Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

December 2019

1 participants
62 discussions

ffmpeg-devel.log.20191130
by burek 01 Dec '19

01 Dec '19

[03:20:17 CET] <cone-902> ffmpeg 03James Almer 07master:18507b48825d: avformat/avc: fix sps buffer offset when calling ff_avc_decode_sps() [11:04:59 CET] <Lynne> I'm suprised, 2 people used to sign their commits via gpg but they stopped some time ago [13:31:43 CET] <thardin> https://www.youtube.com/watch?v=ym1brc2OcYQ Kdenlive is a Garbage Fire (but there's a better tool) [13:31:48 CET] <thardin> does libmelt not use multithreading? [13:32:04 CET] <thardin> last time I poked with it it seemed p good [13:33:01 CET] <ubitux> i confirm kdenlive is garbage [13:33:36 CET] <Lynne> I heard blender is pretty good [13:34:20 CET] <ubitux> last thing i tried was olive, it was ok, even though it was pretty basic with basically nothing [13:35:00 CET] <ubitux> last shitty kdenlive bug i had was not being able freeze a frame and crop at the same time [13:35:21 CET] <ubitux> seems it was a bug present in libmelt for years [13:35:25 CET] <thardin> yeah kdenlive is terrible, but I had at least thought libmelt was decent enough [13:35:48 CET] <thardin> I've successfully used it for the command-line editing [13:35:54 CET] <thardin> s/the/some/ [13:46:17 CET] <durandal_1707> how do you create vtable inside ghidra? [13:47:23 CET] <Lynne> what made you change from ida? the disasm looks worse than that [13:48:17 CET] <durandal_1707> ida is piece of shit [13:49:07 CET] <durandal_1707> the only ida i have does not play well with class inspector/informer, unless you buy me latest version i will continue to use ghidra [13:51:12 CET] <durandal_1707> i tried various scripts that do this for ghidra, but they all basically fail [13:51:15 CET] <ubitux> durandal_1707: there is a help for advanced stuff like this available in the doc [13:51:30 CET] <ubitux> ah, you meant automatic [13:51:56 CET] <ubitux> always done it manually so i dunno [13:52:06 CET] <durandal_1707> ubitux: even manual if good to know [13:52:36 CET] <JEEB> has ghidra been improved during the time post-initial release? [13:52:42 CET] <durandal_1707> i see in assembly view just list of functions [13:52:43 CET] <ubitux> durandal_1707: check GhidraDocs/GhidraClass/Advanced/improvingDisassemblyAndDecompilation.pdf [13:52:54 CET] <ubitux> there are a few slides about this [13:52:56 CET] <durandal_1707> i read that.. [13:53:22 CET] <ubitux> JEEB: yeah it improved quite significantly between releases [13:53:27 CET] <ubitux> at least wrt bugs [13:53:38 CET] <JEEB> nice [13:54:11 CET] <ubitux> they still haven't adressed my simd issue though [13:54:40 CET] <ubitux> https://github.com/NationalSecurityAgency/ghidra/issues/249 [13:55:21 CET] <ubitux> i should probably open an issue without the question label, since i'm unable to drop it [13:55:59 CET] <durandal_1707> that is not garbage [13:56:10 CET] <durandal_1707> its valid decompiler code [13:56:12 CET] <ubitux> it's doing better with arm assembly though [13:56:19 CET] <ubitux> durandal_1707: that's not the point [13:57:09 CET] <durandal_1707> do you want original code with intrinsic stuff? or just better output [13:57:32 CET] <ubitux> something close to the original intrinsic [13:57:46 CET] <ubitux> but whatever the syntax they want as long as it is readable [13:57:52 CET] <ubitux> they actually do that for neon [14:00:51 CET] <cone-349> ffmpeg 03Andreas Rheinhardt 07master:76e0ecec0bfa: avcodec/extract_extradata_bsf: Don't unref uninitialized buffers [14:02:28 CET] <durandal_1707> for creating vtable structure, i need to know singature of each function in vtable, this is not possible to guess i think? [14:05:21 CET] <ubitux> i haven't played with a c++ codebase for a while, but i remembed having to create individual types for each method prototype [14:05:40 CET] <ubitux> and then use them in the struct [14:05:57 CET] <ubitux> there is probably a more efficient way of doing that nowadays [14:06:02 CET] <ubitux> and likely a plugin to help with that [14:10:10 CET] <durandal_1707> i tried plugin and it help only little [15:06:48 CET] <Lynne> one thing which I find kinda sketchy about gpu filters is that they're sometimes not bitexact [15:07:10 CET] <Lynne> like our nlmeans_opencl filter doesns't output the same results compared to the cpu one [15:07:36 CET] <Lynne> and I think it even differs between vendors, since I can vaguely remember it behaving differently under nvidia and intel [15:23:12 CET] <ubitux> Lynne: because float [15:23:40 CET] <ubitux> also yeah, you can get random results depending on the chip, version of the drivers, etc [15:23:47 CET] <ubitux> just like with compilers, systems, cpus [15:24:30 CET] <ubitux> float and bitexact are a pita to deal with, you need to rely on thresholds for all the tests [15:24:39 CET] <Lynne> its not like gpus don't have integer vectors too, you know [15:25:22 CET] <Lynne> but you can't sample images then, so that's bad [15:27:03 CET] <Lynne> and not sampling images means your code also runs slower because caching just isn't a big thing with reading storage images (from memory either nvidia or intel is slower, but not both) [15:29:00 CET] <ubitux> let's be honest, they're not designed for dealing with ints at all [15:57:06 CET] <kierank> you can get different results depending on some value of float [15:57:31 CET] <kierank> I have seen some crazy optimisations that really matter between 98 and 99 or something [15:57:34 CET] <kierank> but all other values were fine [16:03:48 CET] <cone-349> ffmpeg 03Andriy Gelman 07master:99d78e4f4249: lavc/extract_extradata: Use bytestream api [16:03:49 CET] <cone-349> ffmpeg 03James Almer 07master:e5db2e96c02d: avformat/avc: add missing return error value [23:26:44 CET] <durandal_1707> fun, braw uses internally 420 subsampling [00:00:00 CET] --- Sun Dec 1 2019

1 0

ffmpeg.log.20191130
by burek 01 Dec '19

01 Dec '19

[00:59:01 CET] <sector_surfer> Hi. [00:59:25 CET] <sector_surfer> What commandline can I use to slow down the download rate of a stream when downloading form an url? [01:00:13 CET] <sector_surfer> for example, I want to limit the download rate of ffmpeg to say 15 k, because I've got other programs who's download is more importantthat are [01:00:31 CET] <sector_surfer> whoops [01:01:22 CET] <sector_surfer> whenever I download from a facebook livestream via .mpd, there are parts of the video that repeat itself [01:03:38 CET] <sector_surfer> https://video.fmel5-1.fna.fbcdn.net/hvideo-prn1-ftw/_nc_cat-103/v/rVU7tYWnF… [01:09:54 CET] <DHE> I suggest something like: curl http://<your-url> --limit-rate=15k | ffmpeg -i - [...] [01:15:17 CET] <sector_surfer> I'm using windows, not linux [01:20:00 CET] <LevelDrain> hello [01:20:14 CET] <LevelDrain> i have a better way for add list of musics to a video? [01:21:31 CET] <b1rk0ff> is there a filter to filter out parts of an audio stream that are below certain 'decibels' or sound level ? [01:26:13 CET] <kepstin> b1rk0ff: https://ffmpeg.org/ffmpeg-filters.html#agate [01:26:25 CET] <kepstin> hmm, well, that might not be what you want actually [01:26:41 CET] <kepstin> that just allows you to completely mute sections below a certain sound level [01:27:28 CET] <kepstin> b1rk0ff: maybe https://ffmpeg.org/ffmpeg-filters.html#silenceremove [01:28:04 CET] <b1rk0ff> I want to actually remove/not reencode low sound level sections [01:28:56 CET] <b1rk0ff> silencremove is only for silence on the beginning/end [01:43:14 CET] <LevelDrain> The only way I can find is to concatenate all the music in a single music file, and then put the music together with the video. Isn't there a way to do it all in one step? [01:44:21 CET] <kepstin> b1rk0ff: might i suggest re-reading the silenceremove docs :) [06:17:20 CET] <void09> how would I extract dvdsub subtitles from an mkv file ? [06:17:45 CET] <void09> ffmpeg -i "input.mkv" -map 0:s:0 -c copy subtitle_file.vob.sub [06:17:57 CET] <void09> [microdvd @ 0x5606cee3e180] Exactly one MicroDVD stream is needed. [06:17:58 CET] <void09> Could not write header for output file #0 (incorrect codec parameters ?): Operation not permitted [06:26:55 CET] <furq> there is no vobsub muxer [06:26:59 CET] <furq> mkvextract will do it iirc [06:30:05 CET] <furq> void09: mkvextract foo.mkv tracks 2:bar.idx [06:30:14 CET] <furq> or whatever the track id from mkvmerge --identify is [06:33:13 CET] <void09> got it :) [06:33:17 CET] <void09> writing everything down [06:33:28 CET] <void09> is there no cheatsheet with all this stuff in one place [06:33:30 CET] <void09> everything to everything [06:33:47 CET] <furq> that'd be more than one sheet [06:33:48 CET] <nicolas17> there's like 300 formats [06:34:05 CET] <void09> well, then a smart sheet, input type, output type, object type [06:34:21 CET] <void09> DVD SUBS TO .VOB/.SUB [06:34:21 CET] <void09> mencoder VTS_01_0.IFO -nosound -ovc frameno -o /dev/null -vobsuboutindex 0 -sid 0 -vobsubout subEN.vob [06:34:37 CET] <void09> see, i write everything down as I find it. sometimes the answers are not easy to find in the first google result page [06:34:47 CET] <void09> well, they usually have windows programs as answers [06:35:10 CET] <furq> is mencoder still developed [06:35:57 CET] <void09> probably not, but this is what i found and works [06:36:15 CET] <furq> saying that i'd probably use tcdemux for that and that definitely isn't developed any more [06:36:56 CET] <void09> I'll have to wind a near fool proof way of converting .vob/sub to .srt oen day [06:37:08 CET] <furq> there's no good non-ocr way of doing it [06:37:21 CET] <void09> ocr is the only way, what do you mean [06:37:31 CET] <furq> even the ocr methods aren't great [06:37:39 CET] <void09> what non-ocr methods are there? [06:38:01 CET] <void09> I find it pretty amazing what AI can do these days, and still there's no good way of turning dvd subs to text ? lol [06:38:37 CET] <furq> i meant no good non-gui method of doing it [06:38:51 CET] <furq> the cli stuff i know of uses tesseract but it has some kind of database of common fonts [06:38:59 CET] <furq> which is still technically ocr, just less tedious [06:39:01 CET] <void09> right, and why does that not work well ? [06:39:20 CET] <nicolas17> dvd subs have crappy resolution, for a start [06:39:45 CET] <void09> well those can be upscaled easily i think [06:40:16 CET] <furq> honestly i normally just leave them as image subtitles now [06:40:40 CET] <furq> l'll [06:40:52 CET] <furq> sorry i mean l,ll [06:44:02 CET] <void09> yeah I leave them as images too, until i find a good way to convert [09:21:09 CET] <funnybunny2> Is there a table which shows which audio encodings support planar, interleaved, or both sample formats? [12:16:36 CET] <Arnob> Hi, could someone tell me what I'm doing wrong here. When using the following command to generate a waveform and blend it with a pic, if I input a mp3 as audio it works fine but if I input a WAV file I've got errors. [12:16:50 CET] <Arnob> The command I use: ffmpeg -y -i 12.WAV -loop 1 -i b.jpg -filter_complex "[0:a]showwaves=s=1280x720:mode=line,colorkey=0x000000:0.01:0.1,format=yuva420p[v];[1:v][v]overlay[outv]" -map "[outv]" -pix_fmt yuv420p -map 0:a -c:v libx264 -c:a copy -shortest output1.mp4 [12:18:48 CET] <Arnob> the errors I receive when using a wav : https://arnauld.org/zerobin/?ee89459bd8bd6f9e#qGDPJmgMLRQlpzBAWcrzO0f/uZd90… [12:20:00 CET] <c_14> can't copy wav into mp4 [12:20:04 CET] <c_14> reencode the audio [12:21:26 CET] <Lynne> funnybunny2: AVCodec->sample_fmts[] [12:22:04 CET] <Arnob> So I should reencode to wav? [12:22:45 CET] <c_14> It's already wav (pcm) you're going to have to encode to mp3, aac, alac (in mov) or opus [12:22:56 CET] <c_14> or use .mkv instead of .mp4 [12:25:33 CET] <Arnob> ok, so to get the best audio, is it better to reencode to mp3 first then input the mp3 or keep the wav and output to mkv? [12:25:42 CET] <funnybunny2> Lynne: I'm not talking about a single file. I'd like to know in terms of the specifications of audio codecs, which allow planar, interleaved, or both [12:25:57 CET] <c_14> Arnob: you'll get the best audio by putting it in mkv. Because then you won't have to reencode it [12:27:16 CET] <funnybunny2> I'm trying to decide if I need to support both planar and interleaved, or if the encodings I'm interested in all support one format or the other [12:27:24 CET] <Arnob> c_14: ok, thank you! [12:27:58 CET] <pink_mist> c_14: eh, if it's wav it's already lossless, so any other lossless encoding would work exactly as well [12:28:06 CET] <pink_mist> such as alac or flac [12:28:19 CET] <c_14> sure, but alac in mov isn't particularly common [12:28:25 CET] <c_14> at that point might as well just stick to mkv [12:29:04 CET] <funnybunny2> Am I misunderstanding? Does it depend on the specific decoder FFmpeg uses rather than the file format whether interleaved, planar, or both sample formats are supported? [13:26:07 CET] <funnybunny2> Arnob: You asked about getting the best audio. If you have a lossless [13:26:24 CET] <funnybunny2> Arnob: ...wav file, it's obviously going to be worse if you encode it as mp3 [13:28:05 CET] <funnybunny2> Rereading, it seems like you're going to put the wav in an mkv. If that works, then good [14:41:14 CET] <safinaskar> nut has very cool feature: it can store various types of uncompressed raw video: https://git.ffmpeg.org/gitweb/nut.git/blob/670ff4339ed54f779ece2086fe8dbe9c… . are there any other containers with similar feature? [14:46:04 CET] <DHE> mkv is probably the "best" container in terms of features... [15:13:05 CET] <LevelDrain> i have a better way for add list of musics to a video? [15:13:11 CET] <LevelDrain> The only way I can find is to concatenate all the music in a single music file, and then put the music together with the video. Isn't there a way to do it all in one step? [15:17:33 CET] <DHE> so you want a video to have an audio track that is a concatenation of all the music files you have? [15:19:49 CET] <LevelDrain> DHE, yes, a video with a list of musics in the background... [15:24:13 CET] <DHE> there are a few 'concat' options... file format, filters [17:20:32 CET] <wowa_putin> test [17:21:54 CET] <wowa_putin> hello everybody! How to fight problem of time drift from rtp h264 stream from IP Camera [17:25:29 CET] <wowa_putin> for the moment i use the following code to obtain timestamp: picture->pkt_dts * av_q2d(stream->time_base) [17:26:29 CET] <wowa_putin> it works fine for 15-20 mins. then there appears time drift that seems to left behind real time [17:26:33 CET] <wowa_putin> what to doen? [17:28:05 CET] <JEEB> so the pts/dts values stop going upwards properly? [17:28:44 CET] <JEEB> RTP should have timestamps so it sounds like the input gets funky [17:32:18 CET] <LevelDrain> DHE, what?... [17:32:27 CET] <wowa_putin> hm, i am gonna give u more details [17:37:09 CET] <wowa_putin> I have got 3 devices: very Jai go speed camera that gives me raw frames with timestamps, doppler effect radar that also gives me kind of frame (no pictures) with timestmap, and IP Camera. All timestamps of those devices are normalized to monotonic clock of pc. So, radar and jai camera work together just fine. Ip camera also for 10-20 mins, but then starts being behind. Do i make myself clear? [17:42:20 CET] <wowa_putin> so, having that fact that Jai camera and Doppler radar work fine, i concluded there is problem with rtsp camera. [17:46:53 CET] <shiin> I'm looking for a cli tool to search an audio track for noise spikes, to then print the timestamp of the audio/video file. Can ffmpeg do that? [18:01:59 CET] <cousin_luigi> Greetings. [18:03:45 CET] <cousin_luigi> I have a mpeg 4 video that corrupts during seek: what can I do to find out what's causing it and what can I do to improve the situation? [18:05:54 CET] <LevelDrain> Greetings, mortal. Are you ready to die? [18:07:33 CET] <LevelDrain> oh... it's ffmpeg... sorry... [18:07:47 CET] <LevelDrain> bye... [18:29:41 CET] <cousin_luigi> Can I add keyframes to a mpeg4 stream without recoding everything? [18:30:23 CET] <pink_mist> not a chance [18:30:58 CET] <cousin_luigi> ok [18:48:32 CET] <cousin_luigi> ok, recoding fixed it. bbl! [21:19:01 CET] <dinki> Hey guys. I'm trying to upgrade to 4.2 on my Raspberry Pi Zero running rasbian buster and am having a whale of a time doing it. Is the only way to get it running by building it? [21:19:44 CET] <JEEB> there are some armv7 binaries built by some guy linked on the download page, but in general yes. I do recommend setting up a cross-compilation setup, though [21:20:02 CET] <JEEB> so that you can use debian's ARM toolchain with a raspi sysroot [21:20:06 CET] <JEEB> to build stuff on x86_64 [21:20:15 CET] <JEEB> and then just moving the binaries to the raspi [21:20:38 CET] <DHE> rather than letting the Pi chew on building ffmpeg for an hour or two (?) [21:21:10 CET] <JEEB> I always wondered why raspbian didn't offer something like that or documentation on how to get it set up [21:21:30 CET] <JEEB> although in theory it shouldn't be too hard [21:21:35 CET] <JEEB> 1) get sysroot under some dir [21:21:38 CET] <dinki> TBH, I'm a fairly experienced linux user, but I'm very clumsy when it comes to building from source. I tried using instructions from https://trac.ffmpeg.org/wiki/CompilationGuide/Ubuntu but am getting dependency issues and other isues [21:21:42 CET] <JEEB> 2) try using the debian cross-compiler [21:23:13 CET] <dinki> I do have an x86 debian server (Open Media Vault). [21:23:50 CET] <furq> yeah it's pretty easy on debian/ubuntu/etc [21:24:47 CET] <furq> pretty much just install crossbuild-essential-armhf and then install the deps you need with multiarch [21:24:50 CET] <furq> https://wiki.debian.org/Multiarch/HOWTO [21:25:37 CET] <dinki> Linux sign1 4.19.75+ #1270 Tue Sep 24 18:38:54 BST 2019 armv6l GNU/Linux [21:25:45 CET] <dinki> So I'd build for armv6l ? [21:25:50 CET] <JEEB> yup [21:26:54 CET] <furq> i forget if you need armel or armhf on raspbian [21:27:26 CET] <DHE> is that reported somewhere? /proc/cpuinfo or uname -a ? [21:27:50 CET] <dinki> I'm pretty sure I've downloaded both precompiled binaries for armel and armhf and neither of them ran on my rpi [21:28:17 CET] <dinki> @dhe see the uname I posted just above [21:28:30 CET] <DHE> ah there it is [21:28:36 CET] <furq> oh nvm it's a pi zero [21:28:39 CET] <furq> that is armel then [21:28:51 CET] <dinki> So I can download a precompiled armel version? [21:28:57 CET] <furq> that should work [21:29:01 CET] <furq> i've never used a zero though so shrug [21:29:14 CET] <furq> the precompiled armhf version runs on my (non-raspbian) pi 2 [21:29:15 CET] <dinki> hmm. Okay, let me try that again. Pretty sure it seg faulted [21:30:19 CET] <dinki> linux-armel , right? [21:30:59 CET] <furq> https://johnvansickle.com/ffmpeg/releases/ffmpeg-release-armel-static.tar.xz [23:07:59 CET] <dinki> @forq I tried that version and it seg faulted as before [23:09:31 CET] <JEEB> yea, so you'll need to compile or cross-compile (latter preferably) it for the older ARM arch that the zero still utilizes [23:17:49 CET] <dinki> @JEEB is there a guide for compiling from source on the raspberry pi? I'd rather just bite the bullet and do it rather than set it up to crosscompile only to be disappointed if it doesn't work. [23:19:18 CET] <dinki> by the way, this is the error I'm getting if anyone can tell me what it might be from: [23:19:19 CET] <dinki> /usr/bin/ffmpeg -loglevel error -y -f v4l2 -i /dev/video0 -f pulse -i default -codec:v libx264 -pix_fmt yuv420p -preset veryfast -b:v 500k -g 60 -codec:a aac -b:a 128k -ar 44100 -maxrate 500k -bufsize 250k -strict experimental -f flv [23:19:19 CET] <dinki> rtmps://live-api-s.facebook.com:443/rtmp/111103643332502?s_bl=1&s_ps=1&s_sw=0&s_vt=api-s&a=AbzwuWpQlJ3859ds [23:19:19 CET] <dinki> default: No such process [23:25:06 CET] <darsie> hi [23:26:37 CET] <darsie> https://pastebin.com/G45Y9NAB Unable to find a suitable output format for '1:1' [23:26:38 CET] <darsie> 1:1: Invalid argument [23:27:53 CET] <furq> -map 0:0 -map 1:1 [23:28:02 CET] <darsie> ffmpeg -i 0.mp4 -i 1.mp4 -map 0:0 1:1 -c copy out.mp4 [23:28:08 CET] <darsie> ah, thx. [23:28:49 CET] <lavalike> what could it be if mpv playing an hls stream via youtube-dl prints out "cannot decode block" and "cabac" error messages every once in a while, disrupting the playback but not killing it altogether? [23:29:18 CET] <JEEB> decoder having issues with decoding of the stream [23:29:43 CET] <furq> yeah that sounds like the hls chunks have junk data at the start [23:29:49 CET] <lavalike> it's boggling because on macOS nothing of the sort happens, on win10 it happens every time, with the same streams [23:30:02 CET] <JEEB> is one of them using hwdec? [23:30:08 CET] <lavalike> mac is [23:30:08 CET] <JEEB> hwdec might just silently derp and not log [23:30:15 CET] <JEEB> since that is not FFmpeg's decoder [23:30:19 CET] <lavalike> good point [23:30:56 CET] <JEEB> cabac is basically something like zip to make the coded data smaller in H.264/HEVC :P [23:31:05 CET] <JEEB> so yes, issues trying to decode the coded packets [23:31:07 CET] <lavalike> I see (: [23:31:21 CET] <lavalike> I shall see if I can use any other decoder on windows I think [23:33:11 CET] <darsie> marvellous, audio in sync perfectly till the end. thx! :) [00:00:00 CET] --- Sun Dec 1 2019

1 0