Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

June 2015

1 participants
60 discussions

ffmpeg-devel.log.20150609
by burek 10 Jun '15

10 Jun '15

[00:03:15 CEST] <jamrial> ubitux, BBB: win64 disass of clamp_mv in vp8.c is currently like this: http://pastebin.com/d0gZihxK, and with av_clip_int16 it becomes http://pastebin.com/FPn7cGtG [00:03:49 CEST] <jamrial> i'm thinking that sse2 packss instructions are going to be faster. that's an awful amount of branches [00:03:55 CEST] <BBB> yeah [00:03:57 CEST] <BBB> thats a lot of code [00:04:50 CEST] <jamrial> of course that's after forcing the function to not be inlined, but still [00:06:33 CEST] <ubitux> i suppose it doesn't help if you drop the -fno-tree-vectorize? (yeah i'm making an obsession about this) [00:09:57 CEST] <jamrial> doesn't look like [00:22:28 CEST] <jamrial> ubitux, BBB: same function with an av_clip_int16 implemented with packssdw looks something like this http://pastebin.com/dUxYT8bn [00:23:18 CEST] <BBB> you do two packssdws [00:23:20 CEST] <BBB> you only need one [00:23:26 CEST] <BBB> (it can do both mvs at once) [00:23:59 CEST] <BBB> and ubitux has a point maybe were obsessing too much :-p [00:24:30 CEST] <jamrial> i'm doing one packssdw per av_clip_int16, and there are four of those in clamp_mv right now [00:24:31 CEST] <jamrial> heh, maybe [00:28:42 CEST] <philipl> WOOO fixing bugs [00:33:35 CEST] <BBB> theres no reason clipmv has to use clip_int16 [00:33:47 CEST] <BBB> it can use a custom clip implementation on x86 that does something AWESOME [00:33:54 CEST] <BBB> (I mean this is all just for testing right?) [01:02:42 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:028c59c17b14: avcodec/jpeg2000dec: Fix high bit depth branch sample shift [01:02:43 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:12ba1b2b4d55: avcodec/jpeg2000dec: Check that coords match before applying ICT [01:02:44 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:074159ed70ac: avcodec/jpeg2000dec: Fix subsampled decoding [01:06:37 CEST] <cehoyos> mcihaelni: Which example do you mean? [01:07:21 CEST] <cehoyos> There are several open tickets containing samples that can be decoded with libopenjpeg but not the native decoder, link in ticket 3619 [01:07:43 CEST] <cehoyos> To the best of my knowledge, there is no j2k sample that decodes correctly with the native decoder. [01:08:07 CEST] <cehoyos> For any non-testing use, the native decoder is useless: Why should it be the default if libopenjpeg was enabled? [01:08:54 CEST] <michaelni> i meant the example provided by the reporter of the ticket [01:09:03 CEST] <cehoyos> The only reason there are only a handful of related tickets open, is that I don't open more tickets: There are already enough open. [01:09:09 CEST] <michaelni> also libopenjoeg does not decode it here [01:10:22 CEST] <cehoyos> This is ridicilous: I always request a sample and one was given, but the ticket isn't about a sample: It is about a non-working experimental decoder being the default even if a working decoder is available. [01:12:21 CEST] <cehoyos> I just tested decoding the sample from the ticket with libopenjpeg and it works fine here. [01:12:40 CEST] <michaelni> iam sure you have a more recent libopenjpeg [01:13:19 CEST] <michaelni> i can reopen the ticket if you like but i dont know which sample libopenjpeg decodes better [01:13:29 CEST] <cehoyos> Decoding is bitexact here. [01:13:32 CEST] <cehoyos> with libopenjpeg [01:14:09 CEST] <cehoyos> Decoding is not-bitexact (nona) with the native decoder. [01:14:59 CEST] <michaelni> was that ever reported ? [01:15:10 CEST] <cehoyos> It decodes every single sample better because there is no sample for which the native decoder works correctly (afaik) [01:15:42 CEST] <cehoyos> Yes, there is a ticket that states that our native decoder does not work for any sample and that it therefore should not be the default if libopenjpeg is enabled: [01:15:46 CEST] <cehoyos> Ticket 3619 [01:15:53 CEST] <cehoyos> There is also a mailing list thread. [01:16:06 CEST] <cehoyos> http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/177508 [01:16:33 CEST] <cehoyos> And sinse the patch was not accepted, I sent a new one today (because of the Debian discussion: They need such a patch and one for aac) [01:16:54 CEST] Action: michaelni doesnt get it [01:17:05 CEST] <michaelni> i fix 2 long standing bugs and get flamed [01:17:13 CEST] <cehoyos> Our decoder was and is experimental and does not work for any sample, sorry. [01:18:01 CEST] <cehoyos> And while I would have been thankful if we had a capable GSoC student for j2k, there is none and I don't think you will be able to make our decoder work with a few patches (unfortunately) [01:18:13 CEST] <cehoyos> There was not even a mentor iirc. [01:19:57 CEST] <michaelni> how can i reproduce the issue that it is not bitexact ? [01:20:55 CEST] <michaelni> keep in mind libopenjpeg does not work here so i cant compare to it [01:26:39 CEST] <cehoyos> ./ffmpeg -threads 1 -i matrixbench_mpeg2.mpg -vframes 1 -f framecrc - [01:26:51 CEST] <cehoyos> ./ffmpeg -threads 1 -i matrixbench_mpeg2.mpg -vframes 1 -vcodec libopenjpeg -format j2k test.j2k [01:27:06 CEST] <cehoyos> ./ffmpeg -vcodec jpeg2000 -i test.j2k -f framecrc - [01:27:19 CEST] <cehoyos> ./ffmpeg -vcodec libopenjpeg -i test.j2k -f framecrc - [01:30:20 CEST] <michaelni> libopenjpeg does not work for decodiing here, i dont really trust it to work for encoding bitexactly [01:31:18 CEST] <cehoyos> Afaiu, we both agree that very old libopenjpeg does not work. [01:31:25 CEST] <michaelni> yes [01:31:32 CEST] <cehoyos> I tested with a three year old version and it works fine. [01:31:54 CEST] <cehoyos> If you believe this makes a difference, I can test 1.4 [01:32:12 CEST] <cehoyos> But I honestly don't understand what difference this would make. [01:37:05 CEST] <michaelni> i dont think testing 1.4 really would help [01:37:47 CEST] <michaelni> ubunto 12.04 ships 1.3 [01:38:19 CEST] <Compn> lol ubuntu fail then [01:38:38 CEST] <michaelni> so does ubuntu 14.04 if iam not looking at the wrong box [01:38:50 CEST] <Compn> probably not even high bit jp2k supported then [01:42:48 CEST] <jamrial> so what's the issue here? the last couple commits fixed decoding of one kind of j2k samples, but there are still more that can't be decoded? [01:43:24 CEST] <cehoyos> openjpeg 1.4 works fine here: Is bitexact for the matrixbench command line you provided [01:43:41 CEST] <cehoyos> There is no sample that decodes correctly with the native j2k decoder afaik [01:46:17 CEST] <jamrial> the one from the fate test does [01:46:24 CEST] <jamrial> and apparently so does the 420 sample from ticket 2871 now [01:46:58 CEST] <jamrial> unless they are not supposed to look like this [01:47:08 CEST] <cehoyos> The samples from ticket 2871 do not show artefacts now. [01:48:21 CEST] <jamrial> so the native decoder does work with some samples then [01:48:36 CEST] <cehoyos> You are right, the fate sample is decoded correctly, I wasn't aware. [01:49:27 CEST] <cehoyos> ffmpeg -i fate-suite/lena.pnm out.j2k is not decoded bit-exact though [01:49:46 CEST] <cehoyos> Same for the first frame of matrixbench [01:50:44 CEST] <jamrial> the native decoder may use float dsp [01:52:44 CEST] <cehoyos> Ticket 2871 samples also decode bitexact now, the sample from ticket 2792 does not work. [01:53:51 CEST] <cehoyos> Same for ticket 2586 [01:55:47 CEST] <cehoyos> Samples from ticket 1211 is not bitexact yet. [01:59:24 CEST] <cehoyos> 2872 is not bitexact yet (but no more artefacts). [02:01:43 CEST] <cehoyos> Sample from ticket 2869 does not work yet. [02:25:04 CEST] <cone-068> ffmpeg 03Rodger Combs 07master:ecefce41d9f9: lavf/tls_securetransport: fix SNI support when not verifying [02:25:05 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:7c9fcdfabd4c: avcodec/jpeg2000dec: Fix some 5/3 bitexactness issues [02:51:07 CEST] <jamrial> mmh, those psnr values look worse now [02:55:47 CEST] <michaelni> yes, but its just testing our encoder against our decoder so thats not a reliable way to test [02:56:00 CEST] <michaelni> unless we would know that our encoder is correct [02:56:55 CEST] <jamrial> ah ok [04:32:53 CEST] <jamrial> michaelni: some of your freebsd fate clients have their clocks wrong [04:33:21 CEST] <jamrial> for example http://fate.ffmpeg.org/report.cgi?time=20150604005949&slot=x86_32-debian-kf… which says it was run on June 3 using a commit that's only two hours old [04:36:53 CEST] <michaelni> should be fixed [11:44:34 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:e1009665759d: avcodec/x86/h264_weight: handle weight1=128 [11:45:35 CEST] <ubitux> ok, i'm really confused by all this apple crap [11:45:52 CEST] <ubitux> can anyone enlight me on the purpose to support both vda and videotoolkit? [11:46:11 CEST] <ubitux> vda seems to be implemented and available on osx [11:46:20 CEST] <ubitux> but obviously it's not available on ios (fuck you apple) [11:47:00 CEST] <ubitux> where you are supposed to use videotoolbox [11:47:31 CEST] <ubitux> but videotoolbox might actually be available on osx (didn't check but assuming vda is a wrapper over videotoolbox as i read...) [11:47:43 CEST] <wm4> ubitux: videotoolbox used to be a private API [11:47:43 CEST] <ubitux> so what would be the reason to prefer vda over videotoolbox? [11:47:52 CEST] <ubitux> yeah, but it's been about 3 years now [11:47:54 CEST] <wm4> but I think they're moving towards making it public (or it already is) [11:48:01 CEST] <wm4> and VDA is just a wrapper around VT [11:48:10 CEST] <wm4> so the answer is, no reason to prefer VDA (anymore) [11:48:10 CEST] <ubitux> right, so far that's my understanding [11:48:15 CEST] <ubitux> OK [11:48:33 CEST] <ubitux> so it makes sense to have videotoolbox and wipe vda? [11:48:52 CEST] <wm4> yeah [11:48:53 CEST] <ubitux> or is there benefits/features of vda over vt? [11:48:54 CEST] <wm4> I'd say so [11:49:04 CEST] <ubitux> ok [11:49:10 CEST] <wm4> except that VDA is part of the public API in FFmpeg [11:49:29 CEST] <ubitux> yeah sure i mean other than our retro compat issues [11:49:33 CEST] <wm4> (so not killing it immediately might make sense) [11:49:39 CEST] <ubitux> and eventually compat with things before vt was public [11:49:43 CEST] <wm4> is there someone working on VT support? [11:49:50 CEST] <ubitux> probably me [11:50:03 CEST] <ubitux> i mailed the guy who submitted a patch 2.5 years ago [11:50:15 CEST] <wm4> (I wonder what VDA actually does...) [11:50:23 CEST] <ubitux> (who is actually the vda maintainer) [11:50:35 CEST] <wm4> we have a VDA maintainer? [11:51:04 CEST] <ubitux> [~/src/ffmpeg]- grep -i 'vda\*' MAINTAINERS [11:51:05 CEST] <ubitux> vda* Sebastien Zwickert [11:51:20 CEST] <ubitux> apparently :p [11:51:27 CEST] <ubitux> he's the one who submitted a VT patch [11:51:52 CEST] <wm4> btw. using VDA from ffmpeg is absolutely trivial now [11:51:55 CEST] <wm4> can this be kept with VT? [11:53:35 CEST] <ubitux> probably [11:55:15 CEST] <nevcairiel> is VT available on all relevant versions of OSX, ie. those were VDA is available today? [11:55:22 CEST] <nevcairiel> For me, I would care about 10.7+ [11:55:52 CEST] <ubitux> "This patch adds support of VideoToolbox to the HWAccel infrastructure. [11:55:54 CEST] <ubitux> VideoToolbox framework became public since MacOS 10.8. [11:55:56 CEST] <ubitux> " [11:56:22 CEST] <nevcairiel> if there is a distinct advantage in using VT over VDA, i guess I could sell not supporting 10.7 [11:57:13 CEST] <rcombs> VT also includes encoding [11:57:22 CEST] <rcombs> or, I'm pretty sure it does [12:01:57 CEST] <ubitux> https://github.com/eugeneware/ffmpeg-progress-stream lol [12:02:06 CEST] <ubitux> > Generate a stream of progress events from the stderr stream of ffmpeg [12:02:34 CEST] <ubitux> (note: we have a -progress option) [13:34:31 CEST] <cone-815> ffmpeg 03Rodger Combs 07master:e60b018c7416: lavf/segment: add an option to allow breaking on non-keyframes [14:53:33 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:b6ee1912f98d: avcodec/jpeg2000dec: Handle Psot = 0 [15:54:31 CEST] <kierank> lglinskih: hello [15:56:47 CEST] <av500> http://blogs.windows.com/buildingapps/2015/06/05/using-ffmpeg-in-windows-ap… [15:58:02 CEST] <lglinskih> kierank: hi! [15:58:07 CEST] <kierank> how's it going? [16:10:05 CEST] <BBB> av500: omg [16:10:12 CEST] <BBB> I see pigs flying, theyre everywhere [16:10:32 CEST] <nevcairiel> its a terrible managed c++ api though =( [16:10:43 CEST] <BBB> FLYING PIGS! [16:10:55 CEST] <BBB> who cares thats is C++, THEY FLY!!!! IN THE AIR!!! [16:11:09 CEST] <nevcairiel> i dont care that its c++, i care that its managed code [16:11:15 CEST] <nevcairiel> managed c++ is so ugly [16:11:19 CEST] <nevcairiel> use c# if you want managed :( [16:11:35 CEST] <BBB> BUT THEY FLY!!! [16:11:57 CEST] <BBB> ok fine if you care that flying pigs are managed, then I wont make a big deal about it, but .. I mean .. THEY FLY! [16:12:17 CEST] <nevcairiel> its the new microsoft, now with flying pigs, ffmpeg, and ssh support [16:14:51 CEST] <wm4> "We have released a set of Compilation Instructions for a WinRT compatible build of FFmpeg." [16:14:54 CEST] <wm4> they have? [16:14:57 CEST] <wm4> THEY? [16:15:12 CEST] <nevcairiel> they tried to push some evil long document as a patch into the repository some tiem ago [16:15:26 CEST] <nevcairiel> it was full of bad advice and generally in the wrong place, but shrug [16:15:38 CEST] <wm4> so I guess that means MS is suggesting the use of MSYS2 [16:16:11 CEST] <nevcairiel> unfortunately, yes [16:19:01 CEST] <BBB> https://trac.ffmpeg.org/wiki/CompilationGuide/WinRT [16:19:09 CEST] <BBB> that is an insanely long document [16:19:33 CEST] <nevcairiel> mostly because there is lots of duplication between the different compiling methods [16:19:39 CEST] <lglinskih> kierank: I got ill a little bit, so yesterday and today I sleep a lot.=/ I give up with makefiles for now. So I'll start with fixing my flac test and writing new tests. [16:20:33 CEST] <kierank> ok [16:20:46 CEST] <lglinskih> should I start with test for H264? [16:24:47 CEST] <kierank> yes [16:25:39 CEST] <lglinskih> ok [16:32:31 CEST] <kierank> you'll need to write an independent version of framemd5 [16:32:40 CEST] <kierank> but that's not too hard [16:45:15 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:45db92180693: avcodec/jpeg2000dec: Add placeholder for PLT parsing [16:45:16 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:44327cbc9a8c: avcodec/jpeg2000: Disable special case for JPEG2000_QSTY_SI [16:47:52 CEST] <Daemon404> BBB, they tried to put that + a stupid .bat in our repo [16:47:53 CEST] <Daemon404> i rejected it [16:47:57 CEST] <Daemon404> for obvious reasons [16:48:21 CEST] <BBB> I missed that [16:48:22 CEST] <BBB> fun though [16:48:41 CEST] <Daemon404> [15:31] <@Daemon404> im also a little miffed they made no mention of the work wbs, bbb, etc did for the windows / winrt support [16:48:44 CEST] <Daemon404> [15:31] <@Daemon404> which was massive [16:48:47 CEST] <Daemon404> [15:31] <@Daemon404> "hey look what ms did!" [16:48:49 CEST] <Daemon404> ^ related [16:48:58 CEST] <BBB> I missed that too :-p [16:49:15 CEST] <BBB> anyway, Im happy ms finally admits their bubble is dead and theyre coming out of zombieland into the real world [16:49:22 CEST] <Daemon404> indeed [16:49:24 CEST] <BtbN> "Set the following environment variables" uhm, ok oO [16:49:36 CEST] <Daemon404> their compiler uses them [16:49:42 CEST] <Daemon404> it;s standard ms [16:49:49 CEST] <Daemon404> including the massive copypasta [16:49:54 CEST] <BtbN> I'm more concerned about the incredibly long stuff that follows [16:49:55 CEST] <Daemon404> with confusing or no explanations [16:50:18 CEST] <Daemon404> thats how you breed the cargo-cult win32 developer [16:50:28 CEST] <Daemon404> this line does magic. [16:50:53 CEST] <Daemon404> also i doubt all that kind of crap is needed... there is a winrt fate instance [16:51:17 CEST] <BtbN> hm, that copy&paste is that long because it reads registry keys and builds the PATHs from that [16:51:35 CEST] <BtbN> there are some version specific ones a bit further down, they look not that bad [16:51:55 CEST] <BtbN> But why isn't there some file you can just source? [16:52:02 CEST] <BtbN> Or is the windows shell not able to source stuff? [16:52:06 CEST] <nevcairiel> there is [16:52:11 CEST] <nevcairiel> no idea why they dont use them [16:52:17 CEST] <Daemon404> yeah. [16:52:29 CEST] <Daemon404> thats how all the fate instances work [16:53:12 CEST] <BtbN> MS developers not knowing their own stuff :/ [16:53:59 CEST] <nevcairiel> they do seem to use them, at least they mention the shortcut in the start menu [16:54:12 CEST] <nevcairiel> but that should make the setup of the env vars redundant [16:54:13 CEST] <nevcairiel> oh well [16:54:18 CEST] <nevcairiel> i dont set them up manually [16:54:38 CEST] <nevcairiel> i just go like, "call "C:\Program Files (x86)\Microsoft Visual Studio 12.0\VC\vcvarsall.bat" x86_amd64" [16:54:46 CEST] <Daemon404> yeah [16:54:48 CEST] <nevcairiel> can probably not hardcode the path somehow =p [16:55:15 CEST] <nevcairiel> maybe the store libs are not included in any of the batch files? [16:55:17 CEST] <nevcairiel> idunno [16:55:50 CEST] <Daemon404> that would be pretty stupid [17:11:10 CEST] <gnafu> Speaking of pigs flying: http://www.washingtonpost.com/news/morning-mix/wp/2015/06/09/2200-piglets-s… [17:12:47 CEST] <gnafu> Thanks, Microsoft! Now 300-400 piglets are dead. [17:13:56 CEST] <Daemon404> mmmm pork [17:19:54 CEST] <cone-815> ffmpeg 03Rodger Combs 07master:4b082bc2419d: doc/muxers: document new break_non_keyframes option [17:29:14 CEST] <michaelni> Does anyone want to write release notes for 2.7 ? [17:35:03 CEST] <cone-815> ffmpeg 03Janne Grunau 07master:fb1473080223: aac_parser: add required padding for GetBitContext buffer [17:35:04 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:5f48a73bb099: Merge commit 'fb1473080223a634b8ac2cca48a632d037a0a69d' [17:42:15 CEST] <cone-815> ffmpeg 03Janne Grunau 07master:09447f2b0faf: ac3_parser: add required padding for GetBitContext buffer [17:42:16 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:840465ebf979: Merge commit '09447f2b0fafac6d9565aab82a4c5f16fc99ee5e' [17:47:47 CEST] <ubitux> so, we still do not have any doc about hwaccels? [17:48:14 CEST] <ubitux> i see the pattern <codec>_<accel>_{decoder,hwaccel} [17:48:34 CEST] <Daemon404> g 42 [18:08:39 CEST] <cone-815> ffmpeg 03Janne Grunau 07master:210921722bf8: imc: add required padding for GetBitContext buffer [18:08:40 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:551813a963b2: Merge commit '210921722bf828b3b895ebcbc34374e6c4452c6f' [18:16:48 CEST] <wao813> whois [18:17:34 CEST] <wao813> Hi, I am trying to package h264 encoded data to file, but seeing following error, anyone knows what to do? [18:17:38 CEST] <wao813> [h264 @ 0x7fc014012000] missing picture in access unit with size 73 [18:18:02 CEST] <wao813> [h264 @ 0x7fc014012000] no frame! [18:18:37 CEST] <cone-815> ffmpeg 03James Almer 07master:b380337020e2: mpjpegdec: don't try to alloc an AVIOContext when probe is guaranteed to fail [18:18:38 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:962654f25137: Merge commit 'b380337020e271c5431aa8ef8f8e9dfda5e919b2' [18:30:47 CEST] <cone-815> ffmpeg 03Luca Barbato 07master:b14086ca38ef: mkv: Correctly report the latest packet had been flushed [18:30:48 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:dac7b2780264: Merge commit 'b14086ca38efa1a86cb0f0c6aa147b05f698877b' [18:55:50 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:32b1131fc16e: avformat/oggenc: Fix return code in case of flushing [18:55:52 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:dbbb31e5bb88: avcodec/jpeg2000: remove unused variable [20:17:15 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:adbe1d7de446: avcodec/libopenjpegdec: Mark as experimental if <= 1.3 [20:17:16 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:6bf8d9e1649d: fate: Force jpeg2000 decoder for dcinema [20:44:47 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:0916938a8dfa: avcodec/jpeg2000dec: Fix gain regression with gray16 [21:10:10 CEST] <cehoyos> Anybody with a long history? wm4 reported a dca/dts sample from samples.ffmpeg.org that prints "sample requested" when decoded with FFmpeg - I forgot which sample it was... [21:10:46 CEST] <wm4> http://samples.ffmpeg.org/evob/Subtitle-sample.evo [21:10:56 CEST] <nevcairiel> evo uses dts? [21:11:18 CEST] <nevcairiel> i thoughts eac3 is the evo-level madness [21:11:30 CEST] <cehoyos> thank you [21:11:31 CEST] <wm4> well this file does [21:11:50 CEST] <wm4> the file has vc1/dts/eac3/ac3 [21:11:52 CEST] <wm4> all the good stuff [21:11:58 CEST] <nevcairiel> evo is primarily vc1 [21:12:05 CEST] <nevcairiel> thank god hd-dvd didnt make it [21:12:05 CEST] <nevcairiel> :D [21:12:49 CEST] <JEEBsv> :D [21:24:16 CEST] <jamrial> didn't hd-dvd use h264? [21:24:47 CEST] <jamrial> i remember back in the early years hd-dvd was h264 and some blu-rays were mpeg2, even [21:24:50 CEST] <nevcairiel> the same formats as BD are supported [21:24:58 CEST] <nevcairiel> vc1, h264 and mpeg2 [21:25:11 CEST] <nevcairiel> but i've seen most hd-dvds using vc1 for some reason [21:25:49 CEST] <nevcairiel> similarly, it often uses eac3, despite also supporting the typical blu-ray audio formats like ac3, truehd, dts, dts-hd [21:28:42 CEST] <nevcairiel> the terrible part about evo is that it encodes 24p video using 3:2 soft pulldown flags, and appropriate timestamps [21:28:54 CEST] <nevcairiel> its a terrible format, so much dvd legacy left in it [22:21:17 CEST] <cone-815> ffmpeg 03Vignesh Venkatasubramanian 07master:34ae98a77189: lavf/webm_chunk: Remove AVFMT_ALLOW_FLUSH [22:43:22 CEST] <jamrial> so about 130 warnings when compiling ffmpeg with gcc 5.1 are bogus (-Warray-bounds) [22:43:41 CEST] <nevcairiel> i still didnt bother getting gcc 5.1 [22:43:53 CEST] <nevcairiel> i dont trust first releases of a new line [22:44:14 CEST] <philipl> I can't believe they haven't run out of warnings to add after all this time. [22:44:18 CEST] <jamrial> they got fixed anyway. not backported to gcc5 branch yet but will before gcc 5.2 is released [22:44:36 CEST] <Daemon404> so 12 months from now [22:44:54 CEST] <nevcairiel> i wonder where gcc 4.9.3 disappeared to, it was originally scheduled for like march/aprilk [22:44:56 CEST] <nevcairiel> -k [22:44:57 CEST] <jamrial> you can always just compile a snapshot from the gcc5 branch :p [22:45:15 CEST] <jamrial> some distros do that and never bother with actual releases [22:45:28 CEST] <Daemon404> i look forward to -Wfalse-warn [22:45:39 CEST] <Daemon404> "warning: previous warning may be false" [22:50:20 CEST] <philipl> heh [22:56:41 CEST] <cone-815> ffmpeg 03Simon Thelen 07master:83307a32eb0c: libavutil/channel_layout: Correctly return layout when channel specification ends with a trailing 'c'. [23:15:56 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:6ddb109b9096: doc/APIchanges: Add 2.7 separator [23:15:57 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:a5f44bc4460b: doc/APIchanges: fill in missing stuff [23:15:58 CEST] <cone-815> ffmpeg 03Michael Niedermayer 07master:3a99f6e79fa3: Changelog: Add 2.7 entry [23:23:18 CEST] <cone-815> ffmpeg 03Andreas Cadhalpun 07master:58995f647b5f: sonic: set avctx->channels in sonic_decode_init [23:47:31 CEST] <cone-815> ffmpeg 03Andreas Cadhalpun 07release/2.7:HEAD: sonic: set avctx->channels in sonic_decode_init [00:00:00 CEST] --- Wed Jun 10 2015

1 0

ffmpeg.log.20150609
by burek 10 Jun '15

10 Jun '15

[00:00:27 CEST] <onefix> I know that there was some work on it, but last I checked, conventional wisdom was that a multi-core system was better [00:21:39 CEST] <rikai> Hm... Are none of the prores encoders available in ffmpeg performant enough for realtime encoding? [00:25:26 CEST] <pzich> probably depends on your hardware and the input [00:27:42 CEST] <rikai> mmm, trying to use prores_ks in the new version of obs that uses ffmpeg for it's backend. Getting 95% frameskipping regardless of what settings i give it, which seems... unusual. [00:27:52 CEST] <brontosaurusrex> rikai: what ffmpeg version you got? (I think i get about 5x realtime) [00:28:41 CEST] <brontosaurusrex> rikai: also, certain decoders may be botleneck [00:30:58 CEST] <haasn> I want to reduce CPU usage of music playback. One of the sources is ReplayGain - I have files tagged with RG data, which the device then has to adjust for during playback. Would it be possible to instead strip the RG data and modify the files themselves, so that my device just has to do 1:1 playback? [00:34:01 CEST] <rikai> brontosaurusrex: hm... it appears that obs is using a build from git. "N-70634-g3bedc99" so..... http://git.videolan.org/?p=ffmpeg.git;a=commit;h=70634-g3bedc99 [00:36:01 CEST] <rikai> And i'm pretty sure the issue isn't the decoder, but rather encoding. OBS-MP is telling me that 95+% of frames were skipped during decoding. Not sure what the issue could be, since cpu usage ins't terribly high and i'm recording to an ssd... I can't imagine it's a throughput issue, because then changing what profile(proxy/lt/hq) would have an effect, wouldn't it? [00:36:14 CEST] <Mavrik> prores probably isn't optimized... like at all. [00:36:33 CEST] <rikai> Thats the conclusion i was coming to, yeah. [00:36:57 CEST] <rikai> but he said he was getting 5x realtime encoding, so i was thinking i might be mistaken... [00:37:41 CEST] <rikai> Especially since someone with a somewhat more powerful system (newer i7) tested for me and got essentially the same results. [00:38:45 CEST] <rikai> Just trying to figure out if it's an issue in the software or a matter of prores not being efficent enough in ffmpeg, since if it's the former, i can bug people to look at it. ;) [00:39:28 CEST] <Mavrik> Honestly I've never used it, so I can't really tell you much, but afaik prores didn't get much attention [00:39:48 CEST] <haasn> Ah, -af volume="replaygain=album" [00:40:27 CEST] <haasn> Is there a way to set the volume of tracks which have missing RG information? [00:40:28 CEST] <rikai> brontosaurusrex: mind if i ask which of the prores formats you're using ? Snice there are 3... [00:42:05 CEST] <rikai> -af "volume=#dB" [00:42:06 CEST] <rikai> isnt it? [00:45:56 CEST] <rikai> I'm so close to being able to remove another bit of mac software from our production workflow that i can taste it... :P [00:49:40 CEST] <rjp421> rikai, ^5 [00:49:56 CEST] <rikai> rjp421: hm? [00:55:24 CEST] <haasn> Hmm. ffmpeg does not seem to want to read the replaygain information from vorbis files [00:55:37 CEST] <haasn> I can set the album gain to like -10 manually but -af volume="replaygain=album" just ignores it [01:03:59 CEST] <haasn> Seems like an ffmpeg bug; the tags are definitely present, it even prints eg. REPLAYGAIN_ALBUM_GAIN: -6.17 dB as the metadata for the files [01:04:06 CEST] <Canar> >not using R128gain [01:04:09 CEST] <Canar> =3 [01:04:36 CEST] <rikai> rjp421: well sadly it depends on being able to record realtime to prores, which OBS now has the ability to do, but it seems lacks the performance optimizations to do so reasonably. We've got the largest linux podcast out there, so we're slowly trying to move to all open tools as a point of pride and principle, but we need prores support for the moment, so it's kind of a hard block on switching anything else over for the moment. :/ [01:04:52 CEST] <haasn> Canar: that's not even present in my distro's repositories [01:04:53 CEST] <rikai> Looking into perhaps a bountysource of some sort, if we've got some funding available for it... [01:05:10 CEST] <Canar> haasn: it's built into ffmpeg now [01:05:22 CEST] <Canar> but i was mostly just being silly [01:05:25 CEST] <haasn> oh [01:05:42 CEST] <rikai> Didn't the project change names, as well? [01:05:45 CEST] <haasn> well I'm open to alternatives [01:05:58 CEST] <rikai> Yeah, its bs1770gain now, it seems. [01:06:05 CEST] <haasn> Right now I'm just going to live with not having .ogg files on my phone, this only affects like two albums [01:06:16 CEST] <Canar> hm [01:06:19 CEST] <Canar> that is news to me [01:06:24 CEST] <Canar> need to get with it [01:07:36 CEST] <Canar> 5~meh [01:07:40 CEST] <Canar> terminology [01:09:52 CEST] <Canar> seems that only really peter belkner's projects did the name change [01:10:01 CEST] <Canar> which is fine, as they're shit anyhow =P [01:10:25 CEST] <Canar> https://ffmpeg.org/ffmpeg-filters.html#ebur128 haasn rikai [01:11:19 CEST] <haasn> I'm not sure what use this is supposed to be, at least to me [01:13:11 CEST] <Canar> think of it as a successor to replaygain; handles bass and gating better [01:13:28 CEST] <Canar> ...but that doesn't fix/answer your query [01:13:59 CEST] <Canar> as a replaygain user, might be of interest [01:14:21 CEST] <haasn> Canar: One important point of consideration for me is to preserve loudness across track boundaries that are within the same album [01:14:44 CEST] <haasn> This is why I want to use ReplayGain's album tag capabilities, since I can just scan albums as a unit and apply the appropriate album tags [01:14:59 CEST] <haasn> So I'm not looking for a solution that will adjust each file in isolation and break track borders [01:17:52 CEST] <Canar> haasn: I get that; you can hack around the tag-loading issue using the appropritate command-line-fu. [01:17:57 CEST] <Canar> appropriate* [01:18:19 CEST] <haasn> Canar: I could do it manually for these few albums I guess [01:18:28 CEST] <haasn> But I'd rather the bug be fixed [01:18:49 CEST] <Canar> Album mode just treats the album as one long track, FYI. [01:19:44 CEST] <Canar> i wonder if there's something weird about the tag [01:19:57 CEST] <Canar> not like Vorbis comments are complicated, tho [01:29:34 CEST] <haasn> Canar: Yes, but I'm invoking ffmpeg on a per-file basis [01:29:43 CEST] <haasn> so it can't just analyze the whole album [01:58:45 CEST] <DrBotato> I'm wondering if it is possible to do the Ken Burns technique with FFMPEG filters on a video. For example. given a vertical video with aspect ratio 9:16 @ 1080x1920, i wish to create a viewframe of the video that is 16:9 @ 608x1080, starting from the top of the video, and slowly pan down across the height of the video to the bottom over the duration of the video [01:59:27 CEST] <DrBotato> sorry that should have said: "16:9 @ 1080x608" [02:01:14 CEST] <DrBotato> oh, looks like zoompan might be able to do what i want. nifty. [02:02:30 CEST] <DrBotato> damm people and holding their phone vertically while filming. lol. [02:05:41 CEST] <rikai> Right, so... Here's a conondrum. I wan't to open a bounty on bountysource to improve prores encoding performance within ffmpeg to the point where it's usable for realtime encoding with software such as OBS... However, bountysource requries there to be an issue on ffmpeg's tracker to post a bounty on... It's not really a bug per-se though... So i'm not really sure what to do... Could someone the works on the project advise me as to how i might proceed? [02:07:12 CEST] <pzich> I don't know for sure, but you might be able to file as a "feature request" instead of a bug [02:14:32 CEST] <rikai> Hm... i wonder what a "sponsoring request" is... Probably not what i'm looking for, but i noticed it in the type field... [02:43:12 CEST] <tpw_rules> hey. i have a whole crapton of baseline DV video that i'm hoping to re-encode into h265. what CRF would preserve it? the wiki says 18 is lossless but i don't imagine DV has the best quality [02:44:44 CEST] <tpw_rules> what other options might give me the best quality? [04:07:53 CEST] <rehunted> is possible to disable 'disk write' every second? and make it only when the file is finished ? (or in segmented steps?) [04:30:21 CEST] <kyleogrg> hi [04:31:23 CEST] <kyleogrg> I did IVTC on a video (fieldmatch,yadif,decimate) and now it's gone from the duratio of 2:40:00 to 2:09:00 [04:31:57 CEST] <kyleogrg> vlc still displays the longer (original) duration, but the video is actually shorter [04:32:05 CEST] <kyleogrg> why is this? [04:38:08 CEST] <DrBotato> can anyone point me in the direction to understanding what this does: format=pix_fmts=yuva420p [08:42:33 CEST] <chama> hi, I need to add and an image to the end of a video for 2 second duration using a single command... Is there anyway to do this using ffmpeg filter_complex? [10:31:25 CEST] <Valeriu> hello everyone [10:31:56 CEST] <Valeriu> i'm trying to extract and rencode a portion from a ts file using the following args: [10:32:03 CEST] <Valeriu> cmdArgs = @"-i ""{0}"" {1} -ss {2} -c:v libx264 -profile:v high -b:v {3}K -vf yadif=0:-1 -y -t {4} ""{5}"""; [10:32:42 CEST] <Valeriu> it works very well, but sometimes i get an extra audio portion at the begining of the file [10:32:49 CEST] <Valeriu> which shouldnt be there [10:33:17 CEST] <Valeriu> do you have any idea if there is some parameter i can specify in order to repair the audio desync? [10:37:49 CEST] <brontosaurusrex> Valeriu: try moving -ss after input, upgrade your ffmpeg [10:44:11 CEST] <Valeriu> unfortunately, i have the latest ffmpeg version (windows build) and the -ss is after the input [10:44:46 CEST] <brontosaurusrex> what version is that? [10:52:28 CEST] <elmargol> What is the default x264 profile ffmpeg is using? [10:54:28 CEST] <JEEBsv> s/ffmpeg/libx264/ [10:54:31 CEST] <JEEBsv> and that depends on your settings [10:54:45 CEST] <JEEBsv> it starts with high or high10 depending on if you're using a higher bit depth libx264 or not [10:54:52 CEST] <JEEBsv> and then selects the lowest your parameters let it do [10:55:13 CEST] <chama> hi, I need to add and an image to the end of a video for 2 second duration using a single command... Is there anyway to do this using ffmpeg filter_complex? [10:55:23 CEST] <JEEBsv> (or some other profile in case you're not encoding 4:2:0) [10:55:28 CEST] <JEEBsv> but the idea is the same [10:56:13 CEST] <chama> simply i need to add a outro... [10:56:26 CEST] <chama> what is the best way? [10:56:58 CEST] <elmargol> JEEBsv, -c:v libx264 -crf 18 -vf "format=yuv420p" out.mkv [10:57:38 CEST] <elmargol> Just wondering If this is ok for archiving my videos [10:58:06 CEST] <elmargol> There is some lenscorrection and deshaking aswell [10:59:07 CEST] <JEEBsv> elmargol: it depends on your requirements. basically first decide on the CRF by actually doing the checking yourself with your eyes instead of copying a value from somewhere, and then check which is the slowest preset you can take [10:59:37 CEST] <Valeriu> can i find the x264 profile ffmpeg is using by executing a "ffmpeg -info, -help" command? [11:00:17 CEST] <Valeriu> i get ffmpeg static build from http://ffmpeg.zeranoe.com/builds/ [11:01:35 CEST] <elmargol> how about -crf 18 -maxrate 10m [11:12:27 CEST] <Hugo_> Hey! With which parameter can I change the output FPS? :) [11:13:07 CEST] <elmargol> -r 25 [11:16:17 CEST] <Hugo_> Thank you! [11:17:58 CEST] <framerate345097> It doesn't work for me, all the frames are dropping [11:21:35 CEST] <relaxed> framerate345097: what doesn't work? [11:21:52 CEST] <framerate345097> Changing the framerate to 5 [11:22:19 CEST] <framerate345097> I want 5 fps for faster rendering because I only have to render a image in a video [11:22:31 CEST] <relaxed> ffmpeg -r 5 -i input ... [11:25:33 CEST] <framerate345097> Thank you very much! [11:34:50 CEST] <zhanshan> hey with command ffmpeg -f image2 -i 'III-kongste-miao_%07d.jpg' I still get an error [11:34:52 CEST] <zhanshan> '[image2 @ 0x2f9bc60] Could find no file with path 'III-kongste-miao_%07d.jpg' and index in the range 0-4 [11:35:01 CEST] <zhanshan> what's wrong? [11:35:35 CEST] <zhanshan> the files are named: III-kongste-miao_0088547.jpg ... [11:35:44 CEST] <relaxed> zhanshan: give me an example file name [11:36:45 CEST] <zhanshan> starts at III-kongste-miao_0080500.jpg and ends at III-kongste-miao_0092571.jpg [11:37:41 CEST] <zhanshan_aw> sorry gotta go, be back at in two hours [11:37:59 CEST] <relaxed> zhanshan_aw: look at "ffmpeg -h full | less +/image2" [11:38:39 CEST] <relaxed> or, ffmpeg -h demuxer=image2 [11:56:33 CEST] <t4nk587> Hello All, [11:56:43 CEST] <t4nk587> I am a beginner using FFMPEG [11:56:50 CEST] <t4nk587> I have pasted my problem over here: [11:56:51 CEST] <t4nk587> http://pastebin.com/embed_js.php?i=94Z1NQZC [11:57:21 CEST] <t4nk587> If anyone would kindly have a look [11:57:32 CEST] <t4nk587> and tell me what I am missing here. [11:58:38 CEST] <c_14> The concat protocol doesn't work with mp4s [11:58:45 CEST] <c_14> https://trac.ffmpeg.org/wiki/Concatenate [11:59:24 CEST] <t4nk587> I even tried it with creating a file of videos and it did not work in that case as well [12:00:50 CEST] <c_14> Probably mp4 being bitchy. try remuxing to mpegts [12:01:50 CEST] <t4nk587> Thanks, can you reference how can I do that. [12:01:58 CEST] <t4nk587> Though I searching over ther internet now. :) [12:02:05 CEST] <c_14> ffmpeg -i video -c copy -map 0 -f mpegts out.mpegts [12:02:38 CEST] <t4nk587> Thanks I am going to try it now. [12:17:10 CEST] <t4nk587> I tried to do the following and it merged the two video timestamp and played the first video completly. However, only some green dots appear on the screen for the second video. [12:17:16 CEST] <t4nk587> I tried the following commands: [12:17:38 CEST] <t4nk587> ffmpeg -i input1.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts intermediate1.ts ffmpeg -i input2.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts intermediate2.ts ffmpeg -i "concat:intermediate1.ts|intermediate2.ts" -c copy -bsf:a aac_adtstoasc output.mp4 [12:20:33 CEST] <t4nk587> Following is the output link: [12:20:33 CEST] <t4nk587> http://pastebin.com/znsbBfYb [12:21:51 CEST] <c_14> It might be that one of the videos is baseline and the other isn't, or that the tbcs differ [12:23:13 CEST] <t4nk587> Any thoughts about making both videos baselined? [12:23:41 CEST] <t4nk587> How can check it? [12:23:57 CEST] <c_14> ffmpeg -i video -c:v libx264 -profile:v baseline [12:25:28 CEST] <t4nk587> http://pastebin.com/4q2Gi8yM [12:26:11 CEST] <t4nk587> I tried executing and it show the following. Just trying to see the difference. [12:26:15 CEST] <c_14> You have to specify an output file. And you probably want to keep -c:a copy. And you should only have to do it for the 2.mp4 [12:27:06 CEST] <t4nk587> Thanks for assisting me. I am going to do that now. [12:36:20 CEST] <t4nk587> http://pastebin.com/qHCUL83x [12:36:38 CEST] <t4nk587> It might be that I have not done it right. But it did not work :( [12:40:23 CEST] <c_14> Hmm, why is it using constrained baseline now... Have you tried reencoding during concat with the concat demuxer? [12:41:51 CEST] <t4nk587> I tried the following sequence: [12:41:53 CEST] <t4nk587> ffmpeg -i 2.mp4 -c:a copy -profile:v baseline 2b.mp4 [12:41:58 CEST] <t4nk587> converting the video to baseline [12:42:13 CEST] <t4nk587> ffmpeg -i "concat:intermediate1.ts|intermediate2.ts" -c copy -bsf:a aac_adtstoasc output.mp4 [12:42:38 CEST] <c_14> ye, I saw that. What I meant was if you could try it with the concat demuxer aund without -c copy (at least for the video). See if that works. [12:42:45 CEST] <t4nk587> and ofcourse creating concat:intermediate1 and concat:intermediate2 before the final concat [12:43:47 CEST] <t4nk587> :) and how would I do that? [12:44:07 CEST] <t4nk587> I am sorry, I am just learning FFMPEG. [12:44:15 CEST] <t4nk587> So quite a beginner :( [12:44:50 CEST] <relaxed> t4nk587: (MP4Box is part of gpac) MP4Box -cat 1.mp4 -cat 2.mp4 -new combined.mp4 [12:45:05 CEST] <c_14> you can try that too [12:47:44 CEST] <t4nk587> Thanks, I am going to try it now. [12:49:06 CEST] <t4nk587> No, concat demuxer did not work with the baselined video [12:50:53 CEST] <t4nk587> http://pastebin.com/iduS3SCp [12:52:37 CEST] <t4nk587> Maybe, I am approaching a problem in the wrong direction. I wanted to add a title slide to a video. First a created a video from the image and then tried to merge it. Maybe if anyone has better suggestion. [13:08:43 CEST] <relaxed> t4nk587: you'll need to create it with the same settings as the audio/video streams in the original video [13:31:05 CEST] <ruby_on_tails> can someone explain why "ffmpeg -i grab.avi -vcodec libx264 -vpre medium -crf 22 -threads 0 grab1.mp4" gives me a file which is approx 3.5mb per minute but "avconv -i grab.avi -vcodec libx264 -preset medium -crf 22 -threads 0 grab1.mp4" gives me a file which is 1mb per minute? [13:31:13 CEST] <ruby_on_tails> grab.avi is the same video [13:31:19 CEST] <ruby_on_tails> and the quality of the videos are almost same [13:31:28 CEST] <ruby_on_tails> 3.5mb vs 1mb is a huge difference [13:40:56 CEST] <relaxed> ruby_on_tails: the "-vpre medium" syntax means that ffmpeg version is really old [13:41:59 CEST] <ruby_on_tails> relaxed: i just installed it 2 weeks ago using instructions from a website on my rmbp [13:42:21 CEST] <ruby_on_tails> well wait [13:42:24 CEST] <ruby_on_tails> let me check version [13:42:31 CEST] <ruby_on_tails> maybe that command worked on a different machine [13:42:51 CEST] <ruby_on_tails> relaxed: i am on ffmpeg 2.6.3 [13:44:40 CEST] <ruby_on_tails> relaxed: i just checked, i have to use -preset instead of -vpre on this new installation [13:44:50 CEST] <ruby_on_tails> but still the file size is something like that only 3.5mb vs 1mb [13:45:48 CEST] <[-T-]> hey [13:45:56 CEST] <[-T-]> anyone using qsv around? [13:46:30 CEST] <relaxed> ruby_on_tails: if you want a specific size use two pass, or limit the it with -maxrate [13:47:30 CEST] <ruby_on_tails> relaxed: i dont want a specific size but the quality is almost same [13:47:37 CEST] <ruby_on_tails> i want to understand why the huge difference? [13:53:08 CEST] <ruby_on_tails> sorry got disconnected, did i miss anything? [13:53:33 CEST] <zhanshan> relaxed do I have to choose a pattern type, is that the problem? [13:55:40 CEST] <c_14> ruby_on_tails: you didn't miss anything. Are you sure they're both linking against the same version of libx264, is there audio in the file? What is your complete console output for both commands? [13:56:22 CEST] <harisk92> hey guys is it possible to use relative path in output ? [13:56:26 CEST] <c_14> zhanshan: you need -start_number (I think that's what it was called, it's in the docs) [13:56:31 CEST] <c_14> harisk92: sure [13:56:34 CEST] <c_14> Why shouldn't it be. [13:56:50 CEST] <harisk92> well iam getting error [13:56:56 CEST] <harisk92> i have folder [13:56:58 CEST] <harisk92> videos [13:57:01 CEST] <harisk92> and inside it [13:57:06 CEST] <harisk92> i have 3 subfolders [13:57:20 CEST] <ruby_on_tails> c_14: yes there is audio [13:57:28 CEST] <ruby_on_tails> c_14: running the commands now to compare the outputs [13:57:31 CEST] <c_14> ruby_on_tails: try encoding without audio. Might be part of it. [13:58:50 CEST] <harisk92> http://pastebin.com/d2KTi2AL [13:58:57 CEST] <harisk92> this is my ffmpeg command [13:59:16 CEST] <relaxed> oh my [13:59:52 CEST] <c_14> What's the error you're getting? Complete console output in a pastebin please. [14:00:59 CEST] <harisk92> http://pastebin.com/DbEmnMzP [14:01:13 CEST] <zhanshan> c_14 that was it, thanks to both of you! [14:01:58 CEST] <harisk92> c_14 [14:02:09 CEST] <harisk92> I must specify file exstension ? [14:02:17 CEST] <harisk92> cause that's the problem [14:02:20 CEST] <c_14> probably [14:02:30 CEST] <c_14> Otherwise the segment muxer won't know what format to use [14:02:32 CEST] <relaxed> you don't if you add -f $format [14:02:42 CEST] <c_14> Though you could use -segment_format [14:03:08 CEST] <harisk92> relaxed: if I add -f $format will that append .ts to end of my file ? [14:03:29 CEST] <c_14> harisk92: in your case you already have -f ssegment [14:03:30 CEST] <relaxed> I think c_14 was right about -segment_format [14:03:57 CEST] <harisk92> yeah [14:04:27 CEST] <harisk92> iam trying to get list of chunks from directory via Lua [14:04:42 CEST] <harisk92> and create playlist on the fly [14:05:00 CEST] <harisk92> so I tought it would be better without .ts exstension [14:05:36 CEST] <harisk92> I just need to convert segment name to number and get all segments between two timestamps [14:05:46 CEST] <harisk92> then create playlist.m3u8 [14:06:17 CEST] <harisk92> hey c_14 [14:06:20 CEST] <BtbN> You know ffmpeg can output to hls? [14:06:23 CEST] <harisk92> is it possible for ffmpeg [14:06:46 CEST] <harisk92> BtbN: yeah I know [14:06:53 CEST] <harisk92> BtbN: but this what I do [14:07:33 CEST] <harisk92> BtbN: will allow user to watch content from before couple of days [14:07:42 CEST] <harisk92> BtbN: and between time intervals [14:08:21 CEST] <harisk92> BtbN: for example you want to watch some tv show that was aired 3 days ago and that was played between 1.00 AM and 1.30 [14:08:56 CEST] <harisk92> c_14: ffmpeg can't create paths on it's own ? [14:09:05 CEST] <c_14> no [14:09:46 CEST] <harisk92> ok ty [14:09:48 CEST] <harisk92> for your help [14:12:18 CEST] <harisk92> Iam pretty new when it comes to ffmpeg [14:12:31 CEST] <harisk92> so I apologize for my ignorance [14:12:36 CEST] <harisk92> and stupid questions [14:16:21 CEST] <harisk92> shouldn't this testsrc show timestamp not somekind of counter that goes from 0 to xxx ? [14:53:00 CEST] <bushart> Hello, everyone. I am trying implement ping-pong-loop effect to my video. But i can't make a smooth-gluing to video fragments. What I can be read on this subject? [15:11:16 CEST] <termos> is there a way of adding a delay of X seconds when streaming a video with ffmpeg? [16:22:45 CEST] <harisk92> hey guys is this filter ok "drawtext=expansion=strftime:fontfile=/usr/share/fonts/dejavu/DejaVuSerif.ttf:text='%Y-%m-%d %H:%M:%S':fontsize=20:fontcolor=red:x=20:y=20" [16:23:17 CEST] <harisk92> [Parsed_drawtext_3 @ 0x341b800] Both text and text file provided. Please provide only one [16:24:02 CEST] <harisk92> Error initializing filter 'drawtext' with args 'expansion=strftime:fontfile=/usr/share/fonts/dejavu/DejaVuSerif.ttf:text=%%Y-%%m-%%d %%H:%%M:%%S:fontsize=20:fontcolor=red:x=20:y=20' [16:25:33 CEST] <k_sze> Does the perceptual compression of libx264 do anything funky about gamma? [17:54:50 CEST] <JonG> I am having an issue with my ffmpeg command...: ffmpeg -i input1.mp4 -i input2.mp4 -filter_complex ... [bla1][bla2]overlay=main_w-overlay_w:main_h-overlay_h:format=yuv420[outv] -c:v libx264 -preset ultrafast -b:v 45000k -aspect 1.7777778 -map [outv] -map [a] -c:a libfdk_aac -ac 2 -b:a 128k -t 24 output.mp4 ... I receive an error "Unrecogn [17:54:50 CEST] <JonG> ized option 'map [outv]'.". I am using: "ffmpeg version git-2015-02-25-b0d3322 Copyright (c) 2000-2015 the FFmpeg developers" with "configuration: --arch=arm --cpu=cortex-a8 --target-os=linux --enable-runtime-cpudetect --enable-pic --disable-shared --enable-static --cross-prefix=/home/user/android-ndk-r10d/toolchains/arm-linux-androideabi-4.8/prebu [17:54:50 CEST] <JonG> ilt/linux-x86_64/bin/arm-linux-androideabi- --enable-cross-compile --sysroot=/home/user/android-ndk-r10d/platforms/android-9/arch-arm/ --extra-cflags='-I../x264 -mfloat-abi=softfp -mfpu=neon -fPIE -pie' --extra-ldflags='-L../x264 -fPIE -pie' --enable-version3 --enable-gpl --disable-doc --enable-yasm --disable-decoders --enable-decoder='mpeg4,aac,h2 [17:54:50 CEST] <JonG> 64,mpegvideo,mpeg1video,mpeg2video,png,mjpeg' --disable-encoders --enable-encoder='mpeg4,aac,h264,libx264,mpeg1video,mpeg2video,png' --disable-parsers --enable-parser='aac,mpeg4video,ac3,h261,h264,vc1,mpegvideo' --disable-demuxers --enable-demuxer='aac,h264,mpegvideo,m4v,mov,vc1,mp4,concat,image2' --disable-muxers --enable-muxer='h264,mpeg1video,mp [17:54:50 CEST] <JonG> eg2video,m4v,mov,vc1,md5,mp4' --enable-protocols --enable-indev='v4l,v4l2' --disable-filters --enable-filter='aresample,scale,movie,overlay' --enable-avfilter --disable-indevs --enable-indev=lavfi --disable-outdevs --enable-hwaccels --enable-ffmpeg --disable-ffplay --disable-ffprobe --disable-ffserver --disable-symver --enable-network --enable-libx [17:54:51 CEST] <JonG> 264 --enable-zlib". Am I missing a configuration option to include map? Where would I find this information of what options are required for the map command? Or is my ffmpeg just too old? [18:02:29 CEST] <onefix> JonG: What is the issue? [18:02:57 CEST] <JonG> I receive an error "Unrecognized option 'map [outv]'." [18:03:19 CEST] <JonG> only on this custom build. The same command seems to work fine on my ubuntu laptop. [18:03:50 CEST] <onefix> JonG: Are you sure you're using ffmpeg and not avconv? [18:05:01 CEST] <JonG> onefix: definitely [18:05:06 CEST] <onefix> ok [18:12:21 CEST] <JonG> onefix: My version is "ffmpeg version git-2015-02-25-b0d3322 Copyright (c) 2000-2015 the FFmpeg developers" ... I assume that is ffmpeg, not avconv! [18:12:48 CEST] <hefest> trying to create video from one image: ffmpeg -r 1/30 -i avatar.png -c:v libx264 -r 25 -pix_fmt yuv420p -y out.mp4 [18:19:14 CEST] <JonG> hefest: Does this work for you: ffmpeg -loop 1 -i avatar.png -c:v libx264 -t 10 -y out.mp4 ? http://stackoverflow.com/questions/5887311/ffmpeg-1-image-1-audio-file-1-vi… [20:27:41 CEST] <Lencl> hello [21:32:42 CEST] <Parsec300> Hi people, I can't seem to find how to offset subtitles from an srt file by 4 seconds and have them burnt into the movie. I'm using latest stable compiled version on a Mac from the terminal. I've tried to use -itsoffset, but it doesn't work [21:33:21 CEST] <Parsec300> ffmpeg -i Input_file.mkv -itsoffset 4 -i subtitles=Subs.srt -f avi -c:v libxvid -b:v 600k -g 300 -vf scale=640:-1 -bf 2 -c:a libmp3lame -b:a 128k output_file_ffmpeg.avi [21:34:21 CEST] <Parsec300> ffmpeg -i Input_file.mkv -itsoffset 4 -vf subtitles=James.srt -f avi -c:v libxvid -b:v 600k -g 300 -vf scale=640:-1 -bf 2 -c:a libmp3lame -b:a 128k output_file_ffmpeg.avi [21:34:49 CEST] <Parsec300> Both of those commands don't get the desired result and the documentation isn't getting me anywhere, nor searches on the internet [21:35:02 CEST] <Parsec300> Does somebody know? [22:06:08 CEST] <DRAWNBOX> Hello? [22:06:24 CEST] <DRAWNBOX> So, I tried to ffmpeg and do some conversion [22:06:44 CEST] <DRAWNBOX> and what I ended up with was not right [22:06:56 CEST] <Parsec300> I'm not very familiar with ffmpeg, but I will try to help [22:07:31 CEST] <DRAWNBOX> http://pastebin.com/sZ0iaE4v [22:07:36 CEST] <DRAWNBOX> That was my command [22:07:54 CEST] <llogan> you forgot to include the complete console output [22:08:00 CEST] <llogan> and a description of the issue [22:08:03 CEST] <DRAWNBOX> oh, sorry about that [22:08:57 CEST] <DRAWNBOX> http://pastebin.com/LX2JhbJx [22:09:46 CEST] <llogan> remove profile and pix_fmt [22:10:34 CEST] <DRAWNBOX> Also for 60 FPS what am I doing wrong? [22:10:53 CEST] <llogan> nothing that i notice [22:11:01 CEST] <DRAWNBOX> The output said 12 [22:11:10 CEST] <DRAWNBOX> console* [22:11:25 CEST] <llogan> that was how fast it was encoding at that time [22:11:30 CEST] <DRAWNBOX> oh. ok [22:12:03 CEST] <llogan> if you want to view the output in non-FFmpeg based players then add -pix_fmt yuv420p. [22:12:26 CEST] <llogan> if you are presenting the video via progressive download then add "-movflags +faststart" [22:12:36 CEST] <llogan> (both as output options) [22:13:09 CEST] <llogan> and you can probably get away with -crf 18, or maybe 17 [22:13:16 CEST] <DRAWNBOX> The video that it makes isn't right [22:13:36 CEST] <DRAWNBOX> well, it looks like it took the average color of the frame or something [22:14:04 CEST] <DRAWNBOX> nvmd [22:14:09 CEST] <DRAWNBOX> just the player I used [22:16:53 CEST] <DRAWNBOX> Thanks [22:17:28 CEST] <llogan> if you don't mind the wait you can use a slower -preset than the default [22:18:35 CEST] <DRAWNBOX> Nahhh [22:20:59 CEST] <DRAWNBOX> Are there any video formats that do transparency? [22:21:00 CEST] <DRAWNBOX> Or will I need to do some green screen type thing? [22:21:16 CEST] <JEEBsv> usually stuff that is used in the middle in production or so [22:21:38 CEST] <JEEBsv> nothing you'd use for final distribution [22:24:53 CEST] <Parsec300> Hi llogan and JEEBsv, any ideas for my problem? [22:24:56 CEST] <Parsec300> http://pastebin.com/fg2p8Ruv [22:25:14 CEST] <llogan> you forgot the complete console outputs [22:25:24 CEST] <llogan> as per tradition, apparently [22:26:12 CEST] <Parsec300> Well, there isn't an error associated with it, other than the subtitles not being present in the movie or not in sync with the audio [22:26:41 CEST] <llogan> how do we know if you're not using a version from 2004? [22:28:56 CEST] <Parsec300> http://pastebin.com/SMRs9XJ5 [22:31:59 CEST] <Parsec300> The other one does create an error: http://pastebin.com/mjmyvmAi [22:32:02 CEST] <llogan> i don't think subrip can go into avi, and that's why the subtitle stream wasn't mapped by default, but I'm not much of a subtitles user. [22:33:35 CEST] <Parsec300> But without the --itsoffset in the second paste, it does work, but then subtitles are about 4 seconds early [22:35:06 CEST] <llogan> i don't know. maybe the timing is off in the srt file [22:36:05 CEST] <llogan> and itsoffset is not an output option [22:46:19 CEST] <Parsec300> Perhaps I could adjust the srt [22:50:11 CEST] <Parsec300> llogan, thanks for trying [22:58:27 CEST] <DRAWNBOX> Another question [22:58:44 CEST] <DRAWNBOX> From what I understand avi has alpha channels [22:58:58 CEST] <BtbN> avi has nothing, avi is a container. [22:59:05 CEST] <DRAWNBOX> oh [22:59:29 CEST] <DRAWNBOX> another question [22:59:57 CEST] <DRAWNBOX> Is there a pixel format I am supposed to use for alpha channel? [23:00:36 CEST] <BtbN> video codecs usualy don't support alpha channels [23:00:39 CEST] <BtbN> h264 definitely doesn't [23:01:24 CEST] <BtbN> For WebM/VP8 there seems to be some support, but no idea how common it is [23:01:56 CEST] <DRAWNBOX> Well, Im gonna be throwing this in a video editor [23:02:00 CEST] <BtbN> There aparently is support for it in ffmpeg though [23:02:54 CEST] <BtbN> utvideo is propably best then, no quality loss, supports an alpha channel [23:04:12 CEST] <DRAWNBOX> So would my command be ffmpeg -framerate 60 -i Intro%04d.png -c:v utvideo -profile:v main -crf 15 -pix_fmt yuv444p Intro.mp4 [23:04:27 CEST] <BtbN> yuva or something like that [23:05:01 CEST] <BtbN> And check if your editor even supports it [23:06:19 CEST] <DRAWNBOX> Whats the extension? [23:06:40 CEST] <DRAWNBOX> Its not very many frames I need converted [23:06:57 CEST] <BtbN> no idea which container supports utvideo. [23:07:09 CEST] <DRAWNBOX> im just gonna look it up [23:07:10 CEST] <BtbN> ffmpeg will propably tell you if the one you use doesn't [23:09:42 CEST] <llogan> just do "-c:v utvideo output.avi" and get rid of the other option opions. [23:09:50 CEST] <llogan> *output options [23:11:12 CEST] <DRAWNBOX> no support for utvideo [23:11:16 CEST] <llogan> if you're in windows, install utvideo so yuor editor can use it. http://umezawa.dyndns.info/archive/utvideo/?C=M;O=D [23:11:35 CEST] <voltagex> Hi - why can't I convert dvb_teletext/subtitles to srt? http://sprunge.us/EXIO [23:20:17 CEST] <llogan> voltagex: your build probably needs libzvbi for teletext support [23:20:38 CEST] <voltagex> blah. I still haven't been able to build on Windows [23:20:41 CEST] <voltagex> thanks though [23:21:42 CEST] <llogan> you can request zeranoe to do add it [23:22:06 CEST] <voltagex> was requested in 2013 [23:45:41 CEST] <rehunted> pkzxa7bd [23:53:25 CEST] <anoop_r> hi [23:53:36 CEST] <anoop_r> how to fork and rename ffmpeg ? [00:00:00 CEST] --- Wed Jun 10 2015

1 0

ffmpeg-devel.log.20150608
by burek 09 Jun '15

09 Jun '15

[00:36:00 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:7856afef524d: avformat/hdsenc: Change duration from single to to double precision [00:55:26 CEST] <cone-151> ffmpeg 03Luca Barbato 07master:9b56ac74b170: mpjpeg: Initial implementation [00:55:27 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:8985e7c56130: Merge commit '9b56ac74b170d12027fbc81f581a451a709f1105' [01:19:13 CEST] <cone-151> ffmpeg 03Luca Barbato 07master:252d6200c36e: avio: Add avio_put_str16be [01:19:14 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:ce838ad950fc: Merge commit '252d6200c36e7eaa79f8d5205b7d731179e94897' [01:30:47 CEST] <cone-151> ffmpeg 03nu774 07master:677c804aa3a7: aac: correctly map 7.1ch-wide AAC from FDK AAC encoder [01:30:48 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:8eb2c411c1e5: Merge commit '677c804aa3a78d61b21e6423165a252846c20f0e' [01:43:01 CEST] <cone-151> ffmpeg 03Sebastian Dröge 07master:a188108ebf28: aac: Support channel configurations 11 and 12 [01:43:02 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:476692abdb32: Merge commit 'a188108ebf28ebac9d2b8fc7d5b391aef45698b3' [02:42:19 CEST] <cone-151> ffmpeg 03Luca Barbato 07master:bc76c4694327: aac: Wait to know the channels before allocating frame [02:42:20 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:153d23ee39db: Merge commit 'bc76c46943272515805d7ac48ca39f14826d1fed' [02:42:22 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:990605768c86: avcodec/aacdec: Do not return a uninitialized value [03:54:55 CEST] <cone-151> ffmpeg 03Caligula useraccount 07master:51ac1f616f58: avformat: Add single jpeg muxer [03:54:56 CEST] <cone-151> ffmpeg 03Caligula useraccount 07master:3b89a673155f: ffserver: Use singlejpeg muxer for jpeg [03:54:57 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:5cf84e595d7c: doc/ffserver: Add entry for missing jpeg variant [03:59:05 CEST] <lglinskih> michaelni: hi! Not as good as I expected. I'm trying to locate my test in "tests" directory. And also I want to build it there) [04:01:31 CEST] <lglinskih> the magic of makefiles is not easy for me [06:57:21 CEST] <thardin> hey, does ffmpeg have an anti-sourceforgization thing in its license? [06:57:25 CEST] <thardin> because it should [07:30:01 CEST] <rcombs> it's LGPL [07:30:13 CEST] <rcombs> with optional GPL components and no OpenSSL exception [07:35:48 CEST] <thardin> myes, I know [07:42:11 CEST] <ubitux> michaelni__: thx for avio* [12:06:05 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:ac2dad9690ad: avformat/version: Bump version for single jpeg muxer [12:06:06 CEST] <cone-068> ffmpeg 03Max Poliakovski 07master:d765e07322a1: atrac3plus: add support for GHA phase inversion. [12:06:07 CEST] <cone-068> ffmpeg 03Max Poliakovski 07master:4b343f7c359b: atrac3plus: give the phase_shift flag a better name. [13:17:05 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:7630cce4b309: avformat/mxfenc: Allow overriding /manual setting of the signal standard [13:43:44 CEST] <lglinskih_> michaelni: I'm sorry! it was 5 a.m. in my city and I didn't expect you to answer to me. So I went to sleep) [13:44:56 CEST] <michaelni> sure, np [13:45:19 CEST] <michaelni> could you resolve the problem with the test ? [13:47:59 CEST] <lglinskih_> About makefiles: my test (api-flac-test) is already integrated in "make fate", but the command for running test is placed in libavcodec.mak in /tests/fate [13:50:22 CEST] <michaelni> yes, i dont know what kierank prefers (he is your mentor so better ask him than me) but i would just leave it in libavcodec.mak if that works [13:56:36 CEST] <lglinskih_> My first wish was to locate all source files of API tests in one place. Kieran didn't argue with me, but now I think that I can't do it by myself.=/ [13:57:29 CEST] <kierank> lglinskih_: hmm [13:58:42 CEST] <kierank> if you're getting stuck then perhaps you should just leave things as they are and continue with libavcodec.mak [13:59:26 CEST] Action: kierank gets lunch [14:20:23 CEST] <Daemon404> man that ticket... [14:21:55 CEST] <cehoyos> I tested this extensively last time and I was unable to see any (even a small) difference in time. [14:22:18 CEST] <cehoyos> Now why -f aac should be slower than -f mp4 is not easy to understand... [14:31:09 CEST] <Compn> i'm curious why his curl and wget times are so different [14:31:16 CEST] <Compn> is curl using multiple connections or something ? [14:31:52 CEST] <Daemon404> curl does not, no [14:32:00 CEST] <Daemon404> his pipe is probably not stable. [14:32:44 CEST] <Compn> heh [14:33:27 CEST] <cehoyos> I cannot reproduce a different time between wget and curl (curl is slower here if at all) [14:33:47 CEST] <Compn> he could have a broken wget or something [14:34:03 CEST] <cehoyos> But I can reproduce the "-vn" issue;-( [14:35:09 CEST] <Compn> i dont doubt there is a bug there somewhere, thats why i asked for more info (and bench) :) [14:35:30 CEST] <cehoyos> I don't understand: If you saw a bug why didn't you explain how to reproduce it? [14:35:35 CEST] <cehoyos> I wasn't able to reproduce it... [14:38:38 CEST] <cehoyos> It's a regression;-( [14:39:35 CEST] <cone-068> ffmpeg 03Rodger Combs 07master:6dd5371e34c6: lavf/tls: let the user specify what name to verify against [15:39:49 CEST] <cone-068> ffmpeg 03Donny Yang 07master:ed09bb3782e2: avcodec/apng: Dispose previous frame properly [16:03:26 CEST] <TimNich> Is the mailserver working OK at the moment? [16:04:10 CEST] <Daemon404> was it that one of your mails never went through? [16:06:01 CEST] <cone-068> ffmpeg 03Donny Yang 07master:0ab1c46fe07f: avcodec/apng: Add blending support for non-alpha pixel formats [16:08:36 CEST] <TimNich> its that I havent seen it, but michaelni obviously has.. [16:10:06 CEST] <Daemon404> spam collection maybe [16:10:14 CEST] <Daemon404> i often see a pile of TimNich mails in my spam folder [16:10:17 CEST] <Daemon404> due to yahoo mail address [16:11:28 CEST] <TimNich> Ah, I see an Excess bounce in the spam folder& I fixed the DMARC issues ages ago... [16:13:37 CEST] <TimNich> Nothing from me via ffmpeg-devel should come from me. [16:14:01 CEST] <Daemon404> thats true [16:14:07 CEST] <Daemon404> we, wait no its not [16:14:12 CEST] <Daemon404> man my brain is broken [16:15:41 CEST] <TimNich> Hmmm. I also patched up the new server ready for the change over, so maybe it got unpatched& [16:24:29 CEST] <TimNich> no, the adress munging is still happening from avserver2 [16:25:48 CEST] <cone-068> ffmpeg 03Donny Yang 07master:33292c07fe19: avcodec/apng: Add support for blending with GRAY8A pixel format [16:37:20 CEST] <cone-068> ffmpeg 03Donny Yang 07master:130a6c04a466: avcodec/apng: Add partial support for blending with PAL8 pixel format [16:37:41 CEST] <TimNich> OK all good now , so I guess a hiccup at the change over. [19:01:01 CEST] <Luigi12_work> hello :o) I would like to do one more (hopefully last) ffmpeg security update for Mageia 4. Would it be possible to roll one more 2.0.x ffmpeg release with the current CVE fixes? [19:11:14 CEST] <iive> Luigi12_work: michaelni is the person responsible for most of that. he should answer (and this would highlight him :) [19:11:44 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:653bf3c5a150: avcodec/hq_hqa: Fix signness of tag [19:13:04 CEST] <llogan> i wonder how many other downstreams use 2.0 (no others are listed on wiki downstreams page) [19:13:31 CEST] <michaelni> Luigi12_work, ill see what i can do [19:14:43 CEST] <llogan> Luigi12_work: do you have an EOL date yet for Mageia 5? [19:16:48 CEST] <Luigi12_work> llogan: not yet, since it hasn't been released yet. It should be this week though, so I will update the wiki page. It'll be release date + 18 months. [19:17:11 CEST] <Luigi12_work> michaelni: thank you! [19:17:30 CEST] <llogan> Luigi12_work: ok, thanks. [19:17:36 CEST] <Luigi12_work> glad my downstreams wiki page has been useful :o) [19:51:07 CEST] <cehoyos> michaelni: Your commit also fixed ticket 4510 for me. [19:53:06 CEST] <cehoyos> Am I the only one who gets "libswresample/swresample.c:696:5: warning: "ASSERT_LEVEL" is not defined [-Wundef]" ? [19:54:15 CEST] <michaelni> cehoyos, cant reproduce 4510 either, feel free to close [19:55:26 CEST] <cehoyos> You cannot reproduce on x86-64 or on x86-32? [19:55:43 CEST] <cehoyos> Because I just tested again, and it always crashed here... [20:10:55 CEST] <jamrial> cehoyos: no, i also get that warning [20:14:17 CEST] <michaelni> cehoyos, i tried the ticket with 32 ad 64bit after the patch and couldnt repro, i didnot try before the commit [20:24:30 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:56f0fe6b8434: swr: Fix ASSERT_LEVEL warning [20:24:59 CEST] <michaelni> cehoyos, fixed the warning [20:49:33 CEST] <cehoyos> michaelni: Thank you! [20:49:56 CEST] <cehoyos> The crash with x86-32 was only reproducible before your patch. [20:50:52 CEST] <cone-068> ffmpeg 03Luca Barbato 07master:a6f19d6a9f8d: configure: Support MSVC 2015 [20:50:53 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:a4557b7a98f7: Merge commit 'a6f19d6a9f8d1e08653d9d77581e8c823f4955c2' [21:04:33 CEST] <jamrial> the mpjpeg probe function made valgrind and asan go nuts, lol [21:04:49 CEST] <cone-068> ffmpeg 03James Almer 07master:34d278f9838e: mpjpegdec: fix memory leak in probe function [21:04:54 CEST] <jamrial> there [21:06:19 CEST] <wm4> jamrial: this must have been very evil [21:06:40 CEST] <jamrial> http://fate.ffmpeg.org/report.cgi?time=20150608135429&slot=x86_64-archlinux… [21:06:56 CEST] <jamrial> about 2000 tests failed because of that :p [21:07:12 CEST] <nevcairiel> another quality libav implementation [21:07:16 CEST] <nevcairiel> wait, i shouldn't troll [21:09:28 CEST] <jamrial> huh, they fixed it already [21:09:42 CEST] <jamrial> now it looks like i stole attribution for it, ouch [21:10:01 CEST] <wm4> http://git.libav.org/?p=libav.git;a=blobdiff;f=libavformat/mpjpegdec.c;h=72… [21:10:21 CEST] <wm4> well [21:10:27 CEST] <wm4> you didn't remove the blank line [21:10:36 CEST] <wm4> copyright violation averted! [21:10:48 CEST] <wm4> (copyright is stupid) [21:11:04 CEST] <nevcairiel> i would consider that change trivial and therefor not protected under copyright =p [21:11:14 CEST] <jamrial> michaelni: when you merge Janne Grunau's fix for the mpjpeg leak don't reference my commit, or at least don't do it making it look like mine was first [21:11:25 CEST] <jamrial> because it wasn't. i was seven hours late :p [21:12:48 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:da0c8664b4dc: mpegvideo: Move various temporary buffers to a separate context [21:12:49 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:db8ae37a783a: Merge commit 'da0c8664b4dc906696803685f7e53ade68594ab8' [21:16:05 CEST] <rcombs> http://blogs.windows.com/buildingapps/2015/06/05/using-ffmpeg-in-windows-ap… huh [21:16:30 CEST] <nevcairiel> we saw that yesterday already [21:16:34 CEST] <nevcairiel> its kinda lol [21:16:42 CEST] <nevcairiel> also their managed API is all kinds of terrible [21:16:45 CEST] <rcombs> and, uh, https://github.com/Microsoft/FFmpegInterop [21:16:54 CEST] <nevcairiel> (also, i hate managed c++, its just terrible, use C# if you want managed) [21:18:06 CEST] <Compn> did it get posted to twitter ? [21:19:41 CEST] <rcombs> a coworker just linked it to me [21:21:15 CEST] <nevcairiel> dunno where it showed up first, but it was linked here yesterday i think [21:23:09 CEST] <wm4> rcombs: I already trolled on the issue tracker [21:23:53 CEST] <nevcairiel> your troll was stupid though, some developer of a media interop isnt going to run to the kernel devs and demand they implement pthread already [21:24:47 CEST] <nevcairiel> (not to mention that ffmpeg works perfectly fine without pthreads) [21:24:53 CEST] <wm4> too bad if my troll isn't good enough for them [21:25:20 CEST] <cehoyos> It may even be faster... [21:26:12 CEST] <nevcairiel> I'm not sure your troll even makes sense, even pthreads in ffmpeg needs the silly lockmanager [21:27:15 CEST] <nevcairiel> (and the default implementation of the lockmgr in ffmpeg works fine with w32threads, so all they probably need to do is ensure w32threads actually works for them, and then request threading) [21:27:16 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:f8716a1408f4: mpegvideo: Rework frame_size_alloc function [21:27:17 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:e05fda99f7a4: Merge commit 'f8716a1408f4f4ec63857b7015fbd62f9eac344a' [21:27:49 CEST] <wm4> nevcairiel: native pthreads could avoid it, because static mutex initializers work [21:27:59 CEST] <wm4> nevcairiel: just ffmpeg's pthread "emulation" can't [21:28:41 CEST] <nevcairiel> if you claim to only support one locking mechanism, you could of course just get rid of the lockmgr and and hardcode the mutex calls, but that doesn't really change much at all [21:29:51 CEST] <wm4> I think there's a strong point that if win supported pthread, we'd just make it mandatory [21:39:19 CEST] <jamrial> or you could ask them to implement a way to statically initialize critical sections (aka, mutex) [21:44:37 CEST] <cone-068> ffmpeg 03Andreas Cadhalpun 07master:b18eac7ff223: vp9: change type of tile_size from unsigned to int64_t [22:04:42 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:9bb11be0e5a7: mpegvideo: Split picture allocation for encoding and decoding [22:04:43 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:3244a176505b: Merge commit '9bb11be0e5a75782c3139ad058c2b571499aa37d' [22:04:53 CEST] <ubitux> FYI, apparently LOTRO is using ffmpeg, avcodec 53 (0.10?) with the following flags: http://pastie.org/pastes/10230189/text [22:05:44 CEST] <ubitux> also this in the strings: "e:/src/ffmpeg-head/ffmpeg/source/patched-ffmpeg/libavcodec/options.c" [22:06:10 CEST] <JEEBsv> yeah, a lot of stuff uses lavf/lavc [22:08:35 CEST] <nevcairiel> BtbN: do you know if nvenc has a particular max bitrate it cannot/would not exceed? [22:13:52 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:925b80d64029: mpegvideo: Move OutFormat enum to mpegutils.h [22:13:53 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:83e3516e64dc: Merge commit '925b80d64029d41962e5998d7d901226c3a9baea' [22:25:20 CEST] <philipl> nevcairiel: we can run it with increasing bitrates and see when it stops paying attention [22:25:43 CEST] <nevcairiel> well its hard to judge if the content is just perfectly coded at that point [22:25:52 CEST] <nevcairiel> need high frequency noise pattern! [22:35:45 CEST] <cone-068> ffmpeg 03Janne Grunau 07master:caf7be30b112: mpjpgdec: free AVIOContext leak on early probe fail [22:35:46 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:402b18afcc51: Merge commit 'caf7be30b11288c498fae67be4741bfbf083d977' [22:36:52 CEST] <jamrial> michaelni: thanks [22:37:07 CEST] <michaelni> np [22:40:11 CEST] <baptiste> that's kinda cool though, microsoft/ffmpeg who would have thought ? [22:40:20 CEST] <baptiste> can they finish the vc1 decoder ? :) [22:40:51 CEST] <nevcairiel> microsoft has changed quite a bit over the last two-ish years, they are quite different now [22:41:23 CEST] <philipl> nevcairiel: Did you read the SDK docs? I don't remember it being discussed, but maybe I missed it. [22:42:07 CEST] <baptiste> yeah, it's changing [22:42:30 CEST] <jamrial> or maybe they could make wmalossless 24bit work [22:45:47 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:8ef98855d25e: sctp: Always initialize outmsg struct [22:45:49 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:f7e932473314: audiointerleave: Always initialize new_pkt [22:45:49 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:01a6ae1396b5: Merge commit '8ef98855d25e457094468e2e1a79d9b10d6445b2' [22:45:50 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:d1f7b313ac34: Merge commit 'f7e932473314e6ca4c851d49cbde8570b6e66383' [22:56:01 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:bc1eace1b365: jack: Check memory allocation [22:56:02 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:77510a969861: Merge commit 'bc1eace1b3654c490cb2c226b3c80854244dbb9a' [22:59:44 CEST] <philipl> nevcairiel: And their SDK doc never discusses encoder performance except for providing numbers for 720p. Nothing else. [22:59:53 CEST] <philipl> Not even to state if you can hope 4k60 can be done. [23:00:37 CEST] <nevcairiel> yeah i didnt find many numbers [23:00:51 CEST] <nevcairiel> shadowplay goes to 130mbit, so i'm going to assume that at least works [23:01:48 CEST] <philipl> You're asking out of curiosity or you need to support multi-hundred-megabit encoding? :-) [23:02:11 CEST] <nevcairiel> mostly curiousity [23:02:33 CEST] <nevcairiel> was also wondering if it would increase the quality to just boost it into the sky [23:02:39 CEST] <nevcairiel> 50mbit still has noticeable loss [23:02:44 CEST] <nevcairiel> but that may be 4:2:0 .. [23:02:51 CEST] <philipl> Probably not. They probably don't pay much attention above 50mbit [23:02:54 CEST] <nevcairiel> (game recording) [23:03:07 CEST] <philipl> (I mean, don't pay attention to how well it uses the bits) [23:03:27 CEST] <philipl> In general nvenc quality is pretty pants. [23:03:57 CEST] <nevcairiel> doesnt it do 444 encoding [23:03:58 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:6308cd4868d2: mov: Check memory allocation [23:03:59 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:3d6635749ab2: Merge commit '6308cd4868d2bd5fdf8bfa8dd10856c9a91874f5' [23:04:01 CEST] <nevcairiel> or did i misremember [23:04:21 CEST] <philipl> nevcairiel: It does. But I'm pretty sure that won't do anything for painfully visible blocking. [23:04:33 CEST] <nevcairiel> i dont get blocking, just a somewhat soft image [23:04:38 CEST] <nevcairiel> which may be the 4:2:0 effect [23:04:43 CEST] <philipl> possibly. [23:05:10 CEST] <philipl> It's basically a toy as far as I can see. Quality is too terrible for any non-realtime uses IMHO. [23:05:42 CEST] <nevcairiel> yeah my only use-cases for it are realtime, game recording through shadowplay, and realtime encoding for mobile streaming [23:06:00 CEST] <nevcairiel> (mobile in my use-case being the tablet on my nightstand <.<) [23:06:27 CEST] <philipl> You're re-encoding for bandwidth reasons? [23:06:40 CEST] <nevcairiel> nah [23:06:46 CEST] <nevcairiel> mostly just convenience [23:07:02 CEST] <philipl> I mean, there isn't much material out there today that isn't natively supported on a tablet. [23:07:13 CEST] <philipl> in terms of codecs and features. [23:07:23 CEST] <nevcairiel> in a generic case, its better to transcode everything and have everything play, rather than transcode nothing and have a couple things not play [23:07:42 CEST] <philipl> True, but a good server can be selective. [23:07:49 CEST] <nevcairiel> well, i was lazy [23:08:10 CEST] <nevcairiel> .. and i do watch a lot of old TV shows which are mpeg4 [23:08:26 CEST] <philipl> android tablets support that :-) [23:08:36 CEST] <philipl> And actually, ipads do as well. [23:08:54 CEST] <philipl> Sort-of. I think they screw up packed b frames. [23:08:55 CEST] <nevcairiel> dont think i ever really tried, the android player i'm using right now certainly doesnt [23:09:06 CEST] <philipl> Kodi should handle it. [23:09:18 CEST] <philipl> I know my android devices expose hardware mpeg4 asp [23:09:39 CEST] <philipl> Heck, I got mpeg2 working on my nexus 6 just by adding it to the xml file. [23:10:21 CEST] <nevcairiel> Well, strictly speaking its all not my stuff, but $work software, including the android player app, which uses Google's ExoPlayer, which is really nice, but doesn't have very broad format support yet [23:10:39 CEST] <nevcairiel> but supporting passthrough h264 on the server is certainly a goal for the next push on streaming features [23:11:48 CEST] <nevcairiel> not sure i'll bother with mpeg4-asp [23:12:20 CEST] <nevcairiel> transcoding will always need to be an option for bandwidth reasons when you are actually out, and not just using it over 300mbit wifi in your flat =p [23:12:22 CEST] <cone-068> ffmpeg 03Vittorio Giovara 07master:4733a12dd17a: rtpdec_asf: Check memory allocation and free memory on error [23:12:23 CEST] <cone-068> ffmpeg 03Michael Niedermayer 07master:d5a645625d87: Merge commit '4733a12dd17a91d606e0079ff9bb48b9f419cbef' [23:27:51 CEST] <cone-068> ffmpeg 03James Almer 07master:1382add59df1: mpjpegdec: don't try to alloc an AVIOContext when probe is guaranteed to fail [23:31:16 CEST] <cone-068> ffmpeg 03Andreas Cadhalpun 07master:6fdbaa2b7fb5: vp8: change mv_{min,max}.{x,y} type to int [23:34:13 CEST] <llogan> c_14: in 43c1d82 you changed the example from pan=1 to pan=1c. was this due to d96e377? [23:35:03 CEST] <llogan> with 1c i just get silence [23:38:27 CEST] <c_14> IIRC, yes. [23:38:47 CEST] <c_14> Let me test. [23:44:12 CEST] <c_14> llogan: using 1c works correctly on 43c1d82b, looks like a regression somewhere afterwards [23:44:57 CEST] <llogan> i thought i tested on 43c...but i already forgot and now am recompiling for something else [23:58:54 CEST] <jamrial> ubitux: i wonder if implementing av_clip_int* with sse2 packss instructions would be faster than the c version [23:58:58 CEST] <c_14> Wait, hmm. Now it's not working. [00:00:00 CEST] --- Tue Jun 9 2015

1 0

ffmpeg.log.20150608
by burek 09 Jun '15

09 Jun '15

[00:00:11 CEST] <kodur> nevermind, i think I got it. [00:08:34 CEST] <kodur> thanks anyway :) [00:25:45 CEST] <animax> hi all. whttp://www.pasteall.org/58869 hat do the expressions '%%a' and '%%~na' mean? [00:25:57 CEST] <animax> sorry [00:26:02 CEST] <animax> link: [00:26:04 CEST] <animax> http://www.pasteall.org/58869 [00:26:43 CEST] <animax> what do the expressions '%%a' and '%%~na' mean? [01:23:23 CEST] <zumba_addict> hey folks, a friend of mine gave me this mp4 file and ffmpeg reported this - Stream #0:0: Video: h264 (Constrained Baseline), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 1k tbn, 50 tbc [01:23:56 CEST] <zumba_addict> mplayer cannot play it. What is it? [01:27:47 CEST] <zumba_addict> here is a complete ffmpeg -i report - http://pastebin.com/rnTAhLj5 [02:28:36 CEST] <chungy> zumba_addict: https://en.wikipedia.org/wiki/H.264/MPEG-4_AVC#Profiles [02:29:13 CEST] <chungy> seems odd that mplayer can't handle it; does it work with ffplay, or other players like vlc or mpv? [04:29:31 CEST] <zumba_addict> chungy: it's ok now. It just took so long for mplayer to play it [04:29:37 CEST] <zumba_addict> but it's playing well now [04:41:59 CEST] <HellsZNC> join server irc.deadchildren.info join channel #HELLNET [04:42:00 CEST] Last message repeated 1 time(s). [04:42:02 CEST] Last message repeated 2 time(s). [04:42:02 CEST] <HellsZNC> v [04:42:03 CEST] <HellsZNC> join server irc.deadchildren.info join channel #HELLNET [06:49:04 CEST] <bindi> greetings, will nvenc work on ubuntu and a gt 520? [09:06:07 CEST] <zachary> [flv @ 0x1c53260] Video codec wmv3 not compatible with flv [09:06:35 CEST] <zachary> could anyone help on how to push AVI video with ffmpeg [09:07:08 CEST] <zachary> this is the command: ffmpeg -re -i ~/nrf/sq.avi -c copy -f flv -strict -2 rtmp://127.0.0.1/live/mystream [09:20:07 CEST] <harisk92> hello guys [09:20:32 CEST] <harisk92> I just tried to install ffmpeg from sources on Debian Jessie [09:21:12 CEST] <harisk92> when I enter into ffmpeg folder which I cloned from git//git.videolan.org [09:21:20 CEST] <harisk92> and try to configure [09:21:33 CEST] <harisk92> Android studio opens [09:21:52 CEST] <harisk92> this is so weird [09:55:55 CEST] <cehoyos> harisk92: "and try to configure" <- what do you mean exactly? [09:56:43 CEST] <harisk92> cehoyos: ./configure [09:58:50 CEST] <cehoyos> And that opens Android studio? [10:16:25 CEST] <harisk92> cehoyos: yes [10:16:34 CEST] <harisk92> cehoyos: I removed [10:16:46 CEST] <harisk92> cehoyos: android studio and it went trought [10:17:00 CEST] <harisk92> cehoyos: now I have another issue [10:17:03 CEST] <harisk92> mkdir: cannot create directory /usr/local/share/ffmpeg: No such file or directory [10:17:42 CEST] <cehoyos> But this is an installation issue, not a build issue or do I misunderstand? [10:18:07 CEST] <harisk92> cehoyos: yup [10:18:14 CEST] <cehoyos> Note that I am curious how your original issue happened: Did configure run through and Android Studio was just opened additionally? [10:18:48 CEST] <cehoyos> Short answer: run make install as root (or sudo), long answer is that I never install (and I build practically every day), it depends on your usecase. [10:19:26 CEST] <cehoyos> The ffmpeg binary that you created should run by itself (if you did not specify "--enable-shared", you can put it into any directory you want. [10:19:31 CEST] <harisk92> cehoyos: Iam building .deb file from sources using checkinstall [10:20:00 CEST] <cehoyos> Ok, since I am not a Debian user myself, I cannot really comment but sudo (or running as root) should fix that. [10:20:05 CEST] <harisk92> cehoyos: but Iam having hard time I guess it's just a part of process xD [10:20:17 CEST] <harisk92> cehoyos: did it [10:41:05 CEST] <harisk92> cehoyos: everything went fine [10:41:15 CEST] <harisk92> cehoyos: I had to manually create folder :/ [10:45:52 CEST] <cehoyos> How can I reproduce this? [10:50:48 CEST] <cehoyos> I tested "./configure --prefix=/some/path" and make install did create "/some/path/shared/ffmpeg" [10:51:01 CEST] <cehoyos> s/shared/share [11:05:11 CEST] <aiena> I had a script which worked fine earlier rerunning that script on a fresh install does not work I get this strange error "[x11grab @ 0xf22680] Capture area 1366x741 at position 0.0 outside the screen size 1366x768" [11:05:32 CEST] <aiena> but 1366x741 is smaller than 1366x768 [11:05:37 CEST] <aiena> i dont know what they mean [11:12:12 CEST] <aiena> hmm got it it needs to be 1365 is scren res is 1366 [11:18:25 CEST] <chungy> aiena: it's a bug [11:18:48 CEST] <chungy> See the tip of the releases/2.6 branch for a fix. Should be safe to compile from that, too. [11:26:45 CEST] <aiena> chungy thanks [11:26:52 CEST] <aiena> er info ffmpeg [11:26:57 CEST] <aiena> erm sorry [11:27:21 CEST] <aiena> terminal command partial output accidentally went here [11:29:43 CEST] <mkilivan> Hi all, I am trying to convert g722 file to wav file. The default input frequency is 16000 Hz, I couldn't change this value. Here is command and respond. [11:29:46 CEST] <mkilivan> >ffmpeg -f g722 -ar 8000 -i input.g722 output.wav [11:30:02 CEST] <mkilivan> ... [11:30:03 CEST] <mkilivan> Option sample_rate not found. [11:31:22 CEST] <mkilivan> Can anyone point me in the right direction for doing that? [12:00:36 CEST] <durandal_1707> havent you read documentation and list demuxer options [12:06:56 CEST] <animax> hi again. what is the meaning of '%%a' and '%%~na' in this expression: http://www.pasteall.org/58875? working on win 7 ult 64bit. I want to batch encode a couple of video files. all are in the same folder with nothing else in it. [12:11:43 CEST] <mkilivan> durandal_1707 I had a look at the documentation but it didn't work for me. Perhaps I wasn't looking at the right section. I would be glad if you could help. [12:42:22 CEST] <harisk92> hey guys I have this command ffmpeg -f lavfi -i 'testsrc=size=704x480:duration=30' -vcodec libx264 -f ssegment -hls_list_size 10 -segment_list playlist.m3u8 -segment_time 10 -segment_list_flags +live -strftime 1 "%Y%m%d%H%M%S.ts" [12:42:36 CEST] <harisk92> how can I add testsrc audio to it ? [12:49:47 CEST] <durandal_1707> mkilivan: ffmpeg -h demuxer=g722 [12:53:03 CEST] <cehoyos> harisk92: ffmpeg -f lavfi -i testsrc -f lavfi -i sine [12:53:39 CEST] <harisk92> cehoyos: thx mate you are life saver [12:53:55 CEST] <cehoyos> harisk92: Why did you try to specify the input sample rate for your g722 file? Did "ffmpeg -i input.g722 out.wav" not work? [12:56:27 CEST] <harisk92> cehoyos: I didn't [12:56:50 CEST] <harisk92> cehoyos: that's mkilivan guy I think :D [12:57:26 CEST] <mkilivan> cehoyos it worked but generated wav file looks to be about 2x the frequency [12:57:58 CEST] <mkilivan> And I am sure my input file sample's rate is 8000 Hz. [12:58:56 CEST] <durandal_1707> have you tried my company? [12:59:12 CEST] <durandal_1707> command [12:59:52 CEST] <durandal_1707> there is setrate filter [13:02:00 CEST] <mkilivan> durandal_1707: I tried. It didn't print help for g722. The output is here http://pastebin.com/S4qFT4ZJ [13:04:18 CEST] <durandal_1707> then try filter [13:07:12 CEST] <cehoyos> mkilivan: Please provide your input file. [13:09:21 CEST] <cehoyos> Wikipedia says: "G.722 sample audio data at a rate of 16 kHz" [13:09:34 CEST] <cehoyos> This is not necessarily true but it is the only information (except FFmpeg) we have so far. [13:09:40 CEST] <spaceone> hi [13:09:49 CEST] <spaceone> i get ":0.0: Protocol not found" [13:10:04 CEST] <spaceone> ffmpeg -f alsa -i pulse -f x11grab -framerate 10 -s '1920*1080' '1920*1080' -i :0.0 -c:v libvpx -deadline realtime -qmin 10 -qmax 20 -b:v 2M -c:a libvorbis -q:a 4 ./2015-06-08-125657_screen.webm [13:10:05 CEST] <mkilivan> cehoyos: Here is the g722 file http://pastebin.com/RhrijRMc [13:10:17 CEST] <spaceone> how can i fix this? [13:11:27 CEST] <cehoyos> How can I use this?? [13:12:07 CEST] <cehoyos> spaceone: Please post your complete cosole output on pastebin [13:12:41 CEST] <cehoyos> mkilivan: Did you produce your file with FFmpeg? Is there an application that plays it correctly? [13:13:37 CEST] <harisk92> cehoyos: how can I see advanced help for -v option ? [13:13:58 CEST] <harisk92> cehoyos: I tried ffmpeg -help -v [13:14:28 CEST] <cehoyos> You find it on https://ffmpeg.org/ffmpeg-all.html if you search for "loglevel" [13:15:00 CEST] <mkilivan> cehoyos: Sorry, you can downlaod the g722 file from here. [13:15:10 CEST] <mkilivan> http://filebin.ca/24WO7KvynP21/FH_R0300_3.g722 [13:15:35 CEST] <spaceone> cehoyos: http://nopaste.linux-dev.org/?543550&download [13:16:57 CEST] <cehoyos> What is the following supposed to do? -s '1920*1080' '1920*1080' [13:17:26 CEST] <spaceone> screen resolution [13:17:32 CEST] <spaceone> I just copied it [13:17:51 CEST] <cehoyos> Please use instead;. -video_size hd1080 [13:18:05 CEST] <cehoyos> Or -video_size 1920x1080 [13:18:26 CEST] <spaceone> okay [13:18:28 CEST] <cehoyos> Or even -s 1920x1080 [13:19:03 CEST] <spaceone> nice! it works then! [13:19:18 CEST] <cehoyos> mkilivan: How did you produce the file? [13:19:32 CEST] <cehoyos> FFmpeg currently has a bug: It allows to write G722 files that are not 16kHz [13:19:46 CEST] <spaceone> "currently" == which version? [13:21:08 CEST] <mkilivan> cehoyos: This file is captured from an g722 voice recorder. I played this file on Audacity after it has been converted to wav. I don't know if there an application to play g722 directly. [13:24:30 CEST] <cehoyos> So the file you uploaded was produced by a hardware encoder? [13:24:53 CEST] <cehoyos> And is there software or hardware that plays it differently than ffplay? [13:26:15 CEST] <cehoyos> Could you tell the name of the voice recorder? [13:26:29 CEST] <mkilivan> cehoyos: Yes, this file produced by a hardware encoder. But there is no software/hardware to plat this. [13:26:39 CEST] <cehoyos> Not even on the recordere? [13:26:39 CEST] <mkilivan> *play [13:27:34 CEST] <mkilivan> Not even recoder. [13:27:58 CEST] <cehoyos> What is the name / type of the recorder? [13:28:04 CEST] <mkilivan> The only interface on the recorder is uart interface [13:28:15 CEST] <cehoyos> (This issue would be very easy to fix but if the file is invalid it may be better not to fix it...) [13:29:52 CEST] <mkilivan> It's not on the shelf as it is still under development. [13:30:51 CEST] <cehoyos> So is there a possibility that the device is simply buggy and produces incorrect files? [13:32:53 CEST] <mkilivan> yes, it is possible. That's why I need to play back on ffplay. [13:33:45 CEST] <mkilivan> If you say 16Khz is a standard value it is must than I would accept this device is faulty. [13:34:30 CEST] <cehoyos> No, I said that wikipedia claims that G722 is 16kHz (both English and German). [13:34:51 CEST] <cehoyos> I believe the specification is available, please search there! [13:35:33 CEST] <mkilivan> If I could change the sample rate to 8Khz it would be easy to confirm that device is works fine. Currently I don't have any tool/software to do that. [13:36:18 CEST] <cehoyos> FFmpeg has a filter to do this... [13:37:26 CEST] <mkilivan> Is this filter for only output? Because I did change the output file bitrate but not input? [13:37:28 CEST] <cehoyos> -af aresample=8k [13:37:42 CEST] <cehoyos> Filters only affect the output file, never the input file [13:39:42 CEST] <mkilivan> So, it seems there is no way to change input sample rate at the moment? [13:39:54 CEST] <cehoyos> ffmpeg -i yourfile -af aresample=8k out.wav [13:39:55 CEST] <cehoyos> ? [13:40:39 CEST] <cehoyos> No, sorry, I misunderstood. [13:40:52 CEST] <cehoyos> you have to edit libavformat/g722.c and replace 16000 with 8000 [13:41:58 CEST] <mkilivan> Alright, I will have a look at the standard to see if 8Khz is supported. Also I will try to change device sample rate to 16Khz (If I can). [13:42:09 CEST] <mkilivan> Thanks for your time cehoyos. [13:42:11 CEST] <cehoyos> No. [13:42:17 CEST] <cehoyos> Found the filter: asetrate [13:42:34 CEST] <cehoyos> -af asetrate=8k [13:42:44 CEST] <cehoyos> Does this produce the correct output? [13:42:57 CEST] <mkilivan> aah, that is good advice, thanks. [13:47:47 CEST] <anshul_mahe> does any one know of some test videos which are most difficult to maintain constant bitrate for h264, and other video codec [13:48:30 CEST] <cehoyos> You mean -f rawvideo -i /dev/random ? [13:49:47 CEST] <anshul_mahe> I dont know, it looks like it would be worst video to maintain cbr for any codec. and best for my test [13:50:35 CEST] <anshul_mahe> I am just playing with ffmpeg and x264 preset file [13:57:30 CEST] <harisk92> cehoyos: I have question this is my command which produces ffmpeg -f lavfi -i 'testsrc=size=704x480' -f lavfi -i sine -vcodec libx264 -preset ultrafast -profile:v baseline -level 3.0 -pix_fmt yuv420p -acodec libfaac -f ssegment -segment_list_size 5 -segment_list chunklist.m3u8 -segment_time 10 -segment_list_flags +live -strftime 1 "%Y%m%d%H%M%S.ts" [13:58:10 CEST] <harisk92> cehoyos: I set up nginx webserver to call chunklist.m3u8 [13:58:23 CEST] <harisk92> cehoyos: and I opened it locally in Safari [13:58:54 CEST] <harisk92> cehoyos: and what I notice is that video and audio constantly freezing [13:59:13 CEST] <harisk92> cehoyos: at random intervals between 5-15 seconds [13:59:26 CEST] <harisk92> cehoyos: do you know reason behind this ? [14:01:59 CEST] <harisk92> cehoyos: Iam playing .m3u8 file with default Debian player and it freezes after 1 minute or so [14:08:06 CEST] <harisk92> looks like iam missing -re flag [15:27:25 CEST] <_vaibhavingale_> http://irc.lc/freenode/chromium/t4nk@@@ [15:28:57 CEST] <_vaibhavingale_> I have converted flac file into mp3 @ 320 kbps. When I play it in browser throught it takes 20-25 seconds delay to buffer before it could start [15:29:11 CEST] <_vaibhavingale_> is there any way I can reduce that delay? [15:38:01 CEST] <chungy> faster network connections or getting Chromium to have a smaller buffer, or use a bitrate that produces a smaller file. I don't imagine there's much else control you can have. [15:50:01 CEST] <_vaibhavingale_> Okie, so what if I make mp3 segments from some streaming server. [16:45:59 CEST] <rjp421> http://www.streamingmedia.com/ConferenceVideos [18:01:45 CEST] <NoTiTo> morning [18:02:24 CEST] <NoTiTo> How dow I workout my ffmpeg version fro this response? >>ffmpeg version N-72743-g130a6c0 [18:02:57 CEST] <NoTiTo> this is what I get from running "ffmpeg -version" after compiling from the git repo [18:05:56 CEST] <anoop_r> how to set profile in libx265 [18:06:18 CEST] <anoop_r> i want to set profile to main444-10 [18:06:46 CEST] <anoop_r> i had used this -x265-params profile=main444-10 but not working [18:06:49 CEST] <anoop_r> why ? [18:08:19 CEST] <chungy> NoTiTo: That's because there is no version number associated with the Git HEAD :P [18:08:30 CEST] <chungy> "git show 130a6c0" would give you the exact commit [18:10:24 CEST] <NoTiTo> chungy: thanks, my commit is "130a6c04a4660e77d5eaf2e84e9fe751e091449b". How do I work it out from there? [18:11:22 CEST] <chungy> What are you trying to work out? [18:13:50 CEST] <NoTiTo> The version number e.g. 2.6... [18:14:10 CEST] <chungy> There is none [18:14:18 CEST] <anoop_r> i had used latest ffmpeg to compile [18:14:22 CEST] <chungy> It'll eventually be in 2.7 and later versions, but that doesn't exist yet. [18:14:29 CEST] <NoTiTo> I pulled fresh source via "git pull origin master" [18:16:20 CEST] <chungy> there is no version number. [18:16:52 CEST] <NoTiTo> I pulled from this repo git://source.ffmpeg.org/ffmpeg.git - master branch [18:17:23 CEST] <NoTiTo> would that be the latest stable? Recommended? [18:17:46 CEST] <chungy> the latest stable and recommended is 2.6.3. If you want to check it out from your repo, just type "git checkout n2.6.3" [18:18:43 CEST] <NoTiTo> once I've done that do I just rebuild? ./confugure + make, etc? [18:18:49 CEST] <chungy> yes [18:24:24 CEST] <NoTiTo> chungy: thanks mate, trying it out now [19:01:11 CEST] <Guest55906> hi [19:03:12 CEST] <Guest55906> I'm pretty new to ffmpeg and I'm trying to convert multiple videos(with different aspect ratios and encodings) to a default aspect ratio(let's say 4x3) and default encoding [19:05:20 CEST] <Guest55906> do I need to go and check the aspect ratios and encodings for each video separately and then based on the information to build the ffmpeg command or there is a way I can "force" encoding and aspect ratio(by padding if necessary). What is the best practice? [19:27:13 CEST] <llogan> Guest55906: are you also trying to scale to a specific width and/or height? [19:29:09 CEST] <NoTiTo> chungy: I switched branches to n2.6.3, rebuilt and I'm still not getting 2.6.3 in my version [19:42:52 CEST] <chungy> NoTiTo: that shouldn't be, but it might be because you configured it on the master branch before. [19:43:13 CEST] <chungy> "git clean -xfd" will remove all files not in the repository tree, if you want to try it from a completely clean slate. [19:43:37 CEST] <NoTiTo> would I still need to run "make clean"? [19:45:16 CEST] <chungy> It would remove all the generated files, you'd need to do ./configure again [19:45:31 CEST] <chungy> so, no [19:45:46 CEST] <NoTiTo> cool, running ./configure again [19:56:43 CEST] <NoTiTo> ~ ffmpeg -version [19:56:43 CEST] <NoTiTo> ffmpeg version n2.6.3 Copyright (c) 2000-2015 the FFmpeg developers [19:56:56 CEST] <NoTiTo> chungy: it worked :) [19:57:01 CEST] <chungy> woo :) [19:57:06 CEST] <NoTiTo> w00t! [19:57:09 CEST] <NoTiTo> thanks mate [19:57:48 CEST] <NoTiTo> now, to try and get MLT Framework & Melted (server) working [20:02:53 CEST] <anoop_r> ffmpeg1 -pix_fmt yuv444p -i Air.mkv -c:v libx265 -preset veryfast -x265-params crf=28 -c:a aac -strict experimental -b:a 128k output.mp4 [20:03:02 CEST] <anoop_r> but Option pixel_format not found. [20:03:23 CEST] <anoop_r> can anybody tell me how to use -pix_fmt [20:06:19 CEST] <anoop_r> ok [20:06:24 CEST] <chungy> you put -pix_fmt as an input file option. you want it as an output option [20:07:12 CEST] <[-T-]> just a small patch [20:07:17 CEST] <[-T-]> oups [20:07:31 CEST] <llogan> anoop_r: also, -crf should work with libx265 now instead of using -x265-params [20:07:34 CEST] <anoop_r> i want to convert 4:2:2 video to 4:4:4 AND to provide it to libx265 [20:08:50 CEST] <Guest55906> @llogan I'm trying to scale to 4x3, and yes, it can have some default width and height e.i. 720x480 [20:09:12 CEST] <Guest55906> llogan I'm trying to scale to 4x3, and yes, it can have some default width and height e.i. 720x480 [20:10:34 CEST] <anoop_r> http://pastie.org/10229959 [20:10:40 CEST] <anoop_r> please see this [20:13:53 CEST] <llogan> anoop_r: "<chingy> you put -pix_fmt as an input file option. you want it as an output option" [20:14:08 CEST] <llogan> that was supposed to be "chungy" [20:14:30 CEST] <anoop_r> ya but i don't get it [20:14:54 CEST] <anoop_r> you seen my command line [20:15:08 CEST] <llogan> http://ffmpeg.org/ffmpeg.html#Synopsis [20:16:00 CEST] <chungy> the order of arguments matters in ffmpeg [20:16:30 CEST] <chungy> -pix_fmt for input only makes sense for raw video streams [20:17:22 CEST] <llogan> and even then it should be -pixel_format for rawvideo demuxer [20:18:06 CEST] <anoop_r> now i understand [20:26:56 CEST] <anoop_r> i had encoded using -pix_fmt yuv444p [20:27:24 CEST] <anoop_r> but the output videos profile is unknown now [20:27:31 CEST] <anoop_r> why [20:27:43 CEST] <anoop_r> Format/Info : High Efficiency Video Coding [20:27:55 CEST] <anoop_r> Format profile : Unknown(a)L3.1 [20:29:09 CEST] <llogan> Guest55906: maybe you want something like this? http://superuser.com/a/547406/110524 [20:33:03 CEST] <anoop_r> i am not using raw video .also my output video is playable with no problem.the only thin i notice the profile name turned to unknown [20:41:39 CEST] <kyleogrg> hi [20:43:09 CEST] <kyleogrg> best IVTC setting, in your opinions? [20:44:00 CEST] <JEEBsv> correct ones for your source [20:44:16 CEST] <JEEBsv> I mean, IVTC is not exactly limited by anything else than you getting the right pattern :P [20:44:29 CEST] <JEEBsv> there is literally no "best" settings [20:44:40 CEST] <JEEBsv> just correct ones and incorrect ones [20:46:40 CEST] <kyleogrg> my source is a movie which was probably 24p (film) [20:47:09 CEST] <kyleogrg> That is, that's the source of the DVD I'm ripping from. [20:47:19 CEST] <kyleogrg> The DVD itself is 29.97 interlaced [20:47:21 CEST] <JEEBsv> I recommend previewing with something like mpv and testing out the swscale IVTC filters [20:47:27 CEST] <JEEBsv> uuh [20:47:30 CEST] <JEEBsv> libavfilter I mean [20:47:38 CEST] <JEEBsv> that way you can have a visual look at it [20:47:56 CEST] <kyleogrg> ok, what's mpv? [20:48:03 CEST] <JEEBsv> it's the sanest mplayer fork [20:48:43 CEST] <JEEBsv> the filters you'd be testing are: https://www.ffmpeg.org/ffmpeg-all.html#fieldmatch and https://www.ffmpeg.org/ffmpeg-all.html#decimate [20:49:18 CEST] <kyleogrg> okay, the reason i ask is I have a video encoding in ffmpeg [20:49:34 CEST] <kyleogrg> my IVTC setting is fieldmatch, yadif, decimate [20:49:51 CEST] <kyleogrg> and the output looks good, except it seems slightly jumpy [20:49:53 CEST] <JEEBsv> sounds like a sane'ish combo, although in general you don't need the yadif there [20:50:17 CEST] <JEEBsv> anyways, I recommed doing field-by-field / frame-by-frame checks [20:50:20 CEST] <kyleogrg> so actually, i just want to know why it's jumpy [20:50:21 CEST] <JEEBsv> with mpv or so [20:50:38 CEST] <JEEBsv> (or you could test a frameserver like vapoursynth with vapoursynth editor) [20:50:51 CEST] <JEEBsv> but mpv would let you test avfilter chains [20:50:58 CEST] <JEEBsv> which you can then use in ffmpeg [20:51:14 CEST] <kyleogrg> offhand, do you know why output might seems slightly jumpy or uneven? [20:51:50 CEST] <JEEBsv> no, because it can be everything from "that's how it is when you have stuff at 24/1.001fps" to "you're field matching it wrong" or "you're decimating it wrong" [20:53:18 CEST] <JEEBsv> which is why the first thing you do is something a la separatefields() -> selecteven() and go through the content in a few spots. note the type of content you're dealing with. possibly check with the selecteven() removed and check how the fields go. then test according to the documentation and preview before actually doing encodes [20:53:38 CEST] <kyleogrg> Okay... Source is NTSC DVD. FFmpeg IVTC setting is: "fieldmatch,yadif=deint=interlaced,decimate" [20:53:59 CEST] <JEEBsv> unfortunately it's not as simple as that, as I said you have to dig in and look at your source [20:54:07 CEST] <kyleogrg> ok [20:54:09 CEST] <JEEBsv> with mpv or vapoursynth or something [20:54:15 CEST] <JEEBsv> that gives you a good look [20:54:27 CEST] <kyleogrg> those functions you just mentioned - what program are those for? [20:54:33 CEST] <kyleogrg> I only am familar with ffmpeg [20:54:45 CEST] <ChocolateArmpits> vapoursynth is a frameserver, mpv is a media player [20:54:46 CEST] <JEEBsv> those are vapoursynth, but similar stuff is possible with avfilter most definitely as well as with vapoursynth [20:55:11 CEST] <kyleogrg> ok thanks [20:55:14 CEST] <JEEBsv> ChocolateArmpits: yes, but coupled with vapoursynth editor it gives you a rather good way of doing preview of your work :P [20:55:29 CEST] <ChocolateArmpits> It's missing one thing [20:55:31 CEST] <ChocolateArmpits> play button [20:55:48 CEST] <JEEBsv> I was talking about frame-by-frame shit and you can open the shit in mpv if you really want to [20:56:24 CEST] <JEEBsv> so what I noted in mpv to begin with is to be able to preview your avfilter filtering chains since the result is pretty much the same between it and ffmpeg in that regard [20:56:54 CEST] <ChocolateArmpits> Why not use ffplay to preview the filtering ? [20:57:00 CEST] <JEEBsv> because ffplay is fucking shit [20:57:10 CEST] <JEEBsv> it's a proof of concept piece of shit [20:57:27 CEST] <JEEBsv> mpv might actually have other uses and the lavf/lavc/whatever usage is generally close enough to ffplay [20:57:28 CEST] <ChocolateArmpits> I never had problems with it and it's a good way to go from the filtergraph there to ffmpeg [20:57:48 CEST] <JEEBsv> yes, except it's not the only thing capable of that [20:58:12 CEST] <kyleogrg> you can't step field-by-field in vlc, can you? [20:58:17 CEST] <JEEBsv> nope [20:58:25 CEST] <JEEBsv> use a player that lets you use avfilter chains [20:58:29 CEST] <JEEBsv> mpv, ffplay, whatever [20:58:30 CEST] <JEEBsv> :P [20:58:46 CEST] <kyleogrg> ok [20:58:57 CEST] <kyleogrg> can you give me a command line example for ffplay (never used it)? [20:59:03 CEST] <JEEBsv> I don't use ffplay [20:59:10 CEST] <JEEBsv> I just mentioned it since ChocolateArmpits happened to mention it [20:59:13 CEST] <ChocolateArmpits> well yadif deinterlacing could technically be substituted for a field previewer ? [20:59:19 CEST] <kyleogrg> then mpv? [20:59:28 CEST] <JEEBsv> see the mpv manual regarding avfilter chains [20:59:32 CEST] <JEEBsv> the manual is pretty good [20:59:50 CEST] <JEEBsv> http://mpv.io/manual/master/#video-filters [20:59:52 CEST] <kyleogrg> ok [21:00:17 CEST] <JEEBsv> it does IIRC have its own filters too, but you create an avfilter filter chain as well [21:00:27 CEST] <JEEBsv> yeah lavfi=graph[:sws-flags[:o=opts]] [21:00:35 CEST] <JEEBsv> http://mpv.io/manual/master/#video-filters-lavfi [21:02:18 CEST] <kyleogrg> why isn't pullup more simple? [21:02:41 CEST] <kyleogrg> because isn't the pulldown method pretty standard? [21:02:51 CEST] <JEEBsv> you'd be surprised :P [21:03:28 CEST] <kyleogrg> theoretically, if there's an uninterrupted pulldown pattern, can't it be simply reversed? [21:03:44 CEST] <JEEBsv> yes, you just have to set the pattern then and that's it [21:03:51 CEST] <JEEBsv> unfortunately, in a lot of cases that's not how it goes [21:03:57 CEST] <kyleogrg> hmm ok [21:04:00 CEST] <anoop_r> please see this http://pastie.org/10230072 [21:04:02 CEST] <JEEBsv> which is why algos usually look at possible patterns [21:04:12 CEST] <anoop_r> profile is not changing [21:04:14 CEST] <JEEBsv> of course there's also the completely fucked up shit [21:04:24 CEST] <JEEBsv> which depends on your level of masochism [21:04:27 CEST] <kyleogrg> but this is an entire film on dvd. should it have a consistent pulldown pattern? in other words, it's unedited [21:04:50 CEST] <JEEBsv> like 30i credits on 24p content and stuff like that [21:05:08 CEST] <kyleogrg> ok [21:05:11 CEST] <JEEBsv> (I'm not even going to mention complete frame rate conversion fuck-ups [21:05:18 CEST] <kyleogrg> this is a 1957 hollywood movie [21:05:24 CEST] <kyleogrg> everything should be the same frame rate [21:05:32 CEST] <ChocolateArmpits> but what's the source for the movie? [21:05:41 CEST] <ChocolateArmpits> Are you sure it's film negative ? [21:05:47 CEST] <ChocolateArmpits> and not some vhs [21:05:48 CEST] <JEEBsv> well, I'm just going to say that it's either as simple or as hard depending on the content [21:05:54 CEST] <JEEBsv> which is why you take a look at it [21:05:55 CEST] <kyleogrg> It's an official Bluray/DVD release [21:06:23 CEST] <kyleogrg> yes, it's remastered/rescanned from the film [21:06:47 CEST] <kyleogrg> JEEBsv: yeah. in this case then, it's probably simple? [21:06:52 CEST] <kyleogrg> cause it's from film? [21:06:54 CEST] <JEEBsv> that's possible [21:07:02 CEST] <JEEBsv> but you never, ever decide that before looking [21:07:12 CEST] <kyleogrg> yeah [21:07:32 CEST] <JEEBsv> but basically if you look at the source, look at what your final filter chain outputs and then you know if you've done something wrong when it feels "uneven" at the end [21:08:12 CEST] <JEEBsv> because yunno, it could be everything from you feeling it being like that due to video rendering being shit in whatever you're testing to it actually being like that originally [21:08:35 CEST] <kyleogrg> right [21:11:16 CEST] <kyleogrg> as for x265 in ffmpeg, why won't it accept -x265-params tune=grain? [21:12:30 CEST] <JEEBsv> dunno [21:13:06 CEST] <JEEBsv> I think that param is supposed to just pass the options through to libx265 as-is [21:13:25 CEST] <JEEBsv> so either your binary is old/whatever or they just don't do that stuff like that [21:13:31 CEST] <JEEBsv> also I think tunings could be set with just -tune ? [21:13:40 CEST] <JEEBsv> a la libx264's wrapper [21:14:13 CEST] <kyleogrg> hmm, i haven't tried -tune with libx265. binary could be old. the tune options are officially listed for x265. [21:16:05 CEST] <brontosaurusrex> kyleogrg: bluray should run at 24p, so no pulldowns [21:16:39 CEST] <JEEBsv> he's using the DVD as source as far as I can tell [21:16:39 CEST] <kyleogrg> i'm ripping from the DVD that came with it [21:16:50 CEST] <JEEBsv> also blu-ray supports 30i mighty fine [21:16:58 CEST] <JEEBsv> and some people have (ab)used that fact [21:17:09 CEST] <brontosaurusrex> JEEBsv: yes, but would they really do that? [21:17:15 CEST] <brontosaurusrex> i see [21:17:32 CEST] <JEEBsv> the fuck I know, I just know that I have some shit like that due to someone in the chain being a faggot :P [21:17:55 CEST] <brontosaurusrex> I'd definately just return that [21:17:58 CEST] <JEEBsv> (and then there's the cases of "no, we cannot re-edit the credits to be 24p. stop crying." [21:18:13 CEST] <JEEBsv> (and then the whole thing is 30i with mostly 24p content due to that) [21:18:23 CEST] <kyleogrg> okay, what about CRF in x265? how does it correspond to x264's? [21:18:32 CEST] <JEEBsv> in no way whatsoever [21:18:36 CEST] <JEEBsv> you should stop comparing the two [21:18:41 CEST] <brontosaurusrex> kyleogrg: it doesn't [21:18:46 CEST] <JEEBsv> they're made by completely different people etc [21:18:56 CEST] <kyleogrg> then what is generally considered visually lossless for x265? [21:19:16 CEST] <JEEBsv> nothing is considered like that for either of the two encoders supporting the CRF rate control [21:19:28 CEST] <JEEBsv> encode a few thousand pictures and decide yourself [21:19:42 CEST] <JEEBsv> like, cut 500 pictures in five or so places [21:19:54 CEST] <JEEBsv> then check the result between different values [21:20:03 CEST] <kyleogrg> there's no basic range? maybe people say 18-23 for x264 (i know it's not the same) [21:20:20 CEST] <JEEBsv> no they don't [21:20:31 CEST] <JEEBsv> and if they do they have no idea what they're talking about [21:20:35 CEST] <JEEBsv> I know people love to run from work [21:20:43 CEST] <kyleogrg> and i mean nearly lossless [21:21:13 CEST] <kyleogrg> https://trac.ffmpeg.org/wiki/Encode/H.264 [21:21:14 CEST] <JEEBsv> anyways, it depends on the content, resolution of content and the resolution at which you will be watching it [21:21:20 CEST] <JEEBsv> among other things [21:21:26 CEST] <kyleogrg> "Consider 18 to be visually lossless or nearly so: it should look the same or nearly the same as the input but it isn't technically lossless. " [21:21:37 CEST] <JEEBsv> kyleogrg: then whomever wrote that is a fucking idiot [21:21:40 CEST] <brontosaurusrex> kyleogrg: just rip the doovd and keep as is [21:21:47 CEST] <sfan5> https://trac.ffmpeg.org/wiki/Encode/H.265 says "The CRF of 28 should visually correspond to libx264 video at CRF 23" [21:22:10 CEST] <JEEBsv> sfan5: wow, some other bullshit [21:22:28 CEST] <kyleogrg> sfan5: that's actually what I did. crf 18 is (supposedly) generally visually lossless for x264. So I did crf 13 for x265. [21:22:38 CEST] <JEEBsv> ... [21:22:46 CEST] <sfan5> uhh [21:22:54 CEST] <kyleogrg> JEEBsv: but it's unreliable numbers? [21:23:22 CEST] <JEEBsv> yes, please stop copying numbers and just do some prework for fuck's sake [21:23:47 CEST] <kyleogrg> JEEBsv: sure. that's why i'm asking - I don't know which source to trust [21:24:11 CEST] <sfan5> kyleogrg: if x265 crf 28 is (-5) x264 crf 23 then x264 crf 18 should be (+5) x265 crf 23 (not 13) [21:24:33 CEST] <JEEBsv> the default crf is where you start, 23 with x264 and 26 with x265 IIRC [21:24:36 CEST] <kyleogrg> sfan5: ooh good catch [21:24:41 CEST] <JEEBsv> then you go down if shit looks bad [21:24:43 CEST] <bindi> greetings, will nvenc work on ubuntu and a gt 520? [21:24:46 CEST] <JEEBsv> you go up if it looks good [21:25:09 CEST] <kyleogrg> yeah [21:25:10 CEST] <JEEBsv> I'm not even gonna say you have to try the non-integer numbers [21:25:24 CEST] <JEEBsv> just fucking do the work for your eyes [21:25:37 CEST] <sfan5> bindi: the gt 520 supports nvenc so, yes, probably [21:25:40 CEST] <JEEBsv> and if you need lossless you just do lossless although with DVDs that makes no fucking sense [21:25:42 CEST] <kyleogrg> JEEBsv: gotcha [21:25:49 CEST] <kyleogrg> speaking of lossless, [21:26:04 CEST] <kyleogrg> I wonder what compression ratio you might get in x265? [21:26:06 CEST] <JEEBsv> because the lossless will end up bigger than the DVD [21:26:17 CEST] <JEEBsv> depends on content, preset among other things [21:26:27 CEST] <JEEBsv> I've never tested x265 with lossless so far [21:26:38 CEST] <JEEBsv> it's way too slow for any of my use cases where I use lossless [21:26:45 CEST] <sfan5> kyleogrg: H.265 is supposed(!) to give same quality as H.264 with half the file size [21:27:10 CEST] <kyleogrg> yeah, i know [21:27:26 CEST] <kyleogrg> it's super slow thoguh [21:27:27 CEST] <JEEBsv> that's lossy [21:27:37 CEST] <kyleogrg> yes, it's lossy [21:27:55 CEST] <JEEBsv> also it's a prediction based upon the spec and results from the reference implementation [21:28:07 CEST] <JEEBsv> which got ~33% or whatever, don't remember the exact numbers [21:28:43 CEST] <JEEBsv> the "same quality with half rate" is a marketing thing at this point which uses the fact that the spec is probably good. implementations much less so [21:29:08 CEST] <JEEBsv> I mean, I only got x265 beating x264 with placebo+ a month or so ago [21:30:04 CEST] <ChocolateArmpits> Does x265 look simply better at sub-1mbps bitrates compared to x264 or is it just my eyes? [21:30:38 CEST] <JEEBsv> low rates is where it had been giving its best results for some time now [21:30:53 CEST] <JEEBsv> basically if both looked utter shit, it probably looked a bit less shit [21:32:28 CEST] <ChocolateArmpits> Well the Tears of Steel comparison given here looks much better https://x265.com/hevc-video-files/ [21:32:45 CEST] <ChocolateArmpits> Artifacts are there sure [21:34:10 CEST] <JEEBsv> someone posted some pics for comparison of my first random test encode batch from may, I should do a page like this for the rest of them https://i.fsbn.eu/pub/jeeb-2015/index.htm [21:34:37 CEST] <JEEBsv> I encoded x265 with crf at first since I don't trust its 2pass algo yet [21:34:46 CEST] <JEEBsv> then matched its rate with x264's 2pass [21:36:02 CEST] <JEEBsv> although I'm still worried about random artifacts that x265 seems to push out every now and then [21:36:47 CEST] <ChocolateArmpits> At specific bitrates or in general ? [21:36:58 CEST] <JEEBsv> in general [21:37:05 CEST] <JEEBsv> some are in chroma, some in luma [21:37:42 CEST] <JEEBsv> although with different rate control decisions it might differ [22:07:20 CEST] <bindi> http://ubuntuforums.org/showthread.php?t=2265485 does this still apply? [22:15:31 CEST] <bindi> ugh, now i need 347.09 [23:42:25 CEST] <onefix> Anyone used comskip to transcode videos from a HDHomerun or similar device? [23:43:02 CEST] <onefix> I'm trying to figure out if I should stick with running the Windows version in Wine or migrate to the Linux version... [23:45:29 CEST] <pzich> sounds like something worth testing? [23:48:05 CEST] <onefix> pzich: I'm really trying to figure out if there is a way to tell ComSkip to skip the "coming soon" commercials that include the channel's logo [23:48:41 CEST] <pzich> hmm, I haven't used comskip in forever but that sounds familiar [23:50:52 CEST] <onefix> I have a Linux box that runs TVHeadend, ZoneMinder, Plex, SubSonic , and Domoticz ... the main reason for running TVHeadend is just so I can schedule recordings and run conversions afterwards... [23:52:52 CEST] <onefix> Once TVHeadend is done with it, it passes the recording to a perl script which runs (optionally) ComSkip, then a custom parse routine and FFMPEG with options based on the size of the input and output [23:53:53 CEST] <onefix> If (for instance) I just record the 6:00 news, I want commercial skip deinterlace and downscale to X:480 [23:56:17 CEST] <onefix> It would be nice if I could create a search in TVHeadend that would automatically schedule new series based on a wildcard [23:57:15 CEST] <onefix> The way some series are reported in MC2XML, it doesn't record all episodes of series / mini-series [23:59:41 CEST] <onefix> Also, I think I know the answer, but is there any (reasonable) way to speed up H264 / X264 transcoding using hardware? [00:00:00 CEST] --- Tue Jun 9 2015

1 0

ffmpeg-devel.log.20150607
by burek 08 Jun '15

08 Jun '15

[00:00:41 CEST] <jamrial> nevcairiel: there's a function called interleaveBytes_sse2 that's used when you convert yuv420p -> nv12 [00:01:36 CEST] <nevcairiel> oh well, could still be optimized by doing it on-the-fly while filling the bugger [00:01:39 CEST] <nevcairiel> buffer* [00:02:31 CEST] <BtbN> Hm, using swscale there would mean linking lavc to sws [00:04:45 CEST] <BtbN> philipl, on my gt630, on linux, with 352.09, yuv420p works [00:04:51 CEST] <BtbN> on windows, my compiler broke [00:05:32 CEST] <BtbN> only the cygwin one still works [00:06:18 CEST] <BtbN> interesting that my 40¬ GT630 has a higher SM version (3.5) than my GTX760 (3.0) [00:08:21 CEST] <nevcairiel> they build a few "new" 630s with a newer chip [00:09:04 CEST] <BtbN> Yes, i made sure i have one of those. The original 630 isn't Kepler [00:09:12 CEST] <BtbN> It's the cheapest possible card with nvenc support [00:09:18 CEST] <BtbN> propably around 30¬ now [00:18:09 CEST] <philipl> BtbN: guess its a driver improvement then [00:20:03 CEST] <nevcairiel> so a user dug up a 1519x733 rgb24 video file ... where do these people find such odd things [00:23:14 CEST] <philipl> Selection bias. what other project would even pretent to consider supporting such a format [00:23:20 CEST] <philipl> :-) [00:26:59 CEST] <wm4> sounds easy to handle? [00:27:58 CEST] <nevcairiel> not if you want to encode it into 420 h264 :D [00:28:09 CEST] <nevcairiel> apparently, some part of the chain didnt like that resolution [00:29:02 CEST] <philipl> its certainly not aligned! [00:29:14 CEST] <BtbN> It non-even [00:29:21 CEST] <BtbN> h264 doesn't even support that iirc [00:29:28 CEST] <nevcairiel> sure it does [00:29:31 CEST] <nevcairiel> with cropping metadata [00:29:50 CEST] <BtbN> Non-even resolutions in yuv420 mode? [00:30:21 CEST] <nevcairiel> the encoder might not allow it, but there is no reason the cropping metadata couldnt be set to allow that [00:30:23 CEST] <BtbN> It supports non-power of 8 resolutions that way, but non even? [00:30:40 CEST] <BtbN> How would you even get that converted to yuv420? [00:30:53 CEST] <BtbN> which shared the U/V data for 4 pixels each [00:31:12 CEST] <nevcairiel> that is the more interesting question, thats probably the area of the code that didnt like this file [00:31:16 CEST] <nevcairiel> hope the user shares it [00:32:09 CEST] <nevcairiel> personally i would probably just pretend its 1px wider and higher and add black =p [00:33:33 CEST] <BtbN> Or something would automaticaly insert a scaler and it would ruin the quality by doing a linear scale [00:44:03 CEST] <philipl> To get that as your input format, there's a very good chance it's already ruined. [00:44:31 CEST] <philipl> "I have a sequence of screencap bmps..." [00:56:22 CEST] <michaelni_> durandal_1707, are the other apng patches from donny yang ok to be applied (that is #3 & #4 i think) or they are on hold because of the thread support removial #1 patch ? [01:21:37 CEST] <philipl> BtbN: do you understand how the 2 pass nvenc mode works? I don't see how it can actually run two passes. [01:21:44 CEST] <philipl> the documentation is terrible [01:21:58 CEST] <BtbN> They must have an interesting interpretation of what 2pass means [01:22:08 CEST] <philipl> Quite. [01:22:39 CEST] <BtbN> low-latency hq with 2pass enabled gives the best quality for live streaming though [01:22:54 CEST] <BtbN> No idea how that makes any sense [01:23:07 CEST] <BtbN> I'd expect low-latency to result in worse quality [01:23:15 CEST] <BtbN> and 2pass should not be a thing for live streaming [01:23:36 CEST] <philipl> "3.1.2 Input Buffers Allocated Externally" [01:23:41 CEST] <philipl> right [01:29:12 CEST] <kierank> hardware can run two passes in realtime [01:29:32 CEST] <kierank> similarish to x264 lookahead but it does the actual encoding [01:29:52 CEST] <philipl> Not true 2pass though - it can't evaluate the entire stream [01:30:01 CEST] <BtbN> But it'd have to keep some buffer for that to even work at all [01:30:17 CEST] <BtbN> And low-latency usualy means keeping as little of a buffer as possible [01:31:02 CEST] <kierank> it can encode the frame and encode it again [01:31:15 CEST] <kierank> 2-pass is usually marketing though [01:31:35 CEST] <BtbN> It's some kind of obscure quality booster [01:32:02 CEST] <BtbN> which you just have to turn on, and you get better quality for a lower encode speed, which doesn't matter at all if you are doing live streaming [01:33:00 CEST] <philipl> Strangely, at least for 4k60, llhp is slower than regular hp, or even hq. [01:33:04 CEST] <philipl> Makes it sub-realtime. [01:33:10 CEST] <philipl> Rendering the ll pointless. [01:33:24 CEST] <BtbN> yes, ll is slower and has a better quality than the non-ll profiles [01:33:36 CEST] <BtbN> It makes no sense at all [01:34:02 CEST] <philipl> Ours not to reason why. [01:35:27 CEST] <BtbN> The nvenc encoder in OBS just has an "Automatic" mode, which just selects the best quality mode which is known to work in realtime for the selected resolution/fps [01:35:38 CEST] <BtbN> To safe the users from having to deal with that chaos [01:35:46 CEST] <philipl> A good idea. [01:35:59 CEST] <kierank> damn OBS [01:36:01 CEST] <kierank> stealing my name [01:36:36 CEST] <BtbN> There is also another, secret, mode. [01:36:44 CEST] <philipl> oh yes? [01:37:03 CEST] <BtbN> If you query the nvenc api, how many profiles it supports, it lists two more than there are GUIDs in the public header [01:37:26 CEST] <philipl> but without guids you can't use them? [01:37:30 CEST] <BtbN> yep [01:38:06 CEST] <BtbN> I did some wiretapping on the ShadowPlay app, and when you use it to stream to twitch, it uses a GUID for the encoding preset and profile that are not in the public header [01:38:30 CEST] <philipl> It would make sense that it's shadowplay related. And if you pass those GUIDs? [01:38:30 CEST] <BtbN> But just using that GUID results in an initialization error [01:38:40 CEST] <philipl> heh [01:39:16 CEST] <BtbN> I guess it's NvIFR/FBC related, and a special mode to encode on-gpu [01:39:55 CEST] <philipl> right. [01:42:06 CEST] <BtbN> If you google a bit, you can actualy find api headers and libs for an older version of those [01:42:23 CEST] <BtbN> They came with some old nvidia driver release, that's still mirrored on some sites [01:42:39 CEST] <BtbN> But they don't work anymore [02:54:51 CEST] <cone-046> ffmpeg 03Ganesh Ajjanagadde 07master:42db4aaaa6e4: vf_colormatrix: calculate coefficients only once [02:56:44 CEST] <durandal_1707> michaelni_: Yes because of threads [04:07:54 CEST] <klaxa> i'm currently looking at http listen and chunked encoding and also maybe creating headers in a less static way [04:08:09 CEST] <klaxa> does it even make sense to not use chunked encoding for serving files? [04:08:53 CEST] <klaxa> i don't think i can tell the stream-size while encoding [04:09:15 CEST] <klaxa> or muxing or whatever [04:11:34 CEST] <Compn> what [04:11:47 CEST] <Compn> who is the target for your videos klaxa ? [04:11:58 CEST] <klaxa> an http client, e.g. wget [04:12:08 CEST] <Compn> just you? 3million people? 41 nerds at a lan party? [04:12:34 CEST] <Compn> or do you mean for developing chunked encoding for ffmpeg ? [04:12:39 CEST] <Compn> sorry, i am confused :P [04:13:07 CEST] <klaxa> i am working on http.c in libavformat [04:13:11 CEST] <Compn> oh ok :) [04:13:13 CEST] <Compn> ignore me then [04:13:29 CEST] <klaxa> heh, ok [04:14:26 CEST] <klaxa> for now it only accepts one client, but nicolas george suggested we could write it to support multiple clients, but first i have to clean up the code for a single client and make it work for most usecases [04:14:57 CEST] <klaxa> and a client sending a file without chunked transfer encoding is a pretty valid usecase i think [05:29:46 CEST] <cone-046> ffmpeg 03Philip Langdale 07master:cc904353fa9a: avcodec/allcodecs: Re-order nvenc encoders below x264/5. [06:20:53 CEST] <jamrial> should every demuxer, muxer and protocol remain enabled if i run configure with --disable-avformat? because that's what happens [06:21:56 CEST] <jamrial> same with libavdevice and in/outdevs, and i assume with the other libraries as well [06:22:25 CEST] <jamrial> guess configure is missing some dependency checks? [06:25:48 CEST] <cone-046> ffmpeg 03James Almer 07master:e225f5f23253: fate: add missing avdevice dependency to closed caption test [12:50:52 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:d6039063aaba: ffserver: Check allocations (likely not all) [13:31:18 CEST] <mattlea> Hello, does anyone have any experience with psychoacoustic models? [15:15:34 CEST] <anshul_mahe> is there anything I can help to get my hls webvtt patch in main stream [15:16:25 CEST] <BtbN> Did you send it to the ML? [15:20:39 CEST] <anshul_mahe> yes its on ml from 6 months [15:20:56 CEST] <anshul_mahe> I am tired of merging it with new git [15:23:34 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:dc55477a64ce: avformat/ffmdec: Check ffio_set_buf_size() return value [15:23:35 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:0023ea4e20b0: avformat/aviobuf: Check for ffio_set_buf_size() failure [15:23:36 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:ddda9cee1c4b: ffserver: Check for ffio_set_buf_size() failure [15:24:49 CEST] <BtbN> It propably won't be merged before 2.7 is branched, but i didn't follow the ML on that patch [15:33:36 CEST] <anshul_mahe> yes I dont have any problem even if it get merged later, here is this month mail https://ffmpeg.org/pipermail/ffmpeg-devel/2015-June/173813.html [15:34:38 CEST] <anshul_mahe> but I came here asking to atleast get experts comment to make that patch better [18:03:15 CEST] <ubitux> anshul_mahe: + {"hls_subtitle_path", "set path of hls subtitles", OFFSET(subtitle_filename), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E}, [18:03:26 CEST] <ubitux> can't you pick an input subtitles stream instead? [18:03:32 CEST] <ubitux> that sounds wrong at first sight [18:10:04 CEST] <ubitux> ah that's the output one [19:05:14 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:7c453277a399: avdevice/iec61883: Check pthread init for failures [20:21:41 CEST] <cone-151> ffmpeg 03Clément BSsch 07master:40cc3be73c2c: avfilter: fix a few 5 spaces indent [20:26:48 CEST] <ubitux> i'm writing a filter that will take up to 4 * 9 settings, or 4 adjustments for 9 different region/colorranges [20:27:04 CEST] <ubitux> you will generally want to adjust the 4 settings for one or 2 regions [20:27:20 CEST] <ubitux> what would ppl prefer as interface? [20:27:54 CEST] <ubitux> 4*9 options, requiring to fill with 0 all the unaffected regions (or explicit 4 or 8 (or more) keys) [20:28:34 CEST] <ubitux> or having the region as option key, and a quad of floats? (region1=0.1|0.6|0.3|0.4:region2=...) [20:29:42 CEST] <ubitux> (first case being expliciting region[0..8]_setting[0..3] option -- note: no digit in the final names, they are more meaningful/uniques) [20:30:31 CEST] <ubitux> i personally prefer the second solution (because more convenient for the user -- cli at least) [20:31:13 CEST] <ubitux> which will require to just sscanf "%f|%f|%f|%f" on every region option [20:31:36 CEST] <ubitux> ...which actually brings me to the question: shouldn't we add array-like options in AVOption? [20:33:52 CEST] <beastd> ubitux: good question. such an avoption extension would probably be useful in more instances. [20:34:00 CEST] <wm4> there are already array avoptions, and unfortunately I even had to use one of them [20:35:02 CEST] <beastd> i remember sth about arrays in avoption, IIRC it came up in lavd discussions. but i may be mistaken [20:35:19 CEST] <ubitux> ah? damn i time wrapped or something, seems i missed a few changes [20:35:34 CEST] <ubitux> what's the type name? [20:36:00 CEST] <ubitux> ah, avoptionrange mmh [20:39:31 CEST] <kierank> wm4: https://ffmpeg.org/pipermail/ffmpeg-user/2015-May/026824.html [20:39:33 CEST] <kierank> lol [20:39:34 CEST] <wm4> ah I think I removed my array usage, and replaced it with something simpler and Libav compatible [20:39:54 CEST] <kierank> stream-ids aren't based on pmt's they are based on when the data appears in the stream... [20:40:00 CEST] <wm4> kierank: wait, everything is detecred as mp3? [20:40:10 CEST] <kierank> no but the stream order isn't based on what the ts says [20:40:15 CEST] <kierank> it's based on stupid probing as usual [20:40:26 CEST] <wm4> ah [20:41:05 CEST] <wm4> just today I was wondering how to stop libavformat from detecting ELF files as mp3 [20:41:09 CEST] <wm4> probing sure is a mess [20:43:12 CEST] <kierank> poor user [20:44:50 CEST] <JEEBsv> every now and then I decide to see if I can care enough to make lavf's mpeg-ts demuxer support PID switching. But then I always lose my will to poke it after opening the file [20:45:41 CEST] <wm4> I remember how the mpeg-ts demuxer had issues with h264 timestamps, and the bug was actually in obscure utils.c code [20:48:08 CEST] <nevcairiel> why would someone rely on stream order anyway o.o [20:50:16 CEST] <wm4> there's no real way to identify a stream, other than the ffmpeg index [20:50:19 CEST] <wm4> which is annoying [20:50:27 CEST] <nevcairiel> in mpegts you also get the pid [20:50:31 CEST] <wm4> for some formats, like mpeg-ts, you could of course special-case it, and get the real id [20:50:49 CEST] <wm4> but for other formats, the field may just be 0 (for all streams!) or something random [20:50:52 CEST] <nevcairiel> and formats with actual headers have their order based on that [20:50:58 CEST] <nevcairiel> its just mpegts which doesnt necessarily [20:56:11 CEST] <kierank> mpegts has order in pmt [20:56:15 CEST] <kierank> that's why that user isn't happy [20:56:21 CEST] <kierank> because ffmpeg produces a random order each time [21:11:13 CEST] <durandal_1707> ubitux: what filter does? [21:11:23 CEST] <ubitux> PS selective color [21:12:15 CEST] <ubitux> http://dsp1i8etdrigy.cloudfront.net/wp-content/uploads/2011/08/4-selective-… [21:24:39 CEST] <anoop_r> while encoding i get Past duration 0.619041 too large why? [21:26:08 CEST] <anoop_r> onybedy [21:26:13 CEST] <anoop_r> anybody [21:26:14 CEST] <anoop_r> help me [21:28:18 CEST] <nevcairiel> you need to ask in #ffmpeg, you ahve been told several times that this is the wrong channel [21:31:57 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:4c4c3d5d5abb: avcodec/libstagefright: Check for pthread_create() failure [21:31:57 CEST] <anoop_r> no body answers [21:32:04 CEST] <nevcairiel> then wait [21:32:15 CEST] <nevcairiel> that doesnt give you the right to ask here, bencause you wont get answers here either [21:32:48 CEST] <anoop_r> but its an ffmpeg bug .x265 also confiremed it [21:33:21 CEST] <JEEBsv> no, you were just told that it's ffmpeg pushing out the message [21:33:54 CEST] <anoop_r> why its logging if its not a bug [22:09:39 CEST] <cone-151> ffmpeg 03Michael Niedermayer 07master:0dbea4642f92: avformat/rmenc: Remove float usage [00:00:00 CEST] --- Mon Jun 8 2015

1 0

ffmpeg.log.20150607
by burek 08 Jun '15

08 Jun '15

[00:00:06 CEST] <JEEBsv> you could probably go with one of the fastest presets even [00:00:16 CEST] <rafael2k> but that could cope encode in realtime [00:01:10 CEST] <JEEBsv> basically you could cut some of that kind of content and see how well/quickly it compresses with libx264's -q:v 0 -preset veryfast -c:v libx264 or so [00:01:25 CEST] <rafael2k> I'll do that [00:01:42 CEST] <JEEBsv> and -pix_fmt yuv422p10 if it tries to do 4:2:0 by default [00:01:42 CEST] <rafael2k> I bought a Dektec ISDB-T modulator for me to play [00:01:44 CEST] <rafael2k> ; ) [00:02:31 CEST] <rafael2k> I think I'll convert to 4:2:0 for my tests, less cpu intensive [00:03:59 CEST] <JEEBsv> do you need it to be intra-only, short gops or you can just leave the default gop length on (or even use the mode where libx264 only puts intra pictures in when that's the best option) [00:04:20 CEST] <rafael2k> I'd like to have low latency [00:04:33 CEST] <JEEBsv> that's separate [00:04:35 CEST] <rafael2k> no temporal compression at all [00:05:10 CEST] <JEEBsv> short GOPs or intra only mean that clients can connect right away if they didn't have a connection already and thus get all of the packets [00:05:23 CEST] <JEEBsv> but yeah, if you need intra only then -g 1 or so [00:05:45 CEST] <JEEBsv> latency is separate [00:05:52 CEST] <rafael2k> good [00:06:10 CEST] <JEEBsv> of course I will have to say that you will have to also optimize ffmpeg itself for low latency [00:06:23 CEST] <JEEBsv> I can tell you to use -tune zerolatency for the least latency options in x264 [00:06:35 CEST] <JEEBsv> but that will not make the rest of ffmpeg possibly optimal for it already [00:06:38 CEST] <JEEBsv> just keep that in mind [00:06:48 CEST] <rafael2k> I'll use it, tks! I'm thinking about using less protocols in the networkk stack also, write the data redirect to ethernet frames [00:07:06 CEST] <JEEBsv> and I just don't remember of all the caching etc. stuff ffmpeg does to the input [00:07:16 CEST] <JEEBsv> if you want least amount of latency you'll have to minimize that :P [00:07:37 CEST] <rafael2k> forget, I'll use x264 directlt [00:07:39 CEST] <JEEBsv> libx264 itself makes it easy (the zerolatency tuning), but you're using it within ffmpeg [00:07:39 CEST] <rafael2k> directly [00:07:59 CEST] <JEEBsv> well I don't exactly know how the x264 command line tool does it either [00:08:21 CEST] <JEEBsv> just don't go about yelling that libx264 can't do low latency since that's been made rather simple with the library ;) [00:08:32 CEST] <rafael2k> I remember when I worked with DTV, it was pretty small the latency [00:08:49 CEST] <JEEBsv> yeah, libx264 itself can do really low latency stuff [00:09:05 CEST] <rafael2k> let me check [00:09:07 CEST] <JEEBsv> I mean, I have played games nicely with it :D [00:09:22 CEST] <rafael2k> serious? goooood [00:09:29 CEST] <JEEBsv> yes [00:09:36 CEST] <JEEBsv> the library has no issues with latency [00:10:02 CEST] <JEEBsv> if you have issues with latency when using libx264 in something it is either the rest of the encoding process somewhere else, or on the player side [00:10:21 CEST] <JEEBsv> which is why I kept saying to you that you would have to optimize whatever you use libx264 in for low latency as well [00:10:37 CEST] <JEEBsv> libx264 itself does it as soon as you set the zerolatency tuning [00:13:58 CEST] <DrBotato> dumb question. if i stretch a video from 720p to 1080p with ffmpeg and then play it back at 720p. will it look worse than just playing it at 720p? [00:14:06 CEST] <DrBotato> without the pre-stretching [00:17:16 CEST] <atomnuker> DrBotato: nope, not unless you rescale the video 2x original size using nearest neighbour and then scale it down the same way [00:18:12 CEST] <DrBotato> nope = it will be worse? nope = it won't be worse. [00:18:19 CEST] <DrBotato> i'm confused by your double negative [00:19:32 CEST] <atomnuker> 1280x2 != 1920 so nope, it will 'look' worse, detail will be lost [00:20:31 CEST] <DrBotato> ah ok i see what you mean. [00:20:40 CEST] <DrBotato> thanks [07:24:47 CEST] <chama> Hi, is there anyway that i can concat a video with a image [07:24:59 CEST] <chama> video + (image for 2 seconds) [17:33:40 CEST] <amoz_> hello. I am newbie. I have gone through ffmpeg.org but got completely lost there. If i want to contribute then, how shall i get started ? [17:34:28 CEST] <c_14> Fixing bug tickets on trac is always a good start. [17:36:02 CEST] <amoz_> c_14: thank you for responding. I referred the documentation. but i am still confused. has anyone blogged about it in a simpler way :) [17:37:13 CEST] <beastd> amoz_: heh. there is lots of blogged information of ffmpeg out there. some is spot on/some outdated/some plain wrong. just use google and you will find lots of hits about ffmpeg. [17:37:34 CEST] <beastd> amoz_: You have specific reason you want to contribute? [17:37:59 CEST] <amoz_> beastd: yes. planning for gsoc 2016. [17:38:32 CEST] <amoz_> actually I am a voluntary contributor to the open source projects of Mozilla community. [17:38:56 CEST] <beastd> amoz_: also there is an ffmpeg wiki (you can find it easily from ffmpeg homepage) [17:39:20 CEST] <amoz_> beastd: ohhkk...I will do an exhaustive reading [17:40:38 CEST] <amoz_> thank you c_14 and beastd [17:42:14 CEST] <incal> when I concat files with 'ffmpeg -f concat -i $videos -c copy $new' the second part gets a (somewhat) different color tone even tho the original videos look identical in their original files - themselves setup like this: ffmpeg -f concat -i $videos -c copy \ [17:42:14 CEST] <incal> -c:v libx264 -tune stillimage \ [17:42:14 CEST] <incal> -c:a aac -strict experimental -b:a 192k \ [17:42:14 CEST] <incal> -pix_fmt yuv420p -vf scale=1280:720 \ [17:42:17 CEST] <incal> -shortest $new [17:44:46 CEST] <beastd> amoz_: Should you want to proceed at some point: Read the homepage. Clone the source code. Try to build ffmpeg current git master on your machine. If that's done, study the documentation and other information sources. Try things you are interested in and try finding the corresponding source code in ffmpeg code base. That is what I would recommend to get familar with the project. [17:45:43 CEST] <amoz_> beastd: ohhkk....seems a bit lengthy procedure though :P [17:45:46 CEST] <beastd> amoz_: reading/subscribing to ffmpeg-user (and maybe other mailing lists) can also help in finding out things [17:46:08 CEST] <amoz_> yes. definitely. [19:26:42 CEST] <anoop_r> <anoop_r> hi i got error Past duration 0.904747 too large [19:26:42 CEST] <anoop_r> <anoop_r> while using x265 to encode [19:26:42 CEST] <anoop_r> <anoop_r> x265 10bit with ffmpeg [19:26:50 CEST] <anoop_r> please see this http://pastebin.com/9L60nhYi [19:29:31 CEST] <Darby_Crash> hi to all [19:30:01 CEST] <Darby_Crash> i have a question about ffserver [19:30:27 CEST] <Darby_Crash> i want re-stream a HLS live channel [19:31:43 CEST] <Darby_Crash> and i want set ffserver.conf for restream the channel with same codec and quality for video and audio [19:33:29 CEST] <Darby_Crash> how can i set ffserver.conf? [19:33:50 CEST] <Darby_Crash> i have read about documentation [19:34:29 CEST] <Darby_Crash> but i haven't found nothing about this [19:36:19 CEST] <Darby_Crash> someone can help me? [19:54:08 CEST] <Darby_Crash> i have solved [19:54:22 CEST] <Darby_Crash> but i have another question [19:54:51 CEST] <Darby_Crash> my stream have a problem with buffering [19:57:24 CEST] <Darby_Crash> http://clbin.com/MyhOe [19:57:56 CEST] <Darby_Crash> this is my ffserver.conf [19:58:40 CEST] <Darby_Crash> how can resolve for buffer? [21:26:43 CEST] <animax> hi all. can I encode a batch of video files in one single process? settings for libx264 shall all be the same for each input file. the files are all contained in the same folder nothing else there in it. afk for a couple minutes ... [21:27:45 CEST] <c_14> mkdir conv; for video in *; do ffmpeg -i $video [stuff] conv/$video; done [21:34:48 CEST] <anoop_r> while encoding i get Past duration 0.619041 too large why? [21:34:58 CEST] <anoop_r> why ffmpeg is messaging like that [21:35:14 CEST] <anoop_r> if its not a bug then why it logging that [21:37:23 CEST] <animax> c_14: I'm on windows ... (because of the '$') [21:37:50 CEST] <c_14> animax: it's still possible. you'll need batch script for it though (or a scripting language) [21:38:25 CEST] <animax> c_14: really? no command matching my purpose? [21:39:05 CEST] <animax> never made a batch script ... [21:39:11 CEST] <c_14> Well, you _could_ built one giant monster of an ffmpeg command. But batch script'll be easier. [21:39:20 CEST] <animax> where can I find instructions? [21:42:59 CEST] <animax> ok. so google is your friend ... [21:47:42 CEST] <Darby_Crash> hi c_14 can you help me? [21:48:31 CEST] <c_14> With what? [21:48:54 CEST] <Darby_Crash> i must re-stream a hls stream with ffserver [21:49:31 CEST] <Darby_Crash> if is possible with same video and audio [21:49:47 CEST] <Darby_Crash> this is my ffserver.conf [21:49:49 CEST] <Darby_Crash> http://clbin.com/MyhOe [21:50:12 CEST] <c_14> Unless you can use codec copy, you'll just have to pick something. [21:51:12 CEST] <Darby_Crash> i have re encoded it with x256 and liblame [21:51:53 CEST] <Darby_Crash> because i want more high possible quality [21:52:09 CEST] <BtbN> you can't enhance the quality, you can only keep it or make it worse [21:53:22 CEST] <Darby_Crash> with this conf i have a good quality but the stream have buffer problems [21:54:20 CEST] <BtbN> Not too surprising with x265, propably overloads your cpu [21:55:17 CEST] <Darby_Crash> on the server the stream don't have problem [21:55:57 CEST] <Darby_Crash> when i open the stream with a client like vlc on another machine it have many stop and go [21:57:21 CEST] <c_14> Then the problem is probably network bandwidth. [21:58:15 CEST] <Darby_Crash> you can say me another way to have the same quality of the source? [21:58:53 CEST] <Darby_Crash> how can i change my ffserver.conf? [22:02:16 CEST] <Darby_Crash> i wish a good quality [22:04:44 CEST] <Darby_Crash> if not the same of the source, a middle [22:23:16 CEST] <Darby_Crash> ok [22:23:26 CEST] <Darby_Crash> i have a solution [22:24:02 CEST] <Darby_Crash> c_14, can you help me a little more? [22:24:16 CEST] <c_14> With what exactly? [22:25:36 CEST] <Darby_Crash> https://clbin.com/MNNaT [22:25:49 CEST] <Darby_Crash> this is my new ffserver.conf [22:26:59 CEST] <Darby_Crash> i run ffserver [22:27:22 CEST] <Darby_Crash> after run ffmpeg with the feed in this way [22:30:03 CEST] <Darby_Crash> livestreamer "hlsvariant://http://manifest.m3u8" best --stdout | ffmpeg -i pipe:0 -vcodec copy -acodec copy -f mpegts -map 0:0 -map 0:1 http://127.0.0.1:8090/live.ffm [22:30:09 CEST] <Darby_Crash> c_14, [22:31:55 CEST] <c_14> you'll probably want -override_ffserver [22:33:45 CEST] <Darby_Crash> https://clbin.com/rPFcm [22:34:14 CEST] <Darby_Crash> this is the output c_14 [22:37:13 CEST] <c_14> do you get that error when outputting to file? [22:37:15 CEST] <Darby_Crash> i use livestreamer only to force the stream to best quality [22:38:47 CEST] <Darby_Crash> c_14, here you can read output https://clbin.com/rPFcm [22:39:24 CEST] <c_14> change the http://127.0.0.1:8090/live.ffm to out.mpegts and see if you get the same error [22:40:04 CEST] <Darby_Crash> ok [22:41:22 CEST] <Darby_Crash> work fine [22:41:31 CEST] <Darby_Crash> what is the problem? [22:41:38 CEST] <Darby_Crash> c_14, [22:44:35 CEST] <c_14> I don't know. The command/config look fine to me. [22:54:37 CEST] <Darby_Crash> av_interleaved_write_frame(): Broken pipe [22:56:25 CEST] Last message repeated 1 time(s). [22:56:25 CEST] <Darby_Crash> frame= 2 fps=0.0 q=-1.0 Lsize= 1kB time=00:00:00.08 bitrate= 105.8kbits/s [22:56:25 CEST] <Darby_Crash> video:0kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 422.222229% [22:56:25 CEST] <Darby_Crash> Conversion failed! [22:56:52 CEST] <Darby_Crash> c_14, here there's the problem [23:58:42 CEST] <kodur> could anyone help me out on rotating a movie and converting it to webm? [00:00:00 CEST] --- Mon Jun 8 2015

1 0

ffmpeg-devel.log.20150606
by burek 07 Jun '15

07 Jun '15

[01:12:34 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:51080dfa0843: avcodec/s302m: Check for non PCM Streams [01:12:35 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:42c41e96ff6d: avformat/utils: Do not select audio streams with unknown sample rate in av_find_best_stream() [01:12:36 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:802cca5905ab: avcodec/s302m: Only set the sample rate when some data is output [02:50:20 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:f07376402cdd: avformat/mxfenc: Add mxf muxer avclass [10:08:23 CEST] <ubitux> michaelni: i prefer our version from 81a91269a2fbeb87de53e395c6bed0c70c310cae [10:08:38 CEST] <ubitux> assuming it doesn't lack anything, i'd prefer if we keep ours when you'll merge the one from luca [10:09:40 CEST] <ubitux> you might want to take the doxy part [12:14:52 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:be1862ffb06a: doc/filters: fix inaccuarte/inaccurate typo [12:17:15 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:c5a08956a3b1: swresample: fix initilaize/initialize typo [12:50:16 CEST] <kierank> can someone pm me the incoming read password [12:51:43 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:aa74401af8a4: lavf/http: Document method option. [12:51:44 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:bbcee92b6d59: lavf/http: Process HTTP header before sending response. [12:51:45 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:8cfaa76a5e77: lavf/http: Rudimentary error handling for HTTP requests received from clients. [12:51:46 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:a7e7c68b0e26: lavf/http: Properly process HTTP header on listen. [12:51:47 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:290b23755673: lavf/http: Indent else-clause. [12:51:48 CEST] <cone-604> ffmpeg 03Stephan Holljes 07master:44d19212008e: lavf/http: Add simple autodetection for client HTTP method, based on AVIO_FLAG_READ. [12:51:49 CEST] <cone-604> ffmpeg 03Michael Niedermayer 07master:7897ed0d3cb9: Merge remote-tracking branch 'cigaes/master' [12:53:42 CEST] <kierank> cehoyos: ping, you definitely know [13:16:31 CEST] <cone-604> ffmpeg 03Ronald S. Bultje 07master:ade5684cda9c: hevc: fix typo (mpv -> mvp). [13:17:43 CEST] <kierank> lol [13:18:37 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:0f6118c58169: ffplay: use av_clip() instead of nested min & max [13:18:38 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:622ef80e3f5f: avcodec/mpegvideo: use av_clip() instead of nested min & max [13:18:39 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:533c37db8ddb: avcodec/snowenc: use av_clip() instead of nested min & max [13:18:40 CEST] <cone-604> ffmpeg 03Clément BSsch 07master:272f87fc5ca9: avdevice/x11grab: use av_clip() instead of nested min & max [13:26:14 CEST] <cehoyos> kierank: ? [13:26:19 CEST] <kierank> never mind [13:26:23 CEST] <kierank> 11:50 AM <"kierank> can someone pm me the incoming read password [13:26:38 CEST] <cehoyos> Sorry, missed it... [13:26:43 CEST] <cehoyos> Which sample are you interested in? [13:26:49 CEST] <kierank> elemental one [13:26:52 CEST] <kierank> but it's corrupt still [13:27:06 CEST] <cehoyos> Yes. [13:33:39 CEST] <cehoyos> michaelni: https://trac.videolan.org/vlc/ticket/14811 J-B thinks this is a regression in FFmpeg, I cannot test vlc atm, not reproducible with ffmpeg. [13:48:54 CEST] <michaelni_> cant reprodue with the vlc ive here [13:49:13 CEST] <michaelni_> but thats some random git version not 2.2.1 [13:56:11 CEST] <BtbN> Are the RGB to YUV functions this guy on the ml is trying to improve even used anywhere? [13:56:47 CEST] <kierank> michaelni_: i reported that bug to you [13:56:52 CEST] <kierank> and a fix [13:56:57 CEST] <kierank> it has been fixed iirc [13:59:18 CEST] Action: michaelni_ doesnt remember exactly but if it has been fixed thats good, anything i should backport ? [14:02:44 CEST] <kierank> https://trac.ffmpeg.org/ticket/4533 [14:02:48 CEST] <kierank> michaelni_: it's that [14:06:46 CEST] <michaelni_> ok, this only affects 2.6 and fix is already in 2.6.3, vlc 2.2.1 uses a fixed hash based checkout from master it seems [14:08:29 CEST] <wm4> "ffmpeg -i somefilewithaac.mkv -c:a copy -f spdif out.raw" fails with "Wrong AAC file format" [14:08:53 CEST] <wm4> so apparently ffmpeg's aac parser uses a different format than spdifenc expects? [14:09:48 CEST] <nevcairiel> i never tried to use aac over spdif, what devices accept that? [14:09:49 CEST] <wm4> so if you use ffmpeg's aac encoder [14:10:05 CEST] <wm4> nevcairiel: I have no idea, and I was curious to see what my receiver does with it [14:10:23 CEST] <nevcairiel> it needs ADTS, apparently [14:10:36 CEST] <nevcairiel> thats what the code wants, at least [14:12:02 CEST] <wm4> there's only a aac_adtstoasc BSF [14:12:21 CEST] <kierank> use the adts mux [14:12:21 CEST] <nevcairiel> there is a adts muxer [14:12:31 CEST] <wm4> can ffmpeg chain that? [14:12:34 CEST] <kierank> no [14:12:47 CEST] <kierank> ffmpeg cannot chain anything afaik [14:12:51 CEST] <kierank> apart from some tricks in rtp [14:13:01 CEST] <nevcairiel> some formats do that automatically, ie. try to mux raw aac into mpegts, and it'll automatically use the adts muxer to convert it [14:13:09 CEST] <nevcairiel> but there is no generic chaining support [14:13:52 CEST] <rcombs> does segment count [14:14:40 CEST] <wm4> ok let's try what the receiver does with it [14:15:06 CEST] <nevcairiel> i suppose some may support it, considering that there is a spec for it [14:15:19 CEST] <rcombs> and ff_write_chained is a thing [14:15:22 CEST] <rcombs> of some kind [14:15:32 CEST] <rcombs> (but it's not what mpegtsenc uses) [14:15:51 CEST] <nevcairiel> mpegtsenc does it manually, since it needs the output itself [14:16:04 CEST] <nevcairiel> write_chained works for things like segment, where you dont need the output [14:16:16 CEST] <rcombs> not sure which is the case for spider [14:16:19 CEST] <rcombs> *spdif [14:16:30 CEST] <nevcairiel> it would need th mpegts way [14:16:40 CEST] <rcombs> ~yay~ [14:16:57 CEST] <nevcairiel> its really not that bad [14:17:07 CEST] <nevcairiel> just no "official" API to do that easily [14:17:28 CEST] <rcombs> wonder if it'd be worth having if we wanted it for spdif [14:17:46 CEST] <rcombs> (except as far as I'm aware nobody cares about AAC in SPDIF so&) [14:17:51 CEST] <nevcairiel> yeah [14:17:57 CEST] <nevcairiel> i dont think i was ever asked [14:18:07 CEST] <nevcairiel> i think more people asked me about DSD over HDMI than AAC [14:18:10 CEST] <nevcairiel> so ... [14:18:14 CEST] <rcombs> wm4: something sounding rather unpleasant? [14:18:26 CEST] <rcombs> nevcairiel: how even [14:18:31 CEST] <rcombs> (same way SPDIF works?) [14:18:54 CEST] <nevcairiel> how wah [14:18:54 CEST] <wm4> the receiver is just sitting there and pretends not to receive any data [14:19:00 CEST] <cehoyos> nevcairiel, kierank: Please elaborate, I am tired and drunk and I would really, really like to understand! [14:19:18 CEST] <nevcairiel> HDMIs compressed format support works the same way as SPDIF, ie. the same transmission protocol is used [14:19:29 CEST] <nevcairiel> HDMI just supports a bunch higher-bandwidth modes for HD audio formats [14:19:31 CEST] <cehoyos> Could you explain the sample you just uploaded? [14:19:46 CEST] <nevcairiel> its the sample the guy tried to upload to the ftp [14:19:53 CEST] <nevcairiel> well, shortended a bit to get below 2.5mb [14:19:56 CEST] <cehoyos> Why do you have it? [14:20:02 CEST] <cehoyos> (Sorry, I am really just tired.) [14:20:12 CEST] <nevcairiel> Because I directed him to report it to ffmpeg =p [14:20:18 CEST] <cehoyos> Ok. [14:20:19 CEST] <cehoyos> Sorry. [14:20:25 CEST] <nevcairiel> I send all the people here when they report problems to me xD [14:21:00 CEST] <nevcairiel> original sample is also here: http://www.mediafire.com/watch/s76856vq2jnx7i5/test_bref.mp4 [14:21:17 CEST] <nevcairiel> the one i uploaded cuts away audio and 5 seconds of video or so [14:21:25 CEST] <nevcairiel> but the issue is clearly visible anyway [14:21:36 CEST] <cehoyos> It is a regression. [14:21:52 CEST] <cehoyos> And your sample is of course much "better" than a longer one. [14:21:55 CEST] <nevcairiel> not sure, i didnt bother to check older versions [14:21:55 CEST] <cehoyos> (At least imo) [14:22:02 CEST] <cehoyos> I did. [14:22:05 CEST] <nevcairiel> I only checked other decoders, and those work [14:26:24 CEST] <nevcairiel> in unrelated news, android's exoplayer supports matroska files with h264+opus now, and it got a wrapper for a libopus decoder when no opus hardware support is present, i should really try to use opus now instead of the native aac encoder :d [14:26:48 CEST] <wm4> sounds pretty nice [14:26:59 CEST] <wm4> does that mean opus is allowed in webm now? (not that this means much) [14:27:09 CEST] <nevcairiel> thats been allowed a while now [14:27:16 CEST] <nevcairiel> vp8/vp9 + vorbis/opus [14:27:55 CEST] <wm4> apropos webm, did they specify how alpha transparency works yet? (for ages there have been 3 different ways) [14:28:05 CEST] <nevcairiel> until recently, the exoplayer webm reader didnt allow generic matroska files though, just a couple days ago it learned how to handle matroska with h264 and aac/ac3 audio .. and of course vorbis/opus, since those were supported for webm [14:28:38 CEST] <nevcairiel> matroska is a odd format for streaming though, but it may just work! [14:35:05 CEST] <cone-604> ffmpeg 03Michael Niedermayer 07master:440fa7758b68: Changelog: Drop duplicate NVENC entry [14:43:06 CEST] <nevcairiel> that is one old regression [14:44:05 CEST] <cehoyos> I was actually one commit wrong, the one I mention is missing the asm file... [14:48:11 CEST] <cehoyos> Darwin award: http://tirol.orf.at/news/stories/2714860/ (German) [15:53:16 CEST] <wm4> "If you want to help, upload a sample of this file to ftp://upload.ffmpeg.org/incoming/ and contact the ffmpeg-devel mailing list. (ffmpeg-devel(a)ffmpeg.org)" [15:53:26 CEST] <wm4> except this file is on sampels.ffmpeg.org [15:53:29 CEST] <wm4> *samples [15:55:07 CEST] <cehoyos> ftp upload.ffmpeg.org works fine here... [15:55:32 CEST] <cehoyos> You cannot upload files to samples.ffmpeg.org [15:55:34 CEST] <cehoyos> ? [15:55:39 CEST] <wm4> who said that [15:57:16 CEST] <cehoyos> You can connect to samples.ffmpeg.org (which may be a bug) but you are not allowed to upload there via anonymous ftp. [15:59:41 CEST] <wm4> maybe you should read again what I wrote [16:00:07 CEST] <wm4> it prints this warning for a file which is already somewhere on samples.ffmpeg.org [16:00:13 CEST] <wm4> so what's the point of that warning [16:00:24 CEST] <cehoyos> I fail to understand you, sorry. [16:00:30 CEST] <cehoyos> Which file are you talking about? [16:00:43 CEST] <cehoyos> And I agree that if we request samples that we don't need, it is a bug [16:01:11 CEST] <wm4> http://samples.ffmpeg.org/evob/Subtitle-sample.evo [16:01:13 CEST] <cehoyos> Bug as you certainly know, the only solution that makes sense is to fix the bug by submitting a patch that removes the request for the given case. [16:03:28 CEST] <cehoyos> Ranting here (and as you know, I also like rants from time to time) makes no sense in this case. [16:12:12 CEST] <Compn> wm4 : we need more samples, unless you think we have enough ? [16:13:16 CEST] <cehoyos> wm4: I will commit later, gtg, sorry. [16:13:34 CEST] <Compn> obviously just not that sample that we already ahve ;p [17:08:41 CEST] <cone-604> ffmpeg 03James Almer 07master:1b4dd59e5fbd: configure: improve the checks for gmtime_r and localtime_r [18:03:05 CEST] <cone-604> ffmpeg 03Paul B Mahol 07master:a03b69478b7f: avcodec/exr: fix crash caused by merge [19:12:55 CEST] <wm4> ah, so we have nvenc_a.c and nvenc_b.c [19:12:58 CEST] <wm4> just so absurd [19:14:41 CEST] <jamrial> and nvenc_b is based on nvenc_a, but with some differences and a new name [19:14:52 CEST] <cone-604> ffmpeg 03Michael Niedermayer 07master:7944b0ce94bc: avutil/colorspace: Remove RGB_TO_Y/U/V [19:21:08 CEST] <philipl> jamrial: but without proper credit, of course. [19:23:32 CEST] <jamrial> no, they did give credit, at least. their commit says it's "Partially based on the work of Timo Rothenpieler" [19:23:44 CEST] <philipl> But not in the file. [19:24:10 CEST] <philipl> So the copyright statement is dubious at best. [19:24:22 CEST] <nevcairiel> wm4: send a patch to the ML to remove the new one and rename the original back, i would support it [19:25:10 CEST] <philipl> Michael asked Timo what he wanted to do and he said drop theirs and see what might be beneficial to backport. [19:25:17 CEST] <wm4> would result in a useless flame [19:25:48 CEST] <philipl> who would flame? I don't think anyone is arguing for it to stay. [19:26:11 CEST] <wm4> what nevcairiel suggested [19:26:32 CEST] <nevcairiel> not sure anyone would have a reason to flame [19:26:40 CEST] <philipl> That's what I mean. [19:26:54 CEST] <nevcairiel> except maybe michael, since it may make future merges harder if libav works on theirs =p [19:27:00 CEST] <philipl> No one wants this around. The only question is what's the best way to deal with it in the long term. [19:27:11 CEST] <philipl> Michael merged it for that reason but he's not wedded to it. [19:28:03 CEST] <philipl> Has libav ever merged anything back properly? As opposed to gratuitous rewrites? [19:28:20 CEST] Action: michaelni_ iam fine with droping the rewritten nvenc [19:28:39 CEST] <jamrial> some things [19:28:57 CEST] <philipl> michaelni_: do you think we should alias the codec names before we release? [19:29:01 CEST] <jamrial> philipl: we could drop theirs and rename ours (so they are called the same in both projects), but there's the issue of clashing avoptions [19:29:12 CEST] <philipl> That's the main issue, yes. [19:29:22 CEST] <wm4> I don't consider it an issue at all [19:29:29 CEST] <wm4> not for something as obscure as nvenc [19:29:31 CEST] <michaelni_> philipl, up to you and BtbN [19:29:33 CEST] <philipl> Well, not rename - we have our own compatibility to think about. [19:29:54 CEST] <philipl> BtbN: around? [19:30:15 CEST] <jamrial> ah right, nvenc h264 is already part of a release [19:30:43 CEST] <jamrial> but not h265 [19:30:46 CEST] <philipl> So, I'd propose the following: 1) Revert the merge, that gets us back to where we started. 2) Add their codec names as aliases to ours. 3) Ignore option compatibility for now. BtbN will look at it at some point. [19:30:51 CEST] <philipl> 265 was released, I thought. [19:31:19 CEST] <jamrial> shows up in Changelog as part of version next [19:31:25 CEST] <philipl> Ok. [19:31:27 CEST] <BtbN> yeah, i'm around. 265 would be in 2.7, not in a relase yet [19:31:33 CEST] <philipl> Well, I'm ok with renaming. [19:31:49 CEST] <philipl> I called in nvenc_h265, but it seems the whole world has gone with hevc, so just as well. [19:32:05 CEST] <nevcairiel> makes for a cleaner distinction with h264 [19:32:50 CEST] <michaelni_> its h261, h262, h263, h264, hevc, i dont know why, but that is indeed what people prefer [19:33:06 CEST] <nevcairiel> i dunno, people also use mpeg2 more than h262 [19:33:21 CEST] <nevcairiel> or mpeg4 for h263 [19:33:22 CEST] <BtbN> Propably whatever gets used more first wind [19:33:26 CEST] <BtbN> *s [19:33:28 CEST] <philipl> mpegh forever! [19:33:55 CEST] <michaelni_> i am not 100% sure about mpeg2 but mpeg4 ASP (which people use) is not identical to h263 [19:34:06 CEST] <philipl> it's a superset. [19:34:29 CEST] <philipl> Anyway, BtbN: is that plan ok with you? [19:34:44 CEST] <BtbN> Dropping the libav nvenc encoder and adding aliased names to our one, and then continouing to port usefull features? [19:34:51 CEST] <philipl> yes. [19:34:56 CEST] <BtbN> Yeah, sounds good [19:34:57 CEST] <philipl> although for hevc, I'll just rename ours. [19:35:02 CEST] <philipl> so one alias. [19:35:16 CEST] <BtbN> Yes, as it's not released yet, that might be better [19:35:16 CEST] <philipl> Is there a cheap way to do aliases or does it need another codec declaration? [19:35:20 CEST] <nevcairiel> our hevc and h264 names should be consistent [19:35:24 CEST] <nevcairiel> whats their name vs ours anyway? [19:35:42 CEST] <philipl> Ours today are nvenc and nvenc_h265. Theirs are nvenc_h264 and nvenc_hevc. [19:35:50 CEST] <nevcairiel> ah [19:35:57 CEST] <nevcairiel> i guess having the codec in the h264 as well is a bit cleaner [19:36:22 CEST] <philipl> I'll have a patch series up shortly. [19:36:35 CEST] <BtbN> When i started with this, there was no support for any other codec than h264 in nvenc, so even on short term, it's kind of a historic reason [19:36:39 CEST] <nevcairiel> unfortunately i dont think t here is a good way to do aliases [19:36:42 CEST] <nevcairiel> its probably going to be ugly [19:36:44 CEST] <jamrial> philipl: theirs are h264_nvenc and hevc_nvenc [19:36:56 CEST] <philipl> they are? [19:37:06 CEST] <nevcairiel> thats terrible names then [19:37:13 CEST] <philipl> jamrial: other way round [19:37:19 CEST] <philipl> I just looked. It's what I wrote. [19:37:29 CEST] <philipl> .name = "nvenc_hevc", [19:37:34 CEST] <nevcairiel> its ... inconsistent [19:37:39 CEST] <nevcairiel> AVCodec ff_h264_nvenc_encoder = { [19:37:39 CEST] <nevcairiel> .name = "nvenc_h264", [19:37:44 CEST] <jamrial> allcodecs and configure show what i mentioned above [19:37:46 CEST] <jamrial> so meh [19:37:54 CEST] <philipl> Well, I meant the command line visible name. [19:37:55 CEST] <nevcairiel> how un-libav of them =P [19:38:49 CEST] <nevcairiel> they did do the thing we didnt want to do, shoving the levels in the options table, heh [19:39:50 CEST] <philipl> because it looks neat. [19:42:11 CEST] <philipl> I just learned how to revert a merge. yay. [19:51:48 CEST] <philipl> revert patch series sent. [19:52:44 CEST] <Daemon404> http://blogs.windows.com/buildingapps/2015/06/05/using-ffmpeg-in-windows-ap… [19:52:47 CEST] <Daemon404> my god. [19:52:59 CEST] <phh> that's called game [19:53:10 CEST] <Daemon404> FFmpegMSS = FFmpegInteropMSS::CreateFFmpegInteropMSSFromStream(readStream, false, false); [19:53:15 CEST] <Daemon404> they somehow made our api even worse [19:53:44 CEST] <Daemon404> my main issue is that they make it out like they did any work at all making it work on winrt [19:53:49 CEST] <Daemon404> they wrotet he convouted wiki page. [19:53:55 CEST] <Daemon404> wbs, bbb, etc did all the work. [19:54:20 CEST] <kierank> do you want me to troll them [19:54:41 CEST] <Daemon404> doubt it gets anywhere [19:55:03 CEST] <Daemon404> that wiki page is :V though [19:55:07 CEST] <Daemon404> like 90 pages of copypaste [19:55:18 CEST] <Daemon404> i rejected it from including it in oru texi [19:55:20 CEST] <Daemon404> our* [19:55:45 CEST] <wm4> lol https://github.com/Microsoft/FFmpegInterop/issues/2 [19:55:51 CEST] <philipl> I'm sure it does hilarious things if you try and use hardware acceleration. [19:55:57 CEST] <wm4> (I bet they just massed up API usage) [19:56:28 CEST] <Daemon404> to be fair, i cant blame them [19:56:35 CEST] <Daemon404> its not exactly easy [19:58:21 CEST] <wm4> actually it's pretty simple now [19:58:42 CEST] <wm4> you just have to ignore all the wrong official ffmpeg examples and all those "tutorials" [19:58:56 CEST] <Daemon404> i wouldnt call it simple [19:59:01 CEST] <Daemon404> certainly not ffms2-simple [20:00:41 CEST] <wm4> IMO the most confusing thing by now is shuffling the data between the sub-APIs [20:00:45 CEST] <wm4> connecting libavformat and libavcodec [20:01:03 CEST] <wm4> and then getting the data in a format you want, which may involve other sub APIs [20:01:56 CEST] <Daemon404> i think youve got a bit of tockhom [20:02:00 CEST] <Daemon404> stockholm* [20:02:35 CEST] <wm4> not really... it could be e.g. vaapi-bad [20:03:01 CEST] <Daemon404> sure worse thigns exist [20:03:04 CEST] <Daemon404> but it doesnt make it good [20:03:11 CEST] <Daemon404> or eays [20:03:14 CEST] <Daemon404> easy even [20:05:14 CEST] <kierank> personally I would abstract away demux and decode [20:05:26 CEST] <wm4> abstract away how? [20:05:35 CEST] <kierank> pipelines I guess [20:05:47 CEST] <Daemon404> wm4, like ffms2 does maybe [20:05:49 CEST] <Daemon404> for a higher level api [20:05:49 CEST] <kierank> it's the only real way you can do chained muxing and demuxing [20:07:36 CEST] <wm4> well you could just make it a thing with filters and connections between them, like almost every media framework does [20:20:36 CEST] <philipl> all changes posted. [21:44:25 CEST] <Compn> i have a questioooon [21:44:49 CEST] <Compn> what would microsoft use for codecs on winrt if ffmpeg didnt exist ? [21:44:58 CEST] <Compn> did they port the mediaframework crap ? [21:45:20 CEST] <Compn> i guess they could compile h264 , wmv stuff for winrt phones and tablets [21:46:32 CEST] <Compn> https://playerframework.codeplex.com/ [21:46:52 CEST] <Compn> so apple, builds an os on opensource bsd, refuses to share anything? [21:46:57 CEST] <Compn> microsoft, goes open source ? [21:47:02 CEST] <kierank> apple do share stuff [21:47:33 CEST] <Compn> webkit? [21:50:45 CEST] <Daemon404> most of the os core [21:50:55 CEST] <Daemon404> its just in terrible apple places [21:53:45 CEST] <Compn> ah [21:55:08 CEST] <Mavrik> Compn, all of those have HW decoders [21:55:21 CEST] <Mavrik> so I guess just like on apple stuff you'd be stuck with whatever hardware supports [21:57:56 CEST] <Compn> true [22:16:16 CEST] <nevcairiel> Daemon404: i never found ffmpeg API to be that bad, its about the complexity I expected when dealing with media frameworks ... maybe it "helps" that I wrapped DShow around it, which has similar separation between demux, decode and rendering =p [22:16:34 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:8a6536cbc0a9: Revert "Changelog: Drop duplicate NVENC entry" [22:16:34 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:728c82a53253: Revert "avcodec/Makefile: fix checkheaders for nvenc_b" [22:16:34 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:58b49f78cf34: Revert "nvenc: remove cuda.h requirement for nvenc_a" [22:16:34 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:d47de79372d3: Revert "Merge commit 'b08caa87c35a768ec0abb16b1e99c3a85f1df28e'" [22:16:34 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:c10e6bcb43ad: Revert "avcodec: Rename nvenc.c to nvenc_a.c, to avoid conflict with the other implementation" [22:16:35 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:e79c40fe7247: avcodec/nvenc: Rename nvenc_h265 to nvenc_hevc [22:16:35 CEST] <cone-659> ffmpeg 03Philip Langdale 07master:7e4661174ae1: avcodec/nvenc: Add 'nvenc_h264' as an alternative name for 'nvenc' [22:20:23 CEST] <Daemon404> nevcairiel, maybe [22:20:39 CEST] <wm4> I'm also dealing with my own demux/decode separation [22:20:49 CEST] <wm4> not having this separation would actually be a pain for me [22:59:25 CEST] <cone-659> ffmpeg 03Michael Niedermayer 07master:f30a7d9861af: avcodec/mpegvideo: Merge thread context initialization loops [22:59:26 CEST] <cone-659> ffmpeg 03Michael Niedermayer 07master:7ddedd23625a: avcodec/mpegvideo: Clear thread_context array before allocating [22:59:27 CEST] <cone-659> ffmpeg 03Michael Niedermayer 07master:46428ea332d8: avcodec/mpegvideo: Use av_memdup() for allocating thread_context [23:06:08 CEST] <philipl> BtbN: Do you remember the report that our nvenc is noticeably slower than the nvidia reference encoder - particularly, it ends up slower than real time for 4Kp60 content. [23:07:00 CEST] <BtbN> you mean shadowplay? [23:07:28 CEST] <philipl> No, there was that guy who was trying to transcode 4k content and he couldn't get it to go at realtime, but the nvidia example app could. [23:07:33 CEST] <philipl> Just normal nvenc [23:08:07 CEST] <BtbN> Hm, the only difference i can think of then is that the reference app doesn't use CUDA for interop, but some DirectX/D3D code [23:08:09 CEST] <philipl> Now, the example app only takes in raw yuv frames, so I wonder if he was decoder bottle-necked. [23:08:26 CEST] <philipl> I can't remember what his source format was or where the data was coming from. [23:08:37 CEST] <BtbN> Usualy the bottlenack when using nvenc in ffmpeg is the yv12 to nv12 conversion [23:09:45 CEST] <BtbN> Someone was here, who made a CUDA based yuv420p->nv12 filter, which resulted in massive speedups [23:10:00 CEST] <BtbN> Might actualy be worth it do implement something like that, with OpenCL [23:10:14 CEST] <nevcairiel> using sse2 for yv12->nv12 is usually really not much slower than a memcpy [23:10:23 CEST] <nevcairiel> because that conversion is so extremely trivial [23:10:26 CEST] <philipl> Yeah, if that's the difference between realtime and slower than realtime, it's definitely worth it. [23:10:48 CEST] <philipl> If the input data was nv12, can you avoid a memcpy on the input path? Can it reuse the caller's buffer? [23:10:48 CEST] <BtbN> hm? nv12 is the least trivial conversion of all, as you have to inverleave the u/b planes [23:11:14 CEST] <BtbN> no, the buffer has to be allocated by nvenc [23:11:32 CEST] <philipl> sad. [23:11:36 CEST] <BtbN> There's no way to make it use a pre-existing buffer [23:12:04 CEST] <philipl> It seems rather unfortunate that it can't consume the most common surface format... [23:12:05 CEST] <nevcairiel> yv12->nv12 is a load, a unpack, and a store sse operation [23:12:16 CEST] <nevcairiel> its super trivial o.o [23:12:29 CEST] <BtbN> Hm, isn't yv12 3 planes? [23:12:38 CEST] <philipl> yeah, yv12 is three plans [23:12:41 CEST] <philipl> planes [23:12:52 CEST] <nevcairiel> there is a sse2 instruction to interleave bytes [23:13:21 CEST] <BtbN> I don't have to much of an idea about sse, but even with avx and everything supported, the conversion still is the slowest part of the chain, if you use hw decoding and encoding [23:13:40 CEST] <philipl> So, rewinding slightly - nvenc is already initialising cuda, so we should just get that guys routine and use it. [23:13:42 CEST] <BtbN> Can very well be the bottleneck when using 4K60 [23:13:57 CEST] <nevcairiel> so you do: load u, load v, unpack*2 (since it operates on half a reg), store uv into nv12 buffer *2 [23:14:16 CEST] <nevcairiel> not sure thats going to be faster if you copy it back and forth between the gpu [23:14:29 CEST] <nevcairiel> unless you can keep it on the gpu if done in cuda [23:14:30 CEST] <BtbN> the copy to GPU happens internaly in nvenc [23:14:38 CEST] <BtbN> you just memcpy to another system-ram buffer [23:14:51 CEST] <philipl> can nvenc take a cuda buffer for the input? [23:15:08 CEST] <BtbN> It should, but i never looked into that [23:15:15 CEST] <philipl> I thought I read that. [23:15:30 CEST] <philipl> so yeah, in theory you could use cuda for the conversion and keep the result in GPU memory to pass to nvenc. [23:15:33 CEST] <BtbN> It can take D3D surfaces, which is interesting if you are using it in a game, or you are steam and are hooked into a game and want to encode it [23:16:04 CEST] <philipl> I'm on linux, and there's no equivalent GL interop. [23:16:08 CEST] <philipl> :-P [23:16:24 CEST] <BtbN> Nope, there's nothing like that on linux [23:16:32 CEST] <BtbN> although NvFBC should work on linux [23:16:36 CEST] <BtbN> But it's NDA-Only [23:17:39 CEST] <BtbN> That's how ShadowPlay works, it basicaly grabs the framebuffer on the GPU, and sends it to the encoder [23:17:46 CEST] <nevcairiel> I assume the big problem here is that sws just has no optimized yv12->nv12 path, which makes it slow, although since you need to memcpy anyway, it might be ideal to do this conversion in nvenc with a specialized function that transfers "in-flight" with simd if available [23:18:18 CEST] <BtbN> The "libnvenc" thing does that, although in the most non-optimal way possible [23:18:23 CEST] <philipl> Yeah. You'd want to do the conversion as part of the copy that you're already committed to [23:18:26 CEST] <BtbN> just a few for-llops [23:18:32 CEST] <BtbN> *loops [23:18:46 CEST] <BtbN> But even that is already slightly faster than was sws does [23:18:55 CEST] <philipl> I did some brief comparisons between your encoder and the one the nvidia people wrote and they both performed about the same. [23:19:10 CEST] <nevcairiel> In my decoder, I always convert yv12->nv12, because GPUs are much happier with nv12, and its never had a noticable performance impact, since its such an easy conversion [23:19:10 CEST] <philipl> (the ffmpeg encoder they wrote) [23:19:14 CEST] <BtbN> The difference gets more aparent when you are on a non-sse3 CPU [23:19:34 CEST] <nevcairiel> and sse2 is like present everywhere [23:19:38 CEST] <BtbN> Someone was here with an AMD Phenom, which doesn't support some sse thing. [23:19:58 CEST] <BtbN> The conversion was awfully slow for him, couldn't even handle 720p60 [23:20:25 CEST] <jamrial> BtbN: ssse3, not sse3 [23:20:56 CEST] <philipl> BtbN: hang on. nvenc supports yv12 input surfaces. [23:20:57 CEST] <jamrial> and yeah, amd was late to add that [23:21:02 CEST] <BtbN> philipl, nope, it doesn't [23:21:04 CEST] <BtbN> The API does [23:21:08 CEST] <BtbN> But nvenc doesn't [23:21:13 CEST] <philipl> ah. lovely. [23:21:16 CEST] <BtbN> it's not in the list of supported formats if you query i [23:21:17 CEST] <BtbN> *it [23:21:23 CEST] <philipl> *groan* [23:21:46 CEST] <BtbN> I just hardcoded it to nv12, because you have to initialize the nvenc chain quite a bit until you can get that information from it [23:22:04 CEST] <BtbN> and doing that in the init-static-data function wasn't so pretty [23:22:12 CEST] <nevcairiel> interestingly the libav encoder offered that in the pixfmt list, what did it do, just fail? :d [23:22:39 CEST] <BtbN> If they are based on the libnvenc encoder, they do in-place conversion when copying to the nvenc buffer [23:22:46 CEST] <BtbN> That's what it did [23:23:23 CEST] <BtbN> If they don't do that, and just send yv12 to nvenc, it will just fail to initialize at some point [23:23:41 CEST] <BtbN> Or nvidia added support for yv12 with some driver update, and it actualy does work now [23:23:51 CEST] <philipl> I'm trying to establish that. [23:24:03 CEST] <BtbN> Easy enough to check, add yuv420p to the list of formats, the code to use it is already in place [23:24:09 CEST] <philipl> Yes, I saw. [23:24:15 CEST] <philipl> It's working here, with no difference in speed. [23:24:18 CEST] <philipl> ... [23:24:24 CEST] <BtbN> interesting [23:25:03 CEST] <BtbN> If they did add support for it, that's a bit of a problem. [23:25:18 CEST] <BtbN> Adding it to the list of supported formats will screw people with older drivers [23:25:23 CEST] <philipl> Yeah. I'm seriously seeing exactly the same encode framerate. [23:25:32 CEST] <philipl> How much do we care about that? [23:26:31 CEST] <philipl> If you're on the nvidia bandwagon, you update drivers monthly, don't you? [23:26:50 CEST] <BtbN> If you're on some stable distro like ubuntu, you don't :/ [23:26:59 CEST] <BtbN> On Windows it shouldn't be an issue [23:27:20 CEST] <BtbN> I'll have to check on my hardware, maybe it's a maxwell thing, and not only the driver [23:27:33 CEST] <philipl> This is probably where you remember that the only legitimate way to get nvenc support is to compile for yourself, so you're probably qualified to use the right ppa for the driver. [23:27:42 CEST] <philipl> Yeah, maxwell here. [23:27:56 CEST] <BtbN> I only have Kepler cards here [23:28:08 CEST] <BtbN> And last time I checked, yv12 didn't work [23:28:35 CEST] <philipl> Well, the 'good' news is that supporting yv12 doesn't make anything faster :-P [23:29:16 CEST] <JEEBsv> NV12 seems to be the new favourite child around GPUs, no? [23:29:27 CEST] <philipl> How do I check if ffmpeg is doing a format conversion implicitly? [23:29:44 CEST] <JEEBsv> -v debug or so? [23:29:58 CEST] <nevcairiel> it should tell you the input format on the format line [23:30:07 CEST] <nevcairiel> input to the encoder [23:30:11 CEST] <BtbN> It should be quite obvious with debug logging, or even just verbose [23:30:14 CEST] <philipl> Yeah, it does. [23:31:02 CEST] <philipl> Anyway, I'm just so puzzled that it's made no difference. [23:31:19 CEST] <philipl> Might be decoder-bound? My sample source here is a vp9 4k stream from youtube. [23:31:40 CEST] <BtbN> try disabling the encoder, and check how fast it goes [23:31:49 CEST] <BtbN> Just pipe raw output to /dev/null or something [23:32:38 CEST] <BtbN> I can't imediately think of a reason why there would be a major speed difference between the ffmpeg and reference-implementation [23:33:01 CEST] <philipl> Decodes at ~97 [23:34:05 CEST] <BtbN> hm [23:34:41 CEST] <BtbN> maybe it's actualy a memory bandwidth thing, and copying that much data around twice just can't get any faster [23:34:53 CEST] <philipl> and the reference implementation doesn't support yv12. [23:34:56 CEST] <BtbN> But getting rid of the conversion should have done something about that [23:35:25 CEST] <philipl> I'd originally suggested to the guy who asked that maybe the hardware performance fell off non-linearly but he swore the reference implementation could do realtime. [23:36:13 CEST] <BtbN> Did he try to use a diffrent nvenc profile? [23:36:28 CEST] <BtbN> Maybe the reference one was in some high-performance mode? [23:37:49 CEST] <philipl> hp preset takes us from 36 -> 43. [23:38:03 CEST] <philipl> (for hevc) [23:38:26 CEST] <BtbN> i also think i read somewhere that 4K, 60 fps isn't yet supported? [23:38:32 CEST] <BtbN> But 4K30 is [23:38:34 CEST] <philipl> for h264, it goes from ~60 to 75 [23:38:57 CEST] <philipl> The claim was that he was doing 4k hevc. [23:39:01 CEST] <philipl> but what do I know. [23:39:09 CEST] <philipl> Heck, the reference encoder doesn't do hevc. [23:39:49 CEST] <BtbN> That would explain the difference at least [23:40:34 CEST] <philipl> Ok. so with h264, supporting yv12 gets you ~60fps on hq preset. [23:40:38 CEST] <philipl> With nv12 only you get ~55. [23:40:56 CEST] <philipl> For hevc, you see no difference at all and it's stuck around 36. [23:41:09 CEST] <nevcairiel> and that difference could probably be fixed by improving the conversion [23:41:39 CEST] <philipl> So, I'm going to conclude the following: hevc encoding simply can't do 4k60 until someone shows me proof that it can somewhere. [23:41:50 CEST] <philipl> And h264 can do 4k60 if not bottlenecked. [23:42:23 CEST] <BtbN> I'm quite sure i read somewhere that h265 with 4K60 isn't supported until the next GPU generation [23:43:47 CEST] <philipl> Well, 35fps would make a lot of sense in that context, [23:45:05 CEST] <BtbN> I'll check how my gtx760 likes yv12 [23:49:07 CEST] <philipl> I'm on a gtx980 with the linux 352.09 drivers here for comparison. [00:00:00 CEST] --- Sun Jun 7 2015

1 0

ffmpeg.log.20150606
by burek 07 Jun '15

07 Jun '15

[00:04:21 CEST] <t4nk336> Agh. I told ffmpeg to build x64 and it built x86. Who does that?! [00:08:31 CEST] <BtbN> It will build as whatever msvc shell you used. [00:11:21 CEST] <t4nk336> Ah, thanks BtbN; that solved it, I think. [00:20:02 CEST] <t4nk336> Now it seems to be building x64... Except for things built by yasm... [00:24:17 CEST] <t4nk336> Maybe I just need to reconfigure [00:43:54 CEST] <t4nk336> Wowee - I finally compiled and got exes with pdbs [00:44:31 CEST] <t4nk336> And it only took almost 2 hours. A new record perhaps? [00:51:06 CEST] <klaxa> t4nk336: no [00:51:38 CEST] <t4nk336> And debugging, it looks like the issue is that the parser uses a different H264Context than the codec context for some reason... [00:51:39 CEST] <klaxa> i can assure you people (including me) have spent way waaaaay more time trying to get a sofware to run [00:51:57 CEST] <t4nk336> I meant record in the other direction ;) [00:52:27 CEST] <t4nk336> As the parser sets its H264Context's is_avc, but the codec context's H264Context's is_avc remains cleared [00:54:01 CEST] <t4nk336> In fact, I think that's a libavcodec bug [00:54:17 CEST] <t4nk336> So I guess it's a good thing I have the source so I can workaround it locally [00:55:42 CEST] <t4nk336> Oh or maybe I need to set the decoder's "is_avc" field myself or something. [00:57:49 CEST] <t4nk336> Ohh - I needed to set the extradata/extradata_size *before* I opened the codec, not after [00:58:04 CEST] <t4nk336> Now it works fine. [00:59:29 CEST] <t4nk336> Gotta love struct-based APIs [01:01:14 CEST] <BtbN> you don't even want to know what the libav* apis would look like if everything were function parameters. [01:06:06 CEST] <t4nk336> True, but I do want to know what the libav* apis would look like if everything were APIs that operated on the AVContext/Codec/etc. objects - setters, getters, etc. [01:36:21 CEST] <klaxa> C has no objects and that is good [01:36:24 CEST] <klaxa> but that's just my opinion [01:39:17 CEST] <t4nk336> Just look at something like libpng (modern) [01:39:46 CEST] <t4nk336> it's c code, but it's object-oriented rather than requiring the user to poke around in a struct [01:40:17 CEST] <t4nk336> And it just reduces a good amount of user mistakes as well as allows for better binary/source compatibility [02:04:04 CEST] <chungy> glib is also object-oriented and plain C. GTK/GNOME are built from it :P [02:04:23 CEST] <chungy> (those devs have actively expressed hatred towards C++, haha) [02:10:35 CEST] <gnfz> I'm having trouble finding much information on creating chapters using ffmpeg (command line). Can someone point me in the right direction? Something along the lines of -metadata:c:# key=value is spitting out an error message that the chapter index is invalid. (with # an int and key an element name from looking at matroska's section on chapter element name's) [02:13:13 CEST] <gnfz> I'm just blindly guessing though as I couldn't find any specific examples in the documentation. [02:53:37 CEST] <gnfz> I ended up just dumping the metadata from a file I knew had chapters and used map_metadata. If there's a way to create chapters without using a separate metadata configuration file, I'm still interested. [03:53:17 CEST] <MINIMAN10000> Alright so I was looking for the lowest bitrate I could get my video and then ran the encode. This is what I ended up when trying to find the lowest bitrate. ffmpeg -i input.mkv -preset slowest -c:v libvpx-vp9 -c:a libvorbis -q:a 0 -b:a 0k -b:v 1k output.webm [03:54:37 CEST] <MINIMAN10000> are there any other things that can make it smaller while still having both video and audio? [05:42:11 CEST] <cehoyos> MINIMAN10000: Generally, to get the lowest bitrate, choose the highest quantiser supported by a codec (don't know if this is also true for libvpx) [05:42:48 CEST] <MINIMAN10000> cehoyos, on libvpx the if I remember correctly ended up being around 50% larger [05:43:58 CEST] <MINIMAN10000> oh wait nvm [05:44:00 CEST] <MINIMAN10000> i used crf [05:44:09 CEST] <MINIMAN10000> unless crf is the quantizer? [05:49:32 CEST] <cehoyos> It is not exactly the same iirc, but choosing the worst possible quality with crf should produce the smallest possible files. [06:39:13 CEST] <MINIMAN10000> cehoyos, your right setting min and max quantizer does produce the smallest thanks [06:39:26 CEST] <MINIMAN10000> but now my audio is larger than my video lol [06:41:54 CEST] <MINIMAN10000> got ogg vorbis set to quality of 0... as low as it goes jeeze [06:49:51 CEST] <MINIMAN10000> alright manually set it to 96 kb by punching in smaller numbers til one worked and its smaller than q 0 [06:51:47 CEST] <MINIMAN10000> Alright the new smallest output command line is. ffmpeg -i input.mkv -preset slowest -c:v libvpx-vp9 -c:a libvorbis -ab 96k -qmin 63 -qmax 63 output.webm [06:57:31 CEST] <MINIMAN10000> huh according to opus own codec chart they have higher quality at what looks to be 128 kb/s [07:04:49 CEST] <MINIMAN10000> not entirely sure why 48 or 64 k didnt work with vorbis but it may be worth a shot to try opus after this finishes [07:13:18 CEST] <MINIMAN10000> ... now that im on the wiki and went down further [07:13:36 CEST] <MINIMAN10000> they also have 80 64 45 officially [07:14:39 CEST] <rjp421> http://www.streamingmedia.com/ConferenceVideos [09:45:47 CEST] <asimas> Hello. I've found this command http://pastebin.com/9irnTLPe that uses a complex filter to concat and then scale a video. Is there any way I could first scale the videos and then concat then? I'd still like to do it in a single command. [09:47:39 CEST] <Mavrik> asimas, switch the order? [09:47:58 CEST] <asimas> Right [09:51:05 CEST] <asimas> Also the input files might need to be scaled differently, so the question is how can I refer to specific input's video stream and use it for the scale filter [09:53:52 CEST] <Mavrik> you can't afaik [09:53:57 CEST] <Mavrik> but that shouldn't matter [09:54:03 CEST] <Mavrik> since you need to get them all to the same size right? [09:56:35 CEST] <asimas> Mavrik, couldn't I do something like this? http://pastebin.com/B3iw8KjP [09:58:11 CEST] <asimas> Mavrik, I need them all at the same size but I want to scale them without losing aspect ratio and adding padding where necessary. That's why I think I need to use separate filters for each input before concatenating. [09:59:09 CEST] <Mavrik> ah [09:59:29 CEST] <Mavrik> yes, you'll have to do it for each stream first and then concat, but I don't know the syntax for that I'm afraid :/ [10:01:26 CEST] <asimas> Mavrik, that's ok. Thanks for your help :) I'll try to come up with something myself. [10:30:49 CEST] <asimas> Well if anyone's interested here's how I managed to scale first and then concat in a single command: http://pastebin.com/xdTKrkw7 [11:19:39 CEST] <hendry> anyone know how to capture the fullscreen with -f x11grab ? it seems to require "-i :0.0" now, but that doesn't capture whole screen [14:50:17 CEST] <MINIMAN10001> alright im back [14:51:53 CEST] <cehoyos> hendry: Command line and complete, uncut console output missing. If this is a regression (if a command line used to work with an older version of FFmpeg but does not work anynore), I am especially interested. Please don't forget to test current FFmpeg git head before reporting issues / asking questions. [14:52:17 CEST] <MINIMAN10001> so by changing from bitrate a target of 1kb to a quantizer setting of 63 and to opus 96kbps im hitting just under 20 MB from the previous 26 MB [14:54:19 CEST] <MINIMAN10001> with an input of 1.09 GB [14:57:10 CEST] <cehoyos> 96kb sounds a lot for opus if you use quantizer 63... [14:57:25 CEST] <MINIMAN10001> i tried 48 and it sounded like it was missing the hightones [14:57:34 CEST] <cehoyos> And how did it look like?? [14:57:36 CEST] <MINIMAN10001> it was extremely distracting from the original [14:57:46 CEST] <MINIMAN10001> honestly better than i thought considering its the worst it can get [14:57:47 CEST] <cehoyos> I mean the video. [14:58:11 CEST] <cehoyos> 96kb is still a (very) high bitrate if you are trying to get as low as possible. [14:58:20 CEST] <cehoyos> But opus is said to beat aac [14:59:04 CEST] <MINIMAN10001> http://puu.sh/iexOf/1adb9c3065.jpg this vs original http://puu.sh/iexQ0/17cce1fcb7.jpg [15:00:30 CEST] <MINIMAN10001> but yeah now id like to make another one that aims to have higher quality video. Last used command. ffmpeg -i input.mkv -preset slowest -c:v libvpx-vp9 -c:a libopus -b:a 96k -qmin 63 -qmax 63 output.webm [15:00:58 CEST] <MINIMAN10001> so just up the qmin and that should do it? [15:03:31 CEST] <cehoyos> Why are qmin and qmax necessary? [15:03:43 CEST] <cehoyos> Did you try -q:v 63 ? [15:03:51 CEST] <MINIMAN10001> qmax wasnt nessassary but i didnt want to leave it open to interpretation at least. [15:03:59 CEST] <c_14> You know, if you want to make the video smaller, just downscale? [15:04:01 CEST] <MINIMAN10001> i thought q:v was presets [15:04:25 CEST] <MINIMAN10001> i wasnt sure if a downscaled would result in higher quality [15:04:31 CEST] <MINIMAN10001> for the same datarates [15:05:56 CEST] <cehoyos> Is there a difference (md5sum-wise) if you use -q:v 63 or qmin and qmax? [15:06:39 CEST] <MINIMAN10001> not sure how to obtain the md5 but i have no idea which is why i set both qmin and max [15:08:05 CEST] <cehoyos> Is there a difference with the output filesize if you use -q:v or qmin/qmax? [15:08:27 CEST] <MINIMAN10001> encoding a 10 second segment on qmin/qmax atm [15:08:53 CEST] Action: c_14 got a 3byte difference over 5 seconds [15:09:28 CEST] <MINIMAN10001> i swear if its 3 bytes smaller for q:v imma throw a fit lol [15:09:29 CEST] <c_14> Eh, 3kilobyte that is [15:09:34 CEST] <MINIMAN10001> kb* [15:09:43 CEST] <MINIMAN10001> which is quite significant for the sizes on a 5 sec i imagine [15:10:35 CEST] <c_14> Nah, it is bytes. or blocks whatever `ls -l' outputs [15:10:42 CEST] <MINIMAN10001> q:v was around 100% larger [15:11:50 CEST] <MINIMAN10001> was the difference between 100 kb and 200 kb for qmin/max and q:v respectively [15:12:43 CEST] <cehoyos> Sorry, it is -crf 63 [15:13:30 CEST] <cehoyos> There is a kind of stress field between "q:v" and "crf" and qmin/qmax seem to be unable to decide on which side they are;-) [15:14:40 CEST] <cehoyos> (Some codecs like vpx ignore -q:v but accept -crf,, others like mpeg4 don't know crf but respect -q:v and the fact that qmin maps to both in different codecs makes the issue worse.) [15:15:18 CEST] <MINIMAN10001> crf is 50% larger than min/max [15:18:27 CEST] <JEEBsv> cehoyos: i think the funniest part is that only libx264 and libx265 actually have crf [15:18:54 CEST] <JEEBsv> libvpx's thing is different iirc, but somehow was mapped to it [15:18:58 CEST] <cehoyos> The given report (and the source code) indicates that libvpx also has it... [15:19:18 CEST] <cehoyos> I am not talking about semantics, only command line usage. [15:19:55 CEST] <cehoyos> And I still don't think that any user cares about "lamda" in this context. [15:20:57 CEST] <JEEBsv> dunno, all I remember is that libvpx's thing to which -crf was mapped was different. might have even been different between vp8 and vp9 [15:21:28 CEST] <JEEBsv> as in, how that libvpx option works [15:21:43 CEST] <JEEBsv> but I don't remember the details :) [15:24:50 CEST] <MINIMAN10001> alright [15:25:14 CEST] <MINIMAN10001> im no longer interested in minimizng the file size, like i did with opus id like to meet a certain quality level... how should i do that. [15:29:47 CEST] <MINIMAN10001> lol dangit sounds like a pain according to this [15:30:11 CEST] <MINIMAN10001> says you should use crf in conjunction with bitrate in order to manipulate quality [15:30:24 CEST] <c_14> just use crf [15:30:27 CEST] <c_14> What are you reading? [15:30:43 CEST] <MINIMAN10001> ffmpeg vp8 encode wiki [15:30:47 CEST] <MINIMAN10001> https://trac.ffmpeg.org/wiki/Encode/VP8 [15:31:01 CEST] <Mavrik> *sigh* [15:31:05 CEST] <c_14> Aren't you using vp9? Read https://trac.ffmpeg.org/wiki/Encode/VP9 [15:31:09 CEST] <Mavrik> I don't think there's a way of making VP8 not crap. [15:31:16 CEST] <MINIMAN10001> " Note that you shouldn't leave out the -b:v option as the default settings will produce mediocre quality output." [15:31:36 CEST] <MINIMAN10001> eh thought the vp9 page wasnt complete [15:31:40 CEST] <MINIMAN10001> guess that isnt the case lol [15:32:02 CEST] <MINIMAN10001> oh says set b:v 0 [15:32:04 CEST] <c_14> It's been complete for 2 month [15:33:08 CEST] <c_14> +s [15:33:33 CEST] <MINIMAN10001> welp [15:33:45 CEST] <MINIMAN10001> crf 10 might have been higher than i wanted lol [15:35:01 CEST] <MINIMAN10001> its practically the same bitrate as the original with a newer encoder... sounds like a recipe for wasted space [15:35:06 CEST] <c_14> crf might be a bit high, I usually use 31 or so for short clips [15:35:22 CEST] <c_14> Not sure why I wrote a crf of 10 there [15:35:40 CEST] <MINIMAN10001> crf 10 is used in all the examples on the libvpx pages [15:39:13 CEST] <c_14> Oh well, not like it really matter.s [15:42:02 CEST] <MINIMAN10001> so i was reading on streaming ealier and it said a iframe was sent every 250 frames and nolatenecy mode makes every frame an iframe. does that mean the point between iframe is considered latency? [15:44:25 CEST] <MINIMAN10001> i became interested in latency when i heard it can be adjuted from 5 to 120 ms in opus [15:44:31 CEST] <MINIMAN10001> adjusted* [15:45:06 CEST] <c_14> https://web.archive.org/web/20141118095252/http://mewiki.project357.com/wik… [15:45:09 CEST] <c_14> that's for x264 though [15:46:07 CEST] <MINIMAN10001> oh [16:01:03 CEST] <doki_pen> I'm trying to cut a video, and for the most part it's working, but I get a few seconds of grey screen at the begining of the output: ffmpeg -i cut.avi -ss 10 -t 00:06:25 -c copy final.avi [16:01:20 CEST] <doki_pen> any ideas on how to fix? [16:20:07 CEST] <MINIMAN10001> for me using -ss in conjunction with -t works just fine [16:22:17 CEST] <doki_pen> yeah, must have something to do with the encoding. no matter where I clip from, the first few seconds are grey screen. audio is fine [16:44:17 CEST] <klaxa> doki_pen: it might be related to the types of frames in the video, you are probably trying to start from a non-i-frame [16:45:35 CEST] <klaxa> an i-frame is a full frame that can be decoded without referencing earlier or later frames [16:45:53 CEST] <klaxa> you can try to re-encode (which will reduce quality) or cut at i-frames [16:57:40 CEST] <zhanshan> hi [16:58:11 CEST] <zhanshan> is there an option in ffmpeg to reorder massive JPG files to render them properly to video? [16:58:59 CEST] <zhanshan> I made the big mistake to name the files '002_wild0001.jpg' - '002_wild32465.jpg' [16:59:23 CEST] <zhanshan> so in the end it is one more digit than in the beginning [17:01:06 CEST] <c_14> Not in ffmpeg, but you can use something like `perl-rename 's/([0-9]{4})/0$1/' *' [17:03:15 CEST] <zhanshan> c_14 thanks but I don't know how to use this [17:04:33 CEST] <intracube> hi, is '*.jpg' in: ffmpeg -f image2 -i *.jpg -c:v ... expanded by the shell (on linux) or handled directly by ffmpeg? [17:06:07 CEST] <jarainf> intracube, it's handled by the shell [17:07:08 CEST] <intracube> jarainf: thanks [17:09:40 CEST] <c_14> zhanshan: the command I gave you should work (assuming you have perl-rename installed). You might want to add -n for testing though (that'll just tell you what it'll rename). [17:16:05 CEST] <zhanshan> ok thanks! [17:20:11 CEST] <zhanshan> I don't get it, is the command 'rename' or 'perl-rename'? [17:20:18 CEST] <zhanshan> I only have rename installed [17:22:04 CEST] <zhanshan> $ rename 's/([0-9]{4})/0$1/' *' -n [17:22:12 CEST] <zhanshan> there's one ' missing I guess [17:22:20 CEST] <zhanshan> the command doesn't work like this [17:24:07 CEST] <zhanshan> ok I got it [17:24:26 CEST] <zhanshan> but there are still files with different digit-length 002_wild002389.jpg 002_wild0023899.jpg [17:24:49 CEST] <zhanshan> the command seems to just add one '0' in front after '002_wild' [17:25:07 CEST] <zhanshan> I actually performed the command twice already :P [17:56:29 CEST] <spectromas> is there anything unusual about this file? https://bpaste.net/show/46c40c13e59f my tv is having problems with it but I can't see anything that might be a problem [18:02:31 CEST] <Mavrik> does your TV handle MKVs? [18:03:34 CEST] <spectromas> Mavrik: it does usually yes [18:03:41 CEST] <dsl420> spectralsun, DAR looks a bit weird, but i could be wrong though [18:04:17 CEST] <Mavrik> is it perhaps concatenated from multiple files? [18:05:09 CEST] <spectromas> it's possible, i don't know the source of the file [18:06:02 CEST] <spectromas> it plays fine on my computer with mpv though, that's the annoying thing [18:18:03 CEST] <c_14> zhanshan: eeeh, right. The regex wasn't entirely correct. it should have been s/[^0-9]([0-9]{4}/0$1/. Now you'll have to update the {4} to whatever the shortest digit length is [18:37:28 CEST] <MINIMAN10001> alright figured it out [18:37:36 CEST] <MINIMAN10001> the reason why audio sucked is it's 5.1 channel [18:39:11 CEST] <MINIMAN10001> weird a cfr of 63 is relatively high quality [18:39:26 CEST] <MINIMAN10001> crf* [18:40:27 CEST] <MINIMAN10001> some significant artifacting [19:17:21 CEST] <zhanshan> pls see https://gist.github.com/zhanshan/b72fd3a750830dbe9520 [19:17:30 CEST] <zhanshan> a new error from ffmpeg [19:20:55 CEST] <zhanshan> c_14 about the issue before: in 002_wild000001.jpg the digit length would be 6 or 1? [19:21:56 CEST] <zhanshan> 002_wild001650.jpg, 002_wild0016509.jpg -> 6, 7? [19:22:02 CEST] <zhanshan> so I'd go for 6 [19:22:05 CEST] <zhanshan> I guess [19:39:45 CEST] <zhanshan> [image2 @ 0x40c2c60] Could find no file with path '*.jpg' and index in the range 0-4 [19:39:46 CEST] <zhanshan> *.jpg: No such file or directory [19:39:48 CEST] <zhanshan> :P [19:41:55 CEST] <zhanshan> with '-pattern_type glob' it seems to work now [19:42:00 CEST] <klaxa> i remember having troubls with long numbers too. maybe making a subdirectory and creating symlinks like 000001.jpg -> ../002_wild000001.jpg would make it easier? [19:44:18 CEST] <klaxa> and if all fails try: cat some/dir/*.jpg | ffmpeg -f image2pipe -i - [options] out.mkv [19:46:04 CEST] <zhanshan> klaxa I think the problem seems to be that ffmpeg accepts only four digit-numbering 0000? [19:46:13 CEST] <zhanshan> more than that causes problems? [19:46:43 CEST] <klaxa> i have an ffmpeg command in my history with %05d [19:46:52 CEST] <klaxa> so i guess 5 works at least [19:46:53 CEST] <intracube> zhanshan: if the file sequence starts at 00000001.jpg onwards you should be able to do: -i $08d.jpg [19:46:57 CEST] <intracube> for example [19:47:51 CEST] <zhanshan> alright [19:48:36 CEST] <klaxa> just tested it and %06d works too for me [19:49:12 CEST] <zhanshan> ok, thanks! [21:50:38 CEST] <feliwir> hey, is there any official FindFFMPEG.cmake script? [21:52:39 CEST] <feliwir> everytime i use a script from the web i get linker errors -.- [22:26:31 CEST] <Emmanuel_Chanel> Hello! [22:27:31 CEST] <Emmanuel_Chanel> I'm seeking how to add Track2.acc to Version1.mkv with 1 video track and 1 audio track. Do you know where I should see? [22:28:59 CEST] <c_14> ffmpeg -i Track2.acc -i Version1.mkv -map 0 -map 1 -c copy out.mkv [22:33:48 CEST] <Emmanuel_Chanel> c_14: It says Could not write header for output file #0 (incorrect codec parameters ?): Invalid argument [22:47:47 CEST] <anoop_r> hi i got error Past duration 0.904747 too large [22:47:48 CEST] <anoop_r> while using x265 to encode [22:47:48 CEST] <anoop_r> x265 10bit with ffmpeg [22:53:31 CEST] <anoop_r> please see this http://pastebin.com/9L60nhYi [23:17:00 CEST] <MINIMAN10001> hmm [23:17:13 CEST] <MINIMAN10001> not sure how to run x265 [23:18:06 CEST] <c_14> The binary or encode to H.265 using ffmpeg? [23:18:35 CEST] <MINIMAN10001> playing a video encoded using ffmpeg to use h.265 [23:18:49 CEST] <c_14> what? [23:19:41 CEST] <MINIMAN10001> i create a h265. i have no idea how to play it [23:19:56 CEST] <c_14> ffplay file [23:19:56 CEST] <c_14> mplayer file [23:19:56 CEST] <c_14> mpv file [23:19:57 CEST] <c_14> vlc file [23:20:02 CEST] <MINIMAN10001> ooh [23:20:05 CEST] <MINIMAN10001> i used mp4 [23:20:33 CEST] <MINIMAN10001> vlc sounds promising [23:20:58 CEST] <MINIMAN10001> oh... vlc not suitable for libx265 [23:21:11 CEST] <MINIMAN10001> either is mpv [23:22:35 CEST] <iive> libx265 is only encoder. ffmpeg libavcodec has the decoder and that is what everything uses. [23:23:33 CEST] <MINIMAN10001> for example i encoded using libx264 and decoded using google chrome [23:23:54 CEST] <MINIMAN10001> i encoded with libx265 but google chrome and windows media player is unable to decode it. [23:24:14 CEST] <MINIMAN10001> the file was a .mp4 [23:28:48 CEST] <JEEBsv> so do you want to play back or encode? [23:28:52 CEST] <JEEBsv> you sound confused [23:30:22 CEST] <MINIMAN10001> i want to play back [23:30:33 CEST] <MINIMAN10001> h265 [23:30:52 CEST] <JEEBsv> then I'd just grab VLC 2.2 or newer, or mpv from https://mpv.srsfckn.biz/ [23:30:58 CEST] <JEEBsv> both should support HEVC :P [23:31:11 CEST] <JEEBsv> latter you drag and drop stuff on top of the exe [23:34:57 CEST] <MINIMAN10001> alright vlc works fine [23:41:34 CEST] <MINIMAN10001> hmm i give up that task lol [23:42:07 CEST] <MINIMAN10001> i wanted to try and get a bitrate that would playable realtime by a 56k modem [23:42:26 CEST] <MINIMAN10001> but at low bitrates video quality jumps off a cliff [23:43:12 CEST] <MINIMAN10001> hmm yeah [23:43:18 CEST] <MINIMAN10001> problem is probably the resolution [23:43:59 CEST] <JEEBsv> and frame rate, since every picture has to have its headers etc [23:44:23 CEST] <MINIMAN10001> i was running 4 fps 1920x1080 and 265 only created 2 frames at 50 kbps [23:44:34 CEST] <MINIMAN10001> 264 just sorta did its best [23:45:59 CEST] <JEEBsv> uhh [23:46:10 CEST] <JEEBsv> I'm really not sure what exactly you're trying to say [23:46:30 CEST] <MINIMAN10001> that 264 had 4 fps they just looked unrecognizable [23:46:41 CEST] <MINIMAN10001> 265 only created 2 frames. [23:46:47 CEST] <MINIMAN10001> over 10 seconds [23:46:52 CEST] <JEEBsv> uhh [23:47:11 CEST] <JEEBsv> are you talking about it being slower or are you talking about either encoder dropping frames? [23:47:22 CEST] <JEEBsv> because yes, it's lol slow [23:47:24 CEST] <MINIMAN10001> encoder dropped frames on 265 [23:47:30 CEST] <MINIMAN10001> all of them but 2 [23:47:31 CEST] <JEEBsv> ok, did it tell you about it [23:47:53 CEST] <MINIMAN10001> dropped frames some odd 260 yeah [23:48:03 CEST] <JEEBsv> can you post a log? [23:48:08 CEST] <JEEBsv> on pastebin [23:48:17 CEST] <MINIMAN10001> it appears out of habbit i closed the cmd prompt lol [23:48:45 CEST] <JEEBsv> because I was trying to make sure you're just not having an issue between the chair and the keyboard [23:49:16 CEST] <JEEBsv> (of course it is possible that libx265 drops frames but that would be rather anticlimatic due to libx264 specifically being made so that it doesn't drop anything) [23:49:31 CEST] <JEEBsv> (and I would be surprised if libx265 dropped that design) [23:49:40 CEST] <JEEBsv> (although... it's made by MCW so the fuck I know :V) [23:50:03 CEST] <MINIMAN10001> so [23:50:08 CEST] <MINIMAN10001> how do i even give you he console output anyways [23:50:40 CEST] <MINIMAN10001> ah [23:50:42 CEST] <MINIMAN10001> select all enter [23:50:45 CEST] <MINIMAN10001> didnt know that existed [23:50:55 CEST] <MINIMAN10001> for some reason i always assume you can't highlight in cmd prompt lol [23:51:15 CEST] <JEEBsv> also if you want to just redirect an app's standard output/error stuff into a file that's also possible [23:51:32 CEST] <JEEBsv> ffmpeg uses stderr so you add " 2> file.txt" to the end of your command [23:51:37 CEST] <JEEBsv> without the "" [23:51:58 CEST] <JEEBsv> 2 is "standard error, please", > is redirect and then you have the file name :P [23:52:41 CEST] <JEEBsv> default is 1 and you get it with just a >, but ffmpeg doesn't use it when encoding so you can use it for output purposes [23:53:10 CEST] <MINIMAN10001> http://pastebin.com/ZbzpZKMd [23:53:21 CEST] <MINIMAN10001> i forgot the bitrate limits this time but it replicated either way [23:57:56 CEST] <JEEBsv> that drop is funky, esp. since you didn't even set a bit rate limit there [23:58:06 CEST] <rafael2k> People, which codec do you recommend for squeezing a HD-SDI 1.5Gb signal thru 1Gb ethernet without losing quality? [23:58:10 CEST] <JEEBsv> so it wouldn't be the lack of emergency mode, either [23:58:28 CEST] <JEEBsv> rafael2k: if it fits through something lossless, that [23:58:34 CEST] <JEEBsv> what kind of image is it? [23:58:46 CEST] <JEEBsv> as in, pixel format and bit depth [23:58:47 CEST] <rafael2k> It's for TV contribution [23:59:00 CEST] <rafael2k> I'm trying to save some money using 1Gb ethernet [23:59:02 CEST] <rafael2k> ; ) [23:59:14 CEST] <rafael2k> 10bit 422 [23:59:19 CEST] <JEEBsv> ok [23:59:40 CEST] <JEEBsv> your best bet is probably -q:v 0 -c:v libx264 with a 10bit libx264 linked in [23:59:46 CEST] <rafael2k> I was thinking about using huffman or something very simple [00:00:00 CEST] --- Sun Jun 7 2015

1 0

ffmpeg-devel.log.20150605
by burek 06 Jun '15

06 Jun '15

[03:04:20 CEST] <cone-291> ffmpeg 03Vignesh Venkatasubramanian 07master:7be0f48a3215: lavf/matroskaenc: Write Block Keyframe correctly [03:52:29 CEST] Action: llogan loves screenshots of text [03:53:08 CEST] <llogan> proper method should be console > image > OCR > libflite [04:17:43 CEST] <cone-291> ffmpeg 03Michael Niedermayer 07master:c93602ae5bd4: avcodec/rdft: Use more specific error codes [13:11:25 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:c7ce16e716a8: avutil/softfloat: Fix thresholds in av_normalize_sf() [13:11:26 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:904ea60afacd: avutil/softfloat: Add more tests for the normalization [13:51:49 CEST] <cone-824> ffmpeg 03Nedeljko Babic 07master:19ed3e35a5cd: libavutil/doc: Changes in documentation due to changes to fixed_dsp [13:51:50 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:d3585c53debe: avutil/softfloat: Move av_sincos_sf() from header to c file [17:14:48 CEST] <cone-824> ffmpeg 03Andreas Cadhalpun 07master:5bf84a584e9c: arm: only enable setend on ARMv6 [17:56:41 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:6690ca22b36e: avutil/softfloat: Fix alternative implementation of av_normalize_sf() [17:56:42 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:af8d63e756e6: avutil/softfloat: Fix alternative implementation of av_normalize1_sf() [18:23:11 CEST] <pps> any info about building ffmpeg with msvc tollchain and --disable-optimizations option? [18:23:54 CEST] <pps> it seems that all the deadcode is till referenced by ms compiler when compiled in debug/non-optimizing mode [18:24:15 CEST] <pps> is there a way to make it work without optimizations? [18:24:30 CEST] <pps> (e.g. without manually passing -O1 or something like that) [18:26:49 CEST] <nevcairiel> pps: no, that is not possible, without optimizations doesnt work with msvc .. if you are developing code and using msvc for it, you can manually disable optimizations for individual files using #pragma commands, which is what I usually use when i work on code inside ffmpeg, or debug a particular problem [18:28:12 CEST] <wm4> (this ffmpeg "requirement" is so stupid) [18:28:24 CEST] <pps> i have this idea in mind. For all code that has for if(HAVE_SOMETHING) { call_something(); } configure might automatically generate a c file with this kind of stuff: [18:29:12 CEST] <pps> #ifndef HAVE_SOMETHING [18:29:14 CEST] <pps> char call_something; [18:29:15 CEST] <pps> #endif [18:29:17 CEST] <pps> ... [18:29:46 CEST] <wm4> I'd rather just have #idef in the code [18:29:50 CEST] <wm4> directly [18:30:01 CEST] <pps> that way, it won't result in link errors in debug builds [18:30:08 CEST] <pps> in some places you cannot. [18:31:26 CEST] <pps> like in the function that registers all the codecs. Macro expands there to if(HAVE_##CODEC) avregister(codec)l; It's obviously possible to modify that macro to avoid referencing dead code (e.g. to expand to nothing instead) [18:33:07 CEST] <pps> in short, making proper ifdefs will result in many changes all over the place. Generating automatic .c file with dummy variables doesn't have any harm: if they aren't referenced they won't be included, if they are actually referenced it will be a link error, e.g. it cannot result in code trying to call a char variable as if it was a function [18:33:42 CEST] <wm4> pps: allcodecs.c should just be generated by configure or so [18:33:59 CEST] <pps> no, not a good idea IMO [18:34:22 CEST] <pps> there you can see all supported codecs, and you don't need to go to some configure script etc [18:35:06 CEST] <nevcairiel> wm4: but configure gets its list of codecs from allcodecs.c :D [18:35:18 CEST] <wm4> nevcairiel: I know... which is just backwards [18:35:28 CEST] <pps> anyways, allcodecs is minor issue compared to others. There are like hundred other places where similar if(0) call_something(); code exists [18:35:48 CEST] <pps> nevcairiel: exactly, also true :0 [18:35:50 CEST] <wm4> pps: you could also try some preprocessor magic, which wouldn't even require changes to most of allcodecs.c [18:36:11 CEST] <nevcairiel> You are unlikely to find much support for actually changing this in the code base [18:36:40 CEST] <pps> allcodecs is minor, registration macro could be expanded to c preprocessor majic to completely remove code for macro taht are 0. It'seasy. [18:36:54 CEST] <pps> but there are MANY other places that are all different. [18:38:00 CEST] <pps> forget abotu allcodecs.c, it's a simple fix. All other places aren't likely to get support for that kind of change from if(HAS_SOMETHING) to #ifdef HAS_SOMETHING [18:38:46 CEST] <nevcairiel> its unfortunate the optimizer can't just optimize these things away and leave everything else in place, however with the recent improvements to release-build-debugging features, its not too bad [18:38:56 CEST] <nevcairiel> and in fact, if you try to debug a gcc build, you run into the same problems [18:39:03 CEST] <nevcairiel> a *lot* of local variables optimized out =p [18:39:04 CEST] <pps> because in many places it's like 5 lines and each like call functions for some platform. FOr example if(HAVE_ARM) register_dsp_arm(); next like is already for x86 etc [18:39:16 CEST] <pps> with #ifdefs it will look like mess [18:40:25 CEST] <pps> the reason I want to ahev debug build is that I'm trying to debug code and I cannot see values of some variables i deeply nested code because of optimizations. [18:40:31 CEST] <nevcairiel> i'm probably one of the few regular contributors who actually use msvc primarily, and I don't even mind that much, i'm just happy i dont have to mess with gdb, is all :D [18:40:50 CEST] <nevcairiel> pps: like i said, wrap that particular code in #pragmas that disable optimizations [18:41:32 CEST] <nevcairiel> #pragma optimize("", off) [18:42:06 CEST] <pps> i know, i can find a workaround, i was just courious if a generic solution could be added to ffmpeg build [18:42:21 CEST] <BtbN> MSVC does not have a better way to turn off optimizations [18:42:41 CEST] <BtbN> gcc has -O0, but MSVC simply doesn't [18:42:52 CEST] <nevcairiel> honestly the situation is no different in gcc, people just have been dealing with it for years for ffmpeg already so they dont complain =p [18:42:55 CEST] <pps> -Od [18:43:08 CEST] <nevcairiel> i dont think gcc builds with disabled optimizer [18:43:22 CEST] <pps> i think it builds [18:45:09 CEST] <cehoyos> nevcairiel: You cannot disable dead code optimization for gcc [18:45:14 CEST] <cehoyos> (afaik) [18:47:18 CEST] <pps> cehoyos: regarding automatically generating by configure refs to deadcode to enable linking in msvc/icl. Would something like this be acceptable in ffmpeg? [18:48:12 CEST] <cehoyos> I cannot answer this question but what's wrong with using -Od or what is necessary to activate dead code elimination in msvc? [18:48:53 CEST] <nevcairiel> you need O1 [18:49:05 CEST] <nevcairiel> or well, global optimization anyway [18:49:10 CEST] <nevcairiel> which is part of O1 [18:49:15 CEST] <nevcairiel> but also removes all the local variables [18:50:26 CEST] <cehoyos> Why is this not set as cflags_noopt in configure? [18:51:34 CEST] <nevcairiel> it makes no difference [18:52:26 CEST] <cehoyos> It makes no difference if the user has to pass an option instead of automatically setting it? [18:52:43 CEST] <cehoyos> And while there: Why is _cflags_size not set to Os? [18:53:45 CEST] <cehoyos> Ah, Og is deprecated in favour of =1 [18:53:48 CEST] <cehoyos> of O1 [18:54:34 CEST] <nevcairiel> O1 is maximize speed, which should be the default for the size thing i think? [18:54:45 CEST] <nevcairiel> eh minimize size [18:54:47 CEST] <nevcairiel> i got confused [18:55:19 CEST] <cehoyos> Same here, sorry. [18:56:07 CEST] <nevcairiel> noopt could be set to Og if you really wanted to, but O1 is probably similar in generated code, the other options make no huge difference [18:56:15 CEST] <nevcairiel> Od would be appropriate, but that doesn't build [19:09:33 CEST] <pps> yes, in msvc there are no options to disable dead code elimination without optimizations. [19:10:05 CEST] <pps> ffmpeg totally relied on gcc to do so even without optimizing code, which is wrong imo (for gcc to do so) [19:10:41 CEST] <pps> there is linker option in msvc /OPT:REF, but it doesn't help [19:13:02 CEST] <nevcairiel> if the compiler still thinks its to be used, then the linker cant fix that [19:14:24 CEST] <wm4> can't you make the linker to ignore unresolved symbols? [19:15:00 CEST] <wm4> I even remember how the old Borland Pascal compiler could output object files even if the source file had semantic errors [19:20:46 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:06b49375a2cd: avutil/ppc/cpu: add include avassert.h [19:28:47 CEST] <BBB> michaelni__: why were the fixed_dsp functions added to APIChanges? that doesnt seem like it belongs there [19:30:16 CEST] <michaelni__> they will be used by the aac patches and they are in libavutil so they have to be public [19:30:27 CEST] <michaelni__> if they arent ATM then the APIChanges entry was premature [19:34:27 CEST] <michaelni__> ac3dec uses AVFixedDSPContext so i would assume its public already [19:38:09 CEST] <jamrial> softfloat.h is not an installed header, though [19:38:39 CEST] <jamrial> sorry, fixed_dsp.h [19:41:42 CEST] <wm4> these APIchanges entries are completely bogus [19:42:15 CEST] <wm4> oh, they're callbacks in the DSP context [19:43:07 CEST] <wm4> they're not public anyway, and fixed_dsp.h is indeed not an installed header, so they should be removed [20:09:25 CEST] <jamrial> does it need to be installed for that matter? it's essentially the same as float_dsp, and that one has always been internal [20:12:34 CEST] <michaelni__> jamrial, thats a philosophical question, ill just accept what the community prefers on that [20:23:02 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:6fd15d6e90b1: Revert addition of a few fixed_dsp functions to APIChanges [20:24:18 CEST] <BBB> I meant what jamrial said, yes [20:25:00 CEST] <BBB> I think stuff that is in the avpriv_ symbol range (or should be there), should be treated as ff_* for documentation purposes, i.e. not public api and not for use in applications" [20:25:18 CEST] <BBB> (not that we can tell apps what to do and what not to do, but anyway) [20:26:20 CEST] <michaelni__> yes, i misunderstood what you meant [20:43:13 CEST] <pps> regarding linking in msvc/icl. What about: instead of: http://pastie.org/10225658 it could be: http://pastie.org/10225663#3 [20:44:19 CEST] <wm4> that's the preprocessor hack I mentioned [20:44:25 CEST] <Daemon404> i think it's not going to be popular to work around msvc like that when no other compiler needs it [20:44:51 CEST] <wm4> this is about doing standard C instead of relying on compiler specifics [20:45:08 CEST] <Daemon404> it's been raised a billion times about relyign on DCE [20:45:10 CEST] <pps> agree [20:45:13 CEST] <Daemon404> i doubt this time will end any different [20:45:16 CEST] <wm4> heh [20:45:30 CEST] <wm4> ffmpeg will be ffmpeg [20:45:32 CEST] <pps> agree with wm4 O ean [20:46:31 CEST] <pps> well, then it could be added a configure step that adds all these function into a dummy file for msvc/icl debug builds only [20:46:36 CEST] <pps> that way no code change is needed [21:04:23 CEST] <jamrial> the font-awesome characters are apparently not working on fate.ffmpeg.org anymore [21:30:56 CEST] <michaelni> jamrial, do you know why ? some error /warning somewhere ? [21:36:32 CEST] <baptiste> michaelni, CORS issue I believe [21:42:25 CEST] <michaelni> thx, jamrial, beastd, should be fixed [21:45:57 CEST] <michaelni> hi lglinskih_, how is the outreachy work going ? [21:46:00 CEST] <michaelni> any patches ? [21:56:55 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:b2fecce3c16e: avformat/flvdec: increase buffer size for parsing metadata string key [21:56:56 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:790a3cdf7687: avformat/flvdec: Add TYPE_ONCAPTIONINFO [21:56:57 CEST] <cone-824> ffmpeg 03Michael Niedermayer 07master:108b738db1fd: avformat/flvdec: Show unknown tags at debug level [21:56:58 CEST] <cone-824> ffmpeg 03Ganesh Ajjanagadde 07master:a86928d2abf8: flvdec: fix lack of duration for some files [22:03:05 CEST] <BBB> so Im looking at ff_hevc_cabac_init(), do we ever carry over cabac state from one frame to the next? [22:03:10 CEST] <BBB> I mean, does that ever happen? [22:04:25 CEST] <BBB> (it looks like we dont?) [22:48:16 CEST] <nevcairiel> re: earlier discussion, having avpriv in avutil seems like a wrong design. Everything avutil should be generically useful, or it seems to not be the right fit.. [22:53:45 CEST] <jamrial> well, float_dsp can't be moved to lavc and made fully internal there because it's also used by lavfi [22:54:58 CEST] <jamrial> float/fixed_dsp could be made public, in a similar fashion as avfft is [22:55:01 CEST] <jamrial> but if nobody ever cared about using them outside of libav* then i don't think it's worth it [22:56:24 CEST] <jamrial> heh, another ticket that is solved by using pkgconfig :p [22:56:57 CEST] <nevcairiel> another thing we arent using pkgconfig for? meh [22:58:10 CEST] <nevcairiel> its odd though, the ticket claims we use pkgconfig for libx264 :d [22:58:36 CEST] <nevcairiel> or did that get the hacky workaround-y thing [22:59:05 CEST] <nevcairiel> apparently it did get a workaround-y thing with fallback for non-pkgconfig [22:59:13 CEST] <jamrial> we do, but there's a fallback [22:59:14 CEST] <jamrial> yeah [22:59:44 CEST] <nevcairiel> but yeah, pkg-config is useful, even if some people are unwilling to accept that [22:59:54 CEST] <nevcairiel> even (and maybe especially?) if you have custom build setups [23:23:39 CEST] <baptiste> OS X does not have pig-config out of the box [23:24:04 CEST] <baptiste> that's one reason I don't like pig-config :) [23:25:14 CEST] <nevcairiel> its a tiny binary, just install it [23:25:25 CEST] <nevcairiel> also, i think mine had it, i dont r emember installing it [23:25:27 CEST] <nevcairiel> maybe i just forgot [23:25:36 CEST] <baptiste> well, yeah, and actually reading .pc file is simple, you could do it in configure :) [23:27:33 CEST] <wm4> or you could just install it [23:27:51 CEST] <wm4> I didn't even have to install it when I ran osx, it somehow came with homebrew [23:57:23 CEST] <Plorkyeran> I think homebrew just assumes anything built from source needs pkgconfig [00:00:00 CEST] --- Sat Jun 6 2015

1 0

ffmpeg.log.20150605
by burek 06 Jun '15

06 Jun '15

[01:35:03 CEST] <bbloom> is there a way to set a AV_OPT_TYPE_BINARY option via AVDictionary? [01:51:02 CEST] <bbloom> oh wow, ok gotta hex encode it [02:15:42 CEST] <dgordon> Afternoon everyone - I have a question about the fastest way to scale video, I need to downres a lot of video and am currently using libx264 - I tried using nvenc but the scaling is not good with that library - its super fast but the quality is horrible when I scale - any suggestions? [02:19:35 CEST] <pzich> also please upload some screengrabs if you can [02:19:46 CEST] <dgordon> for my nvenc? or for my libx264? [02:19:51 CEST] <pzich> both [02:25:01 CEST] <dgordon> http://pastebin.com/7mhqPUYP [02:26:25 CEST] <pzich> so actual size libnvenc and libx264 look good, scaled libx264 looks good, but scaled libnvenc does not? [02:26:59 CEST] <dgordon> correct [02:28:10 CEST] <dgordon> working on some images - need to get a generic source file [02:28:13 CEST] <pzich> does it look better/different if you scale it with libx264, then actual size encode with libnvenc? [02:28:34 CEST] <dgordon> the actual size encode for libnvenc looks good [02:28:59 CEST] <pzich> right, but what about at 900 across, but not doing the -vf scale at the same time? [02:29:38 CEST] <pzich> I suspect the problem is with encoding libnvenc with those settings/resolution, not the scale itself, but this will help diagnose that [02:29:39 CEST] <dgordon> not following the question - [02:30:08 CEST] <pzich> you're doing libnvenc and the scale at the same time, right? And seeing a problem? [02:30:13 CEST] <dgordon> yes [02:30:19 CEST] <pzich> let's try them one at a time [02:30:32 CEST] <pzich> scale it with a different encoder, then encode libnvenc with it already at 900 [02:30:32 CEST] <dgordon> the source files are already 264 [02:30:44 CEST] <pzich> but not 900? [02:30:57 CEST] <dgordon> no - 1920x1080 [02:31:13 CEST] <pzich> right, let's try libnvenc with no -vf flag with an already 900 input [02:31:19 CEST] <dgordon> my end goal is to drop the res - was hoping the nvenc would do it because its so fash [02:31:44 CEST] <pzich> yep, that'd be awesome [02:31:50 CEST] <pzich> let's try to figure out what's wrong [02:31:53 CEST] <dgordon> that will look fine - the nvenc reencode from 1080p to 1080p looks good [02:32:06 CEST] <pzich> perhaps, by removing the scale step, but operating on something that's already at scale [02:32:21 CEST] <dgordon> right but the purpose here is to scale it :) [02:32:29 CEST] <pzich> yep, and that works fine in libx264? [02:32:39 CEST] <dgordon> would -s 900x505 be any different than the -vf option? [02:32:50 CEST] <dgordon> uses the same library right? [02:32:53 CEST] <pzich> it does scaling a bit differently, but I think you're better off with -vf [02:35:33 CEST] <dgordon> hmm - just crashed my machine [02:37:42 CEST] <pzich> I'm about to head out, so best of luck. My suggestion is that, since you can encode scaled with libx264, but have problems with scaling and libnvenc, try to diagnose if the problem is libnvenc or the scaling by pre-scaling with x264 and then nvenc encoding at actual size. If it looks great, something is up with scaling, if it looks shitty, it's because that's what you get with nvenc with those settings and that resolution. [02:38:09 CEST] <pzich> If the problem is with the latter, you may need to adjust your encoding settings, or it may be that you just don't get good quality at that speed, or that encoder. [02:38:38 CEST] <dgordon> is there a way to scale with 264 and encode with nvenc on the same pass? [02:39:07 CEST] <pzich> not that I know of, I'm suggesting writing it out as a file for a piece of test footage just to see [02:39:13 CEST] <c_14> x264 isn't doing the scaling [02:39:19 CEST] <pzich> right [02:45:50 CEST] <dgordon> also the end result video crashes windows media player and doesnt load into adobe products :( [02:48:34 CEST] <llogan> that's an odd git hash. i don't see it in git log [02:50:29 CEST] <dgordon> I havent been able to build this from source yet in windows - can do it easy on my mac but am running into issues on windows [02:51:40 CEST] <dgordon> found this version from here: https://github.com/MediaBrowser/MediaBrowser/wiki/GPU-Tanscoding [02:51:55 CEST] <dgordon> working on a cross-compiler on my mac right now [02:53:21 CEST] <llogan> i think you're using a fork [02:53:46 CEST] <dgordon> k - do you know of any static builds with nvenc that I can download? [02:54:25 CEST] <dgordon> (for windows) [02:54:27 CEST] <llogan> no. isn't it nonfree? if it is then it is non-redistributable [02:54:35 CEST] <dgordon> it is non-free [02:54:49 CEST] <llogan> there there should be no distributed binaries [02:54:52 CEST] <llogan> *then there [02:55:06 CEST] <dgordon> and building in windows sucks [02:55:34 CEST] <llogan> i don't know anything about that [03:22:12 CEST] <foxtrotniner> this video encoded by ffmpeg won't play on IE 11, using this line: ffmpeg -framerate 1 -i foo%d.jpg -c:v libx264 -pix_fmt yuv420p -r 25 -crf 38 test.mp4. Although it works if I use the -s hd1080 flag, but its not the resolution I want [03:23:00 CEST] <foxtrotniner> I put a custom resolution -s 4172x2000 but IE refuses to play it in html5 [03:23:53 CEST] <c_14> You sure IE supports that resolution? [03:24:34 CEST] <foxtrotniner> not sure, but chrome seems to [03:24:44 CEST] <dgordon> llogan - my problems came out of the forked version - I just built a cross compiled version on my mac and it looks great [03:24:59 CEST] <c_14> foxtrotniner: If at first you don't succeed, blame IE. [03:25:27 CEST] <foxtrotniner> yes IE is to blame for all problems [03:25:45 CEST] <pzich> it's almost certainly too big for IE, looks like a lot of 264 decoders don't support a resolution that big [03:26:47 CEST] <pzich> have you tried 4096 or 2560? not sure where the cutoff is [03:26:52 CEST] <foxtrotniner> what would be a good resolution? [03:27:01 CEST] <foxtrotniner> i guess its trial and error from here on out [03:27:11 CEST] <pzich> basically :-/ [03:27:31 CEST] <klaxa> i vaguely remember something about windows media player being unable to play portait videos, i wouldn't be surprised if IE has some stupid restictions like that too [03:28:43 CEST] <pzich> might be 4096x2304 https://msdn.microsoft.com/en-us/library/windows/desktop/dd797815(v=vs.85).… [03:29:12 CEST] <pzich> looks like the max is 1920 for Win7 [03:31:11 CEST] <foxtrotniner> hmm [03:31:21 CEST] <foxtrotniner> lets try 4096x2304 [03:32:33 CEST] <llogan> dgordon: we don't need no stinkin' forks [03:38:03 CEST] <foxtrotniner> hmm 4096x2000 seems to work [03:38:25 CEST] <foxtrotniner> but even though it still plays, IE quality can't compare to chrome [03:38:43 CEST] <foxtrotniner> i still think IE is downing the quality back to 1080 [04:56:13 CEST] <diffuse> I've been debugging an issue and it seems i am leaking memory and cant seem to figure out if it is my use of the ffmpeg api or osmething in ffmpeg itself. [04:57:23 CEST] <diffuse> i am calling avformat_alloc_context, avformat_open_input, avformat_find_stream_info, avformat_close_input, avformat_free_context [04:57:58 CEST] <diffuse> and valgrind claims avformat_alloc_context is leaking memory with calls to av_malloc [04:58:49 CEST] <klaxa> that sounds somewhat unspecific, can you provide code? [05:00:54 CEST] <diffuse> sure [05:03:03 CEST] <diffuse> http://pastebin.com/Uzx0PuFU [05:05:45 CEST] <klaxa> hmm... avformat_free_context() should clean everything up [05:05:48 CEST] <klaxa> weird [05:05:55 CEST] <diffuse> it definitely doesnt [05:06:44 CEST] <diffuse> i run the same file about a dozen times in my test [05:06:53 CEST] <diffuse> i end up leaking a little over 500k [05:07:16 CEST] <diffuse> i've got a similar piece of code that i use my own avio context, the leak doesnt occur [05:09:48 CEST] <diffuse> just as a test, i just do avformat_alloc_context/free_context, everything is good as expected [05:11:21 CEST] <diffuse> okay, the leak doesnt actually occur unless i call avformat_find_stream_info [05:13:18 CEST] <klaxa> hmm... [05:13:50 CEST] <diffuse> i've tested with the latest release and also HEAD [05:19:44 CEST] <diffuse> klaxa: so...I just explicitly set pFormatCtx->probesize2 = 4096; [05:19:50 CEST] <diffuse> and reran, no leaks [05:19:56 CEST] <klaxa> the best suggestion i can give is -- what [05:20:10 CEST] <diffuse> no other changes [05:20:51 CEST] <diffuse> wait...hold that thought [05:22:14 CEST] <diffuse> ok, no leaks when setting probesize2 explicitly [05:25:58 CEST] <klaxa> looking through the code i can't find any unfreed mallocs [05:26:17 CEST] <klaxa> at least not in combination with unset probesize2 [05:27:19 CEST] <diffuse> hm [05:27:40 CEST] <diffuse> does ffmpeg always using memalign? [05:28:11 CEST] <diffuse> i guess that doesnt matter [05:28:19 CEST] <diffuse> but i find it odd my leak doesnt occur if i set it [05:33:09 CEST] <klaxa> well the best suggestion i can give is to step through it with a debugger and try to find where memory is allocated when you don't set probesize2 [05:37:36 CEST] <diffuse> yeah, i will have to rebuild with debug flags [05:37:55 CEST] <diffuse> i dont feel like going through the assembly [05:38:03 CEST] <klaxa> i suggest adding --disable-optimizations [05:38:11 CEST] <klaxa> it helped me follow the debugger [05:38:15 CEST] <diffuse> rgr [06:15:27 CEST] <Prelude2004c> hey everyhone good evening [06:15:42 CEST] <Prelude2004c> question.. is there a good way to get ffmpeg to playout files in sequence without delay? [06:15:56 CEST] <Prelude2004c> like if i want to play one file after another without any delay ? [06:16:04 CEST] <Prelude2004c> like for a playout server [06:16:38 CEST] <klaxa> what about the concat demuxer? [06:16:54 CEST] <klaxa> https://trac.ffmpeg.org/wiki/Concatenate [07:26:31 CEST] <foxtrotniner> ffmpeg is giving an unrecognized option framerate, is there an alternative? [08:19:35 CEST] <DrBotato> i'm using the following command to try to cross fade two 1080p videos. http://kopy.io/AFjl5 [08:19:49 CEST] <DrBotato> the cross fade works fine. but the output video aspect ratio is wonky [08:20:20 CEST] <DrBotato> the strangest part is when i open the output video in VLC it is still at 1080p in the media information [08:20:45 CEST] <DrBotato> but it defaults to playing at 5:4 [08:20:50 CEST] <DrBotato> instead of playing at 16:9 [08:21:09 CEST] <DrBotato> is there some sort of header i need to add to the output video file to force it to default to 16:9 ? [08:42:22 CEST] <LigH> Hello [08:42:45 CEST] <anoop_r> hi [08:45:52 CEST] <anoop_r> how to add avisynth to ffmpeg [08:46:00 CEST] <LigH> Just compiled ffmpeg with jb_alvarado's media-autobuild_suite; ffplay refuses to play AviSynth scripts and complains that my AviSynth version is too old; but I have the brand new version 2.60 final installed. [08:46:53 CEST] <LigH> Should be a compile-time option: --enable-avisynth [08:48:03 CEST] <anoop_r> did i have to compile this first [08:48:03 CEST] <anoop_r> http://sourceforge.net/projects/avisynth2/files/AviSynth%202.6/AviSynth%202… [08:48:42 CEST] <LigH> No, you don't need to compile AviSynth as frameserver on its own. You only need to enable ffmpeg to use it when it is installed. [08:49:15 CEST] <LigH> Just found my mistake: I should not have used the 64-bit version. [08:49:22 CEST] <anoop_r> no way to add it into ffmpeg like x265 or x264 [08:50:12 CEST] <LigH> AviSynth is a frame server installed into Windows. You can't integrate its filters into ffmpeg (yet?), ffmpeg has own video filters. [08:50:12 CEST] <anoop_r> ? [08:50:38 CEST] <LigH> And x264 or x265 don't have AviSynth integrated either. [08:51:07 CEST] <LigH> You can only enable them to use the AviSynth API to load video through an installed version. [08:51:16 CEST] <anoop_r> ok [08:51:30 CEST] <anoop_r> i had build ffmpeg using mingw64 [08:51:53 CEST] <anoop_r> so did i have to use 64bit avisynth [08:52:16 CEST] <LigH> If you build a 64 bit version of ffmpeg, it would look for a 64 bit AviSynth. [08:52:39 CEST] <LigH> Only AviSynth+ (avs-plus.net) is a stable 64 bit AviSynth version. [08:53:05 CEST] <LigH> The original AviSynth (sf.net/projects/avisynth2) is 32-bit only. [08:53:11 CEST] <anoop_r> does ffmpeg works with avisynth+ [08:53:44 CEST] <LigH> Let me check. [08:53:59 CEST] <anoop_r> ok [08:55:12 CEST] <LigH> Well, it should, but the error message I get is confusing. [08:55:23 CEST] <LigH> [avisynth @ 0000000000356f80] AviSynth version is too old. Please upgrade to either AviSynth 2.6 >= RC1 or AviSynth+ >= r1718. [08:55:38 CEST] <LigH> There is no AviSynth+ r1718 available, as far as I know. [08:55:57 CEST] <LigH> Still the last build on their homepage and in their github releases is r1576 [08:56:17 CEST] <LigH> I'm already asking if there is a newer release, and where to get it... [08:57:05 CEST] <LigH> http://www.mediafire.com/download/hvnjb8wy54ys7gl/AviSynth%2B_v0.1.0_r1825-… [08:57:06 CEST] <anoop_r> did you tried avisynth 2.6 with 64 bit ffmpeg [08:57:21 CEST] <LigH> I don't have any 64-bit AviSynth 2.60 final. [08:57:31 CEST] <LigH> There is only a 32-bit AviSynth 2.60 final. [08:57:41 CEST] <LigH> So I can only use 32-bit ffmpeg with it. [08:58:01 CEST] <anoop_r> just try 32 bit with 64bitffmpeg [09:00:48 CEST] <LigH> It is not possible. A 64 bit application cannot use a 32 bit DLL. [09:01:07 CEST] <anoop_r> ok [09:01:43 CEST] <anoop_r> does AviSynth+_v0.1.0_r1825-MT works with ffmpeg 64bit [09:01:54 CEST] <LigH> Installing ... please wait ... [09:02:26 CEST] <LigH> OK, at least ffplay plays Version.avs. [09:02:46 CEST] <LigH> Now remember, AviSynth+ 64 bit cannot use 32-bit plugins. [09:03:55 CEST] <anoop_r> ffmpeg 64 uses 32 bit plugins ? [09:05:24 CEST] <LigH> No [09:07:14 CEST] <anoop_r> so this version of avisynth+ works perfectly with 64bit ffmpeg [09:07:48 CEST] <LigH> ffmpeg 32 bit will use AviSynth 32 bit which can load 32 bit plugins; there are plenty of. [09:08:11 CEST] <LigH> ffmpeg 64 bit will use AviSynth+ 64 bit which can load 64 bit plugins; there are only few, yet... [09:10:25 CEST] <anoop_r> i am trying to make a gui for ffmpeg [09:10:49 CEST] <anoop_r> so any way to filter out ffmpeg output [09:11:29 CEST] <anoop_r> like i need media info resolution fps etc form the cli [09:12:45 CEST] <LigH> Another GUI? [09:12:55 CEST] <LigH> There are already so many... [09:13:04 CEST] <LigH> Like Hybrid by Selur [09:13:18 CEST] <anoop_r> no its just for my acdemic project [09:13:35 CEST] <anoop_r> a transcoder for x265 [09:14:48 CEST] <LigH> Then I would recommend to ask that in a forum where he is a member, e.g. doom9 or videohelp. Capturing console outputs is a pretty usual task for GUI authors-. [09:16:59 CEST] <mrskman> hi, any reason why http://www.ffmpeg.org/releases/ returns 403? [09:18:56 CEST] <anoop_r> my net got disconnected [09:19:03 CEST] <anoop_r> so any idea or method [09:19:18 CEST] <LigH> Then I would recommend to ask that in a forum where he is a member, e.g. doom9 or videohelp. Capturing console outputs is a pretty usual task for GUI authors-. [09:19:47 CEST] <LigH> mrskman: No idea. zeranoe builds are fresh. [09:20:26 CEST] <anoop_r> i had build ffmpeg+x265 using msys2 [09:20:43 CEST] <anoop_r> any methode to do it in windows using msvc [09:22:03 CEST] <anoop_r> i had build it seperatly using msvc [09:22:19 CEST] <anoop_r> but no methode found to build it together [09:23:46 CEST] <LigH> I don't know how hard it would be. Since media-autobuild_suite by jb_alvarado builds both in MSYS2/MinGW64 without problems, I am satisfied. [09:24:53 CEST] <anoop_r> i too build it using mingw64 [09:25:16 CEST] <anoop_r> but failed in building using msvc [09:32:19 CEST] <LigH> Bye [09:34:12 CEST] <anoop_r> see you later [10:17:22 CEST] <nick234> Hi All, Does anyone know equivalent ffmpeg C API call for command "ffmpeg -ss start -i input.mp4 -t duration output.mp4" [10:19:08 CEST] <cehoyos> nick234: Have you looked at the files in the doc/examples folder? [10:19:40 CEST] <cehoyos> There is no answer to your question and your question indicates that you have to put some effort into a solution. [10:19:40 CEST] <nick234> no [10:22:03 CEST] <cehoyos> DrBotato: Complete, uncut console output missing. [10:25:27 CEST] <cehoyos> mrskman: The server is currently being replaced, http://ffmpeg.org/download.html works [11:39:31 CEST] <anoop_r> is it possibel to add support for heaac to ffmpeg ? [11:40:10 CEST] <BtbN> HE-AAC v2 is supported since ffmpeg 0.8 [11:40:27 CEST] <anoop_r> which codec [11:40:55 CEST] <anoop_r> means which encoder [11:41:33 CEST] <BtbN> no encoder, decoder [11:41:42 CEST] <BtbN> libfdk can propably encode it though [11:41:56 CEST] <anoop_r> i want encoder [11:42:19 CEST] <BtbN> If you don't want to use libfdk_aac, you'll have to write one first. [11:42:55 CEST] <anoop_r> libfdk have heaac [14:03:08 CEST] <termos> I'm trying to set some metadata in my flv stream by doing av_set_dict(&format_ctx->metadata, "some key", "some value", 0) but it's not showing up in my rtmp dump of the stream. Am I missing something? [14:07:21 CEST] <termos> ah, I need to pass the AVDictionary to avformat_write_header(format_ctx, &format_ctx->metadata) like so [14:43:19 CEST] <hefest> guys, any idea why when im merging 3 mp3 files of total length 25 secs , merged file is 25.75 secs long? [14:43:28 CEST] <hefest> im doing it with ffmpeg -f concat [14:53:45 CEST] <cehoyos> hefest: Please post your command line including the complete, uncut console output on pastebin [14:56:17 CEST] <hefest> cehoyos: can you tell me how to specify Hz, mono and bitrate with: ffmpeg -filter_complex "aevalsrc=0:d=5" "silence_5.mp3" ? [14:56:45 CEST] <cehoyos> I don't understand your question, sorry. [14:57:21 CEST] <hefest> cehoyos: im generating silence of particular length, i'd like it to have following spec: Stream #0:0: Audio: mp3, 22050 Hz, mono, s16p, 32 kb/s [14:57:35 CEST] <hefest> cehoyos: because im concatenating it later with the mp3 file with the same spec [15:03:30 CEST] <hefest> cehoyos: nvm, got it [15:44:20 CEST] <KrzysOsw> Hello, i need add to my camera to ffmpeg something like news bar (with scrolling text). I`m trying using drawtext from file, and it`s works,but the text does not move. How to scroll it? [15:55:55 CEST] <hefest> what does this "und" mean? Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 640x480 [SAR 1:1 DAR 4:3], 5 kb/s, 25 fps, 25 tbr, 12800 tbn, 50 tbc (default) [15:58:34 CEST] <hefest> https://gist.github.com/akomic/2f6e7a0e47dbfcf0865a [15:58:37 CEST] <hefest> any ideas? [16:04:34 CEST] <relaxed_> hefest: -map 0 -map 1 [16:06:03 CEST] <relaxed_> or -map 0:0 -map 1:0 [16:09:29 CEST] <relaxed_> you tried -map 0:1, which means the second stream of the first input, but it only contains a video stream (0:0) [16:16:01 CEST] <spaam> hefest: undefined? [16:22:52 CEST] <zumba_ad_> hey folks, i have a 1080i file. I need to deinterlace it. what is the param for it? [16:26:32 CEST] <c_14> https://ffmpeg.org/ffmpeg-filters.html#yadif [16:27:15 CEST] <zumba_ad_> also found an answer but a bit confusing - http://superuser.com/questions/253691/how-to-convert-108050i-72050p-using-f… [16:32:04 CEST] <c_14> The easiest case is just -vf yadif, if you want one frame per field, -vf yadif=mode=1, if you're not sure all frames are interlaced -vf idet,yadif=deint=interlaced, if you want you can throw an mcdeint after the yadif [16:32:26 CEST] <c_14> If the video is also telecined you'll need to use an inverse telecine like pullup/decimate [16:33:26 CEST] <zumba_ad_> got it working, ffmpeg -i c.mov -vf yadif=1:-1:0 -vcodec libx264 c.mp4 [16:35:23 CEST] <zumba_ad_> i'll try your params too [16:36:03 CEST] <zumba_ad_> it worked and it produced a smaller file. I haven't played it yet [16:36:17 CEST] <c_14> You might want to adjust the crf etc [16:36:51 CEST] <zumba_ad_> i notice the playback was smoother on the first one i made [16:37:29 CEST] <hefest> relaxed_: got it, thanks. i though -map 0:1 means 0 stream from input goes to 1 stream to output, and position of the -map switch on the command line determines the input. obviously wrong [16:38:47 CEST] <KrzysOsw> What with my news bar? :) [16:41:27 CEST] <c_14> KrzysOsw: update the "news ticker" 2 characters at a time. Delete a character from the front and add one to the back [16:41:32 CEST] <c_14> Might make it look like it's scrolling [16:41:37 CEST] <c_14> You'd have to get the timing down right though. [16:42:31 CEST] <zumba_ad_> what is this warning folks? [h264 @ 0x7fdcc5649600] mmco: unref short failure [16:43:15 CEST] <KrzysOsw> Too bad that ffmpeg not support something like a news ticker... [16:45:16 CEST] <zumba_ad_> does this output tell that the video is interlaced? I know it's interlaced because I see the lines when i play it [16:45:19 CEST] <zumba_ad_> Stream #0:0(eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], 11576 kb/s, 29.97 fps, 29.97 tbr, 2997 tbn, 59.94 tbc (default) [16:45:47 CEST] <zumba_ad_> or is there a better tool for checking interlace or not [16:45:59 CEST] <c_14> ffmpeg -i video -vf idet -f null /dev/null [16:46:13 CEST] <zumba_ad_> ok, trying it [16:46:17 CEST] <[-T-]> hi, by change, has anyone here played with QSV ? [16:46:38 CEST] <zumba_ad_> c_14: it's like it's playing it [16:46:47 CEST] <c_14> zumba_ad_: ja, it'll take a while [16:46:53 CEST] <zumba_ad_> ah :D it's done [16:47:09 CEST] <c_14> It'll tell you how many frames are tff bff and progressive [16:47:14 CEST] <zumba_ad_> http://pastebin.com/4E1reQQV [16:47:30 CEST] <zumba_ad_> i don't know how to analyze the output [16:47:53 CEST] <zumba_ad_> so progressive 0 which means, it's interlaced [16:48:16 CEST] <c_14> tff = top field first [16:48:20 CEST] <zumba_ad_> am I right since 1 has always been mapped to true [16:48:43 CEST] <zumba_ad_> what is top field first [16:49:10 CEST] <c_14> You know how interlacing splits a frame into fields? [16:49:18 CEST] <c_14> tff means that the top field is sent first [16:49:24 CEST] <zumba_ad_> ah, check this out [16:49:26 CEST] <zumba_ad_> [Parsed_idet_0 @ 0x7fdf38d00000] Single frame detection: TFF:4 BFF:2 Progressive:264 Undetermined:40 [16:49:38 CEST] <zumba_ad_> i ran it against the one i recently created. Progressives has non-zero [16:50:09 CEST] <zumba_ad_> thanks for the syntax [17:31:37 CEST] <DrBotato> cehoyos: console output: http://kopy.io/vHVpO [17:32:59 CEST] <DrBotato> Input ... 1920x1080 [SAR 1:1 DAR 16:9] [17:33:09 CEST] <DrBotato> output ... 1920x1080 [SAR 3:4 DAR 4:3] [17:33:17 CEST] <DrBotato> same pixel amount, but aspect ratio changes [17:33:21 CEST] <DrBotato> how do i force 16:9 ? [17:33:51 CEST] <DrBotato> (this is the command i'm using: http://kopy.io/AFjl5) [17:36:38 CEST] <JonG> I have a number of ffmpeg command line scripts that do things like cropping, merging, fading between videos etc. I was hoping to include the same methodology in the Ffmpeg C API library. Are there any good resources for effectively translating between the two systems? I found the tutorials by Stephen Dranger (e.g. https://github.com/chelyaev/ffmpeg [17:36:38 CEST] <JonG> -tutorial), but they seem to be focused on displaying the video, rather than editing. [18:25:31 CEST] <Anoia> JonG: does the source for the ffmpeg utility give you a clue? [18:27:48 CEST] <cehoyos> DrBotato: [2:v]scale=1920x1080,trim=duration=142[over] is missing setdar=16/9 [18:28:03 CEST] <DrBotato> thanks! [18:28:12 CEST] <cehoyos> You are using the "color" input as background for overlay and (luckily) the background is used as reference for the output. [18:28:39 CEST] <cehoyos> And please remember to test current FFmpeg (that is current FFmpeg git head) when asking questions or reporting bugs [18:28:46 CEST] <DrBotato> so like this: [2:v]scale=1920x1080,trim=duration=142,setdar=16/9 [18:29:02 CEST] <DrBotato> yeah i might be on an older ffmpeg [18:29:06 CEST] <cehoyos> I didn't test test but yes, that looks as if it works. [18:29:06 CEST] <DrBotato> been a while since i installed it [18:29:23 CEST] <DrBotato> thanks for your help [18:29:57 CEST] <cehoyos> np [18:30:42 CEST] <DrBotato> yeah, looks like head is 2.6 [18:30:45 CEST] <DrBotato> and i'm on 2.3.3 [18:30:56 CEST] <DrBotato> probably should upgrade, maybe its a bit faster ;) [18:35:13 CEST] <cehoyos> DrBotato: My point was that "current FFmpeg" is NOT 2.6 [18:35:23 CEST] <cehoyos> But current FFmpeg git head. [18:38:18 CEST] <JonG> Anoia: So for example if I want to implement the equivalent of -filter_complex fade, you are suggesting look at the source code of filter_complex and look up the fade option? Is that correct? [18:44:47 CEST] <cehoyos> JonG: No, you should look at doc/examples/filtering_video.c [18:48:55 CEST] <JonG> cehoyos: OK, thanks, I'll take a look at those examples. [19:25:58 CEST] <DrBotato> If i want to add a couple frames of text on a colored background with a certain font, is this possible with ffmpeg? or should i create a still frame first and then use the still frame with ffmpeg [19:26:43 CEST] <c_14> DrBotato: I think the box option of drawtext should do it [19:33:17 CEST] <DrBotato> [AVFilterGraph @ 0x7fd512d014c0] No such filter: 'drawtext' [19:33:43 CEST] <DrBotato> http://kopy.io/r6tVW [19:33:43 CEST] <c_14> Did you build with libfreetype support? [19:33:54 CEST] <DrBotato> i just got the latest version from homebrew. [19:33:58 CEST] <DrBotato> so i'm on 2.6.3 [19:34:05 CEST] <DrBotato> not sure if that includes libfreetype [19:34:47 CEST] <c_14> Looks like it doesn't, you probabli need --with-freetype or something [19:35:06 CEST] <DrBotato> ahh i see [19:41:05 CEST] <DrBotato> ok sweet that fixed it. thanks! [20:00:36 CEST] <DrBotato> ok i've mostly got this figured out, but with the timeline editing capabilities of drawtext, is it possible to insert frames instead of having the drawtext overlap on frames? ie. video for 10 frames. draw text for 4 frames. video resumes. [21:56:23 CEST] <Zucca> Does/will ffmpeg have support for lagarith encoding? [21:57:42 CEST] <durandal_1707> Nope [21:58:24 CEST] <DrBotato> Can anyone tell me whats wrong with my ffmpeg command [21:58:25 CEST] <DrBotato> http://kopy.io/8B6Ko [22:03:36 CEST] <t4nk885_> Hi. I'm trying to decode H264 data that was extracted from an mp4 file using libavcodec. (Not by me) [22:03:38 CEST] <t4nk885_> I've created a codec context for CODEC_ID_H264 (and initialized its width and height) [22:03:41 CEST] <t4nk885_> I also have the avcC data, which I've placed in the codec context's extradata (& extradata_size) [22:03:45 CEST] <t4nk885_> I've created an av_parser for the codec and can successfully feed it data, having it return packets to me. [22:03:54 CEST] <t4nk885_> However, avcodec_decode_video2 seems to always fail for those frames, with the error "no frame!". [22:03:57 CEST] <t4nk885_> Is there something I'm missing? E.g. something else I need to initialize from the original mp4 file? [22:08:41 CEST] <t4nk885_> Sadly I don't have libav's pdbs (not downloadable?) and I doubt I'd be able to compile from original source easily (it seem to use gcc extensions?) [22:09:10 CEST] <t4nk885_> So I'm sort of relegated to poking at the apis and reading the libav sources to try to figure out what's going on. [22:25:51 CEST] <zumba_ad_> hey guys, i can't properly merge my 2 mp4 files. They have the same encoding parameters [22:26:00 CEST] <zumba_ad_> but i'm using MP4Box [22:26:10 CEST] <zumba_ad_> i used this tool few years ago and it worked well [22:26:12 CEST] <zumba_ad_> any suggestions? [22:52:35 CEST] <t4nk336> Maybe someone has a libavcodec x64 exe+pdb pair lying around? [22:53:01 CEST] <t4nk336> What are the chances of that? [22:54:23 CEST] <zumba_ad_> hey folks, what's wrong with this? it's not combining the 2 m4a files - ffmpeg -i new.m4a -i output128.m4a combined.m4a [22:54:39 CEST] <c_14> -map 0 -map 1 [22:54:41 CEST] <zumba_ad_> i mean, it created combined.m4a but the audio from second file is missing] [22:54:44 CEST] <c_14> and probably -c copy [22:54:50 CEST] <zumba_ad_> oh [22:54:53 CEST] <zumba_ad_> thanks [22:55:27 CEST] <zumba_ad_> no, didn't work, it still didn't append it [22:55:37 CEST] <c_14> https://trac.ffmpeg.org/wiki/Concatenate [22:58:54 CEST] <t4nk336> I've done pretty much anything the h264dec.c example did but I still can't decode my data, and the only error is "no frame". (I'm the guy who left a few minutes ago talking about the same thing) [23:00:57 CEST] <t4nk336> Then again mine is an avc1 h264 stream, while that example seems concerned with raw ones - but I'm pretty sure I did all that's required to handle avc1 (set extradata). [23:01:02 CEST] <t4nk336> Of course, I can't be sure without the pdbs [23:05:19 CEST] <t4nk336> I guess I'll try to build ffmpeg on windows [23:06:18 CEST] <zumba_ad_> gheez, i can't combine mp4 files properly :( i've been troubleshooting it since morning [23:07:27 CEST] <DrBotato> heh i'm in the same boat [23:07:41 CEST] <DrBotato> as soon as first clip ends, all remaining frames drop [23:10:38 CEST] <zumba_ad_> i was able to concatenate the m4a files but not the non-audio mp4 files [23:10:47 CEST] <zumba_ad_> ffmpeg -f concat -i video.txt -c copy v.mp4 [23:11:06 CEST] <BtbN> remux them to mpeg-ts [23:11:22 CEST] <zumba_ad_> what tool and what are the params please? [23:11:26 CEST] <BtbN> ffmpeg [23:11:32 CEST] <zumba_ad_> k [23:11:33 CEST] <BtbN> just stream copy to a .ts file [23:12:08 CEST] <zumba_ad_> how is that done? [23:12:59 CEST] <zumba_ad_> got it [23:13:14 CEST] <DrBotato> BtbN: got any idea what i'm doing wrong here? http://kopy.io/ I'm trying to take a 4 second video clip and add a couple seconds of text at the end (after the video clip ends, and extending the timeline of the resulting video). [23:13:22 CEST] <zumba_ad_> it worked [23:13:46 CEST] <DrBotato> its gives this erro: Past duration 0.999992 too large [23:13:47 CEST] <zumba_ad_> now i need to combine this new mp4 and m4a, so -vcodec copy -acodec copy right [23:13:50 CEST] <DrBotato> and then drops all other frames [23:20:57 CEST] <DrBotato> anyone? [23:21:02 CEST] <DrBotato> this is driving me totally crazy [23:33:34 CEST] <t4nk336> oh boy [23:33:37 CEST] <t4nk336> As I thought [23:33:51 CEST] <t4nk336> There's no way I'm going to compile the damned thing in my MSVC [23:33:57 CEST] <t4nk336> Which isn't even a supported version [23:34:37 CEST] <t4nk336> Wait nevermind (yet), those are warnings rather than errors. maybe this will even work(!) [00:00:00 CEST] --- Sat Jun 6 2015

1 0