Ffmpeg-devel-irc

Download

ffmpeg-devel-irc@ffmpeg.org

July 2015

1 participants
62 discussions

ffmpeg-devel.log.20150705
by burek 06 Jul '15

06 Jul '15

[01:56:45 CEST] <michaelni> anyone alive ? Compnn maybe ? [01:57:07 CEST] <michaelni> ive just switched samples.ffmpeg.org to the new server if peolpe want to test [01:57:31 CEST] <BtbN> oh, there already is a new server? [02:03:48 CEST] <BtbN> Seems to work and also seems decently fast [02:04:48 CEST] <Compnn> yes? [02:05:30 CEST] <Compnn> michaelni : Tracing route to samples.mplayerhq.hu [192.190.173.55] [02:05:43 CEST] <Compn> samples.mphq resolving to old ip [02:05:52 CEST] <michaelni> ye i just see [02:05:53 CEST] <Compn> theres lots of domains to switch around :P [02:06:13 CEST] <Compn> thanks for working on it michaelni [02:06:35 CEST] <michaelni> samples should be the new IP in a few min [02:08:05 CEST] <michaelni> git.ffmpeg.org switched to new IP too [02:08:41 CEST] <BtbN> shows the ffmpeg homepage now. Or is that what it allways did? [02:09:00 CEST] <michaelni> probably [02:09:29 CEST] <BtbN> Wasn't it a redirect to the videolan git? [02:09:56 CEST] <michaelni> source.ffmpeg.org was and is [02:10:01 CEST] <BtbN> ah [02:14:10 CEST] <Compn> are the git.mplayerhq.hu server up ? [02:14:17 CEST] <Compn> mirror of mplayer svn [02:14:22 CEST] <Compn> not important just curious [02:15:56 CEST] <michaelni> mplayer svn & git are still on the old box someone still needs to look in to them [02:16:16 CEST] <Compn> ask reimar/beastd i guess [02:17:38 CEST] <michaelni> yes, if noone else does i might look too but its very long i touced svn [02:24:33 CEST] <michaelni> www subdomain switched too [02:39:39 CEST] <jamrial> where was everything hosted until now and why the sudden move? [02:39:45 CEST] <jamrial> it's not really explained in the ml thread [02:48:55 CEST] <llogan> management changes in the IT department at the university in Hungary. or something like that. [02:49:23 CEST] <llogan> IIRC, internal hosting is being moved to "cloud" service (I hate that term). [02:50:43 CEST] <jamrial> ah, i see [02:54:21 CEST] <jamrial> pretty crappy it ultimately happened with a one day warning [02:56:02 CEST] <llogan> we had a notification with no known deadline on July 1 [02:56:16 CEST] <llogan> i was hoping we would have ~2 weeks [03:15:31 CEST] <Compn> cloud is so shitty [03:15:42 CEST] <Compn> just constant downtime. [03:18:49 CEST] <rcombs> does that mean it rained? [03:34:19 CEST] <Compn> llogan : tell arpi to check the managment cv , usually people with these "great ideas" are fibbing about their degrees... [03:34:20 CEST] <Compn> ;p [04:39:40 CEST] <philipl> wm4: It's not in the latest qsv patches? ;-) [13:40:38 CEST] <lookatmeyou> help [13:40:43 CEST] <lookatmeyou> how to convert a raw nv21 video to bgr24 uncompressed avi? [13:40:54 CEST] <lookatmeyou> ffmpeg -s 480x640 -pix_fmt nv21 -i video.nv21 -vcodec rawvideo -pix_fmt bgr24 video.avi [13:41:02 CEST] <lookatmeyou> I use this command, but it prints "video.nv21: Invalid data found when processing input" [13:41:09 CEST] <lookatmeyou> Is there anything wrong? [13:44:09 CEST] <wm4> #ffmpeg [13:45:19 CEST] <lookatmeyou> Thanks [13:54:08 CEST] <BtbN> philipl, lossless behaves fine on Windows/MSVC on my 760 [13:54:43 CEST] <BtbN> Some error messages might need a higher level, so it doesn't fail without any indication what went wrong. [13:54:54 CEST] <BtbN> Testing on my linux box now [14:01:08 CEST] <cone-828> ffmpeg 03Philip Langdale 07master:671bdd4b0977: avcodec/nvenc: Add support for H.264 High 444 Predictive encoding [14:01:32 CEST] <BtbN> philipl, works as expected. [14:02:18 CEST] <wm4> philipl: what do you mean by that? [14:02:27 CEST] <wm4> <philipl> wm4: It's not in the latest qsv patches? ;-) [14:04:20 CEST] <BtbN> And i finaly fixed my cygwin msvc build env. [14:04:30 CEST] <BtbN> Needed more quotes [14:10:13 CEST] <nevcairiel> cygwin is just a terrible way to do that :-( [14:11:40 CEST] <BtbN> Best way i found [14:11:46 CEST] <BtbN> better than a windows command prompt [14:13:12 CEST] <wm4> is powershell useable these days? [14:13:36 CEST] <BtbN> If you invest enough time to learn all the strange commands it has [14:13:44 CEST] <BtbN> I prefer a unix-like shell [14:14:32 CEST] <nevcairiel> yeah if you know how to then powershell is fine [16:52:12 CEST] <cone-828> ffmpeg 03Rostislav Pehlivanov 07master:e06578e392e3: aacenc: use the new function for setting special band scalefactor indices [16:52:13 CEST] <cone-828> ffmpeg 03Rostislav Pehlivanov 07master:117b15f4a8d9: aaccoder: remove previous PNS implementation from twoloop [16:52:14 CEST] <cone-828> ffmpeg 03Rostislav Pehlivanov 07master:38fd4c2e664a: aaccoder: add a new perceptual noise substitution implementation [17:11:30 CEST] <cone-828> ffmpeg 03Rostislav Pehlivanov 07master:0b233900fa36: aacenc: add support for coding of IS spectral coefficients [17:11:31 CEST] <cone-828> ffmpeg 03Rostislav Pehlivanov 07master:e8576dc8dfc9: aacenc: implement Intensity Stereo encoding support [17:41:57 CEST] <philipl> wm4: It was a joke about the fact that qsv presumably wraps the hevc decoder [17:42:04 CEST] <philipl> and we're getting lots of random qsv patches [17:42:09 CEST] <wm4> ok [17:42:17 CEST] <wm4> I wonder why qsv is so popular [17:42:30 CEST] <philipl> because it's the only option on windwos for encoding? [17:42:49 CEST] <wm4> yeah, but the latest patches were for linux too? [17:43:02 CEST] <BtbN> Yeah, because it works there, too [17:43:07 CEST] <BtbN> And there is no libva encoder [17:43:28 CEST] <BtbN> So of course people want to use it, even though it's just another stupid wrapper [17:44:50 CEST] <philipl> Well, none of us is volunteering to write the native encoder, for understandable reasons. So either its nothing or a wrapper. [17:45:20 CEST] <BtbN> I already did write a libva based encoder, for vlc though. And i couldn't figure out how to do B-Frames, and rate-control. So i gave up. [17:45:28 CEST] <philipl> and qsv is a better choice of wrapper than yami. [17:45:45 CEST] <wm4> what rules libyami out for encoding? [17:46:03 CEST] <philipl> We scared the intel guy away :-) [17:46:03 CEST] <BtbN> It's just a wrapper around libva [17:46:20 CEST] <BtbN> with questionable future [17:46:36 CEST] <BtbN> specialy as QSV exists for free on linux now. [17:46:38 CEST] <wm4> why not just use gstreamer [17:46:38 CEST] <philipl> and given that qsv is already a windows dependency for a good reason, it makes much more sense to enable it on linux than use a completely different wrapper [17:47:13 CEST] <BtbN> you mean a generic gstreamer wrapper in lavc? ;) [17:47:26 CEST] <philipl> lavc -> gstreamer -> lavc sounds pretty awesome to me. [17:47:28 CEST] <wm4> something like this *shudder* [17:48:05 CEST] <philipl> Does qsv expose the vp8 encode/decode? [17:48:31 CEST] <BtbN> The primary problem with QSV is, that it most likely will never be packaged by distributions [17:48:49 CEST] <wm4> why not? legal shit? [17:49:06 CEST] <BtbN> Basicaly, yes [17:50:24 CEST] <BtbN> no idea if it's still not possible to download the SDK without an intel account, but last time i checked, you still needed one [19:05:28 CEST] <klaxa> BtbN: i downloaded it yesterday, you still need to create an account [19:44:12 CEST] <ubitux> so the aac patch is merged? :o [19:45:21 CEST] <nevcairiel> no, different patch [19:45:31 CEST] <ubitux> oh, ok [19:45:35 CEST] <BtbN> there were a lot of aac patches recently [19:45:58 CEST] <ubitux> 'made some (slow) progress on the selective color for those interested [19:47:08 CEST] <ubitux> remaining is black param, relative adjustement, and the usual round trip of fate/doc/test [19:47:25 CEST] <ubitux> then it should be good for a first draft [19:47:37 CEST] <BtbN> Would be nice if the ffmpeg encoder could get a bit closer to libfdk [19:50:53 CEST] <nevcairiel> psy optimizations are closely kept secrets for aac apparently [19:52:30 CEST] <BtbN> You can keep those secret even if the source of the encoder is 100% open? [19:52:46 CEST] <wm4> open binary! [19:56:41 CEST] <wm4> ( http://www.vapoursynth.com/2012/10/open-binary-introducing-a-practical-alte… ) [19:57:11 CEST] <nevcairiel> BtbN: you can read the source, understand the algorithms, and re-implement it in fresh code, but of course you cant just take the code =) [20:12:16 CEST] <cone-828> ffmpeg 03Michael Niedermayer 07master:47496eb97cff: avcodec/rawenc: Use ff_alloc_packet() instead of ff_alloc_packet2() [20:31:06 CEST] <rcombs> where do you even get the linux qsv lib [20:31:26 CEST] <rcombs> and can you actually redistribute binaries linking to it without angering the GPL lords? [20:34:23 CEST] <durandal_1707> is there specification that dictates various volume metering? [20:41:05 CEST] <ubitux> durandal_1707: ebu r128? [20:41:48 CEST] <durandal_1707> there are others too [20:42:37 CEST] <durandal_1707> I want to add them to avf_showvolume [20:42:55 CEST] <J_Darnley> replaygain, maybe? [20:43:37 CEST] <durandal_1707> no, that is something else [21:43:10 CEST] <wm4> always nice to see when FFmpeg adds new public API without any documentation: dc92464aee01a78c9963f86f24600ddf9c825258 [21:43:39 CEST] <wm4> you can tell just by looking at the code, and running git blame always confirms it [21:43:56 CEST] <wm4> Libav still has a good reason to exist after all [21:45:15 CEST] <kierank> rcombs: feel free to pay for medialooks if you don't like the gpl lords [21:45:27 CEST] <kierank> I'm sure those binary blobs will work well [21:45:39 CEST] <kierank> or you could pay $1000 for their ffmpeg wrapper [21:46:00 CEST] <rcombs> kierank: quality [21:47:40 CEST] <wm4> what $1000 ffmpeg wrapper? [21:48:25 CEST] <kierank> wm4: http://blog.medialooks.com/0LyGa6/ [21:48:54 CEST] <wm4> ah, that one [21:49:03 CEST] <kierank> aka a lot of broadcast software [21:49:07 CEST] <kierank> ffmpeg with a flashy gui [22:55:49 CEST] <philipl> I just realised cuda has vdpau interop. So it's theoretically possible to do vdpau -> nvenc without readback. [22:55:56 CEST] <philipl> I wonder how much you gain by doing it. [23:00:22 CEST] <nevcairiel> less cpu usage, probably no speed up [23:00:27 CEST] <nevcairiel> nvidia is quite efficient with that [23:05:54 CEST] <Compn> isnt that what the whole quicksync intel qsv is supposed to do ? [23:06:04 CEST] <Compn> once its up and committed... [23:08:33 CEST] <cone-828> ffmpeg 03Michael Niedermayer 07master:0e7bbdbeaf7f: avcodec/avuienc: Initialize output data [23:08:34 CEST] <cone-828> ffmpeg 03Michael Niedermayer 07master:c0db6320a639: avcodec/avuienc: Use ff_alloc_packet() [23:08:59 CEST] <cone-828> ffmpeg 03Michael Niedermayer 07release/2.7:a7fa1c9b2b18: ffmpeg_opt: Fix forcing fourccs [23:24:07 CEST] <philipl> I went and built the latest nvenc sdk samples and we're way slower than the example encoder. [23:27:31 CEST] <philipl> I can't gte the encode frame rates that nvidia document (1111fps) but I can get 854 from the example and we're ~390 [23:31:02 CEST] <Compn> wasnt there an nvenc from nvidia ? and someone rewrote their own implementation ? [23:53:08 CEST] <philipl> Compn: I mean, the example app in the nvenc sdk is much faster than using nvenc through ffmpeg [23:53:23 CEST] <philipl> BtbN and I have discussed this before. There's no obvious reason. [00:00:00 CEST] --- Mon Jul 6 2015

1 0

ffmpeg.log.20150705
by burek 06 Jul '15

06 Jul '15

[00:05:13 CEST] <Seylerius> So, I'm trying to figure out how to produce any of these formats with ffmpeg, so I can easily import some things into Lightworks: http://lightworks.wikidot.com/formats [00:05:14 CEST] <polyphagia> why not smo_ [00:05:26 CEST] <Seylerius> I've got these: http://sprunge.us/LciO [00:05:57 CEST] <Seylerius> avi with mpeg2video is registering as audio-only to lightworks. [00:06:17 CEST] <Seylerius> Perhaps this is because some additional option would be necessary to make it mpeg2 with i-frame [00:06:25 CEST] <Seylerius> I'd appreciate any suggestions. [00:06:55 CEST] <Seylerius> Also, for the sake of limited processing power on the machine running lightworks, I'd prefer to get out of h.264. [00:07:08 CEST] <user> hello :) [00:07:13 CEST] <user> anybody here? :) [00:07:44 CEST] <Guest86769> ^^' [00:07:57 CEST] <Guest86769> can i ask just a small question ? [00:08:35 CEST] <klaxa> don't ask to ask, just ask [00:08:39 CEST] <Guest86769> xD [00:09:00 CEST] <Guest86769> well i successfully compiled following this guide ffmpeg on centos https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [00:09:04 CEST] <Guest86769> i followed all the guide [00:09:23 CEST] <Guest86769> there's any chance i can produce also a windows x64 executable ? [00:10:19 CEST] <emk-> Guest86769: you'd need cross-compilation toolchain installed. [00:10:42 CEST] <Guest86769> emk-: do you have a guide i can follow? [00:10:46 CEST] <klaxa> https://trac.ffmpeg.org/wiki/CompilationGuide/CrossCompilingForWindows [00:11:08 CEST] <Guest86769> klaxa: thank you ;) [00:11:12 CEST] <klaxa> it might be easier to compile it on a windows machine since i'm pretty sure that if you compiled all the libraries for linux you can't build a windows executable from them [00:11:13 CEST] <Guest86769> i will try [00:11:50 CEST] <Guest86769> klaxa: so if i use that guide it will not work? [00:11:56 CEST] <Guest86769> i'm using centos as alternative [00:12:08 CEST] <Guest86769> because i'm having some problems with mingw [00:12:51 CEST] <klaxa> i don't know if it will fail [00:13:02 CEST] <klaxa> you can give it a try though [00:13:27 CEST] <Guest86769> i will try because at least till now on centos nothing failed :D [00:21:27 CEST] <Compress0r> Hi guys. I have an issue with converting my Canon EOS 700D h264 movies anything else. I always get washed out colors in the result. [00:22:17 CEST] <Compress0r> I first thought its an issue with libx264 output, but even using huffyuv I see the same washed out colors. Specifing color_range, pix_fmt didn't help either. [00:22:51 CEST] <emk-> Compress0r: can you give ffprobe output of one of the movies? [00:23:06 CEST] <Compress0r> VLC is playing the video in shiny colors, but anything that went thought ffmpeg just looks washed out. [00:23:10 CEST] <Seylerius> I'm trying to put 1080x1920 into dvvideo, but it's complaining about lacking a DV profile. Which one should I use (to preserve quality), and how do I specify it? [00:23:12 CEST] <Compress0r> sure give me a sec [00:23:39 CEST] <Seylerius> Current iteration of the command is `20150628_101238_01.mp4 -codec:v dvvideo -codec:a pcm_s16le 20150628_101238_01.avi` [00:23:50 CEST] <Compress0r> Relevant part should be thing, right? Video: h264 (Constrained Baseline) (avc1 / 0x31637661), yuv420p(pc, bt709/bt470bg/bt709), 1920x1080, 44656 kb/s, 25 fps, 25 tbr, 25k tbn, 50k tbc (default) [00:24:03 CEST] <Compress0r> yuv420p(pc, bt709/bt470bg/bt709) [00:25:43 CEST] <Compress0r> sorry...actual output is this: Video: h264 (Constrained Baseline) (avc1 / 0x31637661), yuvj420p(pc, bt709), 1920x1080, 44656 kb/s, 25 fps, 25 tbr, 25k tbn, 50k tbc (default) [00:26:05 CEST] <Compress0r> i messed around in the coding already to force full range and different color profiles. [00:26:14 CEST] <emk-> Compress0r: shenanigans with bt709 color space :( [00:26:19 CEST] <Compress0r> nothing helped [00:26:57 CEST] <Compress0r> yes I also think that this has to do with bt709 color space [00:27:05 CEST] <emk-> did you try -x264opts colorprim=bt709:transfer=bt709:colormatrix=bt709:fullrange=off ? [00:27:32 CEST] <Compress0r> yes i think i found this on some mailing list already...let me try again [00:28:02 CEST] <Compress0r> funny thing is its happening with huffyuv output...so i guess x264opts are just for output right? [00:28:29 CEST] <emk-> yep [00:28:48 CEST] <Compress0r> with huffyuv ALSO, so I have tried anything like converting it to rgb24... [00:28:59 CEST] <Compress0r> i really think that the input gets messed up really early [00:29:02 CEST] <emk-> probably huffyuv also does colorspace conversions... [00:29:27 CEST] <emk-> or maybe the video is not really bt709, and hence weird results [00:29:52 CEST] <emk-> colorspace conversions are performed when they shouldn't [00:30:25 CEST] <Compress0r> which AVColorSpace i could try to force? [00:30:31 CEST] <Compress0r> AVCOL_SPC_RGB? [00:31:05 CEST] <Compress0r> i just force it libavcodec/h264_slice.c [00:31:52 CEST] <Compress0r> but rgb doesn't sound right, and I can't find anything in AVColorSpace that sounds like "don't mess with my colors" [00:31:57 CEST] <emk-> Compress0r: did you mess with all the combinations of https://ffmpeg.org/ffmpeg-filters.html#colormatrix on input? [00:32:20 CEST] <Compress0r> colormatrix is a -vf? [00:32:44 CEST] <emk-> yes [00:33:01 CEST] <Compress0r> I just feared that the video gets messed up before some filters are applied... [00:33:22 CEST] <Compress0r> can a filter interfer with how the video is interpreted? [00:33:39 CEST] <emk-> And how does the video look in ffplay? [00:37:37 CEST] <Compress0r> also "washed out" [00:38:19 CEST] <Compress0r> but actually I'm not sure if ffmpeg is correct and vlc is just "missing" the bt709 conversion and showing a way to oversaturated image... [00:39:24 CEST] <Compress0r> maybe vlc is intepreting the video as MPEG range and ffmpeg actually does correctly see it as JPEG range? [00:39:55 CEST] <Compress0r> I thought vlc is also using libav* [00:40:34 CEST] <emk-> you need some kind of test board ;) [00:40:54 CEST] <klaxa> maybe you have some filters enabled in vlc? [00:44:23 CEST] <Compress0r> nope...double checked that :) [00:44:51 CEST] <Compress0r> also i play both, the original and the ffmpeg edited video in vlc. so the same filter should apply then ;) [00:44:57 CEST] <klaxa> hmm true [00:45:10 CEST] <klaxa> what does ffprobe say about your encoded file? [00:45:21 CEST] <klaxa> (especially regarding the pixel format and color space) [00:45:54 CEST] <Compress0r> Video: h264 (Constrained Baseline) (avc1 / 0x31637661), yuvj420p(pc, bt709), 1920x1080, 44656 kb/s, 25 fps, 25 tbr, 25k tbn, 50k tbc (default) [00:47:01 CEST] <klaxa> that's identical to your input file [00:47:10 CEST] <klaxa> are you sure it's the right paste? [00:50:35 CEST] <Compress0r> ah the encoded output...sorry [00:54:22 CEST] <Compress0r> give me a sec. I think I'm onto smth [01:04:22 CEST] <Compress0r> ouh...now samba was playing tricks on me and giving me old encodes over and over cause there have been open handles on the old file -.- [01:04:59 CEST] <Compress0r> note to myself: never reuse filenames when trying something like this :) [01:08:40 CEST] <Compress0r> actually I more and more think that VLC is wrong with its output! [01:17:00 CEST] <Compress0r> wow...color management sucks... [01:17:05 CEST] <Compress0r> how hard can it be... [01:17:38 CEST] <Compress0r> so media player classic looks like encoded ffmpeg output, but the whites are slightly more blue... [01:17:52 CEST] <Compress0r> PowerDVD is as dark as VLC [01:18:40 CEST] <spl33n_> hey all, i convert video to gif in hight quality but video size is 4.6Mo and gif file 59Mo, how can i reduce this size without modify too much quality ? [01:18:41 CEST] <Compress0r> but I'm not sure if the blue could be also related to color profiles of my monitor in windows [01:19:04 CEST] <polyphagia> spl33n_: GIFs aren't for video [01:19:07 CEST] <emk-> Compress0r: maybe some apps take your monitor Color Management settings into account if you have some... [01:19:37 CEST] <emk-> nvm, yoe've written it above ;) [01:24:16 CEST] <Compress0r> omg...the AMD driver also does its magic on the video output... [01:26:14 CEST] <spl33n_> polyphagia: i don't understand [01:26:47 CEST] <BtbN> reduce the resolution, make it shorter. That's it. [01:27:00 CEST] <BtbN> Or reduce the framerate. [01:27:22 CEST] <Compress0r> okay...media player classic was using the amd video driver output optimizations, vlc don't care about it... [01:27:41 CEST] <Compress0r> this means media player classic looks exactly like ffplay output now [01:27:55 CEST] <BtbN> Compress0r, sounds like you are fighting with limited vs. full range HDMI? [01:28:29 CEST] <Compress0r> PowerDVD and VLC looks more saturated, but darker and whites are overexposed... [01:28:53 CEST] <BtbN> yes, that's what full range RGB on a screen expecting limited looks like. [01:29:02 CEST] <Compress0r> yes, looks like an limited/full range issue...I'm still not sure what my input is, limited or full range :) [01:29:17 CEST] <BtbN> Videos don't care about that. [01:29:29 CEST] <BtbN> It's the job of your player and graphics driver to figure that out [01:30:28 CEST] <Compress0r> so strange thing is: original video from my canon eos played in VLC is interpreted in limited range, but after conversion in ffmpeg as full range [01:31:08 CEST] <Compress0r> actually I guess the video coming from the cam isn't correctly interpreted by VLC and ffmpeg fixes this [01:31:31 CEST] <BtbN> videos are usualy YUV, they don't realy care about RGB color ranges [01:31:37 CEST] <Compress0r> thats why I think I've "lost some colors" [01:36:00 CEST] <Compress0r> but still I think there is a bug in ffmpeg. It's a bit hard to discribe. Just encoding with -i INPUT.MOV -c:v libx264 gives this warnings: No pixel format specified, yuvj420p for H.264 encoding chosen. [01:36:01 CEST] <Compress0r> Use -pix_fmt yuv420p for compatibility with outdated media players. [01:36:28 CEST] <Compress0r> and output is still dark [01:36:47 CEST] <Compress0r> actually specifing -pix_fmt yuv420p makes the output brigher [01:36:59 CEST] <Compress0r> -color_range 1 or 2 doesn't change anything [01:37:17 CEST] <Compress0r> that's why I'm confused [01:37:41 CEST] <Compress0r> yuv420p with color_range 1 should be equal to yuvj420p [01:44:44 CEST] <durandal_1707> using yuvj pixfmt output should not change anything [01:45:26 CEST] <durandal_1707> ffplay is not perfect in displaying colors use mpv [02:00:09 CEST] <Guest86769> :) [02:00:21 CEST] <Guest86769> Compress0r: hello [02:01:25 CEST] <Compress0r> hello :) [02:01:36 CEST] <Guest86769> Compress0r: i used the following command for some conversion from screen recorders to h264 -> -c:v libx264 -pix_fmt yuvj420p -crf 24 -profile:v high444 -level 4.2 -preset fast [02:02:04 CEST] <Guest86769> Compress0r: idk if it can help :) [02:02:08 CEST] <Compress0r> I think I fell into an ugly VLC bug! [02:02:13 CEST] <Guest86769> Compress0r: however yeah [02:02:21 CEST] <Guest86769> Compress0r: is actually an nvidia cards bug [02:02:28 CEST] <Guest86769> Compress0r: do you have a nvidia card ? [02:03:04 CEST] <Guest86769> Compress0r: in this case just try disabling the hardware acceleration pixel format on VLC u will see the colors correctly ;) [02:03:29 CEST] <Compress0r> If I select Direct2D or Windows GDI as Output Module I see "Full Range" output, if I select Direct3D or OpenGL i see limited range output for the original video [02:03:36 CEST] <Guest86769> Compress0r: or u can try the trick i said u above to encode with full range yuv but is not a good way of work imho [02:04:33 CEST] <Guest86769> Compress0r: yeah it's a mess between VLC / video card drivers and os xDD [02:05:25 CEST] <Compress0r> I just thought its ffmpeg that is messing with my video, but this is a new level of messing around with my videos :) [02:06:23 CEST] <Guest86769> Compress0r: i think the problems is not of ffmpeg [02:06:31 CEST] <Guest86769> problem* [02:06:46 CEST] <Compress0r> no its not...actually ffmpeg solves the problem :) [02:07:23 CEST] <Compress0r> the videos coming from my Canon EOS look different for the various output modules in VLC, the encoded video looks the same on all modules [02:09:42 CEST] <Guest86769> lol i see [02:10:29 CEST] <Guest86769> i'm trying to build the win64 version of ffmpeg with all the libs on [02:10:32 CEST] <Compress0r> thats what I mean with new level of messing up :) [02:10:46 CEST] <Guest86769> cuz of this lib aac fdk :p [02:11:11 CEST] <Guest86769> learning new things for now ^^ [02:11:56 CEST] <Guest86769> well i go sleep [02:12:00 CEST] <Guest86769> is europe tz here [02:12:02 CEST] <Guest86769> yawn [02:12:04 CEST] <Guest86769> night [02:12:37 CEST] <Guest86769> btw [02:12:45 CEST] <Guest86769> why need of recompress eos videos? [02:12:57 CEST] <Guest86769> they are h264 already ^^' to fix the pixel format ? [02:12:57 CEST] <Compress0r> its too hot too sleep anyways :D [02:13:10 CEST] <Guest86769> Compress0r: oh well i see u have my same problems xD [02:13:16 CEST] <Guest86769> Compress0r: where u from me italy :D [02:13:25 CEST] <Compress0r> germany :) [02:13:42 CEST] <Guest86769> Compress0r: xD all europe is kinda 37°C [02:13:50 CEST] <Guest86769> >.< [02:14:12 CEST] <Compress0r> Mh...but its late. I try to get some sleep :) [02:14:14 CEST] <Compress0r> See you [02:14:17 CEST] <Guest86769> same here [02:14:19 CEST] <Guest86769> byez [06:26:15 CEST] <benschwarz> Hi there, I'm using x11grab + ffmpeg to record my screen, but I can't get the videos to play under quicktime. [06:26:36 CEST] <benschwarz> the command that I'm using is `ffmpeg -y -r 30 -s 1280x1024 -f x11grab -draw_mouse 0 -i :0 -g 30 -preset ultrafast -movflags +faststart -crf 0 -vcodec libx264 -pix_fmt yuv420p video.mp4` [06:27:12 CEST] <benschwarz> I've read a whole bunch of docs about quicktime, but I still can't seem to get the command right. any pointers? [06:31:29 CEST] <klaxa> i'm not sure crf 0 is compatible with -pix_fmt yuv420p [06:31:50 CEST] <klaxa> check ffmpeg's output to see if pix_fmt was changed [06:31:59 CEST] <klaxa> benschwarz: ^ [06:32:20 CEST] <benschwarz> klaxa: hmm, I'll check [06:34:00 CEST] <relaxed> benschwarz: you want -framerate 30 instead of -r 30 [06:34:11 CEST] <benschwarz> relaxed: oh yeah? how come? [06:34:41 CEST] <relaxed> because it's the correct option, see ffmpeg -h demuxer=x11grab [06:34:57 CEST] <klaxa> ah right, i ran into that earlier today again too [06:35:06 CEST] <benschwarz> relaxed: oh right ! [06:35:40 CEST] <relaxed> and maybe quicktime can't decode lossless? try -crf 16 [06:36:31 CEST] <benschwarz> anything else look wrong? [06:36:45 CEST] <benschwarz> & is there a way to get more info from quicktime as to why the video won't open? [06:37:32 CEST] <relaxed> looks fine otherwise [06:37:39 CEST] <benschwarz> hey! It works now [06:39:54 CEST] <benschwarz> Thanks relaxed && klaxa [07:22:54 CEST] <Piedro> Hi, guys! [07:23:47 CEST] <Piedro> Got some troubles with MPD seeing my .dts files. [07:24:15 CEST] <Piedro> He sees one, but not another. [07:24:51 CEST] <Piedro> FFMPEG output for .dts file visible in MPD says: [07:25:10 CEST] <Piedro> Input #0, wav, from 'file1.dts' [07:25:18 CEST] <Piedro> Stream #0:0: Audio: dts (DTS) ([1][0][0][0] / 0x0001), 44100 Hz, 5.1(side), fltp, 1411 kb/s [07:25:39 CEST] <Piedro> For file not visible in MPD: [07:25:51 CEST] <Piedro> Input #0, dts, from 'file2.dts' [07:25:59 CEST] <Piedro> Stream #0:0: Audio: dts (DTS), 44100 Hz, 5.1(side), fltp, 1411 kb/s [07:27:29 CEST] <Piedro> The question is what is the meaning of wav and dts in Input #0 line and how could I change it with FFMPEG? [07:53:04 CEST] <Piedro> Still looking for an answer :3 [09:49:15 CEST] <Piedro> Anyone alive? [09:54:17 CEST] <Fjorgynn> no [10:19:55 CEST] <Piedro> :-( [10:20:06 CEST] <Piedro> Anyone undead to give me some help? [10:26:32 CEST] <durandal_1707> mpd bug? [10:31:15 CEST] <nyuszika7h> why does reencoding a video (-c:v libx265 -c:a copy -c:s copy) change the subtitle font? (ASS) [10:31:49 CEST] <nyuszika7h> other options (probably irrelevant): -preset veryfast -crf 23 -t 00:00:10 [10:31:55 CEST] <Nitori> possibly the fonts where embedded and not copied over? [10:37:52 CEST] <Nitori> @nyuszika7h [10:37:58 CEST] <nyuszika7h> maybe [10:38:31 CEST] <Nitori> have you checked? (ffprobe and see if there are fonts embedded in source and output) [10:44:31 CEST] <nyuszika7h> sec [10:45:09 CEST] <nyuszika7h> input: http://dpaste.com/0JGTPS8 [10:46:34 CEST] <nyuszika7h> output doesn't have the font [10:46:39 CEST] <Nitori> yea [10:46:44 CEST] <Nitori> so, according to this: https://trac.ffmpeg.org/wiki/How%20to%20use%20-map%20option#Example4 [10:46:57 CEST] <Nitori> you should be able to just do: -map 0 -c copy -c:v libx265 .... [10:47:04 CEST] <Nitori> to copy everything but reencode the video [10:52:09 CEST] <nyuszika7h> yeah that works, thanks [11:53:04 CEST] <nyuszika7h> is there any way to make ffmpeg x264 encode not lag my system so much? I set ffmpeg.exe to low priority in task manager but that didn't help [11:53:57 CEST] <durandal_1707> use faster preset? [13:01:02 CEST] <user> : 3 [13:01:25 CEST] <Guest14853> hello [13:02:10 CEST] <Guest14853> i was wondering there's a way to get the size of single streams via ffprobe? [13:02:34 CEST] <Guest14853> because the only size it shows is the size of whole file, is something i can actually know already with my os XD [13:02:51 CEST] <Guest14853> instead i want to know how much the audio part for example weights [13:03:38 CEST] <Guest14853> i could -map 0:1 stream -a:copy -> and check the size of new file but i think is something stupid to do :D [13:37:11 CEST] <lookatmeyou> help [13:37:41 CEST] <lookatmeyou> how to convert a raw nv21 video to bgr24 uncompressed avi? [13:37:52 CEST] <lookatmeyou> ffmpeg -s 480x640 -pix_fmt nv21 -i video.nv21 -vcodec rawvideo -pix_fmt bgr24 video.avi [13:38:25 CEST] <lookatmeyou> I use this command, but it prints "video.nv21: Invalid data found when processing input" [13:38:47 CEST] <lookatmeyou> Is there anything wrong? [13:40:26 CEST] <lookatmeyou> Hello? [13:42:52 CEST] <Filarius> try to specify input codec too, not only pixel format [13:43:19 CEST] <Guest14853> Filarius: hello [13:44:38 CEST] <Filarius> add -f rawvideo [13:45:10 CEST] <lookatmeyou> OK, let me try [13:46:42 CEST] <Filarius> hello Guest, and I`m too noob to give direct answer. ffprobe can make json data output or else formats, try from this direction [13:46:42 CEST] <lookatmeyou> It worked. Thank you very much! [13:47:57 CEST] <Filarius> np [14:26:00 CEST] <Filarius> Do anybody know nice app for Windows where I can look at graphs of rate factor, quality and bitrate, what changes during file ? [15:01:59 CEST] <Guest14853> Filarius: hello [15:02:11 CEST] <Guest14853> Filarius: maybe u have a solution to my question above? [15:02:23 CEST] <Filarius> no [15:02:24 CEST] <Guest14853> Filarius: i'm trying with ffprobe to get the size of single streams [15:02:37 CEST] <Filarius> try to look at ffprobe json output [15:02:45 CEST] <Guest14853> hmm [15:02:47 CEST] <Guest14853> ok [15:04:00 CEST] <copulator> yow [15:04:23 CEST] <copulator> how do i mux x264 and mp3 into mp4? [15:04:41 CEST] <Guest14853> nope [15:04:41 CEST] <copulator> i use -acodec:copy and -vcodec:copy but it still re-encodes shit [15:04:45 CEST] <Guest14853> it doesn't work [15:05:43 CEST] <Guest14853> copulator: try using also -map 0:0 -map 0:1 [15:06:11 CEST] <lookatmeyou> Help, I use this command "ffmpeg -s 352x288 -f rawvideo -pix_fmt nv21 -r 15 -i Qq.nv21 -vcodec rawvideo -vframes 50 -pix_fmt bgr24 Qq4.avi" [15:06:29 CEST] <Guest14853> and? [15:06:44 CEST] <lookatmeyou> to convert a nv21 rawvideo to bgr24 avi uncompressed rawvideo [15:07:21 CEST] <lookatmeyou> it works fine, except add metadata to the output file which I don't want [15:07:51 CEST] <lookatmeyou> like metadata: encoder : Lavf56.13.100 [15:08:39 CEST] <lookatmeyou> matlab aviread can't read this kind of video with metadata. [15:09:07 CEST] <Guest14853> lookatmeyou: i think you could follow this: [15:09:13 CEST] <Guest14853> lookatmeyou: http://askubuntu.com/questions/139570/how-do-i-remove-all-metadata-from-a-q… [15:09:34 CEST] <lookatmeyou> Let me see [15:12:01 CEST] <lookatmeyou> like this command? "ffmpeg -i Qq3.avi -map_metadata -1 -vcodec copy Qq5.avi" [15:12:30 CEST] <Guest14853> i guess simply -map_metadata -1 does the trick [15:12:30 CEST] <lookatmeyou> No, the output Qq5.avi still has meta_data [15:12:44 CEST] <Guest14853> lookatmeyou: try just do it in sequence [15:12:49 CEST] <Guest14853> first encode the stuff [15:13:09 CEST] <Guest14853> then another passage into ffmpeg with just video and audio copy and the metadata mapping out [15:15:04 CEST] <Guest14853> i read the stuff inside that link and it seems u have to do some extra passages to wipe out all metadata [15:16:11 CEST] <lookatmeyou> A bit of trouble, Is there a simple way? [15:16:59 CEST] <Guest14853> i think not as always with metadata stuff [15:17:09 CEST] <Guest14853> same for me with exiftool i had to prepare a bat myself [15:17:32 CEST] <Guest14853> in order to copy a blank xml file preset to the images in order to wipe all out definitively [15:17:40 CEST] <Guest14853> prepare a nice bat file and u will live happy [15:17:51 CEST] <Guest14853> all the files u convert them are post -> nuked metadata with all the procedure [15:19:57 CEST] <lookatmeyou> Thanks, I still want to find a simple way. [15:27:12 CEST] <koz_> What should I add to my conversion command if I get this: Video uses a non-standard and wasteful way to store B-frames ('packed B-frames'). Consider using the mpeg4_unpack_bframes bitstream filter to fix it. [15:28:13 CEST] <Guest14853> koz_: i suppose it suggests a filter [15:28:22 CEST] <Guest14853> like [15:28:33 CEST] <Guest14853> -filter:v "mpeg4_unpack_bframes" [15:28:41 CEST] <Guest14853> try in case [15:28:45 CEST] <Guest14853> and see if it works [15:29:00 CEST] <koz_> Guest14853: OK. What will this supposedly do? [15:29:20 CEST] <Guest14853> koz_: https://www.ffmpeg.org/ffmpeg-bitstream-filters.html there [15:29:56 CEST] <koz_> Ah. So 'fix Losedows shite', got it. [15:30:26 CEST] <Guest14853> but need to be used as that one say [15:30:29 CEST] <Guest14853> ffmpeg -i INPUT.avi -codec copy -bsf:v mpeg4_unpack_bframes OUTPUT.avi [15:32:15 CEST] <koz_> Thanks Guest14853! [15:32:34 CEST] <Guest14853> koz_: u welcome [16:19:01 CEST] <chaz321> Hello. I have a filter_complex question. I am using ffmpeg -drawtext with timecode. The text is displayed as: 09:25:44:12 Is it possible to remove the frame field from the display? We would prefer if the text was displayed as: 09:25:44 [16:32:55 CEST] <maslen> How can I rotate a video 90 degrees? If I try using vf="rotate=PI/2", it cuts off part of the video. [16:33:19 CEST] <maslen> (chops off their heads,, so to speak) [16:42:19 CEST] <klaxa> maslen: -vf transpose [16:42:34 CEST] <klaxa> see: https://ffmpeg.org/ffmpeg-filters.html#transpose [17:01:37 CEST] <maslen> Strange. So the issue I'm having is that even though VLC plays it properly (portrait), WMP plays it in landscape, which is 90 degrees off :( [17:02:40 CEST] <maslen> Is there some metadata that I'm not aware of? Or is WMP just doing its own thing. [17:20:18 CEST] <klaxa> it's probably wmp [17:20:28 CEST] <maslen> So nothing I can do then. Oh well. [18:22:55 CEST] <nashgul> hi all, how can i see the parameters of each preset for h264 codec? [18:23:02 CEST] <nashgul> my OS is Archlinux [18:23:30 CEST] <nashgul> i'm looking on the web (ffmpeg.org, the wiki, ...) but i don't find anything [18:24:05 CEST] <nashgul> h264 presets: ultrafast, veryfast, medium, faster, slow, .... [18:24:22 CEST] <JEEBsv> those are built-in things in libx264 [18:25:06 CEST] <nashgul> yes, but on the wiki appears this: To list all possible internal preset and tunes: [18:25:07 CEST] <nashgul> ffmpeg -f lavfi -i nullsrc -c:v libx264 -preset help -f mp4 - [18:25:27 CEST] <nashgul> i execute that command but only returns an error [18:26:39 CEST] <JEEBsv> yes, it will [18:26:52 CEST] <JEEBsv> and that is a hack [18:27:04 CEST] <JEEBsv> it will list you all the presets and tunes if you set an invalid value [18:27:10 CEST] <JEEBsv> "help" is an invalid value [18:27:20 CEST] <nashgul> ummm [18:27:26 CEST] <JEEBsv> so libx264 will output them, as you can see with the lines starting with [libx264] [18:28:13 CEST] <nashgul> JEEBsv: what is the command to see the presets? [18:28:53 CEST] <JEEBsv> that thing you just noted happens to work, just see the lines starting with [libx264 [18:31:01 CEST] <nashgul> JEEBsv: i see those lines, but what i want to know is: what are the parameters stablished for each preset? [18:32:39 CEST] <JEEBsv> try x264 --fullhelp &> x264_settings.txt [18:33:03 CEST] <JEEBsv> not sure how much use you will get out of the details of the presets, though [18:33:09 CEST] <JEEBsv> they were designed relatively well [18:36:50 CEST] <nashgul> JEEBsv: thanks, x264 shows me, more or less, what i want [18:59:28 CEST] <lvml> Hi, I would like to instruct ffmpeg to use "libdcadec" as a decoder (rather than "dca") for decoding all DTS content. However, I found no other way to do this than by modifying the source code (registering "dca" as only an encoder in allcodecs.c). Isn't there any way to configure a decoder preference/priority with ffmpeg? [19:04:24 CEST] <JEEBsv> there is a thing for that, although people tend to just disable the dca decoder [19:07:40 CEST] <lvml> "Disable the dca decoder" - without source code modification? How? [19:10:47 CEST] <JEEBsv> configure script option [19:11:24 CEST] <JEEBsv> --disable-decoder=dca [19:11:32 CEST] <lvml> Ok - but that also disables dca as an encoder, right, and since libdcadec can only decode, you lose the ability to encode dca then. [19:11:49 CEST] <JEEBsv> no, that would only disable the decoder as far as I can tell :P [19:34:51 CEST] <bmcginty> Does anyone know if there's a way to have FFMpeg determine an output extension of a file automatically? Basically, I'm downloading a file, audio or video, and I don't know what internal codec it's using. However, All I'm interested in is the audio, so I use -acodec copy, and set the output extension to what ffprobe returns. I can't run ffprobe on a downloading file as far as I know, and if I don't set FFMpeg's output ... [19:34:57 CEST] <bmcginty> ... extension/format correctly, the audio doesn't get just copied, it gets copied and transcoded, which is slow and wasteful for my needs. I can reexplain this if it's unclear. [19:56:10 CEST] <bmcginty> Ah! I can pipe the copied audio into the nut container format, at least with what I've tried so far. FFMPeg rocks! [20:05:30 CEST] <klaxa> you could also try matroska [20:05:51 CEST] <klaxa> not sure nut is implemented widely, then again if things are using libavformat, they probably support nut [20:15:28 CEST] <Kaedenn> Can I ask questions about avconv here? [20:17:09 CEST] <Kaedenn> Well I can use either, really. I have a bunch of pictures, frame_%04d.png, that I wish to combine into a single .avi. Initially, everything is fine, but the resulting AVI has a bunch of dithering effects and is quite low quality compared to the pngs. [20:18:23 CEST] <Kaedenn> My command is, avconv -f image2 -i frames_0%03d.png -r 30 4_random_short.avi [20:19:13 CEST] <Kaedenn> How do I increase the quality of the resulting AVI? [20:19:50 CEST] <durandal_1707> use different codec and ffmpeg [20:20:13 CEST] <Kaedenn> I thought ffmpeg was deprecated, and what codec should I use? [20:20:43 CEST] <sfan5> either use a different codec like h264 or increase quality [20:21:47 CEST] <BtbN> It's quite amazing this deprecated-brainwashing is still in place and works. [20:22:16 CEST] <Kaedenn> I run ffmpeg, and it states: *** THIS PROGRAM IS DEPRECATED *** [20:22:16 CEST] <Kaedenn> This program is only provided for compatibility and will be removed in a future release. Please use avconv instead. [20:22:26 CEST] <sfan5> see the link posted by fflogger [20:24:09 CEST] <Kaedenn> ...are you serious. [20:24:10 CEST] <Kaedenn> x_x [20:24:40 CEST] <Kaedenn> So the version of ffmpeg I have is a wrong one? [20:25:13 CEST] <sfan5> you have Libav instead of FFmpeg installed [20:26:54 CEST] Action: Kaedenn downloads the static build [20:28:38 CEST] <Kaedenn> Thank you. [20:28:46 CEST] <Kaedenn> The manpages now state how to do what I want. [20:28:53 CEST] <Kaedenn> ffmpeg -h encoder=libx264 <rest of my args> # Yes? [20:29:22 CEST] <sfan5> that shows the help [20:29:26 CEST] <Kaedenn> ...oh. [20:29:28 CEST] <Kaedenn> Duh. [20:29:29 CEST] <Kaedenn> XD [20:29:56 CEST] <Kaedenn> Okay so I have output/frames_%04d.png and media/Bowser.wav that I'd like to combine into a movie. [20:30:04 CEST] <sfan5> if you want to use libx264 pass -c:v libx264 to ffmpeg [20:30:29 CEST] <Kaedenn> Has anyone made a graphical UI to ffmpeg? [20:30:38 CEST] <Kaedenn> For idiots like me who have difficulty wading through options. [20:33:28 CEST] <Kaedenn> Woo, using libx264 works like a charm. [20:34:09 CEST] <Kaedenn> How do I add a wav file to my output? [20:37:42 CEST] <sfan5> just add -i <your wav file> [20:39:48 CEST] <Kaedenn> ...the images are upside-down [20:40:22 CEST] <Kaedenn> Oh that's amusing, they're upside-down in my generator too. [20:41:36 CEST] <Kaedenn> Oh hey I think I found a bug in SDL. It's probably between the keyboard and the chair, but it's a bug! [20:41:39 CEST] <Kaedenn> Thank you all for your help. [21:34:12 CEST] <Kaedenn> And, lastly, how do I make the program exit after processing my input at 30fps, and not at the end of the audio track? [21:34:47 CEST] <BtbN> -shortest [21:46:06 CEST] <^bit> Hi, how can I force a smaller resolution when recording a live feed? This is the command I'm using currently http://pastebin.com/USvW9i0X [22:01:35 CEST] <BtbN> you can't without re-encoding. [22:07:04 CEST] <Guest8503> hello guys [22:07:43 CEST] <^bit> BtbN: you can't do it in real time while recording the stream? [22:07:48 CEST] <Guest8503> I'm trying to transcode from h264 to dnxhd [22:08:29 CEST] <BtbN> if you have enough cpu power, of course. [22:08:42 CEST] <Guest8503> i don't know why i'm still getting this msg:[dnxhd @ 00000000030c9800] picture could not fit ratecontrol constraints, increase qmax [22:37:01 CEST] <^bit> BtbN: but I'm looking for the command to do that with [22:46:24 CEST] <antiatom> Hi all, I am using Kdenlive, and I notice that the render profiles have a different syntax than FFmpeg does on the CLI. I am trying to figure out how to best choose encoding options for x264 and Opus in MKV file for output [22:51:00 CEST] <Chuckles32> Hello. I have a filter_complex question. I am using ffmpeg -drawtext with timecode. The text is displayed as: 09:25:44:12 Is it possible to remove the frame info from the display? We would prefer if the text was displayed as: 09:25:44 [22:57:32 CEST] <durandal_1707> Chuckles32: open feature request on bug tracker [23:00:51 CEST] <Chuckles32> durandal_1707: OK, so there is no native support for dropping the frame info from the timecode. Perhaps there is another way to skin the cat.... We are just looking to timestamp videos. We are running in a bash script and we are passing our start time to FFMPEG as timecode variables, but if there's another way to draw a clock on the video, i'd love to know.... [23:03:06 CEST] <durandal_1707> you can write into file and let drawtext read from it [00:00:00 CEST] --- Mon Jul 6 2015

1 0

ffmpeg-devel.log.20150704
by burek 05 Jul '15

05 Jul '15

[01:47:15 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:8255b14c685f: swscale/output: fix input indexing in yuv2ya8_2_c() [01:47:16 CEST] <cone-987> ffmpeg 03Kevin Coyle 07master:126271138803: YUV->BGR32 MMX support [02:28:22 CEST] <cone-987> ffmpeg 03George Boyle 07master:4385a1ce8f53: fate/api-tests: Tests that need samples should only run if SAMPLES is set [10:56:45 CEST] <cone-569> ffmpeg 03Paul B Mahol 07master:c3a5702ebf1a: lavfi: add (a)drawgraph filter [10:56:45 CEST] <cone-569> ffmpeg 03Paul B Mahol 07master:e658b1036c0b: avfilter/avf_showvolume: make it possible to use current channel number in color expression [10:56:45 CEST] <cone-569> ffmpeg 03Paul B Mahol 07master:5233f2534ca5: avformat/yuv4mpegenc: use avio_printf() [14:57:47 CEST] <BBB> so, uhm, what happened to ffmpeg.org? I dont understand the ML messages [14:57:50 CEST] <BBB> theyre pretty terse [14:58:09 CEST] <BtbN> Nothing? [14:58:13 CEST] <BtbN> It just needs a new hoster [14:58:20 CEST] <BBB> whos hosting it now? [14:58:32 CEST] <BBB> didnt videolan host us? [14:58:37 CEST] <BBB> or is the website hosted by someone else? [14:59:38 CEST] <durandal_1707> can switch everything to github [15:08:40 CEST] <Compn> BBB : arpi is currently hosting it at his place, got new boss, might not host it in future [15:08:49 CEST] <Compn> arpi = mplayer founder [15:09:27 CEST] <Compn> videolan hosts the git (although we could have hosted it too, since we have had git server for a while) [15:10:35 CEST] <Compn> ffmpeg/mplayer/rtmpdump is mailing lists, website, mplayer svn (and git mirror), rtmpdump svn... maybe ffmpeg trac and mplayer trac? i dont remember. [15:12:40 CEST] <durandal_1707> ask sponsors? [15:13:24 CEST] <Compn> and samples repo. [15:15:10 CEST] <Compn> we could post it to the news section on ffmpeg/mplayer sure [15:15:25 CEST] <Compn> not sure what other sponsors you are talking about? [15:15:37 CEST] <Compn> g**gle? samsung? [15:16:15 CEST] <durandal_1707> apple, *cough* [15:17:16 CEST] <durandal_1707> seriously, how much time is left? [15:19:14 CEST] <Compn> probably >1 week going by michael's last mail [15:19:38 CEST] <Compn> more likely ... monday. [15:19:45 CEST] <Compn> if we are lucky [15:20:36 CEST] <Compn> guess i'll make a news entry [15:21:07 CEST] <michaelni> durandal_1707, time is unknown [15:21:58 CEST] <durandal_1707> ah [15:23:02 CEST] <Compn> michaelni : should i make a news entry or post patch for it? [15:25:57 CEST] <Compn> i'll post a patch. [15:31:02 CEST] <michaelni> i dont have time to review patches [15:31:39 CEST] <durandal_1707> for how long? [15:31:53 CEST] <michaelni> maybe until monday ybut this is a guess [15:32:32 CEST] <michaelni> i mean the servers will possibly be shutdown mo but it could be sooner or later [15:32:41 CEST] <michaelni> i wont have time until this is resolved [15:40:33 CEST] <Compn> news entry posted [15:40:40 CEST] <Compn> to ffmpeg [15:41:44 CEST] <Compn> if anyone finds a problem feel free to fix ... [15:57:36 CEST] <ubitux> Compn: we have a new server already [15:58:52 CEST] <Compn> say whuh ? [15:59:00 CEST] <ubitux> ? [15:59:35 CEST] <ubitux> admins are migrating stuff as i'm talking [15:59:47 CEST] <durandal_1707> who is boss now? [16:00:00 CEST] <Compn> ubitux : whos the host ? [16:00:05 CEST] <ubitux> who wants to be? [16:00:56 CEST] Action: kierank has offered [16:01:14 CEST] <ubitux> Compn: i ordered a box at a random provider [16:01:41 CEST] <Compn> ubitux : dont we need a sponsored/free host as well ? [16:02:33 CEST] <ubitux> it's not expensive, that's fine with me for now [16:02:48 CEST] <kierank> well I have a proper datacentre [16:02:53 CEST] <kierank> can show you a photo if you want [16:03:02 CEST] <Compn> kierank : what country is it hosted in ? uk ? [16:03:42 CEST] <kierank> yes [16:03:51 CEST] <kierank> se11 5jh [16:03:53 CEST] <kierank> there [16:05:17 CEST] <Compn> kierank : probably get a server setup then, we can use it as a backup, i think. although i am not admin do not take anything i say as speaking for anyone... [16:05:53 CEST] <Compn> then we can switch over to it if ubitux's "random provider" takes a shit, as many hosts do :\ [16:06:51 CEST] Action: Compn curses anyone who uses cloud hosting. [16:09:43 CEST] <Compn> i believe we need better backup system anyway, so having kierank as a backup would be great :) [16:12:00 CEST] <kierank> ubuntu installer has locked up [16:12:02 CEST] <kierank> which isn't good [16:14:18 CEST] <BBB> you guys arent very technical all of a sudden [16:14:28 CEST] <BBB> you should compare the features offered by kierank and ubitux providers [16:14:34 CEST] <BBB> like, speed, data cap, etc. [16:14:43 CEST] <BBB> and see which one you think you like better [16:14:47 CEST] <BBB> and then use the other as a backup [16:14:48 CEST] <ubitux> we're probably going to use both [16:14:56 CEST] <ubitux> with different services [16:16:05 CEST] <Compn> BBB : well i dont want to insult ubitux or kierank by talking about it in detail :P [16:19:27 CEST] <Compn> and i'm waiting patiently for a "big host" to offer [16:19:45 CEST] <Compn> 24/7 staff mmmm [17:28:25 CEST] <kierank> wm4: seriously? [17:29:48 CEST] <wm4> kierank: what? [17:29:55 CEST] <kierank> you want drawtext to use libass [17:30:00 CEST] <wm4> yes [17:30:14 CEST] <wm4> and that's less insane than it sounds [17:30:17 CEST] <kierank> when you have a hammer everything looks like a nail I guess [17:30:38 CEST] <wm4> because libass has a relatively light weight implementation of a complete text renderer including stuff like arabic text layout etc. [17:32:01 CEST] <wm4> you could also use pango, but then the dependencies will kill you [18:23:57 CEST] <Plorkyeran> libass's API is kinda weird for just drawing some text on video [18:24:38 CEST] <Plorkyeran> but otherwise the only thing that makes it not exactly what you want is that it uses a bespoke markup language [18:38:19 CEST] <BtbN> Didn't jessie have real ffmpeg? [18:39:05 CEST] <nevcairiel> dont think it made jessie [18:40:07 CEST] <beastd> no ffmpeg in jessie [18:40:44 CEST] <BtbN> hrm [18:41:20 CEST] <nevcairiel> debian moves slow like that [18:41:39 CEST] <atomnuker> politics [19:00:13 CEST] <BtbN> interesting webcam. It can do 1080p via h264 or mjpeg. Or 2304x1536 via RAW YUYV [19:30:50 CEST] <philipl> BtbN: Does it come with a 10GBe link to stream that? :-P [19:31:26 CEST] <philipl> Did you get a chance to look at the nvenc 444 stuff? [19:32:51 CEST] <BtbN> not yet, i'm actualy at work, nowhere near my nvidia machines [19:33:00 CEST] <BtbN> Just because i have an AC here [19:33:34 CEST] <philipl> heh. [19:34:02 CEST] <BtbN> I still think it needs a static init, to disable the yuv444p support when in non-lossless mode. Otherwise you might break things for Kepler hardware [19:34:33 CEST] <philipl> So you think that kepler supports lossless yuv444p but not lossy, and that's fixed in Maxwell? [19:35:02 CEST] <philipl> I've got a 670 in a box, but I don't think I can be arsed to dig it out. [19:35:24 CEST] <philipl> I really should sell that before the value goes to zero... [19:36:44 CEST] <BtbN> i think kepler doesn't support yuv444 at all [19:37:04 CEST] <philipl> Then it's fine right? [19:37:12 CEST] <philipl> Oh, I see. [19:37:18 CEST] <philipl> Well, maybe. [19:37:26 CEST] <philipl> So, the init will fail if the smver is too low. [19:37:53 CEST] <philipl> but you'd rather that it triggers implicit conversion of yuv444p instead. [19:37:59 CEST] <BtbN> Well, it could also try to do lossy yuv444p [19:38:04 CEST] <BtbN> there's nothing that would prevent that [19:38:32 CEST] <philipl> But I mean, you're saying Kepler doesn't support for yuv444p at all, and maxwell, for me, supports lossy and lossless. [19:38:44 CEST] <philipl> So I'm confused as to which case would be exposed to buggy lossy yuv444p [19:39:14 CEST] <BtbN> yuv444p is in the static list of supported formats [19:39:20 CEST] <BtbN> but on Kepler HW, it's not supported [19:39:31 CEST] <philipl> Right, and the init will fail when the cuda version check fails. [19:39:33 CEST] <BtbN> At least i think it isn't, which is what i want to test [19:39:44 CEST] <BtbN> Not for lossly mode, as that checks for 3.0 [19:39:56 CEST] <philipl> That's not how i wrote the check... [19:40:04 CEST] <philipl> + target_smver = avctx->pix_fmt == AV_PIX_FMT_YUV444P ? 0x52 : 0x30; [19:40:10 CEST] <BtbN> oh, right. Thought it checked for lossless mode [19:40:34 CEST] <BtbN> hm, i wonder if that check could be moved into a static init function [19:40:41 CEST] <BtbN> of if initializing cuda would be too much there [19:40:51 CEST] <philipl> I would have thought it would be too much. [19:41:13 CEST] <philipl> The main advantage is that you can trigger implicit conversion if it's not in the pixfmt list, but that's probably not a big deal in practice. [19:42:06 CEST] <philipl> Initialising cuda all the time seems scary to me. [20:00:27 CEST] <BtbN> Doesn't need to be initialized, just device enumeration [20:31:51 CEST] <BtbN> Great, the entire RPi crashed, hardware wise. Streaming a USB Webcam via USB Wifi seems to be too much for it. [20:33:07 CEST] <wm4> I managed to crash the rpi so often [20:33:16 CEST] <wm4> in many cases, only the GPU crashed/froze [20:33:47 CEST] <BtbN> GPU is not involved here [20:34:19 CEST] <BtbN> USB Webcam that outputs h264 + alsa capture, libfdk_aac for aac, and then streaming it via rtmp [20:34:39 CEST] <BtbN> The pi just froze now [20:36:34 CEST] <BtbN> And the, slightly more expensive, replacement-NUC is lacking RAM. Great. [20:43:04 CEST] <michaelni> Compnn, and others if anyone noties problems with the servers ping me (in ffmpeg-dev best so iam not multi pinged) [20:56:27 CEST] <rcombs> wm4: thanks, giant binary firmware? [20:58:01 CEST] <wm4> *shrug* [21:05:57 CEST] <wm4> has anyone looked at vaapi HEVC decoding yet? [22:57:28 CEST] Action: durandal_1707 porting removegrain [00:00:00 CEST] --- Sun Jul 5 2015

1 0

ffmpeg.log.20150704
by burek 05 Jul '15

05 Jul '15

[02:07:30 CEST] <joltman> I'm following the compiliation guide for Ubuntu here: https://trac.ffmpeg.org/wiki/CompilationGuide/Ubuntu [02:08:15 CEST] <joltman> and I'm running into an issue where ffmpeg's ./configure line is telling me "ERROR: libx264 not found" [02:08:41 CEST] <c_14> Did you install libx264? [02:08:48 CEST] <joltman> I'm pasting the entire output of config.log to pastie in a moment [02:08:51 CEST] <joltman> c_14, yes i did [02:08:57 CEST] <joltman> matter of fact, i re-ran it just to be sure [02:09:17 CEST] <joltman> i can see libx264.a in the ~/ffmpeg_build folder [02:09:32 CEST] <c_14> Can you see an x264.pc as well? [02:09:35 CEST] <joltman> i can see libx264.a in the ~/ffmpeg_build/lib folder [02:09:38 CEST] <joltman> hmm [02:09:45 CEST] <c_14> should be in lib/pkgconfig [02:09:51 CEST] <joltman> no i can't...interesting [02:10:00 CEST] <joltman> ah [02:10:12 CEST] <joltman> that's in the ~/ffmpeg_build/lib/pkgconfig folder [02:10:17 CEST] <joltman> so yeah, it's there [02:10:43 CEST] <c_14> Did you export PKG_CONFIG_PATH=$HOME/ffmpeg_build/lib/pkgconfig ? [02:11:10 CEST] <joltman> as i said, i ran the exact command that's listed on that page. [02:11:12 CEST] <heap__> hi [02:11:18 CEST] <heap__> is there any 2.7 package for raspbian? [02:11:31 CEST] <joltman> this is the ./configure command i ran: PATH="$HOME/bin:$PATH" PKG_CONFIG_PATH="$HOME/ffmpeg_build/lib/pkgconfig" ./configure --prefix="$HOME/ffmpeg_build" --pkg-config-flags="--static" --extra-cflags="-I$HOME/ffmpeg_build/include" --extra-ldflags="-L$HOME/ffmpeg_build/lib" --bindir="$HOME/bin" --enable-gpl --enable-libass --enable-libfdk-aac --enable-libfreetype --enable-libmp3lame --enable-libopus --enable-libtheora [02:11:31 CEST] <joltman> --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-nonfree --enable-x11grab [02:12:59 CEST] <c_14> joltman: I'll probably need to see the config.log [02:13:05 CEST] <joltman> ok [02:13:13 CEST] <c_14> heap__: none I know of, never checked though [02:13:14 CEST] <joltman> what is the preferred method of getting it to you? Pastie? [02:13:19 CEST] <heap__> c_14: ; / [02:13:35 CEST] <heap__> what is the module for hw accelaration i have to enable during compliation on rpi2 ? [02:14:02 CEST] <c_14> joltman: any pastebin service'll work. Some might have length limits though. [02:14:20 CEST] <joltman> yeah, pastie doesn't like anything greater than 64k [02:14:40 CEST] <c_14> You can probably cut the middle portion. [02:14:48 CEST] <c_14> just denote the part you cut with [..] or something [02:15:00 CEST] <heap__> RROR: gnutls not found using pkg-config [02:15:04 CEST] <c_14> The interesting bits are usually at the beginning or end. [02:15:05 CEST] <joltman> http://pastebin.com/NnG9LERa [02:15:34 CEST] <heap__> i installed gnutls [02:15:37 CEST] <c_14> heap__: you probably need gnutls-dev or something [02:15:55 CEST] <heap__> both [02:15:57 CEST] <heap__> installed [02:16:45 CEST] <c_14> heap__: what does pkg-config --exists --print-errors return? [02:16:47 CEST] <c_14> eh [02:16:49 CEST] <heap__> no [02:16:50 CEST] <heap__> wait [02:16:57 CEST] <c_14> pkg-config --exists --print-errors gnutls that is [02:17:49 CEST] <heap__> what about that HW accelaration? [02:18:56 CEST] <c_14> heap__: what are you looking for? hw instruction sets or something like vdpau [02:19:18 CEST] <c_14> joltman: you downloaded a recent git snapshot of ffmpeg, right? [02:19:25 CEST] <joltman> when i compile z264, i get this error: Warning: gpac is too old, update to 2007-06-21 UTC or later [02:19:26 CEST] <joltman> c_ [02:19:34 CEST] <joltman> c_14, yes, i grabbed today's version [02:19:38 CEST] <heap__> http://emby.media/community/index.php?/topic/16869-mediabrowser-server-swit… [02:19:51 CEST] <heap__> ppl mentioned there is some hw acceleration in ffmpeg on rpi2 [02:20:01 CEST] <c_14> joltman: what's the output of ./version.sh (in the source directory of ffmpeg). Also, you can ignore that gpac error [02:20:47 CEST] <heap__> wi35 [02:21:05 CEST] <joltman> hmmmmm [02:21:08 CEST] <joltman> 2.3.git [02:21:10 CEST] <joltman> so it's very old [02:21:16 CEST] <joltman> lemme try removing that directory [02:23:01 CEST] <joltman> i removed the ffmpeg source directory, re-grabbed the ffmpeg-snapshot.tar.bz2, decompressed and ran the same ./configure [02:23:03 CEST] <joltman> same issue [02:23:07 CEST] <heap__> c_14: ?:)idea? [02:23:19 CEST] <joltman> still showing 2.3.git as the version. [02:24:01 CEST] <c_14> heap__: try --enable-mmal [02:25:51 CEST] <c_14> joltman: I just downloaded the snapshot and it shows the correct version for me (N-73399-g866404d [02:25:53 CEST] <c_14> ) [02:26:44 CEST] <heap__> thanks [02:36:34 CEST] <heap__> c_14: whats that mmal? [02:36:41 CEST] <heap__> gpu hw acceleration? [02:38:49 CEST] <c_14> heap__: h264 hardware decoding using whatever mmal is [02:39:50 CEST] <heap__> ?:) [02:41:16 CEST] <joltman> then what the heck am i doing wrong [02:41:17 CEST] <joltman> hmmm [02:41:22 CEST] <heap__> c_14: cant get it [02:41:44 CEST] <c_14> heap__: can't get what? [02:42:10 CEST] <joltman> i'm an idiot [02:42:24 CEST] <joltman> i deleted the folder, but forgot to delete the old tar.bz [02:42:26 CEST] <joltman> running again [02:42:52 CEST] <joltman> miraculously, it works [02:42:59 CEST] <joltman> thanks for helping me through my dumbness [02:43:10 CEST] <c_14> heap__: mmal is the hardware accel thing for rpi2 though, just cross-referenced with the ml [02:47:21 CEST] <heap__> ok [02:47:23 CEST] <heap__> thx [02:53:52 CEST] <heap__> so on rpi2 should i configure it with -arch=armel ? [03:00:06 CEST] <klaxa> if you compile it on the pi itself the architecture should be autodetected afaik [03:04:27 CEST] <heap__> ok [03:20:06 CEST] <joltman> c_14, it built correctly. Thanks again! [03:29:27 CEST] <spl33n_> hello, i come back, i use ffmpeg to convert video to hight quality gif, durandal_1707 suggest me to use palettegen and paletteuse, i look this here http://blog.pkh.me/p/21-high-quality-gif-with-ffmpeg.html but not work i have this error [AVFilterGraph @ 0x80a431040] No such filter: 'palettegen' [03:33:57 CEST] <prelude2004c> hello everyone [03:34:17 CEST] <prelude2004c> anyone know why ffmpeg would be so slow to start -i < input > and then -f flv rtmp...out [03:34:31 CEST] <prelude2004c> when is tart ffmpeg, it takes a few seconds to get stream and then it starts very slow.. and speeds up [03:34:35 CEST] <prelude2004c> very odd behavior [03:34:43 CEST] <prelude2004c> when i play the input with vlc it works right away [03:34:47 CEST] <prelude2004c> no slow downs to start at all [04:45:16 CEST] <pandb> I'm trying to apply a filter to a frame with av_buffersrc_write_frame() and av_buffersink_get_frame(), but av_frame_ref(), which is called by ..._get_frame() fails because ->buf[0] is null in the AVFrame* I've passed to ..._get_frame() [04:45:27 CEST] <pandb> however, the AVFrame* in question is perfectly fine when I encode it and write it to output [04:45:38 CEST] <pandb> this frame was generated with sws_scale() [04:46:07 CEST] <pandb> also, many of the attributes of this frame are uninitialized, like height, width, format and others. [04:46:37 CEST] <pandb> is there a function I should have called before or after sws_scale() (and sws_getContext()) that would take care of all or most of these things? [04:46:54 CEST] <pandb> The actual video resulting from these sws_scale'd frames displays perfectly fine, however. [04:47:56 CEST] <pandb> (I posted this exact same sequence of questions to #libav, I hope that's acceptable, since it seems well within the purview of both of these channels) [06:33:12 CEST] <durandal_1707> spl33n_: compile your own ffmpeg [11:10:20 CEST] <emk-> hello [11:10:57 CEST] <emk-> is it possible to recode variable frame rate -> constant frame rate correctly with ffmpeg? [11:11:38 CEST] <emk-> I've done it using mencoder, but no success using ffmpeg [11:12:23 CEST] <durandal_1707> what you tried? [11:13:33 CEST] <emk-> i've tried -vsync 1/2 [11:13:54 CEST] <emk-> but ffmpeg doesn't duplicate frames when the video goes to low framerate [11:14:39 CEST] <emk-> basically -vsync 1 should duplicate frames, but it just passes them, as normal framerate [11:17:32 CEST] <durandal_1707> ask on user mailing list [13:12:10 CEST] <user01> ffmpeg 2.6.3 on debian -- the only way i could get a upside down video to flip that had metadata for 180 was to so 2 operations of transpose=1 . . . was there a more elegant solution like removing the 180? [13:12:24 CEST] <user01> s/so/do [13:13:52 CEST] <user01> i had tried this but it didnt work: ffmpeg -i 20150702_223047.mp4 -c -metadata:s:v:0 rotate=0 output.mp4 [13:14:32 CEST] <user01> tried this and it didnt work: ffmpeg -i 20150702_223047.mp4 -vf "transpose=2,transpose=2,format=yuv420p" -metadata:s:v rotate=0 -codec:v libx264 -codec:a copy output.mp4 [13:15:31 CEST] <user01> tried this and it didnt work: ffmpeg -i 20150702_223047.mp4 -vf vflip,hflip out.mp4 [19:30:20 CEST] <Seylerius> I'm trying to convert an mp4 (ffmpeg info: http://sprunge.us/LciO) into something that Lightworks will accept (http://lightworks.wikidot.com/formats) I'm on ArchLinux. Any recommendations? [19:38:00 CEST] <Seylerius> Lightworks can't see any video in the output of `ffmpeg -i 20150628_101238_01.mp4 -codec:v mpeg2video -codec:a aac 20150628_101238_01.avi` [19:38:23 CEST] <Seylerius> Maintaining quality is good, although there are some mild space concerns. [19:39:13 CEST] <Seylerius> I have 7.3G of mp4 video like that, and about 70G of free space. [19:58:56 CEST] <DHE> I thought quicktime H.264 was what your input file already is... [20:00:01 CEST] <DHE> oh, never mind me then [20:06:22 CEST] <BtbN> What the hell is this webcam doing. Non-monotonous DTS, tons of them. [20:17:43 CEST] <relaxed> Seylerius: try, ffmpeg -i input.mk4 -c:v mpeg2video -q:v 2 -c:a pcm_s16le output.avi [20:18:03 CEST] <relaxed> mp4* [20:19:55 CEST] <relaxed> Seylerius: also try stream copying to .mov [20:37:18 CEST] <Seylerius> relaxed: Stream copying being changing to mov with codecs set to copy? [20:39:28 CEST] <Seylerius> relaxed: The line you suggested produced an avi, but Lightworks again didn't detect that it contained video, only audio. [20:39:34 CEST] <Seylerius> relaxed: trying stream-copy [20:42:21 CEST] <Seylerius> relaxed: that didn't work either. [20:42:32 CEST] <Seylerius> Complained about codecs lacking global headers [20:53:41 CEST] <relaxed> Seylerius: they list ffmpeg, but don't give a single example :/ [21:09:38 CEST] <Seylerius> relaxed: It was ffmpeg that complained with the stream-copy. [21:10:02 CEST] <Seylerius> I did this: `ffmpeg -i 20150628_101238_01.mp4 -codec copy 20150628_101238_01.mov` [21:38:59 CEST] <BtbN> Is there a way to just throw away timestamps(dts) when stream copying, and make up new ones? This webcam is realy bad with its timestamps aparently [21:53:04 CEST] <asimas> When concatenating files I'm receiving an errror saying that SAR values must be equal. The values are 1:1 and 65536:655536. As I understand these are equivalent. Is there any way to quietly avoid this error or do I need to setsar? [22:23:01 CEST] <edoceo> Hey, I've got a matroska/webm file that I cannot decode, I get a " Read error at pos. 37" [22:23:25 CEST] <edoceo> So, is it possible to get ffmpeg to ignore this somehow and proceed through the rest of the file? [22:24:23 CEST] <edoceo> Here's my full command & output: http://edoceo.io/paste?p=OBy2nu [22:25:01 CEST] <edoceo> Oddly, these .webm files play in VLC, just w/o sound [22:27:01 CEST] <smo_> hi [22:35:22 CEST] <smo_> i open some youtube videos wih commands line like : http://pastebin.com/SPHfXSXj [22:35:55 CEST] <smo_> 2 youtube links (video/audio) But it takes aaaaaaaaggges to start the streaming .... how can i speed it up please? [22:36:12 CEST] <smo_> links opened in the browser are opened very fast [22:36:37 CEST] <sfan5> why not just use youtube-dl [22:36:46 CEST] <sfan5> youtube-dl does exactly what you want [22:36:52 CEST] <sfan5> download video & audio and mux it into 1 file [22:37:19 CEST] <sfan5> also using -preset ultrafast or -deinterlace with -c:v copy won't do a thing [22:37:56 CEST] <sfan5> "copy" just copies the video data [22:38:06 CEST] <sfan5> without re-encoding you can't do deinterlacing [22:38:30 CEST] <sfan5> and without encoding specifying a preset doesn't make sense [22:40:53 CEST] <smo_> i know... but in stream it in live in my app [22:41:01 CEST] <smo_> just use ffmpeg to merge the 2 files [22:41:23 CEST] <smo_> then pipe results thru nodejs in my video player in live [22:41:37 CEST] <smo_> just take too long to start the stream [22:58:16 CEST] <sfan5> smo_: this is a little hacky but starts doing something in about 3 seconds http://pastebin.com/M42e6t9d [23:00:57 CEST] <smo_> very nice sfan5 but don t think i can do it thru nodejs [00:00:00 CEST] --- Sun Jul 5 2015

1 0

ffmpeg-devel.log.20150703
by burek 04 Jul '15

04 Jul '15

[00:03:11 CEST] <kierank> don't you live in hamburg? [00:03:27 CEST] <kierank> wasn't there a vlc meeting there [00:03:32 CEST] <kierank> at google hamburg? [00:03:40 CEST] <nevcairiel> i do, and no idea [00:03:47 CEST] <Daemon404> nevcairiel is the competetition [00:03:56 CEST] <kierank> loool [00:03:58 CEST] <kierank> yeah true [00:04:11 CEST] <kierank> so is chrome in a way [00:04:13 CEST] <nevcairiel> what do i care, i dont make a player [00:04:31 CEST] <nevcairiel> well $work does, but its an entirely different target audience than vlc [00:06:52 CEST] <Daemon404> nevcairiel, wrong. vlc's target audience: https://trac.videolan.org/vlc/ticket/35 [00:07:29 CEST] <nevcairiel> as long as they dont have a full media library and 10ft browsing interface, its not that audience =p [00:08:21 CEST] <cone-448> ffmpeg 03Paul B Mahol 07master:2778fdbe5424: swscale: implement YA8 output [01:27:44 CEST] <rcombs> does Benoit Fouet hang out on IRC? [01:50:37 CEST] <kierank> erm [02:45:23 CEST] <klaxa> okay, i have a round of patches ready and i want to send them with git send-email, but i'm afraid i will generate a lot of noise if i do it incorrectly [02:45:47 CEST] <klaxa> i sent test emails to myself but they do not appear correctly in the thread [02:46:10 CEST] <klaxa> does --in-reply-to require the patches to be sent to the mailing list for it to be displayed correctly? [02:46:40 CEST] <c_14> You need the message you're replying to in your inbox. [02:46:47 CEST] <klaxa> well it is [02:47:18 CEST] <c_14> Are you sure the in-reply-to is correct? [02:47:43 CEST] <klaxa> i copy pasted it, but i'll try again with it as a command line parameter, git send-email asked me for it so i figured i could also use that [02:48:27 CEST] <klaxa> maybe you can confirm that the message id for this email https://ffmpeg.org/pipermail/ffmpeg-devel/2015-July/174937.html is 20150701141008.GA805947(a)phare.normalesup.org ? [02:49:31 CEST] <klaxa> using --in-reply-to as a command line option did not change the behavior :( [02:49:41 CEST] <c_14> ye, that should be correct [02:52:35 CEST] <c_14> If it's a final or close-to-final version, you could probably just resend the whole patchset as a new thread. [03:39:45 CEST] <cone-722> ffmpeg 03wm4 07master:1316df7aa98c: lavu: add an API function to return the Libav version string [03:39:45 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:e15e78f391ab: Merge commit '1316df7aa98c4784f190d107206d0bb12c590b89' [03:44:07 CEST] <jamrial> michaelni: IMO having both libavutil/ffversion.h and avversion.h is silly. we should probably choose one and get rid of the other [03:45:45 CEST] <jamrial> or maybe moving ffversion.h to top level dir [03:48:01 CEST] <michaelni> i didnt mere avversion.h except the cleanup code but feel free to improve/change [03:50:45 CEST] <cone-722> ffmpeg 03Vittorio Giovara 07master:910247f1720c: lavc: Deprecate avctx.{inter,intra}_quant_bias [03:50:46 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:a8ab64d2f700: Merge commit '910247f1720c6aae422723c05dac6d0b19f20bec' [03:51:40 CEST] <jamrial> michaelni: my bad, saw you merged the distclean addition but not the actual part that creates the header [03:54:56 CEST] <jamrial> as in, saw the distclean addition, so i assumed you merged it all. didn't notice the rest wasn't there [03:55:14 CEST] <jamrial> it's good as is, then [03:55:48 CEST] <michaelni> good [03:58:43 CEST] <cone-722> ffmpeg 03Alexandra Hájková 07master:9752d2e6cc9b: asfdec: prevent possible memory leak in the asf_read_metadata_obj [03:58:44 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:7755a5744005: Merge commit '9752d2e6cc9b9e8070ec515db8ed8374683d0856' [04:06:01 CEST] <cone-722> ffmpeg 03Alexandra Hájková 07master:016cac75c606: asfdec: prevent the infinite loop in detect unknown_subobject [04:06:02 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:30ffbeb04abd: Merge commit '016cac75c6061a1c03f812ddf258b8baefe70b00' [04:14:41 CEST] <cone-722> ffmpeg 03Andreas Cadhalpun 07master:76d4c62734fb: webp: Make sure enough bytes are available [04:14:42 CEST] <cone-722> ffmpeg 03Andreas Cadhalpun 07master:0762152f7af6: webp: fix infinite loop in webp_decode_frame [04:14:43 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:3ffa38580190: Merge commit '76d4c62734fbb8a9f497712812f30ff5c27e787f' [04:37:03 CEST] <cone-722> ffmpeg 03Luca Barbato 07master:80f955c90867: vda: Check the correct pointer for buffer allocation [04:37:04 CEST] <cone-722> ffmpeg 03Michael Niedermayer 07master:daff49ccf3e5: Merge commit '80f955c90867561dcce769216bc497e13281eb38' [07:31:42 CEST] <prelude2004c> hey everyone [07:31:47 CEST] <prelude2004c> question.. i dont know if this is a bug or not [07:31:59 CEST] <prelude2004c> i am using nginx with rtmp [07:32:33 CEST] <prelude2004c> when i try to publish something with ffmpeg > nginx's rtmp stream server it goes very very slow [07:32:43 CEST] <prelude2004c> complaining about realtime not working or something [09:03:34 CEST] <prelude2004c> looking for some help [09:03:37 CEST] <prelude2004c> anyone around ? [09:23:27 CEST] <durandal_1707> prelude2004c: ? [10:08:44 CEST] <mateo`> hello there o/ i'm wondering if there is a conveniant way to get the plane width given a specific pixel format and the plane number (and not the linesize allocated by av_frame_get_buffer) ? [10:10:02 CEST] <durandal_1707> mateo`: see filters, like vf_ssim [10:12:13 CEST] <mateo`> durandal_1707: thanks for the reference [10:15:31 CEST] <mateo`> durandal_1707: so it's just a matter of rshiftting the frame width with log_chroma_w ? [10:17:39 CEST] <wm4> that's wrong [10:17:49 CEST] <wm4> it needs to be a rounding-up rshift at least [10:18:11 CEST] <wm4> (because the awful way ffmpeg handles cropped video, you can get odd widths with subsampled yuv) [10:24:15 CEST] <mateo`> my goal is relatively simple i would like to be able to copy an AVFrame to disk (or a chunk of memory containing the whole frame) [10:25:40 CEST] <mateo`> i'm looping through each plane, but i need to retreive the actual width to copy and not the linesize [10:26:29 CEST] <nevcairiel> then just get the pixfmt descriptor, it tells you the subsampling factor [10:26:37 CEST] <nevcairiel> and apply it to the width [10:28:34 CEST] <durandal_1707> mateo`: filters use FF_CEIL_RSHIFT and that is correct [10:31:44 CEST] <mateo`> nevcairiel: the subsampling factor is stored as log2_chroma_w and log2_chroma_h right ? [10:31:55 CEST] <nevcairiel> yea [10:34:02 CEST] <mateo`> thanks a lot for your help guys. [10:34:45 CEST] <wm4> <mateo`> my goal is relatively simple i would like to be able to copy an AVFrame to disk (or a chunk of memory containing the whole frame) <- rawvideo encoder/muxer? [10:38:04 CEST] <mateo`> wm4: I want to avoid to go that route since i only want to perform a color conversion using swscale to which I provide an AVFrame allocated on my side [11:55:23 CEST] <cone-987> ffmpeg 03Paul B Mahol 07master:b74ebd09c719: avfilter/vf_lut: >8 bit depth planar yuv support [12:48:54 CEST] <durandal_1707> wtf nv16 without any support in swscale [13:18:48 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:8f4cfda9721f: avutil: add missing bswap include [13:18:49 CEST] <cone-987> ffmpeg 03Rostislav Pehlivanov 07master:8e607c747eca: aacpsy: use a different metric for the spread of a band [13:30:06 CEST] <cone-987> ffmpeg 03Rostislav Pehlivanov 07master:57848ef3c6a1: aaccoder: fix M/S coding [14:54:56 CEST] <kierank> durandal_1707: yes [14:55:00 CEST] <kierank> I wrote that [14:55:07 CEST] <kierank> because x264 accepts interleaved chroma [16:20:18 CEST] <cone-987> ffmpeg 03Rostislav Pehlivanov 07master:9f4f5787046d: aacenc: reset marked IS and M/S bands upon frame encoding [17:07:17 CEST] <asdf293472938748> Hi. I have some MPEG-4 Part 2 encoder hardware that I'm trying to decode with ffmpeg. It seems to be producing VLCs which the decoder doesn't recognise. Wondered if there was anyone here familiar with MPEG-4 whom I could ask some advice from? [18:16:20 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:d554715f673d: avcodec/jpeg2000dec: Fix decoding of subsampled multi tile images [18:32:52 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:9f653e6d369b: avcodec/j2kenc: Support user specified tile dimensions [18:34:39 CEST] <cone-987> ffmpeg 03Paul B Mahol 07master:d3836b603e7e: avfilter/af_astats: export metadata [18:34:40 CEST] <cone-987> ffmpeg 03Paul B Mahol 07master:c0d676f9773a: avfilter/af_astats: implement recalculation of stats after each X frames [19:03:08 CEST] <cone-987> ffmpeg 03Paul B Mahol 07master:94cfb6db7d6c: avcodec/shorten: use init_get_bits8() [19:03:09 CEST] <cone-987> ffmpeg 03Paul B Mahol 07master:866404df2d46: avfilter/vf_lut: fix oversight [19:59:47 CEST] <durandal_1707> michaelni: just webserver or others too? [20:00:57 CEST] <michaelni> others too, except fate server (which is seperate) and git which is on videolan [20:13:17 CEST] <Compn> arpi hosted for a few years, dang server switch time. [20:15:19 CEST] <Compn> humm why didnt michael's mail hit rtmpdump list [20:15:23 CEST] <Compn> not in mod queue either [20:18:39 CEST] <kierank> michaelni: how many RU and how much bandwidth? [20:18:49 CEST] <kierank> or do you need hw too [20:19:35 CEST] <j-b> michaelni: you're using mailman, right? [20:19:48 CEST] <michaelni> j-b, yes [20:20:00 CEST] <michaelni> kierank, we need new hw too if we have to move [20:20:07 CEST] <kierank> specs? [20:21:02 CEST] <michaelni> mom, i forgot some stuff must write a 2nd mail, ill also list what i know about specs [20:21:32 CEST] <kierank> I have a LOT of servers [20:21:35 CEST] <kierank> that are doing nothing [20:21:36 CEST] <kierank> like 50 [20:21:42 CEST] <kierank> but they are old and probably waste electricity [20:21:52 CEST] <Compn> we use a lot of bandwidth [20:22:08 CEST] <kierank> got 250meg here [20:24:17 CEST] <Compn> dont remember how much data / month though. i've been out of the loop [20:24:30 CEST] <michaelni> i need to subscribe to rtmpdump list it rejects my mails [20:26:51 CEST] <Compn> everyone is also modded on that list so will have to approve mails manually. [20:35:14 CEST] <nevcairiel> Daemon404: can we remove -strict -1 requirement for 422/444 x265 encoding by now? [20:39:45 CEST] <kierank> ohohohoho [20:39:47 CEST] <kierank> 300 cad [20:39:48 CEST] <kierank> lool [20:44:28 CEST] <J_Darnley> Hell, I'd spend 290 of that 300 to buy a computer that I can test on just to earn the remaining 10 [20:44:50 CEST] <nevcairiel> i have all the hardware, but why would i care [20:47:15 CEST] <J_Darnley> it is money [20:50:51 CEST] <nevcairiel> not enough to care [20:53:59 CEST] <wm4> I once accidentally got a 100¬ ffmpeg bounty for implementing something which someone else happened to want [20:54:09 CEST] <wm4> but I didn't claim it because it sounds like a pain to do so [21:00:50 CEST] <durandal_1707> and you prefer to be anonymous [21:12:23 CEST] <Mista_D> Upped to $500 CAD, just encoding and VPP scaling at least. [21:18:34 CEST] <cone-987> ffmpeg 03Vittorio Giovara 07master:832129431fd5: lavu: Add version information for av_version_info() [21:18:35 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:d563e13a7c7d: Merge commit '832129431fd5c693b12c32a1563944c631feaf36' [21:41:54 CEST] <wm4> what is AV_PIX_FMT_AYUV16? [21:42:09 CEST] <wm4> is it what I think it is? [21:42:22 CEST] <wm4> and why does it have 16 in the name? [21:42:52 CEST] <wm4> oh, so it's packed 4:4:4 yuv with alpha and 16 bits per component [21:42:57 CEST] <wm4> is there a 8 bit variant? [21:42:59 CEST] <wm4> durandal_1707: ping [21:44:03 CEST] <durandal_1707> there is also 8 bit variant..... [21:44:22 CEST] <wm4> what's its name? [21:44:43 CEST] <durandal_1707> ayuv [21:46:02 CEST] <durandal_1707> maybe instead of 16 it should be 64 [21:47:20 CEST] <wm4> there's no ayuv [21:47:33 CEST] <wm4> is it a pending patch? [21:48:26 CEST] <cone-987> ffmpeg 03Vittorio Giovara 07master:7a5902c556d8: lavc: Disable deprectation warnings coming from options table [21:48:27 CEST] <cone-987> ffmpeg 03Vittorio Giovara 07master:f046c3b5ac36: lavc: Move deprecation warning disabling to files including the table [21:48:28 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:3b03186d5679: Merge commit 'f046c3b5ac36848cce824b008e0347c621523041' [21:48:39 CEST] <durandal_1707> what? this one? [21:51:35 CEST] <wm4> 8 bit 4:4:4 packed [21:53:25 CEST] <durandal_1707> no [21:54:18 CEST] <durandal_1707> maybe yuva ones how are they put in tiffs [23:18:49 CEST] <cone-987> ffmpeg 03Michael Niedermayer 07master:ab80d3fb3a75: swscale/output: fix null pointer dereference in yuv2ya8_2_c() [23:44:07 CEST] <philipl> durandal_1707: the vdpau 444 formats are 8bit yuva and ayuv (or was it avuy?) [23:44:35 CEST] <philipl> no vdpau implementation supports them yet. [23:50:30 CEST] <durandal_1707> dunno, this one is for "apple" [00:00:00 CEST] --- Sat Jul 4 2015

1 0

ffmpeg.log.20150703
by burek 04 Jul '15

04 Jul '15

[00:31:47 CEST] <janicez> h [00:47:45 CEST] <Nolski> Hey, for some reason when I try to use the overlay filter I get 'Option 'overlay_w' not found' [00:48:01 CEST] <Nolski> but that doesn't make sense. overlay_w looks to be an option in the docs [00:51:01 CEST] <Nolski> Same happens with using 'w' as an option [00:51:27 CEST] <c_14> That's not an option, that's a variable. iirc [00:51:39 CEST] <c_14> yep [00:51:47 CEST] <c_14> It's a variable that you can use in the x and y expressions [00:52:17 CEST] <Nolski> Ah, so if I want to change the width of the video, I need to do that before hand? [00:52:17 CEST] <c_14> If you want to control the size of the overlay, scale/crop/pad it before the filtler. [00:52:22 CEST] <c_14> *filter [00:52:28 CEST] <Nolski> cool cool. That makes sense. [00:52:44 CEST] <Nolski> Thanks c_14 I honestly don't know what I'd do without ya haha [03:45:15 CEST] <polyphagia> what is gop_size? [03:45:19 CEST] <polyphagia> for -g [03:45:41 CEST] <c_14> https://en.wikipedia.org/wiki/Group_of_pictures [03:46:37 CEST] <polyphagia> does it have a default value? [03:46:45 CEST] <c_14> Depends on the encoder [03:47:19 CEST] <polyphagia> oh alright this makes sense, thanks c_14 [03:55:59 CEST] <DHE> aka the keyframe interval [06:05:33 CEST] <MarkCourtenay> Does anyone know of a good way to run a playlist through ffmpeg without concatenating a bunch of files into one? [06:14:12 CEST] <relaxed> MarkCourtenay: what are you trying to do? [06:16:01 CEST] <Darkfire113> Hey guys! I just recently got ahold of ffmpeg and I notice that just using -i input.mp3 output.m4a works, but I was wondering if it's possible to get an even smaller file. Is there anything I can do or..? I'm very new to codecs and ffmpeg as an fyi. [06:18:42 CEST] <relaxed> ffmpeg -i input.mp3 -map 0:a -b:a 64k output.m4a [06:20:09 CEST] <Darkfire113> Thanks! Though may you explain it? I'd like to properly learn instead of just use. [06:21:52 CEST] <relaxed> map only the audio to output (exclude cover art if there is any), and set the bitrate to 64k [06:22:51 CEST] <relaxed> Darkfire113: https://trac.ffmpeg.org/wiki and https://ffmpeg.org/documentation.html [06:24:02 CEST] <Darkfire113> Is it possible to keep the covert art? I'm willing to lose a bit of space for it. Also.. Do you have any suggestions on were to start off? [06:24:13 CEST] <Darkfire113> Regarding the wiki/docs [06:25:59 CEST] <MarkCourtenay> relaxed: I'm trying to essentially simulate a TV channel...play one episode of a show (ie, 1 file), then another, then another in a specified order. Right now, I can concat all the files together into one big file but I am getting errors in playback. And that is not ideal in terms of HDD space. [06:26:55 CEST] <relaxed> Darkfire113: what does this return, ffmpeg -codecs 2>&1| awk '/DEA.*aac/' [06:28:01 CEST] <Darkfire113> awk isn't recognized, running Win7. [06:28:27 CEST] <relaxed> oh, nevermind then [06:28:54 CEST] <Darkfire113> Ehe, trust me I'd be running Debian if all my games supported it. [06:28:59 CEST] <relaxed> remove -map 0:a if you want to keep the cover art [06:29:42 CEST] <Darkfire113> Alright, thanks. Now to read lots of doc and wiki's.. [06:29:53 CEST] <relaxed> MarkCourtenay: why not create a playlist for your media player? [06:30:14 CEST] <MarkCourtenay> relaxed: I'm trying to run this from server command line [06:30:28 CEST] <MarkCourtenay> to RTMP output so I can stream on my site [06:34:21 CEST] <relaxed> write a for loop that sends each video through rtmp [06:38:33 CEST] <MarkCourtenay> OK...would that leave a gap between each file in playback? [06:38:58 CEST] <relaxed> probably [06:39:00 CEST] <MarkCourtenay> Or would the shows play one immediately after the other? [06:39:09 CEST] <MarkCourtenay> hmm [06:39:33 CEST] <relaxed> give it a shot with some samples [06:41:20 CEST] <relaxed> if they're all encoded the same way you could pipe each one to an ffmpeg instance that's sending the rtmp stream [06:44:57 CEST] <MarkCourtenay> relaxed: Thanks! I [06:45:01 CEST] <MarkCourtenay> I'll give that a shot. [06:45:33 CEST] <relaxed> which is basically concat :) [06:51:12 CEST] <prelude2004c> hey guys.. quick question.. i don't know if this is a dev problem or something.. i am trying to ffmpeg -i m3u8 location and output to rtmp [06:51:21 CEST] <prelude2004c> when playing back the stream it stuttering and cutting out like crazy [06:51:32 CEST] <prelude2004c> when i input the m3u8 straem to vlc directly it goes on quick and right away [06:51:36 CEST] <prelude2004c> not sure why [06:51:45 CEST] <prelude2004c> ffmpeg doesn't like the m3u8 for some reason? [06:51:54 CEST] <prelude2004c> flc seems to be ok with it [06:55:44 CEST] <thomedy> okay so i need to know when i am adjusting dimensions i am best servd by what consistent dimensions [06:56:57 CEST] <thomedy> so for example the user inputs a phone video.... that may be tall not wide [06:57:06 CEST] <thomedy> is it custom to adjust that or ckeep dimensions [06:57:17 CEST] <thomedy> and if so what is the common dimensions to adjust im thinking big to shrink [09:48:53 CEST] <bertieb> Huh. Why might ffmpeg get stuck generating one frame (incrementing drop number) when invoking overlay filter as specified by the 0.3 red example in 'filters' (http://www.ffmpeg.org/ffmpeg-filters.html#Examples-55) ? [09:50:38 CEST] <bertieb> Or does the filter just keep producing frames until told not to? [09:55:03 CEST] <bertieb> Yep, I missed the "repeatlast" option, my bad [12:34:47 CEST] <rrooij> Hello, I was wondering if there are any plans to add a demuxer for the "smooth streaming" ISML playlists? ffmpeg is already capable of muxing them, but it is not capable of demuxing them. [12:34:53 CEST] <rrooij> Example file: http://mediadl.microsoft.com/mediadl/iisnet/smoothmedia/Experience/BigBuckB… [12:35:19 CEST] <rrooij> It would enable people to move away from Silverlight and use other media players to play back streaming content. [12:35:28 CEST] <rrooij> (Except the ones protected by DRM, of course) [12:36:09 CEST] <BtbN> Is this another DASH/HLS thing? [12:36:31 CEST] <sfan5> "It would enable people to move away from Silverlight" [12:36:36 CEST] <sfan5> not really [12:36:47 CEST] <sfan5> the media server would still be running something silverlight related [12:37:14 CEST] <rrooij> BtbN: It is the same concept, but another format, as far as I know. [12:37:25 CEST] <rrooij> sfan5: Yes, that's true. I was talking about the clients. [12:40:47 CEST] <rrooij> Of course, Silverlight is already deprecated and servers are already moving to DASH/HLS. So I don't know if it is worth the effort at all. Probably not, but I was curious if there were some people working on it. [15:33:52 CEST] <pothibo> Does ffmpeg allow me to load a stream and trigger recording at moment's notice? The goal would be to be connected to a stream and a user could start/stop recording for later replay. [15:37:29 CEST] <pothibo> Basically, I would like to have the library loaded in memory and streaming configuration already parsed so when a user press start, it could record it right away [17:42:14 CEST] <pothibo> I'm trying to get ffmpeg to capture and record a multicast stream (UDP) https://gist.github.com/anonymous/af5c2228f13432ef08a2 ANy help? [17:43:34 CEST] <DHE> no @ required [17:44:09 CEST] <pothibo> I know it was a quick copy, the problem doesn't lie there anyway [17:44:27 CEST] <DHE> a few errors at the start is actually kinda normal since joining the feed will not start with a keyframe [17:45:39 CEST] <pothibo> ok [17:45:46 CEST] <pothibo> so I can wait and it will resolve itself? [17:46:38 CEST] <pothibo> oh yeah, it works. good [17:46:48 CEST] <pothibo> just thought they were critical errors ;( [17:47:31 CEST] <pothibo> Can I tell ffmpeg to start recording when it catches a keyframe? [19:03:02 CEST] <hwd> hello, when using libavcodec avfft (in my case rdft -- DFT_R2C), is the calculated rdft post-scaled , or do I need to do that (and by post scaling meaning something like real and imag. /= 2 * fft_size) !? [19:33:23 CEST] <bofh> hi there! What is the best way to merge multiple audio streams with no overamplifying? I use "pan=stereo:c0=c0+c2+c4+c6:c1=c1+c3+c5" and that results in "cracking" sound effect [19:38:58 CEST] <durandal_1707> you can't just sum up values [19:40:02 CEST] <durandal_1707> pan doesn't clip values [19:40:30 CEST] <bofh> ok, so what do I do then? [19:45:07 CEST] <bofh> ffmpeg -y -i ad5e9655f912bb49c969ba25.mp3 -i 9d5a96551f88fc9ba97f095a.mp3 -i 9d5a96551f88fc9ba97f095a.mp3 -filter_complex [0]aformat=sample_fmts=fltp:sample_rates=44100:channel_layouts=stereo,volume='if(between(t,2.20,159.69),0.00000 * (t - 2.20) + 1.00,1)':eval=frame,atrim=2.20:12.20,apad[0:out];[1]aformat=sample_fmts=fltp:sample_rates=44100:channel_layouts=stereo,volume='if(between(t,0.00,8.57),0.00000 * (t [19:45:13 CEST] <bofh> - 0.00) + 1.00,1)':eval=frame,atrim=0.00:6.40,adelay=3600|3600|3600|3600,apad[1:out];[2]aformat=sample_fmts=fltp:sample_rates=44100:channel_layouts=stereo,volume='if(between(t,0.00,8.57),0.00000 * (t - 0.00) + 1.00,1)':eval=frame,atrim=0.00:8.57,apad[2:out];[0:out][1:out][2:out]amerge=inputs=3,pan=stereo:c0=c0+c2+c4:c1=c1+c3+c5,atrim=duration=10.00 -shortest -ac 2 -vn track.mp3 [19:45:19 CEST] <bofh> oh, sorry [19:45:28 CEST] <bofh> http://pastie.org/private/jbo1ncwgckxw6z7dgmqeqw [20:33:28 CEST] <durandal_1707> bofh: use clip in expressions of channels [20:36:02 CEST] <bofh> durandal_1707: can you please elaborate? [20:37:52 CEST] <durandal_1707> clip(c0+c1...., -1, 1) [20:38:21 CEST] <bofh> oh, coo, didn't know about that [20:38:54 CEST] <bofh> I added this (from examples) Ò compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2 [20:38:58 CEST] <bofh> seems to works fine, too [20:41:53 CEST] <durandal_1707> there is another maybe better example after that one [20:42:59 CEST] <durandal_1707> but what you are trying to do? [20:47:39 CEST] <spl33n_> hello all i want to convert video to gif but output gif file has a very low quality. i use this command, ffmpeg -i input -pix_fmt rgb24 output.gif , how can i have dif file has a hight quality ? [20:49:28 CEST] <kepstin-laptop> if you want high quality, don't use a gif ;) that said, you'd want to look into the two-pass mode that precalculates a palette, and play with the dithering options. [20:50:37 CEST] <durandal_1707> spl33n_: look at palettegen and paletteuse filter [20:53:27 CEST] <bofh> durandal_1707: I just need to overlay several audiotracks [20:57:24 CEST] <spl33n_> durandal_1707: ok i look this, i come for feedback [22:22:31 CEST] <StatiQ> ii found this site: https://github.com/JayH5/android-ffmpeg-cmdline but my andraied is 2.2 not 4 to use the interface. the ffmpeg cmd instead is working on here and now i have reall aac music :D [22:23:33 CEST] <StatiQ> very interesting. static unix. it will work in some another arch? soon i will try. [22:25:22 CEST] <chungy> android 2.2? damn [22:25:37 CEST] <chungy> static binaries generally only work on the OS and arch they're built for. [22:46:18 CEST] <Q3> Hello, I was wondering if "ffmpeg -i test.dtshd -c:a flac out.flac" was using the lossless part or if I was doing lossy -> lossless? [22:48:41 CEST] <chungy> "-c:a flac" is unnecessary for an out.flac file (though it does no harm). If your input has multiple streams, it will be using the first one; use -map to select a different one. ffprobe can tell you what streams are in the file [22:49:34 CEST] <Q3> I mean one stream of dts-hd ma, or dts-ma, but they have a lossless core and a lossy fallback; I just don't know if libdca can use the lossless core. [22:51:06 CEST] <durandal_1707> force libdcadec decoder [22:51:33 CEST] <Q3> Is this mandatory? How do I know when the core of the fallback is used? [22:52:24 CEST] <durandal_1707> native dca xll is not bitexact [22:52:43 CEST] <StatiQ> chungy: i have two versions of ffmpeg in static bin now. one is from 2012 without aaclib and now this version 2013 from this link. and one other more in threaded emulation with shared libraries like unix, :D [22:53:14 CEST] <durandal_1707> paste debug output to see which decoder is used [22:53:17 CEST] <StatiQ> 3 differente versions and none compiled in my os [22:54:41 CEST] <StatiQ> unix is unix. [22:55:06 CEST] <kepstin-laptop> StatiQ, however, linux isn't really unix at all ;) [22:55:07 CEST] <Q3> Input #0, mpegts, from '00001.m2ts': [22:55:07 CEST] <Q3> Duration: 00:29:10.75, start: 6009.000000, bitrate: 44499 kb/s [22:55:08 CEST] <Q3> Program 1 [22:55:08 CEST] <Q3> Stream #0:0[0x1011]: Video: h264 (High) (HDMV / 0x564D4448), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], 2 [22:55:10 CEST] <Q3> Stream #0:1[0x1100]: Audio: pcm_bluray (HDMV / 0x564D4448), 48000 Hz, stereo, s16, 1536 kb/s [22:55:13 CEST] <Q3> Stream #0:2[0x1101]: Audio: dts (DTS-HD MA) ([134][0][0][0] / 0x0086), 48000 Hz, 5.1(side), fltp [22:55:16 CEST] <Q3> Stream #0:3[0x1102]: Audio: dts (DTS) ([130][0][0][0] / 0x0082), 48000 Hz, stereo, fltp, 448 kb/s [22:55:19 CEST] <Q3> Stream #0:4[0x1200]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090) [22:55:21 CEST] <Q3> Stream #0:5[0x1201]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080 [22:55:24 CEST] <Q3> [matroska @ 0x7f912705f940] Codec for stream 0 does not use global headers but container format requires [22:55:27 CEST] <Q3> Output #0, matroska, to 'Rurouni Kenshin - Meiji Kenkaku Romantan - Tsuiokuhen/02.mkv': [22:55:30 CEST] <Q3> Metadata: [22:55:33 CEST] <Q3> encoder : Lavf56.36.100 [22:55:34 CEST] <kepstin-laptop> Q3, *pastebin*, don't flood the channel :( [22:55:35 CEST] <Q3> Stream #0:0: Video: h264 (H264 / 0x34363248), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 23.98 fp [22:55:39 CEST] <Q3> Stream #0:1: Audio: flac ([172][241][0][0] / 0xF1AC), 48000 Hz, 5.1(side), s16, 128 kb/s [22:55:42 CEST] <Q3> Metadata: [22:55:44 CEST] <Q3> encoder : Lavc56.41.100 flac [22:55:47 CEST] <Q3> Stream mapping: [22:55:49 CEST] <Q3> Stream #0:0 -> #0:0 (copy) [22:55:52 CEST] <Q3> Stream #0:2 -> #0:1 (dts (dca) -> flac (native)) [22:55:54 CEST] <Q3> Sorry, but i was aked to paste, how should I do it? [22:56:18 CEST] <durandal_1707> See native decoder is used [22:56:54 CEST] <Q3> So that doesn't work? [22:57:07 CEST] <StatiQ> kepstin-laptop: i know. but if i do something besed on unix , the only different thing during compilation of binaries and "archs" is the directory tree. mathematics things are only one [22:57:56 CEST] <durandal_1707> Compile ffmpeg with libdca enabled and dca decoder disabled [22:58:15 CEST] <StatiQ> i am not talking about this aberration android. [22:59:15 CEST] <Q3> durandal: Well, if libdcadec can't do it, it won't even be useful. I tried the VLC website, but it wasn't talking about the features. [23:01:40 CEST] <kepstin-laptop> Q3, according to the summary on https://github.com/foo86/dcadec the external libdcadec supports lossless decoding. [23:02:25 CEST] <Q3> Thanks a lot [23:03:53 CEST] <StatiQ> kepstin-laptop: try to run this: https://github.com/JayH5/android-ffmpeg-cmdline extract the raw ffmpeg only and put in some executable directory from your phone, set chmod to 755 an callit like linux in terminal emulator. [23:04:42 CEST] <kepstin-laptop> StatiQ, sure, that's a linux binary compiled for arm arch, so I assume it would work on most android phones [23:05:04 CEST] <kepstin-laptop> now, if I had one of the (relatively rare) intel based phones, it wouldn't work ;) [23:05:25 CEST] <StatiQ> kepstin-laptop: try first he says android 4.1 but i am on 2.2 [23:06:18 CEST] <kepstin-laptop> the fancy android gui frontend probably needs android 4.1 for some libraries. But the ffmpeg binary doesn't really care about the android version... [23:06:53 CEST] <StatiQ> kepstin-laptop: i know! is where unix is unix [23:06:58 CEST] <StatiQ> :D [23:08:34 CEST] <kepstin-laptop> ... [23:08:51 CEST] <StatiQ> kepstin-laptop: i am using this binary with shell scripts like zenity (SManager) with all conversion rules in a widget folder or nenu shortcuts [23:10:08 CEST] <StatiQ> androind needs only two things to be acceptable: zenity (xdialog) and XUL nothing more [23:10:23 CEST] <StatiQ> .. [23:11:29 CEST] <kepstin-laptop> XUL? I'm kind of wondering why android needs mozilla's xml user interface markup language for anything ;) [23:12:51 CEST] <StatiQ> kepstin-laptop: is because XUL is REAL source opened and we are able to create and edit things without re-compilation/installation. [23:13:15 CEST] <durandal_1707> Q3: it can [23:15:14 CEST] <StatiQ> kepstin-laptop: i have in XUL an interface app (created by me) for convert media using static ffmpeg. i have in extension for mozilla browser, and in app format to use ffmpeg without terminal dependency [23:16:13 CEST] <StatiQ> only because this. XUL is REAL OPENED and simply editable [23:16:56 CEST] <StatiQ> Zenity is similar to xul [23:17:08 CEST] <StatiQ> but only for unix... [23:18:57 CEST] <kepstin-laptop> not actually true, you can use zenity on linux, windows, mac os, as well as unix ;) [23:22:19 CEST] <StatiQ> kepstin-laptop: >>Actually<< ... [23:23:04 CEST] <StatiQ> zenity is the .net framework of linux. [23:23:25 CEST] <StatiQ> i will use. net to create windows apps [23:23:47 CEST] <StatiQ> i prefer XUL because it runs in all OS [23:25:27 CEST] <chungy> zenity just lets you create gtk dialgos [23:25:47 CEST] <chungy> but yeah, why does android need any of that? it has a perfectly fine UI toolkit on its own [23:26:46 CEST] <kepstin-laptop> heh, apparently the firefox folks are working towards reimplementing the ui in html instead of xul. [23:27:13 CEST] <kepstin-laptop> they already try to get extension devs to use apis for ui controls rather than include xul stuff. [23:28:34 CEST] <StatiQ> kepstin-laptop: is serious. i have a panel to manipulate all media type from simple. amv to aac+ is something like this: https://addons.mozilla.org/pt-br/firefox/addon/video-downloadhelper/ [23:30:17 CEST] <StatiQ> xul is a Jit compiler we don't need to compile nothing more than the static-unix-binaries/and libraries. ah they have compiler too [23:30:37 CEST] <StatiQ> we don't nedd only gcc [23:31:04 CEST] <StatiQ> gcc is the error of "archs" [23:32:21 CEST] <StatiQ> xul is something similar to java port to unix on phones but is horrible [23:32:35 CEST] <StatiQ> java is horribility [23:32:54 CEST] <StatiQ> javascript is the xul engine [23:33:21 CEST] <kepstin-laptop> well, xul is only useful if you have xulrunner, which is a rather gigantic C++ project that includes a javascript engine (spidermonkey), layout & rendering engine (gecko), image libraries, xml and html libraries, and must be compiled with a c++ compiler like gcc's to native code. [23:35:10 CEST] <StatiQ> kepstin-laptop: not so giant than these horrible MB occupied by Navigators in phone. my xulrunner have 15MB and my ffmpeg 10MB and my app 35 MB [23:36:12 CEST] <StatiQ> xulrunner in andraid is a menace to gogle. this is why they don't have [23:36:37 CEST] <kepstin-laptop> StatiQ, technically, you can run xulrunner on android, https://www.mozilla.org/en-US/firefox/android/ ;) [23:37:30 CEST] <StatiQ> kepstin-laptop: i know. i tried but i have to get 512MB OF RaM to get freedom. [23:37:34 CEST] <kepstin-laptop> (what's really funny is that the browser builds of xulrunner can, in some cases, actually link in ffmpeg as native code - indirectly, via gstreamer, but still) [23:38:11 CEST] <kepstin-laptop> ("via gstreamer" means it's often actually libav rather than ffmpeg, but whatever) [23:38:22 CEST] <StatiQ> i don't have 2000 dollar today to buy a ne phone [23:38:54 CEST] <StatiQ> kepstin-laptop: firefox is for 512MB of RaN [23:39:34 CEST] <StatiQ> i will not pay stupid price in a ridiculous phone for me these apparatus have not value [23:40:10 CEST] <StatiQ> samsung is a lie motorola is another worst lie. etc.. [23:41:02 CEST] <StatiQ> xulrunner pure without firefox or apps. we don't have. [23:41:24 CEST] <StatiQ> for phones like pc xulrunner haves [23:41:58 CEST] <kepstin-laptop> it's actually kind of interesting that in firefox os, they explicitly chose not to expose xul to phone apps, but rather require them to use html. [23:42:30 CEST] <StatiQ> Xhtml [23:42:49 CEST] <StatiQ> :D [23:43:01 CEST] <kepstin-laptop> nah, the xml serialization of html has fallen out of favour. it added complexity with really no gain. [23:43:58 CEST] <StatiQ> kepstin-laptop: ok. correct [23:45:40 CEST] <StatiQ> kepstin-laptop: i tried an example app using firefox to see if xulrunner is frozen on andraid and i seen that works. any suggestion without firefox in phone? [23:46:42 CEST] <StatiQ> allright! they won't will construct xulrunner for phones. i don't know why. [23:46:57 CEST] <StatiQ> won't or will? [23:49:22 CEST] <StatiQ> andraeid 2.2 armv6 Vfp modded firefox /28/29 version (highly slow). if was only the xulrunner it may works [23:50:16 CEST] <StatiQ> if i found xulrunner for phones i will construct an interface for ffmpeg like i don on pc in my pendrive. [00:00:00 CEST] --- Sat Jul 4 2015

1 0

ffmpeg-devel.log.20150702
by burek 03 Jul '15

03 Jul '15

[00:00:02 CEST] Action: Compn considers reinstalling win2k [00:00:15 CEST] <wm4> a liability for those companies [00:00:16 CEST] <JEEBsv> ugh, that doesn't even have aligned malloc, IIRC [00:00:29 CEST] <JEEBsv> I think the nice stuff in msvcrt came in... xp sp2? [00:01:58 CEST] <kierank> not really [00:02:11 CEST] <kierank> if they have programs that for whatever reason only run on xp [00:02:19 CEST] Action: kierank sadly has customers with that [00:02:27 CEST] <wm4> so what, many companies have that with DOS [00:02:31 CEST] <wm4> (or worse) [00:02:45 CEST] <kierank> yes but xp is much larger [00:02:49 CEST] <kierank> orders of magnitude [00:03:07 CEST] <wm4> so you have customers who need to run bleeding edge ffmpeg on XP machines, or something [00:03:07 CEST] <Compn> millions of machines [00:03:36 CEST] <nevcairiel> kierank: any corporate environment that still uses xp is dumb [00:03:40 CEST] <nevcairiel> its a security nightmare now [00:03:41 CEST] <Compn> your idea to drop xp also affects ffmpeg libs, which affects vlc and j-b and all other tools that use lavc [00:03:46 CEST] <kierank> wm4: well it's more vlc [00:03:48 CEST] <kierank> yes exactly [00:04:13 CEST] <Compn> kierank : i dont think he cares, so why are we arguing with him ? [00:04:13 CEST] <wm4> j-b: do you care about XP support? [00:04:14 CEST] <Compn> heh [00:04:23 CEST] <nevcairiel> I think many downstreams will be happy if we give them an excuse to drop xp support =p [00:04:43 CEST] <wm4> Compn: I don'T care about you, but I care a lot about getting rid of XP [00:05:57 CEST] <Compn> your precious w7 is obsolete as well. every time i point this out to w7 users they shrug it off [00:06:09 CEST] <wm4> does anything you say ever make sense [00:06:11 CEST] <Compn> you are going to have to make a choice about new windows, very soon... [00:06:12 CEST] <Compn> bah [00:09:57 CEST] <j-b> wm4: yes, I do. [00:10:24 CEST] <kierank> BURN [00:10:28 CEST] <j-b> wm4: but I wouldn't care if I had no hw threading on XP [00:11:35 CEST] <wm4> whatever that means [00:11:50 CEST] <wm4> compiling vlc against pth? [00:12:00 CEST] <Compn> j-b : some devs dont care about vlc, but i care about vlc, i want a list of things vlc wants from us. like on a wiki. :) [00:13:26 CEST] <j-b> wm4: all winpthread implems are shit, so far [00:14:22 CEST] <nevcairiel> w32thread in ffmpeg/libav isnt too bad, if you discount the xp compat layer, but oif course its just minimal enough for the things ffmpeg needs [00:14:53 CEST] <wm4> j-b: because of win xp [00:14:58 CEST] <j-b> wm4: no [00:15:06 CEST] <Compn> i thought there were native threads .... hmm [00:15:10 CEST] <j-b> wm4: because most of them can't read the fucking documentation [00:15:14 CEST] <j-b> because many require init [00:15:20 CEST] <j-b> like ptheadGC2 [00:15:42 CEST] <j-b> because some expose other POSIX functions as __inline at the same time (see mingw64) [00:15:49 CEST] <j-b> and fuck up the configures + builds [00:16:28 CEST] <nevcairiel> pthread-w32 (ie. pthreadGC2) is one of the least terrible at least, imho, the one mingw-w64 build is worse [00:17:42 CEST] <nevcairiel> but ffmpeg is the only pthread thing i frequently build/use, and it doesn't require either, which is better for me [00:22:35 CEST] <jamrial> pthreads on win32 seems pointless to me when the native critical section stuff maps almost perfectly [00:24:22 CEST] <nevcairiel> a small subset maps perfectly, but some of the more advanced pthreads stuff gets complicated, but luckily ffmpeg doesnt use that [00:41:40 CEST] <cone-394> ffmpeg 03Janne Grunau 07master:a31c4b2cbef9: fate-g2m3: disable the audio stream [00:41:40 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:03b2b40fd7f0: Merge commit 'a31c4b2cbef9aee15910fc3df52519aef46760de' [04:58:15 CEST] <cone-575> ffmpeg 03rogerdpack 07master:a1c03b9d5882: ffmpeg_filter: log more information on failure to init simple filter graph [05:15:59 CEST] <cone-575> ffmpeg 03John Adlum 07master:c8eca438a953: avformat/asfdec_f: factor error checking out of main header parsing loop [05:16:00 CEST] <cone-575> ffmpeg 03John Adlum 07master:59fffefdb4dd: avformat/asfdec_f: Use dynamic allocation in asf_read_metadata() instead of a fixed size buffer [05:16:01 CEST] <cone-575> ffmpeg 03John Adlum 07master:72cad800164c: avformat/asfdec_f: Add ASFDataType, use named types for metadata [05:18:47 CEST] <jamrial> where are these patches coming from, and why is the old asfdec suddenly getting a lot of development only after a replacent showed up? [05:23:11 CEST] <jamrial> that last one especially. this all just reeks of NIH [05:31:02 CEST] <philipl> BtbN: The 'docs' imply that only lossless 444p works. [05:31:04 CEST] <philipl> That's what I tested. [05:34:08 CEST] <philipl> BtbN: No luck on identical frames though. Looks like some colourspace difference. odd. [05:55:11 CEST] <philipl> ah, got it. Needed to make sure my source was really yuv444p [05:55:26 CEST] <philipl> So yeah, it works. [06:04:55 CEST] <philipl> Hmm, and while the nvenc packed 444 was a false alarm, in vdpau, the 444 output mode is definitely packed yuva. Fun times. [08:24:41 CEST] <durandal_170> michaelni: from where those asf_f patches are coming? [08:46:58 CEST] <jamrial> michaelni: for that matter, 59fffefd added an av_malloc inside a loop without an av_free [08:47:04 CEST] <jamrial> asan and valgrind are of course complaining [09:50:42 CEST] <cone-448> ffmpeg 03Michael Niedermayer 07master:1f69b7baa10c: avformat/asfdec_f: Fix memleak [10:18:59 CEST] <durandal_170> are reviewers alive? [10:35:50 CEST] <BtbN> philipl, did you check if yuv444p works for the non-lossless modes? I remember it beeing useable excluviely for the lossless mode. So it'd might need a static init function, that sets the supported pixel formats based on the mode. [11:07:21 CEST] <cone-448> ffmpeg 03Stefano Sabatini 07master:8cbce1001db3: doc/muxers/segment: fix formatting of segment_list_type option [11:11:34 CEST] <Newbie> hi [11:12:37 CEST] <Guest92195> Helloo [11:12:50 CEST] <Guest92195> HIiiiiiiiiiiiiiiiiiiiiiiii [11:13:02 CEST] <Guest92195> help me please!!!!!!!!!!!!!!!! [11:13:14 CEST] <phh> no [11:13:35 CEST] <Guest92195> hi phh [11:13:36 CEST] <phh> well, anyway, noone can [11:13:48 CEST] <phh> since noone knows your problem [11:14:03 CEST] <Guest92195> i have problem when i capture Screen [11:14:11 CEST] <wm4> Guest92195: go to #ffmpeg [11:14:17 CEST] <wm4> your question is offtopic here [11:15:02 CEST] <Guest92195> no no [11:15:07 CEST] <[-T-]> hey [11:15:08 CEST] <[-T-]> https://github.com/TheTroll/FFmpeg/commit/16787892da578c29cf2578a7af4a1eac8… [11:15:12 CEST] <Guest92195> I develop software using FFMPEG library [11:15:22 CEST] <ubitux> then it's offtopic [11:15:25 CEST] <[-T-]> it works nicely in tvh [11:16:09 CEST] <nevcairiel> Guest92195: this channel is for developing ffmpeg itself, not using ffmpeg, not developing software with ffmpeg [11:16:14 CEST] <wm4> [-T-]: that looks pretty clean [11:16:29 CEST] <Guest92195> Sorry ALL [11:16:44 CEST] <j-b> Why doing changes in qsvdec_h264.c [11:16:52 CEST] <j-b> in a commit for qsv_mpeg2? [11:17:08 CEST] <j-b> but, yes, that's what I remembered from Luca, and that was a sane patch [11:18:32 CEST] <wm4> j-b: yeah, and the changes in that file are wrong [11:19:50 CEST] <nevcairiel> thats what happens when some random guy picks old patches from someplace without really knowing wtf is going on [11:20:01 CEST] <nevcairiel> also, i dont understand why everyone obsesses a bout qsv decoders that much [11:20:10 CEST] <nevcairiel> its just a wrapper around vaapi or dxva2, depending on platform [11:20:51 CEST] <j-b> +10 [11:20:55 CEST] <nevcairiel> (and in my experience, not a bug free wrapper) [11:20:59 CEST] <[-T-]> f'wm4: well this one is pretty clean, but it's using a less clean patch [11:21:45 CEST] <[-T-]> nevcairiel: wtf are you saying ? [11:21:55 CEST] <[-T-]> first of all, calm down with your words [11:22:11 CEST] <[-T-]> i am not doing P/R [11:22:26 CEST] <[-T-]> I know you guys don't give a shit about QSV [11:22:44 CEST] <j-b> libyami [11:22:45 CEST] <j-b> lol [11:22:47 CEST] <nevcairiel> and neither should you, it just adds another layer in between ffmpeg and the hardware, and more layers mean more bugs [11:23:02 CEST] <wm4> <nevcairiel> its just a wrapper around vaapi or dxva2, depending on platform <- oh, so that's its dirty secret? [11:23:18 CEST] <j-b> well, in cases, it fallbacks to software [11:23:25 CEST] <nevcairiel> yeah it can also do that [11:23:26 CEST] <j-b> but I did not say it. [11:23:40 CEST] <nevcairiel> at least on windows it also has software decoders for all the codecs [11:23:44 CEST] <[-T-]> (and yes, you are right, there should not be changes in h264_qsv in this commit, it's just that I fixed a problem that impacted both qsv decoders...) [11:23:45 CEST] <nevcairiel> but you obviously dont want to use them [11:24:08 CEST] <[-T-]> 11:18 < f'wm4> j-b: yeah, and the changes in that file are wrong [11:24:13 CEST] <[-T-]> can you elaborate ? [11:24:34 CEST] <nevcairiel> "ret < 0" is the correct check, just if (ret) is wrong [11:24:36 CEST] <wm4> ret = AVERROR(ENOMEM); [11:24:39 CEST] <wm4> return AVERROR(ret); [11:24:42 CEST] <wm4> makes no sense [11:24:43 CEST] <[-T-]> yes sorry wm4 [11:24:45 CEST] <[-T-]> you are right [11:24:48 CEST] <j-b> it does make sense [11:24:51 CEST] <[-T-]> it's correct in the mpeg2 file though [11:24:59 CEST] <[-T-]> nevcairiel: no [11:25:00 CEST] <j-b> AVERROR(AVERROR(AVERROR(AVERROR(ENOMEM)))); [11:25:21 CEST] <[-T-]> because AVERROR(EGAIN) = 11 [11:25:23 CEST] <[-T-]> and 11 > 0 [11:25:30 CEST] <wm4> [-T-]: wrong [11:25:33 CEST] <[-T-]> it leads to a crash [11:25:40 CEST] <wm4> it's negative [11:25:46 CEST] <[-T-]> well .. not in my tests [11:25:50 CEST] <nevcairiel> all error codes are negative [11:25:51 CEST] <wm4> then your tests are broken [11:26:06 CEST] <[-T-]> let me double check that ... [11:28:00 CEST] <wm4> do you really think that at least 50% of all code in ffmpeg is wrong? [11:28:32 CEST] <nevcairiel> you get such problems if you do thinks like AVERROR(ret) [11:28:37 CEST] <nevcairiel> because it may turn it back positive [11:29:03 CEST] <[-T-]> what do you mean ? [11:29:12 CEST] <[-T-]> oh ok... [11:29:18 CEST] <[-T-]> you are sooooo funny [11:33:58 CEST] <[-T-]> ok it's always negative ... not sure what went wrong in my log last time... [11:34:02 CEST] <[-T-]> i ll fix this [11:43:04 CEST] <j-b> https://blogs.gnome.org/uraeus/2015/06/30/introducing-pulse-video/ [11:45:30 CEST] <[-T-]> a "better" one [11:45:30 CEST] <[-T-]> Add QSV MPEG2 decoding support [11:45:30 CEST] <[-T-]> Correct qsv_h264.c DecodeHeader when more Data is needed [11:45:30 CEST] <[-T-]> QSV Decoding is based on drocon11 ffmpeg fork (thanks!) [11:45:33 CEST] <[-T-]> oups [11:45:38 CEST] <[-T-]> https://github.com/TheTroll/FFmpeg/commit/18b3b8fdfd13c72469629840e3d29df67… [11:45:42 CEST] <[-T-]> this one :) [11:45:51 CEST] <[-T-]> again, it's not meant for PR [11:46:02 CEST] <[-T-]> but it may be usefull for some people [11:48:08 CEST] <__gb__> A way to import native { DXVA2, D3D11, VA-API } surface into an mfxSurfaceFrame or whatever it's called should be more appropriate to avoid MFX decoders [11:49:00 CEST] <__gb__> using the native functions would be much faster, as you'd probably also avoiding checking for e.g. NAL boundaries (start code) multiple times [11:49:45 CEST] <__gb__> in some situations, looking for the next start code is actually the most CPU intensive residual task when HW decode is involved... [11:52:21 CEST] <[-T-]> __gb__: why part of the layers is looking for the start codes ? [11:52:26 CEST] <[-T-]> is it done in SW ? [11:52:43 CEST] <nevcairiel> j-b: at least its only Video cameras, not output or some shit [11:55:56 CEST] <j-b> nevcairiel: for now... [11:56:08 CEST] <[-T-]> why only video cameras ? [12:00:11 CEST] <[-T-]> here is a "perf report" of a tvheadend transcoding from mpeg2_qsv 576i to h264_qsv 720p [12:00:14 CEST] <[-T-]> http://pastebin.com/FF3bjwLZ [12:16:15 CEST] <wm4> j-b: it would be awesome if video decoding APIs would be made privleged, and everything had to go through PulseVideo to use it! [12:16:44 CEST] <j-b> wm4: don't laugh [12:25:14 CEST] <durandal_170> move it into kernel [13:16:04 CEST] <Compn> [-T-] : you'll have to put up with nevcairiel , hes being a real pain as of late. [13:17:52 CEST] <nevcairiel> because I'm saying the truth? Go deal with it =p [13:18:06 CEST] <Compn> John Adlum (April 29, 1759 March 14, 1836) was a pioneering American viticulturalist who was the first to cultivate the Catawba grape. [13:18:30 CEST] <Compn> sour grapes. [13:21:58 CEST] <durandal_170> michaelni? [13:23:01 CEST] <michaelni> durandal_170, ? [13:23:08 CEST] <durandal_170> after Elvis is gone... [13:26:15 CEST] <durandal_170> michaelni: should we care for backward compatibility when converting find_rect and cover_rect to dualinput? [13:26:25 CEST] <[-T-]> nevcairiel: the truth is that instead of give me advices on how to improve what I did, you indirectly provocted/insulted me [13:26:38 CEST] <[-T-]> you must be a very nice guy irl.. [13:27:05 CEST] <nevcairiel> when i told you wahts wrong, you said "no, I'm right", so shrug [13:27:33 CEST] <[-T-]> is it a joke ?:) [13:27:35 CEST] <[-T-]> 11:19 <+f'nevcairiel> thats what happens when some random guy picks old patches from someplace without really knowing wtf is going on @f'sireta~ [13:27:41 CEST] <[-T-]> this is your first message [13:28:27 CEST] <nevcairiel> if you find that insulting, you should really not visit the internet [13:28:36 CEST] <[-T-]> ... [13:29:09 CEST] <[-T-]> I have nothing more to say :) [13:32:17 CEST] <__gb__> [-T-], looking for start codes is needed to extract NAL units and their related headers where you extract additional info to pass to the HW [13:32:32 CEST] <[-T-]> yep ok [13:32:50 CEST] <[-T-]> where is it currently done ? [13:32:57 CEST] <__gb__> unless there is a way to tell MSDK that you feed it with e.g. NAL units instead of whole access units, it would scan for start codes again there, imho [13:34:33 CEST] <__gb__> the point is, it's totally useless to supply QSV-based decoders, the better option is to work on interop from native hwaccel & qsv for e.g. encoding or other video processing [13:35:01 CEST] <__gb__> especially since, in Linux, you have to supply a VADisplay anyway, so, you could just be fine with ffmpeg vaapi decoders for instance [13:35:17 CEST] <nevcairiel> not only useless but also a new source for bugs. If you have the alternatives of: ffmpeg -> qsv -> vaapi -> hardware, or ffmpeg -> vaapi -> hardware ... the second is going to be a better choice [13:35:42 CEST] <__gb__> some additional helpers akin to newer vdpau's from courmisch would be useful [13:36:04 CEST] <michaelni> durandal_170, how hard/messy is backward compat? [13:36:18 CEST] <michaelni> if its easy, i think compatibility is always a good ide [13:36:19 CEST] <michaelni> a [13:38:53 CEST] <__gb__> nevcairiel, yes, and I have also mentioned the performance aspect wrt. scanning for start codes again at the MSDK level [13:38:54 CEST] <[-T-]> totally useless is maybe a bit exagerated, it's a 15%CPU gain on my machine [13:39:25 CEST] <nevcairiel> compared to software decoding? certainly. compared to vaapi decoding? likely not [13:39:25 CEST] <[-T-]> but I get your point [13:39:26 CEST] <Compn> [-T-] : i think he means useless vs using vaapi or vdpau [13:39:34 CEST] <[-T-]> oh ok [13:39:36 CEST] <[-T-]> sorry [13:39:41 CEST] <Compn> or dxva2 [13:39:46 CEST] <[-T-]> yep ok [13:39:47 CEST] <nevcairiel> the point is that qsv just uses vaapi internally [13:39:53 CEST] <nevcairiel> so you can cut out an extra layer in between [13:39:57 CEST] <nevcairiel> it only adds overhead [13:40:00 CEST] <[-T-]> yes ok [13:40:17 CEST] <Compn> nevcairiel : does it add stability or ease of use ? [13:40:59 CEST] <[-T-]> btw, I get the exact same kind of gain using VPDAU [13:41:21 CEST] <__gb__> in my older measurements, scanning for startcodes is number 1 function hotspot for the whole decoding processing with vaapi [13:41:26 CEST] <[-T-]> I mean, VDPAU and QSV decoding bring about the same gain over sw [13:41:55 CEST] <__gb__> and optimizing it, or removing a couple of occurrences of it, could reduce the overall amount of instructions by ~15+% [13:42:34 CEST] <[-T-]> the code that looks for SC is in libavcodec/utils ? [13:42:54 CEST] <__gb__> yes, and ffmpeg [13:43:04 CEST] <__gb__> 's version is quite optimized already [13:43:10 CEST] <[-T-]> ok [13:55:04 CEST] <durandal_170> find_rect is so slow [13:56:27 CEST] <nevcairiel> Compn: in my experience on windows, it adds bugs, so using dxva2 directly is favorable to me .. although I suppose its easier to use. But if you have avcodec already anyway, t hen the ease of use argument melts away since s omeone else has already done the heavy lifting [14:44:26 CEST] <Daemon404> philipl, the target_smver stuff sure is lulzy [14:44:34 CEST] <Daemon404> there are no enums or defines for the magic numbers? [14:46:27 CEST] <BtbN> nope [14:46:55 CEST] <Daemon404> that's pretty awful [14:46:57 CEST] <Daemon404> but hey. nvidia. [14:47:30 CEST] <BtbN> It's not part of the nvenc api though [14:47:38 CEST] <BtbN> that part is CUDA [14:47:42 CEST] <Daemon404> still nvidia. [14:47:55 CEST] <BtbN> You don't have to use CUDA for nvenc [14:47:59 CEST] <BtbN> unless you are on linux [15:07:22 CEST] <cone-448> ffmpeg 03Ivan Uskov 07master:db89f45535aa: avcodec/qsv: Extending QSV/MFX session initialization for the linux platform where a display handle is required. [15:21:05 CEST] <[-T-]> Thanks Ivan! [15:21:12 CEST] <[-T-]> are you on the chan ? [16:10:47 CEST] <philipl> BtbN: It seems to work in lossy mode. [16:31:58 CEST] <durandal_170> Looks like everybody hates me and my patches [16:32:19 CEST] <wm4> nevcairiel: is it possible that when using opengl on windows in exclusive mode (fullscreen, the driver forces it), you can't use d3d? [16:32:25 CEST] <ubitux> durandal_170: what do you need review on? [16:32:35 CEST] <nevcairiel> wm4: no idea, never dealt with ogl fullscreen exclusive [16:32:40 CEST] <wm4> nevcairiel: actually I'm trying to use dxva2, but that of course requires d3d [16:33:21 CEST] <wm4> and a similar problem doesn't exist with d3d exclusive mode? [16:33:46 CEST] <nevcairiel> well it can [16:34:01 CEST] <nevcairiel> you have to create the d3d device before the exclusive mode locks stuff up [16:34:33 CEST] <nevcairiel> but if you render in d3d anyway, you can also just use the render device [16:35:08 CEST] <durandal_170> ubitux: ya8 astats (a)drawgraph lut lavfi [16:35:15 CEST] <wm4> a bit hard with ogl and if the application starts in fullscreen [16:45:42 CEST] <BtbN> philipl, i'll test it later today or tomorrow on my Kepler hw [16:49:10 CEST] <philipl> BtbN: cheers. [16:49:25 CEST] <nevcairiel> Daemon404: the smver thing isnt all that magic really though, its just 4 bits of major and 4 bits of minor version [16:49:30 CEST] <BtbN> Currently too busy melting to test anything [16:49:33 CEST] <nevcairiel> 0x30 = 3.0, 0x52 = 5.2 [16:50:10 CEST] <nevcairiel> and yeah. f'ing summer [16:50:12 CEST] <nevcairiel> who invented this [16:50:24 CEST] <Daemon404> it's only bad because europe is a third world continent [16:50:26 CEST] <nevcairiel> its supposed to get even hotter tomorrow [16:50:31 CEST] <Daemon404> and doesnt have things like screens on windows [16:50:32 CEST] <Daemon404> or AC. [16:50:47 CEST] <nevcairiel> ACs are not really worth it for 6 days of hot weather a year [16:50:57 CEST] <nevcairiel> but during those 6 days, we do want them =p [16:51:11 CEST] <BtbN> I'll have a lot of work in our server rooms the next weeks [16:51:13 CEST] <Daemon404> i am seriously thinking of getting screens *made* for my place. [16:51:22 CEST] <Daemon404> i dont get why europeans dont have window screens [16:51:25 CEST] <Daemon404> to keep bugs out [16:51:30 CEST] <Daemon404> as soon as i open ym windows -> YAY BUGS [16:51:38 CEST] <nevcairiel> plenty people install bug screens [16:51:39 CEST] <BtbN> I have those on every single window? [16:52:00 CEST] <nevcairiel> i dont have them myself, but bugs arent that bad in this area [16:52:04 CEST] <BtbN> Not only against bugs, but against pollen, because of allergies. [16:52:11 CEST] <Daemon404> nevcairiel, none in any place ive been [16:52:18 CEST] <RiCON> screens stop pollen? [16:52:21 CEST] <Daemon404> yes [16:52:22 CEST] <nevcairiel> some can [16:52:23 CEST] <Daemon404> they do. [16:52:25 CEST] <J_Darnley> uh, a bug screen is far too large to stop pollen [16:52:26 CEST] <BtbN> there are special ones that do [16:52:51 CEST] <Daemon404> BtbN, what country? [16:52:54 CEST] <BtbN> germany [16:52:58 CEST] <Daemon404> same as nevcairiel [16:53:15 CEST] <Daemon404> i have never seen one in the UK, ireland, netherlands, belgium, france... [16:53:25 CEST] <Daemon404> italy i am told does not [16:53:25 CEST] <nevcairiel> my parents have bug screens on all bedroom windows, so they can open those at night and stuff and still keep them out [16:53:27 CEST] <BtbN> I installed them myself, but they are not that uncommon [16:53:29 CEST] <nevcairiel> not on all windows though [16:53:43 CEST] <Daemon404> maybe its more common in .de [16:54:06 CEST] <Daemon404> meanwhile in the UK, insulated housing is a luxury [16:54:50 CEST] <nevcairiel> we do have proper insulation most of the time, which means houses don't heat up that quickly during summer either [16:55:05 CEST] <RiCON> Daemon404: you'd hate portugal then, don't come here on vacation [16:55:15 CEST] <Daemon404> vacation is fine [16:55:17 CEST] <Daemon404> hotels have AC [16:55:36 CEST] <Daemon404> nevcairiel, i am moving into a newly built property this weekend (i bought a house) [16:55:40 CEST] <Daemon404> it is actually insulted [16:55:44 CEST] <Daemon404> my previous 3 places werent [16:55:50 CEST] <Daemon404> 2 of which didnt even have double glazed windows. [16:55:52 CEST] <nevcairiel> why do you insult your new house [16:56:07 CEST] <Daemon404> i did? [16:56:15 CEST] <nevcairiel> you said its insulted [16:56:19 CEST] <nevcairiel> i assume you did the insulting! [16:56:22 CEST] <Daemon404> oh [16:56:23 CEST] <Daemon404> lol [16:56:24 CEST] <Daemon404> typo. [16:57:00 CEST] <Daemon404> nevcairiel, germany also has actual winters in some bits [16:57:07 CEST] <Daemon404> so i assume you'd need it. [16:57:10 CEST] <nevcairiel> but yeah, we usually tend to have double glazed windows at least, and any house or apartment building constructed in the last two decades or so will have proper insulation [16:58:27 CEST] <nevcairiel> in fact, if you wanted to build something new, you would probably not get a license if its not properly insulated [16:58:36 CEST] <nevcairiel> no idea how old those laws are [16:59:05 CEST] <Daemon404> i know houses here probably wouldnt pass building codes in other countries [16:59:16 CEST] <Daemon404> e.g. exterior walls have pipes in them [16:59:20 CEST] <Daemon404> or out them [16:59:27 CEST] <Daemon404> which is an awesome way to freeze/explode your pipes [16:59:29 CEST] <Daemon404> in winter. [16:59:34 CEST] <nevcairiel> sounds like fun [16:59:53 CEST] <wm4> I thought UK doesn't have real winters [17:00:01 CEST] <Daemon404> scotland does [17:00:05 CEST] <Daemon404> here? no. [17:00:22 CEST] <BtbN> in my last house, the insulation was litteraly made of straw... [17:00:32 CEST] <Daemon404> straw is better than air [17:00:34 CEST] <BtbN> I was living under the roof, and had 45°C on the hottest days [17:01:22 CEST] <nevcairiel> directly under the roof is often quite painful, especially if the insulation is crap like that [17:01:25 CEST] <nevcairiel> i couldnt stand that [17:01:37 CEST] <BtbN> I'm still living under the roof now, but it's propperly insulated. [17:01:41 CEST] <Daemon404> BtbN, also there houses here with roofs MADE of straw [17:01:50 CEST] <Daemon404> and theyre listed as historical. you cant remove it. [17:01:54 CEST] <Daemon404> (it's ugly as shit) [17:02:04 CEST] <nevcairiel> we have that too [17:02:37 CEST] <nevcairiel> same historical BS [19:07:06 CEST] <cone-448> ffmpeg 03Michael Niedermayer 07master:9dc0bac9719a: avcodec/motion_est_template: Fix undefined shifts in CHECK_MV() [19:07:07 CEST] <cone-448> ffmpeg 03Michael Niedermayer 07master:c9220d5b0653: avcodec/mjpegdec: Reorder operations to avoid undefined behavior [19:46:24 CEST] <cone-448> ffmpeg 03Ivan Uskov 07master:6e5864ab294c: avcodec/qsvenc_h264: Change the set of performance presets to match with the MFX library constants. [19:47:11 CEST] <klaxa> how can i properly debug ffmpeg? i compiled it with --enable-debug --disable-optimizations --enable-openssl --enable-shared --enable-pic, when i run it in my debugger (nemiver) not all symbols can be resolved and i can't view the code from the libraries [19:47:27 CEST] <klaxa> it looks like this: http://dedi.klaxa.eu/public/2015-07-02-194524_3200x1800_scrot.png [19:47:52 CEST] <klaxa> when i debug my code from doc/examples/ i can view all source files [19:49:43 CEST] <J_Darnley> Does ffmpeg link witht the correct shared libraries? I see that the paths there start with /usr/loc [19:49:47 CEST] <klaxa> ah, i have the feeling enabling shared libraries might be an issue [19:50:05 CEST] <klaxa> i install my freshly compiled libraries in /usr/local/ [19:50:41 CEST] <klaxa> because i couldn't figure out how to tell the makefile/ld to use the libraries i just built when i built my code in doc/example [19:51:34 CEST] <klaxa> i'm compiling it without shared libraries right now to see if that helps [19:53:40 CEST] <c_14> klaxa: the `ffmpeg' binary? That's stripped by default, either use the ffmpeg_g binary or --disable-stripping [19:53:47 CEST] <klaxa> i'm using ffmpeg_g [19:54:12 CEST] <klaxa> i can see the ffmpeg.c code and everything but nothing from libavformat (which i want to debug :P) [19:54:26 CEST] <klaxa> s/and everything// [19:55:18 CEST] <c_14> Might be the shared libraries, because the installed libraries are stripped and then the unstripped ffmpeg_g uses the stripped libraries. Either try static or --disable-stripping [19:56:07 CEST] <klaxa> okay thanks, i'll try that [20:02:35 CEST] <klaxa> ah well after switching to the correct branch and recompiling the segfault i wanted to inspect doesn't occur anymore [20:04:57 CEST] <klaxa> and i also can properly debug it now, thanks! [21:32:58 CEST] <durandal_170> michaelni: what to do with lut highbit depth support fate test? [21:33:47 CEST] <durandal_170> just kill pixfmt one and add new ones or something else ? [21:43:31 CEST] <kierank> a wild netflix appears [21:47:12 CEST] <michaelni> durandal_170, not sure i understand the question [21:48:56 CEST] <durandal_170> michaelni: pixfmt tests can't work with >8 bits formats [21:50:02 CEST] <michaelni> why? [21:51:52 CEST] <durandal_170> on bigendian you get be fmts on litttle le fmts [21:52:23 CEST] <durandal_170> filter operates in native format only [21:53:27 CEST] <michaelni> converting to fixed endiannness might work [21:53:30 CEST] <durandal_170> there is no switch to use only le fmts [21:54:48 CEST] <J_Darnley> I thought that -pix_fmt *le gave you a little endian format. [22:09:23 CEST] <cone-448> ffmpeg 03Paul B Mahol 07master:7ff5a345a46e: avfilter: use AVFILTER_DEFINE_CLASS() [22:09:24 CEST] <cone-448> ffmpeg 03Paul B Mahol 07master:96953e2ef653: avfilter/vf_mpdecimate: remove packed formats [23:33:32 CEST] <cone-448> ffmpeg 03Andreas Cadhalpun 07master:d0eff8857cef: wavpack: limit extra_bits to 32 and use get_bits_long [23:47:40 CEST] <kierank> Daemon404: do my emails appear as spam [23:52:13 CEST] <jamrial> kierank: i'm having a lot of emails being sent to the spam folder lately. is this something happening to other people as well? [23:52:19 CEST] <kierank> yes [23:52:23 CEST] <kierank> @googlemail.com [23:52:31 CEST] <kierank> and a few others [23:53:03 CEST] <Daemon404> kierank, no [23:53:05 CEST] <J_Darnley> heh, someone probably couldn't figure how to unsubscribe to they started hitting the spam button [23:53:22 CEST] <jamrial> kierank: yeah [23:53:44 CEST] <nevcairiel> google seems to have screwed up their SPF records or something flagging googlemail.com as invalid [23:54:31 CEST] <nevcairiel> or the dmarc header in this case, apparently [23:55:13 CEST] <nevcairiel> people that use such an address should report to google [23:56:26 CEST] <Daemon404> does google have a german office? [23:56:42 CEST] <j-b> yes [23:57:38 CEST] <nevcairiel> several, depending on which branch you want [23:58:15 CEST] <nevcairiel> i th ink there is an accounting office in my city [00:00:00 CEST] --- Fri Jul 3 2015

1 0

ffmpeg.log.20150702
by burek 03 Jul '15

03 Jul '15

[00:26:38 CEST] <hanshenrik> all files have the exact same encoding (they're actually clipped from the same source file, with ffmpeg with codec copy); and i get no errors, BUT the files is not concatenated, only the first file is encoded and it has the same filesize as the first file only. ... any idea what's wrong here? (only foo1.mp4 is in the resulting file); [00:27:08 CEST] <hanshenrik> ffmpeg -i "concat:foo1.mp4|foo2.mp4|foo3.mp4" -c:a copy -c:v copy full.mp4 [00:27:30 CEST] <hanshenrik> i dont get any errors, but it does not concat, it only adds the first file to full.mp4 [00:30:59 CEST] <hanshenrik> ..looks like a bug to me x.x [00:31:35 CEST] <hanshenrik> (the codecs are mpeg4 video + mp3 audio ) [00:40:33 CEST] <relaxed> hanshenrik: the concat protocol only works with mpeg type containers [00:41:45 CEST] <relaxed> hanshenrik: did you see https://trac.ffmpeg.org/wiki/Concatenate ? [00:42:44 CEST] <relaxed> or, (MP4Box is part of gpac) MP4Box -cat 1.mp4 -cat 2.mp4 -new combined.mp4 [00:45:03 CEST] <hanshenrik> relaxed, i did barely see it, yes; and i thought .mp4 container qualified? :p [00:45:26 CEST] <relaxed> you thought wrong :p [00:47:28 CEST] <hanshenrik> apparently. funny that the files are called .mp4 in the examples on https://trac.ffmpeg.org/wiki/Concatenate then [00:48:16 CEST] <c_14> Because you ignored the remuxing to mpegts step? [00:48:23 CEST] <c_14> s/ignored/overread/ [00:49:19 CEST] <hanshenrik> yup, that seems to be it. [01:05:54 CEST] <relaxed> this works too http://pastie.org/pastes/10268974/text [01:14:19 CEST] <hanshenrik> relaxed, it generates really bad quality and spews errors every second about "codec copy not implemented. contributions are welcome!". but MP4box worked great! thanks [01:15:05 CEST] <relaxed> it's because you have mpeg4/mp3, but I'm glad mp4box worked [09:46:49 CEST] <Abhijit> how can i concatenate segment files on the fly into single video while frontend is still playing the video? [10:20:29 CEST] <gurpartap1> I am compiling ffmpeg on angstrom.While doing ./configure , It is showing ---> ./configure: line 422: pr: command not found .What does it mean... [10:23:19 CEST] <chungy> it means you're missing pr. [10:23:37 CEST] <chungy> it's a program from coreutils. [10:24:27 CEST] <Pixer> hey everyone [10:24:40 CEST] <Pixer> is ffmpeg project still active? [10:25:10 CEST] <JEEBsv> yes [10:25:25 CEST] <JEEBsv> both ffmpeg and libav are continuations of ffmpeg-old [10:25:41 CEST] <Pixer> i'm using it to extract an audio but it says it has deprecated to give way to avconv [10:26:35 CEST] <JEEBsv> in the libav project, yes [10:26:41 CEST] <JEEBsv> so if you are using libav, you use avconv [10:26:45 CEST] <Pixer> JEEBsv: could you please help me with this task i'm on. need to extract audio from an .mp4 file [10:26:47 CEST] <JEEBsv> if you use ffmpeg, you use ffmpeg [10:27:17 CEST] <JEEBsv> since you're on libav, `avconv -i input.mp4 -c copy out.file` [10:27:29 CEST] <JEEBsv> also since you still have the ffmpeg binary, I recommend you update to a newer thing [10:27:32 CEST] <JEEBsv> be it libav or ffmpeg [10:27:46 CEST] <JEEBsv> because that usually means that you're on libav 0.8, which is _old_ [10:29:35 CEST] <Mavrik> Pixer, in short what's JEEBsv telling you: don't use the ancient ffmpeg package in Ubuntu/Debian [10:30:45 CEST] <Pixer> Mavrik: ok [10:30:52 CEST] <Pixer> JEEBsv: thanks [10:31:16 CEST] <Mavrik> even though, for extracting audio I guess the avconv bundled should work just fine :) [10:34:57 CEST] <Pixer> < anconv -i sample.mp4 -vn -acodec 'copy' sample.aac> is this command correct? [10:37:30 CEST] <JEEBsv> Pixer: for libav, yes. also you don't need the '' there [10:37:38 CEST] <JEEBsv> I would also prefer -c:a [10:37:40 CEST] <JEEBsv> not acodec [10:38:06 CEST] <Pixer> thanks . did it. [10:39:15 CEST] <Pixer> < anconv -i sample.mp4 -vn sample.mp3> worked fine. [10:39:28 CEST] <JEEBsv> although the -vn might not be needed if you want a raw audio thing, since that can't even have video tracks [10:39:38 CEST] <JEEBsv> yes, that will re-encode it with *something* [10:39:57 CEST] <Pixer> what's command for raw audio? [10:40:15 CEST] <JEEBsv> I meant raw audio bitstream [10:40:29 CEST] <JEEBsv> dot-mp3 and dot-aac by default map to raw formats [10:40:50 CEST] <Pixer> ok [10:42:25 CEST] <Pixer> yeah -vn is not necessary. [11:10:28 CEST] <JEEBsv> it will be if you will be using a container that can handle video as well, though [11:25:30 CEST] <koz_> Does anyone know which open formats for audio and video offer the best compression? [11:48:48 CEST] <chungy> /dev/null is pretty good [11:51:04 CEST] <DeadSix27> heh [17:22:40 CEST] <mateo`> hello o/, is there a conveniant way to determine the plane width (not the stride) depending on the AVPixelFormat ? [17:34:37 CEST] <bertieb_> Quick q- is there a quick (semantic) way of stripping the final frame (audio and video) from a file using stream copy; or do I have to ffprobe it and give a -t duration that stops one frame short? [17:40:25 CEST] <bertieb> (The docs for 'seeking' state that a cut will always occur at an I-frame when doing stream copy, but I assumed that meant the first frame would definitely be an I-frame) [18:12:16 CEST] <bertieb> Can anyone confirm (or deny) that *both* ends will always be cut at an I-frame when doing a stream copy? Would like to save myself and the ml some hair pulling :) [18:13:13 CEST] <BtbN> I don't see any reason why the end would have to be at an iframe [18:13:28 CEST] <BtbN> unless the codec uses forward references of course [18:13:28 CEST] <bertieb> (attempting a stream copy with a -t duration of less than entire duration produces a video with the same length as the original, but I don't want to treat that as canonical ) [18:15:33 CEST] <bertieb> BtbN: The video stream is h264 if you know off the top of your head (a quick search gave me a resounding maybe) for forward references [18:16:27 CEST] <BtbN> I wouldn't be surprised if it's implemented to only cut at iframes, but i'd have to look at the code first to be sure. [18:22:20 CEST] <bertieb> Fair enough; I wouldn't be sure whereabouts to start digging in the code (having glanced at its github)! [18:55:14 CEST] <StatiQ> on where i find static builds of ffmpeg phone ARMV6? [18:56:19 CEST] <StatiQ> i have one, but is missing libfaac and at least i don't know the developer [18:56:50 CEST] <StatiQ> ARMV6 ANDRUG PHONE [19:01:35 CEST] <StatiQ> i have another version with thread emulation. (shared libraries) like real linux. but it shows an error while trying to encode to .aac relative to wrong params. [19:02:29 CEST] <StatiQ> may i have enabled aac? or i have to get another binaey? [19:03:51 CEST] <StatiQ> so much users here to Wait [19:05:03 CEST] <StatiQ> i still don't believe that linux still not created an universal binary standard that works independently of archs [19:05:07 CEST] <StatiQ> xc [19:06:25 CEST] <bertieb> StatiQ: The mess that you want is called java [19:06:29 CEST] <StatiQ> incredible. all broken. is arm v1.2.3.4.5.6, amd, intel, x86 (and on billion broken more). lamentable [19:06:45 CEST] <bertieb> But that doesn't solve your problem [19:07:31 CEST] <StatiQ> bertieb: armv6 of ffmpeg runs on armv7 or no? [19:07:47 CEST] <bertieb> StatiQ: I have no idea [19:08:11 CEST] <StatiQ> i have armv5 libs running in my phone [19:08:33 CEST] <StatiQ> i have some armv7 runnin on my armv6 phone [19:08:47 CEST] <StatiQ> why this is all broken? [19:09:28 CEST] <StatiQ> i has neon running in my armv6 [19:10:26 CEST] <StatiQ> so. if these all works i have a unix binary of ffmpeg that have to run in all archs [19:10:42 CEST] <StatiQ> but is missing libfaac. [19:11:33 CEST] <StatiQ> if i convert to m4a or mp4a i need to rename to aac. but this is not legal [19:13:46 CEST] <StatiQ> what i am doing wrong? the ffmpeg (threaded version) is saying "wrong params " i tried 128k_mp3 and pcm_s16le_wav. [19:17:11 CEST] <StatiQ> the static version, i have to convert to mp4a to rename to aac after [19:18:14 CEST] <StatiQ> well the static is showing D aac -formats so is "disabled" ? :D [19:18:30 CEST] <bertieb> StatiQ: D is for decoder [19:19:03 CEST] <StatiQ> yes if was enebled Did showed DE [19:19:36 CEST] <StatiQ> but the aac lib for andraeid still is broken [19:19:52 CEST] <StatiQ> like standsrd linux [19:20:28 CEST] <StatiQ> in my live linux i have an special separated aac encoder... [19:26:14 CEST] <bertieb> BtbN: Glancing through the source code for stream copy handling as you mentioned, I'm not seeing overt references to mandate ending on and i-frame; but I am likely missing something [19:27:38 CEST] <bertieb> But then I've yet to figure out exactly how it determines end time in any case [23:39:15 CEST] <_Vi> How do I make FFmpeg scan the entire file and extract things that look like a media? [23:40:26 CEST] <klaxa> you... don't [23:40:46 CEST] <klaxa> you can try foremost or scalpel or other file recovery software though [23:41:41 CEST] <JEEBsv> well you can make the probesize or whatever it was really big if you just want ffmpeg to probe more data from the file [23:41:56 CEST] <_Vi> When I specify the format explicitly like "-f mp3" and "-f ogg", it does. [23:42:44 CEST] <JEEBsv> that just forces an input "container" to be set, ignoring whether or not it looks like it [23:43:01 CEST] <JEEBsv> doesn't mean it actually is anything valid [23:43:09 CEST] <_Vi> So I want it to look for various containers. [23:43:21 CEST] <JEEBsv> it does that by default, as it probes [23:43:40 CEST] <JEEBsv> you can try making the probesize bigger if the standard X megabytes is not large enough [23:43:55 CEST] <_Vi> ... but don't stop after the first found one, instead continue probing the trailer of the file. [23:44:04 CEST] <JEEBsv> uhh [23:44:26 CEST] <JEEBsv> if it probes something it takes the whole input as that [23:44:46 CEST] <JEEBsv> if you have concaterated random things you won't get too far [23:45:14 CEST] <JEEBsv> *if the probe finds something "matching enough" by its checks [23:46:15 CEST] <_Vi> photorec does save a bunch of ogg files, but that are broken. [23:48:47 CEST] <JEEBsv> most probably because they got corrupted along the way :P [23:49:02 CEST] <JEEBsv> congratulations, you have most probably failed in a similar way as I did [00:00:00 CEST] --- Fri Jul 3 2015

1 0

ffmpeg-devel.log.20150701
by burek 02 Jul '15

02 Jul '15

[00:00:00 CEST] <lglinskih_> kierank, wm4: comparing crcs of a frames will work, right? I can make manually few reference files. [00:00:29 CEST] <wm4> nevcairiel: posts gsoc http patches [00:00:34 CEST] <nevcairiel> oh a gsoc guy [00:00:41 CEST] <llogan> ^ klaxa [00:00:42 CEST] <nevcairiel> he did post them using send-email on the devel mail once before [00:00:47 CEST] <wm4> lglinskih_: yeah, I guess so [00:01:00 CEST] <kierank> lglinskih_: yes but it's not really clear how you would know whether the seeking worked to begin with [00:01:28 CEST] <klaxa> wm4: sorry, i was told i can either send them with git-send-email or as attached git format-patch patches [00:01:46 CEST] <wm4> klaxa: I think it makes it pretty hard to read and review them [00:01:48 CEST] <klaxa> i don't know how to reply to emails with git send-email, is there an easy way to do that? [00:01:56 CEST] <wm4> klaxa: and git send-email should be more convenient to you too [00:02:02 CEST] <wm4> ah [00:02:03 CEST] <klaxa> i would very much like to do it like that [00:02:30 CEST] <wm4> you can use --in-reply-to= and the id of the mail you want to reply - dunno if there's a more convenient way [00:02:56 CEST] <nevcairiel> yeah replying with a fixed up patch is a bit inconvenient, but shrug [00:03:10 CEST] <klaxa> where do i get the id from? unfortunately i'm using the gmail webclient [00:03:15 CEST] <wm4> making a new thread is fine too, as long as you don't post 10 per day [00:03:35 CEST] <nevcairiel> id is in the headers, look at show original or something in the gmail web [00:03:58 CEST] <klaxa> okay, i will try that in the future, thanks for telling me! [00:05:38 CEST] <lglinskih_> kierank: I'm sorry, I don't understand( [00:06:04 CEST] <kierank> lglinskih_: For example if the seeking produced a green frame (i.e broken) you wouldn't be able to tell with a crc [00:06:15 CEST] <kierank> but it's not a big problem right now [00:06:47 CEST] <nevcairiel> (note that broken frames in ffmpeg are mostly gray) [00:07:03 CEST] <wm4> lglinskih_: what we might want to do is comparing the same frames produced after different seeks [00:07:28 CEST] <wm4> actually, we definitely want to do this [00:09:06 CEST] <lglinskih_> kierank, wm4: I can start with calculation crcs of all frames in file without seeking. And after that reopen file and make a lot of different seeking=) [00:09:18 CEST] <kierank> yes good idea [00:10:40 CEST] <wm4> yep, great idea [00:11:27 CEST] <lglinskih_> thx ^_^ [00:35:28 CEST] <cone-342> ffmpeg 03Andreas Cadhalpun 07master:32a5b631267e: pthread_frame: forward error codes when flushing [00:35:29 CEST] <cone-342> ffmpeg 03Andreas Cadhalpun 07master:cd64ead8d96b: ffmpeg: exit_on_error if decoding a packet failed [00:35:30 CEST] <cone-342> ffmpeg 03Andreas Cadhalpun 07master:bd0f14123fd8: ffmpeg: only count got_output/errors in decode_error_stat [01:01:47 CEST] <cone-342> ffmpeg 03rogerdpack 07master:4ebb43f19c41: ffmpeg: windows: respond to logoff and ctrl+break messages as well [02:35:07 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:79a98294da6c: avcodec/aacsbr: check that the element type matches before applying SBR [02:35:08 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:2e13a45b1a9a: avcodec/aacsbr: Assert that bs_num_env is positive [02:39:47 CEST] <Compn> carl : think we should close this? http://trac.ffmpeg.org/ticket/1870 [03:12:21 CEST] <Compn> http://trac.ffmpeg.org/ticket/3352 this one still a problem ? [03:14:18 CEST] <cone-342> ffmpeg 03Martin Storsjö 07master:e2bd03a14a4e: fate: Avoid unnecessary pixel format conversions [03:14:19 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:3974889614e0: Merge commit 'e2bd03a14a4e3366df0b1ee8e284a97165be1f3c' [03:20:57 CEST] <Compn> carl : maybe close this one as theres only one or two samples?, http://trac.ffmpeg.org/ticket/3077 [03:36:32 CEST] <cone-342> ffmpeg 03Kostya Shishkov 07master:08c2d8f0aa67: Go2Meeting decoder [03:36:33 CEST] <cone-342> ffmpeg 03Diego Biurrun 07master:4d1229dabf7a: g2meet: Add FATE tests for all three G2M variants [03:36:34 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:68939f7682de: Merge commit '08c2d8f0aa679c2f060721d1d0d4f33d2ae32368' [03:36:35 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:9e93e544dc2e: Merge commit '4d1229dabf7a7e3b6a7b326afd79102256c3b008' [03:49:59 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:3f872c9bfa8e: lavc: Add missing API guard to dtg_active_format option [03:50:00 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:0f87f9b4fcee: lavd: Add library identifier [03:50:01 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:0d449c81b3dd: lavfi: Add library identifier [03:50:02 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:16dd0426e568: Merge commit '3f872c9bfa8ee1032058cfa48068b5a73ef27b5e' [03:50:03 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:818ff7ff5a19: Merge commit '0f87f9b4fceee854f09da2d1ef329245196775f8' [03:50:04 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:9c010ba66800: Merge commit '0d449c81b3dd25835ae6ac849cdd150f35b9c5c6' [04:05:55 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:df22e30172b0: dump: Use the correct abs() version [04:05:56 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:a8e137a32286: Merge commit 'df22e30172b09cda4d6f7d4f43508284be65848a' [04:13:19 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:2eef75fd7e1a: mov: Adjust variable types to fix format warnings [04:13:20 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:35c8dda5c2db: Merge commit '2eef75fd7e1ac96ab9ca63bb4523078c908bc9b1' [04:22:57 CEST] <chocoladedevelop> I'm using c# with ffmpeg and pipes and i'm pushing frames and create in real time compressed mp4 video file. [04:23:08 CEST] <chocoladedevelop> The problem is that after few seconds the ffmpeg.exe memory usage is raising to over 1GB of memory and the hard disk raise up to over 53MB/s [04:23:17 CEST] <chocoladedevelop> Is that a known problem ? [04:23:22 CEST] <cone-342> ffmpeg 03Vittorio Giovara 07master:a1e2caa93e4f: mov: Log format rather than fourcc in stsd in trace mode [04:23:23 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:55a04a5d7a90: Merge commit 'a1e2caa93e4f8102666a21222f01b74838b6497f' [04:25:12 CEST] <chocoladedevelop> I posted my code in csharp in the ffmpeg forum here: http://ffmpeg.gusari.org/viewtopic.php?f=11&t=2204 and also here: http://stackoverflow.com/questions/31149752/why-when-using-ffmpeg-to-create… [04:37:12 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:dec728888f88: avcodec/elsdec: Remove EOVERFLOW [04:48:47 CEST] <Compn> chocoladedevelop : sounds like bug [04:48:58 CEST] <Compn> memleak somewhere ? [05:11:33 CEST] <chocoladedevelop> BUG IN FFMPEG.EXE OR IN MY CSHARO CODE ? [05:11:39 CEST] <chocoladedevelop> oops sorry for caps [05:11:53 CEST] <chocoladedevelop> csharo = csharp [05:12:50 CEST] <chocoladedevelop> cant figure out i didn't find in my code any mem leak. Maybe someone can look at the links i provided to see my code. [05:23:46 CEST] <cone-342> ffmpeg 03Michael Niedermayer 07master:8750aef3d65c: ffmpeg_opt: Fix forcing fourccs [10:04:55 CEST] <wm4> so the libswresample docs claim that e.g. lfe_mix_level is in db [10:05:14 CEST] <wm4> but it actually seems to be linear? [12:02:00 CEST] <wm4> sometimes I feel like michaelni intentionally adds this kind of code http://git.videolan.org/?p=ffmpeg.git;a=blob;f=libavutil/utils.c;h=0b765ed0… to give Libav a point that there is insane code in FFmpeg that isn't in Libav [12:02:06 CEST] <wm4> or maybe I'm just becoming paranoid [12:03:06 CEST] <wm4> j-b: so when will you fork ffmpeg [12:08:45 CEST] <ubitux> i'm pretty sure it helps triggering real case issue [12:08:53 CEST] <ubitux> typically while doing merges [12:09:15 CEST] <wm4> bullshit argument [12:09:29 CEST] <ubitux> ? [12:09:52 CEST] <wm4> there are so many potential ABI issues that could happen, and you test one of the least important/easiest to detect ones [12:09:53 CEST] <ubitux> also any contributor trying to insert a pix fmt in the middle or something [12:10:10 CEST] <wm4> also source code shouldn't contain safeguards for stuff that could happen due to merges [12:10:14 CEST] <wm4> ... [12:10:21 CEST] <wm4> BULLSHIT [12:10:54 CEST] <wm4> why doesn't it check whether you add a field to AVCodecContext? [12:11:01 CEST] <j-b> wm4: depends. Do you join, if I do? [12:11:03 CEST] <wm4> why doesn't it check whether a function changes its signature? [12:11:35 CEST] <wm4> j-b: sure I would, but most likely I wouldn't be able to spend much effort to do all the work that it'll require [12:12:06 CEST] <ubitux> are these safeguard causing any problem to you? [12:14:38 CEST] <wm4> ubitux: yes [12:15:57 CEST] <wm4> many why is there suddenly a huge patch set against the old ASF demuxer by someone I've never seen before [12:16:24 CEST] <wm4> how does anyone in here manage not to go mental? [12:16:47 CEST] <nevcairiel> you get annoyed because some weirdo posts a patch set? [12:17:01 CEST] <nevcairiel> looks like the majority is for the muxer though [12:17:17 CEST] <nevcairiel> and 90% of the patch is unrelated crap changes [12:17:29 CEST] <wm4> oh, it's mainly a ASF muxing thing [12:19:19 CEST] <wm4> maybe software development is a too stressful endeavor for me and I should just become a gardener or something [13:08:58 CEST] <Compn> wm4 : gardening is pretty stressful, i have a groundhog that is eating everything in my garden. all of my sunflowers! [13:09:32 CEST] <wm4> unlike with software developers, you can just kill the groundhog [13:11:00 CEST] <Compn> aww but hes so cute and fuzzy wuzzy :( [13:11:09 CEST] <Compn> also whoever this asf patch guy is hes using msys git... [13:11:25 CEST] <Compn> a vindows users. [13:11:26 CEST] <Compn> -s [13:40:02 CEST] <durandal_170> michaelni: I triggered assert in nutdec line 1331 [13:42:45 CEST] <durandal_170> I used cat to concatenate 2 nuts with single raw yuvs [13:43:37 CEST] <michaelni> durandal_170, how can i reproduce ? [13:46:19 CEST] <durandal_170> you must use cat and pipe [13:47:22 CEST] <durandal_170> cat 1st.nut 2nd.nut Š ffplay - [13:48:31 CEST] <durandal_170> and its line 1301 [13:51:17 CEST] <michaelni> tried random rawvideo in nut but seems not triggering te assert [13:53:34 CEST] <durandal_170> did you used pipe and cat with ffplay? [13:56:38 CEST] <durandal_1707> michaelni: ffmpeg -f lavfi -i testsrc -pix_fmt yuv420p -c:v rawvideo -frames 1 /tmp/1.nut [13:56:42 CEST] <durandal_1707> michaelni: ffmpeg -f lavfi -i testsrc -pix_fmt yuv420p -c:v rawvideo -frames 1 /tmp/2.nut [13:57:05 CEST] <durandal_1707> michaelni: cat /tmp/1.nut /tmp/2.nut | ffplay - [13:57:14 CEST] <durandal_1707> michaelni: seek back with left arrow [13:58:02 CEST] <michaelni> durandal_170, thanks [14:03:01 CEST] <cbsrobot_> durandal_170: still looking for new filters to implement? see http://shenidam.org/ [14:09:38 CEST] <Compn> auto sync ? [14:12:23 CEST] <cbsrobot_> yeah - similar to pluraleyes [14:38:16 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:2f8c81637c47: avformat/matroskadec: Fix undefined shift in read_sint() [14:38:17 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:56fd4705c049: avcodec/motion_est_template: Fix undefined shifts in CHECK_MV_DIR() [14:38:18 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:0ea099ad3e6d: avcodec/mpegvideo_enc: fix undefined shifts in ff_dct_quantize_c() [14:38:19 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:4eee685a212a: avcodec/motion_est: Fix undefined shifts in cmp_inline() [14:38:20 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:60ec3007e69c: avformat/nutdec: Check ff_gen_search() for failure [14:38:34 CEST] <michaelni> durandal_170, fixed [14:55:59 CEST] <cone-302> ffmpeg 03Vadim Belov 07master:db64af639559: avformat/concatdec: copy stream metadata when using concat [15:12:12 CEST] <cone-302> ffmpeg 03wm4 07master:f91126643a91: lavu: add an API function to return the FFmpeg version string [15:53:48 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:ac78014f0b1f: avcodec/motion_est: Fix some undefined shifts [15:53:49 CEST] <cone-302> ffmpeg 03Michael Niedermayer 07master:53fd70579bc2: avcodec/h264_mvpred: Fix undefined shifts in MAP_F2F [17:04:27 CEST] <durandal_170> stop bloating lavu [17:29:09 CEST] <Zeranoe> Holy cow FFmpeg struggles to play this https://dl.dropboxusercontent.com/u/30310240/flourish.mid [17:34:52 CEST] <wm4> well that's a midi file? [17:39:43 CEST] <Compn> yeah midi [17:41:37 CEST] <j-b> WTF is Extended ASF functionality [17:41:43 CEST] <Compn> lol [17:42:02 CEST] <philipl> "We extended the functionality. Pray we don't extend it again." [17:42:14 CEST] <j-b> why does it require 3 new headers file? [17:43:28 CEST] <Compn> there must be millions of weird patches for ffmpeg floating out there [17:48:42 CEST] <Compn> wonder if the virtualdub asf demuxer was any good [17:48:52 CEST] <Compn> or if that was asf muxer only ? mhm [17:50:19 CEST] <Compn> curious why people didnt adopt jp2k instead of ffv1 :P [17:50:43 CEST] <Compn> nice to see open source ffv1 take over though [18:05:04 CEST] <cone-394> ffmpeg 03Shivraj Patil 07master:2eb28e889d9c: avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for mpegvideo functions [18:05:17 CEST] <rcombs> is Benoit Fouet around on IRC? [18:06:48 CEST] <rcombs> Zeranoe: AFAIK it doesn't play at all? I'm not aware of ffmpeg having a MIDI demuxer or decoder [18:07:11 CEST] <Zeranoe> rcombs: it tried, and sounded quite interesting to say the least [18:07:38 CEST] <rcombs> I also tried, and just got AVERROR_INVALIDDATA [18:08:14 CEST] <Zeranoe> rcombs: with a recent build? [18:08:53 CEST] Action: rcombs git pull [18:10:31 CEST] <nevcairiel> decoding MIDI is annoying, as you need a synth bank [18:10:39 CEST] <c_14> Probably libmodplug? [18:10:51 CEST] <nevcairiel> oh yeah we have a library that can do it [18:12:20 CEST] <nevcairiel> not sure what kind of formats it supports however [18:13:20 CEST] <jamrial> somebody sent a huge patch for the old asf demuxer [18:13:29 CEST] <jamrial> no end in sight for this whole debate, haha [18:13:58 CEST] <nevcairiel> that huge patch is full of crap as well [18:17:03 CEST] <nevcairiel> he also doesnt specify at all what it does [18:19:09 CEST] <rcombs> it enhances the support, duh [18:30:11 CEST] <philipl> who wouldn't want enhanced support? [18:38:10 CEST] <Daemon404> [17:13] < jamrial> somebody sent a huge patch for the old asf demuxer <-- and i dont even know what it does [18:38:15 CEST] <Daemon404> "extend asf" [18:38:17 CEST] <Daemon404> yeah ok.. [18:38:24 CEST] <Daemon404> (too lazy to read. probably metadata) [18:38:25 CEST] <rcombs> oh, "extends" [18:39:00 CEST] <nevcairiel> i immediately went "wat, this patch is broken" when i saw what it all changed, without even wondering what its supposed to do [18:56:41 CEST] <cone-394> ffmpeg 03John Adlum 07master:089a818bd3cd: avcodec/pthread_frame: Correcting typo of "occurred" [18:56:42 CEST] <cone-394> ffmpeg 03John Adlum 07master:28206b75e81c: avformat/asfdec_f: Correct skip to key code [18:56:43 CEST] <cone-394> ffmpeg 03John Adlum 07master:811008b8eeaa: avformat/asfdec_f: Assert that packet positions match in asf_read_pts() [19:09:15 CEST] <cone-394> ffmpeg 03Janne Grunau 07master:007e27d363ba: avcodec: add missing CODEC_CAP_DR1 to codecs using get_buffer() [19:09:16 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:9cf95654ac90: Merge commit '007e27d363ba7d994019dc897dc9c39071bb204a' [19:17:52 CEST] <cone-394> ffmpeg 03Janne Grunau 07master:4ccccd6c40a6: g2meet: use an unsigned type for the djb hash [19:17:53 CEST] <cone-394> ffmpeg 03Janne Grunau 07master:9eec23b8a7fd: g2meet: use av_ceil_log2 instead of a custom function [19:17:54 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:92c858ae667a: Merge commit '4ccccd6c40a6d0ce85e96a6e37f558236e2a6a75' [19:17:55 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:6205143bb3fe: Merge commit '9eec23b8a7fd0f91827bbc3ed0792c39a8cc9a8a' [19:20:17 CEST] <cone-394> ffmpeg 03Paul B Mahol 07master:be35b8b9a9fb: avfilter/vf_extractplanes: support more pixel formats [19:20:18 CEST] <cone-394> ffmpeg 03Paul B Mahol 07master:fc40cdbf4904: avfilter/vf_extractplanes: rename misleading variable [19:20:19 CEST] <cone-394> ffmpeg 03Paul B Mahol 07master:17e6d7b4008f: avfilter/vf_extractplanes: use faster path for input formats with only one component [19:36:40 CEST] <BBB> durandal_170: ? [19:37:00 CEST] <BBB> oh the version string [19:37:04 CEST] <BBB> thats a few bytes [19:38:13 CEST] <durandal_170> but each time need to recompile whole tree [19:43:39 CEST] <BBB> get a faster computer :-p [19:43:50 CEST] <BBB> (ok I admit I get annoyed every time that happens) [19:43:55 CEST] <BBB> (so I just end up updating less often) [19:46:57 CEST] <cone-394> ffmpeg 03Janne Grunau 07master:f91fe24e9bd6: g2meet: force simple idct for identical results over all fate configs [19:46:57 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:5142963b7d93: Merge commit 'f91fe24e9bd6912c29bbb03d8afe878e045f9721' [19:52:59 CEST] <wm4> michaelni: Libav wants av_version_info() instead of avutil_version_info(); would you agree to delete the old new symbol and replace it with the one with the new name? [19:53:47 CEST] <lglinskih_> Is pkt.pts a good parameter as a timestamp to seek with avformat_seek_file? [19:54:44 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:4e0819310e2d: elsdec: Replace EOVERFLOW with INVALIDDATA [19:54:45 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:f8e038f9a07d: Merge commit '4e0819310e2d2eff60be2d6df28335f0739712b9' [19:56:12 CEST] <wm4> lglinskih_: that's not really clear [19:56:24 CEST] <wm4> in most situations, it probably is [19:56:53 CEST] <michaelni> it might make sense to add similar functions to the other libs, user apps might want to know if they are grossly mismatching each other [19:57:12 CEST] <wm4> michaelni: no, they can use the integer version number functions [19:58:51 CEST] <michaelni> honestly i think the name as it is is better as it really described the revission from which avutil is build not neccessarily the one from all or the other libs but i dont really care much if you prefer to rename it [20:00:00 CEST] <wm4> then we'll rename [20:10:32 CEST] <lglinskih_> wm4: when I set min_ts and max_ts like this avformat_seek_file(fmt_ctx, str_index, pts, pts, pts, AVSEEK_FLAG_ANY) it seeks to a frame with bigger pts. [20:10:39 CEST] <lglinskih_> is it normal? [20:11:34 CEST] <wm4> probably [20:11:39 CEST] <wm4> note that there are 2 seek APIs [20:11:46 CEST] <iive> what does av_version_info() return? [20:11:50 CEST] <wm4> and I think avformat_seek_file() is the worse one [20:12:41 CEST] <lglinskih_> but its value is out of the interval! [20:12:46 CEST] <iive> imho, such function should return the release number. e.g. (ffmpeg) 2.7.1 [20:13:05 CEST] <jamrial> iive: it does when you build any of the release branches afaik [20:14:01 CEST] <iive> then what does avutil_* one returns? the same or api version? [20:15:11 CEST] <nevcairiel> wm4: internally the APIs work mostly the same way, they just have different semantic [20:15:32 CEST] <nevcairiel> the new seeking methods were never implemented [20:16:19 CEST] <jamrial> iive: actually there's only one, and it's avutil_version_info [20:17:26 CEST] <iive> does it return the same info release/git_hash? [20:19:16 CEST] <jamrial> it returns what i said above. for example n2.7.1 when you compile the tagged release, n2.7.1-x-g****** when you compile newer commits from the same release branch, N-xxxxx-g****** for master branch [20:20:08 CEST] <durandal_170> do we have reviewers? [20:20:14 CEST] <jamrial> see the stuff under "rev" in http://fate.ffmpeg.org/, that's supposedly what it will return [20:20:15 CEST] <lglinskih_> wm4: do you mean I should use av_seek_frame? [20:58:38 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:ce81e47c911f: avcodec/mss2: Fix integer overflow [20:58:39 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:06a0d5ef5ce3: avcodec/h264dsp_template: Fix undefined shifts [21:11:21 CEST] <durandal_170> working on libswscale can be fun [21:39:31 CEST] <anoop_r> does ffmpeg encodes audio and video simultaniously [21:39:47 CEST] <BBB> in the same process? yes [21:41:39 CEST] <anoop_r> so while encoding some media files ffmpeg takes more time even if all frames are encoded [21:41:43 CEST] <anoop_r> why [21:45:26 CEST] <BBB> I dont understand the question [21:45:44 CEST] <BBB> when you compare something (more time), you need to say what youre comparing to what (what takes more time to encode than what) [21:46:52 CEST] <anoop_r> suppose the sample media have 300 frames [21:47:39 CEST] <anoop_r> the progress shows frame=300 but i have to wait for at least 10 or 20 seconds to complete the process [21:49:00 CEST] <BBB> its the input frame counter [21:49:12 CEST] <BBB> so imagine that I have an encoder that queues frames internally [21:49:15 CEST] <BBB> if I input 300 frames [21:49:18 CEST] <BBB> theyre not all done yet [21:49:24 CEST] <BBB> ffmpeg doesnt know this [21:49:33 CEST] <anoop_r> it's not the progress bar [21:49:35 CEST] <BBB> what are you encoding? [21:49:49 CEST] <BBB> Im not talking about ffmpegs queue, Im taking about the encoder queue [21:49:54 CEST] <BBB> encoders typically have an internal delay [21:49:58 CEST] <BBB> this is for rate control purposes etc. [21:50:07 CEST] <BBB> its quite common [21:50:16 CEST] <anoop_r> i am using x265 [21:50:27 CEST] <BBB> and so even if ffmpeg gave the encoder 300 frames, the encoder is still processing some of these 300 frames internally [21:50:41 CEST] <anoop_r> ok [21:50:46 CEST] <BBB> so its expected [21:50:53 CEST] <anoop_r> anyway to show real progress [22:02:26 CEST] <cone-394> ffmpeg 03Rodger Combs 07master:c190fdf65de8: lavu: Makefile: skip atomic.c if native atomics are available [22:04:20 CEST] <cone-394> ffmpeg 03Paul B Mahol 07master:5165c600ebea: avformat/rtmpproto: increase hardcoded url/path lengths [22:36:08 CEST] <philipl> BtbN: blarghly argle. [22:36:45 CEST] <philipl> the nvenc 444p mode takes packed 444, and ffmpeg doesn't have a packed 444p as far as I can tell. Will need to pack it as part of the copy. :-( [22:37:26 CEST] <JEEBsv> ouch [22:38:05 CEST] <JEEBsv> btw, can you set colorimetry info for the output stream? [22:38:14 CEST] <JEEBsv> or is it hardcoded to BT.709/YCbCr? [22:38:31 CEST] <durandal_170> I can add packed 444... [22:39:08 CEST] <JEEBsv> because if you can set the BGR colorimetry you can actually encode RGB with the 4:4:4 mode [22:39:26 CEST] <philipl> JEEBsv: that would be cute. I'm not sure, but I *think* so. [22:40:04 CEST] <philipl> https://github.com/jp9000/OBS/blob/master/ObsNvenc/inc/nvEncodeAPI.h#L1009 [22:40:07 CEST] <JEEBsv> well, of course you could just override those parameter set values in output in case of hardcoding [22:40:26 CEST] <JEEBsv> needs some parameter set parsing magic, but should be doable [22:40:39 CEST] <philipl> durandal_170: It would seem a reasonable format to support. [22:40:47 CEST] <philipl> I'm surprised it hasn't come up before. [22:41:25 CEST] <Daemon404> nothing uses it [22:41:26 CEST] <BBB> is packed 444p just like rgb24? [22:41:36 CEST] <BBB> yuvyuvyuvyuvyuv...? [22:41:43 CEST] <philipl> BBB: Yes. [22:41:50 CEST] <BBB> I thought we had that... [22:42:01 CEST] <philipl> I'd have thought so too but it's not the pixfmt list. [22:42:28 CEST] <BBB> I guess not [22:42:33 CEST] <BBB> surprising [22:42:41 CEST] <Daemon404> it's not, because nothing practical uses it. [22:43:03 CEST] <philipl> And yet here we are :-) [22:44:04 CEST] <JEEBsv> it really sounds like it was meant for RGB :D [22:44:34 CEST] <philipl> I could tell it accept RGB on our side and see what silliness results... [22:44:34 CEST] <wm4> wouldn't packing it as 32 bit be better? [22:44:53 CEST] <philipl> NV_ENC_BUFFER_FORMAT_YUV444_PL = 0x1000, /**< Planar YUV [YUV separate bytes per pixel] allocated as serial 2D buffer. */ [22:44:59 CEST] <philipl> Hang on. [22:45:04 CEST] <philipl> That's confusing. [22:45:30 CEST] <philipl> That's a contradictory comment. [22:45:33 CEST] <Daemon404> no it isnt [22:45:39 CEST] <Daemon404> it means it's in a contiguous buffer [22:45:46 CEST] <philipl> They call it planar [22:45:48 CEST] <Daemon404> all Y, then all U, then all V. [22:45:52 CEST] <Daemon404> in one 2d buffer. [22:45:53 CEST] <philipl> but describe it as packed. [22:46:05 CEST] <Daemon404> where is it described as packed [22:46:06 CEST] <philipl> NV_ENC_BUFFER_FORMAT_YV12_PL = 0x10, /**< Planar YUV [YUV separate planes] allocated as serial 2D buffer. */ [22:46:18 CEST] <philipl> NV_ENC_BUFFER_FORMAT_IYUV_PL = 0x100, /**< Packed YUV [YUV separate bytes per pixel] allocated as serial 2D buffer. */ [22:46:40 CEST] <philipl> So now I don't know what they mean. [22:46:52 CEST] <Daemon404> uh [22:46:55 CEST] <Daemon404> IYUV is planar as well. [22:47:17 CEST] <Daemon404> IYUV is like YV12, but with the two chroma planes switched. [22:47:34 CEST] <philipl> Which is what I thought. [22:47:40 CEST] <philipl> So I guess I have no idea what these comments mean. [22:47:41 CEST] <nevcairiel> so iyuv == yuv420p [22:48:01 CEST] <philipl> I do know that right now I just get solid green assuming it's planar. [22:48:09 CEST] <philipl> (for black frames) [22:49:27 CEST] <philipl> Ah, who knows. [22:50:40 CEST] <BBB> Daemon404: thats avi/iyuv, you dont know if thats what they meant here [22:50:47 CEST] <BBB> they might just be trolling/confusing [22:51:27 CEST] <Daemon404> note the _PL. [22:52:14 CEST] <durandal_170> yea [22:52:37 CEST] <Daemon404> i feel like its just nvidia being shoddy [22:52:42 CEST] <BBB> see, if only we all commented our code and clearly named our variables like this [22:52:54 CEST] <BBB> we would be flametrolling for the rest of our lives [22:53:49 CEST] <philipl> No doubt. [22:53:55 CEST] <philipl> And the 'documentation' is just as awesome. [22:54:05 CEST] <Daemon404> ... that is the documentation [22:54:10 CEST] <Daemon404> i am looking at nvidia's pdf [22:54:18 CEST] <Daemon404> the comment is a copypaste of that/ [22:54:34 CEST] <BBB> oh so its like googles webm spec [22:54:37 CEST] <BBB> oops [22:54:38 CEST] Action: BBB runs [22:55:08 CEST] <Daemon404> BBB, i was straight up told my one of the managers that the rason for no spec is that it makes it easier for people to sue over patents. [22:55:17 CEST] <Daemon404> reason* [22:56:24 CEST] <jamrial> isn't webm just "take matroska, remove 4/5 of its optional elements and limit it to a handful of codecs"? [22:56:45 CEST] <Daemon404> well i meant vpN [22:56:49 CEST] <Daemon404> s/N/X? [22:56:52 CEST] <Daemon404> fff typos. [22:57:52 CEST] <BBB> jamrial: I meant vpx, yes [22:58:31 CEST] <philipl> Ah, got it. [22:58:35 CEST] <jamrial> ah [22:58:37 CEST] <philipl> You have to separately set the chroma format. [22:59:00 CEST] <philipl> It will happily store high444pp and mark the chroma as 4:2:0. [22:59:08 CEST] <philipl> I got valid output. How cute. [22:59:25 CEST] <philipl> So yes, it's planar. [22:59:36 CEST] <philipl> Thanks, nvidia comments! [23:01:15 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:52c5521877aa: ffmpeg_opt: Favor streams that had packets [23:01:16 CEST] <cone-394> ffmpeg 03Michael Niedermayer 07master:838c5f3df7e9: avformat/utils: Redesign scoring in av_find_default_stream_index() [23:39:23 CEST] <philipl> lossless 444p appears to work. but still need to verify frame correctness. [23:45:06 CEST] <BtbN> philipl, according to some forum posts on the OBS forums, turning on lossless mode with nv12 input seems to work, but produce dodgy output [23:45:23 CEST] <BtbN> To my understanding, it shouldn't work at all [23:57:48 CEST] <Compn> lol wm4 wants to drop winxp? [23:57:50 CEST] <Compn> you fool! [23:58:02 CEST] <Compn> worded differently, i disagree. [23:58:26 CEST] <wm4> look how much I care [23:59:20 CEST] <Compn> i'd have pity for your users if they werent all weeaboos and audiophiles [23:59:28 CEST] <kierank> I also disagree [23:59:33 CEST] <kierank> winxp is still good [23:59:37 CEST] <kierank> and used in many corporate environments [00:00:00 CEST] --- Thu Jul 2 2015

1 0

ffmpeg.log.20150701
by burek 02 Jul '15

02 Jul '15

[04:20:24 CEST] <chocoladedevelop> Hi [04:20:56 CEST] <chocoladedevelop> I'm using c# with ffmpeg and pipes and i'm pushing frames and create in real time compressed mp4 video file. [04:21:26 CEST] <chocoladedevelop> The problem is that after few seconds the ffmpeg.exe memory usage is raising to over 1GB of memory and the hard disk raise up to over 53MB/s [04:21:53 CEST] <chocoladedevelop> Is that a known problem ? [04:25:20 CEST] <chocoladedevelop> I posted my code in csharp in the ffmpeg forum here: http://ffmpeg.gusari.org/viewtopic.php?f=11&t=2204 and also here: http://stackoverflow.com/questions/31149752/why-when-using-ffmpeg-to-create… [05:32:47 CEST] <fred1807> I have to need to always convert mp4 to raw h264 for my application. I use this cmd , ffmpeg -i input.mp4 output.h264 (sometimes with resolution scale included). What would be the best way to automate this process ? Can I create a shell window where I can simply throw my input files over it? [05:50:35 CEST] <godofgrunts> I'm struggling to figure out how to burn .ass subtitles into an mp4. I suspect my problem is that I don't know the name of the .ass file, but I'm not sure how to get it. trac.ffmpeg.org tells me to do ffmpeg -i video.mkv -vf "ass=subtitle.ass" out.mkv but I get the following output http://pastebin.com/eujhxdYx [05:51:19 CEST] <godofgrunts> BTW ffmpeg.pastebin.org can't be resolved for me [05:54:47 CEST] <godofgrunts> as far as the video info in concerned http://pastebin.com/BKbGxGfM [06:35:02 CEST] <chungy> godofgrunts: I think you're going to have to extract it to a separate file first; ffmpeg -i video.mkv -c copy -map 0:2 video.ass [06:36:27 CEST] <chungy> (-c copy is probably unnecessary, but it keeps the ass stream as-is) [06:56:53 CEST] <godofgrunts> chungy, seems to be working, I'll let you know once the video is done [07:01:30 CEST] <godofgrunts> just produces a black screen for some reason [07:19:14 CEST] <Mista_D> Looking for old FFmpeg source files, git version 31792 [07:20:37 CEST] <anoop_r> anybody know how to use x265 multilib with ffmpeg [07:25:00 CEST] <godofgrunts> anoop_r, do you have --enable-libx265? [07:25:18 CEST] <anoop_r> yes [07:25:37 CEST] <anoop_r> anything else needed [07:27:12 CEST] <godofgrunts> Forgive my ignorance, but are you trying to compile ffmpeg or just encode something with x265? [07:27:55 CEST] <anoop_r> x265 mulilib means 8bit and 10 bit builds together [07:28:20 CEST] <anoop_r> so i want to add x265 multilib to ffmpeg [07:28:28 CEST] <godofgrunts> Oh I see. Sorry, I have no idea [07:29:11 CEST] <anoop_r> i too have no idea [07:32:07 CEST] <godofgrunts> anoop_r, this is old and for x264, but maybe it'll help https://ffmpeg.org/pipermail/ffmpeg-user/2013-January/012769.html [08:38:29 CEST] <anoop_r> does ffmpeg encodes audio and video simultaniously ? [08:47:56 CEST] <anoop_r> i had encountered a problem [08:48:12 CEST] <anoop_r> i am using x265 10bit + ffmpeg [08:49:15 CEST] <anoop_r> the encoding process will only finish after some times even after the frame count reaches the maximum farme count [08:49:45 CEST] <chocoladedevelop> ffmpeg.exe have a memory leak ? [08:50:10 CEST] <anoop_r> but for a certain build encoding process outs output file exactly when frames reaches maximum farme of out put [08:50:29 CEST] <anoop_r> file [08:50:40 CEST] <anoop_r> i don't know [08:51:58 CEST] <anoop_r> does ffmpeg have any performance builds [08:52:54 CEST] <anoop_r> it's very hard to explain [08:53:00 CEST] <anoop_r> let me try again [08:53:17 CEST] <anoop_r> see the progress is show as frame= 30 [08:53:36 CEST] <anoop_r> let say the media have 300 frames [08:54:11 CEST] <anoop_r> when frame= 300 it should immediatly complete encoding and output file [08:54:35 CEST] <anoop_r> but its taking some more time to finish [08:54:54 CEST] <godofgrunts> I've had that issue with VP9 encoding as well [08:55:34 CEST] <anoop_r> but a certain previous source was able to provide output at the same time when frame= 300 [08:55:37 CEST] <anoop_r> any idea [08:55:50 CEST] <godofgrunts> I'm 99% sure that ffmpeg does lazy parrel pocessing in that it will x cores to encode video y cores to encode audio, etc [08:56:19 CEST] <godofgrunts> I don't know enough about muxing to be certain, but I feel the delay comes in the remuxing [08:56:43 CEST] <anoop_r> so which version of ffmpeg source will give performance [08:57:22 CEST] <godofgrunts> I'm encoding a video with 3 streams right now and I'm using 6 cores so I assume each stream is assigned two cores [08:57:58 CEST] <godofgrunts> I don't know that to be honest. I'm using static builds from zeranoe [08:58:53 CEST] <anoop_r> which ffmpeg source gives optimum performance [08:59:09 CEST] <godofgrunts> I couldn't tell you. [08:59:43 CEST] <chungy> x265 is slow in general, you'll just have to be patient with it [09:00:07 CEST] <anoop_r> i know that [09:00:26 CEST] <anoop_r> but i got a build which gives zero delay [09:01:00 CEST] <anoop_r> but after using number of updated source i lost it [09:01:18 CEST] <anoop_r> anybody else encountered any such issue [09:02:28 CEST] <pandb> I keep getting the error: "unknown type name AVPixelFormat" and I can't figure out why. I know it's an enum defined in libavutil/pixfmt.h, but including that or avutil.h doesn't seem to make a difference [09:06:17 CEST] <chocoladedevelop> I remember asking about 2-3 years ago about the ffmpeg.exe memory leak but nothing. No one gave a solution or any info. [09:06:24 CEST] <chocoladedevelop> And now i have the same problem. [09:06:57 CEST] <anoop_r> x265-43afbde189f3 [09:07:06 CEST] <chocoladedevelop> Strange it seems i'm the only one with this problem. [09:07:30 CEST] <anoop_r> this commit gives more performance but it will not give output without delays [09:08:29 CEST] <chungy> chocoladedevelop: what leak [09:09:29 CEST] <chocoladedevelop> chungy in c# using pipe to compress batch of images and create mp4 video file in real time make the ffmpeg.exe file memory usage to jump after 3-5 seconds from 0.1MB to over 1GB usage. [09:09:38 CEST] <chocoladedevelop> Also the hard disk usage raise to over 53MB/s [09:10:05 CEST] <chungy> not necessarily a leak. [09:10:08 CEST] <chocoladedevelop> I see it in the Task Manager the ffmpeg.exe just jump in seconds to over 1GB memory usage. [09:10:28 CEST] <chocoladedevelop> What can it be else ? [09:10:36 CEST] <chungy> right, if the uncompressed image data takes over 1GB, then that's what it is... that wouldn't be a leak. [09:11:09 CEST] <chungy> you may want to ask the mailing list though to be certain [09:13:36 CEST] <chocoladedevelop> Ok i will try to ask. But if what you say is true then what can i do ? not sure. [09:13:50 CEST] <chocoladedevelop> How do i get ot the mailing list ? [09:15:17 CEST] <pandb> ahhh 'enum AVPixelFormat...' :) [09:15:49 CEST] <chungy> https://ffmpeg.org/contact.html#MailingLists [09:15:55 CEST] <pandb> im not even mad that everyone was silent :) [09:17:55 CEST] <chocoladedevelop> Thanks, [09:54:15 CEST] <equanox> Hey, i have a question about interlaced encoding [09:55:09 CEST] <equanox> i try to reencode a video from a interlaced source in a mts container [09:55:27 CEST] <equanox> to a interlaced mp4 [09:56:15 CEST] <equanox> after reencoding the metadata of the output file is 'scan_type' is progressive [09:56:57 CEST] <equanox> i'm not able to see the parameter 'scan_type' with ffprobe [09:57:15 CEST] <equanox> but mediaarea shows this meta params [09:57:51 CEST] <equanox> the fottage inside the output mp4 file is still interlaced [09:58:17 CEST] <Mavrik> IIRC that's expected [09:58:37 CEST] <equanox> but due to the metadata winmedia player will not detect it correctly [09:59:29 CEST] <equanox> is there a way to set the metadata manually? [10:03:30 CEST] <Mavrik> equanox, does manually setting field order with "-field_order tt" help? [10:03:46 CEST] <Mavrik> do you explicitly set interlaced output for the encoder? [10:19:08 CEST] <claz> if i try to cut the last frame of a video by seeeking to it's duration it often doesn't work [10:19:17 CEST] <claz> it doesn't error or anything [10:19:24 CEST] <claz> just doesn't create the output file [10:19:35 CEST] <claz> is there any way to cut the last frame without probing to see it's timestamp? [11:20:55 CEST] <equanox> i found a answer to my question [11:20:58 CEST] <equanox> -flags +ildct+ilme -x264opts tff=1 [11:21:02 CEST] <equanox> is doing the job [11:21:18 CEST] <Mavrik> ah [11:21:24 CEST] <Mavrik> so you were actually encoding a progressive video [11:21:30 CEST] <equanox> the problam is . mp4 does not traditionall interlaced [11:21:54 CEST] <equanox> instead they use MBAFF [11:22:28 CEST] <equanox> no the video was interlaced [11:23:01 CEST] <JEEBsv> if it was interlaced and your source noted it correctly, it should be encoded through interlaced [11:23:14 CEST] <JEEBsv> MBAFF is interlaced if the tag is there in the input bit stream [11:23:25 CEST] <JEEBsv> both PAFF and MBAFF should come up as interlaced in case it is interlaced [11:23:36 CEST] <JEEBsv> if it is not, well then you can blame the creator of the original file :P [11:23:40 CEST] <Mavrik> ffmpeg doensn't do PAFF at all afaik [11:23:43 CEST] <Mavrik> or was it x264? [11:23:45 CEST] <JEEBsv> x264 [11:23:56 CEST] <JEEBsv> since the benefits of PAFF are like none [11:24:08 CEST] <JEEBsv> and the amount of work needed to add PAFF coding is on the other hand... [11:28:20 CEST] <equanox> the sources "scan_type" was named as interlaced [11:28:58 CEST] <equanox> but the output files "scan_type" was progressive (but containing the interlaced video) [11:29:09 CEST] <equanox> yes the encoder was x264 [11:34:27 CEST] <JEEBsv> how old is your ffmpeg? [11:35:04 CEST] <equanox> only a few months old [11:36:13 CEST] <JEEBsv> ffmpeg -version ? [11:37:30 CEST] <equanox> ffmpeg version N-69972-g6c91afe Copyright (c) 2000-2015 the FFmpeg developers [11:37:37 CEST] <Mavrik> equanox, did you set the field type flag? [11:37:48 CEST] <JEEBsv> Mavrik: shouldn't it grab that from the decoded picture? [11:38:02 CEST] <JEEBsv> unless ffmpeg is retarded and will not take that into mention when passing things into libx264 [11:38:16 CEST] <Mavrik> When in doubt, I rely on ffmpeg being retarded [11:38:25 CEST] <Mavrik> I see that it reads the field_type AVContext in movenc.c [11:38:31 CEST] <equanox> i tried setting the field type flag [11:38:36 CEST] <Mavrik> But I'm not entirely sure if it can reliably set that on H.264 [11:38:36 CEST] <JEEBsv> movenc!? [11:38:43 CEST] <equanox> but it doesn't had any effect [11:38:46 CEST] <JEEBsv> the container should have nothing to do with it [11:38:52 CEST] <JEEBsv> it's a bit stream thing [11:39:18 CEST] <Mavrik> Unless you want the container flag set :) [11:39:56 CEST] <JEEBsv> no, I mean the decoder should output pictures with the interlacism flag set, and then libx264 should be passed that information [11:40:02 CEST] <JEEBsv> (unless ffmpeg is retarded enough not to do that) [11:40:25 CEST] <JEEBsv> if the decoded picture is not interlaced then of course there's no way to pass that information :P [11:40:32 CEST] <RobertNagy> Does ffmpeg somehow encode gapless meta-data for aac in mp4? i.e. front padding, end padding and real samples? [11:41:28 CEST] <JEEBsv> RobertNagy: if the encoder outputs negative PTS at first, that will be written in the mp4 file [11:41:33 CEST] <RobertNagy> similar to how itunes uses "iTunSMPB" meta headers. I can see that ffmpeg parses "iTunSMPB" but I can't see any such data being written when muxing. [11:41:36 CEST] <JEEBsv> as that would be the encoder delay [11:41:48 CEST] <JEEBsv> yeah, ffmpeg uses standard stuff in the container [11:42:01 CEST] <RobertNagy> so how would I get those 3 values? [11:42:15 CEST] <RobertNagy> front padding = 0 - first pts? [11:42:31 CEST] <RobertNagy> real samples = duration? [11:42:34 CEST] <RobertNagy> end padding = ??? [11:42:42 CEST] <JEEBsv> there shouldn't be any padding at the end [11:42:49 CEST] <JEEBsv> the stream has duration X and then you have the encoder delay [11:43:09 CEST] <RobertNagy> I assume the encoder delay is not included in the duration? [11:43:37 CEST] <JEEBsv> I don't exactly remember the way ffmpeg outputs the stuff, but IIRC it does take it into mention [11:44:02 CEST] <JEEBsv> the encoder delay is for the encoder's padding that's in the front, basically [11:44:20 CEST] <JEEBsv> so anything sane will not play those samples [11:44:24 CEST] <JEEBsv> (usually silence) [11:44:36 CEST] <JEEBsv> and then ffmpeg doesn't do by itself any other checking of what you feed to it [11:44:44 CEST] <JEEBsv> so it's the stream length and that's it [11:44:50 CEST] <Mavrik> JEEBsv, it does, but equanox is complaining about MP4 file flags, not the actual video - he is saying that it's indeed interalced :) [11:44:53 CEST] <RobertNagy> yea, I need to know the padding for playback using Media Extensions in Chrome [11:45:19 CEST] <JEEBsv> Mavrik: of course, but he has also not given any info [11:45:40 CEST] <JEEBsv> but as I noted, I have no idea if ffmpeg does the Right Thing [11:45:49 CEST] <JEEBsv> it could have some retarded logic that encodes everything as progressive [11:46:04 CEST] <JEEBsv> but on the other hand, it could just be that his bit stream is actually flagged as progressive [11:46:47 CEST] <RobertNagy> JEEBsv: actually there is one problem. If the media doesn't start a pts=0 I have no way of knowing the padding. [11:47:08 CEST] <RobertNagy> Though, I could guess it based on the previous segment. Maybe that would work. [11:51:49 CEST] <RobertNagy> Actually, ffprobe gives me 'start: 0.000000', which can't be right... [11:56:00 CEST] <RobertNagy> will have to use afconvert for now then... [11:56:43 CEST] <JEEBsv> I only know that since around 2013 encoders that actually give negative PTS should have their encoder delay stored correctly [11:56:47 CEST] <JEEBsv> not how it's shown [12:07:04 CEST] <RobertNagy> so how can I read it? [12:07:22 CEST] <RobertNagy> bruteforce with ashowinfo? [12:10:25 CEST] <JEEBsv> I will probably test later and check how it's shown [12:10:43 CEST] <JEEBsv> for API users it should be rather simple, your decoded audio packets would have a negative PTS until the encoder delay finishes [12:11:11 CEST] <RobertNagy> yea, it just feels overkill to create a whole new application just for this [12:11:26 CEST] <RobertNagy> and you are sure there is no end padding with libfdk_aac? [12:11:30 CEST] <RobertNagy> nvm [12:11:36 CEST] <RobertNagy> use duration... [12:11:48 CEST] <RobertNagy> given that the duration is sample accurate... [13:13:25 CEST] <theeboat> Hello, im using ffmpeg to capture video from a multicast stream. Does anybody know if its possible to force a black frame when there are no frames from the multicast stream? [14:26:39 CEST] <zhanshan> hi [14:27:12 CEST] <zhanshan> can mkv take wav format? and If input is 32bit float what should I use to make the resulting file properly playable? [14:28:11 CEST] <zhanshan> given command line is: $ ffmpeg -f image2 -pattern_type glob -i '*.jpg' -i 'sound-in.wav' -pix_fmt yuv420p -c:v libx264 -crf 18 -x264opts "keyint=50" -c:a wav 'output.mkv' [14:30:45 CEST] <Mavrik> zhanshan, you probably mean PCM not WAV [14:30:54 CEST] <Mavrik> raw audio right? [14:31:38 CEST] <zhanshan> it's lossless wav 32bit float [14:31:54 CEST] <zhanshan> I didn't know wav is called "raw" [14:32:08 CEST] <zhanshan> but if that's so I've learned something new [14:32:36 CEST] <Mavrik> huh [14:32:43 CEST] <Mavrik> WAV is a container [14:32:51 CEST] <Mavrik> that contains non-compressed audio [14:33:02 CEST] <Mavrik> and non-compressed audio is usually referred to as PCM [14:33:27 CEST] <zhanshan> maybe I should use -c:a pcm_s16le ? [14:33:33 CEST] <Mavrik> if you have 32-bit float [14:33:41 CEST] <Mavrik> you should use the format that sets that :) [14:33:44 CEST] <Mavrik> pcm_f32le [14:33:57 CEST] <Mavrik> DEA..S pcm_f32le PCM 32-bit floating point little-endian [14:34:02 CEST] <zhanshan> original recording it was 24bit/48KHz [14:34:32 CEST] <zhanshan> that'd make pcm_s24le ? [14:34:37 CEST] <zhanshan> does that exist? [14:34:58 CEST] <Mavrik> yes [14:35:04 CEST] <Mavrik> look at ffmpeg -codecs [14:37:06 CEST] <zhanshan> so many!! [14:37:19 CEST] <zhanshan> 4 different pcm_s24 [14:37:38 CEST] <zhanshan> be, le, daud, le_planar [14:37:45 CEST] <zhanshan> what's the meaning of all that [14:37:51 CEST] <godofgrunts> btw, mkv is the only container that can hold pretty much anything. [14:37:52 CEST] <zhanshan> so many choices [14:38:01 CEST] <zhanshan> godofgrunts thanks for that! [14:38:04 CEST] <chungy> if your input file is a *.wav, you can just use "-i file.wav -c:a copy" [14:38:07 CEST] <zhanshan> I love mkv I guess [14:39:20 CEST] <zhanshan> chungy that's a good one, too, but since I drove a bit too far by exporting the audio from blender to 32bit float [14:39:55 CEST] <zhanshan> and the recorded source was "only" 24bit I want to keep it was little as possible with the about "original" quality [14:40:21 CEST] <JEEBsv> well if it's in float you did something to it, right? [14:40:23 CEST] <chungy> fwiw, flac is usually a good choice. It's lossless but often is much smaller than the raw stream :P [14:40:45 CEST] <JEEBsv> in that case you've already derped it and you should not decide anything on the original bit depth [14:40:57 CEST] <zhanshan> blender doesn't seem to have to ability to export audio in 24bit? [14:41:00 CEST] <JEEBsv> it's the same issue as with people wanting to re-encode shit with the same bit rate :P [14:41:09 CEST] <zhanshan> so what should I've done to do it right? [14:41:22 CEST] <zhanshan> export it to 32bit signed or 16bit signed instead? [14:41:28 CEST] <JEEBsv> whatever you goddamn like :) [14:41:34 CEST] <zhanshan> to be as close as possible to 24bit signed? [14:41:45 CEST] <zhanshan> JEEBsv that's a good hint, too [14:42:03 CEST] <chungy> pcm_s24le is as close as possible to 24-bit. [14:42:21 CEST] <JEEBsv> unless you did only lossless changes to the source, then just don't try to decide something through 'the original bit depth was this' [14:42:25 CEST] <zhanshan> JEEBsv "same" in that case I guess means to not loss too much of the original holyness of sound and of originality [14:42:29 CEST] <zhanshan> :P [14:42:38 CEST] <JEEBsv> well you already did some filtering to it, right? otherwise it wouldn't be in float [14:42:48 CEST] <JEEBsv> thus that is now a completely different source [14:42:54 CEST] <zhanshan> chungy yes that's right for ffmpeg, but in blender I didn't find that export option [14:43:06 CEST] <theeboat> Hello, im using ffmpeg to capture video from a multicast stream. Does anybody know if its possible to force a black frame when there are no frames from the multicast stream? [14:43:08 CEST] <zhanshan> that's why I shouldn't export anything from blender I guess [14:43:16 CEST] <zhanshan> but use Ardour instead [14:43:22 CEST] <chungy> do you have your original source? [14:43:33 CEST] <zhanshan> but that workflow didn't establish in my work routine so far [14:43:46 CEST] <zhanshan> I don't really know how to connect both programs [14:44:02 CEST] <zhanshan> I can use/show video in Ardour and I can show audio in Blender.. [14:44:43 CEST] <zhanshan> chungy the original source are maybe 15 different files and they've been edited (i.e. cut, fading etc.) in blender [14:44:54 CEST] <zhanshan> that's why I don't like to start from the beginning right now [14:45:05 CEST] <zhanshan> chungy yes I have them [14:47:43 CEST] <zhanshan> JEEBsv: "well you already did some filtering to it, right? otherwise it wouldn't be in float" what has happened there. is it something bad that happens if I work and save like: signed->float->signed? [14:47:46 CEST] <zhanshan> more noise? [14:49:34 CEST] <zhanshan> chungy I like flac a lot but many programs can neither show the length of the flac-file, nor seek them. That's why this time I chose wav_pcm. I have to check the lengths in time of the files easily [14:50:05 CEST] <chungy> should be irrelevant inside a mkv container [14:52:49 CEST] <zhanshan> chungy yes, but I have to re-check the files before encoding [14:55:32 CEST] <c_14> zhanshan: don't work with float if you don't have to [14:57:06 CEST] <zhanshan> c_14 all this knowledge about the different wav forms really goes over my understanding right now. my head is full of video editing, scenes, composition... [14:57:50 CEST] <c_14> If you're working with wav, it's pretty simple. Stay away from float, don't decrease the bit depth, and try not to mess around with the samplerate. [14:58:35 CEST] <zhanshan> so keep inside the same specs for all time.. [14:58:56 CEST] <c_14> If you do that your audio should stay lossless. [14:59:25 CEST] <zhanshan> technically that sounds resonable [14:59:59 CEST] <zhanshan> but i.e. I recorded files in 24bit signed, blender can only export 16s or 32f/s [15:00:01 CEST] <zhanshan> or flac [15:00:11 CEST] <zhanshan> then I should choose preferrably flac? [15:00:55 CEST] <c_14> IF you care about it staying lossless 32s or flac. [15:01:14 CEST] <c_14> You probably won't notice the difference between 24bit and 16bit signed pcm though [15:07:36 CEST] <zhanshan> thanks, I appreciate you answer! [15:20:52 CEST] <fiacha> Hi, question, is it possible to process/request only the video of an RTP input stream? I have an IP cam and I want to dump the video but it also sends audio with lots of "Non-monotonous DTS in output stream" errors even though it doesn't have a mic [15:28:29 CEST] <godofgrunts> I don't have any experience with rtp streams, but can you not just pass -an? [15:50:37 CEST] <godofgrunts> fiacha, I just tried it with an rtp stream and it worked for me [17:07:23 CEST] <fiacha> godofgrunts: yes, -an works, thank you [20:52:42 CEST] <fred1807> why some files I convert from mp4 to h264 result in bigger files , and sometimes result in smaller sizes? [20:52:47 CEST] <fred1807> from .mp4 to raw .h264 [20:53:25 CEST] <fred1807> and why mediainfo cannot read bitrate information from raw .h264 I create with ffmpeg? [20:59:51 CEST] <JEEBsv> well, it probably doesn't want to parse all of the pictures to get a picture count, and then try to note if there's any frame rate kind of flag around [21:00:52 CEST] <JEEBsv> because you would have to a) calculate how many pictures there are all in all b) get all the possible information on how fast the pictures are supposed to be shown c) use the file size of the raw stream on that [21:01:01 CEST] <JEEBsv> anyways, I wouldn't trust mediainfo too much on various stuff [21:01:14 CEST] <JEEBsv> you have to know how mediainfo comes to various conclusions to trust it [21:01:33 CEST] <JEEBsv> because then you know what kind of information or guesses it is giving to you [21:22:29 CEST] <Bleakwise> can someone tell me where ffdshow's avisynth plugin folder is by default? [21:22:57 CEST] <JEEBsv> aand that has nothing to do with FFmpeg [21:33:14 CEST] <anoop_r> hi [21:34:58 CEST] <anoop_r> ffmpeg -i hd.mp4 -strict experimental -c:v libx265 -c:a -c:a libfdk_aac -b:a 320k output.mkv [21:35:34 CEST] <anoop_r> if i use like this , does the video and audio is encoded concurrently [21:35:39 CEST] <DHE> why did you put "-c:a" twice? [21:35:46 CEST] <anoop_r> sorry [21:35:54 CEST] <anoop_r> ffmpeg -i hd.mp4 -strict experimental -c:v libx265 -c:a libfdk_aac -b:a 320k output.mkv [21:37:26 CEST] <anoop_r> does this encode audio and video simultaniously ? [21:39:49 CEST] <DHE> you mean multi-threaded? [21:40:09 CEST] <anoop_r> yes [22:04:05 CEST] <Nolski> Hey, if I'm using the overlay filter and I want the overlaid video to start 30 seconds into the input video do I just do ffmpeg -ss 00:00:30 -i input.mkv -complex_filter "overlay filter goes here" out.mkv? [22:04:30 CEST] <Nolski> or do filters not really take into account whether you scan to a position or not? [22:05:16 CEST] <c_14> Do you want the overlay to not have the first 30 seconds of video, or do you want the overlay to show up 30 seconds into the main video? [22:06:20 CEST] <Nolski> show up 30 seconds into the video [22:06:26 CEST] <Nolski> c_14: ^ [22:06:42 CEST] <c_14> use the enable argument to the overlay filter [22:06:55 CEST] <c_14> https://ffmpeg.org/ffmpeg-filters.html#Timeline-editing [22:07:16 CEST] <Nolski> ah right, I always forget about the enable argument since it's never right under the filter options in the docs [22:07:21 CEST] <Nolski> thanks :) [23:35:37 CEST] <RobertNagy> I'm a bit confused by ffmpeg while encoding opus into webm. I have a wav file with exactly 240000 samples @ 48kHz. When i encoding it: ffmpeg -i 0.wav -c:a opus 0.webm. The resulting file gives me the following information: Duration: 5.01, Start 0.007, discard 648/900 samples, 240312 samples decoded. [23:36:04 CEST] <RobertNagy> If I have several of these files how would I play them seamlessly without gaps? [23:37:12 CEST] <RobertNagy> i.e. in a browser I've tried: sourceBuffer.timestampOffset = 5 * n - 0.007; sourceBuffer.appendWindowStart = 5 * n; sourceBuffer.appendWindowEnd = 5 * (n+1); [23:37:22 CEST] <RobertNagy> however, there are audible gaps. [23:37:52 CEST] <RobertNagy> How many samples am I actually supposed to discard? 0.007 * 48000, 648, or 240312 - 240000? [23:40:49 CEST] <adamellsworth> can ffmpeg output SVG files? [23:41:13 CEST] <c_14> no [00:00:00 CEST] --- Thu Jul 2 2015

1 0