July 2018 - Ffmpeg-devel-irc - lists.ffmpeg.org

ffmpeg-devel.log.20180719
by burek 20 Jul '18

20 Jul '18

[01:06:40 CEST] <jkqxz> tmm1: (From yesterday.) I haven't touched it in a while, but rebased and put on <https://github.com/fhvwy/FFmpeg/commits/cbs>. [01:18:18 CEST] <tmm1> jkqxz: thanks [02:15:15 CEST] <tmm1> jkqxz: would love to see those patches merged. i was able to use the filters to add simple (i.e. pts reordering based) closed caption support to the mediacodec hardware decoders: https://paste.ubuntu.com/p/kRWQ5tsn8J/ [02:29:23 CEST] <nicolas17> hmm is there a 'growing string' in libavutil? where appending to the string may reallocate if there isn't enough capacity [02:34:45 CEST] <tmm1> bprint? [02:36:23 CEST] <nicolas17> tmm1: yes, perfect! [11:54:50 CEST] <durandal_1707> Compn: ping! [13:59:31 CEST] <Compn> durandal_1707 whaaaaaaaaaa [14:00:23 CEST] <JEEB> < InTheWings> still waiting access to mailing list.. [14:00:32 CEST] <JEEB> someone needs to let this guy onto the ML [14:00:40 CEST] <JEEB> for some reason he's not joining this channel [14:01:03 CEST] <JEEB> you're on #videolan so if you need to speak to him specifically (if you can help with the ML) you can poke him there [14:01:38 CEST] <Compn> i can help with ml but whats problem [14:02:21 CEST] <Compn> he not recieve confirm subscription mail or so? [14:02:26 CEST] <Compn> ok i ask :P [14:02:32 CEST] <JEEB> cheers [14:07:54 CEST] <gagandeep> i am using ff_thread_release_buffer in update_thread_context so as to release the buffer for the frame, or am i supposed to use it somewhere else [14:08:13 CEST] <gagandeep> currently i am getting error when a thread is completing it's work [14:09:10 CEST] <durandal_1707> gagandeep: release buffer when you no longer need it? [14:09:55 CEST] <gagandeep> so i think update_thread_context is a good enough place to release the buffer for the thread that submit packet will be called on [14:42:55 CEST] <durandal_1707> I need money asap from ffmpeg! [14:43:39 CEST] Action: DHE gives durandal_1707 around 5 bux [14:44:28 CEST] <Compn> durandal_1707 : we have some money, put in formal request what money is for, invoice it [14:44:33 CEST] <Compn> and then mail to ml for reviews [14:44:34 CEST] Action: Compn afk [16:49:22 CEST] <durandal_1707> atomnuker: write av3 when? [16:55:08 CEST] <atomnuker> when av2 gets crushed under the weight of all the filtering and insane ideas people come up to avoid having vector quantization [16:56:12 CEST] Action: gnafu starts work on AV4 to get ahead of the curve. [16:56:23 CEST] <gnafu> Maybe my first frame will finish encoding by the time AV3 is finalized. [17:04:14 CEST] <kierank> atomnuker: why can't they do vector quant? [17:06:47 CEST] <durandal_1707> lair of multimedia guru - where is multimedia content here hidden ? [17:28:19 CEST] <cone-077> ffmpeg 03Baptiste Coudurier 07master:b420fead53b1: avformat/mxfdec: only call mxf_free_metadataset when ctx_size is != 0, otherwise ctx == mxf [18:18:01 CEST] <cone-077> ffmpeg 03Aman Gupta 07master:37644f047703: avcodec: parse options from AVCodec.bsfs [18:26:15 CEST] <cone-077> ffmpeg 03Thomas Guillem 07master:93e157f40f41: avcodec/videotoolboxenc: fix undefined behavior with rc_max_rate=0 [18:26:54 CEST] <cone-077> ffmpeg 03Thomas Guillem 07release/4.0:db923b3fbd3a: avcodec/videotoolboxenc: fix undefined behavior with rc_max_rate=0 [20:18:24 CEST] <cone-077> ffmpeg 03alexander schmid 07master:b23c4a9dbd8e: lavu/x86/cpu: Fix aesni detection [20:53:54 CEST] <mont3z> I'm using custom avio to parse a RTP stream. Although I can see that ffmpeg is parsing NAL units at some point av_read_frame blocks for ever. Does anyone knows any reason for av_read_frame to never return? [21:00:02 CEST] <mont3z> sry, wrong room [23:14:19 CEST] <cone-077> ffmpeg 03Carl Eugen Hoyos 07master:9cb3d8fcb7eb: lavfi/af_afir,af_aiir: Remove a variable that is always -1. [00:00:00 CEST] --- Fri Jul 20 2018

1 0

ffmpeg.log.20180719
by burek 20 Jul '18

20 Jul '18

[04:04:33 CEST] <jnollette> anyone notice some processors perform way less with ffmpeg [04:04:59 CEST] <jnollette> three 4790s. one gets 1/2 the peformance [04:05:48 CEST] <DHE> with the exact same binary version, input file and commandline? [04:06:06 CEST] <jnollette> yes [04:06:33 CEST] <jnollette> maybe 3/4 of the average [04:06:42 CEST] <DHE> could be lots of things. systems that are busy or other bottle necks, BIOS and CPU clock settings, memory channels [04:07:00 CEST] <DHE> 2 sticks of RAM (properly inserted of course) can give around 30% performance boost over 1 stick [04:07:22 CEST] <jnollette> only difference is motherboard and ram brands [04:24:13 CEST] <zx_> en? [04:24:26 CEST] <zx_> this can not save the chat history ? [04:26:01 CEST] <nicolas17> zx_: this particular channel is publicly logged: https://lists.ffmpeg.org/pipermail/ffmpeg-devel-irc/2018-July/005150.html [05:58:35 CEST] <kepstin> intel cpus throttle themselves based on power/heat, so a poorly installed hsf can do that sort of thing [05:59:28 CEST] Action: kepstin would suggest using the 'turbostat' tool while ffmpeg is running to check clock speeds and temperatures. [09:54:21 CEST] <th3_v0ice> What is the best way to generate audio PTS and DTS? Is this correct way to do it: pts = dts = ((1/(fps.num/fps.den)) * timebase.den * count) / frame->nb_samples? [09:54:58 CEST] <Mavrik> Why would you base audio pts/dts on frames? [09:55:03 CEST] <Mavrik> Audio doesn't really have fps [09:55:16 CEST] <Mavrik> If you have a track that's not timestamped, use sample count and samplerate to timestamp them [09:58:03 CEST] <th3_v0ice> I need to sync it with video [09:58:36 CEST] <Mavrik> So you're going to have gaps where audio cuts out and breaks players in between? :) [10:11:40 CEST] <th3_v0ice> I am not following you. This frame is an audio frame, and it has 1024 samples in it. Wouldnt that mean that in 24fps sequence, this audio frame lasts 0.04 seconds? [10:13:32 CEST] <JEEB> samples/sample_rate amount of time [10:17:32 CEST] <th3_v0ice> Ok, thanks guys [10:25:38 CEST] <automatical> Hey folks, I'm consuming a TCP stream that has a fairly unreliable frame rate, then pushing this to a live streaming service, is there a way to have ffmpeg smooth this framerate out when creating the output? [10:26:03 CEST] <JEEB> add the fps video filter I guess [10:26:07 CEST] <JEEB> if you need a constant frame rate [10:26:10 CEST] <automatical> I'm finding that the stream is speeding up and the delay from capture to output is getting smaller over time, and then I eventually see buffering when the delay reaches zero [10:26:35 CEST] <automatical> I'm using that, but I'm not sure if it's detecting the input framerate correctly [10:26:56 CEST] <JEEB> there's not much to detect, each input frame should have a timestamp [10:27:05 CEST] <JEEB> of course it could be a random IP camera that has no concept of timestamps [10:27:11 CEST] <JEEB> in which goot luck and all that crap [10:27:31 CEST] <automatical> ah :/ [10:27:44 CEST] <automatical> it's a raspberry pi camera [10:27:54 CEST] <automatical> creating the tcp stream with raspivid [11:17:12 CEST] <azaki> i'm trying to play around with av1, i encoded a file over several days, and now i'm trying to mux it into an mkv, but it's not working, the file ends up empty. i know the mkv mapping isn't finalized yet, but i figured it'd be using the draft for now or something [11:17:26 CEST] <azaki> i'm wondering if this is normal for now or if i've run into some bug? [11:17:55 CEST] <azaki> (this is ffmpeg 4.0.1) [11:31:40 CEST] <th3_v0ice> How can I remove stream delay using API for example: "Stream #0:0: Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, delay 1024, 64 kb/s"? [11:50:54 CEST] <InTheWings> still waiting access to mailing list.. [11:52:39 CEST] <durandal_1707> InTheWings: which mailing list? [11:52:51 CEST] <InTheWings> ffmpeg-devel [11:54:00 CEST] <durandal_1707> InTheWings: is there specific reason why you need it? [11:54:20 CEST] <InTheWings> need to comment on av1 in mp4 patch [11:55:36 CEST] <durandal_1707> InTheWings: visit #ffmpeg-devel and complain there to llogan or Compn [12:07:35 CEST] <JEEB> yeh, you'll have to poke people with access to the review list [13:38:30 CEST] <fsphil> is there any reason not to use the same sws context for scaling two fields in an interlaced frame? it's working for me, but all the examples I've seen use two separate contexts [13:38:39 CEST] <fsphil> so now I'm worried I've missed something [13:51:43 CEST] <bencoh> .41 [14:24:04 CEST] <th3_v0ice> While streaming to Twitch from FFmpeg C API twitch shows N/A for video codec. Does anyone know what could be the issue? Outputting to file works like a charm. Muxer is flv, and output is to rtmp. [14:29:19 CEST] <DHE> fsphil: if the settings for building the sws are identical it should be fine. the sws library should be memoryless (unlike a filter chain) [14:30:42 CEST] <fsphil> DHE: thanks. yeah the setup is identical for both [14:30:57 CEST] <DHE> you're not using multithreading are you? [14:31:20 CEST] <fsphil> not for this bit, both fields are scaled in the same thread one after the other [14:33:54 CEST] <olspookishmagus> hello, despite using: ffmpeg -i foo.AAC_126kbps_CBR.m4a -codec:a libmp3lame -q:a 6 foo.mp3 , I get a ~= 45.8kbps VBR MP3 ? I was expecting something with higher bit rate. [14:37:49 CEST] <DHE> olspookishmagus: I believe ffmpeg uses a scale where -q:a 0 is the best quality and 9 is the worst. your setting of 6 is fairly high [14:37:54 CEST] <DHE> have you tried with a value like 2? [14:40:42 CEST] <olspookishmagus> I used this a reference: https://trac.ffmpeg.org/wiki/Encode/MP3 which is close to what I have [14:41:36 CEST] <DHE> is the audio relatively simple? eg: just normal talking vs music with lots of instruments? [14:52:09 CEST] <olspookishmagus> DHE: it's actually one of the most simple audio clips, a 6-hour long audio clip of white/pink noise that's supposed to be: Spaceship Engine Reactor Noise (6 Hours) [youtube, yvUXAqLakQw] [15:15:54 CEST] <olspookishmagus> DHE: well that did it, I had to play with the various Q values to get somewhere close to the original bitrate [15:35:17 CEST] <kepstin> if the audio is mono or there's very little high frequency content, i'd expect lame's vbr mode to use less bitrate, that's pretty normal. [17:03:03 CEST] <th3_v0ice> Nobody with twitch streaming experience with the API? Or streaming in general. [18:24:59 CEST] <ayohmang> hi can i ask about configure command here [18:26:25 CEST] <iive> seems to be the right channel for that [18:28:55 CEST] <ghoti> totally [18:30:07 CEST] <BobCat> Don't ask if you can ask, just ask. [18:30:09 CEST] <ayohmang> im building on solaris 11.3 sparc, configure doesnt work due to sed [18:32:40 CEST] <ghoti> ayohmang: was there any particular error, or did your computer just vanish into a singularity after you typed the command? [18:33:05 CEST] <ghoti> ayohmang: also, what version of {everything} are you using? [18:40:12 CEST] <ayohmang> sed: illegal option -- E , FFMPEG 3.4.git , gcc 4.8.2 [18:41:11 CEST] <ghoti> Righto. The `-E` option for sed works in Linux, macOS, *BSD, etc, but the ancient sed on Solaris apparently doesn't include it. [18:41:27 CEST] <ghoti> If your package manager allows you to install a more modern alternative, like GNU sed for example, you'd get the option. [18:42:33 CEST] <ayohmang> i will try compile gnu sed [18:42:35 CEST] <causasui> I have an opus (audio) file that I'd like to turn into a webm. I don't need any image, just need to play it in a web player that can only play webm video. can ffmpeg do that, and how? [18:44:52 CEST] <ghoti> ayohmang: probably easier to go with a package manager first. ALso, I think Solaris 11 may already include the GNU sed command, perhaps in `/usr/gnu/bin/sed`. [18:45:51 CEST] <Shibe> timestamps are unset in a packet of stream 0. what might this mean? [18:45:54 CEST] <Shibe> that pts is unset? [18:50:48 CEST] <ayohmang> thank ghoti. [19:02:32 CEST] <ChocolateArmpits> Shibe, it would seem that dts is unset [19:02:50 CEST] <ChocolateArmpits> though I don't know where that message originated from [19:03:50 CEST] <ChocolateArmpits> causasui, if you want to play it extra safe then you can do it like this: ffmpeg -i audio.opus -acodec libvorbis -b:a 128k audio.webm [19:04:03 CEST] <ChocolateArmpits> or do you need a blank vidoe alongside? [19:12:17 CEST] <causasui> ChocolateArmpits: I'll try that and see if it works [19:13:28 CEST] <causasui> ChocolateArmpits: victory, thanks [19:28:25 CEST] <mont3z> I'm using custom avio to parse a RTP stream. Although I can see that ffmpeg is parsing NAL units at some point av_read_frame blocks for ever. Does anyone knows any reason for av_read_frame to never return the packet? [19:39:17 CEST] <DHE> causasui: it's possible that "ffmpeg -i audio.opus -c copy audio.webm" will also do the job and be faster if the browser also supports Opus directly [19:42:27 CEST] <ChocolateArmpits> DHE, more browsers support vorbis though and it's still good enough for most cases [19:43:18 CEST] <causasui> 4chan still doesn't support opus, just ogg :\ [19:43:21 CEST] <causasui> but it's not 4chan [19:43:24 CEST] <causasui> so i'm okay :) [19:48:14 CEST] <iss_> hello, I am trying to recode some data in blender internal process - proxies for videos. This was done by encoding them to MJPEG. [19:48:54 CEST] <iss_> they lack alpha channel, so I changed the codec, which was OK [19:49:13 CEST] <iss_> to HUFFYUV [19:49:32 CEST] <iss_> but I can not change pixel format to encode alpha [19:50:08 CEST] <iss_> when I look in avcodec_find_encoder(AV_CODEC_ID_HUFFYUV)->pix_fmts there is only AV_PIX_FMT_YUV422P [19:50:21 CEST] <furq> huffyuv supports bgra [19:50:30 CEST] <furq> it doesn't support any yuv format with alpha though [19:51:13 CEST] <kepstin> if you don't need compatibility with apps that don't use ffmpeg, try ffvhuff instead [19:51:31 CEST] <furq> ffvhuff and ffv1 both support all the yuva formats [19:51:36 CEST] <iss_> hmm thanks I will try that... [19:52:00 CEST] <furq> i don't know any other lossless codecs that do yuva [19:52:40 CEST] <iss_> I don't need lossless, actually lossy format may be better [19:53:09 CEST] <iss_> I do need indexing, but that is container issue I guess? [19:53:35 CEST] <kepstin> yeah, just throw whatever you encode into mkv and you should be fine there. [19:54:27 CEST] <iss_> It worked, great :) thanks a lot! [19:55:20 CEST] <kepstin> as far as good lossy codecs with alpha, I, uh, really don't know any :/ [19:58:01 CEST] <kepstin> maybe libopenjpeg (a jpeg2000 encoder)? [19:58:32 CEST] <kepstin> if you were previously using mjpeg, that should behave fairly similarly. [20:00:51 CEST] <iss_> I mean we can have options -"small" files, but no alpha, or larger files, but alpha... [20:01:18 CEST] <iss_> if someone wants both they can write a codec :) [20:13:37 CEST] <furq> kepstin: vp8 does alpha [20:13:42 CEST] <ghoti> I'd like to take a 1080p video and resize and crop it to play on a 2x2 video wall, where cells are 960x960 pixels. There's a bezel between screens that I've calculated to be approximately 48 pixels (~6.5mm). [20:13:42 CEST] <furq> maybe vp9 as well [20:13:48 CEST] <ghoti> I think I can handle the math to take care of cropping out four videos from the original, one for each monitor, then pasting them back together again (hstack/vstack). But is there a better way? [20:14:11 CEST] <furq> !filter tile @ghoti [20:14:17 CEST] <furq> uh [20:14:24 CEST] Action: ghoti holds his breath [20:14:27 CEST] <ghoti> ;) [20:14:44 CEST] <furq> !filter tile @ghoti [20:14:44 CEST] <nfobot> ghoti: http://ffmpeg.org/ffmpeg-filters.html#tile [20:14:58 CEST] <furq> that takes care of padding for you [20:15:16 CEST] <furq> oh nvm tile is for successive frames [20:15:21 CEST] <furq> that probably doesn't work for you [20:15:56 CEST] <ghoti> Could "successive frames" be expanded to "every frame in the file"? [20:16:39 CEST] <ghoti> Or .. oh, multiple definitions of "frame" happening here... [20:17:10 CEST] <ghoti> But thanks, I didn't know about this filter, I'll see if I can learn more in order to ask more specific questions. [20:25:54 CEST] <ghoti> I guess what I'm thinking I want is to crop out (remove) pixels 937 to 984 in each of X and Y, then either offset everything 24 pixels down+right so that things remain centred. [20:26:30 CEST] <ghoti> ... or if not offset, then scale everything up, back to 1920x1920, so that the cut line remains at the bezel. [20:27:21 CEST] <furq> if you just want the stuff occupied by the bezel to be missing then you could just scale [20:28:15 CEST] <ghoti> Yes, missing is better than having diagonal lines and faces look distorted... But .. scale what? [20:28:19 CEST] <furq> or you could just play back the original video, so i'm guessing that's not the answer [20:28:25 CEST] <furq> it's too hot today [22:29:41 CEST] <Exagone313> Does ffmpeg supports TAK decoding? I don't see any compile time option for it in configure --help, but it seems to have been supported at some point (?) [22:30:22 CEST] <c_14> yes [22:30:44 CEST] <c_14> it's an internal decoder [22:30:44 CEST] <furq> Exagone313: builtin codecs aren't listed in configure [22:30:58 CEST] <furq> they're all enabled by default [22:30:58 CEST] <c_14> --enable-decoder=tak, but it should be enabled by default [22:31:03 CEST] <furq> ^ and yes it is [22:31:43 CEST] <Exagone313> oh, ok, I think it's not what I thought [22:31:49 CEST] <Exagone313> I get this error from ffplay: TAK codec type 0 is not implemented. [22:32:11 CEST] <Exagone313> ffmpeg 4.0.1 [22:33:02 CEST] <Exagone313> it's just not implemented I guess. I don't know this format, it does not seem really used and I think the encoder is proprietary (probably patented) [22:33:57 CEST] <furq> i don't have any tak on hand to test [22:35:11 CEST] <JEEB> Exagone313: TAK is not patented, just someone thought it was a good idea to make his own thing and of course make it magical closed source stuff [22:35:29 CEST] <JEEB> also it's just possible that whomever made that decoder didn't have the samples for it [22:35:42 CEST] <JEEB> if you have a sample, post one on the trac [22:35:55 CEST] <JEEB> so that if someone cares about TAK codec type 0, they can hack it up [22:35:55 CEST] <Exagone313> I'm not allowed to have that file so I can't share it [22:36:36 CEST] <Exagone313> I'll see if I can use the proprietary decoder in a vm or something xD [22:36:54 CEST] <Exagone313> if it can decode to pcm [22:37:12 CEST] <Exagone313> thanks anyway! [22:37:14 CEST] <JEEB> well yea, stuff like foobar2000 etc have support for the decoder AFAIK [22:37:56 CEST] <JEEB> also if you just cut the sample and post it as a sample media file with a generic file name, nobody really cares :P [22:38:10 CEST] <JEEB> unless you search trac and there's already a sample posted [22:38:49 CEST] <JEEB> ok, seems like there's samples already https://trac.ffmpeg.org/ticket/2837 [22:39:06 CEST] <Exagone313> opened 5 years ago :D [22:39:15 CEST] <durandal_1707> $$$$$$$ [22:39:38 CEST] <JEEB> if you want ¬¬¬¬ kierank already noted what he'd want [22:39:44 CEST] <Exagone313> I don't have the knowledge to code the thing :/ [22:39:55 CEST] <Exagone313> neither the money to hire [22:40:08 CEST] <durandal_1707> you have money sure [22:40:32 CEST] <JEEB> durandal_1707: you're gonna get money out of the broadcast people, not random warez folk :P [22:40:48 CEST] <JEEB> of course what those two groups people want are different [22:41:05 CEST] <Exagone313> fuck the guy who thought about using tak instead of flac or wav [22:41:09 CEST] <durandal_1707> warez folks make cracked software [22:41:42 CEST] <atomnuker> warez folks have money since they don't buy stuff [22:42:18 CEST] <Exagone313> what if warez folks are students? :P [22:42:25 CEST] <furq> i just tried a few different presets and ffmpeg handled them all fine [22:42:31 CEST] <furq> it's not at all clear from the encoder what type 0 is [22:42:40 CEST] <durandal_1707> old version [22:42:42 CEST] <JEEB> furq: with that lol old encoder in that trac issue? [22:42:48 CEST] <furq> oh i missed that [22:42:56 CEST] <JEEB> yea modern versions encode newer versions of the format [22:43:01 CEST] <durandal_1707> only stereo... [22:43:01 CEST] <furq> fun [22:43:30 CEST] <furq> i guess all the tak you'll find in the wild is some japanese guy from 2005 who uploaded it to perfect dark and then did a murder-suicide [22:43:50 CEST] <JEEB> I think at that point it was still share or even winny where it was at [22:43:53 CEST] <furq> and left us to clean up his mess [22:43:58 CEST] <durandal_1707> optimfrog [22:44:07 CEST] <furq> i've never seen anyone use optimfrog [22:44:09 CEST] <furq> or LA [22:44:29 CEST] <Exagone313> lol this is japanese music [22:44:34 CEST] <furq> of course it is [22:44:56 CEST] <furq> if it's tak, tta, or wavpack, it's from japan [22:45:00 CEST] <furq> if it's ape then it's either from japan or russia [22:45:26 CEST] <furq> and if it's shn then it's a grateful dead soundboard from the internet archive [22:45:34 CEST] <atomnuker> :) [22:45:44 CEST] <durandal_1707> alac? [22:45:51 CEST] <furq> i try not to think about alac [22:46:03 CEST] <JEEB> ALAC at least got fully opened if nothing else [22:46:17 CEST] <JEEB> as in, the reference encoder/decoder set [22:46:22 CEST] <Exagone313> yes I did see an ape from russia too [22:46:41 CEST] <JEEB> also there's at least one store where you can pick between AAC, FLAC and ALAC [22:47:02 CEST] Action: JEEB bought some music from that place, too bad lossless = audiophile mostly :< [22:47:11 CEST] <durandal_1707> you can pick aiff [22:47:26 CEST] <JEEB> and I'm just anal about not buying non-physical that's worse than than the CDs [22:48:32 CEST] <Exagone313> I'd like to know because either I get physical copies (ok but sometimes higher quality is available, also it's very limited for unknown/foreign artists), either I'm not sure if it's worth buying if you get something worse than physical (like mp3) [22:49:27 CEST] <durandal_1707> buy lossless, transcode at will [22:49:44 CEST] <JEEB> yeh [22:49:58 CEST] <Exagone313> for example, on itunes, you don't know what you get [22:50:23 CEST] <JEEB> ototoy is nice, lets you pick [22:50:44 CEST] <JEEB> although it's rather specific [22:51:04 CEST] <Exagone313> it's not written if you get their alac or aac [22:52:05 CEST] <Exagone313> there is a french company named qobuz that sell lossless tracks, but they of course don't have unknown artists :P [22:52:58 CEST] <furq> just be glad it's not tidal mqa [22:54:18 CEST] <JEEB> oh wow, ototoy finally started adding region limiting. I thought they didn't have it because nothing seemed limited, but now it's just per-item :< [22:54:32 CEST] <JEEB> which is I guess better than some other stores which are just not letting you buy stuff without a proxy [00:00:00 CEST] --- Fri Jul 20 2018

1 0

ffmpeg-devel.log.20180718
by burek 19 Jul '18

19 Jul '18

[01:39:38 CEST] <jdarnley> > Ticket #7318 (Freezing glitch) [01:39:53 CEST] <jdarnley> All programs in windows cmd do that. [01:40:33 CEST] <jdarnley> I would guess that writes to the terminal block when you're using it. [01:40:47 CEST] <jdarnley> well, "terminal" [01:42:41 CEST] <kierank> linux does that as well I think [11:56:50 CEST] <cone-193> ffmpeg 03James Darnley 07master:088b07427dfe: fate: add more vc2 encoder tests [13:28:10 CEST] <atomnuker> jdarnley: wtf [13:28:15 CEST] <atomnuker> I didn't ack the commit [13:29:26 CEST] <jdarnley> I was unaware that you had to. [13:29:52 CEST] <atomnuker> I'm still the maintainer, so I _have_ to [13:30:14 CEST] <durandal_1707> !drama time! [13:31:46 CEST] <nevcairiel> he didnt even touch your precious encoder =p [13:32:22 CEST] <jdarnley> Feel free to ask anything you want here or on the list. [13:33:43 CEST] <atomnuker> in any caase its fine [13:38:30 CEST] <jdarnley> durandal_1707: sorry, no drama this time [13:47:24 CEST] <gagandeep> guys, ff_thread_get_buffer allocates buffer for all the planes required by just calling it once or just one plane [13:48:40 CEST] <durandal_1707> jdarnley: revert! revert! [14:01:42 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07master:706c9beaead7: avcodec/lagarith: Remove redundant varianble init [14:01:43 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07master:91578f5798d3: avcodec/tscc: Move reget buffer to after decoding [14:01:44 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07master:4a657acaefd5: avcodec/cfhd: Use bytestream2 for peaks [14:01:45 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07master:1809f1cda956: avcodec/dvbsub_parser: Do not discard data [14:05:20 CEST] <gagandeep> nevermind, finally found the default function used for this [15:05:37 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/4.0:0a155c57bd8e: Update for 4.0.2 [15:29:50 CEST] <cone-193> ffmpeg 03Rahul Chaudhry 07release/3.4:a8c5c71a0489: swresample/arm: rename labels to fix xcode build error [15:29:51 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/3.4:2b8d4f6f0186: avformat/movenc: Write version 2 of audio atom if channels is not known [15:29:52 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/3.4:15990a43b023: avcodec/dirac_dwt_template: Fix signedness regression in interleave() [15:29:53 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/3.4:6ea8b39c9057: avcodec/dvdsub_parser: Init output buf/size [15:29:54 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/3.4:6f354ca5fd5e: avcodec/dvdsub_parser: Allocate input padding [15:29:55 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07release/3.4:8df2c7dde6af: Update for 3.4.4 [16:14:28 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07n4.0.2:HEAD: avcodec/dvbsub_parser: Do not discard data [16:43:18 CEST] <cone-193> ffmpeg 03Michael Niedermayer 07n3.4.4:HEAD: avcodec/dvbsub_parser: Do not discard data [16:46:36 CEST] <durandal_1707> atomnuker: many of my projects depends on your work! [17:10:18 CEST] <atomnuker> sorry if I get hit by a truck [17:10:53 CEST] <kierank> durandal_1707: don't you know main reason for maintainer is to block other people changes and to force in crap code [17:29:55 CEST] <durandal_1707> kierank: looks like you gonna do full conversion to Nicolas mode? [17:30:19 CEST] <kierank> durandal_1707: why [17:30:29 CEST] <kierank> durandal_1707: you should come to no time to wait [17:31:12 CEST] <durandal_1707> too far away [17:32:07 CEST] <atomnuker> if you want to know how too far away is you should try a 12 hour flight to San Francisco [17:34:58 CEST] <kierank> durandal_1707: 2hr by plane [20:07:13 CEST] <tmm1> looks like AVCodec.bsfs doesn't work with multiple comma separated filters [20:08:44 CEST] <jamrial> tmm1: odd, it should [20:08:52 CEST] <jamrial> bsfs_init() in decode.c has code to handle that [20:09:35 CEST] <tmm1> yea i think it has some sort of bug.. it says non-existing bitstream filter '' [20:09:39 CEST] <jamrial> no decoder currently uses more than one, though [20:10:28 CEST] <jamrial> post your code? [20:16:46 CEST] <tmm1> i set bsfs = "h264_mp4toannexb,h264_metadata" [20:16:52 CEST] <tmm1> it looks like the tokenzier isn't working correctly [20:17:18 CEST] <tmm1> first iteration is "h264_mp4toannexb", second iteration is "" [20:18:38 CEST] <tmm1> off by one i think [20:18:59 CEST] <tmm1> bsfs_str = ",h264_metadata" on second iteration, so the parser returns "" before the comma [20:22:10 CEST] <jamrial> tmm1: does https://pastebin.com/D1XT2pRS fix it? [20:23:58 CEST] <tmm1> i did https://pastebin.com/t4RzsyVT [20:25:09 CEST] <jamrial> my patch copies the behavior of the similar loop in ffmpeg_opt.c [20:25:59 CEST] <tmm1> ok let me verify it works [20:26:34 CEST] <tmm1> jamrial: yep that works [20:26:38 CEST] <tmm1> do you want to email/commit it? [20:26:55 CEST] <jamrial> also, what decoder would use h264_metadata? you can't even pass option arguments for it to do anything but packet passthrough [20:27:05 CEST] <jamrial> i'll send the patch to the ml, yeah [20:30:43 CEST] <tmm1> yea that's next on my list, figure out how to pass options to it =) [20:31:19 CEST] <tmm1> i'm trying to use it from h264_mediacodec to extract a53 data [20:32:33 CEST] <jamrial> you'd have to add an extra av_get_token() call looking for "=" after the comma, then use av_dict_parse_string() or such to parse the options part of the string [20:33:46 CEST] <tmm1> that doesn't seem to bad [20:33:51 CEST] <tmm1> how would you pass the options into the bsf itself? [20:34:41 CEST] <jamrial> as part of the avcodec.bsfs string. so instead of "h264_mp4toannexb,h264_metdata" it would be "h264_mp4toannexb,h264_metdata=opt=value:opt2=value2" and similar [20:35:31 CEST] <tmm1> right i get that part, but once parsed into a dict how do i use it with av_bsf_init() [20:36:37 CEST] <tmm1> i guess just set the options onto the AVBSFContext* [20:37:04 CEST] <jamrial> yes, in AVBSFContext->priv_data [20:37:35 CEST] <jamrial> with av_opt_set_dict() [20:38:15 CEST] <tmm1> thanks, i'll try it out [20:38:16 CEST] <jamrial> alternatively, do what ffmpeg_opt.c does in new_output_stream() [20:38:23 CEST] <jamrial> using av_opt_set_from_string() to do both at the same time [20:38:39 CEST] <tmm1> oh cool [20:43:56 CEST] <jamrial> it also lets you omit the option name in the string if it's the first one and if you set a shorthand, so you could do "h264_metadata=value:opt2=value" [20:49:11 CEST] <tmm1> cool it works [20:49:17 CEST] <tmm1> https://paste.ubuntu.com/p/2r5rJDNHbP/ [20:52:57 CEST] <jamrial> i'd set the error to AVERROR_BUG and change the printed message accordingly, since much like the one about the missing filter, wrong options in the string should not happen [20:53:30 CEST] <tmm1> right ok [20:54:34 CEST] <jamrial> it'd be a developer error, not a user error or something unexpected like oom [20:55:52 CEST] <tmm1> makes sense [20:56:07 CEST] <tmm1> i'm just going to combine the other parsing fix into this patch and send it to the list [20:56:15 CEST] <jamrial> alright [20:57:41 CEST] <jamrial> although av_opt_set_from_string() can return ENOMEM, which should be returned as is [20:58:46 CEST] <tmm1> if (ret != AVERROR(ENOMEM)) ret = AVERROR_BUG; ? [21:03:35 CEST] <jamrial> if (ret < 0) { if (ret == enomem) { av_log("no memory"); } else { av_log("bug"); ret = bug; } goto fail; } [21:03:40 CEST] <jamrial> or such is probably better [21:06:04 CEST] <jamrial> the "Error parsing options for bitstream filter %s\n" message in your patch for the enomem case should be fine [21:06:08 CEST] <jamrial> the actual "no memory" error will be printed later based on the return value [21:09:48 CEST] <tmm1> ok [21:13:33 CEST] <tmm1> thanks for the help [21:25:09 CEST] <jamrial> no prob [22:16:37 CEST] <durandal_1707> d9 is down! no more AS and VS [22:45:00 CEST] <DHE> still? [22:45:23 CEST] <JEEB> most likely: swede's on holidays and the box in his basement went down [22:45:34 CEST] <JEEB> nordics in general is far away from home these months [22:46:36 CEST] <DHE> months?! [22:46:39 CEST] <nevcairiel> yeah crazy nordics and their long summer vacation, all at the same time [22:58:26 CEST] <JEEB> that's why the Finnish railroad company can just start breaking people's routes during July [22:58:35 CEST] <JEEB> because most people are Somewhere Esle [22:58:37 CEST] <JEEB> *Else [23:03:22 CEST] <jdarnley> Sounds similar to the Dutch Belgians and some Germans. [23:03:32 CEST] <jdarnley> Well, not the railroad part. [23:03:44 CEST] <jdarnley> They clog up roads through France on a few weekends over the summer. [23:04:37 CEST] <jdarnley> "School's out, kids. We're migrating south." [23:28:20 CEST] <cone-519> ffmpeg 03James Almer 07release/3.2:e870255e4e3d: avformat/utils: fix mixed declarations and code [23:28:20 CEST] <cone-519> ffmpeg 03Rahul Chaudhry 07release/3.2:a1d4de0559cc: swresample/arm: rename labels to fix xcode build error [23:28:20 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07release/3.2:2c0e98a0b478: avformat/movenc: Write version 2 of audio atom if channels is not known [23:28:20 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07release/3.2:fe73afd730be: avcodec/dirac_dwt_template: Fix signedness regression in interleave() [23:28:20 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07release/3.2:ffc9ef8f4c74: avcodec/dvdsub_parser: Init output buf/size [23:28:21 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07release/3.2:94fda4f9ec95: avcodec/dvdsub_parser: Allocate input padding [23:28:21 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07release/3.2:a911f234e26e: Update for 3.2.12 [23:44:03 CEST] <cone-519> ffmpeg 03Michael Niedermayer 07n3.2.12:HEAD: avcodec/dvbsub_parser: Do not discard data [00:00:00 CEST] --- Thu Jul 19 2018

1 0

ffmpeg.log.20180718
by burek 19 Jul '18

19 Jul '18

[02:59:36 CEST] <bob123> Cracki thanks for the reply [02:59:49 CEST] <bob123> I will check those links out [03:00:19 CEST] <bob123> I found out that you can export color curves from adobe photoshop and load them in libavfilter [03:00:30 CEST] <bob123> so I have been experimenting with that [05:07:31 CEST] <killown> [mov,mp4,m4a,3gp,3g2,mj2 @ 0xc0a280] moov atom not found [05:07:43 CEST] <killown> using ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 [12:38:53 CEST] <King_DuckZ> hi, anyone knows if 32 bit rgb float is supported in ffmpeg? I want to convert from that to yuv444p using swsScale [12:41:21 CEST] <durandal_1707> nope [12:41:44 CEST] <durandal_1707> only way is using zscale filter via libavfilter [12:42:47 CEST] <King_DuckZ> hm I just found a AV_PIX_FMT_FLAG_FLOAT defined in libavutil/pixdesc.h, is it what it's used with? [12:44:43 CEST] <remlap> Possibly the right place to ask, is there any video editor for linux that can do smart h264 encoding as in only recodes from cut to the next keyframe only [12:45:38 CEST] <durandal_1707> King_DuckZ: yes, but that is just helper, there is pixel format, but no swscale support [12:47:10 CEST] <King_DuckZ> durandal_1707: so what's the difference between that and the zscale thing you mentioned earlier? it's not like I must use swscale either... [12:50:00 CEST] <durandal_1707> King_DuckZ: zscale filter interface is much simpler [12:50:32 CEST] <durandal_1707> but you need to use special filtergraph incarnation [12:51:55 CEST] <King_DuckZ> ok, I have no idea what you're talking about and in the docs I can only see zscale_class, zscale_options and ZScaleContext, is there a usage example somewhere? one with explanations, if possible? [12:53:38 CEST] <durandal_1707> King_DuckZ: if you do not how to use libavfilter library do not use it [12:55:02 CEST] <King_DuckZ> I thought you just said that's what I should do [12:55:15 CEST] <King_DuckZ> <durandal_1707> only way is using zscale filter via libavfilter [13:00:40 CEST] <durandal_1707> King_DuckZ: have you ever used libavfilter? [13:01:11 CEST] <durandal_1707> for zscale you will need to install/have zimg library too [13:05:31 CEST] <InTheWings> can I haz confirmation on mailing list plz [13:36:30 CEST] <King_DuckZ> durandal_1707: I don't think I did, but if that's the way forward... I guess my other alternative otherwise would be to convert to 8 bit RGB myself first, then pass that buffer on to swscale... which sounds very slow [13:37:47 CEST] <King_DuckZ> or I could just do the conversion myself... I could check how slow that would be maybe [13:54:55 CEST] <zap0> ffprobe -gimme_all_details outputs piles of text.... what character encoding is that? [14:00:51 CEST] <JEEB> it should be utf-8 [14:01:08 CEST] <JEEB> except maybe with windows where it might be windows unicode (UCS-2/UTF-16) [14:01:37 CEST] <JEEB> also I recommend if you're going to parse ffprobe's stuff that you utilize a specific output format [14:01:40 CEST] <JEEB> -of json, for example [14:02:20 CEST] <zap0> just been reading lots of manual, -print_format flat is in "shell output escaped" (wtf that means i don't know) [14:02:48 CEST] <zap0> the -print_format ini says it is in UTF-8.. which i might be able to parse with similar effciency [14:03:34 CEST] <zap0> i was using flat. and i say non-ascii-7bit in the output of 1 video file [14:03:42 CEST] <zap0> (which is the cause of my problems) [14:04:16 CEST] <zap0> there is json and xml.. but i'd rather eat glass than use those [14:04:42 CEST] <JEEB> python etc just make usage of JSON pretty simple and iterable entities are nice [14:04:53 CEST] <JEEB> from json import loads as load_json [14:05:02 CEST] <JEEB> load_json(ffprobe_output) [14:05:05 CEST] <JEEB> and you get a dictionary [14:05:16 CEST] <zap0> https://www.ffmpeg.org/ffprobe-all.html#flat 4.3 flat. why does that not say with more clarity the encoding [14:06:11 CEST] <zap0> (i'm using C++) [15:10:10 CEST] <zx_> I have download the ffmpeg source 4.0.1,and build it on my Mac, it works. when I try to use the example in the doc dir , the example is decode_video.c. I had tried building decode_video.c using "make example " or using Xcode. Both will be ok. But when I supply tow mp4 file for input , Error info is https://pastebin.com/32SkNzYp [15:12:10 CEST] <zx_> [15:14:20 CEST] <zx_> quit [15:54:32 CEST] <shinsh> hello~ [15:57:25 CEST] <shinsh> When will I see an article about compiling https://ffmpeg.zeranoe.com/forum? [15:57:45 CEST] <DHE> huh? [15:59:18 CEST] <shinsh> When can I see an i wrote article in the Compiling category of https://ffmpeg.zeranoe.com/forum? [15:59:43 CEST] <zap0> why are you asking in here? [16:01:15 CEST] <shinsh> I'm sorry it was not IRC for forum of ffmpeg.zeranoe.com. [16:02:46 CEST] <JEEB> no, this is the actual FFmpeg project's IRC channel [16:02:58 CEST] <JEEB> Zeranoe is Zeranoe [16:20:28 CEST] <Tu13es> hi all. I've started ripping old family VHS tapes to mp4 via an Elgato Video Capture. it just creates one large .mp4 file for each tape. can I use ffmpeg to somehow split the file based on a scene change and/or black frames and/or static? [16:29:17 CEST] <DHE> that's not really an ffmpeg feature in one shot. you can probably do it in two shots using the blackdetect filter for pass 1, then run another pass with the segment output target and a timestamp listing [16:32:52 CEST] <Tu13es> i see. what about using scene detection? [16:33:05 CEST] <Tu13es> seems like i'd do what you mentioned, then do it again except with scene detection instead of blackdetect? [18:11:22 CEST] <ChocolateArmpits> Does ffmpeg cache a single still image input or does it read continuously ? [19:33:51 CEST] <kepstin> ChocolateArmpits: I'm pretty sure that ffmpeg's image input with the -loop option will re-read the image sequence [19:34:38 CEST] <ChocolateArmpits> kepstin, heh don't remember if I used -loop or not [19:34:41 CEST] <ChocolateArmpits> for the still [19:34:54 CEST] <ChocolateArmpits> So I can hope that the drive at best has the file cached? [19:34:59 CEST] <kepstin> if you don't use -loop, it'll decode a single frame then give end of file [19:35:02 CEST] <kepstin> and exit [19:35:19 CEST] <kepstin> (you can use the "loop" filter to repeat the frame in-memory if you prefer) [19:35:51 CEST] <ChocolateArmpits> what do you mean in-memory, would it not read the image from the disk under those circumstances? [19:36:29 CEST] <kepstin> filters are completely separate from the input/demuxer [19:36:43 CEST] <ChocolateArmpits> oh I missed the "filter" thing [19:38:11 CEST] <ChocolateArmpits> hmm would the process not quit when the end of file was met with this filter somehow? [19:40:21 CEST] <ChocolateArmpits> just to state my situation, I want to overlay an image over a video [19:41:02 CEST] <JEEB> I think the overlay filter had an option regarding what to do if one input EOFs [19:41:03 CEST] <ChocolateArmpits> so the overlay is absolutely clear, I just don't want the drive to be pointlessly read [19:41:31 CEST] <JEEB> https://www.ffmpeg.org/ffmpeg-all.html#overlay-1 [19:41:39 CEST] <JEEB> https://www.ffmpeg.org/ffmpeg-all.html#framesync [19:41:49 CEST] <JEEB> > repeatlast=1 [19:42:18 CEST] <JEEB> also eof_action=repeat should be the same [19:44:52 CEST] <kepstin> yeah, if you're doing overlay, the overlay filter's builtin eof stuff is the way to go for sure. [21:40:55 CEST] <ChocolateArmpits> ok it seems that overlay does it all right from the start [00:00:00 CEST] --- Thu Jul 19 2018

1 0

ffmpeg-devel.log.20180717
by burek 18 Jul '18

18 Jul '18

[01:26:17 CEST] <nicolas17> why is the hls demuxer hardcoded to accept only file and http URLs? [01:27:42 CEST] <nicolas17> eh, I'll deal with that later [01:30:24 CEST] <atomnuker> nicolas17: because its short for http live streaming? [01:31:40 CEST] <nicolas17> atomnuker: with that argument it shouldn't allow file: [01:32:20 CEST] <atomnuker> what else do you want it to accept? [01:43:08 CEST] <Compn> nicolas17 : it accepts file because the file is a hls playlist m3u8 usually [01:43:17 CEST] <Compn> which, again, links to http stream lol [01:44:03 CEST] <Compn> nicolas17 : what do you want it to accept ? [01:44:18 CEST] <Compn> you want to rebuild thousands of mp4 files? iirc theres a php script for that ? [01:44:55 CEST] <Compn> but if you want to add that functionality to ffmpeg , good idea too [02:39:38 CEST] <nicolas17> does ffmpeg have unit tests? [02:42:50 CEST] <jamrial> nicolas17: https://ffmpeg.org/fate.html [02:47:47 CEST] <nicolas17> hmm, what's the scope of those tests? [02:49:25 CEST] <nicolas17> ohh I was only looking at ffmpeg/tests [02:49:43 CEST] <nicolas17> I missed eg. ffmpeg/libavutil/tests which is more unit-like [04:28:22 CEST] <nicolas17> is there a function to do URL escaping / percent encoding? [11:43:23 CEST] <cone-733> ffmpeg 03Gyan Doshi 07n3.3.8:HEAD: avformat/mov: only set handler_name from mdia->hdlr [17:24:39 CEST] <durandal_1707> atomnuker: you still working avfft ? [17:52:07 CEST] <atomnuker> yep [18:25:10 CEST] <akravchenko188> jkqxz: hello. it is reminder. could you please review updated patches of scaler_amf?http://ffmpeg.org/pipermail/ffmpeg-devel/2018-July/232111.html ? Thanks [18:36:18 CEST] <jamrial> akravchenko188: ping the patch instead. others may also be able to review it [18:53:53 CEST] <jkqxz> It's in my queue of stuff I should really look at soon, behind vulkan. [18:56:42 CEST] <jamrial> jkqxz: also, found another issue in cbs av1 aside from the fixes in the patch i linked the other day. added it to https://pastebin.com/0V0G2vRp [18:57:00 CEST] <jamrial> with that still pictures using the reduced header work now [18:58:12 CEST] <jamrial> i still wonder if the file that seems to have global motion fields truly has them, or if it's another parsing error and it's reading a bit at the wrong offset. coulnd't find any mistake in that regard, though [19:04:52 CEST] <jkqxz> Thanks, I'll fix that. [19:05:05 CEST] <jkqxz> It is very possible that there are other funny parsing errors there. [19:07:51 CEST] <jkqxz> Things like deciding whether skip_mode_present is there or not are somewhat nontrivial. [19:10:32 CEST] <TD-Linux> the encoder does search and use global motion so I'd expect it to appear [19:11:01 CEST] <jkqxz> (I ended up instrumenting the decoder to write some of this.) [20:03:50 CEST] <durandal_1707> kierank: you are evil! [20:04:48 CEST] <kierank> Why [20:05:06 CEST] <kierank> durandal_1707: are you going to vdd [20:05:40 CEST] <durandal_1707> no, those are for evil people, they gather do drink blood, etc [20:20:08 CEST] <j-b> right [20:21:10 CEST] <kierank> durandal_1707: why am I evil then [21:16:18 CEST] <durandal_1707> kierank: because what you said yesterday [21:16:40 CEST] <kierank> To whom [21:16:46 CEST] <kierank> Was I drunk [21:19:59 CEST] <kierank> durandal_1707: ?????? [21:21:10 CEST] <durandal_1707> [22:47:05 CEST] <kierank> durandal_1707: play for your country gootball team [22:01:52 CEST] <durandal_1707> why vdd always need to be in paris? [22:02:39 CEST] <atomnuker> it was in berlin once and it wasn't nice [22:08:34 CEST] <durandal_1707> its unfair, it should be every time in different country [22:09:25 CEST] <atomnuker> it'll invariably end up in the US at least once and it'll completely blow [22:09:42 CEST] <atomnuker> maybe if its in the EU [22:10:23 CEST] <atomnuker> I hear slovakia's cheap in the autumn, bratislava's nice [22:11:06 CEST] <durandal_1707> i accept only slovenia and italy [22:13:54 CEST] <atomnuker> durandal_1707: but surely going to say, rome would take you many more hours by train than it would take you to fly to paris [22:16:58 CEST] <durandal_1707> ? i don't get what you are trying to say [22:18:47 CEST] <atomnuker> well, afaik croatia isn't well connected to italy by train [22:20:17 CEST] <durandal_1707> lol, we have helicopters and planes too, why you travel only by train? [22:25:43 CEST] <atomnuker> because I thought the reason you didn't want to go to paris was because you have to fly there [22:28:12 CEST] <durandal_1707> ah, it's not because of that, but because of burning cars [22:30:37 CEST] <atomnuker> I haven't seen a single burning car in paris [22:31:23 CEST] <atomnuker> the tube system sucks though, its crowded and you always need to transfer and the tickets don't always work and are expensive and like to leave you stranded at airports and the stations are dirty and... [22:32:12 CEST] <atomnuker> and cdg terminal 3 is a cross between an airport terminal, a bus station and hell itself [22:34:23 CEST] <kierank> atomnuker: which one is 3 [22:34:27 CEST] <kierank> there's one that's really crap i remember [22:34:30 CEST] <kierank> 2f I think [22:34:45 CEST] <durandal_1707> see, thats why i prefer staying home [22:34:54 CEST] <atomnuker> the one where we flew back together in august 2016 [22:35:14 CEST] <kierank> atomnuker: yes that one [22:35:15 CEST] <kierank> well remember [22:35:18 CEST] <kierank> that was bloody awful [22:35:52 CEST] <tmm1> jkqxz: is there a git branch somewhere with the userdata cbs patchset? [23:06:18 CEST] <January> durandal_1707: croatia beat england tho [23:08:22 CEST] <January> atomnuker: vdd in vegas [23:55:11 CEST] <tmm1> is dynarray the only data structure helper in avutil? [23:55:54 CEST] <nicolas17> tmm1: there's tree.h too [23:56:12 CEST] <DHE> there's a few others, like a thread queue (blocks the current therad when the queue is empty) [00:00:00 CEST] --- Wed Jul 18 2018

1 0

ffmpeg.log.20180717
by burek 18 Jul '18

18 Jul '18

[10:09:07 CEST] <Hail_Spacecake> howdy [10:09:23 CEST] <Hail_Spacecake> I want to use some software that will let me stream videos on my computer to my own browser, like a bespoke youtube [10:09:27 CEST] <Hail_Spacecake> is ffmpeg the right tool to use for this? [10:16:18 CEST] <JEEB> FFmpeg usually gets used for use cases like that in some form, but no idea what you want to do specifically [10:16:41 CEST] <Mavrik> In most cases you want a streaming server to go with ffmpeg though [10:16:46 CEST] <Mavrik> E.g. nginx-rtmp-module or something [10:17:27 CEST] <JEEB> yes, or if you just use FFmpeg's HLS output with HTTP posts just normal nginx [10:17:54 CEST] <Mavrik> ohh, forgot about that option [10:18:06 CEST] <JEEB> which reminds me I have to ask the author(s) of that how you're supposed to do multiple audio languages in it - I have done a single line change which works but I've no idea if that's what they want :P [10:18:26 CEST] <bencoh> wait, how does that http/post output works? [10:18:26 CEST] <JEEB> also re-using a stream in multiple muxes "conveniently" crashes [10:18:44 CEST] <JEEB> bencoh: it pushes the generated segments and playlist straight to the server [10:18:49 CEST] <JEEB> Akamai seems to love that [10:18:56 CEST] <JEEB> (they're one of the authors of that thing) [10:19:04 CEST] <Mavrik> Yep, just dump the playlist and segment files somewhere [10:19:04 CEST] <bencoh> they do, yeah [10:19:15 CEST] <Mavrik> And point the player to the m3u8 [10:19:19 CEST] <bencoh> but then you need more than just "normal nginx" to handle the POST I guess [10:19:22 CEST] <JEEB> no [10:19:30 CEST] <JEEB> just normal nginx which is configured for file uploads [10:19:32 CEST] <Mavrik> Nope, it's literaly just dumping files somewhere and then serving them [10:19:32 CEST] <bencoh> ah [10:19:43 CEST] <JEEB> no magicks [10:19:43 CEST] <bencoh> forgot it could do that :) [10:19:56 CEST] <bencoh> (file uploads, that is) [10:19:59 CEST] <Mavrik> Too bad the two major desktop browsers don't support HLS :( [10:20:10 CEST] <JEEB> hls.js seems to work "alright" [10:20:21 CEST] <JEEB> and everyone seems to be testing with it anyways [15:56:13 CEST] <FishPencil> Which deshake program/filter provides the best results right now? I know vidstab is pretty good, but are there better ones? (in ffmpeg or not) [15:58:11 CEST] <Hello71> youtube's is allegedly better [15:58:32 CEST] <Hello71> according to google they have some algorithm for removing moving shutter [16:00:30 CEST] <FishPencil> http://www.guthspot.se/video/deshaker.htm claims to do rolling shutter as well [16:46:11 CEST] <th3_v0ice> Why is encoder producing PTS that is less then the DTS? [17:07:30 CEST] <DHE> note that PTS and DTS can be negative. are you using the correct type for the PTS and TS? [19:18:54 CEST] <ted__> Hello all~ I have a question about the libvpx-vp9 encoder AVOptions. [19:20:25 CEST] <ted__> Which one is correct? -arnr-type (default -1) or -arnr_type (default 3) [19:23:52 CEST] <Hail_Spacecake> JEEB: I was hoping for some tool that would just provide a browser video player video + controls for selecting videos [19:24:00 CEST] <Hail_Spacecake> and whatever backend you need to make the streaming happen [19:28:27 CEST] <ted__> The -arnr- options of the VP9 have still been wrong? If so, I hope devs will fix them at the next version. [19:29:00 CEST] <ted__> Thank you devs for your help as always. [19:29:19 CEST] <th3_v0ice> DHE: Yes actually. The type is int64_t, the problem is that PTS is 49 and DTS is 500 for example. What could throw the encoder to produce these results? [19:38:29 CEST] <DHE> th3_v0ice: interesting... that is definitely not right... [19:57:11 CEST] <th3_v0ice> DHE: Also the DTS are not monotinically increasing, they are jumping all over the place :) For example 10, 20, 60, 70, 80, 30, .... [20:03:45 CEST] <FishPencil> How should I rotate a video 180 degrees [20:04:50 CEST] <Hello71> transpose=2,transpose=2 [20:14:03 CEST] <th3_v0ice> How can i properly copy the packet and its buffers? [20:40:45 CEST] <FishPencil> How do I take a 4:3 image, add black letterboxes up to 16:9, resize to 1920x1080, and display it for 3 sec? [21:00:55 CEST] <Hello71> have you considered just using an image viewer program [21:01:01 CEST] <Hello71> or even just mpv [21:02:15 CEST] <FishPencil> I'm trying to make a video [21:08:40 CEST] <FishPencil> This seems to work: ffmpeg -loop 1 -i i.jpg -vf "scale=1920:1080:force_original_aspect_ratio=decrease,pad=1920:1080:(ow-iw)/2:(oh-ih)/2" -c:v ffv1 -t 3 -r 60 o.mkv [21:11:42 CEST] <Hello71> sure [21:11:54 CEST] <Hello71> why does it need to be 60 fps [21:13:49 CEST] <FishPencil> I'm concatenating it with other 60fps video [21:14:46 CEST] <Hello71> so in conclusion, XY problem? [21:15:28 CEST] <FishPencil> huh? [21:42:02 CEST] <bob123> I really need a de-fog / de-haze filter for my live stream but I don't know where to start.. [21:42:57 CEST] <bob123> I've looked at libavfilter and libxcam libraries [21:43:33 CEST] <bob123> I'm not sure if libavfilter comes with a defogging filter [21:46:02 CEST] <bob123> libxcam seems like it's not done yet for FFMPEG use [22:21:30 CEST] <Cracki> anyone care for an auto white-balance filter, like handheld video cameras and phone cameras have? involves some summation and some multiplication [22:22:10 CEST] <Cracki> I figure I'd ask here before I ask in #vlc because at least here someone knows what color balance is :P [22:23:29 CEST] <Cracki> plenty of "movies" have been colorgraded and low-contrasted to death and I want something to correct that. I know the math but not how to make a filter for any player I use. [23:07:06 CEST] <bob123> Does anyone know of a defog filter for FFMPEG? [23:08:43 CEST] <Cracki> i've read a paper on de-fog [23:09:12 CEST] <Cracki> involves estimating fog strength (desaturation) and then continuing from there. different approaches look more or less good [23:11:57 CEST] <Cracki> Real Time Fog Removal Technique with Improved Quality through FFT - IEEE Conference Publication https://ieeexplore.ieee.org/document/6821512/figures [23:11:57 CEST] <Cracki> A fast method of fog and haze removal - IEEE Conference Publication https://ieeexplore.ieee.org/document/7471871/ [23:11:57 CEST] <Cracki> Fog removal techniques from images: A comparative review and future directions - IEEE Conference Publication https://ieeexplore.ieee.org/document/6884973/ [23:13:24 CEST] <Cracki> https://www.sciencedirect.com/science/article/pii/S2215098616305067 [00:00:00 CEST] --- Wed Jul 18 2018

1 0

ffmpeg-devel.log.20180716
by burek 17 Jul '18

17 Jul '18

[07:53:41 CEST] <cone-892> ffmpeg 03Gyan Doshi 07master:b8c4d2b2ed22: doc/filters: add missing details to aphasemeter [13:14:03 CEST] <cone-826> ffmpeg 03Gyan Doshi 07n2.8.15:HEAD: doc/filters: add missing details to aphasemeter [13:23:54 CEST] <kierank> http://obe.tv/about-us/obe-blog/item/54-a-look-at-the-bbc-uhd-encoder-in-20… [13:23:55 CEST] <kierank> BBB: ^ [13:25:03 CEST] <kierank> crap need to link the full report [13:25:18 CEST] <kierank> http://downloads.obe.tv/Parabola-Explorer-BBC-Russia-UHD.pdf [13:27:04 CEST] <atomnuker> yeah, that is pretty shoddy, I wonder if they're using their own encoder for it [13:32:08 CEST] <kierank> atomnuker: elemental [13:33:20 CEST] <BBB> wellIve always said that making a realtime software encoder isnt difficult [13:33:27 CEST] <BBB> just look at x264 ultrafast [13:33:49 CEST] <BBB> shoddy is the right word indeed <3 [13:34:19 CEST] <JEEB> yea, being realtime is generally not the issue [13:34:36 CEST] <JEEB> you can just write the minimal bit stream required to be correct + some VBV/HRD [13:34:36 CEST] <BBB> I recall a conversation at a webm summit a while ago about realtime hardware encoders [13:34:46 CEST] <BBB> the question was how does quality compare to software encoders?" [13:34:51 CEST] <BBB> and the answer was pretty good [13:35:31 CEST] <BBB> all I remember is that there was this disbelief on both sides of the Q&A [13:35:38 CEST] <JEEB> lol [13:35:48 CEST] <BBB> you dont have bdrate numbers? vs. what is this thing called quality?" [13:46:32 CEST] <kierank> https://news.ycombinator.com/item?id=17540384 [13:46:35 CEST] <kierank> please upvote i guess [14:10:44 CEST] <Compn> kierank : who makes the parabola explorer ? [14:10:45 CEST] <Compn> ehe [14:10:47 CEST] Action: Compn afk [16:32:19 CEST] <jamrial> jkqxz: ping [18:03:53 CEST] <cone-880> ffmpeg 03Gyan Doshi 07master:8aa6d9a8d37b: avformat/mov: only set handler_name from mdia->hdlr [18:07:13 CEST] <JEEB> durandal_1707: are you interested in bink2? :D [18:08:32 CEST] <JEEB> I have KB2f and KB2i at least [18:09:18 CEST] <durandal_1707> JEEB: yes [18:09:54 CEST] <JEEB> the sample posted on the issue tracker seemed to be KB2j [18:10:35 CEST] <JEEB> oh, found one KB2g [18:11:45 CEST] <JEEB> will push somewhere [18:14:32 CEST] <nevcairiel> how the f does one build autotools things on osx targeting an older version of the OS, setting the env var or passing the compiler flag doesn't work on autotools configure [18:15:37 CEST] <JEEB> haven't tried doing that yet with my 10.13.5 and autotools [18:23:45 CEST] <JEEB> durandal_1707: KB2f: https://0x0.st/sOjF.bk2 KB2g: https://0x0.st/sOeo.bk2 KB2i: https://0x0.st/sOeM.bk2 [18:24:18 CEST] <JEEB> KB2j I also happened to find, but the issue you commented on had seemed to have a sample for that [18:42:44 CEST] <akravchenko188> jkqxz: hello. just reminder. could you please review updated patches of scaler_amf http://ffmpeg.org/pipermail/ffmpeg-devel/2018-July/232111.html ? Thanks [19:04:19 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:2c404cc11a02: avcodec/mjpegdec: Check for end of bitstream in ljpeg_decode_rgb_scan() [19:04:20 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:84280dc7cfa4: avcodec/magicyuv: Check bits left in flags&1 branch [19:04:21 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:10e5302db413: avcodec/h264_parser: Reduce needed history for parsing mb index [19:04:22 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:d00548f2c150: avcodec/escape124: Check buf_size against num_superblocks [19:04:23 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:21732c1adc3d: avcodec/cscd: Check output buffer size for lzo. [19:04:24 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:bf3e331b7633: avcodec/ra144: Fix integer overflow in ff_eval_refl() [19:04:25 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:9f76f0fab80c: avcodec/escape124: Fix spelling errors in comment [19:04:26 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:841e1399e6d0: avcodec/dpx: Check elements in 12bps planar path [19:04:27 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:20c5fb972101: avformat/movenc: Check that frame_types other than EAC3_FRAME_TYPE_INDEPENDENT have a supported substream id [19:04:28 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:0003ace83b18: avcodec/ac3_parser: Check init_get_bits8() for failure [19:04:29 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:76105382244e: avformat/movenc: Do not pass AVCodecParameters in avpriv_request_sample [19:04:30 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:679d749eaba9: avcodec/mpeg4videodec: Check read profile before setting it [19:04:31 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:d3536ce8393d: avformat/movenc: Use mov->fc consistently for av_log() [19:04:32 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:2b13c136c4c6: avcodec/jpeg2000dec: Check that there are enough bytes for all tiles [19:04:33 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:e21e5c95c1ce: avcodec/jpeg2000dec: Fixes invalid shifts in jpeg2000_decode_packets_po_iteration() [19:04:34 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:c4b23793d4fe: avcodec/shorten: Fix undefined integer overflow [19:04:35 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:de0a1d01baae: avcodec/shorten: Fix undefined addition in shorten_decode_frame() [19:04:36 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:5fd1dce39a70: avcodec/mpeg4videodec: Remove use of FF_PROFILE_MPEG4_SIMPLE_STUDIO as indicator of studio profile [19:04:37 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:1361e4abb81f: avcodec/ac3dec: Check channel_map index [19:04:38 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:3bf80c7b223e: avcodec/indeo4: Check for end of bitstream in decode_mb_info() [19:04:39 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:4439d6aa6956: avcodec/mpeg4videodec: Check for bitstream end in read_quant_matrix_ext() [19:04:40 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:a28ab09e2a2a: avformat/mxfdec: Fix av_log context [19:04:41 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:a21703ca5d42: avformat/asfdec_o: Check size_bmp more fully [19:04:42 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:37f505cc853f: avformat/rmdec: Do not pass mime type in rm_read_multi() to ff_rm_read_mdpr_codecdata() [19:04:43 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:6f4b82cc3a87: avformat/pva: Check for EOF before retrying in read_part_of_packet() [19:04:44 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:6d992a51c75a: avformat/mms: Add missing chunksize check [19:04:45 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:fc92ca5b8e3b: avcodec/vp8_parser: Do not leave data/size uninitialized [19:04:46 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:d8c4b2ae57e0: avcodec/mjpegdec: Check for odd progressive RGB [19:04:47 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:0981dfee7d41: avformat/movenc: Check input sample count [19:04:48 CEST] <cone-880> ffmpeg 03Rahul Chaudhry 07release/4.0:5db47b3983ef: swresample/arm: rename labels to fix xcode build error [19:04:49 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:fd53179f4a71: avformat/movenc: Write version 2 of audio atom if channels is not known [19:04:50 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:48479937c3e9: avformat/mov: remove modulo operations from mov_estimate_video_delay() [19:04:51 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:052edeec55fd: avformat/mov: Eliminate variable buf_size from mov_estimate_video_delay() [19:04:52 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:6b65f46673d8: avformat/mov: Break out of inner loop early in mov_estimate_video_delay() [19:04:53 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:670b565ba2b1: avformat/mov: Simplify last element computation in mov_estimate_video_delay() [19:04:54 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:0561cde1289b: avcodec/dirac_dwt_template: Fix signedness regression in interleave() [19:04:55 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:40ed40902ade: avcodec/dvdsub_parser: Init output buf/size [19:04:56 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/4.0:3ef38c414e79: avcodec/dvdsub_parser: Allocate input padding [19:48:10 CEST] <durandal_1707> ms killed github :( [19:48:27 CEST] <JEEB> oh so that's why my fetches didn't work :D [19:48:54 CEST] <DHE> github is currently suffering from a bad case of unicorns [19:49:01 CEST] <DHE> (joke makes sense in context) [19:49:11 CEST] <Mathieu_Du> seems to be back up [20:33:28 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:f4e2ac296dac: doc/APIchanges: Fix typos in hashes [20:33:29 CEST] <cone-880> ffmpeg 03Rahul Chaudhry 07release/3.3:f322de901d2d: swresample/arm: remove unintentional relocation. [20:33:30 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:4b0181dc009d: avcodec/cinepak: move some checks prior to frame allocation [20:33:31 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:5d42300731b1: avcodec/cinepak: Skip empty frames [20:33:32 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:45f03502c93d: avcodec/dfa: Check dimension against maximum [20:33:33 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:bd6a181d5a49: avcodec/dsicinvideo: Propagate errors from cin_decode_rle() [20:33:34 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:7784a7c1d8df: avcodec/dsicinvideo: Fail if there is only a small fraction of the data available that comprises a full frame [20:33:35 CEST] <cone-880> ffmpeg 03Stephan Holljes 07release/3.3:2d975fff0548: lavf/http.c: Free allocated client URLContext in case of error. [20:33:36 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:bcc47312d4fd: avcodec/ffv1enc: Check that the crc + version combination is supported [20:33:37 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:77121f6bff4d: avcodec/vc1_block: simplify ac_val computation [20:33:38 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:3d29170013b1: avcodec/elsdec: Fix memleaks [20:33:39 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:69f861be42ec: avcodec/h263dec: Check slice_ret in mspeg4 slice loop [20:33:40 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:986747c9e25d: avcodec/error_resilience: Fix integer overflow in filter181() [20:33:41 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:bb7b9238b412: avcodec/wavpack: Fix integer overflow in DEC_MED() / INC_MED() [20:33:42 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:90e9b76d3a9b: avcodec/flac_parser: Fix infinite loop [20:33:43 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:6315215e5c0b: avcodec/g2meet: Change order of operations to avoid undefined behavior [20:33:44 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:083c48e6d9db: avcodec/jpeg2000dec: Skip init for component in CPRL if nothing is to be done [20:33:45 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:716ab0f2f456: avcodec/jpeg2000dec: Fix undefined shift in the jpeg2000_decode_packets_po_iteration() CPRL case [20:33:46 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:d48738bd674f: avcodec/g2meet: Check RGB upper limit [20:33:47 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:38fd80c656b4: avcodec/fic: Check available input space for cursor [20:33:48 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:8283586dce95: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_DD97iH0 / COMPOSE_DD137iL0 [20:33:49 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:c48ac14c067d: avformat/mov: Only fail for STCO/STSC contradictions if both exist [20:33:50 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:b0207c5b2a82: indeo4: Decode all or nothing of a band header. [20:33:51 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:fdab123c0312: avcodec/mlpdec: Only change noise_type if the related fields are valid [20:33:52 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:d1bac7f77772: avcodec/mpeg4videoenc: Use 64 bit for times in mpeg4_encode_gop_header() [20:33:53 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:45ce622f9093: avcodec/g723_1dec: Clip bits2 in both directions [20:33:54 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:e95a0e261fe9: oavcodec/aacpsdsp_template: Use unsigned for hs0X to prevent undefined behavior [20:33:55 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:6d8859e23474: avcodec/aacdec_fixed: use 64bit to avoid overflow in rounding in apply_dependent_coupling_fixed() [20:33:56 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:3be0bcbef6aa: avcodec/g2meet: ask for sample with overflowing RGB [20:33:57 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:05e253684ebc: avcodec/fic: Avoid some magic numbers related to cursors [20:33:58 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:aa50d67cc7d0: avformat/mov: Break out early if chunk_count is 0 in mov_build_index() [20:33:59 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:9ecb1998b034: avformat/mov: replace a value error by clipping into valid range in mov_read_stsc() [20:34:00 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:6577ae7119eb: avcodec/amrwbdec: Fix division by 0 in find_hb_gain() [20:34:01 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:412872e6cbf6: avcodec/opus_silk: Change silk_lsf2lpc() slightly toward silk/NLSF2A.c [20:34:02 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:0f1e6771d03c: avcodec/truemotion2: Fix overflow in tm2_apply_deltas() [20:34:03 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:bb6d47cf3583: avcodec/mjpegdec: Fix integer overflow in ljpeg_decode_rgb_scan() [20:34:04 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:8da3d6916316: avcodec/shorten: Check non COMM chunk len before skip in decode_aiff_header() [20:34:05 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:f51163b1666e: avcodec/shorten: Sanity check nmeans [20:34:06 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:b7134d7fb679: avcodec/shorten: Fix a negative left shift in shorten_decode_frame() [20:34:07 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:c2d16aafbb15: avcodec/shorten: Fix undefined shift in fix_bitshift() [20:34:08 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:3f66c3386c24: avcodec/shorten: Fix multiple integer overflows [20:34:09 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:719931c036e7: avcodec/wavpack: Fix overflow in adding tail [20:34:10 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:a16a4fefb6d1: avcodec/xwddec: Use ff_set_dimensions() [20:34:11 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:5f771a74b23f: avcodec/h264_mc_template: Only prefetch motion if the list is used. [20:34:12 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:f4f9bd655d42: avcodec/h264_ps: Move MAX_LOG2_MAX_FRAME_NUM to header so it can be used in h264_sei [20:34:13 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:98f10ca12ccc: avformat/mov: Only set pkt->duration to non negative values [20:34:14 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:b5d51d23a068: fftools/ffmpeg: Fallback to duration if sample rate is unavailable [20:34:15 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:96cf8a9cf993: avutil/common: Fix undefined behavior in av_clip_uintp2_c() [20:34:16 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:c66723ec4980: avcodec/dirac_dwt_template: Fix undefined behavior in interleave() [20:34:17 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:516f8dbc58a2: avcodec/aacdec_fixed: Fix undefined integer overflow in apply_independent_coupling_fixed() [20:34:18 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:4fb69a60f6c0: avcodec/mjpegdec: Check for end of bitstream in ljpeg_decode_rgb_scan() [20:34:19 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:76c4838a11a6: avcodec/h264_parser: Reduce needed history for parsing mb index [20:34:20 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:b0c14b02b057: avcodec/escape124: Check buf_size against num_superblocks [20:34:21 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:3db270a07862: avcodec/cscd: Check output buffer size for lzo. [20:34:22 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:eedde18f1a8f: avcodec/ra144: Fix integer overflow in ff_eval_refl() [20:34:23 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:228c05ee547d: avcodec/escape124: Fix spelling errors in comment [20:34:24 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:672ada0f179b: avformat/movenc: Do not pass AVCodecParameters in avpriv_request_sample [20:34:25 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:0bc74532a947: avcodec/jpeg2000dec: Check that there are enough bytes for all tiles [20:34:26 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:bba9bb735a53: avcodec/jpeg2000dec: Fixes invalid shifts in jpeg2000_decode_packets_po_iteration() [20:34:27 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:739e3ff47533: avcodec/shorten: Fix undefined integer overflow [20:34:28 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:01a694d4823c: avcodec/shorten: Fix undefined addition in shorten_decode_frame() [20:34:29 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:c43f5e0dce1b: avcodec/indeo4: Check for end of bitstream in decode_mb_info() [20:34:30 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:34654d41d4bd: avformat/rmdec: Do not pass mime type in rm_read_multi() to ff_rm_read_mdpr_codecdata() [20:34:31 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:9bb3047060c3: avformat/pva: Check for EOF before retrying in read_part_of_packet() [20:34:32 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:4d77a4a54d2f: avformat/mms: Add missing chunksize check [20:34:33 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:68fb93be04a4: avcodec/vp8_parser: Do not leave data/size uninitialized [20:34:34 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:78b1fbca3404: avformat/movenc: Check that frame_types other than EAC3_FRAME_TYPE_INDEPENDENT have a supported substream id [20:34:35 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:314ca5672d37: avcodec/mjpegdec: Check for odd progressive RGB [20:34:36 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:6a0a16e563f0: avformat/movenc: Check input sample count [20:34:37 CEST] <cone-880> ffmpeg 03Rahul Chaudhry 07release/3.3:0e9ecdabf3a4: swresample/arm: rename labels to fix xcode build error [20:34:38 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:87ddf73e52b4: avformat/movenc: Write version 2 of audio atom if channels is not known [20:34:39 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:c8ca54961dcc: avcodec/dirac_dwt_template: Fix signedness regression in interleave() [20:34:40 CEST] <cone-880> ffmpeg 03Simon Thelen 07release/3.3:28e0e494ee64: avcodec/imgconvert: fix possible null pointer dereference [20:34:41 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:db85ff35788d: avcodec/dvdsub_parser: Init output buf/size [20:34:42 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:eae9ab019301: avcodec/dvdsub_parser: Allocate input padding [21:24:30 CEST] <cone-880> ffmpeg 03Michael Niedermayer 07release/3.3:18c9d5d3e80d: update for 3.3.8 [21:26:05 CEST] <kurosu> so, what's the status on ffav1? Has anyone started ? [21:27:26 CEST] <kurosu> I'm curious what the slowest part is in there in "normal" bitstreams [21:27:57 CEST] <kurosu> I suspect entropy decoding [21:30:26 CEST] <atomnuker> someone's already finished, lol, its just that the code isn't public because $$$ [21:31:29 CEST] <atomnuker> they'll get their dosh very soon though [21:32:10 CEST] <atomnuker> depends what you consider entropy coding though [21:33:06 CEST] <kurosu> I don't even know if it's a "boolean"/range/arithmetic coder in there [21:33:31 CEST] <kurosu> is that 100% or some profiles ? [21:33:46 CEST] <kurosu> like the global compensation left out [21:35:27 CEST] <j-b> I doubt we have samples for the whole spec [21:35:40 CEST] <j-b> kurosu: what part was the slowest in VP9? [21:36:32 CEST] <nevcairiel> its pretty much always entropy decoding on medium to high bandwidth [21:36:34 CEST] <kurosu> no idea, but likely not as disproportionate as in codecs with arithmetic coders [21:37:04 CEST] <kurosu> vp9 has fixed probabilities afaik => no need to update that part of the state [21:38:42 CEST] <kurosu> at some point I ran perf on an hevc decoder and spent some time organizing the data [21:38:50 CEST] <kurosu> I don't think I've kept that around [21:39:23 CEST] <kurosu> obviously cabac was higher there, like 20% just on a single type of bits (significance map) [21:40:15 CEST] <kurosu> mc was weird, but then again its code size makes it not fit in icache [21:46:06 CEST] <atomnuker> the entropy coding system is quite lightweight and efficient, and adaptation is very fast and simdable [21:46:23 CEST] <atomnuker> but coeficients are coded through another layer called lvmap [21:47:48 CEST] <kurosu> simdable? you need simd? or does that allow you to output multiple symbols ? [21:47:51 CEST] <atomnuker> which isn't exactly fast and I think is the main bottleneck that limited the spec's max bitrate [21:48:02 CEST] <atomnuker> its multisymbol [21:48:15 CEST] <kurosu> if it's like hevc, it's all for h/w, and fuck s/w [21:48:47 CEST] <nicolas17> what is URLProtocol.default_whitelist? I kinda know what the whitelist is for in general terms, but I don't understand the exact implication of protocol A having default_whitelist="B" [21:49:09 CEST] <atomnuker> if anything there's hardly much in the codec made to make hardware easy and software terribly slow [21:49:11 CEST] <kurosu> ok, anyway, some profiling will be nice, but then anyone can run it if it gets posted [21:49:22 CEST] <atomnuker> rather, everything gets screwed equally [21:50:20 CEST] <atomnuker> there's a slight tweak in the entropy coding system to reduce the range coder's multiply size for hardware but it didn't affect efficiency much (0.01%) and its barely a shift and an add (I think) per symbol [21:50:27 CEST] <kurosu> bah, as long as it's real-time on 4 cores for 4K60p... (which I think hevc isn't) [21:50:47 CEST] <kurosu> (or rather ffhevc) [21:51:13 CEST] <atomnuker> EC is mostly frequency bound [21:51:37 CEST] <kurosu> yup, can't get that spectre-enable optimization working its magic [21:52:03 CEST] <atomnuker> spectre-enabled optimization? [21:52:14 CEST] <kurosu> speculative execution [21:52:23 CEST] <nicolas17> the hardware optimizations that enable the spectre vulns :P [21:53:35 CEST] <kurosu> or is that another frequency than the cpu's you're speaking of ? [21:53:39 CEST] <atomnuker> welp, I don't think the fix affects EC systems much, after all usually all your EC reading functions are a single function (with maybe inline asm magick) [21:54:05 CEST] <atomnuker> so retpoline doesn't take place until you return [21:55:14 CEST] <kurosu> what I meant is that it's highly unpredictable and in tight loops, so you can't really get much concurrent operations going [21:55:49 CEST] <kurosu> retpoline is, if I'm not mistaken, a mitigation at the end of syscalls ? [21:56:01 CEST] <kurosu> (in kernel space) [21:56:23 CEST] <atomnuker> I thought it was after every function, after all gcc and clang default with it enabled now [21:56:31 CEST] <kurosu> possible [21:56:56 CEST] <atomnuker> "-mindirect-branch, -mindirect-return and -mindirect-branch-register" [21:57:06 CEST] <kurosu> but as you said, you likely don't want function calls in EC [21:57:44 CEST] <kurosu> (or rather, "inline asm magick") [22:00:09 CEST] <nicolas17> it *can* be done after every function [22:00:25 CEST] <nicolas17> but it's only enabled where it matters, such as the kernel [22:00:58 CEST] <nicolas17> hmm my gcc 7.3 manpage doesn't even have -mindirect* [22:08:32 CEST] <kierank> j-b: I think argon have the whole spec [22:09:32 CEST] <kurosu> weren't they the ones finding loopholes for vp9 - I remember BBB changing some transforms because of unexpected coefficients or the like [22:10:02 CEST] <kierank> yes, that's why they are part of the process now as far as i can tel [22:10:03 CEST] <BBB> I believe they made the compiler-generated conformance streams [22:10:22 CEST] <kurosu> yeah, I saw that about their methodology [22:10:29 CEST] <kurosu> very impressive [22:42:25 CEST] <durandal_1707> shit, not going to find job here [22:46:28 CEST] <j-b> kierank: samples for all of those? [22:46:48 CEST] <kierank> j-b: can get through aomedia, no? [22:46:50 CEST] <kierank> or pay argon monies [22:47:05 CEST] <kierank> durandal_1707: play for your country gootball team [22:48:09 CEST] <j-b> kierank: when is next madness? 2 years for Euro? [22:48:22 CEST] <kierank> yes or rugby next year [22:48:48 CEST] <kierank> or next falcon heavy launch [22:48:50 CEST] <j-b> Japan? [00:00:00 CEST] --- Tue Jul 17 2018

1 0

ffmpeg.log.20180716
by burek 17 Jul '18

17 Jul '18

[00:44:54 CEST] <kubast2> Does libaom-av1 use h.264 https://i.imgur.com/x8CdYDo.png ? [00:45:14 CEST] <kubast2> I never seen it print out hex too [00:45:26 CEST] <kubast2> I didn't tried like 1080p encode before [00:45:33 CEST] <kubast2> or usage of -threads 4 so [00:46:18 CEST] <furq> kubast2: that message is from the decoder [00:46:41 CEST] <kubast2> oh [00:47:18 CEST] <furq> i'm not sure what the hex dump is but i guess that's some libaom debug thing [00:54:08 CEST] <nicolas17> is it possible to know which of two videos has better quality, without having a max-quality "original" to compare to? I suspect A was transcoded into B, and possibly B has higher bitrate (thus bigger filesize), but obviously B will have lost quality vs A [00:56:18 CEST] <furq> nothing better than just using your eyes [00:56:35 CEST] <nicolas17> okay [00:56:48 CEST] <nicolas17> that's what I thought... [01:08:57 CEST] <nicolas17> when someone streams to Twitch at 720p30, the player lets you select "720p" or "720p (source)", they correspond to HLS .m3u8 URLs containing /720p30/ or /chunked/ respectively, I *suspect* the source/chunked one is the "original" and all other variants (even the one with the same resolution!) are transcoded from it [01:10:56 CEST] <BtbN> The one called Source is always the best possible. [01:11:19 CEST] <nicolas17> I wonder why they transcode it to "720p30" with *higher* bitrate [01:12:03 CEST] <BtbN> Because they only have a defined set of transcodes. They never increase res/framerate, but otherwise they just have their fixed set [01:12:26 CEST] <DHE> maybe in case someone encodes at yuv444 so they produce yuv420 ? [01:12:36 CEST] <BtbN> They will straight up reject that [01:12:55 CEST] <BtbN> Your original input stream has to be playable by browsers [01:14:00 CEST] <nicolas17> oh wow I just compared the two, they don't even have the same brightness [01:14:10 CEST] <nicolas17> possibly because: [01:14:22 CEST] <nicolas17> 720p60/1000.ts: yuv420p(progressive) [01:14:32 CEST] <nicolas17> chunked/1000.ts: yuv420p(tv, bt470bg/unknown/unknown, progressive) [01:15:08 CEST] <nicolas17> that's color space right? [12:31:30 CEST] <blair_> hi all [12:32:27 CEST] <blair_> is it possible to use av_read_packet with h264 then pull out the frames with av_parser_parse2? I want to get raw frames and then send to cuda [12:33:13 CEST] <blair_> I have a camera I am reading the frames from in a loop then just need to pull out the uncompressed data [12:33:25 CEST] <blair_> or am i missing something [12:34:40 CEST] <DHE> just a raw h264 bitstream? [12:34:54 CEST] <blairdavidson> believe so [12:36:02 CEST] <blairdavidson> it all works when i decode it with ffmpeg but i want to move the work now to the gpu so i want to skip this step [16:52:54 CEST] <utack> does anyone know what is going on at doom9? [16:53:14 CEST] <JEEB> swede's apartment went offline again? [16:53:17 CEST] <JEEB> if it's down [17:02:59 CEST] <DHE> seems down for me... [17:03:45 CEST] <JEEB> but yes, it's hosted in the swede's basement (without any jokes) [17:05:37 CEST] <DHE> oh my... it is... [17:08:00 CEST] <utack> lol JEEB [17:08:03 CEST] <utack> i did not know that [17:11:51 CEST] <DHE> which is a shame because I was just googling something with doom9 results... [17:12:04 CEST] <DHE> oh well, cached view ftw [17:41:52 CEST] <Hello71> been down since two days ago I think [17:46:55 CEST] <DHE> :( [21:16:56 CEST] <kgrid03> Hi! I am new to command prompts, and only started using it for ffmpeg because that is the only way I can convert .flv files to .avi or another useable format for premiere pro. I downloaded a static 64 bit version of ffmpeg and set up my path in my PC's extended variables. I had some trouble with "access is denied" and "permission is denied" so I changed my system 32 ownership to admin and my user, changing the settings to [21:16:56 CEST] <kgrid03> full control, and I also turned of my UAC, and only run command as an admin. I also had problems with "there is no such file in this directory" and "this is not a valid command (giving me 3 options of what would be". This is the closest I have gotton for it to work, as it seems to want to convert the file, but I cannot find a converted file anywhere in my PC. This is what the command prompt is giving me, and I can't exactly [21:16:56 CEST] <kgrid03> tell what it is.C:\WINDOWS\system32>ffmpeg -i c:\users\kgrid03\videos\whereclips\opening\CarlS.flv CarlS.avi [21:16:56 CEST] <kgrid03> ffmpeg version 4.0.1 Copyright (c) 2000-2018 the FFmpeg developers [21:16:56 CEST] <kgrid03> built with gcc 7.3.1 (GCC) 20180710 [21:18:37 CEST] <Hello71> so... "I reverted to windows xp nonsecurity" [21:32:33 CEST] <DHE> shoutouts to irccloud for just flooding people out [21:52:34 CEST] <automatical> hi folks, I'm tryin to do some live encoding of a network camera that I'm capturing over TCP - is there a way to have ffmpeg duplicate the last good frame when a frame is dropped? [22:11:21 CEST] <Hello71> maybe but why [00:00:00 CEST] --- Tue Jul 17 2018

1 0

ffmpeg-devel.log.20180715
by burek 16 Jul '18

16 Jul '18

[02:58:26 CEST] <kierank> atomnuker: the transform was broken in the "undefined behaviour fixes" [03:15:59 CEST] <atomnuker> kierank: which one? was it fixed [03:16:19 CEST] <kierank> https://ffmpeg.org/pipermail/ffmpeg-devel/2018-July/232287.html [03:31:27 CEST] <atomnuker> oh, this should have broken practically all transforms [03:46:44 CEST] <nicolas17> what is URLProtocol.default_whitelist? [04:55:32 CEST] <nicolas17> I know what the whitelist is in general terms [04:56:16 CEST] <nicolas17> but I don't understand the exact implication of protocol A having default_whitelist="B" [10:21:24 CEST] <cone-988> ffmpeg 03Stephen Seo 07master:0ea20124b710: Add lensfun filter [10:24:54 CEST] <cone-988> ffmpeg 03Paul B Mahol 07master:5d8df3cc4ac0: Update Changelog for lensfun addition and bump minor [12:23:08 CEST] <JEEB> anyone objects me cherry-picking 8c2c97403baf95d0facb53f03e468f023eb943e1 into 3.4 ? [12:23:23 CEST] <JEEB> because it seems like we are currently getting segfaults in that branch [12:23:31 CEST] <JEEB> https://github.com/mpv-player/mpv/issues/5989#issuecomment-405046588 [15:21:47 CEST] <atomnuker> BBB: what file did you hate the most from libaom? [15:21:56 CEST] <atomnuker> mvref_common.c? [15:29:01 CEST] <BBB> yes, it got better since they merged mfmv but its still dumb [15:34:15 CEST] <atomnuker> mfmv? when was it merged [15:34:58 CEST] <JEEB> anyone has any opinions on pushing a cherry-pick of a segfault fix to release/3.4 and making a new release with it, since 3.4.3 seems to crash in `avcodec_find_best_pix_fmt_of_list` ? [15:35:19 CEST] <atomnuker> just push it [15:35:33 CEST] <JEEB> k [15:36:25 CEST] <cone-651> ffmpeg 03Simon Thelen 07release/3.4:c1e172c2e14e: avcodec/imgconvert: fix possible null pointer dereference [15:36:30 CEST] <JEEB> there [15:37:23 CEST] <JEEB> some gentoo user using 3.4.x releases reported a crash on my side and I had him test cherry-picking that patch [15:37:27 CEST] <JEEB> and it fixed it for him [15:39:37 CEST] <JEEB> I wonder if any other release branch got that first back-port in the imgconvert line, but not the crash-fixing one [15:45:31 CEST] <JEEB> ok, 3.2.11 has the same issue [15:48:19 CEST] <cone-651> ffmpeg 03Simon Thelen 07release/3.2:c773ce32e1f0: avcodec/imgconvert: fix possible null pointer dereference [15:48:23 CEST] <JEEB> there we go [15:52:47 CEST] <JEEB> michaelni: since you handled the last 3.2/3.4 releases I guess I recommend another one to fixup the segfault in avcodec_find_best_pix_fmt_of_list ? [18:55:19 CEST] <atomnuker> ubitux: congrats [18:55:23 CEST] <atomnuker> durandal_1707: condolences [18:55:32 CEST] <ubitux> :/ [18:56:22 CEST] <ubitux> you have no idea the level of chaos there is in my neighborhood [18:56:44 CEST] <nevcairiel> i was happy when germany failed out, no noise anylonger [18:56:48 CEST] <JEEB> :D [18:56:57 CEST] <ubitux> yelling, honking, firecrackers [18:57:04 CEST] <ubitux> it's gonna be that for the next hours [18:57:10 CEST] <ubitux> along with burnt cars and shit [19:00:16 CEST] <atomnuker> they burn cars for this? [19:00:54 CEST] Action: DHE switches to baseball now... [19:02:35 CEST] <kierank> france will burn cars for anything [19:03:55 CEST] <j-b> Here, it is insane [19:25:03 CEST] <atomnuker> even here in montreal there were a few people on the streets waving french flags [19:34:50 CEST] <michaelni> JEEB, i had that bug backported in more branches locally, ill backport the fix as well to all of them. About releases i plan to make one from 2,8 next as that had no release since quite a while. Ill look into 3.* after that [19:35:26 CEST] <JEEB> roger [20:05:40 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07master:7334985ffae8: libavcodec/cfhd: Fix signed overflow in shift [20:05:40 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07master:8d21ab4d128d: avcodec/lagarith: Check that the range coded data stream is consistent when the probabilities indicate no data could have been coded. [20:05:40 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07master:9e6c84377616: avcodec/dvdsub_parser: Init output buf/size [20:05:40 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07master:2e6ea3938669: avcodec/vp8_parser: Remove redundant output initialization [20:05:40 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07master:cd86b5cfe278: avcodec/dvdsub_parser: Allocate input padding [20:09:09 CEST] <jamrial> jkqxz: https://pastebin.com/SFBq4F7v [20:12:28 CEST] <jamrial> jkqxz: with those fixes it works great for all the samples i tried except two that seem to have global_motion_params which you haven't yet implemented, so it just fails to parse the unit [20:43:16 CEST] <cone-496> ffmpeg 03Rahul Chaudhry 07release/2.8:9ffbda72c768: swresample/arm: rename labels to fix xcode build error [20:43:17 CEST] <cone-496> ffmpeg 03Simon Thelen 07release/2.8:03af6ab540cc: avcodec/imgconvert: fix possible null pointer dereference [20:43:18 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07release/2.8:d25463c515cd: avcodec/dvdsub_parser: Init output buf/size [20:43:19 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07release/2.8:83679bd293c4: avcodec/dvdsub_parser: Allocate input padding [20:43:20 CEST] <cone-496> ffmpeg 03Michael Niedermayer 07release/2.8:48e104a8240d: update for 2.8.15 [00:00:00 CEST] --- Mon Jul 16 2018

1 0

ffmpeg.log.20180715
by burek 16 Jul '18

16 Jul '18

[00:01:51 CEST] <killown> ffmpeg -i test.mp4 -ss 00:00:00 -to 00:0:21 -strict -2 -c copy out.mp4 hmmm [02:57:38 CEST] <analogical> killown, you could use MkvToolnix but the output will an mkv-file [06:35:17 CEST] <stmps> hey guys, i'm trying to hardcore/burn in .srt subtitles for a media player that doesn't support separate sub files. using ffmpeg like so: https://pastebin.com/XmmfErNE [06:36:02 CEST] <stmps> testing the .srt with VLC works fine, but when trying to hardcore with ffmpeg, the output media doesn't have any subs burned in [06:37:16 CEST] <stmps> am I doing something wrong? my ffmpeg seems to have been compiled with --enable-libass, as is required by the subtitle filter [11:02:52 CEST] <Henry_> hi [15:05:48 CEST] <fiftysix> hi! [15:06:08 CEST] <fiftysix> is there a way to run a video file that freezes somewhere in the middle through ffmpeg in the hopes of repairing it somehow? [15:42:43 CEST] <kepstin> depends on the reason why it's freezing. if it's a corrupt file that a player using ffmpeg libraries (e.g. mpv) can't play through, then probably not. [15:42:55 CEST] <kepstin> but sometimes... maybe [15:43:46 CEST] <kepstin> er, wait, fiftysix already left [16:06:54 CEST] <Croepha> hello, im trying to stream from ffmpeg to ffplay on a lan, on ffmpeg output I have: rawvideo (UYVY / 0x59565955), uyvy422(top first), 1920x1080, q=2-31, 995328 kb/s, 30 tbr, 90k tbn, 1000k tbc but on ffplay i have: rawvideo (UYVY / 0x59565955), uyvy422, 1920x1080, 829440 kb/s, 25 tbr, 25 tbn, 25 tbc the video is just streaks, so im thinking the rate is off... would that make sense? if so how do I fix ffplay? [16:26:34 CEST] <kepstin> what network protocol? that's basically a gigabit/s of video, so you're probably having issues due to packet loss if udp, or just bottlenecks otherwise [16:39:08 CEST] <Croepha> kepstin actually I got it working, I needed -f rawvideo on recv side [16:39:19 CEST] <Croepha> now im just trying to tune latency [16:40:09 CEST] <Croepha> i have this on transmit: ./ffmpeg -fflags nobuffer -flags low_delay -format_code Hi60 -f decklink -i 'UltraStudio Mini Recorder' -c:v copy -an -f rawvideo "udp://192.168.4.140:12334" and this on recv: ffplay -fflags nobuffer -flags low_delay -s 1920x1080 -pix_fmt uyvy422 -f rawvideo -i "udp://:12334?" [16:40:24 CEST] <Croepha> still seems to have about 1 second of latency not sure why [17:03:06 CEST] <Mavrik> But... why. [17:03:35 CEST] <Mavrik> Streaming to the limit of performance this way is not going to be stable O.o [17:13:36 CEST] <Croepha> Mavrik: well, its a lan connection, I can make some assurances that packet loss will not be a big issue [17:14:22 CEST] <Croepha> i basically need a VNC like solution... imagine, you were remotted into a machine, and you got to see every keystroke a second later [17:15:16 CEST] <Croepha> but its a little more complicated, because I cant simply run a vnc server on what i want, so I have to capture it through hdmi [17:16:22 CEST] <kepstin> encoding with e.g. x264 in zerolatency mode should only add a small number of ms more per frame, and significantly reduce bandwidth [17:17:29 CEST] <Mavrik> Croepha: at 1GBit packet loss will be an issue no matter what kind of connection you have [17:17:39 CEST] <Mavrik> any kind of small queue issue will cause artifacts [17:18:50 CEST] <Mavrik> Use at least some simple lossless compression [17:18:59 CEST] <Mavrik> It'll save you like 5x as much bandwidth [17:21:59 CEST] <kepstin> I mean, if you're running 10g ethernet it might work ok :) [00:00:00 CEST] --- Mon Jul 16 2018

1 0