previously mentioned wiki page): http://slhck.info/articles/crf [19:52] <mjsmith> so anyone have an idea how to do this? Basically ffmpeg reading audio from stdin. If nothing is coming from stdin then write padding [19:53] <Hfuy> 21 certainly is not the default value. [19:53] <Hfuy> 21 looks OK. [19:53] <Hfuy> Omitting -crf produces unwatchable results. [19:53] <mjsmith> i think crf 23 or 24 is the default [19:54] <Hfuy> Whatever the default is, it doesn't produce useful results. [19:54] <Mavrik> hmm, is your ffmpeg old? :) [19:55] <Hfuy> Ten minutes is old for ffmpeg [19:55] <Hfuy> ffmpeg version N-57245-gf6b56b1 Copyright (c) 2000-2013 the FFmpeg developers [19:55] <Mavrik> hmm [19:56] <Mavrik> that's probably not a sane default then [19:56] <sacarasc> October, eh? [20:01] <Hfuy> Not a clue. [20:01] <Hfuy> Sounds reasonable. [20:01] <Hfuy> Most people just include a date, you know. [20:01] <sacarasc> Not reasonable... [20:01] <sacarasc> Fri, 18 Oct 2013 21:38:54 +0000 (23:38 +0200) [20:01] <sacarasc> commit f6b56b1f26a1e40a47d85d24b42cd5475cd3b04f [20:02] <sacarasc> Exact. [20:02] <sacarasc> It's too old, update. \o/ [20:02] <Actium> that's not very old by distribution standards :) [20:05] <Hfuy> I think this behaviour has been reasonably normal for some time [20:06] <Hfuy> Does anyone know if mplayer supports the whole nvidia ion mp4 acceleration thing, and if so how to use it? [20:07] <Actium> sorry, no clue about that windows stuff [20:08] <Hfuy> It's just that my netbook is capable of playing 1080p24 h.264 in realtime under windows, but it has ubuntu at the moment. [20:08] <Actium> vdpau [20:08] <Mavrik> you need to make vdpau work [20:09] <Hfuy> I just installed mplayer on it, and it did mention vdpau. [20:09] <Hfuy> Perhaps there are other problems. [20:10] <Mavrik> make sure you have nvidia drivers (not the noveau stuff) and vdpau libraries installed [20:10] <sacarasc> Try mplayer -vc ffvdpauh264 -vo vdpau blah.mkv [20:10] <Mavrik> then see if mplayer uses vdpau to play [20:10] <sacarasc> If that doesn't work, then you can't. \o/ [20:11] <Hfuy> it's actually a lot better if I encode it as pix_fmt yuv420p [20:12] <Mavrik> using anything else than YUV 4:2:0 won't be compatible with most HW decoders [20:12] <Hfuy> that may be the difference, then [20:13] <Hfuy> sacarasc's suggestion just gets me "cannot find codec matching selected -vo..." [20:13] <Hfuy> Still a bit marginal, though [20:14] <Actium> what's the benchmark say? VO still the bottleneck? [20:15] <Hfuy> I'm running a benchmark right now, wait one. [20:15] <Hfuy> This is a different laptop, y'understand. [20:15] <Actium> now i do :) [20:15] <Hfuy> this one's under ubuntu [20:15] <Hfuy> which is probably a bit ambitious for a little atom netbook [20:15] <Hfuy> ubuntu's showy GL UI may not be helping [20:16] <Hfuy> 84% of time in VC. It's presumably the codec. [20:16] <Actium> which is? [20:17] <Hfuy> ffmpeg -i foo.avi -c:v libx264 -crf 22 -pix_fmt yuv420p bar.mp4 [20:17] <Hfuy> oh, and "-tune fastdecode" [20:17] <Actium> what profile? baseline/main/high? [20:17] <Hfuy> No idea. [20:17] <Hfuy> Can I tell? [20:18] <Actium> ffprobe [20:18] Action: Hfuy sudo apt-get install libav-tools [20:18] <Actium> wrong channel :P [20:19] <Mavrik> yeah, not using ffmpeg will hurt our chances of helping :P [20:19] Action: Actium gives Mavrik a high five [20:20] <Hfuy> Video: h264 (High) [20:21] Action: Hfuy adds -profile baseline [20:21] <Mavrik> ugh [20:21] <Mavrik> do main if you have 1080p content PLEASE :P [20:21] <Actium> try -profile:v main [20:22] <Mavrik> baseline downright murders any quality at a given bitrate :) [20:22] <Hfuy> well, I've done baseline now [20:22] <Hfuy> let's see how revolting it is [20:22] <sacarasc> If you're reencoding anyway, why not resize? [20:22] <Hfuy> I'm not reencoding, I'm encoding for the first time. Source is uncompressed AVI. [20:23] <Hfuy> Looks OK to me. [20:23] <Hfuy> OK problem solved moving on [20:30] <Hfuy> Now the next problem is something you probably can't help me with. I'm driving three displays on the set of a movie. [20:30] <Hfuy> And I've run out of laptops :) [20:31] <Hfuy> Do you think people will still be using laptops in 2090? [20:31] <Actium> i probably won't care due to being 6 feet under by then :) [20:31] <Hfuy> Well that's a defeatist attitude! [20:34] <Actium> even I'll have reached triple digit age by then [20:35] <Actium> but 75 years is a long time for post-post-post-tablets to wipe out notebooks [20:35] <Hfuy> Yeah [20:36] <Hfuy> I just need a way of getting more video displays into this movie set. [20:36] <Hfuy> I have two identical 24" TFTs. Could have got a third, but don't have another lappy to drive it with. [20:37] <Hfuy> Annoying. [23:38] <curfont> what decodes faster, a series of gzipped BMPs, or PNGs [23:39] <curfont> or any alternative for lossless monochrome video? [23:40] <sacarasc> Time for a test! [23:40] Action: sacarasc goes off to not do it. [23:43] <curfont> on another note.. h264 decoding is so slow :( [23:44] <curfont> I can only decode my CCTV footage at 100FPS, so to analyse the footage for 24 hours it takes me 8 hours [23:44] <sacarasc> Get a bigger, beefier CPU! [23:45] <curfont> I have an i7, one of the latest ones [23:45] <curfont> Does ffmpeg use the GPU? [23:46] <curfont> Its strange because the decoding isnt using much cpu (13%) [23:46] <sacarasc> What resolution is this video? [23:48] <curfont> 1080p [23:48] <curfont> Main(a)L4.1 [23:49] <sacarasc> You should be able to decode at >100FPS with a recent i7... Are you using all the threads? [23:50] <curfont> Maybe not, I am not too sure on the parameters. Ideally I would like to use it in an app (as a library) instead of as an executable [23:50] <curfont> But it wouldnt work multithreaded in that case right? [23:58] <sacarasc> You should be able to do it multithreaded... 4 threads to decode the video, 4 to do whatever else you need... [00:00] --- Sun Mar 23 2014

1 0

ffmpeg-devel.log.20140321
by burek 22 Mar '14

22 Mar '14

[00:25] <cone-796> ffmpeg.git 03wm4 07master:8b6136d3d18f: vaapi: switch ff_vaapi_get_surface_id from Picture to AVFrame [00:25] <cone-796> ffmpeg.git 03Michael Niedermayer 07master:19139d8516dd: Merge remote-tracking branch 'qatar/master' [00:34] <wm4> now vdpau works again [03:46] <cone-558> ffmpeg.git 03Michael Niedermayer 07master:45ad7bc42328: avcodec/snowdec: provide information about the wavelet with -debug 1 [03:46] <cone-558> ffmpeg.git 03Michael Niedermayer 07master:1cd107f63792: avcodec/x86/snowdsp: add missing clobbers to inner_add_yblock_bw_8_obmc_16_bh_even_sse2() and inner_add_yblock_bw_16_obmc_32_sse2() [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:87d335c00756: avcodec/parser: reset indexes on realloc failure [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:ea23dcc4986e: avfilter/ff_insert_pad: fix order of operations [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:84e1aa7768a2: avcodec/jpeglsdec: check err value for ls_get_code_runterm() [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:c44ac4d81e86: dnxhdenc: fix mb_rc size [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:e0f351a97955: avcodec/vmnc: Check that rectangles are within the picture [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:605c3f051540: avcodec/vc1: reset fcm/field_mode in non advanced header parsing [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:356f8634e991: wmalosslessdec: make mclms arrays big enough for whats written into them. [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:954ce59a0abb: avcodec/wmalosslessdec: fix mclms_coeffs* array size [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:d234e1d333c9: avformat/mpegtsenc: Check data array size in mpegts_write_pmt() [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:573d5fdedae7: avcodec/msrle: use av_image_get_linesize() to calculate the linesize [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:85162a60b8f9: avcodec/snow: split block clipping checks [05:52] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:3926ebc7029d: vorbisdec: Check bark_map_size. [05:57] <cone-558> ffmpeg.git 03Michael Niedermayer 07release/0.9:d55059d048bd: update for 0.9.4 [06:20] <cone-558> ffmpeg.git 03Michael Niedermayer 07fatal: ambiguous argument 'refs/tags/n0.9.4': unknown revision or path not in the working tree. [06:20] <cone-558> Use '--' to separate paths from revisions [06:20] <cone-558> refs/tags/n0.9.4:HEAD: avcodec/x86/snowdsp: add missing clobbers to inner_add_yblock_bw_8_obmc_16_bh_even_sse2() and inner_add_yblock_bw_16_obmc_32_sse2() [09:40] <ubitux> BBB: so, i grabbed the raw of etv, i'll try to remake the encode in the next days [09:40] <ubitux> ...if i figure out how bluray supports works in ffmpeg [09:59] <ubitux> BBB: btw, i wonder if we should test with a --tune grain for that sample [10:14] <ubitux> ah but well that won't do with ssim tuning [10:28] <j-b> 'morning [11:46] <BBB> ubitux: you want to re-encode for 5 days again? [11:46] <ubitux> it's in progress [11:46] <BBB> (x N encodes) [11:46] <ubitux> :p [11:46] <ubitux> i'm redoing all the encode currently [11:47] <BBB> well that's a ton of work, but I guess it's good [11:47] <BBB> we should write a faster encoder ;) [11:47] <ubitux> it's fine the script is already done [11:47] <ubitux> i just have to wait [11:48] <ubitux> i've roughly the same sample (maybe a few frame mismatch) [11:48] <ubitux> cropped the same, so same resolution [11:49] <ubitux> that should provide more accurate results :) [11:49] <ubitux> and we'll get fixed on the high bitrate quirck [11:59] <BBB> yeah, that was the one where x264 beat libvpx-vp9, right? [12:11] <ubitux> BBB: yes [12:12] <ubitux> the sample is about 4 seconds off the previous one [12:12] <ubitux> shouldn't affect much the results [12:21] <BBB> as long as you re-do the x264 encodes also, it's fine, yes [12:21] <BBB> but you were going to do that already so I suppose it's fine [12:24] <ubitux> yes i'm redoing all of them [12:39] <Compn> people love fighting about benchmarks :D [12:39] <Compn> big companies like nvidia and ati even spend time to cheat at 3d benchmark software :P [12:43] <ubitux> it's not benchmark here, it's all about the quality of then encode [12:43] <ubitux> s/then/the/ [12:43] <nevcairiel> you could also "benchmark" the quality, its not always about speed [12:43] <ubitux> the results were weird and we wanted to know if that was because the source was h264 [12:45] <BBB> yeah, I think it's a fair concern [12:45] <BBB> so good to check [15:22] <cone-58> ffmpeg.git 03Anshul Maheshwari 07master:6d25e90262e3: doc/examples/avcodec: information regarding gop_size ignored if frame->pic-type is AV_PICTURE_TYPE_I [15:28] <saste> WTF why was it renamed? [15:30] <nevcairiel> what was renamed? [15:32] <saste> encoding_decoding -> avcodec [15:32] <saste> anyway, i don't care [15:58] <michaelni> saste, see b339182eba34f28de5f1a477cdd2c84f1ef35d90 [15:59] <michaelni> i can rename it back if you prefer [15:59] <saste> michaelni, i'll send a patch in case i feel the urge [16:05] <Waraqa> Hi, Who can help me with my problem? [16:06] <Waraqa> I opened ticket #3485 [16:15] <ubitux> it's just been closed [16:15] <Waraqa> I know but I need help [16:15] <ubitux> upgrade? [16:15] <Waraqa> no, the video doesn't work with vlc also [16:16] <ubitux> vlc is probably using that ffmpeg version as well [16:16] <ubitux> just upgrade [16:16] <ubitux> also, it's a #ffmpeg question probably [16:16] <Waraqa> Ok, I will move there [17:11] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:9dca02ee5411: ffmpeg: dont call exit_program() from a signal hander [17:57] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:db55365d0f93: configure: Try to fix detection of ARM Compiler 5.04 [18:41] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:c3272674c943: avcodec/mpegvideo: make mc_mb_var_sum_temp / mb_var_sum_temp 64bit [18:42] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:e92a78a4095d: avcodec/mpegvideo: make mc_mb_var_sum / mb_var_sum 64bit [18:42] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:b4356c9cc281: avcodec/ratecontrol: make (mc_)mb_var_sum(_last) 64bit [18:42] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:2fd14f062cfc: avcodec/snowenc: allow larger mb variances [18:59] <llogan> or an info can say "-qscale is ignored with libx264. Use -crf instead. Default of -crf 23 will be used." i don't know how to implement that though. [19:01] <llogan> and 99% of users shouldn't be messing around with -qp, so defnintely don't map to that [20:25] <J_Darnley> llogan: Are you working on a patch for the qscale-libx264 ticket? [20:36] <llogan> J_Darnley: no. please do so if you would like to. [20:44] <cone-58> ffmpeg.git 03wm4 07master:76c8fbc5f22c: http: never send 'Cookie: (null)' to the server [21:40] <J_Darnley> llogan: You suggested "-qscale is ignored with libx264. Use -crf instead. Default of -crf 23 will be used." on trac. [21:40] <J_Darnley> I am using "-qscale is ignored with libx264, use -crf instead. Default of -crf 23 will be used.\n" [21:40] <J_Darnley> Do you or anyone else want to make sueggestions here? [21:41] <J_Darnley> *suggestions [21:46] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:8a5ea5b9efca: avcodec/snowenc: simplify spatial_decomposition_count seting code [21:46] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:296aa4ac0168: avcodec/snowdec: return more specific error codes [21:46] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:2fb388fc9d0f: avcodec/snowdec: forward error codes [21:46] <cone-58> ffmpeg.git 03Michael Niedermayer 07master:1b4f2fb165a3: avcodec/snowdec: also print qbias and mvscale in debug output [21:54] <J_Darnley> Wait a minute... Why isn't ffmpeg printing any colours in the log [21:57] <J_Darnley> WTF! Who made cygwin be treated like a native Windows application? [22:21] <J_Darnley> If anyone would like to see my current proposal: https://gitorious.org/ffmpeg/jdarnley-ffmpeg/commit/049ed4fe09189b67405bd14… [22:21] <J_Darnley> 1 - It is a warning [22:22] <J_Darnley> 2 - It is just shoved in at the top on the init function [22:25] <michaelni> assuming crf and qp is set then somethig else than 23 would be used [22:26] <J_Darnley> Ah, true [22:26] <J_Darnley> Do you have a suggestion for that case? [22:27] <J_Darnley> Reading the user set crf value and printing that? [22:36] <J_Darnley> michaelni: How about this "-qscale is ignored with libx264, using -crf %f instead." after we set the libx264 crf parameter? [22:37] <michaelni> LGTM [22:37] <michaelni> if it works ... [22:39] <J_Darnley> I believe so. If only -qscale is given 23 (x264's default) is printed. If both -qscale and -crf are given the crf value is printed. [22:47] <J_Darnley> Okay done. Do you want me to send a patch to the mailing list? [22:51] <michaelni> yes thats an option [22:55] <J_Darnley> Well, would you prefer that rather than getting it from gitorious? [23:35] <llogan> J_Darnley: sorry, i was AFK. the latest sentence looks good to me. thanks. [23:36] <llogan> what happens if a user tries -qscale and -qp? [23:55] <J_Darnley> llogan: with my patch ffmpeg will print the message "-qscale is ignored with libx264, using -crf %0.0f instead." and then continue encoding using crf [23:55] <J_Darnley> my patch only makes it print a warning [00:00] --- Sat Mar 22 2014

1 0

ffmpeg.log.20140321
by burek 22 Mar '14

22 Mar '14

[00:19] <voip> hello guys [00:21] <voip> I am using ffmpeg -f flv -i "$INPUT" -acodec copy -vcodec copy -vbsf h264_mp4toannexb -f mpegts "$OUTPUT" [00:21] <voip> Now i need convert stream to 720x480 and frame rate 29.97. [00:21] <voip> How can i do that ? [00:24] <synth_> wouldn't you just need to add "-r 30000/1001 -s 720x480" after "mpegts" ? [00:26] <voip> <synth_> what mean -r 30000/1001 ? [00:27] <voip> i got it , frame rate [00:33] <caral> Is there a good comparison website with benchmarking different cpus with ffmpeg? [00:45] <voip> <synth_> i made cahnges: ffmpeg -f flv -i "$INPUT" -acodec copy -vcodec copy -vbsf h264_mp4toannexb -f mpegts -r 30000/1001 -s 720x480 "$OUTPUT" & [00:45] <voip> but in output stream i dndnt see any changes , there is same format [01:00] <llogan> voip: you can't change frame rate and stream copy at the same time [01:00] <voip> yah, you right ! [01:01] <synth_> sorry, my bad [01:01] <voip> can you suggest right parametrs, i just need change frame size and frame rate for flv stream [01:05] <voip> guys, its my first day in irc , shuld i install pastebin plugin in my irc client ? [01:06] <llogan> no. just got to pastebin.com or some other similar site [01:06] <voip> then, i shuld put link to pastebin site ? [01:07] <llogan> synth_: yes [01:07] <voip> ok [01:07] <llogan> voip: yes [01:07] <llogan> synth_: can you provide a sample file? [01:09] <synth_> llogan are you referring to my colorspace issue? [01:10] <llogan> yes [01:10] <synth_> sure, let me render one out real quick. [01:10] <llogan> i may not have been clear: one of the exr images you're using as an input [01:12] <synth_> yes, i know, i'm rendering out an exr file that i use as the input [01:13] <synth_> sorry, confusing. i work in vfx, so we generate exr files with our software [01:13] <llogan> what is the software? [01:13] <synth_> Nuke 8.0v3 [01:14] <synth_> http://www.thefoundry.co.uk/products/nuke-product-family/ [01:15] <synth_> http://mellow.me/tmp/frames.0001.exr here's a single frame with a color wheel. i was using this for color comparisons [01:15] <voip> ok, i have http://pastebin.com/n4h7QT9Q . Its just coping stream. Now i need to change frame size and frame rate for output stream. [01:16] <llogan> synth_: ffmpeg recognizes size, etc without issue for this single image [01:17] <llogan> voip: re-read the message from fflogger [01:17] <synth_> yeah i understand. my issue earlier was due to a non cropped exr. there was extra data in overscan [01:18] <synth_> identify showed 2000_0080_comp_master_v0010.0001.exr EXR 1942x1104 1942x1104+0+0 16-bit DirectClass 8.7MB 0.000u 0:00.000; however, mediainfo showed 1920x1080 [01:18] <synth_> so even defining the -s size of 1920x1080, ffmpeg was throwing errors [01:19] <llogan> mediainfo is not considered to be completely trustworthy here [01:19] <synth_> i tried specifying 1942x1104 as well though and it threw the same error [01:19] <voip> <llogan> my old script works perfect, i just need new script [01:19] <synth_> technically we shouldn't be rendering out these frames with extra data in overscan anyway; we'll figure out a way to auto crop these before they are written out [01:20] <llogan> voip: we don't support scripts here. only FFmpeg cli tools and libraries. [01:20] <voip> i mean cli [01:20] <llogan> synth_: is the extra data in overscan a common thing for exr? (I know little about this format) [01:21] <llogan> voip: i want to see an actual ffmpeg command you've tried and the complete console output [01:21] <voip> ok [01:22] <synth_> llogan it's optional. final composite frames shouldn't have them though and we would be using final comps for ffmpeg to convert. as for 3d assets to deliver to the compositors, adding overscan to a cgi "element" allows the artist the ability to apply lens distortions to the cgi that match the film lens without worry of not having enough "element" data if for some reason it's squeezing the image [01:22] <synth_> inward [01:23] <synth_> so 3d element exrs can sometimes have a larger bounding box than the actual data window [01:24] <llogan> if you think this is something that ffmpeg should support you can file a feature request on the bug tracker. [01:24] <synth_> so data can exist in overscan. it would just be displayed using the data window however. using Nuke, we can access the overscan data [01:25] <synth_> also, do you know if ffmpeg handles multi channel exrs? sometimes we condense everything into a single file so that it has more layers other than rgba [01:26] <llogan> i don't know. Paul B Mahol would probably be the developer to contact if you want to sponsor this as a feature. [01:27] <synth_> typically you would only want to generate media using rgba, however, it would be nice if we could choose a different layer in case someone wants to visualize the specular, diffuse, zdepth, etc.. [01:27] <synth_> ah okay [01:28] <voip> I have http://pastebin.com/ZLc8PpSW. Command made stream copy. How i can change output for 720x480 and frame rate for 29.97 ? [01:29] <llogan> voip: remove -vcodec copy. add -vcodec libx264. add -r 30000/1001 (or -r ntsc). add -r 720x480 [01:29] <llogan> that last one should be -s 720x480 [01:30] <voip> -s or -r ? [01:30] <llogan> synth_: i think there is a layer option [01:30] <voip> ok, thank you ! [01:31] <llogan> synth_: http://git.videolan.org/?p=ffmpeg.git;a=commit;h=3d20260157cb65582cb547befe… [01:31] <llogan> undocumented [01:32] <synth_> oh awesome [01:38] <voip> llogan i made suggested changes, but recieve error: http://pastebin.com/vNSxJ6RY [01:38] <llogan> synth_: as for your color issue you may want to see if chroma subsampling is occurring [01:39] <llogan> voip: you're using an ancient build [01:39] <llogan> use a newer ffmpeg [01:40] <Moussekateer> Hi guys. I'm getting this issue remuxing an mkv to an mp4 and I'm not sure why http://pastie.org/private/1qyto20ed9kdgos7hhrgra [01:41] <Moussekateer> I can copy over the audio stream fine but the video stream is the one crashing [01:42] <voip> I know. Its not my server, and they owner used apt install for ffmpeg. I am not shure can i change ffmpeg. You think there is no way for using this version ? [01:42] <llogan> no, it is unsupported here. [01:43] <llogan> just download, extract, and run a static build [01:43] <llogan> no need to install [01:43] <llogan> see the links i gave you [01:43] <voip> ok, thanks llogan! [01:45] <synth_> llogan, chroma subsampling within nuke? or using ffmpeg? [01:48] <llogan> synth_: ffmpeg [01:50] <voip> can i use Static FFmpeg builds on 32bit linux ? [01:50] <llogan> yes [01:52] <voip> one more question :) is it ok if i make apt-get remove ffmpeg for remove old version? [01:53] <synth_> how would one do that with ffmpeg? [01:55] <llogan> voip: ask your admin. [01:55] <synth_> llogan http://mellow.me/tmp/mjpg_h264_linear_difference.mov here's an example of what i'm seeing between h264 and mjpeg [01:55] <voip> :) [01:55] <llogan> synth_: view the console output [01:55] <llogan> voip: but you won't need to remove the old ffmpeg. you can use the static build independently [01:56] <voip> nice! [02:00] <voip> llogan, intresting but i cannot execute binary file [02:01] <voip> ffmpeg [02:02] <voip> may be problem 32bit ubuntu ? [02:02] <llogan> cd /to/path/containing/binary && ./ffmpeg [02:03] <llogan> or /full/path/to/ffmpeg -i input ... [02:03] <llogan> or move it somewhere in your PATH [02:03] <synth_> llogan http://pastebin.com/XaJFWcUw i'm not seeing anything in there about chroma subsampling [02:04] <llogan> synth_: your input is rgb. the output is YUV planar 4:2:0. your mjpeg is probably yuvj444p. [02:05] <llogan> that's the cause of the "degraded" colors [02:05] <synth_> interesting [02:05] <llogan> probably [02:05] <synth_> http://i.imgur.com/zLaQmZZ.png a better look at it in a linear colorspace [02:05] <synth_> so is there a way to use yuvj444p with h264? [02:06] <llogan> those nodes remind me of shake [02:06] <llogan> see ffmpeg -h encoder=libx264 [02:06] <synth_> shake was what a lot of us compositors learned in school or used professionally :) [02:06] <voip> llogan i tryng from that directroy.. http://pastebin.com/Q40djyNt [02:07] <llogan> you are using a 32-bit system but downloaded a 64-bit binary. [02:07] <llogan> i thought you asked if those links contain 32-bit builds. gusari does. [02:08] <voip> beaccuse i aseced , my linux 32 [02:08] <llogan> you will need a 32-bit build. [02:08] <voip> ok [02:09] <llogan> but your kernel migh tbe too old for the gusari ones. [02:09] <voip> yes kernel 2.6 [02:09] <llogan> then you'll have to compile [02:09] <llogan> or use a less ancient server [02:09] <voip> :) [02:10] <llogan> https://trac.ffmpeg.org/wiki/UbuntuCompilationGuide [02:12] <voip> acctualy its CentOS release 6.5 (Final) [02:12] <voip> and comes with 2,6 kernel [02:12] <llogan> synth_: you're forcing yuv420p with -pix_fmt. by default, with libx264 at least, ffmpeg will attempt to minimize chroma subsampling. however the output may not be playable by non-FFmpeg based stuff. why do you want H.264? you're not using it as an intermediate file are you? [02:13] <llogan> voip: https://trac.ffmpeg.org/wiki/CentosCompilationGuide [02:14] <llogan> it is a "non-destructive" guide, so nothing is installed to system [02:14] <synth_> llogan http://pastebin.com/x3TYs4qA [02:14] <llogan> other than dependencies [02:14] <llogan> remove -pix_fmt -pix_fmt yuvj444p [02:14] <synth_> ah crap [02:14] <synth_> i didnt' see that [02:15] <synth_> the reason we're generating these is to show clients previews or to view them from our internal intranet page [02:15] <synth_> kind of as a proxy [02:15] <synth_> but our software will pull up the raw exrs for review in our theatre [02:15] <llogan> then you'll need to use yuv420p, or they'll need to use a non-retarded player, or you'll need to use another encoder. [02:15] <synth_> sigh, okay [02:16] <llogan> but the capable player will convert to yuv420p on playback anyway, IIRC. i can't remember. [02:16] <synth_> color is huge in this industry so if anything is slightly off they wig out. that's why i was trying to go for the most accurate color with lower file size [02:16] <llogan> i should know this stuff but i have a shitty memory [02:16] <synth_> mjpeg should be fine in this case then if h264 doesn't work out [02:17] <synth_> and i guess mjpeg would be better for all platforms too [08:30] <amigojapan> hi relaxed [08:30] <amigojapan> I used the following command to concatenate a bunch of videos I made with my software: ffmpeg -f concat -i concat.txt -c copy output.mkv Here is an example of the contents of the concat.txt file http://pastebin.com/1ie6wFEb , here is the output video http://www.youtube.com/watch?v=ICU32z3RmiY , as you can see, by the end of the video, the lips are out of sync. and as you can see here,... [08:30] <amigojapan> ...(the last one) of the original videos , the lips are synced correctly http://youtu.be/xVJtAVvgQBQ [08:32] <relaxed> amigojapan: concat with mkvmerge: mkvmerge -o combined.mkv 1.mkv +2.mkv +3.mkv +4.mkv [08:33] <amigojapan> ok, let me try that relaxed ,. thanks [08:34] <relaxed> mkvmerge is the better tool when it comes to matroska [08:35] <amigojapan> relaxed: ok, I will try it out, I hope it it multiplatform.... [08:35] <relaxed> it is [08:35] <amigojapan> good, thanks [08:36] <DeadSix27> amigojapan-> http://www.bunkus.org/videotools/mkvtoolnix/downloads.html | sudo apt-get install mkvtoolnix mkvtoolnix-gui [08:36] <amigojapan> thanks DeadSix27 [08:55] <amigojapan> ok guys, that worked! thanks a lot! [09:16] <anshul_> I was trying to encode video using h264 encoder, i am getting all I frame, I have set AVCodec -> gop = 15, Isnt it suppose to come 1 I frame per 15 frame [09:35] <Mavrik> anshul_, are you setting frame type before passing it to encoder by any chance? [09:48] <anshul_> if you mean frame->key_frame parameter its always 1 of passed frame in avcodec_encode_video2 [10:02] <steinchen> hiho... wont you setup a paypal account for donations (e.g. donation(a)ffmpeg.org) ? i really think more ppl will donate than - at least i would [10:15] <Mavrik> anshul_, no, there's a "type" field on an AVFrame structure you're passing into encode [10:15] <Mavrik> and libx264 will listen to it if it's set to anything that's not AVFRAME_TYPE_NONE [10:15] <Mavrik> so if you set type to AVFRAME_TYPE_I you'll always get I-frame out even if gop size is different [10:26] <anshul_> Mavrik, i didn't got anything like AVFRAME_TYPE_I but if you mean AV_PICTURE_TYPE_I = 1; when I pass frame to av_encode_video2 there frame->type is always = 1 [10:27] <Mavrik> that's your problem then. [10:27] <anshul_> I did not set that explicitly but this frame is what i get from buffer filters [10:31] <anshul_> I checked same thing with ffmpeg -i test.ts -vcodec libx264 out.ts ; over here at avcodec_encode_video2 frame->type is always passed as 1; and in out.ts there are p frames present [10:34] <Mavrik> anshul_, http://www.ffmpeg.org/doxygen/2.0/libx264_8c_source.html#l00152 [10:34] <Mavrik> do I have to spell it out for you? [10:34] <Mavrik> the property name is pict_type [10:35] <Mavrik> checking out ffmpeg source and doc will really save you alot of time when solving such issues ;) [10:35] <anshul_> thanks [10:35] <Mavrik> (type it seems is deprecated property from older versions) [10:36] <anshul_> I think problem arise from sample_aspect_ratio in my code it is set to 0,1 while ffmpeg set it to 1,1 [11:20] <anshul_> Mavrik, Thanks you saved my day. [12:46] <steinchen> hey, how can i add a watermark/picture to the encoding? like "-vhook '/usr/lib/vhook/watermark.so -f /usr/lib/vhook/Logo.jpg'" in earlier versions [12:47] <jnvsor> steinchen: Use an overlay filter http://www.ffmpeg.org/ffmpeg-filters.html#overlay-1 [12:48] <steinchen> ty [12:56] <DrSlony> Hi, which version of ffmpeg is the first to support libvpx-vp9? [12:59] <steinchen> jnvsor, can i use a jpg as watermark? [12:59] <ubitux> yes [13:00] <jnvsor> steinchen: You'll probably want to use a png though. Higher quality + alpha values [13:14] <steinchen> ok.. ty ill try that [14:02] <DrSlony> How do I change the framerate from 29.970 to 15 so that playing time remains the same? [14:04] <DrSlony> -r15 before the output file name? [14:09] <DrSlony> What is the recommended intermediate format when capturing a 1920x1080 screencast? i used libx264 -preset ultrafast when my final file was to be h.264 too, but now the final file will be vp8, so what is the recommended intermediate? [14:09] <JEEB> well you want to have your intermediate to be lossless and fast to encode, so lossless H.264 still sounds like the best thing :P [14:16] <DrSlony> JEEB ok, mplayer says that h264 file is YV12, does VP8 not prefer i420 instead? [14:16] <JEEB> same thing called differently :P [14:17] <DrSlony> ok [14:26] <steinchen> soo.. i tried "-i logo.png -filter_complex 'overlay=10:main_h-overlay_h-10'" ... but there is no logo in the encoded videofile [14:26] <steinchen> i also use -deinterlace and encode in mp4 - if that matters [14:28] <steinchen> oh seem deinterlace was the prob [14:28] <jnvsor> steinchen: you need to specify the inputs and outputs [14:28] <steinchen> ye sure [14:28] <jnvsor> steinchen: overlay takes 2 inputs (The overlay and the video) and an output stream [14:28] <steinchen> thats what i did [14:29] <steinchen> now without -deinterlace the logo is in the dest file [14:29] <jnvsor> It's not in the filtergraph you pasted :/ [14:29] <steinchen> that was just a part of it [14:30] <steinchen> ./ffmpeg -y -i $1 -i Logo.png -filter_complex 'overlay=10:main_h-overlay_h-10' -acodec aac -ar 44100 -ab 96k -r 25 vcodec libx264 -b 1800k -f mp4 -threads 8 -strict -2 -s 1920x1080 $1-noqt.mp4 [14:30] <steinchen> thats the whole command [14:32] <jnvsor> I'd be surprised if that worked, last I checked filter_complex was very whiny about having the right filtergraph [14:32] <jnvsor> Something like: http://pastebin.com/N0nLRZd9 [14:34] <steinchen> mmh but it worked fine.. i took it from here: http://ffmpeg.org/ffmpeg-filters.html#Examples-13 [14:36] <jnvsor> Cool, it usually complains to me if I leave something out [14:51] <steinchen> cant i use -deinterlace with the filter_complex? there is no logo then [14:55] <wodim> hello, i've used ffmpeg with no parameters to cut a part of a video but the output looks bad; i thought it would use the same quality settings as the input [14:55] <wodim> but it doesn't [14:55] <wodim> can you help? [14:55] <wodim> i don't want to force a quality setting, i want to use the same exact settings so it looks more or less like the same [14:57] <steinchen> '-c copy' should work [14:57] <wodim> let's see [14:57] <wodim> (thanks) [14:58] <wodim> now it looks great but there is no sound. :-) [15:03] <jnvsor> wodim: pastebin your command [15:03] <wodim> jnvsor: i've decided to use -q instead, but thanks [15:10] <steinchen> what happens with the logo when i use deinterlace.. cant i combine that in one command? [15:13] <jnvsor> steinchen: give me a paste of your command? [15:13] <steinchen> just add -deinterlace to the above one [15:13] <steinchen> ./ffmpeg -y -i $1 -i Logo.png -filter_complex 'overlay=10:main_h-overlay_h-10' -acodec aac -ar 44100 -ab 96k -r 25 vcodec libx264 -b 1800k -f mp4 -threads 8 -strict -2 -s 1920x1080 -deinterlace $1-noqt.mp4 [15:14] <steinchen> ./ffmpeg -y -i $1 -i Logo.png -filter_complex 'overlay=10:main_h-overlay_h-10' -acodec aac -ar 44100 -ab 96k -r 25 -vcodec libx264 -b 1800k -f mp4 -threads 8 -strict -2 -s 1920x1080 -deinterlace $1-noqt.mp4 [15:14] <steinchen> ah sry [15:14] <jnvsor> steinchen: you want output deinterlace right? the man page says `-deinterlace` is for inputs not outputs [15:18] <jnvsor> steinchen: try this: [15:18] <jnvsor> http://pastebin.com/dB49HRPG [15:21] <steinchen> ok thanks, sec [15:24] <steinchen> ok works thanks ... now with the logo on the right side? [15:44] <steinchen> got it .. thanks so far! [16:33] <Waraqa> Hi, How can I encode h264 file that is backward compatible with old players? [16:34] <DrSlony> old players use divx [16:34] <Waraqa> I mean old ffmpeg [16:35] <Waraqa> I used recent ffmpeg to encode it but it's not working with ffmpeg 0.8 [16:36] <Waraqa> I see only solid grey when I play the video [16:37] <DrSlony> if upgrading that badly outdated ffmpeg is impossible, try -profile:v baseline -level 3.0 [16:38] <Waraqa> DrSlony: I used to play every h264 successfully, this is the first time I get this [16:38] <Waraqa> You can see the output errors in this ticket (which is closed) https://trac.ffmpeg.org/ticket/3485 [16:38] <JEEB> oh [16:38] <JEEB> you encoded 4:2:2 H.264 [16:38] <DrSlony> and maybe -pix_fmt yuv420p [16:38] <JEEB> yeah, that's newer than that release [16:39] <Waraqa> How can I disable that? [16:39] <JEEB> DrSlony, you don't need to limit profile or level for lavc [16:39] <JEEB> what DrSlony said the last [16:39] <JEEB> -pix_fmt yuv420p [16:48] <Waraqa> Thanks [16:48] <Waraqa> Is this option related to the resolution? [16:48] <JEEB> not specifically, it sets the pixel format to YCbCr, 4:2:0 subsampling [16:49] <JEEB> but yes, 4:2:0 limits you to mod2 width/height because of the subsampled chroma [16:49] <Waraqa> If I'm encoding HD video, Do I need to change anything? [16:49] <JEEB> just make sure your output width and height can be divided by 2 :P [16:50] <JEEB> (you have one value per 2x2 area, whereas with 4:2:2 you only limited yourself to mod2 width, every line had its own samples) [16:52] <Waraqa> Thanks again for your information [16:56] <Waraqa> What makes ffmpeg chooses 4:2:2 by default? [16:56] <ronlut> Hi. I want to use ffmpeg in an android app. I compiled ffmpeg with NDK, and now I have [libavcodec, libavformat, libavfilter, libavutil, libswscale] SO files. I have the command line arguments to do what I need, but how do I transfer it to a proper C code? I see functions are mentioned in the documentation, but where can I learn how to properly use them? [16:57] <Bombo> i got a problem with the aspect of a target x264.mp4: i'm encoding a mpeg2 (tv-dvb) file, 'Video: mpeg2video (Main), yuv420p(tv), 720x576 [SAR 64:45 DAR 16:9]' with '-vf yadif,crop=in_w-(14+14):in_h-(4+4):14:4,scale=640:368' getting a 'Video: h264 (libx264) ([33][0][0][0] / 0x0021), yuv420p, 640x368 [SAR 15916:15975 DAR 5536:3195]' and when i play the mp4 (mplayer) it is 4:3 even when i add '-aspect 16:9' to the ffmpeg line [16:58] <Bombo> when i encode just the first 60s from the mpg, the resulting mp4 is 16:9. but when i encode tzhe whole file i end up with 4:3 [16:59] <Bombo> ronlut: good question, i want to learn that too (got it on my todo list ;) my plan is to study the ffmpeg main tool source [17:01] <Bombo> ronlut: it's hopefully readable, but i would like to have a really small documented example too, like for getting infos about the source first (i would try to understand ffprobe.c for that) [17:08] <ronlut> Bombo: My problem is I really don't come from this world, and have no idea how to proceed. [17:08] <ronlut> Bombo: I mean from the ffmpeg / video world. I am a programmer. [17:09] <Mavrik> ronlut, do what you want in C on your desktop first [17:09] <Mavrik> documentation is online in doxygen and you have examples in doc/examples in ffmpeg souce [17:09] <JEEB> ronlut, see the examples under doc [17:09] <Mavrik> dragging in Android NDK before you even understand what you're trying to do will lead to a catastrphe [17:12] <ronlut> Ok thanks guys, would try to achieve this in simple c before proceeding to android. [17:12] <iive> ronlut: there should be ffmpeg program too. you may also have ffplay if you have libsdl. [17:17] <Mavrik> ronlut, also I recommend static linking for what you're trying to do, smaller binaries and easier to use on Android :) [17:18] <Bombo> ronlut: if you're a programmer you should know how to do this ;) [17:18] <Bombo> ronlut: by reading source/examples [17:23] <ronlut> Mavrik: You mean having an libffmpeg.so file and executing command line params? [17:30] <Mavrik> ronlut, no, having libyourlibrary.so that exposes as simple interface as possible [17:30] <Mavrik> and with as little Java/JNI bridge calling as possible [17:37] <Bombo> just found the ffmpeg/doc/examples/*.c ... interesting :) [17:38] <Bombo> Mavrik: what do you want to do? encode videos on your phone? [17:39] <Mavrik> O.o [17:39] <Mavrik> Why are you asking me that? :P [17:40] <Bombo> oops i meant ronlut [17:40] <Bombo> ;) [17:40] <Bombo> sorry Mavrik [17:41] <Bombo> ronlut: what do you want to do? encode videos on your phone? [17:41] <Bombo> ;) [18:03] <share> my capture card only supports raw output [18:03] <share> how to encode using x264 on fly [18:10] <llogan> share: ffmpeg -i input -c:v libx264 output [18:12] <share> /dev/video0: Invalid data found when processing input [18:13] <share> ok [18:13] <share> llogan: https://pastee.org/4fw2f [18:14] <share> do you also want v4l2-ctl --all [18:14] <llogan> sure [18:14] <llogan> you forgot "-f v4l2" as an input option [18:14] <share> oops [18:15] <llogan> is it a webcam? [18:15] <share> no, capture card [18:15] <share> https://pastee.org/vyqpp [18:15] <llogan> oh, you mentioned that before and i instantly forgot [18:16] <share> check that :) [18:17] <share> I can capture in raw mode using VLC but that's like 200mb/s [18:17] <share> sorry bad calculation [18:18] <share> 20mb/s [18:18] <llogan> in vlc you capture it as a v4l2 device? [18:18] <share> I dont see v4l2 in vlc [18:19] <share> advanced options show v4l [18:19] <llogan> i don't know what type of device it is, so you'll have to look at http://ffmpeg.org/ffmpeg-devices.html [18:19] <share> capture device /dev/video0 [18:21] <share> hm [18:21] <share> it shows Video camera [18:21] <share> llogan: vlc shows :v4l2-standard=PAL :live-caching=300 [18:21] <share> yeah it uses v4l2:// [18:22] <llogan> maybe it is /dev/video1 [18:22] <share> i only have /dev/video0 [18:24] <llogan> oh, i confused myself with the v4l2-ctl output. i guess i didn't get enough sleep last night. show: ffmpeg -f vfl2 -list_formats all -i /dev/video0 [18:25] <share> you mean v4l2 [18:25] <llogan> yes [18:26] <share> https://pastee.org/qn4rm [18:29] <llogan> that's progress at least, but i don't know why you get "invalid data" [18:31] <share> this /dev/video0: Immediate exit requested ? [18:31] <share> what invalid data [18:31] <llogan> no. earlier you said "/dev/video0: Invalid data found when processing input". try ffmpeg -f v4l2 -video_size 720x576 -framerate 25 -i /dev/video0 output.mp4 [18:32] <share> ok [18:32] <share> it's working.. [18:32] <llogan> what brand/make and model is this device? [18:32] <share> ok it captured [18:32] <share> this http://linuxtv.org/wiki/index.php/Easycap#USBTV007_EasyCAP [18:33] <share> only audio isnt supported in the kernel but it does work https://github.com/simon3z/usbtv [18:34] <llogan> you should add -pix_fmt yuv420p if you want the output to work in non-FFmpeg based players when using libx264 with this type of input [18:34] <llogan> also see https://trac.ffmpeg.org/wiki/x264EncodingGuide [18:35] <share> why yuv420p [18:36] <llogan> it will attempt to minimize chroma subsampling, but most players can't handle anything else [18:37] <share> how to change the container [18:37] <llogan> output.mkv [18:37] <share> to use mkv instead, do i just change the extension [18:37] <share> k [18:40] <share> llogan: some interesting messages https://pastee.org/85s2r [18:40] <share> when recording [18:40] <share> you were right :P [18:46] <share> why did .mp4 used less space [18:46] <llogan> it has more Chinese orphan children making more calculations per second. [18:47] <llogan> -preset urchins [18:47] <share> vlc shows Lavf55.19.104 on the first capture [18:47] <llogan> i don't know. for each command you'll have to... [18:47] <share> mkv file doesnt show that [18:47] <share> so changing the extension alters it [18:48] <llogan> and are you capturing the same thing (content and duration) each time? [18:48] <share> no :P [18:48] <share> yeah the bitrate changed i guess [18:48] <share> the first capture was more static image [18:49] <llogan> they should be hardly any difference in size between mp4 and mkv if all things are equal. [18:52] <share> "FFmpeg today doesn't support any means of GPU encoding, outside of libx264." [18:52] <share> how do I know if my GPU supports that [18:53] <share> my guess is that it only supports decoding [18:53] <share> with vdpau [18:53] <share> h264 high 4.1 [18:56] <llogan> add -x264opts opencl as an output option if your build supports it, then refer to the console output [18:56] <llogan> otherwise, don't worry about it [19:37] <ronlut> Bombo: sorry. wasn't here. The only thing I need to do is overlay one video over another one. That's it. I'm really lost here with all the outdated tutorials / ways to do that spreaded all over the internet. :( [19:39] <Bombo> ronlut: like a logo for watermarking? [19:40] <voip> hello guys [19:41] <voip> i am converting stream to ntsc format http://pastebin.com/A9Y2KtNM but converted stream propertie shows 50frame rate [19:43] <llogan> according to what? [19:52] <voip> whats the problem its shuld be 25 fps [20:02] <voip> ? [20:09] <share> llogan: do I need opencl-nvidia for that? [20:17] <Venti^> WINDOW 20 [20:18] Last message repeated 1 time(s). [20:28] <voip> i am converting stream to ntsc format http://pastebin.com/A9Y2KtNM but converted stream propertie shows 50frame rate [20:38] <llogan> voip: console output shows ntsc like you requested [20:39] <voip> llogan, but i requested 29.97fps, in VLC i see 50 [20:41] Action: llogan blames VLC [20:43] <voip> llogan, are you shure in my oputput 29.97fps, and this VLC problem ? [21:25] <ronlut> Bombo: yes. like a video watermark (one video on another) [21:34] <mjsmith> Hey guys. Im using ffmpeg-20140320-git-19139d8-win64-static and when I load in a properly formed (verified in audacity) raw pcm s16le and try to save it as a wav (or anything) it says Format mp3 detected only with low score of 1, misdetection possible! Is this a bug? [21:35] <mjsmith> Here's my command. ffmpeg.exe -loglevel debug -format s16le -sample_fmt s16 -ar 44100 -ac 2 -i raw.pcm file.wav [21:35] <Zeranoe> mjsmith: Please paste the full output to pastebin [21:35] <mjsmith> will do [21:37] <mjsmith> Zeranoe, http://pastebin.com/sS2KWygF [21:38] <Zeranoe> your input file should be first if all your trying to do is convert to wav [21:39] <Zeranoe> ffmpeg.exe -i raw.pcm -loglevel debug -format s16le -sample_fmt s16 -ar 44100 -ac 2 file.wav [21:40] <mjsmith> still says format mp3 detected [21:40] <mjsmith> i'll paste new output [21:41] <mjsmith> Zeranoe, http://pastebin.com/Gd0PgPN7 [21:43] <klaxa> try: ffmpeg -f s16le -i raw.pcm instead [21:43] <klaxa> the command syntax is: ffmpeg[.exe] <input options> -i <input> <output options> output [21:44] <mjsmith> yes klaxa that worked thanks. sorry for my stupidity :) [21:45] <Zeranoe> klaxa: Why does that work? Is it really an mp3 steam? [21:46] <mjsmith> sound is distorted though. gotta fine tune the options [21:46] <mjsmith> its definitely not an mp3. it's a fwrite sound buffer [21:58] <mjsmith> Ok now for my next problem. Audacity spits out a proper wav at length 32 seconds. ffmpeg spits out one at 39 seconds. Both reported as 16bit pcm 2 channel 44100Hz [00:00] --- Sat Mar 22 2014

1 0