Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

May 2018

1 participants
62 discussions

ffmpeg-devel.log.20180430
by burek 01 May '18

01 May '18

[00:05:22 CEST] <Chloe> durandal_1707: what you writing asm for? [00:10:59 CEST] <durandal_1707> nlmeans filter function [02:33:17 CEST] <atomnuker> oh look its another soname bump for libx265 [11:42:45 CEST] <Chloe> should I be using av_mallocz_array over just av_mallocz(size*count)? [11:43:03 CEST] <Chloe> err, im being silly. I should just use calloc [11:46:04 CEST] <Chloe> so why does av_mallocz_array exist [11:48:26 CEST] <durandal_1707> Chloe: Libav NIH syndrome [11:54:35 CEST] <iive> Chloe, i think all memory allocated by ffmpeg routines should be memory aligned to largest SIMD word [11:54:59 CEST] <Chloe> iive: but why av_mallocz_array over av_calloc? [11:55:35 CEST] <iive> that i have no idea. [12:04:16 CEST] <Squeak> hello [12:04:58 CEST] <Chloe> Squeak: guten Tag [12:05:17 CEST] <Squeak> can tell me someone how can i use unicode version of function "avformat_open_input" ? [12:05:46 CEST] <Squeak> exist any unicode version ? [12:06:11 CEST] <Chloe> does it not just take utf8? [12:06:30 CEST] <iive> do you mean windows ucs2 ? [12:06:38 CEST] <Squeak> yes [12:06:52 CEST] <iive> aka utf16. [12:06:58 CEST] <Squeak> yes [12:07:24 CEST] <Chloe> Squeak: I dont think ffmpeg has any utf16 functions [12:07:41 CEST] <Squeak> :-) [12:09:46 CEST] <SqueakPL> can i use "_wopen" to open FILE, ans then use FILE structure ? [12:10:27 CEST] <nevcairiel> you can use whatever you want, but not with ffmpeg [12:14:04 CEST] <iive> does windows support utf8 for the CreateFileA() ? [12:14:21 CEST] <iive> or that would depend on locale settings? [12:14:43 CEST] <nevcairiel> unicode with the A functions doesnt work, no [12:14:52 CEST] <nevcairiel> the A functions use the system locale [12:15:31 CEST] <nevcairiel> but ffmpeg automatically translates string to wide-character and uses the W functions if you use its file open API [12:16:06 CEST] <Chloe> nevcairiel: so ffmpeg only takes in native or does it take utf8? [12:17:08 CEST] <nevcairiel> ffmpeg understands utf8, it converts it to wchar for the Windows API, so you dont h ave to worry about it [12:18:43 CEST] <Chloe> I guess that makes it annoying if you're working with wchar on windows since you have to convert to utf8 only for ffmpeg to convert it back >.< [12:19:34 CEST] <nevcairiel> you're not supposed to care what ffmpeg does with it internally [12:22:10 CEST] <Chloe> good point [13:56:20 CEST] <jdarnley> Blame Windows and everyone else (like Lua) who thinks it is okay to have a NUL byte in a string! [14:01:06 CEST] <nevcairiel> its a wchar, its not bytes, its words [14:01:11 CEST] <nevcairiel> no null words in there [14:28:40 CEST] <durandal_1707> how to store only first 4 bytes from xmm register into memory? [14:35:57 CEST] <durandal_1707> pextrd? [14:37:11 CEST] <nevcairiel> movd [14:46:28 CEST] <BBB> movd [14:46:32 CEST] <BBB> pextrd is very slow [17:50:08 CEST] <kurosu> durandal_1707, even if you are doing integer maths, and it is said switching between float and int ops now has a penalty, I'd benchmark using movss too [18:09:45 CEST] <atomnuker> there's really no penalty for e.g. using xorps on simd registers which were just used for integer ops [18:09:58 CEST] <atomnuker> on intel CPUs since core 2 I think [18:15:59 CEST] <jamrial> if you're only going to handle ints, then the only float instructions that make sense using if needed are shufps and some of the complex movs that have no int version [18:17:05 CEST] <atomnuker> yeah, mainly the movs are useful [19:18:32 CEST] <Gramner> shufpd, movmskps, movmskpd as well [20:41:44 CEST] <cone-189> ffmpeg 03Michael Niedermayer 07master:34dbdcfc20d6: avformat/m4vdec: Fix detection of raw MPEG-4 ES Studio [20:41:45 CEST] <cone-189> ffmpeg 03Michael Niedermayer 07master:0f176bb8e089: avformat/m4vdec: Use the same constant names as libavcodec [20:41:46 CEST] <cone-189> ffmpeg 03Michael Niedermayer 07master:a47bd1cd1c71: avcodec/mpeg4video_parser: Fix incorrect spliting of MPEG-4 studio frames [20:41:47 CEST] <cone-189> ffmpeg 03Michael Niedermayer 07master:c0aa89eeee2a: avcodec/mpeg4video_parser: Avoid litteral 0x1B6, use named constant instead [20:46:53 CEST] <cone-189> ffmpeg 03Paul B Mahol 07master:da55304cafea: avfilter/vf_mix: initialize last to silence possible warning [21:28:17 CEST] <durandal_1707> Gramner: you solution give extra boost [21:29:31 CEST] <Gramner> durandal_1707: nice. also what's up with all the function parameters that aren't used? [21:34:41 CEST] <Gramner> durandal_1707: oh, and in scalar code you can do movzx regd, byte [mem] instead of zeroing and doing a byte load [21:36:23 CEST] <durandal_1707> Gramner: those are for cases for subsamples planes, when one averages alphas [21:39:41 CEST] <durandal_1707> i thinking how to write yuv422 case for chroma planes [21:39:52 CEST] <durandal_1707> and C version looks wrong [22:33:27 CEST] <cone-189> ffmpeg 03Marton Balint 07master:c60a824ee87a: avformat/qtpalette: parse color table according to the QuickTime file format specs [22:33:28 CEST] <cone-189> ffmpeg 03Marton Balint 07master:e894d958fce6: avcodec/anm: fix palette alpha [22:33:29 CEST] <cone-189> ffmpeg 03Marton Balint 07master:4c501bafc08c: avcodec/hnm4video: fix palette alpha [22:33:31 CEST] <cone-189> ffmpeg 03Marton Balint 07master:56b081da578f: avutil/pixdesc: add AV_PIX_FMT_FLAG_ALPHA to AV_PIX_FMT_PAL8 [22:33:32 CEST] <cone-189> ffmpeg 03Marton Balint 07master:7033654f7f3c: Use AV_PIX_FMT_FLAG_ALPHA for detecting transparency where nb_components was used [23:04:27 CEST] <durandal_1707> Gramner: instruction to take every 2nd byte and other places to be zeroes? 01 00 02 00 03 ... [23:05:45 CEST] <durandal_1707> use movu and pand ? [23:06:05 CEST] <Gramner> pand to mask away unwanted bits, yes [23:06:34 CEST] <Gramner> if you only care about avx then you can combine the movu and pand [23:06:50 CEST] <Gramner> since avx memory args can be unaligned [23:07:17 CEST] <Gramner> e.g. pand dst, const_mask, [mem] [23:07:44 CEST] <iive> keep the mask in register and let the pand use memory argument? [23:08:11 CEST] <Gramner> yes [23:09:20 CEST] <Gramner> assuming the pand is done multiple times. if it's only done once then there's no point to having the mask in a register [23:11:01 CEST] <jamrial> this is where avx512 k registers would come into play i guess? [23:11:26 CEST] <Gramner> that's one use case of them, sure [23:12:09 CEST] <Gramner> you can conditionally mask away elements or blend the result with the previous value of dst [23:12:12 CEST] <Gramner> for free [23:12:29 CEST] <Gramner> on basically every instruction [23:13:18 CEST] <Gramner> or well, setting up the kmask requires some instructions I guess, but that can be done once and reused a million times in a loop [23:16:19 CEST] <cone-189> ffmpeg 03Marton Balint 07master:75d1529c6e90: avdevice/decklink_dec: do not copy video data [23:16:20 CEST] <cone-189> ffmpeg 03Marton Balint 07master:649087fa83a5: avdevice/decklink_dec: unref packets on avpacket_queue_put error [23:27:56 CEST] <cone-189> ffmpeg 03James Almer 07master:e3866ea20d55: fftools/ffmpeg: fix mixed code and declarations [00:00:00 CEST] --- Tue May 1 2018

1 0

ffmpeg.log.20180430
by burek 01 May '18

01 May '18

[00:42:53 CEST] <TheAMM> https://ffmpeg.org/ffmpeg.html says on -vsync: "With -map you can select from which stream the timestamps should be taken." [00:43:29 CEST] <TheAMM> Do I understand that corrently, in that I could use two inputs (image2 and a video) and map the video's frame-timestamps to the image2 input? [00:53:00 CEST] <JEEB> TheAMM: umm, vsync IIRC only manages the weird vsync logic in ffmpeg.c unless there's a custom vsync AVOption in image2 etc [00:54:55 CEST] <TheAMM> I have a video which I've extracted the frames out as an image2 sequence, then edited the frames, and I want to encode those frames back to a video [00:55:27 CEST] <TheAMM> The kicker is that the video may be VFR, and I can't blindly set the framerate either (because container lies) (because it's mpv and 1000fps) [00:56:01 CEST] <JEEB> image2 and VFR sounds like not the best of things without extra tools :P [00:56:24 CEST] <TheAMM> Yap [00:56:52 CEST] <TheAMM> (to be specific I'm using image2pipe and no files don't touch the disk) [00:56:53 CEST] <JEEB> anyways, vsync is just the thing that controls how ffmpeg.c tries to duplicate and drop frames [00:57:05 CEST] <TheAMM> Yeah, it is [00:57:11 CEST] <JEEB> ok, if you're doing piping just output raw images in NUT or something? [00:57:22 CEST] <JEEB> that way you can have timestamps, too [00:57:25 CEST] <TheAMM> I was asking if -map allows me to map the timestamps from the video input to the image2 input [00:57:32 CEST] <TheAMM> NUT? [00:57:39 CEST] <JEEB> not that I know re: mapping [00:57:52 CEST] <JEEB> NUT is the container to use for piping raw images :P [00:57:58 CEST] <JEEB> pretty simple to generate/parse [00:58:07 CEST] <JEEB> has timestamps for each sample [00:58:38 CEST] <TheAMM> Sounds great [00:58:47 CEST] <TheAMM> Because it was going to be the next thing I'd ask for [00:59:49 CEST] <JEEB> but now that I thought of it, if your video is matroska you could just extract v1 or v2 timecodes (actually timestamps) with mkvtoolnix's tools [01:00:04 CEST] <JEEB> then just output matroska file with the image2 and constant frame rate [01:00:17 CEST] <JEEB> and finally apply the timestamps from the other thing with mkvtoolnix [01:00:25 CEST] <JEEB> if you're OK with coding some, do the NUT+piping way [01:01:16 CEST] <TheAMM> I'm not fond of adding dependencies [01:01:33 CEST] <TheAMM> Eventual goal is to do it all with mpv, since it wraps ffmpeg [01:12:24 CEST] <TheAMM> Ooookay, this is not a 15-minute 2AM coding session [04:48:36 CEST] <SpeakerToMeat> Can quicktime files have plain xml inside? [04:48:41 CEST] <SpeakerToMeat> I explain, somebody wants help with non playing "undeleted" video files. I suspect simply sectors of the file where overwriten after deletion, or it was non contiguous and somehow a contiguous block was recovered. the file has the moov at the start so mediainfo and ffprobe show its data, but nothing plays it. simply editing the file in an editor to check it "looks ok" (a.k.a. human giberish) I find in th [04:48:43 CEST] <SpeakerToMeat> emidle a short but complete xml, but the xml contents are congruent with camera metadata so I wonder if it's a sign of corruption, or if .mov is capable of having textual metadata inside in the form of human readable xml. [04:48:58 CEST] <SpeakerToMeat> Well actually it's a .mp4... almost the same thing but there might be an important distinction, and has a moov header (at the start) [05:04:22 CEST] <cc2> I'm hitting a configure error when building GStreamer for ios. The error pointed me here. It looks like an argument being passed to clang that it doesn't support. Here's a link to the log: https://www.dropbox.com/s/a52pj5tf57jvm2i/config.log?dl=0. Any pointers would be appreciated! [05:12:04 CEST] <cc2> this is the piece that seems to matter: clang: error: unknown argument: '-mincoming-stack-boundary=4' [10:23:38 CEST] <jcelerier1> hi :) [10:23:56 CEST] <jcelerier1> is there a list of the API-breaking changes in ffmpeg 4 somewhere ? [10:24:33 CEST] <jcelerier1> when linking my app with ffmpeg 3.4 everything works, but when using ffmpeg 4 I can't load any sound [10:24:48 CEST] <jcelerier1> (I'm getting "protocol not found" errors) [10:35:02 CEST] <durandal_1707> jcelerier1: what configure options did you used? perhaps you need to whitelist protocol you are using. [10:36:42 CEST] <jcelerier1> oh ? that would make sense, but did this change between ffmpeg 3 -> 4 ? (I used almost the same configure line) [10:37:10 CEST] <jcelerier1> durandal_1707: ./configure --arch=x86_64 --cpu=x86_64 --disable-doc --disable-ffmpeg --disable-ffplay --disable-debug --prefix=/usr/local --pkg-config-flags="--static" --enable-gpl --enable-version3 --disable-x86asm --disable-openssl --disable-protocols --disable-securetransport --disable-videotoolbox --disable-network --disable-iconv --disable-lzma [10:37:27 CEST] <jcelerier1> yeah, I guess --disable-protocols is the problem [10:39:30 CEST] <durandal_1707> indeed [10:39:40 CEST] <durandal_1707> you need atleast file protocol [10:40:07 CEST] <durandal_1707> and --disable-x86asm is it really needed? [10:40:22 CEST] <jcelerier1> durandal_1707: I didn't want to handle the nasm dependency [10:40:39 CEST] <jcelerier1> but if it's not too painful to set-up maybe I can add it back [10:43:32 CEST] <jcelerier1> also, apparently av_register_all / avcodec_register_all are now deprecated but the code comments still say that you have to call them ? [10:51:19 CEST] <durandal_1707> jcelerier1: buidling with nasm will give you quite slow decoding [10:51:29 CEST] <durandal_1707> *without nasm [10:53:58 CEST] <Mavrik> jcelerier1: there's an APIchanges file in doc/ [10:54:16 CEST] <Mavrik> Also running ffmpeg without asm is pretty insane :P [10:55:07 CEST] <jcelerier1> I must admit I was too lazy to devise how to set it up on ms windows [10:55:39 CEST] <Mavrik> Isn't it like a single .exe file? :) [10:55:50 CEST] <Mavrik> Anyway most CPU optimizations are in ASM so you'll have a severe performance penalty [10:56:39 CEST] <jcelerier1> laziness knows no bounds :p [11:33:59 CEST] <dragmore88> hi, anyone that can take a peak on my syntax and see why the x265 options arent passed along to x265? https://pastebin.com/2eSXCtZT [12:26:16 CEST] <Nethe> Hi Guys! Question about decoding & encoding videos on Nvidia GPU. When I'm having mpeg2video input, trying to run decoder & encoder using this command: https://pastebin.blesmrt.net/pkgvmwd4f/dkbj4x . But it is giving me this error: https://imgur.com/a/t27qnx2 [12:26:26 CEST] <Nethe> Can you please help me to solve the issue? [12:58:10 CEST] <friki_> Nethe: probably you should post also the full command [12:58:56 CEST] <Nethe> friki_: ok, just a second [13:00:15 CEST] <Nethe> friki_: here it is: ffmpeg -y -c:v mpeg2_cuvid -gpu 3 -vsync 0 -deint 2 -drop_second_field 1 -surfaces 10 -i 'udp://@227.10.20.149:1234?fifo_size=100000000' -map i:0x12c -gpu 2 -r 25 -filter fps=30,scale_npp=640:360:format=same:interp_algo=lanczos -aspect 16:9 -g 96 -vcodec h264_nvenc -b 400K -minrate 400k -maxrate 400k -f mpegts 'udp://@239.0.0.125:1000?overrun_nonfatal_option=1' (also posted on the first link in my original message) [17:09:19 CEST] <leewdch> is handbrake using ffmpeg to convert videos? [17:09:33 CEST] <JEEB> yes, it is an application that is an API client for FFmpeg [17:09:51 CEST] <leewdch> Im having troubles converting an swf file into mp4/mkv [17:10:06 CEST] <leewdch> I tried both via cli with ffmpeg and with handrbake but since it uses ffmpeg it's obvious the result [17:11:17 CEST] <durandal_1707> swf can be various crap [17:11:29 CEST] <leewdch> indeed, I wonder if I'm doing anything wrong [17:11:51 CEST] <leewdch> I just used -i input.swf output.mp4 for test and the result wasn't ok [17:12:06 CEST] <leewdch> it was speeded up from 1 minute down to 4 seconds, audio was fine tho [17:12:31 CEST] <leewdch> this means that it played video really fast for the first 4 seconds and then I had a still frame with the normal audio playing in the background [17:18:21 CEST] <durandal_1707> leewdch: report it on our bug tracker [19:13:52 CEST] <oerg866> ffmpeg just doesn't want me to complete my tasks these days it seems XD [19:14:06 CEST] <oerg866> it looks like -shortest is a bit buggy / inaccurate [19:14:25 CEST] <oerg866> when I look at the length ffprobe gives me and I load the audio in an audio editor there's differences [19:16:34 CEST] <oerg866> ffprobe says for example [19:16:53 CEST] <oerg866> video stream duration=275.310556 / audio stream duration=275.454033 [19:16:58 CEST] <oerg866> on a file generated with -shortest while muxing [19:17:15 CEST] <oerg866> and since I'm using a concat filter afterwards on all files that means the audio desyncs quite badly at some point [19:26:33 CEST] <durandal_1707> oerg866: shortest is not buggy, what are your commands? [19:52:21 CEST] <oerg866> oh whoops I think I did something very stupid [19:52:22 CEST] <oerg866> hang on [19:54:20 CEST] <oerg866> is there a way to force it to consider the -shortest parameter even if the output file has only one stream [19:54:33 CEST] <oerg866> like this: ffmpeg -y -i a.vob -i a.wav -vcodec copy -an -shortest C:\output\tmp.vob [19:57:45 CEST] <ariyasu> why "-i an audio file" then do "-an" to omit audio tracks from output? [20:01:59 CEST] <colekas> hello friends, does anyone know a way of maintaining packet dts when getting a frame? The documentation seems to indicate that if I need a "P" picture to get a "B" picture out of the decoder, the DTS of the "B" frame I receive will be that of the "P" picture rather than the DTS of the "B" picture. [20:04:36 CEST] <klaxa> are you sure [20:04:47 CEST] <klaxa> dts is decoding timestamp [20:05:34 CEST] <klaxa> wait what [20:07:34 CEST] <colekas> int64_t AVFrame::pkt_dts [20:07:35 CEST] <colekas> DTS copied from the AVPacket that triggered returning this frame. [20:07:43 CEST] <colekas> https://ffmpeg.org/doxygen/3.2/structAVFrame.html#a0452833e3ab6ddd7acbf8281… [20:10:01 CEST] <colekas> so if I pass in a B frame to my decoder that has a dts of 10 and a pts of 9, and then a P frame that has a dts of 11 and a pts 12, then the B frame that I get back will have a pts of 9 and a dts of 11, instead of the dts of 10 [22:20:02 CEST] <oerg866> <ariyasu> why "-i an audio file" then do "-an" to omit audio tracks from output? [22:20:25 CEST] <oerg866> Because I want to truncate the video to the audio's size if it's longer [22:20:26 CEST] <oerg866> :S [23:45:39 CEST] <ovi> hello [23:45:52 CEST] <JEEB> ohai [23:46:13 CEST] <ovi> what should i do when i get the following error with ffmpeg [23:46:14 CEST] <ovi> [libvpx @ 0xa4736c0] CQ level 25 must be between minimum and maximum quantizer value (5000-50) [23:46:34 CEST] <JEEB> make the value be between 50 and 5000? [23:46:36 CEST] <JEEB> seemingly [23:47:06 CEST] <ovi> the crf value? [23:47:21 CEST] <JEEB> whatever you set to 25, yes [23:57:14 CEST] <ovi> Value 70.000000 for parameter 'crf' out of range [-1 - 63] [23:57:29 CEST] <JEEB> :D [23:57:53 CEST] <ovi> it is strange.. if I put 60 it says it must be between 50 and 5000 [23:58:11 CEST] <ovi> if I put 70 the error is it must be between -1 and 63 [23:58:15 CEST] <ovi> :( [23:58:59 CEST] <JEEB> the funny fact is that both of those seem to come from FFmpeg's libvpx wrapper [23:59:40 CEST] <JEEB> also weird, the message is supposed to have min first [23:59:42 CEST] <JEEB> and max later [23:59:47 CEST] <JEEB> yet in your paste it's 5000-50 [23:59:51 CEST] <JEEB> that... makes no sense [00:00:00 CEST] --- Tue May 1 2018

1 0